Copyright 2003 by Jeremy Silman and Pal Benko

All rights reserved. No part of this book may be used

or reproduced in any manner whatsoever without written permission from
the publisher, except in the case of brief quotations embodied in critical articles
and reviews.
First Edition
10 9 87 6 5 4 3 2 1

Library of Congress Cataloging-in-Publication Data

Benko, Pal, 1928Pal Benko: my life, games, and compositions / by Pal Benko and Jeremy Silman;
with an opening survey by John Watson. -- 1st ed.
p.cm.
Includes indexes
1. Benko, Pal, 19282. Chess players--Hungary--Biography.
2. Chess--Collection of games.
I. Silman,Jeremy. II. Watson,John L. III. Title
GV1439.B43 2003

794.1'092-dc22 [B]

2003061556

ISBN: 1-890085-08-1

Cover design by Heidi Frieder/Art Head

Cover photograph by Demeter Balla

Printed and bound in the United States of America

SHes Press
3 624 Shannon Road
Los Angeles, CA 9002 7

To

my wife Gizella and my children Palma and David.

Foreword by Susan Polgar

Preface by Jeremy Silman
Acknowledgements XV1tt
Introduction XIX

X11l
xv

PART ONE

MY LIFE AND GAMES

1 IN THE BEGINNING 3
( 1 ) T Feldman - Benko (Budapest, 1 945)
(2) Szigeti - Benko (Budapest, 1 945)

(3) G Barcza - Benko (Hungarian Championship, 1 946)

13

(4) Benko - Szilagyi (Hungarian Championship, 1 946)

2 AFTER THE WAR 19
(5) Tagmann - Benko (Correspondence, 1 947)
(6) Benko - Lounek (Vienna, 1 947)

25

(7) Foltys - Benko (Budapest, 1 948)

28

(8) Polzer - Benko (Bad Gastein, 1 948)

15

20

32

(9) Benko - Fiister (Hungarian Championship, 1 948)

35

( 1 0) Benko - Ragozin (Budapest vs. Moscow Match, Moscow, 1 949)

( 1 1 ) Prins - Benko (Hungary vs. Netherlands Match, 1 949)
( 1 2) Benko - Opocensky (Bucharest, 1 949)
(1 3) Benko - Turi (Budapest, 1 949)

39

43

45

3 PROVING MYSELF AT THE MAROCZY MEMORIAL 47

( 1 4) Benko - Dr. Ogaba (Hungarian Championship, 1 950) 47
(1 5) Benko - I Molnar (Hungarian Championship, 1 950)
( 1 6) Korody - Benko (Budapest, 1 9 5 1 )

51

53

( 1 7) Szabo - Benko (Hungarian Championship, 1 9 5 1 )

( 1 8) Benko - Petrosian (Budapest, 1 952)

59

( 1 9) Botvinnik - Benko (Budapest, 1 952)

62

(2 0) Benko - Stahlberg (Budapest, 1 952)

64

55

4 FALL FROM GRACE (THE DARK YEARS)

67
(2 1 ) G Barcza - Benko (Hungarian Championship, 1 954)
(2 2) Benko - Kluger (Budapest, 1 95 5)

75

(2 3) Portisch - Benko (Hungarian Championship, 1 95 5)

VI I

71
78

37

PAL B E N KO: MY LI FE, GAMES A N D COMPO S ITI O N S

5 ESCAPE TO FREEDOM! 81
(24) Benko - Giustolisi (Dublin, 1 9 5 7)
VIII

82

6 THE NEW WORLD 85

(2 5) Benko - Fischer (Portoroz Interzonal, 1 9 5 8)

89

(26) Benko - Averbach (Portoroz Interzonal, 1 95 8)

92

(2 7) R Weinstein - Benko (U.S. Championship, 1 959)

96

(2 8) Benko - F Olafsson (Zagreb Candidates Tournament, 1 959)

(2 9) Benko - Smyslov (Zagreb Candidates Tournament, 1 959)

98
100

(3 0) F Olafsson - Benko (Zagreb Candidates Tournament, 1 959)

(3 1 ) Benko - Fischer (Buenos Aires, 1 960)

102

106

(32) Benko - Eliskases (Buenos Aires, 1 960)

110

(3 3) Benko - Taimanov (Buenos Aires, 1 960)

113

(34) Uhlmann - Benko (Buenos Aires, 1 960)

115

(3 5) Benko - Reshevsky, 1 st Match game (New York, 1 960)

118

120

(3 6) R Byrne - Benko (U.S. Championship, 1 960)

(3 7) Benko - Mednis (U.S. Championship, 1 96 1 )

123

7 THE HUNT FOR THE WORLD CHAMPIONSHIP 125

(3 8) Benko - German (Stockholm Interzonal, 1 962) 128
(39) Benko - Uhlmann (Stockholm Interzonal, 1 962)
(40) Benko - Gligoric (Interzonal Playoff, 1 962)

131

135

(41 ) Benko - Fischer (Candidates Tournament Curaao, 1 962)

(42) Benko - Keres (Candidates Tournament Curaao, 1 962)

138
142

(43) Benko - Korchnoi (Candidates Tournament Curaao, 1 962)

(44) Benko - Tal (Candidates Tournament Curaao, 1 962)
(45) Filip - Benko (Candidates Tournament Curaao, 1 962)
(46) Benko - Darga (Varna Olympiad, 1 962)

156

158

(47) Benko - Pachman (Varna Olympiad, 1 962)

8 A PROFESSIONAL AT LAST 163
(48) Benko - Najdorf (Piatigorsky Cup, 1 963)
(49) Benko - Panno (Piatigorsky Cup, 1 963)
9 BACK TO HUNGARY 173
(50) Benko - R Byrne (New York, 1 964)

151 '

160
165
168

174

(5 1 ) Benko - Bisguier (New York playoff, 1 st Game, 1 964)

176

(52) Benko - Bisguier (New York playoff, 3 rd Game, 1 964)

179

(5 3 ) Bisguier - Benko (New York playoff, 4th Game, 1 964)

181

(54) Benko - Bisguier (New York playoff, 5th Game, 1 964)

183

(55) Benko - Suttles (Boston, 1 964)

185

(56) Benko - Matanovic (Belgrade, 1 964)

(57) Benko - Parma (Belgrade, 1 964)

190

190

145

CO NTE NTS

(58) Benko - Darga (Tel Aviv Olympiad, 1 964)

194

(59) Benko - Tringov (Tel Aviv Olympiad, 1 964)

196

1 0 TRAVEL, TRAVEL, AND MORE TRAVEL 199

(60) Benko - R Byrne (U. S. Open, 1 965) 200
(6 1 ) Lombardy - Benko (Seattle, 1 966)

IX

203

(62) Benko - Bisguier (U.S. Championship, 1 966)

(63 ) Benko - Medina (Havana Olympiad, 1 966)

206
209

(64) Bisguier - Benko (Manhattan Club Championship, 1 967)

(65) Benko - Bogdanovic (Sarajevo, 1 967)

213

(66) Benko - Diickstein (Sarajevo, 1 967)

216

(67) Mestrovic - Benko (Sarajevo, 1 967)

218

(68) Bisguier - Benko (USA, 1 967)

220

(69) Benko - Kagan (Winnipeg, 1 967)

223

(70) Benko - Larsen (Winnipeg, 1 967)

226

(7 1 ) Benko - Matanovic (Winnipeg, 1 967)

229

1 1 CREATING THE BENKO GAMBIT 233

(72) Vukic - Benko (Sarajevo, 1 967) 234
(7 3 ) Gross - Benko (U.S. Open, 1 968)
(74) Benko- Berry (Canada, 1 9 7 1 )

236

238

(75) Portisch - Benko (Palma de Mallorca, 1 97 1 )

(76) Vranesic - Benko (Toronto, 1 97 1 )

242

(77) Aspler - Benko (Vancouver, 1 97 1 )

245

(78) Camara - Benko (Sao Paulo, 1 97 3 )

247

(79) Gligoric - Benko (Lone Pine, 1 975)

249

(80) I Zaitsev - Benko (Szolnok, 1 97 5)

251

(8 1 ) Kovacs - Benko (Debrecen, 1 97 5)

255

(82) Gordon - Benko (National Open, 1 976)

1 2 EUROPE AT MY FEET 261
(8 3) Benko - R Byrne (USA, 1 968)
(84) Benko - Formanek (USA, 1 968)

240

258

262
265

(8 5) Benko - Horowitz (U.S. Championship, 1 968)

267

(86) Benko - Reshevsky (U.S. Championship, 1 968)

(87) Benko - Ostojic (Wijk aan Zee, 1 969)
(88) Benko - Tatai (Malaga, 1 969)

274

277

(89) Burger - Benko (U.S. Championship, 1 969)

(90) R Byrne - Benko (USA Championship, 1 969)
(9 1 ) Benko - Lengyel (Venice, 1 969)
(92) Benko - D Byrne (Vrsac, 1 969)

270

284
285

(93) Benko - Geller (Wijk aan Zee, 1 969)

288

279
281

211

PAL BEN KO: MY LI FE, GAM ES A N D C O M POSITI O N S

(94) Benko - Sigurjonsson (Caracas, 1 970)

291

(95) Benko - Donner (Wijk aan Zee, 1 970)

294

(96) Benko - Platonov (Wijk aan Zee, 1 970)

(97) Durao - Benko (Malaga, 1 970)

296

300

(9S) Benko - Popov (Reggio Emilia, 1 970)

303

(99) Benko - Czerniak (Reggio Emilia, 1 970)

( 1 00) Bukal- Benko (Sarajevo, 1 970)

303

305

( 1 0 1 ) Benko - Jansa (Siegen Olympiad, 1 970)

( 1 02) Benko - Kavalek (U.S. Open, 1 970)

307

309

( 1 03) Benko - Csom (Palma de Mallorca, 1 97 1)

312

( 1 04) Benko - Barczay (Vrnjacka Banja, 1 97 1)

315

( 1 05) Benko - Dominguez (Las Palmas, 1 972)

318

( 1 06) Benko - Menvielle (Las Palmas, 1 972)

( 1 07) Benko - Pomar (Las Palmas, 1 972)

318

321

( l OS) Benko - Planinc (Novi Sad, 1 972)

323

( 1 09) Benko - Emma (Skopje Olympiad, 1 972)

325

( 1 1 0) Benko - Malich (Skopje Olympiad, 1 972)

329

( 1 1 1 ) Benko - Bellon (Malaga , 1 97 3 )

331

( 1 1 2) Benko - Pomar (Orense, 1 97 3 )

335

( 1 1 3) Martz - Benko (Torremolinos, 1 97 3 )

337

( 1 1 4) Benko - Browne (U.S. Championship, 1 97 3 )

( 1 1 5) Kane - Benko (U.S. Championship, 1 97 3 )
( 1 1 6) Martz - Benko (U. S. Championship, 1 97 3 )
( 1 1 7) Benko - Janosevic (Vrnjacka Banja, 1 97 3 )
( 1 1 S) Basman - Benko (Hastings, 1 97 3 )
( 1 1 9) Benko - Pytel (Hastings, 1 97 3 )

343
344
348

352

354
356

1 3 RETIREMENT 361
( 1 2 0) Benko - Commons (U.S. Championship, 1 974)
( 1 2 1 ) Benko - Evans (U.S. Championship, 1 974)
( 1 22) Benko - Gilden (U.S. Championship, 1 974)
( 1 2 3) Benko - Cosulich (Venice, 1 974)

371

( 1 24) Benko - Kushnir (Lone Pine, 1 975)

( 1 2 5) Benko - Pinter (Szolnok, 1 97 5)

373

376

( 1 2 6) Benko - Kopec (World Open, 1 975)

( 1 2 7) Benko - Krnic (Sombor, 1 976)

378

381

( 1 2 S) Benko - Brasket (Lone Pine, 1 977)

383

( 1 2 9) Benko - Chavez (Sao Paulo, 1 977)

387

( 1 3 0) Benko - Peters (Lone Pine, 1 975)

( 1 3 1 ) Benko - Seirawan (Lone Pine, 1 975)

390
391

362

366
368

CO NTE NTS

394

( 1 3 2) Fedorowicz - Benko (Lone Pine, 1 979)

(1 3 3) Zaltsman - Benko (USA, 1 983)

396

( 1 3 4) Benko - Trosclair (New York, 1 984)

399

( 1 3 5) Benko - Harrison (Australian Open, 1 985)

( 1 3 6) Benko - Griinfeld (New York, 1 986)

XI

401

404

( 1 3 7) Benko - Garcia (New York Open, 1 988)

( 1 3 8) Benko - F Roeder (Augsburg, 1 99 1 )

406

408

PART TWO

INTERVIEWS by Jeremy Silman

Interview with National Master RONALD GROSS

Interview with Grandmaster LARRY EVANS
Interview with Grandmaster PAL BENKO

413

421
425

PART THREE

PAL BENKO'S CREATIVITY, AN OPENING SURVEY by John

Watson

INTRODUCTION 435
BENKO AS WHITE 439
The English Opening 439
Flank Openings with the b-pawn: l .b3 and 1 .b4
Benko System: l .g3

464

King's Indian Attack

King Pawn: l .e4

463

465

468

d-pawn Openings

484

BENKO AS BLACK 497

The English Opening 498
Flank Openings

503

l .e4 Openings

506

l .d4 Openings

533

PART FOUR

CHESS PROBLEMS AND COMPOSITIONS

INTRODUCTION
567
TWO MOVE MATES 569
THREE- (AND MORE) MOVE MATES
LETTER PROBLEMS 593
ENDGAMES 607
HELPMATES 639
PUZZLES 653

581

A Selection of Pal Benko's Significant Tournaments and Matches

Index of Names 661
Index of Openings 668

659

Standing: Klara Po lgar, Pal Benko, S u san Polgar, L. H azai , M. Movay.

Sitti ng: M. Mari n ,Ju dith Polgar, Sofie Po lgar.

drnast" Pol Benko i, a living legend. AlwaY' a man that did thinl!'
his own way, Benko learned chess from his father at the very late age of ten but
didn't take part in his first real tournament until the age of sixteen. Starting seri
ous play at such an advanced age is unheard of among chess professionals today,
yet he remarkably won the Hungarian National Championship a few short years
later at twenty.
It's clear that he possessed an enormous natural talent. Unfortunately, his
progress came to a sudden stop when he was interned in a hard-labor camp for
trying to escape to the West-his political views did not match the views of the
Communist regime. Mter many hardships, at the age of twenty-nine, Pal did
indeed seek, and get, political asylum in the United States.
Benko feels that he missed out on his best chess years during his struggle for
freedom, yet this didn't prevent him from making his presence felt once he settled
in his new country: He was a candidate for the World Championship in 1 959
and 1 962 , he won the U.S. Open an unparalleled eight times, he had an out
standing record as a player and as a Team Captain in several Chess Olympiads,
he was a world-class chess composer, a world-class chess theoretician, one of the
world's great endgame authorities, and an award-winning chess author. These
tremendous accomplishments got him a trip to the Chess Hall of Fame-he was
inducted in 1 993 .
All these things are very impressive, yet one of his biggest gifts to the United
States is somewhat forgotten today. In 1 970, Pal Benko gave up his earned spot
in the World Championship Interzonal tournament to Bobby Fischer so that
Bobby could compete for the title. This led to one of the most dramatic events
in sports history when Fischer defeated Spassky in 1 972 and became Chess Cham
pion of the World. This would not have happened without Benko's sacrifice,
and his unselfish act clearly demonstrates his character and personality.
I myself had the good fortune to meet Grandmaster Pal Benko more than
twenty years ago. One of my first memories of him was at the Hungarian
XI I I

PAL B E N KO: MY LI FE, GAM ES AN D COMPO SITI O N S

XIV

National Team Championship when he represented one of our rival teams. I

vividly recall that, after his game was finished, he demonstrated his latest com
position to the players and spectators. A big crowd was surrounding his table,
yet no one was able to find the solution. This was the highlight of the day dur
ing that match.
Grandmaster Benko was one of the very few people who were willing to help
my sisters and I for no charge whatsoever. At that time he lived half the year in
the United States and half in Hungary, yet whenever he came to Budapest he
would visit our family. Pal played countless blitz games with the three of us and,
even though he was well beyond his prime, he was still a fierce, energetic com
petitor at this quick form of chess.
Benko always motivated us with his enthusiasm and love for the game. He
often shared his incredible chess compositions-most were in their final ver
sions, but some were still in various stages of development and we would fine
tune them together. I remember how much pleasure he got from showing these
amazing creations to other top-level players, and this pleasure turned to pure
joy whenever they couldn't find the answers.
Possessing a wonderful sense of humor, Pal loved to tell us jokes. We also
enjoyed listening to his accounts of other world-class players-he knows them
all. One of his most fascinating subjects was his close friend Bobby Fischer, and
we would never tire of hearing about their adventures together.
In 1 994, Benko was chosen as Captain of the Silver Medal Hungarian Woman's
Olympiad team that consisted of two of the three Polgars (Sophia and me). He
was again the Captain of the 1 996 Hungarian Women's Olympiad Team with
Sophia.
Aside from chess, Benko also has a deep love of mathematics. Both of his chil
dren and his wife play chess and all three of them are professors in mathematics.
One of his quirkier hobbies is arm wrestling. Pal, now over seventy, loves to
challenge athletic young men to arm wrestling duels, beating them consistently.
I suppose this, plus his jet-black hair and playful disposition, allows him to dem
onstrate his enormous vitality and eternal youthfulness.
Seen by his fans as being a skilled teacher, player, writer, composer, and endgame
specialist, Pal Benko the person is, more than anything else, a wonderful human
being and a real gentleman. Always a big supporter of the Polgar sisters, Benko
has helped us in many ways throughout the years. I am deeply indebted to this
great man for the many kindnesses he has shown to my family.
This book about Grandmaster Benko's life, games and compositions is long
overdue. International Master Jeremy Silman is a famed chess author and player
himself. When Pal Benko and Jeremy Silman collaborate together, chess players
around the world can only benefit a great deal. This makes this book a must
have in any chess player's library.
Susan Polgar, World Champion and Olympic Champion
New York, 2 003

ung player, I came to know Pal Benkn though the pages of

Chess Life magazine. His games, his dealings with Fischer, his compositions, and
the attachment of his name to so many openings (the Benko Gambit in particu
lar) combined to make the Hungarian-American grandmaster a legendary figure
in my mind.
In the early '70s, the legend became flesh and blood when I finally saw him
in a tournament in New York. My first "Benko moment" occurred when I was
sitting next to one of his games. I watched his much weaker opponent lovingly
pull out a new, very expensive chess set he had just received as a gift. I watched
their game progress, and I watched Benko "succeed" in getting into his usual
horrific time pressure. Captivated, I more or less forgot about my own game as
the grandmaster's frantic bid to make the necessary number of moves led to an
other Benko "custom": he hung a Rook. The opponent, of course, couldn't be
lieve his good fortune, but it was all too much for Benko, who went berserk,
picked up the whole set and board, and flung everything across the room! A bit
ironic: the set's first foray into over-the-board play led to a huge scalp and, si
multaneously, its complete annihilation. "The guy's crazy!" I thought, as I fi
nally turned back to my own position.
Two decades later I arrived in Bratislava, Slovakia as acting head of the u.S.
delegation for the World Youth Championships. The other American coach was
none other than Pal Benko, and I was thrilled to get the chance to work with
him and, hopefully, see the man behind the chess playing exterior. I've retained
many memories of this junior tournament: horrifying chess parents (the bane of
all junior events!), the crooked organizers who embezzled as much money as
they could from the contestants, the mini-dramas that many of our players went
though, and the list goes on and on. Foremost in my mind, though, was Benko's
professionalism throughout the event (The more insane things became, the calmer
he seemed to be!), and his "amused confusion" at not having his hotel paid in
xv

PA L B E N KO: MY LI FE, GAMES AN D COMPO S I TI O N S

XVI

advance. For reasons that were never made clear, the United States Chess Fed
eration had forgotten to arrange lodgings for Pal, seemed unable or unwilling to
wire us the necessary funds and, despite many frenzied calls for help, ultimately
left us to fend for ourselves.
After haggling with the hotel management, they agreed to let Benko's non
payment slide for the length of the tournament, but I eventually realized that a
serious problem was developing (Having just escaped Communist rule, our Slo
vak hotel wasn't willing to accept our credit cards, and we didn't have enough
cash to cover the bill.). Finally, with just one day left and payment past due, I
was forced to enact a scene right out of an old spy movie: Four in the morning, a
cab pulls up to the curb and patiently waits two blocks from the hotel. I quietly
lead a lone figure through the hotel's halls and out the back door. Having suc
cessfully made our exit, I drag the horrified grandmaster to the waiting cab, which
streaks him away to the Vienna airport and a successful escape. When the orga
nizers eventually realized that something was amiss, I just played dumb and
shrugged my shoulders. Such intrigue wasn't part of the job description, but a
coach has to be ready for any eventuality!
Leaping forward several years, I again had the opportunity to work with Benko:
first in the charming Hungarian city of Szeged, and then at Palma de Menorca
(an island off the coast of Spain). It was during our stay at Palma that Pal and I
were finally able to spend some private time together. After many conversations,
a twenty game blitz match (I still remember his post-match comment: "Jeremy,
you are the most boring player I have ever encountered ! "), watching him break
dance to Madonna tunes (remarkable for a man approaching seventy!), marvel
ing as he crushed Tal Shaked (a strong young man who was a fraction of his age)
at arm wrestling, and being unable to solve any of the dozens of chess problems
he placed in front of me, I was able to come face to face with the "real" Benko
his enormous love of chess, his humor, his creativity, and his Old World charm.
Having drawn heavy fire for my book review comments concerning chess
biographies-what is supposed to be a book about a player's life and games in
stead is nothing more than a chronology of one tournament after another, with
out any real "life" or personality being visible-I realized that Benko offered
me the perfect opportunity to demonstrate my own vision of how such a book
should be presented. Here was a man that I liked and admired, who was a part of
chess history, who played many beautiful games, and who lived a colorful life
that transcended mere chess concerns -life-and-death struggles, sexuality, fi
nancial stability, etc.
From the book's "birth" in Palma de Menorca (where we first discussed the
idea), to my visiting Benko in Budapest to interview him and begin the actual
book-writing process, I couldn't have imagined how difficult such a project would
turn out to be! In fact, it took an astounding five years to go from contract sign
ing to publication! In the end, I learned an important lesson-I'll never criticize
another writer for taking the easy road when putting together a life and games
collection !

PREFAC E BY JE REMY S I LMAN

Hard work aside, though, I must admit that I'm very happy with the final
result: a personal look at one man's struggle from humble (and often tenuous!)
beginnings to the status of chess icon; a study of his life's work, ideas, compositions, and games. The oft-repeated question: "Is chess an art, a science, or a
sport? " is easily answered when asked of Benko. Pal Benko is an artist, and I
consider it an honor to have helped make his achievements something that can
be shared, understood, and enjoyed by chess players of every strength and age.
Jeremy Silman, International Master
Los Angeles, 2 003

XVII

thanks '0 in,em,tional M",e, John W",on fm hi,; ti,e[ess effort,;

and assistance; to Grandmaster Larry Evans and Ronald Gross for sharing a piece
of themselves in the interviews; to Harold van der Heijden for his input on the
compositions; to Dr. Anthony Saidy for giving us access to his huge chess lib ray;
and to Grandmaster Susan Polgar for her kindness in supplying this book with a
foreword.
We would like to thank those individuals who supplied us with photographs:
Ronald Gross; Hannon Russell (of www. ChessCafe.com); International Master
John Donaldson; and Jami L. Anson of the USCF who kindly allowed us free ac
cess to their photo database. Special thanks to Dr. Richard Cantwell, who gave us
the magnificent series of photographs he shot in Curac;ao, 1 962 . The hospital pho
tos of Tal (who was forced to leave the tournament due to illness) and Fischer are
particularly striking.

XVI I I

was mated with very specific goals in mind. Most importandy,

I wanted the prose concerning Benko's life to be chronologically merged with
the games of the same period. In this way the reader can see the progression of
Benko's style, and also ascertain whether the ups and downs of life away from
the chessboard (I insisted that quite a bit of his "non-chess" adventures be pre
sented!) affected his tournament performances.
A large number of his games ( 1 3 8 !), all deeply analyzed, have been given.
Pal, who is fond of every one of them, understood that I might cut some. How
ever, after a lot of soul-searching, I finally decided that I couldn't deprive Pal or
the reader of the delights they offer.
One of the more idyllic times in Pal's life was his stay in Los Angeles with
Larry Evans and Ron Gross. I conducted interviews with both these gentlemen
so that the reader could enjoy deeper insights into this period of time-person
ally, I found the behind-the-scenes interactions of Benko, Larsen, Evans, and Gross
to be extremely interesting. Ronald Gross in particular has the gift of gab, and
his stories about Evans, Larsen, and Reshevsky are sure to grab your attention.
"While mulling over this project, I made a thorough study of Benko's games
and discovered that Pal (though he insisted it wasn't the case) was always at the
forefront of opening theory. Rather shocked by this revelation, I approached
International Master John Watson (one of the world's finest chess writers) and
asked him to look into this matter. He kindly agreed, and then went beyond
anything I imagined by writing a very detailed opening survey that proves, be
yond a shadow of a doubt, that Benko was indeed a theoretical dynamo ! The
material in Watson's section will prove useful to people who employ the same
openings that Benko did, and for those who simply want a feel for the kind of
preparation that was necessary in topflight grandmaster chess at that time.
The final part of this book presents 3 00 of Benko's compositions. Simple "play
and mate" situations, magnificent endgames, puzzles, helpmates, letter problems,
XIX

PAL B E N KO : MY L I F E. GAMES AN D COMPOSITI O N S

XX

twins, Christmas tree motifs, duplexes, sets - suffice it to say that I had never
heard of many of these terms! Nevertheless, I discovered firsthand that a player
could immerse himself in these 3 00 artistic creations for months and months!
The list of games is included in the table of contents at the beginning of the
book, while an index of names and an index of openings is given at the book's
end. I decided to create a partial record of Pal's tournament results: Benko's par
ticipation in endless U.S. Swiss System events made a complete record too com
plex and, in my estimation, too meaningless to warrant inclusion.
Though many books have specific photo sections, I felt that strategically pre
senting pictures all through the book would create a more organic and energetic
effect. Occasionally the placement of a particular photo won't be chronologi
cally correct, and a number of interesting pictures (the negatives having long
been lost) were taken from old magazines and thus were not as clear as I would
have liked. But, if a player's image turns a mere name into a human being, then I
offer no apologies.
This project took over five years to complete. I designed it to be unlike any
other chess biography- to be fun, instructive, insightful, and at times offering
genuine surprises. If you, the reader, finds that it brought the game/sport/art of
chess and the grandmasters that play it to life, then I'll consider those five years
to be time well spent.

here does the creative urge come from? Why are some people happy
to work in a factory, in the military, or in a thousand different labor-intensive ca
reers, while others need music, art, literature, and yes -even chess -to make sense
out of life? In my case the creative gene was unquestionably passed down by my
father who, though an engineer by profession, desperately wanted to succeed as an
artist. Pursuing this dream, he had my family move all over Europe. This explains
why I was born in Amiens, France in 1 928, but grew up in Hungary when my
mother made it clear that she was far from happy to be living out of a suitcase. Even
after we settled down in my parent's native Hungary, my father's wanderlust
couldn't be contained. He continued to travel by himself, had many adventures,
and was even arrested in Germany (trying to cross over the French border without
proper papers!) at one point. Yes, I turned out to be very much like my father!
Unlike many players who were prodigies, I didn't learn to play chess until I
was ten. Even then I wasn't consumed by the game, considering it nothing more
than a hobby. Nevertheless, I played in the park from time to time and picked
up a few ideas in this way.
Overall, I had a great childhood. I was athletic, inquisitive, and even took a
job as a messenger when I was twelve. Everything was fine before 1 940, but then
my whole world turned to hell. The war hit Hungary hard. Food shortages left
everyone starving, and ration cards became a necessity. I still remember the bread
lines: thousands of people would begin lining up at midnight (families usually
did this "tag team" style, having a family member take over so the original line
man could rest) for a loaf of bread that was only passed out at 7 A.M!
The lack of food wasn't the only problem. There was no coal to heat the schools,
so everyone stayed home throughout the colder times of year. A citywide infes
tation of lice also made things unpleasant, since in those days we didn't have
3

PAL B E N KO: MY LIFE, GAMES AN D COMPO SITI O N S

anti-louse medications (very, very hot saunas and washing one's clothes in scald
ing water was the only way to eradicate the little monsters). Then, as if famine
and parasites and freezing to death were not enough, the Americans started to
bomb Budapest relentlessly.
Even though fire rained down from the sky and misery was a normal state of
mind, I was still able to play a bit of chess: A school tournament was thrown
together, various games in the park were contested, and I avidly studied a book
of Capablanca's 3 5 0 best games (this was my first chess book and the Cuban
quickly became my chess hero). In 1 943 , a strong club tournament was held,
This event showed me that I was already very strong, and I was leading by quite
a bit when the tournament was cancelled due to most of the players being
drafted!
The Germans invaded Hungary in March of 1 944 and, as might be ex
pected, the war effort became even more frenetic. Thus, it wasn't too surpris
ing that, when I turned sixteen, I was also drafted and assigned to a regiment.
I never fought, but I did have to dig endless ditches and put up with even
more hardships. After about four months of this, my regiment started head
ing for Austria. Finally I decided that enough was enough and I escaped, hid
ing in the homes of peasants as I tried to make my way back to Budapest and
my family. This was pretty terrifying since I would have been executed if I'd
been caught!
After three days on the run, the Russian army appeared and policemen were
everywhere. Suddenly I was hiding from the Hungarian army (who would shoot
me on sight), the police (who would turn me over to the Hungarian army), and
the Russian army (who were fond of shipping wayward males back to Russia as
slave labor). As it turned out, the Russians did, indeed, catch me. They immedi
ately turned me into a laborer, making me work all day on a bridge. They said I
could go home afterwards, though I knew they were lying. Fortunately, a bomb
scare forced them to turn all the lights off and, in the evening, I escaped again in
the pitch darkness.
When I returned to Budapest in early 1 945, I found that the Russians had
arrived two months before. At first the Hungarian people didn't think much of
it; in fact, we were happy. The Germans, whom most Hungarians hated throughout
the war, were gone, and we really thought things were about to improve.
The papers said some horrible things about the Russians, but most Hungarian
civilians didn't believe it, thinking it was nothing but propaganda. In fact, why
would we allow ourselves to believe more negative news? We wanted to embrace
a better future, and this need for a happier life temporarily put blinders on our
eyes. After a short time, though, we quickly saw that things were worse than the
papers had said! So, before we knew it, all the people were hiding from the new
threat: the Russians!
This was an incredibly trying and painful time. Once I made my way back to
our apartment (which was bombed out and open to the elements) in Budapest, I
found that my father and brother had been shipped to Russia as slave labor. This

I N T H E BEG I N N I N G

was actually ironic, since my father had been arrested by the Hungarian govern
ment at the end of 1 943 for refusing to join the army. When the Nazis took over
Hungary, they released him, thinking (erroneously) that he was against the former
government and was thus pro-Nazi. The Germans didn't realize that he belonged
to the leftist party and helped Jews by giving them work permits in a small machine parts factory he owned. When the Russians (our "saviors") arrived, they
quickly took over the factory, tossed my father and brother on a train, and a
major part of my family simply vanished.
When the war ended, the situation became even worse than before. After Ger
many invaded Hungary we weren't shocked to see the SS, in competition with
the Hungarian political police, race to confiscate anything of value (art in par
ticular) from Jewish families. However, we didn't expect the Red Army to con
tinue the looting! In fact, chaos ruled the day: women everywhere were being
raped (Men made sure their wives and daughters stayed indoors!), and if a Rus
sian soldier asked for your coat or shoes, you wouldn't hesitate to give it to him
since the alternative was a horrible beating or even death! Once again I found
myself in hiding, this time because my mother feared that I'd be taken away to
Russia if! set foot outside. Naturally, life was much safer if you joined the Com
munist Party, as Szabo did. But I had principles, and just didn't believe in their
rhetoric. Besides, how could I even entertain joining a group that had enslaved
my father and brother?
So far, my sixteenth birthday had brought an almost surreal amount of horror
with it. Fortunately, I was allowed a break from the hellish events that plagued
me: I got a temporary job in construction and, with things apparently calming
down, I accepted an invitation to my first real chess tournament (see games 1
and 2). It was an eighteen-player event that featured ten masters, seven candi
dates, and me. Clearly, I was expected to come in last (All the more so since,
going into this event, I had never played a tournament game against a master!),
but I surprised everyone by winning first prize, and because of this I was awarded
the master title. I still have the diploma on my wall today.
Sadly, I still had to live through one last, crushing shock before turning seven
teen: my mother, at the age of forty-one, died. The lack of shelter, the cold win
ter, the scarcity of food, and the loss of her husband and oldest son (my brother
is one and a half years my senior) proved too much for her.
How can I describe my state of mind at that time? Depression is far too mild
a word, while an emotion like anger demanded a lot more energy than I pos
sessed. The only term that paints the proper picture is devastation-pure,
overwhelming, devastation. Still a child, I wanted to crawl under a rock and
give up- somehow make it all go away. "Resignation" wasn't possible, though;
my little sister depended on me and I wasn't going to let her down by accept
ing failure. Realizing that the situation in Budapest didn't offer me any oppor
tunities to improve our lot, I left my sister in the safe hands of a relative,
packed up a few possessions, and made my way to the lovely little Hungarian
town of Szeged.

PA L B E N KO : MY L I F E , GA M ES A N D C O M POSITI O N S

When I arrived in Szeged, I was starving and penniless. However, a small Mas
ters' tournament led to the acquisition of some desperately needed food (I won
flour and bacon!). My result impressed the chess aficionados there, and I was
invited to teach them and was given a place on the local chess team. In turn, this
new, relatively safe environment allowed me to finish secondary school.
(I) T Feldman - Benko (Budapest, 1 945)
I .c4 Nf6 2.Nc3 e6 3 . e4 d5 ! 4.cxd5

Nowadays this is known to be inaccurate. The main line now revolves

around the immediate 4.e 5 , when 4 . . . d4 ( . . . Ne4 is no longer correct)
5 .exf6 dxc3 6.bxc3 Qf6 7.d4 e5 leads to reasonable play for Black.
4 . . . exd5 5.e5 Ne4!

Initiating a strong pawn sacrifice that was first played in this game.
6.Nxe4

White couldn't resist! Eventually theory recognized that accepting the pawn
was too risky, and the calmer 6.Nf3 Nc6 7.Bb5 Bd7 8.0-0 became the
indicated course. Even here, though, Black gets equal chances by 8 . . . a6
9.Be2 Be6 1 0.d4 Be7.
6 . . . dxe4 7.Qa4+ Nc6 8.Qxe4 Qd4!

Information traveled slowly in the '40s, so I can't be sure if Unzicker was

aware of my game with Feldman four years later. Though my 8 . . . Qd4! has
withstood the test of time and is considered to be Black's strongest move,
Unzicker (as Black against Kuppe, Oldenburg 1 949) preferred 8 ... Be6, and
achieved a crushing position after 9.Bb5 Bd5 1 0 .Qg4 a6 I l .Bxc6+ Bxc6
1 2 .Ne2 Qd5 1 3 .d4 Bb4+ 1 4.Bd2 Bxd2 + 1 5 .Kxd2 Qb5 .
9.Qxd4

In Hubner-Christiansen, Munich 1 99 1 White tried 9.Bd3 , but got noth

ing after 9 . . . Qxe4+ 1 O.Bxe4 Nxe5 1 1 .Nf3 Nxf3 + 1 2 .Bxf3 Bc5 .
9 . . . Nxd4 1 0.Bd3 Be6 I 1 .Be4

An important alternative is I l .Ne2 0-0-0 1 2 .Nf4 Bd7 (actually, I don't see

anything wrong with 12 ... g5 1 3 .Nxe6 fxe6 1 4.0-0 Nf3 + 1 5 .gxf3 Rxd3, when
Black had a fantastic position in Tartakower-Horne, Hastings 1 9 5 3 , though
he later found a way to lose) 1 3 .0-0, when 1 3" . g5 1 4.Nh5 Be6 is recom
mended by theory as being okay for Black, but 1 3" . Re8 1 4.b3 Nc6
1 5 .e6 fxe6 1 6.Bb2 , Varga-Lukacs, Budapest 1 992 also seems fine if Black
now continues with 1 6 . . . e5.
1 1 . . . 0-0-0 1 2 .f4

Also comfortable for Black is 1 2 .Ne2 Bc5 .

1 2 . . . Bd5 1 3 .Bxd5 Rxd5

I N TH E B EG I N N I N G

White i s a pawn up (if you can really consider that pathetic creature o n d2
a pawn), but Black's lead in development, active pieces, and pressure down
the d-file combine to give him all the chances.
14.Kd l gS ! ?

Balogh recommended 1 4 . . . BcS in his notes to this game. This i s certainly

a good, sane, move, but I was young and impetuous and felt that opening
up lines with 14 . . . gS ! ? simply had to be worthwhile.
l S .fxgS

Balogh pointed out that I S .d3 is met by I S . . . Nc6 ! , when Black has the
superior game.
l S . . . Bb4! 1 6.Nh3

This might not be White's best defense, but other moves would have also
left me with a threatening initiative: 1 6.Nfl (simply bad is 1 6.a3 Bxd2
1 7 .Bxd2 Nb3 , while 1 6.b3 ! ? Rhd8 1 7 .h4 RxeS 1 8.Nfl RedS 1 9.Rh2 Bd6
20.Rh3 Bf4 2 1 .Nxd4 Rxd4 22.Kc2 Bxd2 also promises Black all the chances)
1 6 . . .Nxf3 1 7 .gxf3 Bxd2 1 8 .Kc2 Bxc 1 1 9.Raxc 1 RxeS , with an obvious
advantage.
1 6 ... Rhd8?

An inaccurate move that gives White more options than he deserved. Cor
rect was 1 6 ... RxeS ! 1 7 .Rfl Rd8 ! (Also possible is 1 7 . . . Rhe8 1 8 .b3 , when
Black has several promising ways to play the position. For example, 1 8 ... R8e7
[Or 1 8 . . . BaS ! ? , stopping White's idea of a3 and b2 -b4. Then 1 9.Rb l is
strongly met by 19 ... cS 20.Bb2 Re2 2 1 .Bc3 Bxc3 2 2 .dxc3 NbS 2 3 .Rc 1 Rxa2 ]
1 9.a3 [19.Bb2 Re2 ] 1 9 ... Bd6! 2 0.Bb2 RdS 2 1 .Bxd4 Rxd4 2 2 .g3 RdS , with
excellent play.) 1 8 .Rxf7 (bad is 1 8 .d3 Rde8 1 9.Bxd2 Bxd2 2 0.Kxd2 Re2 +
2 1 .Kc3 cS, but 1 8.Rf4 also leads to a Black advantage: 1 8 ... cS 1 9.Rxf7 Rde8
2 0.Rfl Re2 2 1 .Nf4 Rf8 22 .a3 BaS 2 3 .g3 Rxh2 24.b4 Bd8 2 S .bxcS BxgS
2 6.Rb l hS 2 7 .Re l Bxf4 2 8 .gxf4 Rxf4) 1 8 ... Rde8 1 9.Rfl Re2 2 0.Nf4 Rf8,
and White is gagged, bound, and suffering.
1 7.Nf4

PAL B E N KO : MY L I F E , GAMES AN D COMPO SITI O N S

Worse is 1 7 .Rfl RxeS 1 8.Rxf7 Rde8 1 9.Rfl Re2 2 0.Nf4 Rf8 2 1 .h4 Rxg2 , a
position we've already seen in the note to Black's 1 6th move. On 1 7 .Re 1 ,
Black continues to call the shots after 1 7 . . . Nc6. One sample of the kind of
play that could occur is: 1 8.Re2 RxeS 1 9.RxeS NxeS 20.Nf4 Rd4 2 1 .Ne2
Rg4 2 2 .g3 Nfl 2 3 .h4 Bd6 24.a4 Nd4 (since White is helpless, Black might
also consider 24 . . . aS ! ?) 2 S .Nxd4 Rxd4, and White's kingside pawns will
fall like ripe fruit.
1 7 . . . RxeS 1 8.h4?

Missing the only chance for survival: 1 8.Nd3 ! RbS 1 9.Nxb4 (The tempt
ing 1 9.a4 Rb6 2 0.aS RbS 2 1 .Nxb4 Rxb4 is worse since the weakness of the
b3-square will come back to haunt White after 2 2 .Ra3 [More accurate is
2 2 . Rfl , but Black would still be happy after 2 2 . . . RdS 2 3 . Rxf7 RxgS
24.g3 Nb3 .] 22 . . . RdS 2 3 .h4 Nb3 24.Rfl RcS 2 S .Rxb3 Rb3 26.Rxf7 Rb4!
2 7 .d3 Rxh4.) 1 9 . . . Rxb4 2 0.Rfl RbS 2 1 .Rxf7 (on 2 1 .d 3 , Black must avoid
2 1 . ..Ne6? 2 2 .Rxf7 NxgS 2 3 .Rg7, and instead play the calm 2 1 . . .Rd7 when
2 2 .Rf6 Ne6 2 3 .Rh6 Rxd3 + 24.Ke2 RddS ! gives Black all the chances)
2 1 . . . RxgS 2 2 . g 3 (Better than 2 2 . Rf2 Re S 2 3 . b 3 Rde8 2 4 . Rfl Re2
2 S .Bb2 Nc6, when 26.Rf8 is met by 26 ... Re l + 2 7.Kc2 Nb4+ 2 8.Kc3 Nxa2 +.)
2 2 ... RhS 2 3 .Rf2 (and not 2 3 .h4 RfS !), and now 23 ... ReS leads to the kind
of position Black wants to avoid after 24.d3 Nc6 2 S .Bd2 Rxd3 2 6.Kc2 Rfl
2 7 .Rafl . Instead, 2 3 ... Nc6 Ieads to more lively play where Black has plenty
of compensation for the sacrificed pawn : 2 4 . a4 N eS 2 S . Re2 N d 3
2 6.Ra3 RfS 2 7 .Kc2 Nb4+ 2 8 .Kb l Rfl 29.Rc3 Rd4 3 0.b3 Kd7.
1 8 . . .NfS !

White's game now collapses like a house of cards.

1 9.Kc2

Black's pieces would exert tremendous pressure on the enemy position

after 1 9.d3 Ng3 20.Rgl Rd4 2 1 .a3 Bd6.
1 9 . . . RcS + 20.Kd l ?

White had to try 2 0.Kb3 Bxd2 2 1 .Bxd2 Rxd2 , though Black's superiority
is obvious.

I N TH E B E G I N N I N G

2 0 . . . Rxc l + ! 2 1 .Kxc l Bxd2 + 22 .Kc2 Bxf4 2 3 .g4

Other moves also lose: 2 3 .Rae l Rd2 + 24.Kb3 Nd4+ 2 5 .Kc3 Ne2 + 2 6.Kb3
Be5 2 7 . Rhfl Rxb2 + 2 8 .Ka3 b5, and it's all over.
2 3 . . . Ne3 + 24.Kc3

After 24.Kb l Rd2 , Black would maintain a material plus and a strong at
tack.
24 . . . Be5+ 2 5 .Kb3 Rd3 + 26.Ka4

White gets mated after 2 6.Kb4 Bd6+ 2 7 .Ka4 Nc4 2 8.Kb5 Nxb2 2 9.Rac l
a6 3 0.Ka5 Ra3 mate.
2 6 . . . Bxb2

Even stronger was 2 6 . . . Nc4 ! , but the text move is more than adequate.
2 7.Rab l Nc4 2 8.Kb4 Rd4 29.Rxb2

This and 2 9.Kb3 (which loses to 2 9 . . . Nd2 + 3 0.Kxb2 Rb4+ 3 1 .Kc2 Nxb l
3 2 .Rxb l Rxb l B.Kxb l Kd7) are the only ways to defend against the mate.
29 . . .Nxb2 + 3 0.Kc3 Rd l 3 1 .Rxd l Nxd l + 3 2 .Kd4

The rest is clear and easy.

32 . . . Nfl 3 3 .Ke5 Nxg4+ 34.Kf5 Kd7

Simplest.
3 5 .Kxg4 Ke6 3 6.Kh5 Kf5 3 7.Kh6 Kg4 3 8 .h5 c5 3 9.Kxh7 Kxh5, 0- 1 .

Dr. Balogh concluded: "Played in the style of a grandmaster. Even Flohr

could not have performed more simply or smoothly. " Maroczy originally
awarded this game the first brilliancy prize, but the jury overturned his
decision and gave it the second prize after Hajtun-Feldman. Maroczy's
judgment was as follows: "An excellent game. White seemed to be supe
rior, but his collapse was interesting and surprising. It is a game of theo
retical value; Black's attack is too strong."
(2) Szigeti - Benko (Budapest, 1 945)
l .d4 d5 2 .c4 e6 3 .Nc3 c 6 4.NB Nf6 5 .Bg5 dxc4

Those that wish to avoid the chaos that ensues after 5 . . . dxc4 often choose
5 . . . h6, when 6.Bxf6 Qxf6 7.e3 Nd7 (known as the Moscow Variation) has
proven to be extremely resilient. Of course, attacking maniacs aren't tak
ing this lying down, and now the so-called "Anti-Moscow" with 6.Bh4!?
has become topical.
6.e4 b5 7.e5 h6 8.Bh4 g5

This is the Botvinnik Variation, which became well known when he used it
to beat Denker at the Soviet Union vs. U.S.A. radio match in 1 945 . That
match was carried out in September, while my game with Szigeti was played


BENKO, PAL (HUN) (USA)

GM

?
the
U E

Pal Benko fifty years ago earned the Hungarian

chess master title. During that time he won

Hungarian Championship, e ight times the

.;" BenkO.PaI HU .) (US"')GM

Open Championship. He became twice Worldl.:::i,

I a te. S'IX t'Ima s h a was O lymplC-;'
C and'd
.
Champlon

f!!!!.

B(nk6 Pil
1akknagym('slt'r
team member and s even times team captain. .. d
.
_ "U .prsI8.199S d tC
.
He achieved the international grand master htle

XXIWSADIlIMPlAUIIIISAIIIlS

;I
'"
14.
'"

(1958) and beside that he attained the problem

composing international master title

(1995).

Benko - Fischer
(1962. Curacao)
World Champion Candidates Tournament
1 g3Nf6
2 Bg2 g6
3e4 d6
4 d4 Bg7
5Ne2 ()"O
6()"0 e5
7 Nbc3c6
8 a4 Nbd7
9 a5 exd4
10 N.d4 Nc5
11 h3Re8
12 ReI N fd7
13 Be3 Oc7
14f4RbII

IS Qd2 b5
16 axbll axb6
1 7 b4 Ne6
18 b5Nxd4
19 B.d4 Bxd4
20 Oxd4 c5
21 Qd2 Bb7
22 Radl Re6
2385 B.g2
24 Kxg2 01>7+
25Kf2 Rd8
26exd6Nf6
27 Rxe6fxe6
28 0e3 Kn

29 Qf3 QbII
30 Ne4 N.e4+
31Oxe4Rd7
32Qc6Qd8
33 KI3 Kg7
34 g4 e5
35fxe5 Rn.
36 Kg2 Qh4
37 Rfl Rxfl
38 Kxf1 Oxh3+
39 0g2 0e3
400e2 Qh3+
black gives up
1-0.

38- 50

in July. However, neither Botvinnik or I can claim to be the first to use this
system. Reti played it in 1 92 0 (!), while Szigeti (yes, the same one) had the
White side in 1 93 1 . E. Griinfeld had Black against Van Scheltinga in
Amsterdam 1 936, and it was used in the game Szabo-Euwe in Hastings 1938.
To the best of my knowledge, Botvinnik first used it (twice) in 1 944. Today,
the line is extremely popular among tactical players who enjoy cutting-edge
theoretical discussions. At the time of this game, though, the variation rep
resented an unusual, unexplored vista.
9.Nxg5 hxg5

The earliest games featuring 9.NxgS saw Black play 9 . . . Nd S , when

1 O.Nxf7 (the most dangerous reply) was seen in Szigeti-Michlo, Budapest
1 93 1 ( 1 -0 in 3 7), while the more placid 1 0 .ND was Leif-Reti, Kristiania
1 92 0 : 1 0 . . . QaS I 1 . Qd2 Bb4 1 2 . Rc 1 Nd7 1 3 .Be2 N7b6 1 4.Bd l Nxc3
I S . bxc3 B a 3 1 6 . B f6 Bxc 1 1 7 . Qxc 1 Rg8 I S . Nd2 NdS 1 9 . Ne4 b4
2 0.BhS bxc3 2 1 .0-0 Qxa2 2 2 .Bxf7+ Kxf7 2 3 .Qxh6 Qa3 24.Nd6+ Qxd6
2 S . Qh7+ KeS 26.exd6 Rg4 2 7 .QhS+ KfS 2 S.Be7+ Nxe7 29.dxe7+ Kxe7
3 0.Qxg4 RbS 3 1 .Qg7+ KdS 3 2 .h4 c2 B. QhS+, 1-0.
l O.Bxg5 Nbd7 1 1 .exf6 Bb7 1 2 .Be2

Years of theoretical research and lots of practice eventually proved that

1 2 .g3 was White's best move, but we were still busy "inventing the wheel."
1 2 . . . Nxf6

Also playable are 1 2 . . . Qc7 and 1 2 . . . Qb6, preparing to castle queenside as

quickly as possible.
1 3 .BfJ

I N TH E B EG I N N I N G

It's interesting to note that 1 3 .a4 was recommended by Korchnoi i n the

ECO. Apparently, he wasn't aware that this had already been played in
Van Scheltinga- Grunfeld, Amsterdam 1 9 3 6 : 1 3 . . . B e 7 1 4.axb5 cxb 5
1 5 .BB Nd5 1 6.Bxd5 Bxd5 1 7 .Bxe7 Kxe7 I S.0-0 Qd6 1 9.Nxd5+ Qxd5
20.Re l RagS 2 1 .Rxa7+ Kf6 2 2 .B Qh5 2 3 .Kf2 Rxg2 +, 0- 1 .
1 3 . . . Be7 14.Bxf6

I had a good laugh when this game appeared with notes in the Hungarian
Chess World (the number one Hungarian chess magazine until 1 9 5 1 ). Who
ever annotated the game gave 1 4.Bxf6 a question mark and claimed that
White's best move was 1 4.Nxb 5 . Naturally, this "best move" loses at once
to 14 ... Qa5+ 1 5 .Nc3 Qxg5 .
14 . . . Bxf6 1 5 .Nxb5 cxb5 1 6.Bxb7 Rb8 1 7.Bc6+ Ke7

Pachman assessed this position as equal in one of his books. One might
think that White should be better since he's a pawn up and Black isn't able
to castle. This opinion, though, is completely mistaken. Black's King is
actually very safe on e7. He will easily win his pawn back when, despite
the opposite-colored Bishops, Black will have all the chances thanks to his
queenside majority and the fact that the dark-squared Bishop will prove
stronger than its white counterpart.
1 8.Qd2 Qd6 1 9.BO Rh4!

Black isn't in a hurry to win the pawn back. First, he wants to make sure
every piece is as active as possible.
20.g3 Rxd4 2 1 .Qa5 c3 ! ?

An optimistic attempt to knock my opponent out. The straightforward

2 1 . . .Rd3 , in connection with . . . Bxb2 , was simpler.
22.Qxa7+

And not 2 2 .bxc3 ? ? Ra4, when Black wins the house.

22 . . . KfB? !

This natural move turns out to be inaccurate. Correct was 2 2 . . . KeS ! , when
2 3 .Bh5 fails to 2 3 . . . Qe5+ 24.Kfl Rd7.

II

PA L B E N KO: MY LI FE, GAMES AN D COMPOSITI O N S

2 3 .0-0? !

12

White returns the favor. He had to try 2 3 .Bh5 Be7 (A fun alternative is
2 3 . . . Qe5+ 24.Kfl Qxh5 2 5 ,QxbS+ RdS 26.Qf4, when both 2 6 ... Qh3 + and
26 . . . cxb2 are interesting, though I'm not sure if either offers more than a
draw,) 24.bxc3 Qe5+ 2 5 .Kfl Ra4 26.f4 Qxf4+ 2 7 .gxf4 Rxa7, when Black is
still better, though not nearly as much as in the game (i.e., 2 S .h4 h4!).
2 3 . . . cxb2 24.Rab l Rd3 2 S .Be4

Black is still much better after 2 5 .Bh5 Be7, since 2 6.Rxb2 ? ? Ra3 wins right
away.
2 S ... Ra3 26.Rbdl Rxa7 27.Rxd6 Rxa2

White succeeded in rescuing his Queen, but the resultant endgame doesn't
promise him anything but misery-the b-pawn is simply too strong and,
in a very interesting way, it is helped by its doubled brother on b5.
2 S.Rb l

Blockading with the Bishop fails to 2 S .Bb l Ra l 29.Rdd l ReS, when Black
wins easily by 30 . . . Be7, 3 1 .. .Ba3 , and 32 . . . Rc 1 .
2 S ... Ra l 29.Rdd l ReS 3 0.Kfl Re l 3 1 .Ke2 Bc3 3 2 . g4? BeS

Black has defended himself against ... f7-f5 , however the advance of the
g-pawn accelerated his inevitable demise since the f4-square is now avail
able to Black's Bishop. Nevertheless, the game was lost in any case since,
if White had employed a passive defense, Black would have broken the
blockade by marching his King to a2 .
3 3 .h4 Bf4 34.Bd3 b4 3 S .Be4 Kg7 3 6.gS b3 3 7 .Bd3 Re2 + 3 S .m Be l ,

0- 1 .
The editor of Magyar Sakkvilag wrote: "A beautiful game ! The conges
tion of pieces in this endgame is spectacular. "
While I was going through my own drama, my brother, father and scores of
other Hungarian men were being worked to death in prison camps. Many died of
hunger, though they were told that the Hungarian "workers" were eating as well
as the peasants (In other words, the average Russian was also starving at that time! ).

I N TH E BEG I N N I N G

Under such conditions, i t was no surprise that my father became ill and, since he
was unable to work, the Russians released him and my brother. You can imagine
my relief when they returned to Budapest. Soon my family was together again,
and this bit of stability allowed me to begin university (During my first year it
was called the University of Economy. In year two the communists took over and
renamed it the Karl Marx University.), where I majored in economics.
One thing that still stands out in my mind is the inflation, which was rampant
throughout Hungary. Money became so worthless that workers preferred being
paid in food! In fact, I played in the Hungarian Championship where the prizes
were food rather than money-I was absolutely delighted to get the chance to
win something so valuable !

(3) G Barcza - Benko (Hungarian Championship, 1 946)

l .NO dS 2 .b3 cS 3 .Bb2 ? f6 !

It seems that White's third move is already a mistake ! When playing a

regular Queen's Indian ( l .d4 Nf6 2 .c4 e6 3 .Nf3 b6), the strong move 4.f3
isn't available to White since his Knight occupies that square. Here, though,
Black can take advantage of the fact that his Knight isn't on f6 to build a
powerful center and block the enemy b2-Bishop. In the Fischer-Petrosian
candidates match in 1 97 1 , White (Petrosian) played 4.c4 d4 S.d3 eS and
was obviously not trying to prevent Black from executing his plan. He
should at least have attempted to emulate a defensive system used by Black
against the Samisch variation in the King's Indian: l .d4 Nf6 2 .c4 g6 3 .Nc3
Bg7 4.e4 d6 S.f3 0-0 6.Be3 b6 7 .Bd3 Bb7 8.Nge2 cS 9.dS e6 1 0 .0-0 exdS
I I .cxdS Ba6, although I must confess that personally I could never fully
appreciate this variation positionally. Petrosian-Fischer continued: 6.e3 Ne7!
(Fischer wisely brings his g8-Knight to the still unoccupied c6-square) 7.Be2
Nec6 8 .Nbd2 Be7 9.0-0 0-0 1 0.e4 a6, and it is easy to see that Black has
all the play on the queenside and more central space, while White's
b2-Bishop is "biting on granite."
To return to my game, White tried to counter Black's plan more ac
tively. In this game we see that Fischer's plan was not new and probably

13

PAL B E N KO : MY LI FE, GA M ES AN D COMPO S ITI O N S

14

was not new even when I played it in 1 946, since it is possible to find these
ideas over the board, as I did. Why Petrosian played 3 .Bb2 and didn't em
ploy the superior 3 .e3 is a mystery. However, now it's too late, for 4.e3 e5
5 .d4 cxd4 6.exd4 e4! again blocks the b2-Bishop and mobilizes a danger
ous kingside pawn majority. Incidentally, I had some difficulty naming this
opening since, at the time this game was played, no one had ever heard of
the Larsen Opening- in fact, no one had even heard of Larsen! Perhaps
"Larsen Opening" is an anachronism.
4.d4 cxd4 S .Qxd4 eS 6.Qd2 Ne6 7.e3 Bb4!

An original plan that activates this Bishop and stops White's c2 -c4. S.c3
BaS 9.b4 Bb6 1 0.a4

The "aggressive" 1 0.c4 fails to 1 0 . . . dxc4 I I .Bxc4 Nxb4! .

1 0 . . . Be6

Still making sure that White can't play c3 -c4.

1 1 .aS Be7 1 2 .a6 bxa6 1 3 .Bxa6 Nge7 14.Na3 0-0 l S .O-O?

This natural move costs a pawn, but Black's position was already superior
when you take his strong center into consideration. White's only break,
c4, was bad because of 1 5 ... Nxb4! .
l S . . . e4! 1 6.Ne l Nxb4! 1 7.NbS

Since 1 7 .cxb4 Qd6 threatens both ... Qxh2 mate and ... Qxa6, White tries
to console himself with possession of d4.
1 7 . . . Nxa6 l S . Rxa6 Bb6 1 9 .Nd4 BeS 2 0 . Ra2 NfS 2 1 .Nee2 Nxd4
2 2 .Nxd4 Qd7

This defends the e6 and f5 squares and allows me eventually to activate

my light-squared Bishop by . . . a7-a5 followed by . . . Ba6.
2 3 .0

White has to gain a bit of activity in this fashion before Black plays
. . . a7-a5 , . . . Bc8-a6-c4, and then . . .f6-f5-f4 with a complete crush.
23 . . . exO 24.Rxf3 ReS 2 S .Qd3 as 26.Ba3 Ba6 27.QfS

Interestingly, Barcza trades Queens even though he's a pawn down.

Objectively, of course, it is better to keep the pieces and trade pawns (when
you're in trouble in an endgame, try and trade off as many pawns as pos
sible !), but Barcza has faith in his endgame technique.
2 7 ... QxfS 2 S.NxfS Be4 29.Rd2 RadS 3 0.m g6 3 1 .Nd6

Again, longer resistance was possible after 3 1 .Nd4, avoiding the trade of
pieces.
3 1 . . .Re6 3 2 .Nxe4 dxe4 3 3 .RxdS+ BxdS 3 4.Rf4 Re6 3 5 .Kf3 fS

I N T H E BEG I N N I N G

15

Stopping the white King from reaching an active position and threaten
ing 3 6 . . . Bg5 3 7 .Rd4 Bf6. Black has the right Bishop (I was very careful
not to allow the creation of opposite-colored Bishops!), his pawns are on
the right squares, and with his extra passed pawn the win is assured-with
the right technique.
3 6.Ke2 BgS 3 7.Rfl a4 3 8.Kd2 Re6 3 9.BeS ReS !

Forcing the white Bishop to give up its blockade of the passed pawn.
40.Bd4 RaS 4 1 .Ke2 a3 42 .Kh l Kf7 !

The King can't wait to reach e4!

43 .Ka2 Be7

By defending a 3 , my Rook is free for more aggressive action.

44.Rh l

Otherwise, 44 . . . Rb5 and a black King march will decide.

44 ... Ra6 4S .Rh7 Re6 46.Bh6 Ke6 47.BaS hS

Now my King is free to roam since everything is guarded: the Bishop

defends a3 and the Rook defends c4 and g6. Notice how Black is taking
his time. Patience is a very important quality in the endgame.
48.Rh l

If 48 .Bb4 Bxb4 49.Rxb4 Kd5-e4 wins easily.

48 . . . KdS 49.Bh4

Since ...Ke4 would end the game, White makes one last desperate attempt.
49 . . . Bxh4 SO.exh4 c3 S 1 .Kxa3 e2 S2 .Rc 1 Ke4 S 3 .hS Re8 S4.h6 Kd3
S S .Kh3 Rh8, 0- 1 .

(4) Benko - Szilagyi (Hungarian Championship, 1 946)

l .e4 Nf6 2 .Ne3 dS 3 .d4 eS 4.e3 cxd4 S .exd4 e6 6.Nfl dxc4?!

We've transposed into a Panov-Botvinnik Attack in the Caro-Kann De

fense. Nowadays nobody would make this mistaken capture because it de-

PA L B E N KO : MY LI FE, GAMES AN D C O M POSITI O N S

velops White's Bishop . Instead, Black usually plays 6 . . . Be7 o r 6 . . . Bb4, only
taking on c4 after the fl -Bishop moves, thereby making it move twice .
16

7 . Bxe4 B e 7 8 . 0 - 0 0 - 0 9 . Bg5 Nb d7 1 0 . Q e 2 Nb6 1 1 . B b 3 Nbd 5

1 2 .Rad l b6?

The position in the diagram represents a common middlegame situation:

White has an isolated pawn, but this "weakness" gives him more central
space and control over the key e 5 - square (the f3 -Knight can't wait to set
up residence there) . White's lead in development and more active pieces
creates tactical possibilities and chances for an attack against the black King;
in fact, White often wins isolated d-pawn positions in a violent manner.
Of course, if White's attack doesn't crash though, Black can drag the first
player down by: 1) Dominating the d5 -square, thus fixing the isolated pawn
and creating a stationary target.

2) Trading minor pieces. This ends White's

attacking hopes. 3 ) Targeting d4 for destruction by training his Queen

and Rooks against the iso.
Black, an experienced master, has just played . . . b 7 -b6, intending to
fianchetto his Queen-Bishop and cement his control over the very impor
tant d5 -square . However, the somewhat vulnerable state of the a8 -Rook
allowed me to turn my back on White's usual attacking dreams and in
stead steer the game towards a favorable endgame. It's interesting to note
that, even at the tender age of eighteen, I still preferred the iron logic of
endgames to the uncertainty of unclear middlegame complications.

1 3 .Nxd5!
Now the players will have matching d-pawns , as 1 3 . . .Nxd 5 1 4. Bxd5 ! ,
hitting a 8 , wins material.

13 ... exd5 1 4.Rfe 1 Be6 ? !

H e doesn't see the coming storm, but 1 4 . . . Bb4 (if 1 4 . . . Re8, White ruins
Black's kingside by 1 5 . Bxf6) 1 5 . Bxf6 Qxf6 1 6 .Bxd5 Rb8 1 7 . Rfl (White also
retains some winning chances with 1 7 . Q e 5 Qxe 5 1 8 . Rxe5) 1 7 . . . Bg4 Ieaves
Black a pawn down and suffering, though he does retain some drawing
chances in the opposite-colored Bishop endgame.

1 5 .Ne5 Re8?

I N TH E B EG I N N I N G

Black only notices one threat, but l S . . . h 6 1 6 .Nc6 Qd6 1 7 .Nxe7 + Qxe 7
1 8 .BxdS hxgS 1 9.Bxa8 Rxa8 20.dS is also insufficient. He could try 1 5 . . .Qd6,
although 1 6 .Bf4 is a strong reply.

1 6.Nxf7 ! Bxf7 1 7.Qxe7 Re8 1 8.Qxd8

could have captured his a7-pawn as well, but why complicate when the

exchange of Queens leads to a dead won, completely safe, endgame?

1 8 . . . Rcxd8

White 's pawn plus and Bishop pair should make it possible to win in more
than one way.

decided to play for exchanges- first the Rooks, then my

dark-squared Bishop for the Knight-and to set up an advantageous pawn

formation, forcing Black's pawns onto light squares so they would be vul
nerable to attack by my light-squared Bishop . Black will be left with an
inferior minor piece (on top of being down a pawn) , and these positional
and material advantages should prove to be more than enough. Watch how
it's done !

1 9.0

A useful move that keeps the Black Knight off e4 and g4, and also lets my
King enter the game on f2 .
1 9 . . . Kf8 20.Kf2 Rxe l 2 1 .Rxe l Re8 22 .Rel Rxe l
Black had no choice. Otherwise, White's Rook would enter via the c-file.

2 3 . B xc l

Ke 7 2 4 . Bf4 a6 2 S . g4 h6 2 6 . B e S Ke 6 2 7 . B e2 g6

28.Bd3 bS
There is nothing better. White can win the queenside pawns at his leisure
after 2 8 . . . aS 2 9 .Bc7 Nd 7 3 0 .Bb S .

29.Kg3 Be8 30.h4 a s 3 1 .Be7!

By forcing the a-pawn onto the wrong color, White hacks out a road for
his King.

3 1 . . .a4 3 2 .BeS Ng8 H .Bf4 hS 34.BgS

Black's Knight is trapped and must eventually be exchanged for the White
dark-squared Bishop .

17

PAL B E N KO : MY LI FE, GAMES AN D COMPOSITI O N S

3 4 . . . b4 3 S .a3
Fixing the target on a4 and clearing away the c3 -square so that White 's
18

King can eventually penetrate on the queenside.

35 . . . bxa3 3 6.bxa3 Nf6 3 7.Bxf6 Kxf6 3 8 .g5 +

Making sure that all the black pawns are permanently stuck on light squares.
This concept of "fixing" enemy pawns on undesirable squares is an im
portant one .

3 8 . . . m 39.Kf4 Ke6 40.Bc2

Black's position is beyond salvation. One simple winning plan is f3 -f4-f5
followed by Kf4. In fact, White would win even without the f-pawn, as in
the line 40 . . . Kd6 4 1 . B b i Bf7 42 .Bd3 Be8 43 .Bc2 Ke6 44.Ke 3 , when Black
can't stop the march of the white King to b4. Still, it's nice to have a pawn;
in chess, there 's no such thing as overkill !

40 . . . Kd6 4 1 .Ke3 Ke6

On 4 1 . . .Kc7 , White wins by 42 .Kd2 Kb6 43 .Kc3 Kb 5 44.f4 and 45 .f5 .

42 .Kd2 Kd6 43 .Kc3 Ke7 44.Kb4 Ke6 45.Bxa4, 1 -0 .

As w e have seen, the key t o winning a same-color Bishop endgame with
the "good" Bishop is to attack two weaknesses, thereby tying down the
opposing forces and penetrating with the King, often with the aid of

Zugzwang. If you are the defender, try to keep your pawns on the opposite
color of the enemy Bishop, don't let the enemy King penetrate into your
camp, and keep your position as flexible as possible .

very Httle ehe" activity in po,twac Hungary. This fact, and my

efforts to complete my education, left me dying to squeeze in a game whenever

possible. Because of the lack of playing opportunities, I agreed to play six corre
spondence games for the Hungarian team in the semifinals of the Correspon
dence Team Championship (these games began while I was still in Szeged) . I
won four of those games and drew the other two. One of the drawn games was
against an Australian player -having an extra pawn and the Bishop pair, I of
fered a draw because the winning procedure would have been too long and too
expensive !
It is curious that one rarely finds correspondence games in the pages of this
type of book, although such games should be on an even higher level of play
than those with more restricted time limits. Perhaps the main reason for this
state of affairs is that grandmasters mainly participate in over-the-board tourna
ments and neglect correspondence play entirely. However, correspondence chess

has played

an important role in the development of many strong players -the

names of Alekhine and Keres can be mentioned in this respect. In fact, Keres
used to recommend correspondence chess as a wonderful way to study both the
openings and other phases of the game. This form of chess is particularly useful
and important for those who aren't able to take part in normal chess events in
person.
I admit that I am able to count all the correspondence games I've played in my
life on my fingers. The game against the Swiss player Tagmann (the second drawn
game from the six) , though, was very exciting and has remained fresh in my mind
over the decades. I hope you find it as interesting as I do.

19

PAL B E N KO : MY LI FE, GAMES AN D COMPO S ITI O N S

(5) Tagmann - Benko (Correspondence, 1 947)

20

l .e4
Years later (in 1 9 7 3 ) a Mr. Heinen offered to play two postal games against
me for a fee . He wanted Black in both games. Since I'd been away from
postal for so long, I decided to accept his offer. Here is one of the games :
l . e4 c5 2 .NB Nc6 3 . d4 cxd4 4.Nxd4 Nf6 5 .Nc3 d6 6 . Bg5 e6 7 . Qd2 Be7

s .o-o-o 0-0 9 . f4 Nxd4 1 0. Qxd4 h6 I I . Bh4 Qa5 1 2 .Bc4 RdS (Black's most
popular and best move is 12 . . . Bd7) 1 3 .Rhfl ! (This is considered the strongest
continuation. White aims for f4-f5 which, after . . . e 5 , gives him control of
the d5 -square . If Black avoids . . . e6-e5 and allows White to play fxe6, Black's
King- Rook will be missed from the kingside and the f6-square will be
come the target of a sacrificial frenzy.) 1 3 . . . a6? (Too slow. 1 3 . . . Bd7 seems
to be the most natural move) 1 4.f5 b5 1 5 .Bb3 Bd7 1 6 .fxe6 fxe6 1 7 . Bxf6
Bxf6 I S .Rxf6 ! gxf6 1 9 . Qxf6 Kh7 2 0 . Qf7+ Kh S 2 1 . Qg6, 1 -0 . Black can't
defend against Qxh6+ followed by Rd3 .

1 . . .Nf6
The Alekhine Defense never became a permanent part of my repertoire .
However, I thought it would be fun to give it a try on this occasion.

2 .eS NdS 3 .c4

Now it's known that the sternest test of Alekhine's Defense is 3 .d4 d6 4.NB .

3 . . . Nb6 4.d4 d6 S .f4

The Four Pawns Attack is a try at outright refutation. However, Black
gets plenty of counterplay against White's center pawns, and this aggres
sive line has now fallen out of favor.

S . . . dxeS 6.fxeS Nc6 7.Be3 BfS 8.Nc3 e6 9.Nf3

Instead of the text move , old theory recommended 9.Be2 . Bringing the
Knight to B used to be considered doubtful because of 9 . . . Bg4, until the
Swiss master Henneberger proved White 's superiority after 1 0 . Qd2 .

9 . . . Qd7
The modern choice is 9 . . . Be7, when Black gets comfortable equality, though
many of the lines are very complicated and require thorough study.

l O.dS
This aggressive move is also of Swiss origin, but it seems to be dubious.
The more restrained 1 0 .Be2 is White 's only real hope for an opening
advantage .

l O . . . exdS l l .cxdS Nb4

And not 1 1 . . .Na 5 ? 1 2 . Bxb6 axb6 1 3 . a 3 , when the threat of b2 -b4 leads to
the win of material for White .

l 2 .Nd4

AFfE R T H E WAR

21

A position full of excitement and tension has been reached. 12 . . . Bg6 was
recommended at the time, but it seemed to be too passive and I rejected
it. In fact, as I discovered later, this whole variation had been thoroughly
mined by the Swiss analysts , while my only compass was my chess intu
ition. Henneberger "proved" that 12 . . . Bg6 was indeed too slow: 1 3 .Bb5 !
c6 1 4. dxc6 bxc6 1 5 .Nxc6 ! Nxc6 1 6 .Qf3 ReS 1 7 . Rc 1 , and "White will re
gain the piece on c6 . " Years went by before I looked at this position again,
and then I realized that 1 7 . . . Nc4! I S .Bf4 Bd3 ! ! turns this whole line up
side down . So much for established theory!
Instead of 1 2 . . . B g 6 , the game Tomasits - D r. Bata continued with
12 . . . 0-0-0, and there followed 1 3 .Bb5 ! c6 1 4.dxc6 bxc6 1 5 .0-0. Now 1 5 . . . Be6
was recommended as the best solution, but this was refuted by Henneberger
(against Emden in 1 946) by 1 6 . a 3 ! cxb 5 1 7 . axb4 Bc4 I S . Qc2 KbS ! 1 9 . Rfd l
Nd5 ( 1 9 . . . Bxb4 2 0 .e6!) 2 0 .Ndxb 5 Bxb4 2 1 .Bxa 7 + . Since 1 2 . . . Bg4 10ses a
piece immediately to 1 3 .e6 ! , Black is left with 1 2 . . . N6xd 5 , though this looks
very dangerous . However, after prolonged thought I finally decided to do
just that, daring White to punish me.

12 ... N6xd5 1 3 .Nxd5 Qxd5 ? !

Unlucky thirtee n - my move turns out to b e dubious . The correct
1 3 . . . Nxd5 1 4.Nxd5 0-0-0 would have led to unfathomable complications.
One oft- quoted line is 1 5 . Q d 3 g6 1 6 .Ng3 Nxe 3 1 7 . Qxe3 Bc5 I S . Qf4
RheS 1 9 .Be2 Qd5 2 0 . Rd l Qxa2 2 1 . Rxd S + Rxd S , when White is suffer
ing. The reader, of course, should beware . Before playing any of this you
simply have to do your own detailed analysis of the position after 1 5 . Q d 3 .
Incidentally, an analysis by Ernst Griinfeld in the

Wiener Schachzeitung

in 1 945 clarified some variations in connection with the game Gallia

Keller. He demonstrated that the continuation 12 . . . N4xd5 (instead of my
1 2 . . . N6xd 5) 1 3 .Nxd 5 (Griinfeld didn't mention 1 3 . Nxf5 ! ? , but perhaps
this is White 's most accurate choice: 1 3 . . . Qxf5 1 4.Nxd 5 Qxe 5 [Black
probably should try 1 4 . . . 0-0-0 and hope that he gets some compensation
for the piece after 1 5 .Nxb6+ axb6] 1 5 .Nxc7+ Qxc7 1 6 . B b 5 + Nd7 1 7 .0-0,
with a very strong attack) 13 ... Qxd 5 ? ! (of course, 13 . . . Nxd 5 is correct)
1 4.Nxd5 Bb4+ (White is also better after 1 5 . . . Qxe 5 1 6 . Q d 3 ) 1 5 . Kf2 Qxe 5

PA L B E N KO : MY LI FE, GAM ES AN D C O M POSITI O N S

1 6 . Qg4 Bc5 1 7 . B b 5 + c 6 I S . Rhe l is i n White's favor, a n opinion that I

have to agree with .

22

1 4.Nxf5 Qxe5
I thought that Black had a superior position. He has two pawns for the
Bishop and will probably be able to win back his piece. No sudden attacks
are possible : 1 5 . Qf3 Nc2 + 1 6 . Kf2 Nxa l 1 7 . B b 5 + Qxb S I S . Qe4+ Kd 7
1 9.Rd l + KcS ! 2 0 . Rxa l Bc5 , and Black keeps his material advantage .

1 5 .Rc l ! !

An excellent and unexpected move !

I didn't know it at the time, of course,

but this too was Henneberger's analysis. It paradoxically takes the

initiative, although it not only doesn't give check, it promises to return
the piece and even to sacrifice further material. Now 1 5 . . . RdS would be
answered by 1 6 . Qf3 , so Black has no choice but to take the piece
"biting the bullet, " as some would say- and hope that things work out.

1 5 . . . Qxf5 1 6.Rxc7 Qe6 1 7.Bb5+ Nc6

Now I S . Qd2 would be met by I S . . . B c 5 , and I S . Qe2 by I S . . . Bb4+ , after

which Black could safely castle and get his King out of danger.

I B.O-O! !
The alternative is I S .Kf2 Be 7 1 9 . Rxb 7 , but Black comes out on top after
1 9 . . . 0-0 2 0 . Q d 7 (not 2 0 . Bxc6 Qxc6 2 1 . Rxe7 Qf6+) 20 . . . Qf6+ 2 1 .Ke2
(2 1 .Kg l runs into 2 1 . . . RadS 2 2 . Qg4 Qe6 ! ) 2 1 . . .Na 5 .

I B . . . Qxe 3 +
Forced. I f I S . . . B e 7 1 9 . Re l , I couldn't castle: 1 9 . . . 0 - 0 ? 2 0 . Bxc6 bxc6
2 1 . B f2 , a n d Wh i t e wi n s . Al s o u n p l e a s a n t fo r B l a c k is I S . . . B d 6
1 9 . Rxb 7 0-0 ( 1 9 . . . Qxe 3 + 2 0 .Kh l 0 - 0 2 1 . Qxd6 i s worse) 2 0 .Qb3 Qxb 3
2 1 . axb 3 , when White wins the a-pawn and retains excellent winning
chances.

1 9.Khl RdB 20.Qh5 g6

The only move . If 2 0 . . . Q e 6 2 1 . Rxb7 (far better than 2 1 . B c4 Qxc4
2 2 . Re l + Be7 2 3 . Rexe 7+ Nxe 7 24.Rxc4 0-0, when Black has a solid posi
tion and good chances to hold the draw) 2 1 . . . RcS 2 2 . Bxc6+ is crushing:

I
III
fi
II
<t
l!!
I!I

'-;7-M!'-lii

White d raws

H e l p m a te in t hree

22 . . . Rxc6 2 3 .Rb8+ Ke7 (or 2 3 . . . Kd7 24.Qa5 , with a winning attack) 24.Qh4+
g5 2 5 . Qd4, and Black has to hang it up .

2 1 .Rxc6 ! !
This is a better winning try than the very interesting 2 1 .Rxb7 gxh5 2 2 .Bxc6+
Rd7 n . Rd7 Qe6 ! 2 4.Rd6+ Ke7 2 5 . Rxe6+ Kxe6 ! 2 6 .Be8 f6 2 7 . Bxh5 Bd6
2 8 .Bg4+ Ke7 2 9 .Bf5 Be5, with excellent drawing chances.

2 1 . . . bxc6 22 .Bxc6+ Ke7 2 3 .Qa5 !

My opponent comes up with another fine non-checking move . Not good
was 2 3 .Qh4+ g5 24. Qb4+ Ke6 2 5 . Qc4+ Kd6 2 6 . Q d 5 + (2 6 . Rd l + or 2 6 . Rf6+
are both met by 2 6 . . . Ke5 !) 2 6 . . . Kc7 2 7 .Rxf7 + B e 7 , when Black has sur
vived "\Vhite's assault and will win the game.

23 ... Rd6?
Sacrificing the Queen in order to put an end to the attack. At the time I
thought this was the only defense, but it turns out that another, much
safer, choice existed: 2 3 .. .f5 ! 24.Qc7+ (2 4.Re l wins the Queen but ends
the attack. However, "\Vhite still retains some initiative after 2 4 . . . Qxe l
2 5 . Qxe l + Kf7 2 6 .Ba4! Bd6 2 7 .Bb3 +) 2 4 . . . Kf6 2 5 . Qxd8+ Kg7 2 6 . Rd l Be7
2 7 .Qa5 Rc8 2 8 . Re l Qc5

=.

I should mention that the "tricky" 23 ... Qf4? ?

would be refuted by 24.Qe l + .

24.Qc7+!
He doesn't even think o f 24.Re l , which wins the Queen but ends the at
tack and merely restores material equality (once again, "\Vhite would have
a slight initiative after 24 . . . Qxe 1 + 2 5 . Qxe 1 Re6 2 6 . Qb4+, but I would have
welcomed this after the hair-raising horrors I'd been through ! ) . After
24.Qc7+, I think that Black is losing.

24 ... Ke6 2 5.Qxf7+

PAL B E N KO : MY LI FE. GAM ES AN D C O M POSITI O N S

24

A position worth looking a t ! Black has a n extra Rook, but his King i s per
ilously placed in the center. The King is safe at the moment, but White's
attack isn't over.

26.Bd7
It doesn't help White to check aimlessly and chase the King to a safer
port. This move covers the e6-square and creates a strong threat of 2 7 . QeS+.

26 . . . Kd4
The complications arising from 2 6 . . . Qd2 ! ? 2 7 .Ba4! are also interesting, but
Black still seems to be in serious trouble. I'll only give one line: 2 7 . . . Qxb2
2 S .Qf4+ Kd5 2 9.Bb3+ Kc6 3 0 . Qa4+ Kb6 3 1 . Qb4+, and it's over.

27.h4? !
Finally White goes wrong. The best move was 2 7 .Bg4 ! , when

would

have been hard pressed to resist the attack: 2 7 . . . Bh6 (Black also goes down
in flames after both 2 7 . . . Kc5 2 S . Qxa7+ Rb6 2 9 . Qc7+ Kb 5 3 0. a4+ Ka6
3 1 . B c S , and 2 7 . . . a6 2 S . Q a 7 + Kd 3 2 9 . Qc7 Kd4 3 0 . Q a 5 ! Bg7 3 1 . Rd l +)
2 S . Qxa7+ Ke5 2 9 . Qe7+, 1 -0.

2 7 . . .Qd2 2 8 .Bg4 Bh6 !

This Bishop finally gets developed, though the cost turns out to be high:
Black will soon lose his Queen.

29.Rd l
Not s o good i s the tempting 2 9 . Qxa7+ Kc3 3 0.Qa3 + Kc4 3 1 .Qb3 + Kb5 , and
the King slips away: 3 2 .a4+ (a better try than 3 2 .Rd l ReS !) 32 . . . Kb6 3 3 .Qc4
RfS 3 4 . a 5 + Kb 7 ! 3 5 .a6+ Rxa6 3 6 .BcS+ (or 3 6 . RxfS BxfS 3 7 . BcS+ KbS
3 S .Bxa6 Qe l +, with a draw) 3 6 . . . KbS 3 7 .Bxa6 Rxfl + 3 S . Qxfl Qxb4

=.

29 ... Qxd 1 + 3 0.Bxd l Re8 !

Now the Rook finally gets to be heard.

3 1 .Qxa7+
Best. Black holds after 3 1 . Qe7 BfS 3 2 . Qxa7 + Kd3 , when the threat of
3 3 . . . Re I keeps him alive and well.

3 1 . . . Kc3 3 2 .Bg4? !

AFTE R T H E WAR

Also inadequate is 3 2 .Bb3 Kb2 3 3 .h4 Rc 1 + 3 4.Kh2 Bf4+ 3 5 . g3 Rc3 ! . How

ever, White should have tried 3 2 . Q a 3 + Kd2 3 3 .Qh3 ! , when the exchange
of Bishops leaves White, who has a Queen and two connected passed pawns
versus two Rooks, with excellent winning chances.

32 . . . Kb2 ! , 1 /2_112 .
Now that Black's King has found shelter, his colleague begins to feel the
danger. I offered a draw here and sent my opponent the following line:
3 3 . Bxc8 (3 3 .h3 leads to mate, and 3 3 . g3 is not sufficient due to 3 3 . . . Rc2 )
3 3 . . . Rd 1 + 3 4 . Q g 1 Rxg 1 + 3 5 .Kxg 1 Be3 + 3 6 . Kf1 Kxa2 , and the opposite
colored Bishops ensure a peaceful result.
Who says draws are boring? Somehow I came to disregard correspondence
chess . I should also add that, during this period of Stalinism in Eastern Europe,
it was not wise to engage in correspondence with other countries.
As I've already stated, 1 947 wasn't a busy chess year for me. The one impor
tant exception was my first international competition: the Vienna-Budapest team
match, held in Vienna in the American Zone (game six) .
Since gaining the master title in 1 945 , I was now recognized as a strong player.
When this recognition and respect led to a trip outside of Hungary, I began to
see chess as a potential stepping-stone to bigger and better things.

(6) Benko - Lounek (Vien na, 1 947)

l .Nf3 d5 2 .d4 Nf6 3 .c4 e6 4.Nc3 c5 5.cxd5 Nxd5 6.e4
During the Second World War, especially in its final years, almost all chess
activity ceased in Hungary. I could acquaint myselfwith master games only
from a few books. Fortunately for me, one of those books was about the
Alekhine-Euwe 1 9 3 7 match. As will be seen, this had a direct influence on
this game.

6 . . . Nxc3 7.bxc3 cxd4 8.cxd4 Bb4+ 9.Bd2 Bxd2 + 1 0.Qxd2 0-0

In the early to mid '60s this position was considered equal, and it was only
the Spassky-Petrosian game (mentioned below) that made everyone change
the assessment. Though I, as the game against Lounek proves, never be
lieved that this line gave Black equality, it was always fun to try a new
opening idea that left my opponent completely on his own. One such idea
was played by me against Bisguier at the 1 966 U. S . Open. The move or
der, which held the KN's development back for a while, was l .c4 Nf6 2 . d4
e6 3 .Nc3 d 5 4.cxd5 Nxd 5 5 . e4 Nxc3 6.bxc3 c 5 . Now I played my new
move : 7 .a3 ! ? This is an attempt to steer away from drawish variations by
preventing the exchange of the dark-squared Bishops, while still main
taining many options. I've never seen this move in this position, so I can't
say whether or not it's original, but at the time it seemed well worth try
ing. After 7 . a 3 , the B i s guier game continu e d : 7 . . . cxd4 8 . cxd4 B e 7

25

PA L B E N KO : M Y LI FE, GAM ES AND C O M POSITI O N S

9.NB 0 - 0 1 O.Bd3 Nd7 (More natural is 1 0 . . . Nc6, but 1 1 .Bb2 Q a S + 1 2 .Qd2

gives White the better endgame. Also possible is 1 2 .Kf1 ! ? , with the idea
of playing for an attack with h2 -h4.) 1 1 .0-0 b6 1 2 . Qe2 Bb7 1 3 .Bf4 Nf6

26

I 4.a4 White now stands better due to his well-defended center and good
kingside attacking prospects, as well as queenside action with a4-aS . My
advantage grew after. . . 14 . . . NhS l S .Bd2 Qd6 1 6.aS Nf4 1 7 .Bxf4 Qxf4 1 8 .a6
( 1 8 . axb6 also favors White) 1 8 ... Bc6 1 9 .NeS Rac8 2 0 . g3 Qh6 2 1 . Rac 1 Ba8
2 2 . Rc4 Rfd8 2 3 . Rd 1 g6 2 4. Qc2 Rxc4 2 S .Bxc4 Qg7 2 6 .BbS Bf6 (Bisguier
offered a draw here, but his troubles were far from over.) 2 7 .Nd7 ! Be7
(and not 27 ... Bxd4? 2 8 .Qc7 Re8 29.eS and wins) 2 8 . Bc6 Rc8 2 9 . d S , and
White eventually won on the S 2 nd move.

I l .Bc4 Nc6 1 2 .0-0 b6 1 3 .Rad l !

To quote Gligoric's notes to the game Polugaevsky-Tal, USSR ch. 1 969

(they appeared in the Jan. 1 97 0 issue of

Chess Life):

"This move was rec

ommended by Botvinnik back in 1 9 3 7 but it was Spassky who was first to

play it, thirty-two years later, in the fifth game of his recent match with
Petrosian. " Not only Gligoric, but commentators all over the world hailed
this "new" move, and even the

Chess Informant

gave the move an "N,"

which signifies a theoretical novelty. Of course, one does not have to be a

world champion, either past or present, to think of such a move, espe
cially if one has studied the games of the Alekhine-Euwe return match of
1 9 3 7 , as I have. In the 1 8th game of that match, Alekhine played 1 3 . Rfd 1 .
When looking over that game it struck me that 1 3 . Rad 1 was more natu
ral. I was quite happy to get a chance to test that opinion !

1 3 . . . Bb7
In the Polugaevsky-Tal game, Black chased the Bishop off its active di
agonal by 1 3 . . . NaS . However, after 1 4.Bd3 Bb7 I S . Rfe 1 ! Rc8 1 6 .d S ! ! exdS
1 7 .eS Nc4 1 8 . Qf4 Nb2 1 9. Bxh7 + ! , White got a murderous attack and won
in brilliant style. Lately, this kind of reversal has prompted Black to try
1 1 . . .Nd 7 , so as to leave the c-file open and to transfer the Knight to f6
where it can defend the King. Nevertheless, the same Rook setup com
bined with an eventual dS break still looks good for White.

1 4.Qf4

AFT E R T H E WAR

Actually, 1 4. Rfe l ! was played here in the above-mentioned Spassky

Petrosian game. I'm probably justified in calling 1 4. Qf4 a novelty - I say
probably because it is a natural move and might have been played in some
unknown, earlier gam e . After 1 4. Rfe l ! , Spassky-Petrosian continued
14 . . . RcS I S . dS exd S 1 6 . BxdS NaS (In a game between Hort and Unzicker
in 1 969, Black tried to improve with 1 6 . . . Qc7 ! ? 1 7 . QgS [Is this Queen
adventure necessary? I would prefer 1 7 .Re 3 .] 1 7 . . . h6 I S . Qg4 RfdS 1 9. QfS ?
[ 1 9.h3 has been recommended] 1 9 . . . Nb4, and Black got the upper hand.)
1 7 . Qf4 Qc7 I S . QfS BxdS 1 9 . exdS , and the central passed pawn became
decisively strong.

14 . . . Rc8 1 S .dS exdS 1 6.BxdS Qe7 1 7.NgS NeS

White's threat was to capture three times on f7 and then continue with
Rd7 + . Another defensive try was 1 7 . . . Ba6, but after I S . Rfe l Ne S 1 9 . QfS
g6 (Not 1 9 . . . Ng6? 2 0.Nxh7 Kxh7 2 1 .QhS+ KgS 2 2 .Qxg6) 2 0 .Qf4 h6 2 1 .NB
NxB + 2 2 . QxB , White's position is obviously superior (Black's weakened
kingside structure will be highlighted by e4-eS -e6).

1 8.Bxb7 Ng6 19.QfS Qxb7 20.Rd7 Qa6

This position is almost identical to the oft-mentioned Alekhine-Euwe game.

The difference is that White's fl -Rook stood on al in that game, making
the back rank rather vulnerable. Alekhine played 2 1 .M (avoiding 2 1 . Rxf7
Qxa2 !), and after 2 1 . . .RcS the game ended in a draw after a hard fight. In
my game, the Rook stands on fl - a clear improvement (because now
2 1 .Rxf7 is possible) ! However, I could still get over-ambitious and throw
the game away with 2 1 . Qh 3 ? ? Qxfl + ! 2 2 .Kxfl Rc 1 + 2 3 .Ke2 Nf4+ .
Amazingly, Black can do better than 2 0 . . . Qa6 by playing the risky looking 2 0 . . . Rc7 ! . The second player is alive and kicking after 2 0 . . . Rc7
2 1 . Rfd i h6 2 2 .NB (2 2 .Nxf7 ? Rxd7 2 3 .Rxd7 Qc6 favors Black) 2 2 . . . Rxd7
2 3 .Rxd7 Qa6.

2 1 .Rxf7 Qxa2 22 .h3 Rxf7

White retains the advantage after 2 2 . . . RcS 2 3 . RxfS+ NxfS 2 4 . e S , since his
maj ority of pawns is far more threatening than Black's.

27

PA L B E N KO: MY LI FE, GAMES AN D C O M POSITI O N S

2 3 .QxcS+ Rf8 24.Qd7 h 6 2 S .Ne6 Rfi 26.QeS+ Nf8 2 7.NdS Rd7 2S.eS !
Will this pawn decide the game? It seemed so at the time. The point is

28

that 2 7 . . . QdS fails to 2 S .e6 RxdS 2 9 .Qf7+ Kh7 3 0 .e7 ! Qxf7 3 1 . exdS=Q. 1t
is interesting to note that in this as well as other variations, the white Rook
is only an onlooker.

2 S . . . Qa3 ? ?
Black fails t o find the only defense. He had to try 2 S . . . Rc7 ! 2 9.e6 Qe2 ! ,
when it's remarkably hard to demonstrate the expected white advantage :
3 0 . Rb 1 b S ! ! (and not 3 0 . . . Qd2 ? 3 1 .Nc6 ! ) 3 1 . Qxb 5 Rc 1 + 3 2 .Kh2 Qxf2
3 3 .Rxc 1 Qf4+ .

29.Nc6
Now it's all over. The threat of eS -e6 (or Ne7+, if Black moves his Queen
off the a 3 -fS diagonal) is overwhelming.

29 . . . Rc7 3 0.e6 Kh7 3 1 .e7 Ne6 3 2 .Qf7 Qd6 3 3 .QfS +

Good enough, but the strongest move was 3 3 . Re 1 ! , when 3 3 . . .NgS loses
miserably to 3 4.QfS + g6 3 S .eS=Q (Also crushing is 3 S . QfS Ne6 3 6 .Rxe6)
3 S . . . gxfS 3 6 .Ne7 ! .

3 3 . . . g6 3 4.Qf7+ Ng7 3 S .eS=Q

I could have won a piece by 3 S .eS=N or 3 S . Qxg7 + Kxg7 3 6 .eS=N+ . These
moves are prettier, but promoting to a Queen is the fastest winning method.

3 S ... Rxf7 36.Qxf7 Qxc6 3 7.Qxa7, 1 -0 .

I participated in m y first international tournament in Budapest in the spring
of 1 945 . In this event, the best players of the so-called People's D emocracies
participated, along with Grandmaster S. Tartakower (who represented France)
and B.H. Wood from England. This sixteen-player event ended with Laszlo Szabo
taking top honors, in front of Gligoric, Foltys, Pachman, and Tartakower. I fin
ished in the middle of the field with six wins, five draws, and four losses. My
defeats were mainly a result of poor theoretical preparation, something I simply
didn't have time for since I had been occupied with university exams before this
tournament started. The following game was my best performance.

(7) Foltys - Benko (Budapest. 1 948)

l .e4 cS 2 .NB Nc6 3 .d4 cxd4 4.Nxd4 Nf6 S .Nc3 d6 6.Be2 e6 7.0-0 Be7
S.Kh l 0-0 9.f4 Qc7 1 0.BB RdS ! ?
I had my own ideas about how the Scheveningen should b e handled! The
text lets me answer a quick wing attack by g2 -g4 with a central counterstrike
by . . . d6-dS or . . . e6-e S .

1 l .Nb3 a 6 1 2 .a4 b 6 1 3 .Be3 RbS 1 4.Qe l NaS !

AFT E R TH E WAR

29

This is the point of Black's setup . Now 1 5 .Nxa5 bxa5 gives Black a strong
initiative thanks to the open b- and c-files, and it's doubtful if the doubled
a-pawns will ever become a liability (In a way, Black gets B enko Gambit
type queenside play without sacrificing a pawn ! ) . At the time, . . . Na5 was a
new idea ( . . . Nb4 was usual) . However, today (in many different forms and
positions) it's accepted as a normal maneuver.

1 5 .Rd l Nc4 1 6.Bc 1 bS 1 7.axbS axbS 1 8.Nd4

I reached the same position twenty years later in the game Reyes-Benko,
Lugano 1 965, only this time White moved his Knight back to e2 . It hardly
turned out to be an improvement, though, and I got a winning advantage
after: I S .Ne2 e5 1 9. Qg3 Bb7 2 0.Nc3 b4 2 1 .Nd5 Bxd5 2 2 . exd5 e4 2 3 .Be2
Nb6 2 4.Nd4 Nfxd5 2 5 .f5 Bf6. I went on to score the full point.

1 8 ... b4 1 9.Nce2 e S !
After having driven White back o n the queenside, Black now takes over
the initiative both in the center and on the kingside.

20.Nb3
Of course, 2 0 .Nf5 loses to 20 . . . Bxf5 2 1 . exf5 e4.

20 ... dS ! 2 1 .fxeS NxeS 22. exdS Nxf3 2 3 .Rxf3 NxdS

Grandmaster Geza Maroczy wrote the following in the tournament book:
"Black played very skillfully, and now he stands somewhat better. " As a
young man with virtually no international experience, it pleased me a lot
to read these kind words by a legend like Maroczy!

24.Rfd3 Bb7 2 S .Nf4 Nxf4! 26.Bxf4 Qc6

Of course, 26 . . . Qxf4? ? walks into a back-rank mate after 2 7 . RxdS+ RxdS
2 S .RxdS+ BxdS? (2 S . . . BfS lets Black play for a while) 2 9 . QeS mate .

27.Qe2 Rxd3 2 8.Rxd3 Re8 29.NaS ? !

It's understandable that White wants to get rid o f the strong battery down
the as-h I diagonal, but he didn't foresee the counter-blow. The tourna
ment book gave 2 9 .Re3 as an adequate defense: 29 . . . RdS 3 0 .Rd3 Rxd 3
3 1 .cxd3 Qd5 3 2 .h3 BfS 3 3 .Nd2 Qd4 3 4. B e 5 Qh4 3 5 . Qe3 . Probably the

PA L B E N KO : MY LI FE, GAMES AN D COM POSITI O N S

game i s about equal after this, but Black could have played better: 3 2 . . . g5 ! ,
instead o f 3 2 . . . B fS , is far more effective and leaves White under serious
pressure .

30

29 ... BgS !
Also possible is 2 9 . . . Qe4 ! ? , since 3 0 .Re3 Qxf4 3 1 .Nxb7 Qc7 leaves the
Knight trapped. So 2 9 . . . Qe4 100ks pretty good, but White had other op
tions that might not have been so clear. Thus, I decided to cash in my
positional advantage by trading down to an endgame with an extra pawn.

3 0.Re3
Black wins after 3 0 .Nxc6 Rxe2 3 1 .Kg I Bxc6 3 2 .Bxg5 Rxg2 + B .KfI h6.
Even worse is 3 0 .Qf2 Qxg2 + 3 1 . Qxg2 Re I mate.

30 ... Rxe3 3 1 .Bxe3 Qxg2 + 3 2 .Qxg2 Bxg2 + n .Kxg2 Bxe3 3 4.m Bd4
3 S .Nc6
White has no time for 3 5 .b 3 (going after the h4-pawn) , since the black
kingside pawns would be too fast. Instead, he tries to exchange as many
pawns as possible.

3 S ... Bxb2 3 6.Nxb4 BeS 3 7.h3 fS 3 S.Nc6 Bd6 3 9.c4 Kf7 40.Nd4 Kf6
4 1 .Nb3 gS 42 .cS BeS 43 .c6 Ke6 44.NcS+ KdS 4S .Nd7 Bd6 46.c7 !
This advance of the c-pawn is his best chance. Completely hopeless is
46.Nf6+ Kxc6 47 .Nxh7 B e 7 .

46 ... Bxc7 47.Nf6+ KeS 4S.Nxh7 BdS 49.NfS Kd6

SO.Ke3
White's plan has almost succeeded, but in the process his Knight has ven
tured into unsafe territory. I expected him to play the most challenging
line: 5 0.Ng6 Bf6 5 1 .h4 ! ? g4+ ! (After 5 1 . . .gxh4 5 2 .Nf4, White is closing in
on the draw. If he can sacrifice his Knight for the f-pawn, Black won't be
able to win since a classic Bishop and wrong-colored rook pawn position
would be achieved.) 5 2 .Kf4 Ke6 5 3 .h 5 (5 3 .NfS+ Kf7 54.Nd7 Bd4 5 5 .NbS
B e 5 + 5 6 . Kxe 5 g3 wins for Black) 5 3 , . . Bd4! 5 4.Nf8 + Kf7 5 5 .Kxf5 (5 5 .Nh7
Bf6 ! ) 5 5 . . . g3 5 6 .Ne6 Be3 5 7 .h6 g2 5 S .h7 g I =Q 5 9 .hS=Q Qg6+ 60.Ke5
Qxe6 mate . These lines are very much like a composed study.

A FTE R T H E WAR

50 ... Bf6
I toyed with the idea of cornering the Knight on the other side of the
board: 50 . . . Ke7 5 1 .Ng6+ Kf6 5 2 .NfS Be7 5 3 .Nd7+ Ke6 54.Nb8 Kd6 5 5 .Kd4
f4 5 6 . Ke4 Kc7 5 7 .Na6+ Kb7 , and the Knight is lost. However, I finally
decided to stick to my original plan.

5 1 .1ffi
This was his last chance to try 5 1 .h4 ! ? g4! 5 2 .Ng6 (52 .Kf4 B e 5 +) 5 2 . . . Bg7,
though Black still wins with some care . After 5 1 .Kf3 , I force the exchange
of minor pieces when the King and pawn endgame offers no difficulties.

5 1 . . .Ke7 52 .Nh7 Kf7 5 3 .Nxf6 Kxf6 54.Kf2 Kg6 5 5 .Kg2 Kh5 56.Kg3
f4+ , 0- 1 .
Unfortunately, the most interesting part of this endgame could only be
published in my notes. In any case, Maroczy considered it the best endgame
of the tournament.
The Budapest tournament had hardly ended when I was invited to participate
in another event, this time in the beautiful Austrian spa town of Bad Gastein.
Originally, various Soviet masters were invited, but they chose not to play. Per
haps they couldn't because Bad Gastein was situated in the American zone. What
ever the reason was, their refusal diminished the prestige of the tournament,
and this led to the prize fund being cut by quite a bit. In response, the Swiss
master Grob withdrew in protest.
Even though the start was chaotic,
the site turned out to be very comfort
able, thanks in part to the Marshall
Plan (U. S . Secretary of State General
George Marshall's plan to offer free
loans to war-torn countries that needed
to rebuild their economies. Austria
made use of it, but Czechoslovakia,
Poland, and Hungary declined.). After
so many years of hardship, the luxury
of a spa (combined with the very use
ful experience I picked up at the pre
vious grandmaster tournament) did
wonders for my play-I started with
a perfect 4-0 and, in the fifth round,
found myself paired with Polzer, who
was in l a s t . Full of c o n fi d e n c e , I
couldn't have guessed that I was in for
a very unpleasant surprise !
(Photo courtesy USeE)

31

PAL B E N KO : M Y LI FE, GAMES AN D C O M POSITI O N S

(8) Polzer - Benko (Bad Gastein, 1 948)

l .NO Nf6 2.c4 d6 3 .d4 Nbd7 4.Nc3 g6 5 .e4 e5 6.dxe5 dxe5 7.Bg5
32

Bg7 S.Nxe 5 ? !

I have t o confess that, upon seeing this move, I almost fell out o f my chair!
Naturally, I suspected that I had walked into some sort of well-known trap,
and I began to scold myself for not knowing theory.

S . . . Nxe5 9.QxdS+ KxdS l O.Nd5

Our table was now surrounded by excited spectators who shared my view
that I'd fallen into a trap. This impression was verified when, some hours
later, I was told that my opponent had already won a game from this same
position.

l O . . . Be6 ? !
Giving the piece back s o I could reestablish material equality. However,
since this gives White the advantage, and since l O o o .Ned7 ? 1 1 .0-0-0 leaves
Black busted, I decided the trap was correct and, at the first opportunity,
that I should use it myself as White against some other poor, unsuspect
ing, fool.
I must admit that I tried to spring it on Grandmaster Lilienthal but,
luckily for me, he avoided this line by answering BgS with an immediate
o o .h7-h6. I didn't come close to catching someone in this line again until I
gave a 1 949 simultaneous exhibition in Moscow against a group of college
students . In one of the games, my young opponent allowed me to reach
the position after l O.NdS (yes, I did experience a certain glee when I played
S .NxeS). Sure that this was going to be easy, I made my moves in the other
games and finally got back to my "trapped victim, " who calmly tossed out
1 0 o o .Neg4! and wiped me off the board !
It turns out that 1 0 . o oNg4 refutes S .Nxe S , since 1 1 .0-0-0 ( l 1 . Rd l Bd7
1 2 . e S loses to 1 2 o o .ReS) 1 1 . . .Bd7 ! (The tempting 1 1 . . .Nxf2 isn't as strong:
1 2 .Be2 Nxd l 1 3 .Rxd l Be6 1 4.eS ! BxdS I S . exf6 BfS 1 6 .RxdS +) 1 2 .eS (He
should try 1 2 .Be2 c6 1 3 .Nxf6 Bxf6 1 4.Rxd7+ Kxd7 I S . Bxg4+ Ke7 , though
this is, of course, also miserable for White.) 1 2 o o .Bh6 ! 1 3 .Bxh6 NxdS forces
White to resign. Remarkably, where I and other masters failed to find the

I do battle with G randmaster Kotov, Budapest vs. Moscow Match 1 949.

proper defensive idea, an unknown Russian youth in a simultaneous exhi

bition succeeded ! I have no idea how strong he was, but I do know that in
simultaneous exhibitions I've never come across more dangerous oppo
nents than Russian college students !

1 1 .0-0-0 Ke8 1 2 .Bxf6 Bxf6 1 3 .Nxf6 Nxe4 1 4.Bxe4 Bxe4 1 5 .b3 Be6

Though Black has regained his pawn, the Knight on f6 is very strong. In
addition, the a8-Rook is cut off from the game. My next few moves address
this problem.

1 6.f4 a5 1 7.g4 Ra6 1 8.e5 a4 1 9.Kb2 axb3 20.axb3 Rb6 2 1 .Rd3 Rd8
22 .Rxd8+?
Of course, 2 2 .Rhd l ? ? would have been a mistake due t o 2 2 . . . Rxb 3 + . How
ever, White could have

given

me a hard time with 2 2 .Ra l ! Ra6 2 3 .Rxd8+

Kxd8 2 4 . Rxa6 bxa6 2 5 . g5 , followed by

Nxh7

when he has a winning

PAL B E N KO : MY LIFE, GAM ES AND C O M POSITI O N S

endgame. Fortunately, my opponent respected m e too much and suddenly

began playing for a draw!
34

22 ... KxdS 2 3 .Rd l + Ke7 24.NgS+, If2_11z.

He asked me for a draw that, of course, I couldn't refuse! Since 24 ... Kf8
2 5 .RdS+ Kg7 2 6.Nf6 walks into a mating net, I would have been forced to
play 24 . . . KeS and grovel for my very life !
This game, a humiliating near-loss against the eventual last-place
finisher, taught me to never underestimate an opponent. In the end, I man
aged to share the second and third places with Rossolimo, losing only to
Lundin, who took first.
Another memory I have of Bad Gastein was meeting a very beautiful young
woman during the tournament. She was wonderful, and we spent a lot of time
together and became pretty close. As was so typical of those times, we somehow
lost track of each other.
Twenty years went by and, by then, I was living in the United States. One day
a letter arrived. Amazingly, it turned out to be from her ! She told me that she
owned a hotel in Switzerland and she invited me to visit. I was polite and sent
her a postcard, but a second letter followed where she admitted that she now
had four children and was twice her previous size ! Again I replied and politely
refused her invitation. Sometimes it's best to leave the past alone.
In 1 945, I was making a bit of extra money working in a textile factory. Since
many of the workers there lifted weights, I joined them and also had some fun
playing for their chess team. This newfound physical strength turned out to be
useful. Around this time, I competed in an international event in Bucharest, Ro
mania. It turned out that the Romanian gymnastic team (which mostly consisted
of Hungarians ! ) was staying in the same hotel. When some free time appeared, I
went with them to work out and made a bet with one of the professional gym
nasts that I could beat him to the top of the rope. I won, and was rewarded with
two kilos of Halva.
With work, school, tournaments, and simultaneous exhibitions wearing me
down, I decided to take some time off and enj oy a trip to the beach with a girl
friend. We were hanging out by a swimming pool and ran into another boy
friend of hers. He and his friends began to make rude remarks and then one of
them pushed me into the water, trying to humiliate me in front of the girl. Natu
rally, I couldn't let him get away with this kind of behavior so I hit him. He took
a swing or two and I just grabbed him and forced his head underwater, and I was
so strong that he wasn't able to do anything about it. He pulled my hair and
struggled, but soon he went limp and it took several police to pull me off of him.
The guy was half dead, but since he started it, they let me go (though I was
forced to pay a fine) .
When I went back to Budapest a couple of days later, a friend asked me about
my fight with the boxer. I was stunned; how could he have known about this?
And why did he think this guy was a boxer? Then he showed me the newspaper,

AFT E R TH E WAR

and the headline read: " Chess master beats up professional boxer on the beach . "
I n the story, it said that chess players might be tougher than people thought!
The rest of 1 948 and all of 1 949 was spent tending to my university studies,
with just a little chess on the side (games 9- 1 3 ) . One highlight was my winning
the Hungarian Championship. This success eventually helped me realize that a
very important decision had to be made: should I continue my education or put
my energies into chess? It was a no-contest: I left school and, in 1 9 5 0 , became a
bookkeeper. I had hoped that I would be able to work and study chess too . The
j ob served its purpose, bringing in some much-needed money, but it turned out
that I had no time whatsoever to improve my game.

(9) Benko - Fiister (Hungarian Championship, 1 948)

l .e4 cS 2 .Nfl Nc6 3 .d4 cxd4 4.Nxd4 Qc7
More usual is 4 . . . Nf6, which hinders c2 -c4 since the best reply, 5 .Nc3 ,
blocks the c-pawn.

S .NbS! QbS 6.c4

White managed to successfully carry out c2 -c4 and, as a result, Black has a
cramped position. He can't easily free his game with a subsequent . . . d7 -d5
and, in addition, his usual pressure on the c-file has also been nullified.

6 . . .Nf6 7.NSc3 !
The natural 7 .N l c3 allowed 7 . . . a6 8 .Nd4 Nxe4! 9.Nxe4 Qe5 .

7 ... e6 S.Be2 Be7 9.0-0 0-0 1 0.Be3 RdS 1 1 .f4!

In the game Ran-Flohr, Moscow 1 93 6, White answered 1 0 . . . Rd8 with I I .a 3 ,

i n order t o prevent . . . Nb4 after the expected . . . d7 -d5 . The annotator put
a question mark to my 1 1 . f4 due to the possible . . . d7 -d5 reply. Luckily, I
didn't know that game or the analysis of this position, so I innocently en
tered the minefield that was discussed so many years earlier. When I played
1 l .f4, Fiister replied very quickly, which is something he rarely did. Right
away I realized that I'd walked into his preparation, but it was too late to
back down now!

35

PA L B E N KO : MY LI FE, G A M E S AN D COM POSITI O N S

1 1 . . .dS ? ! 1 2 . cxdS exdS 1 3 .NxdS!

This allows the b l -Knight t o leap t o c3 and participate i n the fight for d S .
36

13 . . .NxdS
And not 1 3" .Nxe4? ? 1 4.Nxe7 + Kf8 I S .Nxc6.

1 4.exdS Nb4 I S .Nc3 BfS 1 6.g4! Be6

Avoiding the poor 16 . . . Nc2 ? 1 7 .Bf2 .

1 7.BB Qc8 1 8.fS BcS 1 9.Bf2 BxdS 20.NxdS NxdS 2 1 .BxdS

Also strong was 2 1 .Rc l Bxf2 + (Hopeless is 2 1 . . .Ne3 2 2 . Qe2 Nxf1 2 3 .RxcS .)
2 2 .Rxf2 Nc7 2 3 .Qb3 Rb8 24. Rfc2 Rd7 2 S .Qe3 Qd8 2 6 . Qxa7 , with a pawn
more and a winning position.

2 1 . . .Bxf2 + 2 2 .Rxf2 QcS 2 3 .Bxf7+ Kxf7

The position has cleared up, with White emerging with an extra pawn.
However, he must play cautiously due to his open kingside.

24.Qb3 + Kf8
Black shows no interest in going into a pawn-down Rook endgame after
2 4 . . . QdS .

2 S .f6 Rd2
On 2 S . . . gxf6 , White should avoid 2 6 . Qe6? due to 2 6 . . . Rd6. However,
2 6 . Qxb7 is simple and strong.

26.fxg7+ Kxg7 27.Qxb7+ Kh6 2 8.QB Rxf2 29.Qxf2

Black's horribly exposed King prevents him from refusing the Queen ex
change .

29 . . . Rc8 3 0.QxcS RxcS 3 1 .Rfl KgS 3 2 .RfS + !

Returning a pawn s o I could enter a winning, and rather interesting, King
endgame.

32 ... RxfS 3 3 .gxfS KxfS 3 4.Kf2 Ke4 3 S .Ke2 Kd4 3 6.Kd2 hS 3 7.h4 Ke4
This is his only chance. Obviously, moving his King to the queenside would
have been a hopeless cause.

3 8 .Kc3 Kf4 3 9.b4 Kg3 40.bS Kxh4 4 1 .a4 Kg3 42 .aS h4 43 .b6 axb6
44.a6 ! , 1 -0 .

AFT E R T H E WAR

The point of White's play (44.axb6 is a theoretical draw) . Now 44 . . . h3

4 5 . a 7 h2 46 .a8=Q is easy, so Black gave up.
The following game was played in a team competition between Budapest and
Moscow. The Russian team consisted of a very powerful eight-player lineup that
included (aside from Ragozin) Smyslov, Bronstein, Kotov, and Lilienthal. I had
to play all of them twice, once in Budapest and once in Moscow. When the event
concluded, I was honored to have scored the best result on the Hungarian team.

( 1 0) Benko - Ragozin (Budapest vs. Moscow Match, Moscow, 1 949)

l .d4 Nf6 2 .c4 e5
At the time, the Budapest Gambit was somewhat rare. Nowadays it's gained
a measure of respectability and is quite common in open events, and is
occasionally seen at the highest levels of competition.

3 .dxe5 Ng4 4.Bf4 Nc6 5 .NfJ Bb4+ 6.Nbd2

Avoiding the much analyzed 6 .Nc3 Bxc 3 + (in the past, 6 . . . Q e 7 ? was con
sidered to be Black's correct move order, but now it's known that 7 . Re l !
Ngxe5 8 .Ngxe5 Nxe 5 9.a3 ! is very nice for White, while 7 . Qb 3 ! Na5 ?
8 . Qc2 Nxc4 is just bad for Black: 9 . a 3 BaS 1 0 . e 3 Ncxe 5 I l .b4 Bb6 1 2 .Nd5
Qd6 1 3 . Qe4 f5 1 4.Bxe5 fxe4 1 5 .Bxd6 cxd6 1 6 .Ng5) 7 . bxc3 Qe7 8 . Q d 5 f6
9 . exf6 Nxf6 1 0 .Qd3 d6, when White has an extra pawn, but the weakness
of White's pawn structure gives Black reasonable compensation.

6 . . . Qe7 7.a3 Bxd2 + ? !

O f course, theory has gone though many changes since 1 949 ! Now it's
known that the correct move order for Black is 7 . . . Ngxe 5 (threatening
. . . Nd 3 mate) 8 .Nxe5 Nxe 5 9 . e 3 Bxd2 +, and we end up back in the game.

8.Qxd2 Ngxe5 9.Nxe5 Nxe5 l O.e3

White could have punished Black's inaccurate seventh move by l O.c5 ! 0-0
(12 ... Qxc5 1 3 . Re l Qd6 1 4. Qxd6 [ 1 4.e3 ! ? ] 1 4 ... cxd6 1 5 .Be2 [ 1 5 .M ! ? , with
the idea of 1 6 .Rh 3 is a serious alternative] followed by 0-0 and Rfd l , leaves
White with more than enough pressure for the pawn.) 1 1 .Re l , with an edge.

l O . . . d6

37

PAL B E N KO : MY L I F E , GAMES A N D C O M POSITI O N S

We've reached one o f the key positions i n the Budapest Gambit. White's
Bishop pair gives him a slight advantage. Black's game, though solid, is
also a bit passive, and this kind of situation is not to the taste of most
38

Budapest gambiteers. I was sure that Ragozin, a very tactical player, couldn't
have been too comfortable here.

I 1 .Be2
Also common is 1 1 .Qc3 and 1 1 .Rc l .

1 1 .. .Bd7
A bit out of the ordinary, but not necessarily bad. More common is
1 1 . . .0-0 1 2 .0-0, when both 12 . . . aS ! ? and 12 . . . b6 are quite playable.

1 2 .0-0 Bc6
Worse is 12 . . . 0-0 ? ! 1 3 . cS ! , when 1 3 . . . dxcS ? ? 1 4. Bxe S QxeS I S . Qxd7 wins
a pIece.

1 3 .Rac 1 0-0 1 4.Bg3 ! ?

This avoids possible moves like . . . Ng6 o r . . . g S with tempo. Now I was
wondering what Ragozin would play - would he remain calm (quite out
of character!), or would he find some way to sharpen things up?

1 4 ... fS ? !
H e couldn't resist making this premature kingside demonstration. White
immediately counterattacks on the queenside and in the center.

I S .b4!
The threat is 1 6 .bS Bd7 1 7 . Q d S + , winning the b7 -pawn, while c4-cS is
also hanging over Black's head.

IS ... Kh8 1 6.bS Be8 1 7.cS !

Opening as many lines as possible for my Rooks and Bishops.

1 7 . . . Rd8
He accepts a weakness on d6. However, 1 7 . . . dxcS 1 8 .QdS was far worse.

1 8.cxd6 cxd6 1 9.Rfd l Bti 20.Qd4 b6

White has placed his pieces on excellent squares and isolated Black's d
pawn. This sounds very impressive, but how does White make progress?

A FT E R T H E WAR

2 1 .h4!
The immediate 2 l . Bh4 was met by 2 1 . . . g5 . Now, however, White threat
ens to play 22 . h 5 , when both Bh4 and h6 will cause Black some uneasy
moments .

2 1 . . .Bb3 22 .Rd2 Nfl 2 3 .hS Rd7 24.BB Qf6 2 S .Qb4!

The endgame, which I'm usually happy to go into, was also good for White,
but I decided that keeping the Queens on offered me even more.

2 S ... Be6 26.Bc6 RddS

This loses the Exchange, but 2 6 . . . Rc7 2 7 . Bxd6 would have left White with
both an extra pawn and a huge positional advantage.

2 7.Bh4 Qh6
If 27 . . . Ng5 , White would win material by 2 S .KfI (and not 2 S .f4? Nh3 +)
followed by f4.

2S.BxdS RxdS 29.Qh4 gS 3 0.Qd4+ Qg7

No better was 3 0 . . . KgS 3 l . Bd5 .

3 1 .BdS
The rest is just a matter of technique.

3 1 . . .BxdS 3 2 .QxdS f4 3 3 .e4 B

D esperation, but 3 3 . . . Ne 5 3 4 . Qxd6 ! breaks Black's back.

3 4 .Rc7 RfS 3 5 . gxB Qf6 3 6 . QfS QxfS 3 7 . exfS Kg7 3 S . Rxd6 KgS
3 9.Rdd7 h6 40.f6 g4 4 1 .f4 g3 42 .fxg3 , 1 -0.
A very smooth win against a strong Soviet player.

(I I ) Prins - Benko (H u ngary vs. Netherlands Match, 1 949)

1 .e4 cS 2 .NB Nc6 3 .d4 cxd4 4.Nxd4 Nf6 S .Nc3 d6 6.BgS e6 7.g3

An unusual continuation.

7 . Qd2 is seen in the vast maj ority of games.

7 ... h6 S.Be3 Be7 9.Bg2 0-0 1 0.0-0 Bd7 1 l .a4 a6 1 2 .Nxc6

Normally White doesn't play this move since it allows Black to bring a
side pawn towards the center (via . . . bxc6). In this case, though, White wishes
to make use of the weakened b6-square by a4- a 5 , Na4 and Nb6 . He will
also apply some pressure to the a6-pawn (if Black recaptures with his b
pawn) with moves like Qd3 or Qe2 and/or Bfl .

1 2 ... bxc6
A tough choice. More cautious is 1 2 . . . Bxc6, when the b6-square can be
guarded by . . . Nd7 . The text, though, is also quite playable, and leads to a
more complex battle than 1 2 . . . Bxc6 .

1 3 .aS ! ? RbS

39

PAL B E N KO : M Y L I F E , GAM ES A N D C O M POSITI O N S

I f it hadn't been a team match (which means that you have a responsibility
to your teammates) I would most likely have tried 1 3 . . . c5 1 4. e 5 dxe 5 , sac
rificing the Exchange for a promising position (i.e., a pawn and play against
40

White's weakened kingside light-squares) . After 1 3 . . . Rb8 , I get a cramped

but solid position where it's possible to try calmly (and safely!) to outplay
my opponent.

1 4.Bb6 Qc8 I S .Re l eS

A good move that gains space and allows my light-squared Bishop to move
to a more active position on e6. Once this Bishop moves, then . . . Nd7 is
possible, challenging White's domination of b6.

1 6.Qd3 Be6 1 7.f4 exf4 1 8.gxf4 Nd7 1 9.Na4

A tense position has been reached where Black has to stay on top of the
pressure against d6, c6, b6, and a6. Though many moves are possible here,
I decided to end threats against a6 in the most dynamic manner possible.

1 9 . . . cS
Now 2 0 . Bfl can be met by 20 . . . c4.

20.eS?
This central break (which might have been more effective later in the game)
brings Black's pieces to life. White had many calm alternatives like 2 0 . Bfl
or 2 0 . Rad l or 2 0 . Kh l . Also very interesting is the ugly 2 0 . f5 ! ? , giving the
e 5 -square to Black's Knight: 2 0 . . . Ne5 2 1 .Qg3 Bc6 (It might be better to
play 2 1 . . .Bd7 2 2 .Nxc5 Bb5 2 3 .Nb3 Qxc2 24.Nd4 Qxb2 2 5 .Rab l Qd2 , with
a complex struggle ahead.) 2 2 .Nc3 (better than 2 2 .b3 Bb5 2 3 .Nc3 Bc6
2 4.Bxc5 dxc5 2 5 . Qxe5 B f6 2 6 . Qxc5 Bb7 2 7 . Qxc8 B d4+), when the dis
gusting 22 .. .f6 might be called for. Normally (after 22 .. .f6) White would
stand considerably better in such a position, but the fact that the Bishop
on b6 is locked out of the game makes things tolerable (the piece is inac
tive and vulnerable to a well timed . . . Nd7).

20 . . . dxeS 2 1 .fS ?
I expected 2 1 .fxe5 c4, when Black stands better. White's choice, 2 1 .f5 ? ,
just loses a pawn. Apparently h e thought h e was winning a piece, but my
reply brings him back to reality.

AFTE R T H E WAR

2 1 . . .BxfS 2 2 .QxfS Nxb6

Evidently, White forgot that his Queen on f5 would be unprotected.

2 3 .Qxc8 Nxc8 24.RxeS Bf6 2 S .RdS

Desperately trying to get a bit of counterplay. It's important to note that
White would draw if he could trade his three queenside pawns for Black's
two while also making sure to trade the Knights and all the Rooks. The
resultant Bishop and three versus Bishop and one is not winnable . Natu
rally, I was well aware of this and would never allow it to happen. In gen
eral, when there are opposite-colored Bishops on the board the stronger
side should try to retain as many pieces as possible in order to highlight
their attacking potential (i. e . , my Bishop can attack things that his Bishop
can't defend) .

2 S . . . RbS 26.c3
And not 2 6 .Rxc5 ? ? Rxc5 2 7 .Nxc5 Bd4+, winning a piece.

26 . . . Be7
The threat of . . . Rxa5 forces White to play aggressively.

27.c4 RxaS 28.b4 Rxa4

A complete answer to White 's tirade.

29.Rxa4 Nb6 3 0.Rxa6 NxdS 3 1 .BxdS cxb4

White is two pawns down and must hope that the opposite-colored
Bishops give him some drawing chances.

32 .Kf1
White would like to play 3 2 . Ra7 (or even 3 2 . Rb6, stopping Black's Rook
from getting behind his passed pawn) , but both moves lose to 32 . . . Bxc 5 + .

3 2 . . . Rb8 !
Just i n time. I f the b- and c-pawns were off the board, White would play
3 3 . RaS and force the drawn position discussed earlier. Unfortunately for
White, the existence of these pawns now makes the pure Bishop endgame
an easy win for Black.

3 3 .Ra7 b3 3 4.Rxe7 b2 3 5 .Be4 Kf8 3 6.Ra7 b l =Q + 3 7 .Bxb l Rxb l +

3 8 .Kf2 ? !
White should have tried to advance his passed pawn by 3 S .Ke2 . Black is
only one pawn up but he still retains good chances to win because he can
create two connected passed pawns on the kingside. Mter 3 S .Ke2 , I would
most likely have played 3 S . . . g5 . The greedy 3 S . . . Rb2 + 3 9 .Kd3 Rxh2 is very
tempting, but it accelerates the advance of White's c-pawn. Here's one
possible line: 40 .c5 Rb2 4 1 .c6 RbS 42 .Kd4 h5 43 .Kc5 h4 44. c7 RcS 45 .Kd6
g5 46 .Kd7 Rxc7 + 47.Rxc7 h3 4S . Rc2 g4 49 . Rh2 , when the game appears
to be drawn.

41

PA L B E N KO : MY L I F E , GAM ES A N D C O M POSITI O N S

3 8 . . . Rc l 3 9.Ra8+
White could have put up more resistance with 3 9.Rc7, though Black should
42

would still win after 3 9 . . . g5 40.h3 Kg7 .

3 9 . . . Ke7 40.Ra7+ Ke6 4 1 .Ra6+ Ke5 !

Intending to take the game into a winning King and pawn ending.

42 .Ra5+ Kf6 43 .Ra6+ Kg5 44.Ra7 Rxc4 45.Rxfi Rf4+ 46.Rxf4 Kxf4

Less experienced players usually think that such positions are always easy
to win, but g- and h-pawn versus h-pawn is often drawn, depending on
the position of the pawns and on the activity of the respective Kings . The
stronger side must be very careful before entering such endgames because
the slightest detail can turn an apparent win into a depressing draw. In the
present case Black wins because his King is dominant, and also because
his g-pawn stands on g7. This gives him critical tempo moves that wouldn't
exist if the pawn stood on g6 . Yes, it's amazing but true: if Black's pawn
stood on g6 the game would not be winnable!

47.Kg2 Kg4 48.Kf2

No better is 48.h3 + Kf4! 49.Kf2 h5 5 0 . Kg2 h4 5 1 .Kf2 g6 5 2 .Kg2 Ke3 ,
and Black wins the h3 -pawn and the game.

48 ... Kh3 49.Kgl h5

The trick to winning this position is to advance the h-pawn to h4 and
then, noting the position of White's King, to advance the g-pawn in such
a way as to create a won King and pawn versus King situation.

50.Kh l h4 5 1 .Kgl g6 !
Black can push his g-pawn one or two squares, and this allows us to see why
the pawn had to be on g7 when we first entered the King and pawn endgame.
If 5 L .g5 , White draws after 5 2 .Kh l g4 (Black would also be unable to win
after 52 . . . Kg4 5 3 .Kg2 Kf4 54.Kf2) 5 3 .Kg l g3 5 4.hxg3 hxg3 5 5 .Kh l , taking
the opposition and saving the game. By moving the pawn to g6 first, White
won't be able to take the opposition at the critical moment.

52 .Kh l g5 5 3 .Kgl g4 54.Khl g3 5 5 .Kgl

AFT E R T H E WA R

A final j oke . 5 5 .hxg3 hxg3 5 6 . Kg l g2 5 7 .Kf2 Kh2 was also the end.

55 . . . g2 , 0- 1 .
Not falling for 5 5 . . . gxh2 + ? ? 5 6 . Kh l with a draw.

( 1 2) Benko - Opocensky (Bucharest, 1 949)

l .d4 d5 2.c4 c6 3 .Nf3 e6 4.Nc3 f5
The text takes the game into a Stonewall variation of the Dutch. How
ever, the move order is a bit suspect and allows White to gain a small but
solid plus .

5 .Bf4!
Usual is 5 . g3 , but this move (not possible from normal/accurate Dutch
move orders) followed by e2-e3 is very nice for White .

5 ...Nf6 6.e3
This position first occurred in Staunton-Saint Amant, London 1 843 !

6 ... Be7
I would have answered 6 ... Bd6 with 7.Bd3 , since 7 ... Bxf4 8 .exf4 gives White
a lock on the important e 5 -square.

7.Bd3 0-0 8.0-0 Qe8

A known maneuver that envisions swinging the black Queen over to h5
with an attack against the white King. It's hard to imagine this being suc
cessful, all the more so when you consider Black's positional imperfec
tions, i . e . , the hole on e 5 and the horrible state of Black's light-squared
Bishop.

9.h3 Kh8
And not 9 . . . Qh5 1 0 .Ne5 . White would be happy to exchange Queens and
stop Black's attack before it even begins.

1 0.Bh2 Ne4 1 1 .Ne5 Nd7 1 2 .f3

At the time, I thought this was very promising. Perhaps a better way to
take advantage of White's lead in development is 1 2 .Bxe4 fxe4 1 3 .D Nxe5
1 4.Bxe5 exD 1 5 .RxD .

1 2 . . .Nxc3 1 3 .bxc3 Nxe5 1 4.Bxe5 b6

Black, in threatening to activate his Bishop by . . . Ba6, wants to force White
to capture on d 5 . This capture will give Black's light-squared Bishop more
squares to choose from and will, in some lines, open up the e-file, allow
ing Black's Queen to place pressure against e3 .

1 5 .cxd5 exd5 1 6.c4 Be6!

43

PA L B E N KO : MY LI FE, GAM ES A N D C O M POSITI O N S

44

Black begins t o defend extremely well. Far worse was 1 6 . . . B a 6 1 7 .cxd 5

Bxd3 1 8 . Qxd 3 cxd 5 , when the weakness of d5 assures White the better
chances .

1 7.cxdS BxdS 1 8.Qc2 Be6!

Defending f5 and preventing White from advancing the e-pawn with tempo.
Now 1 9 .Rac 1 can be met by either 19 ... c5 or 19 ... Ba3 .

1 9.Qc3 Bf6 20.f4 ! ? Qd7 2 1 .a3 BdS 2 2 .Rac 1 bS 2 3 .Qc2 Be7!

Another fine move ! Instead, 2 3 . . .Bxe 5 2 4.fxe5 gives White an obvious
advantage . Now the a 3 -pawn is under annoying pressure, and Black's
queenside pawns are free to roll and cause White all kinds of problems.
Throughout all this, White's "powerful" e 5 - Bishop has no impact on the
queenside battle.

24.Ra l
I considered sacrificing the a-pawn and going all out o n the kingside via
2 4.Kh2 followed by g2 -g4, but decided that it was too slow.

24 ... aS 2 S .e4! ?
Black has outplayed me positionally, but I still have trumps to use: This
move gets rid of the black f-pawn and allows me to pursue a kingside
attack via f4-f5 - f6 . Once my attack gets underway, my Bishop (which has
had no influence on the queenside) will become a hero.

2S ... fxe4 26.Bxe4 Bxe4 27.Qxe4 b4 28.axb4 axb4 29.Rab l Ra7 !

Defending e7 and g7 . The immediate 2 9 . . . Qd5 ? failed to 3 0 . Bxg7 + .

3 0.fS QdS 3 1 .Qg4 Bd6

The situation has changed, and now Black wants to trade Bishops since
White's is simply too strong an attacking piece.

3 2 .f6! BxeS H .dxeS gxf6 34.exf6 cS?

Too optimistic. I t appears that White's attack is over, so Black felt i t was
time to decisively advance his queenside pawns.

3 S .RfS Qe6 3 6 .Qf4

Of course, 3 6 . Rxc5 ? ? hung the Rook to 3 6 . . . Q e 3 + .

A FT E R TH E WAR

36 . . .c4?

Black still thought he was safe, and decided to shove his pawns to victory.
Naturally, 3 7 .Rxb4? ? lost to 3 7 . . . Qe l +. As it turns out, he had to play
3 6 . . . Raf7, even though it would cost a pawn after 3 7 .RxcS Rxf6 3 S.Qxb4.
3 7.f7 !

A real lightning bolt! Both players were in severe time pressure, and this
made it almost impossible for Black to hold the position.
3 7 ... Raxf7?

Losing immediately. Also hopeless was 3 7 . . . Rfxf7 3 S.Qd4+ KgS 3 9.RgS+,

leading to mate or loss of the black Queen. However, Black could have
still put up resistance with 3 7 . . . Rd7 .
3 S.Qd4+ Rf6
As horrible as it is, Black was forced to walk into this pin.

3 9.Rbfl Kg7 40.RgS + KhS 4 1 .g4! ! , 1 -0.

This is the point of White's play! Now the loss of Black's Rook is inevi
table since the threat of 42 .RgfS Kg7 43 .gS is impossible to deal with. If
4 1 . . .h6 42 .Rg6 ends things, while 4 1 . . .c3 42 .RgfS c2 43 .Rxf6 Rxf6 44.Rxf6
c 1 =Q+ 4S.Rf1 + also forces resignation.

( 1 3) Benko - Turi (Budapest, 1 949)

l .e4 e6 2 .d4 dS 3 .Nd2 cS 4.exdS QxdS

This capture has become very popular at all levels of the game. Black loses
a bit of time with his Queen, but he avoids an isolated d-pawn (which
would come about after 4 . . . exdS) and ensures a reasonably easy develop
ment for his forces.
S .NgB cxd4 6.Bc4 QhS

The usual Queen move is 6 . . . Qd6, while 6 . . . QdS is also seen from time to
time. With 6 ... QhS, Black is hoping to exchange Queens after Nxd4. How
ever, White doesn't have to go along with this.

45

PAL B E N KO : MY LI FE, GAM ES AND C O M POSITI O N S

7.0-0 Bd6?

46

This looks nice and active, but it's actually a serious mistake. Much better is
7 ... Nc6 S.Nb3 Nf6 (and not S ... eS? 9.NxeS Qxd 1 1 0.Rxd l NxeS 1 1 .Re 1 f6
1 2 .f4, as seen in Tarrasch-Thorold, l S90), with a playable position.
8.Ne4 BeS 9.b4! Be7

Also awful is 9 ... Bxb4 1 0.Qxd4 (Though the threat of 1 1 . Qxg7 is obvious,
there is no satisfactory way to guard the g-pawn.) l O ... BfS ( 1 O . . . Qg6 would
lose a piece after 1 1 .BbS+, while 1 O .. .f6 1 1 .Bxe6 Nc6 1 2 .Qc4 Bxe6 1 3 .Qxe6+
is also very much in White's favor.) 1 1 .NeS, when Black is hopelessly be
hind in development: l 1 . . .f6 1 2 .BbS+ Nc6 1 3 .Nxc6 QxbS 1 4.QdS+ Kf7
l S.Qc7+ Kg6 1 6.NdS, when the dual threat of 1 7 .Qg3 + and 1 7 .Qf7+ gives
White a winning attack.
1 0.Qxd4 Nf6 1 1 .Nd6+ Bxd6 1 2 .Qxd6

With two active Bishops and a lead in development, White's advantage is

obvious.
1 2 ... bS?

Now the game is over. However, even after other moves, Black's game
would be miserable: 12 ... Ne4 1 3 .Qd4 Nf6 1 4.BgS Nc6 ( 1 4 ... Nbd7 l S.Qd6)
l S.Qb2 ; 1 2 . . . Bd7 1 3 .Bb2 Bc6 ( 1 3 . . . Nc6 1 4.Rad 1 RdS l S.bS) 1 4.Bxe6! Bxf3
(no better is 14 . . . Ne4 1 S.Bxf7+! Qxf7 1 6.Rae l ) l S.Bxf6 gxf6 1 6.Rfe 1 , with
a winning attack for White.
l 3 .Ne S ! Bd7

Also hopeless is 1 3 ... bxc4 1 4.Nc6 (threatening mates on e7 and dS) 14 ... Nxc6
l S. Qxc6+ , picking up the Rook on a s , and 1 3 . . . Ba6 1 4. Bxe6 fxe6
l S.Qxe6+ KfS 1 6.Be3 with total carnage.
1 4.Bd3 a6 H .Be3

Black can't deal with the threat of BcS.

H ... NdS 16.Be4, 1 -0.

There's simply no reply, so Black wisely resigned. One of the shortest games
in my career!

oUPle of ye", ,"w me oementing my putation " tho be" pI.ye<

in Hungary (along with Szabo) . In 1 9 5 0 , FIDE created the international master

title. They only awarded this title to a few players, and I was quite proud to be
one of them. Unfortunately, I had no chance for the grandmaster title as long as
I lived in Hungary due to the lack of qualifYing tournaments (the powerful Maroczy
Memorial was a rare exception) .

Benko vs. Dr. Ogaba brings back happy memories for me. It features cutting
edge opening play (for the time ! ) and deep strategic planning, culminating in a
nice tactical display. In fact, games 1 4- 1 7 all show that my style during this pe
riod was quite sharp .

( 1 4) Benko - Dr. Ogaba (Hungarian Championship, 1 95 0)

l .e4 e6 2 .d4 d5 3 .Nd2 Nc6
This line of the French seems anti-positional since Black is blocking his c
pawn, thus preventing his thematic . . . c7-c5 counter-strike in the center.
However, the advance of the c-pawn is not permanently ruled out, and
Black's Knight on c6 adds power to . . . f7 -f6 (after White pushes his pawn to
e5), with a frontal attack that will challenge White's central domination.

4.NgfJ Nf6 5 .e5 Nd7 6.Bd3 ! ?

Since the position after 6.Bd3 hadn't been that deeply explored a t the time,
I decided to turn away from the recommended theoretical continuations
47

PAL B E N KO : M Y LIFE, GAMES A N D COMPOSITI O N S

(i.e., 6.b3 , 6.Nb3 , 6.Be2 , and 6.Bb5) in an effort to create new and origi
nal problems for my opponent.
48

6 ...Nb4

Also reasonable is 6 . . .f6.

7.Be2 cS S.c3 Nc6 9.0-0 Qb6

Trying to show that White's Knight on d2 is misplaced. However, this

doesn't turn out very well for Black, and later experience proved that his
best plan was 9 . . . cxd4 1 0.cxd4 f6 with approximate equality.
1 0.Nb3 !

After I completed this game, I learned that the position after 9 . . . Qb6 had
already been reached in an encounter between Dr. Balogh and Dr. Ogaba.
There, 1 0.Qb3 was played. This Knight move is a clear improvement, and
was later used by Geller to win several nice games.
1 0 . . . cxd4

Black tried t o . . . a5 in the game Geller-Vaganian, USSR ch. 1 97 5 . White

got a clear advantage after 1 1 . dxc 5 Qc7 1 2 .Bf4 Nxc5 1 3 .Nbd4.
1 l .cxd4 as 1 2 .a4

Forced. Now Black controls the b4-square, which seemingly assures him
of an equal game. However, this view is shallow. White's many plusses far
outweigh the single hole on b4: more central space, chances on the kingside,
control over b5, and he can also play to occupy c5, the possession of which
later proves decisive.
1 2 ... Be7

Reserving the b4-square for his Knight.

1 3 .BgS !

Black is faced with a difficult choice. Either he allows the trade of dark
squared Bishops, which would leave the c5-square fatally weak (in this sce
nario, Black's c8-Bishop and d7-Knight would both occupy bad positions),
or he pushes his pawn to f6, when his King will have trouble finding a safe
place to live.

1 become top dog in H u ngary, 1 950.

1 3 ... f6

Black doesn't want to be pushed around, so he decides on the more dy

namic answer.
14.exf6 gxf6

1 4 . . ,Nxf6 would leave Black having to care for an additional weakness

on e5 .
1 5 .Bh6 NfS 1 6.Nh4 Ng6 1 7.Bh5 Rg8 1 8.Qd3
As they say, strike while the iron is hot! My last few moves (Nh4, Bh5, and

Qd3) were all designed to provoke .. .f6-f5 . In that case, I would have quickly
brought my pieces back towards the center and initiated play against the
permanent weaknesses on e5 and e6.
1 8 ... Kd8

Rejecting the ugly 1 8 .. .f5 , and falling out of the frying pan and into the
fire. This King move ends the annoying pin against the g6-Knight, but
throws away his "life insurance policy," since queenside castling is now
impossible. Naturally, I would have been happy to see 1 8 . . . Nb4 1 9.Qg3
Bd6 2 0.Bf4 Bxf4 2 1 . Qxf4, since the threats of Qxf6 and Qh6, plus the fact
that c5 has completely fallen into White's hands, makes Black's prospects
very bleak.
19.Nf3

The pin has done its job, and now it's time to bring my pieces back to the
center and prepare to put pressure against the target on e6. Note that
1 9.Nxg6? hxg6 20.Bxg6 ? ? would be suicide, since 2 0 ... Rxg6 2 1 . Qxg6 Qxb3
picks up material.
1 9 ... Bd7 20.Rfe l

Once again, my attention turns to that juicy c5 -square.

PAL BEN KO: MY LI FE, GAMES AN D COMPOS I TI O N S

20 . . . Nb4 2 1 .Qe3 Na6

He's kept me out of cS, but the Knight isn't well placed on a6.
50

22.Bg4! Nh4

Black hopes that an exchange of Knights will bring a bit of relief from the
constant pressure.
2 3 .Bh3 Nxf3 + 24.Qxf3 Re8 2 5 .Bd2

The first hammer-stroke! Now Black is forced to dance back to b4 with

his Knight, once again giving me control over cS .
2 5 . . . Rxc 1 + 26.Rxc 1 Nb4 27.Ne5 !

Suddenly Black is lost! The threat is 2 S .Nxd 7 Kxd7 2 9 . Bxb4 Qxb4

3 0.QxdS+.
2 7 ... f5

After the game concluded, my opponent put a question mark after 2 7 . . . fS

and, on the side of the scoresheet, wrote 2 7 ... BcS, giving it an exclamation
point, I felt a bit sorry for him, but I was duty bound to tell him that I was
actually hoping he'd play 2 7 . . . BcS. White would then beautifully crash
through by 2 S.Bxb4 axb4 29.Nxe6+ Bxe6 30.QxdS+ BxdS 3 1 .RcS mate.
28.Qf4? !

Threatening 2 9.QbS+, but much crisper was 2 S .BxfS ! exfS (2 S . . . BxcS

29.Bxh7) 29,Nxd7, when 29 .. ,Kxd7 30,Bxb4 Bxb4 3 1 .QxdS+ is easy for
White.
2 8 ... Be8?

The only defense was 2 S . . . Nc6. Now Black gets overrun.

29.Qh6 Re8
As we like to say in Hungary, this move is the beginning of death agony!

30.Bxf5 !

Making use of the Queen-to-Queen pin along the sixth rank.

30 . . . Na2 3 1 .Nxe6+ ! Bxe6 32.Bxa5 ! , 1 -0.

The second, annihilating hammer-stroke ! Black resigned, having no de

sire to face 32 . . . QxaS 3 3 .Qxe6.

P ROV I N G MYS E LF AT TH E MARO CZY M E M O RIAL

( 1 5) Benko

I Molnar (Hungarian Championship, 1 95 0)

l .e4 e5 2 .NfJ Nc6 3 .d4

The text was a major surprise for my opponent. At the time (long before
Kasparov turned it into a serious weapon), theory considered the Scotch
harmless, and for this reason it seldom occurred in master praxis. However,
in an age where young players only study fashionable lines, unearthing an
old system like this can often pay huge dividends. Why? Because he might
not be acquainted with its subtleties, and his experience with it will un
doubtedly be small. Bent Larsen used this philosophy often in the prime of
his career, successfully resuscitating many old but positionally sound varia
tions. The following quote by Perez about Larsen is worth repeating: "Pre
paring for this tournament, the other participants have studied Boleslavsky's
latest innovations, but Larsen has studied Greco and Philidor! "
3 . . . exd4 4.Nxd4

The Scotch Gambit comes about by 4.Bc4, when the simple L .Bc5 4.c3
d3 ! , or the more complicated 4 ... Nf6 are both sufficient for equality.
4 . . . Nf6
In 1 950, the textbooks gave this move an exclamation mark, at the expense of

4 ... Bc5 (Zukertort's line). I never believed this to be valid (and modern prac
tice thinks highly of both 4 ...Nf6 and 4 ... Bc5). The game Stoltz-Bronstein,
Saltsjobaden 1 948 showcased 4 ... Bc5 : 5 .Be3 Qf6 6.c3 (Blumenfeld's 6.Nb5 is
very questionable: 6 ... Bxe3 7.fxe3 Qe5 [Even 7 ... Qh4+ 8.g3 Qd8 9.Qg4 Kf8
is possible.] 8.Qd5 Kd8 ! 9.Qxf7, and now both 9 ... Nge7 and 9 ... Nh6 are
good for Black.) 6 ... Nge7 7.Nc2 d6! ? 8.Bxc5 ? ! dxc5 9.Ne3 0-0 10.Be2 Rd8
1 1 .Qc2 Be6 1 2 .0-0 Ng6, and Black got the advantage.
5 .Nxc6
I

felt this was the only way to play for an advantage, and, more recently,
Kasparov verified this with several fine victories.

5 ... bxc6 6.e5 Ne4? !

This move takes the game far from well-trodden paths. At that time, the
main line was 6 ... Qe7 7.Qe2 Nd5 , when the game Mieses-Forgacs, Ostende
1 907 was often quoted : 8 . f4 f6 9 . c4 Ba6 1 0 . b 3 fxe 5 1 1 . fxe5 0-0-0
1 2 .Bb2 Qg5 ! , with an edge for Black. Better, of course, is the modern 8.c4.
7.Qf3 Ng5 8.Qg3 Be7

Black could have considered 8 . . . Ne4, expelling the Queen from its good
position on g3 . White, if he doesn't want to reconcile himself to the draw
by repetition (9.Qf3 Ng5 , etc.), would have to play 9.Qf4 (9.Qe3 ! ?), when
Black could try to gain a tempo by hitting White's Queen with 9 . . . Nc5
followed by ... Ne6. However, in this case, 1 O.Bc4 Ne6 1 1 .Bxe6 seems prom
ising for White.
9.Bd3 0-0 1 0.0-0 d5 ! ?

51

PAL B E N KO : M Y LIFE, GAMES AN D COMPOSITI O N S

Black shows that he isn't afraid of White's pawn storm. More cautious is
1 0 . . .f5 .
1 l .f4 Ne4
52

White answers 1 1 . . .Ne6 with the strong 1 2 .f5 Bh4 1 3 .Qe3 Bg5 1 4.Qf2 .
1 2 .Bxe4 dxe4 1 3 .fS

White is staking everything on a kingside attack. While it might seem like

gambling due to his undeveloped army, the reality is that White can't hesi
tate, even for a moment! For example: 1 3 .Nc3 f5 ! (and not 1 3 ... Bf5 1 4.Be3 ,
when White's positional advantage is obvious. He could even try to force
the win of a pawn by 1 4.Nc3) 14.exf6 Bxf6 1 5 .Nxe4 Qd4+ 1 6.Nf2 Ba6
1 7 .Rd l Qb6, and Black's Bishops give him the advantage. After 1 3 .f5 , the
game is balanced on a razor's edge.
1 3 ... Bh4!

Black attacks and defends skillfully at the same time. Any show of passiv
ity would be punished: 1 3 . . . Re8 1 4.Bh6 g6 1 5 .e6, with a rapid decision.
14.Qg4

Though 1 3 ... Bh4 appears ultimately to lose a tempo, the fact is that White's
Queen doesn't have any comfortable square to run to. The seemingly strong
1 4.Qf4 ensures Black an advantage after 1 4 ... Ba6. One sample: 1 5 .f6 g5
1 6.Qg4 Qd4+! 1 7.Khl Bxfl 1 8.Bxg5 Be2 ! , and Black wins. Thus, for want
of a better alternative, I moved my Queen to g4, which, unfortunately, self
pins my f-pawn, making the thematic f5-f6 advance very difficult to achieve.
14 ... Qd4+ l S .Khl Be7

It seems that White's attack has run into a roadblock since my e- and f
pawns aren't able to safely advance. However, this turns out to be a tem
porary state of affairs, and my sleeping queenside reserves wake up and
tum the battle in my favor.
1 6.Bd2 !

The most natural move. 1 6.Qg3 ? fails to 1 6 . . . Bxf5 1 7 .Bh6 Bg6.

1 6 ... Rd8 1 7.Qg3

A bit more preparation for Bc3 , which was premature right away due to
1 7 . . . Qe3 .Thus far, Black has defended himself extremely well. But now,

P ROV I N G MYSELF AT TH E MARO CZY M EM O RIAL

due to time pressure, he fails to find the best defense and, as a result, his
position quickly goes downhill.
1 7 . . . Qc4?

A decisive blunder. Correct was 1 7 . . . Bxf5 , after which 1 8.Rxf5 ? Qxb2 wins
due to the weakness of White's back-rank. After 1 8.Bc3 ! Qb6! 1 9.e6 f6
20.Nd2 Bg6 (2 0 . . . Bxe6? 2 1 .Nxe4 Kh8 2 2 .Nxf6 ! gxf6 2 3 .Rxf6) 2 1 .Rae 1 ,
the battle rages on.
1 8.Na3 ! Bxa3 1 9.Bh6

Suddenly the white attack blows Black away.

1 9 . . . Bf8 20.Bxg7 Bxf5

Black gets mated after 2 0 . . . Bxg7 2 i .f6 Kf8 2 2 .Qxg7+ Ke8 2 3 .Qg8+ Kd7
24.Rad 1 + Ke6 2 5 .Qg4+ Kxe5 2 6 .Qf4+ Ke6 2 7 .Qf5 mate.
2 1 .BxfB+ Bg6

Black also falls victim to a winning attack after 2 1 .. .Kxf8 2 2 .Rxf5 Rd7
2 3 .Rafl Rad8 24.Qh4 (24.Qf4? Qxf1 +!).
22 .Bh6 Qxc2

Black would last a bit longer with 2 2 . . . Re8, though the end would still
come swiftly and violently: 2 3 .h4 Qe2 24.e6 Rxe6 2 5 .Qc3 f6 26.Rxf6 Re7
2 7 .Rafl .
2 3 .Qg5 5 24.Qe7 Bf7 2 5 . Qf6, 1 -0.

( 1 6) Korody - Benko (Budapest, 1 95 1 )

l .d4 d5 2 .c4 c6 3 .NB Nf6 4.Nc3 e6 5.e3 Nbd7

I've played both sides of this very rich, complicated, system.

6.Bd3 dxc4 7.Bxc4 b5 8.Bd3

The best move, though both 8.Bb3 and 8 .Be2 have their supporters.
8 ... Bb7

At the time, this was a new idea. Black intends to do without ... a 7 -a6, instead
hoping for a quick ... b5-b4 followed by ... c6-c5 (opening the light-squared
Bishop's diagonal).

53

PAL B E N KO : MY L I F E . G A M E S AND C O M POSITI O N S

9.0-0

54

White's sharpest move is 9.e4, when the position after 9 ... b4 1 0.Na4 c5 I l .e5
Nd5 has been very deeply analyzed.
9 . . . b4 1 0.Ne4

The main line, but 1O.Na4!? also deserves consideration.

1 0 . . . c5 ! ?

Initiating a very sharp, risky battle. More usual is 1 0 . . . Be7 or 1 0 . . .Nxe4

1 1 . Bxe4 Be7.
1 1 .Nxf6+

Less threatening is I l .Nxc5 Bxc5 1 2 .dxc5 Nxc5 1 3 .Bb5+ Ke7, when Black
has a solid position. For example, 1 4.Nd4 Qd5 1 5 .8 Rhd8 would be more
than comfortable for the second player.
1 1 . . . gxf6

And not 1 1 ...Nxf6? 1 2 .Bb5+. The move I chose, 1 1 . ..gxf6, ruins my kingside
pawn structure but opens up the g-file and, as a result, gives me some at
tacking chances. White's correct plan would now be a quick e4 (closing off
the diagonal of Black's light-squared Bishop) followed, possibly, by d4-d5 .
The position after 1 2 .e4 cxd4 (stopping White's dreams o f a d4-d5 advance)
1 3 .Nxd4 is a very complicated one, but practice has favored White and
Black rarely enters this line anymore.
1 2 .Qe2 Qb6 1 3 .a3

Black's King is insecure, so White is wisely trying to open up attacking

lines so his pieces can penetrate into my position.
1 3 . . . Bd6

I think 1 3 . . .b3 , keeping White's pieces contained, was a better idea. How
ever, I was very young at that time and, as is common for youth, couldn't
resist the siren call of a sharp tactical battle.
1 4.axb4 cxd4

Of course, 1 4 ... cxb4? would leave White with a clear advantage due to the
open a-file and the possibility of e3 -e4.
1 5 .exd4 Rg8

P ROV I N G M Y S E LF AT T H E MARO CZY M EM O RIAL

Now 1 6.Bxh7? is well met by 16 . . . Bxh2 + 1 7 .Kxh2 RhS, winning back the
piece with a good game due to Black's play on the open h-file.
1 6.bS ? ?

Much better was 1 6.Be4, closing off the dangerous diagonal. White wanted
to save the b-pawn and stop any possibility of . . . Qc6, but he overlooked
Black's main threat. After 1 6.Be4, White would have obtained the advan
tage, but now the tide suddenly turns.
1 6 . . . Qxd4!

A bolt from the blue. The Queen leaps into the attack, and it can't be
taken since 1 7 .Nxd4 Rxg2 + I S .Kh I Rxh2 + 1 9.Kgl Rh l is mate.
1 7.h3

Trying to stop the ... Qg4 threat or a mix of ... Rxg2+ followed by ... Qg4+.
But in any case, White no longer had a good defense against Black's attack.
1 7 NeS !
..

A pretty move that creates a whole new series of mates.

1 8.Nxd4 Rxg2 + 1 9.Khl Rh2 + ! , 0- 1 .

The finish would have been 2 0.Kxh2 Ng4+ 2 1 .Kgl Bh2 mate. This flashy
game was published more times than any game I've played before or since.
Oddly enough, I was partying all Saturday night, got virtually no sleep,
and then rushed off to play this game (for a team match) Sunday morning!
( I 7) Szabo - Benko ( H ungarian Championship, 1 95 1 )
l .e4 cS 2 .NB Nc6 3 .d4 cxd4 4.Nxd4 Nf6 S .Nc3 d6 6.Bc4 e6

Many years later, I introduced 6 ... Qb6 into top-flight competition. This
popular system (6 ... Qb6) still bears my name today.
7.0-0

At that time, the inventor of the dangerous Velimirovic Attack (7 .Be3 fol
lowed by S.Qe2 and 9.0-0-0) was still playing with marbles and couldn't have
suspected the impact he'd eventually make on the 6.Bc4 system.
7 . . . Be7 8.Be3 0-0 9.Bb3 Bd7 1 0.f4

White's plan, a favorite of Fischer's, is f4-f5 . This attack against e6 will more
or less force ... e6-e5, when White's b3-Bishop suddenly springs to life and
the d5 -square falls into White's hands.
1O . . . Nxd4 I 1 .Bxd4 Bc6 1 2 .Qd3

One year later, everyone realized that 1 2 .Qe2 was the accurate way to handle
this position. In fact, Fischer was very successful with this move. One sample:
1 2 ... b5 1 3 .Nxb5 Bxe4 1 4.Nxa7 e5 1 5 .fxe5 dxe5 1 6.Be3 QbS 1 7 .Nb5 Bc6
I S.Nc3 Qb4 1 9.Rad l , with a clear advantage for White in Fischer-Saidy,
New Jersey 1 95 7 .
1 2 ...b S

55

PA L B E N KO : MY L I F E , GAM ES A N D COM POSITI O N S

56

At the time, this plan (an invention of mine) against the Sozin was new.
Theory marches on and, as mentioned in the last note, 1 2 .Qe2 became
the standard move. About four years after my Szabo game, Geller started
to play my . . . b7-bS idea against 1 2 .Qe2 . Naturally, the Russians called it
the Geller Variation!
1 3 .Nxb5

Equality results from 1 3 .eS dxe5 1 4.fxeS Nd7 I S .Ne4 Bxe4 1 6.Qxe4 NcS .
1 3 . . . Bxe4

Now we can see the difference between 1 2 . Qd3 and 1 2 . Qe2 . If the Queen
stood on e2 , 1 3 , . . Bxe4 wouldn't be with tempo! Another important
possibility is 1 3 , . .BxbS (the best move if White's Queen were on e2)
1 4.QxbS Nxe4 I S .fS . This position has been debated for many years, but
it seems that Black is perfectly all right after 1 5 . . . e S 1 6 . B e 3 BgS !
1 7 .Qd3 Bxe3+ 1 8.Qxe3 Nf6.
1 4.Qe2 Qd7

It might have been better to play 14 ... a6, since I S .Nc3 doesn't allow White
to make active use of his c-pawn.
1 5 .c4! d5

This seemed good at the time, but it only took a couple moves for me to
realize that I'd done something wrong. Better was I S . . . Bc6.
1 6.Rad l Qb7 1 7.cxd5

White could also have tried 1 7 .Nc3 ! ? dxc4 1 8.Bxc4, but I think Black's
okay after 1 8 . . . BfS ( 1 8 ... Bc6 1 9.fS gives White chances).
17 ... Bxd5 1 8.Bxd5 Nxd5 1 9.f5

White now has a dangerous initiative.

19 . . . a6 20.Nc3 Nxc3 2 1 .bxc3 exf5 2 2 .Rxf5 Rad8

I could have neutralized the powerful d4-Bishop by 2 2 . . . f6 . After

2 3 .Qe6+ Kh8 24.RdS Rad8 2 S .c4 White would stand better, so I decided
to take my chances in the tactics that follow.
2 3 .Re l Bd6 24.Qg4 g6 2 5 .Qg5

PROV I N G MYSELF AT TH E MARO CZY M E M O RIAL

57

White has a serious attack and Black has to show great resourcefulness if
he wants to survive. Instead of 2 5 .Qg5 , White could have considered the
interesting 2 5 .Rh5 ! ? However, Black is able to defend by 2 5 . . . RfeS 2 6.Rfl
(2 6.RxeS+? RxeS 2 7 .Rxh7 fails to 2 7 . . . Qb 1 +) 2 6 . . . BfS ! .
2 5 ... Bb8 ! ?

This move allows Black to get rid o f White's

powerful attacking Bishop by . . . Rxd4 or
... Ba7. Now 2 6.Qf6 Rxd4 2 7 .cxd4 gxf5 only
leaves White with perpetual check.
26.Bf6 Rd6

Not fearing 2 7 . Qh6 (2 7.Be7 ReS) 2 7 . . . Qb6+

2 S .Bd4 Rxd4 29.cxd4 Qxd4+ 3 0.Rf2 Ba7.
27.Khl

zlo Szabo: Though Laszlo was

a good player, not many people liked

him. He was a tough communist at
that time, a real party guy, and he took
full advantage of it. He once told me,
when I was twenty-four, that if! didn't
behave and do what he told me to do,
he would make sure I was drafted into
the army a second time! Can you be

A very sane move . If White had played

lieve that guy? I had already done my
27 .Refl , Black would have answered with
army nightmare stint when I was six
27 ... ReS !, when back-rank counterthreats lead
teen, so I certainly had no desire to be
to salvation: 2 S.Qh6 Qb6+ 29.Bd4 (2 9.Kh I
introduced to that kind of thing again !
Rxf6 3 0.Rxf6 Qxf6 3 1 .Rxf6 Re l + and mates)
29 ... Rxd4 3 0.cxd4 Qxd4+ 3 1 .Kh I (or 3 1 .R5f2
Ba7, threatening 3 2 ... Qxf2 + ! 3 3 .Rxf2 Re I mate) 3 1 .. .gxf5 .
2 7 . . . Re6 2 8.Refl

It looks like Black doesn't have a defense against Qh6. For example, 2 S ... Re2
29.R5f3 wins.
28 ... Qc7 !

Suddenly Black has a powerful counterattack! Now the advantage moves

to my side.
29.g3

This avoids 29.R5f4 Qxf4 3 0.Rxf4 Re l + and mates. Best, though, was
29.Qh6 Qxh2+ 3 0.Qxh2 Bxh2 , when White can play the endgame a pawn
down.
29 ... Qc6+ 30.R5f3 ?

PAL B E N KO : MY L I F E , G A M E S A N D CO M POSITI O N S

An exasperated Szabo falls apart. The only move was 30.R l f3 , though
3 0 . . . Be5 3 1 .Bxe5 Rxe5 stops the attack and leaves White on the edge of a
precipice.
58

3 0 ... Re l !

This double threat of 3 1 ...QxB + or . . . Rxf1 + is a killer.

3 1 .Rxe l

We get the same finish as in the game after 3 1 .Kg 1 QxB .

3 1 ...Qxf3+ 3 2 .Kgl Ba7+ H .Bd4 Qxc3 ! , 0- 1 .

Early in 1 952 I was invited to play in the exceptionally strong Maroczy Memo
rial (games 1 8-2 0), held in Budapest. Here I faced such chess giants as Botvinnik,
Geller, Keres (who eventually won the event), Petrosian, and Smyslov.
I was able to observe many interesting players at this tournament. Stahlberg, for
example, liked to drink. (Actually, most players, past and present, have had this
habit. Personally, I've never had any interest in alcohol.) The organizers wanted
the big Soviet names to win, and they were worried that Stahlberg might do well
and spoil their plans. To ensure this didn't happen, they allowed him to drink any
thing he wanted for free, thinking it would affect his play. He happily took them
up on this offer and almost won the tournament!
Seeing that the liquor wasn't having any effect on him at all, the organizers
desperately offered me extra money if I could beat Stahlberg in the last round.
Since I intended to try to win anyway, I accepted money wasn't something you refused in those days !
I won this game and picked up a nice, extra "prize. "
Herman Pilnik and O'Kelly were both entertain
ing. Pilnik was a real ladies' man, an incredible char
acter. He was constantly chasing after women, al
ways trying to attract their attention.
O'Kelly had a very different pick-up strategy: he
brought nylon stockings and panties with him, and
these were worth a lot in Hungary because they were
only available in the West-women would do any
thing to get their hands on these things ! He called
it "Hungarian currency."
I remember the East German player Platz, who lost
H e rman Pil n i k, the "lad i es' man."
the
first nine games. I beat him in the tenth, with a
(Photo courtesy USCF.)
King's Indian using . . . Nas. I think this was a new idea
at the time. He also lost the eleventh game. In the twelfth game he played the
World Champion, Botvinnik, and Botvinnik got a horrible position and was very
lucky to draw. This guy loses eleven in a row and then beats up the champ! The
lesson I learned against Polzer (in game 8) comes to mind!

PROVI N G MYSELF AT T H E MARO CZY M E M O RIAL

Po l ugayevsky (left to
right) , O ' Kel ly,
Trifu n ovich, and
Geller in 1 9 64. Back
in 1 952, O ' Ke l ly had
his own form of
" H u ngarian
cu rrency."

(Photo courtesy
USeF.)

( 1 8) Benko - Petrosian (Maroczy Memorial, 1 95 2)

l .NO Nf6 2 .g3 d6 3 .d4 g6 4.b3 Bg7 5.Bg2 0-0 6.Bb2 a5 !

A good move that gains queenside space and prepares . . . a5-a4.

7.0-0 c6 8.a4?!

Ending any discussion about Black's potential . . . a5 -a4 advance. Unfortu

nately I have to pay a price for this: the b4-square has been weakened.
8 ... Na6

Letting me know that c2 -c4 will hand over b4 forever.

9.Nbd2 d5 ! ?

Black has evidently decided that I shouldn't b e allowed to play e2 -e4. At

this point it became difficult to find a good plan and, out of necessity, I
concluded that I should accept Black's challenge and force the e2 -e4 ad
vance no matter what!
1 0.Ne5 Bf5

Continuing to clamp down on e4.

l l .O ! ?

Creating some weaknesses in my own camp, but finally forcing through

the move that I've worked so hard to achieve.
1 1 ...c5 !

Increasing the central tension and forcing \Vhite to accept a pawn weak
ness if he still insists on the e2 -e4 push.
1 2 .e4 Be6 1 3 .exd5

\Vhite accepts an isolated d-pawn (No, you're not hallucinating. The iso
won't appear for several more moves.) in exchange for domination over the
c4-square. I was tempted to try the sharp 1 3 .c4 dxe4 1 4.d5, however, my
opponent was already famous for his defensive skills so I pushed my impul
sive tendencies out the window and kept the game under control.

59

PAL B E N KO : MY L I F E , GAM ES A N D C O M POSITI O N S

1 3 . . .Nxd5 1 4.Ndc4 Qc7 1 5 .Qd2 Ndb4 1 6.f4 Rad8 1 7.c3 Nc6 1 8.Qe3
cxd4 1 9.cxd4
60

The isolated pawn finally arrives. As compensation, White's pieces have

become very active.
1 9 ... Nab4 20.Rac 1 Bd5

Neutralizing my powerful g2 -Bishop. Here I spent a lot of time looking

deeply into the position. Black's control over d5 and pressure against d4
could prove annoying unless I did something about it. But what can be
done? Eventually I came up with an interesting plan based on some key
minor piece exchanges.
2 1 .Rfd l Rfe8 2 2 .Ba3 !

The start of my plan. Black's Knights are the key to his control over d5, so
trading off my poor dark-squared Bishop for one of these steeds makes
excellent sense.
22 . . . Bxg2 2 3 .Kxg2 e6 24.Qe4 Bf8 2 5 .Bxb4!

Now Black only has one Knight left. Note that his remaining Bishop has
little impact in the fight over d4 and d 5 .
2 5 . . . Nxb4

Worse was 2 5 ... Bxb4 2 6.Nxc6 Qxc6 2 7 . Qxc6 bxc6 2 8 .Ne5 c5 29.dxc5 Rxd l
3 0.Rxd l Bxc5 3 1 .Rd7, when White stands better.
26.Ne3 Qe7 2 7.Rc5

I could have gotten rid of my isolated pawn and gained approximate equality
by 2 7 .d5, but I decided to try for more. With 3 7 ,Rc5 , I'm bringing an
other piece to bear on the embattled d5-square and also intend to double
Rooks on the c-file. Both sides were already in time trouble due to the
difficult positional decisions that had to be made,
2 7 . . .b6

This move forces my Rook off of c5, but it also weakens the c6-square.
The flashy 2 7 ... Rxd4 Ieads to equality, and therefore might be best: 2 8.Rxd4
Qxc5 2 9 . Qxb7 (Also interesting is 2 9.N3 g4! ? ) 2 9 . . . Be7 3 0.Rd7 Qxe3
3 1 .Rxe7, with a draw.

28.Rcc 1 Bg7 29.Rc4 Qd6 3 0.Rdc 1 f6

White's control of the c-file is worth

more than Black's pressure against d4.
The text move chases away my power
fully placed Knight, but it also creates
new weaknesses in Black's camp.
3 1 .Nf3 Re7 3 2 .Rc8 Red7 H .Rxd8 + ? !

The lack of time starts affecting both

players' moves. Far stronger was 3 3 .Nc4
Qd5 34.Qxd5 Nxd5 3 5 .Rc6, winning a
pawn.
3 3 . . . Rxd8 34.Qb7

Now 34.Nc4 no longer gives White any

thing: 34 . . . Qd5 3 5 .Re l ReS ! (and not
3 5 . . . Kf7 ? 3 6.Nxb6 Qxb3 3 7 .d5 !).
34 . . . Rb8 3 5 .Rc8+ Rxc8 36.Qxc8+ Bf8
3 7.Qb7?

Low on time, I misjudged the strength

of 3 7 .Nc4. In fact, this move virtually
forces a pawn-up endgame after 3 7 ... Qd5
3 S.Nxb6 Qxb3 3 9.Nd7 Qc2 + 40.Qxc2
Nxc2 4 1 .Nxf6+.
37 . . . Nd5 3 8.Nc4 Qb4?

The clock also affects Tigran. Correct

was 3 S . . . QdS, when he must have been
afraid of 3 9.Qc6 Kf7 40.Nce5+, missing
that 40 . . . Ke7 ! was a satisfactory reply
(40 . . . fxe5 ? ? 4 1 .Ng5 + wins for White.).
3 9.Qd7

Now Black's position is extremely un

pleasant.
3 9 . . . Q e 7 4 0 . Q c 6 h5 4 1 . Nxb 6 Qd6
42 .Qxd6 Bxd6 43 .Nc4

I avoided 43 .Nxd5 exd5 since I wanted

my King eventually to penetrate on the
c4 or e4 squares. Naturally, this would
be impossible after the Knight exchange.

Tigran Petrosian (to p) , with h i s wife Rona ( m i d d l e ) ,

with h i s son Vartan (botto m ) , 1 9 66.

(Photos by Z. M i l utinevich. Courtesy USCF)

an Petrosian: We were friendly

when I lived in Hungary. However,

Petrosian began to treat me poorly af
ter I escaped, trying to win points with
his government.
He had a very strange, solid, style that
was typical of many Armenians (though
he did it better than anyone else) . In
fact, in his prime, he was almost unbeat
able.
Many people thought that his wife,
Rona, was very pushy. In Curac;ao, she
came over and asked me why I wasn't
married.
"Well, there are so many nice women
that it's hard to make up my mind . "
S h e s a i d , "This shouldn't p o s e a
problem. Just marry the richest one . "

PA L B E N KO : MY L I F E , GAM ES A N D C O M POSITI O N S

4 3 ... Be7 44.Kf2

62

White has won a pawn, but the ending is only winnable if my King can
somehow creep into the queenside. This proves to be far from easy to
accomplish.
44 ... Kfi 4S.Ke2 Nb4 46.Kd2 Ke7 47.Ne3 Bd6 4S.Ng2 !

A fine move that is designed to keep Black's King tied up guarding g6

(after Ngh4) and thus far away from the queenside. Playing the f-Knight
to h4 would allow . . . g6-g5 .
4S . . . NdS 4 9 . Ngh4 Kf7 S O . Kd 3 Nb 6 S l . Ng2 Ke 7 S 2 . N e 3 B e 7
S 3 .Ke4 BdS S4.dS !

At last White gets to rid himself of the isolated pawn. More importantly,
the advance of the d-pawn makes the d4-square available to my King, al
lowing it to penetrate into the queenside via d4 and c5 .
S4 fS + S S .Kd4 Nd7 S6.Ne4
.

The battle is finally decided, but Petrosian plays it out for a few more
moves.
S 6 . . . exdS S 7 . KxdS Nf6+ S S . Ke6 Ne4 S 9 . Kb S Ke6 60.NxaS Nc3 +
6 1 .Ke4 Nxa4 62.bxa4, 1 -0.

Black gave up since he has no possibility for counterplay (his King is still
stuck guarding g6 !). White will move his King to b5 and push his a-pawn
to victory.

( 1 9) Botvinnik - Benko (Maroczy Memorial, 1 95 2)

l .Nf3 Nf6 2 .e4 e6 3 .d4 dS 4.exdS exdS

The Exchange Variation is a popular anti-Slav weapon among positional

players who enjoy plodding away in a symmetrical position with an extra
tempo. White is often able to create a certain amount of positional pres
sure without having much chance of losing. Of course, against a well-pre
pared opponent White's chance of winning is also fairly low.
S.Nc3 Ne6 6.Bf4 e6 7.e3 Bd6 S.Bd3 ! ?

P ROVI N G MYSELF AT T H E MARO CZY M E M O RIAL

An interesting way to try to infuse

some life into the game. Also seen are
S.Bxd6 (this gives Black an easy game)
and S.Bg3 , retaining some tension
since the capture on g3 opens up the
file for "White's h I-Rook.
8 ... Bxf4

Breaking up "White's pawn formation,

but giving him permanent control
over the e5 -square. Botvinnik had al
ready used this idea against Kottnauer
in Moscow 1 947 : S . . . O-O (avoiding
Botv i n n i k i n 1 9 6 1 .
S . . . Nxd4? 9.Nxd4 e5 l O.Bb5+) 9.0-0
(Photo courtesy USeE)
b6 l O.Rc 1 Bb7 I l .a3 RcS (Perhaps he
should have tried I l ...Nxd4! ? 1 2 .Nxd4
e5 1 3 .Bg5 exd4 1 4.exd4 h6, with simplification) 1 2 .Qe2 , with a slight edge for "White.
9.exf4 0-0

Since the one problem with "White's S.Bd3 is the isolation of the d-pawn,
Black should jump on this fact with 9 ... Qb6 ! . The game Andersson-Donner,
Amsterdam 1 979, continued: l O.a3 Bd7 I l .Na4 Qc7 1 2 .g3 0-0 1 3 .0-0 RfcS
1 4.Rc 1 QdS 1 5 .Nc5 Rc7 1 6.Rc3 RacS 1 7 .Qc2 g6 1 S.Rc 1 BeS I 9.Qd2 Qe7
2 0.b4 NbS 2 1 .Ne5 b6, with complete equality.
1 0 . 0 - 0 B d 7 I 1 . Rc l Na5 1 2 . N e 5 Re8 1 3 . Q e 2 Ne6 1 4 . Nf3 g6
1 5.Rfd l Ne8

At this point I realized I was going to be squeezed to death if I didn't

come up with something interesting. As it turned out, Botvinnik didn't
appreciate what I was doing until his advantage began to fade.
1 6 .h4 Nd6

My plan is clear: I will bring this Knight to f5 where it will attack d4.
Then I'll take the e5- and g5 -squares away from "White's B -Knight with
. . . f7 -f6 and, if allowed, will cement my control over f5 with . . . h 7 -h5 (stop-

63

PAL B E N KO : M Y L I F E , GAM ES A N D CO M POSITI O N S

ping g2 -g4). If all this can be accomplished, Black will have an excellent
game!
1 7.g3 f6 1 8 .Kg2
64

Too much preparation! White had to play 1 8.h5 if he wanted to stop my

dream setup.
18 ... NfS 1 9.Bb l hS ! 20.Qd2

Completely losing the thread. His last chance to keep the initiative was
20.Bxf5 gxf5 2 1 .Ne l Qe8 2 2 .Nb 5 .
20 ...Nce7 2 1 .Re l Qb6 2 2 .b3 Qb4, liz_liz .

I accepted Botvinnik's offer of a draw. Too bad! White has run out of tar
gets to attack and now Black is in the driver's seat. I should have squeezed
him by doubling Rooks on the c-file and keeping him tied down to his
weak d4-pawn. All this could have been done without risk, but I was young
and my opponent was the World Champion!

(20) Benko - Stahlberg (Maroczy Memorial, 1 95 2)

l .NO dS 2 .g3 Nf6 3 .Bg2 g6 4.0-0 Bg7 S.d4 0-0 6.c4 c6 7.Nbd2
In the '80s and '90s, White has tried to milk the rather dull position arising

after 7.cxd5 cxd5 8.Nc3 Nc6 9.Ne5 .With 7.Nbd2 , I'm not trying to claim
an opening advantage. Instead, I'm simply trying to create an interesting
strategic battle that will give me room to outplay my opponent.
7 ...BfS 8.b3 Ne4 9.Bb2 as 10.h3

I've had Black in this position on a couple of occasions (see game 2 7 ,

Weinstein-Benko, U.S. ch. 1 959). Usual i s 1 0.Nh4 Nxd2 1 1 .Qxd2 , but I
don't think White has much after 1 1 . . . Be6. Of course, 1 0.h3 is just some
thing I came up with over the board, and certainly doesn't improve on
1 O.Nh4.
10 ... Na6 I 1 .Nh4 Be6?

It's usually not a wise decision to give up a center pawn for an h-pawn.
This rule is especially true here, since White can use the open h-file for
an attack. Better was 1 1 . . .Nxd2 1 2 .Qxd2 Be6, when I would have played
1 3 .cxd5 cxd5 1 4.Rfc l . White then has a very slight edge due to the mis
placed black Knight on a6 (it would much prefer to be on c6).
1 2 .Nxe4 dxe4 1 3 .Bxe4 Bxh3 14.Bg2 Bxg2 1 5 .Kxg2

I was quite pleased with my position here. White has a central space
advantage and kingside chances based on Rh l .
1 5 ... a4 16.NO

The Knight wasn't doing anything on h4 and had to be brought back into
play. Never leave your pieces on squares that don't offer them a bright
future.

P ROV I N G MYSELF AT TH E MARO CZY M E M O RIAL

16 ... Qc7 1 7.RhI RfdS I S.Qc l !

A nice move that defends b2 , gets the Queen off the dangerous d-file, and
envisions an eventual invasion into h6.
I S . . .NbS

Black doesn't realize that he's on the edge of a precipice.

1 9.dS !

By trading off Black's only kingside defender (the g7-Bishop), White starts
a powerful attack.
1 9 . . . Bxb2

Black couldn't avoid the exchange with 1 9 . . .f6, since 2 0.Nd4 followed by
Ne6 would be too strong.
20.Qxb2 hS

Simply bad is 20 . . . a3 (or 20 . . . cxd 5 ? 2 1 .Rxh7!) 2 1 .Qc3 cxd5 2 2 .Rxh7! d4

2 3 .Nxd4 Kxh7 (also hopeless is 2 3 ,..Qe5 24.Rh4 g5 2 5 .Rg4) 24.Rh l + Kg8
2 5 .Ne6 Qc6+ 2 6.f3 f6 2 7 .Qc2 ! Qe8 2 8.Nxd8 .
2 1 .g4

Of course, 2 1 .e4 wasn't bad, but why not dismember your opponent when
the chance arises?
2 1 . ..cxdS 22 .gxhS Ra6 !

Black finds the only way to slow down White's attack. On 2 2 ... dxc4, White's
attack becomes overwhelming after 2 3 .hxg6 c3 24.Qc2 ! .
2 3 .hxg6

Also very strong was 2 3 .h6.

2 3 ... Rxg6+ 24.KfI d4! ?

This tricky move closes the diagonal, but the d-pawn is vulnerable here.
The alternative, 24 . . . Rg7 , can be met in two different ways. First, White
can claim a safe advantage by 2 5 .cxd5 Rxd5 2 6.bxa4 Nc6 2 7 .Rb l . More
interesting, though, is 2 5 .Qd4 dxc4 26.Qh4 Kf8 (Black's best defense is
26 .. .f6, when White has to settle for 2 7 .bxc4 Nc6 2 8 .Rb l , with a plus)

65

PA L B E N KO : M Y L I F E , GAM ES A N D CO M PO S I TI O N S

27.Ng5 ! RgS 2S.Qh7 Qf4 (or 2 S ... Rxg5 29.Qh6+ Rg7 3 0.Rgl) 29.Rgl Qxg5
3 0.Rxg5 .
66

2 S .Qc2 !

Though 2 5 .Nxd4 seems tempting, Black gets too much counterplay after
25 ... Qe5 2 6.Rd l Nc6 27.e3, and now 27 ... Qe4? is a mistake: 2 S,RhS+ ! KxhS
2 9.Nxc6+ Kh7 3 0.RxdS Rf6! 3 1 .Qd4 Qb l + 3 2 .Qd l Qxd 1 + B .Rxd l Rxc6
34.Ke2 axb3 3 5 .axb3 Rb6 3 6.Rd 3 , when White can torture Black in the
pawn-up endgame. However, 2 7 . . . Rdd6 ! (swinging the Rook over to the
kingside in many variations) leaves White in a real bind. For example:
2 S . Qd2 (or 2 S.Re l axb3 29.axb3 Qe4 3 0.f3 Rdf6) 2S . . . Nxd4 2 9.exd4 Qe4,
and White's position starts to fall apart. My 2 5 .Qc2 ! intends to surround
the d-pawn in a much more controlled, dynamic, way.
2 S ... Qf4 26.Rh4 Qf6 27.Rd l axb3 2 8.axb3 Nc6 29.Qe4

All of White's pieces are taking aim at d4.

29 ... eS 30.b4!

Initiating a final, game-winning assault against Black's center. Now

30 . . . Nxb4 3 1 .Nxe5 ReS 3 2 .Nd7! Qe7 3 3 .Nf6+ wins for White, as does
30 . . . ReS 3 1 .b5 NdS 3 2 .Rxd4.
3 0 ... d3

Desperation in a lost position. I would have answered 30 . . . Qe6 (hoping

for 3 1 .b5 f5) with 3 1 .Rh5 ! , stopping Black's . . . 7-f5 ideas and adding more
fuel to the attack against e5.
3 1 .Rxd3 Ra8 32 .Rd l Qg7 B .bS Nd4 34.Rh l

Not falling for 3 4.Qxe 5 ? ? Rgl + ! .

34. . .NxB 3 S.Qxf3 Re8 36.Qxb7 e4 3 7.Qd7 QeS 3 8.Qh3 Qg7 39.cS RgS
40.Qe3 Kf8 41 .b6, 1 -0.

Here the game was adjourned, but Black decided to resign without re
suming.

o'

ce in a while a pe"on feel, on top of the wodd, nevee mlizing that,

in the blink of an eye, he might be turned into some form of road-kill. Usually
when success appears, it creates a warm glow, ego sets in, and you are mercifully
deprived of seeing sides of yourself that are, perhaps, better left unknown. Would
any man wish to feel supremely confident to the point of arrogance one day,
then wake up a few weeks later filled with a new self-image based on hopeless
ness, terror, and despair? The following is how this metamorphosis occurred in
my life.
My fall from grace occurred after a pleasant team match in Goerlitz, East Ger
many in March, 1 952 . For some reason that I've never comprehended, the trip
home took us to East Berlin, though it wasn't close to where we wanted to go.
Sitting in the train station with many hours between us and the departure of the
train to Budapest, my fellow team members Fiister and Szilagyi asked me if I
would like to go with them to West Berlin for some relaxation and shopping. At
the time, this wasn't a problem since the war was over-you just get on the sub
way and you're in West Germany. (You knew you were in West Berlin when you
saw sex magazines for sale. Such periodicals were not permitted in the Eastern
half of the city.)
Walking on the streets of West Berlin, we began to tentatively talk about defect
ing. Soon the discussion became serious and, having made up our minds, we made
our way to the U.S. Consulate. Sadly, it closed at 4:00 P.M. and we got there several
minutes after this time. We stood and stared at the building- what to do? Szilagyi
had a son in Hungary so he decided to go home. Fiister, though, said, "Why not
stay the night here and then we can go to the embassy in the morning? "
This was actually a tough decision for m e because I had qualified to play in
the Interzonal, and after that the Olympics, and those two events gave me an
67

PAL B E N KO : MY L I F E , GAM ES A N D C O M POSITI O N S

68

excellent chance to get safely to the West. If our plan in W. Berlin didn't work
out, though, I'd be doomed. After all, West Berlin was surrounded by East Ger
many-almost like an island- so I wasn't sure how we'd get out. Ultimately I
decided that it was best to just give up on our immediate plans to escape since,
clearly, we hadn't thought this through too well!
So we returned to the East Berlin train station and, as we were walking, I sud
denly changed my mind again! The Korean War was going on at that time and I
couldn't be 1 00% sure if another opportunity to escape would really arise- those
tournaments might be cancelled or I might not be allowed to attend for some
unknown reason (I wasn't a good boy politically, so anything could happen).
As if we were living out a scene from some silver-screen comedy, we once again
headed for the subway when a East German sport official appeared who knew
us. He wanted to know where we were going, we told him that we intended to
buy some cigarettes and whatnot, and he told us that we should return to our
team to avoid missing the train. We were trapped!
This created a real-life case of time pressure and, as anyone who has followed
my career knows, I've never been able to make decent decisions when the clock
is loudly ticking. We began to go back to the train with him, but I whispered to
Fiister, "Look, when we reach that comer, you run right and I'll run left and if
we're both lucky, I'll meet you at the consulate in West Berlin. " (A typical time
pressure plan!)
Things became a bit surreal here. We hit the comer, I started to run, and
Fiister just stood there! Suddenly there were cries of, "Halt! Halt!" Policemen
appeared, they eventually ran me down, and I was arrested and taken to prison.
In the meantime, while they were hunting for me, nobody was paying attention
to Fiister. He quietly waited until everyone had dashed off, and then calmly walked
to the subway and escaped! He could have told me that he wasn't going to run,
but no . . . We never did discuss this fiasco, but I have to admit that, at the time,
I felt like I had been set up.
I was kept for three days in an East German prison where I was questioned.
Of course, I couldn't deny anything-I was running in the opposite direction so
they caught me red-handed. After this, I was taken to Prague for three more
days, though this was a good deal since other Hungarian prisoners had been
detained for up to six months! Then a group of Hungarian secret police ap
peared in a car and took me to Budapest, which was actually a mercy because the
food in the Prague prisons was the worst I've ever experienced!
Back in Hungary, I was accused of being an American spy. An almost non
stop, three-week interrogation began that was designed to break me down men
tally so I would confess to all my crimes. During this time I was kept alone in a
prison cell - there was no contact with anyone other than the people question
ing me. Sleep wasn't allowed and bright lights were constantly in my face.
At first they asked me about the CIA, but I didn't know what that was. A poorly
pronounced "CIA" actually sounds like the Hungarian word for cats, so I thought
they were talking about an infestation of cats - it made no sense to me at all.

FALL FROM G RAC E

Finally I said, "I don't own any cats ! " This enraged them and they threatened to
start physical beatings, but I managed to calm them down by saying 1'd sign
anything they put in front of me!
During these interrogations, I was forced to write down what I had been doing throughout my life-virtually a minute-by-minute review! They demanded
to know about such things as whom I talked to, what my political and religious
beliefs were, and anything else that entered their tiny minds.
After searching my apartment for proof that I was trying to take down the
government (of course, there was nothing to find), the torturing dimwits re
turned with mail devoted to my postal games. They were sure that the chess
notation was secret code, and they demanded to know how to break the code so
they could see the real meaning behind it all! In particular, the game against
Tagmann (game five) caught their attention since, in their minds, the march of
my King through the middle of the board clearly had secret significance. It didn't
take long for me to realize that these men were not very bright, but I was in no
position to do anything but suffer and hope that they didn't start grinding my
bones to dust!
Finally this phase of my incarceration ended. Without ever having seen a law
yer or a judge, I was dragged out of my cell and taken to concentration camp
once they got the information they wanted (which in my case was nothing), they
would just lock you up and forget about you completely.
The "camp"- a large, dark building-had many small rooms, each of which
was crammed with twenty or more people. The windows throughout the build
ing were all blackened, and no sunlight was ever allowed to seep in. The vic
tims inside had no idea of time -if someone got sick they were ignored and
left to die, and if their teeth went bad we would just pull them out. While I was
there, many succumbed to starvation, and I remember one unfortunate man
who became so depressed that he tried to commit suicide by swallowing spoons
and anything else he could get down his throat. Sadly, the poor bastard sur
vived, but was in constant agony as the swallowed obj ects ate through his
insides.
At that time, normal citizens were having trouble finding proper food. Not wish
ing to waste what little existed on scum like us, we were given old bread and mea
ger amounts of stinking slop, the likes of which most dogs would gag at. Some
how, I only lost twenty pounds, but I was young and very strong and this allowed
me to survive. Nevertheless, is life in hell really a life at all? Looking into the
endless gloom, one would see walking "skeletons" hobbling by, and many of these
bags of bones would simply drop dead of starvation -one moment a living hu
man being, the next a rotting corpse that we would be forced to step over.
There were no visitors allowed, of course, and we were only permitted one or
two hours a day to read or rest. Eventually I made chess pieces from some bread
and played a game or two with other prisoners - anything to keep from going
insane! What little food we did get was mixed with saltpeter so that our sex drives
would fade into nothing. This helped to keep us all docile.

69

PA L B E N KO : MY L I F E , GAM ES A N D CO M POSITI O N S

70

My family didn't know I was locked away, they thought I had escaped! There
is a chessplayer named F. Benko in Argentina, and they thought this was me.
After a couple of months I was allowed to send them a postcard. I asked for one
kilo of lard, one kilo of jam (this was common among poor people- putting
lard and jam with a bit of pepper on bread gives you some energy), and that was
all I was allowed. It was only then that my family realized I was in prison. They
couldn't do anything for me, but it was still nice to know that I hadn't been
completely forgotten.
The one person who did know that I was locked away was Grandmaster Szabo.
He was the political editor of the top Hungarian chess magazine. The first page
had nothing about chess on it at all, just political ravings about the wonders of
communism. This mean-spirited person had no interest in helping me out; in
fact, he was happy about my being arrested!
Some of the prisoners had already been there for six years. No time frame had
been mentioned about my stay, and this implied that I was serving a life sen
tence. I became increasingly certain that I was going to rot there until I died
one moment a national chess hero, the next a broken creature relegated to an
existence of perpetual night.
Obviously, conversation became the main mode of entertainment. In time,
you would hear the life story of everyone in your room . . . then, you would
hear it again.
Many of the people I met hadn't done anything wrong at all. I knew one man
whose son had escaped from the army and had run to Austria. He was out of
official reach, so the powers that be arrested his mother, his father, and his brother
and placed them all in prison camp as a warning to anyone else that was even
thinking about breaking the law! These people had nothing to do with their
son's escape, but they were sacrificed and made an example of.
Another man, an editor of a sport magazine, was seen happily singing a song
in a restaurant. For some reason (perhaps it was a western tune), they said this
made him a traitor, and he was also locked away. Just for singing! Other people
were heard telling jokes that were determined to be "inappropriate," and this
was enough to give them life in prison camp.
Every day we would line up, and the guards would count everyone. Nobody
ever escaped. If a person collapsed, he would be dragged out and we wouldn't
see him again. I remember doing some physical work for a particular guard, and
he rewarded me by giving me six cigarettes. However, I don't smoke so I gave
them away to prisoners that do. Someone reported me - it was a crime to give
away a gift from a guard! They made me run up and down the staircase twenty
times and I learned my lesson. You have to realize that people would literally kill
each other for a cigarette. When I gave the cigs away, I thought, "We are all
prisoners . . . all equal . . . all brothers. " The man who reported me made it clear
that some were more "equal" than others.
I had been living like a diseased troll for a year and a half when a miracle
occurred: Stalin died. Shortly after this, President Nagy, who wanted to test the

FALL F RO M G RACE

Soviet's tight control, gave amnesty to most prisoners. Nagy was a communist
but was also a nationalist, so he tried to modify the Soviet mandate. The Soviet's
didn't like this and came into Hungary with tanks. They executed Nagy in 1 956.
After our release (in October, 1 95 3), we were given a lecture about not saying
anything, and given enough change to get to Budapest on a streetcar. We thought
it was a trap, but we happily risked being shot or followed-anything was better
than fading away in that huge tomb!
When I got out, I stayed with my brother. Everything had a wonderful glow
to it, the food tasted like nectar, and women seemed so beautiful that I had to
date as many as possible. Prison camp really makes you appreciate things that
you might have been oblivious to before!
I wasn't allowed to play chess for a couple of months - it took time for the
people in power to trust me again. In fact, I was constantly watched by the se
cret police, and they filed weekly reports on what I said, what I did, and whom I
talked to. Since chess was denied me, and since bookkeeping (my profession be
fore my arrest) was too dry, I became a draftsman. Of course, I wanted to get out
of Hungary more than ever, but any kind of travel was strictly forbidden and I
was forced to wait patiently until things quieted down.
I think I changed a lot after I got out of camp. Before, I thought I'd succeed
no matter what I did. I was full of confidence, voiced my opinions to anyone
who wanted to listen, and felt unstoppable. Afterwards, I learned to keep my
mouth shut. Camp had given me a heavy dose of mortality, and I was painfully
aware that my life could be snuffed out at any moment.
Eventually I was deemed to have "learned my lesson," and things calmed down.
My chess privileges were returned, I was given a passport, and I was allowed to
go to Moscow with our team. I did well there and at home, and that ultimately
put me in a position to represent Hungary in the zonals.
(2 1 ) G Barcza - Benko (Hungarian Championship, 1 954)
l .NO Nf6 2.c4 g6 3 .g3 Bg7 4.Bg2 0-0 5 .0-0 c5 6.d4 d6

From a potential Reti, to an English, we've finally reached the King's In

dian, Yugoslav Variation. Now White usually chooses between the calm
waters of the Symmetrical Variation (7 .dxc5) or the complexities of 7 .Nc3
Nc6 8.d5.
7.b3 ? !

A rare move that had been played on a few occasions i n Yugoslavian tour
naments.
7 ... d5 ! ?

An innovation that, in my mind, gives Black a fairly easy game. Other con
tinuations are 7 . . . Nc6 8.Bb2 Ne4, and 7 . . . Nc6 8 .Bb2 cxd4 9.Nxd4 Bd7 .
8.cxd5

71

Budapest 1 9 54. Stand i ng (left to right) : K. Honfi. B. Pogats. B. Berger, u n known. Dr. J. Ban. u n known. L.
Porti sch. P. Dely. S itti ng: I. B i lek. Gy. Kluger, P. Benko.To u rnament D i rector, G . Barcza. Dr. Szily. B. Sandor.
My chess priv i l eges were fi n a l ly returned!

White also can't expect anything from 8.Bb2 . In the game P. Nikolic
Krasenkov, Elista 1 998, Black equalized effortlessly after 8 ... cxd4 (of course,
8 . . . dxc4 is also good) 9.cxd5 Nxd5 1 0.Bxd4 Nf6 1 1 .Bb2 Qa5 1 2 .Na3 Nc6
1 3 .Qc1 Bf5 1 4.Nc4 Qa6.
8 ... Nxd5 9.Bb2 Nc6 l O.Qd2

Bad is 1 O.e3 ? Bg4 1 1 .h3 Bxf3 1 2 .Qxf3 Ndb4, when Black already stood
better in Gausel-Djurhuus, Munkebo 1 998.
l O ... Nc7

I avoided the tempting 1 0 ... cxd4 1 1 .Nxd4 Nxd4 1 2 .Bxd4! Nf4, since 1 3 .Bxg7
Qxd2 1 4.Nxd2 Nxe2 + 1 5 .Kh 1 Kxg7 1 6.Rfe 1 1eaves White with a very nice
endgame.
1 1 .Rd l Bg4

A hard choice. I wasn't interested in 1 1 . . . cxd4, though this was played four
years later in the game Raizman-Dunkelblum, Munich 1 9 5 8 . There fol
lowed 1 2 .Nxd4 Nxd4 1 3 .Bxd4 Bxd4 1 4.Qxd4 Qxd4 1 5 .Rxd4 Nb5 1 6.Rb4? ?
(Better is 1 6.Rd2 Rb8 1 7 .a4 Nd6 1 8.Nc3 Be6, with approximate equality.)
1 6 ... Rd8 ! 1 7 .Nd2 Nd4, and White is dead lost. However, I did take 1 1 .. .Be6
seriously. After 1 2 .d5 (Better is 1 2 .Na3 cxd4 1 3 .Nxd4 Nxd4 1 4.Bxd4 Bxd4
1 5 .Qxd4 Qxd4 1 6.Rxd4 Rad8 1 7 .Rxd8 Rxd8 1 8.Bxb7 Rd2 , when the ac
tive Rook on the seventh rank will ensure a draw.) 1 2 . . . Qxd5 1 3 .Nd4 Qd6
(during the game, I completely missed the stronger 1 3 . . . Qe5!) 1 4.Nxc6
Bxb2 1 5 .Nxe7+ Qxe7 1 6.Qxb2 Rad8 1 7 .Nc3 b6, White has nothing. Since

FALL FROM G RAC E

I wanted to play for a win, I finally decided that 1 1 . . .Bg4 gave me more
chances to reel in the full point.
1 2 .d5

Of course, 1 2 .dxcS ? ? loses instantly to 12 . . . Bxb2 .

1 2 ... Bxb2 1 3 .Qxb2 e6 1 4.d6

Perhaps this is a bit over-optimistic. The safer alternative was 1 4.Nc3 exdS
I S .Ne4 BxB (A year later, I S . . . d4? ! was played in the game Stahlberg
Beni, Hamburg 1 9 S 5 . White got a clear advantage after 1 6.NxcS Ne6?
1 7 .Nxb7, and went on to win.) 1 6.exB ( 1 6.BxB Nd4 1 7 .NxcS NxB + is
nothing for White, while the extremely tempting 1 6.Nf6+ leads nowhere
after 1 6 . . . Kh8 1 7 .NxdS+ Nd4 1 8 .Nxc7 Qxc7 1 9.BxB QeS) 1 6 . . . d4, and
Black's extra "half" pawn compensates for White's more active pieces.
1 4 ...Nd5

We've reached the key moment in the game. The result will revolve around
one question: "Is the d6-pawn strong or weak? " Black threatens to take on
d6 at once, so White is forced to find some way to make use of the
dynamic factors in his position.
1 5 .h3

I expected something more dynamic, but I have to admit that White can
still equalize after this move. Since the straightforward I S .e4 fails to I S ... Qf6
1 6.Qxf6 Nxf6, when Black will occupy d4 (or win d6!) and gain a clear
advantage, White only has two other moves to choose from: I S .Nc3 or
I S .Nbd2 : I S .Nc3 Qxd6! ( I S ... Qf6 1 6.Na4 Qxb2 1 7.Nxb2 Nc3 1 8.Rd2 Ne4
1 9.Rdd l Nc3 leads to a draw by repetition of moves) 1 6.Ne4 ( 1 6.e4 BxB
1 7 .BxB Nd4) 1 6 . . . Qe7 1 7 .Rac 1 ( 1 7 .RxdS is well met by 1 7 .. .fS !) 1 7 . . . BxB
1 8.Bxf3 b6 1 9.RxdS exdS 2 0.Nf6+ Kh8 2 1 .Nd7+ Nd4! 2 2 .NxfS QeS ! , gives
Black serious winning chances; I S .Nbd2 is a reasonable try, though White
fails to get any advantage after I S . . . BxB (worse is I S .. .fS 1 6 .NeS Bxe2
1 7 .Nxc6 bxc6 1 8.Re l Ba6 1 9.Rxe6 Bc8 2 0.BxdS cxdS 2 1 .Re7 d4 2 2 .Nc4)
1 6.Bxf3 Qxd6 1 7 .Nc4 Qe7 1 8 .BxdS exdS 1 9.RxdS Rad8, and Black has no
problems at all.

73

PA L B E N KO : MY LI FE, GAM ES A N D C O M POSITI O N S

i s ...B xf3 1 6.Bxf3 Qxd6 1 7.Nc3

There is nothing to be gained by 1 7 .e4 QeS ! .

74

1 7 . . .Nd4 1 8.BxdS exdS 1 9.e3 ?

White lets his last chance pass him by. He could have equalized with 1 9.Kg2
Qc6 2 0.e3 Nxb3 2 1 . Qxb3 d4+ 2 2 .QdS ! dxc3 2 3 .Qxc6, when the game should
be drawn. He avoided 1 9.Kg2 because he thought that 1 9 ... QeS was strong,
overlooking that 2 0.Rd2 ! RadS 2 1 .Rad 1 was possible, when 22 .e3 will follow.
1 9 . . . Nf3 + 20.Kg2 d4!

This refutes 1 9 .e3 .

2 1 .Kxf3

Even worse was 2 1 .Ne4 Qc6 2 2 .KxB fS .

2 1 . .. Qf6+ 2 2 .Kg2 dxc3 2 3 .Qe2

On 2 3 .Qc2 , Black plays 23 . . . bS followed by . . . bS-b4, when the extra pawn

is enough to win the game.
23 . . . Rad8

Black's pawn on c3 ties down the white army and, for this very reason,
Black strives to penetrate with his Rook and create a winning attack against
White's King. Instead of this, a more methodical and cautious plan was
2 3 ... Qc6+ followed by . . . b7-bS -b4, when the extra pawn should once again
prove decisive.
24.Rac1 Rxd l 2 S .Qxd l bS 26.QdS

The only try, but now my Rook achieves the penetration that I was striv
ing for.
26 ... Rd8

Completely wrong is 26 . . . RcS? 2 7 . Qb7 RdS 2 S.QxbS Rd2 29.QbS+ Kg7

3 0.Qf4.
27.QxcS Rd2 2 8.e4

Black gets a winning endgame after 2 S . QcS+ Kg7 2 9 . Qxc3 Rxf2 +

30.Kg1 Rxa2 .
2 8 ... Rd3 29.f4

FALL FRO M G RACE

Hopelessly opening up his King, but the game would also end quickly
after 29.QxbS Qf3+ 3 0.Kgl c2 ! 3 1 .QeS+ Kg7 3 2 .QeS+ Kh6.
29 . . . Qe6! 3 0 . e S Rd2 +

The rest of the game is a bloodbath.

3 1 .Kf3 Qxh3 32.Qxc3 Qg2 + H .Kg4 Qe2 + 3 4.QO

I would have preferred 3 4.KgS QhS+ 3 S .Kf6 QfS + 3 6.Ke7 Qe6 mate.
34 . . . hS+ , 0- i .

(22) Benko - Kluger (Budapest, 1 95 5)

l .c4 Nf6 2 .Nc3 e6 3 .NO dS 4.d4 cS S.cxdS cxd4

A rare move. Usual is S . . . NxdS .

6.Qxd4 exdS

And here, 6 ... NxdS is the most common response, though 7 .e4 Nxc3 S.Qxc3
is known to give White a small but bothersome edge.
7.e4! Nc6

On 7 . . . dxe4, White gets the better game with S.QxdS+ KxdS 9.NgS Be6
1 0.Nxe6+ fxe6, and now Bisguier-Hearst, NY 1 9S4 continued: I I .BgS (even
stronger than I l .Bc4 Nbd7 1 2 .Bxe6 Bb4 1 3 .Ke2 ReS 1 4.Bxd7 Kxd7 I S .Rd l +
Ke6 1 6.NbS RacS 1 7 .Be3 a 6 I S .Nd4+, which also favored White in
Bolbochan-Madema, Mar del Plata 1 9S3) I l ...Bb4 1 2 .0-0-0+ KeS 1 3 .Nxe4!
Nxe4 1 4.RdS+ Kf7 I S .RxhS NxgS 1 6.Bc4, when Black was in bad shape.
8.BbS

A very sharp situation has appeared by move eight!

8 ... a6? !

The game Geller-Keres, Zurich 1 9 S 3 , went S . . . Nxe4? 9.0-0 Nf6 and

now, instead of Geller's 1 O.Re 1 +? , Najdorf dealt with this line in the tour
nament book where he offered a detailed analysis of the best move:
1 0.NeS (Later, Keres himself claimed a clear advantage for White with
the straighforward 1 0.BgS Be7 I 1 . Bxf6 Bxf6 1 2 . Rfe l + Be6 1 3 .QcS)
10 . . . Bd7 (if 10 . . . Qc7 , White crashes through with I 1 .Bf4 Bd6 1 2 .Nxc6

75

PA L B E N KO : MY L I F E , GAM ES A N D CO M P OSITI O N S

76

bxc6 1 3 .Nxd5 Nxd5 14.Qxd5) I 1 .Nxd7 Qxd7 (The greedy 1 1 . . .Nxd4

leads to a fun finish: 1 2 .Nxf6+ Ke7 1 3 .Nfxd5+ Kd6 [Black won't find any
joy on the kingside: 1 3 . . . Ke6 1 4.Re 1 + Kf5 1 5 .Bd3+] 1 4.Bf4+ Kc5 1 5 .b4
mate.) 1 2 .Bxc6 bxc6 1 3 .Bh6 ! 0-0-0 ( 1 3 . . . c5 1 4.Rfe l + Be7 1 5 .Qe5 0-0
1 6.Qg5) 1 4.Rac 1 KbS 1 5 .Na4, with more than enough for the sacrificed
pawn.
However, all this was apparendy made moot in 1 9 5 5 when Vukovic
pointed out that Black should play S ... Bd7 9.Bxc6 Bxc6 1 0.exd5 (1 0.e5 Ne4)
1 O ... Bxd5, when he claimed equality. The position after 1 O . . . Bxd5 occurred
in Polugaevsky-1. Zaitsev, USSR 1 96 3 , and White quickly showed that
Black wasn't out of the woods by any means: 1 1 .0-0! Bc6 (l l ...BxB 1 2 .Qe3 +,
and 1 1 .. .Be7 1 2 .Nxd5 Qxd5 1 3 .Re l , are both obviously good for White)
1 2 .Re l + Be7 1 3 .Qe5 .
Black's best way to handle the position after S.Bb5 is probably S . . . dxe4
(Played by Lasker in 1 907 against Marshall!) 9.QxdS+ KxdS 1 O .Ng5 , when
White's advantage is manageable.
9.Bxc6+

Vukovic felt that White gets a greater advantage by 9.exd 5 ! axb5 1 0.dxc6.
His analysis went as follows: A) 1 O . . . Qe7+ I 1 .Be3 bxc6 1 2 .0-0 b4 1 3 .Rfe l
bxc3 1 4.Qxc3 , and Black's King i s stuck in the middle o f the board; B)
1 0 ... bxc6 I 1 .QxdS+ KxdS 1 2 .Ne5 b4 1 3 .Ne2 ! KeS 1 4.Nd4 c5 1 5 .Nb5 RbS
(Vukovic failed to mention 1 5 . . . Nd5 ! , which seems like a much stronger
defense.) 1 6 .Nc7 + KdS 1 7 .Nxf7+ Kxc7 I S .Bf4+, and White wins; C)
1 0 . . . Qxd4 I 1 .Nxd4 bxc6 1 2 .Nxc6 Bb7 1 3 .Nd4 Bxg2 1 4.Rgl followed by
Nxb5. Is Vukovic's analysis correct? Perhaps, perhaps not. As usual, I wasn't
aware of its existence when I was playing this game.
9 ... bxc6 10.NeS

White's best is 1 O.exd5 Nxd5 1 1 .0-0, with a clear advantage. However, I

saw the possibility of some interesting tactics and wanted to see how they
turned out.
10 . . . Bb7 l 1 .exdS NxdS 1 2 .0-0 Be7 1 3 .Nxc6 ! ?

As I mentioned i n my note to the tenth move, I just couldn't resist!

13 ... Bxc6 1 4.Qxg7 Rf8

According to Vukovic, 1 4 . . . Bf6 1 5 .Re l + Kd7 1 6.Qxf7 + KcS 1 7 .Qe6+

( 1 7 .Re6 BeS!) 17 ... Kb7 I S.Ne4 ReS, was the real test of my sacrificial con
cept. However, instead of 1 7 .Qe6+, I think the immediate 1 7 .Ne4 is a better
continuation.
I S .Re 1 Qd6

This same position had been reached in Najdorf-Keres, Zurich 1 95 3 . There,

Najdorf played 1 6.Ne4? ! , when both players agreed that a draw would be
the best result!
1 6.NxdS QxdS?

Early in my career my style was very sharp.

This obvious move turns out to be an error. Better was 1 6 . . . Bxd5 1 7 .Bg5
Be6 ( 1 7 .. .f6 1 S.Bf4) I S.Rad l Qc5 1 9 .Bh6 Qf5 ( 1 9 ... RdS 20.Qxh7) 20.Qg3
Qg6 (Even 20 ... RhS is possible since 2 1 .Qc3 f6! holds.) 2 1 .Qf3 RcS 2 2 .BxfS
KxfS, when Black is alive and kicking.
1 7.Bf4! !

On the immediate 1 7 .Bh6, Black might castle long when I S .Rxe 7 loses to
I S . . . RgS.
1 7 . . . Rd8

Now castling long is a mistake: 1 7 . . . 0-0-0 I S .Rxe7 RgS 1 9.Rc7+ KbS

20.Rxf7+, and Black will be mated.
1 8.Bh6 !

This might look like a loss of tempo but, in reality, Black's inability to
castle long now makes Bh6 very strong. White threatens 1 9 .QxfS+, and
the only way Black can deal with this is to give up his Queen.
1 8 ... Kd7 1 9 .Rad l

The rest of the game is not too difficult.

PAL B E N KO : MY L I F E , GAM ES A N D C O M POSITI O N S

1 9 . . . RgS 20.RxdS+ BxdS 2 1 .Rxe7+

Forcing exchanges and making the win easy.

78

2 1 ...Kxe7 22 .QeS+ Be6 2 3 .BgS+ RxgS 24.QxgS + KeS 2 S .h4

Not only stopping back-rank mate, but also moving this pawn closer to
the queening square. I now intend to win the h7-pawn and push my h
pawn to the promised land.
2 S ... Bxa2 26.QgS+ Ke7 27.Qxh7 BdS 2 S.Qg7 Bc6 29.hS Rd l + 30.Kh2
RdS 3 1 .g4 Rd l 3 2 .h6, 1 -0.

Was it all sound? I'm not sure. The real test was probably 16 ... Bxd5. How
ever, the great Mikhail Tal wouldn't have hesitated to sacrifice with
1 3 .Nxc6! ? , and as the game showed, the problems Black faces proved im
possible to deal with in over the board play.

(23) Portisch - Benko (H ungarian Championship, 1 95 5)

l .d4 Nf6 2 .c4 g6 3 .Nc3 Bg7 4.e4 d6 S.O 0-0 6.Be3 eS 7.dS cS

Black's best choices are 7 . . . Nh5 followed by . . . f7 -f5 , or 7 . . . c6. Wanting to

label my move as a mistake, White stops me from carrying out that ma
neuver.
S.g4! hS !

In turn, I try to point out some of the disadvantages of 8.g4. More com
mon plans are 8 . . . Ne8 followed by 9 . . . f5 , or 8 . . . Na6 followed by . . . Nc7,
. . . a7-a6, . . . Rb8, and . . . b7-b 5 .
9.h3

I would have met 9.g5 with 9 . . . Nh7 followed by . . . f7-f6.

9 ...Nh7

A year later, the game Tal-Georgadze, USSR 1 956, continued 9 ...Na6 (pre
maturely moving this Knight away from the e5-square [via d7] , thus al
lowing White to play for a quick, effective, f3 -f4 advance.) 1 O.Qd2 Nh7
1 1 .0-0-0 h4 1 2 .f4! exf4 1 3 .Bxf4 Nc7 1 4.Nf3 , and White had a large ad
vantage.
10.Qd2

This gives Black a comfortable game. Better was 1 0.h4 ! ? or 1 O.gxh5 gxh5
(and not 10 ... Qh4+ l 1 .Bf2 Qxh5 1 2 .Nb5 ! when Black loses material) 1 1 .Rh2 ,
though Black's game is still playable: 1 1 . ..f5 1 2 .Rg2 ( 1 2 .exf5 ! ?) 1 2 .. .f4 1 3 .Bf2
Kh8 1 4.Rg6 Rg8 1 5 .Qd2 Bf6, with a reasonable position.
10 . . . h4 l 1 .gS f6 1 2 .gxf6 Qxf6

Also good is 1 2 . . . Bxf6, since 1 3 .f4 leads to a clear Black edge after 1 3 . . . exf4
1 4.Bxf4 Bg5 1 5 .Nge2 Nd7 followed by . . . Ne5 .
1 3 .Rh2 Na6 1 4.Rg2 Nc7 I S .Bd3 KhS 1 6.0-0-0 gS

FAL L FROM G RAC E

79

Necessary, else White would play Rfl followed by f3 -f4. Now White must
make a major decision since passive play allows Black to choose between
two good plans: he can start a queenside attack by . . . a 7 -a6 followed by
. . . b7-bS , or he can play positionally and strive to bring a Knight to f4 by
. . . Bh6 followed by . . . Nc7-e8-g7-hS .
1 7.Rxg5 ! ?

White couldn't prevent the plans just listed, s o this i s his only chance to
grab the initiative.
1 7 . . .Nxg5 1 8.Bxg5 Qg6 1 9.f4 Bh6 20.Nf3

White decides to turn the game into a wild berserker slugfest. A less dras
tic approach to the position was 2 0.Bxh6 Qxh6 2 l.fxeS (2 l .fS ! ?) 2 1 . . . Qxd2+
2 2 .Rxd2 dxeS 2 3 .Be2 , when the realization of the extra Exchange would
not be an easy task for Black.
20 . . . Bxg5

Also good was 2 0 . . . Rxf4, which

leads to a similar position after
2 1 .Bxh6 Qxh6.
2 1 .Nxg5 Rxf4 22 .Rgl Bd7 2 3 .Ne2
Rf2 ! 24.Qe3 Qf6

Cle arly the best move . On

24 ... Raf8, White can mix things up
with 2 S .Ne6 Rf1 +? (Black is still
in charge after 2 S . . . R8f3 ! , but I
thought it best to avoid the follow
ing complications: 26.Rxg6 Rxe3
2 7 . Kd 2 Rff3 2 8 . Rh 6 + Kg8
29.Rg6+ Kf7 3 0.Rg7 + Kf6 3 1 .Rxd7
Rxd3+ 3 2 .Kc2 Rde3 3 3 .Nd Nxe6
3 4 . Rxd6 Rf2 + 3 S . Kb 3 Rxh 3
3 6.Rxe6+) 2 6.Kc2 ! Rxgl 2 7 .Nxf8,
and Black has been tricked!

Lajos Portisch, 1 9 66.

(Photo courtesy USC F.)

PA L B E N KO : M Y LIFE, GAMES AN D C O M POSITI O N S

2 S .Nf7+

80

Since Black didn't fall for White's

trap, Portisch is now forced to add
fuel to the fire and hope I slip up
in the coming tactical complica
tions.
2S . . . Qxf7 26.Qh6+ Qh7 2 7 . Qxd6
Ne8 28.QxeS + Nf6

os Portisch: Lajos moved up

the ranks slowly. But eventually, by

working very hard on chess every day,
he became one of the best players in
the world. He was famous for taking
a briefcase of chess notes and opening
ideas everywhere he went.
Tactically he wasn't very good. In

2 8 . . . Ng7 allows 29.Qd6, with the

threats of 3 0.Qxd7 and 30.Rg6 fol
lowed by 3 1 .Rh6.
29.Qc3 Rg8 30.eS

fact, he once told me that an average

Hungarian master was better than him
tactically. Of course, this is an exag
geration, but his enormous positional
skills are what made him special .

Down a whole Rook, Portisch at

tacks like a man who is looking death
in the face !
3 0 . . . Rxg l + 3 1 .Nxgl BfS !

I remember acting as Portisch's sec

ond for an event in Bad Kissingen. I
wasn't used to Germany at that time,
so when I walked in the sauna and saw
a nude woman there, I thought that I

This counterattack puts an end to the

battle, though White isn't quite ready
to accept his fate.
3 2 .Qe l Qh6+ H .Kd l Bxd3 34.Qxf2
Ne4 3 S .Qg2

must have gone in on ladies day! Em

barrassed, I began to retreat when I saw
Portisch showering, and he explained
that it was, indeed, men's day. Shrug
ging my shoulders, I once again headed
back into the sauna, this time running

No better was 3 5 .Qel Qf4,

into two gorgeous eighteen-year-old

girls. I was thoroughly confused, and

3 S . . . Qe3 36.e6 Nf2 + , 0- 1 .

even suspected that Portisch might

have been playing some sort of trick
on me. Now I know better-in Ger
many, it's very common for nude men
and women to hang out together in
hotel pools and saunas. This is one
great thing about chess: it's such an
international game that it gives you
wonderful opportunities to come face
to face with many different cultures
and customs.

Lajos Porti sch vs. G l igoric, Vrnjacka

Banja 1 9 66.

(Photo courtesy USeE)

feW

Y"'" wm devoted to one thing' "e'ping from Hung''Y! I

told my father of my intentions, and shared my fears that my escape might get
him into trouble. He calmed me by saying that I had to do what I thought was
right. Fortunately, he managed to get away before I did, and that gave me a
green light to increase my own efforts.
My ticket to freedom was, naturally, chess. First on my agenda was to qualify
for an out-of-country zonal. During the tournament that decided this, I gave
myself some uneasy moments by playing too well ! The problem was a bit hu
morous: If I came in first, I would qualify for a zonal that was going to be held in
a Soviet-controlled country. This wasn't acceptable, and my sights were set on
the zonal in Ireland, which required a second-place finish. Ironically, I was win
ning the qualifier and desperately went out of my way to draw the last three
games (avoiding victory like the plague !) so I could ensure the coveted second
position!
In the end, everything worked out well and I did, indeed, get sent to Ireland.
Nevertheless, my plans changed when I was invited to the World Student Cham
pionships, to be held in Reykjavik, Iceland (I played first board, Portisch was on
board two). I decided this was a better place to actualize my defection, so after
Ireland I calmly traveled to Luxembourg where I hung out for a month giving
exhibitions and whatnot, and then made the fateful trip to Reykjavik.
The only people that knew of my intentions were my brother and sister, who
remained in Hungary. They gave me their best wishes and, on the way to Ice
land, I sent a telegram saying, "My uncle is sick." This was a secret code telling
them to clean out my apartment because I wasn't coming back.
In July 1 95 7 , I walked into the American embassy in Reykjavik and asked for
asylum. The Americans seemed quite happy about it, no doubt realizing that I
81

From l eft: Fo l d i , Novarovaki, Portisch, and Benko i n Reykj avik 1 957

was a useful political tool. I gave a press conference and explained why I didn't
want to go back. It almost seemed too simple, but I was free. All that remained
was to wait until I was given clearance to travel to the United States.

For a while, I tried to reach reversed King's Indians, Pircs, and Griinfelds with a
tempo more. The following game is one of my early experiments with this idea.

(24) Benko - G iustolisi (Dublin, 1 95 7)

l .c4 Nf6 2 .Nc3 g6 3 .g3 Bg7 4.Bg2 0-0 S.d3 cS 6.NO dS 7.0-0

White has reached a reversed Yugoslav Variation of the King's Indian with
an extra tempo ( l .d4 Nf6 2 .c4 g6 3 .NB Bg7 4.g3 0-0 5 .Bg2 0-0 6.0-0 c5
7 .Nc3 Nc6 is how it is achieved when Black plays the KID).
7 ... d4? !

Black enters a sharp main line with a tempo less. More reasonable would
have been 7 . . . dxc4 8.dxc4 Nc6, when we've suddenly reversed colors again
(I would have the white side of a Yugoslav Exchange Variation) !
8.Na4 Nfd7 9.e3 Nc6 1 0.exd4 cxd4

If 1 O . . . Nxd4 1 1 .Nxd4 Bxd4 1 2 .Be3 , Black would want to avoid 1 2 . . . e5 due

to 1 3 .Bh6 (White would also gain an edge by 1 3 .Bxd4 followed by playing
for a b2 -b4 advance.) 1 3 , ..Re8 1 4.Nc3 .

ESCAPE TO F RE E D O M

1 1 .a3 a5 1 2 .c5 !

83

Gaining queenside space, clamping down on the b6-square, giving my pieces

access to the c4-square (NB -d2 -c4 is possible), and preparing Bd2 fol
lowed by b2 -b4-b5 with a huge advantage.
12 . . .Rb8 1 3 .Re l Re8 1 4.Bd2 ! h6

On . . . b7-b5 or . . . b7-b6, White would trade: 1 4 . . . b5 1 5 .cxb6 Nxb6 1 6.Ng5

Bd7 1 7 .Nc5 , and Black is faced with several threats. For example, on 1 7 ... h6
1 8.Nxd7 Qxd7, White has 1 9.Nxf7 ! Kxf7 2 0.QB +, picking up the Knight
on c6.
1 5 .b4 axb4 1 6.axb4 b5

Black couldn't let White play b4-b5 .

1 7.cxb6 Ba6

The natural 1 7 . . . Nxb6 loses the Exchange: 1 8.b5 Na7 1 9.Ba5 .

1 8 .Qb3 ! Bb5

1 8 . . . Nxb6 is still a poor idea. This time it leads to the loss of a pawn after
1 9.Ng5 hxg5 2 0.Bxc6 Rf8 2 1 .Bxg5 .
1 9.Qd5 ! Bxa4 20.Rxa4 Qc8?

Of course, 20 ... Nxb6? ? loses two pieces for a Rook after 2 1 . Qxc6 Nxa4
2 2 . Qxa4. However, Black missed his last chance to make a fight of it:
20 . . . Rxb6 2 1 .b5 (2 1 . Rea 1 ! ?) 2 1 . . .e6 ! (A very annoying move that extricates
Black from his immediate problems, though more long-term worries re
main.) 2 2 .Qc4 (The only move. 2 2 .Qa2 hangs the b-pawn, while 2 2 .Qb3
allows 2 2 . . .Nc5 .) 2 2 , . .Nce5 2 3 .Nxe5 Nxe5 2 4.Rxe 5 ! Bxe5 2 5 .Ba5 Bc7
26.Bxb6 Bxb6 2 7 .Qc6 ! RfS 2 8.Ra6, and the breaking of the blockade al
lows White's passed pawn to push powerfully forward.
2 1 .b7 Rxb7

2 1 . . .Qxb7 2 2 .Nxd4 Bxd4 2 3 .Qxc6 is also winning for White.

2 2 .Ng5 !

PAL B E N KO : MY L I F E , GAM ES AN D C O M POSITI O N S

84

The simple 2 2 .Nxd4! Bxd4 2 3 .Qxc6 Rc7 24.Qe4 is also good enough.
2 2 ... hxgS

Easy for White is 2 2 . . . Nd8 2 3 .Nxf7 Nb6 24.Nd6+ Nxd5 2 5 .Bxd5+, but
2 2 . . . Nde5 would have put up a lot more resistance.
2 3 .Qxc6 Rc7

Better than 2 3 . . . Qxc6 24.Bxc6 Rc7 2 5 .b5 Bf6 2 6 .Ba5 Ra7 2 7 .b6 Nxb6
2 8 .Bxb6 which, of course, is resignable for Black.
24.QbS Bf6 2 S .RaS

The rest is just technique, and doesn't require comments.

2 S . . . NbS 26.BxgS BxgS 27.QxgS Rc2 2 S.h4 Qc3 29.Rxe7 Rc 1 + 3 0.Kh2
Rf8 3 1 .BdS, 1 -0.

I was in Iceland for three months, and was treated very well there. I had two
girlfriends during that time, and they made my stay less alienating. The first
didn't last long, and we broke up. The second, though, was a very different mat
ter. We became very close. This, combined with well-wishers telling me that my
chess skills would be far more appreciated in Iceland because the game wasn't
popular in the United States, made me seriously consider settling down there.
This young lady and I became inseparable, and I remember being invited to her
parent's home for dinner. I immediately saw wedding bells in their eyes (at that
time, there were only 1 50,000 people in Iceland, and everyone wanted children)!
Ultimately, I decided that my destiny lay in the United States, and when my visa
arrived I broke the news to her that I had to leave. She came to the airport to say
goodbye, tears flowed, but we never saw each other again. This is all part of the
tournament traveling experience - at times it can be very hard, almost bittersweet.
I eventually heard she got married and had two children. Life goes on.

landed in New yo,k on Octobe< [7, [957. I was bmke, but not as
one might suspect , isolated. Thanks to chess, I had made many good friends all
over the world, and this "habit" continued in the United States. Meeting people
was natural and easy, but money was quite another matter! With only $350 to
my name (courtesy of a victory at an Icelandic event) and no place to live, I kept
my expenses down by renting a room at a YMCA and eating only two small
meals a day.
Now it was time to plan my future, but it quickly became clear that the chess
landscape painted an ugly picture. There were few tournaments, the prizes were
small in the events that did exist, and the United States Chess Federation (with
only 6,000 members) had limited funds-they offered me lots of moral support,
but I couldn't pry a penny from their fingers. This chess desert extended far
beyond my own life. The U.S. team couldn't go to the Olympics in Moscow due
to lack of funds, and the concept of a chess professional was non-existent:
Reshevsky had a job, Lombardy became a priest, Rossolimo was driving cab,
and Evans was into all sorts of things. With a heavy heart, I realized that I might
have to give up chess completely.
I left New York after only three weeks, traveling to Akron, Ohio to be with
my father. I was there a month when my father lost his job as an engineer (work
was very hard to come by at that time), and we both moved to Cleveland. Though
I was now based in Cleveland, I wanted to see the rest of the country. A traveling
opportunity arose when I got word that an international tournament in Dallas
was about to start, but the invitations had been sent out long ago and I wasn't
able to play. However, the organizers asked me to come anyway as an alternate.
If everyone showed up (which they did), they would make sure I played a few
games by arranging a match against the champion of Texas, Ken Smith.
85

PAL B E N KO: MY LI FE, GAM ES A N D CO M PO S I TI O N S

86

This match started easily for me, as I won the first three games. At that point
the organizers approached me and said, "You can't win every game ! It won't look
good. Next time, give him a draw." Not wanting to be rude to my hosts, I agreed
to their demand. We played twenty moves to make things look normal, and then
I proposed peace. He said, "No ! I stand better. " This really bothered me since I
hadn't been trying to win - if I had gained a large advantage it would be hard to
convince the spectators that the draw was legitimate.
The game, of course, continued, though now I was paying attention to the moves
I made! Soon I was on top and he offered a draw, but this time I refused. Looking
at me as if! had done something despicable, he screeched, "You promised! "
What could I do? I gave him the draw and walked away muttering to myself.
Later in that match I actually lost a game by hanging a Rook (a "habit" that I
eventually became famous for). In the end, the organizers were happy- even
though I had won by a wide margin, their champion had gotten his pound of
flesh and was able to retain a sense of dignity.
After my match with Smith, I continued west (Bent Larsen, who had played in
the Dallas tournament, went with me), stayed at Larry Evans' motel in Los An
geles for a month, and continued on to San Diego. Then I returned to Cleve
land for a few months, and finally (permanently) moved to New York where I got
an apartment for $9 a week. Still strapped for cash, I ate at the automats for a
dollar or so to keep costs down. It looked like life was going to be a continuous
struggle, all the more so when I was invited to the Portoroz Interzonal, only to
find that the USCF couldn't pay anything towards the trip.
At this point - like a western where the cavalry comes riding to the rescue
the Hungarian Sport Organization took me under wing, insisted I play in Portoro,
and paid for all my expenses. Led by Mr. Friedrich Gorog- the very kind presi
dent of the Hungarian Jews Federation, to whom lowe a great debt- the goal
of this organization was to help disenfranchised Hungarian athletes in the United
States get on their feet. When I returned from the tournament, they got me a
job in a brokerage firm, and also arranged paid vacations whenever I needed to
play internationally. I was suddenly blessed with an incredible situation.
I did reasonably well in Portoroz and in the Zagreb Candidates Tournament,
though I didn't prepare at all. In fact, I was already working in the U.S . before
Zagreb and was no longer able to pay much attention to chess. Who could? My
job was very demanding - I had to rise at 6 A.M. every morning, fully concen
trate on my duties, and also find the time to study English.
My immediate financial worries were over, but the same could not be said for
the many desperate players in Eastern Europe. A good illustration of this is the
1959 Candidates Tournament in Zagreb, Yugoslavia. I arrived there a week be
fore the tournament started so I could get acclimatized. I didn't have an early
reservation at the international hotel where the event was going to be held, so I
found a smaller place to stay while waiting for everything to begin. This hotel
officially cost $1.25 a day, but on the black market you could arrange a price of
one dollar. Anyway, I was very happy about this price, of course, and then began

Enjoying the laid -back Cal ifornia lifestyle.

to search for reasonably cheap restaurants to eat at. Finally someone asked, "Why
are you eating out when you've already paid for your food? " I was shocked to
discover that the dollar a day price included both lodging and food-three full
meals every day!
When the tournament began, I ran into a Yugoslav friend who was acting as
second to one of the players, and he was looking for a place to stay. I recom
mended this amazing little hotel but he refused, telling me that a dollar a day
was much too much-it was far beyond his means ! In the end, he shared a room
with four other desperate Yugoslavs and they paid a quarter each.
I became acquainted with Fischer at the picturesque town of Portoroz, situ
ated along the Yugoslavian coast. He was just a teenager at that time, a nice kid.
Sometimes he would cry if he lost a game, and I quickly became quite fond of
him, almost protective. Once I asked him what he wanted to be, and he said, "I
want to be an international playboy, just like Benko ! " Unfortunately, he was al
ways very shy and never learned how to communicate with women.
This tournament was a very important one: Taking any of the top six places
ensured participation in the Candidates Tournament, so you can imagine how
tense the proceedings often were. Twenty-one of the world's strongest players
gathered together for this event-I had qualified for it in the Hungarian cham
pionship and the zonal tournament in Dublin.

PAL B E N KO : MY L I F E , G A M E S A N D CO M PO S I TI O N S

88

How grandmasters relax (in Belgrade).Tal (far left) , Petrosian, Averbach (holding on for
dear life in the middle) , and Keres (right) whose bow-tie became the subject of a
running series of jokes in the Yugoslav press.
(Photo courtesy USCF.)

Though I had seen Fischer's gentler side, I was also aware that he was an op
ponent to be reckoned with: He was only fifteen and was already U.S. Cham
pion and world famous. In fact, he was considered to be virtually unbeatable on
his home soil, so I was really looking forward to seeing his stuff. We were paired
in the fourth round with me having White. His performance in the event thus
far hadn't been spectacular, since he had played three of the weaker participants
and had only scored one win and two draws. He clearly didn't possess the kind,
of experience that only years of international competition produces, and this made
him vulnerable.
In his win (as Black) against the Hungarian-Canadian Fiister (yes, the same
Fiister that played such an important role in my arrest!), I spotted an improve
ment for White. Hoping that his self
confidence would lull him into a false
sense of security, thereby preventing
him from analyzing this game, I de
cided to use this same opening, posi
tive that he would play his favorite
King's Indian Defense. I intended to
test my new idea and drag him into un-

Garry Moore, the emcee of the popu lar

TV game show "I've Got a Secret" gives
fifteen-year-old Fischer round tri p tickets to
Moscow (where he played a series
of exh ibition matches) , thus al lowing h i m
t o participate i n t h e Interzonal by
circu mventing the price of airfare .
(Photo courtesy USC F.)

T H E N EW WORLD

known territory, forcing him to solve difficult problems on his own. The follow
ing game was my first of three career victories over Fischer.

(25) Benko - Fischer (Portoroz I nterzonal, 1 958)

l .d4 Nf6 2 .c4 g6 3 .Nc3 Bg7 4.e4 d6 S.3 eS 6.Nge2
White would gain nothing by taking on eS and trading Queens, since the
hole on d4 would guarantee Black good long-term chances.

6 ... 0-0 7.BgS

Fiister played 7 .Be3 , but this pin of the f6-Knight caught my eye immedi
ately. Now 7 . . . h6 8.Be3 allows White to gain a tempo with a subsequent
Qd2 , hitting h6.

7 ... exd4
My surprise move did its work, and Bobby gives up his hold on the center.
He is hoping to exploit the absence of my dark-squared Bishop from e3 ,
but this doesn't lead to the desired result. More interesting would have
been 7 . . . Nc6 8.dS, and now 8 . . . Nd4 ! ? 9.Nxd4 exd4 1 O.Qxd4 h6 1 1 .Bxf6
Bxf6 leaves Black with some compensation for his pawn minus in the form
of an active Bishop and firm control over the dark-squares.

8.Nxd4 Nc6 9.Nc2

Getting off the dangerous h8-a l diagonal, the Knight eventually hopes to
jump to e3 and d S . Worse was 9.Be2 ? ! Nxd4 l O.Qxd4 Nxe4! , when the
resulting exchanges would lead to an equal game. By avoiding trades, White
leaves Black with a cramped position.

9 . . . Be6 l O.Be2
It's important to point out that 1 O.Qd2 wouldn't stop Black from playing
l O . . . h6, since 1 1 .Bxh6 Nxe4 1 2 .Nxe4 Qh4+ would make the second player
quite happy.

l O ... h6 l l .Bh4
Moving the Bishop back to e3 would be illogical, since that square is re
served for the c2-Knight.

l 1 ...gS

89

PAL B E N KO : M Y LIFE, GAM ES AN D CO M PO S I TI O N S

90

The impatience of youth! Here Bobby couldn't stand the pin on his f6Knight any longer, and breaks it at a huge cost: his f5-square is now seri
ously weakened. An experienced tournament player would think twice be
fore making such a move. Though f5 can't be exploited right away, it is
bound to lead to further concessions in the long run . I will say that finding
good plans in cramped positions is never easy, and a player of that age rarely
has the defensive technique necessary for such a daunting task.

1 2 .Bf2 Ne5 1 3 .Ne3 c6

Defending d5 but weakening d6.

1 4.0-0 Qa5
A surprise. This offers a pawn sacrifice: 1 5 .Qxd6 RfdS I 6.Qa3 Qxa3 I 7 .bxa3 ,
when my queenside is a shambles. However, the natural I 5 .Qc2 RfdS lets
him force the freeing . . . d6-d5 with good play. So this position proved to be
full of tricks, and I began to appreciate the skills of my young opponent.

1 5.Qd2
Crossing Black's plans and taking aim at g5 .

1 5 ... Rfd8 1 6.Rfdl

Still clamping down on the d5 -square. The other Rook stays on the
queenside in order to curtail Black's play in that sector.

1 6 ... a6 1 7 .a4 Qc7

The more active I 7 . . . Qb4 isn't wise, since after I S.a5, Black couldn't cap
ture on c4 due to I 9.Ra4.

1 8.a5
This nails down Black's queenside pawns and threatens a timely Bb6.

1 8 ... c5
Defending the b6-square, but at the cost of weakening d 5 . The progres
sive weakening of Black's position now compelled me to start my attack,
while Bobby's central probe and queenside play has failed.

-{l\(JM

'\SoISH Y ?-,:,

/., __

..

....,

A very
young,
innocent
Fischer:

T H E N EW WORLD

1 9.h4 Qe7 20.hxgS hxgS 2 1 .NfS

Finally occupying the hole on f5 .

2 1 ... BxfS 2 2 .exfS g4 2 3 .Bh4

The annoying Bishop is back on the h4-d8 diagonal! Now the obvious
threat of Nd5 forces Fischer to take preventive measures.

2 3 ... Qf8
Black steps out of the pin and creates possibilities of . . . Bh6. Here I
refrained from 24.Bxf6 Bxf6 since, even though I win a pawn, the result
ing Bishops of opposite colors would give Black good defensive chances.

24.fxg4 Nexg4 2 S .Bxg4 Nxg4

Black appears to be getting some activity, since . . . Bd4+ might prove trouble
some.

26.QgS
A complete answer. Now g4 is hanging and the g7-Bishop is pinned.

26 ...Nf6 2 7.Rd3

This isn't subtle, but the threat of swinging the Rook over to the kingside
(g3 and/or h3) is devastating.

2 7 ...Nh7
Even worse is 2 7 . . . Kh7, since White can employ an obvious but decisive
Queen sacrifice: 2 8 .Qxf6 ! Bxf6 2 9 .Bxf6, when the threat of Rh3 ends the
game.

2S.Qg4 f6
Sad, but forced. Clearly, Bobby has lost the strategic fight. Now all White
has to do is occupy some of the holes, when Black's game will implode
positionally and/or tactically.

29.NdS Qf7 3 0.Re l ReS 3 1 .Rde3 ReS 3 2.Bg3 Rxe3 3 3 .Rxe3 ReS 34.Re6
Black's d6-pawn is lost with or without the exchange of Rooks.

34 ...NgS 3 S.Rxd6 Re4 36.RdS+ Kh7 3 7 .Bf4

Black's position is, of course, quite lost.

3 7 ... Bh6 3 S .Rd7

91

PA L B E N KO: M Y L I F E , GAM ES A N D C O M POSITI O N S

This wins the black Queen since 38 . . . Qxd7 allows 3 9.Nxf6+.

38 ... Re l +
92

Now a beginner would see that 3 9.Kh2 ends the game immediately. How
ever, I was in my usual time pressure.

39.Kf2 Ne4+
This defends the f6-square and allows Fischer to save his Queen. Fortu
nately for me, it doesn't change the result of the game.

40.Kxe l Qxd7 4 1 . Qg6+, 1 -0.

In this game we saw the effect an opening surprise can have, even on a
very strong player. Bobby played with his usual optimism, but the accep
tance of several weaknesses wasn't justified. Once his tactical tricks were
parried, his position virtually fell apart.

(26) Benko - Averbach (Portoroz I nterzonal, 1 958)

l .c4 c5 2 .g3 g6 3 .Bg2 Bg7 4.Nc3 Nc6 5.d3
One move of many. Other choices are 5 .e3 , 5 . e4, 5 .NB , and 5 .a3 .

5 ... d6 6.NB e5
The first move that doesn't copycat White's. It weakens the d5 -square,
but gives Black's forces some chances for a kingside attack.

7.a3
At the moment there's not much going on in the center, and Black is the
only one who can even consider kingside action (the space-gaining pawn
on . . . e7-e5 allows Black to expand in that sector with .. .f7-f5 -f4) . Due to
these things, White is pretty much forced to play for queenside expansion
by b2 -b4, gaining space and opening files.

7 ... a6
Instead of going all out on the kingside, Black prepares his own queenside
advance.

8.0-0 Nge7 9.Ne l

The Knight was no longer doing anything on B . By jumping to e 1 , it frees
the g2 -Bishop's diagonal and prepares to hop into c2 where it helps with
b2 -b4 and also envisions ganging up on the d5-square via Ne3 .

9 ... Rb8 1 0.Nc2

Normally White throws in Rb I when preparing for b2 -b4. However,
I felt that it wasn't necessary in this position. Besides, in many lines the
a-file will be opened when the Rook stands far better on a 1 .

1 0 ... b5 1 1 .b4!

T H E N EW WORLD

93

White can't allow Black to take the initiative on the queenside, so he lashes
out in an effort to wrest control of that sector from his opponent. The fact
that Black hasn't castled also helps White's cause, since any opening up of
the position could easily place the enemy monarch in extreme danger.

1 1 ...cxb4
The main alternative was 1 1 .. . e4, when I intended to sacrifice a piece by
1 2 .Nxe4! fS (much too risky is 1 2 . . . Bxa l 1 3 .Nxa l fS 1 4.Bb2 , when Black
is horribly weak on the dark-squares) 1 3 .Nxd6+ Qxd6 1 4.Bf4 BeS I S.bxcS
Qd7 1 6.BxeS NxeS 1 7 .cxbS RxbS 1 8 .d4, leaving White with three strong
pawns and the initiative for the Knight.

1 2 .Nxb4 Nxb4 13.axb4 0-0

Black doesn't dare accept the pawn sacrifice by 1 3 . . . bxc4 14.dxc4 Rxb4,
because White gets a dangerous attack after I S .Ne4 Rb6 1 6.Ba3 .

1 4.cxb5 axb5 1 5.Ra5

White's control of the a-file, pressure
against bS, and play against the weakened
dS-square combine to give me a clear ad
vantage.

1 5 ... Bd7 1 6.Bg5 ! h6

Allowing me to trade off Black's dark
squared Bishop for the Knight can't be
good, since dS falls into my hands. If I
were B lack, I would have preferred
16 . . .f6, though White stands better af
ter 1 7 .Qb3 + Kh8 1 8.Bd2.

ri

Averbach: I played Yuri three

t imes, winning once and making draws

in the other two games. We always got
along very well and I l iked h im qu ite a
b it . We h a d the s a m e int e r e s t i n
endgames, he wasn't crazy about the
communists, he enjoyed the company
of beautiful women, and he had a pleas
ant personal ity.

1 7.Bxe7 Qxe7 lS.Qb3 RfeS 19.Bd5

White has an iron grip on the position: all his pieces are more active than
their Black counterparts, and Black is stuck defending weaknesses on b S ,
d 6 , and d S .

1 9 . . . BeS 20.Ne4
Intending to attack d6 via Ra6.

PA L B E N KO : MY L I F E , GAM ES A N D C O M POSITI O N S

20 ... Kh8

94

Black is finally ready to give himself a bit of freedom by . . . f7 -f5 , but he's
never quite able to do it.

2 1 .Ra6 Rd8 22 .Bc6!

Now 2 2 . . . f5 fails to 2 3 .Bxe8 ! .

22 . . .Bxc6 2 3 .Rxc6 dS
Black has managed to chase White's pieces off of d5 and e4, but now other
squares becomes vulnerable.

24.NcS e4 2 S.e3 exd3 26.Qxd3

White is still very much in charge: his Knight is stronger than Black's Bishop,
and the enemy pawns on b5 and d5 are both targets.

26 ... QeS
If Black had tried to dissolve the isolated d-pawn by 2 6 ...d4, White would
have played 2 7.e4, when the kingside majority is more dynamic than the
blocked black pawn. You should also notice the fact that (after 2 6 ... d4 2 7.e4)
Black's Bishop is blocked and no longer controls the a I -square. This al
lows the other white Rook to grab that file with Ra l .

2 7.Na6
Low on time, I throw in a few harmless repetitions to kill off a few moves.

2 7 ... Ra8 2 8.NcS Rab8 29.Rdl Kg8 30.Na6 Rb7 3 1 .Nc5 Rbb8 3 2 .Na6
Rb7
My repetitions have gotten us closer to the 40th move and a new time
control. Feeling less hurried, I decide that it's now time to cash in and win
a pawn.

3 3 .RcS Qd6
No better are 3 3 . . . Qf6 3 4.Nc7, or 3 3 . . . Ra7 3 4.Qxb 5 .

34.Rxd5 QxdS 3 S.QxdS RxdS 3 6.RxdS

Though a pawn down, Black has apparently decided that this endgame
gives him some hope for a successful defense. Both of us were considered
to be endgame experts, so I was quite pleased to accept the challenge and
prove that Black is, indeed, lost.

T H E N EW WORLD

3 6 ... BfS
Black's dream is to trade the b-pawns and the minor pieces and play a
Rook ending with four vs. three on the kingside. Such an endgame is usu
ally drawn. Dreams are one thing, though, and reality is quite another. I
would never allow such a situation to come to pass.

37.NeS Ra7
After 3 7 . . . Bxc5 3 8.Rxc5, White's pawn plus and active Rook should lead
to a win.

3S.Nd3 Ra1 + 3 9.Kg2 Rdl 40.KB Rd2 4 1 .e4 hS 42 .eS h4

Another pawn is about to fall so Black tries one last idea: if he can trade as
many pawns as possible, he might be able to sacrifice a piece for White's
remaining pawns.

43.gxh4 Rdl 44.Ke2 Rhl 4S.RxbS Rxh2 46.RbS Rxh4 47.bS Kg7 4S.b6
Rhl 49.Rb7 Rbi SO.f4
I rejected the win of a piece by 50.e6 KgS 5 1 .e7 (or 5 1 .exf7+ Kg7 , and the
white Rook is tied up) 5 1 .. .Bxe 7 52 .Rxe 7, since Black would have two pawns
against one on the kingside- a difficult technical task that I wanted to
avoid.

SO . . . KgS S 1 . Kd2 B a 3 S 2 . Ke2 Rb S S 3 . Kc 3 B fS S 4 . Ke4 Rbi

S S.RbS gS
Black keeps trying to trade pawns. On 5 5 . . . Kg7 , I intended 5 6.RxfS ! KxfS
5 7 .Nb4, with a winning game.

S6.fxgS Kg7 S7.Rxf8 Rxb6

Also winning for White is 5 7 . . . KxfS 5 8.Nb4 Rc 1 + 59.Kb5 Rc8 60.Nc6.

SS.RdS Rg6 S9.Rd7 Kf8 60.KdS RxgS

The rest of the game is simple technique.

6 1 .Ke4 Rgi 62 .Nf4 Re l + 63.KfS Rf1 64.Ra7 KeS 6S.Rb7 Kf8 66.KgS
KeS 67.NdS RgI + 6S.Kf4 RfI+ 69.Ke4 Kf8 70.Ne3 Rf2 7 1 .NfS KeS
72 .Nd6+ Kf8 73 .Re7 Rf1 74.KdS Rdl + 7S.Ke6 Re1 + 76.Kb7 Rxe7+
Desperate, but 76 ... Rb l + 7 7 .KcS wouldn't change anything after 7 7 ... Rfl
7S.Rxf7+ Rxf7 79.Nxf7 Kxf7 SO.Kd7, and the pawn will promote.

77."Kxe7 Ke7 7S.Nxf7 ! , 1 -0.

At the end of the tournament, I shared third and fourth places with the future
World Champion, Tigran Petrosian. Bobby shared fifth and sixth with Olaffson.
We both qualified for the Candidates Tournament, and we also earned the grand
master title. Bobby was the youngest player ever to achieve this lofty goal, a
record that wasn't broken until Judit Polgar, Etienne Bacrot, Peter Leko, and
Ruslan Ponomariov appeared in the 1 990s.

95

PAL B E N KO : M Y LI FE, GAM ES A N D CO M PO S I TI O N S

(27) R Weinstein - Benko (U.S. Championship, 1 95 9)

96

l.d4 Nf6 2.e4 g6 3.g3 Bg7 4.Bg2 0-0 S.NO e6 6.0-0 dS

Choosing to play a Griinfeld Defense. Often the choice between this and
a King's Indian (via 6... d6) is a matter of taste or mood.
7.Nbd2

Quite a reasonable system, though it doesn't offer any real chance for a
meaningful advantage. White also tries 7.Qb3 and 7.b3 from time to time,
but the modern way of playing for a safe (but very small) edge is 7.cxd5
cxd5 8.Nc3.
7...BfS

An excellent alternative is 7... a5 intending to meet 8.b3 with 8... a4 or

8... Ne4. However, to me 7... Bf5 has always seemed more to the point.
S.b3 Ne4 9.Bb2 as

Also fully acceptable is 9... Nd7.

10.Nh4 Nxd2 11.Qxd2 Be6

This is more accurate than 11...Bc8, as seen in Spassky-Najdorf, Santa

Monica 1966. Now the threat against c4 limits White's options.
12.Rfcl

Sharper is 12.e4, sacrificing a pawn. In Stein-Benko, Stockholm 1952, I

refused to accept the sacrifice and got adequate play after 12... dxe4 B.Bxe4
Nd7 14.f4 Bh3 15.Rel a4. The game eventually ended in a draw. Later it
became apparent that the sacrifice wasn't quite sound: 12.e4 dxc4! B.d5
Bxb2 14.Qxb2 cxd5 15.Rfdl, and now 15...Qb6, 15...Qc7 and even 15...cxb3
are all in Black's favor.
12...Nd7 13.e4 dxe4 14.Bxe4 a4 15.Ng2

After this White soon runs into trouble. Safer was 15.Qc2, though White
can't expect much out of the opening.
IS...NeS! 16.Be2 BfS!

A strong move that gives my Knight access to e4.

17.BxfS gxfS IS.Nf4

T H E N EW WORLD

White should still try 18.Qc2, though he would stand worse after 18... a3
19.Bc3 Ne4.

I S ...a3

19.Bxa3 Ne4

Far stronger than 19... Rxa3 20.Qb4.

20.Qb4 Qxd4 21.c5

ymond Weinstein:

This young International Master was considered very tal

ented, but something went horribly wrong and he committed a murder. Diagnosed as
being mentally ill, in 1964 he was institutionalized in the Kirby Forensic Psychiatric
Center in New York City, where he has been ever since.
I played Raymond in a couple of u.s. Championships. I remember one game in
particular: There was a defective clock and my flag fell four minutes early! A friend of
his who was watching pointed to the clock and told him that I had forfeited on time.
Naturally, it was clear that the clock was badly defective so the game continued and I
won. I forgot about this episode and concentrated on my other games. In the last round
I was playing Reshevsky and someone told me that Weinstein had protested my vic
tory over him, saying that he had been disturbed during the clock incident and, as a
result, wanted to return to that point and replay the game !
Naturally, I was incredulous. Nobody had told me about this, and it seemed insane
to consider it in any serious way during the last round ! Nevertheless, in the middle of
my game with Reshevsky, I was officially informed that Weinstein and I would have to
play the game again from the point of the disturbance, even though we had both signed
the scoresheets.
Now I was upset. I no longer knew how
many points I had and, due to that, if I
should play for a draw or a win against
S a m my. Extremely agitate d , I gave
Reshevsky a draw from a good position,
then played Weinstein again and, once
again, won!
Weinstein was, for the most part, very
nice. He was polite and quiet, but on oc
casion he did things that were a bit odd. I
remember Edmondson telling me in his
hotel room, "Something's wrong with this
Weinstein. "
"What? "
"He was here before you. W hen he
came in he left the door open. I asked him
to close it but he refused and said, 'I can't.
I have to see if anyone is following me.'"

Weinstein and his mother

(Photo Dr. R. Cantwell.)

97

PAL B E N KO : MY LI FE, GAM ES A N D C O M POSITI O N S

Black wins a piece after 2 1 .Rfl c5, though 2 1 ...Nxf2 is also devastating.

2 1 ...Qxf2+ 2 2 .Kh l QB+ 2 3 .Ng2

98

Another possibility was 2 3 .Kg l Rxa3 24.Qxa3 Bd4 mate.

23 ... Bxal 24.Rxal Rfd8, 0- 1 .

I n such positions it's best t o resign and get away from the board a s fast as
possible!

(28) Benko

F Olafsson (Candidates Tournament, 1 95 9)

l .e4 c5 2 .NB d6 3 .d4 cxd4 4.Nxd4 Nf6 5.Nc3 a6 6.Bg5 e6 7.f4 Be7
8.QB Qc7 9.0-0-0 Nbd7 1 0.f5
The usual moves here are 1 0.g4 or 1 0.Bd3. I wanted to avoid main-line
book analysis so I tried out a new idea.

10 ... e5 I 1 .Nde2 b5 1 2 .Ng3 b4

More cautious was 12 ...Bb7.

1 3 .Bxf6 Nxf6 1 4.Nd5 Nxd5 15.exd5 0-0 1 6.Ne4

My plan has worked perfectly! I've traded dreams of a Knight on d5 for

the reality of a dominating Knight on e4. Now, with a space advantage in
the center and on the kingside, and secure in the knowledge that my supe
rior minor piece will give me good chances in any endgame, I can calmly
train my sights on the enemy King.

1 6 . . .f6 1 7 .h4 Bd7 1 8.g4 as

Black's counterattack on the queenside has to be taken seriously, but I was
sure that, with a little care, my own plan would ultimately prevail.

19.Kb l a4 20.Bd3 Ra5

More natural is 2 0...b3. Black has to get White's attention on the queenside
or White will run his opponent over on the opposite wing.

2 1 .g5 Kh8 22 .gxf6 Bxf6 2 3 .Ng5

The play now becomes very sharp. I was still quite confident since, aside
from tempting Ne6 possibilities, I threaten 24.Qh5 h6 2 5.Qg6, winning.

T H E N EW WORLD

However, chess isn't a one-man game and Olafsson, a strong and experi
enced grandmaster who wasn't going to go down without a fight, found
an interesting way to gain some counterplay.
99

2 3 ... e4!
Opening up the f6-Bishop's diagonal. Its ability to take part in the queenside
attack far outweighs the loss of a pawn.

24.Bxe4
Now 24.QhS no longer works: 24 . . . Bxg5 2 5 .hxg5
Bxf5 2 6.g6 h6.

24 ... Be8
S topping Qh5 and ending White's attack. Of
course, there's a price: Black will be down an Ex
change and a pawn. This means that his survival
will depend on his queenside attack - if it doesn't
succeed, Black will have to give up.

2 S.Ne6 Qe7 26.NxfB QxfB

2 6 . . . Qe5 fails to 2 7 .c3 .

2 7.Rd3 ReS 28.Re3 Bd4 29.Rhe l !

Giving back the Exchange is the best solution, since
it forces Black to part with his powerful dark
squared Bishop. Far weaker was 29.Rd3 Qf6.

29 ... Bxe3

Olafsson was the only

other player at the
Candidates Tou rnament
whose time trouble was
as bad as m ine!
(Photo courtesy USCF.)

Now 2 9 . . . Qf6 10ses to 3 0.Bd3 .

3 0.Qxe3 a3 3 1 .Bd3 Bf7 3 2 . Qe7 !

'White makes use of Black's back-rank problems to force a winning endgame.

Naturally, the exchange of Queens ends the black attack once and for all.

32 ... Re8 H .QxfB+ Rxf8 34.Re4

Showing that Black's attacking pawns are now endgame weaknesses.

34 ... Rb8 3 S.Rd4 Re8 36.c3 Re l + 37.Kc2 bxc3 3 8.bxc3 Rhi 39.Kb3 Kg8
40.Kxa3 Kf8

PA L B E N KO : MY L I F E , GAM ES A N D CO M PO S I TI O N S

Black, who only had seconds left, didn't have time t o resign!

4 1 .Kb4 Ke7 42.a4 KdS, 1-0.

1 00

Black gave up without waiting to see my reply.

(29) Benko - Smyslov (Zagreb Candidates Tou rnament, 1 95 9)

l .e4 cS 2 .NO e6 3 .d4 cxd4 4.Nxd4 a6 S .c4 Nf6 6.Bd3
This avoids 6.Nc3 Bb4, but I don't think it causes Black much trouble.

6 ... Nc6 7.Nxc6 dxc6 S.O-O eS 9.Qc2 BcS

A glance might convince the reader that Black already has the better game:
White has a hole on d4, his light-squared Bishop is horrible, and his pieces
lack active posts. However, I now employ a plan that changes the land
scape on the queenside and gives White the upper hand.

1 0.Nd2 Be6
It's possible that 10... Qe7 was more exact, since then Il.Nb3 Ba7 12.c5
could be answered by 12 ... a5.

I l .Nb3 Ba7 1 2 .cS

Suddenly White has more queenside space, the "horrible" light-squared

Bishop is showing signs of activity, Black's once mighty dark-squared Bishop
is now entombed within its own position, and the squares on a5, c4, and
d6 have all been targeted for White domination.

1 2 ... aS
Instead, 12... Qe7 13.Bc4 secures the c5-pawn.

1 3 .Bd2 ! a4 1 4.NaS Qe7 I S .b4! axb3 1 6.axb3 0-0

Black lost the battle over the c5-pawn since 16... Bxc5 17.Nxc6 Rxa1 18.Nxe7
Rxfl+ 19.Kxfl Bxe7 20.Bb5+ is winning for White.

1 7.b4 RadS I S .Bc3 BbS 19.Rfdl NhS

White's plan has been highly successful and Black clearly stands worse on
the queenside. Therefore, he turns his attention to the other side of the
board.

T H E N EW WORLD

2 0 . g3 BeS 2 1 . Ne4 Q g5 2 2 . Bf l B g4 2 3 .RxdS RxdS 2 4.RaS Rf S

25.Qb2 Bf3
If 25 .. .6, I intended 26.Nxe5! (The tournament book gave the more mun
dane 26.b5 cxb5 27.Qxb5 Qcl 2S.Bb2 Qbl 29.Qxb7 as winning.) 26 ... fxe5

101

27.Bxe5, because Black can't play 27... Bxe5 due to 2S.Qa2+

and mates.

26.Bxe5
I decided to play for the win of a pawn and a very favor
able endgame. However, if I had known that it was go
ing to be so difficult, I' d have tried 26.Nd2 Bg4 27.b5.

26 ... Bxe5 27.Qxe5 Qxe5 2S.Rxf8+ Kxf8 29.Nxe5 Bxe4

30.Ne4! Nf6 3 1 .Nd6 Bf3 3 2.Be4 Nd5 H .Bxd5 Bxd5
34.Nxb7
Though White has won a pawn and also seems to enjoy
the superior minor piece, the win proves to be a long
way off.

3 4 ... Ke7 35.f4 Ke6 36.Kf2 f6 3 7.Ke3 Bh l 3 S.Nd6 Bg2

3S... Kd5 39.NeS is much easier for White.

3 9 . Kd4 Bh 3 40.Ne4 B g2 4 1 .Nd6 Bf l 42 . Kc 3 g5

43 .Kd4 gxf4 44.gxf4 Ba6

Vassily Smyslov i n 1 958.

(Photo L. Velikzhan, courtesy
USeE)

The game was adjourned here and after analyzing at leisure, I realized the
win would take quite a long time. If victory is to be achieved, the white
King will have to find a way to penetrate. Unfortunately, Black's King and
Bishop make this difficult. As a result, I decided that it was time for a bit
of cat and mouse: I would patiently maneuver, hoping to make Smyslov
less alert. I knew that such long-term torture from a passive position would
be very hard for Black to deal with, so I sat down and resumed the game
ready for a lengthy siege.

45.f5+
The first order of business is to push back the black King. It isn't optimal
to place my pawn on a vulnerable white square, but I had no choice in the
matter.

PA L B E N KO : MY L I F E , GAM ES A N D CO M PO S I TI O N S

45 ... Kd7 46.Ne4 Bb5 47.Kc3 Ba6 48.Kb3 Bb5 49.Nd6 Be2 50.Kc3

1 02

Continuing my cat-and-mouse strategy, but more straightforward was

5 0.Ka4 Bd3 5 1 .Ka5 Kc7 5 2 .h4 h6 5 3 .h5 Be2 (one pretty line is 53 . . . Kd7
54.Kb6 Bc2 5 5 .Kb7 Bd3 5 6.Nf7 Bxf5 5 7 .Nxh6 Be6 5 S .Nf7 ! Bxf7 59.h6
Bg6 60.b5, and White wins) 54.NeS+ KdS 5 5 .Nxf6, winning.

50 ... Ke7 5 1 .Ne4 Kd7 52 .Ne3 Ke7 53 .Ne2 Kd7 54.Nd4 Bft 55.Kd2 Be4
56.Ke3 Bf7 57.Kf4 Be4 58.Kg4 Bf7 59.Kh4 Ke7?
Black finally cracks and commits a losing error. He should have played
59 . . . BeS, but Smyslov assumed that, after 59 . . . Kc7 , I wouldn't be able to
win the h-pawn because my Knight gets trapped.

60.Ne6+ Kd7
60 . . . KcS 6 1 .NfS BgS 62 .Kh5 is easy for White.

6 1 .Nf8 + ! Ke7 62 .Nxh7 Bg8 63 .Ng5 ! fxg5+ 64.Kxg5

Black did, indeed, win my Knight, but now my pawns are unstoppable.

64 ... Kf7 65.h4 Kg7 66.f6+ Kf8 67.h5 Be4 68.Kf4 Kf7 69.Ke5 Bd3 70.h6
Black's game is hopeless. He has to worry about the h-pawn, the f-pawn,
and the c-pawn (after a b4-b5 break).

70 ... Kg6 7 1 .Ke6 Be4+ 72.Ke7 Bd5 73 .f7 Bxf7 74.h7, 1 -0.

(30) F Olafsson - Benko ( Candidates Tou rnament, 1 95 9)

l .e4 e5 2 .NB Ne6 3 .Bb5 a6 4.Ba4 Nf6 5.0-0 Be7 6.Re 1 b5 7.Bb3 0-0
8.c3 d6 9.h3 Bb7
If Black wants to play the Breyer Variation, the correct order is 9 . . . NbS
1 O .d4 Nbd7.

1 0.d4 Nb8
As stated in my note to Black's 9th move, this isn't the best way to get into

the Breyer Variation! I played this way because I wanted to avoid certain
bothersome systems that White could play after 9 . . . NbS , but Keres, in a
later round, cured me of this desire.

1 l .Nbd2

In 1958, Botvi n n i k beat

Smyslov and regained the
title of World Champion.
Here we see him
overwhelmed by fans
and autograph seekers.
(Photo courtesy USCF.)

7Ut'l1i,.l!Rfld,'IIQ
""

T he players in the Candidates Tournament were vying to chal lenge Botvinnik.

This is where Keres demonstrated the antidote: I I .dxe5 dxe5 (Also bad is
1 1 . . .Nxe4 1 2 . e6 fxe6 1 3 .Bxe6+ Kh8 1 4.Bd5, as pointed out by Gligoric.)
1 2 .Qxd8 Bxd8 1 3 .Nxe5 Nxe4 1 4.Be3 Bf6 1 5 .Ng4, and Black's position is
quite unpleasant.

1 1 ...Nbd7
Now we're back in regular Breyer lines.

1 2 .Bc2 ReS B .Nfl Bf8 14.Ng3 g6 I S .h4? !

White tries to break up Black's kingside with this over-optimistic pawn
move, but all he ends up doing is weakening his own kingside structure.
Better was 1 5 .a4, when White retains chances of coming out of the open
ing with a slight edge.

I S ... dS !

Following the tried and true rule: The best reaction to an attack on the
wing is a counterattack in the center.

PAL B E N KO : MY L I F E , GAM ES AN D C O M POSITI O N S

1 6.dxe5
VVh.ite gets less than nothing from 16.Bg5 h6 17.Bxf6 Qxf6 18.exd5 exd4
1 04

19.Ne4 Qf5! 20.Nxd4 Qxd5.

1 6 ... dxe4 1 7.Nxe4 Nxe4 1 8.Bxe4 Bxe4 19.Rxe4 Nxe5 20.Bg5 Qxd 1 +
2 1 .Rxdl Nxf3+ 2 2 .gxf3 Bd6 2 3 .Rde l Rxe4 24.fxe4
VVh.ite has managed to undouble his f-pawns, but his inferior kingside pawn
structure will still cause him problems for a long time to come.

24 ... Re8 25.f3 f5 !

Nibbling at e4 and allowing Black's King to rush into play via t7 and e6.

26.m Kf7 27.Ke3 fxe4 2 8.fxe4 Ke6

Black's advantage lies in his ability to create an outside passed pawn on

the kingside. This means that any King and pawn endgame will be a clear
win, while a Bishop endgame will also offer Black good chances of suc
cess. VVh.ite's best chance to draw is to exchange Bishops and enter a pure
Rook endgame. Unfortunately, he isn't able to achieve this.

29.h5
Black would have answered 29.Bf4 by 29...Be7. The text gets the pawn off
the vulnerable dark square and prepares Bf4.

29 ... Rf8 !
Stopping VVh.ite's Bf4 idea once and for all.

30.hxg6 hxg6 3 1 .Kd3 Rf3+ 3 2.Re3 Rf l H .Rh3 Rd1 +

Black drives VVh.ite's King further away from his passed pawn.

34.Kc2 Rgl 35.Bd2 g5

It's now clear that the passed pawn is very dangerous.

3 6.Rh6+ Kf7
Not letting VVh.ite's Rook get behind the passed pawn via Rg6.

3 7.Kd3
According to the tournament book, "It was a must to control the g5-pawn
by 37.Be3 Rg2+ 38.Kd3 g4 39.Bd4, and VVh.ite has activated his Bishop."

TH E N EW WORLD

37 ... Rg3 + 3 8 .Ke2 Rg2+ 39.Kd3 g4 40.eS

White decides to trade as many pawns as possible, usually a good defen
sive idea in inferior endgames.

40 ... BxeS 4 1 .Rxa6 g3

Threatening to end the game immediately with 42 ... Rxd2 + 43 .Kxd2 g2 .

42 .Be3 Rxb2 43 .a4

White's plan is now clear: he wants to trade off Black's final two queenside
pawns and sacrifice his Bishop for Black's passed g-pawn. The resulting B+R
vs. R endgame would be a draw. Of course, I didn't allow this to occur.

43 ... b4!
This prevents White's Rook from getting back to the kingside.

44.cxb4 g2 4S.Ke4 Re2 46.KB Rxe 3 + 47.Kxg2 Rb3

Black won White's Bishop and still has a pawn on the board. This means
that the win is a simple matter of technique.

48.bS Bd6 49.Kf2 Ke6

All Black has to do is guard c7, stop White from trading pawns by b5 -b6,
and bring his King over to the queenside. Once he achieves these goals,
the game will be over.

50.Ke2 KdS S 1 .Ra8 Kc4 S2.Ra6 Bc5 S 3 .Rc6 Re3 + 54.Kd2 Re7, 0- 1 .
Black has firmly defended his final pawn and stopped White from exchang
ing it. Now nothing can be done about . . . Kc4-b4 and . . . Bb6 when both of
White's pawns will be devoured.

I was now working full time and playing in tournaments whenever the opportu
nity arose. In general, I had excellent results, though it was clear to me that my
best years had been lost during my struggles to escape from Hungary.
My main event in this time period was the international tournament in Buenos
Aires, and it's here that I gained my second victory over Fischer. By that time,
Bobby was a shining star. The fact that, three months earlier in Mar del Plata, he
had tied (with future World Champion Boris Spassky) for first with the incredible
score of 1 3 1/2 points from a possible 1 5 , proved that he was head and shoulders
over other players of his age. I could tell he enjoyed traveling to far away places
and that Argentina excited him. In fact, he wanted to play day and night.
One morning (at 4A.M.to be exact!) I was woken up when someone relentlessly
pounded on my door. There stood Fischer and Larry Evans, still up and run
ning after exploring the city throughout the night. They wanted to talk-it seemed
that, for them, sleep was out of the question. Naturally, I didn't take kindly to
this blindly selfish behavior, and I blasted them for acting in such a manner. I

1 05

PAL B E N KO: MY LI FE. GAM ES A N D CO M PO S I TI O N S

1 06

was scheduled to play Reshevsky in four hours time. He was a very strict Jew
and, in consideration for the Sabbath, requested an unusually early start. \Vhen
1 explained this to Fischer, he said, "I would never agree to such a change in
schedule ! " Apologies were never one of his strong points. It's interesting to note
that Fischer eventually became religious himself. As a result, he also refused to
play on the Sabbath, and tournament organizers had to create extra time slots
for him, thus forcing others to change their schedules to conform to his needs.
All in all, the nightly adventures in Buenos Aires took their toll on the young
grandmaster, and he had the worst performance of his international career: ty
ing for thirteenth to fifteenth places with a minus score. This tournament was
particularly grueling. In a field of twenty, fourteen grandmasters and several strong
Argentinean players were present (two of them earned their GM title at the fin
ish). Bobby and 1 were paired in the final round, with me having the white pieces.
By that time, frayed nerves, exhaustion, and anxiety were common among the
participants. Naturally, such extra tension and pressure often causes costly mis
takes. Perhaps, as Botvinnik once suggested, last rounds should be eliminated!
The game against Fischer was my final chance to improve on my rather me
diocre score. 1 had been spending too much time away from the board with a
good-looking local Hungarian girl and, like Fischer, Evans, and many others,
forgot that 1 was there to play in a chess tournament! 1 must say that Bobby also
liked her and asked if it would be okay if he dated her after the tournament was
over. One night, 1 decided to have some fun with Bobby and said (in jest, of
course), "Look kid, I'll arrange it, providing that 1 can have the point! " \Vhen
we finally sat down to play, 1 told Bobby that 1 was going to beat him without
getting him the date. With some luck, I managed to make good on this promise,
but soon rumors got back to me suggesting that, while Bobby had honored our
deal, 1 had failed to come through!
Naturally, such light-hearted teasing is a normal part of tournament life. No
body took it seriously because it's well known that Bobby dislikes draws and never
agrees to prearranged results of any kind. His vehement stance on the illegality
of Soviet deals among their players shows his moral position on this issue. Per
sonally, I see nothing wrong with quick "grandmaster draws. " This kind of non
game allows both contestants to take a day off in the midst of an exhausting
schedule. However, losing on purpose is indeed deplorable.
In the actual game, I once again tried a new opening idea, hoping to steer play
away from memorized schemes and into more intuitive channels. Unfortunately,
in this instance my preparation was badly flawed.

(3 1 ) Benko - Fischer (Buenos Ai res, 1 960)

l .d4 Nf6 2 .c4 g6 3 .g3 Bg7 4.Bg2 0-0 S .Nc3 d6 6.NB Nc6
This is a well-known line of the KID (known as the Panno Variation), and
Bobby's choice brought back a lot of memories for me. In the 1952 Maroczy
Memorial tournament in Budapest, I used this same move as Black against

Fischer (left) with Tal and

Petrosian at the Candidates
Tournament, 1 959.
(Photo courtesy USeF.)

the German Master Platz. After 7.d5, I created quite a stir with 7 . . . Na5 , a
virtually unheard of idea at that time since 7 . . . N e5 or 7 . . . NbS were consid
ered to be the only playable choices. Now it's common knowledge that the
Knight's placement on the edge of the board (i.e., after . . . Na5) gives rise to
many interesting possibilities. Here's the Platz-Benko game from Budapest
1 95 2 : 6 . . . Nc6 7.d5 Na5 S.Qd3 c5 9.0-0 a6 1 0.Rb l Qc7 l 1 .b3 Bd7 l 2 .Bd2
RfbS 1 3 .a4 b5 1 4.axb5 axb5 l 5 .cxb5 Nxb3 l 6.Rxb3 c4 l 7.Qb l cxb3 l S.Qxb3
Qc5 1 9 .Rb l Ra3 2 0.Qb4 Qxb4 2 1 .Rxb4 Rxc3 2 2 .Bxc3 Nxd5 , 0- 1 .

7.h3
At that time, 7.d5 and 7 .0-0 were the accepted moves. This ill-fated idea
of mine prevents . . . Bg4 and also allows me to leave my pawn on e2 and
play Be3 without having to fear . . . Ng4. I expected normal "Panno" play
from Black via 7 . . . a6 and S . . . RbS . Instead, Bobby began active operations
in the center, quickly exploiting my loss of time.

7 ... eS
Eight years later, Korchnoi against Ciocaltea in the Lugano tournament
refined my idea by castling on the 7th move. Then, after 7 . . . a6 (commit
ting to the Panno idea of queenside expansion), he made use of my plan
with S .h3 RbS 9.Be3 b5 1 O.Nd2 . Korchnoi continued to utilize this idea
with great success, and soon other grandmasters came to realize its strength.
Now 7 .0-0 a6 S.h3 is known to be one of White's most promising choices
against the Panno Variation.

8.0-0
Hastily abandoning my original idea of S.Be3 , since I didn't like the look
of S . . . ReS 9.0-0 exd4 1 0.Nxd4 Rxe 3 , when Black has good play for the
sacrificed Exchange.

8 ... exd4 9.Nxd4 Nxd4 1O.Qxd4 Be6

Now White must lose time avoiding a discovered attack on his Queen by
the g7-Bishop. Black already has the advantage.

1 1 .Qh4
Threatening both Bh6 and Bg5 . Worse was l 1 .Qd3 Nd7 , when . . . Nc5 or
. . . Ne5 would prove very annoying for White.

1 l Nd7!
..

PAL B E N KO : MY L I F E , GAM ES A N D CO M POSITI O N S

Simple and strong. Black opens the g7-a l diagonal, places his Knight on a
flexible square, and offers an exchange of Queens. White can't comply, since
such a trade would make it hard for him to defend his queenside pawns.
1 08

1 2.Bg5 f6 1 3 .Be3
White has temporarily shut off the g7 -Bishop's long diagonal, while plac
ing his own dark-squared Bishop on an active square. This sounds pretty
good, but Fischer's reply feeds me a dose of reality. Instead, seeking to
trade Bishops by 1 3 .Bh6 would have been a mistake, because after 1 3 . . . g5
1 4.Qh5 Bxc4 1 5 .Bxb7 Rb8 1 6.Bxg7 Kxg7 1 7 .Be4 f5 1 8.Bg2 Rxb2 , White
would lose a pawn, while the weakening of Black's King position wouldn't
offer enough compensation.

1 3 ... g5 !

This move has been adorned in many publications with a question mark,
though no alternative was ever suggested. The truth of the matter is quite
the reverse: this fine advance is the only way for Black to prevent White
from consolidating his position.

14.Qd4 f5 15.Qd2
Hoping to gain some time by attacking the g-pawn.

15 ...f4!
The remorseless Fischer continues to play with his customary energy. Notice
how he hasn't given me a chance to recover from my opening transgres
SIOns.

16.gxf4 gxf4 1 7.Bxf4

The pawn had to be accepted since 1 7 .Bd4 Ne5 1 8 .Bxe5 Bxe5 followed by
. . . Qh4, leaves White with a lost game.

1 7 ... Nb6
The point of his maneuver. Now 1 8.b3 is strongly met by both 18 . . . Rxf4
and 1 8 . . . Qf6, in both cases with a black advantage.

1 8.Qe3
This places my Queen on a better square and "suggests" that he should
capture on c4 with his Bishop.

T H E N EW WORLD

1 8 ... Qf6
Creating a variety of threats. Worse was 1 8 . . . Qd7 1 9.c5 ! dxc5 2 0.Qg3 ,
when White is better.

19.BgS Qg6
By keeping the Bishop under a watchful eye, White is still prevented from
playing 20.b3 due to 20 . . . Bxc3 .

20.Qg3 Nxc4?
A bad move that was hastily played (no doubt a result of too many late Buenos
Aires nights). Instead, this capture should have been prepared by 20 . . . c6,
when Black will regain the sacrificed pawn with an excellent position.

2 1 .NdS

Turning the tables. This strong piece threatens both Nxc7 and Ne7+, and
can't be captured because 2 1 . . .Bxd5 22 .Bxd5 + picks up the enemy Knight.

2 1 . .. Qf7
Played immediately, as ifhe expected 20.Nd5 . However, White would have
been better even after 2 1 . . .Rf7 2 2 .Nf4 Qf5 2 3 .Bxb7 ! Rb8 24.e4.

22.Bh6
A powerful move that pins the g7-Bishop and threatens Bxg7 followed by
Nxc7.

22 ... c6
As good as anything else, even though it loses the Exchange. There was

no good way to defend c 7 , as 2 2 . . . Rac 8 ? 2 3 . Bxg7 Qxg7 2 4. Ne 7 +

demonstrates.

2 3 .Bxg7 Qxg7 24.Qxg7+ Kxg7 2 S.Nc7 Kf6 26.Nxa8 Rxa8 2 7.b3 , 1 -0.
Bobby held out his hand in resignation. This was a bit of a surprise to me,
since I had expected him to put up a long, tough, defense with 27 . .. Na3 .
In the end, though, I was positive that I was going to reel in the full point
(and why should Bobby spend his time defending a losing cause when, in
the tradition of this event, he could hit the streets and have some fun?).
Looking back on the play in this game, my poor opening and Fischer's
mistakes from a superior position give me nothing to be proud of. In fact,

1 09

PAL B E N KO : MY LI FE, GAM ES A N D C O M POSITI O N S

it's more an example of Fischer's active style than any positive features
that I might possess.
1 10

(32) Benko - Eliskases (Buenos Aires, 1 960)

l .c4 g6 2 .d4 Bg7 3 .Nc3 d6 4.NB Nf6 5.e4 0-0 6.Be2 e5 7.0-0 Nc6
S.d5 Ne7
In many ways, this particular position represents the very heart and soul
of the King's Indian Defense. White intends to overrun the queenside with
a quick c4-c5 (prefaced by b2 -b4 or NB -e l -d3) while Black will try to
knock out the white King by . . . Ne8 and then . . .7-f5 -4. I've often been
asked, "Why would White play a line that gives his opponent a very strong
kingside attack? " This is actually an important question! The answer lies
in the importance of long-term and short-term advantages. In the short
term, White might crash through on the queenside and win, or Black might
decapitate White's King. However, the long-term advantage rests with
White, who usually has a significant advantage in most endgames. In a
way, then, we can say that White is being given endgame odds, and you
can imagine how much this would please me and many other players who
believe that their King will never be dragged down! Unfortunately, this
variation has now been analyzed so deeply that certain lines don't allow
original play until the 3 0th move or even much later! Such theoretical
discussions were never to my taste, but at the time this whole King's In
dian terrain was still undiscovered country.

9.Ne 1
Both this and 9.Nd2 were White's main choices for many years. Nowa
days the straightforward 9.b4 is considered promising for White.

9 ... Nd7 1 0.0 f5 1 1 .g4!?

I introduced this move at Portoroz 1 9 5 8 against Pachman. Though it looks

aggressive, I I .g4 is actually a prophylactic move designed to stop Black's
usual attacking plan based on . . . 7-f5-f4 with . . . g6-g5-g4 to follow.

1 1 ...f4

T H E N EW WORLD

I had always considered this move to be a strategic mistake because White

can permanently close the position on the kingside. The first player will
then have a free hand on the other side of the board. This game proves
the validity of my view, though its complete implementation will take time
and patience. It's now known that Black's best and most flexible answer to
my system is 1 1 . . .Kh8 followed by 1 2 . . . Ng8 and . . . Bh6.

I2 .h4!
Now White can close things up in case of 12 ... g5 B .h5 or 12 ... h5 B .g5 .

I 2 ...aS
Black has tried several other 1 2 th moves over the years, but all of them
have verified the impression that the position after 1 2 .h4 is simply better
for White.

1 3 .Ng2 NcS I4.Bd2 Kh8 I S .Bel Bd7 I 6.Bf2

A great square for the Bishop. Here it defends my King and the h4-pawn,
while staring balefully towards the queenside (no doubt dreaming of fu
ture conquests in that area).

I6 ... b6 I 7.Ne l
White's next plan is to kick the c5-Knight off of its strong perch. Of course,
while playing to open queenside lines, I always have to keep an eye on the
kingside. Black can't be allowed to penetrate there, and potential enemy
piece sacrifices must always be taken into account.

I 7 ...Ng8 I 8.Kg2
Preparing to answer 1 8 . . . Bf6 with 1 9.Rh l .

I 8 ... Nf6 I9.Nd3 Qe7 20.Rb l rub8 2 1 .b3

Preparing to chase away the evil c5-Knight by a3 and b3 -b4. Of course,
2 1 .a 3 ? would end my queenside aspirations quickly after 2 1 ... a4.

2 1 . .. Be8 22.Ncl
The immediate 2 2 .a3 would allow Black to mix things up a bit with 2 2 . . . a4
2 3 .b4 Nb3 . My n .Nc 1 prevents this possibility.

22 ... h6 2 3 .a3
Having stopped all enemy counterplay on the kingside and queenside, I
can finally continue with my plan of queenside expansion.

23 ... Nfd7 24.b4 Nb7

I I I

PAL B E N KO : MY L I F E , G A M E S A N D C O M POSITI O N S

The first phase is over. Phase two consists in weakening the light-squares
in Black's camp by exchanging White's bad Bishop (on e2) for Black's good
one (on e8).
1 12

25.Bd3 Nf6 2 6.Be2 Bd7 2 7 .Ba4 Bxa4 2 S .Nxa4 axb4 29.axb4 NdS
30.Nc3 Nfi 3 1 .Nb3 Ra6 32 .Qe2 RbaS
Phase two is complete, and White now trades Rooks so Black can't gener
ate any counterplay on the open a-file.

B .Ra l Rxal 34.Rxa l QdS 3 S .RxaS QxaS 36.NbS NeS 3 7.Qb2 Bf6
3 S.Qal Qxal 39.Nxal
White confidently steps into an endgame, secure in the knowledge that
he will retain good winning chances due to his better Bishop and the ever
threatened c4-c5 pawn break.

3 9 . . . NdS 4 0 . Ne2 Nb7 4 1 .Na7 KgS 42 .Na3 Kf7 4 3 .Ne6 NdS

44.NbS Nxe6 4S.dxe6
The trade didn't help Black because now the white Knight can make use
of the dominating square on d 5 .

4S ... Ke7 46.Nc3 KdS 47.NdS KeS 4S.hS gS

Black has to close the kingside and stop White's King from penetrating
there. Now the only break is by c4-c5, so White's King heads for the
queenside. Of course, a move like Nxf6 wouldn't even be a consideration.
Who in his right mind would exchange a dominating Knight for the "tall
pawn" on f6?

49 . Kfl BdS SO .Ke2 Ng7 S l .Kd3 Ne6 S 2 .Kc3 Kb8 S 3 .Kb 3 Ka7
S4.Ka4 Ka6 SS.Bgl Ng7 S6.Bf2 Ne6
White first makes the time control before breaking through on the
queenside. This is a wise decision. Black can't do anything but helplessly
sit back and wait, so why shouldn't I take a bit of time to put everything in
order? By calmly placing all my pieces on their optimum squares and by
putting any thoughts of time pressure behind me, I leave Black with virtu
ally no chances of survival.

S 7.Ka3 Ng7 SS.Kb3 Ne6 S9.Ka4 Ng7 60.eS

With my time pressure a thing of the past, I was finally ready to make this
long awaited pawn advance!

60 ... bS+
Also losing is 60 ... dxc5 6 1 .bxc5 b 5 + 62 .Kb3 ! Ne6 63 .Kb4 Ng7 (6 3 ... Nd4
64.Bxd4 exd4 65 .e5 is easy for White) 64.Be l Ne6 65 .Bc3 , and Black's
defenses fail. Note that 60 . . . bxc5 6 1 .bxc5 dxc5 62 .Bxc5 forces mate with
Nb4.

6 1 .Kb3 Ne6 62.Kc3

The natural 62 .cxd6 just frees the black Bishop. At the moment that Bishop
is in a cage, and I had no intention of setting it free.

62 ... dxeS 63 .bxcS KaS

T H E N EW WORLD

Not letting my King into b4.

64.Kb3
Black is in Zugzwang. 64 ... Ka6 allows 6S.Kb4, while 64 ... Ng7 fails to 6S .Be l +
followed by Bc3 .

64 ... b4 65.Kc4
And not 6S .Nxb4 KbS when Black frees his forces.

65 ... b3 66.Be l + Ka6

Avoiding 66 . . . Ka4 67.Bb4 with mate next move.

67.Bc3 Nd4 6S.Nh4+ Ka5 69.Nd3 + Ka6

Still avoiding the mate after 69 . . . Ka4 70.Bb4 and 7 1 .Nb2 .

70.Nxe5 Ne2 7 1 .Nf7 Be7

7 1 . . .Nxc3 n .Kxc3 was just as hopeless.

72 .Nxh6 Bxc5 73 .Nf7 Bd4 74.Bxd4 Nxd4 75 .Kc3 Nxf3 76.h6 Nh4
77.Ne5
Not allowing Black to stop my h-pawn by . . . Ng6. Now he has to give his
Knight up on this square if he wants any hope at all.

77 ... Ng6 7S.Nxg6 f3 79.h7 f2 SO.hS=Q f l =Q S l .QaS+

Black will be mated by force.

S l ...Kb5 S2 .Qb7+ Kc5 S3.Qb4+, 1 -0.

A pity! I was looking forward to 8 3 . . .Kxc6 84.NeS mate !

(33) Benko - Taimanov (Buenos Aires, 1 960)

l .c4 e6 2 .g3 d5 3 .Bg2 Nf6 4.Nf3 Nc6
Very strange. Black will need to use his c-pawn (on c6 or c5), so blocking
it is rarely a good idea in Queen-pawn or flank openings. More usual is
4 . . . Be7 .

5.0-0 Be7 6.d4 0-0 7.cxd5 exd5

Taimanov:

I beat Mark in this

game, but he later avenged h imself by beat

ing me on the white side of a Benko Gam
b it, wh ich was one of my very rare losses in
that open ing! He was a talented player, and
moved very qu ickly, almost by instinct. He
was also a terrific p ianist. I heard h im play
p iano many times, and was always amazed
at h is mastery of th is instrument.

Mark Taimanov
in 1 954.
(Photo courtesy
USeF.)

1 13

PAL B E N KO : MY L I F E, GAM ES A N D C O M POSITI O N S

1 14

The trade on d5 has steered the game into a kind of reverse Caro-Kann
position where the Knight on c6 will prove to be misplaced. Yes, this allu
sion to the Caro-Kann might raise some eyebrows, but take a look at l .e4
c6 2 .d4 d5 3 .exd5 cxd5 4.Nc3 (a mistaken move in this formation because
the Knight has no future here) 4 . . . Nf6 5 . Nf3 g6 6 . B e 2 B g 7
7 .0-0 0-0 with a mirror image o f the actual game position!

8.Bg5 h6 9.Bxf6 Bxf6 1 0.Nc3 Ne7 1 1 .Rcl c6 1 2 .h4!

Starting an immediate minority attack. This well-known plan is designed
to create permanent weaknesses in Black's queenside structure.

12 ... a6 1 3 .a4 Nf5 14.h5 Re8? !

I'm sure that Black later regretted his failure to trade by 1 4 . . . axb 5 . That
would have left him with only one weakness (on c6). The disadvantage of
14 . . . axb5 1 5 .axb5 is that it clears away the a4-square for a white Knight.

15.hxc6 hxc6 1 6.e3 h5? !

Taimanov was always an optimist, but this lunge on the kingside is going a
bit too far. More advisable was a defensive strategy based on . . . a6-a5 and
. . . Ba6.

17.Ne2 Qd6 1 8.Nf4 h4 19.94!

This stops Black's kingside attack and forces him to weaken his position
further since any Knight move would allow g5, winning a piece.

19 ...g5 20.Nd3 Ne7 2 1 .h3 Ng6 22.Qc2

The only way Black can get anything going on the kingside is by an even
tual . . . f7 -f5 advance. By attacking c6, I force his pieces into a defensive
stance and never give him time to prepare the advance of the f-pawn.

22 ... Bd7 2 3 .Nc5

Simply threatening Nxd7 followed by Qxc6.

2 3 ... Rec8 24.Rh l Be8 25.a5 !

Fixing the target on a6. Now the game almost plays itself.

25 ... Rc7
Black's game would also be unpalatable after 2 5 . . .Bd8 2 6.Nb7 Qe7 2 7.Nxd8
Rxd8 2 8.Rb6 Bd7 29.Rfb l .

T H E N EW WORLD

26.Rb6 Rca7 27.Rfhl Ne7

White would have met 27 . . . Bd8 with 2 8 .Nb7 .

2S.Bfl
Ending the life of the a6-pawn, which was fixed (by a4-a5) and then sur
rounded.

2 S . . . NcS 2 9 . Rxa6 Rxa6 3 0 . Bxa6 QdS 3 1 . BxcS QxcS 3 2 . a6 B e 7

3 3 .Rb7 QdS
Or B . . . Bxc5 3 4.Qxc5 Rxa6 3 5 .Qe7, with an easy win for White.

34.a7
It's clear that Black is lost. My opponent plays on for a few more moves
before giving in to the inevitable.

34 ... BfS 35.Qf5 Be7 36.Nxg5 !

A fun way to put an end to the game.

36 ... Bxg5 37.Ne6 Qf6 3 S.Qxf6 Bxf6 39.Nc7 Rxa7 40.Rxa7, 1 -0.

(34) Uhlmann - Benko (Buenos Aires, 1 960)

l .d4 d5 2 .c4 dxc4 3 .NB Nf6 4.e3 e6 5.Bxc4 c5 6.0-0 a6 7.Qe2 b5 S.Bb3
Bb7 9.a4 Nbd7
Also possible is 9 . . . b4, though I didn't want to give White access to the
c4-square without a fight. The move I chose, 9 . . . Nbd7, leads to a much
more complex and interesting battle.

10.axb5
Best. Neither the gambit move 1 O.e4 nor the slower 1 0.Rd l are threaten
ing to Black.

10 ... axb5 l 1 .RxaS QxaS 1 2 .Nc3

Of course, White wanted no part of 1 2 .Qxb5 BxB , while 1 2 .Nbd2 c4 is
known to be comfortable for Black.

12...b4 1 3 .Nb5 Qa5

In this well known variation, Black must choose between 1 3 . . .Qb8 and
1 3 . . . Qa5 (The game Yusupov-Anand, Las Palmas 1 993 saw Black try a
third idea: 1 3 . . . BxB 1 4.gxB Qb8. Though the Indian genius equalized in

1 15

PAL B E N KO : MY L I F E , G A M E S AN D C O M POSITI O N S

this game, I don't like the idea of giving up Black's strong light-squared
Bishop so early). Since I wanted to win this game, I chose the more ag
gressive but riskier I 3 . . . Qa5 .
1 16

14.e4
By far the most popular move. However, 1 4.Ne5 ! ? is also interesting.

1 4 ... Be7
Taking the e-pawn leaves Black dangerously behind in development. Not
wanting to leave my King in the center (White's threat of Bf4 is quite
annoying), I decided to get the rest of my pieces out and shuttle my King
to safety as quickly as possible.

15.e5
The alternative, 1 5 .d5, is known to be fine for Black after 1 5 . . . exd5 1 6.Bxd5
Nxd5 1 7 . exd 5 0-0 1 8.Qxe7 Qxb 5 .

15 . . .Ne4!
Better than 1 5 . . . Nd5 , because 1 6.Bg5 would then be strong.

1 6.Bc2
After this game, players with the white pieces started to try 1 6.Rd l with a
sharp, interesting battle ahead. It's interesting to note that VV'hite has yet
to demonstrate an opening advantage against 1 3 . . .Qa5 .

1 6 ... Bc6
The Knight on e4 is hanging in the air but so is VV'hite's counterpart on b5.

1 7.Bxe4 Qxb5 1 8.Qc2 c4! 1 9.Bg5

In later years, some theoreticians have claimed that 1 9.d5 ! ? Bxd5 20.Bxd5
exd5 2 1 .Nd4, gives VV'hite a serious initiative for the pawn. An honest look
at this line, though, will show that 1 9.d5 is completely unsound. The oft
quoted, "Play what the grandmasters do, not what they recommend." should
always be kept in mind when you see lines like 1 9.d5 recommended but
never actually employed.

1 9 ... Bxe4 20.Qxe4 Bxg5 2 1 .Nxg5 h6 22 .Qa8+ Nb8 2 3 .NB

Stronger is 2 3 .Ne4, but 2 3 . . .0-0 24.Nd6 Qd5 2 5 .Qxd5 exd5 still favors
Black, though not as much as in the game.

23 ... 0-0 24.Rc1 Nc6 25.Qa2 b3 26.Qa3 Rd8

T H E N EW WORLD

Wolfgang Uhl mann, though not

wel l known to new generations
of U.s. players, was extremely
strong and played a World
Championship Candidates Match
against Larsen in 197 1 .

1 17

(Photo courtesy USCF.)

It's clear that something went wrong for White in the opening. His back
ward d-pawn can't be defended since 2 7 .Rd l Nxe5 2 8 .Nxe5 Qxe5 wins a
pawn. White now resorts to desperation, but all his attempts to compli
cate the game fail.

27.Ral Nxd4 2 8.Nxd4 Rxd4 29.Qa8+

White is completely lost, and it's not surprising to see how quickly his
position falls apart.

29 ... Kh7 30.Qf3 Qxe5 3 1 .Qxf7 Rf4 3 2 .Qb7 Qxb2 3 3 .Rfl Qc2 34.g3 Rf6
3 5.f4 b2 36.Rf2 b l =Q+, 0- 1 .

My ten-game match with Samuel Reshevsky came about when someone in my

company asked me how I would do in a multi-game, head-to-head contest with
the legendary player. I felt confident that I would be victorious, so my backers
came up with the money and the match was on. I was on the verge of winning
several games, but blunders, my usual time pressure, and his amazing defensive
ability allowed him somehow or other to survive. In the end, Sammy won three
games to my two.
Some years later, I acted as his second in his matches versus Korchnoi and
Hort. While preparing for Korchnoi, I visited his home and asked, "Where are
your chess books ? " To my horror I found he only owned three or four, and they
were ones that he wrote ! (Years later, at the age of sixty, he picked up a few more
books and proudly announced that he was finally ready to study!)
One problem Sammy had was his memory, which was terrible. During prepa
rations for the Korchnoi and Hort matches, we would study openings all day,
and by the evening he wouldn't remember anything we had looked at. Thus, he
was never able to really learn openings in depth, and always used up vast amounts
of time in the beginning phase.
Once I realized just how bad his memory was, I was able to have some good
natured fun with him. For example, I showed him a game once and asked, "What
do you think of this game? "
H e said, "It's nothing special at all. These guys weren't very good."
"But Sammy, this is one of your own games ! "

PAL B E N KO: MY L I F E . GAM ES A N D CO M PO S I TI O N S

(35) Benko - Reshevsky, 1 st match game (New York, 1 960)

1 18

l .d4 Nf6 2 .c4 g6 3 .g3 Bg7 4.Bg2 d5 5.cxd5 Nxd5 6.NO 0-0 7.0-0 Nb6
S.Nc3 Nc6 9.d5
Also popular is 9 .e3 .

9 ...Nb4
Later, 9 . . . Na5 became all the rage.

lO.e4 c6 l l .Qb3
I just dreamed up this move over the board because I didn't like the more
natural l l .a3 Na6 1 2 .Be3 , since 1 2 . . . Nc4 would then be possible. I try to
avoid this Knight jump to c4 with the text move. Later theory confirmed
that I l .a3 Na6 1 2 .dxc6 bxc6 1 3 .Qc2 gave White a small plus, though it's
certainly nothing to write home about.

l l .. .Nd3 l 2.Be3 c5
Black tries to reinforce his Knight on d3 . Naturally, both sides had to take
1 2 . . . Nxb2 into consideration. Many complicated variations can then oc
cur, though 1 3 .Qxb2 (The initial 1 3 .dxc6 ! ? is critical, since then 1 3 . . . bxc6
1 4. Qxb2 gains in strength. However, Black can try to stir the pot with
1 3 . . .N2c4 when things are far from clear.) 1 3 . . . Na4 1 4.Nxa4 Bxb2 1 5 .Nxb2 ,
with three pieces for the Queen suggests itself.

l 3 .Ne l c4 l 4.Qc2 Nxe l l5.Rfxe l

White, having gotten rid of Black's advanced Knight, can be more than
satisfied with the opening due to his strong central plus. Black has to do
something about the positional threats of a2 -a4-a5 or f2 -f4 followed by
e4-e5 .

l5 ... e6 l 6.dxe6 Bxe6 l 7.Radl Qe7

White traded on e6 because now Black's Queen doesn't have any com
pletely comfortable squares. The move Black chose in the game ( 1 7 . . . Qe7)
turns out poorly, but 17 . . . Qf6 I S.Bc5 RfdS 1 9.e5 Rxd l 2 0.Nxd l , is very
strong. Perhaps Black should have considered the passive but solid 1 7 . . . Nd7
(though White is obviously better) or 1 7 . . . Qc7, when the simple I S .Bf4

T H E N EW WORLD

has considerable punch (though it looks bad, Black is still fighting after
I S .Nb5 Qe5 I 9 .Nd4 Bd7 20.4 Qa5).

1 8.Nd5
Forcing Black to capture on d5 and give "White a monster passed pawn.
Suddenly Black's position has reached critical mass.

1 19

1 8 ...Nxd5 19.exd5
The creation of this powerful passed pawn leaves Black in a strategically
lost position. Simply put: the defender's chances are grim if such a pawn
can't be blockaded before it safely reaches the sixth rank.

ue1 Reshevsky:

Sammy was a strange guy, very religious, but I was on good terms

with him until the 197 5 U.S. Championship. We were scheduled to play in the last round.
When the time arrived, he approached me and said that the game was very important for him
the result determined whether or not he quali
fied for the Biel Interzonal in Switzerland.

I told him that I'd give him a draw if it helped

him qualify. Somehow, though, it seems he mis
understood me (he might have even felt that I was
willing to lose on purpose which, of course, I'd
never do). When the round started, Byrne and
Rogoff both drew quickly, ending Reshevsky's
hopes for qualification. After that, Sammy offered
me a draw and I refused. This made him very an
gry. I explained I would only have given him

draw if it affected his qualification. Otherwise,

what's the point? So he began to make noise, to
bang the pieces down and slam the clock. We ad-

Reshevsky and Fischer with actor

Jose Ferrer (right), 1 9 6 1 .
(Photo courtesy USCF.)

journed in a winning position for me, but he ran

to the tournament director and told him that we agreed to a draw ahead of time ! Imagine that.
He got me so upset that I screwed up and let him escape with a half-point.
After that, we didn't talk to each other for quite a while, but eventually he approached me
and asked to make peace.
Fischer always respected Reshevsky, and vice-versa. Reshevsky always offered him an early
draw and Fischer would always refuse right away with a simple, "No ! " Fischer especially didn't
like the way Reshevsky would make the offer-Sammy would often say, "Are you playing for a
win?"
Najdorf was also known to ask this question from time to time. Once, Naj dorf asked an
opponent if he was playing for a win. The guy said, "No, I'm not. Would you like a draw ? "
When Najdorf refused, his opponent became very irate and wanted t o know why h e asked if he
was playing to win in the first place. Najdorf said, "I was just wondering."
I heard that Reshevsky once got into a fist-fight with Najdorf. Sammy wasn't a big guy in
any sense of the word, but he attacked with such ferocity that the far larger Najdorf turned and
ran for his life !

PAL B E N KO: MY L I F E , GAM ES A N D C O M POSITI O N S

19 . . . Bg4 20.0 BfS 2 1 .Qxc4

1 20

Black gave up a pawn in order to create some counterplay. Of course,

2 1 . . .Bxb2 loses the Exchange after 2 2 .Bh6.

2 1 ...RacS 22.QbS a6 23 .Qb6 RfeS 24.Bf2 Qd7 2 S.RxeS+ RxeS 26.d6

Even Reshevsky's legendary defensive powers couldn't save him here. Other
than crude and ultimately unsatisfactory threats, Black has no real plan.
White, however, knows exacdy what he has to do: free the g2 -Bishop by
B -f4, take on b7, and then break the d7-blockade with Bc6.

26 ... Re2 27.b3 gS

If 27 ... Rxa2 , 2 8 .f4 is strong. Black vainly tries to stop this, but in the end
the passed d-pawn carries the day.

2S.f4 gxf4 29.Qxb7 fxg3 30.hxg3 Rxa2 3 1 .QaS+ Bf8 32 .Bc6

The blockade has finally been broken. Black tries one last threat, but this
allows White to play to the gallery with some pleasing, albeit obvious,
fireworks.

32 ... Qe6 3 3 .d7 Qe2 34.Qxf8+!

The simple 3 4.Rfl was also sufficient.

34 ... Kx8 3 S .dS=Q+ Kg7 3 6.Bd4+ Kh6 3 7.Qf6+, 1 -0.

Black had no desire to experience 37 . . . Bg6 3 8.Qf4+ KhS 3 9 . Qh4 mate.

(36) R Byrne - Benko (U.S. Championship, 1 960)

l .d4 Nf6 2 .c4 g6 3 .Nc3 Bg7 4.e4 d6 S .O 0-0 6.Nge2 Nbd7 7.Be3 cS
S.Ng3? !
The Samisch system against the KID was very popular in the late ' SOs and
early '60s. However, the best lines for both sides hadn't been worked out
yet. Today, players don't think twice about playing the superior 8.Qd2 or
8.dS.

S ... cxd4 9.Bxd4 NeS 1 0.Be2 Be6 l 1 .NdS

White, who is behind in development and hasn't casded, should accept
the exchange of Bishops and Queens by 1 1 .BxeS (worse is I l .b3 Nc6)

T H E N EW WORLD

1 1 . . .dxeS 1 2 .Qxd8. Of course, giving up control of the dark squares in this

fashion didn't appeal to Byrne, but what happens in the game turns out to
be even worse.
121

1 1 ...Rc8 1 2 .Qa4 Bd7 !

This sends the white Queen scampering back to d l since 1 3 .Qxa7 loses
material after 1 3 . . .Nc6 1 4.Qb6 (Also possible is 1 4.Bxf6 Nxa7 I S .Nxe7 +
Kh8 1 6.Nxg6+, when \Vhite's three pawns can't compete with the extra
black piece.) 1 4 . . . NxdS (This wins the Exchange, but 1 4 . . . Nxd4 I S .Qxd4
NxdS , with a powerful initiative, might be even stronger.) I S .Qxd8 Rfxd8
1 6.Bxg7 Ndb4.

1 3 .Qd l
Dismal, but 1 3 .Qb3 NxdS (with the threat of .. .Nxf3 +) 1 4.BxeS QaS+ I S .KfI
Ne3 + 1 6.Qxe3 BxeS, leaves Black with an overwhelming position.

1 3 ... NxdS
I could have won a pawn by 1 3 . . . Nxc4 1 4.Nxf6+ exf6, when Black is obvi
ously better, but my poor pawn structure makes it impossible to talk of a
definite win. The move I chose ( 1 3 . . . NxdS) shoots for more by retaining
the initiative and giving \Vhite difficult problems to solve.

14.cxdS QaS+

bert Byrne:

Robert was

very strong-he did well in the

197 3 Leningrad interzonal, but cl
timately lost to Spassky in the

Balcerowski vs. R
Byrne (right) i n
Poland, 1 9 62.
(Photo courtesy
USCF.)

1974 Candidates Matches. He was

a no-nonsense, serious man.
I acted as his second at one
point in his career. He hired me
when he saw how useful I'd be in
adjourned endgames. Apparently,
he came to this conclusion after
we played a game where I had a

A 1 9 6 1 U.s.
Championsh i p
moment: Robert
Byrne plays Pal Benko
(front right) , wh ile
Reshevsky (seated
next to Benko) takes
on Lombardy.
(Photo Raoul
Echeverria, courtesy
USeF.)

s l i gh t e d g e i n a d r awn R o o k
endgame. I won, and afterwards
he was upset and said, "This was
a draw! " I told him to prove it, but
he just couldn't make a draw-I
won over and over in our analy
sis. Finally I showed him how to
save the position and, evidently,
this made an impression.

PAL B E N KO : MY L I F E , GAM ES A N D C O M POSITI O N S

1 22

Believe it or not, White is on the verge of defeat.

15.Bc3?
Better was 1 5 .Kf2 , though Black would retain a strong attack with 1 5 . . . f5 .
Naturally, 1 5 .Qd2 10ses to 1 5 ... Qxd2 + 1 6.Kxd2 NxB + followed by 1 7 ... Bxd4.

15 ... Rxc3 ! 1 6.bxc3

If 1 6.Qd2 Ra3 1 7 .bxa3 , Black wins the white Queen (and the game) by
1 7 . . . NxB + 1 8 .gxB Bc3 .

1 6 ... Qxc3 + 1 7 .Kf2

Even worse is 1 7 .Kfl NxB ! ( 1 7 . . . Ng4 is also good) 1 8 .Re I Nxh2 + 1 9.Rxh2
Qxg3 20.Rh l (2 0.Kgl Be5 is the end of the universe for White) 20 . . . Qf4+
2 1 .BB f5 , when Black has a crushing attack.

1 7 ...Ng4+ !
This blow makes the white game hopeless. A huge mistake would be
1 7 . . . NxB ? ? , because 1 8 .gxB gives the white King access to g2 after
1 8 . . . Bd4+.

1 8.fxg4 Bd4+
The point. White's King is forced to block the h I-Rook, thereby leaving
the a I -Rook inadequately defended.

19.Kf l Qxal 20.Qxal Bxal 2 1 .Kf2 Bd4+ 22.KB Re8 2 3 .Rdl Be5
An extra pawn, two strong Bishops, and possession of the only open file

combine to make the win very easy.

24.Rd2 Re3 + 25.Kf2 Bf4 26.Rb2 Be3 + 27.Ke l Rcl + 28.Bdl Bxg4
White could have safely resigned here, but he chose to suffer for a few
extra moves.

29.Ne2 Re7 30.Rb3 Bb6 3 1 .Kd2 Bd7

Once a plan gives you a decisive material advantage, it's not uncommon
for your pieces to be a bit off balance. When this happens, it's a good idea
to consolidate your position by firmly guarding any and all weaknesses,
and to make sure your pieces once again interact well with each other.
The last few moves have seen me firmly defend b 7 and place all my pieces
on safe yet flexible squares.

T H E N EW WORLD

3 2.Kd3 ReS
Now it's time to push forward again, maximizing the strength of my Rook
and two Bishops.

3 3 .Rc3 BbS + 34.Ke2 Be4 3 S.Kb2 RbS+, 0- 1 .

White has no reason to play out such a hopeless position.

(3 7) Benko - Mednis (U.S. Championship, 1 96 1 )

l .e4 eS 2 .NO Ne6 3 .d4 exd4 4.Nxd4 Nf6 S.Nc3 d6 6.BgS e6 7.Qd2 a6
s.o-o-o Bd7 9.f4 Be7 1 0.NO bS

According to theory, White's best is now 1 0.Bxf6, forcing Black to sacrifice

a pawn or allow his pawn structure to be compromised. I rarely like to enter
mainstream theory, so I made a point of trying a different continuation.

1 1 .eS b4
It was already known that 1 1 . . . dxe5 wasn't a good idea: 1 2 .fxe5 Nd5 (lvkov
Ciocaltea, Belgrade 1 95 6 was a complete disaster for Black after 1 2 . . . b4
1 3 .exf6 bxc3 1 4.Qxd7+ Qxd7 1 5 .Rxd7) 1 3 .Nxd5 exd5 1 4.Bxe7 Nxe7 1 5 .Bd3
Nc6 1 6 .Kb l Be6 1 7 . Qf4 h6 1 8 .h4, and White has the better position.

1 2 .exf6
Sacrificing a piece by 1 2 .exd6 bxc3 1 3 .Qxc3 Bf8 is interesting but prob
ably unsound.

12 ... bxe3 1 3 .Qxc3 gxf6 14.Bh4

White has also tried 14.Bh6 and 1 4.f5 ! ? , but neither promises an advantage.

14 ... dS 15.a3
This is a cautious move that aims to control the b4-square. It also avoids
the heavily analyzed lines that arise from 1 5 .Kb l (which, to be honest, is
clearly best) .

15 ... NaS
Black plans . . . Rc8 followed by . . . Nc4. However, the immediate 1 5 . . . Rc8 is
more accurate since 1 6.Bxa6? Nb4 wouldn't make White happy.

1 6.Be 1 Nb7 1 7.Nd4 NeS 1S.Nb3 Ne4 19.QO fS 20.g4

The black Knight has ended up on a good central square, so White tries
to prove that it's not completely secure there. My 2 0th move undermines
the enemy Knight and also keeps the black King in the center.

20 ...fxg4 2 1 .Qxg4 Qb6 22.Bd3 ReS 2 3 .Kb 1 Ba4 24.Bxe4 dxe4 2 S.Bh4
White could avoid the loss of a pawn by 2 5 .Qg3 , but I decided to toss all
thoughts of passive defense out the window. For some reason, I became
consumed with a seek-and-destroy mentality.

2 S ... Bxh4 26.Qxh4 Qe7

Black could, and probably should, accept the pawn, though White would
have enough compensation because Black's King is trapped in the center.

1 23

PAL B E N KO: MY L I F E , GAM ES A N D CO M PO S I TI O N S

1 24
Edmar Mednis (left) plays
Fischer at the 1 9 5 9 U.s.
Championship.
(Photo courtesy USCF.)

If he grabbed the pawn by 26 . . . Bxb3 2 7 .cxb3 Qxb 3 , White would do quite

well. For example, 2 8.Rc 1 Rb8 2 9.Qf6 Rf8 3 0.Rc7 Qd3 + 3 l .Ka l Qd8
3 2 .Qc3 Qd6 B .Rc i . A more testing way to win a pawn was 2 6 ... Rxc2
2 7 . Kxc2 Bxb 3 + 2 8 . Kb l Bxd l 2 9 . Rxd l e 3 3 0 . f5 , with interesting
complications.

27.KaI Qe7 28.Qe l

White doesn't trade Queens because his King is safer than his opponent's.
Now 28 ... Rxc2 2 9 . Qxe4 wins for White.

28 ... Bxb3 29.cxb3 fS

This seems too loosening. H e should have preferred 2 9 . . . Qb7.

30.Qe2

The two King positions make the difference. If Black protects the a6-pawn,
then Rgl would leave the black King without a safe haven. This, in turn,
would leave the black Rooks unconnected. Due to these considerations,
Mednis gives up the pawn so he can castle, but his King seems doomed to
suffer no matter where it runs to.

30 ... 0-0 3 1 .Qxa6 Ra8 3 2 .Qc4 Qf6

If 3 2 . . . Rfc8, White wins by B .Rhgl + and Qd4.

H .Rd6 Rfe8
White also wins after 3 3 . . . Rxa 3 + 3 4.Kb l followed by Rxe6.

34.Ka2 Rac8 3 S .Rg1 + Kh8 36.Rxe6! , 1 -0.

A satisfying way to end the game.

inten" ,he" yeac I've ever h,d w"' 1 962 . A.ide from the Varn,
Olympiad, it featured two exceptionally strong tournaments: Stockholm and
Cura<;ao (a beautiful island in the Dutch Indes).
Fischer had bounced back from his disaster in Buenos Aires with a brilliant
victory at Bled, where he was undefeated. Next came the Stockholm Interzonal.
This was a very important event for both of us - the top eight finishers would
be seeded into the Cura<;ao Candidates Tournament, and this would produce
Botvinnik's challenger for the title of World Champion.
Fischer's performance at Stockholm was simply incredible. He finished two
and a half points ahead of his nearest rival in an extremely strong field. During
that event, we spent countless hours together analyzing our adjourned games.
Bobby, in his determination to win every game, exhausted himself by playing
one long, difficult endgame after another. In one instance, after making an error
in a game versus Yanofsky, he fought on for 1 1 2 moves, refusing to let his oppo
nent deprive him of an extra half-point.
My enjoyment of our post-mortem analysis faded on one occasion. Against
Gligoric, Bobby had adj ourned with a pawn minus. However, after hours of
joint analysis we became convinced that the position could be saved. By that
time it was early morning and I needed to start work on my game, which was
also adjourned. Against the late Leonid Stein, I held the advantage of owning a
minor piece for two pawns. "You've got no problems, " Bobby declared. "Let's
get some sleep. "
The next day, after avoiding several devilish traps that we had foreseen in our
analysis, Bobby managed to hold on and secure the draw. In my game, Stein let
loose with a carefully designed shot, and in the ensuing tense complications he
managed to steal a draw. Nevertheless, all turned out well: Bobby and I both
1 25

Awards
ceremony at
Cura<;:ao: (left to
right) Benko,Tal,
two organizers,
and Fischer.
(Photo D R.
Cantwell.)

qualified (Bobby won, while I shared sixth place with eight wins, eleven draws,
and three losses).
WIth the chess world applauding Fischer's genius, he was labeled the favorite
in Curac;ao, though a host of legendary players were also participating. This tour
nament assumed mammoth proportions, with twenty-eight scheduled rounds (we
had to play four games against every other contestant) ! I played White against
Bobby right at the outset (see game 4 1 ) . Within the short span of time between
the Stockholm and Curac;ao events, deep theoretical preparation was impossible.
To minimize this problem, I made use of an opening setup that avoided main
stream theory and soon became known as the Benko System. Naturally, I used it
at every opportunity-the positions I usually reached with it being comfortable
for me and somewhat unfamiliar for my opponents.
Fischer, positive that he would win the tournament, did terribly in Curac;ao.
Travel problems, a late arrival, and the burden of enormous expectation stressed
him (in fact, his behavior bordered on panic!), and soon he was walking an emo
tional tightrope. First, he began to turn his back on many grandmasters that had
always been friendly to him in the past. Next he mercilessly attacked the orga
nizers about the conditions of the tournament. Finally, he claimed that the Rus
sians had conspired to make key draws in their games between themselves. This
is easy to believe, since many people make pre-arranged draws - it's no big deal.
However, Fischer also thought that some games had been lost on purpose in
Curac;ao, even claiming that Korchnoi might have tossed a game to Petrosian. I
can't claim to know the truth about Fischer's charges, but there were mini-teams
working together. For example, Geller and Petrosian, who were close friends,
did everything in their power to make sure Keres did not come in first.
Curac;ao was a great event. I went swimming before all the games, enjoyed the
company of many nice women, and had a wonderful time. The trick to the tour
nament was to know when to start having fun. The formula for this was simplic-

T H E H U NT FO R T H E WO RLD C H A M P I O N S H I P

ity itself: if you could n o longer qualify after the first half was over, then you
should relax and enjoy yourself. But Fischer refused. Even though his dreams of
conquest had been shattered, he still thought of only chess.
It was at Curaao that Fischer and I had our one and only altercation, and I've
always felt that the USCF was, at the very least, partially to blame. Kasper, the
federation official who was responsible for giving us expenses and arranging trainers and/or analysts, explained that funds were low and that they could only afford one second. Thus, Bisguier would serve as a shared second, with the provision that, if Fischer needed him or if both of us had adjourned games at the
same time, Bisguier would help Bobby and ignore me. I wasn't happy about this
at all, but what choice did I have?
I started well in Curaao, so every game had great significance for me. In the
fifth round, I adjourned against Petrosian with excellent winning chances. Natu
rally, I hunted Fischer and Bisguier down and asked them to help me analyze the
position. Why wouldn't I? Bobby was my friend and we always analyzed together
at tournaments, while Bisguier was hired for this very purpose ! However, when
Bobby realized that I intended to make use of "his" second, he said, "I forbid it! "
I was outraged. "How can you forbid this? It's in the agreement. "
It turns out that Kasper talked to both of us separately. Apparently, he didn't tell
Fischer that we had to share Bisguier. In fact, when I discussed this with Bobby
sometime later, he told me that he would never have accepted such a deal.
During our face-off, though, I was sure Bobby knew what was going on. Un
fortunately, in the heat of the moment, with both of us convinced that we were
in the right, things got ugly. Bobby got insulting and upset, I got more and more
angry, Bobby goaded me, and bang, I hit him.
Looking back on this mix-up, I'm ashamed of the whole affair. In fact, I be
came so guilt-ridden for punching someone I genuinely cared about that I could
never play well against him again.
This tournament wasn't only a nightmare for Fischer. The great Paul Keres
also had some experiences that would prove traumatic. In the hunt for first place
throughout, his demise effectively began during our third game, with him play
ing White against my Sicilian. I got the better game, but I spoiled my chances
by incorrectly sacrificing a piece in horrible time pressure. This left me with
nothing better than a perpetual check. With only seconds to spare, I made the
move that forced the perpetual, but it was slightly off-square. He punched my
clock and said, "Adjust the pieces ! " Surprised, my clock ticked for a second or
two before I realized what was going on. Then I desperately reached out to fix
the position of the piece, glanced at the clock, and watched in horror as my flag
fell and I was forfeited.
I didn't complain, but I was very angry and thought, "I'm going to beat this
guy when it's the most painful for him. " Sure enough, our final game was critical
to his whole career, since a draw would allow him to conduct a playoff against
Petrosian to see who played Botvinnik for the World Championship, while a
win would make him the outright challenger.

1 27

Petrosian and
Tal, Curac;ao.
(Photo Dr. R.
Cantwell.)

In this all-important game, I was a bit better, and adjourned. A while later,
Petrosian and Geller came to me in secret and offered to help me beat their own
countryman! I was disgusted. Telling them that it would be a draw with best
play, I demanded that they leave. However, when we resumed, Keres made an
error and I won.
This result crushed poor Keres, but his suffering wasn't over yet! In my next
and final game I played Geller. Oddly, they had already printed the last round
results in the bulletins before the final game had been completed: Petrosian first,
Keres second, Geller third. Against Geller, I was up two pawns in a Queen
endgame. I had one move left to play and wanted to be sure he didn't get a per
petual check. Sickeningly, my flag fell as I made my final move and Geller won!
Because of this, Keres had to play a match (which he won) with Geller to deter
mine second place, and he later wrote that I had deliberately lost to Geller to
"screw him."
Naturally, I'd never do anything like that, but Keres believed otherwise and
probably went to his grave with this erroneous impression. One can understand
his feelings -losing that one game against me influenced chess history in a very
big way: Petrosian got the match against Botvinnik, won, and became the new
champion. And poor Keres never did get to play the championship match that
meant so much to him.

(3 8) Benko - German (Stockholm I nterzonal, 1 962)

l .e4 e6 2 .d4 d5 3 .Nd2 Nf6 4.e5 Nfd7 5.c3 c5 6.f4
A greedy system that claims a huge amount of space. If White can repulse
Black's counterattack against the pawns on d4 and e 5 , and if White can
find a safe haven for his King, then an eventual f4-f5 will lead to a clear
advantage.

6 ... Nc6 7.Ndf3 Qb6

Black quickly brings all his soldiers to bear against d4. White has to be
very careful, if his center implodes, then his lack of development and the

Petrosian, maki ng sure

that Keres did not come
i n fi rst at Curac;ao.
(Photo Dr. R. Cantwell.)

potentially weak squares on e4 and along the b6-g 1 diagonal will tell in
Black's favor.

8.g3 ! ?
As far as I know, I introduced this setup in international tournaments in the

Dublin Zonal event in 1 9 5 7 against L. Schmid. Now 8 . . . cxd4 9.cxd4 Bb4+

can be met by 1 O.Kf2 f6 I l .Kg2 , when White is better thanks to his obvi
ous advantage in space. Not wanting to die without a fight, recent games
have seen Black increase the tension with the berserk 1 O . . . g5 ! (instead of
1 O .. .f6), when a highly complicated battle ensues with mutual chances.

8 ... f6 9.Bd3
Mter this game with German, White usually played 9.Bh3 . However, I
don't see anything wrong with my choice.

9 ... cxd4 l O.cxd4 Bh4+? !

Tempting, but i t actually helps the white King reach a safe haven o n g2 .
Better was 1 0 . . . Be7, though I still prefer White. Of course, 1 0 . . . Nxd4? ?
would have lost to I l .Nxd4 Qxd4 1 2 .Bg6+ hxg6 1 3 .Qxd4.

1 1 .Kfl f5
Closing the kingside. White would obtain a dangerous attack after 1 1 . . . fXe5
1 2 .fXe5 0-0 1 3 .Kg2 Be7 1 3 .h4.

1 2 .Ne2

PA L B E N KO : MY L I F E . GAM ES AN D C O M POSITI O N S

1 30

With my d4-pawn firmly defended, I'm now free to seek a kingside deci
sion by expanding in that sector via h2 -h3 followed by g3 -g4. It's clear
that any play Black creates on the queenside can't compare to the huge
gains White will be making on the opposite side of the board.

12 ... Be7
The Bishop no longer had a role to play on b4. By going to e7, the b4square is freed for a possible invasion by . . . Nh4.

1 3 .h3 NfS 1 4.g4 g6 15.Nc3

The Knight is much more aggressively posted here than it was on e2 . Now
Black must constantly worry about possible Knight leaps to a4 or b 5 . An
other point of 1 5 .Nc3 is that Black won't be able to answer gxf5 with . . . exf5
since that would leave the d-pawn hanging.

15 ... Bd7 1 6.Be3 a6

Preparing to trade some pieces on b5 by . . . Na7, followed by . . . N (or B) to
b 5 . Black can't capture the b-pawn because 1 6 . . . Qxb2 1 7 .Nb5 Rc8 1 8 .Rh2
Qb4 1 9.Bd2 traps the Queen.

1 7.Rh2
This Rook is heading for g2 where it will perform double duty: preparing
to attack down the g-file while simultaneously defending b2 .

1 7 ...Nh4 I S.Be2 Ne6 19.Rb l Na7 20.Rg2 ReS 2 1 .Nd2

An odd-looking move that has two purposes: the first clears the d l -hS
diagonal so that, after White plays gxf5 , and after Black exchanges some
pieces on b 5 , White's Queen will be free to leap into h 5. The other point
of Nd2 is that, once White's Queen jumps to h5, the Rook on b l will be
in need of a defender. 2 1 .Nd2 takes care of that problem.

2 1 ...Bb5 22 .gxf5 gxf5 2 3 .Bxb5+ Nxb5 24.Qh5+ Kd7 25.Rg7 QdS

The Queen runs back so it can defend against threats like 26.Qh6 and 26.Na4.
I wasn't worried about 2 5 . . .Nxc3 , since 26.bxc3 instantly brings my inactive
Rook into the game (note that the Knight on d2 is guarding the b l -Rook). If
Black had tried 2 5 . . . Kd8, I would have broken through with 26.Qh6 Nxc3
(2 6 . . . Rc7 fails to 2 7 .Nxd5) 2 7 .bxc3 Qc7 2 8.c4 dxc4 29.d5, when the dual
threat of 3 0.Bb6 and 3 0.d6 snuffs out Black's resistance.

T H E H U NT FO R TH E WO RLD C H A M P I O N S H I P

26.NxbS axbS 27.Nb3 b6

Keeping VV'hite's Knight out of cS .

28.Qe2 Rc4
The threat was 29.QxbS+ Kc7 30.NaS ! Qd7 3 1 .Rc 1 +, so Black gives up the
Exchange in a last-ditch effort to keep VV'hite's pieces out of his position.

29.Nd2 Qa8 3 0.Nxc4 dxc4 3 1 .Qg2

Nipping any and all Black counterplay in the bud.

3 1 . .. Qxg2+ 3 2.Kxg2 Ng6

VV'hite purposely allowed his Rook to get locked in so that Black would
try to win it. The resultant gain of time allows VV'hite to penetrate on the
queenside.

3 3 .Rc1 Ke8 34.a4 bxa4

Also hopeless is 3 4 . . . Kf8 3 S .Rxg6 hxg6 3 6.axbS , followed by 3 7 .Rxc4.

3 S.Rxc4 Kf8 36.Rxg6 hxg6 3 7.Rxa4 Kf7 3 8.Ra6 Rb8 39.Kf3

Rushing my King over to the embattled queenside.

39 ... Rb7 40.Ke2 Ke8 4 1 .Kd3 Kd7

If Black tries 4 1 . . . Bd8 with the intention of . . . Rh 7 , VV'hite strikes first with
42 .dS ! exdS 43 .Ra8 Ke7 44.Bf2 Bc7 4S.Bh4+ Kd7 46.Kd4, when the win
is no longer in doubt.

42 .dS !
A well-prepared breakthrough that ends the game.

42 ... exdS 43 .Bxb6 gS 44.fxgS BxgS 4S.e6+ Ke7 46.Kd4 4

On 46 . . . Bf6+ 47 .KxdS Bxb2 48.BcS+ Kf6, I had intended 49.Ra2 ! , chasing
the Bishop off the a l -h8 diagonal. Then 49 . . . Bc1 SO.Ra8 with the threat
of S I .Rf8+ (SO . . . Bh6 S I .Rh8) leads to an easy win.

47.KxdS f3 48.BcS+ Kf6 49.Ra3 Rxb2

A blunder, but the game was resignable anyway.

SO.Bd4+, 1 -0.

(3 9) Benko - Uhl man n (Stockholm I nterzonal, 1 962)

l .e4 e6 2 .d3
I had no desire to challenge Uhlmann to a theoretical battle in his favor
ite French Defense (Uhlmann was long considered to be one of the great
experts on this opening). Of course, I knew that he had a deep under
standing of the black side of the KIA, but now the scales were even be
cause I also had quite a bit of experience with this system-I successfully
employed it four times in Stockholm, though this was by far my hardest
test.

2...dS

131

PAL B E N KO : MY L I F E , G A M E S A N D CO M PO S I TI O N S

Uhlmann always entered the main line as a matter of principle. Nowa

days, many players postpone . . . dS for a bit and play 2 . . . cS 3 .Nf3 Nc6 4.g3
g6 S . Bg2 Bg7 6.0-0 Nge7, with a satisfactory position.
1 32

3 .Nd2 Nf6
Another popular system is 3 . . . cS 4.Ngf3 Nc6 S . g3 Bd6 6 . B g2 Nge7
7 .0-0 0-0.

4.g3 c5
Still heading for the line that was typical for that time. A much sharper
reply is 4 . . . dxe4 S . dxe4 eS 6.Ngf3 BcS 7 .Bg2 Nc6.

5.Bg2 Nc6 6.NgO Be7 7.0-0 0-0 S.Re l Qc7

In later years players began dispensing with this move and going for the
immediate 8 . . . b S . However, White still achieved excellent results: 8 . . . bS
9.eS Nd7 1 0.Nfl as 1 1 .h4 b4 l 2 .Bf4 a4 1 3 .a3 ! bxa3 1 4.bxa3 Ba6 l S .Ne3
NaS l 6.Bh3 ! (stopping Black from challenging the e-pawn by . . .f6) l 6 . . . d4
l 7 .Nfl Nb6 l 8.NgS NdS 1 9.Bd2 BxgS 20.BxgS Qd7 2 1 .QhS Rfc8 2 2 .Nd2
Nc3 2 3 .Bf6, and White's attack crashed through in Fischer-Miagmasuren,
Sousse 1 967.

9.Qe2
When Fischer reached this position, he played 9.eS Nd7 1 0.Qe2 . How
ever, I wanted to gain a tempo by attacking Black's Queen with Bf4 before
I pushed my pawn to e S .

9 . . .b5
Black's play lies on the queenside, so this space-gaining advance is much
better than the tame 9 . . . b6.

10.Nf l Ba6
More common is 10 . . . aS, when the a-pawn can be used in Black's queenside
attack. Uhlmann had a very different idea in mind.

I l .Bf4
So far, all according to plan. Having gained a tempo with this Bishop move,
I'm finally ready to close the center with e4-eS and start an assault against
the black King.

1 1 . .. Qb6 12.e5 Nd7 1 3 .h4

T H E H U NT FO R TH E WO RLD C H A M P I O N S H I P

A very interesting position has been reached: White's pawn o n e5 cuts the
board in half and stops black defenders from using the f6-square. By over
protecting this important pawn, White prevents his opponent from challenging the e5-monster by . . .f6 or .. .f5 since then exf6 would instantly free
White's Queen and e l -Rook (down the e-file), create a weak black pawn on
e6, and give White's f4-Bishop and B -Knight access to the e5 -square. Since
White's e5 -pawn prevents central counterplay, White will be free to probe
Black's kingside for weaknesses by h4-h5 -h6, when the dark squares surrounding black's King will be compromised. Then moves such as Nfl-h2 g4 and Qd2 followed by Bg5 will fight for access to the newly weakened f6square. All this sounds wonderful for White, but Black isn't without his own
chances. By playing . . . c5-c4xd 3 , Black will open the c-file and create a potentially weak pawn on d3 (it can be attacked by . . . b5-b4, opening up the
a6-d3 diagonal, followed by . . . Nd7-c5).

13 ... e4 14.h5 cxd3 1 5.exd3 Rae8

Of course, 1 5 . . . h6 begs for a sacrificial explosion on h6 after Nfl-h2 -g4.
Also possible (if Black plays . . . h6) is g3 -g4-g5 when the kingside gets
ripped open.

1 6.h6 g6
Both players have actively pursued their respective goals: White trying
for an attack against the enemy King, and Black striving to open lines on
the queenside. I've always believed that White's chances in this kind of
position are superior. Uhlmann, naturally, had a different opinion.

1 7.Be3 !?
I decided to get my pawn to d4. This will lead to a strengthening of the
e5 -pawn and will also deny the black pieces access to c 5 . The more the
matic 1 7 .Ne3 b4 1 S.Ng4, was also possible.

17 ... Qb7
Black must have been tempted to play 1 7 . . . d4 (far worse is 1 7 . . . Bc5, since
White wants to exchange dark-squared Bishops and weaken Black's grasp
on f6), but he ultimately decided against it since it gives White eventual
access to the e4-square. Nevertheless, I feel that 1 7 . . . d4 was Black's best
move, when, after I S .Bf4 Nb4 1 9.Rec 1 ( 1 9.Red l ! ?) 19 . . . Rxc 1 (the simple
1 9 . . . Bb7 might be preferable) 2 0.Bxc 1 RcS 2 1 .Bg5 ! , the game becomes
very sharp and unclear: 2 1 ...Rc2 (or 2 l . ..Bxg5 22 .Nxg5 Qc7 2 3 .Qd2) 22 .Qdl
Bxg5 2 3 .Nxg5 .

1 8.d4 Nb4 1 9.Recl Rxcl 20.Bxcl Re8 2 1 .Bg5

White would love to exchange these Bishops since that would leave the
squares f6 and g7 very weak.

2 1 ...BfB
Too dangerous is 2 1 . . .Rc2 since 2 2 . Q d l Bxg5 2 3 .Nxg5 Rxb2 24.QB Nb6
2 5 .Qf6 mates. Black needs his dark-squared Bishop for defense. A very
interesting alternative, though, is 2 1 . . .Nc2 when 2 2 .Rd l b4 2 3 .Qd2 BfS

1 33

PAL B E N KO : MY LI FE. G A M E S AN D C O M POSITI O N S

leads to a balanced position. However, I intended to answer 2 1 . . . Nc2 with

2 2 .Bxe7 ! Nxa 1 2 3 .Ng5 with a very strong attack due to the weakness of f7
and the squares f6 and g7 .
1 34

22 .Ne3
Simultaneously threatening Ng4, and keeping Black's pieces out of c2 .

22 ... Nc6 2 3 .Bf1 !?

The immediate 2 3 .Ng4 also left White with the superior position.

23 ...Ncb8 24.Ng4 b4 25.Qd2 Bxfl 26.Rxfl

Black has succeeded in exchanging his bad Bishop, and his position on the
queenside appears to be quite threatening. However, during the game I
felt that the vulnerable position of Black's King would give White good
long-term chances in the middlegame and in the endgame.

26 ... b3 27.axb3 Qxb3 28.Kg2 Rc2 29.Qa5

This leads to interesting positions, but the more natural move is 29.Qf4,
when Black's King is far from safe: 1) 29 . . . Qxb2 3 0.Bh4 Nc6 3 1 .Ng5 f5
3 2 .exf6 Nxd4 3 3 .Ne5 Ne2 3 4.Qa4, and White wins. 2) 2 9 . . . Nc6 3 0 . Bh4
Qb8 3 1 .Rc 1 ! Rxc 1 3 2 .Qxc 1 , when the threat of B . Qf4 followed by Ng5
is annoying. 3) 29 . . . Rxb2 (this gives Black the defensive maneuver . . . Qd3 f5) 3 0.Bh4 Qd3 3 1 .Rc 1 Na6 (3 1 ... Qf5 3 2 .RcS leaves Black in serious trouble)
3 2 .Be7 ! Bxe7 B .RcS+ (even the shocking B .Qf6 works !) 3 3 . . .Bf8 34.Qf6,
and Black gets mated.

29 ... Qb7
At first, 29 . . . Nc6 3 0.Qa6 Qxb2 seems to be promising for Black. How
ever, my earlier statement about the miserable black King still holds true,
and White demonstrates this after 3 1 .Ne3 Rc3 3 2 .Nd 1 Qa3 3 3 .Qb7 RxB
34.Qxc6 NbS 3 5 .QeS (also strong is 3 5 .QcS Rb3 36 .Ne3 Qa6 3 7.Rc 1 QxcS
3 S.RxcS Na6 3 9.RdS Rb7 40.Ng4) 3 5 . . . Rb3 3 6.Ne3 Rb7 3 7 .Ng4 Nd7
(3 7 . . . a5 3 S .Rb 1 ! ! crashes through) 3 S .Rc 1 , and White wins.

3 0.Rcl Nc6 3 1 .Qa4 Rxcl 32.Bxcl a6 B .Ng5 Be7 3 4.NB Qb5 35.Qc2
White's play is directed against Black's vulnerable King. Due to this, White
has no interest in exchanging the Queens.

35 ... Qc4 36.Qd l 5 37.exf6 Bxf6 38.Be3 Be7 39.b3 Qc3 40.Qe2 Qxb3
4 1 .Qxa6 Nd8

TH E H U NT FO R TH E WO RLD C H A M P I O N S H I P

The time control has been reached, and I was able to calm down and study
the position. White is better because his King is safer than Black's. My
first order of business is to trade those dark-squared Bishops.

42 .Bg5 Kf8 43.Qa7 Qb7 44.Bxe7+ Kxe7 45.Qa3 + Ke8 46.Ng5 Qb l

Black is losing a pawn, but his Queen (which is badly needed for defense)
is able to scamper back and avoid immediate disaster.

47.Qa4 Ke7 48.Qa3 + Ke8 49.Qf3 Qf5 50.Nxh7 Nfl 5 1 .Nhf6+ Nxf6
52 .Nxf6+ Ke7 53 .Ng4 Qxf3 + 54.Kxf3 Kf8 55.Kf4
The extra pawn, combined with the far superior King position, adds up to
an easy win for White.

55 ... g5+ 56.Kf3 Kg8 57.Nf6+ Kf8 58.h7 Kg7 59.Nd7

Now 59 . . . Kxh7 is met by 60.Nf8+ Kh6 6 1 .Nxe6 Nd6 62 .Kg4, and wins.

59 . . . Kg6 60.Nf 8 + Kf5 6 1 . g4 + Kf 6 62 . Ke 3 Nh 8 6 3 .Nd 7 + Kg7

64.Ne5 ! Kxh7
White gives back his pawn plus in order to reach a winning pawn endgame
(thanks to his outside passed pawn).

65.f4 gxf4+ 66.Kxf4 Kh6 67.Kf3

Whenever the black Knight emerges from its hole, White will trade it
and win. For example: 67 ... Ng6 68 .Nxg6 Kxg6 69.Kf4 Kf6 70.g5+ Kg6
7 1 .Ke5 Kxg5 72 .Kxe6 Kg6 7 3 .Kxd 5 .

67 ... Kg7 68.Kg3 Kh7 69.Kh4 Kh6 70.g5+ Kh7 7 1 .Kg4 Kg7 72.Kh5, 1 -0.

(40) Benko - Gligoric (I nterzonal Playoff, 1 962)

l .d4 Nf6 2 .c4 g6 3 .Nc3 d5
A bit of a surprise since Gligoric usually employed the King's Indian De
fense and was considered to be the world's greatest authority on that
opening.

4.Bf4
The most principled answer is 4.cxd 5 Nxd5 5 . e4, but I didn't wish to en
ter a sharp line that my opponent probably knew better than I. Instead, I
aim for a small but safe edge with the text move.

4 ... Bg7 5.e3 0-0

Daring me to enter the complications of 6.cxd5 Nxd5 7 .Nxd5 Qxd5 8.Bxc7,
when Black has a lead in development and White's King is still stuck in the
center. Though it's by no means clear that this pawn sacrifice is completely
sound (and no definite assessment has been reached even today!), I decided
to avoid his preparation and steer the game into more placid waters.

6.Qb3
An old continuation that dates back to the '40s !

6 ... c5 7.cxd5

1 35

{j?tetozar Gligoric:

Gligo was an extremely

strong player who ruled Yugoslav chess for de

cades. I played him several times, and time pres
sure almost always played a part in the result. A
good example is from round one of the 1 95 9 Can
didates Tournament. Gligoric played badly in the
opening and, in my desire to wipe him out, I used
huge amounts of time on obvious moves, declin
ing a draw in the process.
Unfortunately, my opponent refused to be
blown away and by move 3 5 (I had to reach move
40 to make the time control) I realized I had no
time left at all. Still much better, I offered a draw
and Gligoric stopped the clocks and accepted. As
we got up to leave, a curious spectator rushed up

G l igoric plays Herman Steiner i n

Hol lywood, 1 946.

to our table and pressed my clock to see how

much time I had left and, my flag instantly fell!

I wanted no part of 7 .dxc5 Ne4 8.cxd5 Qa5 9.Ne2 Nxc5 , when Black has a
very active position.

7 ... cxd4 8.exd4

We 've now transposed into a Caro-Kann (usually reached by l . e4 c6
2 .d4 d5 3 .exd5 cxd5 4.c4 Nf6 5 .Nc3 g6 6.Qb3 Bg7 7.cxd5 0-0) where
White's options are limited due to the early Bf4. Nevertheless, I felt that
accurate play would still give me chances for a small plus.

8 ... Nbd7 9.Be2 Nb6 1O.Bfl e6

S a fer is 1 0 . . . B g4, which wins back the pawn after I l .Bxg4 Nxg4
1 2 .ND Nf6 1 3 .d6. White would then only possess a slight advantage.

1 1 .d6!
The transpositions have served their purpose since Gligoric wasn't really
familiar with this line. On the other hand, I knew this position extremely
well -I suffered through the black side of it in 1 954! After I l .dxe6 Bxe6,
Black has ample compensation for the pawn. Now, with 1 1 .d6, it will be
much harder for Black to demonstrate sufficient play for his sacrifice. The
whole game revolves around this question: can Black get something going
before White consolidates? Theory (and my own painful experience !) says
the answer is no. However, a second look suggests that things might not
be as simple as they appear.

1 1 ...Ne8
In the game Kluger-Benko, Budapest 1 954, I didn't like the look of 1 1 ... Ne8,
and instead tried 1 1 . . .Nfd7. Unfortunately, White gained a clear advan
tage after 1 2 .Nge2 eS 1 3 .dxeS NxeS 1 4.BxeS BxeS I S .Rd l . No wonder I
was so happy to repeat this same line with the white pieces!

1 2 .NbS! NdS

T H E H U NT FO R T H E WO RLD C H A M P I O N S H I P

I (and the theory books) had always thought that I 2 . . . a 6 1 3 .Nc7 was sim
ply crushing. However, while writing the notes for this book it suddenly
dawned on me that things were far from clear: 1 3 . . . Nxc7 I 4. dxc7 (worse is
1 4. Qxb6 Nb 5 ! ) 1 4 . . . Qxd4 I 5 .Ne2 Qxb2 1 6 . Qxb2 Bxb2 1 7 . Rb 1 Nc4
( 1 7 . . . Na4 is worse), Black has an extra pawn but White's pieces are more
active and the c7-pawn is a major force in the game. I'll leave it up to the
theoreticians to figure out what's really going on, but here's a sample line:
I S .Bh6 Bg7 (far better than IS ... ReS 1 9.0-0 Ra7 2 0.Rfd 1 , with advantage
for White) I 9.Bxg7 Kxg7 20.Rc 1 Ne5 , and Black has turned things around.

1 3 .Bxd5 exd5 14.Ne2 a6 15.Nbc3 Be6 1 6.Qxb7

White, who doesn't mind defending for a while, is glad to trade his doubled
pawn. Now I 6 . . . RbS fails to I 7 .Qe7.

1 6 ...Nxd6 1 7.Qb4 Nc4 I S.0-0 ReS 1 9.Qb3 Ra7

Of course, the bS-square is off limits due to White's Bishop. Perhaps Black
should have tried 19 . . . g5 20.Bc 1 , though both 20 . . . Bf5 and 20 . . . RbS are
answered by 2 1 .Qd 1 , when I prefer White.

20.Radl Bg4 2 1 .Rfe l

Having a n extra pawn, all I have t o d o i s consolidate my position by de
fending my pieces and getting my Queen to safety. Once that is done, I'll
be ready to start more active operations.

2 1 . .. Re6 22 .Qc2 BfS 23 .Qc1 Rc6 24.h3 Be6 25.Na4!

So far Black hasn't been able to create any real threats. Now White tries
to take over the initiative by occupying the c5 -square and kicking the
c4-Knight with b3 .

25 ... Qa5 26.Nec3 Rb7 2 7.Re2 Rb4 2 S.Nc5 ! QdS

White meets 2 S . . . Bxc5 with 2 9 . dxc5 Rxc5 3 0 . b 3 followed by B d 6 ,
winning the Exchange. Even worse for Black i s 2 S . . . Bxc5 2 9.dxc5 Qxc5
3 0.Nxd5 Bxd5 3 1 .ReS+ Kg7 3 2 .Bh6+.

29.b3 Nb6 30.Nxa6 QcS 3 1 .Nc5

Of course, 3 1 .Nxb4? ? Bxb4 dooms my c3 -Knight due to the pin along the
c-file.

3 1 . .. Bxc5 3 2 .dxc5 Rxc5 3 3 .Be5

1 37

PAL B E N KO : MY L I F E . GAM ES A N D CO M PO S I TI O N S

The exchanges have left White with two passed pawns and a good Bishop.
As usual, White's low on time, but if he can avoid some tactical tricks he
should be able to reel in the point.
1 38

H ...Nd7 34.Qa3 ! Qb7 3 5.Na4

With my flag hanging, I saw 3 S .Bd6 but got spooked by 3 S . . . Rxb3 36. Qxb3
Qxb3 3 7 .axb3 Rxd . White would be winning after 3 8.b4, but I decided to
play it safe and not do anything risky until I reached the time control.

3 5 ...RcS 3 6.Bal Rf4 3 7.Qd6 Rf5 3 S.Bd4

It's not just my connected passed queenside pawns that bother Black. He's
also worried about the weak dark squares around his King.

3S ...Rc6 39.Qa3 Qa6 40.Red2 Rf4

White made the time control and Black's position is completely hopeless.
Now my play becomes more purposeful and I'm able to put my opponent
away.

4 1 .Qe7 QcS 42 .Qg5 Rf5 43 .Qh6 Ne5 44.Qe3

Not falling for 44.f4? RhS .

44. . .Qc7 45.Nc5 g5 46.Rc2 BcS 47.Rdc1

The rest is an execution and doesn't need any comment.

47 . . . f6 4S .Na6 Bxa6 49.Rxc6 Nxc6 5 0 . Q e S + Kg7 5 1 .Rxc6 Qb7

52 .Qe6, 1 -0.

(4 1 ) Benko - Fischer (Candidates Tou rnament Curaao, 1 962)

l .g3
A flexible move that intends the fianchetto of the light-squared Bishop,
and allows many transpositional opportunities that could easily take the
opponent into positions that he's not completely familiar with. As we shall
see, the system demands sophisticated positional technique, and Black can't
afford to develop in a vacuum. Two rounds later, I took a point from the
former World Champion M. Tal with the same opening. This first move
by White can't be considered as an original opening, of course, since Black
has many ways to lead play back into known realms.

T H E H U NT FO R TH E WO RLD C H A M P I O N S H I P

1 ...Nf6 2 .Bg2 g6 3 .e4 d6 4.d4 Bg7

White has managed to gain space by occupying the center with pawns.
However, Fischer isn't concerned because he's very familiar with the King's
Indian positions that result.

5.Ne2 0-0 6.0-0 e5 7.Nbc3

I had designed this move to catch Bobby by surprise. Usually 7 .c4 was
played, transposing into a King's Indian Defense. However, though this
setup for White wasn't really known against the Pirc at that time, it has
since become a popular way for the first player to handle the position.

7 ... c6
Faced with a new twist, Bobby reacts
with a common King's Indian move.
After the game Fischer suggested
7 . . . Nc6, but in that case I intended
8.dxeS dxeS 9.NdS , with a slight positional advantage for White.

'U

CHE SS LI FE

8.a4
S topping Black from entertaining
ideas of queenside expansion via ... b 7bS. White's 8.a4 also grabs queenside
space and lets Black worry about the
pawn's continued journey to a s .

8 ... Nbd7 9.a5

Sharpening the play and gaining even
more queenside space. Tal prevented
this advance by 8 . . . aS, though I still
gained a pull by 9 . b 3 followed by
l O.Ba3 with pressure against d6.

Fischer and I col lide in Cura<;ao.

(Photo courtesy USeE)

9 ... exd4
A very common King's Indian plan. Black accepts a weak d6-pawn, but in
compensation hopes to gain a lot of activity for his pieces. In this posi
tion, however, Black's d7-Knight doesn't have permanent access to the
key cS -square (White can always chase it away by b2 -b4), and this makes
the d6-pawn a serious liability since Black won't be able to generate any
real pressure against White's e4-pawn (a black Knight on cS plays a ma
jor role in any serious attack against e4).Now we can see why Tal played
8 . . . a S . By taking control over the b4-square, Black is then able to capture
on d4 and place his Knight on cS without fear of it being hit by the
advance of White's b-pawn.

1 0.Nxd4 Nc5 1 1 .h3

Rejecting 1 1 . Bf4, since 1 1 . . .Re8 1 2 .Re l Ng4 followed by 1 3 . . . NeS would
show that the f4-Bishop isn't really doing anything. Instead, I decided to
develop my Bishop on the solid e3 -square where it will defend d4 and cast

1 39

PA L B E N KO : MY L I F E . GAM ES A N D C O M POSITI O N S

a hungry eye towards the queenside and, potentially, c5 and a7. Also note
that by keeping the Bishop off f4, I'm able to place a pawn there and thus
deprive Black's pieces of access to e 5 .
1 40

1 1 ...Re8 1 2 .Re l Nfd7

This eases the pressure on my e-pawn, which allows me to develop my
Bishop smoothly to e3 . However, a move like 1 2 ...Qc7 would have allowed
me to play 1 3 .Bf4, when the d-pawn is quite uncomfortable. Black could
have tried to rid himself of his weakness by 12 ... d5, but after 1 3 .exd5 Rxe l +
1 4.Qxe l Nxd5 1 5 .Nxd5 Bxd4 (Black would have been left with an isolated
d-pawn after 1 5 ... cxd5.) 1 6.Qb4 Ne6 1 7 .Ne7+ Kg7 1 8.a6! (Also good for a
small but long-lasting edge is 1 8.Nxc8 Qxc8 1 9.c3 Bf6 20.a6, when White
has a pair of Bishops and a favorable pawn formation.) 1 8 ... Bc5 1 9.axb7 Bxb7
20.Qxb7 Bxe7 2 1 .c3 White has an obvious advantage.

B .Be3 Qc7 14.f4 Rb8 1 5 .Qd2 b5

Fighting for some air on the queenside. Of course, 1 5 . . .Nf6 would have
been calmly met by 1 6.Bf2 .

16.axb6 axb6
It was probably better to try 1 6 . . . Nxb6, but Bobby's commanding victory
at Stockholm took away his sense of danger, and he didn't see that his
position was quickly becoming critical.

1 7 .b4
Forcing the c5 -Knight away from its active perch. A strong alternative
was 1 7 .e5.

1 7 ... Ne6 1 8.b5

Destroying the c6-pawn's guard over d5, and thus giving White's Knight
access to that very fine central post.

18 ... Nxd4 19.Bxd4 Bxd4+ 20.Qxd4 c5

Black's lack of development, plus the weaknesses on d5 and d6, are enough
to guarantee his demise.

2 1 .Qd2 Bb7 22 .Rad l Re6 2 3 .e5

This wins material because Black's d-pawn is pinned.

TH E H U NT F O R TH E WO RLD C H A M P I O N S H I P

2 3 ... Bxg2 24.Kxg2

Simpler is 24.exd6 Rxe 1 + 2 5 .Rxe l Qb7 26.Qxg2 , when the win won't prove
too difficult to achieve.

24 ... Qb7+ 25.Kf2 Rd8 26.exd6

So White has won Black's d-pawn, but the game isn't over since the pawn
is blocked and in need of support. Also, my King is exposed. Transform
ing a winning position into the full point is often very difficult, and this
game is yet another reminder of this fact.

26 ...Nf6
Also hopeless is 2 6 . . . NfS 2 7 .Rxe6 Nxe6 2 S .Qd5 ! Qxd5 2 9 .Nxd 5 Nd4
3 0.d7 ! KfS 3 1 .Nxb6.

27.Rxe6 fxe6 28.Qe3 Kfi 29.Qf3 Qb8

Trading Queens wouldn't make any sense for Black, since then White's
central King would become advantageously placed (an exposed King is
only bad if there are heavy pieces on the board that can scare it). Now
White has to deal with his threatened d-pawn.

30.Ne4 Nxe4+ 3 1 .Qxe4 Rd7

Unfortunately for Fischer, 3 1 . ..Rxd6 wasn't possible due to 3 2 .Qe5 RdS
(32 . . . Ke7 3 3 .Qg7+ KdS 3 4.QfS+ ends the game quickly.) 3 3 .Rd7+, win
ning the black Queen.

32 .Qc6 Qd8 3 3 .Kf3 Kg7 34.g4

A simple move that gains kingside space. Seeing that he can't maintain his
passive defensive posture (moves like 3 4 . . . KgS and 34 . . . KfS are both answered by the crushing 3 5 .Ra l Rxd6 3 6.RaS, while 34 . . . Kf7 3 5 .Ra l Rxd6
3 6.Ra7+ KfS 3 7 .Qb7 also wins easily), Bobby sacrifices another pawn in a
desperate bid for counterplay against the white King.

34 ... e5 35.fxe5 Rf7+ 36.Kg2

Much better was 3 6.Kg3 Qg5 3 7 .Qe4, when the game would end quickly.
Unfortunately, I was in my usual time trouble and allowed my opponent
to keep a certain amount of hope smoldering in his chest.

36 Qh4 3 7.Rf l
..

141

Bobby in 1 9 62.

Avoiding his final trap: 3 7 .Rd2 ? ? allows 3 7 . . . Qe l , when Black will achieve
a draw by perpetual check.

3 7 ... Rxfl 3 S.Kxfl Qxhh 39.Qg2 Qe3 40.Qe2

Guarding all my pawns and securing the win. Here the game was adjourned.

40 ... Qh3 + , 1 -0.

This was the sealed move, but Bobby resigned when he realized that there
wasn't a perpetual check. For example : 4 1 .Kf2 Qh2 + 42 .Kf3 Qhh 43 .Kf4
Qh6+ 44.Ke4 Qh l + 45 .Qf3 Qe l + 46.Kd5 Qd2 + 47 .Kc6, and White's King
is immune to further harassment and his passed pawns can't be stopped.
Losing a game like this so early in the event had a very negative effect on
Fischer, and in the next round against Geller he lost again as Black. When
the tournament ended and the smoke had cleared, his great performances
in Yugoslavia and Sweden were overshadowed by seven losses, and he ended
up three full points behind the leaders.

(42) Benko - Keres (Candidates Tournament Curaao, 1 962)

l .Nf3 dS 2 .g3 Bg4
Usually Black employs this Bishop move after 2 . . . Nf6 3 .Bg2 c6 4.0-0 Bg4,
but there's certainly nothing wrong with this quicker version. The point
of this move order is to keep Black's options open concerning the g8-Knight,
which can be developed on e7 in some lines (the possibility of an eventual
. . . f7 -f6, defending e 5 , is also used).

3 .Bg2 Nd7 4.0-0 c6 S.d3 eS 6.h3 BhS 7.c4

Playing for e4 is probably a bit stronger, though it's not clear if White can
get more than a very small plus. A recent example: 7 .e4 (I prefer 7 .Nbd2
Bd6 8 .e4 Ne7 9.exd5 cxd5 1 0.c4!, with an initiative) 7 . . . dxe4 8.dxe4 Bc5

T H E H U NT FO R TH E WO RLD C H A M P I O N S H I P

9.Nbd2 f6 1 O.Nc4 Nb6 I l .Na5 Qc7 1 2 .Bd2 Rd8 1 3 .Qe2 , with mutual
chances in Art. Minasian-Bareev, Montecatini 2 000.

7 ... dxc4 8.dxc4 Ngf6 9.Be3

White prevents a possible ... Bc5 .

9 ... Qc7 10.Nc3 Bb4? !

This places the Bishop on a vulnerable square and leads to a loss of time. I
would have preferred 1 0 . . . Be7, though White retains a slight edge with
1 1 .Nh4.

1 1 .Qb3 a5
On 1 1 . . .Ba5, I would have played 1 2 .Qa3 , when Black's position is quite
uncomfortable.

1 2 .Na4!
Stopping Black's planned . . . Bc5 and threatening c4-c5 , trapping the Bishop.

12 ... Be7
The Bishop has to go back, but now the other Knight leaps into play.

1 3 .Nh4!
This sacrifice of a pawn (e2 is hanging) forces the gain of the two Bishops
that, thanks to the hole on b6, will guarantee White an advantage.

1 3 ... 0-0
Black wants no part of 1 3 . . . Bxe2 ? 1 4.Rfe l Bh5 1 5 .Nf5 Bf8 1 6.f4, when
White has a raging attack.

14.g4 Bg6 1 5.Nxg6 hxg6 16.Rfdl Rab8 1 7.c5

Clamping down on b6.

17 ... Nh7
Black is hoping to swing this Knight around to e6 by . . . Nh7-g5-e6. How
ever, I had prepared a very strong retort.

1 8 .Nb6! RfdS
If 1 8 . . . Nxc5, then White wins the Exchange by 1 9.Bxc5 Bxc5 2 0.Nd7.

1 9.Qc3

Defending c5 and eyeing both e5 and a s . Black can't prevent the loss of a
pawn.

1 43

PA L B E N KO : MY L I F E , G A M E S A N D C O M POSITI O N S

19 ... BgS 20.Nc4?

1 44

Not best. There wasn't anything wrong with 20.Bxg5 Nxg5 2 1 .Nxd7 (and
not 2 1 . Qxa5 Ne6) 2 1 .. .Rxd7 22 .Rxd7 Qxd7 2 3 .Qxe5 Qd8 24.e3 , when Black
would have absolutely no compensation for the lost pawn.

2 0 . . . Bxe3 2 1 . Qxe3 Re8 2 2 .Qa3 Ra8 2 3 .Rd2 Nhf 8 24.Rad l Red8

2 S.Qe3
White is still clearly better, even after the error on the 2 0th move. Now
White tries to stop Black from playing . . . Ne6, which would be answered
by 2 6.Nb6 Nxb6 2 7 .cxb6 Rxd2 2 8 .Rxd2 Qb8 2 9.Rd7, when White is ob
viously in control.

2S ... Re8 26.b3

Preparing a3 followed by b4, when the c5 -pawn is rock solid.

26 ... Rab8 27.a3 Ra8 28.b4 axb4 29.axb4 Ra4 30.Qc3 Ra6
Black is helpless.

3 1 .Rd6 Nf6?
Keres never liked passive positions so it's no surprise that the pressure
finally gets to him. His 3 1 . . .Nf6 loses a pawn.

32 .Nb6? !
It's hard to believe that, even in extreme time pressure, I would miss a
move like 3 2 .Nxe5 . Of course, my 3 2 .Nb6 still maintains an iron grip on
the position.

32 ... e4 3 3 .e3 gS 34.Qd2

It's always been my policy to strengthen my position quietly and avoid
risk when I'm low on time.

34 ... Qe7 3 S.Rd8 Ra3 36.Rxe8 Qxe8 3 7.Qb2 Ra7 3 8.Nc4 Qe6 39.Bf l
Better was 3 9.Nd6, going after the e4-pawn.

39 ...NdS 40.Nb6 Nxb6 4 1 .cxb6

The game was adjourned here. White still retains the advantage, though
it's not nearly as large as it was earlier in the game.

4 1 . .. Ra4? !

 Keres:

Everyone liked Keres.

At first he was a great tactician, but he

turned more positional as the years
passed. Before the war, he was likely the
best player in the world, only Fine
might have been his equal.

[Left Young giants in 1 940: (left to right) Levenfish, Keres, Botvi n n i k, and Kotov.
(Photo courtesy USCF.)

[Right] Keres points out a variation to Petrosian, Cura<;:ao 1 9 62.

(Photo Dr. R. Cantwel l.)

The Rook turns out to be vulnerable to attack on this square. A more

solid choice was 4 1 . . .RaS.

42 .b5 Ra2
Black would lose a pawn after 42 . . . c5 43 .Qc2 hitting c5, e4, and a4. Also
rather miserable is 42 ,..cxb5 43 .Qxb5 RaS (and not 43 .. .Ra6 44.RdS) 44.Bg2 .

43.Qb l ! cxb5?
After this mistake the game unexpectedly ends, but 43 . . .c5 44.RdS Ra5
45.RbS, was also a losing cause for Black.

44.RdS !
Suddenly Black has no good defense against the threat of 45. Qb4.

44 ... f5 45.gxf5 Qf7 46.Bxb5 g6 47.RcS, 1 -0.

There is no answer to White's intended Bc4.

(43) Benko - Korchnoi (Candidates Tournament Cu raao, 1 962)

l .d4 Nf6 2 .c4 e6 3 .NB d5 4.Nc3 c5 5.cxd5 Nxd5 6.g3
Sharper is 6.e4 Nxc3 7 .bxc3 , but I've always enjoyed the positional battle
that arises from 6.g3 .

6 ... cxd4
A playable sideline. More usual is 6,. .Nc6 7 .Bg2 Be7 S.O-O 0-0, when both
9.Nxd5 and 9.e4 were the main lines for many years. In the 'SOs, the mys
terious 9 .Rb I ! ? also made its mark on theory.

Victor Korchnoi in 1963.

(Photo courtesy USCF.)

Korchnoi:

Victor has a very strange style that, in some ways, is similar to the great

Emanuel Lasker's. Amazingly, he's still going strong in his 70's - nobody has ever played like
him at such an advanced age.
I never had many personal dealings with him, though he was always against the hard-line
communist stances. I remember, long before he defected, reading in a paper how Jews were
being mistreated in Russia and I asked him if it was true. He verified it, though other players
would have feared the consequences of being so outspoken.
I beat him one of the four games we played in Cura(.:ao. In one of the games, we adjourned
with me holding all the winning chances, though I thought he could draw with best play. I
played a tricky sealed move, but I was sure his helpers would take all my possibilities into
consideration and find the correct defense. On the next day, there was an excursion, and I had
to agree to draw our game without resuming if ! wanted to do the trip. I accepted this because
I didn't have much doubt that it would have been drawn anyway.
During the excursion, I told Korchnoi my sealed move and asked how he would have an
swered it. "Oh, we never even considered that! " Needless to say, this really destroyed my en
joyment of the excursion ! In the next game with him (game 43), we adjourned again, and he
offered me a draw. I was a bit better, but it was still a draw with best play. However, this time I
played it out and actually got the victory in the endgame. I learned a big lesson here: "Never
take it for granted that your opponent, or his helpers, knows what's going on ! " In fact, Fischer
lived by this rule, and won many drawn positions from the world's best players.
I remember adjourning with Korchnoi in another tournament. I was slightly worse, but didn't
think there was much real danger. Of course, I intended to look at the position in detail, but I
heard some Russians talking at the hotel, and they said Korchnoi thought it was a dead draw.
Then my Yugoslav girlfriend said, "Let's go to a movie ! " and, well, what's a guy supposed to
do? Mer all, if Korchnoi himself thought it was drawn, then why disappoint the young lady?
Of course, when we resumed I lost the game . If I had analyzed it, I would have drawn easily.

T H E H U NT FOR TH E WO RLD C H A M P I O N S H I P

7.Nxd5 Qxd5
In a later game, Korchnoi (against Krogius, USSR ch. 1 965) tried 7 . exd5,
and got a good position after 8.Nxd4 Qb6 9.Nb3 Nc6 1 O.Bg2 Be6 1 1 .0-0
d4 1 2 .Bd2 Rd8.
. .

8.Qxd4 Qb5
For 8 . . . Qxd4, see my game against Darga at the Varna Olympics, 1 962
(game 5 8). All of these lines give \\!hite a slight endgame edge, and there
fore suit my style very well.

9.e4 Qb4+
Black has to trade Queens because there's no good place to run. For ex
ample, 9 . . . Qh5 1 0 .Ne5 wouldn't give Black any joy.

lO.Qxb4 Bxb4+ 1 1 .Bd2 Bxd2 + 1 2 .Kxd2 Nc6 1 3 .Ke3 Ke7

\\!hite has a bit more space, but Black has no weaknesses. Nevertheless, I
was quite happy to play this kind of position against Korchnoi.

14.Rcl Bd7
This position is considered to be equal by theory, but in my opinion \\!hite
has a very small but long-lasting edge.

1 5 .a3
A simple but flexible move that takes away the b4-square from the black
Knight, prepares b2 -b4, and allows the white Bishop to run to a2 after it
moves to c4.

1 5 .. .Rhc8 16.Bc4 Na5 1 7.Ba2 Bb5 1 8.Nd4 Nc4+ 19.Bxc4 Bxc4

I was willing to trade my Bishop for the Knight because I felt that, ulti
mately, my Knight would prove itself to be more useful than the enemy
Bishop.

20.Rc2 Ba6 2 1 .Rhc l Rxc2 22 .Rxc2 Kd7 2 3 .e5

In such positions, little things make all the difference. Here I'm gaining
central space, getting my pawns off the color of Black's Bishop, and (by
preventing an eventual . . . e6-e5) making sure the d4-square will be perma
nently available for use by my Knight or King.

1 47

PAL B E N KO : MY L I F E , GAM ES A N D C O M POSIT I O N S

2 3 ...ReS 24.RxeS KxeS

1 48

The Rooks are traded and, to the uninitiated, the game appears to be headed
for a draw. However, the position is actually very pleasant for White be
cause only he can play for a win while Black will be defending for a long
time to come. Why is White better? The flexibility of the Knight is one
huge factor (the d6-square in particular is calling to it!), and my central
space advantage and superior King also combine to make Black's life un
pleasant.

25.NB
Opening up d4 for my King, while threatening both Ng5 and NB -d2 e4-d6.

25 ... h6 26.Nd2 Kd7 2 7.Ne4 b6 2S.f4 Bf l 29.Kd4 Bg2 30.Nd6 f5

White's position has improved, but Black has everything covered. Now
starts a long game of cat and mouse: I make little j abs at his weak points
and torment him endlessly.

3 1 .Nb5 Bf l 32 .Nd6
I avoided 3 2 .Nxa7 Kc7 B .a4 Kb7 3 4.Nb5 Bxb5 3 5 .axb5, because the pawn
endgame would be drawn since White's King isn't able to penetrate.

32 ... a5 H .b4 axb4 34.axb4 Bh3

Black's King isn't in a hurry to move up since that would allow me to at
tack his kingside pawns by Ne8.

35.Ne4 Ke6 36.Nd6 Kd7 3 7.Ne4 Ke6

Of course, Black isn't willing to advance his pawn to b5 because that gives
White a chance to penetrate on the c5-square.

3S.Ne3
By dominating Black's Bishop, I cut his options down considerably.

3 S ... Kb5 3 9.Kc3 g5

Not bad at all, but he could also have considered remaining passive with
3 9 . . . Kc6 40.Kc4 Bg4, since 4 1 .Nxg4 fxg4 followed by . . . g7-g6 is just a draw.

40.Ne4 gxf4 41 .gxf4 Ke6

The game was adjourned here and Korchnoi offered a draw. I declined,

T H E H U NT FO R T H E WO RL D C H A M P I O N S H I P

hoping that I could still squeeze something from my advantage.

42 .Kd4 h5 43 .Nd6 h4 44.Nc8 Bg2 45.Ne7+ Kb5 46.Kc3 h3

Now his pawn is fixed on h3 . My chances reside in his targets on h3 , e6,
and b6.

47.Ng8 Kc6 48.Nf6 Bf3 49.Kd4 Kb5 50.Nd7 Bd5

He didn't want any part of SO . . . Kxb4 S l .Nxb6 KbS S 2 .Nd7 Kc6 S 3 .NfS
BdS S4.Nh7 Bg2 S S .NgS Kd7 S6.Ke 3 , when I would march my King over
to g3 and snap off his h-pawn.

5 1 .Kc3 B a2 5 2 . Nf8 Kc6 5 3 .Nh7 Kd5 54.Nf6+ Kc6 55 .Kd4 Kb5

56.Kc3 Kc6
White safely made the time control because Black couldn't do anything
but repeat moves (this made it easy for me to flick out a bunch of solid
moves without any risk whatsoever). Now I play my last big card: I will
bring my Knight to gS where it will attack both h3 and e6.

57.Nh7 Kd5 58.Ng5 b5 59.Kd3

Suddenly Black is facing some serious problems. White is willing to enter
the race by giving up his b-pawn because the passed e-pawn or h-pawn
(whatever one White creates) will prove faster. Note that the immediate
S9.Nxh3 ? would have been an error due to S9 . . . Ke4.

59 ... Bc4+ 60.Ke3 Bfl 6 1 .Kf2 Bg2

If this was a problem, it would read: White to play and win.

62 .Kg3?
Missing my chance to grab victory by 62 .Ke2 ! . This powerful move forces
Black to part with either his h-pawn or e-pawn. In either case, White wins
the game, though many of the lines are extremely complicated: 62 . . . Kc4
(By far the best chance. 62 . . . Be4? 63 .Nxh3 , followed by NgS is easy for
White. More interesting is 62 . . . Bh l 63 .Nxh3 , when I'll just give one line:
6 3 . . .Kc4 64.NgS Kxb4 6S .h4 Kc3 66.hS b4 67 .h6 b3 6S.h7 b2 69.hS=Q
b l =Q 70.QcS+ Kd4 [70 ... Kb2 7 1 . QbS+ trades Queens and easily wins for
White] 7 1 .Nxe6+ KdS [7 1 .. .Ke4 n .Qc6 mate] 72 .Qd7+ Kc4 7 3 .Qc7+ KdS
74.Qd6+ Kc4 7 S .QcS+ Kb3 76.Qb6+, and the game is over.) 63 .Nxe6 Kxb4
64.Nd4 KcS 6S .NxfS b4, and now 66.e6 leads to lots of fascinating varia-

1 49

PAL B E N KO : MY L I F E . GAM ES A N D CO M PO S I TI O N S

1 50

tions (i.e., 66 . . . b3 67.e7 Bc6 [67 . . . b2 6S.eS=Q b l =Q 69.QfS+, should win

for White, though this is still a bit touch-and-go] 6S .Kd3 Kb4 69.Nd6
Ka3 70.Kc3 Ka2 7 1 .Nc4 Bd7 n .f5 BeS 7 3 .6 Bh5 74.Nb2 B1 7 5 .Nd3 ,
and White wins) but 66.Kd2 might prove even stronger: 66 . . . Be4 (There
are several moves to examine here, we'll settle for this one.) 67 .Ne4 Kd4
6S.e6 Bc6 69.Nc2 + Kc5 (69 . . . Ke4 70.Nxb4 Be8 7 1 .Na6 Kxf4 n .Nc7 Bc6
7 3 .Ke2 ! , ends the game) 70.5 b3 7 1 .6 Kd6 12 .1 Ke7 73 .Nb4 Be4 74.Kc3
Bf5 7 5 .Nc6+ KfS 76.Nd4, and it's over.

62 ...Kc4 63 .Nxe6 Kxb4 64.Nd4

One big difference between the lines here and the lines that would have
come about after 62 .Ke2 ! is that White's King is able to stop the b-pawn
(after 62 .Ke2).

64 ...Kc4??
Korchnoi finally cracks. He had to try 64 . . . Kc5 ! when the game would be
drawn: 6 5 .Nxf5 b4 66.Ne3 b3 67.Nd l Kd4 6S.e6 Bd5 69.e7 B1 70.Kxh3
Kd3 7 1 .5 (7 1 .Nb2 + Kc3 n .Na4+ Kb4, =) 7 1 . .. Kd2 n .Nb2 Kc3 , and a
peaceful conclusion can't be avoided.

6 S .Nxf S b4 66.Ne 3 + K d 3 6 7 .Nxg2 hxg2 6 S .Kxg2 b3 6 9 . e 6 b2

70.e7 b l =Q 7 1 .eS=Q
White is two pawns up and Black doesn't have a perpetual check. The rest
IS easy.

7 1 ...Qa2+ 72.Kg3 Qa7 73 .QbS+ Ke4 74.QeS+ Kd3 7S.fS Qh7 76.h4 QgS+
77.Kf4 Qc4+ 78.KgS QgS+ 79.Kh6 Qf8+ SO.Qg7 Qd8
Black also loses after SO . . . Qxf5 S l .Qg6 Ke4 S2.h5.

8 1 .f6, 1 -0.

After I beat Fischer in the first round of the Candidates Tournament, I found
myself facing Tal in the third and seventeenth rounds. I looked forward to these
games with relish since our first meeting in 1 95 7 ended in an unsatisfactory man
ner. Here's the position from the 1 95 7 contest.

White to Move

T H E H U NT F O R T H E WO RLD C H A M P I O N S H I P

This game, from the World Student Team Championship (I was playing for
the Hungarian team), had been going well for me, and why not? Though Tal
was already Russian Champion and occupied their first board, I was older and
considered myself to be more experienced. I was determined to win from the
start! In the diagram, I had to make my 40th move. I was low on time (time
pressure is mandatory in my games!) and had to choose between two King moves.

40.Kg3??
This is the wrong one. I could have won a piece and the game with 40.Kg l ,
since Black couldn't stop both the mate threat and Nxa6. Play would have
then gone: 40 . . . Nd4 4 l .Nxa6 Rxb2 42 .Bc7 ! (stopping . . . Nh2 +), and vic
tory would have been mine.

40 ... Re3 + 41 .Kf2??

I tried to repeat the position so I could correct my mistake and play Kg l .
Of course, Black declined. Better was 4 1 .Kf4 (and not 4 1 .Kh4 h6 !), but
after 41 . . .Rd3 42 .Re l Rd4+ 43 .Kg3 Rd3 +, Black has a draw by perpetual
check. I saw this line but could not bear to let Black escape with a draw.

41 ...Rd3 42 .Re l?
Continuing to self-destruct. White should at least trade Rooks and try to
draw due to the opposite-colored Bishops. Now the game becomes
hopeless.

42 . . . Rd2 + 4 3 . Kg3 Be2 ! 44.Na6 c4 45 .Nc5 Nxc5 46. Bxc5 Rxb2

47.Bd4 Rd2 48.Be3 Ra2 49.Kf4 f6 50.h4 c3 5 1 .g5 fxg5+ 52.hxg5 Kf7
53 .Ke4 c2 54. Bel Rat 55.Ke3 Rxel , 0- 1 .
This was my first taste of Tal's legendary luck. I never forgot this night
marish finish, and it continued to haunt me every time I played him.
I wanted revenge very badly for that 1 95 7 abomination (not to mention sev
eral other losses against him in Portoroz 1 95 8 and the 1 959 Candidates Tourna
ment) ! This was finally achieved in the third round of Curaao.

(44) Benko - Tal (Candidates Tournament Cu raao, 1 962)

l .g3 g6 2 .Bg2 Bg7 3 .d4 d6 4.e4 Nf6 5.Ne2 0-0 6.0-0 Nbd7 7.Nbc3
We've now transposed into a popular line against the Pirc Defense (usu
ally reached by 1 .e4 d6 2 .d4 Nf6 3 .Nc3 g6 4.g3 Bg7 5 .Bg2 0-0 6.Nge2
Nbd7 7.0-0) . At the time, this set-up was rather new and Black's best re
plies were not known (which explains why both Fischer and Tal did poorly
against it). Of course, I was quite happy with this opening position be
cause I felt that it wouldn't be to Tal's taste.

7 ... c6 8.a4 a5 !?
Tal's improvement over the Fischer game. This typical King's Indian move
stops White from advancing to a5 and secures the c5-square for a black
Knight after . . . e5 and . . . exd4.

15I

PA L B E N KO : MY L I F E . G A M E S A N D C O M POSITI O N S

9.b3
An interesting plan: \Vhite intends to play Ba3 , when the pressure along
1 52

the a3 -f8 diagonal will prove to be quite annoying for Black. \Vhite usu
ally plays h3 followed by Be3 with a slight edge. In fact, it's now known
that this setup (the Knight on d7 combined with . . . c6 and . . . a5), with an
eventual . . . e 5 , doesn't give Black equality against \Vhite's system.

9 ... Re8 1 0.Ba3 Qc7 1 1 .Qd2 e5 12 .Rad l

\Vhite's moves are obvious, to the point, and strong (In other words, you
don't have to be flashy to win a game !).

1 2 ... exd4 1 3 .Nxd4 Nc5 14.f3 b6 l S.Nde2 Bf8

\Vhite must stand better if Black has to defend the d-pawn like this.

16.Bb2
The Bishop no longer had a future on a3 so it steps back to the wide-open
a l-h8 diagonal.

16 ... Qe7 1 7.Nd4 Bb7 1 8.Rfel Bg7 19.f4 Rad8 20.Bf3

\Vhite's central space advantage has grown over the last few moves. 2 0.Bf3
is very useful since it keeps the black Knight from jumping to g4 and also
prepares to place the white Queen on g2 when e4-e5 , targeting the c6pawn, will hang over the black position like a dark cloud.

20 ... Qd7 2 1 .Qg2

Black is under considerable pressure. Right now the threat is e 5 . If Tal
played the natural 2 1 . . . Qc7, I intended to become aggressive with 2 2 .g4.
Tal was never happy defending a passive position, so he tries to free him
self by lashing out in the center.

2 1 ...d5 22 .e5 Nfe4 23 .Nxe4 dxe4 24.Be2

Naturally, I avoided 24.Bxe4 due to 24 ... Nxe4 2 5 .Qxe4 c5 when Black wins.

24 ... Qe7 25.Ba3

Suddenly the a3 -f8 diagonal is once again useful.

25 ... f6 26.Bc4+ Kh8 27.Ne6

Black's position is critical . If 2 7 . . . Rxd 1 2 8 .Rxd l Xe5 29.Nxc5 (also good is
2 9.Nxg7 followed by Bb2 when \Vhite's domination of the a l -h8 diago
nal will cause Black lots of pain) 29 . . . bxc5 30.Qxe4, Black's queenside is a

ai1 Tal:

Tal was a genius, there is no question about this. He had an amazing ability

to calculate long, intricate variations. Everyone liked Tal, in fact, I never ran into a player who
wasn't fond of him (I certainly was) .
Once I played him with sunglasses on because he often stared at people, as if he was trying to
hypnotize them. Some journalists convinced me to do it, as a j oke. Afterwards, a story circu
lated around the world that I had been seriously worried about Tal putting me under his spell !

"I can see in your palm, M r. Tal, that the next World Champion wi l l be . . . Bobby Fischer" Framed
between Tal and Fischer. we see in both photos, Lombardy and Weinstein.T he lady in the
foreground is M rs. Smyslov. T he laughing gentleman at the extreme left is Tal's trainer and second,
A. Koblentz. (Photos courtesy USCF.)

shambles and 3 0 . . . exf4 loses to 3 1 .Qxe7 followed by RdS+. Because I was

in my usual time pressure, Tal tries to confuse matters by sacrificing the
Exchange. However, this time he wasn't able to escape.

2 7 ... RdS 28.BxdS cxdS 29.Nxg7 Kxg7 30.exf6+ Qxf6 3 1 .Qf2

Covering the important d4-square and threatening the cS-Knight. Black
is completely busted.

3 1 . ..Ne6 3 2 .Qxb6 Ba8

Tal is still hoping to create chances for himself with ... d4 followed by ... e3
when the opening of the as-h i diagonal will put White's King in jeopardy.

H .Bd6 QfS 34.QxaS

Stopping . . . d4 again since 3 S .QxfS gxfS 3 6.BeS+ followed by 3 7 .Bxd4 is
game over.

34 ... Kh6 3 S .c4

A strong move since Black can't afford to exchange Queens.

35 ... Rd8 36.Be7

Other moves (like 3 6.cxdS Rxd6 3 7 .QxaS Nxf4 3 S.Rfl ) also won, but I
preferred the simplicity of the text move.

36 ... e3 3 7.Rxe3 Re8 38.BgS+ Kg7 39.Rde l NxgS 40.fxgS

A simple solution. Another crushing continuation was 40. Qc3 + d4 4 1 . Qxd4+
Kf7 42 .Qa7+, and Black must resign.

Tal was hospital ized during the Curaao event, and it was said that the only player to visit him was Fischer.
(Photos Dc R. Cantwell.)

PAL B E N KO : MY L I F E , GAM ES A N D CO M PO S I TI O N S

40. . . Rf8 41 .Qa7+, 1 -0.

White safely made the time control so Black resigned.
1 56

(45) Filip - Benko (Candidates Tournament Cu raao, 1 962)

l .d4 Nf6 2 .c4 g6 3 .Nc3 Bg7 4.e4 d6 5.0 0-0 6.Nge2 Nc6 7.Be3 a6
S.Qd2 Bd7
Black's main alternative is S . . . RbS.

9.Nc l
Other critical White moves are 9.Bh6 and 9.h4. This whole line is ex
tremely complicated and always leads to a full-blooded fight.

9 ... e5 1 0.Nb3
More usual here is 1 0.dS Nd4 1 1 .Nb3 Nxb3 1 2 .axb3 .

1 0 ... exd4 I 1 .Nxd4 ReS 1 2 .Be2 Nb5 !?

A multi-purpose move that opens up the g7-Bishop's diagonal, eyes the

f4-square, and makes . . . Qh4+ or . . . f7-fS possible.

1 3 .0-0
Modern books on theory more or less dismiss this line by claiming a White
advantage after 1 3 .Nxc6 bxc6 1 4.0-0 (on 1 4.g4 Nf6 l S .cS, Black gains ex
cellent counterplay with l S . . . dS) 1 4 . . . cS (the active 1 4 . . . Qh4, intending to
answer l S .BgS with l S . . . Bd4+, is well met by l S .Rf2 ! , when l S . . . BeS fails
completely to 1 6.g4!) l S .Rad l Bc6 1 6.BgS . However, 1 6 . . . Bd4+ 1 7 .Kh1
Bxc3 ! l S .bxc3 f6 1 9.Bh4 Qe7 seems quite acceptable for the second player.
As always, lines used by top players in over-the-board play deserve respect,
while disembodied analysis deserves suspicion!

1 3 ... Nf4
This little tactical device emerged a couple of decades later in the Sicilian,
Accelerated Dragon: l .e4 cS 2 .Nf3 Nc6 3 .d4 cxd4 4.Nxd4 g6 S .c4 Bg7
6.Be3 Nf6 7 .Nc3 0-0 S.Be2 b6 9.0-0 Bb7 1 0.f3 NhS 1 1 . Qd2 (a more vio
lent expression of this idea comes about after 1 l .f4? Nxf4!) 1 1 . . .Nf4 ! . Of
course, 1 3 . . .Nxd4 1 4.Bxd4 Bxd4 I S .Qxd4 Nf4 was also playable.

14.Bxf4 Nxd4 15.Bg5 Nxe2 +

T H E H U NT F O R TH E WO RLD C H A M P I O N S H I P

Taking the bad Bishop, but Black didn't really have a choice since 1 5 . . . Qb8
1 6.Nd5 is unpleasant.

1 6.Nxe2 Qb8
It might be hard to believe, but this move envisions the opening of the
b-file !

1 7 .Nc3
More testing is 1 7 .c5. After 1 7 .Nc3 , Black gets a comfortable position.

17 ... Be6 1 8 .Nd5 Bxd5 19.cxd5 c5 20.dxc6 bxc6

Suddenly the Queen on b8 is doing a great job putting pressure on the b
file and defending d6!

2 1 .Rac1 Qb5

Black has to play actively if he wishes to counterbalance the static weak

ness of his center pawns.

22 .Rf2 d5 2 3 .exd5 cxd5 24.h4 d4

This passed (but isolated) pawn could be weak in the endgame, but in the
middlegame it's a tower of strength!

2 5 .Bh6 Bf6 26.Bg5 Be5

Black is not satisfied with a draw so he provokes White into advancing his
f-pawn. Is this pawn advance strong or is White overextending himself?
The battle will rage around this question.

2 7.f4 Bg7 28.f5 d3

Avoiding the drawish 28 . . . gxf5 2 9. Rcfl (2 9.Bh6 ! ?) 29 . . . d3 3 0. Rxf5 Qxb2
(At first, 30 . . . Re2 seems very nice. However, 3 1 .Qf4 Qxb2 3 2 .Qg3 re
minds Black that his King is vulnerable. One sample is enough to show
some of the pitfalls hiding in the position: 3 2 . . . Kh 8 3 3 . Rxf7 Rg8
34.Rxg7 ! ! Qxg7 3 5 .Qd6.) 3 1 . Qxd3 Qd4+,
After my 28 ... d3 , the fight
becomes very sharp .
=.

29.fxg6
White could also have tried 2 9.f6 Bf8, but I can't believe that Black would
have anything to be afraid of-White's King is a little loose, and the d 3 pawn i s striking panic into the hearts o f the White army.

1 57

PAL B E N KO : MY L I F E , GAM ES A N D CO M PO S I TI O N S

29 ...hxg6 3 0.RB Re2

Black's attack comes first after 3 1 . Qf4 Qxb2 3 2 . Qxf7+ Kh8 , with threats
like . . . Rxg2 + and . . . d2 .
1 58

3 1 .Qxd3 Qxb2

32 .Rxf7
Now 3 2 . . .Kxf7 3 3 .Qd5+ wins for White. Instead of 3 2 .Rxf7 , White could
have tried to defend by 3 2 .Rg3 , but then 32 . . . Bd4+ 3 3 .Kh I Be5 3 4.Qd5
(34.Rg4 Qxc 1 + mates) favors Black because of the simple 3 4 ... Rf8 ! . The
attempt at beauty via 34 ... Qxc 1 +?? fails to a problem-like refutation: 3 5 .Bxc 1
Bxg3 3 6.Qxa8+ Kh7 and now, instead of 3 7 . Qa7 Re l + 3 8.QgI Bxh4, White
has the wonderful 3 7 .Bd2 ! ! Rxd2 (3 7 . . . Rf2 3 8 .Qxa6 Rxd2 3 9 . Qf1 ) 3 8 .Qf3 ,
winning.

32 ... Rxg2 + 3 3 .Kf l RxgS 34.Rxg7+ Kxg7 3 S.Rc7+??

An unfortunate blunder. White could have forced a draw by 3 5 .Qd7+.

3S ... KhS
Suddenly the game is over.

36.hxgS Rf8+ 37.Ke l QeS+ 3S.Kdl Qxc7

Ending the fun. White is left with a few spite checks.

39.Qd4+ Qg7 40.Qh4+ KgS 41 .Qc4+ Qf7, 0- 1 .

(46) Benko - Darga (Varna Olympiad, 1 962)

l .c4 Nf6 2.g3 e6 3 .Bg2 dS 4.NfJ Be7 S.O-O 0-0 6.d4 c6
It's now known that . . . Nbd7 (here or on the next move) gives Black far
better chances to reach eventual equality. One common line is: 6 . . . Nbd7
7 . Qc2 c6 8 .Nbd2 b6 9.e4 Bb7 l O.eS Ne8 1 1 .b3 Rc8 1 2 .Bb2 c5, when both
sides have chances in this complicated, tense, position.

7.Qc2
The sharpest plan is 7 .Nc3 b6 8 .Ne5 Bb7 9.e4, with an edge for White.

7 ... b6 S.NeS Bb7 9.Rd l QcS? !

TH E H U NT FOR TH E WO RLD CHAM P I O N S H I P

The closed variation against the Catalan i s hard to beat, but this i s too
passive. Black refrains from 9 . . . Nbd7 1 0.cxd5 cxd5 because I l .Nc6 gets
one of his Bishops (whether this means anything in a position with such a
fixed central structure is another matter). I was more concerned with
9 . . . Nfd7, which is almost certainly Black's best chance to keep White's
edge to a bare minimum.

1O.Nc3 Nbd7 I 1 .Bf4 NxeS 1 2 .dxeS Nd7 1 3 .Qd2 !

A strong move that cuts down Black possibilities like . . . dxc4 or . . . g5 , and
threatens to blast open the center with e2 -e4.

1 3 ...f6? !
Such a weakening move can't be good, but Black didn't like the look of
1 3 . . . Rd8 1 4.cxd 5 cxd5 1 5 .Rac 1 Qb8 1 6.Nb 5 .

14.cxdS! NxeS
There isn't anything better. After 1 4 . . . exd 5, 1 5 .Nxd 5 ! cxd5 1 6.Bxd5+ is
strong, while 14 . . . fxe 5 1 5 .dxc6 wins.

I S .d6 Rd8
Avoiding 1 5 . . . Bd8 1 6.Bxe5 (also strong are 1 6.Qe3 and 1 6.d7 Qc7 1 7 .Qd6)
1 6 . . . fxe5 1 7 .d7 Qc7 1 8 .Bh3 , which is absolutely horrible for Black.

16.BxeS fxeS 1 7.Ne4 cS

This allows White to finish nicely, but it's the only way to try and get out
of the horrible mess that Black has found himself in. It is easy to see that
1 7 . . . Bf6 1 8 .d7 Qc7 1 9.Bh3 wasn't very appealing.

1 8.dxe7
I could have kept control of the game with 1 8. Qc2 Bf6 ( 1 8 . . . Bf8 1 9.Ng5)
1 9.b4 ! , but 1 8.dxe7 was too tasty to resist.

18 ... Rxd2 1 9.Rxd2 Qc7

Trying to avoid 1 9 . . . Qe8 2 0.Nd6, when White gets two pieces and a Rook
for the Queen.

20.Radl
Black is completely lost. 2 0 . . . Bxe4 fails to 2 1 .Rd8+, and 2 0 . . . Qxe7 loses
the b 7 -Bishop after 2 1 .Rd7, so he vainly tries to close off the d-file.

20 ... BdS 2 1 .Nc3 , 1 -0.

1 59

PA L B E N KO : MY L I F E , G A M E S A N D C O M POSITI O N S

1 60

Black didn't want to suffer anymore, so he gave up. Since 2 1 . . . Re 8

2 2 .NxdS exdS 2 3 .BxdS+ Kh8 24.Bf7 wins immediately for White, that
only left 2 1 . . . Rb8. Unfortunately, 2 1 . . .Rb8 2 2 .BxdS exdS 2 3 .NxdS Qc6
2 4.Nc7 ! Kf7 2 S .Rd7 Re8 2 6.Nxe8 Kxe8 2 7 .Rxa7 would make resignation
seem desirable.

(47) Benko - Pachman (Varna Olympiad, 1 962)

l .e4 c5 2 .NO e6 3 .d3
The King's Indian Attack is only effective if Black has played . . . e 7 -e6 (that's
why White occasionally uses it against the French Defense). It's interest
ing to note that Fischer often played the KIA against 2 . . . e6.

3 ... Nc6 4.g3 g6 5.d4!?

The first surprise (played for the first time in this game); White moves
the d-pawn twice in the first five moves, trying to show that Black's dark
squared weaknesses more than make up for the lost tempo. This move
soon became quite popular, and a long list of strong players have tried it
over the years.

5 ... cxd4
The usual response, but S . . . Bg7 6.dxcS has also been played quite a bit,
with results usually favoring White. 1 should mention that 6 . . . dS ! ? also
deserves consideration.

6.Nxd4 d5
At first it seems that 6 . . . Bg7 7 .NbS is just bad for Black, but it turns out
that this is a real test of White's whole concept. The complications that
arise from 7 . . . dS 8.exdS exdS 9.QxdS are quite interesting. During the
game 1 couldn't figure everything out, but 1 felt that White wouldn't be
in any real danger if he played carefully, so 1 was willing to enter this
sharp battle. The main line runs as follows: 9 . . . Qe7 + (also critical is
9 . . . QxdS 1 0 .Nc7 + Kf8 1 1 .Nxd S BfS [I also prefer White after both
1 1 . . .Bg4 1 2 .Bg2 Re8+ 1 3 .Ne3 Nd4 1 4.Na 3 , and 1 1 . . .Nd4 1 2 .Ne3 ND +
1 3 .Kd l ! Bd7 1 4.Bg2 Bc6 l S .c3 Nf6 1 6.Ke2] 1 2 .Ne3 Be4 1 3 .Bg2 Bxg2
1 4.Nxg2 Re8 I S .Ne3 , and White comes out on top) 1 0. Be2 Nb4 (I think
1 0 . . . Bg4! is Black's most dangerous move. I'll leave it up to the theory
hound to work out its ramifications.) I l .Qd2 ! ( l l . Qc4 BfS 1 2 .Nc7+ Kf8
1 3 .BgS f6 favors Black) 1 1 . . .Bg4 1 2 .Nd6+ (perhaps White retains some
chance for an advantage with 1 2 .N l a3 ! Rd8 1 3 .Qe3) 12 . . . Kf8 1 3 .D BxD
(the tempting l L .Rd8 is inaccurate : 1 4. Qxb4 BxD I S .0-0 Qxe2 [or
I S . . . BdS 1 6.Nc3 Rxd6 1 7 .BgS] 1 6.RxD Qd l + 1 7 .Kg2 Qxd6 1 8. Qxb7 , and
White is way on top) 1 4.0-0 BdS ! I S . Qxb4 Qxe2 1 6.Ne4+ ( 1 6 .NfS + leads
to a quick draw after 1 6 . . . Ne7 1 7 . Rf2 Qd l + 1 8 . RfI Qe2 1 9 . Rf2 )
1 6 . . . Ke8 ! (worse is 1 6 . . . Ne7 1 7 .Nbc3 as [not 1 7 . . . Qc4 1 8 .NxdS QxdS
1 9.Nf6 Qc6 2 0 . BgS] 1 8 .QcS Qc4 1 9.Be3 Rc8 2 0 . Qxc4 Bxc4 2 l .Rfd l ,
=

TH E H U NT FO R T H E WO RLD C H A M P I O N S H I P

when White still has some chances for a little something) 1 7 .Nd6+ Kf8 !
with a draw (risky is 1 7 . . . Kd7 1 8 .Qb5+ Qxb5 1 9 .Nxb5 Kc6 2 0.a4! Ne7
2 1 .N l c3 when White retains the initiative) . Throughout all of this,
White was being pushed around a bit. Therefore, less materialistic players should take a close look at 9.Bf4 ! ? , a blatant attempt to refute Black's
6 . . . Bg7 in no uncertain terms. The game I. Glek-Henrichs, Germany
1 995 was a ringing endorsement of 9.Bf4: 9 . . . Bxb2 1 O.Nlc3 Qe7+ I l .Be2
Ne5 1 2 .0-0 Nf6 1 3 .Na4 Bxa l 1 4. Qxa l Ned7 1 5 .Bd6, 1 -0. Of course,
Henrichs played very poorly here, and major improvements for both sides
are undoubtedly lurking behind every move . For example, instead of
10 . . . Qe7+, I can't see how White demonstrates sufficient compensation
after 1 0 . . . d4.

7.Bb5 !?
The second surprise. Instead of the natural 7.Bg2 Bg7, White plays to
isolate Black's d-pawn.

7 ... Bd7 8.exd5 Nxd4!?

It was my turn to be surprised ! I didn't expect Black to sacrifice the Ex

change in this manner.

9.Qxd4 Bxb5 10.Nc3

Avoiding 1 O.Qxh8 ? ? Qxd5 1 1 .Rg i Qf3 1 2 .Nc3 Bc5 ! 1 3 .Rg2 ( 1 3 .Qxg8+
Kd7 is also hopeless) 1 3 . . .0-0-0, with a crushing attack.

1O ... Qf6 I l .Nxb5 Qxd4 1 2 .Nxd4 0-0-0 1 3 .Be3 Rxd5 14.0-0-0

Black managed to avoid the isolated pawn, though he has fallen a bit be
hind in development. Now he should have played 14 ... Nf6, when he's very
close to complete equality. Instead, he makes a tactical error.

14 ... a6??
Trying to keep White's Knight off b5, Black finds himself hit by the final
surprise of the game.

1 5 .Nxe6! fxe6 16.Rxd5 exd5 1 7.Bd4

The point. White gets the piece back and remains a solid pawn ahead.

1 7 ... Bh6+ 18.Kd l Nf6 19.Bxf6 Re8 20.h4 BfS 2 1 .h5

161

PAL B E N KO : MY L I F E . GAM ES A N D CO M PO S I TI O N S

Simpler was 2 1 .Re l or 2 1 .Bd4, in both cases stopping Black from gener
ating any real counterplay.
1 62

2 1 ...BeS 22 .hxg6 hxg6 2 3 .f4 Re3 24.RhS+ Kd7 2S.Bh4 Rf3 26.RdS+ Ke6
Not 26 . . . Kc6? ? 2 7 .Rc8+ Kd6 2 8 . Rxc5 ! , when Ke2 will trap the black Rook.

27.ReS+ Kfi 2S.ReS Rf 1 + 29.Kd2 Rf2 + 30.Re2 Rxe2 +

Black could have put up greater resistance with 3 0 . . . Rf3 , though White
would still have every reason to expect the full point.

3 1 .Kxe2 Ke6 3 2.g4

White's extra kingside passed pawn will eventually decide the issue.

32 ... bS H .Bfl Bd6 34.Be3 Be7

Also inadequate is 34 . . . d4 3 5 .Bd2 Kd5 3 6.Kf3 Be7 3 7 .f5 g5 (no better is
3 7 . . . gxf5 3 8.gxf5 Bf6 3 9.b3 Ke5 40.Kg4) 3 8.c3 ! .

35.Kd3 BdS 36.Kc3 Be7 37.b3 Kd6 3S.Ba7 Bh4 39.BbS+ KeS 40.BeS
Bg3 41 .Bd4+, 1 -0.
The game was adjourned here, but team analysis obviously convinced my
opponent that resistance was futile.

Odcing (two yw, on Wall Slree" then th<ee mo<e ye", ,elHng
mutual funds and real estate as an independent agent) guaranteed me security, I
didn't like getting up every day at a specific time, and much preferred the free
dom of a chessplayer's life. Since I now had some money in the bank and was
comfortably ensconced in my new country, I quit my job and became the only
professional chessplayer (along with Fischer) in the United States.
With super-tournaments like Stockholm and Curaao behind me, I didn't ex
pect to participate in another such event for a long time to come. Imagine my
delight when I was invited to the first Piatigorsky Cup in Los Angeles, a tour
nament sponsored by the world-famous cellist Gregor Piatigorsky and his wife.
Both Mr. and Mrs. Piatigorsky were very nice to us, and Gregor even gave
us a cello recital next to his pool in his swimming trunks. Unfortunately, only
one more Piatigorsky Cup was held, and then they refused to do any more .
Gregor told me that the players continuously complained, and that he
couldn't make anyone happy. At one point, Mrs. Piatigorsky had to carry the
demo board to the event herself (none of the players would move a muscle to
help), and the constant demands for more and more money turned the whole
thing sour.
Though the tournament turned out to be doomed as an ongoing event, all
the players experienced the usual array of thrills, horrors, triumphs, and vexa
tions. One memory that stands out is when Soviet politics and Disneyland rammed
heads. Since we were in Los Angeles, several of the players wanted to go to
Disneyland. Keres and Petrosian, acting like delighted little kids, were very ex
cited about it since Disneyland was a legend in the Soviet Union. However,
when the day came to go there, Petrosian said he felt ill and couldn't go, and
Keres said he'd stay with Tigran.
1 63

PAL B E N KO : MY L I F E . GAM ES A N D C O M POSITI O N S

1 64

It turned out that Soviet Premier Nikita Khrushchev was visiting at that time
and also wanted to go. However, Walt Disney wasn't impressed and said, "He
can come if he buys his own ticket. " Because there was no red carpet for
Khrushchev, Disneyland was suddenly off limits to poor Keres and Petrosian
they didn't dare to go and get the Soviet politicos angry.

Piatigorsky Cup l i neup: PB, Gl igoric.

Olafsson, Najdorf, Petrosian, Reshevsky.
Keres, and Pan no.
(Photo courtesy USCF.)

Petrosian and Keres discuss the real ity

of politics with M rs. Piatigorsky.
(Photo courtesy USCF.)

M rs. and Mr. Piatigorsky with I. Kashdan,

the d i rector of the 1 963 and 1 9 66
events.

Also fresh in my mind is the insanity that

occurred in the very first round against
Reshevsky. We were both in time pressure,
and I was moving so quickly that I only
noticed that his flag had fallen after we
made the fortieth move. At the beginning
of the tournament, it was made clear that
the players didn't have to call the clock
(under FIDE rules), because a director
would always be there to point out any
such forfeiture.
Unfortunately, the director presiding
over our game (a friend of Reshevsky's !)
chose to ignore the time forfeiture. Na
tional master Ronny Gross was watching
the whole thing and was laughing, since
he watched the director stare right at the
clock and deliberately avoid making the
only legal and correct call. In fact, Gross
told me that Sammy's flag had fallen on
move 3 8 !
Once I noticed it myself, I tried to claim
the win. However, I had left a move out
on my scoresheet and it was move forty
one when I pointed to the clock, so he
snubbed my claim. Naturally, I protested,
but none of the officials supported me
since that would, in effect, point to the in
competence of the director (following the
unwritten rule that tells you to never drag
down one of your own). In the end, I lost
this game. It was the first of three straight
losses - certainly a miserable way to be
gin the event! My final score of four wins,
seven losses, and three draws was not an
auspicious start to my new, professional,
career.

A P RO FESS I O NA L AT LAST

(48) Benko - Najdorf (Piatigorsky Cup, 1 963)

l .d4 Nf6 2.c4 c5 3 .d5 d6 4.Nc3 g6 5.e4 Bg7 6.Be2
Another way to handle this position is 6.Bd3 0-0 7 .h3 e6 8 .Nf3 exd5 9.exd5
Re8+ 1 0.Be3 Bh6 1 1 .0-0 Bxe3 I I .fxe3 , with good chances for White, since
taking the pawn on e3 is very risky for Black.

6 ... 0-0 7.Nf3

I could have transposed into the Averbach King's Indian by 7 .Bg5 but, of
course, you can only play one line at a time!

7 ... e5
This takes us into a King's Indian. A main line Modem Benoni could have
occurred by 7 . . . e6 8 .0-0 exd5 9.cxd5 a6 1 0.a4 Bg4.

8.Bg5 h6 9.Bh4 g5
The tournament book considered this move to be premature, and recom
mended 9 . . . a6 1 0.0-0 g5 as more logical. However, White doesn't have to
castle short. Instead, he can play 1 0.Nd2 (saving his Bishop) or 1 0.Qc2 ,
when queenside castling becomes an option. Don't forget, Black's . . . b7-b5
advance isn't so effective here because the g7-Bishop's diagonal is closed
by the pawn on e5 (as compared to the Bishop being open and active in
the Benko Gambit).

1 0.Bg3 Nh5 1 1 .h4!?

An experiment that worked out well in this game, but the usual, and best,
line is I l .Nd2 Nf4 1 2 .0-0 Nxe2+ (else White will play Bg4) 1 3 .Qxe2 , with
a very nice game. Why is White better in this position? Because his play
on the queenside is easy to achieve (via Rfb l and b2 -b4), while Black's
kingside play depends on . . . f7 -f5 which, unfortunately, allows exf5 when
the white Knights gain access to the juicy e4-square.

1 1 ... Nf4
Another possibility is 1 1 . . . g4 1 2 .Nd2 Nxg3 1 3 .fxg3 h5 1 4.0-0, when the
weakness of the f5-square gives White good chances.
Three years after this game, Najdorf found himself facing 1 1 .h4 again.
However, this time he played very sharply: I l . . .Nxg3 1 2 .fxg3 gxh4 1 3 .Nxh4

1 65

l11;e; l Naj dorf:

I only played three games with Najdorf over the course of my career. I had

a win and a draw against him at the 1 963 Piatigorsky Cup. Our only other game was played at
Mar del Plata. It must be understood that Najdorf was an incredibly nice guy, but he would use
every trick in the book to win . During this third game, we adjourned in a position where I was up
a pawn in a Rook endgame. I didn't have a chance to analyze the position, and was thinking about
a move when we resumed on the next day. Suddenly he looked at me- his face showing pain and
some outrage at the same time - and said, "How can you do this to an old man like me? How can
you play this out? I analyzed all night and it's a dead draw! A dead draw! I guarantee it. In fact, I'll
bet you a thousand dollars it's a draw! A thousand dollars ! " I ignored him and tried to think but he
wouldn't shut up, he just kept gibbering on and on. Finally I just gave him the draw, anything to
get some quiet. Then I went upstairs to my room and looked at the position. Instantly I saw that it
was easily winning for me-he had been lying through
his teeth ! So I rushed downstairs and confronted him.
"Why did you lie to me like that? What in the hell's
wrong with you? Why didn't you let me think? "
He just smiled, put his arm around me and said,
"Don't worry about it. Come on, I'll take you out to a
nice nightclub ! " How could you stay mad at a guy like
this?
During the Piatigorsky Cup, we'd play blitz until
three in the morning every single night, though he
did cancel our late night session once when he found
a woman friend to spend some time with . I couldn't

Najdorf and I duke it out at the

Piatigorsky Cup.

figure out when he slept, it always seemed like he was

(Photo Art Zeller. courtesy USCF.)

up twenty-four hours a day!

QgS 1 4.0-0 Qxg3 I S .NfS BxfS 1 6.RxfS Nd7 1 7 . Qd2 Nf6 1 8 .Raft Nh7 ,
Cobo-Najdorf, Havana 1 966. I have t o wonder, though, why White didn't
simply defend the g3 -pawn with 1 4.Qd3 . It seems to me that White's po
sition would then be very promising.

1 2 .hxg5 hxg5 1 3 .Bfl

An interesting alternative was seen in Dzindzichashvili-Jansa, Tbilisi 1 965 :

1 3 .Kfl fS 1 4.exfS BxfS I S .Nd2 Nxe2 1 6.Qxe2 Qf6 1 7 .QhS Bd3 + 1 8 .Kg l
Bg6 1 9.Nde4 BxhS 2 0.Nxf6+ Bxf6 2 1 .RxhS, and White had an obvious
advantage and went on to win the game.

1 3 ... Bg4
Also possible was 1 3 .. .fS , but at this point I like White's position no mat
ter what Black tries.

14.Qc2 BxO 15.gxO Nd7

Black's idea is to bury White's dark-squared Bishop. Of course, taking on
f4 via Bxf4 would be great for Black after . . . exf4, since Black's Bishop would
then rule the board.

16.0-0-0 ReS 17.Bh3

This leaves Black with a difficult choice. He can either give up his good
Knight for the light-squared Bishop, or let it become powerful on fS .

A P RO F E S S I O N A L AT LAST

1 7 ...Nxh3 IS.Rxh3 Nf8

Black has decided on passive defense. I would have preferred the more
usual, and active, plan of . . . a7-a6 followed by . . . b7-b 5 .

1 9.Rdh l Ng6 20.Ndl

Heading for the dream square on f5 .

20 ... RcS
Going into complete defensive mode. More active was 2 0 . . . RbS . White's
plan is simply to occupy the hole on f5 and then penetrate with his Rooks
into Black's kingside fortress.

2 1 .Ne3 Rc7 22.Nf5 Rf8

This move is hard to understand, but in any case Black was running out of
options. The only way to get rid of the dominating f5 -Knight was by
22 . . . Ne7 , but that didn't work because of 2 3 .Nxg7 Kxg7 2 4.f4 ! , when
White's buried Bishop comes strongly into play (e.g., 24 . . . gxf4 2 5 .Bxf4
exf4 2 6.Qc3 +, and White's attack crashes through).

23 .Qd l
Tempting was 2 3 .Rh7, since 2 3 . . .f6 24.Rxg7+ Rxg7 2 5 .Nh6+ KhS 2 6 .Nf7+
KgS 2 7 .NxdS wins. However, I wanted to mate, and the line 2 3 . . . Bf6
24.Nh6+ Kxh7 2 5 .Nxf7+ Kg7 2 6.NxdS BxdS simply wasn't enough for
me. My 2 3 .Qd l intends to swing my Queen calmly over to the kingside
for a mating attack.

2 3 ... f6
Once again, 2 3 . . .Ne7 would have failed to 24.Nxg7 Kxg7 2 5 .f4! exf4 26.Bxf4
gxf4 2 7 .Qg4+ Ng6 2 S .Rh7+ Kf6 2 9. Qf5 + Ke7 3 0.Qe6 mate. Also inad
equate was 2 3 . . .Nf4 24.Rh7 Bf6 (or 24 .. .f6 2 5 .Rxg7+ Kxg7 2 6.Nh6+) 2 5 .Bxf4
exf4 26.Qg l , when the threats of 2 7 . Qxg5+ or 2 7 . Qh2 will lead to mate.

24.f4!
This allows my Queen to penetrate to the kingside.

24 ... exf4 25.Qh5 Ne5 26.Qh7+, 1 -0.

Black had no wish to experience 2 6 . . . Kf7 2 7 .Qxg7+ KeS 2 S .QxfS+ KxfS
2 9.RhS+.

1 67

PA L B E N KO : MY LI FE, GAM ES A N D C O M POSITI O N S

(49) Benko - Panno (Piatigorsky Cup, 1 963)

l .d4 Nf6 2 .c4 g6 3 .Nc3 Bg7 4.e4 d6 5.Be2 0-0 6.f4 c6
1 68

The best way to challenge the Four Pawns Attack is by 6 . . . c5 7.d5 e6.
Panno's move is too slow.

7.NfJ Nfd7
It might seem more natural to play 7 . . . Nbd7, but 8.e5 Ne8 9.Be3 gives
White too much space.

8.Be3 e5 9.fxe5
Also strong is 9.dxe5 dxe5 1 O.f5 , with good attacking chances.

9 ... dxe5 1 0.d5 f5 I l .Qb3

White shows he's ready for a sharp fight by preparing to castle long.
Besides, 1 1 .0-0 is risky since 1 1 .. .f4 1 2 .Bf2 g5 gives Black real chances on
the kingside.

1 1 ...Na6 1 2 .0-0-0

Now Black must make a difficult choice.

1 2 . . . f4
White also lays claim to a clear advantage after 1 2 ... Qa5 1 3 .dxc6 bxc6 1 4.c5+
Kh8 1 5 .Bxa6 Qxa6 1 6.Ng5 f4 17 .Bf2 . More interesting is the somewhat
surprising 1 2 . . . Ndc5 ! ? ( 1 2 . . . Nac5 isn't as good: 1 3 .Qc2 Nxe4 1 4.Nxe4 fxe4
1 5 .Ng5 Nf6 1 6.dxc6 favors White) when White has to show a good deal
of care if he wants to come away with anything from this position: 1 3 .Qa3
( 1 3 .Qc2 Nb4 1 4. Qb l Nxe4 1 5 .Nxe4 fxe4 is bad for White) 1 3 . . .fxe4
( 1 3 . . . Nxe4 1 4.Nxe4 fxe4 1 5 .Ng5 cxd 5 1 6 .cxd 5 Bf5 1 7 . Kb l is good
for White) 1 4.Ng5 (Worse is 1 4. Bxc5 [ 1 4. dxc6 Nd 3 + 1 5 . Bxd 3 exd 3
1 6.Bg5 Qe8 1 7 .Nd5 h6! appears to be quite nice for Black] 1 4 . . . Nxc5
1 5 .Qxc5 exf3 1 6.Bxf3 Bf5 1 7 .Be4 ( 1 7 .Ne4 Qb6) 1 7 . . . Qg5 + 1 8 .Kb l Qxg2)
1 4 . . . Nd3 + 1 5 .Bxd3 exd3 1 6. Rxd3 Bf5 1 7 .Rd2 cxd5 1 8 .Nxd5 Kh8 1 9.b3 h6
2 0.h4, and White is on top.

1 3 .dxc6
In the tournament book, Reshevsky said White had to do this because
1 3 .BtL allows Black to close the center with 1 3 . . . c5 . I don't completely

A P RO F E S S I O N A L AT LAST

agree with this since White still has a significant advantage after 1 3 .Bf2
c5 . However, I very much wanted to win this game, and opening the posi
tion seemed a better way to heat up the battle.

1 3 ... bxc6
Losing is B . . . fxe3 ? ? 1 4.cxd7 Bxd7 1 5 .c5+! Kh8 1 6.c6 bxc6 1 7 .Bxa6, when
White wins a piece and the game.

14.B2 Rb8
White stands better. If 14 . . . Qa5, then 1 5 .c5+ Kh8 1 6.Ng5 is dangerous
for Black ( 1 6 . . . h6 1 7 .Bxa6 Qxa6 1 8 .Ne6), while 1 4 . . . Qe7 1 5 .c 5+ Kh8
1 6.Qa3 Nc7 (and not 1 6 . . . Naxc5 ? ? 1 7 .Rxd 7 !) 1 7 .Rd6 gives White a clear
advantage according to Dolmotov.

I S .Qa3 Qc7 1 6.Rd2 Nf6 1 7.h3 Ne8

White's advantage is obvious: Black's g7-Bishop is inactive, his a6-Knight
can't move, and his c8-Bishop is stuck defending the a6-Knight.

1 8.Rhdl Bf6 19.cS Qb7

Black is trying to stop me from breaking through on the d-file, but this turns
out to be unrealistic. In the tournament book, Reshevsky pointed out that
1 9 ... Nb4 was met by 20.Bc4+ Kh8 2 1 .Qa4 (threatening to win the Knight
by a2 -a3) 2 l . . .a5 (he only gave this defense, but 2 l . . .Qb7, though still very
good for White, puts up a tougher fight) 22 .a3 Na6 2 3 .Bxa6 Bxa6 24.Rd7.

20.Bc4+ Kg7

2 1 . Bxa6
One source claimed that 2 1 .g4 fxg3 e.p. 2 2 .Bxg3 was stronger. Indeed,
this is crushing, but Black doesn't have to capture en passant. Instead,
2 1 ...Kh8 keeps him in the game. However, a bit more accurate was 2 1 .Qxa6!
Qxa6 2 2 .Rd7+ Kh8 2 3 .Bxa6 Bxa6 24.Rxa7 , when White wins a pawn but
avoids the exchange of a pair of Rooks. This prevents Black from getting
any counterplay by . . . Bfl .

2 1 . . . Qxa6 22 .Rd7+ Wi 2 3 .Rxfi+ Kxfi 24.Rd7+ Kg8 2 S .Qxa6 Bxa6

26.Rxa7
White is a clear pawn ahead and should not have any difficulty in scoring the
point. Unfortunately, my usual time trouble "disease" begins to haunt me.

1 69

PAL B E N KO: MY LI FE, GAM ES A N D CO M PO S I TI O N S

26 ... Bfl 2 7.Ne l f3! 28.Nxf3 ?

1 70

The right move was 2 S .gxf3 . Then 2 S . . . Bxh3 would give Black a protected
passed h-pawn, but in the actual game Black's g-pawn and h-pawn be
come passed !

28 ... Bxg2 29.Ngl Bg5+

Though still a pawn down, Black has managed to get compensation by
activating his pieces.

30.Kh l
I didn't like 3 0.Kc2 Nf6 3 1 .Ra4 RdS 3 2 .Be l , because 3 2 . . . Be3 wins back
the pawn and gives Black the advantage.

30 ...Nf6 3 1 .Ra4 Bd2 32 .Nge2 Bxh3

It seems that lady luck has deserted me. Now Black has a great game with
his two Bishops and two connected passed pawns. However, I didn't have
time to get depressed. The clock was still ticking and moves had to be made !

3 3 .Bg3 Nd7 34.Ra6 Re8 3 5 .h4 g5 36.Ke2 Bf4 3 7.a4 Kf7?

The only way you can win a pawn race is to push your passers ! Instead,
Black makes a couple unnecessary King moves and this loss of time costs
him dearly. Even though we were both low on time, Black should have
grabbed his h-pawn and played 3 7 . . . hS .

3 8.Ra7 Kg6?
The correct move was still 3 S . . . hS.

39.a5 h5 40.a6 h4 4 1 .Bf2

I made the time control and the game was adjourned. Imagine my delight
when I discovered that I still had winning chances!

41. ..Be6 42 .Rh7 h3 43 .Bgl g4

White also wins after 43 . . .RhS 44.a7 h2 4S.Bxh2 Bxh2 46.bS ! .

44.a7 g3 45.Nxg3 Bxg3

Black has won a piece and is ready to win another one. Am I lost? No, I
actually think White is close to winning (though Black might have a way
to escape with the half-point) ! Lady luck, who abandoned me earlier, has
staged a triumphant return and suddenly the White pawns begin to show
their strength.

A P RO F E S S I O N A L AT LAST

171

Spassky and Fischer

meet at the 1966
Piatigorsky Cup.

46.b5 cxb5
The tournament book says, "A better try was 46 . . . BtL 47.bxc6 Bxgl 4S.cxd7
Bxd7 49.RbS h2 50.RxcS h l =Q 5 1 .RgS+ Kf7 5 2 .aS=Q Bxc5, with better
chances than in the game. " I can't agree with this because 5 3 .Qd5+ Be6
54.Qxe5 ! KxgS 5 5 .Qxe6+ wins easily-White will pick up Black's second
Bishop with a series of checks.

47.c6

47 ... Bf2 ?
I don't see how White wins after 47 . . . h4! ! 4S.Rxb4 (or 4S.cxd7 Bxd7 49.Rxd7
Rxc3+ 50.Kb2 RcS ! [not 50 . . . Ra3 5 1 .RdS h2 5 2 . Bxh2 Bxh2 5 3 .aS=Q RxaS
54.RxaS Bf4 5 5 .Kc2 ! Be3 56.RbS Bc5 5 7 .Rb5, and White wins. If Black
could have gotten his Bishop to c3 he would have drawn, but White didn't
allow this to happen] 5 1 .Rb7 [5 1 .Rd3 Bel 52 .Rxh3 Bc3 + 5 3 .Kb3 RaS 54.Rh2
Bd4 5 5 .Bxd4 exd4 56.Ra2 d3 47 .Kxh4 d2 5 S .Rxd2 Rxa7, and Black draws]
5 1 ...RaS 5 2 . RbS Rxa7 5 3 .Bxa7 h2 54.RhS Kg5 , and White has nothing. An
other idea is 4S .Rc7 ! ? Rxc7 ! 49.aS=Q, but it's still a draw after 49 . . . Nf6
5 0 .Nd5 Nxd5 5 1 . exd5 Bxd5 5 2 .QdS Rxc6+ 5 3 .Kb2 Be4 54.QgS+ Kf5
5 5 .Qxg3 Rc2 +, =) 4S . . . h2 ! ! 49.Bxh2 BtL ! 50.cxd7 Bxd7 5 1 . Bxe 5 (5 1 .Rb7 Bc6
52 .RbS Bxe4+ 5 3 .Kb2 Rxc3 54.Kxc3 Bxa7, =) 5 l . ..RaS ! , and Black will draw.

PA L B E N KO ; MY L I F E , GAM ES A N D CO M PO S I TI O N S

Perhaps some intrepid reader will find a way to win for White, but at the
moment it seems that Black missed his final chance to save the game.
1 72

48.cxd7 Bxd7 49.Bxfl hZ 50.Rb8 Rxc3 +

Trying to make White's King as vulnerable as possible.

5 1 .Kxc3 h l =Q 5Z.a8=Q
The only question remaining in the game is, "Can Black find a perpetual
check? "

5Z ... Qc1 +
Naturally, 52 . . . Qf3 + 5 3 .Kb4 Qxf2 loses quickly to 54.RgS+.

53 .Kb4 Qc4+
Just as hopeless is 5 3 . . . Qd2 + 54.Kc5 Qxf2 + 5 5 .Kd6 Qd2 + 56.Qd5.

54.Ka5 QaZ + 55.Kb6 Qxf2+ 56.Kc7 Qfi 57.Rf8 Qe6

The endgame after 57 ... Qe7 5S.QdS QxdS 59.RxdS is an easy win for White.

58.Rd8, 1 -0.
His Bishop is trapped.

Santa Monica 1 9 66.

fM:as

a pm!essional plaY'" in the United States demanded that I

embrace a steady diet of open Swiss System tournaments. Playing in these events
allowed me to make a reasonable living, but also had negative effects on the
overall quality of my play. While a safe and sound style (i.e., many draws with
the occasional victory) usually leads to a good result in grandmaster round rob
ins, it's a sure recipe for disaster in open events. The open tournament calls for
as many wins as possible (the weaker competition makes this a must, but it also
wreaks havoc with your own strength), and often one can only achieve this by
taking chances and playing in a slightly incorrect manner. Thus, though I was
now a chess pro, my days as a World Championship candidate were more or less
behind me, though I hadn't completely reconciled myself to this fact at that time.
Aside from my immersion into the world of the open tournament, two events
stand out in 1 964. The first involved the question of qualification into the
Amsterdam Interzonal. The final spot in this event was to be filled by either
Bisguier or myself, depending on who won a match between us. To be honest, I
didn't really want to play in the Interzonal, but an American gentleman put up
the money for the match, so I agreed. This match began on April 26, and it was
to consist of eight games; if it ended in a draw, two additional games were to be
played. If the match went to the full ten-game limit and still ended in a draw,
then Mr. Bisguier would go to Amsterdam to play in the place for which he had
qualified in the U.S. Championship in 1 963 . However, the match only lasted for
six games, ending with the final score of Benko 41f2, Bisguier 1 1h .
The second and most important event i n 1 964 was my return to Hungary!
Amnesty had been declared for all those that had left, and I jumped at the chance
to reunite with my brother, sister, and Gizella, my fiance. Gizella and I had known
each other before I escaped Hungary in 1 95 7 , and though we had corresponded
1 73

PAL B E N KO : MY L I F E , GAM ES A N D CO M PO S I TI O N S

constantly and wanted to get married, political problems forced us to think care
fully about possible ramifications; the Hungarian government would not allow
her to leave the country (she loved it there and had no desire to emigrate) and I
1 74
was still being closely watched by the secret police.
Going back to Hungary was an intense and strange experience. Old emotional
scars opened up, paranoia rekindled, and the fact that I was now Americanized
made me feel like a fish out of water. When I first arrived I met up with an old
friend and we took a train to a well-known spa. After a tiring journey, we arrived
at the hotel and were horrified to find out
. ,
that they wouldn't let us stay because I had
i,'" "
an American passport! Apparently the close
LIFE
proximity of a missile site created this reac
i{;B:i.<;-,{ W2[:s f'
tion, and showed that Cold War mentality
was alive and well. Before returning on the
train, we had dinner, and I was sure I was
being watched. My friend insisted I was para
noid, but a military vehicle stopped us a while
later and demanded papers and explanations.
, "
Suddenly I began reliving some of the hor
ror of my Hungarian past! Visions of prison
filled my mind, and it took a long time for
me to believe that I was actually safe.
This first tentative step back to my home
land only lasted a month, but it reminded
me that I was very much a Hungarian, and
that I could no longer live without close con
On the cover of Chess Life for wi nning
tact with Gizella and my family. As a result,
the 1 9 64 U.s. Open.
yearly trips to Hungary became a normal part
(Photo courtesy useF.)
of my existence.
0;

c;<fjjBs
"

(50) Benko - R Byrne (New York, 1 964)

l .e4 e6 2 .d3
Playing against a French specialist like Byrne, I decided to avoid regular
book lines. The text move heads for a King's Indian Attack, an opening
which I have played with success many times. One example can be found
in my game against Uhlmann in Stockholm, 1 962 .

2 dS 3 .Nd2 Nf6 4.Ngf3 b6 S.eS

.

There is no hurry to make this move, but White was worried that on 5 .g3
dxe4 6.dxe4, Black could eventually play . . . Ba6. Nowadays, most players
answer 4 . . . b6 with 5 .c3 . However, back in 1 964 we were both on our own
by move five !

S ...Nfd7 6.g3 cS 7.Bg2 Qc7

BAC K TO H U N GARY

Better was 7 . . . Nc6, ganging up on e 5 . It looks like Black is winning that

pawn, but this doesn't prove to be the case: 8 .Qe2 Qc7 (Worth consider
ation is 8 . . . g5 ! ? 9.h3 Bg7) 9.0-0 (9.c4 ! ? ) 9 . . . Ncxe5 1 0.Nxe5 Nxe5 1 1 .Re i
Nc6 1 2 .Bxd5, with interesting play.

8.0-0! Nxe5?
Capturing this gambit pawn is too risky since Black is behind in develop
ment. 8 . . . Nc6 was better. White could then continue 9.Re l or 9.c4 with
good prospects.

9.Nxe5 Qxe5 10.c4!

This was the key move for opening up lines in order to exploit White's
superior development. It puts pressure on the e-file and the long diagonal
while freeing the Queen for action on either side of the board. Black no
doubt underestimated the strength of 10. c4.

1 0 ... Qd6?
Black begins to see the dangers confronting him, but fails to find the best
defense. The text move prepares for I I .cxd5 exd5 12 .Re 1 + Be6. Best, how
ever, was 1 O . . . Bb 7. There would follow I l .exd5 Bxd5 1 2 .Nc4, with a strong
attack.

1 1 .cxd5 exd5 1 2 .Nc4 Qd8 1 3 .Qh5 Bb7

Black is forced to admit the error of his ways since 1 3 . . . Be6 1 4.Re 1 gives
White the double threat of 1 5 .Rxe6+ and 1 5 .Bxd 5 .

14.Ne3
This excellent move recovers the pawn and continues the attack, but stron
ger was the fairly simple 1 4.Re l + Be7 1 5 .Bg5 Nc6 1 6.Ne 5 , winning a piece
and the game. Instead, while analyzing 14 . Re l +, my brain went into
"Morphy Drive " and discounted 1 6 .Ne5 in favor of 1 6 .Rxe7+ Nxe7
1 7 .Re l 0-0 1 8 .Bxe7 Re8 ! 1 9.Bxd8 (I failed to notice that White still wins
with 1 9 . Q e2 dxc4 2 0 . Bxb7 cxd3 2 1 . Bxa8 d2 2 2 . Rd l Rxe 7 2 3 . Rxd2)
1 9 ... Rxe 1 + 2 0.Bfl dxc4, when the game is far from easy.

14 ... Nc6?
Now Black goes down in flames. His only defense was 1 4 . . . Bd6 1 5 .Nxd5
Kf8, though Black is still in serious trouble after 1 6.Bg5 .

1 75

PAL B E N KO : MY L I F E . GAM ES A N D C O M POSIT I O N S

1 5.Nxd5 Bd6
The only move, since I S . . . Be7 1 6.Re l leaves Black with no hope at all.
1 76

1 6.Bg5 Qb8
If 1 6 . . . Qd7 1 7 .Bh3 wins the Queen.

1 7 .Rfe l + Kf8
It would appear that Black has survived the worst, but actually the assault
against his King is just beginning.

1 8.Nf6! Qc7
Of course, 1 8 . . . gxf6 allows mate in two.

1 9.Bh6!

Threatening mate in three by 20.Bxg7 +.

1 9 ...Ne5
If 19 . . . BeS, Black gets squashed by 2 0.Bxg7 + Kxg7 2 1 .QgS+ Kf8 2 2 .Qh6+
Ke7 2 3 .NdS+. Also losing is 19 . . . Ne7 20.Bxg7+ Kxg7 2 1 .Rxe 7 ! .

20.Bxb7 Qxb7 2 1 .Rxe5 Bxe5 22.Qxe5 gxh6

Although he is the Exchange ahead, Black's position is hopeless because of
his weakened kingside. White's next move, a quiet one, is quickly decisive.

2 3 .Re l Rd8
If 2 3 . . .Qc6, White plays 2 4.Ne8 ! .

24.Qf4, 1 -0.
Mate is unavoidable. For example: 2 4 . . . Rg8 (24 . . . Kg7 2 S .NhS+) 2 S .Qxh6+
Rg7 2 6.Nxh7+ Kg8 2 7 .Nf6+.

(5 1 ) Benko - Bisguier (New York playoff, 1 st Game, 1 964)

l .g3 e5 2 .c4 d6 3 .Nc3 f5
Black builds up a center similar to the formation of the Dutch Defense,
with the exception that he was able to play ... e7-eS without any muss or
fuss. Since safely playing ... e7-eS is often the ultimate aim of the Dutch,
why wouldn't everyone play in this fashion? There are two answers to this
question. The first is that, in the Dutch, White has already pushed his

Bisgu ier; left, and two Hol lywood

actor-hopefuls in the early '50s.
(Photo Nancy Roos, courtesy USeF.)

pawn to d4. Here, White can play d2 -d3 and then strike in the center
with either e2 -e4 or f2 -f4. The second answer lies in the rapidity of Black's
. . . e7-e 5 . Black has built a big center, but is he ready to defend it? It's this
point that I pursue in the actual game.

4.Bg2 Nf6 5.d4

This quick strike at Black's center forces him to show me that he can handle
the pressure I'm about to impose in that area. The alternative is 5 .d3 .

5 ... Be7
A sensible reply. White gets a nice plus after 5 . . . exd4 6.Qxd4 Nc6 7 . Qd2 ,
followed by b2-b3 and Bb2 .

6.Nh3
This move keeps the hl-a8 diagonal open and, if Black's e-pawn ever moves,
allows the Knight to jump to the very nice f4-square.

6 ... 0-0 7.0-0 Qe8

Obviously Black plans to play . . . Qh5 , initiating a kingside attack, and White
must now counter-balance the impending threat by starting action in the
center.

8.e5 !
Now any pawn capture will leave Black with the worse game.

8 ...Ne6 9.Nb5!
White's action in the center has become very dangerous and Black is al
ready experiencing serious difficulties. The alternatives left open to him
are far from satisfactory. For example: 9 . . . Qd8 I O .cxd6 Bxd6 I l .d5 Nb8
1 2 .Ng5 , is better for White due to the hole on e6. Even worse is 9 . . . Qd7
1 O.cxd6 cxd6 I l .dxe5 dxe5 1 2 .Qb3 + Kh8 1 3 .Rd l , when Black's position is
obviously poor.

9 ...Nxd4
Being aware of his positional problems, Black tries to solve them by a tac
tical sacrifice of the Exchange. However, his attack is predicated on the
assumption that White would capture the Rook immediately. Unfortu
nately, he miscalculated the sequence of White's moves.

10.Nxe7 Qh5

PAL B E N KO : MY LI FE, GAM ES A N D C O M POSIT I O N S

1 78

Black's attack looks imposing after I l .Nxa8 Nxe2+ 1 2 ,Kh l 4, with inter
esting complications (Though I think that 1 3 .cxd6! Bxh3 1 4.dxe 7 ! is ulti
mately winning for "White, many of the lines are hair-raising!). However,
why give him a chance for a knockout when I have a way to put out the
flames and get a safe, overwhelming, advantage?

1 1 .Nf4!
This move serves a dual purpose, attacking and defending at the same time.
Black's next moves are all forced.

1 1 ...exf4 1 2 .Qxd4 fxg3 1 3 .fxg3 RbS

Having admitted that the attack is over, the only recourse left open to
Black is that of sad retreat!

14.cxd6 BdS l S.Qxa7 Nd7 16.BO Qg6 1 7.Bf4

The game, of course, is over since Black is two pawns down and has a
poor position too ! However, he decides to play on and make it look better
"on paper. "

1 7 ... Bf6 lS.NdS KhS 1 9.Nxf6 Qxf6 20.Qa3 NeS 2 1 .Racl NxO + 22 .exO
Be6 2 3 .Rc7 Qd4+ 24.Be3
Obviously, "White has not chosen the best continuation since he's given
Black some faint hope due to the Bishops of opposite colors. Here, in
stead of 24.Be 3 , "White should have played the simpler and stronger 24.Qe3 :
24 . . . Qxe 3 + 2 S .Bxe3 Bxa2 26.d7 h6 2 7 .BcS Rg8 2 8 .Be7, and it's over.

24 ... QdS 2 S.Qc3 Rf7 26.Rxf7 Bxf7 2 7.Qc7 RgS 2S.Rf2 BeS
If Black had one more move he could have built up a good defense with
. . . Bc6. Of course, "White cannot allow this.

29.QcS Qe6
Better than 29 . . . Qxa2 3 0. QxfS .

30.b4 Bc6 3 1 .bS BdS 32 .Rd2 BxO H .d7

The blockade, and Black's resistance along with it, is now broken.

3 3 ... RdS
Also hopeless is 3 3 . . .4 H.d8=Q fxe3 3 S .Qxg8+ Kxg8 3 6.Rd8+ Kf7 3 7 .Qf8+
Kg6 3 8 .Rd6.

34.Qb6 Qe7 3 S .QxdS+ ! QxdS 36.BgS Qb6+ 37.Rf2 , 1 -0.

BAC K TO H U N GARY

(52) Benko - Bisguier (New York playoff, 3 rd Game, 1 964)

l .NfJ Nf6 2 .g3 dS 3 .Bg2 Nbd7 4.0-0 eS S.d3 Bd6 6.Nbd2 e6
We now have a Pirc Reversed, with Black employing Geller's System a
tempo down.

7.e4 0-0 S.Re 1 ReS 9.c3 NeS 1 0.Qe2 Be7

A much safer continuation was ... dxe4.

l 1 .exdS NxdS

From all outward appearances, it does not seem as if Black's game has any
weaknesses. Nevertheless, there are some inherent flaws in the black posi
tion (i.e., the pawn on e5 is a target and the two black Knights are vulner
able to attack by White's pawns). These will become obvious to the reader
as White proceeds to exploit them over the next few moves.

1 2 .h4! Nd7
A bit more accurate is 1 2 . . . Ne6, when 1 3 .Nxe5 ? fails to 1 3 . . . Nxb4! 1 4.cxb4
Nd4, with advantage for Black. After 1 2 . . . Ne6, White should play 1 3 .a3
with a good, flexible, position.

1 3 .a3 NfS 14.e4 Nf6

Black's two proud Knights have been forced back to ineffective squares.

l S .Bh2 Ng6 16.d4

The only weakness in White's position is now dissolved and, with this ma
neuver, he achieves an overwhelming game.

1 6 ... exd4 1 7.RxeS+ NxeS l S.Nxd4 as

Naturally Black wished to exchange his inactive Rook but, in trying to do
so, he merely assists White in carrying out his projected plan of breaking
with b4-b5 -a move that opens up the long diagonal for his Bishop. Black's
best try was 1 8 . . . Bd7.

19.hS exbS 20.exhS Bh6 2 1 .Qc3

1 79

PAL B E N KO : MY L I F E , GAM ES A N D C O M POSIT I O N S

1 80

White shouldn't avoid the trade of Queens (by 2 1 . . .Bxd4 2 2 .Qxd4 Qxd4)
because the endgame is equally bad for Black, whose queenside is cramped
and subject to attack.

2 1 ...Nf6 22 .Nc4 BcS 2 3 .Nb3 Bb6 24.a4

Fixing Black's pawn on a5 so I can surround and devour it at my leisure.

24 ...Nffi 2 S.Nxb6 Qxb6 26.Qd4 Qc7

Also miserable for Black is 26 . . . Qe6 2 7 .Nc5 Qe7 2 S.Ba3 (or 2 S.Ne4 ).

27.Rc1 Qe7 28.NxaS

White now begins to reap the rewards of his strategy, for what was merely
positional advantage now becomes a material advantage.

2 8 ...Ne6
Of course, 2 S . . . Rxa5 wasn't played due to 29.RxcS .

29.Qd2
Stopping . . . Ng5 and giving support to my Knight.

29 ... h6 3 0.Nc4
The b6-square is a perfect home for the Knight, so it hurries to get there.
My opponent finds himself in a lost position so he tries a desperate attack.

30 ... NgS 3 1 .Nb6 Bh3

The alternative is 3 1 . . .Nh3 +, but it proves insufficient after 3 2 .Bxh3 Bxh3
B .NxaS Qe4 3 4.3 .

3 2.Nxa8 Bxg2
Now B .Kxg2 ? ? allows a draw after B . . . Qe4+ 3 4.Kfl Qh 1 + 3 5 .Ke2 Qf3+.

3 3 .Bxf6
Ending Black's attack.

3 3 ... Qxf6
Also possible is 3 3 . . . gxf6 3 4.RcS+ (White can also win with 3 4.Qe3 Be4
3 5 .Kfl ) 3 4 . . . Kg7 3 5 .QdS Qe 1 + 3 6.Kxg2 Qe4+ 3 7 .Kfl , but the checks will
eventually run out and Black will have to give up.

34.Kxg2 Qf3 + 3 S.Kgl

Of course, 3 5 .Kfl ? ? allows a perpetual check.

3 S Nh3 + 36.Kfl Nxf2 37.Qf4, 1 -0.

..

BACK TO H U N GARY

(53) Bisguier - Benko (New York playoff, 4th Game, 1 964)

l .e4 c6 2 .d4 d5 3 .Nc3 dxe4 4.Nxe4 Bf5 5.Ng3 Bg6 6.NfJ Nd7 7.Bd3
This is a quiet way of meeting the 4 . . .Bf5 variation. White hopes that his
central space advantage will give him a little something.

7 ... e6 s.o-o Ngf6

Black doesn't fear 9.Bxg6, since 9 . . . hxg6 would give Black kingside chances
due to the open h-file. In that case, Black would follow up with . . . Qc7,
... Bd6, and . . 0-0-0.
.

9.c4 Be7 1 0.Re l Bxd3

It's advisable for Black to make this exchange if he wants to castle kingside.

1 l .Qxd3 0-0
More precise was 1 1 . . .Qc7, preventing White's dark-squared Bishop from
taking up residence on f4.

1 2 .Bf4 Qa5
A devious move that invites White to weaken his queenside.

B .a3
And not 1 3 .Nf5 ? ? Bb4.

B ... RfeS
This is necessary since White threatened to play Nf5 .

1 4.b4 QdS 1 5.Rab l a5 ! 1 6.b5

Of course not 1 6.bxa5 Rxa5 1 7 .Rxb 7 ? ? Nc5 ! , and Black wins the Exchange.
A more double-edged move for White would have been 1 6.c5, giving up
the d5 -square for more space and control over d6.

1 6 ... c5 1 7.d5 exd5 I S.cxd5 c4!

Black finally frees his position. The white d-pawn would have become very
dangerous if Black had made any other move at this point.

1 9.Qxc4 Nb6 20.Qd4 Nbxd5 2 1 .Bc1

White's a-pawn was hanging so he was forced into 2 1 . a4, which creates a
hole on b4, or this dismal retreat of the Bishop.

181

PAL B E N KO : MY L I F E , GAM ES AN D CO M PO S I TI O N S

2 1 ...Qb6 22.Qxb6 Nxb6 2 3 .NfS Bf8 24.Bb2 Rxe 1 + 2 S .Rxe 1 NfdS

Suddenly White is having trouble holding the balance due to the weak
nesses on a4, a 3 , b 5 , and c4.
1 82

26.Re4
This move circumvented Black's threatened . . . Nc4, and created the
counterthreat of Nh6+ ! .

2 6. . .g6 27.Ne3 Nxe3 28.fxe3

Certainly not 2 8.Rxe3 ? because of 2 8 . . . Nc4, when White loses a pawn.

28 fS 29.Re6
.

Also leaving a bad taste in White's mouth was 29.Rh4 because, though it
stops . . . Nc4, Black would have overrun White's position with 29 . . . Rc8
3 0.Bd4 Nd5 3 1 .a4 Be7 followed by . . . Rc4.

29 ...Nc4 3 0.Bd4 Bxa3 3 1 .h4

White is trying to find good squares for his misplaced Rook which, at the
moment, has nowhere useful to go. The advance of the h-pawn prepares
to loosen Black's kingside structure with h4-h5 .

3 1 ...a4 32 .hS gxhS 3 3 .NeS NxeS 34.RxeS Bf8 3 5.RxfS a3 36.RgS+ Kfi
3 7.RxhS Kg6

Material is once again even, but Black's powerful passed pawn leaves White
in a very bad state.

38.Rhl a2
Having reached a won position, I became so confident that I missed a faster
win with 3 8 . . . Ra5 3 9 .b6 Rh5 , when the resultant Bishop endgame is
resignable. Nevertheless, even after 3 8 . . . a2 the game is just a matter of
technique.

3 9.m Bg7 40.Bxg7 Kxg7 41 .Ral Kf6 42.Ke2 Ra3

It's important to contain the white King. Then my own King can calmly
stroll to the center and end White's resistance.

43 .Kd2 KeS 44.Kc1 RaS 4S.b6

Black's last move prevented 45 .Kb2 , since 45 . . .Rxb5 + 46.Kxa2 Ra5+ 47 .Kb2
Rxa l 48 .Kxa l Ke4 49.Kb2 Kxe3 forces resignation.

BAC K TO H U N GARY

45 Ra6
..

Leaving my opponent without counterplay or hope since White's King

still can't cross over to the b-file.

46.Kc2 Ke4 47.Kd2 h5 48.Ke2 Ra3 49.Kd2 h4

Zugzwang. White can't move his Rook because of . . . a2 -a l =Q, he can't play
SO.Ke2 due to SO . . . Rxe3 + , and any other King move allows . . . Kxe3 .

50.Kcl Kxe3 5 1 .Kb2 Ra6 52.Re l + Kfl , 0- 1 .

(S4) Benko - Bisguier (New York playoff, Sth Game, 1 964)

l .e4 e5 2 .Nc3 Nc6 3 .g3
I had already played this line against Keres in the Candidates Tournament
(Curaao 1 962), but the order of moves was slightly different: I .g3 eS 2 . e4
Nf6 (I was more worried about 2 . . . dS) 3 .Nc3 , and White eventually got
the better game.

3 ... Bc5 4.Bg2 d6 5 .Na4!

White seizes the opportunity to obtain what Fischer calls the "minor Ex
change. "

5 ... Bb6
Bisguier could have tried S . . . Bd4! ? , intending to meet 6.c3 with 6 . . . Bxf2 +
7 .Kxf2 b S .

6.Nxb6 axb6 7.Ne2 f5 ! ?

A double edged choice: Black challenges the center and gives himself
kingside space, but he risks weakening his King and opening up the posi
tion for White's two Bishops.

8.exf5 Bxf5 9.0-0 Nge7

In my opinion, 9 . . . Nf6 would have been a more natural move. Had my
opponent tried the interesting 9 . . . Qd7, my plan was to proceed with the
following continuation: 1 0.d4 Bh3 I I .dxeS Bxg2 1 2 .e6! Qxe6 1 3 .Nf4 Qf7
1 4.Re l + Nge7 I S .Kxg2 0-0 1 6 .Ne6 ! Qxf2 + 1 7 .Kh l , with better chances
for White.

10.d4 0-0
Much too dangerous for Black would have been 10 . . .Nb4 I 1 .BgS Nxc2
1 2 .Rc 1 Nxd4? ? 1 3 .Nxd4 exd4 1 4.Re 1 .

I l .dxe5 Nxe5?!
A far better move for Black was 1 1 . . . dxe S .

I 2 .Nd4
I avoided 1 2 .Bxb7 , since taking the pawn loses time (White doesn't want
to capture on a8 since that leaves him deathly ill on the a8-h8 diagonal)
and gives Black excellent practical chances: 12 . . . c6! (Insisting on the Ex
change sacrifice! Worse is 12 . . . Bh3 1 3 .Bg2 Bxg2 1 4.Kxg2 Qd7 I S .B .) 1 3 .4

1 83

PAL B E N KO : MY L I F E , GAM ES AN D C O M POSIT I O N S

( 1 3 .Bxa8 NB + 1 4.Kh l Qxa8 I S .Qxd6 cS 1 6. Qxe7 Re8 , is horrible for

White) 1 3 . . . Bh3 1 4.Bxa8 Qxa8 I S .Rf2 Ng4 1 6.RB cS 1 7 .Nc3 Re8 1 8 .Bd2
NfS 1 9.NbS Qe4, and White's position is ready to crack.
1 84

12 ... Bg4 1 3 .f3 Bd7 14.f4 NSc6 1 5.Nf3

It would have been sheer folly for White to allow the exchange of his ac
tive Knight, since after I S . . .Nxd4 1 6.Qxd4 Bc6, Black's game would be
eased considerably.

I S ...NfS 1 6.Bd2 Re8?

A serious error! White already threatened to place Black's King under se
rious pressure by NgS , BdS+, and QhS . The text only makes things worse.
After this mistake, Black's game approaches the brink of disaster.

1 7 .NgS ! Nh6
This move covers the f7 - and h 7 -squares.

18.BdS+ Kh8 1 9.Bc3

Now White threatens to play QhS followed by Qxh6.

1 9 ... NeS
The only move that allows Black to continue the battle.

20.Qd2
Also good is 20.QhS Neg4 2 1 .Nf7+ (or 2 1 .BB Nf6 2 2 .Qxh6 gxh6 2 3 .Nf7+
Kg7 24.Nxd8 with an obvious advantage.) 2 1 . . .Nxf7 2 2 . Qxf7 Nf6 (A forced
move that allows Black to escape into a poor endgame.) 2 3 .Rfe l Rf8 24.Bxf6
Qxf6 2 S .Qxf6 Rxf6 2 6.Re7.

20 ... Neg4?
Black's only try was 2 0 . . . Ng6, in order to meet 2 1 .Qd4 with 2 1 . . .NeS .
However, White could still play 2 2 .Bxb7, when the end would merely be
a question of time. After Black's 20 . . .Neg4, there is no further hope for
him.

2 1 .Qd3 , 1 -0.
Black loses copious quantities of material after 2 1 . . .BfS 2 2 . QxfS NxfS
2 3 .Nf7+ Kg8 24.Nxd8+.

BACK TO HU NGARY

(55) Benko - Suttles (Boston, 1964)

1 .NO g6 2 .e4 Bg7 3 .d4 d6

Suttles eventually became famous for his original and inspired handling
of these systems.

4.Nc3 c6
This opening has several modern names but can sail just as well under its
old moniker, Fianchetto del Rey. Of course, more common is 4 . . . Nf6, taking
us into a classical line of the Pirc.

S.Be2 Nd7 6.0-0 Nh6

This is an awkward way of developing the
KN, though very typical of Suttles. How
ever, 6 . . . Ngf6 no longer works: 7.e5 dxe5
S.dxe5 Ng4 (also good for White is S . . . Nd5
9.Nxd5 cxd5 1 O.Qxd5 Nxe5 1 1 . QxdS+ KxdS
1 2 . Rd l +) 9.e6 fxe6 l O .Ng5 . Apparently,
Black has missed the right moment for
. . . Ngf6. But does he really need that move?
This game is no proof that he does, in spite
of the strangeness of the Knight's placement
on the rim.

7.h3 0-0 8.Be3

Curiously, White already has a threat:
9.Qd2 wins a piece.

Duncan Suttles (left) and Bernard

Zuckerman at the I 966 U.S.
Championship.
(Photo courtesy USeE)

8 ... f6 9.Qd2 Nf7 1 0.dS

White starts the hostilities. He prevents the immediate . . . e7-e5 and is also
intent on meeting a later advance of the e 7 -pawn by dxe6, thereby gaining
the half-open d-file.

1 0 ... fS
This thrust is risky and yet perfectly logical. Black aims at eliminating the
enemy center pawns so he can eventually play . . . e7-e5 without exposing
the e-pawn to an undesired exchange.

l 1 .exfS gxfS 1 2 .Nd4 Bxd4 1 3 .Bxd4 cxdS 14.NxdS eS

1 85

PAL B E N KO : MY L I F E , GAMES A N D C O MPO S I T I O NS

Black has obtained his objective: he has a somewhat shaky King position
but enjoys a menacing pawn center. The sharp pros and cons promise tur
moil!
1 86

1 5.Bc3 f4
Naturally, Black prevents the crashing of his center by f2 -f4.

1 6.Bc4
Now, however, White threatens to shatter the enemy position by 1 7 .Nxf4!
exf4 1 8 .Qxd6 ! , with 1 9.Qd4 or Qxf4 to follow.

1 6 Nc5!
.

Black comes up with the right answer. Wrong are 16 ... Nf6 1 7 .Qxf4! ! exf4
( 1 7 . . . Nxd 5 1 8 . Qg3 +) 1 8 .Bxf6 ! , and 1 6 . . . Nb6 1 7 .Nxb6 Qxb6 1 8 .Rad l Rd8
1 9.Ba5 .

1 7.Nxf4
Even so, White's move looks very powerful. Not only does it pick up a
pawn, it also prevents Black from obtaining relief with 1 7 ... Ne4 or 1 7 . . . Na4.
Nonetheless, the result of this combination is rather disappointing.

1 7 Ne4
..

1 7 . . . exf4 leads to mate after 1 8 .Qd4.

1 8.Qe3 Nxc3 1 9.Qg3 +

This Zwischenzug is important, since it's the only way to avoid the loss of
a piece.

1 9 Kh8 20.Qxc3
..

So far, so good - the combination has given White a material advantage.

20 ... Ng5 2 1 .Nh5

The other Knight moves appeared to have similar consequences, but dur
ing the game I felt that 2 1 .Nh5 gave me an appreciable advantage. In ret
rospect, the best move was 2 1 .Ne2 , after which White probably emerges
with an edge.
21...Rf3!!

Now Black brilliantly starts a dangerous counterattack. It is of the rare

sort that does not spring from any apparent advantage.

BACK TO H U N GA RY

22 .Qd2
After 2 2 .gxf3 Nxh3+, White will be happy to get away with a draw. Bad is
2 3 .Kh2 (or 2 3 .Kh l ) 2 3 . . . Qh4, when Black wins: 24.Kg2 (2 4.Ng3 Nf4+
2 S .Kg l Qh3 leads to a quick mate.) 2 4 . . . QxhS 2 S .f4 (Another mating line
occurs after 2 S .Rh l ? Nf4+ 2 6.Kg l Bh3 !) 2 S . . . Nxf4+ 2 6.Kg l Qg4+ 2 7 .Qg3
Nh3 + 2 8 .Kh2 Qxc4. The best answer to 22 . . . Nxh 3 + is 2 3 .Kg2 , though
Black has at least a perpetual after 2 3 . . . Qh4! :

24.f4 (Also interesting is 24.Bf7 ! ? QgS + 2 S .Kh l Qh4 2 6.Qc4 Nf4+ 2 7 .Kgl
dS 2 8 .QbS Be6! 2 9.Nxf4 [2 9.Bxe6 QxhS] 2 9 . . . QgS + 3 0. Kh l Qh4+ 3 1 .Kg l
QgS + ! = ) 24 . . . QxhS 2 S .fxeS (2 S . Rh l ? Nxf4+ 2 6.Kgl Bh3) 2 S . . . Nf4+ 2 6.Kgl
Nh3 + (2 6 . . . Be6 ? ? fails to 2 7 .exd6+) 2 7 .Kg2 Nf4+, with a draw.

22...Rxh3 !
The pawn is recovered and Black's attack rolls on.

2 3 .Be2 Rh4 24.Rad l

White counterattacks in the center.

24 ... Bg4!
But Black continues his own plans.

25.Bxg4 Rxg4
Now Black threatens mate in two.

26.Ng3
On 2 6.f4 (to stop 2 6 . . . Nf3 + and 2 7 . . . Qh4 mate), Black wins by 2 6 . . . Qb6+
2 7 .Rf2 (both 2 7 .Qf2 Nh3 + and 2 7 .Kh2 Rh4+ 2 8 .Kg3 Ne4+ ! also lead to
quick deaths.) 2 7 . . . Nf3 +.
Note that, had White played 24.Rfd l instead of 2 4.Rad l , he could now
proceed with 2 6 . Qxd6. However, Black would have a perpetual again after 2 6 . . . Nf3 + 2 7 .Kfl Nh2 + 2 8 .Kg l etc. since 2 8 .Ke l or 2 8 .Ke2 both lose
to 2 8 . . . Re4+ ! .

26 ... Qb6!
Black continues to play brilliantly. He is threatening 2 7 . . . Rxg3 , and also has
in mind a combination that renders the loss of the d-pawn rather harmless.

27.Qxd6 Nf3 + ! 28.gxf3

1 87

PAL B E NKO : MY L I FE, GAMES AND C O MP O S I T I O NS

Obviously, 2 S . Kh I Rh4 mate wasn't acceptable.

28 ... Rxg3 + 29.Kh2 Rg5!

1 88

Now Black threatens mate in three, starting with 3 0 . . . RhS + . This mate
would occur after 3 0.Qxb6? ?

30.f4!
By opening f3 as an escape hatch for his King, White not only parries the
immediate threat but also stops the enemy attack altogether.

30 ... Rh5+
Perhaps the simplest choice was 30 ... exf4. White has a small but insignifi
cant edge after 30 . . . Qxd6 3 1 .Rxd6 RhS + (or 3 1 ...exf4 3 2 .Rg I ) 3 2 .Kg2 RgS+
3 3 .Kf3 RfS 3 4.Kg4. 3 1 .Qxb6 axb6 3 2 .Rd3 Rxa2 3 3 .Rb3 RcS (Also equal is
3 3 . . . b S 3 4.Rgl RhS + 3 S .Kg2 RcS 3 6.Kf3 Rxc2 3 7 .Rd3 RaS 3 S . Rd7 Rxb2
3 9.Rh I ) 3 4.Rxb6 Rxc2 3 S .Rxb7 Raxb2 3 6.Rxb2 Rxb2 3 7 .Kh3 Kg7 3 S .Kg4
Rb4 3 9.Ra l , with a draw.

3 1 .Kg2 e4
Threatens mate again, but this time the move only weakens his own King
position. He ought to resort to the liquidation just mentioned above by
either 3 1 . . .Qxd6 or 3 1 . . .RgS+.

32 .Rgl!
This move has a concealed sting.

3 2 ... Rg8+?
Still treading the wrong path. A draw would have resulted after 3 2 . . . Qxb2 ! .

H .Kfl Qb5+
Looks like a killer, but it is actually suicide. Still, there is n o good alterna
tive. The endgames reached after 3 3 . . . Rxg l + or 3 3 . . . Qxd6 are very bad
for Black.

34.Rd3!

This "desperate" move is forced but, believe it or not, it wins. White threat
ens mate in two.

34 ... Rxg1 +

BAC K TO HU N GA RY

The mate must be parried somehow, but there is no playable way of doing
so. The only move that guards simultaneously against 3 5 . Qd4+ and 3 5 . Qf6+
is 34 . . . Qxb2 , but then 3 5 .RxgS+ KxgS 3 6 .Qe6+ still leads to the death of
the black King.

3 5.Kxgl, 1 -0.
The threats of 3 6. QfS mate, 3 6.QdS+ Kg7 3 7 .Rd7+, as well as 3 6 . Qf6+
are murderous.

I have long been considered to be an endgame expert, and I learned early that to
understand the endgame properly, you have to study all the classic examples
and memorize the basic positions. For example, I was playing in Budapest in
1 9 5 2 when Botvinnik (as Black) demonstrated his knowledge of endgame theory
against Szabo.

White is two pawns up and threatens f6-f7. How can Black save this position?
Believe it or not, there is a way.

5 1 ...Rxa5 ! !
This appears to be a blunder since it loses the Exchange. However, the
resulting Exchange-down position turns out to be a dead draw.

52 .Nd7+ Bxd7 53 .Rxa5 Bxg4

This position is drawn even without Black's h-pawn.

54.Ke3 Be6 55 .Kf4 Bc4 56.Ra7 h5 57.Kg5 h4

And Black went on to draw the game. The final moves were:

5S.Kxh4 Bb3 59.Kg5 Bc4 60.Rc7 Ba2 6 1 .Rc 1 Bd5 62.Kf5 Kfi 63 .Ke5
Bb3 64.Rc7+ KfS 65 .Rb7 Bc4 66.Rb4 Ba2 67.Kf5 Bd5 6S.Kg6 Bfi+
69.Kg5 Bd5 70.Rh4 Bb3 7 1 .RhS+ Kfi 72 .Rh7+ Kf8 73 .fi Ke7 74.Kg6
Bc4 75.Rg7 Bb3 76.fB=Q+ Kxf8 77.Kf6 Ke8 7S.Re7+ KdS, 112_112.
Did Botvinnik create this idea out of thin air? No, more than 2 00 years ago,
E. del Rio analyzed an almost identical position and concluded that it was a draw.
In the following game I appear to be dead lost. However, because I was present
at the 1 95 2 Budapest tournament and witnessed the Szabo-Botvinnik game first
hand, I didn't lose heart.

189

PAL B E N KO : MY L I F E , GAMES AND C O MPO S I T I O N S

(56) Benko - Matanovic (Belgrade, 1964)

190

White to Move
I had to seal my move here. Black is threatening S 7 . . . Bg3 (and not S 7 . . . Rxf2 +
S 8 .Kxf2 Bgh S 9.Kxf3 ! , drawing). The obvious S 7 .Bxh4? fails to S 7 . . . Bd4, with
the threat of a mating attack by . . . Rg2 or . . . Rh2 . Am I doomed? If you've looked
at the previous examples, you should know how to save the game.

57.Rxe5+ !
An d not S 7 . Bxh4? Bd4 when Black wins.

57 ... Kxe5 5S.Bxh4 Kf4 59.Bf6 Rd2 60.Bc3 Rd5 6 1 .Kf2 Ke4 62.Bb4
White avoids the trap 62 .h4? RcS 6 3 .Bf6 Rc2 + 64.Kfl f2 6S .Kg2 Ke3 ! ,
when Black wins because the Bishop, blocked by its own pawn, can't move
to h4.

62 ... Rb5 63.Bd6 Rh5 64.Bc7 Rh6 65.BbS Ra6 66.Bc7, 1/Z-1f2.
See problem 1 52 in the section on endgame compositions for another look
at this type of position.

(57) Benko - Parma (Belgrade, 1964)

l .d4 Nf6 2 .c4 e6 3 .NB b6 4.g3 Bb7 5.Bg2 Be7 6.Nc3 Ne4 7.Bd2
At the time, this was an experimental try, now it's one of White's most
popular choices.

7 ... d5
A logical move, but modern theory has shown that Black's best options
are 7 . . . 0-0 8.dS fS , with interesting complications, or 7 . . . Bf6 .

S.Ne5
I tried 8.Nxe4? ! dxe4 9.NeS f6 1 0. Qa4+ c6 I l .Ng4 Qxd4 1 2 .Bc3 against
Kuijpers (Tel Aviv 1 964), but 1 2 . . . bS ! is good for Black (though the game
was drawn in 3 6). After playing this unfortunate capture on e4, I devel
oped a liking for 8 .Ne S , and I used it in several games over the years. Of
course, White has many choices here, the most popular being 8.cxd S .

S . . .Nxc3

BAC K TO H U N GA RY

I had the position after 8.Ne5 twice against Rossolimo. In the first (New
York 1 965) Black simply castled, but after 9.Nxe4 dxe4 1 O. Qc2 f5 1 1 .Bc3
Bf6 1 2 .Rd l , White was better (though I eventually managed to lose after
trying too hard for the full point). In the second game (New York 1 966),
Black varied with 8 . . . Nxd2 9.Qxd2 c6 1 O.e4 dxc4 I l .Nxc4 Nd7 1 2 .0-0,
and now he tried to make up for his spatial limitations by playing for the
win of a pawn: 1 2 . . . Ba6 1 3 .b 3 Bxc4 1 4.bxc4 Ne5 1 5 .Rfd l Nxc4 1 6 .Qe2 b 5
1 7 .a4, when I had strong compensation for the sacrificed pawn and won
the game in forty-eight moves.

9.Bxc3 0-0 1 0.cxd5 exd5

Now the pawn formation favors White, especially since Black has already
slightly weakened his queenside. White's next move aims to restrict the
development of Black's Knight.

1 l .Qa4 Bd6
I think 1 1 . . .Qd6, as Karpov played against me in Caracas 1 970, is stronger.

12 .0-0 Qe7 1 3 .Qb3

Taking aim at d5 and preventing Black's . . . c7 -c5 break.

1 3 ... c6 14.e4?
A big mistake that tosses my advantage out the window. Correct was 1 4.Re 1 ,
calmly preparing for e2 -e4. This game is actually an excellent example of
the flaws of written theory. Most opening books give Parma's play as a
good way to reach equality. None of the authors cared to take a critical
look at my fourteenth move !

14 ... c5!
This is the move I had tried so hard to avoid. I forgot that Black could
advance this pawn, since he had just placed it on c6! Now both sides play
several forced moves and the position seemingly becomes very drawish.

1 5.exd5! cxd4 1 6.Bxd4 Bxe5 1 7.Rfe l Nd7 1 8.Bxe5 Nxe5 1 9.d6 Qxd6
20.Bxb7 Rae8 2 1 .Rad l Qc7 22 .Bd5 Ng4 2 3 .Qc4 Qxc4 24.Bxc4 g6

Here many spectators thought that the game would be drawn as soon as
the required thirty moves had been played (at that time, a game had to be

191

PAL BENKO : MY L I F E , G AMES AND C O MPOSITI O NS

at least thirty moves long before a draw could be agreed). Now 2 5 .Rxe8
Rxe8 2 6.Rd7 fails to 26 . . . Ne5 .
192

25.Bb3 Nf6 26.Kg2 Rxel 27.Rxel Re8 28.Rxe8+ Nxe8 29.KB Kf8 30.Ke4
Nc7 3 1 .h4
White still has an annoying edge (better King position and Bishop versus
Knight with pawns on both sides of the board - a fact that always favors
the speedy Bishop), therefore, why not play on and torture Black a bit? I
certainly didn't risk anything by doing so. If I want to win, though, I will
have to find a way to penetrate into the enemy position. The straightfor
ward 3 1 .Ke5 wouldn't have achieved this goal since 3 1 . . .Ke7 followed by
3 2 . . . f6+ pushes me back.

3 1 ...Ke7 3 2 .f4 h6 H .Bc4

Taking my time and further restricting the enemy Knight, which can't move
to a6 or b5 now.

H ... Kf6
Black chooses a waiting policy since, if 3 3 . . . Ne6 3 4.Kd5 Kd7 3 5 .Bb5+,
White's King would penetrate on one side or the other. Also any pawn
moves on the queenside are weakening while 3 3 .. .f6 allows 3 4.h 5 ! gxh5
3 5 .Kf5 .

34.g4 Ke7 3 5.Ke5

Forcing Black to make a pawn move, since both his King and Knight are
tied up.

3 5 ... f6+ 36.Ke4 Kf8 3 7.h5 g5

This may not have been the best try, but it would have been difficult to
hold the game in any event. For example, 3 7 . . .f5 + 3 8 .Ke5 fxg4 3 9.hxg6 g3
40.Bfl Kg7 4 1 .f5 h5 42 .Kf4 h4 43 .Kg5 Nd5 44.Bh3 ! leads to a White
win. His best defense may have been 3 7 . . . Kg7 , though after 3 8.hxg6 Kxg6
3 9.Bd3 , Black would still be in trouble.

3 8.fxg5 fxg5 39.Kf5 Kg7 40.Ke5

Now my King is finally able to penetrate into Black's position. Black's domi
nated Knight and poor King leave him without counterplay, and it's no
surprise that Parma couldn't find a satisfactory defense during adjourn-

BACK TO H U NG ARY

ment. Nevertheless, the actual winning technique is quite interesting:

White's first step is to fix Black's queenside pawns so they can't run away
when I attack them. Next, I have to place my Bishop so that it keeps the
enemy Knight under wraps. Finally, I will advance my King and capture
the a7 -pawn.

40 ... Ne8 41 .Be6 Nf6 42 .b4

Starting step one: my pawns tie the enemy units down and prepare them
for future consumption.

42 ... Ne8 43 .b5 Nf6

On 43 K.f8, I would play 44.Bd7 Nc7 (White also wins after 44 . . . Ng7
45.Kf6) 45 .Kd6.
. . .

44.a4 Ne8
Now it's time for step two: my Bishop practices a bit of BDSM and ties up
his lethargic Knight.

45.Bf5 Nf6 46.Ke6 Ne8

It's time for the greed to begin. My King is primed and ready to rush in
and start the feast.

47.Be4 Nf6 48.Bf3 Ng8 49.Kd6 Kf6

Black loses his Knight after 49 . . .Kf6 50.Bd5 Ne7 5 1 .Be6 ! . A nice example
of how a Bishop can dominate a Knight.

50.Kc6 Ke7 5 1 .Kb7 Kd6 52 .Kxa7 Kc5 53 .Ka6

The easiest way to victory. On 5 3 .a5, Black might be able to disturb me
by bringing his Knight over for the defense.

53 ... Nf6 54.Bc6!

Not letting the Knight approach the queenside.

54 ... Nxg4 55.a5 bxa5 56.b6 Ne5 57.Be8, 1 -0.

This last, exciting bit of Knight domination convinced my opponent to
gIVe up.
Wanting to see more of Israel, I went on an excursion with some Hungarian
friends during the Tel Aviv Olympiad. At one point I happened across a huge,
thickly bearded Arab gentleman. His appearance was quite daunting, and I was
somewhat put off when he approached us since I had no idea what his intentions
were. He took out a camera (ah, how the heart races when a stranger in a war
torn country reaches for something!) and made it clear that he would take a pic
ture of us while sitting on a camel. "How much? " I asked in Hungarian. Sud
denly his face lit up into a smile and he said, in perfect Hungarian, "For you,
free ! "
I couldn't believe it-he was Hungarian ! Many Hungarian Jews ended up in
Israel and, clearly, many assimilated extremely well. This taught me not to put
too much faith in first impressions !

19 3

PAL BE NKO: MY L I F E . G AMES AND C O MPO S I T I O NS

(58) Benko

Darga (Tel Aviv Olympiad, 1964)

l .d4 Nf6 2 .c4 e6 3 .NB d5 4.Nc3 c5 5.cxd5 Nxd5 6.g3

194

Also common are 6 . e 3 (when 6 . . . cxd4 7 . exd4 Nc6 , transposes to the

Panov-Botvinnik of the Caro-Kann) or 6.e4 (see game 6: Benko-Lounek,
Vienna 1 947). Alekhine used to play the text, and who am I to argue with
such a player's tastes? I should mention that the Semi-Tarrasch also oc
curs from an English move order: l .c4 Nf6 2 .Nc3 c5 3 .ND e6 4.g3 d5
5 . cxd5 Nxd5 6.d4.

6 ... cxd4
The main line is 6 . . . Nc6. I discuss this opening in more detail in my notes
to Benko-Korchnoi, Curaao 1 962 .

7.Nxd5 Qxd5 8.Qxd4 Qxd4

Korchnoi tried S . . . Qb5 against me in Curaao 1 962 (see game 43).

9.Nxd4 Bb4+ 1 0.Bd2 Bxd2 + 1 l .Kxd2 Bd7

In Alekhine-Euwe, World Championship Match 1 9 3 7 (2 4th game), Black
tried 1 1 . . .Ke7 1 2 .Bg2 RdS 1 3 .Ke3 , but the second player found himself
facing problems with his queenside development. Alekhine recommended
the text move as an improvement, though neither I nor my opponent was
familiar with all of this.

12 .Bg2 Nc6 1 3 .Nxc6 Bxc6 14.Bxc6+ bxc6 1 5.Rac 1

More common is 1 5 .Rhc 1 , though this will usually amount to the same
thing as the text. After 1 5 .Rhc 1 , Black has shown that his position is defen
sible, though a groveling draw is the best he can hope for: 1 5 . . . 0-0-0+ (in
Flohr-Laurine, Goteborg 1 9 5 7 , White achieved a powerful bind after
1 5 . . . Kd7 1 6.Rc4 RabS 1 7 .Kc2 Rb5 I S .Rd 1 + Rd5 1 9.Ra4 RaS 20.Rd3 Rxd3
2 1 .Kxd3 Kd6 2 2 .Kc4 a6 2 3 .b3 Ra7 2 4.b4 e5 2 5 .Ra3 e4 2 6.Ra5 g6 2 7 .g4)
1 6.Ke3 Kb7 1 7 .Rc4 Rd6 (the game Krogius-Zaitsev, Sochi 1 963 made things
look easy for Black: 1 7 . . . Rd5 I S .Rac 1 RcS 1 9.Rb4+ Kc7 2 0.Rc3 a5 2 1 . Rbc4
Kb6 2 2 .b4 Rb5 , liz_liz) I S .Rac 1 RhdS 1 9.h4 RbS 2 0.b3 Kc7 2 1 .Ra4 Kb6
2 2 .Rc2 RbdS 2 3 .Rb4+ Kc7 24.Rbc4 RbS 2 5 .Ra4 RaS 26.g4, with a White
plus in Opocensky-Foltys, Rakovnik 1 940.

1 5 ... 0-0-0+ 16.Ke3 Kc7

BACK TO H U NG ARY

Alekhine said that Black can easily defend this Rook endgame. Is that so?
Maybe I will argue with him after all ! It seems to me that Black will have
to suffer for a long time before making a draw. During the game, a master
came up to me and remarked that grandmasters used to give draws in such
positions. I replied that grandmasters such as Rubinstein and Capablanca
used to win this kind of endgame with great regularity! That shut him up.
Of course, now I felt pressured because I had to go back to the game and
prove that I knew what I was talking about!

1 7.Rc4 Rd6 l S.Ra4 RaS 19.Rc1 a6 20.b3 Kb7 2 1 .Rb4+ Kc7 22.Rg4
Though this looks like a series of beginner attacks, it's actually a maneuver
designed to create weaknesses in Black's kingside pawn structure.

22 ... g6 23 .Rh4 hS 24.Rf4 fS 2 S.RcS RbS

Black is also under pressure after 2 5 . . . Rd5 2 6 .Rfc4 Rxc5 2 7 .Rxc5 Kd6
2 8 .Ra5 .

26.RaS RbS !
Black correctly seizes his first chance for active counterplay. I f 2 7 . Rxa6
Re 5 + 2 8 . Kf3 g5 2 9 . Rb4 g4+ , Black recovers the pawn with a good
position.

27.Rfa4
And not 2 7 .Rxa6 Re5 + 2 8 .Kf3 g5 2 9 . Rb4 g4+.

2 7 ... gS 2S.f4 gxf4+ 29.gxf4

It looks like 2 9 .Kxf4 is promising, but 29 . . . Rxa5 3 0.Rxa5 h4! (and not
30 ... Kb6 3 1 .Re5, with a passive position for Black) gives Black some drawing
chances. For example, 3 1 .Rxa6 Rd4+ 3 2 .Kf3 hxg3 , and the win is still a
long way off.

29 ... RxaS 30.RxaS Rdl !

Active defense is best in a Rook endgame, even if it costs a pawn. As men

tioned in the last note, 30 . . . Kb6 3 1 .Re5 is too passive for Black.

3 1 .Rxa6 Rhl 3 2 .b4! Kd7?

Better was 32 ... Rxh2 , but Black was worried that White would have two
connected passed pawns after 3 3 .b5 cxb5 3 4.Rxe6 Rh3 + 3 5 .Kf2 . However,

1 95

PAL BENKO : MY LI FE, G AMES AND C O MPOS I T I O NS

3 S , . . Ra3 3 6. ReS Kc6 3 7 .RxfS Rxa2 3 S.RxhS b4, turns the b-pawn into a
powerful force and gives Black excellent chances to hold the game.

3 3 .Kd4 Rd1 +
19 6

Also bad was B . . . Rxh2 3 4.KcS Rxe2 3 S . Ra 7 + KcS 3 6 .Kxc6 Rc2 + (or
36 . . . KbS 3 7 .Rb7+ KaS 3 S .Rh7 KbS 3 9.bS) 3 7 .Kd Rc4 3 S .bS Rxf4 3 9.a4 h4
40.aS Rb4 4 1 .Rh7 ! , winning.

34.KeS RdS + 3 S.Kf6 Rd4 36.e3 ! Rxb4 3 7.Ra3 ! Re4

The pawns are now even, but Black's Rook is stuck in the middle of the
board doing nursing duty to the weakling on e6. White's active King de
cides the issue.

38.Rd3 + Kc7 39.h4

White's plan is simple and unstoppable : he intends to chop off Black's h
pawn and then push his own to victory.

39 ... Kb6
The seemingly more active 3 9 . . . cS 40.KgS Kc6 4 1 .KxhS c4, fails miserably
to 42 .Rc3 KcS 43 .a3 , when Blacks counterplay, and the game, is over.

40.KgS KbS 41 .Rb3 + !

The sealed move. It's better than 4 1 . Rc3 cS 42 .a3 Rc4 ! , when Black has
far more counterplay than he deserves.

41 ...KcS
Moving the King to a4 lets White gain an important tempo with 42 .Rc3 .

42 .KxhS Rc4 43 .Rb2 Rc1 44.KgS KdS 4S.Rh2, 1 -0.

The final touch. Placing the Rook behind the passed pawn ends Black's
resistance.

(59) Benko - Tringov (Tel Aviv Olympiad, 1964)

l .d4 Nf6 2 .c4 cS 3 .dS d6 4.Nc3 g6 S .g3 Bg7 6.Bg2 0-0 7.Nh3
This move was a favorite of Robert Byrne's. I decided to borrow it as a
surprise weapon, counting on the fact that, in the days before the Internet,
games from the United States were not well known in Europe. Byrne had
already played this idea against me in San Francisco ( 1 96 1 U.S. Champi
onship) with the difference that Black had played 6 . . . a6 instead of 6 . . . 0-0.
There followed: 7 ... bS S.cxbS axbS 9.NxbS QaS+ 1 O.Nc3 Ne4 I l .Bxe4
Bxh 3 , and Black had more than enough for the pawn. This game showed
me how effective the . . . b7-bS sacrifice can be, and was one of the factors
in my eventual creation of the Benko Gambit.

7 ... Na6 8.Nf4 Nc7 9.a4

An improvement on the game R. Byrne-Lombardy, U.S. Open 1 96 3 , which
continued: 9.0-0 bS 1 0.cxbS RbS I l .a4 a6 1 2 .bxa6 Bxa6, and Black had

B AC K TO H U NG ARY

sufficient compensation for the pawn. Another early Benko Gambit type
of position !

9... Rb8
In the last round of this tournament, I played this same line against R.
Garcia since I was curious to see what improvement he might have pre
pared. That game continued: 9 . . . a6 1 0.a5 Rb8 1 1 .0-0 b5 1 2 .axb6 Rxb6
1 3 .Ra3 (Preparing to play b2-b3 , which will eventually be necessary to
defend c4.) 1 3 . . . Bd7 (Another point of 1 3 .Ra3 can be seen if Black tried
1 3 . . . e 5 : 1 4.dxe6 [ 1 4.Nd3 Nd7 1 5 .e4 is also more comfortable for White]
14 . . . fxe6 1 5 .Na4 Rb4 1 6.Rd 3 , with a clear advantage.) 1 4.b3 Qb8 1 5 .Bd2
e6 (Black can't place pressure on b 3 with 1 5 ... Qb7, because of 1 6.Na2 Rb8
1 7 .Ba5 .) 1 6.dxe6 fxe6 1 7 .Qc2 Bc6 1 8 .Bxc6 Rxc6 1 9.Ne4 Qe8 2 0.Ba5 e5
2 1 .Nd5 (and not 2 1 .Bxc7 exf4 2 2 .Nxd6 Qe7) 2 1 . . .Ne6 (Black must avoid
2 1 . . .Ncxd 5 2 2 .cxd5 Nxd 5 ? ? 2 3 . Qc4 Qf7 2 4.Ng5 , when White wins .)
2 2 .Ndxf6+ Bxf6 2 3 .Qd3 Nd4 24.Kg2 Be7 2 5 .f3 Qf7 2 6 . e 3 Nf5 (Not
2 6 . . .Nxf3 2 7 . Qd5 ! and wins.) 2 7 .Bd2 Rd8 2 8 .Nc3 Bg5 2 9 .Nd5 , and White
had a clear advantage which led to a win on the 42nd move.

1 0.0-0 eS 1 1 .Nd3
I decided to keep the position closed because, after l 1 .dxe6 fxe6 1 2 .Nb5
Nxb5 1 3 .axb5 a6 1 4.bxa6 bxa6, it would be difficult to attack the Black
center. Now Black has driven away the strong white Knight but it still has
a good square on d3 , where it can support b2 -b4 or f2 -f4.

1 1 ... aS
Prevents the queenside break via b2 -b4, so White prepares f2 -f4 instead.

12 .e4 Nd7 n .Be3 fS 14.Qd2

Maybe it would have been better to go straight into battle with 1 4.f4.

14 ... Na6
Black prepares to exchange White's powerful Knight on d 3 .

l S.BgS B f6 1 6.Bh6 Bg7 1 7.BgS Bf6 1 8.Bh6 Bg7 1 9.Bxg7

Since Black refuses to make a second-best move, White avoids the draw
and exchanges Bishops.

19 ... Kxg7 20.NbS fxe4 2 1 .Bxe4 Nf6 22 .Bg2 Nb4 2 3 .f4 e4 24.Nf2 !
At the time, I thought that this sharp move was the only way for White to
keep the advantage. I didn't trust 2 4.Nxb4 axb4, because White cannot
get his Knight back to attack the e-pawn. However, recently Fritz pointed
out the interesting 2 5 .f5 ! Bxf5 2 6.Qf4, and the d6-pawn falls.

24 ... e3
Black is forced to accept the sacrifice of the Exchange since, after 2 4 . . . Bf5
or 24 . . . Re8, there follows 2 5 . Rae l and 2 6.Nc3 , attacking e4.

2 S.Qxe3 Ne2 26.Qc3 Nxal 27.Rxal Kg8 28.Re l BfS 29.Bh3 !

1 97

PAL BENKO : MY L I FE, G AMES AND C O MPOS I T I O NS

1 98

A very important move ! White must obtain control of the vital square on
e6, otherwise Black might easily get the advantage. This was the long
range idea of the positional Exchange sacrifice. The threats now are 3 0 . g4
or 3 0.Bxf5 , followed by 3 1 .Re6.

29 ... Bxh3 30.Nxh3 Re8 3 1 .Ng5 Rxe 1 + 32 .Qxe l Qe8 3 3 .Ne6

White has achieved his goal - both my Knights are magnificently placed.
Now Black has to deal with his two attacked pawns on a5 and d6.

3 3 ... b6?
Black doesn't want to lose a pawn, but 33 ... Qe7, protecting the d-pawn,
would have offered more resistance. If then 34.Qxa5 b6, though White
has an obvious advantage here too.

34.Nxd6 Qxa4 3 5 .Qe5

Centralization!

3 5 ... Qdl + 36.Kg2 Nh5 3 7.Ne4

The Knight goes back to the kingside to stop any possibility of perpetual
check.

37 ... Re8 38.Nf2 Qc2 39.g4

Winning a piece. The game is practically over.

39 ... Nxf4+ 40.Qxf4 Qxb2 41 .d6

Winning the Exchange, since Black can't capture the Knight. Even stron
ger was 4 1 .Qc7, but I expected my opponent to resign at any time.

4 1 . .. Qe2 42.d7 Qxe6 43.dxe8=Q+ Qxe8 44.Ne4 Qe6

44 ... Qe7 would have held out a bit longer.

45.Kf3
Now that my Knight's protected, my Queen is free to wreak havoc in Black's
camp.

45 ... a4 46.Qb8+ Kg7 47.Qa7+ Qf7+ 48.Qxf7+ Kxf7 49.Nd6+ Kf6 50.Nb5
Stopping the pawn with time to spare. The rest is simple.

50 ... Kg5 5 1 .Kg3 h5 52 .h4+ Kf6 53 .gxh5 gxh5 54.Kf4, 1 -0.

of my life w" filled with tr,vel ",d che". I no longe' h,d youthful
ambitions to conquer the world, but I enjoyed seeing the many faces of the United
States by playing in tournaments in virtually every state. I also played a lot in
Europe, combining chess with my now regular visits to Hungary. Though inter
national tournaments gave me a chance to play more serious chess, the prizes at
that time were almost laughable. That's why I played in so many open events;
the prizes there, though still small, kept me solvent when mixed with some teaching
and lectures.
I have to admit that open tournaments were quite stressful. You could be do
ing well throughout, but a loss in the final game would usually put you com
pletely out of the money! This occurred in my very first open event, played in
Nebraska . Koltanowski was directing this tournament and, for reasons I never
understood, he didn't seem to like me. To make his distaste clear, he gave me the
black pieces in the final three games. Going into the last round a half-point ahead
of the field I lost to Weinstein, which allowed Bisguier to take clear first.
Another Koltanowski adventure (and typical of the madness that often appears
in this type of tournament) occurred in an event in San Antonio. I was playing
Bisguier and we both had fifteen minutes left for many moves. Arthur offered a
draw with his clock still running and I said, "Make your move and I'll consider
it. " He made a move and I calmly thought about the position until most of my
time was gone. At this point I didn't have much choice, so I accepted his offer. I
couldn't believe it when he said, "No, I want to play! " I immediately complained
to the director (Kolty, of course), but he insisted we continue . Horribly upset
(and with no time left), I lost.

199

I made a living from

chess prizes, teaching,
lectures, and simuls like
this one.

(Photos Dr R. Cantwell.)

Fortunately, there were lots of open tournaments and I played in one after
another. If you bombed in one, there was always next weekend ! Eventually I
became known as the "King of the Opens," and I even won the Triple Crown
(i.e., taking first in the U.S. Open, the National Open, and the American Open
all in one year) !

(60) Benko

Byrne (U.S. Open, 1965)

l .d4 Nf6 2 .NfJ g6 3 .c4 Bg7 4.g3 0-0 5.Bg2 d6 6.Nc3 Nc6
Initiating the Panno Variation, a line that represents one of Black's sharp
est tries against the fianchetto system.

7.0-0 Rb8
An unusual move. More common is 7 . . . a6, preparing for . . . b7-bS .

8.d5 Na5 9.Nd2 c5

This position offers Black a good deal of flexibility: he will play for . . . a7a6 followed by ... b7-b S , and can follow up with ... e7-e6 or ... e7-eS . The
one flaw with Black's game is the Knight on as , which can easily end up
out of play since it has no good way to return to the center.

TRAVEL.TRAVEL, AND MORETRAVEL

10.a3
This seemed to be worth a try at the time, though it's by no means stron
ger than the main line that results after I O,Qc2 a6 1 1 .b3 b5 1 2 .Bb2 , with
20 I

sharp play and chances for both sides.

10 ... Qc7
Black has tried several moves in this position:
1O ... Bd7 1 1 .b4 cxb4 1 2 .axb4 Nxc4 1 3 .Nxc4 Qc7
1 4.Qd3 Rfc8 is interesting; 10 ... Nd7 and 10 ... Ne8
are both a bit passive; 10 . . . e6 ! ? 1 1 .b4 Nxd5 1 2 .cxd5
Bxc3 1 3 . Rb 1 Bxd2 1 4. Qxd2 cxb4 1 5 .axb4 Nc4
1 6. Qd4 b5 1 7 .dxe6 fxe6 1 8 .Qxa7 Bb7 1 9.Bb2 Qb6
was seen in the interesting game, Filippov-Ryskin,
Moscow 1 994; 1 0 . . . b6 1 1 .b4 Nb7 which, though
it appears to give White what he wants, has proven
quite resilient in a number of games.

1 1 .Qc2
White gives up on his plan to trap Black's Knight
because 1 1 .Nb5 Qb6 1 2 .b4 Nxc4 leads to Black
winning a pawn.

Robert Byrne i n 1 9 62.

(Photo courtesy USeE)

1 1 ...a6 1 2 .b3 b5 1 3 .Bb2

We' re back to the main line (mentioned in the note to White's 1 0th move),
with the difference that White has swapped a2-a3 for . . . Qc7.

13 ... bxc4 14.bxc4 e5!

A good move that is designed to stop White's eventual plan of e2-e4 fol
lowed by 2-f4 and e4-e5- an idea that has proven effective in many games.

1 5.Rab l
And not 1 5 .dxe6 (this immediately gives the a5 -Knight access to c6)
1 5 . . . Bxe6 1 6.Nd5 ? ? Rxb2 ! .

1 5 . . .Bd7 1 6.h3 Nh5 1 7.Bal f5

Black's position would be excellent if it wasn't for that poor Knight on a5.

1 8.Rxb8 Rxb8 19.e3 Qd8 20.Ne2 e4?

Positional misjudgment. Instead, he should play 20 . . . Rb6 followed by
2 1 . . .Nb 7, getting the Knight back into the game.

2 1 .Bxg7 Nxg7 22 .Qc3 !

This move seizes the long diagonal and puts pressure on Black's misplaced
.
Knight.

22...Be8
If 22 . . . Ba4 2 3 .Nf4 Nb3 , White gains the advantage with 2 4.Nxb3 Bxb 3
(2 4 . . . Rxb3 ? ? loses to 2 5 .Qxg7+) 2 5 .Rb 1 Ba4 2 6.Rxb8 Qxb8 2 7 .Qa 5 .

2 3 .g4!

PAL BE NKO ; MY L I FE. G AMES AND C O MPOSITI O NS

202

A move with a double purpose: to secure f4 for the Knight and to open
the g-file for an attack against Black's King.

23 ... Rb7 24.gxf5 gxf5 2 5 .Nf4 Rf7 26.Rb l Nb7 27.0 exB 2 8.NxB a5
Black's plan is to bring his Knight to b 3 after . . . as -a4 and . . . NaS , but there
isn't enough time for this. A more active try would have been 2 8 . . . QaS .

29.Khl ! a4 30.Ne6
Everything has been prepared and the time is ripe for the final assault.

3 0 ... Nxe6
Black has nothing better than this capture: 3 0 . . . QaS 3 1 .Qb2 ! . By retain
ing the Queens, White places a lot of heat on the black King- threats
like Rg 1 , Nh4, and NfgS leave Black in bad shape. In comparison, 3 1 .QxaS
NxaS 3 2 .Rb8 Re7 3 3 .Nh4Ieads to the win of the Exchange with the bet
ter chances, but the game remains complicated after 3 3 . . .Nxc4 3 4.Nxg7
Kxg7 3 S .NxfS + Kf6 3 6.Nxe7 Kxe7). Other thirtieth moves are even worse:
30 ... Qe7Ioses to 3 1 .NfgS Qf6 3 2 . Qxf6 Rxf6 3 3 .Rxb7, while 30 ... Qc8 also
leads to defeat after 3 1 .NfgS Rd7 3 2 .Qf6 Bg6 3 3 .Nxg7 Rxg7 3 4.Qe6+ Qxe6
3 S .dxe6 h6 3 6.Nf7 .

3 1 .dxe6 Rg7 3 2 .Nh2 !

Black's unfortunate Knight cannot be safely protected. Now a short rush
of tactics leads to Byrne's surrender.

32...Qg5
Black tries to counter with a mate threat, but to no avail. Other moves
also lost: 32 . . . NaS 3 3 . e7 Qxe7 (3 3 . . .Rxe7 3 4.BdS+ Bf7 3 S .Rg 1 + leads to
mate) 3 4.BdS+ Kf8 (34 . . . Bf7 3 S .Rb8+) 3 S .QxaS Qxe3 3 6.Re 1 Qxh3 3 7 .Qd8
is a forced mate; 32 ... Qc8 3 3 .BdS Bc6 3 4.e7+ Bxd S + 3 S .cxd S , and there is
no defense to 3 6.Rxb7 ; 32 . . . Rxg2 3 3 .Kxg2 QgS+ (White gets a mating at
tack after 3 3 . . . Bc6+ 3 4.Nf3 Qh4 3 S . Rg 1 ! ) 3 4. Kh 1 ! Bc6+ 3 S . e4 fxe4
(3 S . . . Bxe4+ 3 6.Nf3 ) 3 6.Rg1 e 3 + 3 7 .Nf3 Bxf3 + 3 8 .Kh2 .

3 3 .Bxb7 Rxb7
An exciting moment for the spectators. But the illusion is soon shattered.

34.Rgl Bc6+ 3 5.e4 Bxe4+ 36.NO , 1 -0.

TR AVE L. T R AVEL, AND MO RE TR AVE L

(6 1) Lombardy - Benko (Seattle, 1966)

l .e4 c5 2.NB d6 3 .d4 cxd4 4.Nxd4 Nf6 5.Nc3 Nc6 6.f4 e6

This position often comes about via a different move order: l .e4 c5 2 .Nf3
d6 3 .d4 cxd4 4.Nxd4 Nf6 5 .Nc3 e6 6.f4 Nc6. The move order in the game
gives Black the option of playing an immediate 6 . . . e 5 , when 7 .Nf3 Be7
S.Bd3 0-0 leads to sharp play with chances for both sides.

203

7.Be3 Be7 8.QB

This setup for White became popular after Tal's brilliant victory in the
final match game against Larsen. White's plan is to castle long and attack
on the kingside, a common procedure in the Dragon but, at the time, some
thing new in the Scheveningen. Black's next
move is an idea that Furman tried against
Gufeld in the 3 3 rd USSR Championship.

8 ... e5
The main alternative is S . . . Qc7 9.0-0-0 a6
1 0 . g4 Nxd4 I 1 . Bxd4 e5 1 2 . fx e 5 dxe 5
1 3 . Q g3 B d 6 1 4 . B e 3 B e 6 ( 1 4 . . . Bxg4 ! ? )
1 5 .Bb5+ KfS, with an extremely sharp battle
that contains quite a bit of danger for Black.

9.Nf5
Probably a new move, but I don't think it
will attract many followers. White's best
line is now known to be 9.Nxc6 bxc6 1 0.f5
(in the Gufeld-Furman game, the continu
ation was 1 0.fxe5 dxe5 I I .Bc4 0-0 1 2 .0-0
Ng4 1 3 . Rad l Nxe 3 ! 1 4. Qxe 3 [ 1 4. RxdS
BxdS favors Black, since 1 5 .Qxe 3 ? ? is met
by 1 5 . . . B b 6 . Kasparov points out that
White should p l ay 1 4 . B xf7 + ! , when
14 . . . Kh S 1 5 . Qx e 3 Q b 6 1 6 . Qxb6 axb6
1 7 .Bc4 Bg4 gives Black sufficient positional
compensation for the sacrificed pawn]
14 . . . Qb6, and Black had a good game .)
1 0 . . . Q a 5 , and now b oth 1 1 . 0 - 0 - 0 and
I I .Bc4 have scored well for the first player.

Lombardy: Bill was a very

strong positional player- a great natu

ral talent. For some reason, he got dis
appointed in the game- perhaps it
concerned money, or the lack of it!
Bill was also a gentleman. I played
him in a San Antonio open and was a
pawn up. I felt very confident until I
looked at my clock and realized that I
had to make ten moves in less than
thirty seconds! Horrified, I offered a
draw and he accepted just as my flag
fell. Most players wouldn't have been
so kind.

William Lombardy.
(Photo courtesy USCF.)

9 ... Bxf5 lO.exf5 Nd4!

The refutation of 9.Nf5 . Black either prevents White from castling on the queenside or forces White to part with
his dark-squared bishop, after which Black has no problems.

1 1 .Bxd4 exd4 12 .Nb5

Probably 1 2 .Ne4 was slightly better. Then 1 2 . . . Qa5 + 1 3 .Kf2 0-0 1 4.Bd3
gives White good chances to hold on for a draw.

PAL BE NKO : MY L I F E . G AMES AND C O MP O S I T I O NS

12 ... Qa5+ B.c3

On 1 3 .Kd l 0-0, Black will have sufficient compensation for the dying
d-pawn.
204

B ... dxc3 14.Qxc3 Qxc3 + 1 5.Nxc3 O-O-O!

Black has a slight but definite advantage due to the weakness of \Vhite's
doubled pawns. But winning the game is far from easy because of the pres
ence of opposite-colored Bishops.

16.Bd3
Fifteen years later, Mariotti stumbled into the same bad position against
Korchnoi (Rome 1 98 1 ). However, instead of 1 6.Bd3 , he tried 1 6.0-0-0
and after 16 ... d5 1 7 .Kb l Bc5 1 8.g3 Rhe8 1 9.Bb5 Re7 20.Rhe l Rxe l 2 1 .Rxe l
a6 2 2 .Bd3 Bb4 2 3 .Re l Bxc3 24.Rxc 3 + Kd7 2 5 .a4 Kd6, Black's superior
pawn structure led to a win in forty moves.

16 ... d5 1 7.Ke2 Bb4 I S.Kf3 RheS 1 9.Racl KbS 20.Rhd l h5 2 l .h3

\Vhite played this to prevent the threatened 2 1 . . .Ng4. Inasmuch as Black
is now able to cripple \Vhite's majority, perhaps 2 1 .g3 should have been
tried, though 2 1 . . .Ng4 2 2 .h3 Re3 + 2 3 .Kg2 Nf6 24.Bc4 Bxc3 2 5 . Rxc3 Rxc3
2 6.bxc3 Kc7 still favors Black.

2 1 . ..h4 22 .Bb5 Re7 23 .a3 Bxc3

Taking advantage of the opportunity to get rid of the opposite-colored
Bishops. Black now has all the winning chances thanks to his passed pawn.

24.Rxc3 ReS 2 5.RxeS+ KxcS 26.Ba4 Ke7 27.Bb3 Kd6 2S.Ba2

TRAVEL.TRAVE L,AND MORE TRAVE L

\Vhite is compelled to offer only passive resistance. The Rook is unable

to move, as on 2 8 .Rd2 Black gets in with 2 8 . . . Re l , or if the Rook leaves
the d-file there follows 28 . . . d4 and then . . . Re3 + followed by . . . Ne4. But
Black, too, has a problem: in order to win he must penetrate with his Rook,
but as long as \Vhite does nothing to alter the position, this is not possible
as there are no penetration points available to the black forces (2 8 . . . Re4?
2 9.Bxd5). Black, therefore, finds it necessary to force open another file on
the queenside, but this operation will cost a pawn.

28 ... a5 29.Bb l b5 30.Bc2 a4 3 1 .Bd3 Rb7 32.Bc2 b4!

The time has come; Black sacrifices a pawn to open a file. \Vhite might
just as well accept the offer, as the file will be opened anyway.

H .Bxa4 Ra7 34.b3 bxa3 3 5 .Ral Rc7

In his efforts to keep the a-file
and b-file closed, \Vhite has had
to concede several squares on the
c-file.

36.Rxa3 Rc3 + 3 7.Ke2 Kc5

A tempting alternative was
37 ... Nh5 . However, there is no
need to force the position. In
stead, I calmly make sure that all
of my forces are working, and I
also make sure that \Vhite's b
pawn can't go anywhere. To put
things in more b a s i c terms :
\Vhen you reach an endgame,
make sure your King is active !

19 66 U.S. Cham pionship: Robert Byrne and

Sammy Reshevsky in front. Background:
Anthony Saidy (left) , PB, and Wi l l iam Add ison.
(Photo courtesy USCF.)

3 8.Ra2 d4 39.Rb2 Nd5

The pawns on the kingside are harmless. But \Vhite's b-pawn must not be
allowed to advance. The black King now makes a decisive entrance.

40.Kd2 Nb4! 41 .Be8 Rg3

Every black piece is superior to its white counterpart.

42 .Kdl d3 43.Bxfi
\Vhite's position, despite his two extra pawns, is hopeless in view of Black's
strong passed d-pawn and powerful centralized position. If, instead of the
text, 43 . Rf2 Kd4 44.Kd2 Re3 45 .Bxf7 , Black can choose between . . . Na2 c3 -e4 or . . . Re7-a7-a2 .

43 ...Kd4 44.Rf2 Kc3 45 .Bc4 Re3 46.RB

The threat was 46 . . . Nc2 , winning at once because of the threat of 47 . . . Re l
mate.

46...Rxf3 47. gxB Nc2

205

PAL BE N KO : MY L I F E . G AMES AN D C O MP O S I T I O NS

Winning the Bishop, as the threat of . . . Ne3 + and pushing the d-pawn is
too strong. Actually, the immediate 47 . . . d2 was even stronger, but the an206

swer I was expecting to 47 . . . Nc2 was 48. Resigns. The rest wasn't really
necessary.

4S.Bxd3 Kxd3 49.Kc1 Nb4 SO.Kh2 NdS S 1 .Ka3 Ke3 S2.Ka4 Kxf3 S3 .KbS
Kg3 S4.Kc6 Kxh3 S S .KxdS Kg4, 0- 1 .
(62) Benko - Bisguier (U.S. Championship, 1966)

1 .g3 dS
One of the most usual replies. Black, playing in classical style, tries to build
a pawn center, which is prevented by White's next move. It must be said,
however, that it is not necessarily disadvantageous for White to permit
Black his pawn center. For instance, after 2 .Bg2 e S , White can play a
Robatsch Defense Reversed with 3 .d3 , or a reversed Hyper-Accelerated
Dragon (in the Sicilian) by 3 .c4. In both cases, the extra tempo would
prove quite useful.

2 .NO Nf6 3 .Bg2 b6

Attempting to neutralize the pressure exerted by White 's powerful
fianchettoed Bishop.

4.c4 e6 S.cxdS exdS 6.0-0 Bd6

This seems more aggressive than 6 . . . Be7, but that would have been the
wiser course, as the sequel shows.

7.d4 0-0 S.Nc3 c6

While examining this position, I was surprised to discover that I was acci
dentally playing a Caro-Kann Defense with colors reversed ( 1 .e4 c6 2 .d4
dS 3 .exdS cxdS 4.Bd3 Nc6 S . d Nf6 followed by . . . g7-g6Ieads to the same
structure), in which Black has made the unnecessary and weakening move
. . . b7-b6. In this type of position, White's usual plan is to play Bf4 in order
to trade off Black's good Bishop. I decided, however, to try to exploit my
opponent's weakened queenside.

9.Ne S ! Bb7
Bad would be 9 . . . BxeS 1 0 .dxe S , as after 1O . . .Ng4 White has the choice
between two very strong moves: l 1 .f4 and I 1 .NxdS ! . I was actually ex
pecting 9 . . . Re8, which I was planning to answer with 1 O.BgS ! BxeS I I .dxeS
RxeS 1 2 .4 Re8 ( 1 2 . . . Re6 is powerfully met by 1 3 .e4 d4 1 4.eS) 1 3 .NxdS !
cxdS 1 4.Bxf6 Qxf6 I S .QxdS Nc6 1 6.Qxc6, with advantage.

1 0.BgS h6
If 1 O . . . Nbd7 I 1 .Nxd7 Qxd7 1 2 .Bxf6, the game would be as good as over.

1 1 .Bxf6 Qxf6 1 2 .f4 RdS 1 3 .e4!

This isolates White's d-pawn, but in view of his far superior development
and aggressively placed pieces, this is a minor consideration.

TRAVE L.TRAVEL, AND MORE TRAVE L

1 3 ... dxe4 14.Nxe4 Qe7 1 5.Qb3

I also considered 1 S .QhS , but the text is stronger.

1 5 ... Na6
A sad place for the Knight, but 1 S ...Nd7 ? ? isn't possible because of 1 6.Nxf7
Qxf7 1 7 .Qxf7 + Kxf7 1 8.Nxd6+. Note that Black couldn't defend the f
pawn by 1 S ...Rf8 ? due to 1 6 .Ng6.

207

1 6.Rac1

Simple chess. The c-pawn is a target so I logically bring more pressure to

bear against it. As always, tactical strength (and tactical tricks abound in
this position !) comes from positional superiority.

16 ... Rac8?
This fails miserably, but how is the c-pawn to be defended? If 1 6 ...cS ? ? ,
White wins by 1 7.Nxf7 ! Qxf7 1 8 .Qxf7 + Kxf7 1 9.Nxd6+ Rxd6 20.Bxb7. If
1 6...Bxe S , White gets a powerful initiative by 1 7 .fxeS (This is
even stronger than 1 7 . dxe S ,
Bisquier often played care
since now the Rook on f1 enters
lessly-he played by instinct without going too
the game .) 1 7 ...Rxd4 1 8 .Nd6.
deeply into a position. His openings were a bit off

ur Bisguier:

The best defense seems to be

16 ... Bc7 1 7.Rfd 1 (Black gets suf
ficient counterplay after 1 7.Nxc6
Bxc6 1 8 . Rxc6 Rxd4) 1 7 ...Bxe S
1 8 .dxeS cS 1 9.Nd6, when White
has strong pressure. And finally,
1 6 ...Nb4? ? hangs the house to
1 7.Nxd6.

beat, and rus style was a mixture of tactics and posi

tional play. I always did very well against rum.

1 7.Bh3 ! c5
Other moves also l o s e . I f
1 7 . . . Rc 7 , White wins b y
1 8 .Nxf7 ! Qxf7 1 9.Be6. The text
appears to prevent this threat be
cause 1 8 .Nxf7 can now be an
swered by 1 8 ...c4.

Bisguier (left) and Tal.

(Photo Dr. R. Cantwell.)

PAL BEN KO : M Y LI F E. G AM ES AN D C O M POSITI O N S

1 8.Bxc8

208

During the game, both Bisguier and I thought that 1 8 .Nxf7 wasn't pos
sible because of 1 8 . . . c4. However, though Black obtains some nasty-look
ing threats, this turns out to be more illusion than reality: 1 9.Rxc4 Qxe4
2 0.Nxh6+ Kh8 (Or 20 . . . gxh6 2 1 . Rxc8+ Kg7 2 2 .Rxd8 Qh l + [2 2 . . . Bd5 2 3 .Qf3
Qxd4+ 24.Qf2 Bc5 2 5 .Rd7+ Kf8 2 6 . Qxd4 Bxd4+ 2 7 .Rf2] 23 .Kf2 Qxh2 +
2 4.Ke3 Qxg3 + 2 5 .Ke2 , when Black has to admit defeat.) 2 1 .Nf7+ Kg8
2 2 .Nxd8 Qh l + (2 2 . . . Rxc4 2 3 .Qxc4+ Bd5 24.Qxd5 + Qxd5 2 5 .Be6+) 2 3 .Kf2
Qxh2 + 24.Ke l Bd5 2 5 .Bxc8 Bb4+ 2 6.Rxb4 Bxb3 2 7 .Rxb 3 , with far too
much for the Queen. As good as 1 8 .Nxf7 might be, the move I chose is
safer, easier to calculate, and completely winning.

1 8 ... Bxc8 1 9.Nxd6

The attempt to win another Exchange is too dangerous: 1 9.Nc6? Qxe4
2 0.Nxd8 Qe8 ! , and the Knight is trapped.

19 ... Rxd6 20.dxcS NxcS

It seems as though Black has some counterplay for the Exchange as his
pieces are becoming very active. But White had prepared a simplifying
maneuver that immediately clarifies the hopeless nature of Black's
position.

2 1 .Qxfi+! Qxfi 22.Nxfi Rd2

If 2 2 . . . Kxf7 , White wins back the piece with 2 3 .b4.

2 3 .NeS
Not 2 3 .b4 Bh3 ! .

2 3 . . .Rxb2 24.Rf2 Rxf2

2 4 . . . Rb4? is met by 2 5 .Nc6, but 24 . . . Rb5 might have prolonged the game,
though Black would have had no real hope.

2 S.Kxf2
The rest doesn't need comment.

2S ... Be6 26.a3 gS 2 7.Ke3 Kg7 28.Rd l as 29.Rd6 gxf4+ 30.gxf4, 1 -0.

TRAVEL.TRAVEL,AND MORE TRAVEL

(63) Benko - Medina (Havana Olympiad, 1966)

l .d4 Nf6 2 .c4 g6 3 .g3 Bg7 4.Bg2 0-0 5.Nc3 d6 6.Nfl Nbd7
Entering the classic main lines of the fianchetto King's Indian. Other sys
tems like 6 . . . cS and 6 . . . Nc6 are also very common.

7.0-0 e5 8.h3
A useful move that intends e2 -e4 (which could be played right away, of
course) followed by Be3 when Black can't harass my Bishop with . . . Ng4.

8 ... c6 9.e4
This position used to be a favorite of Fischer's (as Black) . Black realizes
his d-pawn might become weak, but he expects to compensate for this by
placing pressure against e4 ( . . . exd4 followed by . . . NcS and . . . Re8 piles up
on this pawn) .

9 ... a6
There are several ways for Black to play here, but 9 . . . QaS is the most popu
lar. Then l O.Re l exd4 1 1 .Nxd4 NeS (hitting c4) 1 2 .BfI Re8 leads to a
tension-filled struggle with chances for both sides. Black's 9 . . . a6 aims for
queenside play via a . . . b7-bS advance .

1 0.Be3 Qe7
This isn't a good mix of systems, but the most logical move, lO . . . b S , is
also met by 1 1 .cS ! , when White achieves a very promising position: 1 1 . . .b4
1 2 .Na4 Nxe4 (or 1 2 . . . dS 1 3 .BgS dxe4 1 4.NxeS Qc7 l S .Nc4, with advan
tage for White) 1 3 .cxd6 Nxd6 1 4.Nxe S Nxe S l S .dxeS Bxe S 1 6.BcS , and
Black is in serious trouble.

I l .Qc2 Re8 1 2 .Rfe l b5 1 3 .c5 !

This break usually works well against Black's kingside play. It works even
better in this particular position since Black, after 1 3 . . . dxcS 1 4.dxe S , doesn't
have the usual . . . Ne8 .

1 3 ... dxc5 14.dxe5 Nxe5 1 5.Nxe5 Qxe5 16.f4 Qc7

On 1 6 . . . Qe7, White has the strong 1 7 . e S . But now, after 1 6 , .. Qc7, White
gets his pawn back with a clearly superior position.

1 7.Bxc5 Nd7 1 8.Bf2 Bb7 19.e5

209

PAL B E N KO : MY L I F E . G AMES AN D C O MPO S I T I O N S

Threatening Ne4 and leaving Black with a tough decision. If he plays 1 9 . . . c5

then 2 0.Nd5 is strong, since he will sooner or later have to give up his b72 10

Bishop for the Knight. However, if he doesn't play ... c5 then his b7 -Bishop
will be locked in.

19 ... f5
This keeps White's Knight off e4.

20.Nd5 QeS 2 1 .Nb6

Trading off Black's main defender of the c5 -square.

2 1 ...Nxb6 22.Bxb6 Bf8 23 .Rac 1

White is firmly in control of the c5 -square and, unless Black is willing to
sacrifice a pawn by . . . c6-c5 , his light-squared Bishop will stay buried alive.

2 3 ... Qe6 24.Qb3 !

A simple solution. White plays for the endgame because he will end up
owning the d-file, thanks to his Bishop's control over dS.

24 ... RaeS 25.Qxe6+ Rxe6 26.Red l

Black has no chance of saving this position.

26 ... g5 27.Rd7 Re7 2S.Red l BaS 29.Rxe7 Bxe7 30.Rd7 ReS 3 1 .Ra7
White makes sure that Black will never be able to free himself by . . . c6-c5 .

3 1 ...gxf4 3 2 .gxf4 Bb4 H .Kf2

Black is completely helpless. His Rook is stuck defending the as-Bishop,

his c-pawn can't move, and his King is trapped on the back rank. When
you add these woes with the fact that his pawns on a6, c6, and f5 are weak
and White's passed e-pawn will soon advance and create havoc in Black's
camp, it becomes clear that a White victory can't be too far away.

H ... RbS 34.a3 Bd2

Black would have lasted a little longer with 34 . . . BfS . In that case 3 5 .Rxa6
Rxb6 3 6.RxaS (of course not 3 6. Rxb6? ? Bc5+) 36 . . . KgS 3 7 .RcS c5 3 S .Bc6
(threatening Bd7) wins, but I would have avoided the opposite-colored
Bishops and kept him hog-tied with 3 5 .Be3 Bb7 3 6.b4.

3 5 .Be7 ReS 36.e6

Now it's over, but Black feels the need to let the pain linger a bit.

TRAVE L.TRAVEL,AND MORE TRAVE L

36 ...b4 3 7.Be5 bxa3 3 8.bxa3 Re8 39.Rg7+ Kf8 40.Rxh7, 1 -0.

There's no answer to the threat of 4 1 .Rh8+. For example: 40 ... a5 4 1 .Rh8+
Ke7 42 .Bd6+ Kxd6 43 . Rxe8 .
21 1

(64) Bisguier - Benko (Manhattan Club Championship, 1967)

l .e4 e5 2.NO d6 3 .c3 Nf6 4.Bd3

At the time, this was Bisguier's favorite line against the Sicilian. Rolf Schwarz,
in his famous German book on the Sicilian, calls this system the "Ham
burger Variation, " in honor of two Hamburg masters who analyzed and
played it. I cannot resist comparing this name with the "Fried Liver Attack"
in the Two Knights Defense. Partly as the result of this game, I suggest that
the "Hamburger Variation" be renamed the "Chopped Meat Variation."

4 ... Bg4 5.Be2

The immediate 5 .h3 seems better, though Black certainly shouldn't expe
rience any problems. A couple of examples: 5 . . . Bh5 (also good is 5 . . . BxB
6. QxB Nc6 7 .Bc2 g6 8 .0-0 Bg7 9.d3 0-0 1 O.Qe2 b5 I l .Nd2 Nd7 1 2 .NB
Qb6 1 3 .a3 a5 1 4.Bg5 b4 1 5 .Qd2 bxc3 1 6.bxc3 a4 1 7 .Rab l Qa5 1 8 .c4 Bc3 ,
Galego-Kudrin, Las Palmas 1 997) 6.Bc2 Nc6 7.d3 e6 8 .Nbd2 d5 9. Qe2
Be7 1 O.Nfl c4 I l .Ng3 BxB 1 2 .QxB Ne5 1 3 .Qe3 Nxd3 + 1 4.Bxd3 cxd3
1 5 .Qxd3 dxe4 1 6.Qb5+ Qd7 1 7 .Qxd7+ Kxd7, =, Minasian-Sadler, Pula 1 997.

5 ... d5 !
This transposes to a form of the French Defense. Truly international
cuisine !

6.e5
Naturally, 6.d3 dxe4 7 . dxe4 Qxd l + is not appetizing for White .

6 ...Nfd7 7.d4 e6

Finally arriving at a French Defense that is favorable for Black. It's well
known, of course, that Black's troubles in the French stem from his prob
lem light-squared Bishop, which is usually difficult to develop effectively.
Here the Bishop has already found a good home.

8.Nbd2 exd4 9.exd4 Ne6 I O.h3 Bh5 I l .g4

PAL B E N KO : MY L I F E . G AMES AN D C O MPOSITI O N S

212

This weakening move will be a sad necessity sooner or later to relieve the
pressure against the e-pawn. After the next several moves Black can be
satisfied with his opening, as he is able to trade his bad Bishop for White's
good one.

1 l ... Bg6 12.Bxg6 hxg6 1 3 .Nb3 as

Forcing White to make an unpleasant decision: should Black be allowed
to push this pawn to a4, or should White prevent the advance with a2 -a4
and cede the b4-square to his opponent?

14.a4 Re8 l S .Bd2 Bb4 1 6.Kfl Bxd2 1 7.Qxd2 Qb6

Black's pressure against d4, b4, and b3 gives him all the chances.

1 8.Ncl Nb4 1 9.Ne l 0-0 20.Kg2 Re4 2 1 .Ne2 Rfe8 22 .b3 R4e7 23 .NO
Black is in control of the only open file, but White still has defensive re
sources which Bisguier handles very well.

23 ... Re2 24.Rael ! Qa6!

But not 24 . . . Rxd2 ? ? 2 S .Rxc8+ Nf8 (2 6 . . . Kh7 ? 2 6 .NgS + mates) 2 6 .Nxd2 ,
and the tables have turned. Another poor idea is 24 . . . Qc6 ? ? 2 S .Rxc2 Qxc2
2 6.Rc 1 , and White wins.

2 S .Nf4 Qe6
Now the variation mentioned above is not available for White. After all of
White's efforts to throw pepper in Black's eyes, Black is still in control of
the c-file.

26.Rxe2 Qxe2 27.Qxe2 Rxe2

Even though the position has simplified somewhat, Black still has
pressure.

2 8.Re l Nb8
The Knight had no future on d7, so it heads for c6 where it will eye the
d4-pawn.

29.Re3
It was better to eliminate Black's strong Rook by 29.Re2 . Black retains
some pressure after 29 . . . Nbc6 3 0.Rxc2 Nxc2 3 1 .Ne2 gS 3 2 .Kg3 f6, but
it's very difficult to break White's defenses down.

29 ... N8e6 3 0.Nd3 gS !

A strong move, preparing the center break . . . f7-f6. If now 3 1 .NcS Nxd4
3 2 .Nxb7 Nxf3 3 3 .Kxf3 Nc6, Black gets the better of the bargain because
his d-pawn is mobilized and White's e-pawn is weak.

3 1 .Nxb4 axb4 3 2 .Kfl Kf8

Also good is 3 2 . . . Rb2 , tying White's Rook down to the defense of the
pawn on b 3 .

H .Rd3
White can't escape his problems with 3 3 .Re2 Rc3 3 4.Re3 , since 3 4 .. .f6
3 S .Ke l (a better try is 3 S .Rxc3 bxc3 3 6.Ke2 Nb4, though Black's advan-

TRAVEL.TRAVE L, AND MORETRAVEL

tage can't be questioned) 3 5 . . . fxe5 3 6. dxe5 d4 3 7 . Rxc3 bxc3 3 8 .Nxg5 Ke7

is winning for Black.

3 3 ... f6 34.exf6 gxf6 3 S .h4 gxh4 36.Nxh4 Rd +

Good enough, but even stronger is 3 6 . . . Rc3 , when 3 7 .Ke2 runs into
3 7 . . . Rxb3 ! , while 3 7 .Ng6+ Kf7 3 8 .Nf4 10ses to 3 8 . . . e 5 .

3 7.KeZ Rgi 38.0 Kf7!

White's goose is cooked, a s h e is now i n Zugzwang. The Knight can't move,

the Rook can't move to d2 because of 3 9 . . . Rb l followed by 40 . . . Na5 and,
finally, the King can't go to d2 because of 3 9 . . . Rh l 4O.Ng2 Rh2 . This means
that White is forced to move his King further away from the queenside
and the weak pawn on b3 .

39.Kf2 Rb i 40.f4 NaS 41 .fS Rxb3 4Z .Rxb3 Nxb3 43 .Ke3

White's doomed pawn has fallen, but he could still have made my job more
difficult by exchanging a pawn with 43 .fxe6+. After the text, Black's passed
pawns prove to be better than White's.

43 ... eS 44.dxeS fxeS 4S .gS Nd4 46.Kd3 b6

Zugzwang again. White must weaken his pair of pawns.
47.f6
No better is 47.Ng6 ND ! .

4 7 . . . Ne6 48.KcZ NxgS 49.Kb3 Kxf6 SO.Kxb4 Ne4 S 1 .Kb S Nc3 +

SZ.Kxb6 Nxa4+
The position is now a theoretical win.

5 3 .KbS Nc3 + S4.KcS Ke6 S S .Kb4 Ne4 S 6 .Kb3 d4 S 7 . Kc4 Nd6+

S8.KcS d3 S9.NgZ
If 59.ND , simplest is 5 9 . . . Ne4+ followed by 60 . . . d2 .

S9 ...NfS 60.Kc4 dZ , 0- 1 .
(65) Benko - Bogdanovic (Sarajevo, 1967)

l .e4 cS Z .NO d6 3 .Nc3

An interesting move that forces Black to take into account several trans
positional possibilities.

213

PAL BE N KO : MY L I FE. G AMES AND COMPO SITI O N S

3 ... Nf6

214

On 3 . . .a6, White could go back into main line Sicilian theory with 4.d4
cxd4 5 .Nxd4 Nf6 6.Nc3 . However, what if Black didn't wish to play the
Najdorf (which he's now in)? Of course, White can also answer 3 . . . a6 by
4.g3 with a closed Sicilian where . . . a7 -a6 isn't a move that Black would
normally play. Another move that Black can consider is 3 . . . e 5 , stopping
d2 -d4 once and for all. It's not all happy for Black, though, because White
would answer with 4.Bc4, when the hole on d5 gives White real chances
for an advantage .

4.e5
Taking my opponent out of book and forcing him to think for himself.

4 ... dxe5 5.Nxe5 Nbd7

A good reply. Also common is 5 . . . a6.

6.Ne4
Larsen played a similar setup with the white pieces on a couple of occa
sions (Larsen-Tringov, Amsterdam 1 964, which was drawn, and Larsen
Geller, Copenhagen 1 966, which Larsen won), and the positions he reached
in both games made it worth repeating.
6 a6
..

It's now thought that 6 . . . g6, refusing to weaken the b6-square, is adequate
for equality.

7.a4 Nb6 8.a5 Nxe4 9.Bxe4

A nice position for me: I have a comfortable development and control over
the hole on b6.

9 ... Bg4 10.0

Black thinks that he's making me create weaknesses in my game, while I
was quite happy to use the opportunity to gain kingside space and grab
the initiative.

10 ... Bf5 1 1 .g4 Bd7

I think 1 1 . . .Bg6 was better. Then 1 2 .h4 Qd6 1 3 .d3 Qg3 + 1 4.Kfl h6 1 5 .Qe2
e6 1 6 . Qf2 Qxf2 + 1 7 .Kxf2 gives White a slight edge.

12 .g5 Nh5 1 3 .d4 exd4 14.Qxd4 e6 1 5 .Be3 Re8 16.Ne4

TRAVEL.TRAVEL. AND MORETRAVEL

So White managed to come out of the opening with a superior position.

Black's seemingly active pieces have been pushed back and White now has
more space on both sides of the board. Nevertheless, Black shouldn't panic.
Simple development by . . . Be? and . . . 0-0 is indicated. Instead, Black seeks
salvation in an endgame.

1 6 ... Be6 1 7.QxdS+ RxdS l S.Kf2 g6

Definitely weakening his position by making a hole on f6. It still wasn't
too late for . . . Be? and . . . 0-0.

1 9.Rad l ReS 20.Bb3 h6?

Black has no time for this. He had to move that dark-squared Bishop and
get castled !

2 1 .Nf6+! Nxf6 22.gxf6

Black is lost since his King-Rook isn't able to get into play. For example,
22 . . . Bb4 2 3 .Bb6 0-0 24.c3 traps the Bishop.

22...hS 2 3 .h4 Bh6 24.Bxh6 Rxh6 2 S.Rd2

Simple chess. Black will be doomed if White is allowed to double Rooks
and dominate the d-file. Black prevents this by trading a pair of Rooks,
but this doesn't ease the pressure.

25...RdS 26.Rhdl Rxd2 + 27.Rxd2

Black is completely helpless.

27 ... gS

Desperation, but if Black doesn't try something, White will win mundanely
on the queenside with his pawn majority. The problem for Black is that he
can't move his King to d8 (after ... Bd? followed by . . . Kd8, White picks up
a piece with Ba4) or f8 (which is answered by Rd8). Since his Bishop isn't
doing anything, and his Rook is imprisoned, White is free to improve his
position in any way he wishes (f3 -f4 followed by c2 -c4, Bc2 , and b2 -b4.
An eventual King-march across the board will also take place.O) while Black
shuttles back and forth, waiting to die.

2S.hxgS Rg6 29.f4 h4 30.Kgl RgS

Black also loses after 3 0 . . . e5 3 1 .Re2 e4 3 2 .Rh2 , followed by 3 3 . Rxh4.

2 15

PAL BE NKO: MY LI F E , G AMES AND COMPOSITI O NS

3 1 .Kh2 RhS 32 .Kh3 Bd7 3 3 .Kh2 Rh5 34.c3 h3

Now 3 4 . . . e 5 Ioses to 3 5 .Bxf7+.
2 16

3 5 .Bdl Rh4 36.Kg3 RhS 3 7.Bg4, 1 -0.

White wins the h-pawn, thereby depriving Black of any hope at all.

(66) Benko - Duckstein (Sarajevo, 1967)

l .g3 e5 2.e4 Nf6 3 .Nc3 d5 4.exd5 Nxd5 5.Bg2 Nb6 6.NfJ Ne6 7.0-0
Be7 S.a3
This move, and the idea behind it - to expand on the queenside with b2 b4 while threatening the black e-pawn by b4-b5 - is credited to Botvinnik,
who played it against Duckstein in 1 95 8 (although he played a3 on the 7th
move, without having castled) . The other main line, d3 followed by Be3 ,
doesn't give White much. Black can, of course, hinder White's queenside
plans by . . . a 7 -a s , but at the cost of weakening his own position.

S ... O-O 9.b4 Be6 1 0.d3

Winning a pawn by 1 O.b5 Nd4 I l .Nxe 5 ? ? is bad because of l l . . .Bb3 1 2 .Qe l
Nc2 .

10 ... f5 1 1 .Bb2
Again, I l .b5 Nd4 1 2 .Nxe5 fails, this time to 1 2 . . . Bf6 ! 1 3 .f4 Nb3 with a
double threat of 1 4 . . . Nxa l and 1 4 . . . Qd4+.

1 1 ...Bf6 1 2 .Nd2
Black's game appears to be all right, but this Knight maneuver points out
its drawbacks. After the game, Duckstein told me he had played this line
as Black many times, but it was not until this game that he realized how
strong White's position really was. The Knight aims for the dominating
c5 -square, when Black's b-pawn will come under heavy fire.

12 ... QeS?!
This i s now considered to be very dubious and Black's best move i s , most
likely, 1 2 . . . Bf7, intending to answer 1 3 .Nb3 with 1 3 . . . e4. However, even
after 1 2 . . . Bf7, White should come away with some advantage by 1 3 . Rb l
Rb8 1 4.Ba 1 , giving the b3 -square more support and nipping Black's . . . e5e4 tricks in the bud.

1 3 .Nb3 RdS 1 4.Ne5 BeS 1 5 .Qb3 +

Over twenty years later, it was found that the immediate 1 5 .NbS ! is even
more accurate : 1 5 . . . a6 1 6.Qb3 + Qf7 1 7 .Qxf7+ Kxf7 1 8 .Nxc7, and White
won quickly in Dzindzichashvili-Donaldson, National Open 1 98 8 .

15 ...KhS 16.Nb5 !

TRAVEL. TRAVEL. AND MORE TRAVE L

217

Black, it seems, has developed his pieces and organized a defense against
Nxb 7, but after this move he finds himself in trouble. The threat, in addi
tion to the obvious Nxc7, is 1 7 .Bxc6 followed by Nxa7 and Nxc8 . White
is not worried about giving up his light-squared Bishop, since he will be
depriving Black of his QB too, thus taking the sting out of any possible
kingside attack. Black's next move is forced.

16 ... Nd4 1 7.Bxd4

Of course, 1 7 .Nxd4 exd4 1 8 .Nxb7 Bxb7 1 9.Bxb7 Qxe2 2 0 . Rfe l was also
strong, but a player can only make one good move at a time !

1 7 ... Qxb5
After 1 7 . . . exd4 1 8 .Nxc7, White threatens to win the Exchange by 1 9.N7e6.

18.Bc3 Na8
A sad move, but it was the only way to protect the b-pawn. If 1 8 . . . c6
( 1 8 . . . Nd7 1 9.Ne6 wins the Exchange) 1 9.a4 forces Black to give away a
piece by 1 9 . . . Nxa4.

19.Ne6
After the obvious 1 9.a4 Qb6 20.a5 Qb5, Black is still kicking since 2 1 .a6
doesn't accomplish anything due to 2 1 . . .b6. Because of this, White heads
for the endgame, which was carefully calculated and assessed.

19 ... Bxe6 20.Qxe6 e4

Virtually forced, due to the threats of 2 1 .Qxf5 and 2 1 .a4.

2 1 .a4 Qe8 22.Qxe8 Rdxe8 2 3 .Bxf6 Rxf6 24.dxe4 fxe4

P AL B E NKO : MY LI FE. G AMES AND COMPOSITI O NS

218

White has calculated that this endgame is a win for him because: 1 ) Black's
e-pawn is very weak; 2) Black's pieces will be tied up defending it; 3) White
will control the open d-file; 4) White's Bishop is a far better piece than
Black's Knight.

25 .Rfd l KgS
The threat was 2 6 . Bxe4. The counterattack by 2 5 . . . Ref8 would not have
worked because of 26.Bxe4 Rxf2 2 7 .Bf3 , trapping the Rook.

26.e3
Fixing the weak pawn.

26 ... Nb6 27.Rd4 Rfe6 2S.Rc1 c6 29.b5

"What's the threat? " you may ask. Is it always necessary to threaten some
thing? Black must make moves too, but he cannot improve his position,
while White can continually exert more and more pressure until some
thing in the enemy position cracks.

29 ... cxb5 30.axb5 RSe7

White was threatening 3 1 .Rc7 . The Knight could not be reactivated by
30 . . . Nc8 due to the winning 3 1 .Bh3 . Also losing material was 30 . . . R6e7
3 1 . Ra 1 Nc8 3 2 .Raa4, when the e-pawn falls.

3 1 .Ral g5
Once again, 3 1 . . .Nc8 ? loses to 3 2 .Bh3 .

32 .Rxa7
Threatening 3 3 .Rd8+ followed by Rb8, winning the b-pawn.

32 ... Kg7 H .Rxe4! , 1 -0.

Black, not needing to see B ... Rxe4 3 4.Bxe4 Rxe4 3 5 .Rxb7+ and 3 6.Rxb6,
gave up.

(67) Mestrovic - Benko (Sarajevo, 1967)

l .e4 g6 2.d4 Bg7 3 .Nc3 d6 4.Be2 c6

The original form of this opening for Black included the early develop
ment of the KN to f6. The newer form, that is, the Robatsch, leaves the
Knight at home so the g7 -Bishop and the Queen (after . . . Qb6) can exert
pressure on d4. This somewhat limits White's choices.

S.NO Bg4
Keeping up the pressure on d4.

6.h3 Bxf3 7.Bxf3 Qb6 S.Ne2

If 8.Be3 , Black can take the b-pawn without risk.

S ... Nf6 9.0-0 Nbd7 10.Bf4

This is somewhat naive, but my young Yugoslav opponent loved to attack
at all costs during this phase of his career.

1 0 ... e5

TRAVE L.TRAVEL, AND MORE TR AVE L

He was hoping for 1 O . . . Qxb2 1 1 .e S dxeS 1 2 .Rb l , with a sharp position

that suited his tastes. I avoided this continuation, intending to enter quiet
lines that he wouldn't be comfortable with. However, his whole concept
was probably unsound and I could have taken up the gauntlet (after 1 2 .Rb 1 )
by 1 2 . . . Qxa2 1 3 .dxeS NdS , when White has insufficient compensation for
the sacrificed material.

1 1 .dxe5 dxe5 12 .Be3 Qc7

Now, of course, 12 . . . Qxb2 1 3 . Rb 1 is dangerous for Black since he has not
yet castled.

1 3 .g4?!
Impetuous and impatient - this attacking move creates serious weaknesses
in his own camp (note the hole on f4). Simple development with Qd2 fol
lowed by Rad l was called for.

1 3 ... h5 14.g5 Nh7

This Knight is headed for f8 and e6 where it will eye the newly created
hole on f4, not to mention gS , d4, and c S .

1 5.Qd3 0-0 1 6.Qb3 Rfd8 1 7 .Kh2 ?

It is always dangerous for the King to be placed on the same line as the
enemy Queen. The better move was 1 7 .Kh l . At first glance, the sacrifice
1 7 .BxhS seems attractive, but after 1 7 . . . gxhS 1 8 .g6 Nhf6 1 9. Qxf7 + Kh8,
White doesn't get enough for the piece.

17 ... NhfB 1 8.Rgl Ne6 19.Rad l

White continues his development. If he tried to sacrifice with 1 9 .Bxh S , I
would have played 1 9 . . . Nd4 when White is busted. This demonstrates the
flaw with 1 7 .Kh2 , as after 20.Bxd4 exd4+ White is in check. This allows
Black to safely capture the Bishop. Better than the text was 1 9.c3 , but even
in that case Black has the initiative after 1 9 . . . Ndc S , thanks to the weak
ness of the d 3 -square.

19 ... Nd4 20.Bxd4 exd4+

White is positionally lost due to the weakened dark-squares and the vul
nerable pawns on e4, gS , and h3 .

2 1 .Kh l Nc5 22 .Qc4 b5

2 19

PAL BE NKO: MY L I F E . G AMES AND COMPO SITI O NS

Black wins a pawn with 2 2 . . . QeS, but he can afford to be tricky with the
text move since 2 3 .QxcS Bf8 wins the Queen.

2 3 .Qb4 Qe5 24.Qe l d3 ! 25.b4

220

If 2 S .cxd3 Nxd3 26.Qfl , Black wins with 26 . . . QcS , 26 . . . Qxb2 , or 26 . . . Nxb2 .

25 ... dxc2
White was hoping for 2 S . . . dxe2 2 6.Rxd8+ Rxd8 2 7 .bxcS, with some chance
for salvation due to the Bishops of opposite colors.

26.RxdS+ RxdS 27.bxc5 Rd l 2S.Qb4 Qa l , 0- 1 .

There is no defense to the threat of 2 9 . . . Rxg 1 + followed by promoting
the pawn. A poorly played game by White, but instructive for players who
live in fear of aggressive opponents who go after their King.

(68) Bisguier - Benko (USA, 1967)

l .e4 c5 2 .c3 Nf6 3 .e5 Nd5 4.NO Nc6 5.d4 cxd4 6.cxd4
In the late 1 980s (lasting through the 1 990s), the Alapin Variation of the
Sicilian (2 .c3) burst into popularity due to renewed interest in the posi
tion after 6.Bc4 Nb6 7.Bb3 d6 8 . exd6 Qxd6.

6 ... d6 7.Nc3
Black has no problems at all after this. The best move is 7 .Bc4, when 7 . . . Nb6
8 .BbS (8.Bb3 leads to interesting complications that also turn out to be
fine for Black) 8 . . . dxe S 9.Nxe S Bd7 has been shown to be fully equal. In
case of 7 . exd6, Black gets very comfortable play with 7 . . . Qxd6 8 .Nc3 Bg4
(8 . . . g6 is also good) 9.Be2 e6 1 O.h3 BhS 1 1 .0-0 Be7 1 2 .NbS Qd8 B .NeS
Bxe2 1 4.Qxe2 Qd6 1 S .Rd 1 NxeS 1 6.dxeS 0-0, Lane-Sadler, Torquay 1 998.

7 ... dxe5
And not 7 . . . Nxc3 8.bxc3 dxe S ? 9.dS .

S.dxe5 Nxc3 9.QxdS+ KxdS

Practice has shown that 9 . . . Nxd8 is a safer move, but I've always been
happy to take on d8 in this manner. Black has lost the right to castle, but
the King hopes to find shelter on c7 where it helps to support the queenside
and later joins in the attack on White's weak queenside pawns.

lO.bxc3 h6
It seems necessary to spend a tempo to prevent NgS . If 1 0 . . . Bg4, White
would gain attacking chances with 1 1 .e6 ! Bxe6 1 2 .NgS .

l l .Bb5
In Sveshnikov-Ruban, Kemerovo 1 99 5 , White preferred 1 1 .Be3 , when
1 1 . . .Bg4 1 2 . Rb 1 Kc8 B .Nd4 NxeS 1 4.f4 Nd7 1 S .h3 BhS 1 6.g4 eS 1 7 .gxhS
exd4 1 8 .Bxd4 Kc7 1 9.Bc4 Bd6 2 0 .Bxf7 Rhf8 2 1 .Bg6 Rxf4 2 2 . Bxg7 Bf8
2 3 .Bd4 BcS 24.Rfl left White with some advantage, though Black drew
without too much difficulty. Some books on theory also give a hearty nod

TRAVEL. TRAVEL, AND MORE TRAVEL

to I 1 .Bf4, claiming a White edge after l l . . .e6 1 2 .Nd2 gS 1 3 .Bg3 Bg7 1 4.Nc4
Ke7 1 5.0-0-0, Milner-Barry-Benko, Moscow 1 956. However, the followup
showed that Black is actually doing very well: I S . . . bS 1 6.Nd6 NxeS 1 7 .BxbS
Rd8 1 8 . Rhe l Rxd6 1 9 .BxeS Bxe S 2 0 . Rxe S Bb7 2 1 . 3 BdS 2 2 . Re2 Rc8
B .Rc2 , and now 2 3 . . . RcS ! (instead of my mistaken 2 3 . . . Rc7) can only be
good for Black. Finally, let's take a look
at I 1 .Bc4: 1 1 . . .e6 1 2 .0-0 BcS 1 3 .Re l Bd7
1 4. Rb l Rc8 I S .Nd2 NaS 1 6 . B d 3 Bc6
1 7 .Ne4 Bxe4 1 8 .Bxe4 b6 1 9.g4 Ke7 2 0.h4
Rhd8 , and Black is for choice, Kunte
Rashkovsky, Ubeda 1 999.

1 1 ...Bd7 1 2 .0-0 e6 1 3 .Rd l Kc7 14.c4?!

There should have been no hurry to make
this move, since Black was not yet threat
ening to take the e-pawn.

1 4 ... Bc5 1 5 .Bb2 a6 1 6.Ba4 Rhd8

Black now threatens to take the initiative
with . . . NaS . Black's only real problem is
the hole on d6. White's following maneu
ver, aimed at taking advantage of this, is
logical.

1 7.Nd2 Bd4!

The "King of the Opens" accepts

the National Open trophy in 1 966.
(Photo courtesy USeE)

This apparently further weakens d6 by

giving up the defending Bishop. My
pieces become very active, though, and this far outweighs White's control
of d6.

1 8.Bxd4 Nxd4 1 9.Bxd7 Rxd7 20.Ne4 Rad8 2 1 .Nd6

It seems that White has reached his goal. Feeling that he couldn't lose
this position, Bisguier offered a draw, which I declined. The fact of the
matter is, White's in trouble. Black was threatening 2 1 . . . Ne2+, and if White
defended by 2 1 .Kfl , then 2 1 . . .Nc6 2 2 .Rxd7+ Rxd7 1 3 .4 Rd4 is strong.

2 1 ...f6!?

22 1

PAL BE NKO: MY L I F E,G AMES AND COMPOSITI O NS

222

When declining the draw, I had in mind a different line: 2 1 . . .Nc6 2 2 .f4 f6
2 3 .Rab l Kb8 24.cS fxeS 2 S .fxeS Nxe S , and Black has safely won a pawn,
as on 2 6.Re l , Black wins with 2 6 . . . Nd3 . But this variation (which, in ret
rospect, is best) is a little tricky and, because of the fast tempo of play in
the tournament, I preferred to play something clearer. After the text, White
must lose the base of support for his proud Knight since 2 2 .f4 is not play
able because of 2 2 . . . Ne2 +.

22 .Rxd4 fxe5 2 3 .Nb5+

After 2 3 .Re4 Rxd6 24.RxeS Rd l + 2 S .Re l Rxa l 2 6. Rxa l Rd4, Black has a
great advantage. Or if 2 3 .Nxb 7, either 2 3 . . Rxd4 or 2 3 . . . exd4 favors Black.

23 ... axb5 24.Rxd7+ Rxd7 2 5.cxb5 Kb6 26.a4 Ka5 27.f3 b6

The immediate 2 7 . . . Rd4 gives White counterplay after 2 8 .Rc 1 followed
by 2 9.Rc7 . Now that this is prevented, White must look elsewhere for
active play. His next move is more or less forced.

28.Re l Kxa4 29.Rxe5 Rd5 30.Rxe6 Kxb5

This is the position I was aiming for. The outside passed pawn is very
strong, but White can still secure a draw with accurate play.

3 l .Re7 g5
Also interesting is 3 1 . . . Kc4! ?

32 .h4?
White forsakes the opportunity to chase the Black pawns by 3 2 .Re6! hS
3 3 . Rh6 KcS 3 4.RxhS bS 3 S .h4 b4, as Black's passed b-pawn appears to
rush up the board much too quickly. However, this is just an illusion, and
White can hold the position after 3 6.hxgS b3 3 7 .Kf2 ! , since 3 7 . . . b2 is now
met by 3 8 .Rh l . Having missed this possibility, White's game quickly be
comes lost.

32 ... gxh4 3 3 .Kh2 Kc4 34.Kh3 Rd4!

Naturally, if Black gives up the advanced h-pawn without a fight, it will be
easier for White to get connected passed pawns.

3 5 .Rc7+
Forcing Black's King to step in front of his pawn, for if 3 S . . . Kd3 3 6. Rb7.

TRAVEL. TRAVE L, AND MORE TRAVE L

3 5 ... Kb4 36.Rc6 b5 3 7.Rxh6 Kc3 3 8.g4

This is White's last chance to make a passed pawn, but it's already too late
to save the game thanks to Black's well-posted Rook.

38 ... hxg3 39.Kxg3 b4 40.f4 b3 41 .Kg4

The moment of truth. White, alas, realizes that he must lose time to get
his King across the 4th rank, as if he pushes his pawn at once, his King
will be too far away after the following forced variation: 4 l .fS b2 42 . Rh l
Kc2 43 .6 b l =Q 44.Rxb l Kxb l 4S.f7 RdS 46.Kg4 RfS, and wins.

41 ...b2 42 .Rhl Ra4 43 .Rb l Ral 44.Rxb2 Kxb2 45.Kf5

One tempo makes all the difference ! On 4S .S , Black catches the pawn just
in time after 45 . . . Kc3 46.6 Kd4 47 .KfS KdS 4S.f7 RfS+ 49.Kg6 Ke6.

45 ... Kc3 46.Ke5 Re l +

The rest i s easy.

47.Kd5 Rf1 48.Ke5 Kc4 49.f5 Kc5 50.Ke6 Kc6 5 1 .f6 Re l + 52.Kf7 Kd7
5 3 .Kf8 Ke6, 0- 1 .
(69) Benko - Kagan (Winnipeg, 1967)

l .e4 c6 2.d4 d5 3 .Nc3 dxe4 4.Nxe4 Bf5 5.Ng3 Bg6 6.h4 h6 7.h5 Bh7
8.NO Nd7
Nowadays 7 . . . Nf6 has become all the rage. The idea is to retain the op
tion of placing the bS-Knight on c6 after a quick . . . c6-cS . At the time of
this game, however, 7 . . . Nd7 was considered to be a forced move since NeS
was thought to be a very real threat. It took some time for the chess world
to realize that NeS weakens the d4-square and gives Black good chances
to counterattack in the center.

9.Bd3 Bxd3 1 0.Qxd3 Qc7 I l .Bd2 e6 12 .0-0-0 Ngf6 1 3 .Ne4

Also seen is 1 3 .Qe2 , but the sharp positions that arise after 1 3 . . . 0-0-0
1 4.NeS Nb6 I S .BaS ( I S .c4 and I S .Rh4 have also failed to offer White
anything) I S . . . RdS don't seem to promise White any advantage.

1 3 . . . 0-0-0 1 4.g3

223

PAL BE NKO : MY L I F E , G AMES AND C O MPO S I T I O NS

This move, threatening Bf4 and forcing a Knight exchange on e4, was
first played in Geller-Foguelman, Chile 1 96 5 . Thirty-five years later, it's
still the only way for \\!hite to play for an advantage !
224

l4 ... Nxe4
This allows Black to meet an eventual Bf4 with . . . Bd6. However, it gets
\\!hite's Queen off the sensitive d-file and gives \\!hite more control over
the critical e5 -square.

1 5 .Qxe4 N6 l6.Qe2 Bd6 l 7.c4

This is more accurate than the cautious Kb l . The King move can always
be played at a more appropriate time.

l7 ... c5 l 8.Bc3 cxd4!

This is now known to be the most accurate move. In fact, Kagan's whole
plan (2 0 . . . Rd7) is still considered to be cutting-edge theory! Over the years,
Black has often held off on the d4-capture and instead tried I S . . . RheS, but
Motylev-Slapikas, Swidnica 1 999 showed the right way for \\!hite to meet
this: 1 9.Ne5 cxd4 2 0 . Rxd4! , when the threat of 2 1 .Rhd i forces Black to
go into a slightly worse endgame with 20 . . . Bxe 5 .

1 9.Nxd4
Taking on d4 with the Rook is now harmless because of 1 9 . . . Rd7 followed
by 20 . . . RhdS . Also promising \\!hite little is 1 9.Bxd4. The game R. Byrne
Seirawan, Greenville 1 9S0 went 1 9 . . . Bc5 (Black has also done well with
1 9 . . . Qa5) 2 0 . Bxc5 Qxc5 2 1 .Ne5 Qc7 2 2 .f4 Rxd 1 + 2 3 .Rxd l RdS 24. Qf3
Rxd 1 + 2 5 .Kxd l Qd6+ 2 6.Kc2 Qd4, and Black drew comfortably in thirty
five moves.

19 ... a6 20.Nf3 Rd7

Black's Rooks rush to get to the important d-file. Now 2 1 .Ne5 Bxe5 2 2 . Bxe5
Qc6 2 3 .Rxd7 Nxd7 , is fine for Black.

2 1 .Kb l Rhd8 22 .a3 Qe6

Also tempting is 22 . . . Qc5 ! ? , eyeing h5 and preparing to swing the Queen
over to f5 .

2 3 .Rh4
This keeps Black's pieces off e4. Now Black, apparently tiring of the unre
solved situation in the center, decides to clarify matters with a few trades.

23 ... Be7
Perfectly playable, but 2 3 . . .Qc5 ! ? still deserves serious consideration.

24.Rxd7 Rxd7 2 5.Ne5 Bxe5 26.Qxe5 Ne8 27.Qe2 6 28.Re4 e5

This locks out the Bishop and threatens . . . Nd6. The alternative was 2 S . . . Nd6
2 9 . Rxe6 Qxc4 2 9 . ReS+ Kc7 , when \\!hite's advantage is very small.

29.Bb4 b6! ?

TRAVE L . TRAVEL, AN D MORE TRAVE L

225

An interesting idea. He intends 3 0 . . . aS followed by . . . Nd6, but I'm able to

disrupt his plans (whether this actually gives me anything concrete is an
other matter). Black's safest move was 29 . . . Nd6, when 30.Bxd6 Qxd6 3 1 .Kc2
would have only offered White a microscopic edge.

30.cS !
Did my opponent miss this move, or did he overlook my 3 4th? Now
3 0 . . . bxcS 3 1 .BxcS QxcS ? ? 3 2 .Rc4 picks up the Queen, but 3 0 . . . bS 3 l .f4
Rd4! 3 2 . Rxd4 exd4 seems like a good choice.

30 ... aS?
As mentioned in the previous note, Black had to play 30 . . . bS. Instead, he

thinks he sees a stronger continuation.

3 1 .cxh6
Threatening to win Black's Queen by Rc4.

3 l ... Rd l +?
This fails tactically. Forced was 3 1 . . . Qxb6, though White would have had
a clear advantage due to Black's exposed King.

32 .Ka2 QdS+
Also bad is 32 . . . Rc 1 3 3 .Qa6+ Kb8 3 4.Qa7+ Kc8 3 S .b7+ Qxb7 3 6. Qxb7+
Kxb7 3 7 .Bxa S , with a winning endgame .

3 3 .Qc4+ Kh7
Black is hoping to play the endgame that arises after 3 4.Bf8 Kxb6.

34.BxaS !
This unexpected piece sacrifice completely refutes Black's play.

34 ... QxaS 3 S .Qf7+ Ka6

King moves to a8, b8 or c8 all let White capture the Knight with check,
while 3 S . . . Kxb6 runs into 3 6 .Rb4+ .

36.Qa7+ KhS 3 7.Rh4+

Even stronger was 3 7 .a4+ ! .

37 ... Qxh4 38.axb4 and White, with his strong passed b-pawn and extra Queen,
easily won the game. 1 -0.

PAL BE NKO: MY L I F E . G AMES AND C O MPO S I T I O NS

226

Throughout my chess career, there have been many cases where I lost or drew a
game I should have won. On several of these occasions I have to admit that my
play was not up to the task. Mostly, though, the trouble was time pressure. It has
been said of me by no less a personage than Tigran Petrosian that I play like a
genius in time trouble, seeing quite deeply into the position. Be that as it may, I
do not subscribe to the belief that a player can play better in time pressure than
he can when he has plenty of time - it is simply against all logic. I will agree
that, sometimes, it might seem that someone always finds the best moves when
he's forced to play quickly, but this is just an illusion. Every player wants to con
ceal his true intentions from his opponent, and for this purpose he contrives
devious and complicated plans requiring a lot of time to analyze. In time pres
sure, however, he must make more or less direct moves, and this creates an erro
neous impression that his play is sharper or more incisiv e- but this is not to say
that it is better.
Though tossing away an easy win or simple draw is always traumatic, blaming
the ravages of time pressure is not a valid excuse - there is no one to blame but
myself. However, there are times when I can't find the justice in a system which
allows a player to lose when the board shows a clear win and when his opponent,
who really should have resigned long ago, similarly has no time on his own clock.
The following game is a perfect example. Having completely outplayed Larsen,
and with a dead won position, I was given a loss instead of the richly deserved
win. The circumstances are described at the appropriate point in the game.

(70) Benko - Larsen (Winnipeg, 1967)

l .c4 e5 2 .Nc3 d6 3 .g3 f5

Larsen is setting up a kind of Dutch Defense, a favorite of his (even when
playing \\!hite, he often plays Bird's Opening, a Dutch with colors reversed).
In this system of defense, Black generally is able to advance the e-pawn
only one square at first, and later tries to enforce ... e5 after strenuous prepa
rations. But in this game, Black has already achieved this freeing move.
Does this mean that Black has solved his normal opening problems? No,
there's an important difference between a real Dutch (which arises after
l .d4) and this position from the English Opening: \\!hite hasn't pushed his
d-pawn yet. This fact allows him to play d2-d3 , gaining control of the e4square (a square that often falls into Black's hands in the Dutch).

4.Bg2 Nf6 5.d3

In Benko-Bisguier (game 5 1 ), I tried the sharper 5.d4.

5 ... Be7 6.NfJ

\\!hite's main alternative is 6.4 exf4 7.gxf4.

6 ... 0-0 7.0-0 Kh8

More recently, moves like 7 ...c6, 7 ...Qe8, and even 7 ...Na6 have come into
vogue.

conti nue my end less

travels. Here I am in
Monte Carlo ( 1 965)
i n a life-and-death
battle against Larsen.
(Photo courtesy USCF.)

8.b4
White immediately starts queenside action. Black usually counters with
play on the kingside, but Larsen fails to achieve this in the present game.

8 ... aS 9.bS Nbd7 1 O.Ba3 Qe8 I 1 .Rc1 Rb8 1 2 .e3

This move has a double purpose: it takes the c S -square away from Black's
Knight ( 1 2 . . . NcS 1 3 .d4) and it takes the sting out of Black's . . . QhS , which
can be answered by moving the f3 -Knight, offering to exchange Queens.
(A Queen exchange rarely slows down a queenside attack, but it puts a
real dent in a player's kingside aspirations by getting rid of his potential
knockout punch.)

12 ...Nb6 l 3 .eS!
This move obtains a majority of central pawns.

l3 ... dxeS 14.NxeS Be6 I S.Qe2 Bd6 16.f4 Nbd7 1 7.NB

According to classical principles, most of which are still useful, one should
avoid exchanging pieces when the opponent has a cramped position.

1 7 ...NdS 1 8.e4
Kicking my central majority into action.

18 ... Nb4 1 9.Bb2 e4? !

Obviously 1 9 . . . Nxa2 2 0.Nxa2 Bxa2 2 1 . Bxg7 + is bad for Black, as his kingside
is then seriously weakened. With 1 9 . . . c4, Black is seeking good squares
for his pieces and isn't afraid to sacrifice a pawn to achieve this. However,
a better way of gaining access to the squares on cS and d3 was 1 9 . . . fxe4
20.dxe4 c4! .

20.dxe4 Nb 6 2 1 .Nd2 Rd8 22.Khl

The reply to 2 2 .a3 is 2 2 . . . Nd 3 and if 2 3 .Qxd3 ? ? BcS + .

PAL BE NKO: MY LIFE, G AMES AND C O MPOSITI O NS

22...Bc5 2 3 .Nd5!

228

This strong move was planned when I accepted the pawn sacrifice and
entered the complications that, obviously, Larsen felt would turn out well
for him.

2 3 ... Bxd5
The tactical justification of 2 3 .Nd5 can be seen after 2 3 . . .Nxa2 2 4.Nxc7
Nxc 1 2 5 .Bxg7 + ! (An important Zwischenzug
Black's King can't take the
Bishop as his Queen is then captured with check.) 2 5 . . . Kg8 2 6 . Rxc 1 Qd7
27 .Nxe6 Qxe6 2 8 . Bxf8, and White comes out with two extra pawns while
Black's King is in an unsafe position.
-

24.cxd5 Na4
Flashy but inadequate is 24 . . . Rxd5 2 5 .Nb3 Bd6 2 6.Bxg7+ Kxg7 2 7 .Qb2 +
Kg8 2 8 .exd 5 .

2 5 .Nb3 Bb6 26.Be5 fxe4 27.Bxe4 Nxd5

Black has recovered his pawn at the cost of extreme positional inferiority.
His Knight on a4 is badly misplaced and White's two centralized Bishops
exert tremendous pressure on both sides of the board.

28.Qc4
As pointed out in the note to my 2 9th move, 2 8 .Rfe l ! was stronger.

28 ... Nb2 !
A nice move that saves the Knight since 2 9 . Bxb2 Ne3 3 0 . Qa4 Rd4!
actually wins for Black!

29.Qc2
Both players were already in time pressure at this point. Since I was not
keeping score here, I cannot be sure if the moves were repeated once or
twice (2 9.Qe2 Na4 3 0.Qc4 Nb2), but this position was reached. Actually,
29.Qe2 was the better move, since if 2 9 . . . Na4 3 0.Rfe l threatens 3 1 .Bxd5
followed by 3 2 .Bxg7 + as well as 3 1 .Qc4.

29 ... Na4 30.Rfe l Ne3

Black might have offered more resistance with 30 . . . Qxb 5 , though 3 1 .Qg2 !
Nf6 (3 1 . . .h6 3 2 .Qh3) 3 2 .Bxb7 still gives White a clear plus, but his time
pressure was even worse than mine !

3 1 .Qe2 Qf7
Black was lost no matter what he played. For example: 3 1 . . . Rb8 3 2 .Rc4
Qxb5 (3 2 . . . Nxc4 3 3 . Qxc4 also wins the Knight on a4.) 3 3 .Qh 5 .
229

32.Bxb7 Nd5 H .Qc4

Winning, but even stronger was B .Rc4 Nb4 3 4.Be4, when there's no de
fense to the simple threat of a2-a3 .

3 3 ... Nb4 34.a3 Qxc4 3 5.Rxc4 Nd3 3 6.Re2 Nf2 +

White wins easily after 3 6 . . . Nxe5 3 7 .fxe5 Rd3 3 8 .e6 Rxb3 3 9.e7 Re8 4O.Bc6.

3 7.Kg2
Of course, I could have won simply with 3 7 . Rxf2 Bxf2 3 8 .Rxa4, but there
were only seconds remaining!

3 7 ... Rdl 3 S.Rxf2 Bxf2 3 9.Kxf2 Nb6 40.Rxc7, 0- 1 .

As I made my last move, Larsen said, "Your flag is down." Out of sheer mo

mentum, however, I completed the move and punched my clock, at which

point Larsen's flag fell. Neither of us had been keeping score for the last ten
or twelve moves, so we didn't know how many moves had been made. The
tournament director ruled that Larsen was the winner. When I protested,
he set up a committee (it included both Szabo and Keres, so I couldn't ex
pect a favorable decision) to decide the question and they later upheld his
original ruling. Since my final score was only one point below Larsen's who
tied for first place, the loss of this game meant the loss of a first-place tie . As
for the game itself, Larsen should have resigned some time ago.

(7 1) Benko - Matanovic (Winnipeg, 1967)

l .d4 Nf6 2 .c4 e6 3 .NB b6 4.g3 Bb7 5 .Bg2 Be7 6.Nc3 Ne4 7.Bd2 f5
S.d5
One of the ideas of the 7 .Bd2 system is to lock Black's light-squared Bishop
out of play for as long as possible. Because of this possibility for White, it
might be better for Black to play 7 . . . d5, as Parma did against me in Belgrade
1 964 (see game 5 7) .

S ... Bf6 9.Qc2 Na6

An attempt to improve the line for Black. Inferior is 9 . . . 0-0 1 O.Nxe4 fxe4
I l . Qxe4 exd 5 1 2 . Qc2 Nc6 (better is 1 2 . . . Re 8 , but
Matanovic, Yugoslavia 1 967 showed that Black is still
Qe7 1 4.cxd5 Qc5 1 5 .Qxc5 bxc5 1 6.Ng5 ! , when White
tage) 1 3 .cxd5 Nd4 1 4.Nxd4 Bxd4 1 5 .e3 Qf6 1 6.Bb4
1 8 .Bxf8 , and White won the Exchange and the game.

the game Bukic

in trouble: 1 3 .0-0
had a clear advan
Bxb2 1 7 .Rb l Be5

1 0.0-0
Taking on e4 gives Black good play: 1 O.Nxe4 fxe4 I 1 .Qxe4 Nc5 1 2 .Qc2
exd5 1 3 .0-0 dxc4 1 4. Qxc4 1 5 . Qc2 0-0, Bukic-Sibarevic, Yugoslavia 1 96 8 .

1 0 ...0-0 I l .Nxe4 fxe4 12 .Qxe4 ReS

P A L BE NKO: MY L I F E . G AMES AND C O MPOS ITI O NS

White also has an obvious advantage after 1 2 . . . Bxb2 B .NgS g6 1 4. Qh4

hS I S . Rab l .

1 3 .Qc2 exd5 14.cxd5 Nc5

230

Black can't take either of the offered pawns: 1 4 . . . Bxd S I S .NgS Bxg2
1 6.Qxh7 + Kf8 1 7 .QhS BdS 1 8 .Nh7 + Kg8 1 9.Nxf6+ and 2 0 . QxdS + wins
for White, while 14 . . . Rxe2 I S .d6 threatens both 1 6.NgS and 1 6.Qc4+ (i.e.,
IS ... NcS 1 6 .NgS ! BxgS 1 7 .Qc4+ ! Re6 1 8 .Bxb7 Nxb7 1 9 .dxc7).

1 5 .Rad l Qe7
Both pawns are still poisoned: I S . . . Bxd S ? fails to 1 6.NgS , while I S . . . Rxe2
1 6.b4 Na6 ( 1 6 . . . Ne4 1 7 .Qc4 wins material) 1 7 .Qc4 Re8 1 8.d6+ Re6 1 9.Rfe l
is also awful for Black.

1 6.Rfe l
Simpler was 1 6.Be3 .

16 ... Qe4
White is a clean pawn up after 16 . . . BxdS 1 7 .NgS Ne4 1 8.Nxe4 Bxe4 1 9.Bxe4
Qxe4 20.Qxc7 Bxb2 2 1 .Qxd7.

1 7.Qxe4 Rxe4 1 8.b3 ?

White returns the well-earned pawn too soon. Better was 1 8 .d6 or 1 8 .b4,
with a huge advantage in both cases.

18 ... Bxd5 1 9.Bg5 Bb7 20.Bxf6 gxf6 2 1 .Nd4 Re7 22.Nf5 Re5 2 3 .Bxb7
Nxb 7 2 4 . e4 d6 H . f3 a s 2 6 .Ne 3 Rf8 2 7 . Rc 1 NcS 2 8 .Red l Rf7
29.Rc2 Kf8

TRAVEL. TRAVEL, AND MORE TRAVE L

White's superior pawn structure gives him the better position. The proper
plan now is to prepare queenside action with Rc3 , a3 and b4. This would
force the strong black Knight off its perch and make the c7 -pawn vulnerable to attack.

23 1

30.Kf2
As my previous note pointed out, the
strongest move was 3 0. Rc3 followed by
a2-a3 and b3 -b4.

30 ... KeS 3 l .RdS

Now 3 1 . Rc 3 IS m e t by 3 1 . . . Rx e 4 !
3 2 . Rxc 5 Rxe 3 B . Rxc7 Rxc7 3 4.Kxe3
Kd7 .

xander Matanovic: Matanovic

was a very serious guy, and seldom

smiled. I beat him many times. For some
reason, I always had his number.

3 1 ...Kd7 32 .Red2 Rfe7? 3 3 .Ng4

Black probably overlooked this. Now 3 3 . . . Rxd5 10ses to 3 4.Nxf6+ followed
by 3 5 .Nxd 5 .

3 3 ... RSe6 34.RfS Rfi 3 5 .eS

Suddenly Black's game is terminal !

3 5 ... KeS 36.Rxf6 Rfxf6 3 7.Nxf6+ Kfi 3S.NdS dxeS 3 9.Nxe7 Re7 40.NbS
e4 41 .Nd6+ Kf8 42 .Nxe4 Nxe4+ 43 .fxe4 Rxe4 44.RdS ! a4?
Black's game was lost anyway, but this makes it easy. After 44 . . . Kg7 45 .Rb 5
Re6 46.a4 White, with a n extra pawn and the more active Rook, should
encounter no real problems. Eventually the white King walks over to the
queenside.

4S.RfS+ Kg7 46.Rf4 ReS

Black has an unpleasant choice: trading Rooks and going into a lost King
and pawn endgame, or giving up a second pawn and going into a lost Rook
and pawn endgame. Of course, there is a third choice: resigning.

47.Rxa4 bS 4S.RaS ReS 49.Ke3 Re2 SO.h4 b4 S 1 .Kf4 Kg6 S2 .Ra6+ Kg7
Avoiding 52 . . . Kh5 5 3 .g4+ Kxh4 54.Rh6 mate.

S 3 .Kg4 Rd2 S4.Ra4 Kg6 S S .Rxb4 Rxa2 S6.Rb6+ Kg7 S7.b4, 1 -0.

When asked why I knowi ngly went into my opponent's preparation: " It would have been
coward ly to do anything else."

book on the B'nko Gambit, I mentioned that lin" 'imibr to mod

ern versions of this system appeared as early as the 1 92 0s, and were used after
that by Sweden's Stahlberg, Stoltz, and Lundin. In these early games (Bronstein
Lundin and Szabo-Lundin, both played in Saltsjobaden 1 948), Black would only
sacrifice the pawn after White played e2 -e4 (i .e., 1 .d4 Nf6 2 . c4 c5 3 .d5 d6 4.Nc3
g6 5 .e4 b5), apparently afraid of setups where White leaves his e-pawn on e2 . I
was aware of these games, and began to wonder why Black didn't sacrifice the
pawn earlier, when a quick . . . Bxa6 would make White think twice about e2 -e4
since the first player would then lose the right to castle (after . . . Bxfl ). Of course,
I had no fear of lines where White kept his pawn on e2 since his most threaten
ing plan is based on an e2 -e4-e5 counterstrike in the center.
I admit that I was never able to study openings - it really bored me. I remem
ber being with a girlfriend who asked me to look over some chess openings with
her. I instantly fell asleep. I found that I could always take a nap in any situation
by just looking at some opening variation- my eyes would shut right away.
Endgames, though, fascinated me. I wanted to dig deep for the ultimate truth,
but truth doesn't exist in the beginning phase of a game.
I desperately wanted a new opening that would allow me to avoid mainstream
theory, and would force my opponents to think for themselves. Looking at the
. . . b7-b5 gambit idea in more detail, I soon realized that I was onto something!
Here was a complete system (against l .d4) that had no body of theory (thus no
memorization !), that was largely unknown (my opponents would not know what
was going on), that lent itself well to general ideas and strategic motifs, and that
gave Black excellent chances in most endgames !
I began testing it in American tournaments with incredible success. In 1 967
(against Vukic), I introduced the gambit into international competition. Within
23 3

PAL BE NKO: MY LI FE. G AMES AND C O MPOSITI O NS

a few years, many players added it to their repertoire and Benko Gambit theory

234

developed by leaps and bounds. Eventually I realized that I had created a

monster-it was born to help me avoid opening preparation, but now all my
opponents entered each game booked to the hilt! As a result, I found it neces
sary to turn to other openings in the mid-seventies. Nevertheless, after more
than fifty games with it, I only lost three (and those defeats had nothing to do
with the opening). Today, the Benko Gambit seems more popular than ever, with
players of all strengths making use of it at every opportunity.
The following game signifies the birth of the Benko Gambit in international
tournaments. There were many people who regarded my choice as an extrava
gance, thinking that I had underestimated my opponent by sacrificing (as Black)
a pawn in the opening for the sake of surprise. However, they didn't know that I
had put the gambit into practice in the U.S. several times and, as a result, was
well acquainted with its mysteries.

(72) Vukic - Benko (Sarajevo, 1967)

l .d4 Nf6 2.c4 c5 3 .d5 b5 4.cxb5 a6 5 .bxa6 Bxa6 6.Nc3 d6 7.NO g6 8.g3
Bg7 9.Bg2 0-0
Many years later, GM Lev Alburt introduced 9 . . . Nbd7 1 0.0-0 Nb6, put
ting pressure on the d5 -pawn.

1 0.0-0 Nbd7 I l .Qc2 Qb6

After years of practice, it is still difficult to determine the best move. Black
can consider 1 1 . . .Qa5 (which I've tried), 1 1 . . .Qc7, 1 1 . . .Nb6, and even
1 1 . . . Ra7, intending to place the Queen on a8.

1 2 .Rd l Rfb8 1 3 .Rb l Ne8 14.Bg5 Qd8

Having made room for the KR to swing over to b8, the Queen returns to
its original position in order to give the e7-pawn some support.

1 5 .Bfl h6 1 6.Bd2 Nc7 1 7.b3 Nb6 1 8.e4 Bxfl

White has succeeded in getting rid of the strong a6-Bishop, but this came
with a price: the light squares on the kingside have been weakened.

1 9.Rxfl Qd7 20.Rfe l Klt7 2 1 .Kg2 e6!

CRE AT I NG T H E BE NKO G AMBIT

White has managed to strengthen his queenside. Therefore, Black endeavors

to create new problems for his opponent in the center. This demonstrates
one of the nice qualities of the Benko Gambit: Aside from straightforward
queenside play (where ...c5 -c4 often proves useful), Black can strike in the
center with ...e7-e6 or ...f7 -f5 .

22.dxe6 Nxe6 23 .Ne2 dS 24.Nf4 dxe4 2S.Rxe4 Nd4 26.Nxd4 cxd4 27.a4??
A blunder that lets Black regain the gambit pawn with an overwhelming
position. But White was in difficulties in any case, the threat being 27 ...Rc8
2 8.Qb2 d 3 . If White tries to blockade the pawn with 2 7 .Nd3 , then 2 7 ...Rc8
followed by 28 ...Qb5 or 28 ...Qc6 2 9.f3 Qc2 follows. Perhaps White's most
natural move is 2 7.Rbe 1 . In that case, 2 7 ...Nd5 2 8.Qb l Ra3 gives Black
more than enough compensation for the pawn.

27 ... Qb7 !
Creating a double threat against the Rook ( ...f7 -f5 is hanging over White's
head) and the a4-pawn ( ...Nxa4 utilizes the pin along the b-file).

28.0 Nxa4 29.Rbe l ?

Desperation, but his expected counterplay fails to appear.

29 ... Qxb3 30.Qxb3 Rxb3 3 1 .Re7 Rb2 3 2 .Rle2 Nc3

White is hopelessly lost.

3 3 .Bxc3 dxc3 34.Rxf7 Raa2 3 S .Kfl gS

An old-fashioned greedy move that forces White to shed lots of wood.

36.Rxg7+
This doesn't help, but since 3 6.R2 e7 Rb 1 + also led to a quick defeat, it
didn't really matter what White played.

36 ... Kxg7 3 7.Ne6+ Kf6 3 8.Nd4 Rxe2 3 9.Nxe2 Ra1 +, 0- 1 .

When RHM Chess Publishing proposed that I write a book on this opening, I
asked what I should name it, because at the beginning I called this system the
Benoni Counter Defense. The publisher's answer was to the point: "Name your
opening the Benko Gambit! " To popularize my upcoming book, I gave a tour of
simultaneous exhibitions where I would take the Black side of the gambit if any
one wanted me to. I was undefeated with the Black pieces, but I lost two games
with White !
Walter Browne had been watching me play the Benko Gambit for a while, but
he never showed any real interest in my system. This attitude changed when we
were both playing in a tournament in Spain. He approached me and asked if I
would teach him how to play this gambit. I accepted and spent time showing
him the ideas and variations. He successfully played it for a couple of years, and
then he published his games under the name, the Browne-Benko Gambit. I
couldn't believe my eyes !

235

P A L BE NKO: MY L I F E , G AMES AND C O MPOSITI O NS

(73) Gross

Benko (U.S. Open, 1968)

l .d4 Nf6 2 .c4 c5 3 .d5 b5 4.cxb5

236

My opponent thoroughly prepared for this game, even though some of my

rivals warned him not to allow the Benko Gambit. To his credit, Mr. Gross
answered, "I'm no chicken ! " and rushed right into my favorite opening.

4 ... a6 5.bxa6 g6 6.Nc3 Bxa6 7.e4 Bxfl 8.Kxfl d6 9.Nge2

Defending the c3 -Knight and achieving a solid position. Unfortunately, it
also deprives White of his only realistic attempt for activity: playing for
an eventual e4-e5 advance. Naturally, placing the Knight on f3 would be
more in accordance with that idea.
The game Visier-Benko, Malaga 1 969 saw the highly ambitious 9.g4
Bg7 1 0.f3 (In Avram-Benko, U. S . Open 1 968, White played the illogical
1 O.h3 ? ! , when Black had a pretty easy time of it: 1 0 . . . 0-0 I 1 .Kg2 e6! [Fol
lowing the old rule: The best reaction to an attack on the wing is a coun
terattack in the center.] 1 2 .dxe6 fxe6 1 3 .e5 Ne8 1 4.exd6 Nxd6 1 5 .Qe2 Qd7
1 6.f4 Nc6 1 7 .Nf3 Nd4, with the better game for Black. I went on to win
on the 3 1 st move.) 1 0 . . . 0-0 I 1 .Nge2 e6!

(Ending White's hopes of a kingside attack by creating play in the center.)

1 2 .Kg2 exd5 1 3 .Nxd5 Nc6 (Black has an excellent position.) 1 4.Nec3 (The
pin 1 4.Bg5 isn't dangerous due to 14 . . . Nxd5 .) 1 4 ... Nxd5 1 5 .exd5 Nb4! (This
is much stronger than 1 5 . . . Nd4, because it builds a strong post on d3 by
. . . c5 -c4.) 1 6 .h4 f5 1 7 .g5 ? c4 1 8 .a3 Nd3 1 9.Rb l Qb6 2 0.Qc2 Rfe8, 0- 1 . A
bit early, but White's game is very bad. For example: 2 1 .Kg3 (to avoid the
threatened . . . Ne 1 +) 2 1 . . . Be5+ 2 2 .Kg2 Bxc3 2 3 .bxc3 Qc5 .

9 ... Bg7 1 0.g3 0-0 1 1 .Kg2 Qb6

Black has many good ideas here. In Goldin-Bareev, USSR 1 986, Black
got great play (and a quick victory) with 1 1 . . .Na6 1 2 .Rb l Qd7 1 3 . b3 Nc7
1 4. f3 Rfb 8 1 5 . Qc2 Nb S , while another tempting plan is 1 1 . . .Nbd 7
1 2 .Qc2 Ra6 1 3 .Rd l Q a 8 1 4.b3 e6 ! , as seen i n W. Schmidt-Conquest,
Poznan 1 986.

1 2 .Rb l Na6

CRE AT I NG THE BE NKO G AM BIT

Black forsakes the usual idea of placing pressure on the queenside and in
stead prepares to attack the center with a timely ... e7-e6.

1 3 .b3 Nc7 14.0 e6 1 S.dxe6 fxe6 1 6.Be3 Qc6 1 7 .Qd2 ?

White underestimates the danger. He had to try and neutralize the posi
tion with 1 7 .b4 cxb4 1 8 .Rxb4, though 1 8 . . . dS 1 9.Rb6 Qc4 20.Qb3 Qd3 is
good for Black.

1 7 ... dS 1 8.exdS exdS 1 9.Bf4 d4

Black's game almost plays itself at this point!

20.Na4 NfdS 2 1 .Rhfl Ne6 22.Rf2 d3 !

The position is hopeless for White despite his extra pawn - Black's active
pieces and strong center are more than White can handle. After 2 2 .Rf2 ,
Black could have won in several ways, for example: 2 2 . . . c4 is very strong. I
even toyed with 2 2 ... Rxa4 2 3 .bxa4 c4, when the two connected passed pawns
would be irresistible. Finally I decided in favor of the advance of the d
pawn, which I felt was the shortest and simplest solution.

2 3 .Ng1
2 3 .Qxd3 isn't possible due to 23 ... Rxf4, when Black wins a piece.

23 ... Nexf4+ 24.gxf4 Nxf4+ 2 S .Kh1 Bd4 26.Rff1 Ne2 ! , 0- 1 .

After 2 7 . Qxd3 (2 7 .Nxe2 RxB i s even worse) 2 7 . . .Nxg l 2 8 . Rxg l RxB
2 9 .Qc4+ Kh8 3 0 . Rg2 Rf2 3 1 . Rg l Rxa2 , there isn't any defense against
Black's mating attack.
During this event, I overheard the following conversation in the washroom:
"Who are you playing today? " "Benko." "What opening will you play against him?"
"The Benko Gambit, of course. He won't dare play the best line against it because
then he won't be able to play the gambit anymore ! " After hearing this, I felt it
wouldn't be sporting of me to decline the challenge, but I decided to try a system
shown to me a few days earlier by my friend, Dr. Elod Macskasy, who told me he
had tried it with success after initial bad experiences against the Benko Gambit.

237

P A L BE N KO : MY L I FE. G AMES AN D C O MPO S I T I O NS

Two young American grandmasters

in 1 968: James Taan and Walter
Browne. It was at this time that
Browne "stole" my opening by
naming it the "Browne-Benko
Gambit."

238

(Photo courtesy USeF)

(74) Benko - Berry (Canada, 197 1)

1 .d4 Nf6 2 .c4 c5 3 .d5 b5 4.cxb5 a6 5.bxa6 Bxa6 6.g3 d6 7.Bg2 g6 8.Bd2
Bg7 9.Bc3 ! ?

This i s Macskasy's system. The plan i s to neutralize Black's strong dark

squared Bishop. This idea is reasonable and well motivated, but it has a
serious drawback: White's b l -Knight is deprived of its natural developing
square on c 3 , and the gl -Knight must be developed on h3 to avoid the
loss of the d5-pawn. Anyway, I thought it was worth trying.

9 ... 0-0
After 9 . . . Nbd7 1 0 .Nh3 Nb6 I l .Nf4 Ra7 , White decided to push his h
pawn, which ends eventual . . . g6-g5 ideas by Black. 1 2 .h4 h5 1 3 .Nd2 Qa8
1 4.Ne4, and White was better in Vokac-Ruzicka, Prague 1 992 .

1 0.Nh3 Nbd7 1 1 .0-0 Nb6 1 2 .Nf4 Qd7

The game Stohl-Nunn, Pardubice 1 992 showed an important plan for Black:
1 2 . . . Ra7 1 3 .b 3 Qa8 1 4.a4 Rb8 1 5 .Re l c4 1 6 .a5 Nbd7 1 7 .bxc4 ( 1 7 .b4! ? was
recommended by Karpov, but if Black moves his Knight to e8 and ex
changes dark-squared Bishops, the pawn on b4 looks to be very weak.)
1 7 . . . Bxc4 1 8 .Nd2 Rc8 1 9.Rc 1 Rac7 2 0 . Bxf6 Bxf6 2 1 .Ne6 Bxe2 2 2 .Nxc7,
Ih _ lh , It seems to me that there is still a lot to be discovered in Macskasy's
little-known system !

1 3 .b3
Necessary to prevent . . . Na4.

CRE ATI N G THE BE N KO G AM BIT

1 3 ... Ra7
So far, Black has built up his position in a logical manner, but the text is
probably too slow. Sharper seems 1 3 . . . gS ! (I recommended this right after
the game, but it wasn't tested until 1 99 5 ! ) since 1 4.Nd3 hangs d S . After
1 4.Nh3 h6, the Knight on h3 is stuck, and there isn't much White can do
about the threat of . . .NbxdS : I S .Qd2 ( I S . Re I ! , giving some support to e2 ,
is probably best) I S . . . NbxdS 1 6.BxdS NxdS 1 7 .Bxg7 Kxg7 1 8 .NxgS QfS
1 9.Nc3 Nxc3 2 0.Qxc3 + Kg8 2 1 .Nf3 Bxe2 2 2 .Nh4 Qd 3 , with advantage
to Black, Castro Moler-Magem Badals, Barbera del Valles 1 99 5 .

14.h4
With this move, White denies Black the previously mentioned opportu
nity, but even here Black can try 1 4 . . . Bh6, making the white Knight's fu
ture uncertain on f4.

14 ... Rb8 1 5 .Re l e4

This looks like the only way to open lines on the queenside, but White is
developed enough to meet it.

16.Bd4! Re7 1 7.Nc3 exb3 1 8.axb3 Be8 1 9.Qd3 Reb7 20.b4 Qe8 2 1 .b5!

This pawn, strong o r weak, will decide the issue. White always finds a way
to defend it, but Black should probably try 2 1 . . . Bd7 anyway.

2 1 ...Nbd7 22 .Reb l Ne5 23 .Qd l Bd7 24.b6 Ne4 2 5.Ra7 Qd8

Not 2 S . . . Nxb6? 26.Rxb7 Rxb7 2 7 .Rxb6

26.Rxb7 Rxb7 27.Rb4 Ne5

Admitting his failure to win the pawn. After 2 7 . . . Nxb6? , there follows
2 8 . Qb3 pinning and winning a piece.

28.Qb3 Qe8 29.Bxe5 dxe5 30.Nd3

After this exchange, Black's game collapses because of the threats of dS -d6
and Nxe S . His next move is an act of desperation.

30 ... e4 3 1 .Nxe4 Nxe4 32.Bxe4 Bf5 H .Bxf5 gxf5 34.Re4 Qd8 3 5 .Re6
Bd4 36.Qe4, 1 -0.
The passed b-pawn has been successfully nursed through to victory, but
even during the game I occasionally wondered, "Is it worth all this suffer
ing to keep the extra pawn? "

239

P AL B E N KO : M Y L I F E , G AM ES AN D C O M POSITI O N S

(75) Portisch - Benko (Palma de Mallorca, 197 1)

l .d4 Nf6 2.c4 c5 3 .d5 b5 4.cxb5 a6 5.e3

240

A cautious move that was suggested by Pachman (who gave it an exclama

tion mark), but it was originally intended for use against the system with
. . . e7 -e6 for Black (the Volga Gambit), as he notes in his theoretical works.
As a matter of fact, I had played S . . . e6 (a move that is now considered
dubious) with Black, in a game against R. Gross at the American Open as
early as 1 966. This time I deviated from that line, being curious to find
out what my opponent could offer against the kingside fianchetto, which
is a logical way to continue.
In the '80s, S .e3 became the "modern" way to deal with the Benko Gambit,
and the line was explored very deeply. Then, as Black's correct defensive
measures were discovered and new anti-Benko Gambit ideas lit up the
theoretical landscape, this "modern" system vanished from top-flight events.
In the late '80s and '90s, the hot lines were S .3 (this leads to remark
ably sharp variations and demands a lot of homework from both players)
and S .b6, which tries to cut down on Black's counterplay by denying him
use of the a-file. In fact, both these moves retained their popularity into
the new millennium. Here's an example of my handling of the black side
of S .b6: S . . . d6 (Certainly not Black's only choice. I played S . . . Qxb6 [an
obvious move that is still considered the main line] against Mendoza in
Malaga 1 969, while S ... as [probably not very good] , and S ... e6 [one of Black's
most important ideas] are other possibilities.) 6.Nc3 Nbd7 7 .Nf3 g6 8.e4
Bg7 9.Be2 0-0 1 0.0-0 Nxb6 (Black finally decided to take the pawn. In
positions of this general type, \Vhite usually stands better because of Black's
cramped position. Here, however, Black's game is freer because of the open
b-file - and there was no sacrifice to get it!) 1 1 .Bf4 Bg4 1 2 .Nd2 (On 1 2 .h 3 ,
my plan was 1 2 . . . Bxf3 1 3 .Bxf3 Nfd7 1 4.Be2 c 4 followed by . . . NcS . I n such
positions, \Vhite's Knight on f3 is stronger than Black's light-squared Bishop
because, without this Knight, the e4-eS breakthrough is harder to achieve.)
1 2 . . . Bxe2 1 3 .Qxe2 Qd7 ! (Preparing to leap into action on the queenside
with . . .Na4. \Vhite appreciated this Knight maneuver and stopped it, but
in doing so he overlooked an even more important threat!) 1 4.a4? ! Qg4!

CRE ATI N G T H E BE N KO G AMBIT

(An unpleasant move that forces an exchange of Queens. The resulting

endgame will favor Black because of the weaknesses along the b-file.) 1 5 .Qe3
Nh5 1 6.Bh6 Bxh6 1 7 . Qxh6 Qf4 I S .Qxf4 Nxf4 1 9.b3 RabS 2 0.g3 Nd3 ,
with a clear plus for Black in Pfleger-Benko, Skopje 1 97 2 .

5 ... g6 6.Nc3 d 6 7.NB Bg7 8.bxa6

Eventually theory realized that S .a4 was White's best chance for an ad
vantage, though Black is fine here, too, if he knows what he's doing.

8 ... 0-0 9.e4

White loses a tempo with e3 -e4 in order to gain central space and free his
dark-squared Bishop. The alternative is 9.Be2 , when 9 . . . Bxa6, 9 . . . Nxa6,
and even 9 ... Qb6 have all proven adequate for Black.

9 ... Qa5 ! ?
An interesting adventure, but it's not certain that i t i s necessary.

1 O.Nd2
1 0.Bd2 isn't very good since it interferes with White's normal Nf3 -d2 -c4
maneuver. In the game Benko-de Fotis, U.S. Championship 1 97 1 (I hated
it when they used my own weapon against me !), Black got comfortable
play (which later led to an easy draw) after 10 . . . Bxa6 I I .Bxa6 Qxa6 1 2 .Qe2
Nfd7 1 3 .a4 Qxe2+ 1 4.Kxe2 Na6. Even worse is 1 0.Bd3 ? Nxd5 ! I l .exd5
Bxc 3 + 1 2 .Bd2 Bxd2 + 1 3 .Qxd2 Qxd2 +.

1 0 ... Bxa6 1 1. Bxa6 Qxa6 12.Qe2 Nfd7

This move has proved to be good, as demonstrated in my game against

Kaufman. But I have also played 1 2 . . . Nbd7 successfully against Farago at
Vranjacka Banja 1 97 1 .

1 3 .a4
In Kaufman-Benko, U.S. Open 1 969, White tried 1 3 .Nc4 when 1 3 .. .f5 led

to a sharp position where White immediately made a mistake: 1 4.f3 ? (He

had to play 1 4.exf5) 14 . . . fxe4 1 5 .fxe4 ( 1 5 .Nxe4 was better.) 1 5 ... Nb6 1 6.Nxb6
Qxb6 1 7 .Rfl Nd7 I S .RxfS+ RxfS I 9.Be3 RbS 20.Nd l Ne5 , and Black had
a clear advantage. The finish was: 2 1 .Kfl Qb4 2 2 .Bd2 RfS+ 2 3 .Kg l Qd4+
24.Be3 Qxe4 2 5 .Nc3 Qc4 2 6.Rfl RbS 2 7 .Be l Qxe2 2 S .Nxe2 Nd3 2 9.b3
Nb4 3 0.Bg5 Nxd5 3 1 .Rd l e6 3 2 .Nf4 Bd4+ B .Kh I Nc3 3 4.Rfl RfS 3 5 .Re l

24 1

PAL BE N KO : MY L I F E , G AMES AN D C O MPO SITI O N S

e5 3 6.Ne6 Rf2 3 7 .Bh4 Rxa2 3 8 .Rfl Ra8 3 9.Bd8 Bf2 40.g3 Ra2 41 .Kg2 Ne4
42 .KB d5 43 .Rd l Bd4 44.Nxd4 cxd4 45.Re l Rf2 + 46.Kg4 Rxh2 , 0- 1 .

1 3 ... Qxe2 +
242

This exchange is already forced as \Vhite was threatening 1 4.Nb 5 . How

ever, just because it's forced doesn't mean it's not good! The endgame is
comfortable for Black due to \Vhite's many weaknesses along the b-file.

14.Kxe2 Na6 1 5 .Rb l

Necessary. On the natural 1 5 .Nc4 Rfb8, in connection with . . . Rb4, would
prove strong.

1 5 ... f5 ! , 112_ 112 .

Here I offered my opponent a draw, which he gladly accepted. Many of
my fellow competitors, Larsen for instance, were of the opinion that Black
had more than enough compensation for the sacrificed pawn.

Before I played Vranesic, I was warned by several players that he was well-known
for his expertise on the white side of the Benko Gambit. He'd developed a sys
tem that he had used successfully on many occasions. This was, of course, news
to me, and though I had no idea what his system was, I used my gambit without
hesitation. It would have been cowardly to do anything else.

(76) Vranesic - Benko (Toronto, 197 1)

l .d4 Nf6 2 .c4 c5 3.d5 b5 4.cxb5 a6 5.bxa6 Bxa6 6.Nc3 d6 7.Nf3 g6 8.Nd2

I finally saw his "mysterious" system, but it was hardly new to me. \Vhite's
plan is to play e2 -e4 and then, after . . . Bxfl , recapture on fl with his Knight
(allowing \Vhite to retain his right to castle) which will then hop to e3
and c4.
8 ... Bg7

An interesting possibility is 8 . . . Qa5 ! ? , when 9.e4 Bxfl 1 O.Nxfl ? (he has to

play 1 0.Kxfl ) no longer works due to 1 O . . . Nxe4.

9.e4 0-0
Over the course of my career, I've lost very few games with the Benko
Gambit. One of my least pleasant memories is 9 . . . Bxfl 1 0.Nxfl 0-0 I I .Ne3
Na6 1 2 .0-0 Nd7 1 3 .Qe2 Qc7 1 4.Bd2 Qb7 1 5 .Rab l Nc7 1 6.b3 e6 1 7 .a4
Rfe8 1 8 .Nc4 Qa6 1 9.Qf3 , when \Vhite had a small plus which grew after
an unfortunate error on my part (Taimanov-Benko, Wijk aan Zee 1 970).
After this defeat I switched to 9 ... 0-0, but there's nothing wrong with
the capture on f1 . For example, 1 1 . . .Nbd7 (instead of 1 1 . . .Na6) has given
Black reasonable results, while 1 0 . . . Qa5 I 1 .Bd2 0-0 1 2 .Ne3 Nbd7 1 3 .0-0
Qa6 1 4.Qc2 Ne5 1 5 .b3 Nfd7 1 6.f4 Nd3 1 7 .a3 Bd4! was used successfully
by no less a player than Kasparov (versus Bareev at Linares 1 994).

CRE ATI N G T H E BE N KO G AM BIT

10.Bxa6 Nxa6 1 1 .0-0 Nd7 12 .Nc4

In the game Ghitescu-Benko, Siegen (01) 1 970, White tried 1 2 .Qe2 but
failed to get anything significant after 1 2 ... Nb6 1 3 .NB Nc7 1 4.Rd l Na4
1 5 .Nxa4 Rxa4 1 6.b3 Bxa l 1 7 .bxa4 Bg7 1 8.Qc2 Qd7 1 9.Bd2 Rb8, with equal
chances.

12 ... Nb6
Also possible is 1 2 . . . Nc7 ! ? 1 3 .Qe2 Qb8 1 4.Bd2 Nb5 1 5 .Rfc 1 Nd4 1 6.Qd3 ,
Doroskevic-Georgadze, USSR 1 97 2 , and now 1 6 ... Ne5 ! 1 7 .Nxe5 Bxe5 gives
approximately equal chances.

1 3 .Ne3
1 3 .Qe2 Nxc4 1 4.Qxc4 Qb6 1 5 .Na4 Qb4 1 6. Qc2 Nc7 1 7 .Nc3 Rfb8 1 8 .a3
Qc4 gave Black fine compensation for the gambit pawn in Ivkov-Browne,
RovinjiZagreb 1 970.

13 ... Qd7

Vranesic had already reached the position after 1 3 .Ne3 against Diez del
Corral, Siegen 1 970. In that game Black tried 1 3 . . . Qc8, but White got a
clear advantage after 1 4.a4 Nb4 1 5 .Ra3 Qa6 1 6.Nb 5 . The move I played
improves on that game (though I was completely unaware of it at the time),
but other moves are also possible: 1 3 . . . Qc7 (Walter Browne was success
ful with 1 3 ... c4! ? against Fitzgerald in 1 970) 1 4.Bd2 Rtb8 1 5 .Qe2 c4 1 6.Rfc 1
Nc5 , when both sides had chances in Soos-Jakobsen, Stockholm 1 9 7 1 172 .

14.a4
A committing move that obviously intends to secure the strong c4-square
for the white Knight on e3 by driving away Black's b6-Knight. On the
other hand, the move seriously weakens the squares on the b-file, and it
also makes the two queenside pawns more vulnerable.

14 ... Rfb8 1 5.Ra3

Disclosing the other side of White's plan: to protect the Rook and to make
possible a future Rh3 , which might help in a later kingside attack. Never
theless, I feel that 1 5 .a5 Nb4 1 6.Ra3 would have been more exact.

1 5 ... Bd4 16.Qe2

243

PAL BE N KO : MY L I F E , G AMES AN D C O MPOSITI O NS

In order to prevent the threatened I 6 . . . Bxe 3 I 7 .Bxe 3 Nc4.

16 ... Qb7!
244

Of course, I 6 . . . Bxc3 would win back the pawn, but after I 7 .bxc3 Nxa4
I 8 .Bd2 Nc7 , the position contains only meager winning chances for Black.
White could even try for an attack against the black King because of the
missing Bishop. Black is, naturally, willing to win back his gambit pawn,
but not if it means giving up his positional advantage. The spirit of this
gambit requires that Black retain winning chances after winning the pawn
back! With the text move, Black regroups his pieces in line with White's
following anticipated maneuver.

1 7.KhI Nc7 I S.a5 Nd7 1 9.Nc4 Qa6 20.f4?

White thinks that after his preparations he's ready for a kingside attack,
but it's still too early and he overlooks Black's strong reply. In any case,
the position is already difficult for White as his queenside weaknesses are
about to be exploited.

20 ... Rb4!
This strong move wins at least a pawn (without giving up Black's posi
tional advantage !). The c4-Knight can't be maintained, the Queens get
traded, and White finds himself in a very unpleasant endgame.

2 1 .Nb6 Qxe2 22 .Nxe2 Nxb6 2 3 .axb6 Rxa3 24.bxa3 Rxb6 2 5.Nxd4 cxd4

This is the endgame Black was willing to enter, since he has all the win
ning chances. Believe it or not, it is difficult, if not impossible, for White
to hold this position. If 2 6 . Rd I , Black plays 26 . . . Rb I , setting up a very
unpleasant pin. Also in Black's favor is 2 6.Bd2 Rb2 2 7 .Bb4 Na3 .

26.KgI f5 !
After this typical breakthrough in the center, White cannot avoid the loss
of a pawn.

27.exf5 gxf5 2 S.Kf2

Also hopeless is 2 8 .Rd l Rb I 2 9 .Kf2 NbS 3 0 .Re l Nc3 .

2 S ... Nxd5
Now Black will have two connected passed pawns in the center.

29.Rd l e5 3 0.Rd3 Kf7

CRE AT I N G THE BE N KO G AMBIT

Such a position is, of course, impossible for \Vhite to defend.

3 1 .h3 h5 32 .a4 Rb I 3 3 .Bd2 Ral

Neutralizing \Vhite's passed pawn.

34.a5 Nf6 3 5.Bb4 Ra4 36.Rb3

And not 36.Bxd6? Ne4+, picking up \Vhite's Bishop.

36 ... Ne4+ 3 7.Ke2 Ke6 38.fxe5 dxe5 3 9.Be l Ra2 + 40.KB h4

Threatening 4 1 . . .Ng5 mate. Of course, 4 1 .Bxh4 is answered by 4 1 . . .Nd2 +,
picking up \Vhite's Rook.

41 .Rb6+ Kd7 42 .Bxh4 Ra3 + 43 .Ke2 d3 + 44.Ke3 d2 +, 0- 1 .

\Vhite gave up, preferring not to experience the horror of 45 .Ke2 Ra 1 .

(77) Aspler - Benko (Vancouver, 197 1)

l .d4 Nf6 2 .c4 c5 3 .d5 b5

Walter Browne, who had switched from the Benko Gambit to the Mod
ern Benoni, was sitting next to me and, when he saw me play my gambit,
he pulled me aside and said, "Are you still fiddling around with that? It's
no good against weaker players because they like to hang on to the pawn."
But he added that it's playable against strong opponents because it's good
for a draw!

4.cxb5 a6 5.bxa6 Bxa6 6.Nc3 d6 7.Nf3 g6 8.g3 Bg7 9.Bg2

The fianchetto variation, a favorite line of many positional players, has its
disadvantages. \Vhite's light-squared Bishop is often not active on g2 be
cause his own center pawns limit its scope. Also, since \Vhite's Bishop is
no longer on the fl -a6 diagonal, the c4 and d3 -squares can fall into Black's
hands after \Vhite pushes his e-pawn.

9 ... 0-0 10.0-0 Nbd7 1 1 .Re l

The main line is considered to begin with 1 1 .Qc2 , when Black has a large
choice: 1 1 . . .Ra7 , 1 1 . . .Qc7, 1 1 . . .Qb6, and the aggressive 1 1 . . . Qa5 , which
happens to be Walter Browne's favorite. After 1 1 . . . Qa5 , Hort-Benko, U.S.A.
1 974 continued 1 2 .Rd l Rfb8 (A standard move, and in modern times the
most common reply. In my book, 1 2 . . . Nb6 is recommended instead. I did
not follow my own recommendation here because of psychological con
siderations, thinking that my opponent may have been prepared against
the "best" line. Anyway, 1 2 . . . Rfb8 is excellent and thematic - it is simply
not the sharpest. I still prefer 1 2 . . . Nb6 because \Vhite cannot easily pre
vent the black Knight from occupying c4.) 1 3 .Bd2 Ne8 1 4.Bh3 ! ? Bc8 1 5 .Be l
Nc7 , with plenty of compensation for the sacrificed pawn. Though the
subsequent play in this Hort game was nothing special, the whole thing
became extremely memorable when we arrived at the following situation:

245

PAL B E N KO : M Y L I F E . G A M E S AN D C O M P O S I T I O NS

246

This position favors Black. Correct is 4O ... Rxe2 (Also interesting is 40 ... Ra3 ! ?
4 1 .Rb2 Kg7 followed by 4 2 . . . Kf6, 4 3 . . KeS and 44 . . . Nf4) 4 1 .Bxe2 Rxb3
42 . Ra 1 ! Ke8 43 .aS Rb7 44.a6 Ra7 4S .Rb 1 Kd7 46.Rb8 NeS 47.BbS+ Kc7
48 .Re8 c4! 49.Kf2 (49.Rxe7+ Kb6 is good for Black) 49 . . . Kb6 SO.Rb8+
KcS S l .Rb 7 Ra8 , with good chances for Black. It's clear that Black is in
the driver's seat, but the move I played put an untimely end to the struggle:
40 ... Nf4?? 4 1 .Rxa2 . Hort said, "Sorry" as he gently removed my Rook
from the board. Naturally, I resigned. Sitting there, I didn't know whether
to laugh or cry. To cheer myself up, I recalled my favorite cartoon charac
ter and his famous remark: "Oh Magoo, you've done it again ! "

1 1 . . . Qb6
1 1 . . .Qc7 is now considered to be the most accurate reply to 1 1 .Re 1 .

1 2 .e4? !
Premature. White is intending a thematic breakthrough in the center with
e4-eS, but he should have prepared this plan with h2-h3 .

1 2 ... Ng4!
Not only preventing White's pawn advance, but also threatening to take
aim at d3 by jumping into e S .

1 3 .Qc2 Rfb S 14.h3

Right after this game was played (in 1 97 1), I wrote: "White is obviously
worried about his f2 square. After, for instance, 1 4.Rb 1 or 1 4.b3 , the move
1 4 . . . c4! threatens both 1 S . . . Bxc3 and 1 S . . . Qxf2 + . However, the text move
allows Black to gain another tempo. " Five years later, someone challenged
me by playing 1 4.Rb 1 . Did I follow through with 14 . . . c4 or did I try some
thing else? For the answer, go to the game Gordon-Benko, National Open
1 976.

14 ... Nge5 1 5.Nxe5 Nxe5 1 6.b3 ?

It was absolutely necessary to protect d3 with 1 6.Rd 1 (but not 1 6.Bfl ?
Nf3 +). After this error, White's whole position quickly collapses.

16 ... Nd3 1 7.Rdl c4!

CRE AT I NG THE BENKO G AMBIT

247

The threat is I S . . . cxb3 . If I S .bxc4 Qxf2 + I 9.Qxf2 Nxf2 2 0.Kxf2 Bxc3 traps
the a I -Rook.

1 8.Be3 Qb4
Avoiding I S . . . cxb3 ? , because I 9.axb3 Qxb3 2 0 . Qxb3 Rxb3 2 l .Rxd3 lets
White escape. The text move improves Black's position and cranks up the
pressure.

1 9.Bd2 Qe5 20.RfI

Losing the Exchange, but so does 2 0.Be3 cxb3 2 I .axb3 Qxc3 2 2 . Qxc3 Bxc3
2 3 .Rxa6 Rxa6 2 4.Rxd3 Rxb3 , with an easy win.

20 ... exb3 2 1 .axb3 Nb4 22.Qb2 Re8 !

There's no hurry to win material: 2 2 . . . Bxfl 2 3 .Bxfl Rxa i is an easy win
for Black, of course, but I thought that 2 2 . . . RcS was even stronger.

23 .Qa3
Naturally, B .Rfd Nd3 is crushing. White's best try was probably; 2 3 .Rad
Bxfl 24.Bxfl , when the incredibly tempting 24 . . . Na2 ? ! (Correct is 24 . . . Qd4
2 5 .Be3 Qf6, when White's cause is hopeless.) let's White continue the battle
with 2 5 . Qxa2 ! Bxc3 (2 5 . . . Rxa2 2 6.Nxa2) 2 6 . Qb l .

2 3 ... Bxc3
Of course, 2 3 . . . Bxfl 24.QxaS RxaS?? (Black still wins with 24 . . . Bxg2 2 5 .Kxg2
Bxc3) 2 5 .RxaS+ BfS 26.Bh6 would be too naIve, even for a long-time blun
derer like me !

24.Bxc3 Qxc3 25 .Rfc1 Qd4 26.Bfl Re2 , O- l .

Suffice it to say that my game was over long before Browne's, who was
still struggling to make something happen with his regular Benoni.

(78) M Camara - Benko (Sao Paulo, 1973)

l .d4 Nf6 2 .e4 e5 3 .d5 b5 4.cxb5 a6 5.bxa6 Bxa6 6.Nc3 d6 7.Nfl g6 8.e4
Bxfl 9.Kxfl Bg7 I O.h3
Most of the time this move is necessary to prevent Black's possible . . . Nf6g4-e5 maneuver. But here it also serves another purpose: White wants to
use h2 as a shelter for his King and allow his h I -Rook to get into play.

PAL BENKO: MY L I F E , G AMES AND COMPOSITI O NS

248

This maneuver, originated by Gligoric, is in my opinion not necessarily

better than the old fashioned g2 -g3 and Kg2 - the present system was
merely more fashionable at the time this game was played. In later years,
the g2 -g3 and Kg2 plan once again became "White's main option.

10 ... 0-0 I 1 .Kgl Na6

There is nothing wrong with 1 1 . . . Nbd7, as in the game Kane-Browne, U. S.
Championship 1 97 3 : 1 2 .Kh2 Qa5 (The Queen in more active here than
on b6. Incidentally, by leaving the b6-square open, Black's Knight can now
jump to b6 followed by ... Na4 or ... Nc4.) 1 3 . Re i Rfb8 1 4.Re2 Ra 7 ( 1 4 ... Rb4!
later became very popular) 1 5 .Rc2 Nb6 1 6.Qe2 Na4! 1 7 .Bg5 Rab7 1 8.Nxa4
Qxa4 1 9.Re 1 (Tukmakov later claimed that 1 9.Bxf6 Bxf6 2 0.b3 is a bit bet
ter for "White, but the reverse is true after 20 . . . Rxb3) 1 9 . . . Rxb2 ("White isn't
afraid of 1 9 . . . Qxa2 due to 2 0.b4) 2 0.Rxb2 Rxb2 2 1 .Qxb2 Ng4+ 2 2 .hxg4
Bxb2 and, after some technical difficulties, Black won.

1 2.Kh2 Qb6 n .Qe2

1 3 .Re 1 Nd7 1 4.Re2 was tested in several games, but Black always achieved
good play.

n ... rubS 1 4.Nd2 Nc7

Also possible is 1 4 . . . Nb4 1 5 .Nc4 Qa6, when a move like 1 6 .Bf4 can be
met by 1 6 . . . Nc2 1 7 .Qxc2 Qxc4, with adequate compensation.

I S .Nc4 Qa6 16.Bd2 Nd7 1 7.Rhe l NeS !

A well-timed Knight maneuver. Now 1 8 .b3 Nxc4 1 9. Qxc4 Qxc4 2 0.bxc4
Rb2 is too strong for Black.

I S.NxeS BxeS+ 1 9.4 Bd4

Black stands better: the pawns on a2 and b2 are targets, the d5 -pawn can
be undermined by an eventual . . . 7 -f5 advance, and "White's counterplay is
nowhere to be seen.

20.Bc1 Qxe2
This is the beauty of the Benko Gambit: Black's pressure carries on right
into the endgame.

2 1 .Rxe2 Kf8
By defending e7, I'm ready to play . . . 7-f5 .

CRE AT I NG T H E BENKO G AMBIT

22.Nd l f5 2 3 .exf5 gxf5

The dS-pawn is going to fall, leaving Black with a winning advantage based
on his enormous positional superiority.
249

24.Be3 Nxd5 2 5.Bxd4 cxd4 26.Rd2 e5

The situation grows worse for White with each passing move. The strength
of Black's central mass is obvious, while White's passed pawns on the
queenside are going nowhere.

27.fxe5 dxe5 2S.Nf2 Ke7

Bringing the King up to support its passed pawns. In a way, this gives Black
a material advantage since White's King isn't participating in the battle.

29.a3 Ne3 30.Re l Rb3 3 1 .Nd3 Rb5

Also possible was 3 1 . . .Nc4 3 2 .Rc2 Rxd 3 3 3 .Rxc4 Kd6, when Black's cen
tral mass should win without too much difficulty.

32 .Kgl Kd6 3 3 .Nb4 Nd5 34.Nd3

Of course, 3 4.Nxd S KxdS would allow the black King to dictate the battle
from its powerful central perch.

34 ... e4 3 5 .Nc1 d3 3 6.g4

The game is already decided, so White tries to break down Black's fS -e4d3 pawn chain, though this only accelerates his defeat.

36 ... Nf6 3 7.Rfl RgS 3S.Khl Ke6 39.Na2 fxg4 40.Rdf2 Rg6 4 1 .Nc3 , 0- 1 .
White would have no hope at all after 4 1 . . .RhS . A very thematic Benko
Gambit, well worth studying if you intend to take up this opening.

(79) Gligoric - Benko (Lone Pine, 1975)

l .d4 Nf6 2.c4 c5 3 .d5 b5 4.cxb5 a6 5.bxa6 Bxa6 6.Nc3 d6 7.NO g6

S.e4 Bxfl 9.Kxfl Bg7 1 0.g3
Gligoric pioneered 1 O.h3 , but he was just as likely to
employ the more common 1 0.g3 in any given game.

10 ... 0-0 1 1 .Kg2 Nbd7 1 2 .h3

A good idea, since a Knight leap to g4 will likely prove
bothersome to White : 1 2 .Qc2 Qb6 1 3 .Rb 1 RfcS ! ? (An
interesting new idea that avoids the almost automatic
1 3 . . . RfbS . Black intends to follow up with . . . cS -c4 and
. . . NcS .) 1 4.b3 Ng4! (Of course, 14 . . . c4 fails to either
l S .bxc4 or l S .b4.) l S .h3 NgeS 1 6.NxeS BxeS (Also
deserving of serious consideration is 1 6 ... NxeS. Though
the Knight can be repelled by 1 7 .f4, this would involve
a further loosening of White's position.) 1 7 .Bd2 c4! (M
ter this move, Black stands slightly better. l S .b4 is now
dubious because of l S ... Ra3 .) l S.Rhc 1 cxb3 1 9.axb3 Qb4

Gl igoric in 1 9 65.
(Photo courtesy
USCF.)

PAL B E N KO : MY L I F E , G AMES AN D C O MPO SITI O N S

2 0.Nd l Rxc2 , liz_liz, Donner-Benko, Palma d e Mallorca 1 97 1 . The final

position still favors Black: 2 1 .Bxb4 Rxc 1 22 .Rxc 1 Rb8 , and Black regains
the pawn with a somewhat superior game.
250

12 ... Nb6! ?

I have to admit that you won't find this in my oid book o n the Benko
Gambit, though now, in 2 00 1 , the move is quite common. The main line
was (and still is) 1 2 . . . Qb6, preparing a queenside attack by freeing the way
for the f8-Rook to get to the b-file, while 1 2 . . . QaS , and even 1 2 . . . Qb8
also have supporters. The text has a different strategic idea: to prepare a
central break by . . . e7 -e6, in which case the f8-Rook is better left on f8 . At
the same time, the move prevents White's thematic e4-eS advance because
of the pressure on his d-pawn. Of course, queenside play is not forgotten
since the Knight can come strongly into action by . . . Nc4 or . . . Na4. Clearly,
the theory for Black in the Benko Gambit is far from exhausted!
I do not agree with Bent Larsen, who wrote in his chapter for How to
Open a Chess Game that I use this opening to save thinking time. In fact, it
is my opponent who was known for his practice of playing very well ana
lyzed opening systems deep into the middlegame. I have always been quick
to condemn mechanical chess play, preferring to use my own head when
ever possible, even at the cost of more clock time in the opening. The text
move was dreamed up on the spur of the moment as a psychological ploy
against my opponent, who had used a whole sixty seconds for his first twelve
moves! After 1 2 . . . Nb6, he started to think, and to worry.

1 3 .Re l Qd7 14.Qc2

The game Taylor-Benko, World Open 1 974 went 1 4.eS ? (White wasn't
aware of my game with Gligoric, and was very surprised when I played
1 2 . . . Nb6.) 1 4 . . . dxeS I S .NxeS Qb7 1 6.Qb3 e6! (Also possible is 1 6 . . . NfxdS
1 7 .NxdS Qxd S + 1 8 .QxdS NxdS 1 9.Bd2 Nb4, with the better endgame for
Black: he will soon win both of the queenside pawns in exchange for his c
pawn. However, with four pawns against three on the same side of the
board, a win would be far off. The move I played is stronger.) 1 7 .Be3 Rfc8
1 8 .Nd3 NfxdS 1 9.NxdS c4! 20.Qxb6 Qxd S + 2 1 .Kg l cxd 3 , and Black had
an overwhelming position and eventually scored the full point.

C R E AT I NG T H E BE NKO G AMBIT

14 ... Qb7
In this line the black Queen is well placed on the same diagonal as White's
King. Now Black is ready to break with . . . e7-e6.

l S .BgS
Trying to prevent Black's plan since now 1 5 . . . e6 fails to 1 6.Qb3 ! , when
1 6 . . . exd5 1 7 .Bxf6 Bxf6 1 8 .Nxd5 or 1 6 . . . c4 1 7 . Qb4 are both bad for Black.

l S ... h6
Forcing the following trade, as the Bishop has no good retreat (i.e., 1 6.Be3
Nc4) .

16.Bxf6 Bxf6 17.Qb3

After the game, Gligoric thought this was a mistake. He tried other moves
in the postmortem such as 1 7 .Rab l , 1 7 .Rad l , and 1 7 .b3 , but Black gets
good play on the queenside anyway thanks to his strong Bishop and pres
sure down the half-open a- and b-files. We finally agreed that the text was
probably best after all. Some years later, 1 7 .Rad l was tested: 1 7 ... Nd7 1 8.Re2
Qb4 1 9.a3 Qc4 20.Rd3 Rfb8 2 1 .Nd i Qa6 2 2 .Rde3 Ra7 2 3 .Qd3 Qa4 =,
Naumkin-Salov, USSR 1 98 3 .

1 7 ... Rfb8 1 8.eS

Giving back the pawn and securing the draw. White didn't like the look of
1 8 .Rad 1 Qa6, when Black would have an excellent position.

18 ... dxeS 1 9.NxeS e6 20.Radl

Of course, 2 0 .Nc6 would be a mistake on account of 2 0 . . . Nxd 5 ! 2 1 .Qxb7
(Both 2 1 .Nxd5 Qxc6 and 2 1 .Nxb8 Ne3 +, with mate to follow, are unac
ceptable for White.) 2 1 . . .Rxb7, when White has a very bad endgame.

20 ... exdS, 112 _ 112 .

After 2 1 .Nxd5 Nxd5 2 2 .Qxd5 Qxd5+ 2 3 .Rxd5 Rxb2 24.Rxc5 Raxa2 2 5 .Nd3
or 2 5 .Rfl , all of the queenside pawns are gone and Black's slight theoreti
cal advantage is not enough to win.

(80) 1 Zaitsev - Benko (Szolnok, 1975)

l .d4 Nf6 2.c4 cS 3 .dS bS 4.cxbS a6 S.Nc3

At the time, this was a new line that had been invented by my opponent.
Though not so popular nowadays, it's still rather unexplored and leads to
wild, interesting positions.

S ... axbS 6.e4 b4 7.NbS d6

Zaitsev had published an article in Shakhmatny Bulletin discussing the po
sition after 8.Bf4. Since 8 . . . Nxe4 was thought to be too dangerous (though
that view is now in dispute), 8 . . . Na6 was played in Tukrnakov-Vaganian,
Vilnus 1 97 5 . There followed 9.NB (9.Bc4 ! ? also looks strong) 9 . . . g6 1 0.e5
Nh5 (according to Gufeld, 1O ... dxe5 I l .Bxe5 Bg7 1 2 .d6 0-0 is unclear.

25 1

PAL BENKO : MY L I FE, G AMES AND C O MPOSITI O NS

252

However, 1 3 .Bc4 100ks very nice for White.) I l .Bg5 ? ! f6 1 2 .exf6 exf6 1 3 .Be3
Bg7 1 4.Nd2 f5 1 5 .Nc4 0-0 1 6.Be2 f4 1 7 .Bd B ! 1 8 .gxB ( 1 8 .BxB Qe8+!)
18 ... Nf4, and Black won nicely. Unfortunately, White can improve with
I l .Qa4! (instead of I l .Bg5) 1 1 . . .Bd7 1 2 .e6! fxe6 1 3 . dxe6 Bc6 1 4.Nxd6+
Qxd6 1 5 .Bb5 ! Qxe6+ 1 6.Be5 Bxb5 1 7 .Qxb5+, with a win.
Tseshkovsky-Alburt, Vilnus 1 97 5 saw Black try something different:
8 ... Nbd7 9.NB Nb6, but this turned out poorly: l O.Rd ! (threatening Rxc5 !)
1 0 . . . Nxe4 1 2 .Bd3 Nf6 1 3 .Qe2 e6 14.dxe6 fxe6 1 5 .Ng5 Nbd5 1 6.Nxe6 Nxf4
1 7 .Nxf4+ Be7 1 8 .Bc4 ! , and Black can't castle.
Since these lines appeared to favor White, I did some homework and
was prepared to meet 8.Bf4 with 8 . . . g5 ! , and I mentioned this to Zaitsev
after our game.

He said that he would have played 8.Bf4 if he had known that I would
respond with such a move ! However, I succeeded in convincing him that
Black would have good chances. In a new article (written after our game),
he gave 8 . . . g5 an exclamation mark and wrote that this move is one of the
best weapons against White's system. By the way, it's interesting to note
that, on seeing me make my 7th move so quickly, Zaitsev realized that I
had something new lurking. Therefore, he decided that, instead of walk
ing into my preparation, he'd surprise me with something new on move
eight. Mter 8.Bf4 g5 ! 9.Bxg5 Nxe4 1 0.Bf4 (worse is 1 0.Bh4 Bg7 I 1 .Bd3
Nf6 1 2 .NB Nbd7 1 3 . 0-0 Bb7 1 4.Ng5 h6 1 5 .Ne6 fxe 6 1 6 . dxe6 0-0
1 7 .exd7 Qxd7, with a clear advantage for Black in Halldorsson-Benko,
World Open 1 97 8) 1 0 . . . Bg7 (also interesting is 1 0 . . . Qa5 ! ?) I 1 .Qe2 Nf6
1 2 .Nxd6+ Kf8 1 3 .Nxc8 Qxc8, Black would be the one with the initiative.

S.NO
This move aims to bypass the line with 8.Bf4, and was played here for the
first time. I selected a very cautious continuation because I feared some
new Soviet analysis and did not want to step on unknown paths in such a
sharp variation. The first move I considered was 8 . . . Nxe4, which Zaitsev
analyzed as follows in his second article: 9.Bc4 g6 1 0.Qe2 Nf6 I 1 .Bf4 Ra6
1 2 .Nxd6+ ! Rxd6 1 3 .Bb5+, and White gains the Exchange. However, in
stead of 1 0 . . . Nf6 ? , which seems weak to me, much better is 1 0 . . .f5 ! 1 1 .Ng5
Bg7 , returning the pawn but reaching an excellent position. Zaitsev also

CRE AT I NG T H E BE NKO G AMBIT

considered 8 . . . g6 9.Bc4 (9 .e5 dxe5 1 O.Nxe5 Bg7 I l .Bc4 0-0 1 2 .0-0 Ba6!
gave Black good chances in Gulko-Vasiukov, Yerevan 1 976) 9 ... Bg7 1 0.e5
dxe5 I l .Nxe 5 0-0 1 2 .0-0 Ne4 1 3 .Re l Nd6 1 4.Nxd6 exd6 I S .Nd3 and,
according to Zaitsev, the position is even. Though that would be a moral
victory for Black, I feel that this position actually favors the second player
since \Vhite's d-pawn and a-pawn are weak. Black can begin exploiting
this at once with . . . Nd7-b6.

8 ... Nbd7 ? !
As shown i n the last note, Black's safest course was 8 . . . g6, trying t o castle

as quickly as possible. The text is now known to give \Vhite a clear advan
tage but, at that time, this was all virgin territory.

9.Bf4 NhS
Probably best. Later, Black tried 9 . . . Nxe4, 9 . . . Nb6, and 9 . . . Ba6, with uni
formly horrible results.

l O.BgS Nhf6
Zilberman tried 1 O . . . Qb6 on a couple of occasions (in 1 9 7 5 and again in
1 98 1 ), losing miserably in both cases. After I l .Nd2 g6 1 2 .Nc4 Qb8 1 3 .a4
Bg7 1 4.Bd3 , \Vhite's advantage is obvious.

1 1 .Qe2 ?
\Vhite could have repeated moves with I I .Bf4, but I didn't expect Zaitsev
to do this because he had to win this game to fulfill the grandmaster norm.
Perhaps this explains why he was playing so sharply. Later he suggested
I I .e 5 ! as a better solution, giving the following possible continuation:
1 1 . ..dxe5 1 2 .Qe2 Ra5 1 3 .Nxe5 Nxe5 1 4.Qxe5 Qb6 1 5 .Nc7+ Kd8 1 6.Bf4
( 1 6.d6 Qxd6! 1 7 .Rd l Kxc7 is fine for Black) 1 6 . . . Ra7 1 7 .NbS Rd7 1 8 .Bc4,
when he claims a white advantage in this complicated position. In my opin
ion, Black can improve on the 1 6th move by 1 6 . . . Nd7 . For instance, 1 7 .QfS
(or 1 7 .Qg5) 1 7 . . . Qf6, and after the Queen exchange the endgame is about
even. If \Vhite tries 1 7 . Qe 3 , then 1 7 . . . gS 1 8 .Bg3 Bg7 1eads to an unclear
position. For example, if \Vhite lashes out with the seemingly strong 1 9.d6
exd6 2 0.Nd 5 , Black turns the tables with 20 ... Re8 2 1 .Qxe8+ Kxe8 2 2 .Nxb6
Nxb6 2 3 .Bxd6? Bxb2 .
\Vhite's I l .Qe2 strongly threatens e4-e5. How can Black prevent this?

1 1 ...RaS ! !

25 3

PAL BE NKO : MY LI FE,G AMES AND C O MPOSITI O NS

254

Undoubtedly, \Vhite was not expecting this, although he thought about

1 1 . Qe2 for over half an hour. It often happens that an unnatural move like
Qe2 can be correctly met by a similarly unnatural reply. Probably this origi
nal and fine defense was a major contributing factor to this game being
awarded the brilliancy prize. I also think that a key to finding this move
was a concept expressed in the first part of my book on this gambit: " . . . a
hidden advantage is the fact that Black's Queen Rook stands on a useful
open file without having had to move. "

1 2 .e5
Now or never!

12 ... Ba6!
Mistaken is 12 . . . dxe S 1 3 .Nxe S Ba6 ? ? (Of course, 1 3 . . .Nxe S is better.)
1 4.Nc4, when the threat of either Knight to d6 leads to Armageddon.

1 3 .exf6
Instead, 1 3 .Nxd6+ leads to Black's advantage : 1 3 . . . exd6 1 4.exf6+ Bxe2
I S .fxg7 Bxg7 1 6.Bxd8 Kxd8 1 7 .Bxe2 Bxb2 .

1 3 ... Bxb5 14.fxe7 Bxe7

Simple and good. Taking with the Queen via 14 . . . Qxe7 leads to a tempo
rary extra pawn for \Vhite after I S .Bxe7 Bxe2 1 6.Bxf8 Bxfl 1 7 .Bxg7 , though
Black would not be in any danger here, either.

1 5 .Bxe7
On I S . Qe3 , I intended to play l S . . . Bxfl 1 6 .Kxfl f6 1 7 . Bf4 Qa8 ! (Hitting
a2 and d S , and also defending against 1 8 .Bxd6? ? by 1 8 . . . Qa6+.) 1 8 .Re l
NeS 1 9.NxeS fxeS 2 0.BxeS dxeS 2 1 . QxeS 0-0 2 2 . Qxe7 Qxd S , with a hor
rible position for \Vhite.

1 5 ... Qxe7 16.Qxe7+ Kxe7 1 7.Bxb5 Rxb5 1 8 .0-0 Nb6 1 9.Rfe l + Kd7

The ending is, of course, advantageous for Black because of the weakness
of \Vhite's d-pawn and a-pawn. In this troublesome position, \Vhite tries
to create counterplay since passive defense is hopeless . Note that 2 0.NgS
is useless due to 20 . . . h6, when 2 1 .Nxf7 Rf8 traps the Knight.

20.a4! bxa3 2 1 .Rxa3 Ra8

CRE AT I NG T H E BE NKO G AMBIT

Also possible was 2 1 . . .Nxd5 n .Ra7+ Nc7, but why give him any activity
at all?

22 .Rae3 Nxd5 2 3 .Rd3 e4 24.Rd4 Re8? !

I t was better to protect the c-pawn with a unit o f lesser value. Thus,
2 4 ... Nb6 ! was correct, all the more so since that move prepares ... d6-d 5 .
Unplayable (after 2 4 . . . Nb6) i s 2 5 .Ne5 + Rxe 5 2 6 . Rxe5 Ra l + .

25.Red l Ree5 26.h4? !

After the game I suggested 2 6.Kfl to my opponent, which eliminates the
weakness of the first rank and, at the same time, brings the King closer to
the battle site on the queenside. In his article, Zaitzev gave 2 6.Kfl Kc6
2 7 .R4d2 Ra5 2 8.b4! and if 2 8 . . . cxb3 2 9.Nd4+. I must add that all this would
not really have been so simple for White, since Black had at his disposal
the same combination played in the game, although under less favorable
circumstances.

26 ... Rxb2 !
A very surprising offer of a Knight, completely confusing White in his
serious time shortage.

27.Nd2 ?
This loses immediately. Of course, White had to try 2 7 .Rxd 5 , though af
ter 2 7 . . . Rxd5 2 8 .Rxd5 c3 Black wins easily: 2 9.Nd4 (2 9.Ne l Rb I ; 29.Rd4
c2 3 0.Ne 5 + Ke6 3 1 .Nd3 Rb l +) 29 . . . Rd2 .

27 ... Nc3 28.Nxe4 Nxd l , 0- 1 .

(8 1) Kovacs - Benko (Debrecen, 1975)

l .d4 Nf6 2 .e4 e5 3 .d5 b5 4.exb5 a6 5.bxa6 Bxa6

Some years later, it became apparent that 5 . . . g6 is a slightly more accurate
move. The reason is that White can play a double fianchetto against the
immediate 5 . . . Bxa6 (i .e., 6.g3 d6 7 .Bg2 g6 8.b3 Bg7 9.Bb2), while 5 . . . g6
avoids this possibility since, in that case, taking on a6 with the Knight
turns out to be more desirable.

6.Nc3 d6 7.g3 g6 8.Nh3

8. Bg2 Bg7 9.Qa4+ is an interesting but ultimately flawed idea that was
tried in Barlay-Benko, U.S. Open 1 969: 9 . . . Nbd7 1 0.Qh4 h6! 1 l .f4 Rb8
1 2 .e3 Qa5 1 3 .Nge2 Bd3 , and Black had an excellent position.

8 ... Bg7 9.Nf4 Nbd7 1 0.h4

An interesting, very aggressive, idea. White intends to open the h-file and
will also place his Bishop on the active h3 -square.

10 ... 0-0
A highly provocative decision, but I wanted to see just how strong his at
tack really was. Much safer is 1O . . . h5, when modern praxis has shown that

255

PA L BE NKO: MY L I F E . GAMES A N D COMPOS I T I O NS

256

Black gets good play: 1 0 . . . h5 1 1 .Bh3 (Or 1 1 .Bg2 0-0 1 2 .0-0 Nb6 1 3 .Qc2
Ra7 1 4.b3 Qa8 1 5 .Bb2 Rb8, with comfortable play in A. Petrosian-Leko,
Lippstadt 1 993) 1 1 .. .0-0 1 2 .Qc2 Ne5 1 3 .0-0 Qb6 1 4.Rb 1 Bc8 1 5 .Bxc8 Rfxc8
1 6.b3 Nfg4 1 7 .Kg2 c4 1 8 .Ne4 Qa7 1 9.a3 cxb3 20.Qxb3 Nc4 2 1 .Ng5 Nxa3
2 2 .Qxa 3 Qxa 3 2 3 .Bxa3 Rxa3 , Ih - Ih , Yusupov-Topalov, Vienna 1 996.

1 1 .hS NeS 12 .hxg6 hxg6 1 3 .Bh3 QaS 14.Qc2 RfbS

Also deserving consideration is 1 4 ... Bc8 ! ? , when 1 5 .Bg2 Bf5 1 6.e4 Bg4
gives Black the initiative.

l S .Ne6
Hoping for 1 5 . . .fxe6 1 6 .Bxe6+ followed by f2 -f4, with a very strong
attack.

l S ... BhS
I didn't want to part with my KB , but much stronger was 1 5 . . . Nxd 5 !
1 6.Nxg7 Kxg7

when Ravikumar gives 1 7 .Qd2 (Black also wins after 1 7 .Bd2 Nb4 1 8 .Qd 1
Ned 3 + ! 1 9 .exd3 Nxd 3 + 2 0.Kf1 Nxb2 + 2 1 .Ne2 Nxd 1 2 2 .Bxa5 Rb2 2 3 .Bg4
Bb7 24.Rxd 1 Rxa5 2 5 .Rh2 f5 2 6.Bh3 Ba6 2 7 .f4 Raxa2) 1 7 . . . Nxc3 1 8 .Qh6+
( 1 8 . Bc8 Rxc8 1 9.Qh6+ Kf6 and the King escapes via e6) 1 8 . . . Kg8 1 9.Be6
Nxe2 + 2 0.Kd 1 (2 0.Kf1 Nf4+ mates, as does 20.Bd2 Nf3 + 2 1 .Kfl Nf4+,
and mate next move) 20 . . . Qa4+ 2 1 .Bb3 Nc3 + 2 2 .bxc3 Qg4+ 2 3 .Kc2 Qe4+
24.Kb2 Nc4 mate.

16.f4 Ned7 1 7.fS ?

White unreasonably insists on attacking, although he had many possible
continuations here. 1 7 .Bd2 , 1 7 .Ng5 , or even 1 7 .0-0 are reasonable alter
natives.

17 ...fxe6 l S.fxg6 Nf8 1 9.Bh6 Bg7?!

Overlooking White's next move and giving him chances t o revitalize his
attack. Correct is 1 9 . . . exd 5 .

20.Qcl !
I was expecting only 20.Qd2 , on which I planned 2 0 . . . Rxb2 ! 2 1 .Qxb2 Bxh6
and, having removed White's QB, I would be ready for a murderous coun-

CRE AT I NG T H E BE NKO G AM BIT

terattack. Fortunately, my error wasn't fatal, but it certainly created a lot

of excitement for the kibitzers.

20 ... Rxb2 ! !
The time for subtlety is gone. Now it's simply do or die !

2 1 . Bxg7 Rxe2+ 22 .Kd l

I t appears that Black's counterattack has been played out and that White,
who is threatening Qh6, is in the saddle again.

22 ... Ne4 ! !

The risk Black takes i s not that of losing his Rook (for 2 3 .Nxe2 Nf2 + 24.Kc2
Bd3 + leads to mate), but that White is able to chase the black King all
over the board. Curiously, 22 . . . Nxd S ? ? would not have led to the goal:
2 3 .Bxe6+ Nxe6 24.RhS+ Kxg7 2 S .Qh6+ Kf6 2 6.g7+ KeS (2 6 . . . Kf7 2 7 .QhS+
Kxg7 2 S .Rh7+ mates) 2 7 . QhS+ Kd4 2 S . QxdS + Kxc3 2 9.Rc l + , and Black
gets mated.

23 .Bd4
After long thought, White avoids the main line (resulting from 2 3 .Bxe6+),
realizing that it would not lead to a white mate, and he decides to try a
trick in view of my serious time pressure. Obviously, I had used too much
time calculating all the possible checks in the main line: 2 3 .Bxe6+ Nxe6
2 4.RhS+ Kxg7 2 S .Qh6+ Kf6 2 6.g7+ KeS 2 7 . Qxe6+ Kd4 2 S .Nxe2 + (No
better is 2 S .Qxe4+ Rxe4 2 9.Nxe4 Kxe4 3 0.RxaS Kd3 mates.) 2 S . . . Bxe2 +
2 9.Kxe2 Qd2 + 3 0.Kf3 Qf2+ 3 1 .Kg4 Qxg3 + 3 2 .KfS QgS mate. To use prob
lem terminology, White's Queen on e6 is a self-block.

23 ... Nxc3 +
Played immediately, and avoiding: 2 3 . . .cxd4? ? 24.Bxe6+ Nxe6 2 S .RhS+ Kg7
2 6 . Qh6+ Kf6 2 7 .g7+ Ke S 2 S . Qxe6 mate.

24.Qxc3 Qa4+ 25 .Kcl Qxd4 26.Qf3

Overlooking mate, but the game was lost anyway.

26 Qxa l mate.
..

257

PAL BENKO : MY L I FE, G AMES AND C O MPOSITI O NS

(82) Gordon - Benko (National Open, 1976)

25 8

l .d4 Nf6 2.c4 c5 3 .d5 b5 4.cxb5 a6 5 .bxa6 Bxa6 6.g3 d6 7.Bg2 g6

S.Nfl Bg7 9.0-0 0-0 1 0.Nc3 Nbd7 1 1 .Re l Qb6 12.Qc2 RfbS 1 3 .e4? !
A typical error that weakens the d3 -square. White should preface this central
advance with 1 3 .h3 , keeping Black's Knight off g4.

1 3 ... Ng4 14.Rb l

In my book on the Benko Gambit, and in my original notes to the game
Aspler-Benko, Vancouver 1 97 1 (where 1 4.h3 was played), I suggested that
1 4.Rb l could be met by 14 . . . c4! , threatening 1 5 . . . Qxf2 + 1 6 .Qxf2 Nxf2
1 7 .Kxf2 Bxc3 . The defense 1 5 .Bg5 looks best for White, connecting his
Rooks (which averts Black's threat) and bothering my e-pawn. Black would
have to decide whether to defend his pawn or to sacrifice it by 1 5 . . . h6 ! ?
Then on 1 6.Bxe 7 , Black can try to win the Bishop by 1 6 . . . Re8 . This would
lead to wild complications that seem to favor Black. I decided not to enter
this line because of the necessity of analyzing all those complicated varia
tions (I had to consider the clock, too), and I suspected that my opponent
had done his homework and was prepared with a new idea. As a result, I
preferred to try to exploit the weak d3 -square at once. After all, why walk
into an opponent's prepared analysis when you can avoid his tricks and
guarantee yourself a pleasant position?

14 ... Nge5 1 5 .Nxe5 Nxe5 16.Rd l Qb4!

An exceptionally strong Queen move that contains the hidden threat

1 7 . . . Bd3 ; e.g., 1 8 .Rxd3 Nxd3 1 9.Qxd3 Bxc3 , and the Bishop cannot be
taken because White's Rook is insufficiently protected. If 17 .a3 Qb3 , White's
queenside would be fatally weak.

1 7.Bfl Bxfl
The simple way. Also good is 1 7 . . . ND + 1 8 .Kg2 Bxfl + (There's nothing
wrong with 1 8 . . . Nd4, which leaves Black with enormous positional pres
sure.) 1 9.KxD (or 1 9 .Rxfl Nxh2 2 0.Kxh2 Bxc3 , with an edge for Black)
1 9 . . . Bxc3 , with a beautiful position for Black. It's often difficult to choose
among several good continuations !

CRE AT I NG THE BE NKO G AMBI T

l S.Kxfl Nc4 1 9.Rd3 Bxc3 20.Qxc3

Black wins the Exchange after 2 0 . Rxc3 Na3 . Now, after 2 0 . Qxc3 , Black
wins back the gambit pawn and, as is usual in such cases, White's remain
ing queenside pawn becomes irreparably weak.

20 ...Rxa2 2 1 .Qxb4 Rxb4 22 .b3 Na3

Another hard choice. As before, I chose the simplest possibility. The tempt
ing 2 2 . . . Ne5 2 3 .Rd2 (2 3 .Re3 Ng4) 2 3 . . . Rxd2 24.Bxd2 Rxe4 2 5 .Bc3 leaves
Black a pawn up with the better position, but I couldn't see how to in
crease my advantage with my Rook sitting, out of play, on the awkward
e4-square.

23 .Bxa3 Rxa3 24.Re3 c4 25.Ke2

A common type of Benko Gambit endgame, and a nightmare for anyone

with the White pieces. The pin, combined with Black's more active Rooks,
leave White with little hope of a successful defense.

2 5 ... Ra2 + ! ?
O f course, Black could take the pawn right away by 2 5 . . . Raxb3 2 6 . Rexb3
Rxb 3 . Then 2 7 .Rxb3 cxb3 2 8 .Kd3 leads to an interesting King and pawn
endgame where White can restore the material balance but, by doing so,
he must allow Black's King to penetrate on the other side of the board:
2 8 ... Kg7 2 9.Kc3 Kf6 3 0.f4 g5 3 1 .Kxb3 gxf4 3 2 .gxf4 Kg6 3 3 .Kc4 Kh5 3 4.Kb5
(34.Kd4 Kg4 3 5 .Ke 3 h5 looks hopeless for White) 3 4 . . . Kg4 3 5 .f5 Kf4
3 6.Kc6 f6 (36 . . . Kxe4 3 7 .f6 offers Black nothing but problems) 3 7 .Kd7 Kxe4
3 8 .Kxe7 Ke5 (White will find himself in Zugzwang) 3 9.h3 h6 40.h4 h 5 ,
when it's time for White t o resign. Naturally, this isn't all forced. After
2 5 . . . Raxb3 2 6.Rexb3 Rxb 3 , White doesn't have to go into the lost pawn
endgame but can play 2 7 .Rc 1 , when Black will have trouble retaining his
extra pawn. However, after 2 6 . . . cxb3 (instead of 2 6 . . . Rxb3), Black could
still reach an endgame that is similar to the game. I decided that I could
win the pawn later; first I wanted to keep White tied up while I activated
my forces to the maximum.

26.Ke l Rc2
To stop White's Rc3 .

259

PAL BE NKO: MY L I F E,G AMES AND C O MPO S I T I O NS

27.h4 h5 2S.Kfl Kf8 29.Kg2 KeS

260

Black's King, which is heading for c5, starts to take on an active role. Once
it gets to c 5 , it can go to either d4 or b4, after which Black would play
. , .c4-c3 , deciding the game. VVhite can do nothing but wait for the axe to
falL If 3 0.RB f6, Black is safe. My young opponent reacts as young play
ers typically do-he loses patience and tries to become active, which only
hastens the end,

30.Ra l ? cxb3 3 1 .RaS+ Kd7 32 .Ra7+ Rc7 B .Ra l

The only chance was 3 3 .Rxc7+ Kxc7 3 4.Re l , but after 34. "b2 the black
King ends the struggle by walking calmly to a3 .

B ... b2 34.Rbl Re t , 0- 1 .

ed 1 968 " I d;d rno" ye''' -pbjriog ;o ",riOllS Ame,;,,,, Smsses.

Aside from a few opens, my only other chess events occurred in the latter half of
the year. Clearly, something else was on my mind, and that "something" was a
promise I had made to myself many years before: I had decided that if I wasn't
married by my fortieth birthday, I would
remain single forever. As my birthday on
July 1 5 th approached (and with it, the
magic number of forty), I realized it was
then or never. The first thing that had to
be done was get government approval,
which they gave on the condition that my
wife would remain in Hungary (short trips
were acceptable). Gizella was happy to ac
cept this since she was teaching math in
university and loved her work.
We mixed our honeymoon with the
Lugano Olympiad, and she also accom
panied me to events in Palma de Mallorca
and Monte Carlo. Though I returned to
the United S tates for the 1 9 6 8 U. S .
Championship, I naturally began spend
ing more and more time in Hungary, and
this led to my playing in many more
European tournaments .
Wedding bells for Pal and G izel l a.
26 1

share a laugh with Robert Byrne at the 1 9 68 Lugano Olympiad.

(83) Benko

Byrne (USA, 1968)

l .NfJ cS 2 .e4 d6 3 .d4 cxd4 4.Nxd4 Nf6 S.Nc3 Nc6 6.BgS Bd7
A refreshing change from the usual 6 . . . e6. With 6 . . . Bd7, Black starts to
develop his queenside quickly and tells White that he doesn't fear the dou
bling of his pawns by Bxf6.

7.Qd2
In 1 992 , Byrne played the white side of this line (against Goldin in Phila
delphia) and decided to chop on f6. However, Black got an excellent posi
tion after 7 .Bxf6 gxf6 8.Nf5 Qa5 9.Qh5 e6 1 O .Ng3 Qe5 ! 1 1 .0-0-0 a6 1 2 .Kb l
b 5 1 3 .Bd3 Be7.

7 ... RcS S.O-O-O Nxd4 9.Qxd4 QaS 1 0.Bd2

At this point I became suspicious that Byrne was trying to inveigle me
into some of his private analysis. We had both seen the variation earlier in
the year at the Monte Carlo tournament, in the game Gheorghiu-Hort:
1 O.f4 Rxc3 l 1 .bxc3 e5 1 2 . Qb4 Qxb4 1 3 .cxb4 Nxe4 1 4.Bh4 Nc3 1 5 .Re l
Nxa2 + 1 6.Kb2 Nxb4 1 7 .fxe5 d 5 1 8 . Rd l Be7 1 9.Bxe7 Kxe7 , draw agreed.
This is a very interesting variation, obviously not all forced (later it was
discovered that 1 5 .Rd3 ! Nxa2 + 1 6.Kb2 Nxb4 1 7 .Rb 3 Nc6 1 8 .Rxb7 favors
White, but that Black could have improved earlier by 1 4 . . . g5 ! , with fasci
nating complications), and the final position is unclear. I decided to avoid
it and, after calmly studying the position for a while, came up with this
new move. At least Byrne had to stop and think!

1 0 ... a6

E U RO P E AT MY F E ET

Eventually 1 0.Bd2 e5 turned into one of the main lines, but Byrne's 1 0 . . . a6
still retained some devotees. Less good is 1 O . . . Qc5 ? ! I 1 .Qxc5 dxc5 (Poor.
Better is 1 1 . . .Rxc5 , though White retains a nice spatial plus after 1 2 .f3 !
[and not 1 2 .Be3 Rxc3 1 3 .bxc3 Nxe4, with compensation for the Exchange]
12 ... a6 1 3 .Be3 Rc8 1 4.g4.) 1 2 .Bf4 (Also good for White is 1 2 .f3 e6 1 3 .Bf4.)
1 2 ... a6 1 3 .Be2 Bc6 1 4.Nd 5 , and White stood better in Benko-Klein, Ventura
1 97 1 . The conclusion is entertaining: 1 4 . . .Nxd5 1 5 .exd5 Bd7 1 6.Bf3 f6?
1 7 .h4 h5 1 8 .Rhe l Kd8 1 9. Re6! g5 2 0.hxg5 fxg5 2 1 .Bxg5 Bxe6 2 2 .dxe6+
Ke8 2 3 .Bxb7 Rc7 2 4.Bxa6 Bh6 2 5 .Bxh6 Rxh6 2 6.Bb 5 + Kf8 2 7 . Rd8+ Kg7
2 8 .Rd7 Rc8 2 9 . Rxe7+ Kf8 3 0 .Rf7+ Kg8 3 1 . Rb7, 1 -0.

1 1 .Be4
Also good is 1 1 . f3 .

1 1 . .. Qe5
Instead, Black could win a pawn by 1 1 . . .e5 1 2 .Qd3 Qc5 1 3 .Bb3 Qxf2 , but
White's attack is too strong after 1 4.Rfl .

12 .Qxe5 Rxe5 1 3 .Bb3 g6 14.Be3 Re8

Mter this, Black finds himself in trouble. Had he foreseen what was com
ing, he would undoubtedly have played 14 . . . Rc6. After 14 . . . Rc8 , White is
able to grab the initiative by making use of his lead in development.

1 5.Bd4 Bg7 16.Nd5 Nh5 1 7.Bxg7 Nxg7 1 8.e5 !

Suddenly Black is in trouble !

1 8 ... Re6
Of course, White wins after 1 8 . . . dxe 5 1 9.Nb6 Rc7 20.Rxd7 Rxd7 2 1 .Ba4
Kd8 2 2 .Nxd7 b5 2 3 .Nxe 5 .

1 9.Nb4 Rb6 20.Nd5 Rc6 2 1 .Ne3

Also possible was 2 1 .Rhe 1 and 2 1 .exd6 Rxd6 22 .Rhe 1 Be6 2 3 .Nf6+ Kd8
2 4.Rxd6+. I didn't see how to get more from the position than the isola
tion of Black's d-pawn.

2 1 ...Be6
And not 2 1 . . .dxe5 22 .Bxf7 +.

22.exd6 Rxd6 23 .Rxd6 exd6

263

PAL B E N KO : MY L I F E , GAMES AND C O MPO SITI O N S

White has achieved some success after the opening - he's reached an
endgame in which Black has a permanent organic weakness. But the game
is not yet over - the victory still has to be earned.
264

24.Rd l Bxb3 25.axb3 Kd7 26.Nd5 Ne6 27.b4 Nc7 28.Nf6+ Ke7 29.Ne4
Rd8 3 0.Re l Ne6 3 1 .Kd2 b6?
Under pressure, Black loses patience. The purpose of the text is to pre
pare . . . dS without fearing the answer NcS . Nevertheless, it is a weakening
of the queenside which soon makes itself felt.

32 .Nc3 d5 3 3 .Ra l Nc7 34.Ne2 Ra8 3 5.Nd4 Kd6 36.Ra3 !

The Rook acquires greater activity on the third rank, threatening Rf3 and
Rc3 to try to break into Black's position. Black's answer proves inadequate,
but it is far from easy to find a good defensive plan. Perhaps best was 36 . . . aS .

36 ... Ne6 3 7.Nxe6! fxe6

It would appear that 3 1 . . .Kxe6 offers more resistance, but after 3 2 .bS as
B . Rc3 RdS 3 S .Rc7 Rd7 3 S .RcS White still has winning chances because
of his active Rook and Black's isolated pawn. Understandably, Black was
happy to solve his isolated pawn problem.

3 8.b5 ! Rf8
A desperate try for active counterplay. If 3 S . . . aS 3 9.Rc3 RbS 40'4, Black's
game is hopelessly lost: White will put his Rook on c6 and march his King
to the center.

3 9.bxa6 Rxf2 + 40.Kc3 Rf8 4 1 .Kb4 d4 42 .a7 Ra8 43 .Kb5 Kc7 44.Ka6
In time pressure, White misses the best continuation several times. Here
or on the next move, Rf3 wins much faster. For example: 44.Rf3 Rxa7
4 S . Rf7 + KbS 46.Rxa7 Kxa7 47.Kc4 Kb7 4S.Kxd4 Kc6 49.KeS Kd7 SO.c4,
with an easy win.

44... e5 45 .Ra4 b5 46.Rb4 Kc6

I suddenly realized that my whole plan was wrong-I cannot take the b
pawn because 47 . Rxb6 Rxa7+, and White actually loses. My only winning
chance, therefore, was to advance the other queenside pawns. So the race
is on!

47.c4 bxc4 48.Rxc4+ Kd5 49.Rc7 d3 50.Rd7+ Kc4 5 1 .b4 e4 52.b5 Rxa7+

EURO P E AT MY F E ET

This was Black's sealed move. The position intrigued several players, some
of who were convinced that Black wins ! Later that evening Byrne and I
sat down to analyze the game as the guests ofJack Howard, a most hospi265
table and hard-working tournament organizer. Byrne revealed his sealed
move and, after analyzing the adjourned position, realized he was lost and
promptly resigned. The game could con
tinue: 5 3 . Rxa7 e3 54.Rc7+ Kd4 5 5 .b6 e2
56.b7 e l =Q 5 7 .b8=Q Qa l + 5 8 .Kb7 Qb2 +
59.Kc8 (had Black answered 54.Rc7+ with
5 4 . . . Kd 5 , in order to try for a perpetual
check with 59 . . . Qh8+, White would play
59 .Ka8) 5 9 ... Qxb8+ 60.Kxb8 h5 (not 60 . . . d2
because of 6 l .Rxh7) 6 1 .Kc8 d2 62 .Rd7+ Ke3
6 3 .Kd 8 , and wins because White simply
takes off all the Black pawns. Actually,
5 2 . . . Rxa7+ wasn't the best sealed move, but
White wins anyway-in that case, the game
is decided by just one tempo: 5 2 . . . e3 5 3 .b6
Re8 (5 3 ... e2 54.b7 Rxa7+ 5 5 .Kxa7) 54.Rc7+
Kb4 5 5 .Rc6 ! e2 (5 5 ... Re5 56.b7 Ra5 + 5 7 .Kb6
1 968.
Rb5 + 5 8 .Kc7) 5 6.b7 e l =Q 5 7 .b8=Q+.
(Photo Art Zeller. courtesy USCF.)

(84) Benko - Formanek (USA, 1968)

l .g3 e5 2 .c4 d6 3 .Nc3 f5 4.d4 e4

Black wants no part of 4 . . . exd4 5 . Qxd4, since the pawn on f5 would then
be nothing but a weakness. In my opinion, Black's best option is 4 . . . Be7
since 5 . dxe5 dxe5 6.Qxd8+ Bxd8 offers White very little.

5.0 exB 6.exB

Also possible was 6.Nxf3 , but it seems to me that 6.exf3 controls more
central squares.

6 ... Nf6 7.Bg2 g6 8.Nge2 Bg7 9.0-0

This position simply favors White. A more modern example of White 's
possibilities is 9.Be3 0-0 1 O.Qd2 d5 I l .cxd5 Nxd5 1 2 .Nxd5 Qxd 5 1 3 .0-0
c6 1 4.Rfe l Qf7 1 5 .Bh6 Bxh6 1 6.Qxh6 Nd7 1 7 .Nf4 Nf6 1 8 .Bfl Bd7 1 9.b3
b 5 2 0.Re 5 , with a huge advantage in Vaulin-Grabarczyk, Poland 1 998.

9 ...0-0 1 O.Be3 Re8 l 1 .Qd2 Be6 l 2 .b3 Bf7 l 3 .Rae l

It doesn't take more than a glance to convince oneself that White has a far
superior position: He has more central space, a lead in development, and
possibilities of central expansion (aimed at dominating the e6-square) by
d4-d5 .

l 3 ... Nc6 l4.d5 Ne5 l 5 .h3

Now Black has to worry about both f3 -f4 and g3 -g4.

PAL BE NKO: MY LIFE, G AMES AND C O MPO S I T I O NS

1 5 ... a5 1 6.Nd4
The e6-square is suddenly a tender point in Black's camp.
266

1 6 ... Ned7 1 7.g4!

White has more central space and he has also completed his development
(Black still hasn't found good squares for all of his pieces). In such situa
tions, opening up the position is always in favor of the better-mobilized
side, following the rule that the better-developed side (White has all his
men in play while Black still hasn't found good squares for all his pieces)
should strive to open up the position.

17 ... fxg4 I S.fxg4 Nc5

Black is trying to cover the weak e6-square. If l S . . . NeS , then 1 9.BgS would
have been strong.

1 9.Bg5
Not letting Black trade pieces and relieve his cramped position with . . . Nfe4.

19 ... Rxe l 20.Rxe l Qd7 2 1 .Ne6!

An unexpected and powerful blow. Black can't play 2 1 . . .Nxe6 due to 2 2 .dxe6
Bxe6 2 3 .Bxf6 Bxf6 2 4.Rxe6 Qxe6 2 S .BdS, pinning and winning the black
Queen.

2 1 ...ReS
Trying to shake off the pressure by trading pieces. Other moves are even
worse. For example, 2 1 . . .BhS is strongly met by 2 2 .Nb S ! NeS (2 2 . . . Bxe6
2 3 .dxe6 Qe7 24.Nxc7! Qxc7 2 S .e7 ReS 2 6.BdS+ NxdS 2 7 . QxdS + Kg7 2 S.Rfl
Rxe7 2 9 .Bf6+) 2 3 .Nexc7 Nxc7 24.Re7, and White wins.

22 .Nxg7 Rxe l + 2 3 .Qxe l Kxg7 24.Nb5!

This wins a pawn because 2 S .Qc3 is a serious threat.

24 ... NeS 25.Qxa5 h6 26.Qc3 + K.h7 27.Bcl

Black is busted. He's a pawn down and the loss of his fianchettoed Bishop
has left the dark squares around his King horribly weak.

27 . . . KgS
Trying to run, but it's far too late .

2S.Bb2 Kf8 29.QhS+ Ke7 3 0.Qxh6

In combat against O ' Kel ly, Spain 1 9 68.

Down two pawns, Black no longer has anything to play for. For some rea
son, he continues the game, but things quickly become even worse.

30 ... KdS 3 l .Qe3 BgS 32 .h4 Na4 3 3 .Bd4 e6 34.dxe6 bxe6 3 S .QgS+ KeS
36.Na7+ KbS 3 7.Bxe6, 1 -0.
There's no reason to suffer through more abuse.

(85) Benko - Horowitz (U.S. Championship, 1968)

l .e4 e6 2 .Nfl dS 3 .b3 Nf6 4.g3

The development of the Queen is delayed since it may be played to either
c2 or b3 , depending on circumstances. Although the text seems to allow
Black to play 4 . . . Bg4 5 . Bg2 Bxf3 6.Bxf3 dxc4 7 . bxc4 Qd4, attacking the c
pawn and the Rook, White turns the tables with 8.Qb3 Qxa 1 9.Qxb7 , threat
ening mate as well as the Rook. A similar position arose in the famous
game Reti-Em. Lasker, New York 1 92 4 where Lasker played . . . Bf5 .
Horowitz prefers a more conservative line .

4 ... e6 S .Bg2 Be7 6.0-0 0-0 7.Bb2 as S.a3

Easily parrying Black's attempts at queenside counterplay. If 8 ... a4, White
must avoid the tempting 9.b4 dxc4 1 0.Ne 5 , since 1 0 . . . b5 1 1 .Nxc6 Nxc6
1 2 .Bxc6 Rb8 is quite nice for Black. However, 9.cxd5 exd5 1 O.b4 gives
White a small plus.

S ... eS
This move is an obvious loss of time as the pawn has taken two moves (in
the opening!) to reach a square it could have reached in one. However,
Black had a method behind his madness: he planned to exploit the sup
posed weakness of White's b3 . As we shall soon see, he does manage to do
so, but at heavy cost.

9.exdS exdS 10.d4 Na6

PAL B E N KO : M Y L I F E , G A M E S AND C O M POSITI O N S

A strange place for the Knight, but consistent with Black's idea. I rather
expected 1 0 . . . b6, avoiding the isolated pawn and giving the QB a promis
ing future on a6.
268

l 1 .Nc3 Bf5

In a roundabout way, the opening has transposed into a sort of Tarrasch

Defense to the Queen's Gambit. The important difference is that Black's
Knight is on a6 instead of c6 (where it would pressure both d4 and e5).
White's problem now is how to take advantage of this. I decided that since
the black Knight wasn't influencing the center, I would take the opportu
nity to play Ne 5 , even though by doing so I would "fall into" Black's plan
to attack my b 3 . I intended to refute his plan by combinative means.

1 2 .Ne5 !
Threatening 1 3 .dxc 5 , followed by the win o f the d5 -pawn.

1 2 . ..cxd4 1 3 .Qxd4
Also possible was 1 3 .Nb5, with positional pressure against the isolated pawn,
but the text is sharper.

1 3 ...Nc5
Both players continue according to plan. If White was now obliged to de
fend the b-pawn by Qd l , Black would have good reason to be satisfied
with his position. But White has other ideas.

14.Nxd5! Nxb3 1 5.Qf4!

Interesting but probably insufficient is 1 5 .Nxe7 + Qxe7 1 6 .Qf4 Nxa 1
1 7 . Qxf5 Nb3 1 8 .Ng4 when 1 8 . . . Qe6? is very good for White: 1 9.Nxf6+
gxf6 2 0 .Qf4! Ra6 (losing instantly is 2 0 . . .5 2 1 .Bd5 !) 2 1 .Be4 Rd8 n .Bf5
Qd5 2 3 .Bd3 Qg5 24.Bxa6 Qxf4 2 5 .gxf4 bxa6 2 6.Bxf6. However, instead
of 1 8 . . . Qe6, better is 1 8 . . .Nxg4! 1 9 .Qxg4 g6 2 0 . Qf4 Rfe8 2 1 .Bd5 a4, and
Black can defend, though White still retains a dangerous attack by n .Rd l
when the threat of Rd 3 is hard to deal with. In fact, it is not quite clear
who stands better in this variation.

1 5 ...Nxd5
Black could have avoided White's combination by 1 5 . . . Nxa 1 , but he would
still be overwhelmed : 1 6 . Qxf5 Nb3 1 6 . . . Nxd5 transposes into the game.

EURO P E AT MY F E ET

1 7 .Rd l Qd6 ( 1 7 . . . QeS I S .Nxf6+ Bxf6

1 9 . B e4; 1 7 . . . KhS I S . Rd 3 ) I S .Nc4 !
(Much stronger than I S .Nxe7+ Qxe7
1 9 .Ng4 Nxg4 2 0 . Qxg4 g6, though
Black's position is hanging by a thread
after 2 1 . Rd 7 ) I S . . . Q e 6 ( 1 S . . . Q c 5
1 9 . Bxf6 Bxf6 2 0 .Be4) 1 9 .Qxe6 fxe6
2 0.Nxe7 + Kf7 2 1 . Bxf6 Kxf6 2 2 .Rd7,
and White wins. Black's most prom
ising defense appears to be 1 5 . . . Bc2 ,
but 1 6 .Rad I ! ! still leaves Black facing
incre d i b l e pre ssure : 1 6 . . . Bxd 1
1 7 . Rxd l Kh S I S . Q f5 Nc5 ( I S . . . a4
1 9.Ng4) 1 9.Rd4! gives White a win
ning attack.

269

Famed chess author I A

Horowitz (left) with Bill
Lombardy.
(Photo courtesy USCF.)

16.Qxf5 Nxal
Black cannot try 1 6 . . . g6 because 1 7 .Ng4! wins on the spot. He might
have tried 1 6 . . . Nf6 , but White's advantage after 1 7 . Rad 1 would be
obvious.

1 7.Nxfi ! !

The killer ! Incorrect i s 1 7 .Bxd5 Qxd 5 I S .Ng6 Qc5 (Even better than
I S . . . Qe6 1 9 .Qxe6 fxe6, which also favors Black.) 1 9.Qxc5 Bxc5 2 0 .NxfS
Nc2 2 1 .Nd7 Bxa 3 , when Black wins. After this unexpected Knight sacri
fice, Black has no way to save himself.

1 7 ... Qe8
1 7 . . . Nc7 is answered by I S .Be4 (This move of the fianchettoed Bishop to
the center in order to threaten the enemy kingside is an Alekhine trade
mark - it appears in a number of his games.) I S ... Ra6 1 9.Qh5 threaten
ing 2 0 .Bxh7+ with mate to follow as well as simply 2 0 .NxdS. The Knight
can't be touched. 1 9 . . . Rh6 20.Qxh6 and Black can resign. Actually, best
for Black is 1 7 . . . Rxf7 , but after I S .Bxd5 he's a pawn down with a very bad
position. Black prefers to die with his boots on.

1 8.Nh6+ Kh8 1 9.Qxd5 Ne2

PA L B E N KO : MY L I F E , GAM ES A N D C O M POSIT I O N S

Not that it matters at this point, but 19 ... QeS would have held out a bit
longer. In that case, 2 0.Rxa l would give White a decisive material superi
ority. The text allows a primitive mate.
270

20.Qg8+, 1 -0.
You don't get to play a classic smothered mate too often (2 0 ... RxgS 2 1 .Nf7
mate).

(86) Benko - Reshevsky (U.S. Championship, 1968)

l .d4 Nf6 2 .c4 cS 3 .NO

Black wants to play a Benoni, but White wants to swim the English
Channel.

3 ... cxd4 4.Nxd4 Nc6

In the late seventies, some players began experimenting with 4 . . . e5 5 .Nb5
d5 6.cxd5 Bc5 with real compensation for the sacrificed pawn. Of course,
giving away a pawn so early in the game isn't to everyone's taste.

S.Nc3 e6 6.g3
Possible here is 6.N4b 5 , when Black can play 6 . . . d5 (on 6 . . . Bb4, White gets
a slight edge with 7 .Bf4, while 7.a3 Bxc 3 + S .Nxc3 d5 is only even) 7 .Bf4 e5
S.cxd5 exf4 9.dxc6 bxc6 1 O.QxdS+ KxdS, with an interesting position that
was thoroughly explored in the 'SOs and '90s (roughly equal).

6 ... Qb6
The sharpest reply. Black also plays 6 . . . Bb4 and 6 . . . Bc5 from time to time,
but the text move is by far the richest possibility.

7.Nb3
I must admit (with a touch of embarrassment) that I wasn't sure which
Knight move was correct (b3 or c2). The purpose of 7 .Nc2 is to avoid the
pin that Black now applies and to be able to defend the c-pawn with Ne3 if
necessary. On the other hand, it lets Black develop his dark-squared Bishop
to the active c5 square, and this is the reason 7 .Nc2 has been rejected by
modem theory (7 . . . Bc5 S.e3 d5 ! with a very active game). Because of this,
7 .Nb3 is White's main choice here. It keeps the black Bishop off c5, and
also prepares tactical c4-c5 advances in several variations.

u.s. vs. Soviet Union at

Lugano 1 968. Evans and
I on the right, Spassky
seated on left, while
Petrosian stands.

(Photo courtesy USCF.)

EURO P E AT MY F E ET

7 ... Bh4
The modern main line is 7 . . . Ne 5 8 .e4 Bb4 9.Qe2 with a sharp, unclear
battle in store . However, theory also thinks highly of Reshevsky's choice.
27 1

8.Bg2 d5
Also leading to interesting complications is 8 . . . Qa6 ! ? 9.c5 b6, though the
position after l O.O-O! seems promising for White.

9.cxd5 Nxd5
On 9 . . . exd5 I planned l O.Be3 and 1 1 .0-0. Now White can win a pawn by
l O.Bxd5, but giving up my light-squared Bishop in this way would be quite
risky. Instead, I decided to sacrifice a pawn and play for the initiative.

10.0-0
The simple l O .Bd2 is also playable, but the text seemed crisper. After all,
if Black can't get away with winning the pawn on c3 , then why bother
playing a defensive move?

1 0 ... Nxc3
Most natural. I would have been happy to play the position after l O . . . Bxc3
l 1 .bxc3 0-0 1 2 .c4.

1 1 .hxc3 Be7
If Black accepts the pawn with 1 1 . . .Bxc3 , then 1 2 .Be3 followed by Re I
gives White more than enough compensation. After 1 1 . . . Bxc3 , the move
suggested by several spectators, 1 2 .Ba3 , is unsound : 1 2 . . . Bxa l 1 3 .Qxa l f6
followed by 1 4 . . Kf7 , and White can resign.
.

12 .Be3 Qa6
White has an obvious lead in development to compensate for his broken
pawn formation. He must therefore act at once as otherwise his advantage
will evaporate. Reshevsky offered a draw here and, after quite a bit of con
sideration, I declined because I conceived a long-range plan to obtain the
better endgame.

1 3 .Bc5 0-0 14.Qd3 ! Qxd3 1 5 .exd3 Rd8

Perhaps 1 5 . . . Re8 ! ? is a slightly more accurate defense .

1 6.Bxc6 Bxc5 1 7.Nxc5 hxc6 1 8.f4

PA L B E N KO : M Y L I F E , GAM ES AND CO M PO S ITI O N S

272

White's plan has become clear: he has exchanged all of Black's developed
pieces, leaving him with only an inactive Bishop trapped behind his own
pawns. When you compare this with the active white Knight, you begin
to realize how serious Black's predicament really is.

1 8 . . . Rb8 1 9.Rab l
Making sure that Black doesn't get any counterplay on the b-file.

19 ... Rb5 20.d4 Kf8 2 1 .Kf2 Bd7 22.Rb3 Ke7 23 .c4 Rb6 24.Rili I Rdb8
2 5 .Na4!
This forces the exchange of one pair of Rooks, after which the a-file be
comes important.

2 5 ... Rxb3 26.axb3

Black could defend himself much more easily after 2 6 . Rxb3 Rxb3 2 7 .axb3 .
In the present position, White's Rook will become very active on the a
file, while Black's Rook is nothing more than a passive defensive piece.

26 ... Kd6 2 7.Nc5

The Knight has done its job and now returns to the dominating c5-square
where it defends b 3 , thus freeing the Rook for more aggressive pursuits.

27 ... Bc8 28.Ral a6 29.Ra3 Ra8 30.Ke3 Kc7 3 1 .Ne4 h6 32 .c5

Creating an even better home for the Knight on d6.

32 ... Bd7 3 3 .Nd6 f5

I would have preferred the safer 3 3 . . .f6, but Black was probably afraid of
an eventual f4-f5 breakthrough.

3 4.Nc4
The Knight enjoys the freedom of the board, hopping merrily from square
to square. However, this doesn't really accomplish anything since d6 is
the Knight's best post. The most effective plan was 3 4.Kd2 followed by
walking the King to a 5 . A shortage of time prevented me from playing in
the most accurate manner.

34 . . . Be8 3 5.Nb6 Ra7 3 6.Ra I Bf7 3 7.Ra3 Bh5 3 8 .b4 Be8 3 9.Kd2 Bh5
40.h3 Kd8 41 .Nc4

The sealed move. Although White has wasted some time, he still holds a
significant edge. After considerable analysis, I decided that my plan should

EURO P E AT MY F E ET

be to keep Black's pieces restricted by maneuvering my King to a s . Black

can try to free himself with . . . g7 -gS , but in that case White either forces
an open file for himself or blocks the position by h3 -h4.
273

41 ...BeS 42 .Re3 Re7

And not 42 . . . Ke7 ? , which loses a pawn to 43 .Nd6 .

43 .Ra3 Ra7 44.Ne5 Kc7 45.Ral KdS 46.Kc3 Kc7 47.Kb3 RaS 4S.Nc4
Bh5 49.Re l Bf7
Of course, 49 . . . Re8 ? SO.Nd6 Re7 S l .NxfS loses material.

50.Ka4 e5
Black finally loses his patience, but what else was he to do? Passive de
fense would have lost to KaS , Nd6 and g4, eventually opening a file and
allowing my Rook to penetrate decisively into my opponent's position.

5 1 .Nb6
Better than S I .NxeS, because after S I . . . BdS Black's Bishop is free and White
doesn't have a passed pawn.

5 1 ...ReS 52 .Rxe5 Rxe5 53 .dxe5 Be6

The position is hopeless for Black. In the final phase of the game I play to
maneuver my Knight to d4 when the blockade on e6 will be broken and
the fS -pawn will be attacked.

54.Ka3 Kb7 5 5.Na4 g6 56.Nc3 Bc4 57.Nd l

Not letting Black get some play after S 7 .Kb2 a s .

57 ... Bfl 5S.h4 Bc4 59.Ne3 Be6 60.Ka4 h5 61 .Nc2 , 1 -0.

Black can't prevent Nd4 and Ka S followed by pushing (and sacrificing)
the e-pawn in order to win both queenside pawns.

While 1 968 was a relatively quiet year as far as chess is concerned, 1 969 turned
out to be very busy! Since I was in Hungary, all of Europe lay before me and I
availed myself of this geographical opportunity by playing in the quiet little sea
side town of Wijk aan Zee in Holland. As a man that loves hot climates, the
bitter chill of a Dutch winter was not to my liking, and I made a beeline to Malaga's
Spanish sunshine. After this, I virtually went berserk (in a purely chessic sense !),
hitting tournaments in Monte Carlo, Netanya, Venice and Vrsac. I also did my
American "duty" by playing in the U.S. Open and U.S. Championship!
This pattern continued in 1 970 (Caracas, Wijk aan Zee, Malaga, Reggio Emilia,
Saraj evo, Siegen Olympiad, U.S. Open) , 1 9 7 1 (Malaga, Netanya , Palma de
Mallorca, Reggio Emilia, Vrnj acka Banja, U.S. Championship), 1 972 (Wijk aan
Zee, Las Palmas, Novi Sad, Skopj e Olympiad, U.S Championship), and 1 9 7 3
(Hastings, Sao Paulo, Vrnjacka Banja, Malaga, Orense, Torremolinos, U. S. Cham
pionship), with me playing in many European events while my days of American
opens more or less came to an end.

PAL B E N KO : MY L I F E , GAM ES A N D CO M PO S I TI O N S

274

In 1 97 1 Fischer was at the Grossinger Hotel in the Catskills preparing for his
match with Spassky. He invited me to work with him, but at that time I had
already accepted an invitation to a European tournament and had (rather sadly)
to decline, though I did come up with one counter "variation": I offered to with
draw from the tournament and help him out if he also hired me to be his second
in Iceland. Unfortunately he couldn't make up his mind, so I had to walk away
from this possibility as well (he eventually chose Lombardy).

(87) Benko - Ostojic (Wijk aan Zee, 1969)

l .Nf3 g6 2 .d4 Bg7 3 .e4 d6 4.Nc3 Nf6 5.Be2

This, the Classical System, is one of White's best choices against the Pirc.

5 ... 0-0 6.0-0 Bg4

Other moves are 6 . . . c6, 6 . . . Nc6, 6 . . . a6, 6 . . . Nbd7, and 6 . . . b6. Bringing the
Bishop out to g4, though, has always been a very popular response.

7.Be3
White can also play for an edge with 7.h3 Bxf3 8.Bxf3 Nc6 9.Ne2 with
c2 -c3 to follow.

7 ... Nc6 8.Qd2

Not the only promising move. The straightforward 8.d5 might be just as
strong.

8 ... e5
Another idea is 8 ... Re8 ! ? , an innovation of mine that I used (as Black) against
Ivkov in Caracas 1 97 2 . The idea is that after 9 .Rad 1 e5 1 0.d5 Bxf3 1 1 .Bxf3
Nd4 is now possible due to 1 2 .Bxd4 exd4 1 3 .Qxd4 Nxe4. Unfortunately,
the Rook move might turn into a loss of tempo in some lines. For ex
ample: 8 . . . Re8 9.Rad 1 (Karpov gained a slight plus against Spassky in Ham
burg 1 982 with 9.Rfe 1 a6 1 0.Rad 1 Bxf3 1 1 .Bxf3 e5 1 2 .dxe5 dxe5 1 3 .Na4
Qe7 1 4.c3 b6 1 5 .Qe2) 9 . . . e5 1 O.dxe5 dxe5 1 1 . h3 , when Black's ... Re8 doesn't
have any positive significance.

9.d5
At the time, 9.dxe5 dxe5 1 O.Rad l was all the rage, though, in my opinion,
1 O . . . Qc8 ! doesn't give White any advantage (I played this against Hug in
Sao Paulo 1 97 3 ) . An instructive Black idea is shown after the following
moves: 1 1 .Qc1 Rd8 1 2 .Rxd8+ Qxd8 1 3 .Rd 1 Qf8 1 4.h3 Bxf3 1 5 .Bxf3 h 5 !
1 6.Nb5 Rc8 1 7 .c3 Kh7 ! , when . . . Bh6 will follow with approximate equal
ity (as in P. Cramling-Yrjola, Gausdal 1 984).

9 ...Ne7
We have transposed into a King's Indian type of position, with the differ
ence that White's c-pawn stands on c2 instead of c4. This means that White
can't use this pawn as a queenside lever by c4-c5 but, on the other hand, it
also means that White's d4-square can be covered by c2 -c3 .

EURO P E AT MY F E ET

10.a4
This move gains queenside space and stops possible . . . b7-b5 advances (for
example, 1 O.Rad l Bd7 1 1 .Ne l b5 ! has recently become popular for Black).

10 ... Ne8

275

Keeping the c8-g4 diagonal clear for his light-squared Bishop. The more
natural l O . . . Nd7, as played in Benko-Seirawan, U.S. Open 1 9 7 5 , lets White
trade off his bad Bishop for Black's
good one by I l .Ng5 or 1 1 .Ne l .
In that game I decided to toss in
I l .a5 first, when 1 1 . . .a6 1 1 .Ng5
would follow. However, Vasser
took on f3 instead and soon got
into a difficult position: 1 1 . . . Bxf3
(and not 1 1 . . .f5 ? 1 2 .Ng5) 1 2 . Bxf3
f5 1 3 .Qe2 ( 1 3 .g3 ? is a mistake be
cause of 13 . . .f4!, when 14.gxf4 exf4
1 5 .Bxf4 g5 wins a piece for Black)
1 3 ...Nf6 1 4.Bg5 fXe4 (there was no
hurry to give this square to White.
Vs. Bronstei n in Monte Carlo, 1 9 69.
B e tter was 1 4 . . . a6 or 1 4 . . . f4 ,
(Photo William Lombardy, courtesy USCF.)
though the absence of Bla ck's
light-squared Bishop makes any kingside breakthrough difficult to achieve)
1 5 .Bxe4 Qd7 1 6 .a6 b6 1 7 .Bxf6 Rxf6 1 8 .Nb5 Nf5 1 9.c3 Bh6 2 0 . Ra4 Rf7
2 1 .Bf3 Bg5 2 2 .Rb4 h5 2 3 .Be4 Kg7 2 4.Bxf5 Qxf5 (2 4 . . . gxf5 isn't possible
since that would hang the h-pawn) 2 5 .Rc4 Bd8 2 6 . f4, with a won game. I
eventually scored the full point after making things harder for myself than
I needed to.

l 1 .a5 a6 1 2 .Ra3 Bd7 1 3 .g3

I want to answer his anticipated . . . f7-f5 with Ng5 . This move is necessary
since it stops him from playing . . . f5 -f4.

1 3 ...f5 14.Ng5 ! ?
A very committal move . White heads for the e6-square .

14 ... h6 1 5 .Ne6 Bxe6 1 6.dxe6

PA L B E N KO : M Y LIFE, GAM ES AND C O M POSITI O N S

The game now revolves around the pawn o n e 6 . I s i t strong o r weak?

16 ... Kh7 1 7.exf5 Nxf5

276

If Black had played 1 7 . . . gxfS , I would have had to decide (and it would
have been a hard choice !) between I S .BhS (heading for 7), I S .Re l (in
tending to torment Black's center pawns after I S . . . dS 1 9.BcS d4 2 0.Na4
Nd6 2 1 .c3), and I S .f4.

1 8.Nd5 Nf6 1 9.c4 c6

The straightforward 1 9 . . . ReS also fails to solve all of Black's problems:
2 0.Rb3 QcS (no better are 20 . . . Rxe6 2 1 .Rxb7 c6 2 2 .Nb6 and 20 . . . Nxe3
2 1 .fxe3 Ne4 2 2 . Qc2 NcS 2 3 .e7) 2 1 . Bf3 when 2 1 . . .RbS is met by 2 2 .Ba7
RaS 2 3 .Nxf6+ and 2 4.Bxb 7 .

20.Bb6 Qc8 2 1 .Nc7

This is a strange home for a Knight, but it helps hold onto the e6-pawn.

2 1 ...Rb8 22 .Rd3 Nd4 2 3 .Qd l Ng8

Black could have finally taken the e6-pawn, but 2 3 . . .Nxe6 2 4.Nxe6 Qxe6
2 S .Rxd6 gives White the better game. Since Black can't take the e6-pawn
off the board, he decides that he might as well block it.

24.Bg4 Ne7 2 5.f4 Rf6

Black lost the thread of the game somewhere or other (he probably should
have looked for an improvement on move twenty-three).

26.Rxd4!
The simple 2 6.Re l leaves White with a large, safe, advantage. However, I
couldn't resist the Exchange sacrifice.

26 ... exd4 2 7.Qxd4 h5

If Black tries to return the Exchange with 27 . . . Rxe6, White can refuse the
offer and gain a strong attack by 2 S .Qd3 NfS 2 9 .BxfS gxfS 3 0. QxfS + Rg6
3 1 . Qd3 .

28.Bh3 Qf8 29.Qxd6 Nf5 30.Qxf8

The endgame guarantees White a simple win. His pieces are just too ac
tive, and his passed e-pawn is just too powerful.

30 ... Bxf8 3 1 .Rd l Kg8 32 .Rd7 Ne7 H .Bd4, 1 -0.

E U RO P E AT MY F E ET

(88) Benko

Tatai (Malaga, 1969)

l .c4 Nf6 2.Nc3 g6 3 .e4 d6 4.d4 Bg7 S .f3 eS

An unusual move, in that Black almost always castles in this position. The
flaw with 5 . . . e5 (if any exists) doesn't lie in the tempting 6 . dxe5 dxe5
7 . Qxd8+ Kxd8, which doesn't really give White much of anything, but in
the fact that Black has committed himself to a plan involving a rather early
advance of his e-pawn.

6.Nge2
This and 6.d5 are White's best responses. Ineffective is 6.Be3 , since 6 . . . exd4
7 . Bxd4 Nc6 shows that White's important dark-squared Bishop is vulner
able on d4.

6 ... 0-0
Heading back into well-known main lines. Black has, on occasion, tried to
put off castling by 6 . . . c6 (moves like 6 . . . Nc6, 6 . . . c 5 , and 6 . . . Nfd7 have also
been seen) 7 . Bg5 Nbd7 8 . Qd2 a6, but his results haven't created many
supporters.

7.BgS c6
Trying to improve on my game with Fischer, which went 7 . . . exd4 8.Nxd4
Nc6 9.Nc2 Be6 1 0.Be2 , with an edge for White. See game 2 5 .

8.Qd2 QaS
Another idea is 8 . . .Nbd7 9.d5 Qb6, but the game Petrosian-Bronstein, USSR
1 974 saw White gain some advantage after l O.Be3 Qc7 1 1 .g4 h5 1 2 .g5
Ne8 1 3 .h4 as 1 4.b3 .

9.dS cxdS 1 O.NxdS

Surprise ! I went for an immediate endgame instead of the more pedes
trian 1 0.cxd 5 , which also led to a big advantage for White after 1 0 . . . Na6
1 1 .Nc 1 Bd7 1 2 .a3 Rfc8 1 3 .g4 Qc7 1 4.h4 Nc5 1 5 .h5 Na4 1 6.Nxa4 Bxa4
1 7 .Na2 Qc2 1 8 .Nc3 Qxd2 + 1 9.Kxd2 , Korchnoi-Tal, USSR 1 95 6 . Perhaps
Tatai had an improvement prepared after 1 O.cxd 5 , but my choice quickly
made this question moot.

10 ... Qxd2 + 1 1 .Kxd2 NxdS 12 .cxd5 f6

Stopping my threat of Be7, forking f8 and d6.

1 3 .Be3

277

PAL BENKO: MY LI F E, G AMES AND COMPOSITI O NS

278

White usually has better queenside prospects in the KID (thanks to the
dS -pawn and the extra space it offers in that sector) while Black seeks his
chances (in the form of a kingside attack) on the opposite wing. It stands
to reason then, that an exchange of Queens will diminish Black's kingside
attack without hurting my queenside play in any way, shape, or form.

1 3 ... Bd7 I 4.Nc3 a6

Black has stopped Na4 and NbS possibilities, but now his Knight has no
where to go. In any case, it was hard to find a good way for Black to resist
White's queenside pressure.

I 5.a4
Stopping . . . b7-bS and preparing to claim the b6-square by a4-aS .

I 5 . . .ReS I6.a5 f5 I 7.Bd3 f4 I S .Bf2 Bf6

This Bishop wasn't doing anything on g7 , so it heads for d8 where it will
eye both as and b6.

I 9.Ra3 BeS
Black's Knight will finally be able to enter the game on d7.

20.b4 BdS 2 1 .Rhal

A "mysterious" Rook move that creates future tactics (and guards as) based
on a b4-bS advance.

2 1 ...Nd7
Black had to make use of this piece, but he has blocked the e8- Bishop and
White takes immediate advantage of this fact by moving his own Knight
to a far stronger post.

22 .Na4 Kf8
Black brings his King closer to the embattled area, but it doesn't really
help. If 2 2 . . . Rc7 , then 2 3 .Nb6 would have been strong since 2 3 . . . Nxb6
24.axb6 followed by b4-bS wins a pawn (this shows the point of 2 1 . Rha l ) .

2 3 .Nb2 !
The Knight heads for the very effective c4-square, where it will lash out
at both b6 and d6. After that I can double my Rooks on the c-file without
fearing any exchanges.

23 . . .Nf6 24.Ne4 Be7 2 5.Bh4 Kf7 26.Rc3 Bb5

Black wants to get rid of the dominating Knight, but he ends up walking
into a nasty pin. However, there wasn't any defense. If 26 . . . Bd7 2 7 .Bxf6
Kxf6, White has the very strong 2 8 .Rac 1 , threatening NxeS and winning
at least a pawn.

27.Rac1 Bxe4
White also wins after 2 7 . . . Ne8 2 8 .Nb6 Bxb6 2 9.Rxc8 .

2S.Rxe4 NeS 29.g3 , 1 -0.

Black has no defense against Bfl -h3 . Another winning maneuver was 2 9 .Be2
with the idea of B d l -a4.

EURO P E AT MY FE ET

(89) Burger - Benko (U.S. Championship, 1969)

l .d4 Nf6 2 .c4 c5 3 .e3

A modest move that avoids the Benko Gambit.

3 ... g6 4.Nfl Bg7 5.Be2 cxd4 6.exd4 d5 7.Nc3 0-0 8.0-0 Nc6 9.h3
This position is, as a matter of fact, a Tarrasch Defense with colors re
versed, the question being how White can best use his extra tempo. The
text move avoids the possibility of . . . Bg4, which occurred often (with re
versed colors, of course) in the world championship match between
Petrosian and Spassky.

9 ... dxc4 1 0.Bxc4 b6! ?

A major surprise, a s neither I nor any o f the grandmasters a t this tourna
ment had ever seen it in this particular position. The usual continuation is
1 0 . . . Na5 1 1 .Bd3 Be6.

1 1 .a3
Apparently a logical refutation that paves the way for d4-d 5 and secures a
retreat for the Bishop on a2 .

1 1 ...e6!
In the previous round I had played this variation with White against Evans,
who surprised me with this move and equalized easily with the black pieces.
The aim of 1 1 . . .e6 is twofold: it neutralizes d4-d5 (the position after 1 2 .d5
exd5 1 3 .Nxd5 Nxd5 1 4.Bxd5 Bb7 is just equal), and it lays the ground
work for the blockade of the d-pawn with a subsequent attack against it by
the maneuver . . . Nc6-e7 -f5 . Black's handling of the position (with 1 0 . . . b6
and 1 1 . . . e6) is important since White can adopt it against the Tarrasch
Defense.

1 2 .Bg5 h6 1 3 .Bf4 Bb7 14.Ba2 Ne7 1 5 .Be5 Nf5 1 6.Nh2

Mter the game, my opponent told me that only here did he suddenly real
ize that the moves were identical with the Benko-Evans contest of the day
before ! In that game, too, White was compelled to make this peculiar but
forced Knight move, avoiding the loss of the d-pawn after 1 6 . . . BxB 1 7 .QxB ,
and avoiding the destruction of his pawn structure with 1 7 . gxB . It's clear
that Black's Bishop is much stronger than its counterpart on a2 .

16 ... Rc8 1 7.Rd Qd7 18.Qd3

The first deviation from my game with Evans: 1 8 .Qd2 Rfd8 1 9.Rfd 1 Ne4
2 0.Qe 1 Nxc3 2 1 .bxc 3 , with the better game for Black (White's pawns on
a3 and c3 are weak), though I managed eventually to draw. I was wonder
ing why my present opponent was following my moves so accurately (be
fore he explained that it was all accidental!). I felt it was possible that he
wanted to improve the variation with 1 8 .Qd3 , which prevents ... Ne4. Natu
rally, I had also toyed with this idea when facing Evans, but I rejected it
because I saw that 1 8 .Qd3 exposed White to even greater dangers.

279

PAL BENKO: MY LIFE, G AMES AND C O MPOSITI O NS

1 8 ... Rfd8 1 9.Rfd l Nh5 !

280

This i s the move that made m e reject l S .Qd3 i n the Evans game. Black's
threat to chop on e S and then leap into f4 is very unpleasant.

20.Qe2
There is no better move . Black threatened ... Bxe S . In case of 2 0.Qfl , there
would follow 20 . . . f6.

20 ... Bxe5?
It was only recently, when I finally had Mr. Fritz take a look at this game,
that I realized that 2 0 . . . BxeS was an error. Correct was 2 0 ... Qc6 ! , which
looks risky, but seems to work tactically: 2 1 .dS (Black is up a free pawn
after 2 1 .Qg4 Nxd4 2 2 .Bxg7 Nxg7 , while 2 1 .Qfl BxeS 2 2 .dxeS Nf4 trans
poses into the game.) 2 1 . . .exdS 2 2 .Bxg7 Kxg7 (Also strong is 22 . . . d4 2 3 .BdS
RxdS 2 4.NxdS QxdS 2 S .RxcS+ Kxg7 2 6.QB Qd7 2 7 .RdS Qc7 2 S .RdS Nf4
2 9.Rd7 BxB 3 0.Rxc7 Bxd l ) 2 3 .BxdS RxdS 24.RxdS Nf6 (24 . . . QxdS 2 S .NxdS
Rxc 1 + 2 6.Nfl BxdS 2 7 .Qd2 Rxfl + 2 S .Kxfl Nf6) 2 S .Rcd l RdS 2 6.QeS ReS,
and Black is winning.

2 1 .dxe5?
During the game w e both thought that 2 1 .QxeS Nxd4 was just a clean
pawn for Black. However, 2 2 .Ng4! changes this assessment and brands
20 . . . BxeS as a mistake. After 2 1 .dxe S ? , the game once again follows its
logical course.

2 1 ...Qc6
This diagonal will kill White, since 2 2 .NdS fails to 2 2 . . . Qxc 1 . The result
of 2 2 .B would be the loss of a pawn after 22 . . . QcS+ 2 3 .Qf2 Qxe S .

22 .Rxd8+ Rxd8 2 3 .Qf1 Nf4 24.NB

2 4.NdS fails to 24 . . . Qxc 1 2 S .Nf6+ Kg7 2 6 . Qxc 1 Ne2 +, while 24.B Rd2 is
also lights out for White.

24 ... Nd4!
Black had many good moves (24 . . . Nxh 3 + and 24 . . . Qc7), but this ends the
game quickly.

2 5 .Nxd4 Rxd4 26.Rd l Rxdl 27.Nxdl Ba6!, 0- 1 .

EURO P E AT MY F E ET

(90) R Byrne - Benko (U.S. Championship, 1969)

l .e4 c5 2 .NB d6 3 .d4 cxd4 4.Nxd4 Nf6 S.Nc3 Nc6

This old, classical position in the Sicilian was neglected for quite a long
time before experiencing a revival in the late '60s. After that, it never looked
back and is extremely popular to this day. One reason for the neglect was
Black's fear of 6.Bg5 and 6.Bc4, both of which were considered to be very
dangerous. He went through all kinds of acrobatics to avoid them, until
various schemes against 6.Bg5 suddenly leapt into prominence, and 6 .Bc4
lost a bit of its luster to 6 . . . Qb6 (The Benko Variation) .

6.Bg5 Bd7
Compared to 6 . . . e6, which is the main line,
the text move is more flexible since Black need
not fear the doubling of his pawns by White's
eventual Bxf6 . He gets ample compensation
for the structural inferiority in the form of
increased control of the dark squares (thanks
to the loss of White's dark-squared Bishop).

7.Qd2 Nxd4
A classic shot of Robert Byrne.

More usual is 7 . . . Rc8, as seen in Benko-R.

(Photo courtesy useF.)
Byrne, U.S. Open 1 968. The text is simply
an exploitation of White's loss of time with
his Queen, while it provides an active role for Black's Queen.

8.Qxd4 Qa5 9.Bd2

An important alternative is 9.f4.

9 ... Qc7
This move never caught on and 9 . . . e5 became Black's main choice. How
ever, White scored quite well against the pawn move: 1 0.Qd3 Rc8 I I .Be2
a6 1 2 .0-0 Be6 1 3 .Nd5 Qd8 1 4.Bg5 Bxd 5 1 5 . Bxf6 ! Qxf6 1 6 . Qxd5 Rc7
1 7 .Bc4, with advantage, Tal-Radulov, Skopje 01 1 9 7 2 . I still don't see a
problem with 9 . . . Qc7 . There's no doubt that the Queen has to retreat at
some point, so why not back up to an active square (an established launch
ing pad in the Sicilian), while leaving White's Bishop on d2 - a far from
ideal post.

lO.Bc4
The game moves towards familiar variations where the drawbacks men
tioned earlier (Black's Queen is quite happy on c7 while White's Bishop is
ineffective on d2) will sooner or later come to light. White could have
chosen a more interesting line with 1 O.Nd 5 , as occurred in a later round
of the USSR Championship in the game Gufeld-Savon: 1 0 . . . Nxd5 I I .exd5
Qxc2 (Taking the pawn is a must! White will get a lead in development,
and Black must make sure he grabs some loot for his troubles.) 1 2 .Rc 1

28 1

P A L BENKO : MY L I F E , G AMES AND COMPOSITI O NS

QfS 1 3 .Bd3 Q e S + 1 4. QxeS dxe S 1 S .0-0 e6 1 6 . dxe6 Bxe6 1 7 . Rc7 Rd8

1 8 .BbS+ Bd7 1 9. Rxd7 Rxd7 2 0.Bc3 a6 2 1 .Bxd7 + Kxd7 2 2 .BxeS f6, Black
equalized and even went on to win.
282

1 0 ... e6 1 1 .Bb3 Be7 1 2 .f4? !

The typical active advance is inaccurate here because it takes away White's
option of castling short. Better was 1 2 .0-0, though White can't claim any
great achievement in the opening.

1 2 ... 0-0 1 3 .Rf1

Preparing f4-fS . Mistaken is 1 3 .0-0? d S ! , when the threat of 14 . . . BcS is
annoymg.

1 3 ... b5!

Placing White in an uncomfortable situation b y saying, "You will be fac

ing some serious heat if you castle queenside, but your life won't be a happy
one if you leave your King in the center. "

14.0-0-0?
White should have stayed with his original plan of playing f4-fS . Then
Black would have continued his attack with 14 . . . aS or 1 4 . . . exfS . In either
case he could face the future with confidence. In all these lines, White's
Bishop on d2 plays no active role, as it would, for instance, on gS , when
the threat of e4-eS would always be hanging over Black's head.

14 ... a5
Black's attack is clearly much faster than White's.

1 5.a4 b4! 16.Ne2

If 1 6.Nb S , Black takes over the game very quickly: 1 6 . . . BxbS 1 7 .axbS a4
1 8 .Bc4 ( 1 8 .b6? axb3 ) 1 8 . . . dS 1 9. exdS Rfc8 20.Bd3 NxdS 2 l .fS Bf6 2 2 .Qe4
a3 , and it's all over.

16 ... Bc6 1 7.Ng3 Qd7

Taking immediate aim at White's weak a-pawn, but 1 7 . . . Nd7 followed by
. . . NcS was also strong. It is evident that White's opening has failed - his
Bishop on d2 prevents any counterplay on the d-file and serves no active
function.

E U RO P E AT M Y F E ET

1 8.e5 dxe5 1 9.Qxe5 Bd6 20.Qd4

20.Bxb4 fails to 2 0 . . . Bxe5 2 1 .Rxd7 axb4.

20 ... Bxa4 2 1 .Be3

Of course, 2 1 .Bxa4 Qxa4 2 2 . Qxd6 ? ? runs afoul of 2 2 . . . Qa l mate.

2 1 ...Bxb3 22.Qxd6 Qxd6

I could have retained the Queens by 2 2 . . . Qc8, with an extra pawn and the
better position, when 2 3 .Bc5 is nicely met by 2 3 . . . Nd7 ! . I must admit that
I overlooked the latter move.

2 3 .Rxd6 Rfe8 24.Rd2 Re6

A good move, but both 24 . . . Nd5 and 24 . . . a4 were also worth considering.

2 5.Kb l Bd5 26.b3 h5 !

This move foils all of White's defensive plans and creates new weaknesses
in his position. Black threatens to chase the Knight and, in case of 2 7 .h4,
he could successfully play 2 7 . . . Rc3 (2 8.Re l Ng4). White tries instead to
exploit the drawing chances offered by Bishops of opposite colors, but, as
will be seen, it is all in vain.

27.Bd4 h4 28.Bxf6 hxg3

And not 2 8 . . . gxf6 2 9.Nh5 .

29.Bd4 gxh2 30.Rhl

I t would seem that White will win Black's h-pawn and cut Black's winning
chances to nil. In fact, Black can exploit the absence of the enemy Rook
for a decisive stroke.

30 ... Be4 3 1 .Rxh2

If the white Bishop were posted on b2 (as after 2 9 .Bb2), then 3 1 . . . a4 would
decide: 3 2 .bxa4 b 3 .

3 1 .. .Rd8 ! , 0- i .
White can't avoid the loss of the c-pawn.

283

PAL B ENKO: MY LI FE, G AM ES AND C O M P O S I T I O NS

(9 1) Benko - Lengyel (Venice, 1969)

1 .c4 e6 2 .NB dS 3 .d4 Nf6 4.Nc3 Be7 S .Bf4 0-0 6.e3 Nbd7
284

Nowadays, it's well-known that Black's best move is 6 . . . c 5 . However, at

the time 6 . . . Nbd7 was employed quite a lot.

7.cxdS
In this position, theory recommends 7 . Qc2 or 7.c5, though 7 . cxd5 be
came rather popular in the '90s and is still considered an excellent choice.
In the game Benko-Medina, Palma de Mallorca 1 968, I played 7 .a3 c5
8 . cxd5 exd5 9.Be2 a6 1 O.dxc5 Nxc5 1 1 .0-0 Be6 1 2 .B e 5 , and White had a
slight advantage due to his control of the d4-square. Taking it for granted
that my opponent had carefully studied my earlier games (none of which
featured 7 . cxd5 since, at the time, this capture was a novelty), I suddenly
decided to cross his plans and deviate from the expected continuation.

7 ... NxdS 8.NxdS exdS 9.Bd3

Compared with the normal variations, a new element here is the disap
pearance of White's QN and Black's KN. Naturally, had Black played
7 . . . exd 5 , a normal Exchange Variation would have arisen, which is consid
ered to be a bit better for White.

9 ... Nf6
In case of 9 . . . Bb4+, White planned the simple 1 0.Nd2 .

1 O.Qc2 c6 1 1 .0-0 Bg4

An inaccuracy which will be immediately exploited by White's taking the
initiative on his next move. Of course, 1 1 . . .Be6 is questionable owing to
1 2 .Ng5 , but Black might have considered 1 1 . . .h6 or 1 1 . . . a 5 , parrying any
immediate attempt at attack on the queenside.

1 2 .NeS Be6 1 3 .b4!

This "petite combinaison" ensures White's initiative with the so-called

minority attack (White attacks Black's majority of queenside pawns [a 7/
b7/c6] with his minority [a2/b4] via b2 -b4-b5xc6 in order to create a per
manent weakness in Black's structure). After 1 3 . . .Bxb4 1 4.Rfb l , White re
gains the pawn in all variations (i.e., 14 . . . Qa5 1 5 .a3).

E U RO P E AT MY F E ET

H ... RcS I4.Rab l h6 1 5 .Rfc l Nd7

Black tries to ease the pain by exchanging as many pieces here as possible.
He could have tried I S ... NhS , although his position would be no brighter
even after he plays . . . Nxf4.
I 6.Nf3 b 5 ? !

Black's plan envisions a Knight maneuver with . . . Nb6-c4.

I 7.a4 a6?

Good or bad, 1 7 . . . Nb6 was a must. After 1 8 .axbS cxbS 1 9. Qa2 ( 1 9.Bc7)
19 ... Na4, White's position would be superior, but not as markedly as after
the text.
I S.a5

Depriving Black's Knight of the b6-square and thus stopping . . . Nd7 -b6c4. Now the c6-pawn is a permanent target.
IS ...f5 I 9.h3 Rf6 20.Rb3 KhS 2 1 .Qb I Qf8 22 .Be5 Rg6

2 2 . . . NxeS isn't a good idea since, after 2 3 .dxe S , White's Knight would oc
cupy a commanding position on d4.
n.Rbc3 Nxe5

White would willingly give up his b-pawn for Black's c-pawn since the
latter is more valuable - the fall of the c-pawn weakens the a6- and d S
pawns and allows the white Rooks t o penetrate into Black's position.
24.Nxe5 Rf6 2 5 .Rxc6 RaS 26.Rxe6

This combination results in an extra pawn for White and an irresistible

kingside attack, despite the presence of opposite-colored Bishops (remember
that such Bishops are often drawish in an endgame, but they add a lot of
zip to most attacks since one Bishop can't defend what the other attacks).
26 ... Rxe6 27.Bxf5 Rxe5 2S.dxe5 Bxb4 29.Qd3 Bxa5 30.Qxd5 Bb4 3 1 .Bd3
Be7 3 2 .Qe4 QgS, 1 -0.

Black resigned as soon as he made his move. Some might say it is too
early to accept defeat, but in fact Black's surrender is fully justified. White
has several winning continuations, for instance, 3 3 .Rc7 Ba3 34.f2 -f4-fS
f6 o r 3 4.e6, forcing a quick decision since Black's queenside pawns are
harmless.

(92) Benko

Byrne (Vrsac, 1 969)

l .c4 Nf6 2 .Nf3 g6 3 .b3 Bg7 4.Bb2 0-0 5.g3 c5 6.Bg2 Nc6 7.0-0 RbS? !

Preparing to play . . . a 6 and . . . b S . However, this push isn't necessarily a good

idea if the center is still open.
S.Nc3 a6 9.d4

White directs play towards the middle and away from the wings. This strat
egy brands Black's queenside demonstration as premature.

285

PAL B E N KO : MY L I F E , GAM ES A N D C O M POSITI O N S

9 ... d6

Black should trade by 9 . . . cxd4. The text lets White place immediate pres
sure on Black's position.
286

1 0.dxc5 dxc5 I l.Na4 Nd7

Better was 1 1 . . . b6 since Black will never get a chance to free himself with
. . . b7-b S . Black's reluctance to admit this leaves him with an unpleasant
position.
1 2 .Bxg7 Kxg7 1 3 .Qd2 e6?

This weakens the d6-square and White, of course, instantly goes after it.
Better was 1 3 . . . Qc7, while 1 3 . . .b S ? 1 4.cxbS axbS l S .NxcS NxcS 1 6.Qc3 +
would have led to the loss of a pawn for Black.
1 4.Rfd l Qe7 1 5 .Ng5

Intending to bring yet another piece to bear on d6 via Ne4. This also
gives White the option of Bxc6, leaving Black's pawn structure in ruins.
1 5 ... Nd4

Black avoids Bxc6 and creates the little threat of 1 7 . . . Nxe2+ 1 8 .Qxe2 QxgS .
1 6.Ne4

Getting my Knight off the vulnerable gS-square and creating a counter

threat of 1 7 .NexcS Nxe2 + 1 8 . Qxe2 NxcS 1 9.QeS+, winning a pawn.
16 ... b6 1 7.e3 Nf5 1 8 .Racl

This prevents a future . . . b6-bS (due to cxbS when the cS -pawn falls) and
allows me to begin the plan that follows without having to worry about
black counterplay.
1 8 ... f6 19.Nd6!

White enters a very favorable endgame. Black is already strategically lost.

19 ... Nxd6 20.Qxd6 Qxd6 2 1 .Rxd6 Kf7 2 2 .Rcdl Ke7 2 3 .Rc6!

Again stopping . . . b6-bS and also threatening Rc7 .

2 3 ... Rd8 24.Rc7 Rf8 2 5 .Bc6 Rd8

Black is in a terrible bind. White's game now plays itself.

2 6.f4 f5 2 7 .m h6 2 8.Rd2 Ke8

&.nald Byrne: Donald was very tal

ented, very friendly and nice. His main

problem was time pressure, at times
thinking for an hour over one move in
the opening! It's a shame he died so
young. I played him in Washington D.C.
after he snuck out of the hospital to play
in the tournament. I felt bad for him, so
we agreed to a draw.

USCF Vice-President Jerry Spann (base

of stai rs) shakes hands with Elliot Hearst
as Robert Byrne, Donald Byrne and
Mednis board a plane en route to the
Chess Olympiad. Fischer was already
aboard, wh ile Evans and Benko left later.
(Photo courtesy USCF.)

White could have tried to get a winning pawn endgame by 29.RxcS Rbxc8
3 0.Bxd7+ Rxd7 3 1 .Rxd7 Kxd7 3 2 .Nxb6+ . However, Black would be able to
thwart this by 3 0 . . . Ke7 ! 3 1 .Ke2 (a bit more accurate is 3 1 .Ke l [prevents a
check on d2 in some lines] but even here Black puts up strong resistance
with 3 1 . ..Rc7 3 2 .Nxb6 RbS 3 3 .NcS+ RcxcS 34.BxcS RxcS, when this pawn
up Rook endgame isn't nearly as favorable as my position was earlier)
3 1 . ..Rc7 3 2 .Nxb6 Rb7 , when unwelcome complications arise. In the posi
tion after 2S . . . Ke8, Black can't do anything (he's completely helpless) so I
will take my time, torture him, and only cash in when it ends the game.
29.e4

Playing to squeeze the life out of my opponent quietly and safely. How
ever, a sharper way to end the game was 2 9.b4! cxb4 3 0.c5, when Black
will suffer heavy material losses .
29 ... Ke7

On 2 9 . . . fxe4, I would play 3 0 .Ke3 when Black would have to continue to

tread water.
30.e5 g5 3 1 .h3 gxf4 3 2 .gxf4 h5 3 3 .Kg3

Anything wins. For instance, I could play 3 3 .Rd6 and wait for Black to
run out of pawn moves (multi-piece 2ugzwangs are always fun !). The plan
I chose is more than sufficient, though. I simply intend to march my King
up the board and eat his h-pawn.
3 3 ,. .RgS+ 34.Kh4 KdS 3 5 .Ra7 Rg7 3 6.Kxh5 KeS 3 7.Kh6 Re7 3 S.Kg6

It's over. Now I'm going to push my h-pawn and make a new Queen.
3S . Kf8 3 9.Bxd7 Rg7+ 40.Kf6 Rf7+, 1 -0.
..

After making the time control, Black, facing 4 1 .Kxe6, decided to resign.

PAL BEN KO : MY LI FE, GAM ES A N D C O M POSITI O N S

(93) Benko - Geller (Wijk aan Zee, 1 969)

l .Nf3 c5 2 . c4 g6 3 .g3 Bg7 4.Bg2 Nc6 5 .Nc3 e6
288

Black's 5 ... e6, a favorite of Fischer's and Tal's, is an attempt to give the game
a dynamic stamp. Unfortunately for Geller, I don't allow this to happen.
6.0-0

Some years later, the super-sharp 6.d4! ? achieved a certain measure of popu
larity. Since it virtually tries to refute Black's fifth move, it's still topical to
day. To my mind, the main line is 6 . . .Nxd4 7 .Nxd4 cxd4, and now 8 .Nb5
was dealt a blow in Markowski-Macieja, Warsaw 1 998: 8 . . . Qb6 9.Qa4 a6
1 O.e3 d3 1 1 .0-0 (I 1 . Qa3 BfS is known to favor Black.) I l . . .Ne7 1 2 .Rd l axb 5 !
1 3 .Qxa8 bxc4 1 4.Rb l 0-0 1 5 .Bd2 d5, when Black had tremendous compen
sation for the Exchange. I think 8 .Ne4! ? gives Black more problems.
6 ... Nge7 7.e3

I've always felt that the positions after 7 . d 3 0-0 8.Bd2 d5 are quite com
fortable for Black and would be well suited to Geller's style. WIth the text,
I steer the game into a position that is more to my taste than his.
7 ... 0-0

I was a bit surprised that Geller didn't try the more combative 7 . . . Nf5 .
8.d4 cxd4 9.Nxd4 d5 1 0.cxd5 exd5

Geller was a magnificent attacking player, but this kind of position never
suited him. On the other hand, I tend to be at my best in such positional!
technical situations.

Geller: Just before I lost a game to Geller in some tournament, we adjourned and,
though I was worse (he had been torturing me the whole game) I sealed the wrong move and

w.fim

resigned. The Russians were surprised that I resigned, but

they didn't know that I had blundered while sealing.
Next time we played, he offered a draw. But I had sworn
that I would extract revenge on him, and this time I tor
tured him all the way and won. The need for revenge al
ways makes me play better!
I couldn't talk to Geller because he didn't speak En
glish or German. He was a great attacking player, and had
mastered the King's Indian. Later in life, though, he be
came very positional. I asked him (after he fmally picked
up a few words of English), "Why are you quitting the
KID?" He said, "Because it's no good!" Oddly, Fischer
also told me it was not a very good opening, but he kept
playing it in game after game.

Geller with his son.

(Photo courtesy USCF.)

E U RO P E AT MY F E ET

1 1 .Qb3

Naturally, Black would have been satisfied with the position resulting from
I l .Nxc6 bxc6, even though his pawns would be weak, since his dark-squared
Bishop exerts strong pressure on the long diagonal and his other Bishop
could be actively mobilized on a6. Though this position must have occurred in Soviet tournament practice, Botvinnik called the text move
( l l .Qb3) a new idea ! Up to this point the moves followed each other in
quick succession.
1 1 ...Nxd4

Of course, 1 1 . . . Bxd4 would not win a pawn either because White can eas
ily restore the material after 1 2 .exd4 Nxd4 1 3 . Qd l . White's two Bishops
would then give him a significant edge.
1 2 .exd4 Nf5 1 3 .Qxd5 Qxd5 1 4.Nxd5

Not as obvious as one might think, since White would also maintain pres
sure after 1 4.Bxd5 Nxd4 1 5 .Bf4.
14 ... Nxd4

The position is completely symmetrical and the reader might remark here,
"It is high time to agree to a draw. " However, this is far from correct since,
in chess, complete symmetry doesn't really exist - it is just a temporary
phenomenon because one of the players has to make a move. As a matter
of fact, my opponent offered a draw after the exchange of Queens but I
wasted no time in refusing his offer. All the more so because, at the Lugano
Olympiad, Geller rej ected a similar offer of mine in an apparently drawn
position. You see, revenge plays a greater role in chess than one might
suppose !
1 5 .Bg5 ! Bh3 ! ?

Black wants to get rid o f my strong light-squared Bishop by means o f this

clever move. Symmetry can't be maintained by 1 5 . . . Bg4, since after 1 6.Nf6+
Bxf6 1 7 .Bxf6 Nf3+, White is not obliged to capture the Knight. It would
be equally unpromising to play either 1 5 .. .f6 or 1 5 . . . Ne2 +, with the idea
of winning a pawn.
16.Rad l Bxg2 1 7.Kxg2 Ne6

289

PAL BEN KO : MY LI FE, GAM ES AN D C O M POSITI O N S

After the game, Geller remarked that this move was a n obvious blunder.
He felt that 1 7 . . . Nc6 was better, when White would possibly continue with
1 8 .b4 or 1 8 .Rd2 .
290

1 8.Be7 Rfe8 1 9.Nf6+ Kh8

Naturally, after 1 9 . . . Bxf6 2 0 .Bxf6, White's superiority would be accentu

ated by his control of the d-file and the power of his dark-squared Bishop
(which is making Black's King quite uncomfortable) . It should also be
pointed out that the Bishop, with its long range abilities, will be vastly
superior to Black's Knight because there are pawns on both sides of the
board (a Knight is often better than a Bishop if there are pawns on only
one side of the board).
20.Rd7 b6 2 1 .Rfd l Re2 2 2 .R l d2 Rxd2 2 3 .Rxd2 NfS

Passive, but Black didn't want White's Rook moving to the seventh rank.
Though 2 3 . . . Rc8 24.Rd7 Rc7 may look like a better defense, it allows White
to demonstrate one major flaw of the black position: the second player is,
in effect, a King down ! Thus, White's Monarch would decisively penetrate
into the enemy position after 2 5 .Rxc7 Nxc7 2 6.Kf3 Ne6 2 7 .Ke4.
24.b3 Re8 2 5 .Rd6 !

This move completes the immobilization of Black's camp and prevents

any possible counteraction with . . . Rc6.
2 5 ... h5

And not 2 5 . . . Rc7 ? ? 2 6 .Bxf8 Bxf8 2 7 .Rd8, when White wins a piece be
cause 2 7 . . .Kg7 hangs the Rook to 2 8 .Ne8+. Trying to exchange Rooks
seems logical (after all, White's Rook is much more active than Black's),
but the idea fails because White's King becomes far too strong: 2 5 . . . Ne6
2 6 . Rd7 Rc7 2 7 .Rxc7 Nxc7 2 8 .Kf3 Na6 29.Ke4 Nb4 3 0.Bxb4 Bxf6 3 1 .Kd 5 .
26.a4 Nh7 2 7 .Nd5

We can see that White took all of Black's escape attempts into consider
ation when he played 2 5 .Rd6. Now nothing can be done about White's
upcoming Rd7 .
2 7 . . .Be5 2 8.Rd7 Re2

Attempting to generate some active counterplay. The passive 2 8 . . . Bb8Ieads

E U RO P E AT MY F E ET

to a hopeless position after 2 9.Ba3 Kg7 3 0.Bb2 + KfS 3 1 .h4, when Black is
completely helpless.
29.Ne3

Naturally, 2 9 .Rxa7 would now be a mistake due to 2 9 . . . Bd4.

29 ...Ra2 3 0.Rxa7 Bd4 3 1 .Kf3 Rb2 3 2 .b4 Kg7 3 3 .Ke4?

An unfortunate move that lets my opponent back into the game. I would
have retained a winning endgame with 3 3 .Rd7 or 3 3 .h4.
3 3 ... Bc3 ?

White's last move was not the luckiest choice, but Black failed to exploit
it. He should have tried 3 3 . . . Bf6, with real chances to save the game.
34.NdS Bel 3 S .Nxb6 Bxfl

The final mistake. Geller had to try 3 5 . . . Bxb4, when he would still have
chances to resist. Now it's all over.
3 6.BcS Nf6+ 3 7.Kd3 BxcS 3 8.bxcS Rxh2 3 9.c6

Though material is even, White's passed pawns are far stronger than Black's.
3 9 ... Ne8

White could now win a piece by 40.c7, but why hurry?

40.Re7 Nd6

If 40 . . . KfS , then 4 1 .RxeS+ Kxf8 42 .c7 ends matters quickly.

41 .Rd7 Rb2 42 .Rxd6 Rxb6 43.Kc4, 1 -0.

(94) Benko - Sigurjonsson (Caracas, 1 970)

l.NB dS 2 .d4 Nf6 3 . c4 c6 4.e3 e6 S.Bd3

A tricky move order.

S ... Nbd7

Reaching a position in the Semi-Slav. However, a better choice for Black

might have been 5 . . . dxc4 6.Bxc4 c5, when we've transposed into the Queen's
Gambit Acceptedwhere both sides have lost a tempo. After 5 . . . Nbd7 6.Nc3 ,
the regular Meran Variation would arise. Instead, I use the opportunity to
deviate and take Black into positions that he might not be familiar with.
6.Nbd2 !

Creating greater problems for Black than the usual 6.Nc3 , because if
6 . . . dxc4, White is able to capture with the Knight, giving the game an
entirely new character.
6 ... cS

According to general QP theory, when White plays his QN to d2 , Black

does well to respond with . . . c 5 . However, here Black had already moved
this pawn to c6. Thus, he now loses a tempo. Black's best answer to this
system is probably 6 . . . Bd6, playing for the thematic . . . e6-e5 advance.
7.b3 !

29 1

PAL B E N KO : MY LIFE, GAM ES AN D C O M POSITI O N S

Avoiding a n isolated d-pawn, which can be inflicted after, say, 7 .0-0 cxd4
8 . exd4 (not 8 .Nxd4 e5) 8 . . . dxc4. The text permits a natural development
of the dark-squared Bishop.
292

7 ... Be7

In similar positions (where Black plays . . . c7 -c5 , meaning that he is up a

tempo), Black used to employ a tactical break in the center by 7 . . . cxd4
8.exd4 dxc4 9.bxc4 e5 1 O.dxe5 Nxe5 1 1 .Nxe5 Qd4. However, entering such
a sharp line with a move less is a sure recipe for disaster, so the fact that
Black gets punished shouldn't be surprising: 1 2 .0-0 (Also strong is 1 2 .Qa4+
Nd7 1 3 .Ndf3 Qxa 1 1 4. 0-0, transposing into lines reached via 1 2 .0-0. An
other promising possibility is 1 2 . Q e2 ! ? ) 1 2 . . . Qxa 1 1 3 . Qa4+ Nd7 (or
1 3 . . . Bd7 1 4.Nxd7 Nxd7 1 5 .Re 1 + Be7 1 6.Nf3 0-0-0 1 7 . Qxa7 , with a crush
ing attack) 1 4.Ndf3 f6 (Both 1 4 . . . b5 1 5 .Qxb 5 Rb8 1 6.Qd5, and 1 4 . . . Bd6
1 5 .Nxd7 Bxd7 1 6.Re 1 +, are extremely unpleasant for the second player.)
1 5 .Nxd7 Bxd7 1 6.Qb3 , when Black's Queen is trapped.
8.0-0 0-0 9.Bb2 b6

Black doesn't seem eager to swap pawns in the center by 9 ... cxd4 1 0.exd4
dxc4 1 1 .bxc4, since the resulting hanging pawns formation gives White
more space and excellent tactical chances.
1 0.Qe2 Bb7 1 l .Rfd l

The other possibility here was to build up an attacking position with 1 1 .Ne5
followed by f2 -f4. However, I chose the text because I saw that Black's
Queen has no good square. This is in sharp comparison to White's Queen,
which is very comfortable on e2 .
1 1 . .. Rc8 1 2 .Rac 1 Rc7

This is the best way to solve Black's problem with development, since now
his Queen can be posted on a8 where it exerts pressure on the long diago
nal. The position is full of tension. Black doesn't want to initiate an ex
change of pawns because that would merely activate White's centrally placed
Rooks. This means that White is the one with his finger on the trigger he's the one that will open central lines when it suits him best.
1 3 .dxc5

Deciding that now is the time ! White starts a battle before Black is fully
mobilized.
1 3 ... Nxc5 1 4.Bc2

Also good was 1 4.Bb 1 , but the text allows the possibility of b4 without
permitting the answer . . . Na4.
14 ... Qa8 1 5 .Ng5 !

E U RO P E AT MY F E ET

293

White opts for a direct kingside attack, otherwise his advantage in devel
opment will evaporate. The threat now is 1 6.Bxf6, and any pawn move
would be weakening: 1 5 . . . h6 1 6.Bxf6 gxf6 1 7 .Nh3 , and Black's destroyed
kingside will cause him no end of concern. If 1 5 . . . g6 1 6.cxd5 Bxd5 1 7 .e4,
White gets strong attacking chances. In this line, worse would be 16 . . . Nxd5 ,
because o f 1 7 .Nxh7 ! (stronger than 1 7 . Qh5 Bxg5) .
I S ...Ned7

Because of the many threats, Black decides to play it safe and overlooks
the main one. Black must have considered 1 5 . . . Nce4, when 1 6.Ndxe4 blocks
Black's QB and leaves White with a permanent advantage due to his
queenside pawn majority and kingside attacking chances. If this is all White
had, then Black might have sucked it up and given 1 5 . . . Nce4 a try. Unfor
tunately, White can force the win of a pawn by 1 6.Ngxe4 dxe4 1 7 .Bxf6
Bxf6 1 8 .Qg4, since 1 8 . . . Bb2 1 9.Rb l f5 fails to 2 0 . Qg3 .
1 6.exdS Rfe8

Black clearly placed a lot of stock in this move, which defends the c7Rook and threatens my light-squared Bishop. However, he probably should
have tried 1 6 . . . Bxd 5 . Yes, I think this ultimately loses, but the complica
tions that result would have given White more opportunities to make an
error: 1 7 . Bxh7+ (The tempting 1 7 .Nxh7 is met by 1 7 . . . Rfc8 ! , when Black
is still very much in the game.) 1 7 . . .Nxh7 1 8 .Rxc7 Nxg5 1 9.e4 Qd8 20.Rxa7
Bc6 2 1 .Nc4, and White's position is overwhelming. A sample line: 2 1 . . .Qe8
2 2 .Rc7 Nc5 2 3 .Nxb6 Ncxe4 24.Rc8 Qxc8 2 5 .Nxc8 Rxc8 2 6 . Qa6 Rc7
2 7 . Qb6 Rc8 2 8 .Rc 1 , and wins.]
1 7.dxe6! Rxc2 1 8.exd7 Nxd7

Relatively better was 1 8 . . . Rxc 1 1 9.dxc8=Q+ Rxc8, though this would leave
Black a solid pawn down with no compensation.
1 9.Rxe2 Rxc2

Perhaps Black was relying on this position, in which two white pieces are
en prise. But White's next move destroys his illusions.
20.Qd3 BxgS

The only reasonable defense to mate. However, since he's the Exchange

PAL BE N KO : MY LI FE, GAM ES AN D COM POSITI O N S

down, he has nothing left to play for.

2 1 .Qxe2 Bxg2 22.Qf5 , 1 -0.
294

Black didn't bother to try 2 2 , ..Qd5, since 2 3 .Qg4 Bhl 24.e4 wouldn't leave
any doubt about the result.

(95) Benko - Donner (Wijk aan Zee, 1 970)

l .Nf3 d5 2 .d4 Nf6 3 .e4 e6 4.Nc3 e5 5.exd5 Nxd5 6.g3 Ne6

For 6 , . .cxd4, see game 43 : Benko-Korchnoi, Cura<;ao 1 962 .

7.Bg2 exd4

Black's most common move is 7 , . .Be7. However, on occasion Black will

try this line, which aims - after a quick trade of pieces- to hold the slightly
inferior endgame that results.
S.Nxd4 Nxc3 9.bxc3 Nxd4 1 0.Qxd4 Qxd4 1 1 .exd4 Bb4+

Also played is 1 1 . . .Bd6. One sample of this is Kramnik-Lautier, Horgen

1 99 5 : 1 2 .0-0 (also promising is 1 2 .Bf4 Bxf4 1 3 . gxf4 followed by Kd2)
1 2 , . .Rb8 13 .e4 0-0 1 4.e5 Be7 1 5 .Be3 Bd7 1 6.Rfc l Rfc8 1 7 .Rxc8 Bxc8 1 8.Rc l
Kf8 1 9.Bh3 ! Ke8 2 0.d5 B d 7 2 1 .d6, and White won o n move 64.
1 2 .Bd2 Bxd2 + 1 3 .Kxd2

If Black thought this would be an easy draw, he was mistaken! My Bishop

sweeps across the board, licking its lips as it eyes b 7 , my Rooks will claim
the c-file, and my central majority of pawns will prove more active than
Black's queenside maj ority.
1 3 ... Ke7

Perhaps 1 3 , . .Rb8 1 4.Rhc l Kd8 is a better defensive approach.

1 4.Rhc 1 RdS 1 5 .Re7+ Rd7 1 6.Rac 1 KdS 1 7.Rxd7+ Kxd7 I S.f4

Kramnik also achieved this position against Lautier (Belgrade 1 995), but
instead of my 1 8 .f4, he tried 1 8 .g4. The continuation confirmed that Black's
life isn't easy: 1 8 , . .h6 1 9.f4 Rb8 2 0.g5 b6 2 1 .gxh6 gxh6 2 2 .Rc3 Bb7 2 3 .Bxb7
Rxb7 24.Rh3 , and White won in 42 .
I S ... RbS 1 9.e4 b6 20.Ke3 Bb7 2 1 .Bh3 ! ReS

E U RO P E AT MY FEET

Yet another piece trade (in fact, the last one of the game !), but it doesn't
ease Black's problems. It's important to note that the logical looking 2 1 . . . Kd6
is powerfully answered by 2 2 . d 5 , when 2 2 . . . exd5 2 3 . e 5 + Ke7 2 4.Rc7+ is
crushing for White.

295

22 .Rxe8 Bxe8

Black would be in serious trouble even

after 2 2 . . . Kxc S : 2 3 . f5 exf5 (perhaps
2 3 . . . Kd7 2 4 . fxe6+ fxe6 2 5 . d 5 Kd6
2 6.Bxe6 Ke5 puts up more resistance)
24.Bxf5 + KdS 2 5 . Bxh7 g6 2 6.h4 Ke7
2 7 . h 5 gxh 5 2 S . d 5 ! Kd6 2 9 . Kd4 f6
3 0. Bg6, and White wins.
2 3 .d5 Ke7 24.e5 ! b5

Black doesn't dare go into a pawn

endgame by 24 . . . exd5 2 5 . BxcS KxcS
Donner vs. Portisch, Santa Monica
(Photo courtesy USCF.)
2 6.Kd4, because of the dominating po
sition of the white King. Of course, after the text move White gets a strong protected passed pawn.
2 5 .Kd4 Kb6 2 6.d6 b4 27.Bg4 Bd7 2 8.Bdl Be6 2 9.Bb3 g6

Preferable was 2 9 . . . a5 3 0.f5 a4 3 1 . Bc4 Bd7, when Black's active queenside

majority gives him a bit of counterplay. However, White still wins in the
long run. For example: 32.g4 h6 3 3 .h4 g5 3 4.fxg6 fxg6 3 5 .g5 hxg5 3 6.hxg5
Kc6 3 7 .Bd3 BeS 3 S .Kc4, and it's all over.
3 0.g4 Bf3 3 1 .f5 ! gxf5

White wins after 3 1 . . .Bxg4 3 2 .Bxe6 ! fxe6 3 3 .f6 .

3 2 .gxf5 Bg4 3 3 .f6

White has created a new target on f7 . In fact, in some lines the e6-pawn
also becomes vulnerable to a tactical sacrifice of the Bishop.
3 3" .Bf5 34.Ba4

Another way to win was 3 4.Kc4 as 3 5 .d7 Kc7 3 6.Kb5 Kxd7 3 7 .Kxa5 fol
lowed by 3 S .Kxb4, when the passed a-pawn is too strong.
34 ... Bg6 3 5 .Be8 a6 3 6.a3 !

1 9 66.

PAL B E N KO : MY LI FE, GAM ES AN D COM POSITI O N S

296

This gives the white King access to Black's queenside. The very tempting
3 6,Kc4 a5 3 7 ,d7 Kc7 3 S ,Kb5 isn't as strong, because the Bishop's entombed
position cuts down on White's options considerably: 3 S . . . Bb 1 3 9.Kxa5 Bxa2
40.Kxb4 Bd5 4 1 .Kc5 BD 42 .h4 h6 43 .Bxf7 Kxd7 44.Bg6 Bd 1 , when Black
is drawing.
36 ... bxa3 3 7 .Kc3 h6

Black's problem is that his Bishop is stuck on g6 or h5 in order to protect

f7, and his King is unable to advance since that would allow the d-pawn to
promote. This means that Black is completely passive and must wait and
see how White intends to initiate a final breakthrough.
3 S.Kb3 Bh5 3 9.Kxa3 Kb7 40.Bd7 Kb6 4 1 .Kb4 a5+ 42 .Ke4 Be2 + 43 .Kd4
Bh5

Forced, since White was threatening to play Bxe6 ! .

44.Ke3 Kb7 45 .Kf4 KbS 46.Kg3 Kb7 47.Kf2 Bg6 4S.Ke2

My first idea was to force out Black's Bishop by bringing my King to g4

and playing h2 -h4-h5 . The other possibility was B-a4- d 1 followed by h2 h4-h 5 , but in both these cases Black's a-pawn might give me some prob
lems. In the end, I decided that I could force Black's Bishop away by mak
ing use of a Zugzwang theme. This plan required very little calculation
and seemed safe and artistically pleasing.
4S ... Kb6 49.Kd2 Bh5 50.Kc3 Bg6 5 1 .Ba4 Bh5 52 .Be2

Now a move like 5 2 . . . BD fails to 5 3 .Bg6 ! .

5 2 . . .Ke6 5 3 .Kb3 Kd7

Black willingly gives up the a-pawn because he saw that he'd have to lose
it anyway after 53 . . .Kb6 54.Ka4. For example: 54 . . . BD 5 5 .Bg6 Bc6+ 5 6.Ka3
BeS 5 7 .Bc2 (threatening 5 S .Ba4) 57 ... Bc6 5 S .Bb3 ! (threatening 5 S .Bxe6)
5S . . . Bd7 59.Ba4, and White wins.
54.Ka4 KdS 5 5 .Kxa5 Kd7 56.Kb6 KeS 5 7.Ke5 Kd7 5S.Kd4

Now that Black's a-pawn is gone, I can safely march back to the kingside.
5S ... KeS 59.Ke3 KdS 60.Kf4 KeS 6 1 .Be4, 1 -0.

After 6 1 . . .KdS 62 .BD Bg6 63 .Kg4 KeS 64.Kh4, Black is in Zugzwang

(64 . . . KdS 6 5 .Bh5 or 64 . . .KfS 65 .d7 or 64 . . . Bf5 6 5 .Kh5).

(96) Benko - Plato nov (Wijk aan Zee, 1 970)

-----

l.d4 Nf6 2.e4 e6 3 .NB Bb4+ 4.Bd2 Qe7

This system has gained enormous popularity in recent years.

5 .g3 Ne6

By placing pressure on d4 and preparing . . . Bxd2 +, Black will force White's

Knight to d2 . The b I -Knight would prefer to stand on c3 where it would
eye both e4 and d 5 , but a little tactic makes this impossible to achieve.

E U RO P E AT MY F E ET

6.Bg2 Bxd2 + 7.Nbxd2

White would like to play 7 . Qxd2 , but then the point of Black's previous
moves would be demonstrated: 7 . . . Ne4 8 . Qc2 Qb4+ 9 .Nc3 Nxc3 1 0 . Qxc3
Qxc 3 + 1 1 .bxc3 d6, with a superior position thanks to the weakness (and
lack of flexibility) of White's doubled c-pawns.

297

7 ... 0-0

Black often delays castling, preferring 7 . . . d6 here. However, this blocks

the Queen (ending . . . Qh4+ ideas) and allows White to play an immediate
8 .Nfl . The game Benko-Flesch, Belgrade 1 964 went: 8 . . . 0-0 9.N e3 N e4
1 0. 0-0 Nd8 I l . Qc2 f5 1 2 .d5 Ng5 1 3 .dxe6 NxB + 1 4.BxB Bxe6 1 5 .Nd5
Qf7 1 6.Nf4 c6 1 7 .Rad l g5 1 8 .Nxe6 Qxe6 1 9.Rd4, with a clear advantage
for White (I won on the 3 7th move).
S.Re l ! ?

The most common line i s 8.0-0 d 6 9.e4, though White hasn't been able to
achieve anything after 9 . . . e5 l O.d5 Nb8 . Placing the Rook on e l (which
defends c4 and thus eliminates . . . Qh4+ possibilities) is actually a useful
waiting move that also prepares a small surprise for my opponent.
S ... d6 9.Nf1 !

Platonov was startled by this move. As strange as it may look, it actually

serves a very logical plan: the Knight wasn't doing much on d2 so it jumps
to e3 where it will eye the important d5 and f5 squares.
9 ... e5 1O.Ne3 e4 I 1 .Nd2 Nxd4 1 2 .Nxe4 Nf5 l 3 .Nd5 Nxd5 14.exd5

Black has made every effort to get rid of the strong Knight
on e3 , but now 8 .Re l gains new significance on the c-file.
1 4 ... e5 1 5 .dxe6 bxe6 1 6.0-0

Black is trying to shake off the pressure by tactical means. I

avoided 1 6. Rxc6 due to 1 6 . . . Bb 7 1 7 .Rc4 d 5 . I'm in no hurry
to win a pawn if it gives Black counterplay since I have a long
term structural advantage.
1 6 ... RbS 1 7.Qd2 d5

Interesting is 1 7 . . . c5, when 1 8 .Nc3 Bb7 1 9.Bxb7 Qxb7 2 0 .Qf4

Qd7 2 1 .b3 Rb4 is fine for Black. After 1 7 . . . c5 , White should
play the simple 1 8 .Rfd l , when 1 8 . . . Bb7Ieads a clear endgame
advantage for White: 1 9.Nxd6 Nxd6 2 0 . Qxd6 Qxd6 2 1 .Rxd6
Bxg2 2 2 .Kxg2 Rxb2 2 3 .Ra6.
I S.Ne5 Rb5 1 9.Nd3

Jan Hein Donner

in 1 9 64.
(Photo courtesy
USCF.)

And not 1 9. b4 a 5 , when Black manages to rid himself of the

weak pawn on a 7 .
1 9. . .Bd7 20.a4 RbbS 2 1 .Ne5 BeS 2 2 .b3

Black's attempt at counterplay has been repelled and now he's left with an
inferior pawn structure and a long defensive chore. My last move (2 2 .b3)

PAL BE N KO: MY LI FE, GAM ES A N D C O M POSITI O N S

frees the Queen from having to defend the h-pawn and also takes the c4square away from the enemy Knight in case he plays ."Nd6.
22 ... QeS 2 3 .Rfd l Ne7 24.e4!
298

This is a well-known method of blowing up this particular pawn forma

tion. Now 2 4".dxe4 2 5 .Bxe4 would open the way for Re I or even Qd6.
24 ... Qf6

On 2 4".Rd8, White would play the annoying 2 5 .Qa5 .

2 S .h3

Also possible is 2 5 . exd 5 , but I would have been very sorry to give up my
fine Bishop, even if I won a pawn by doing so.
2S ... Be6 26.exdS cxdS 27.h4 h6 2 S.hS RfdS 29.QaS RaS 30.Nxe6?

The start of time pressure. I could have increased my positional superior

ity with either 3 0 .Nd3 or 3 0 .Na6 . Instead, I helped my opponent by
strengthening his d-pawn and giving him possibilities of counterplay on
the newly opened f-file.
3 0 ... fxe6 3 1 .Rc7 Rf8 3 2 . Qe l ?

As i s s o typical o f time pressure, one bad move follows another. I t would

have been more consistent to play 3 2 .Rd2 , when I had no reason to fear
3 2 ".Qa 1 + 3 3 .Kh2 Q e l on account of 3 4.Rxe7 Rxf2 3 5 .Qc7, with a mating
attack. Of course, this kind of thing is easy to see when you're sitting calmly
at your desk. However, during actual play, when each tick of the clock
brings you closer to defeat, phantoms and hallucinations rule the day over
logic and common sense.
3 2 . . .NfS B.Qc3 Qxc3 3 4 .Rxc3 RacS 3 5 .Rcc 1 Rxc 1 3 6 .Rxc 1 Rf7
3 7.RcS+ Kh7?

Black was also low on time, and this explains why he feared 3 7".Rf8 - my
queenside majority must have taken on epic dimensions. Nevertheless, af
ter 3 7".Rf8 3 8 .Rc5 , I would still have retained some advantage in the
endgame, but Black's chances would be more promising than they are af
ter the text move.
3 S.Bfl Nd6 3 9.Bd3 + g6 40.Rc6 Rd7 4 1 .f4 Kg7 42 .Kf2 gS 43 .Ke3 gxf4+
44.gxf4 Kf6

E U RO P E AT MY F E ET

299

After mutual blunders in our frenzy to make the time control, we could
finally relax and adj ourn the game. Black's 44 . . . Kf6 was his sealed move,
though I expected the superior 44 . . . d4+ (when White would retain a con
siderable edge with 45 .Kd2) .
45 .Kd4 h 5 46.a5 h4 47.a6

Black is in Zugzwang and must give way to the White pieces.

47 ... Ke7

Also losing was 47 . . . Nf5 + 48 .Bxf5 Kxf5 49.b6 axb6 50.Rxb6 Ra7 5 1 .Kc 5 .
48.Rc2

Black would have counterchances after 48 .Kc5 Ne4+, but now he must
try to prevent White's Rook from making inroads into his position on the
g-file.
48 ... Kf6 49.Ra2

Threatening 5 0.b6 axb6 5 1 .a 7 .

4 9 ... Nc8 50.Rg2 Rg7

Otherwise 5 1 .Rg6 followed by Rh6 would follow.

5 1 .Rxg7 Kxg7 52 .Kc5

Also winning is 5 2 . Ke 5 Kf7 5 3 . f5 exf5 5 4 . B xf5 Nb6 5 5 . B e 6 + Kg6

5 6.Bxd5 Kg5 .

An extremely interesting position! If I try to capture the a 7 -pawn at once,

my King would become caged by Black's . . . Kc7 when, despite his two ex-

PAL B E N KO : MY LI FE. GAM ES AND COM POSITI O N S

300

tra pawns, White would be unable to get the full point. During the game,
this convinced me that the position in the last diagram (after 56 . . . Kg5)
wasn't winning. However, afterwards I realized that White can win if he
approaches the a-pawn and simultaneously carries out a time-gaining ma
neuver designed to deflect Black's King from c7: 57.Bb3 Kg6 5S.Kd6 Kf6
59.Ba2 Na4 (59 . . . Kf5 60.Kc6 Ke5 6 1 .Bb3) 60.Bc4 Nb6 (60 . . . Kf5 6 1 .Kc7
Ke5 62.Kc6 Kd4 63 .Bb3 Nb6 64.Kb7 Nd5) 6 1 .Bb3 Kf5 62 .Kc7 Ke5 63 .Kc6
(Black is in Zugzwang-any King move takes him away from c7, while
Knight moves like .. .NaS or . . . NcS allow Kb7 with tempo.) 6 3 . . .Kd4 64.Kb7
Kc5 65.Kxa7 Kxb5 66.Kb7 Nd7 67 .a7 Nc5+ 6S .Kc7.
52 . . . Kf6 5 3 .Kc6 Ke7 54.Kc7 Nb6 5 5 .Be2 d4

If Black avoids . . . d5-d4, then White can win his a7- and e6-pawns after
55 . . . Na4 56.Bg4! Nb6 57 .Kb7 Kd6 5S .Kxa7 Kc7 59.Bxe6 d4 60.f5 d3 6 l .f6
d2 62 .7 when, although Black queens first, White still wins because of
the mate threat on bS.
56.Kc6 Na4 57.Bdl Nb6 58.Be2 Na4 59.Bfl , 1 -0.

On account of Zugzwang, Black will lose his d-pawn. White's task would
then be a matter of simple technique.

(97) Durao - Benko (Malaga, 1 970)

My opponent, a good tactician with a reputation of being hard to beat, had
won the championship of Portugal on several occasions. I had met him twice
before and both games had been drawn. Before this game he asked me if ! wanted
to win. "Of course, " I answered.
l .g3 d5 2 .Bg2 c6 3 .d3 Nf6 4.e4 dxe4

Durao seemed startled by this move, since he thought it indicated that I

wanted to draw.
5.dxe4 Qxd l + 6.Kxd l e5

Evaluating this position, it can be stated that Black has a tiny advantage.
White's King can no longer castle, further loss of time seems inevitable
and the Bishop on g2 doesn't have a great future. This position differs

E U RO P E AT MY F E ET

from the King's Indian with reversed colors in that Black has not played
. . . c5 , which would grant White a stronghold on d 5 . Of course, Black's vic
tory will depend on further mistakes by White, but I hoped my opponent
wouldn't be able to make 1 00 faultless moves, especially since this was the
last round and adjournments were no longer possible .
7.0 Be6 8.Be3 Na6 9.Nd2

Also good for Black is 9. Bfl 0-0-0+ 1 O.Nd2 Bc5 .

9 ... BcS

Black tries to trade off White 's good pieces, even if it simplifies the
position.
1 0.BxcS NxcS 1 l .Bh3 0-0-0 1 2 .Kc 1 Kc7 1 3 .Bxe6 Nxe6 1 4.Nh3 ? !

Inconsistent. Having brought the King to c 1 s o that the Knight o n g l

could safely go t o e2 , White decides to stick the poor beast o n the side of
the board. This error, which blocks the h2 -pawn, allows Black to
create weaknesses in White's kingside pawn structure.
14 ... hS l S .Nc4 Nd7 1 6.a4 h4 1 7.g4 f6

The hole on f4 and the weakness of the f3 -pawn combine to make life
difficult for White.
1 8.c3 ?

Weakening the d 3 -square. Stronger is 1 8 .Rd l Nd4 1 9.Ng l , though

after 1 9 . . . Nc5 his position would be far from encouraging.
18 ... NdcS 1 9.Kc2 Rd3 20.Rhfl

Of course, 2 0 .b4? ? fails to 2 0 . . . Rxf3 2 1 .Ng l Rf2 +.

20 ... Rhd8

White's position is threatened with immediate collapse. The threat is

2 1 . . .Nxe4, and 2 1 .Ng l isn't a good idea because of 2 1 . . .b5 2 2 . axb5 cxb 5 ,
when 2 3 .Rxa7+ Kb8 1 0ses a piece while 2 3 .Na3 Rd2 + i s also horrible.
2 1 .Rad l

Miserable but forced.

2 1 . ..Rxd l 22 .Rxd l Rxd l 2 3 .Kxd l Nxa4

30 I

PAL B E N KO: MY LIFE, GAM ES AN D COM POSITI O N S

O n account o f his inaccuracies, \Vhite has stumbled into a lost endgame.

24.Kc2 b5 2 5 .Ne3 a5 26.Nf5 g5 2 7.Ne7 Nac5
302

This Knight needs to be able to defend f6 by . . . Nd7 .

28.Ngl Nd7 2 9.Ne2 Nf4

\Vhite has no hope of a successful defense because he's a solid pawn down
and because the Black position is devoid of weaknesses.
3 0.Nc l

Taking on f4 would give the other black Knight access to e5 after 3 0.Nxf4
exf4.
30 ... a4 3 1 .Nd3 Nxd3 3 2 .Kxd3 Nc5 + H .Kd2 Kd7

The King rushes to the center and gives f6 some much needed support.
34.Ng8 Ke6 3 5 .Nb6 Nb3 + 3 6.Kd3 Na5 3 7.Nf5 Nc4

Tying \Vhite's King down to the defense of b2 .

3 8.Kc2 Kd7 !

\Vhite suddenly finds himself i n near Zugzwang! A Knight move (3 9.Ng7

or 3 9.Nh6) would be met by 39 . . . Nd6, when the Knights would eventu
ally be traded (the King and pawn endgame is hopeless for \Vhite). An
attempt to get rid of the weakness on b2 by 3 9.b3 fails badly to 3 9 . . . axb 3 +
40.Kxb3 Nd2 + and 4 1 . . .Nxf3 . That leaves \Vhite with nothing to d o but
move his King back and forth from c 1 to c2 .
3 9.Kc l c5 40.Kc2 Ke6 4 1 .Ng7+ Ke7 42 .Nf5 + Kd7

Gaining a bit of time on the clock.

43 .Kb l b4 44.Kc2

I suspected that might try 44.cxb4 cxb4 45.Kc2 , though 45 . . . a3 46.bxa3

bxa3 is completely winning since 47.Kb3 Nd2 + picks up all of \Vhite's
remammg pawns.
44 ... b3+ 45 .Kc l Kc6

I decided that this was the shortest route to victory, though I had many
good continuations to choose from. Now \Vhite's Knight finally frees it
self from f5 , but it is unable to do anything about the bad situation on the
queenside.
46.Ne7+ Kb5 47.Nd5 a3 48.bxa3 Ka4, O- I .

\Vhite can't stop the b-pawn from turning into a Queen.

The tournament at Reggio Emilia was a pleasant affair, though I once again
conformed to "tradition" and left a Rook en prise. This relegated me to second
place and left quite a bitter taste in my mouth since I was easily winning the
game.

E U RO P E AT MY F E ET

(98) Benko - Popov (Reggio Emilia, 1 970)

303

49.Qxf6 !

A sound Queen sacrifice that cannot be accepted because of 49 . . . Rxf6

5 0.Rc7+ Kh6 5 1 .RhS mate.
49 ... Qxe4+ 50.Kh2 ? ?

I was still under the spell o f the combination and I reversed two moves in
my mind. After the correct 50.Kg l Qe 3 + 5 1 .Kh2 Black would have to re
sign. Instead, reality came crashing down on my head after . . .
50 ... Qxc2 +, 0- 1 .

It's often hard to recover from a shock like this !

(99) Benko - Czerniak (Reggio Em ilia, 1 970)

l .d4 d5 2.c4 e6 3 . cxd5 ? !

Premature. It allows Black to comfortably develop his light-squared Bishop.

The text ought to have been prepared by 3 .Nc3 .
3 ... exd5 4.Nc3 c6 5.Nfl Bf5

Now White doesn't gain anything with 6.Qb3 , because Black can answer
with 6 . . . Qb6. This Queen move would also follow 6.Bf4 or 6.Bg5 . There
fore, I decided to explore new paths.
6.g3 Nf6 7.Bg2 h6

Black wants to secure the position of his light-squared Bishop. After 7 . . . Be7
or 7 ... Bd6, White could continue with S.Nh4 Be6 9.Qc2 , setting up the
possibility of Nf5 .
8.0-0 Bd6 9.Ne 5 !

White takes the opportunity t o occupy the only central square a t his dis
posal. Should the Knight be driven off, it would jump to d3 , providing the
possibility of b2 -b4-b5 on the queenside or of breaking up the center with
f2 -3. Also f2 -f4, to secure e 5 , can't be completely excluded. Finally, I had
to reckon with 9 . . . Bxe5 1 0.dxe5 Ng4, when I planned to play 1 1 .Qd4 Qb6
l 2 . Qf4 Be6 l 3 .h3 g5 l 4.Qa4 Nxe5 l 5 .Nxd5 .

PAL B E N KO : MY L I F E , GAMES AN D COM POSITI O N S

9 ... Nbd7?

Natural and bad. Black should have castled. Now White gains a bit of
time by making threats against Black's unprotected dark-squared Bishop.
304

1 0.Bf4!

Threatening Nxc6.
1 0 ... Qe7

Also 10 . . . Bxe5 1 1 .dxe5 Ng4 1 2 .e4 dxe4 1 3 .Nxe4 Bxe4 1 4.Qxg4 would be
to White's advantage. Black's King is far from safe in the middle of the
board.
1 1 .Nxd7 Qxd7 1 2 . Bxd6 Qxd6 1 3 .f3 !

White didn't trade to simplify the position. Instead, these exchanges were
the only way to maintain the initiative. Now my plan is clear: I intend to
push my e-pawn to e4 and build a strong center. This can't be easily par
ried. For example, 1 3 . . . 0-0-0 1 4.e4 dxe4 1 5.fxe4 Bg4 1 6 .Rxf6 would result
in the loss of material.
13 ... Qb4 1 4.e4 dxe4 I S .fxe4 Bg4 1 6.Qd2 0-0-0

The black King hopes to find refuge on the queenside, but this turns out
to be very unrealistic.
1 7.dS Rhe8

Black wisely avoids 1 7 . . . cxd5? 1 8 .Rac 1 ! Kb8 1 9 . Qf4+ ! Ka8 2 0 .e5 Qxf4
2 1 . Rxf4, when White wins a piece and the game.
1 8 .Rac 1 !

Taking aim at the enemy monarch. A little tactic prevents it from run
ning: 1 8 . . . Kb8 ? 1 9.Rxf6 gxf6 2 0.Qf4+.
18 ... Bd7 1 9.a3

Not as good was 1 9 .Qf4, since 19 . . . Qd4+ followed by 20 . . . Qe5 could be

played.
1 9 ... Qe7 20.NbS a6?

Black could have put up stiffer resistance by 20 . . . Kb8 2 1 . Qf4+ (2 1 .d6 Qe5
n .Nc7 Rf8 2 3 . Qf4 Ng4 holds the center) 2 1 . . . Qe5 n .Nd6 Qxf4 2 3 .Rxf4

E U RO P E AT MY F E ET

Rf8 (or 2 3 . . .Re7 24.dxc6 Bxc6 2 S .NfS) 24.dxc6 Bxc6 2 S .e S , when White
wins a pawn, but Black would retain some defensive chances.
2 1 .e5

Though 2 1 .dxc6 Bxc6 2 2 .Na7+ Kb8 2 3 .Nxc6+ would destroy the black
King's position, I found the text move aesthetically more attractive.
2 1 . ..Qxe5

In case of 2 1 . . .axbS 2 2 .exf6 gxf6, White could choose between two very
good endgames via 2 3 .Qxh6 Rh8 24. Qxf6 Qe3 + 2 S . Qf2 , or 2 3 .Qf2 Qd6
2 4.Qxf6 Qxf6 2 S .Rxf6 Re2 2 6 . Rxf7 .
2 2 .Rce l

Unlike 2 2 .Rfe l , this doesn't allow the Rooks to be exchanged with check
in some key variations.
22

..

Qg5

On 22 . . . Qb8, White plays the lethal 2 3 .dxc6 Bxc6 24.Bh3 +.

2 3 .Nd6+ Kb8 24.Qb4

White squeezes the most out of the position.

24 ... Bc8 2 5 .Nxe8 Nxe8 26.dxc6

It's time for Black to resign. He comes to this realization himself in a few
moves.
26 ...Nd6 2 7.cxb7 Be6 2 8.Qb6 Nc4 2 9.h4, 1 -0.

It was actually stronger to play 2 9 . Qxa6, but I was quite happy to force
the exchange of Queens.

( 1 00) Bukal - Benko (Sarajevo, 1 970)

l .g3 d5 2 .Bg2 e5 3 .c4 c6 4.cxd5 cxd5 5 .Qb3 Nf6 6.Nc3

My opponent told me in advance that he would employ the Benko Sys

tem. To be frank, when playing White I rarely allow both enemy center
pawns to advance. This position, now well known, is usually arrived at
from the following move order: l .c4 eS 2 .Nc3 Nf6 3 .g3 c6 4.Bg2 dS S .cxdS
cxdS 6.Qb3 .
6 ... Nc6 !

Though this pawn sacrifice had been previously mentioned by Pachman,

the famous Czech analyst thought that accepting the pawn would lead to
a White advantage after: 7 .NxdS Nd4 8 .Nxf6+ Qxf6 (Later, 8 . . . gxf6 was
thought to be even stronger, though it isn't clear if this is actually the case.)
9.Qd l BfS 1 0.d3 Rc8 1 1 .Kfl . Now, of course, it's known that this whole
line is bad for White. The first inkling of this truth came in the 1 968 cor
respondence game Abramov-Kuuksmaa. Both players enthusiastically en
tered the "advantageous" position given by Pachman, but Black evidently

305

PAL B E N KO : MY LIFE, GAM ES AN D C O M POSITI O N S

306

had his own ideas about what was going on: 1 1 . . . Qa6 ! (threatening . . . Nc2)
1 2 .Bd2 Bc5 1 3 .Bc3 0-0 1 4.Nf3 Nxe2 ! ! (a strong sacrifice that demolishes
the White position) 1 5 .Kxe2 e4 1 6.Nh4 Bg4+ 1 7 .f3 exd 3 + I S .Kfl RfeS
1 9.Qd2 Be3 2 0.Qe l Bd7 2 1 .a4 Bb6 2 2 .Qd2 Re2 2 3 .f4 Rxd2 , 0- 1 .
7.d3

Evidently, my opponent wasn't familiar with this position. After quite a

bit of thought, he finally decides to decline the sacrifice.
7 ... d4 8.Nb l Bb4+ 9.Bd2 as !

Black's superiority is obvious. Of course, 1 0.Bxc6+ bxc6 I I .Bxb4 Be6! (stron

ger than 1 1 . . .axb4 I 1 . Qxb4 Qd5 1 2 .Nf3 ) 1 2 . Qa4 axb4 1 3 . Qxc6+ Nd7
( 1 4.Qd6? Qa5 ! traps the Queen due to the unstoppable threat of . . . Ra6)
would accentuate this advantage.
1 0.a3 Be6 I 1 .Qa4 Be7

Black is willing to give up a pawn in order to eliminate White's light-squared

Bishop, but my opponent was not tempted.
1 2 .Bc1 0-0

I never imagined that I would have such a dream position with the black
pieces after only twelve moves!
1 3 .Nd2 Nd7 1 4.Qd l a4 I S .NgB NeS 1 6.0-0 ReS 1 7.Ne4 f6

White is in dire straits since his queenside is crippled and he is completely

devoid of counterplay. Natural development with I S . Bd2 fails to I S . . . b 5 ,
which i s a constant threat i n connection with . . . Na5 -b3 . But his next des
perate attempt also fails.
I S .b4? ! axb3 1 9.a4 RaS 20.Nfd2 Nxa4 2 1 .Qxb3 Nc3 22.RxaS QxaS

Stronger than 2 1 . . .Nxe2 +, since guarding the pawn (e.g., taking the bait)
gives Black an opportunity for a quick windup.
2 3 .Re l bS 24.Nb6 Bxb3 2 S .Nxa8 BdS ! , 0- 1 .

The Knight is trapped: 2 6 .Nc7 Bxg2 2 7 .Kxg2 ReS 2 S .Na6 RaS 2 9 .Nc7
Ra7 3 0.Ne6 Kf7.

EU ROPE AT MY F E ET

( I 0 I) Benko - Jansa (Siegen Olympiad, 1 970)

l .NfJ g6 2 .e4 Bg7 3 .d4 d6 4.c3

I've always been rather fond of this solid positional system. More popular
is 4.Nc3 Nf6 5 . Be2 .
4 ... Nf6 S.Bd3

When we finished, Jansa told me that he thought 5 .Bd3 was a mistake (a

strange comment when you consider the fact that he lost this game rather
badly!). Others also voiced this opinion, parroting the opening books by
insisting that the Bishop belonged on e2 or c4. However, I continued to
employ 5 . Bd3 with great success, and eventually it was accepted as an ex
cellent way to handle the position.
5 . . .0-0 6.0-0 Nc6

Black's sharpest reply, intending . . . e 7 -e5 when the c6-Knight will help pres
sure d4.
7.Nbd2

More recently, 7 .b4 has been used by such strong players as Speelman,
Leko, and Korchnoi .
7 ... eS

Perhaps Black should consider 7 ... Nd7!? followed by ... e7-e5 . That way
he can capture on e5 twice with a Knight and gain a tempo on my d 3 Bishop.
8.dxeS NxeS 9.NxeS dxeS 1 0.Nc4

I placed my Bishop on d3 so I could quickly get my Knight into action.

One point of 1 O .Nc4 is to prevent Matulovic's idea of . . . Bh6 . This way
Black isn't able to exchange his rather inactive Bishop. Even so, White
can't hope for more than a very slight plus.
1 0 ... Nd7

Black plays cautiously. A more aggressive move is 1 O . . . Nh5 , heading for

the f4-square. From now on White starts to exert some pressure on Black's
queenside.
1 1 .b4

Gaining queenside space and taking the c5-square away from Black's Knight.
1 1 ...Re8 1 2 .Bc2 Qe7 n.Be3 Nb6? !

Preferable was I L.Nf8 . Now Black's position becomes uncomfortable.

14.NaS

Avoiding exchanges and placing annoying pressure against b 7 .

14. . .Rb8 l S .Bb3 Be6 1 6.Qe2 c6

I would have been happy to see I 6 . . . Bxb 3 , since I 7 .axb3 instantly gives
my a I -Rook play on the half-open a-file.

307

PAL B E N KO : MY L I F E, GAM ES A N D C O M POSITI O N S

1 7.g3 Qc7 1 8.Bxe6 Rxe6 1 9.Rfd l

308

My edge is based on a superior Knight, a good Bishop versus Black's bad

one on g7 , more queenside space, and more dynamic queenside pawns .
1 9 ... Ree8 20.Rd3 Rbd8 2 1 .Rad l Bf6

Black is preparing to trade the Rooks along the d-file. His last move, 2 1 . . . Bf6,
guards d8 and gives the Bishop something to do, though it's clearly infe
rior to White's piece on e 3 . While Black is busy trading Rooks, I try to
make some useful territorial gains.
22 .c4 Rxd3 2 3 .Rxd3 Rd8 24.c5 Nc8 2 5 .Nc4 Rxd3 26.Qxd3 a6

If I had the black pieces, I would have preferred 2 6 ... Be7 (and not 2 6 ... Ne7 ? ?
2 7 .Qd6, when White wins) . Despite the exchange o f Rooks, White i s still
exerting quite a bit of pressure on Black's position. The second player now
brings his King towards the center for defensive purposes while I decide
to pile up on his eS -pawn.
2 7.Bd2 Kf8 2 8.Bc3 Ke7 29.Kg2

Taking my time and letting my opponent simmer in his misery. White's

strongest idea was 29.f4 exf4 3 0 .eS, when 3 0 . . . Bg7 3 1 . Qd6+! wins on the
spot. However, 3 0 . . . BgS! keeps Black in the game after 3 1 . Qd6+ Qxd6
3 2 .exd6+ Ke6, though it's hard to believe he'll ultimately be able to hold
body and soul together after 3 3 .Bd2 .
29 ... Bg7 30.f4

One move too late, but still good.

3 0 ... f6 3 1 .fxe5 fxe5 3 2 .h4 Na7 3 3 .a4

Why complicate when you can keep your opponent completely helpless?
3 3 .a4 deprives his Knight of the bS -square and leaves all of Black's pieces
in very passive positions. 3 3 . Qd6+ would be premature due to 3 3 . . . Qxd6
3 4.Nxd6 (Black also escapes after 3 4.cxd6+ Kd7 3 S .Nxe S + Bxe S 3 6.Bxe S
NbS) 3 4 . . . NbS . When your opponent is helpless, make sure you keep him
that way!
3 3 ... Nc8 34.Qd2 Ke6 3 5 .Qa2 Ke7 3 6.Qe2

The black pieces are badly tied up and White is ready to weaken Black's
position by hS and Qg4.

E U RO P E AT MY F E ET

3 6 ... hS

Stopping White's plan, but creating a new break in point on g5 . Black's

game is no longer tenable.
3 7.Qd2 Bf6

This dies a horrible death. If he wanted to continue the fight, he had to

try 3 7 . . . Ke6, when 3 8 .Qg5 Ne7 holds on for a while longer.
3 8.Qh6 Ke6 39.Qxg6 Ne7 40.QxhS Qd8, 1 -0.

Black resigned without waiting to see the devastating 4 1 . Qg4+ . A good

positional game, which was used by Keene (in his book, The Modern De
fense) as an example of how to play White in this line.

( 1 02) Benko - Kavalek (U.S. Open, 1 970)

l .Nf3 Nf6 2.c4 d6 3 .b4! ?

A form o f the Orangutan Opening ( l .b4) that has certain advantages. One
of them is that it's little analyzed, another that it presents Black with prob
lems to solve that are likely outside his normal experience. I do not claim
this to be any better than more usual continuations, and in fact I wasn't
very successful the first few times I tried it. However, once I got used to
the spirit of the system, I did quite well.
3 ... eS 4.Nc3 g6 S.d3 Bg7 6.Bb2 0-0 7.e3

It is possible to fianchetto the other Bishop, but greater flexibility is pre

served by developing it on e2 and having pawns on e3 and d3 -one or
both pawns may advance as the position requires.
7 ...Re8 8.Be2 c6 9.0-0 dS

Kavalek, a classical player, tries to occupy the center immediately with his
pawns. His other main possibility was to stay within King's Indian lines
with . . . Nbd7 , but in that case it would be better for Black's c-pawn to be
on c7 (R. Byrne played this way against me in the same tournament), stop
ping White from opening queenside lines by b4-b 5 .
1 0.cxdS cxdS l 1 .Rc 1

309

PAL B E N KO: MY L I F E , GAM ES AN D COM POSITI O N S

An interesting position. Black's control of the center is offset by White's

lead in development. White must always be on guard against the possibil

310

ity of . . . e5 -e4 or . . . d5 -d4, although either move at this point wouldn't bother
me (the former is met by 1 2 .dxe4 dxe4 1 3 .Nd4, while the latter runs afoul
of 1 2 .exd4 exd4 1 3 .Nb5). In the meantime, White keeps the black center
pawns under pressure and tries to break into Black's position via the open
c-file.
1 1 ...a6 1 2 .a4

Of course, White cannot allow Black to establish a Knight on c6, which

would close the c-file.
1 2 ... Nc6 l 3 .b5 Nb4 1 4.Qb3 a5

Wisely avoiding 14 . . . Nxd 3 1 5 . Bxd 3 e4 1 6 .Nxe4 ! dxe4 (even worse is

16 . . . Nxe4 1 7 . Bxg7 Kxg7 1 8 .Rfd l ) 1 7 .Ng5 , and now the greedy 1 7 . . . exd 3 ? ?
loses to 1 8 . Qxf7+ Kh8 1 9.Bxf6 .
1 5 .Na2

The advanced Knight outpost must be eliminated. At the same time, the
text opens the b2 -Bishop's diagonal, attacking e 5 .
1 5 . . . e4 1 6.dxe4

Also possible was 1 6 .Nd4, with play against the isolated d-pawn if Black
exchanges pawns on d3 . This choice of good possibilities illustrates the

....... . ....... v..S. f.r..

A card sent to long-time Fischer friend, A. Liepnieks, from the 1 970 Siegen O lympiad. It
is signed by Evans, Edmondson, Spassky ( i n Cyri l l i c) , Benko, Reshevsky, Lom bardy, Mednis,
Fischer, and Koltanowski.

E U RO P E AT MY F E ET

faults of Black's Knight maneuver to b4, as well as the instability of his

classical center.
1 6 ... dxe4 1 7.Nd4 Nxa2 l S.Qxa2 Nd7? !

Mter the game, my opponent told m e he considered this to b e the deci

sive mistake. He said he should have played 1 8 . . .Ng4, but he was afraid of
the simple 1 9 . Bxg4. It's true that after 19 . . . Bxg4 2 0.h3 Bd7, White will
quickly occupy the center files and that all resulting endgames will be in
his favor. Even if Black tries for an opposite-color Bishop endgame by
playing . . . Bxd4 at some point, the long a 1 -h8 diagonal will only add to
White's superiority. However, there is a "blip" on this idyllic picture: after
1 8 . . . Ng4 1 9.Bxg4 Bxg4 2 0.h3 , Black can force a draw with 2 0 . . . Bxh3 !
2 1 .gxh3 Qg5+ 2 2 .Kh 1 Qh5 . This means that White must either avoid 20.h3 ,
or try 1 9.h3 (instead of 1 9.Bxg4) 1 9 . . . Ne5 2 0 . Rfd 1 Qg5 2 1 .Kfl , with a
superior position.
1 9.Bc4 Ne5 20.Bd5

Having established this Bishop on a dominating square, Black's position

becomes critical. The main problem is defending the e4-pawn - the Knight
can't move because of the pressure on f7 . As is so often the case, Black is
forced to launch a desperate attack in the hope of gaining some compen
sation for his positional failings.
20 ... Qh4 2 1 .Rc7 Rf8

A sad move, but attempts such as . . . Ng4 are easily dealt with. White can
now win an Exchange with 2 2 .Ba3 , but I preferred to play it safe and not
give Black counterchances.
22 .Ne2 Bd7 2 3 .Ng3 RadS 24.Bxe5 Bxe5 2 5 .Rxb7

White's play has been simplicity itself: win a pawn and avoid complica
tions. Black is reduced to meaningless moves, such as the next, offering a
pawn that I don't bother taking.
2 5 ... KhS 26.Qc4

There was nothing wrong with 2 6 . Bxf7 , but White is satisfied with his
material advantage.
26 ... f5 2 7.Be6 BeS 2S.Bf7 Bd7 29.Qc5 Bg7

31 1

PAL B E N KO : MY L I F E , GAM ES AN D COM POSITI O N S

Forced, due to the threats of QxeS or Rxd7 .

30.Bb3 f4 3 1 .exf4 Qxf4 3 2 .Qa7 Qd2 ?
312

Black's only move was 3 2 . . . Ra8, though after 3 3 .Qe3 he is forced to trade
Queens and enter a lost endgame. After the text, White can even win a
piece.
3 3 .Nxe4?

On 3 3 .Rd l Qb4, I examined 34.Rdxd7 Qxb3 3 4.Rxg7? ? Rd 1 + 3 S .Nfl Rxfl +

3 6.Kxfl Q d l mate. Spooked by this, I didn't bother looking at 3 3 .Rd l
again, though later I realized that 34.Rbxd7 Rxd7 3 S .Rxd7 wins a clean,
safe, piece.
H ... Qb2 34.Rxd7 Rxd7 3 S.Qxd7 Qxb3 36.NcS Qb4 37.Ne6 RgS 3 S.Nxg7
Rxg7 3 9.QeS+ RgS 40.QeS+ Rg7 4 1 .Rc 1 h6 42 .h3 KgS

Simple is 42 . . . Kh7 43 .Rc7 Rxc7 44. Qxc7+ Kg8 4S .b6.

43 .RcS+ Kh7 44.QeS

Since Black refuses to give up, we will have to play for mate!
44 ... gS 4S.QhS+ Kg6 46.Rc6+, 1 -0.

( 1 03) Benko - Csom (Pal ma de Mallorca, 1 97 1 )

l .Nf3 Nf6 2 .g3 dS 3 .Bg2 e6 4.0-0 Be7 S .d3

An aggressive intent is hidden in this seemingly mild move. White is ac

tually playing a reversed King's Indian against which Black's setup is not
the most effective . I have obtained many grandmaster scalps with this sys
tem (for example, see game 3 8 : Benko-Uhlmann, Stockholm 1 962), and
Fischer himself adopted it successfully several times. In fact, he told me
on one occasion, "Today I played like Benko." I have taken that as well
intended praise.
5 ... 0-0 6.Nbd2 cS 7.e4 Nc6 S.Re l Qc7 9.eS Nd7 1 O.Qe2 bS

According to the spirit of this opening, Black tries to find counterplay on

the queenside since White has a space advantage on the kingside.
1 1 .Nfl as 1 2 .Bf4 Ba6 1 3 .h4 b4 1 4.Ne3

My Knight is heading for g4, but taking the road through e3 (rather than
h2) gives me various tactical tricks. For example, Black now has to worry
about I S .NxdS exdS 1 6.e6.
14 ... Nb6

This deals with the threatened NxdS , while 1 4 . . . Rfe8 I S .NxdS exdS 1 6.e6
Bd6 (or 16 . . . Qc8 1 7 . exf7+ Kxf7 1 8 . Qe6+ Kf8 1 9 .NgS BxgS 2 0 . Bd6+ Re7
2 1 . BxdS , when the board will run red with Black's blood.) 1 7 . exf7 + Kxf7
1 8 .NgS + Kf8 1 9 .Nxh7+ Kg8 2 0 . BxdS+, results in a quick snuff. I should
also mention that 14 . . . Ndxe S ? ? loses to l S .NxeS Nxe S 1 6.BxeS Qxe S
1 7 .NxdS Qd6 1 8 .Nxe7 + Qxe7 1 9 .Bxa8.

E U RO P E AT MY FEET

1 5 .Ng4 Qa7

With this move Black gets his Queen off the dangerous f4-b8 diagonal,
thus avoiding future tricks based on Nf6+.
1 6.h5 Rfe8

Here 1 6 . . . h6 was to be taken into consideration, but in this case Black has
to reckon with sacrifices on h6, or White may be able to carry out a break
through with f4-f5 , as in a game Benko-Bisguier, Stockholm Interzonal
1 962 . Interested readers may consult R. Keene's Flank Openings, where
that game is an instructive example.
1 7.h6 g6

The h2 -h4-h5 -h6 plan created both middlegame and endgame chances.
In an endgame, Black's King will find it difficult to participate. Also, his
h7-pawn can easily turn out to be a target. In a middlegame, White wants
to exchange dark squared Bishops and penetrate into Black's kingside on
the weakened dark-squares.
1 8.c3 bxc3 1 9.bxc3 Nd7 20.Bg5 NfS

It is generally recommended that the exchange of dark-squared Bishops

be avoided by Black since, if they do get traded, the squares f6 and g7 can
easily fall into White's hands.
2 1 .Bxe7

In his joy, White exchanges the Bishops quickly, though Black would be
unable to avoid the exchange anyway. Better was 2 1 . Qd2 , since this would
have made it more difficult for Black's Queen to protect the kingside.
2 1 . ..Qxe7 22 .Qd2

Preparing to swing my Queen to the kingside via Qf4.

22 ... Rab8 2 3 .Rab l Nd7 24.Bfl

Having defended d3 and stopped Black from taking over the b-file, I'm
finally ready to give my full attention to the kingside.
24 ... QfS 2 5 .Qf4 Rxb l 26.Rxb l Rb8 2 7.Rxb8 Qxb8 28.Ng5 QfS 29.Nf6+
Nxf6 30.Qxf6

Threatening 3 1 .Nxe6.
30 ... Be8

313

PAL B E N KO : MY LI FE, GAM ES AN D C O M POSIT I O N S

The Bishop hurries to protect the attacked e-pawn.

3 1 .Bh3 NbS?
314

Either 3 1 . . .Nd8 or 3 1 . . .Ne7 was correct, even though White's positional

advantage is obvious anyway. With 3 1 . . . Nb8 , Black is hoping to chase
White's Queen from its powerful post on f6. However, Black missed some
thing.
3 2 .c4?

A pity. Though this isn't bad in itself, there's no excuse for my failure to
play 3 2 .Nxe6 ! ! fxe6 B .Bxe6+ Bxe6 3 4. Qxe6+ Kh8 (Black ends up in
Zugzwang after 34 . . . Qf7 3 5 . Qc8+ Qf8 3 6.Qb7 ! - his Queen and Knight
can't move and all his pawn moves will soon run out. After 3 6 . . . Kh8 3 7 .e6,
White threatens 3 8 .Qf7 Qxf7 3 9.exf7 , when, after Black plays the forced
3 9 . . . Nd7 , White's King can leisurely penetrate into Black's camp while
the enemy Knight is tethered to the f8-square and the enemy King doesn't
have any moves at all.) After 3 4 . . . Kh8, I spent a good deal of time looking
at 3 5 . Qxd 5 . However, in time pressure, I failed to notice (even though
I'm a problem composer!) that taking this pawn is superfluous since Black
is also in Zugzwang in the position after 34 . . . Kh8 (in other words, White's
Queen is better on e6 than on d5). For example, 3 5 .a4 should do the trick,
since 3 5 . . . Qxh6 (3 5 . . . Qd8 3 6. Q f7 is simple for White.) 3 6 . Qc8+ Kg7
3 7 . Qb7+ Kg8 3 8 . Qxb8+ wins easily due to the strength of White's passed
e-pawn.
3 2 ... dxc4 H .dxc4 Nd7 34.Qf4 Qe7?

Correct was 3 4 . . . Bb 7 , not fearing 3 5 .Nxe6 due to 3 5 . . . Qe7 . Unfortunately,

we were both low on time and, as is so typical, a cascade of errors is the
result.
3 5 .Bg2 !

Again it is difficult for Black to free himself from the pressure. The threat
is 3 6.Bc6 in connection with a Knight maneuver through e4 to f6 or d6.
35 ... f6? ! 36.exf6 Qxf6

On 3 6 . . . Nxf6, there would follow 3 7 .Qb8 and Bh3 .

3 7 .Nxe6! ?

E U RO P E AT MY F E ET

For the second time, White has the opportunity to make this move . How
ever, it is possible that 3 7 . Qc7 was even stronger: 3 7 . . . Qxg5 3 8 . Qxc8+ Nf8
3 9 . Qc7 Qxh6 40.Qxa 5 , when the passed a-pawn is very powerful.
3 7 ... Qxf4

Of course, 3 7 . . . Qxe6? ? isn't possible due to 3 8 .Bd 5 .

3 8 .Nxf4 Ne5 3 9.Bd5+ Kf8 40.Kg2 Ke7 4 1 .f3

This is not the most accurate way to play, but neither of us knew if we had
made the 40th move. The consequences of 4 1 . Bg8 are difficult to foresee,
since after Bxh7 Black is able to confine the Bishop by . . . Kf7 . The text
move prevents . . . Ng4, but the plan Nf4-h3 -g5 (threatening both Nxh7
and Ne4 followed by Nxc5) was to be preferred.
4 1 . ..Ba6?

Better was 4 1 . . .Kf6, with the threat of . . . Kg5 .

42 .Kf2?

A loss of tempo. The immediate Ne6 was necessary, but White is thinking
only of Nxg6+ in case Black plays . . . Nxc4. Playing for obvious traps is
never a wise thing to do !
42 ...Kf6 43 .Ne6 g5 ?

Now Black misses his last chance to escape: 43 . . . Bxc4 44.Bxc4 Nxc4, when
45.Nf8 is answered by 45 . . . Kg5 , with a likely draw.
44.NfS Bxc4

Slightly better was 44 . . . g4, although after 45 .f4 Nxc4 46.Nd7+ Ke7 47.Nxc5
Kd6 48.Bg8 ! Kxc5 49.Bxh7, the h-pawn can't be stopped.
45.Bxc4 Nxc4 46.Nxh7+ Kg6 47.Nxg5 ! , 1 -0.

After 47 ... Nd6 48.h7 Kg7 49.Ne6+, the game is over.

( 1 04) Benko - Barczay (Vrnjacka Banja, 1 97 1 )

1 .Nf3 g6 2 . e4 Bg7 3 .d4 d6 4.c3 Nf6 5 .Nbd2

I think that 5 .Bd3 is more accurate than the immediate 5 .Nbd2 .

5 ... 0-0 6.Be2

This modest but solid buildup against the Pirc was used by me on many
occasions. Also possible are 6.Bc4 and 6.Bd3 . An example of me using this
latter move: 6.Bd3 c5 7 .dxc5 (7 . 0-0 is also worth consideration since now
my Bishop will be hanging and I will lose some time) 7 . . . dxc5 8 . Qe2 Nc6
9.0-0 Qc7 l O.Re l b6 1 l .Nc4 Bg4 1 2 .a4 a6 1 3 . Bc2 (the position of the
Bishop resembles a Ruy Lopez, as does the pawn structure to some ex
tent. But an important difference is that Black has not played . . . e5 and he
can therefore defend his d5 -square by . . . e6) 1 3 . . . b5 1 4.Ne3 BxB 1 5 . QxB
e6 1 6.Qe2 c4 1 7 .g3 Nd7 1 8 .f4 Nc5 1 9.Ng4! (It is Black's intention to oc
cupy d3 with a Knight. White's move, a preparation for the f4-f5 break-

315

PAL B E N KO : MY L I F E , GAM ES AN D COM POSITI O N S

316

through, has the additional purpose o f being able to retreat strongly t o f2

if pushed, thus hindering Black's plan.) 1 9 . . .f5 ? ! 2 0 .Nf2 Rfd8 2 1 .Be3 Bf8
2 2 .g4! , with advantage for White. The finish was: 2 2 . . . fxg4 2 3 .axb5 axb5
24.Rxa8 Rxa8 2 5 .Nxg4 Rd8 26.f5 exf5 2 7 . exf5 Nd3 2 8 .Rfl gxf5 2 9 . Rxf5
Bg7 3 0.Nh6+ Bxh6 3 1 . Bxh6 Kh8 3 2 .Rg5 ! Nf4 3 3 . Qf3 Rd7 (desperation,
since 3 3 . . .Ng6 could be answered by 3 4.Qf6+ Kg8 3 5 . Qf8 mate) 34.Rg4
Ne5 3 5 .Qa8+ Rd8 3 6 .Bg7+, 1 -0, Benko-Saverymuttu, England 1 97 3 .
6 ...Nc6 7.0-0 e 5 8.dxe5

White tried to hold his center with 8.Re 1 in Geller-Parma, Siegen 1 970,
but after 8 ... Re8 he continued with the illogical 9.d5 ? ! Nb8 1 0.Bfl c6
1 1 . dxc6 Nxc6 1 2 .Bc4 h6 1 3 .h3 ? ! , and now 13 ... Be6 would have given Black
the better game according to Parma. By capturing on e 5 , I create a pawn
structure where White has a small but safe edge.
8 ... Nxe5 9.Nxe5 dxe5 1 0.Qc2 Be6

The position looks simple, but this obvious-looking developing move is

already a mistake, as we shall see. Parma recommends 1O . . . Bh6 ! , so Black
can trade off his passive Bishop once White moves his Knight.
l 1 .NfJ Nd7 1 2 .Ng5 !

Immediately grabbing the opportunity to get Black's good Bishop. White's

advantage, though not huge, is beyond dispute. It can be mentioned here
that Black offered a draw after his 1 0th move, which I declined.
1 2 ... Qe7 1 3 .Be3 Rfd8 1 4.Nxe6 Qxe6 1 5 .Qb3 !

By forcing the trade of Queens, I get to open the a-file and increase the
activity of my two Bishops. White is accumulating small advantages that,
over time, will finally become decisive.
1 5 . . . Qxb3 16.axb3 a6

Black doesn't sense the danger else he would play the more active . . . a 5 ,
securing the c5 -square for a trade o f Bishops by . . . Bg7-f8-c5 . Of course,
White could try to stop this in various ways, such as Ra4 followed by Rfa 1
and, if Black's Rook is still on a8, b3 -b4.
1 7.b4 Bf6 1 8.g3 Kf8 1 9.Bc4 Ke8 20.f4 Be7

E U RO P E AT MY F E ET

317

Black is already in serious trouble. White threatens to open the position

for the two Bishops and break through on the f-file. The d-file is not im
portant here since all possible points of penetration are controlled by the
white Bishops.
2 1 .Kg2 f6 22 .h4 exf4 2 3 .gxf4 BfS 24.Bg8

Here White has many good plans; I was even considering a pawn sacrifice
by 24.eS fxeS 2 S .fS , with tremendous compensation. Since it's only pos
sible to employ one idea at a time, I decided to play the solid text move.
24 ... h6 2 S .Kf3 cS

This is Black's only possible attempt at counterplay. Passive play offers no

hope for survival at all.
26.Rgl cxb4 27.Rxg6 bxc3 2 8.bxc3 Rac8 29.Bd4 BcS 3 0.Kg4

Of course, 3 0.Rxh6 was good too, as was 3 0 .BdS, when 3 0 . . . Bxd4 3 1 .cxd4
fS tries to break up the white pawns. In that case 3 2 .exfS should win, but I
decided to avoid the possibility with the King move.
3 0 ... Bxd4 3 1 .cxd4 Ke7 3 2 .Rg7+ Kd6

White is trying to gain some time on the clock, but Black does not want
to allow this. Mter 3 2 . . . KfS 3 3 .Rg6 Ke7 H.BdS fS + 3 S .KxfS RfS+ 3 6.Kg4
Nf6+, the Exchange sacrifice with 3 7 .Rxf6 would win because the three
connected passed pawns are simply too strong.
B .eS+ fxeS 34.dxeS+

Instead, H.fxeS+ would force Black to give up his Knight for two pawns:
3 4 . . . NxeS + 3 S .dxeS + KxeS 3 6.Re 1 + Kd4 (Black gets mated after 36 ... Kf6),
and the win is a matter of technique. I took the other way so he would
have the option of not giving up his Knight.
34 ... Kc6

He could have played H . . . NxeS when White wins as in the previous note .
Actually, I was happy that he didn't do this since now the winning process
is considerably shorter.
3 S .Rc 1 + ?

A mistake that forces m e t o put a bit more work into the game. I could
have won immediately with 3 S .Be6 ! .

PAL B E N KO: MY LI FE, GAM ES AN D COM POSITI O N S

3 5 ... Nc5 3 6.Rg6+ Kh5 3 7 .Rh l + Ka5 3 8.Ra l + Kh4

Allowing me to finish up quickly. A better move in a losing cause was

3 8 . . . Kb S .
318

3 9.Rh6+ Kc3 40.Ra3 + Kd2 4 l .Bh7, 1 -0.

In addition to White's two connected passed pawns, the black King is in a

ticklish position (4 1 . . .Ke2 42 .Rh3).

( 1 05) Benko - Dom inguez (Las Palmas, 1 972)

White to Move

In this position any normal human would play 4 1 .BxB RxB 42 .Rfl , when
Black does best to resign (42 . . . Rxd3 43 .Rf7). For reasons I can't compre
hend, I uncorked the ludicrous
4 l .Rh7+ ? ?

I recall my shocked opponent's eyes bugging out o f their sockets !

4 1 . . .Kxh7 42 .Re2

and after
42 .. .Rxg2 +

I was forced to resign. Such displays of insanity have often left me won
dering why I was never invited to play on the U. S . Olympiad team for the
blind.

( 1 06) Benko - Menvielle (Las Palmas, 1 972)

l .NO f5 2 .g3 Nf6 3 .Bg2 g6

Black employs the popular Leningrad Variation.

4.h3 Bg7 5.Bh2 d6 6.d4 0-0 7.0-0 c6 8.c4 Ne4 9.Nbd2 d5

My opponent unexpectedly chooses a Stonewall setup. Instead, I was an

ticipating . . . Qc7 in conjunction with . . . e7-e S .
1 O.Qc2 Nd7 1 l .cxd5 cxd5 1 2 .Rfc l

White's answer to 1 0 . . . Nd7 was quite normal: since Black's QN will now
be unable to close the c-file (via . . . Nc6), White has opened it. Neverthe-

E U RO P E AT MY F E ET

less, this Rook move turns out to be slightly inaccurate since it weakens
f2 . Preferable was 1 2 .Rac l .
1 2 ...Ndf6 1 3 .Ne5

Now Black will be able to drive White's Queen from the c-file. I had in
tended to play 1 3 .Qc7, but then I noticed that Black would get some
counterplay with 1 3 . . . Ng4 (this attack against f2 was made possible by my
inaccurate 1 2 th move). Therefore, White tries to occupy the permanent
weakness in the Dutch-the eS -square. This square is always weak com
pared to White's e4, which can still be controlled by a pawn. In retro
spect, it turns out that 1 3 .Qc7 does indeed pose Black some serious problems: 1 3 . . .Ng4 1 4.Nxe4 fxe4 ( 1 4 . . . Qxc7 I S .Nf6+) I S . Qxd8 Rxd8 1 6.NeS
Rf8 ( 1 6 . . . NxeS 1 7 .dxeS is unpleasant for Black) 1 7 .f4 ( 1 7 . f3 is also good)
1 7 . . . NxeS 1 8 . dxeS Bg4 1 9.Rc7 Bxe2 2 0.Bh3 , with a considerable advan
tage for White.
1 3 ... Be6 1 4.3 Re8 1 5 .Qd l Nxd2

Superior was I S . . . Nd6 to be followed by 2 6 . . . Nf7 . The exchange of White's

inactive Knight is not a good idea.
1 6.Qxd2 Qh6 1 7.Nd3

The Knight is headed for cS to prevent Black from exchanging all the
heavy pieces on the c-file.
17 ... Nd7 1 8.e3 g5 19.Ba3 Bf6 20.f4 g4 2 1 .Ne5

After only a few inaccuracies by Black, he has incurred a difficult position.

White is threatening 2 2 .Nxd7 followed by 2 3 .BxdS+. The annoying Knight
cannot be exchanged because of the loss of Black's a-pawn: 2 1 . . . NxcS
2 2 .BxcS Qa6 2 3 .Bfl bS 24.a4.
2 1 . ..Rfd8

Now I'm winning material by force. The question is, what's the best way
to do this? In general, a player wants to choose a line that gives his oppo
nent the least amount of counterplay, and with this in mind, the decision I
make becomes much easier to understand.
22 .Bh4!

Very tempting was 2 2 .Nxd7 Bxd7 (2 2 . . . Rxc l + is also possible when 2 3 .Rxc l
Rxd7 2 4.BcS Qd8 2 S .Bxa7 picks up a pawn but allows Black to create Bishops

319

PAL B E N KO: MY LI FE, GAM ES AN D C O M POSITI O N S

320

o f opposite colors by 2 5 . . .b 6 26.Rc6 Rxa7 2 7 .Rxe6 Kf7 2 8 . Rc6 e6) 2 3 .Bxd5+,

but 2 3 . . . e6 24.Bg2 Bc6 2 5 .Bc5 Qa6 26.Bfl Bb5 leaves White weak on the
light squares (this is hardly fatal, but why leave myself with any sort of
weaknesses at all?). So 2 2 .Nxd7 is obviously good for White, but the move
I chose wins material and keeps Black passively placed.
22 ... Qe6 2 3 .Na4

After a good deal of thought, I finally made up my mind to win a pawn,

but not the Queen. What would you have done? On 2 3 .Ne4, I felt that
2 3 . . .dxe4 24.Rxc6 Rxc6 would offer Black real drawing chances, especially
if he can settle his Knight on d 5 . Of course, such a decision is motivated
in part by the fact that I had lost in the two previous rounds (hang of a
Rook vs. Dominguez and hanging several pieces to Portisch). In other words,
I didn't want this guy to get away!
2 3 .. .Qa6 24.Bfl b5 2 5 .Nc3 Qb7 26.Nxb5

So I finally got my extra pawn, and Black's position lacks any dynamic
potential. Now I can calmly go about "milking the cow. "
26 ... a6 2 7.Na3 NbS 2 S .Ba5 Rd6 2 9.RxeS+ BxeS 3 0.Rc l Ne6 3 1 .Bc3 h5
3 2 .Ne2 h4 3 3 .Nb4 Qa7 34.Nd3 hxg3 3 5 .hxg3 Bg7 3 6.Ne5 Qb7 3 7.Nxe6
Rxe6 3 S.Ba5 Bd7 3 9.Rxe6 Bxe6 40.Qb4 Kfi 4 1 .Qxb7 Bxb7 42 .a4 e6

The game was adjourned here. I probably could have played better during
the last several moves, but such is life when you're addicted to time pres
sure. For this reason I was not aiming to win more material, but only to
preserve my extra pawn. It is not easy to win this ending as the position is
closed; without the penetration of White's King into Black's position, no
winning plan can be imagined.
43 .Kf2

My King starts its journey to the queenside. Note that, as outlined by my

earlier plan, Black is still completely passive.
43 . . .KeS 44.Ke l Kd7 45.Kd2 Ke6 46.b4 BeS 47.Kc3 Kb7 4S.b5 Bf6
49.Kb3

Capturing the Black a-pawn doesn't help my King penetrate.

49 ... axb5 50.Bxb5

E U RO P E AT MY F E ET

Taking with the pawn would be an irreparable mistake, eliminating all win
ning chances. This is a critical moment. White's Bishop stands very well
on b 5 . The opponent's Bishop cannot move because Bd7 is always threatened. However, White always has to watch out for Black tricks based on
. . . Bh4! .
50 ... Be7 5 1 .Bb4 Bf6 52 .a5 Ke7 5 3 .Ka4 Bb7 54.a6 BaS 5 5 .Ba5+ Kd6

Somewhat longer resistance was possible by 5 5 . . . Kc8, but there would have
followed 5 6 .Be8 and White's King would finally have been able to pen
etrate.
56.Bb6 Bg7 57.BdS ! , 1 -0.

Freezing his King and ending his tactical ... Bh4 hopes. Now it's no longer
possible to keep White's King from entering via as and b6.

( 1 07) Benko - Pomar (Las Pal mas, 1 972)

l .e4 Nf6 2 .Nf3 e6 3 .g3 d5 4.Bg2 e5 5.0-0 Ne6 6.exd5 exd5 7.d4

After skirting through several possible openings, a Tarrasch Defense fi

nally appears on the board.
7 ... Be6

A rare move. Usual is 7 . . . Be7.

S.b3 B e 7 9.Bb2 0 - 0 1O.Nc3 Ne4

A known maneuver, usual when White has fianchettoed his dark-squared

Bishop. The idea is to play . . . Bf6, neutralizing the long black diagonal.
1 1 .dxe5 Bxe5 1 2 .e3 Be7 1 3 .Ne2 ReS 1 4.Nf4

White's control over d4 and his pressure against d5 and e6 give him a
comfortable edge.
14 ... Qd6 1 5 .Qe2 Nb4

Black tries to exploit the illusory weak points on the c-file, which he occu
pies. But this is a premature attacking attempt.
1 6.Nd4 Bf6 1 7 .Rad l !

After long thought, I decided this was the only move to prevent Black from
obtaining a satisfactory game. Now White threatens 1 8 . Bxe4 dxe4

32 1

PAL B E N KO : MY LIFE, GAM ES AND C O M POSITI O N S

1 9.Ndxe6, taking advantage o f the poor position o f Black's Queen. Black

can't play 1 7 . . . Nc3 1 8 .Bxc3 Rxc3 due to 1 9 .Nb S , while the tempting
1 7 . . . Bxd4 1 8.Bxd4 Rc2 1 9.QhS gives White a very strong kingside attack.
322

1 7 ... Qe7?

As we shall see, the Queen stands badly here. Perhaps 17 ... Qb6 was bet
ter, but Black would have a difficult position anyway after the simple 1 8 .a3 .
l S.Bxe4 dxe4 1 9.Ba3 as 20.Nfxe6 fxe6 2 1 .Qg4

After this Queen move, eyeing the two weak e-pawns, Black is already
lost. He now tries his best tactical chance.
2 1 . . .Bxd4

Another try, 2 1 . . .eS , would be answered by 2 2 .NfS Qc7 2 3 .Rd7, when fur
ther resistance is futile.
22 .Rxd4 Rc2 2 3 .Bxb4 axb4 24.Rxe4 Rf6 2 S .Rxb4 Rxa2 26.Ra4 Rb2

Black doesn't wish to trade his only active piece, even if this means the
loss of another pawn.
27.RaS+ Rf8 2S.Rxf8+ Kxf8 29.Qf4+

A typical time pressure inaccuracy (to borrow a line from Fischer: "Patzer
sees check, patzer gives check! "). Correct was 29.Rc l .
2 9 ... Qf6 !

Black offers a pawn in order to prevent White's Rook from becoming

active.
3 0.QbS+ Kf7 3 1 .Qxb7+ KgS 3 2 .b4 hS B .h4 QfS 34.QcS+ Kh7 3 S .Qc3
Rc2 3 6.Qb3 ?

White, by sticking his Queen on the side of the board, breaks one of the
most useful rules in chess: When you have a decisive positional or mate
rial advantage, make sure your opponent's counterplay is kept to a mini
mum (When he's already dead, why allow him to reanimate and crawl out
of his grave?). Correct was 3 6. Qd4, when 3 6 . . . eS 3 7 .QdS would centralize
the Queen and end Black's hopes.
3 6 ... gS !

Not what I wanted to see in time pressure ! Suddenly my easy path to vic
tory has a few thorns on it.

E U RO P E AT MY F E ET

3 7.hxgS h4 3 8.e4!

The only counter, eliminating Black's dangerous kingside play.

3 8 ... Qxe4 3 9.Rd l h3

In view of his opponent's time shortage, Black should have considered

39 . . . Kg6 or . . . Kg7 or even 39 . . . Rc4, though White must win anyway be
cause of the exposed position of the black King.
40.Qd3 h2 + 41 .Kxh2 Rxf2+ 42 .Kh3 Qf5+

On 42 . . . Qxd3 43 .Rxd3 Rb2 44.Rd4 Kg6 45 .Kg4 and Kf4 wins easily (i. e . ,
4 5 . . . e5 46.Rd6+).
43.Qxf5 + Rxf5 44.Rb l Rb5 45.Kg4 Kg6 46.Kf4 Kf7 47.Rb3 Ke7

If 47 . . . Kg6, White wins by 4S .Ke4 Kxg5 49.Kd4 Kf5 50.Kc4 RbS 5 1 .b5
RcS+ 5 2 .Kd 3 RgS 5 3 .Ke2 RbS 54.b6 Rb7 5 5 .Ke3 Ke5 5 6.Rb5+.
48.g6 Rf5 +

Or 4S . . . Kf6 49.Ke4, and the white King will penetrate Black's queenside.
49.Kg4 Rb5 50.RB Rxb4+ 5 1 .Kg5 Rb8 52 .g7 e5 5 3 .Kg6, 1-0 .

( 1 08) Benko - Planinc (Novi Sad, 1 972)

l .e4 g6 2 .d4 Bg7 3 .Nfl d6 4.c3 e6? !

The opening started as a Robatsch Defense, but this move (preparing to

place the Knight on e7) does not fit into that system, 4 . . .Nf6 being more
usual. It is difficult to determine the origin of the present line. The Czech
international master Uj elky adopted it frequently for many years.
S .Bd3 Ne7 6.a4 ! ?

Maybe this looks a little mysterious, s o I will explain it. My only purpose
was to interfere with Black's development. You see, I had noticed in the
games of many Yugoslav players (particularly speed games) that they like
to fianchetto the light-squared Bishop in similar positions. For this rea
son I planned to answer . . . b7-b6 with a4-a5 .
6 ... Nd7 7.0-0 a5 8.Na3 0-0 9.Nb5 !

323

PAL B E N KO : MY L I F E , GAM ES AND COM POSITI O N S

324

Black has succeeded i n preventing White's a4-a5, but White has been able
to plant an annoying Knight on b5 which Black will be unable to drive
away for another seventeen moves, and even then without satisfactory re
sults. Therefore, Black should have considered . . . c7 -c6 at some earlier point
to prevent this Knight move, but then it would have come into play via c4.
9 ... e5

My opponent has decided not to fianchetto his light-squared Bishop by

. . . b7-b6, on which I intended 1 0.Bg5 Re8 I l .Qd2 . The text means a loss
of a tempo, while at the same time White's setup with a4 and Nb5 is well
suited to Black's pawn position.
1 0.dxe5 dxe5

Black now gets into a very cramped position, but he hardly had a choice
since after 1 O . . . Nxe5 I l .Nxe5 Bxe 5 , there would come 1 2 .Bh6, 1 2 .Bg5 , or
the immediate 1 2 .f4 followed by 1 3 .f5 . In every case White would have
excellent prospects.
I 1 .Be3 h6 1 2 .Qb3 b6 1 3 .Rfd l Rb8 1 4.Qc4 Rb7

Black's game is very passive, and he can hold his position together only
through the most cumbersome maneuvers. Naturally, 14 . . . c6 would have
been strongly met by 1 5 .Nd6.
1 5 .b4

I decided to open files on the queenside, but doubling Rooks on the d-file
via 1 5 .Rd2 was also very tempting.
1 5 ... Nb8 1 6.Be2 Bd7 1 7.bxa5 bxa5 1 8 .Bc5 Kh7 1 9.Rd2 Qe8 20.Qa2

Directed against . . . B e 6 . Now Black seizes his only opportunity for

counterplay.
20 ... f5 2 1 .Bd3 Nbc6 2 2 .Re 1 f4 2 3 .Bc4 g5 24.Be6 Bxe6 2 5 . Qxe6

White has managed to prevent Black's counterattack by fairly simple means.

Now Rd7 is threatened.
25 ... Ng6

This leads to the loss of material, but it is difficult to recommend any

thing better. If 2 5 . . . Qc8 White wins by 2 6 . Qxc8 (or 2 6.Rd7 Rf6 2 7 . Qg4)

E U RO P E AT MY F E ET

2 6 . . . RxcS 2 7 .Bxe7 Nxe7 2 S .Rd7 Ng6 2 9.Nd6 cxd6 3 0.Rxb7 Rxc3 3 1 .Rb S .
Also hopeless i s 2 S . . . Rf6, because o f 2 6.Qd7 .
26.Qh3 ! Rf6 2 7.NxgS + KgS 2S .Nf3

White has not only won a pawn, he's won the g-pawn, the lack of which
upsets all of Black's plans for a kingside attack.
2S ...NfS 29.Bxf8 BxfS 30.Qg4+ KhS

I would have met 3 0 . . . Rg6 with 3 1 .Qd7.

3 1 .Nh4 Ne7 3 2 .Red l Qf7 H .RdS hS

The active-looking B . . . Qc4 suffers a meltdown after 3 4.R l d7 Qxe4 3 S .h3

Qh7 3 6.RxfS+ RxfS 3 7 .Rxe7 .
34.QgS Kh7

Black might have quietly resigned, but he's hoping

that the mutual time shortage will lead to a White
blunder. For example, 3 S . QxeS would give Black
counterplay that he doesn't deserve after 3 S . . . Nc6
3 6.QeS NxdS 3 7 .Rd7 c6 3 S .Rxf7+ Rfxf7 3 9.h3 cxbS
40. QxdS bxa4.
3 S .Nfl Bh6 36.Qh4 c6

Too late ! My Knight jumps from one fine square

to an even better one.
3 7.Nd6 Ng6 3 S.Qh3

Of course, 3 S .QxhS was extremely strong, but. . . time

trouble !
3 S ... Rxd6 3 9.RSxd6 Re7 40.QxhS Kg7 4 1 .QfS

(Photo courtesy USCF.)

Nicer was 4 1 .NxeS ! or 4 1 .Nh4 ! , but neither player

saw anything in the time scramble.
41. .. QxfS 42 .exfS , 1 -0.

The time control has been reached, and Black was able to calm down and
accept defeat.

( 1 09) Benko - Em ma (Skopje Olympiad, I 972)

l .c4 Nc6 2.g3 eS

After a little extravagance, we have arrived at the Closed Variation of the

English, one of Black's most aggressive choices. Actually, it is a Closed
Sicilian Reversed with Black a tempo behind.
3 .Bg2 g6 4.Nc3 Bg7 S.e3 d6 6.Nge2 Be6! ?

Both 6 . . . Nh6 and 6 . . . Nge7 are now more highly thought of than the ven
turesome text.
7.d4

325

PAL B E N KO : MY LI FE, GAM ES AND C O M POSITI O N S

326

Instead o f this, I played 7 .Nd5 twice against Hort, but I did not succeed
in reaching an opening advantage after 7 . . . Nce7 ! . This is actually a pawn
sacrifice, though its acceptance seems to be dangerous: 8 .Nxe7 Nxe7 9.Bxb7
Rb8 1 O.Bg2 (Not 1 O. Qa4+ ? ? Bd7 I 1 .Qxa7 Rxb7 1 2 . Qxb7 Bc6) 1 O . . . Bxc4
I 1 . Qa4+ Bb5 1 2 . Qxa7 c5 1 3 .Qa3 ( 1 3 .Nd Nc6 1 4.Qa3 Nb4 is just as de
pressing) 1 3 . . . Nc6 and White's position is awful.
7 ... exd4

Also possible is 7 . . . Bxc4? ! . The line is given by Pachman up to 8.d5 Bxe2

9.Qxe2 Nb8 1 O.Qb5+ Nd7 I 1 . Qxb7, when he claims White is better. But
this judgment was not confirmed by the game Larsen-Suttles, Palma de
Mallorca 1 970, which continued 1 1 . . . Rb8 (White's better pawn structure
and two Bishops give him the edge after 1 1 . . .a6 1 2 . Qc6 Ne7 1 3 . Qc4, Suba
Karlsson, Las Palm as 1 982) 1 2 . Qa6 Nc5 1 3 . Qe2 f5 1 4.e4 Nf6 1 5 .exf5 gxf5
1 6.Bh3 0-0 1 7 .0-0 Nfd7, with a complicated battle that was eventually
drawn. The flaw with this whole line was only demonstrated some years
later: 1 2 . Qxa7 ! Nc5 1 3 . Qa5 f5 (or 1 3 . . . Nd3 + 1 4.Ke2 [ 1 4.Kd2 ! ? might be
even stronger] 1 4 . . . Nxc 1 + 1 5 . Rhxc 1 Rxb2 + 1 6.Kfl Nf6 1 7 .Rab l Rxb l
1 8 .Rxb l 0-0 1 9.Rb7 Ne8 2 0.Nb 5 , with White owning an obvious superi
ority in Rind-Morrison, Manchester 1 980) 1 4.0-0 e4 1 5 .Nb5 Ne7 1 6.Rd l
0-0 1 7 .Bft Rb7 1 8 .Rb l Qb8 1 9. a4 Kh8 2 0.b3 Ng8 2 1 .Ba3 Nf6 2 2 .Bxc5
dxc5 2 3 .Bc4 Ng4 2 4.d6, and White won in a few more moves, King-Wolff,
NY 1 990.
8.Nxd4

Pachman gives this move an exclamation mark. However, it is interesting

to mention that in his earlier theoretical works he had recommended 8.exd4
as a good continuation, but after my present opponent defeated him in
Mar del Plata 1 959, playing 8 . . . Bxc4 ! , Pachman was forced to change his
mind.
8 ...Nxd4

In the game Mikenas-Podgaets, USSR Championship 1 970, White got a

nice edge after 8 . . . Bd7 9.0-0 Nge7 1 O.Nde2 ! 0-0 l 1 .b3 a6 1 2 .Bb2 b5 1 3 .cxb5
axb5 1 4.Qd2 b4 1 5 .Ne4.
9.exd4 Ne7 ! ?

E U RO P E AT MY F E ET

This is a new move. In Quinteros-Hort, Vincovci 1 97 2 , there occurred

9 . . . Qd7 1 0.0-0 ! ( l O .Bxb 7 RbS 1 1 .Bd5 c6 1 2 .Bxe6 Qxe6+ 1 3 . Qe2 Bxd4, =)
l O . . .Ne7 (if l O . . . Bxc4 1 1 .Re 1 + Ne7 1 2 .Bxb7 RabS 1 3 .Bd5, White must
win) 1 1 .Re l 0-0 1 2 .d 5 Bf5 1 3 .h3 QdS? 1 4.Bg5 , and White skillfully exploited his better position. 1 3 . . . hS was recommended as an improvement
for Black but White still has the better position-he is not forced to enter
the complications arising after 1 4.g4.
1 0.dS

A new move that tests Black's 9th. Instead, 1 0 . Bxb 7 is unclear after
lO . . . O-O ! , the move my opponent later said he was planning. I admit I would
not have dared to accept the Exchange sacrifice: 1 1 .BxaS QxaS 1 2 .d S Bh3 ,
and it is difficult to assess the chances without having analyzed this posi
tion ahead of time.
1 0 ...Bd7 1 1 .0-0 0-0 1 2 .Ne4!

While playing d 5 , I counted on this Knight move. Of course, now 1 2 .. .fS

is premature because of 1 3 .N gS and Black's e6 is seriously weakened, while
1 2 . . .NfS 1 3 .BgS would be unpleasant.
12 ... h6 1 3 .Bd2 fS

Of course, 1 3 . . . Bxb2 ? 1 4.Bxh6 is bad for Black.

1 4.Nc3 gS ! ?

A committing move, but White's advantage i s obvious anyway because of

the bad position of Black's Knight.
I S .Ne2

A nice move that keeps the enemy Bishop off of d4, prepares f2 -f4, and
also envisions an eventual migration to e6 via Ne2 -d4-e6.
I S f4! ?
.

Black sacrifices a pawn to get active piece play. On I S . . . Ng6 ( I S . . . Bxb2

1 6.Rb 1 is nice for White), I intended Qc2 followed by f2 -f4 with the bet
ter chances.
1 6.gxf4 NfS 1 7.fxgS hxgS 1 8.Ng3

Now I S . . . Nh4 1 9.QhS is strong for White.

1 8 ... g4! ?

327

Battling Rakic in Novi Sad, 1 972.

Perhaps the best move under the circumstances, preventing the Queen
attack by Qh5 . The position is very complicated and, at the time, it was
not easy to judge whether Black's activity justified his pawn sacrifice. Choos
ing between the capture of the second pawn by 1 9. Qxg4 Ne3 2 0 . Qg6 Rf6
2 1 .Qh5 Nxfl n .Rxfl (which has to be good for White), and the move I
actually played, was difficult. My opponent, seeing me lost in thought,
took the opportunity to offer a draw. I communicated the offer to my team
captain in accordance with the correct procedure in team competition,
adding that I was not sure if ! had the better game. He replied, "You should
continue playing because you're the better player. " But in so doing I strove
to simplify the game so I could play for the win while retaining a draw in
hand.
1 9.Nxf5 Bxf5 20.Bc3 Rf7? !

I must admit I was afraid o f the exchange o n c3 , a s i n that case i t would be

very difficult to exploit the extra pawn because of the weakness on c4. But
Black wanted to gain a tempo for the kingside attack.
2 1 .Bxg7 Rxg7 22 .Rel Qh4 n.Qb3 Rf8 24.Qg3 Qf6 2 5 .Re2 Qd4 26.Rc 1

White has repulsed Black's attacking attempts. O n 2 6 . . . Bd3 I was plan

ning 2 7 . Qe3 ! , avoiding unnecessary complications that could arise after
2 7 . Rd2 Rf3 (though even here, 2 S . Qh4 favors White).
26 ... Bd7 27.Qe3 Qf6 2 8.c5 !

Instead of snatching pawns, White prefers counterplay. On 2 S . . . ReS, White's

plan was 2 9.Qc3 ! when Black's attack would have come to a standstill.
2 8 ... Bb5 ? ! 29.Qe6+ Qxe6 30.Rxe6 Bd7 3 1 .Re2 Bb5 3 2 .Re4 Bd3 3 3 .Re3
Bf5 34.cxd6 cxd6 3 5 .Be4 Kh7

E U RO P E AT MY F E ET

On 3 S . . . Bxe4 3 6.Rxe4 RfS 3 7 . Rcc4 RxdS 3 8 .Rxg4, White wins the Rook
endgame thanks to his two passed pawns .
3 6.Rc8

A little trick in mutual time trouble.

3 6 ... Rf6 3 7.Rc4

Also strong was 3 7 .Re8.

3 7 ... Kg6 3 8.Kg2 Kg5 3 9.Kg3 Bd7?

A mistake in a lost position.

40.Rc7! Rgf7 41 .Rxb7, 1 -0.

White had sealed his 4 1 st move but Black later resigned. Of course, 41 ... Rxf2
is not playable because of 42 . Rxd7 .

( I 1 0) Benko - Malich (Skopje Olympiad, 1 972)

l .g3 e5 2.c4 Nf6 3 .Bg2 Nc6 4.NfJ Bc5

White left his other Knight on bl so as to avoid the fashionable 4 . . . Bb4

line. Of course, 4 . . . e4 would have been met by S .NgS .
5.Nc3

The . . . BcS line is often met by the tactical shot NxeS . However, here S .NxeS
isn't as good as usual since White has to deal with the hanging c-pawn
after S . . . Bxf2 + 6.Kxf2 NxeS .
5 . . .d6 6.d3 h6

In the game Benko-Rossolimo, Novi Sad 1 972 (held just after the Olym
piad), Black tried 6 . . . a S , doubtless to preserve his dark-squared Bishop,
though I never intended to take it! The game continued: 7 . 0-0 h6
(7 . . . 0-0 8 .BgS h6 9.Bxf6 Qxf6 1 0.Ne4 Qe7 1 1 .NxcS dxcS 1 2 .Nd2 is a bit
better for White) 8 . e 3 0-0 9.b3 Bg4? ! 1 0.h3 BfS I l .Bb2 Re8 1 2 .a3 Nh7
(It is difficult for Black to find counterplay because White is threatening
to obtain an advantage in the center in several different ways.) B .NdS
NgS 1 4.NxgS hxgS I S .Kh2 Qd7 1 6 .Qd2 Ba7 1 7 .g4 Be6 1 8 .b4 Bb8 1 9 .bS
Nd8 2 0 . Kh l ! BxdS 2 1 .cxd S ! QxbS 2 2 . f4 gxf4 2 3 . exf4 f6 2 4. Rab l Qa4
2 S .gS ! Ba7 2 6 . Rbe l Bd4 2 7 .Bxd4 Qxd4 2 8 . Re4, and White had a win
ning attack.
7.0-0 0-0 8.Na4? !

This move was made only to intimidate the opponent, as the carrying out
of the threat to take the Bishop would only facilitate Black's defense: he
could prepare the . . . eS -e4 advance because White (after NxcS dxc S , open
ing the d-file for Black) is unable to move his e-pawn due to the weaken
ing of d3 . A better plan is 8.a3 as 9.e3 followed by b 3 , Bb2 , and eventually
d3 -d4.
8 ... Qe7 9.e3 a5

329

PAL B E N KO: MY L I F E , GAM ES A N D C O M POSITI O N S

I was afraid of . . . a6 instead, which might have served t o prepare a later

. . . b7-b S . Having failed to make use of this equalizing opportunity, Black
now drifts into an inferior position.
330

1 0.b3 Re8 1 1 .Bb2 Bf5 1 2 .a3 Ba7 1 3 .Nc3

The Knight wasn't doing anything on a4, so I rush it back to c3 where it

eyes the critical b S and dS squares.
13 ... Qd7 1 4.Re l

Stopping Black from exchanging the light-squared Bishops.

14 ... Bh3 1 5 .Bhl Bg4 1 6.Qc2 Nh7 1 7 .Nb5 Re7

This protects c7, thus freeing the Queen for attack by . . . QfS . But 1 7 . . . Bb6
was more suitable for this purpose. My opponent offered me a draw at
this point. I was later told that the half-point he might have gained would
have given Malich the grandmaster title. When asked if ! would have agreed
to the draw if I had known this at the time, I replied that this was a team
event and I was therefore fighting not only for myself but for my team. In
any case, I am convinced that such titles must be earned and should not
depend on the good will of one's opponents. There was already far too
much wheeling and dealing in connection with international titles in those
days. Unfortunately, such practices are still common at the present time.
1 8.h4!

Necessary so as to provide a place for the KN on h2 .

1 8 ...NfS 1 9.d4 f6

This is passive. Black should probably have tried 1 9 . . . e4 2 0.Nh2 BfS , al

though in this case White would have various possibilities such as Nc3 d S , or the advance of the d-pawn.
20.d5 Nd8 2 1 .Nh2 g5 ?

White already has the better game, but not to the extent that such desper
ate aggression is necessary. The planned kingside attack fails because of
the weakening of the light-squares. Black was obliged to play 2 1 . . .BfS n . e4
Bh7 2 3 .Nxa7 Rxa7 H.cS , with a bad but tolerable position.
2 2 .hxg5 hxg5

E U RO P E AT MY F E ET

I think 2 2 . . . fxg5 was a little better, but Black was unaware of the danger.
2 3 .Nxg4

This is even stronger than the more flashy 2 3 .Nxc7 Bf5 24.e4 Bxe4 2 5 .Qxe4
Qxc7 2 6.Ng4 Kg7 2 7 . Qf5 Rf7 .
2 3 ... Qxg4 24.Be4! Nfl 2 S .Kg2 Qd7

There is no better reply, since I threatened to win Black's Queen by Bf5 ,

and 2 5 . . . Nh6 lost to 26.Rh l Kg7 2 7 .Rxh6.
26.Rh l Nh8

This is a further loss of time, but Black's game was untenable in any case.
2 7.Rh6 Kg7 2 8.Rah l Nfl 29.R6hS Bb6 30.BfS Qe8 3 1 .Qe4 BcS

Black is in the throes of death. Mercifully, I now put him out of his misery.
3 2 .Be6! c6

Black suffers an apocalypse after 3 2 . . . Nxe6 3 3 . Rh7+ Kf8 34. Qg6.

3 3 .Rh7+ Nxh7 34.Rxh7+ Kf8 3 5 .Qg6 Rxe6 3 6.Nc7, 1 -0.

( I I I ) Benko - Bellon (Malaga , 1 973)

l .b4

A seldom-played but enterprising move, the name of which is not entirely

clear. The story goes that, during a free day at the New York 1 92 4 tourna
ment, Tartakower paid a visit to the world famous Bronx Zoo. There he
had a lengthy consultation with their prize orangutan, Susan, following
which he dedicated his next game to her. That game opened with this
same move, and Tartakower called it the Orangutan Opening. In the years
that followed, the American master Anthony Santasiere played and ana
lyzed it and, for a while, the opening was called Santasiere's Folly. Mean
while, Russian chess experts, who are fond of naming all opening systems
after a Russian, call it Sokolsky's Opening. In my opinion, it is better to
name it after the ape, adopting a certain neutrality. This is the name, inci
dentally, which is becoming more and more popular. It is true that books
have recently been published about this strange opening; Andrew Soltis
has written one, for instance. At the end of his work, he mentions that this
move could be a dangerous weapon in the hands of an expert against an
opponent who is not quite prepared for it. I confess that I have studied
none of the existing literature on this opening. In adopting the move, I
was encouraged by the sense of discovering unknown territory, a sense of
adventure. I consider myself one of those masters who lean toward ro
manticism in chess, rather than following only the beaten path. On the
other hand, I chose this move also for the sake of a joke, since my oppo
nent always adopts the first move 1 . b3 . In fact, l .b3 was sometimes shown
on the large demonstration board for Bellon's games, even before he sat
down to play, and that is how he actually opened his games with the White

33 1

PAL B E N KO : MY L I F E , GAM ES AN D COM POSITI O N S

pieces. Onlookers had to smile at my move, thinking that perhaps I was

trying improve Bellon's system by pushing the pawn a little further.
1 . ..d5 2 .Bh2 Bg4 ! ?
332

An interesting move which was popular against l .b3 a t that time. When
this game was played one couldn't find 2 . . . Bg4 in any opening books, yet
it was used in the game Quinteros-Bellon (which began with l .b3) and by
Tal at a US SR Championship. When I originally annotated this game, I
predicted that 2 . . . Bg4 would gain many adherents. Sure enough, it be
came a very common choice against 1 . b4 when a stronger player as Black
wanted to employ a solid but combative setup that would give him real
chances to outplay his opponent. The idea is to place the Bishop actively
and quickly, not allowing it to become shut out of play. If White drives it
away by 3 .h3 and later g4, his own position will be weakened.
3 .h3

The most common reply now is 3 .Nf3 , not fearing 3 . . . Bxf3 4.gxf3 .
3 ... Bh5 4.g3 e6 5 .Bg2 Nf6 6.a3 Nbd7

Black insists on the setup which has proved to be good against l .b 3 , but
to be considered here is 6 . . . cS ! ? , since after 7 . bxcS BxcS Black threatens
both . . . Bxf2 + and . . . Qb6. This maneuver might have caused White some
problems.
7.Nf3 c6 8.d3 a5 !

An unpleasant move which, if White doesn't want to destroy his pawn

structure by 9.bxa S , can hardly be met better than by White's next move.
Of course, this shuts in his dark-squared Bishop, but I was hoping that
this would turn out to be a temporary state of affairs.
9.c3 Bd6 1 O.Nbd2 Qh6 ! ?

This active move tries t o interfere with White's development. Now, of

course, 1 1 .0-0 would be a blunder because of 1 1 . . . Bxg3 . Although White
could save this pawn by playing g4 and only then 0-0, this kind of thing is
considered dubious before Black castles because of the possibility of
. . . h7-h S .
1 1 .Rh l 0-0 1 2 .g4 ! ?

Now that Black has castled, I can make this move without worrying about
. . . h7-hS .
1 2 ... Bg6 1 3 .0-0 Rad8

1 3 . . . Qa7 seems to be more accurate, taking the Queen away from its ex
posed position. For this reason, White could have considered 1 3 .c4! ? in
stead of castling. Another plan for Black is 1 3 . . . hS 1 4.gS Nh7 I S .c4, with
chances for both sides.
14.c4! Bh8

White would regain his pawn after 14 . . . axb4 I S .axb4 Bxb4 1 6 .Bc3 , with
the better game.

E U RO P E AT MY F E ET

1 5 .Qc2 Qc7 1 6.Rfc1 dxc4

In general, it is wrong to exchange a center pawn for a wing pawn, and the
present case is not an exception. However, White already stood better thanks
to his queenside play and the fact that the power of Black's lineup on the
b8-h2 diagonal is more illusory than real. The advance 1 6 . . . eS is weak due
to 1 7 .cxdS, revealing the drawback of the black Queen's new position along
the c-file.
1 7.Nxc4 axb4 1 8.axb4 Nd5 1 9.b5 !

Having succeeded in this breakthrough on the queenside, White has seized

the initiative. White should not fear the taking (rather, the trading) of this
pawn, since both of Black's b-pawns would become weak and White would,
while Black strives to defend them, gain superiority in the center.
1 9 ... f6

Black is afraid of White's occupation of e S . 1 9 . . . Nf4 would have been met

by 2 0.NceS NxeS (White also retains all the chances after 20 . . . cS 2 1 .Bfl )
2 1 .BxeS Q a S 2 2 .Bxb8 Rxb8 2 3 .bxc6 bxc6 24.Rxb8 Rxb8 2 S .Bfl Rc8 2 6 . e 3
N d S 2 7 .QcS , when White's superior pawn structure gives him a serious
advantage.
20.Ba3 c5 2 1 .b6!

Not permitting Black to close the queenside by ... b7-b6.

2 1 .. .Qf4? !

This impatient move is characteristic of young players who do not admit

strategic defeat, preferring to make active moves in any situation. It was
high time to start a passive defense by 2 1 . . .Qc8.
22 .Qd2

Stronger was 2 2 .e3 .

2 2 ... Rc8

Black would have had fewer difficulties if he traded Queens and then oc
cupied f4 with his Knight.
2 3 .e3 !

Removing the black Queen from the scene of action, at least for a while.

333

PAL BEN KO: MY LI FE, GAM ES AN D C O M POSITI O N S

2 3 ... Qh6 24.Na5 Rf7

334

It is not possible to avoid the loss of a pawn, In addition to this annoy

ance, there is no way to obtain any serious counterplay since . . .f6-f5 will
always be met by g4-g5 . If Black had tried 24 . . . Rfe8, I would have played
2 5 .Rb3 ! and not 2 5 .Nxb7 ? ! , when Black charges back with 2 5 . . . c4! .
2 5 .Nxb7 N7xb6 26.Nxc5 Ra7 27.Ra l

It was tempting to play 2 7 .Nxe6, since 2 7 . . . Rxc 1 + 2 8 .Bxc 1 ! , or 2 7 . . . Rxa3

2 8 .Rxc8+ are good for White. However, 2 7 . . . Re8 would introduce some
complications, although 2 8 .Bc5 Rxe6 2 9.Nd4 would lead to a winning
position. Nevertheless, I was of the opinion that it was not necessary to
gain more material. I already had an extra pawn, so why court danger (and
a possible miscalculation) when there was no reason to do so?
27 ... Bd6 2 8.d4 Rca8 29.Bb2 Bf7

At last a defending move ! 29 . . . Nc4 would lose on the spot to 3 0. Rxa7 !

Nxd2 3 1 . Rxa8+ Kf7 3 2 .Nxd2 Nxe3 3 3 .Ra7+ Be7 3 4.Ba3 . However, 2 9 . . . Ra2
should have been tried, when I would have played Qe2 .
3 0.Rxa7 Rxa7 3 1 .Qd3

Getting out of the possible pin and prevent

ing Black's Queen from becoming active
on g6.
3 1 . . . Qg6 3 2 . Qxg6 hxg6 B.Ne l Ra2
3 4.Ned3 Nb4

Black last attempt, but White finds the right

way to exchange pieces.
3 5 .Bc3 Nxd3 3 6 .Nxd3 Ba3 3 7 .Rb l Rc2
3 8.Bb4 Bxb4 3 9.Rxb4 Nd5 , 1 -0.

Black lost on time, saving me the trouble

of performing a not too difficult technical
task. For instance, 40.Rb8+ Kh7 4 1 .Bxd5
exd 5 42 . Rf8 Rc7 43 . g5 ! , and B lack can
hardly move. If 43 ... fxg5 44.Ne5 Be6 45 .Nf3
Bf5 46.Nxg5+ Kh6 47 .Nf7+, it's all over.

Arturo Pomar:
(Photo courtesy USeF.)

E U RO P E AT MY F E ET

( I 1 2) Benko - Pomar (Orense, 1 973)

1 .NB d5 2 .g3 Nf6 3 . Bg2 c5 4.c4 Nc6

He avoids 4 . . . d4, when 5 .b4! ? is possible, creating a Benko Gambit

reversed !
5.cxd5 Nxd5 6.Nc3 Nc7

Black reveals his intentions: he intends to set up a Maroczy Bind, named

after the late Hungarian grandmaster, by . . . e7-e 5 . This preparatory move
is necessary since the immediate 6 . . . e5 is a blunder because of 7 .Nxe5 .
The c 7 -square i s a good one for the Knight, since i t can eventually go to
e6 where it eyes d4.
7.d3 e5 8.Nd2

Modern opening theory has already worked out several methods to fight
against this bind. One of them consists of attacking the c5 -square by a3 and
b4, another aims to undermine the black e-pawn by f4. The text move pre
pares Nc4, adding pressure against e 5 . At the same time, S .Nd2 opens the
diagonal of the g2 -Bishop and threatens Bxc6, destroying Black's queenside.
8 ... Bd7 9.0-0 Be7

A natural developing move. 9 .. .f6, which occurred in Gheorghiu-Padevsky

a few rounds later, isn't as logical.
1 O.Nc4 f6

Black's other choice is W . . . O-O ! ? 1 1 . Bxc6 Bxc6 1 2 .Nxe5 BeS, when White's
loss of his light-squared Bishop gives Black compensation for the sacri
ficed pawn.
l 1 .f4 b5 1 2 .Ne3

1 2 .Bxc6 appears to give White time to weaken Black's e-pawn, but it

doesn't lead to anything after 12 . . . Bxc6 1 3 .Na 5 Bd7 1 4.fxe5 Ne6 1 5 .Nb7
Qb6 1 6 .Nd6+ Bxd6 1 7 . exd6 Qxd6 I S .Ne4 Q e 7 Ree-Polugaevsky,
Amsterdam 1 97 2 .
1 2 ... 0-0
An inaccuracy that gives White the advantage. It's now known that 12 . . . exf4

is Black's best move, when both sides have chances after 1 3 .gxf4.
1 3 .fxe5

Also, 1 3 .f5 offers White good prospects. Now, for a moment, I thought
my opponent was going to recapture with the Knight, sacrificing the Ex
change. This isn't sound (after 1 3 . . . Nxe5 1 4.BxaS QxaS 1 5 .Nf5), although
it would no doubt have offered some tactical chances.
13 ... fxe5 14.Ned5

White must avoid 1 4.Nxb 5 ? due to 14 . . .Nxb5 1 5 .Qb3 + KhS 1 6.RxfS+ BxfS
1 7 . Qxb5 Nd4, when his "cleverness" has backfired.
14 ... Rb8 1 5 .Bd2

335

PAL B E N KO: MY LI FE, GAM ES A N D COM POSITI O N S

336

I did not like the more active I S .Be3 , because after ... Nd4 there is no e2 e3 available to drive the Knight back. White's game is obviously better
here: Black's queenside pawns may prove to be weak and it will be pos
sible for White to post a piece advantageously in front of Black's isolated
e-pawn. Therefore, my opponent begins a trading policy which, to a point,
can't be criticized.
1S ... NxdS 1 6.NxdS Rxf1+ 1 7.Qxf1 BgS ?

Too much is as bad as nothing at all. This attempt at further trades makes
Black's situation rather difficult. Correct was 1 7 . . . Bd6 followed by l S . . . Ne7 ,
ousting White's dominating Knight from dS .
18.BxgS QxgS 19.Qf2 !

It is probable that my opponent expected me to offer the trade of queens

with 1 9. Qc 1 , when 1 9 . . . Qxc 1 + 2 0 . Rxc 1 c4 2 1 . dxc4 bxc4 2 2 .Rxc4 Rxb2
2 3 .Rxc6 Bxc6 2 4.Ne7+ KfS 2 S .Nxc6 Rxa2 would give Black the better
chances despite White's two pieces for the Rook. The text move ( 1 9. Qf2)
wins an important tempo for the attack on the f-file.
1 9 . . .Nd4

Trying for a counterattack by 1 9 . . .RfS fails: 1 9 . . . RfS 20.QxcS Qd2 (2 0 . . . Nd4

loses instantly to 2 1 .Ne7+ KhS 2 2 .Ng6+) 2 1 .Nc7 ! (A surprise that stops
B lack from defending with . . . B e S , and simultaneously threatens both
2 2 .BdS+ and 2 2 .Bxc6. Not as good is 2 1 .Ne7+ Nxe7 2 2 .Qxe7 Qe3 + 2 3 .Kh l
BeS 24.BdS+ B f7 2 S .Be4 Qxe2 2 6 . QxeS , when White i s much better, but
Black can still resist.) 2 1 . . .Nb4 2 2 .a3 Qxb2 2 3 .Rc 1 , and White wins.
20.Rf1 Qd8

Black could find nothing better than this retreat, by which he defends against
the threat of 2 1 . Qf7 + followed by QfS+ or Qxd 7 . The alternative 2 0 . . . Be6
would have been met by 2 1 . e3 NfS (or 2 1 . . .Nc6 2 2 .Nc7 Qe7 2 3 .Nxe6
Qxe6 24.Bxc6, and wins.) 2 2 .Nc7 Bd7 2 3 .BdS+ KhS 24. e4, and Black has
to give up.
2 1.Qe3 !

E U RO P E AT MY F E ET

337

Stronger than 2 1 . Qf7+ KhS 2 2 .e3 ! (and not 2 2 .Be4 Nxe2 + 2 3 .Kh l QeS,
when Black is okay.) 2 2 ... Ne6 (2 2 ... Be6 2 3 .Qxa7) 2 3 .RfS which, of course,
would also have been in my favor.
2 1 . . .Qe8 22 .QgS h6?

A serious blunder, but Black is not to be blamed because the game is lost
in any case. On 22 . . . KhS (2 2 .. .Nxe2 + 2 3 .Kh l KhS 2 4.Nc7 transposes)
2 3 .Nc7 QdS 24.QxeS Nxe2+ 2 S .Kh I Nd4 26.Rf7 NfS 2 7 .Be4 QgS 2 S .Rxd7
Qc 1 + 29.Kg2 , Black's checks will eventually run out and White will win.
2 3 .Nf6+ Kh8 24.Nxe8 hxgS 2 S .Rf8+ Kh7 26.Be4+ g6 2 7.Nf6+

Black had obviously failed to realize that his Rook was undefended. The
win is now only a matter of technique.
27 ... Kg7 28.Rxb8 Kx:f6 29.e3 Nc2 30.m BfS 3 1 .RxbS Nb4 3 2 .Ke2 Nxa2
3 3 .RxcS Nb4 34.Rc4 as 3 S.RcS Bg4+ 3 6.Kd2 BfS 3 7.RxaS g4 38.RbS, 1 -0.

( I 1 3) Martz - Benko (Torremolinos, 1 973)

l .c4 eS 2 .Nc3 Nc6 3 .e3

Less usual than 3 . g3 , but not a bad move. White intends to transpose into
the Scheveningen Variation of the Sicilian with colors reversed and an
extra tempo.
3 ... Nf6 4.NB Bb4

Instead, 4 . . . dS isn't as good because of S . cxdS NxdS 6.BbS .

S.Qc2

A variation that became quite popular in the eighties and early nineties.
S ... O-O

This move allows some complicated lines, so many players simply chop
off the Knight by S . . . Bxc3 6.Qxc3 Qe7, when practice has shown that Black
has a completely adequate game.
6.d3

Black has no problems after this quiet move. Far more interesting is 6.NdS
ReS 7 . QfS d6 S .Nxf6+ gxf6 (the endgame after S . . . Qxf6 9. Qxf6 gxf6 fa
vors White) 9.QhS with sharp play.

PAL B E N KO : M Y L I F E , GAM ES AN D COM POS ITI O N S

6 ... ReS

338

It's still possible to play 6 . . . d5 , perhaps answering 7 .cxdS with 7 . . . Qxd 5 ,

though the loss o f time doesn't promise Black much here. Mter the text,
Black is threatening . . . e4 or . . . Nd4.
7.Bd2 d6

Many years later, Kasparov (against Ehlvest, Reykj avik 1 988) played
7 . . . Bxc3 and got a very comfortable position after 8.Bxc3 d5 9.cxdS NxdS
1 0.Be2 Bf5 1 1 .Rd 1 (Kasparov claims adequate compensation for Black af
ter 1 1 .e4 Nf4 1 2 .exf5 Nd4 1 3 .Bxd4 exd4 1 4.Ng l Q d 5 , but I don't know if
this has been proven in over the board play) 1 1 . . .a5 1 2 .0-0 Qe7 1 3 .a3 a4.
S.Be2 Bf5

Preparing . . . e S -e4 and grabbing the initiative. Of course, White can pre
vent this advance by 9 .e4, but then 9 . . . Bg4 would leave the d4-square weak.
9.0-0

This is obviously the most natural move, but White has also tried a couple
other ideas. The first was 9.Nd S , but in Smejkal-Portisch, Ljublj ana 1 97 3
White got nothing a t all after 9 . . . a5 1 O.Bxb4 Nxb4 1 1 .Nxb4 axb4 1 2 . 0-0
Qd7 1 3 .Nd2 Ra6 1 4.BB c5 , and a quick draw resulted. Two years later,
the game Ribli-Portisch, Ljubljana 1 97 5 saw White give 9 .Ne4 a shot.
However, this also proved unsuccessful: 9 . . . Bxe4 1 O. dxe4 Bxd2+ 1 1 . Nxd2
Qe7 1 2 . 0-0 a5 1 3 .b3 Nd8 1 4.Nb 1 Ne6 1 5 .Nc3 c6 1 6 .Rfd 1 , 112_ 112 .
9 ... e4 1O.Nd4? !

No better is 1 O . dxe4 Nxe4 1 1 .Nxe4 Bxe4 1 2 .Bd3 BxB 1 3 .gxB Qh4, when
White's King is far from happy. The safest reply seems to be 1 0 .Ne 1 , a
move that Martz used against Chellstorp (in Norristown) later that same
year: 1 0 . . . aS (or 1O . . . Bg6 1 1 .Rd l as 1 2 .Nd5 exd3 1 3 .Bxd3 Ne4 1 4.NB
Bxd2 1 5 .Nxd2 Nf6 1 6.Bxg6 hxg6 1 7 .Qb3 b6 =, Panno-Portisch, Petropolis
1 97 3 ) 1 1 .a3 Bxc3 1 2 .Bxc3 Ne5 1 3 .Bxe5 dxe 5 1 4.dxe4 Bxe4 1 5 .Nd3 c5
1 6. Qc3 , 1fz- 1h ,
1 O . . .Nxd4 1 1 .exd4 Bxc3 1 2 .Bxc3 d5

After trading Bishop for Knight, Black has seized the advantage since
White's dark-squared Bishop will be without any prospects. Black now has
a huge positional threat: 1 3" . exd3 1 4.Bxd3 Bxd3 l S .Qxd3 dxc4, leaving

E U RO P E AT MY FEET

White with an isolated d-pawn. When Black places his Knight on d 5 , the
difference between the Bishop and the Knight will be striking. Therefore,
White prevents this possibility with his next move.
n.b3 Qd7 1 4.dxe4 Nxe4 1 5 .Bd3 Nd6

Of course, Black prefers to keep his strong Knight and avoids the possi
bility of opposite-colored Bishops.
1 6.Bb4?

This is too na'ive -I had no intention of letting Martz exchange his Bishop
for the Knight. To make matters worse, the Bishop no longer defends the
d4-pawn and this will cause White to lose valuable time. He should have
tried 1 6.cxd5 Bxd3 1 7 . Qxd3 Nf5 1 8 . Qc4, when it would not be so easy to
regain the pawn under ideal circumstances, e.g. 1 8 . . . Ne7 1 9 .d6 Qxd6
2 0 . Bb4, and White gets to exchange his Bishop for Black's Knight when
the isolated pawn wouldn't be as bad as it is in the actual game (though I
would still retain a small but comfortable edge). For this reason, 1 9.d6
cxd6 (instead of 19 ... Qxd6), deserves serious consideration. Here's a rather
fanciful example of the kind of play that might result: 2 0.d5 Qf5 2 1 . Qd4
f6 2 2 . Rad l Ng6 (White's game is very unpleasant. The Knight is ready to
leap into f4, the d5 -pawn is weak, and Rook moves like . . . Re4 are loom
ing.) 2 3 . Qb4 Rac8 24.Ba l Nf4 2 5 .Qxd6 Rc2 2 6. Rde 1 Nh3 + 2 7 .Khl Nxf2 +
2 8 .Kg l Nh3 + 2 9 .Kh l Rce2 3 0. Rc 1 Rf2 3 1 . Rfd i Rxg2 , and Black wins.
16 ... Bxd3 1 7.Qxd3 Nf5 1 8 .Rad l Re4 1 9.Bc3

Admitting that his 1 6th move was a mistake.

1 9 ... Rae8 20.c5 c6

It was obvious that White's last move was a preparation for 2 1 . Rfe l . I could
have prevented this by 20 . . . Qe6, which may have been better, but I was
afraid of 2 1 .Qb5 . Anyway, I thought that White could also answer 20 . . . Qe6
with 2 1 .h3 , continuing his preparations for Rfe 1 . In the end, I judged the
position to be in my favor after 20 . . . c6, especially the coming endgame.
2 1 .Rfe l Rxe l + 22 .Rxe l Rxe 1 + 2 3 .Bxe l Qe6 24.Bc3 Qe4

White is forced to trade Queens since after 2 5 .Qd l (2 5 .Qd2 allows a check
on b l ) he would lose a pawn to 2 5 . . . Ne3 ! .
2 5 . Qxe4 dxe4

339

PAL B E N KO: MY L I F E , GAM ES A N D C O M POSITI O N S

340

This i s the position I was playing for. Black's advantage i s obvious thanks
to White's backward d-pawn, his poor Bishop and the hole on dS (which
can be used by Black's King and Knight). Because White's d-pawn is fro
zen, Black is, in effect, a pawn up on the kingside. Thanks to my four
versus-three pawn majority, I felt that White's King would be forced to
stay in that sector and perform damage control. This would allow my own
King to penetrate decisively, via the dS -square, into White's queenside.
26.g4 Ne7 2 7 .3 NdS 28.Bd2 exfl

Better than 2 S . . . e3 29.Bc1 g6 3 0 .KfI fS 3 1 .gxfS gxfS 3 2 .f4 (of course, Black
can't be allowed to play 3 2 .. .f4) 32 . . . Kf7 3 3 .Ke2 Nxf4+ 3 4.Kxe3 .
2 9.Kf2 Kf8

It might seem more natural to bring the King up by 2 9 . . .f6 and 3 0 . . . Kf7 .
However, I wanted to keep the f6-square open for my Knight s o i t could
drive the White King off of e4 via . . . Nf6+.
3 0.Kxf3 Ke7 3 1 .Ke4 Ke6 3 2 .Bel

White must also have considered both 3 2 .gS and 3 2 .BgS . On 3 2 .gS , I in
tended to play 3 2 .. .fS+, while on 3 2 .BgS I would have avoided 32 .. .Nc 3 +
3 3 .Kd 3 Nxa2 , because 3 4.Bd2 traps my Knight (later I realized that Black
wins this position after 34 . . . KdS 3 S .Kc3 Kxd4 3 6.Kb2 Kd3 3 7 .Be l Ke2
3 S .BaS b6 3 9.cxb6 axb6 40.Bxb6 Nb4, but at that time I had no interest in
going on any risky adventures). Instead, I would have answered 3 2 .BgS
with 32 ... h6, since 3 3 .Bh4 really does allow 33 ... Nc3 +. For example, 3 4.Kd3
Nxa2 3 S .B e l Nc 1 + would be winning for Black.
32 ...Nf6+ H .Kf4 KdS

The Knight and King have changed places on d S , and this lets me put
enormous pressure on White's d-pawn since my pieces can now blockade
it and attack it.
3 4.Bc3 Nd7

Intending to turn up the heat by . . . NfS-e6.

3 S .Bb2 f6 3 6.gS fxgS + 3 7.KxgS Nf8 3 8.KfS

White could have protected his d-pawn by 3 S .Kf4 Ne6+ 3 9.Ke3 . No doubt,
White considered this position to be lost. Black's plan would be pretty
straightforward: he would advance his kings ide pawns to hS and gS and
then improve the position of his Knight via . . . Ng7 -fS . A further advance
of the kingside pawns would eventually force White's King away from the
d-pawn and the feast would begin (with d4 and cS being the starters) ! There
fore, White decides to sacrifice a pawn, relying on the active position of
his King to give him some chances for salvation.
38 ... Ne6 3 9.h4 Nxd4+ 40.Bxd4

There is no other way, since 40.Kf4 g6 4 1 .KgS (4 1 .Bxd4 is like the game)
4 1 . . .Nfh 42 .Kh6 (42 .Kg4 Ke4!) 42 . . .Nxh4 43 .Kxh7 gS wins for Black.
40 ... Kxd4 4 1 .Ke6

E U RO P E AT MY F E ET

So we find ourselves in an interesting King and pawn endgame. This kind

of ending rarely occurs in modern tournaments, so I was delighted to get
the chance to play such a position.
34 1

Here the game was adjourned and I had to seal my move. A quick calcula
tion convinced me that 4 1 . . .KxcS would be a blunder: 42 .Kf7 bS 43 .Kxg7
hS 44.Kg6 Kb4 4S.KxhS cS 46.Kg6 c4 47 .bxc4 bxc4 48.hS , with a draw. I
must admit that, in the little time I had left, my first plan was 4 1 . . . Ke4,
when 42 .Kf7 KfS 43 .Kxg7 hS would be an easy win. Also winning for me
is 4 1 . .. Ke4 42 .Kd6 as (or 42 . . . hS 43 .a3 Kd4) 43 .Kc7 KdS 44.Kb6 g6 4S.a3
h6 46.b4 axb4 47 . axb4 Kc4 48.Kxb7 KbS ! . Then I noticed that 4 1 . . .Ke4
42 .h4! created some difficulties for me: 42 . . . hS ! 43 .a4! and now:

A) 43 . . .Kd4 44.Kd6 (44.Kd 7 ? as !) 44 . . .Kc4 4S .Kc7 Kxb4 46 .Kxb7 KxcS

47 .Kxa7 (47 . a S ? Kd6 ! 48 .Kxa7 Kc7 49 .Ka6 g6 ! wins) 47 . . . Kb4 48 .Kb6
Kxa4 49.Kxc6 is a draw.
B) 43 . . .Kf4 44.b S ! gS 4S .hxgS KxgS 46.b6! (46.aS ! ? a6 !) 46 . . . axb6 47 .cxb6
h4 48.aS isn't what Black wants. However, Black can improve his chances
with 46 . . . aS ! (instead of 46 . . . axb6) 47.Kd7 h4 48.Kc7 h3 49.Kxb7 h2
S O.Ka8 h l =Q S l .b7 QdS S 2 .b8=Q Qxc S , and now White can force a
draw with S 3 .Qb6! Qc3 S4.Kb7 cS S S .Kc6 c4 S6.QbS+ Kf4 S 7 .KcS =.
C) 43 . . .a6! 44.aS ! Kf4 4S .Kd7 gS 46 .hxgS KxgS 47 .Kc7 h4 48.Kxb7 h3
49.Kxa6 h2 SO.Kb7 h l =Q S l .a6, and a strange position has come about:
White's pawn only stands on the sixth rank (which usually means that
it's time to resign) but, despite this, Black has to show a bit of care if

PAL B E N KO : MY L I F E , GAM ES A N D C O M POSITI O N S

342

he wants t o reel i n the point: 5 1 . . . Qb l ? (Correct i s 5 1 . . . Qh7 + 5 2 .Kb8

Qg8+ 5 3 .Kb7 Qf7+ 54.Kb8 Qf4+ ! , and wins.) 5 2 .a7 Qxb4+ 5 3 .Kc7 (not
5 3 .Kxc6 ? ? Qe4+ followed by 54 . . . Qa8. Also bad is 5 3 .Kc8 ? ? Qxc5
54.a8=Q Qf8+ 5 5 .Kb7 Qxa8+ 5 6.Kxa8 c5 , and wins) 5 3 . . .Qa5+ 54.Kb7
and Black doesn't have time to take the pawn on c 5 .
The position after 43 . . . a6 is full o f interesting variations, e . g . , 44.a5 !
Kd4 (instead of 44 . . . Kf4) 45 .Kf5 Kc4 46.Kg6 Kxb4 47.Kxg7 ! Kxc5 ! 48.Kg6
Kb4 49.Kxh5 c5 50.Kg4 c4 5 1 .h5 c3 5 2 .h6 c2 5 3 .h7 c 1 =Q, and Black should
win. After considering all of this, I came to the conclusion that I must find
an easier solution. As it happens, there is one in the position.
41 . . . g6! !

The "Columbus' Egg," which wins the game by only one tempo. The threat
is . . . h6 and . . . g5 , so White is forced to bring his King to the kingside. Be
fore resumption of play, my opponent and I had supper together. I told him
not to spend too much time on the adjourned position because I had sealed
the best move and I didn't want him to exhaust himself in vain. When the
game resumed, this move came as an obvious surprise to him, as well as to
the spectators. Perhaps he didn't notice it, being wrapped up in the analysis
of the previous lines that centered around 4 1 . . .Ke4.
42 .Kf6 Ke4! 43 .Kg7 Kf4 44.Kh6

The more natural 44.Kxh7 would be met by 44 . . . g5 ! , transposing into the

actual game after 45 .hxg5 . Note that 44.Kxh7 g5 45.h5 fails because White
loses two moves: one move is lost getting his King out of the path of his
h-pawn, and another tempo is lost because Black gets to Queen with check:
45 . . . g4 46.h6 g3 47.Kg7 g2 48.h7 g l =Q+.
44... Kg4 4S .Kxh7 gS !

The point of Black's play. The trade of pawns is forced and Black's King
will be one square nearer the queenside pawns.
46.hxgS KxgS 47.Kg7 KfS 48.Kfi KeS 49.Ke7 KdS SO.Kd7

Also hopeless is 5 0 .b4 Kc4 5 1 .Kd6 (or 5 1 .a3 Kb 3 5 2 .Kd7 Kxa3 5 3 .Kc7
Kxh4 54.Kxb7 Kxc5) 5 1 . . .Kxh4 5 2 .Kc7 Kxc5 5 3 .Kxb7 Kb5 54.a4+ (it's ei
ther this or 5 4.Kxa7 c5) 54 . . . Kxa4 5 5 .Kxc6 as 5 6.Kc5 Kb3 .
SO . . . KxcS S 1 .Kc7 b S S2 .Kb7 as S 3 .a3

White can also resign after 5 3 .Ka6 Kb4.

S 3 . . .b4 S4.a4 Kd4 S S.Kxc6 Kc3 S6.KbS Kxb3 S7.KxaS Ka3 , 0- 1 .

E U RO P E AT MY F E ET

343
Seventeen-year-old Walter
Browne wins the U.S. Junior
Championsh i p. He went on to win
the U.s. Championsh i p six times!
(Photo courtesy USCF.)

( I 1 4) Benko - Browne (U.S. Championship. 1 973)

I've used my endgame knowledge throughout my career to win seemingly
boring games and to draw positions that appeared to be hopeless. The
position in the diagram is a case in point.

White to Move

Despite the Bishops of opposite colors, my position looks bad. I'm a pawn
down and my only remaining pawn is in danger. Can I save this game? Yes! I
had intentionally played for this position because I knew the saving solution.
46.Kg2 ! Bg1 +

This looks very good for Black since he wins quickly after 47 .KB Rf2 + or
47 .Kfl ? e2 + 48.Ke 1 Bd4 49.Rc7 Be5 . However, I had something else in mind!
47.Kxgl ! Rb 1 + 48.Kg2 e2 49.Re7 e l =Q 50.Rxe l Rxe l 5 1 .BfJ

PAL B E N KO : MY L I F E , G A M E S AN D COM POSITI O N S

344

I f you wish t o play endgames a t a high level, it's absolutely necessary to

know many key, or goal, positions (some are elementary, and others are
quite complex). This kind of knowledge often allows you to win games
that might look like certain draws, and to save games by entering seem
ingly hopeless but theoretically drawn positions. In this particular case Black
won the Exchange under the assumption that victory wasn't far off. How
ever, this is a theoretical draw because Black can't chase the Bishop from
the long diagonal. This, combined with the g3 -pawn's control over f4 and
h4, stops the black King from getting close to the pawn on g3 . I had been
playing for this position many moves before it was reached because I was
well aware that it was a dead draw. Browne, though, played on and on.
Finally I couldn't stand it anymore and I asked him, "Don't you know this
is a theoretical draw?" Browne gave me a funny answer: "I know, but I
don't believe it! "
S 1 . ..Re3 S2.Kf2 Rc3 S3 .Kg2 KfS S4.Kf2 KeS S S .Kg2 Kd4 S6.Kf2 Re2 +
S7.KgI Ke3 S8.Bb7 Re7 S9.BdS ReS 60.Bb7 Rc 1 + 6 1 .Kg2 Rb I 62 .Be6
Rb6 63 .BdS Rd6 64.Bb7 Rd2 + 6S .Kg I Kd4 66.BfJ KeS 67.KfI Kf6
68.KgI KgS 69.KfI KfS 70.Ke 1 Rh2 7 1 .KfI KeS 72 .KgI Rd2 73 .KfI
Kd4 74.KgI Ke3 7S .Bb7 gS 76.Be6 Rf2 77.Bb7 g4 78.Be6

Of course, after 7 8 .Bc8? Kf3 79.Bxg4+ Kxg3 , Black has a winning posi
tion because White's King is boxed in.
78 ... Rf3 79.Kg2 ! Rf6 80.Bb7 Rb6 8 1 .Ba8 Rb8 82 .Be6 Kd4 83 .Kf2 Re8
84.Bb7 Re7 8S .Ba8 Ra7 86.Be6 ! , 1/Z- 1f2 .

Black wins after 86.Bg2 ? Ra2 + 8 7 .Kg l Rxg2 ! 8 8 .Kxg2 Ke3 . Black finally
got the message after my 86th move: I am not willing to trade my Bishop
for his Rook! He gave up and agreed to the draw.

( I I S) Kane - Benko (U.S. Championship. 1 973)

l .d4 Nf6 2.e4 g6 3 .Nc3 dS 4.cxdS NxdS S .e4

The classical Exchange Variation is one of the best methods of playing

against the Griinfeld. White builds a strong center, while Black tries to
undermine it and prove that it's actually a target. The clash of these two
strategic and philosophical viewpoints usually leads to a sharp, double
edged fight.
S ...Nxc3 6.bxc3 Bg7 7.Be4

Nowadays 7 .NB is all the rage, but at that time it was thought that White's
only good plan was to place the Knight on e2 .
7 ... b6 ! ?

O f course, I knew the main lines started with 7 . . . c 5 or 7 . . . 0-0, a s every open
ing book (past and present!) would verify. But my idea was precisely to get
away from popular theory. The text was not a new idea, but it was only seen
after 7 . . . 0-0 8.Ne2 , and now 8 . . . b6, but in that case White can launch a
dangerous attack by 9.h4! Bb7 (better is 9 . . . Nc6, though 1 O.Bd5 ! still leaves

E U RO P E AT MY F E ET

Black struggling for equality) 1 O.Qd3 . The white Queen comes into action
effectively on the h-file. Many attempts to improve this line for Black had
been seen in international tournaments prior to this game, but they all ended
disastrously. Therefore I decided to try delaying castling so my King
wouldn't be such a clear target. After considerable thought, Kane came up
with the most challenging answer.
8.Qf3

White takes advantage of the fact that the K-Knight hasn't been devel
oped yet to place the Queen on this very active and threatening square.
Black is now forced to castle, but this allows White to begin tactical op
erations along the D -a8 diagonal. Though 8 .Ne2 doesn't give White any
real chance for an advantage, 8 .ND has found a few supporters : 8 . . . 0-0
(8 . . . Bb7 ? ! isn't a good idea because of 9.Bxf7+ Kxf7 1 O.NgS +) 9.0-0 Bb7
1O.Qc2 ( 1 0.Re 1 cS 1 1 .Bb2 cxd4 1 2 .cxd4 Nc6 1 3 .Qd2 NaS 14.Bfl Rc8 1 S .dS
Bxb2 1 6 . Qxb2 Qd6 1 7 . Rad 1 , with a very pleasant position for White in
Hartston-Gergs, Hamburg 1 97 7) 1 O . . . Nc6 1 1 .Re 1 Qd7 ( l l . . .Qd6 ! ?) 1 2 .Rb 1
e 6 1 3 .Ba3 Rfd8 1 4.h4 NaS l S .Bfl cS 1 6 .dxcS Q c 7 1 7 . Qc 1 bxcS 1 8 .hS, and
White had the superior chances in Korchnoi-Timman, Brussels 1 986.
8 . . . 0-0

Like it or not, I didn't see a better answer. 8 . . . e6 9.eS c6 1 O.Ba3 did not
look appealing.
9.e5

This looks quite strong. Also critical is the straightforward (but for some
reason untried) 9.h4! ? , when Black has several defensive possibilities:
A) 9 ... Nc6 (threatening ... Nxd4) 1 O.BdS Bb7 1 1 .hS e6 1 2 .Bxc6 Bxc6 1 3 .hxg6
is very strong for White .
B) 9 . . . Bxd4 is a crazy possibility. Unfortunately for Black, White can ig
nore it and continue with 1 O.hS ! , with a very strong attack. Here's a
taste of what can happen: 1 O . . . Bg4 (other moves also leave Black on
the edge of the precipice) 1 1 . Qxg4 Bxch 1 2 .Ke2 Qd4 1 3 .BdS c6
1 4.hxg6 cxdS l S .gxf7+ Kh8 1 6.Rxh7+ Kxh7 1 7 .QhS+ Kg7 1 8 .Bh6+ Kf6
1 9 . QfS mate .
C) 9 . . . eS 1 0.hS , and Black is again going to suffer through a powerful White
attack.
D) 9 . . . hS leads to a structure that usually favors White . Indeed, White
has to have an edge after 1 O.Ne2 ( l 0. Qg3 Ba6 is less clear) 1 O . . . Bg4
1 1 . Qg3 Bxe2 1 2 .Kxe2 ! e6 1 3 . Rb 1 Qd7 1 4.Rd 1 , with two Bishops and
a strong center.
Naturally, White can forego the instant aggression of 9.eS or 9.h4 and
instead proceed quietly by 9 .Ne2 , but I wasn't worried about this at the
time because I thought that every red-blooded tournament player would
immediately try to punish my provocative opening. However, Yusupov was
successful with it in his match against Timman (9th game, Tilburg 1 986):

345

PAL B E N KO : MY L I F E , GAMES A N D COM POSITI O N S

346

9.Ne2 Nc6 1 O.h4 (Of course, this isn't necessary. White can try Dolmatov's
1 O.BgS , or standard moves like i O.Be3 or 10.0-0.) i O ... NaS I l .Bd3 eS 1 2 .Ba3
Re8 1 3 .h S Qd7 1 4. Rd l Qa4 I S . B e l c S (The most testing move was
I S . . . Qxa2 , when White still has to prove that his attack is worth the sacri
ficed pawn.) 1 6.dS Qxa2 1 7 .Bh6 Bh8? (His game falls apart after this. He
still could have defended by 1 7 . . . Bxh6, when 1 8 .hxg6 fxg6 1 9.Rxh6 Bd7
2 0 .Qf6 followed by Rxg6+ is a perpetual check.) 1 8 .BbS Rd8 1 9.BgS Qb3
2 0.hxg6 fxg6 2 1 .Rxh7 Kxh7 2 2 .Qf7+ Bg7 2 3 .Bf6 Rg8 24.Be8, 1 -0.
9 ... Ba6!

I don't like to be pushed around, so instead of the passive 9 . . . c6 (which

favors White after 1 O.Ne2), I decided to strive for active counterplay with
an Exchange sacrifice. When Grandmaster O'Kelly annotated this game,
he wrote (of 9 . . . B a6), "This is either home analysis or a stroke of genius. "
It wasn't home analysis.
l O.Qxa8?

I have to admit I was happy to see my opponent grab my undeveloped

Rook without much thought, giving up his only developed piece. At least
he should have tried i O .Bxf7+ Rxf7 I l .Qxa8, which would have been the
real test of my sacrifice. Black could then get back the pawn by 1 1 . . . BxeS
(other moves are also possible) when 1 2 .dxe S ? (Much too materialistic.
White should play 1 2 .NB .) 1 2 . . . Qd3 gives Black tremendous compensa
tion. One sample line is 1 3 .Qxb8+ Rf8 1 4.Qxc7 Qf1 + I S .Kd2 Qxf2 + 1 6.Kd l
Qxg2 , with a win for Black.
In two games against Timman, Yusupov tried 1 0 .BdS c6 I l .Bb3 , and
now
A) Later, at Bugojno 1 986, Timman improved with 1 1 . . .Qd7 1 2 .Ne2
( 1 2 .h4! ? is a suggestion of Timman's.) 1 2 . . . e6 1 3 .0-0 cS 1 4.Rd l Nc6
I S .BgS cxd4 1 6.cxd4 Bxe2 1 7 .Qxe2 Nb4 1 8 .h4 Rfc8 1 9.hS , and White
had an edge, though the game was later drawn.
B) 1 1 . . .Qc7 ? led to a rout in game 7 of Yusupov's and Timman's 1 986
Candidates Match in Tilburg: 1 2 .h4 cS 13 .hS cxd4 1 4.cxd4 gxhS I S .RxhS
Bb7 1 6.Qd3 Rd8 1 7 . Qxh7+ Kf8 1 8 .Ne2 Rxd4 1 9.Bh6, 1 -0.
White has also tried 1 0.Bb 3 , but 1 0 . . . Qd7 I l .Ne2 ( l l .Qxa8 Nc6 1 2 .Qxf8+

E U RO P E AT MY F E ET

Kxf8 favors Black) 1 1 . . .Nc6, gives Black excellent counterplay since 1 2 .BdS
Bb7 threatens both 1 3 ... NxeS and 13 ... Nxd4.
10 . . . Bxc4 1 1 .Qf3

This was White's "safe" plan, preparing Ne2 and 0-0, and getting away
with the Exchange with minimal damage. But it turns out to be not so safe
after all. No matter how White attempts to secure his King, Black's strong
QB is more than enough compensation for the Exchange.
1 1 ...f6

This looks like a logical continuation of the attack, opening more lines. But
here 1 1 . . .cS was tempting. If 1 2 .Ne2 ( 1 2 .Be3 Qd7 1 3 .Ne2 QbS) 1 2 . . . cxd4
1 3 .cxd4 Bxe2 1 4.Qxe2 Qxd4 1 5 .Bb2 Qb4+ keeps tormenting the white King.
However, I would have hated to give up my dominating light-squared Bishop,
even in order to win a pawn with a favorable position.
1 2 .e6 ? !

White tries t o pacify his opponent and keep the lines closed b y giving up
a pawn. But, after winning the guy on e6, Black has material equality while
White still has trouble finding a safe place for his King. Like it or not,
1 2 . exf6 was a must.
12 . . . Qd6

Of course, there is no need to play . . . Bxe6, as the Queen can take the
pawn and activate herself at the same time.
1 3 .Ne2 Qxe6 1 4.Be3 f5

Making it problematic whether White will succeed in castling or not.

1 5 .0-0?

Anyway! But now it looks as though White will lose a piece. I S .Nf4 Qd7
favors Black since White can't stop both ... e7-eS and ... g6-gS . Perhaps White
should have tried I S .h4! ?
1 5 ... Bxe2 1 6.Qxe2 f4 1 7.d5

This little Zwisch enzug is White's last hope.

17 . . . Qe5 1 8.Qc4 b5

Black returns the Zwisch enzug. If 18 ... fxe3 ? 1 9. d6+ Kh8 2 0 . dxc7 , White
wins back the piece.

347

PAL B E N KO: MY LI FE, GAM ES AN D COM POSITI O N S

1 9.Qxb5 fxe3 20.fxe3 RdS

348

Avoiding the exchange of Rooks since I felt that I needed this piece to
defend my back rank. I also refused to snap up the e3 -pawn. Why open up
new files for the white Rooks when there is no need to do so? I will take
that pawn, but only after I get my Knight into play.
2 1 .Rab l Nd7

Black finally develops his last piece and avoids the threat of 2 2 . Qxb8 . Af
ter this, White has no compensation for his lost material.
22.c4 Qxe3 + B .Kh l Be5

The centralized Bishop aims at the white King position and simultaneously
defends the queenside.
24.Rb3 Qd4 2 5 .Rbf3 Bd6 26.Qb3 RbS, 0- 1 .

( 1 1 6) Martz - Benko (U.S. Championship, 1 973)

l .e4 c5 2 .NB e6 3 .Nc3 a6 4.d4 cxd4 5 .Nxd4 d6 6.g3

White uses many different plans against the Scheveningen setup, with his
light-squared Bishop being placed on c4, d3 , or e2 . The fianchetto is the
most positional treatment.
6 ... Nf6 7.Bg2 Qc7

This early Queen move belongs in this system anyway, but its purpose
here is specifically to prevent the fianchetto of White's dark-square Bishop.
S.O-O Be7 9.Qd2

This strange-looking Queen move (which has been used on many occa
sions by the Dutch/Brazilian 1M Van Riemsdijk and Russian GM Kholmov)
prepares the fianchetto of the Q B. More common choices are 9 .Be3 , 9 . Re 1 ,
9.f4, 9.a4.
9 ... 0-0 1 0.b3 Nc6

Black wants to advance on the queenside, but the immediate 1 O . . . b 5 ? ? isn't

possible due to I I .e5 . Black's tenth move usually addresses this problem,
which is why 1 O . . . Bd7 followed by 1 1 . . .Nc6 with . . . b7-b5 to follow is most
popular, while my 1 0 . . . Nc6 also makes good sense. The experimental
1 0 . . . Rd8 was tried in Kholmov-Schlosser, Brno 1 99 1 , but a tactical flurry
left White with the better game: I l .Bb2 b5 1 2 .a4 b4 1 3 .Na2 Bb7 1 4. Rfe l
d5 1 5 .exd5 Bxd5 1 6.Bxd5 Nxd5 1 7 . Rxe6 fxe6 1 8 .Nxe6.
1 1 .Nde2

This retreat restricts the black position by avoiding the exchange of

Knights. In fact, Fischer allowed the trade against Petrosian in Curaao
1 962 (by transposition of moves) , but the white Queen was on the origi
nal d l -square and therefore got to d4 (by recapturing) in only one move
instead of the two.
1 1 ...b5 1 2 .Bb2 Bb7 1 3 .Rfd l RfdS 14.Nf4 RacS 1 5 .Rac1 Qb6

E U RO P E AT MY F E ET

349

Black has been able to develop all his pieces freely, without any interfer
ence from White. Probably the more direct l S . . . Qb8 followed by . . . Qa8
was better here, exerting pressure on the long diagonal- indeed, this setup
eventually occurs in the actual game. Another possibility, 1 5 . . . d S , looks
premature if Black wants more than mere equality. Mter 1 6. exdS Nb4
1 7 .Qe2 followed by a series of trades on d S , a position is reached that
offers little exercise for the imagination.
1 6.Qe l Ne5 1 7.Nd3 Ned7

Black has improve d his position- the Knight no longer blocks the
diagonal of his QB.
I S.Kh I Rc7 1 9.Nb l RdcS 20.c3

Not a good looking move, but it nullifies any immediate pressure on

the c-file.
20 ... Qa7 2 1 .Nd2 QaS 2 2 .3

Another ugly move, but this time White is trying to resist the pressure on
the long diagonal. I've obtained a dream setup for the Black side of the
Sicilian, but I still haven't achieved any tangible, specific success against
White's wait-and-see policy. If 22 . . . dS 2 3 . e S , the position would remain
closed. If 22 . . . e S , then 23 .c4 restricts any further Black advance in the center.
2 2 . . .h6 2 3 .Qe2 a5

Black tries to start some action on the queenside to "keep the ball in play. "
24.Nfl NeS 2 5 .Ne3 Bf6 26.c4

White finally decides to open the position, but this meets with no
obj ection from Black, who is ready for it.
26 ... bxc4 27.Rxc4 Rxc4 2 S.Nxc4 d5 !

This thematic shot, which aims to create a dominating Bishop on d S , gives

Black the initiative.
29.exd5 Bxd5 30.Nde5 Bxe5

A miscalculation. During the game I felt that 3 0 . . . NxeS failed to 3 1 .Nb6,

but ! overlooked that after 3 l . . .BxB ! 32 .BxB QxB + 3 3 .QxB NxB 3 4.Nxc8
Bxb2 3 S . Rd8 Kf8 3 6 .Nd6, Black has 36 . . . Ke7 winning the game. Instead

PAL B E N KO: MY L I F E , GAM ES AN D CO M PO S ITI O N S

of 3 1 .Nb6, better are both 3 1 .Nxe5 a4 and 3 1 .Bxe5 Bxc4, though Black
has an edge in either case.
3 1 .Nxe5 Nxe5 32.Bxe5 a4
350

Now it's clear that Black has all the trumps on the queenside, while White's
Bishop pair has no significance, considering the passive situation of the poor
King Bishop. Perhaps White's best chance here is to trade down to an end
ing with 3 2 .f4 Bxg2 + B . Qxg2 Qxg2 + 3 5 .Kxg2 Rc2 + (and not 3 5 .. .f6?
3 6.Bd4 ! , when the game is equal: 36 ... Rc2 + 3 7 .Bf2 Rxa2 [Black doesn't
achieve anything with 3 7 . . . a3 3 S .RdS Kf7 3 9.Rd7+ Kg6 40. RdS =] 3 S .bxa4
Rxa4 3 9.RdS KfS 40.Bc5+ Kf7 4 1 .Rd7+ Kg6 42 .Re7 Rc4 43 .RxeS Rxc5
44.Rxe6, with a draw) 3 6 .Kh3 f6 3 7 .BbS Rxa2 3 S .RdS Kf7 3 9.Rd7+ KfS
40.bxa4 Rxa4 4 1 .Ba7 Rc4, when Black has won a pawn, but the four-against
three pawn position (all pawns on the same side) requires a lot of hard work
on Black's part, with no guarantee of a win. Of course, it is never an easy
decision to give up material instead of defending stubbornly; the more so in
this case as my opponent had had a bad experience in the same type of
endgame against Browne in an earlier round.
H .Rb l f6 34.Bd4 e5 3 5 .Be3 axb3 3 6.axb3 Rc3 3 7.b4 Qa3

Black's pieces are surrounding the White position and it's becoming in
creasingly difficult for the first player to hold onto his pawns on b4 and
B.

3 8.Re l Qb3 3 9.b5 Bc4

Even in time pressure, Black has been able to push back the opponent
completely. But here, much stronger was 3 9 . . . Nd6 (Once this Knight joins
the rest of Black's army, White's game should fall apart.) 40.b6 Nc4. How
ever, I had no time to figure out how dangerous the white b-pawn would
be and if it was safe to go after the f-pawn. Analysis later showed that the
b-pawn was not something to be feared.
40.Qd2 Rd3 4 1 .Qf2 Qxb5

E U RO P E AT MY F E ET

35 1

Black has won the b-pawn in better circumstances than those possible af
ter my recommendation for White at his 3 3 rd move. At this point some
thing interesting occurred. My opponent raised his hand with the obvious
intention of pushing his f-pawn, but just then the tournament director
stopped him and instructed him to seal his next move. Martz protested,
claiming that he had already indicated what his move was going to be (an
interesting legal question, and don't forget that Martz was a lawyer) . The
director insisted, saying that the playing time was over, and that since it
was White's move, it was his duty to seal it. I told Martz that I could not
know what move he would actually place in the envelope, and that I would
have to analyze every possibility anyway. Of course, in all honesty, I have
to admit that I expected him to play 42 .f4. After considerable thought,
Martz decided on his move, wrote it down, and placed it in the envelope.
Later, while analyzing alone in my room, I regretted not letting him play
42 .f4 on the board, because I found out that it was one of his feeblest
possibilities. I became convinced that, in view of the circumstances, he
had changed his mind and played something else. Nevertheless, my analysis
showed that Black would also win against other moves since he had an
extra pawn and the better position, but it was possible for White to put up
a long struggle. When the game resumed and the envelope was opened, I
received some pleasant news.
42 .f4?

This makes it easier for Black to gain access to the white King by trading
Bishops and using the Knight to attack the weakened light-squares. After
42 .f4 appeared on the board, Martz commented, with a smile on his face,
that his move was probably not best.
42 ...Bd5 43 .g4 ! ?
An interesting idea, trying t o break u p the enemy kingside and trade off as
many pawns as possible.

43 .. .Nd6 44.fxe5 fxe5 45 .g5 Ne4!

I didn't see a better move, because 45 . . . h5 46 . g6 might give White

counterchances against my King.

PAL B E N KO: MY L I F E , GAM ES AN D C O M POSITI O t:J S

46.Qh4?

352

The obvious 46.Bxe4 Bxe4+ 47 .Kg 1 h5 creates opposite-colored Bishops,

but this does not make the game a draw. In fact, because of these Bishops,
Black has a strong attack against the exposed enemy King. At the same
time, the black King enjoys complete safety after . . . Kh7. Nevertheless,
that was White's only good try. The Queen move brings immediate disas
ter. The alternative 46. Qf5 was strongly met by 46 . . . Rxe3 ! 47 . Qc8+ (47 .Rxe3
Qb 1 + 48 .Bfl Ng3 +) 2 7 . . . Kh7 48 .Qf5+ (48.g6+ Kxg6 49.Qg4+ Kh7) 48 . . . g6
when Black has no problems bringing home the full point. After the text,
a few fireworks finish the long strategic fight.
46 ... Rxe3 ! 47.Rxe3 Ng3 +

The Knight is taboo because of 48 . . . Qfl + and mate.

48.Kgl Qb l +, 0- 1 .

White resigned rather than face the forced mate after 49.Kf2 Qc2 + 50.Ke l
Qc l + 5 1 .Kf2 Qd2 +.

( I 1 7) Benko - Janosevic (Vrnjacka Banja, 1 973)

l .c4 Nf6 2 .Nc3 g6 3 .d4 c5 4.NB Bg7 5 .e4 0-0 6.Be2 d6 7.0-0 Bg4

A logical move, striving to get rid of the ineffective light-squared Bishop.

Furthermore, by trading it for White's B -Knight, he intends to weaken
the dark-squares in White's position.
8.d5

White decides to advance the d-pawn and turn the game into a Benoni.
Far less promising is 8.Be3 BxB 9.BxB cxd4 1 O.Bxd4 Nc6, with a com
fortable version of the Maroczy Bind for Black (the B -Bishop doesn't have
any prospects) .
8 ... Na6

Over the years, I've noticed that many amateurs play . . . Nbd7 in this type
of position, apparently not realizing that Black's plan is queenside expan
sion via . . . b7-b 5 . Naturally, 8 . . . Na6 is superior because this Knight can
help support that pawn advance by . . . Na6-c7 .
9.Bf4

Since I have a space advantage in the center, my correct plan is e4-e5 .

This Bishop move helps make this pawn push a reality.
9 ... Nc7 1 0.h3 Bxf3 l 1 .Bxf3 a6 1 2 .a4 Rb8 1 3 .Be2 Nfe8

Black's counterplay on the queenside has been stopped, thus he tries to

free himself in the center with . . . e 7 -e6 or . . . e 7 -e5 . Also to be considered
was 1 3 . . . b6, after which the immediate 1 4.e5 is not promising due to
14 . . . Nfe8, when White is forced to trade on d6.
1 4.a5 b6 1 5 .axb6 Rxb6 16.Ra2

E U RO P E AT MY F E ET

353

More natural looking is 1 6 .Na4, and on 1 6 . . . Rb8 White has 1 7 .Rb l fol
lowed by Bd2 and b4. However, Black could prevent this by 1 7 . . . a5 and
. . . Nc7 -a6-b4, when he might eventually be able to exploit the weakness of
the b-pawn. Therefore, I decided to keep this Rook on the a-file (which
stops Black from advancing his a-pawn) and place the other Rook on b l
for the purpose of carrying out b 2 -b4. It's important to point out that Black's
usual strategy in this kind of position (where White is about to overwhelm
the queenside with his b2 -b4 plan), . . . e7-e5 followed by . . . f7-f5 -f4 with
kingside chances , is not dangerous here because he has traded his
light-squared Bishop -a piece that is needed for the attack. White, how
ever, still has his light-squared Bishop, and it can be utilized effectively in
both attack and defense. Having realized all of this, Black decided on a
surpnsmg move.
1 6 . . . Bxc3 ? !

This trade, which seriously weakens the dark squares around Black's King,
would be fitting if he was able to successfully close the position by . . . e 7 -e5
or to prevent White from cracking open the center with e4-e5 . As it turns
out, Black isn't able to achieve either of these important goals.
1 7.bxc3 e5 1 8.dxe6

White has to make sure the center is open so that his two Bishops remain
active and Black's d-pawn remains weak. Instead of 1 7 . . . e 5 , Black must
have considered 1 7 .. .f6, but White's obvious response would consist of Bh2 ,
f4 and e 5 .
1 8 . . . Nxe6

White also retains all the chances after 1 8 . . . fxe6 1 9.Qd2 Qh4 2 0 .Bg4, when
the Bishops start to show their stuff.
1 9.Bh2 N8g7?

A blunder, after which Black's position becomes critical. More tenacious

was 1 9 . . . Qa8 2 0.f3 Qc6 followed by . . . N8c7 and . . . Rfb8 . Of course, White
would stand better in this case, too, due to his plan of e4-e5 or f2 -f4-f5 ,
with a kingside attack.
20.Rd2 Qh4

PAL B E N KO: MY LIFE, G A M E S AND CO M PO S ITI O N S

After 2 0 . . . N eS, there would follow the obvious and very strong 2 1 .e S . For
this reason, Black tries to create some tactical chances on the kingside.
2 1 .Bxd6 ReS 2 2 .Bg4 Qf6
354

2 2 . . . hS would be met by 2 3 .g3 QgS 2 4.f4, with an easy game for White .
2 3 .e5 Qf4 24.Qe2 h5 2 5 .g3 Qh6 26.Bxe6 Nxe6 2 7.h4 g5

Due to his material and positional disadvantages, Black launches a desper

ate last-chance attack.
2 S.Qe3 Qg6 29.hxg5 Nxg5 3 0.Bxc5 Rxe5

Black has apparently obtained some tactical chances, as 3 1 . QxeS or Qd4

lose to 3 1 . . . Nf3 +. These threats, however, turn out to be illusory due to
the vulnerability of the black King.
3 1 .RdS+ Kh7 3 2 .Bd4! Rbe6

Of course, the Queen could not be captured because of mate. Now I con
sidered the flashy, strong, but complicated 3 3 .f4. However, I decided to
be practical and play a simple and easily winning line .
3 3 .Bxe5 Rxe5 34.RhS+ KxhS 3 5.Qxe5+ Kh7 36.Kg2 h4 3 7.g4 h3 + 3 S.KhI
Kh6 3 9.f4

With his last move Black tried to avoid a Queen exchange, but White's
reply extinguishes all hope.
3 9 . . . Nh7 4 0 . g 5 + Kh 5 4 1 . Q e 2 + Kh4 42 . Qf2 + Kh 5 4 3 . Qf3 + Kh4
44.Kh2 , 1 -0.

( 1 1 8) Basman - Benko (Hastings, 1 973)

l .e4 c5 2 .NO d6 3 .Bb5 + Bd7 4.Bxd7+ Qxd7 5 .0-0

The other choice is S . c4, intending to establish a Maroczy Bind formation

with d2-d4. If Black prevents this by S . . . e S , then White is left with a supe
rior Bishop, queenside play via a2-a3 followed by b2 -b4, and a slight edge.
5 ... Nc6 6.c3 Nf6 7.Re l

A natural move that leads to a very boring game. Much more interesting is
7 . d4 Nxe4 S.dS NeS 9.Re l , with attacking chances for the sacrificed pawn.

E U RO P E AT MY FEET

7 . . . e6 8 .d4 exd4 9.exd4 d5 1 0.e5 Ne4 1 1 .Nbd2 Nxd2 1 2 .Bxd2 Be7

1 3 .Bg5 0-0 1 4.Bxe7 Qxe7
355

This position is quite drawish. As a matter of fact, this was exactly what
my opponent wanted since he only needed half a point to get his interna
tional master title. Due to this peaceful goal, he changed his style and re
fused to play any of the wild openings that he's become famous for (as
Black, he likes l .e4 g5 , while as White he plays 1 .4 and l .Nc3 , though
his favorite is the outrageous l .h3).
1 5 . R e 3 Rfe8 1 6 .Rc3 Q d 7 1 7 . Rac 1 N e 7 1 8 . h 3 h6 1 9 . Q b 3 Rxc 3
20.Rxc3 Re8 2 1 .Qe2 Rxc3 2 2 . Qxc3 Qe6

Black is already somewhat better. If 2 3 .Qa3 Qc 1 + 2 4.Kh2 , then 24 . . . Nc6

is strong.
2 3 .Qe5

This Queen trade is a bad idea. He should have retained the Queens by
2 3 . Qd2 or 2 3 . Qb4.
23 ... Qxe5 24.dxe5 Ne6

Suddenly White is in trouble. His Knight is nailed to f3 to protect

his e-pawn.
2 5 .a3

Trying to get the queenside majority rolling with b2 -b4. Of course, I don't
allow this.
2 5 ... a5 2 6.b3

White didn't want to let Black play . . . a5-a4, when White's majority would
be permanently crippled.
2 6 . . . Kf8 2 7.Kfl

Maybe White should play Kh2 and try to protect e5 with his King.
27 ... Ke7 28.Ke2 Kd7 29.Kd3 Ke7 3 0.g4

This doesn't help. Even 3 0.h4 makes more sense.

30 ... g5

Now White can't move any of his kingside pawns, and his Knight is also
stuck on f3 guarding e 5 .

PAL B E N KO : M Y LI FE, GAM ES AN D COM POSITI O N S

3 1 .Kc3 b 6 3 2 . cxb6+

Probably played with a heavy heart, but 3 2 .b4 bxcS 3 3 .bxcS loses to 3 3 . . .a4,
when the black King is ready to march to bS and pick up the c-pawn.
356

32 ... Kxb6 H .Kd3 Kc5

Things have gotten worse for White. Now, on top of the immobilized
white army on the kingside and my protected passed d-pawn, Black can
also claim a more active King. All I have to do now is find a way to
penetrate into the white position with my King.
34.Ke3

Not what he wanted to do, but 3 4.Kc3 loses to 3 4 ... d4+ 3 S .Kd3 KdS , and
the eS -pawn drops off the board.
34 ... a4!

This pawn sacrifice clears the way for the black King.
3 5 .bxa4 Kc4 3 6.a5 Nxa5 3 7.Nd4 Kc5 3 8.Kd3 Nc4 3 9.f4

The last try. Now 3 9 . . . gxf4 40.Nf3 allows White to create a passed pawn
by h4-hS followed by g4-gS .
3 9 ... Nh2 + ! 40.Kc2

Black also wins after 40 .Kc3 Nd l + 4 1 .Kd3 Nf2 +.

40 ... Kxd4 4 1 .fxg5 hxg5, 0- 1 .

After 42 .Kxb2 KxeS , my King easily stops White's a-pawn, when my two
connected passers will march to glory. Though he failed here, Basman
later got his 1M title anyway.

( I 1 9) Benko - Pytel (Hastings, 1 973)

l .Nf3 d5 2 . c4 c6 3 .e3 Nf6 4.Nc3 e6 5.d4 Nhd7 6.Bd3 dxc4 7.Bxc4 b5
8.Be2 ! ?

We have reached the very popular Meran Defense, a line that has been
the subject of much debate for many years, and continues to fascinate players
of every level right up to the present time. The text move (8 .Bd3 is the
most popular choice) was often adopted by Capablanca, though with a

E U RO P E AT MY F E ET

different idea than mine in this game -Capa probably used the move when
he was in a peaceful mood. The advantage of S .Be2 is that it is little ana
lyzed and, despite its innocuous appearance, has many subtle points. If
Black intends, for example, to answer with an early . . . cS, he gets into trouble:
S ... b4 (one of Black's best replies) 9.Na4 cS ? (correct is 9 ... Bb7) 1 0.dxcS
NxcS I 1 .QxdS+ KxdS 1 2 .NxcS BxcS 1 3 .NeS Ke7 1 4.BD ! NdS I S . Bd2 ,
and White stands better.
8 ... a6

More common is S . . . Bb7 , which, in my opinion, is also better. Pachman

(and modem theory corroborates his view) then recommends 9.0-0 ! , though
without offering any explanation. The reason is that on 9.e4 (Black doesn't
have any difficulties after 9.a3 b4!), Black equalizes the game by 9 . . . b4 !
1 0 .eS bxc3 l 1 .exf6 Nxf6.
Poor is S . . . Be?? 9.a3 ! , when White is just better. Benko-Kaufman, New
York 1 972 continued: 9 . . . a6 l O.b4? ! ( 1 0.e4! was strong at once, when l O . . . cS
could be met by l 1 .eS Ng4 1 2 .h3 cxd4 1 3 . Qxd4. The text ( l O.b4) is a new
plan that prevents Black's liberating . . . c6-cS and thus entombs his light
squared Bishop. Its disadvantage is that White will be compelled to yield
the a-file, while his b-pawn may become weak) lO . . . aS I 1 . Rb l O-O? (Pref
erable was 1 1 . . .axb4 1 2 .axb4 NdS 1 3 .NxdS exdS, in connection with . . . Nd7b6-c4, reducing White's pressure on the half-open c-file [as is commonly
seen in some lines of the Queen's Gambit Declined] . After 1 1 . . . 0-0,
White stands better.) The rest of the game, without notes, was: 1 2 .e4 axb4
1 3 .axb4 Nb6 1 4. Qc2 Bb7 I S .Bd3 h6 1 6 .0-0 ReS 1 7 .Rd l BfS I S .h3 Nfd7
1 9 . B f4 Q e 7 2 0 . Rb 3 R e c S 2 1 . Q e 2 R a 7 2 2 . B e 3 R c a S 2 3 . d S cxd S
2 4. exdS Ra l 2 S . dxe6 BxD 2 6 . Qxf3 Qxe6 2 7 . Rbb l Rxb l 2 S .Bxb l RbS
2 9 .Nxb S Nf6 3 0 .Nd4 Q d S 3 1 .Nc6 Qxf3 3 2 . gxD Rb7 3 3 . Rd S Nbd7
3 4.Be4 RbS 3 S .Bd3 RhS 3 6.bS g6 3 7 .b6 NeS 3 S .Nxe S RxeS 3 9.b7, 1 -0.
9.e4! b4 1 0.e5 bxc3 1 1 .exf6 cxb2

Black's best move is probably 1 1 . . .Nxf6, as suggested by Euwe, when the

continuation 1 2 .cxd4 Bd6 1 3 .Nd2 ! (heading for c4) only gives White a
slight pull. The text move, which takes the game into an almost comical
situation, resulted in some audible snickering by the spectators.
1 2 .fxg7 bxa l =Q

Black should probably try 1 2 . . . Bxg7 , though White retains some advan
tage after 1 3 .Bxb2 QaS+ 1 4.Nd2 RbS ( 1 4 . . . Qb4! ? hasn't been tested, but
Korchnoi's recommendation of I S . Rb 1 makes sense: I S . . . Bxd4 1 6.Ba 1 QcS
1 7 .Rc l Qb6 I S.Bxd4 Qxd4 1 9.Rxc6 Bb7 20.Rc4, and White's superior pawn
structure gives him a slight edge.) I S . Qcl QgS 1 6 .0-0 cS 1 7 .Nb 3 Qxc l
I S .Raxc l Bb7 1 9 .Ba3 cxd4 2 0.Bd6, when Black was under pressure in
Schneider-Kishnev, USSR 1 993 .
1 3 .gxh8=Q

357

PAL B E N KO : MY L I F E , GAM ES AN D C O M POSITI O N S

358

I t i s indeed a rarity t o have four Queens o n the board a t the same time.
This is one of the peculiarities of the line with 8 .Be2 . One can also go into
this type of thing after 8.Bd3 , though then Black's chances are better than
they are here. An evaluation of the diagramed position leads to the opin
ion that White seems to stand better because his Queen, deep in Black's
kingside, is more threatening than Black's Queen on the queenside.
13 ... QaS +

After the text, I imagined both sides capturing the a2 and h7-pawns and
then making new Queens on those files. Then we would have six Queens !
Unfortunately, reality dictates that we'll soon have to go back to just two.
1 4.Bd2

This forces the exchange of one pair of Queens, which seemed to disap
point the spectators, judging by their murmuring. I considered 1 4.Nd2 ,
but after 1 4 . . . Q 5 c3 followed by . . . Qxa2 my Queen on d 1 would have stood
passively. It would have been interesting to try it anyway, but only in a
game where the point was not so important.
14 ... Qxd l + I S .Bxd l QfS 1 6.0-0 Bb7

Black prepares for queenside castling. White's next move was played to
discourage my opponent from doing this, but 1 7 .NgS was also taken into
account. Here I chose the more beautiful solution, a winning attempt be
ginning with a pawn sacrifice.
1 7.dS !

A very satisfying line-opening sacrifice.

1 7 . . . QxdS

The only good move. Now 1 7 . . . exdS 1 8 .Re 1 + is crushing, while 1 7 . . . cxdS
1 8 .Ba4! 0-0-0 1 9.BaS gives White a winning attack since 19 ... Re8 2 0 . Rc 1 +
I S game over.
1 8.Qxh7 cS

Of course, 1 8 . . . 0-0-0 would have been met by 1 9. Qxf7 . With 1 8 . . . cS, Black
tries to get counterplay on the long diagonal. My opponent offered a draw
here, but after such an exciting opening I was not ready to make peace.
1 9.Ba4 0-0-0

E U RO P E AT MY F E ET

Other moves also fail to hold the game: 1 9 . . . Rc8 20.Rd l Bc6 2 1 .BgS Qxa2
(Black is also completely lost after 2 1 . . .Bxa4 2 2 .RxdS exdS 2 3 .QfS) 2 2 .Bxc6
Rxc6 2 3 . N e S and m a t e s ; 1 9 . . . B c 6 2 0 . B xc6 Qxc6 2 1 . N g S 0-0-0
2 2 .Nxf7 Re8 2 3 . Qg6, and Black's position is hopeless.
20.Ba5 Ne5

Trying to mix things up a bit. Black dies without a fight after 2 0 . . . Qxa2
2 1 . Qh4.
2 1 .Ne l !

The climax of the game. White must retreat for the time being. 2 1 .Rd l
was incorrect because of 2 1 . . .Nxf3 + 2 2 . gxf3 QgS +.
2 1 .. .c4

On the possible 2 1 . . .Nd3 2 2 .Qxf7 Bd6 2 3 .Bxd8 Nxe l , White can force
a winning endgame by 2 4 . Qg8 Qxg2 + 2 S . Qxg2 Nxg2 2 6 . Rd l B d S
2 7 .Bf6 Nf4 2 8 .h4.
22.Bxd8 Qxd8

White's advantage of the Exchange is entirely sufficient to win the game,

since the active white Queen pins down the opponent's forces.
2 3 .Qh8 f6 24.Qg8 Qd6 2 5 .Nc2

White needs to get all his pieces to active squares. This move prepares to
bring my Knight to e3 , and also frees my Rook.
25 ... Kc7 26.Ne3 Be7 27.Rd l Qb6 28.Qe8

This leads to a simple and safe win. Also good was 2 8 .Nxc4 Qb4 (2 8 ... Nxc4
isn't possible due to 2 9.Rd7 mate) 2 9.NxeS fxeS (2 9 . . . Qxa4 3 0.Rc 1 + is easy)
3 0 .Qxe6 when Black should resign.
28 ... Bc5

Black overlooks mate, but his game was hopeless anyway. The best de
fense was 28 . . . Qb4 (2 8 . . . QcS falls victim to 2 9 .Nxc4!) 2 9 .Nc2 (2 9.h4 is
also good, ending back-rank threats and intending to turn the pawn into a
Queen) 2 9 . . . QcS 3 0.Nd4 BdS (3 0 . . . Bc8 lets Black play on for a while)
3 1 .Rb l , and the game would be more or less over.
29.Qd8 mate !

359

My son, David, and I play an unusual form of chess, 1 9 80.

eoPle ;n the wodcing ,ectoc ",",I1y m,ke , coo<c;o", effon to

retire, I don't think chessplayers fade away in quite this manner. Instead, we qui
etly play less and less (almost without noticing it!) until all tournament activity
simply ceases.
In my case, 1 974 hit me hard with the death of my father. We were very close,
I was with him throughout his illness, and I had enormous trouble getting over
his loss. Perhaps this caused me to look around a bit, or to become more intro
spective . Whatever the catalyst was, I realized that the constant globetrotting
was wearing me down, that the mass of opening analysis was taking the fun out
of the game, and that the enormous number of "newly minted" grandmasters
(now a horribly watered-down title) made tournaments unpleasant and unim
portant.
My daughter Palma was born late in 1 969, and my son David joined us in
1 97 1 . With a wife and two children in Budapest and responsibilities in the United
States, the need to play in tournaments began to fade - my time was better spent
away from the board.
Of course, I didn't stop right away. It was a long process to wean myself from
the battle cry of competitive chess. At first I played in the U. S . Championship
(out of habit), Lone Pine (the location was wonderful), and the odd interna
tional event from time to time, but every year saw me compete in fewer and
fewer tournaments.
Though I played in a few team tournaments and special events right up to 1 998,
my final tournament was the World Senior Championship, Bad Worishofen, Ger
many, in 1 992 . Here I was doing quite well, and the last-round computer pairing
was placed up the night before for all to see: Benko as White versus Krogius (an
old commie guy from way back). A win would give me a piece of first, and I en36 1

P roud father with

G izella and newly
born daughter
Palma. 1 9 69.

tered the tournament hall ready for battle. Imagine my surprise when I got to my
board and saw Krogius sitting in front of the white pieces! The director had
changed colors for reasons that were never made clear and, in disgust, I offered a
draw rather than make a scene. Krogius refused, so I had no choice but to protest.
A huge argument ensued, insults and accusations were hurled in all directions,
and when the smoke cleared I had the white pieces after all.
By now I was so upset I couldn't think; I wanted to be done with the whole
distasteful affair. We both played a few meaningless moves, he offered a draw,
and I accepted. This left me in a tie for second with Lein, Geller taking first
with half a point more. Later I wished that I had played to win- nobody wants
to end his career on such a sour note. I take some solace in winning my last
serious game (a team event) against Grandmaster Barbero, but I was no longer
the player I once was, and it was with a breath of relief that I officially stopped
all competitive play.

( 1 20) Benko - Commons (U.S. Championship, 1 974)

l .NO Nf6 2 . c4 e6 3 .g3 d5 4.Bg2 Be7 5 .0-0 0-0 6.d4 Nbd7

Deciding on the Closed Catalan. Of course, 6 . . . dxc4 was also possible,

when I would have played 7 . Qc2 , regaining the pawn. An old trap is 7 . . . b S ?
(Best i s 7 . . . a 6 8 . Qxc4 bS 9. Qc2 B b 7 , when Black's pieces are active but his
structure is a little loose , especially the cS -square and the c7 -pawn.)
8.a4 c6 9.axbS cxbS l O.NgS ! .
7.Qc2 c6

This reinforces his center and usually intends . . . b7 -b6, . . . Bb7, . . . Rc8 with
an eventual . . . c6-cS explosion. In the present game, Black chooses an
alternate plan that involves developing the light-squared Bishop to a6.
8.Rd l ! ?

At the time, 8 . Rd 1 was a relatively new move (8 .Nbd2 , 8.Bf4 and 8.b3
were all more common). I first saw 8.Rd 1 at the 1 97 3 Olympiad in the

RETI R E M E N T

game Portisch-Spassky. It seems to be a useful waiting move directed against

the possible S . . . bS variation. In fact, that game continued: S .Rd 1 bS 9.cxbS
(In later years, 9.cS became a more common choice.) 9 . . . cxbS 1 O. Qc6 RbS
1 1 .Bf4 Rb6 1 2 .Qc1 Ra6 (Petursson had this position as White twice in
the same tournament! In both, Black tried 12 ... NeS , but White came out
on top in both instan c e s : 1 3 . Nbd2 B b 7 1 4 . N b 3 Rc6 1 S . Q d 2 b4
1 6 . Rac 1 Qb6 1 7 .Ne l [Or 1 7 .Ne 5 Nxe S I S . dxe 5 Qa6 1 9 . B e 3 Qxa2
2 0.NcS BcS 2 1 .Nd3 Qa4 n .Ra l QbS 2 3 .Rxa7 Rc7 24.Rda l Rxa7 2 S . Rxa7 ,
and White went on to win in Petursson-H. Olafsson, Reykjavik 1 97 5 .]
1 7 . . . Nef6 I S .Nd3 Ne4 1 9.Bxe4 dxe4 2 0.NdcS BcS 2 1 .Nxe4 RdS n .dS exd5
2 3 . Rxc6 Qxc6 24.Qxd 5 Qc2 2 S .Rc 1 Qxe2 2 6 .Nd4 Qxb2 2 7 .Nc6, 1 -0,
Petursson-Larsen, Reykjavik 1 97 5) 1 3 .Nbd2 ! Bb7 1 4.Nb3 (This Knight
move, taking aim at c S , would not be possible if White had played S .b3 .)
14 . . . Rc6 I S .Qd2 Qb6 1 6 .NeS RccS 1 7 .Rac 1 b4 1 S. RxcS RxcS 1 9.Rc 1 Rxc 1 +
2 0 . Qxc 1 Nxe S 2 1 . Bxe S , and the game was agreed drawn, (the team
captains decided that this was best for both teams), even though White
stands a bit better due to Black's weak queenside pawns.
8 . . . b6 9.b3

Zagoriansky's interesting plan, devised in the 1 940s, is: 9.a4 Ba6 (9 . . . Bb7
1 0.aS bxaS I I .cS) 1 0.b3 RcS I I .aS, with better chances for White. Of course,
the move I chose is also quite good.
9 . . . Ba6

Black's light-squared Bishop is usually a problem in the Catalan, and the

text is a way of trying to find a useful diagonal. If 9 . . . Bb 7, the game could
continue, as suggested by Commons, 1 O.Nc3 RcS I l .e4 c5 1 2 .exd 5 exd S
1 3 .Bb2 dxc4? (Better is l L .ReS, when White would only obtain a slight
advantage .) 1 4 . d S ! cxb 3 (Though Commons noticed 1 4. d S in 1 9 74,
Kasparov was the fi rst to use it in actual play: 14 . . . N e S I S .Nb5 a6
1 6.Na7 Rc7 1 7 .Nc6 Bxc6 1 S .dxc6 Rxc6 1 9 .NeS Rd6 2 0.Nc6, Kasparov
Kamsky, New York 1 9S9.) 1 S .axb 3 , and White's Rook is very useful on d l .
1 0.Nbd2 b5

Another plan is 10 . . . RcS intending . . . c6-cS .

1 1 .c5

I d e c i d e d to close the queenside for the time being. Possible is

I l .Bb2 bxc4 1 2 .bxc4, but Black gets active play on the b-file with 12 . . . RbS .
I played Black against Pytel in a similar position a t Skopj e 1 97 2 , and I had
the feeling that cS would have given White the better of it. This game
allowed me to see if this assessment was correct.
1 1 . . .b4 1 2 .e3

The seemingly active 1 2 .e4 isn't good due to 12 . . . dxe4 1 3 .Nxe4 Nxe4
1 4.Qxe4 NxcS I S .Qc2 Nd7 1 6 . Qxc6 NcS ! .
1 2 . . . Qc7 1 3 .a3

White tries to open the a-file and label a7 as a weakness.

363

PAL B E N KO : MY L I F E , GAM E AN D COM POSITI O N S

1 3 ... bxa3 1 4.Bxa3 B b 5 1 5 .Bb4 Rfe8 1 6.h3

3 64

This move stops Black from answering a dxe5 capture (after he tries to
break free in the center with . . . e6-e5) with . . . Ng4. However, it wasn't
really necessary to stop this, so 1 6.Bc3 was a serious alternative: 1 6 . . . e5
1 7 .dxe5 Ng4 1 8 .Nd4! (suddenly e5 -e6, Bxd 5 , and Nxb5 are all threatened)
1 8 . . . Bxc5 1 9.Nxb 5 ! cxb5 20.Bxd5 Rac8 2 1 .Bxf7 + ! Kxf7 2 2 . Qf5 +.
16 ... e5

After the passive 16 ... a6, there would follow 1 7 .Bc3 followed by dou
bling on the a-file.
1 7.dxe5 Nxe5 1 8.Nxe5 Qxe5 1 9.Bc3 Qc7 2 0.b4 Nd7 2 1 .Qb2 Bf6

Black is still trying to bring his Knight to e 5 . I thought about 2 2 .e4! ? for a
long time here. It looks pretty good: 2 2 . . . dxe4 2 3 .Nxe4, opening the g2 Bishop's diagonal and eyeing the d6-square with the Knight. Now I had to
make a difficult decision: should I enter the complicated lines that follow
2 2 .e4, or should I retain a firm grip on the position with the safer 2 2 .Bxf6?
2 2 .Bxf6

I decided to be conservative and not allow my young opponent tactical

opportunities, especially since I was already getting short of time. Despite
these mature considerations, during the game I felt that I was missing the
sharpest continuation. Here's a taste of what I had to look at when decid
ing upon my 2 2 nd move: 2 2 .e4 Ne5 2 3 . exd5 Nd3 2 4. Qc2 cxd5 (The crazy
positions resulting from 24 . . . Nxf2 2 5 .d6 Nxd l 2 6. dxc7 Nxc3 aren't quite
sound, but they are fun to analyze.) 2 5 .Bxf6 (Tempting is 2 5 .Bxd 5 , but the
surprising 2 5 . . . Re2 ! suddenly gives Black a nasty attack.) 2 5 . . . gxf6 2 6 . Qc3 !
(Stronger than 2 6.Qb3 Re2 ! 2 7 . Qxd5 Rd8 2 8 .Rxa7 Rxd5 2 9.Rxc7 Nxf2 )
2 6 . . . Re2 , and though White must be considerably better, there are tricks
all over the place. For example: 2 7 . Qxf6 (Black also puts up a fight after
2 7 .Nfl Nxf2 2 8 .Rxd5 Bc6 2 9 . Rd2 Rae8) 2 7 . . . Rxf2 2 8 .Qg5+ (2 8.Qd6 ! ?)
2 8 . . . Kh8 2 9 . Qxd5 (Too greedy, but Black can resist even after 2 9.Nfl Re8
3 0. Qxd5 Ree2 3 1 .Bf3 Rxfl + 3 2 .Rxfl Qxg3 + 3 3 .Bg2 Nf4 3 4.Rxf4 Qxf4)
2 9 . . . Rd8 3 0 .Rxa7 Qb8 3 1 .Rb7 Qc8 . I think the reader can understand why
I decided on the practical 2 2 .Bxf6.

RETI R E M E N T

22 . . .Nxf6 2 3 .Nb3 a6 24.Nd4 QeS

White is obviously better. His Knight is very strong on d4, aiming at the
weak c-pawn. But it will still be necessary to open a file so my pieces can
invade into Black's position.
2 S.Ra2 Be4 2 6.Ra3 hS 27.Qe2 BbS 28.Rdal Rae8 2 9.NfS

This is not really a part of White's strategic plan, but we were both in
time pressure and I thought that a few tactical tricks couldn't hurt. If ! had
more time, I would have given serious consideration to 2 9 . Rxa6 Bxa6
3 0.Rxa6, when Black can't defend the c-pawn.
29 ... Red8 3 0.h4 Re7 3 1 .Nd4 g6 3 2 .Bf1 !

This is the way to make progress - I had to get rid of Black's good de
fending Bishop - though it's a bad Bishop, it's holding on to Black's weak
pawns on a6 and c6. Many years after this game, Grandmaster Sub a said,
"A bad Bishop often guards good pawns. "
3 2 ... Ne4 3 3 .BxbS axbS 3 4.Ra8 Q e 8 3 S .Rxd8 Qxd8 3 6.Ra6 Qe8 3 7.Qa2
Kh7 3 8.Ra8 Qh3 39.NfJ Qg4 40.Kg2 Nf6

Here I sealed my move and, as usual, it was not the best even after long
thought. There is such a thing as time-pressure hangover!
4 1 .Qb2 ? !

I f I wanted to place the Queen o n the a l -h8 diagonal, then 4 1 . Q a l would

have been more accurate. However, the best move is 4 1 . Qa5 ! , when Black's
game simply falls apart: 4 1 . . .Rd7 (4 1 . . .Re7 42 .Qd8 Re8 43 . Qxf6 ! wins on
the spot, while 4 1 ...Rc8 42 .Ra7 also leaves Black with no hope since 42 ... Kg8
fails to 43 .Rxf7 !) 42 .Qb6 ! , and Black has no defense.
4 1 ...Kg7 42 .QeS Qd7 43 .Nd4 Re8 44.Rxe8

As can be seen by my comments after Black's 44th move, I thought this

won easily. If I had realized this wasn't the case, I'd have looked for some
thing else and found 44. Ra6 ! (threatening Nxb5), when 44 . . . Re8 45 .Qd6
gives White a winning endgame.
44 ... Qxe8

I have to admit that I did not analyze this position. Since Black is completely
tied up, I was sure it would be an easy win. Now, when I started to look at it,

365

PAL B E N KO: MY LI FE, GAM E A N D COM POSITI O N S

I found that things were not so easy. One reason is that 45 . Qd6 can be an
swered by 45 . . . Qe8, threatening perpetual check and other unpleasant things.
So, the first thing I had to do was stop this potential check on e4.
366

45.Kgl ! Qa8

Black tries to activate his Queen. If 45 . . . Qd7, White wins by 46.Qd6. Now
46.Qd6 or Qe7 are possible and would probably win, but Black could still
find ways to make trouble. Mter some searching, I found the most effi
cient way to finish the game.
46.g4!

This kingside breakthrough works despite the limited material.

46 ... hxg4 47.h5

Now White threatens 48.h6+ or 48.hxg6, winning the Knight. Of course,

47 ... gxh5 loses to 48 .Qg5+.
47 ... Qd8

Black tries to confuse matters, speculating on (what else?) my shortage of

time. If 47 . . . Qh8 48.hxg6 Kxg6 (48 . . . fxg6 49.Ne6+ Kh6 5 0 . Qf4+ Kh7
5 1 .Ng5 + Kg8 5 2 . Q b 8 + Kg7 5 3 . Q c 7 + mates or wins the Queen.)
49.Qf5 + Kg7 (49 . . . Kh6 50.Qf4+ Kg6 5 1 .Nxc6) 50.Qg5+ Kf8 5 1 .Nf5 , and
Black is in Zugzwang: 5 1 . . . Ke8 ( 5 1 . . . Ne8 5 2 . Q e 7 +) 5 2 .Ng7 + Kf8
5 3 .Qxf6 Qxg7 54.Qd8 mate.
48.hxg6 Kxg6 49.Nxc6 Qh8 50.Ne7+ Kh7

If 5 0 . . . Kh6, even 5 1 . Qxf6+ wins.

5 1 .Qf5 + Kg7 52 .Qg5 + Kf8 5 3 .Nf5

Arriving at the same Zugzwang position seen in the note to Black's 47th
move, but this time without Black's c-pawn. The rest needs no comment.
5 3 . . . g3 5 4 . fxg3 Ke 8 5 5 .Ng7 + Kd 7 5 6 . Qxf6 Qh3 5 7 . Qxf7 + Kc6
58.Qe6+ Qxe6 59.Nxe6 Kd7 60.Nd4, 1 -0.

( 1 2 1 ) Benko - Evans (U.S. Championship, 1 974)

White to Move

RETI R E M E N T

We arrived at this position after a long, hard, imperfect struggle . Though

Black's game looks solid enough, White still retains some chances for suc
cess. The first order of business is to push Black's King to a passive square.
3 67

4 1 .Re4+ Kd6 42 .Nc4+

Of course, after 42 . Rd4 White can trade down into a King and pawn
endgame, but that would leave the black King in an active position. White's
plan is to make his two pieces as active as possible, tie Black down to the
defense of a7 and f7 , and then to centralize his King (always a major idea
in any endgame) .
42 ... Kc7 43 .Re8 Re7 ! 44.Ra8 !

On 44.Rxe7+ Nxe7 45 .Ne5 Nf5 (much better than 45 . . . Kd6 46.Nxf7+ Ke6
47.Nh8, when White should win) 46.e4 Nd6 Black's f-pawn is defended,
but White is still better. However, would I be able to win this position? I
felt my chances were much greater if ! retained the Rooks - after all, my
Rook is more active than Black's, so it doesn't make a lot of sense to swap it.
44 ... Kc6

This looks like a good centralizing move, but actually 44 . . . Kb7 was better,
after which I'd play 45.Rf8 , with some advantage.
45 .e4 Nc3 ?

Black's last chance was 45 . . . Kb7 , when Evans was worried about 46.Rf8
Nc3 47 .Nd6+ Kc6 48.Nxf7 Nxe4 49.Nh8, when White would win. How
ever, I think he would have had drawing chances in this line if he played
his King to c7 instead: 47 . . . Kc7 ! 48.Nxf7 Rxe4 49.Nh8 Rxh4, when Black

gry Evans: I

played Evans at a 1 960

blitz tournament in Argentina. He had

promo ted his pawn but he didn't replace it with a Qu e en , so the " sup e r- powered
pawn"

was umpin g

all over the board. We were

both very low on time and I tried to

stop the clock and give him a Queen in exchange for his l e apin g pawn (there's noth
ing worse than having a supposed pawn fly across

the board at you!). He didn't know

why I was trying to stop the clock and he hit my hand. I said, "Be careful! " and came
close to striking him. Seeing the look in my eye s, he replied, "Yes, you 'll be famous
for beating all the best players in the world . . . physically! "

I battle Evans (left)

while Mrs. P i atigorsky
watches.

(Photo Art Zellec courtesy

USCF.)

PAL B E N KO: MY LI FE, GAM E AN D COM POSITI O N S

368

is still very much alive. Preferable after 45 . . . Kb7 was 46.Rd8, and now
46 ... Nc3 47 .Rd3 Nb5 (not 47 ... Nxe4? ? 48.Re3) 48.a4 Nc7 (better is 48 . . . Rc7,
but White stays on top with 49.Nd6+! Nxd6 50.Rxd6 Rc3 5 1 .Kf4 Rh3
52 .Rf6 Rxh4+ 5 3 .Ke5 Rg4 54.Rxf7+ Kb8 5 5 .Kf6 Rxe4 56.Kxg6, and White's
g-pawn wins the game) 49.Nd6+! Kb8 50.Rc3 leaves Black in a position
where he can't move a muscle. The Knight move (45 . . . Nc3 ?) was an act of
desperation since Evans thought that he was losing. He felt that his best
chance lay in a Rook ending, even a pawn down.
46.Rc8+ Kd7 47.Nxb6+ axb6 48.Rxc3 Kd6 49.a4 Ke5

After 49 . . . Ra7 5 0.Ra3 followed by Kf4 and/or a4-a 5 , White should win
without any difficulty because his Rook is already behind his passed pawn.
50.Ke3 Rd7 5 1 .Rd3 Rc7 52 .Rd5+ Ke6 5 3 .Rb5 Rc6 54.Kd4

The game is over. An extra pawn and a dominating King is more than
Black can handle.
54 ... Rd6+ 5 5 .Rd5 Rc6 56.e5

Intending 5 7 . Rd6+ Rxd6 5 8 . exd6 Kxd6 59.a5, when White's King walks
over and picks off all of Black's kingside pawns.
56 ... Rc 1

Even though this avoids the trade of Rooks, Black is still hopelessly lost.
57.Rd6+ Ke7 5 8 .Rxb6 Rhl 59.a5 ! Rxh4+ 60.Kd5 Rhl 6 1 .Rb7+ Kf8
62 .a6 Ra l 63 .a7 Kg7 64.e6, 1 -0.

( 1 22) Benko - Gi lden (U.S. Championship. 1 974)

l .Nf3 c5 2.c4 b6

My opponent, a highly original player, likes to avoid mainstream continu

ations. Probably best now would be 3 . d4 (also promising is 3 .Nc3 Bb7 ,
when both 4.e3 and 4.e4 give White the better chances) L . cxd4 4.Nxd4
Bb7 5 . B ! , with transposition into some sort of Maroczy Bind Sicilian.
3 .g3

A modest continuation that shows I have no interest in trying to refute

Black's second move.
3 ... Bb7 4.Bg2 g6

This double fianchetto is quite reasonable, and more in keeping with his
style than the popular Hedgehog formation that comes about after 4 . . . Nf6
5 .0-0 e6 6.Nc3 Be7 7 . d4 cxd4 8 . Qxd4.
5 .0-0 Bg7 6.Nc3 Nf6 7.d4 cxd4 8.Nxd4

Possible is 8 .Qxd4, but in Panno-Ljubojevic, Petropolis 1 9 7 3 , Black got a

very active position after 8 . . . Nc6 9.Qf4 (9. Qh4 is met by 9 . . . h6) 9 . . . Rac8
l O. Rd l Nh5 1 1 . Qe3 Nb4 ! .
8 ... Bxg2 9.Kxg2 Qc8 1 0.b3 Qb7+ 1 1 .Kgl ! ?

RETI REM ENT

3 69

In the game Capablanca - B o tvinnik, Nottingham 1 9 3 6 , 1 1 . f3 d S !

1 2 .cxdS NxdS 1 3 .NxdS QxdS 1 4.Bb2 ( 1 4.Be3 ! is stronger) 1 4 . . . 0-0 was
played. As the tournament book (by Alekhine) says : "White's Knight is
uncomfortably pinned. The next moves are therefore practically forced. "
The continuation was 1 S .Qd3 Rd8 1 6.Rfd 1 Nd7 , and White had to fight
for a draw, which was agreed on the 2 9th move. Fortunately, I didn't re
member that game, but my opponent did. Had I been aware of it, I might
have been influenced by the great names. However, my instincts told me
not to loosen my King position with 1 1 .f3 .
1 1 ...dS?

Had my opponent not been under the influence of the Capablanca

Botvinnik game, perhaps he would not have made this move. This central
action makes sense after 1 1 .f3 because of White's threat to set up a Maroczy
Bind with e4. However, in the present case White isn't threatening to play
e4, so Black should have continued his development by 1 1 . . . 0-0. White
can't prevent . . . d S , so Black should play it at the safest moment.
1 2 .cxdS NxdS 1 3 .NxdS QxdS 1 4.Be3 !

White doesn't go into the pin with the "natural" 1 4.Bb2 . The Bishop is
much stronger on e 3 , looking at Black's queenside pawns and keeping the
option of going to h6 if mating opportunities should turn up. If we exam
ine the position closely, we can see that Black is already in trouble, mainly
because he's behind in development. If 14 . . . 0-0 1 5 .Rc l , and Black's Knight
cannot develop normally because of the weakness of c6 (on 1 S . . . Rd8, 16.NbS
is a strong continuation). Moral : It is better to understand the openings
than to know them!
1 4 ...Nc6 l S .Nxc6 Qxc6 16.Rc l Qe6

Black tries to simplify, but he is still behind in development and will have a
problem defending his second rank. Therefore, 1 6 . . . Qb7 was more prudent.
1 7.Qd3 hS ? !

Black did not like 1 7 . . . 0-0 (which is a better move) 1 8 . Rfd 1 , intending
1 9.Qd7 with unpleasant designs on Black's position. So he tries to keep
his King close to the center and bring his Rook into play artificially.
1 8 .QbS + Kf8 1 9.Rfd l h4 20.QdS

PAL B E N KO : MY LI FE, GAM E AND COM POSITI O N S

Forcing the trade o f Queens and ending Black's hopes for counterplay along
the h-file.
20 ... Qxd5 2 1 .Rxd5 hxg3 2 2 .hxg3 Rh5 2 3 .Rxh5 gxh5 24.Rc4!
370

The Rook threatens to go to either side, and there is no way to avoid the
loss of a pawn.
24 ... Rd8 2 5 . Ra4 b5

He might as well give up the pawn right away because 2S ... Rd7 2 6 . Bxb6 !
wins two o f them. Note how strong the Bishop i s o n e 3 , where i t aims at
both sides of the board at the same time.
26.Rxa7 Bd4 27.Bxd4 Rxd4 2 8.Ra5 Rb4?

I have to admit that I didn't even dream about this move, practically trap
ping his own Rook. He should have tried the more active 28 . . . Rd2 (2 8 . . . b4
is also possible), but in any case he will lose another pawn (i. e . , 28 . . . Rd2
2 9.e3 b4 3 0.RxhS Rxa2 3 1 . RbS). However, as in most Rook endgames,
careful technique would still be required to win.
2 9.Kfl e5 30.e3

White's plan is clear: The King will walk over to c3 and force the win of
the b-pawn. The problem with Black's 28 ... Rb4 is that his Rook is pas
sively placed, and if you don't have some form of active counterplay in
this kind of position, then it's all over.
30 ... Ke7 3 1 .Ke2 Ke6

Giving in to his fate. An attempt to hold onto the b-pawn fails: 3 1 . . .Kd7
3 2 .Kd3 Kc6 3 3 . Ra6+ KcS 34.f3 ! (threatening to trap the Black Rook by
3 S .Kc3) 3 4 . . . h4 3 S .g4 e4+ 3 6 .fxe4 h3 3 7 .Rh6, when the comical situation
of the Black Rook is suddenly obvious.
3 2 .Kd3 Kd5 H .Kc3 Rg4

Of course, 3 3 . . . KcS is hopeless in view of 3 4.f3 ! (taking all the escape squares
away from Black's Rook. A less enjoyable way to win is 3 4.a3 Rg4 3 S .a4)
3 4 . . . h4 3 S . Ra8 hxg3 3 6. Rc8+.
34.Rxb5+ Ke4 3 5 .Kd2

Black's final hopes rest on the kingside. By bringing my King there to

defend f2 , my opponent will be left with nothing to play for.

RETI REM ENT

37 1

Vs. Sines (left) ,

Su botica 1 974.

SINES

3S ... f6 3 6.Ke2 Rg8 3 7 .Rb4+ KfS 3 8.Rc4 Ra8 3 9.a4 Rb8 40.b4 Ra8
4 1 .bS, 1 -0.

( 1 23) Benko - Cosulich (Venice, 1 974)

l .c4 eS 2 .g3 Nf6 3 . Bg2 Nc6 4.NfJ

White deliberately delays Nc3 , since then Black could play . . . Bb4, adopt
ing a very popular system that, incidentally, my opponent enjoyed using.
4 ... BcS S .d3 0-0 6.0-0 d6 7.Nc3 Bd7
An unusual move which showed me that Black doesn't know this system.

More common are 7 ... a6 (giving the Bishop an escape hole on a7) and
7 . . . h6 (stopping BgS).
8.e3 Bb4? !

Obviously, Black has not achieved what he wanted with his . . . BcS , so he
tries to get back to positions he's familiar with.
9.Ne2 ! ?

Not much advantage was promised by 9 .NdS NxdS 1 O.cxdS Ne7 , since I
can't recommend I l .Qb3 followed by Qxb7. However, the simple 9.Bd2
was possible. I thought the text had some psychological motivation, since
Black will have trouble with his b4-Bishop later.
9 ... e4

Trying to secure the position of his dark-squared Bishop, but now the
center is opened advantageously for White.
1 0.dxe4 Nxe4 1 1 .Qc2 Re8 1 2 .a3 BcS 1 3 .b4 Bb6 1 4.Bb2 Qe7 ? !

Black already stands badly since the position o f his Knight o n e 4 i s far from
secure. On . . . BfS White would answer Nh4, so perhaps Black should have
considered 14 . . . Nxf2 , although White's two minor pieces may prove stron
ger than Black's Rook and two pawns (after I S .Rxf2 Bxe3). Even capturing
with the King is to be considered: 14 . . .Nxf2 l S .Kxf2 Bxe 3 + 1 6.Ke l , when
Black doesn't seem to have enough compensation for the sacrificed piece.
l S .Nf4! Nf6?

PAL B E N KO : MY LIFE, GAM E AND COM POSITI O N S

Black's game collapses after this, but things didn't look good anyway. For
example, I S , . .Bg4 1 6.Nh4! intending NdS is very strong.
1 6.Ng5 !
372

Decisive. Black no longer has a way out.

1 6 ... Ne5

The most interesting lines come about after 1 6, . .g6 1 7 .BdS Ne S ! (Both
1 7 , . .Nd8 1 8 .Nxg6 and 1 7 , . . Rf8 1 8 .Nxg6 lose immediately.) 1 8 .Bxf7 + ! !
( 1 8.Nxg6 Ieads to some fun lines, but unfortunately i t doesn't work: 1 8 ,. .hxg6
1 9.Bxf7 + Kf8 ? ? [The simple 1 9, . . Qxf7 ! 20.Nxf7 Kxf7 spoils White's con
cept.] 20.BxeS QxeS 2 1 . Qxg6 Red8 n .cS ! dxcS 2 3 .Ba2 ! Qe7 24.Nh7+ Qxh7
2 S . Qxf6+ Ke8 2 6 . Bb l , and the game is over) 1 8 , . .Nxf7 1 9 .Nxh7 NeS
( 1 9 . . . BfS loses to 2 0 . QxfS !) 2 0 . BxeS Nxh7 2 1. Qxg6+ Kf8 n . Qh6+ Kg8
2 3 .Ng6 Qxe S (A key line is 2 3 . . .Qf7 2 4.Nh8 Qe7 2 S . Qg6+ Kf8 2 6 .Bf4,
with a quick victory.) 2 4.Nxe S , and White's Queen and three pawns should
triumph over Black's three minor pieces thanks, in part, to the continued
vulnerability of Black's King.
1 7.Nd5, 1 -0.

Black resigned rather than face 1 7 .NdS Qd8 1 8 .BxeS dxeS 1 9 .Nxf6+ Qxf6
2 0 . Qxh7 + Kf8 2 1 .Ne4 with c4-cS to follow.
My feelings were mixed when I learned before round two that AlIa Kushnir
would be my opponent (she was the number two woman player in the world for
many years and played three World Championship matches with Nona
Gaprindashvili) . The pleasure of defeating her, as I expected to do, was some
how tinged with the perhaps unreasonable fear that I may draw or even lose,
especially in view of her win in the first round over Grandmaster Larry Evans. I
cringed as I imagined the teasing I would have to endure : "You lost to a woman? "
I f I did go down in defeat, how could I justify the accustomed parading o f my
inflated male ego, especially when I teach or coach women players? All of this
zips through the mind very quickly, of course. Then there is nothing left to do
but quell the chatter of all thoughts and simply play the game.

RETI REM ENT

( 1 24) Ben ko - Kush nir (Lone Pine, 1 975)

l .c4 Nf6 2 .Nc3 e6 3 .NB b6 4.g3

Korchnoi popularized 4.e4 Bb7 5 . Qe2 in his first match with Karpov, us
ing it three times (though each game ended in a draw) . He used it later to
beat both Petrosian and Portisch. I avoided this line because I knew my
opponent was quite conversant with current theory. Why walk into a
minefield when there's no need to do so?
4 ... Bb7 5 .Bg2 d5 6.cxd5 exd5 7.0-0 Be7

Of course, 7 . . . d4? ? fails to 8 .Qa4+ Nc6 9.Nxd4.

8.d4 0-0 9.Bg5

An obvious developing move, but 9.Bf4, 9.Ne 5 , and 9. Qc2 would all pre
serve more tension. Now Black does not hesitate to trade off her good
Bishop in order to free her position.
9 . . .Ne4 1 0.Bxe7 Qxe7 1 1 .Rc 1 Na6 !

This move is better than the more natural 1 1 . . .Nd7 because the Knight
has a better future from a6 (i.e., it defends the c-pawn, and it can go to b4
or, after the c-pawn advances, to c7). Now White has the problem of find
ing a good spot for the Queen, for if 1 2 . Qa4, then after 1 2 . . . Nxc3 White's
e-pawn is hanging while Black's c-pawn is well protected. At this point I
began to realize that I was up against a worthy opponent who understood
exactly what she was doing and why.
1 2 .e3 Rfd8 1 3 .Qa4 Nb4 14.Rfd l

I didn't want any part of 1 4. a 3 N d 3 1 5 . Rc2 Nxb2 1 6 .Nxd 5 Bxd5

1 7 .Rxb2 c 5 .
1 4 . . . a5 1 5 .a3 Na6

Here I was worried about the sharper 1 5 . . . c 5 , on which I spent consider

able time analyzing the consequences. However, my opponent played the
text without much thought and I was annoyed at having used up so much
of the clock. Kushnir had spent a great deal of time on the whole opening
so I felt it would be all right for me to "catch up, " believing that I would
outplay her in mutual time pressure. Here I began to get the feeling-

373

PAL B E N KO : MY LIFE, GAM E AN D COM POSITI O N S

374

which was later confirmed - that m y opponent was deliberately avoiding

sharp tactics, preferring to play strategic chess according to her style. Per
haps this is the reason Gaprindashvili, who is known to be a fine tactical
player, was able to defeat Kushnir in all of their matches.
1 6 .Ne5 Rd6 1 7.Qc2 Nxc3 l S.bxc3 ! ?

The obvious 1 8 . Qxc3 gives White a slight advantage because, sooner or

later, Black will have to play . . . c7 -c5 and will end up with the so-called hang
ing pawns. With 1 8.bxc3 (which threatens to use the pawn actively via c3 c4), I decided to try for a bigger plus and a more complicated position.
l S ... b5?

I expected my opponent to understand that my strategic plan called for

White's c3 -c4. What she did not realize is that for tactical reasons the text
move doesn't prevent that advance. This confirmed my earlier assessment
of her style.
19.c4! c6

If 1 9 . . . bxc4 2 0.Nxc4 and, of course, 2 0 . . . Rc6 2 1 .Bxd 5 .

20.cxb5 cxb5 2 1 .Qb3 b 4 22 .a4?

Just when White has obtained a clear advantage, he gives up his original
plan, which called for 2 2 . axb4 Nxb4 2 3 .Rc5, going after the two weak
nesses that I created in Black's camp - the d 5 - and the a5 -pawns. The text
overestimates White's chances on the Bishop file and gives Black a pro
tected passed pawn. White's play on the c-file can be only temporary, while
Black's passed pawn (and the weakness of White's a-pawn) is permanent.
It is psychologically interesting that just when I caught my opponent, I let
her slip out, as I was to do again later in the game. I attribute this to my
intention to play " even better" against a woman than I would against a
man. The result of this psychological mistake was my failure to trust my
own judgment.
2 2 . ..QeS!

Black doesn't fall for 22 ... f6 2 3 .Nc6 ! Bxc6 24.Rxc6. The text prepares to
oust the Knight by . . . f7-f6 and takes aim at White's a-pawn.
2 3 .Rd2 f6 24.Nd3 Bc6 2 5 .Ra2 KhS 26.Bfl

This move is the start of a maneuver to post the Bishop on b3 , its best
square for both attack and defense.
2 6 . . .Nc7 2 7.Be2 Ne6 2S.h4

I could not allow . . . Ng5 threatening . . . Ne4-c3 .

2 S ... Ra7

At first I didn't understand this move; in fact, since we were both already
short of time, I thought it was a way of "passing." Later I realized that it
prepares the maneuver . . . Nc7 -a8-b6, and that my opponent was also looking
for effective posts for her pieces. I must give her credit for her fine strate
gic feeling, even in time trouble.

RETI R E M E N T

I watch M rs. Selensky and M iss

Korhonen play a game in 1 9 65. At
that time I didn't have a h igh
opinion of women's chess, but ten
years later my views changed
after playing Alia Kushnir.

(Photo courtesy USCF.)

29.Bdl Nc7 30.Qb l Na8 3 1 .Bb3 Nb6 3 2 .Nc5 Qh5 H .Qd3 Ra8 3 4.Bd l
Qe8 3 5 .Bc2 g6

I would have preferred, with Black, to retain the annoying possibility of

. . . QhS .
3 6.Bb3 Nc4 3 7.Re2

Of course no pawn is won, now or later, by 3 7 .Bxc4 dxc4 3 S . Qxc4 BdS .

Now the threat of e3 -e4 induces Black to weaken her position further.
37 ...f5 3 8.Qd l Nb6

Here we were in horrible time pressure. I was amazed at how well she was
playing, given that only seconds remained on her clock.
3 9.Ra2 Nc4 40.Qf3 Qe7 4 1 .Qf4 Rad8 42 .Rac2 Na3

Probably not the best place for the Knight! However, our moves were still
coming fast and furious, with no time for sane reflection.
43 .Rd2 Ba8 44.h5 Qf6 45 .hxg6 hxg6 46.Kg2

My sealed move, signaling an obvious change of strategy. Now I intend to

build up a kingside attack on the h-file. The other possibility was 46. Qh6+
Kg8 47.Nd3 , to try to invade on the c-file.
46 . . . Kg7 47.Rh l Rg8?

A big surprise. I had been expecting 47 . . . RhS or 47 . . . gS . White would re

tain the advantage against both these moves, but the text gives me an op
portunity for a very strong shot.
48.g4?!

PAL B E N KO : MY LI FE, GAM E AND COM POSITI O N S

376

Not a bad move, but much better was 4S .e4! fxe4 49. Qh6+ Kf7 5 0.Nxe4.
It is hard to explain why I chose the text when I realized that I had the
chance for 4S.e4. When my opponent surprised me with her 47th move,
perhaps I felt that she must have seen the possibility of 4S . e4 and had
devised some sort of antidote. Therefore, trusting that she wouldn't lie to
me, I chose something else.
48 ... fxg4 49.Qxg4 Nc4 50.Re2 Qf5 5 1 .Qg3

Of course, I had no desire to trade Queens when Black's King is in such a

vulnerable position.
5 1 ...Qf6 52 .Rh4 Rh8 5 3 .Rxh8 KxhS 54.Re l

It's time to get my remaining Rook onto the h-file !

54 ... Nd2 5 5 .Rd l !

A nice move that forces Black to make a decision with her Knight.
5 5 ...Nc4

A sad retreat, but 5 5 . . . Ne4 56.Nxe4 dxe4 allows my Bishop to enter the
attack powerfully along the a2 -gS diagonal. Also poor was 5 5 . . . Nxb3
5 6.Nxb3 , since my Knight would prove far stronger than Black's bad Bishop.
56.Rh 1 + Kg7 57.Qh3 Kf7 5S.Qg4 Kg7 59.Rh3

For the attack along the h-file to be effective, I have to lead with my Rook.
This and the next couple of moves sets up that configuration.
59 ... Nb6 60.Qg3 Bc6 6 1 . Qh2 KgS 62 .RhS+, 1 -0 .

I was s o happy t o play this crushing move (62 . . . QxhS 6 3 . Qxd6 wins easily)
that I "neglected" the even stronger 62 .Ne4.

( 1 25) Benko - Pinter (Szolnok, 1 975)

l .c4 g6 2 .d4 Nf6 3 .Nc3 d5 4.NfJ Bg7 5.Bg5 Ne4 6.Bh4

This is the Taimanov Variation, named after the Russian grandmaster who
popularized it. The Bishop on h4 exerts annoying pressure on Black's po
sition.
6 ... c5

Nowadays, most Griinfeld aficionados prefer 6 . . . Nxc3 7 .bxc3 dxc4 S . e 3

b 5 , with a complicated game.
7.cxd5 Nxc3 S.bxc3 Qxd5 9.e3 cxd4 1 O.cxd4 Nc6 I 1 .Be2 e5

This has a bad reputati o n . More common i s 1 1 . . . 0 - 0 1 2 . 0 - 0 e 5

1 3 .dxe 5 Qa5 ( 1 3 . . . Qxd l 1 4. Rfxd l is very nice for White) 1 4.Bf6 Bxf6
1 5 .exf6 Qf5 1 6.Nd4 Qxf6 1 7 .Nxc6, when some books call the position
equal, but a sane look will convince anyone that Black will be suffering for
a long time to come.
1 2 .dxe5 Qa5+ 1 3 .Qd2 Qxd2 + 1 4.Kxd2 Nxe5 1 5 .Rab l 0-0 1 6.Nd4

RETI R E M E N T

377

In my opinion, this endgame is unpleasant for Black. He has problems

finding an active post for his c8-Bishop, the h4-Bishop controls the d8square and the e5 -Knight can be chased away with f2 -f4.
1 6 . . . h6

Intending . . . g6-g5 , blocking White's Bishop and securing the e 5 -square

for the black Knight.
1 7.f4 g5 ! ? 1 8.fxg5 Ng6 1 9.Bg3 hxg5 20.Bc7 Nh4

This stops White from placing his Bishop on B . Now 2 1 .g3 is met by
2 1 . . . Nf5 , when Black has some counterplay. My next move stops this ma
neuver and places pressure on Black's weakened kingside.
2 1 .Rb5 Bf6 22 .Rgl

The immediate 2 2 .g3 is also quite strong.

22 . . .Kg7 2 3 .g3 Ng6

And not 2 3 . . . a6? 24.Rxg5 + ! Bxg5 2 5 .gxh4.

24.Bfl a6 2 5 .Rb6

Black has real trouble finding reasonable moves.

25 ... Bd8 26.Rc 1

White could have won a pawn by 26.Bxd8 Rxd8 2 7 .Bxb7 Rb8 2 8 . Rcb8,
but I wasn't in any hurry. I was sure I'd be able to get even more out of
this position.
26 ... Ra7 27.Be4 Bf6 28.Rfl Be7 29.Rc 1 Bf6 3 0.Bd6 Rd8 3 1 .Rc7 Ne5
3 2 .Bd5

All the White pieces are swarming into the enemy position, and it's only a
question of time before he gets something tangible . The threat is now
3 3 .Bxe5 followed by Rxf7 +.
3 2 . . . Nd7 3 3 .Rb l Be5 34.Ke2

Getting my King away from the stare of the d8-Rook.

34 . . . Bxd6

Black could have put up a bit more resistance by 34 . . . Bxd4 3 5 .exd4 Nf6,
though 3 6 .Be5 Rxd 5 3 7 .Rxc8 is still winning for White.

PAL B E N KO : MY L I F E , GAM E AN D C O M POSITI O N S

3 S.NfS+ Kf6 3 6.Nxd6

White's many threats are overwhelming, and Black would have been ex
cused if he had chosen this moment to resign.
378

3 6 ... RaS 3 7.Nxf7 Rf8 3 S.Nd6

The time for subtlety is over. Now all I have to do is chop off as many of
his pieces as possible!
3 S ... KeS 3 9.NxcS Nf6 40.Nb6 RadS, 1 -0.

He resigned without waiting for a reply. Black also loses a piece after
40 .. .Nxd5 4 1 .Rc5 RadS 42 . Rd l .

( 1 26) Benko - Kopec (World Open, 1 975)

l .c4 Nf6 2 .Nc3 g6 3 .d4 Bg7 4.e4 d6 S .Be2 0-0 6.NB eS

Kopec was hoping to go into the sharp lines that result from 7 . 0-0 Nc6
S . d 5 Ne7 . He couldn't have guessed that I had something very different in
mind.
7.dxeS dxeS S.QxdS RxdS 9.BgS

This game was a must-win situation for me, and it might surprise some
readers to see me trading Queens so early, all the more so when one con
siders that Black is considered to be perfectly safe according to theory.
This type of opening, however, doesn't appeal to a tactician like my oppo
nent, while I have always enjoyed systems that lead to an early exchange
of Queens (Reuben Fine would no doubt have concluded that I harbor a
deep-seated hatred of women) . At times you have to push the dictates of
theory aside and delve into the world of psychology-if the position is
equal but isn't to your opponent's taste, then you're already on your way
to a positive result!
9 ... ReS

The main reply. Black has also tried 9 . . . Nbd7 , 9 . .Na6, 9 . . RfS ! ? , and 9 . . . c6.
.

10.0-0-0

For a long time it was thought that 1 O .Nd5 Nxd5 I l .cxd5 c6 1 2 .Bc4 gave
White chances for a small plus. However, it eventually became clear that
Black was quite alright after 1 2 . . . Nc6, while 1 2 . . . Nd7 ! ? is also interesting.
1 0 ... Bg4

This is not highly thought of and shows that my opponent wasn't familiar
with the positions that arise after 7 . dxe5 . Better choices are 1 O . . . h6, 1 O . . . c6,
and 1 0 . . . Na6.
I 1 .NdS NxdS 1 2 .cxdS fS

An aggressive move that seems to solve all of Black's problems. In fact,

now 1 3 .h3 Bxf3 1 4.Bxf3 f4 leaves White's dark-squared Bishop in trouble.
1 3 .Ngl !

RETI R E M E N T

379

This came as a complete surprise to my opponent, who called it a grand

master move in the post-mortem. Actually, 1 3 .Ng ! (exchanging White's
inferior Bishop for Black's good one) is not only tactically necessary, it is
also strategically desirable.
1 3 ... Bxe2 1 4.Nxe2 Nd7 1 5 .Nc3 BfB

Black begins to feel the weakness of his backward c-pawn and hastens to
protect it. It's too late for . . . c6, since this would only create new weak
nesses.
1 6.Kb l

Preparing to line up my heavy artillery against c7.

1 6 ... a6 1 7.Re l Bd6 1 8 .Re2 Kg7 1 9.f3 b5

Questionable because of the resulting weakness of c6, but my opponent

was worried about Be3 followed by Na4-c5 . This is a case of the cure
being worse than the disease!
20.Rhe l Rf8 2 1 .Be3 Nf6 22.Bg5 Nd7 2 3 .Nd l

The repetition of moves has no significance except as a test of my opponent's

understanding of the position. The text move leaves the c-file open in case
an Exchange sacrifice becomes feasible, and the Knight will stay near f2
to protect the e-pawn.
2 3 . . . Rf7 24.Nf2 h6 2 5 .Be3 Nf6 26.Re6 fxe4 27 .fxe4 g5

My opponent later told me that he made this move so he could bring his
King to d 7 . Nevertheless, that plan never gets carried out.
28.Rle2

White, too, wants to bring his King to the center, protecting the e-pawn
and freeing the Knight.
28 ... a5 29.Ke l Nd7

Black doesn't follow his plan consistently, deciding instead to harass my

Rooks with his Knight. Such opportunistic play is typical of tacticians,
who usually do not think in long-range terms.
3 0 .Kd l Nb8 3 1 .R6c3 Na6 3 2 .Re2 Raf8 3 3 .Re l Re8 3 4.Nd3 Nb4
3 5 .Nxb4 Bxb4

PAL B E N KO: MY LI FE, GAM E A N D COM POSITI O N S

O n 3 5 . . . axb4, my plan was 3 6.Kc2 ! c5 3 7 .dxc6 Rxc6+ 3 8 .Kb3 followed by

Bd2 when Black's b pawn is weak.
3 6.Rf2 RefS 3 7.Rxf7+ Rxf7 3 8.Ke2 Bd6 3 9.Re6 Rf8 40.Be5 ! ?
380

My last move i n mild time pressure . I had to calculate (quickly!) the

coming endgame, and finally decided that the two connected passed pawns
I'd ultimately get would be worth allowing Black's Rook access to the sev
enth rank- even though it would win my two queenside pawns. But there
was no hurry to make this decision, as Black could not avoid it anyway.
After the safer 40.g3 , for instance, Black's Rook would not be so easily
activated and White would still be threatening Bc5 or even Bb6, which
would be very hard for Black to meet.
40 . . . Rf4 4 1 . Ke 3 Rfl 42 . Bxd6 exd6 43 . Rxd6 Re 1 + 44.Kd3 Rd 1 +
45 .Ke2 Rb I 46.Re6 Rxb2 + 47.Ke3 Rxa2 48.Rxe5

We 've reached a very difficult Rook endgame. Can B lack hold this
position?
48 ... Kf6

So far the endgame has followed my original calculation, but this move
does not offer the best resistance. A better try is 48 . . . b4 when Black was
obviously worried about 49.d6. However, he could still stop the d-pawn's
further advance by 49 . . . Ra3 + 50.Kd4 Ra l 5 1 .Re7+ Kf6 5 2 .e5+ Kf5 5 3 .d7,
when: A) 53 ... a4 54.Kc4 wins a pawn; B) 5 3 ... Rc 1 fails to 54.Rf7+, because
54 . . . Ke6 (54 . . . Kg6 5 5 .Rf3 ! is gin) 5 5 .d8=N! creates a pretty Knight-mate;
C) 5 3 . . . Rd l + 54.Kc4 Rc 1 + 5 5 .Kb5 Rd l 56.Ka4, and White's King stops
the pawns while Black can do nothing to prevent e6 and Re8; D) 5 3 . . . b3 is
safely answered by 54.Kc3 .
Since these lines don't allow Black to make a game of it, he should prob
ably play 48 . . . Ra l ! , when 49.d6 (49 .Re6 Kf7 5 0 . Rb6 a4! 5 1 . Rxb5 a 3
52 .Ra5 a 2 5 3 .Kf2 Kf6, and Black has n o problem drawing) 4 9. . . Rd l 50.Rxb5
Rxd6 5 l .Rxa5 gives White winning chances.
49.Re6+ Kf7 50.Rb6 b4 5 1 .e5 Ral

Less experienced players might wonder why 5 l . . . Rxg2 isn't good, not re
alizing that the loss of White's kingside pawns has nothing to do with what's

RETI REM ENT

happening in this position. After 52 . Rb 7 +, Black's capture of the g-pawn

would turn out to be nothing more than a critical loss of tempo.
52 .Rh7+ Ke8

This type of position, with Black's King on the first rank, gives White
mating possibilities and invariably leads to a bad result for the defending
side. Black is lost anyway, but he should try 52 . . . Kg6 5 3 .e6 a4 54.e7 Kf7
5 5 .d6 a3 56.RbS Re l + 5 7 .Kf2 a2 5 S .d7 Rxe7 59.dS=Q a l =Q 60. QgS+ Kf6
6 1 .RfS+ Ke 5 62 .QhS+. White would have an opportunity to go wrong in
this line since some of the variations are quite intricate. However, with
best play White always wins.
5 3 .e6 a4

Too late. White doesn't even have to take the pawn. Mate is stronger.
54.Ke4 Re 1 + 5 5 .Kf5 Rd l 56.Ke5, 1 -0.

Black can't stop both Kd6 and d6.

( 1 27) Benko - Krnic (Sombor, 1 976)

l .c4 Nf6 2 .Nc3 d5 3 . cxd5 Nxd5 4.g3 g6 5 .Bg2 Nb6

This and 5 ... Nxc3 are the most usual moves here. But since White hasn't
committed himself to d4, we are dealing with a different strategy than
usual. Black does not have the opportunity for immediate counterplay in
the center, as he does in the Griinfeld.
6.d3 Bg7 7.h4 ! ?

Black can't answer this aggressive move with . . . h 5 since the g5-square would
be weakened and White could establish a very troublesome Knight there.
The disadvantage of White's move, however, is that his own g4 is poten
tially vulnerable, and Black can try to exploit this.
7 ... h6 8.Bd2

More common is S.Be 3 , but it has a drawback: after . . . e7-e5 or . . . c7-c5

Black puts a Knight on d4 and White won't be able to take it with the KN
due to a resultant pawn fork.
8 ... Nc6 ? !

Giving White the chance to destroy Black's pawn structure o n the queenside.
However, if White took on c6 Black would have compensation in his light
squared Bishop, which could give White great trouble. Although this line
is considered advantageous for White, I adopted it twice without much
success. Anyway, White has other promising strategic plans at his disposal.
9.Qc1

The idea is to delay Black's castling for as long as possible by attacking his
h-pawn.
9 ... Nd4 l O.Nfl c6 l 1 .Ne4 Nd5 l 2 .Rh l

38 1

PAL B E N KO : MY LI FE, GAM E AN D COM POSITI O N S

Now 1 2 .Nxd4 Bxd4 1 3 .Bxh6 is a mistake due to 1 3 . . . Qb6 ! , threatening

the b-pawn and, after . . . f7 -f5 , also the f-pawn.
12 ... Qb6 1 3 .Nxd4 Bxd4 14.b4 Bg4
382

An interesting position has arisen in which each side attempts to obstruct

his opponent's castling as long as he can. The theoretical battle will be
won by the player who is first able to castle safely.
1 5 .Nc5 Rd8 1 6.Qc4 Nf6

Worse was 1 6 . . . e5 1 7 .0-0 ! , threatening I S .e3 .

1 7.Rc1

Premature was 1 7 .e3 because of 17 . . . Bxc5 I S .Qxc5 Rxd3 . Therefore, White

prepared the possibility of taking on c5 with the pawn by bringing his
Rook off the loose b I -square .
1 7 . . . Kf8

White threatened Nxb 7 , so Black avoids it and castles by hand. An inter

esting alternative is 1 7 . . . Be6 ! ? , when I S .Qc2 (It's only a draw after I S .Nxe6
Bxf2 + 1 9.Kd l fxe6 2 0 . Qxe6 [threatening 2 1 .Be4] 2 0 . . . Rd6 2 1 . QcS+ RdS
2 2 . Qe6.) I S . . . Bd5 1 9.0-0 Bxg2 20.Kxg2 leads to a slight edge for White
thanks to his queenside superiority.
1 8.e3 Bxc5 1 9.bxc5 Qb2

Black still tries to prevent White from castling, this time by throwing his
Queen into the front lines.
20.Rc2 Qb 1 + 2 1 .Bc1 Be6 22 .Qc3 Qb5 2 3 .Rd2 Kg7 24.0-0

RETI REM ENT

Thus the opening has been concluded in White's favor. He has safely castled
and has a mobile center as well as the two Bishops. His pawns are threat
ening to advance and Black sorely misses his dark-squared Bishop.
24 ... Rd7 2 S .e4 Rhd8 26.d4 Qc4 27.Bb2 b6?

Both sides were already in time trouble and, as is so common, blunders

start to appear. Black's proper plan is 2 7 . . . Qxc3 2 8 .Bxc3 and only then
. . . b6, although even then White would still hold the advantage.
28.cxb6?

Here 2 8 .e5 ! , which seems contrary to the spirit of the position, is very
strong, inasmuch as 2 8 . . . Nd5 is unplayable on account of 2 9 .Qxc4.
28 ... axb6 29.Rc 1 Qxa2 ?

Even now 2 9 . . . Qxc3 is necessary: 3 0.Bxc3 b5 3 1 .Bb2 Bc4 3 2 .a4, etc. Of

course, White would stand much better here too, but at least Black would
be out of immediate danger.
30.Qxc6

Black's game is no longer tenable after losing this pawn, since he will be
unable to prevent d4-d5 .
3 0 . . . Rd6 3 1 . Q b 5 B d 7 3 2 . Qb4 Rc6 3 3 . Rxc6 Bxc6 3 4 . d S Q b l +
3 S .Kh2 Qe l , 1 -0.

Black overstepped the time limit. After 3 6 . Qxe7 the game would have been
over anyway.
In the first two rounds of the 1 97 7 Long Pine tournament I was held to draws
by two young players: Vasser Seirawan of Seattle and David Goodman of En
gland. Naturally, I was dying to finally get my first victory. Brasket had started
with draws against the two expatriate Soviet grandmasters Shamkovich and Lein.

( 1 28) Benko - Brasket (Lone Pine, 1 977)

l .c4 Nf6 2 .Nc3 e6 3 .NfJ cS

In 1 97 6 at the same tournament I played the same opening against the

same opponent, but then Brasket replied 3 . . . Bb4. Both moves are quite
playable, as are 3 . . . d5 and the recently fashionable 3 . . . b6.
4.g3 Nc6 5 .Bg2 dS 6.cxdS Nxd5 7.0-0 Nc7?

Black should continue 7 ... Be7, leading to the Semi-Tarrasch Defense. The
idea of the Knight move is to stop White's d2 -d4. In this type of position,
however, Black's e-pawn belongs on e5 where it controls d4 and helps to
establish a reversed type of Maroczy Bind. Unfortunately for Black, he
has already committed himself with . . . e 7 -e6.
8.Qa4!

Both 8.d3 and 8.b3 give an advantage, the latter perhaps going 8 . . . Be7
9.Bb2 0-0 l O.Rd ! intending Na4, e.g., 10 ... b6 I l .d4! cxd4 1 2 .Ne4 Bb7
1 3 .Nxd4. However, I felt that 8 . Qa4 gave me even more !

383

PAL B E N KO: MY LI FE, GAM E AND COM POSITI O N S

8 ... Bd7 9.Qe4

This very strong Queen maneuver has allowed the Queen to take up an
active, centralized position.
384

9 ... Be7 1 0.d4 cxd4 I 1 .Nxd4 Bf6

Black tries to complicate . On the natural 1 1 . . .0-0, 1 2 .Rd 1 with the threat
of Nxc6 gives White an overwhelming position. Now Black is pushed to
his first three ranks, and White's light-squared Bishop has a particularly
bright future.
1 2 .Nb3

Equally good is 1 2 .Ndb 5 but, on general considerations, I didn't want to

give Black a chance to relieve his crowded position by trading pieces. Now
the Knight aims for c 5 .
1 2 ... Qe7 1 3 .Bf4? !

I spent a lot of time here trying to find a way to get the most out of White's
advantage. On the obvious 1 3 .Be3 Na6 ( 1 3 . . . b6 1 4.Rfd 1 threatens Rxd7), I
couldn't see how to prevent Black from completing his development by cas
tling and bringing his Rooks to the middle. The text move attempts to pro
voke . . . e6-e5 so as to weaken the d5-square: 1 3 . . . e5 1 4.Be3 Na6 is not at
tractive because of 1 5 .Nd5 . Even so, my move is most likely not the stron
gest. Better was 1 3 .Rd 1 ! 0-0 1 4.Bf4 e5 (and not 14 .. .Ne8 1 5 .Be3 Nd6
1 6.Qd3) 1 5 .Be3 Rfd8 ( 1 5 . . . b6? 1 6.Rxd7), when both 1 6.Bc5 and 1 6.Nc5
give White more advantage than in the actual game. I was trying to gener
ate some quick action before Black had castled and hoped he would not
give up his Bishop for my Knight. But Black is practically forced to do this.
13 ... Bxc3 ! 1 4.bxc3 Nd5 1 5 .Bd2 0-0

Black has gotten some air at last, but at the cost of the two Bishops. How
ever, it will be a long process for White to take advantage of this. The
isolation of White's c-pawn does not compensate Black because it is not
effectively blockaded. The way to bring White's positional edge to a deci
sive advantage is to apply pressure in various sectors-to administer "the
squeeze" and hope that Black will crack under the strain.
1 6.c4 Nf6 1 7.Qe3 e5

I get a trim at
Lone P i n e.

(Photo Ron Chan.)

Black is trying for active counterplay, but as a general rule the opening of
the position favors the Bishops. More reliable, though passive, is . . . Rfd8
followed by . . . Be 8 .

1 8.Bc3 Ng4 1 9.Qc5 Qe6 20.Rfd l Rfd8 2 1 .h3 Nh6 22 .Rd6

Black is suddenly under considerable pressure, and the difficulty of his
position is now clear.

2 2 . . .Qe8
Not 22 . . . Qe7 ? ? 2 3 .Rxc6.

2 3 .Rad l Nf5 24.R6d2 b6 2 5 .Qa3 f6 26.c5

Notice how my supposedly weak c-pawn is taking an active part in the
attack.

26 . . . Nfe7 27.Qa6 Be6 28.cxb6 axb6 29.Qb5

White has succeeded in weakening Black's b-pawn, but it would be a mis
take to take it right away: 2 9 . Qxb6 Rdb8 3 0.Qc5 Rxa2 ! 3 1 . Rxa2 Bxb 3 , and
Black reestablishes material equality.

29 ... Rdc8 3 0.Rd6 Bf7 3 1 .Rld2 Rab8 32.Bb2 Rc7?

A definite mistake, but it is hard to find a good way to shake off the
pressure.

3 3 .Rc2 ! Rbc8 34.Qxb6 Bg6 3 5 .Rc4 Bf5 36.Ba3

In time pressure, White gives Black a chance to complicate. 2 6 .Na5 is
more accurate, although Black can still hold on for a while with 36 . . . B d 7 .

3 6 . . . Ra7 ? !

PAL BENKO: MY LIFE, GAME AND COMPOSITIONS

Black grabs his chance to get out of the pin, since White has no time to
figure out the win .
37.Bc1?
386

I could have ended the game quickly by 3 7 . Rcxc6 (Also winning is 3 7 .Bxc6
Rxc6 3 S . Rcxc6 Nxc6 3 9.BcS , but 3 7 .Rcxc6 is even stronger.) 3 7 . . . Nxc6
3 S .Bxc6 Rxc6 3 9. Qxa7 . Instead, I instinctively defend my loose Bishop.
37 Rb8 38.Qe3 Nd4 39.Nxd4 exd4 40.Rcxd4 Rxa2 41.Bd2 Ra1+ 42.Kh2
..

Black has been able to get the best possible drawing chances. Four pawns
against three on the same side in a Rook endgame is drawable, though
with great difficulty. However, with Queens on the board White has
attacking chances, especially because the advance Of Black's f-pawn has
weakened his King position.

42 . . . Ng6?
Considering the comments in the last note, it's easy to see why Black tries
to get to the endgame as quickly as possible. Unfortunately, this attempt
backfires because of his first-rank weakness.

43.BdS+ KhS
Even worse is 43 . . . KfS 44.Bb4, and now 44. . . Qxe3 ? 4 S . RdS is mate, or if
firs t 44 . . . Rxb4 4 S . Qx e S + and 46 . Rxb4 wins e a s ily, o r i f 44 . . . N e 7
4S .Rxf6+ gxf6 46. Qh6 is mate .

44.QxeS+ RxeS 4S .Bc6 Rf8 46.e4 NeS, 1 -0.

Black realizes that 46 ... BcS 47 .RdS loses a piece. The text, trying for a
miracle, fails because after 47 . exfS Black does not have . . . Nf3+, thus he
resigned without waiting for a reply.
After the game my opponent complained that I seemed to make an extra ef
fort against him and had given easy draws to some other players. Well, I can
assure him that no such distinction exists in my mind. I give a draw only when I
feel the game is completely even and without winning chances, or when I have
the worst of the position. It is possible, however, that I had already lost the opti
mism that characterized my play in my younger days, or the patience to wait for
my opponent's eventual mistakes.

RETIREMENT

(129) Benko - Chavez (Sao Paulo, 1977)

l.NB g6 2 .e4 d6 3 .d4 Nf6 4.Nc3 Bg7 5.Be2
By transposition of moves we have arrived at a very popular variation of
the Pirc Defense. White often uses this harmless looking, but positionally
potent, continuation (known as the Classical Variation) even when he has
a chance for the more aggressive Austrian Attack (i. e . , f2-f4).

5 . . . 0- 0 6 . 0- 0 c6 7.a4
This prevents . . . b7-b5 and threatens to gain more queenside space with
a4- a 5 .

7 . . . a5
The good points of this move are obvious: Black prevents a4-a5 and lays
claim to the b4-square (which can be occupied by . . . Nb8-a6-b4). The move's
flaws are more subtle: the b6-square has been weakened and, in many lines,
White can dominate it via NB -d2 -c4. Also, if Black plays an eventual . . . d6d 5 , the b5 -square can become weak after exd5 . . . cxd 5 .

8.Be3
Not only defending the d4-pawn, but also making the first gesture to
wards b6 (i. e . , If Black plays . . . e7 -e5 then dxe5 would unleash the Bishop
along the e 3 -a7 diagonal). Even though this move makes perfect sense,
White almost always plays 8.h3 first, stopping ... Ng4. However, I wasn't
convinced that . . . Ng4 was anything to worry about and, if the Knight leap
does turn out to be harmless, then why waste a tempo stopping it?

8 . . .Ng4
Black can't resist harassing my Bishop, although in my mind I felt he was
forcing it to a more aggressive square. Nevertheless, this move is the only
way to challenge White's decision to avoid h2-h3 since normal plans (like
. . . Na6 followed by . . . Nb4 or . . . Qc7 followed by . . . e7 -e5) would leave me a
tempo up over main lines. For example: 8 . . . Qc7 9.Nd2 e 5 1 0. dxe5 dxe5
1 1 .N c4, and White's control over the b6- and d6-squares gives him a clear
advantage.

9.Bg5 h6
9 . . . f6 was tried in Szabo-Hubner, Amsterdam 1 9 7 5 . After 1 0.Bd2 Nh6
1 1 . Q c 1 Nf7 1 2 .Be3 e6 B.d5 cxd5 1 4. exd5 e5 1 5 .Nd2 Nd7 1 6.Nc4 b6
1 7 .Qd2 f5 1 8.f3, White enj oyed a slight plus.

1 0.Bh4 Qc7
Also possible is 10 . . . Na6 when White , in the game D amj anovic-A.
Fernandez, Alicante 1 978, played in a very forcing fashion: 1 1 .h 3 Nf6
1 2 .e5 dxe5 1 3 . dxe5 Nd 5 1 4.Nxd5 cxd5 1 5 .Bxa6 (White's plan is simple:
he wants complete control over the d4-square . With this in mind, the ex
change on a6 makes sense since Black's Knight was the only piece that
could challenge White on that square via . . . Na6-b4-c6 or . . . Na6-c5 -e6.)

387

PAL BENKO: MY LIFE, GAME AND COMPOSITIONS

1 5 . . , Rxa6 1 6 , Bg3 Qb6 1 7 , b 3 Qc5 1 8 . Qd2 g5 1 9 .Nd4 f6 2 0 . Rae l , and

White's dominating Knight led to an easy win.

1 1 .Nd2 Nf6
388

The only real alternative is 1 1 . . .f5 1 2 .exf5 gxf5 1 3 .h3 Nf6, but 1 4.Re l should
be better for White.

1 2 .Nc4
A very tempting alternative is 1 2 .f4, when 1 2 . . . e6 1 3 .Nc4 d5 1 4. e 5 Ne8
1 5 .Ne 3 led to an obvious White advantage in the game Psakhis-Kochyev,
Groningen 1 990.

12 ... Nbd7
Now Black gets into a cramped position. If he had tried to free himself by
1 2 . . . Nxe4 1 3 .Nxe4 d 5 , White's game would still be preferable after
1 4.Ned2 dxc4 1 5 .Nxc4, but perhaps less so than after the text.

1 3 .Bg3 !

The threat of e4-e5 forces Black to show his intentions in the center.

1 3 . . . e5 14.dxe5
The other possibility, 1 4. d 5 , also favors White, but the text move looks
more logical, hoping for a faster decision in a more open position.

1 4 ... dxe5 1 5 .Qd2 Nh7 1 6.Qe3

Perhaps the immediate 1 6. f4 is even stronger, since if 16 . . . exf4 1 7 .Bxf4
wins the h-pawn, and otherwise there follows f4-f5 when Black's position
is very restricted and passive. I was worried, here and later, that Black might
sacrifice a pawn in order to obtain firm control over the e 5 -square (i. e . ,
I

1 6.f4 exf4 1 7 .Bxf4 Ne5 1 8 .Bxh6 Rd8 1 9 .Qf4 Bxh6 2 0. Qxh6 Be6). /Thls
would give him some compensation for White's slight material advantage
(the isolated e-pawn) . Whether this would prove sufficient to hold the
game is another matter.

1 6 . . .Ng5 1 7.f4
White finally decides on the break: if Black's Knight gets to the ideal e6square before this move (giving him added control over c5 , d4, and f4) , he
will have good chances to resist White's pressure.

RETIREMENT

17 . . . Ne6
If 1 7 . . . exf4 1 8 .Bxf4 NeS 1 9.Qg3 Re8, White can either send back the Knight
with 20.h4, or play 2 0.Rad l , in either case gaining a solid advantage .

1 8.f5
Again, playing to win a pawn does not seem the strongest: 1 8 .fxeS NxeS
1 9 .NxeS Bxe S 2 0 . BxeS Qxe S 2 1 . Qxh6 NcS 2 2 . Rf4 Be6 2 3 . Rh4 Qg7
24.Qxg7 + Kxg7, when Black's pieces are very well placed . The text move
virtually forces Black to sacrifice a pawn on terms that I was more willing
to accept.

1 8 . . . Nf4
An important alternative is 1 8 . . . Nd4 1 9. f6 (also good is 1 9 . Qxd4 exd4
20.Bxc7 dxc3 , when both 2 1 .bxc3 and 2 1 .b3 give White a superior endgame)
1 9 . . . B x f6 ( B l a c k s h o u l d go fo r 1 9 . . . Nxc2 , though 2 0 . Q d 2 Nxa l
2 1 .fxg7 Kxg7 2 2 .Rxa l f6 2 3 . Bf2 followed by Rd l is very nice for White .)
20.Rxf6 Nxc2 2 1 . Qxh6 Nxa l 2 2 .Rxg6+ fxg6 2 3 . Qxg6+ Kh8 2 4. Qh6+ Kg8
2 S .QgS+ Kh8 2 6.NxeS , and White wins.

19.Bxf4
Also promising is 1 9.Rxf4! ? exf4 2 0.Bxf4 Qd8 2 1 .Bxh6.

1 9 ... exf4 20.Rxf4

Of course, 2 0. Qxf4 Qxf4 2 1 .Rxf4 leaves White a pawn up in the endgame,
and Black still has problems with his development. But after his best
d e fens e , 2 1 . . . B d4+ 2 2 . Kh l Nc S , the win d i d not look s o simpl e :
2 3 .fxg6 fxg6 24.Rxf8+ Kxf8 2 5 .Nb6 Rb8 2 6.Rd l Bxc3 2 7 .bxc3 Be6. There
fore, I avoided the trade of Queens so I could keep the attack going, even
though this cost the Exchange.

20 ... Bd4 2 1. Qxd4 Qxf4 22.Rf1 Qc7

I was expecting Black to keep his Queen closer to his King by 22 . . . QgS
but even then, White keeps a large advantage by 2 3 .e5 gxfS 24.Bd3 when,
because of Black's lack of development and White's beautifully placed pieces,
it's hard for Black to find a good move .

2 3 .Kh1
And not 2 3 .Nd6 due t o 2 3 . . . Qb6.

2 3 ... f6
Black wants to create a blockade on the dark squares, but this doesn't give
him the relief that he was hoping for.

24.Rf3
Also reasonable was 24.fxg6, but I wanted to retain my f-pawn which gains
space, restricts the movements of Black's Bishop, and eventually becomes
passed.

24 . . . Ne5 2 5 .Rg3 g5

389

PAL BENKO: MY LIFE, GAME AND COMPOSITIONS

390

This automatic move turns out to be hopeless, since it leads to a passive

game with no real prospects of defense . Personally, I would have tried to
mix things up a bit by 2 5 . . . Rd8 2 6.Qc5 b6 2 7 .Nxb6 Rb8, when Black's
Rooks have finally gained some activity. Of course , White can win this
position in various ways (28.Ncd5 and 2 8.Nc4 are both strong), but the
resulting variations are complicated and present White with more oppor
tunities to go wrong than in the game.

26.Nxe5 Qxe5
On 2 6 . . . fxe5 , White plays 2 7 .Qe3 ! intending h4.

27.Qxe5 fxe5 2 S.h4 Bxf5 ?

A desperation move in time pressure, but in any case White's protected
passed pawn and active pieces promised victory.

29.exf5 Rxf5 3 0.Bd3

Now Black loses more material, but his lack of time prevented him from
finding "resigns" which, clearly, was the best move.

30 . . . Raf8 3 1 .Bxf5 Rxf5 3 2 .hxg5 Rf2 H.gxh6+ Kh7 34.Ne4 Rxc2 3 5 .Rh3
Rc l + 3 6.Kh2 Rfl 3 7 .Ng5+ KhS 3 S.Rd3 Rf8 3 9.Rd7, 1 -0.

(130) Benko - Peters (Lone Pine, 1978)

White to Move

RETIREMENT

The interesting feature of this position is the white Rook's peculiar situa
tion. But I was quite satisfied: the Rook ties up two black pieces. If it was
Black's move here and he played, for instance, L.Bc7, then 2 .RgS wins a
pawn. Maybe my satisfaction dulled my alertness, for I now committed
one of the worst blunders of the tournament.

39.Bg5 ? ?
Of course, I should have played 3 9.Bd2 KcS (3 9 . . . f5 + 40.Kd4) 40.Bb4 Rc7 !
4 1 .Bd6 Kb7 42 .Bxc7 Kxc7 ! (and not 42 . . . KxaS 43 .BdS), when we get an
interesting pawn endgame where White stands better, but it's not clear if
he can find a way to penetrate into Black's position.

3 9 . . . Kc8.
There is no way for White to prevent ... Rd7 followed by ... Kb7 , winning
the Exchange and the game. I lost after a few more useless moves.

(131) Benko - Seirawan (Lone Pine, 1978)

l .NB Nf6 2.c4 c5 3 .Nc3 d5 4.cxd5 Nxd5 5.e4
Another possibility is 5 .g3 but, in that case, Black can aim for a Maroczy
Bind type of position with 5 . . .Nc6 6. B g2 Nc7 followed by . . . e 5 .

5 . . .Nxc3
Black chooses a simple line, avoiding the complications connected with
5 . . . Nh4, which was played twice in this tournament. For example, B enko
Peters, Long Pine 1 97 5 continued (after 5 . . .Nb4) 6.Bc4 (White's main al
ternative is 6 . B b 5 +) 6 . . . Nd3 + (6 . . . Be6 is also possible, though 6 .. .Nd 3 +
has a better reputation) 7 . Ke2 Nf4+ S . Kfl Be6 ? ! (an experiment gone bad.
The fate of this line rests on the viability o f S . . . Ne6) 9 . B b 5 + B d 7
1 O .d4 cxd4 1 1 .Nxd4 Ng6 1 2 .Be3 e6 1 3 .Qb3 Be7 1 4. Rd 1 QcS 1 5 .Rc 1 QdS
1 6.g3 and, after I castled by hand, White had an obvious advantage out of
the opening (though I managed to eventually lose this game due to a hor
rendous blunder - see game 1 3 0) .

6.dxc3

It seems as though White would be satisfied with a draw since he is offer

ing to trade Queens and is even willing to lose his castling privilege to do

391

PAL BENKO: MY LIFE, GAME AND COMPOSITIONS

it. However, this is definitely not the case ! In fact, the Queen exchange is
quite good for White. Why? First, White's King will actually prove to be
very well placed on c2 . Second, the pawn structure favors the first player.
392

Let's compare the merits of both sets of c-pawns and e-pawns: Black's c 5 pawn blocks his own dark-squared Bishop while White's c3 -pawn takes
the b4 and d4 squares away from Black's pieces. White's advanced pawn
on e4 gains central space and restricts Black's light-squared Bishop while
Black's e-pawn doesn't have much of a future -if it moves to e6 it will
block the c8-Bishop and create a hole on d6. If it moves to e5 a permanent
hole will appear on d5 . These factors make it hard for Black to find a tar
get or a plan, while White can take aim at the c5 -pawn and also play to
dominate the c4-square by a2 -a4 and Nd2 -c4.
Of course, White could also play 6.bxc3 , transposing into a Griinfeld
after 6 . . . g6 7 .d4 Bg7 . At the time, this line of the Griinfeld was thought to
be favorable for Black because systems where White's Knight goes to f3
hadn't been properly explored. Now the Nf3 system is the main line, though
I'd still avoid it because the theory is too extensive .

6 . . . Qxd 1 + 7.Kxd l Nc6

Notice how my c3-pawn keeps this piece under wraps. In fact, Black's Knight
really doesn't have much of a future on c6. Because of this, modern theory
recommends 7 . . . Bg4 8.Kc2 Nd7 as giving Black more chances to reach
equality.

8.Be3
Taking aim at c5 and forcing Black to defend his pawn. However, both
defensive moves lead to slight concessions: 8 . . . e6 blocks the c8-Bishop and
weakens d6 (as I mentioned earlier) and 8 . . . b6 allows me to eventually
strike at his pawn chain by a2 -a4-a 5 . The alternative, 8.Bf4, is also known
to give White a promising position.

8 . . . e6
Better is 8 . . . b6 with the idea of . . . Bb7 and . . . 0-0-0.

9.Kc2 b6? !
This looks solid, but the newly weakened light-squares on a6 and c6 will
haunt Black for a long time. Preferable is 9 . . . B d 7 , when White gained a
slight edge in Cvetkovic-Palatnik, USSR-Yugoslavia 1 976 after 1 0.Be2 Be7
1 1 .Rhd 1 0-0-0 1 2 . Rd2 f6 1 3 .Rad l .

l O.Bb5 Bd7
The consistent 10 . . . Bb7 is also possible, though 1 1 .Ne5 Rc8 1 2 . Rhd 1 a6
1 3 . Bxc6+ Bxc6 1 4.Nxc6 Rxc6 1 5 .Bf4, followed by the doubling of Rooks
on the d-file leaves White with a nice edge (better Bishop, lead in devel
opment and control over the only open file).

1 1 .a4!
The immediate 1 1 .Rhd 1 is a mistake due to 1 1 . . .Nb4+, when Black picks up
White's loose Bishop. Now that this Bishop is firmly protected, Rfd 1 is a

In battle agai nst Raicevic,

Majdan pek 1 9 78,

strong threat (when tactics such as Rxd7 would be in the air). Since the
Bishop can't be allowed to take up permanent residence on b S , Black is com
pelled to weaken his queenside pawn formation if he wants to chase it away.

1 1 . . .a6 1 2 .Be2
Now Black's as-Rook is stuck guarding the pawn on a6.

12 ... Bd6 1 3 .Nd2

Heading for c4.

1 3" .Na5
Trying to keep White's Knight at bay. Far worse is 1 3 . . .0-0 1 4.Nc4 Bc7
I S .Rhd l RfdS 1 6 .aS b S (Like it or not, Black has to try the unpalatable
1 6 . . .NxaS 1 7 .NxaS bxaS I S .BxcS.) 1 7 .Nb6 Bxb6 I S .axb6 c4 1 9.b7, and Black
is finished.

1 4.Rhb l !

My favorite move of the game. Suddenly Black (now facing the threat of
b2 -b4) is forced to accept the vast inferiority of his position.

14 ... 0-0 1 5 .b4 exb4 1 6.exb4 Ne6

The tempting 1 6 . . . RfcS+ actually walks into possible forks on b6: 1 7 .Kd l
Nc6 I S .Nc4, when White's advantage is obvious.

17.b5 axb5
I expected 1 7 . . . Nh4+ I S . Kb3 e S , but on 1 9.Nc4 Be6 2 0 . Kc3 the King per
sonally conducts the battl e . For exampl e , 2 0 . . . RacS 2 1 . Rxb4 Bxb4+
2 2 .Kxb4 as+ 2 3 .Kb3 , with a winning position for White.

1 8 .axb5 Nh4+ 1 9.Kb3 Be5 20.Rc1

PAL BENKO: MY LIFE, GAME AND COMPOSITIONS

Also good is 2 0.NB with the idea of Bd2 , though I eventually decided
that my move was stronger. Trying to save a tempo by 2 0.Rxa8 Rxa8 2 1 . Rc 1
isn't a s clean because 2 1 . . . Ra2 (and not 2 1 . . .Na2 ? 2 2 . Ra l Ra3 + 2 3 .Kb2 ,
394

when White wins a piece) creates some complications.

20 ... Rxal 2 1 .Rxal Rd8

Black is trying to save his stranded Knight. Of course, his position is quite
lost.

22 .Bxe5
Another strong continuation was 2 2 .Ra4 B e8 2 3 .Kd ! , when nothing can
be done about 24.Rxb4.

22 . . . bxe5 2 3 .Kc3
This gives much needed support to d3 and d2 .

2 3 ...Kf8 24.Nb3
Now that everything is defended, it's time for White to take aim at Black
weaknesses and cash in.

24 ...Re8 2 5 .Ra7 Ke7

Black should have tried 2 5 . . . Ke8 2 6. Kc4 (another way is 2 6 .Bc4 with the
idea of 2 7 .Nxc5 and Kxb4) 26 . . . Nc2 , when 2 7 .Nxc 5 ? ? Bxb 5 ! + 2 8.Kxb5
Nd4+ just about equalizes. Of course , White doesn't have to fall for this.
Instead of 2 7 .Nxc5 ? ? , the simple 2 7 .b6 wins without too much trouble.

2 6 . Ke4 Ne2 2 7 .Nxe5 Rd8 2 8 .Nxd7 Rxd7 2 9 . Rxd7 + Kxd7 3 0. Kc3 ,

1 -0.
The Knight is trapped after 30 . . . Ne 1 3 1 .Bfl followed by 3 2 .Kd2 . This was
clearly not Yasser's day. Even his wishing me bad luck before we started
didn't help. But in the next round he beat Grandmaster Tarjan, and he ended
up with a 50% score, playing eight grandmasters in nine games! This is
very impressive when you consider that he was only seventeen years old.

(132) Fedorowicz - Benko (Lone Pine, 1979)

l .e4 e6 2 .d3 d5 3 .Nd2 g6 4.NgB Bg7 5 .c3
A rare move. Ljubojevic was having great success with 5 .Be2 at the time
(5.g3 and 6. Bg2 was more common), but the mix of d and Be2 wasn't
something I was familiar with. Looked at logically, this combination of
ideas doesn't make sense. After Black plays . . . e 7 - e 5 , White often advances
his b-pawn two squares and fianchettoes his dark-squared Bishop so it puts
pressure against e 5 . Now, with the pawn on d, this won't be possible.

5 ... e5 6.Be2 Ne7 7.0-0 0-0 8.b4? !

This move seems premature.

8 . . . a5 ! 9.bxa5
Weakening his pawn formation, especially the a-pawn. However, 9.a3 axb4
1 0. cxb4 gives Black a majority of center pawns.

RETIREMENT

9 ... Qxa5 1 O.Qc2 c5 1 1 .a4? !

This weakens the a-pawn even more. Now Black devotes his entire strat
egy to hunting this pawn down !

1 1 .. .Qc7 1 2 .Re l b6 1 3 .Bfl Bb7 14.g3 Nd7

It might seem more natural to place this Knight on c6, but I was reserving
that square for my Bishop so it could begin the assault against a4. One of
the most important things in chess is the creation and executi on of
targets. Here, a4 is begging to be attacked. This turns out to be easy to
do: the Bishop goes to c6 and I can double or even triple on the a-file.

1 5 .Bg2 Bc6 1 6.Bb2 d4

So White ended up fianchettoing both his Bishops. Unfortunately for my
opponent, neither Bishop is particularly active.

1 7.cxd4 exd4 1 S.Nc4 Ra6 1 9.Bc1 Qa7

19 . Rfa8 was also possible, but my Queen was vulnerable on c7 and I wanted
.

to get it away from possible Bf4 shots.

20.Bg5 NcS
And just like that, the a4-pawn is ready to fall.

2 1 .Nb2 b5 22 .h4 bxa4

White has lost his weak a-pawn so he tries to get some compensation by
building up a kingside attack.

2 3 .Bh3 ReS 24.h5 Ncb6

Winning material doesn't mean you can rest on your laurels. My new plan
is centered around a . . . c5 -c4 advance. Thus, my next several moves will
either shore up my kingside or continue preparations for this thrust.

2 5 .hxg6 hxg6 26.Nd2 Bb5

White is aware of Black's plan, and is making his own efforts to take con
trol of c4. However, at the moment White can't occupy it by 2 7 .Nbc4 due
to 27 . . . Bxc4 2 8 .Nxc4 Nxc4 2 9 . Qxc4 Ne5 , when the attack on White's
Queen, plus the threat of . . . N8+, ends the game.

27.f4 f6 2S.Bh4 Qc7

395

PAL BENKO: M Y LIFE, GA M E AND COMPOS ITIONS

396

Black tries to take advantage of the awkward position of the h4-Bishop.

The immediate threat is . . . g5 . Of course, 28 . . . Qc7 also prepares the . . . c5c4 advance (in other words, one should only make a threat if it helps fur
ther your overall plan).

29.Nf3 Ra7
An important move that defends my Queen and the d7-Knight.

3 0.Nd l c4
Finally achieving the break that I'd worked so hard for. Now, with mutual
time pressure affecting both players, the fight becomes sharp and tactical.

3 1 .Nxd4 Qc5 3 2 .Qf2 g5 3 3 .Nxb5 Qxb5 34.dxc4 Qc5

Another good way to handle this position was 34 . . . Qa5 3 5 . Bxd7 Rxd7 ,
when 3 6.c5 can be answered b y 3 6 . . . Rd2 . However, it's long been my habit
to play the most controlled, safest moves in time pressure. With 34 . . . Qc5 ,
I keep my advantage, make sure all my piece are on safe squares, and (by
fixing the c-pawn) prevent any kind of enemy counterplay.

3 5 .Qxc5 Nxc5 3 6.Rb l Nxc4 3 7.fxg5 fxg5 38.Bxg5

White has managed to save his piece, but he had to open the diagonal for
Black's dark-squared Bishop to do so.

3 8 . . . a3
Thanks to my Bishop's control over the a I -square, the passed a-pawn is
now unstoppable.

39.Bfl Ne5 40.Re3 a2 4 1 .Ra l Nfl 42 .e5 Rxe5 , 0- 1 .

Avoiding 42 . . . Nxg5 43 .Bc4+ followed by 44.Rxa2 , when I don't have any
pawns left. After 42 . . . Rxe 5 , I win a whole Rook.

(133) Zaltsman - Benko (USA, 1983)

l .d4 d5 2.c4 e6 3 .Nfl Nf6 4.Nc3 Be7 5.Bf4 0-0 6.e3 c5 7.dxc5 Bxc5
8.cxd5
The more complicated 8.Qc2 Nc6 9.a3 Qa5 I O.Rd i was popular in the
' 70s and '80s . In the '90s, 1 0.0-0-0 became the main line. The text move

RETIREMENT

(8.cxd5) is played by those who want a quiet game where they hope to
retain a safe, small, edge.

8 . . . Nxd5
Though 8 . . . exd5 is also common, I prefer this capture, which gives me the
option of a quick check on b4.

9.Nxd5 exd5 1 0.Bd3

I think 1 O. a 3 , preventing the check on b4, gives White more chances for
an e d g e . Aft e r 1 0 . . . Nc6 I l . B d 3 (thr e a t e n i n g to win a p awn by
1 2 . Bxh7 + Kxh7 1 3 . Qc2 + , picking up the loose c 5 -Bishop) 1 1 . . . B b 6
1 2 .0-0, Black can play 1 2 . . . d4, trying for a dull draw, o r 1 2 . . . Bg4, with a
complicated battle ahead.

1 O ... Bh4+ 1 1 .Ke2

White is willing to give up castling as long as he can develop quickly and

place his pieces on good squares. It's also important to mention that Black's
dark-squared Bishop is better placed on c5 (where it eyes the d4-square)
than on b4. Of course, forcing White to misplace his King can't be wrong,
and both sides will have chances in the following strategic battle. Though
1 1 .KfI was played in the old game Bernstein-Rubinstein, Ostend 1 906,
nowadays I l .Ke2 is considered to be more accurate since it keeps the Rooks
connected .

1 1 .. .Nc6 1 2 . Qc2 h6 1 3 .Rhd l Bg4

A natural move , though m o d e rn the ory thinks highly of 1 3 . . . Q f6
1 4. a 3 B e 7 1 5 . Q c 3 Be6 1 6 .Nd4 Nxd4+ 1 7 . Qxd4 Qxd4 1 8. exd4, draw,
Tukmakov-Geller, Tbilisi 1 978.

1 4.Bh7+ Kh8 1 5 .Bf5

The sharp 1 5 .Be4 would have been answered by 1 5 . . . Rc8, when Black gets
a good game.

1 5 ... Bxf3 +
White would have a pleasant edge after 1 5 . . . Bxf5 1 6 . Qxf5 . In general,
the side with the isolated d-pawn wants to avoid exchanging too many
minor pieces since that deprives him of the dynamic potential the iso usu
ally bestows.

397

P A L BENKO: M Y LIFE, GAM E AND COMPOSITIONS

1 6.gxf3
Is 1 6.KxB ! ? possible? I can imagine someone like Vasser S eirawan playing
this, but mere mortals don't like to rush their Kings out into the middle of
398

the board. After 1 6. gxB , the weakening of White's kingside gives Black
plenty of counterplay.

1 6 . . . Ne7 1 7 .Bc7 Qe8 1 8.Bh3

Some sources have claimed a White edge after I S . a 3 , but I S . . . B a S
1 9.BxaS Q b S + seems fine for Black, while I S .a3 B a S 1 9 .Bd6 RdS i s also
nothing to worry about.

18 . . . BaS ! 1 9.Bd6 Rd8 20.Ba3

Not falling for 20.QcS ? ? b6! 2 1 .Qa3 Rxd6 2 2 . Qxd6 QbS+, when Black
wins.

20 . . . Bb4!

Once White's dark-squared Bishop gets traded (ridding White of his Bishop
pair), Black will be left with a pleasant initiative due to the insecure posi
tion of the white King.

2 1 .Bxb4 QbS + 22 .Rd3 Qxb4 2 3 .Rb3 ?

Not very realistic. White goes after a pawn and ignores the weaknesses
on his kingside. This leads to a remarkably swift debacle. Much better
w a s 2 3 . Ra d l , t h o u gh B l a c k h a s no p r o b l e m s a ft e r 2 3 . . . Rfe S
2 4.Kfl Qh4.

23 . . . Qh4 24.Rxb7 d4! 2 S . exd4 Nc6 !

Suddenly all of Black's pieces are participating in a furious attack against
White's open King.

26.Qxc6, 0- 1 .
White resigned without waiting for a reply since he saw that he'd be forced
to give up his Queen after 26 . . . RfeS+ 2 7 .Kfl (2 7 .Kd3 Qxd4+) 2 7 . . . Qxh 3 +
2 S .Kgl Re6.

RETIREMENT

(134) Benko -Trosclair (New York, 1984)

l .c4 Nf6 2 .Nc3 c5 3 .g3 g6 4.Bg2 Bg7 5.e3
I alternate between this, S .Nf3 , and S . e4.

5 ... 0-0 6.Nge2 e6 7.Nf4

I realized that Black was most likely playing for a position that's known to
be very drawish. This comes about after 7 .0-0 Nc6 8.d4 cxd4 9.Nxd4 dS
1 0.cxdS Nxd4 1 1 .exd4 Nxd S . Not wishing this to happen, I played a move
that is designed to stop Black from pushing his pawn to d S .

7 ... Nc6 8.0-0 Ne7

When it comes to Black's . . . d7-dS advance, he won't take no for an an
swer! Of course, the position after 7 .Nf4 is nothing special for White any
way. The game Larsen-Tal, Bugojno 1 984, showed a far more dynamic
way for Black to play: 8 . . . b6 9.b3 Ba6 1 0.Bb2 dS 1 1 .Re 1 Rc8 1 2 . d3 d4!
1 3 .exd4 Nxd4, with an excellent position.

9.d4 cxd4 1 0.exd4 d5 1 1 .cxd5 Nfxd5 1 2 .Nfxd5 Nxd5

Black has managed to get the same position that I mentioned in my note
to 7 .Nf4.

1 3 .Qb3
White tries to inj ect some life into this extremely dull position. Usual is
1 3 .NxdS exd S 1 4. Q b 3 Be6 I S . Qxb7 Bxd4 I S . Bf4 Qb6 1 6 . Bxd 5 Bxd5
1 7 . Qxd5 Rfd8 1 8 . Qf3 Bxb2 19 . Rab 1 Qf6, when the game is a dead draw.

1 3 ... Bxd4
Black can also consider 1 3 . . . Ne7 (this is probably the most accurate way
to play) 1 4.Rd 1 Nc6, when I S .Be3 Nxd4 1 6.Bxd4 Bxd4 1 7 .Ne2 eS 1 8.Nxd4
exd4 1 9. Qc4 (a better test of Black's defensive powers is 1 9. Q d 5 Qxd5
2 0.Bxd S , but the second player can still equalize by 2 0 . . . Re8! 2 1 . Rxd4 Be6!
2 2 .Bxe6 Rxe6 2 3 .Rd7 Rb6) 19 ... Be6 20.Qxd4 Qxd4 is, once again, a draw.

14.Bh6 Bg7
Much better than 1 4 . . . Re8 1 5 . Rad 1 Bxc3 1 6.bxc3 Qb6 1 7 . Q c4 (Two years
later, Torre took the white side of this position [against Bordonada, Aus
tralia 1 97 5] and played 1 7 .Qa4 Qc6 1 8.Qe4 Qxc3 and Black, who just hung
a piece, resigned.) 1 7 . . . Qc6 1 8.Qe2 b5 1 9.Bg5 Bb7 20.Qe5 Rac8 2 1 . Rfe i
Ba8 2 2 .h4 QcS 2 3 .Bxd5 Bxd5 24.Bh6 f6 2 5 . Qxf6 , when White won in
Cardoso-Torre, Manila 1 97 3 .

1 5 .Bxg7 Kxg7 1 6.Bxd5 ! ?

In the game Andersson -Gheorghiu, Moscow 1 98 1 , Black found a nice way
to defend after 1 6.Nxd5 exd5 1 7 .Bxd S . The maneuver with his a-pawn and
Rook is quite interesting: 1 7 . . . a5 ! 1 8.Rac 1 a4 1 9. Qch Qf6 20.Qxf6+ Kxf6
2 1 .Rc7 Ra5 2 2 .Rd 1 RbS . Many books feel that White's best is 1 6.Rfd 1 Qb6
1 7 .Nxd5 Qxb3 1 8 .axb3 exdS 1 9.Rxd5 , when 19 ... a6 2 0.b4 was a bit better
for White in Smyslov-Petrosian, USSR ch. 1 974. However, I don't believe

399

PAL BENKO: MY LIFE, GAME AND COMPOSITIONS

White will be able to create serious winning chances if Black plays the simple
1 9 . . . ReS (instead of 1 9 . . . a6): 2 0.Rb5 Re7 (and not 20 . . . Re2 2 1 .Bxb7 Bxb7
2 2 .Rxb7 Rxb2 2 3 . Ra6) 2 1 .b4 a6 2 2 .Rb6 Re2 2 3 .Bxb7 Bxb7 24.Rxb7 Rxb2
400

2 5 .b 5 (or 2 5 .Ra 5 ReS 2 6.Rxa6 Rcc2) 2 5 . . . a5

=.

The move I chose, 1 6.Bxd 5 ,

i s a novelty that strives to take advantage o f the weakened dark squares i n

Black's kingside structure.

1 6 . . . exd5 1 7.Nxd5

The Knight on d5 is clearly stronger than Black's Bishop. This, and the
fact that White's Rooks will take the center files faster than Black's, leaves
my opponent with difficult problems to solve.

1 7 . . . f6
Can Black equalize? It's not clear how. The obvious 1 7 . . . Be6? gives White
a clear structural plus after 1 S .Qc3 + f6 1 9.Nc7 . During the game I looked
at 1 7 . . . Qa5 , but soon realized that Black faces some serious suffering after
1 S . Qf3 Bf5 (or 1 S . . . Be6 1 9 . Qf6+ Kh6 20.Qf4+ Kg7 2 1 . Q e 5 + Kh6 2 2 .Rfd l ,
when Black i s i n trouble) 1 9 .b4 Qa6 2 0.Rfd 1 (White gets nothing from
2 0.Nc7 Q d 3 ) 20 . . . RadS 2 1 .g4! Bd7 2 2 .g5 .

l S.Rfe l
A simple move that keeps the black Bishop off of e6 and also threatens to
penetrate to e 7 .

l S . . . Rfi 1 9.Rad1 Bd7 20.Nf4

This is stronger than the tempting 20.Qxb 7 : 20 . . . Bg4 2 1 .Nc7 Qxc7 2 2 .QxaS
Bxd l 2 3 .Rxd l Qc2 2 4.RdS Qc l + ! 2 5 .Kg2 Qxb2 2 6 . RgS+ Kh6, and the
threat of . . . Qb7 (with check!) should enable him to draw.

2 0 . . . Qb6
Safer was 2 0 . . . QcS, though White's advantage would still be obvious. Now
Black gets a nasty surprise.

2 1 .Qxf7 + ! Kxf7 22 .Rxd7+ Kf8 2 3 .Ree7

Black is busted. The two Rooks on the seventh rank, in cooperation with
the Knight, prove to be far more valuable than Black's useless Queen.

23 . . .RdS 24.Rfi+

RETIREMENT

Not settling for 2 4.Ne6+ Qxe6 2 5 .Rxe6 Rxd7 26.Rxf6+. White's extra pawn
would by no means guarantee a victory.

24 ... KgS 2 5.Rg7+ Kf8 26.Rdfi+ KeS 27.Rxh7 Rdl + 2S.Kg2 Qc6+ 29.Kh3
QcS+ 3 0.g4 Qc4
There was no defense.

3 1 .Rge7+ Kf8
Or 3 1 . . . Kd8 3 2 .Ne6+ Kc8 3 3 . Rec7+, and White wins.

3 2 .Ne6+ , 1 -0.

(135) Benko - Harrison (Australian Open, 1985)

l .e4 c5 2 .Nf3 d6 3 .d4 cxd4 4.Qxd4
At times I get bored with the automatic 4.Nxd4 and turn to this move.
Though the Queen looks vulnerable in the center, it can easily turn out to
be strong on d4. Played by Hungarian masters, especially the Steibler broth
ers in the 1 92 0s, it was revived in the ' 50s by Szabo. Since that time, vari
ous grandmasters have used it as a surprise weapon. Tal beat R. Byrne in a
nice game in 1 97 6 , I 've used it for years, and Judit Polgar carried on
Hungarian tradition by employing it in the early to mid- 1 990s.

4 ... Nc6
The game Benko-Vukcevich, Telephone Match 1 978 continued 4 . . . a6 5 .Be3
(aiming at the hole on b6. Theory now feels that the best way to meet
4 . . . a6 is 5 .Bg5 Nc6 6 . Qd2 h6 7 . Bh4 g5 8.Bg3 Bg7 9.c3 , with a small edge
for White.) 5 . . . Nf6 6 .Nc3 Nc6 (6 . . . e5 7 . Qa4+ ! ? Nbd 7 8.Bc4 occurred in
Benko-Szekely, Sombor 1 97 6 . The position looks dangerous for Black.)
7 . Qb6 Qxb6 8.Bxb6 g6 (8 . . . e6 9.0-0-0 leaves d6 weak due to the threat of
Bc7) 9.0-0-0 Bh6+ 1 0.Kb 1 0-0 1 1 .Nd4 Bd7 1 2 .f3 Ne5 1 3 .Nb3 Bg7 ? (bet
ter was 1 3 . . . Bc6, though after 1 4.Na 5 White would retain a slight advan
tage) 1 4. f4 Nc6 1 5 . B e2 Bg4 1 6 .Bxg4 Nxg4 1 7 .Nd5 Nf6 1 8. Rhe 1 , and
White's superiority wasn't in doubt. In the '90s, players realized that after
4 . . . a6 5 . B e 3 , B lack should hold off on . . . Nf6 and instead play 5 . . .Nc6
6 . Qb6 Qxb6 7 . Bxb6 g6 8.Nc3 Bg7 (also possible is 8 . . . Bh6 ! ? 9.Nd5 Kf8
1 0.Be2 Kg7 , as played in Jandemirov-Atalik, Athens 1 99 3 ) 9.Nd5 (9 .0-0-0
can be met by 9 . . . Bxc 3 ) 9 . . . Kf8 1 0.0-0-0 Nf6 1 1 .Bd3 Nxd 5 ! 1 2 . exd5 Nb4
1 3 .Bc4 Bf5 1 4.a3 Rc8 1 5 .axb4 Rc4 1 6.c3 Rg4! 1 7 .Rhg1 Be4, when Shabalov
feels that Black has an edge.

5.Bh5 Bd7
A playable alternative is 5 . . . Qd7 6 .Bxc6 bxc6 7 .c4 e5 8.Qd3 , with an inter
esting game ahead.

6.Bxc6 Bxc6
White is supposed to be a bit better after 6 . . . bxc6 7 . c4 e5 8.Qd3 Qc7 9.0-0.

7.Nc3

40 I

PAL BENKO: MY LIFE, GAME AND COMPOSITIONS

Though I prefer the sharp piece play that this move offers, I have also
experimented with 7 . c4. The game Benko-Grefe, Lone Pine 1 9 7 7 contin
ued 7 . . . Nf6 S .Nc3 g6 9.0-0 Bg7 1 O. Qd 3 0-0 I l .Nd4 (Also interesting is
402.

I I . Be3 Ng4 1 2 . B d4 NeS 1 3 .BxeS dxe S , when Black has the two Bishops
but White's queenside maj ority looks more impressive. In Damjanovic
Gligori c , 1 9 6 3 B lack p l ayed more carefully a fter 1 1 . B e 3 : 1 1 . . . a 6
1 2 .Rfd l RcS 1 3 . B d4 b S 14.cxbS axbS I S .a 3 Q d 7

) 1 1 . . . Qb6 ! (an improve

ment over Padevski-Ghitescu, 1 96 3 , which went I l .. .RcS 1 2 .b3 Nd7 1 3 .Bb2 ,

with the better game for White) 1 2 .Be3 (What else? The threat was . . . Bxe4.
If 1 2 .Nc2 Ng4, the black pieces are too active and . . . f7 -fS is even in the
air.) 1 2 . . . Ng4 (This is playable, but 1 2 . . . Qxb2 1 3 . a 3 Ng4 is most testing.
If 1 4.Nd S , then 14 . . . NeS is the best answer.) 1 3 .Nxc6 Qxc6 1 4.NdS RfeS
I S .Bd4 Bxd4? ( I S . . . NeS is correct, with equality) 1 6. Qxd4 NeS 1 7 .b 3 a6
I S . Rac 1 RacS 1 9. Rfd l Nd7 2 0.Qd2 , and White enjoyed a clear advantage
and went on to win the game.

7 . . . Nf6 S.BgS e6 9.0-0-0 Be7

If 9 ... Q a S , White can reply with 1 0.h4. The idea is that if Black chases the
gS-Bishop away by . . . h6 (giving White a target to strike at), then a later
g2 -g4-gS will rip Black's kingside apart.

1 0.Rhe l 0-0
This, the main line, promises sharp play thanks to the fact that both sides
have castled on opposite sides.

I 1 .Kb l
White has also tried I l . Q d 3 , I l . Q d2 , and 1 1 .e S .

1 1 . . .QaS 1 2 .Qd2
Giving gS more support and threatening 1 3 .NdS Qxd2 ? ? 1 4.Nxe7+.

12 ... KhS
Black wanted to avoid any NdS tricks (he usually does this by 1 2 . . . Qa6,
12 ... Qb6, and 12 ... Qc7), but occasionally he plays 12 ... RfdS and dares White
to carry out his threat. Oddly, most players don't seem to have faith in
1 3 .NdS (the position after l L. Qxd2 1 4.Nxe7 + KfS I S .Nxd2 Kxe7 1 6. f4
h6 1 7 .Bh4 eS offers White very little) and instead continue with their
kingside aspirations by 1 3 .Nd4, when f2 -f3 followed by g2 -g4 will follow,
and N dS is in the air.

1 3 .Nd4 RacS
In Ciric-Moser, Baden-Baden 1 9S 5 , B lack tried 1 3 . . . RfeS 1 4. f4 RadS ,
but after I S . h4 Q a 6 1 6 . fS e S 1 7 .NB R d 7 I S . Bxf6 Bxf6 1 9 .NdS BxdS
20.QxdS RcS 2 1 .g4, White had a winning advantage.

1 4.f4
The other plan is 1 4. f3 RfdS I S .h4 BeS 1 6.g4, with a sharp game.

14 ... h6 I S .h4!?

Teach i n g a
you n g p u p i l at
the Marshall
Ch ess Club.

A common idea. White dares Black to open up the h-file and give White a
direct road to his King.

1 5 ... Nxe4
This doesn't work out, but taking the Bishop is extremely dangerous. One
sample: 1 5 . . . hxg5 1 6.hxg5 Nh7 1 7 . Rh l g6 l S . g4 Kg7 1 9. Rxh7 + Kxh7
2 0 . Qh2 +, and mates. Perhaps Black had to try something like 1 5 . . . RfdS,
though White's attack remains very strong: 1 6. g4 hxg5 1 7 .hxg5 Nxg4
l S .Rh l + KgS 1 9. Qg2 Nf6 2 0 . Qh2 KfS 2 1 .QhS+ NgS 2 2 .Rh7, and the rest
would be nothing but violence.

1 6.Rxe4! hxg5
This isn't pleasant, but 1 6 . . . Bxe4 1 7 .Bxe7 is too easy for White.

1 7.Nxc6 Rxc6 1 8.hxg5 Qf5

Black is still experiencing enormous difficulties because of the open h
file. This Queen move doesn't help, but perhaps there wasn't a good de
fense.

1 9.94! Qg6
No better is 1 9 . . . Qxg4 2 0 . Rh l + KgS 2 1 . Qh2 f6 2 2 . Rg l Qf5 when 2 3 .g6
ends the game.

20.Rh 1 + Kg8 2 1 .Qh2 f5

It looks like Black's King will be able to escape by the f7 -square.

PAL BENKO: MY LIFE, GAME AND COMPOSITIONS

22 .Rxe6! Qxe6 2 3 .Qh7+ Kf7 24.gxf5 , 1 -0 .

Black resigned because 24 . . . Qc4 2 5 .Qh5+ KgS 2 6 . g6 leads t o mate. This
game earned me a brilliancy prize.
404

(136) Benko - Grunfeld (New York, 1986)

l .NO Nf6 2 .g3 g6 3 .Bg2 Bg7 4.0-0 0-0 5 .d4 d6 6.Nc3!?
Of course, more natural is 6.c4, with a common King's Indian Defense.
By playing 6.Nc3 , I intend to follow up with e2 -e4 with transposition to a
Fianchetto Pirc, a line that I have a lot of experience with.

6 ... d5!?
Black stops e2 -e4 at the cost of a tempo, counting on the fact that my
Knight on c3 isn't well placed since it's blocking the c-pawn.

7.Ne5 Be6
Probably the most solid choice is 7 . . . c6, which was played in Ivkov-Fischer,
Santa Monica 1 966 (and in many other games) .

8.e4 c5?!
Griinfeld plays enterprisingly. Safer alternatives are S ... c6 and S ... dxe4.

9.Be3!
I would gain nothing from 9.dxc5 Nxe4.

9 ...Nc6
Perhaps 9 . . . Nxe4 1 O.Nxe4 dxe4 1 1 . Bxe4 cxd4 1 2 . Bxd4 Qc7 was a better
choice.

1 0.Nxc6 bxc6 1 1 .dxc5 Rb8?!

My opponent continues to play for the initiative, but this move might be
stretching things . Better was 1 1 . . .Nxe4 1 2 .Nxe4 dxe4 1 3 .Qc 1 (Far better
than 1 3 .Bxe4 Bxb2 1 4. Bxc6 RbS 1 5 . QxdS RfxdS 1 6 . Rad l Bxa2) 1 3 . . . QbS
1 4.c3 , with a complicated game in store which, in my opinion, offers White
slightly the better chances .

1 2 .b3!
White accepts the challenge and hangs on to his extra pawn.

1 2 ... Nxe4 1 3 .Nxe4 dxe4 14.Bxe4!

RETIREMENT

This Exchange sacrifice is the point of my play.

14 ... Qa5
After 14 . . . Bxa 1 1 5 . Qxa 1 , White threatens to win the Exchange back by
Bh6. If Black defends against that threat, then Bxc6 gives White more
than enough for the sacrifice. Therefore, Griinfeld chooses to complicate
the issue .

1 5 .Bxc6 Bh3 1 6.Re l Bc3 1 7.Bd4!

Now the situation begins to clear up.

1 7 ... Rbd8
After 1 7 . . . Qc7 ! ? 1 8 .Bxc3 Qxc6 1 9. B Qxc 5 + , White has the strong reply
2 0 .Bd4! . My opponent also had no stomach for taking either of the Rooks.
For example, 17 ... Bxe 1 1 8 . Qxe 1 is very nice for White.

1 8.Bxc3 Qxc3 19.Qe2 Rd2

White ends up with a clean extra pawn after 1 9 . . . Qxc5 2 0 .Bg2 (2 0.BB is
also good.) 20 . . . Bxg2 2 1 .Kxg2 Rd7 2 2 .Rad l .

20.Qe5
Far stronger than 2 0 . Qxe7 Rxc2 2 1 .Bg2 Bxg2 2 2 .Kxg2 Qxc5 2 3 .Qxc5 Rxc 5 .
Why should I let Black capture my powerful passed c-pawn?

20 ... Qxc2 2 1 .Be4

With this attack, I force simplification into a superior endgame.

2 1 . . .Qb2 22.Qxb2 Rxb2 2 3 .Reb l Rd2 24.b4

The win is now just a matter of simple technique.

24 ...f5 2 5 .Bg2 Bxg2 26.Kxg2 e5 27.Rd l

I must act quickly because Black threatens . . . f5 -f4 followed by . . . e 5 -e4.

The first rule in such endgames is, "No counterplay allowed!"

27 ... Rb2
Black would gain no relief after either 2 7 . . . Rc2 2 8 .Rac 1 ! or 2 7 . . . Re2 2 8.Re l .

28.Rab l !
Returning the pawn i s a shortcut to victory.

28 ... Rxa2 29.Ra l ! Rxa l

Black would fare no better after 2 9 . . . Rb2 3 0. Rxa7 Rxb4 3 1 .Rdd 7 .

405

PAL BENKO: MY LIFE, GAME AND COMPOSITIONS

30.Rxa l Rf7 3 1 .c6 Kf8 3 2 .b5 Ke8 H.Rxa7!

I planned to play this combination in response to any black King move . I
can also cash in my advantage by 3 3 .b6 ! .
406

3 3 . . . Rxa7 3 4.b6, 1 -0 .
O n e of the white pawns will turn into a Queen.

(137) Benko

Garcia (New York Open, 1988)

l .c4 Nf6 2 .Nc3 e6 3 .NB Bb4 4.Qc2 0-0 5.a3

White sacrifices a couple tempi to seize one of Black's Bishops.

5 . . .Bxc3 6.Qxc3 d6 7.e3

Also possible were other developing moves like 7 .g3 (the most popular
choice) or 7 . b4 ! ? Less logical is 7 . d4, since White surrenders control of
the e4-square without having any good reason to do so.

7 ... e5 8.d3 Bg4 9.Be2 e4 1 0.dxe4 Nxe4 I 1 .Qc2 Re8 1 2 .b3

Notice that White is avoiding h2-h3 , since that would only force the Bishop
to where it wants to go (i . e . , . . . Bg4-h5-g6).

12 ... Nd7 1 3 .Bb2 Qe7 1 4.0-0 Bh5 1 5 .Nd4 Bxe2 ? !

Black has more or less equalized, but this exchange gives White some ini
tiative. Preferable was 1 5 . . . Bg6 . This pawn formation is generally better
for White, especially if he can post his Queen Knight on d 5 . Here, how
ever, that Knight has already been traded away. Now, the most logical play
would be 1 6.Nxe2 followed by Nf4, but I was interested in seeing how my
opponent would defend against the Nf5 threat and my advance in the center.

16.Qxe2 f6 1 7.3 Nec5 1 8.e4 Qf7 1 9.Qc2 Ne5 20.Rfd l as 21 .Rac 1

Black's problem is that White controls all the pawn breaks in this position
- the second player must constantly be alert to b3 -b4 (chasing the Knight
from c5) or f3 -f4 (chasing the other Knight from e5). My last couple of
moves were designed to keep Black passive . His Knights can't jump into
d 3 , . . . c7 -c6 is met by Nf5 , and my c4-pawn is given protection in antici
pation of an eventual b3 -b4 push.

RETIREMENT

2 1 . . . Nc6 22.Nb5
White declines the trade, expecting great things from his Knight at b 5 .
However, more accurate would have been 2 2 .Nf5 with the same plan.

22 ... a4! ? 2 3 .h4 Nb3

This Knight may become either an asset or a liability later on. For ex
ample, if war is declared on the kingside, the b3 -Knight will find itself
stuck far away from the field of action.

24.Rh l Ne5
Simpler was 24 . . . Na7 , forcing the exchange of White's remaining Knight.

25.Bxe5 Rxe 5 ? !
Better was 2 5 . . . fxe 5 , securing the d4-square for the Knight a t b3 .

26.Qd3
I want to play Nc3 , but first I have to give my c-pawn some support.

26 . . . f5
Black must play actively because of the threatened Nb5 -c3xa4.

27.Nc3 fxe4 28 .fxe4 Qe8 29.Rfl !

I could have won a pawn by 2 9 . Re 1 followed by Nxa4, but this would
have left me with serious technical difficulties to overcome. Instead, I de
cided to take advantage of Black's absent Knight and go for a kingside
attack.

29 ... Qc6 30.Nd5 Qd7 3 1 .Rf4!

White is ready to break through on the f-file, and Black is too late to pre
vent it.

3 1 . . .Ree8?
This leads to a quick demise, but 3 1 . . .Rf8 is no better: 3 2 . Rxf8+ Kxf8
3 3 .Rf1 + Kg8 3 4. Qg3 Qd8 3 5 .Nf6+ Kh8 3 6.Nd7 ! Re8 3 7 .Rf7 Rg8 3 8 .Nf8 !
(Threatening 3 9.Ng6+ hxg6 40. Qh 3 + , and mates.) 3 8 . . . Qxf8 3 9 .Rxf8 Rxf8
40.Qg5 Nd4 4 1 .h3 , with an easy win for White. Just as hopeless for Black
is 3 1 . . . Rae8 3 2 . Qg3 Kh8 3 3 . Rbfl . He should probably try 3 1 . . . Q d 8
3 2 .Rbfl c 6 , though 3 3 .Rf8+ (3 3 .Nc3 i s even stronger, but the lines after
3 3 . Rf8 + are quite interesting) 3 3 . . . Qxf8 3 4 . Rxf8 + Rxf8 3 5 .Nc3 Re6

407

PAL BENKO: MY LIFE, GAME AND COMPOSITIONS

3 6.Nxa4 Rxe4! 3 7 .h3 (The endgame after 3 7 . Qxe4 Rf1 + 3 8 .Kxfl Nd2 +
3 9 . Ke 2 a l s o gives White g o o d c h an c e s . ) 3 7 . . . R e 1 + 3 8 . Kh 2 Nc 1
3 9. Qxd6 Rfe8 40.NcS still leaves Black in real trouble.
408

3 2 .Rbfl Rf8 H.Qh3 ! , 1 -0.

Black is losing his overloaded Queen (3 3 ".Rfd8 3 4.Rf8 + ! ) since 3 3" . Qxh3
3 4.Ne7 + forces mate. Garcia pointed out that 3 3 .Nb6 ! also wins (3 3 . . . cxb6
3 4. Q d S +) , though in that case Black could cut his losses via 3 3" . Qe6
3 4.Nxa8 Rxa8 .

(138) Benko

F Roed er (Augsburg, 1991)

l .e4 c5 2 .NO Nc6 3 .d4 cxd4 4.Nxd4 Nf6 5 .Nc3 e5 6.Ndb5 d6 7.Nd5
A respectable positional system that avoids the well analyzed lines that
result from 7 . BgS .

7 ... Nxd5 8.exd5 Ne7

The alternative is 8 . . . Nb8 .

9.c4 Ng6
More common is 9 . . . NfS , but Black didn't want to block his f-pawn.
A well-known blunder is 9 . . . a6? ? (9 . . . g6? ? meets with the same fate) 1 O.Qa4 ! ,
when White wins.

1 0.Be3 a6 I 1 .Nc3 Be7 1 2 .Qd2

Stopping a possible . . . BgS trade. If Black continues to insist on the swap
by 1 2 . . . h6, then 1 3 .Bd3 would follow.

12 . . .f5 1 3 .0 0-0 1 4.Be2

The plans are clear for both sides: White will try to make use of his
queenside spatial plus (and his majority of pawns in that sector) by playing
for a c4-cS advance. Black will attempt to strike back on the kingside.

14 ... Bd7
Too pedestrian. Black had to sharpen things up by 14 . . . Bh4+ I S .g3 f4
1 6. Bf2 BgS 1 7 . Qc2 Bh3 , with a double-edged position.

1 5 .0-0 f4 1 6.Bf2 Bh4 1 7 .Bd3 Bxf2 + 1 8.Rxf2 Nh4

Black succeeded in trading off his bad Bishop for White's strong piece,
but he had to give up the e4-square to accomplish it. However, it's not
easy for White to occupy this square because Ne4 would be met by . . . NfS
when Black's Knight would also have a future.

1 9.Rc1
Continuing t o make preparations for the " obligatory" c4-cS advance.

1 9 . . . a5 20.Ne4 Bf5 ?
B lack h a d t o try 2 0 . . . NfS , though 2 1 . cS would still l e a d t o a White
advantage. After 20 . . . BfS , Black's Knight becomes stranded on the edge
of the board.

RETIREMENT

2 1 .cS
White takes the initiative and never lets it go.

2 1 . . .Bxe4 22.Bxe4 dxcS 2 3 .d6!

409

It's clear that Black's strategy has completely failed. White's Bishop is far
stronger than Black's Knight, White's passed d-pawn is a monster, and
Black's pawns on b 7, c S , and eS are all vulnerable.

2 3 . . .Ra6 24.QdS+ Rf7 2 S .Rd2!

Far stronger than 2 S .Bxh 7 + Kf8 2 6.Be4 Rxd6 2 7 . QxcS , which is also good
for White.

2 S ... Ng6 26.RxcS NfS 27.Rc7 Nd7 28.Qe6, 1 -0 .

Black has n o answer t o the threat o f 2 9.BdS .

Often people are asked if they would do it all over again if given the chance .
Though I could have done without the horrors of war and imprisonment, I would
happily embrace the life of a chess player again. Travel to exotic lands, the brac
ing rush of mind versus mind over those sixty-four squares, the deep satisfaction
that only artistic creativity can bring, and the companionship of many close friends
the world over-what more could a man ask from life? When I add this to my
fortune in meeting my wonderful wife, combined with my two amazing children
(my son has a Ph. D . in math while my daughter has a Masters in math and is
working towards her Ph.D.), I can't feel anything but deeply blessed.
Though my chessboard battles are now behind me, I still lead a rich life . I am
working on some books, I follow all chess news, meet with old friends, travel,
coach young prospects , and exercise my creative urges by composing new
endgames.
I hope you have found the tales of my interaction with various chess legends
interesting. If my life story, with its struggles and triumphs, has given the reader
insight into human nature and the subculture of chess, and if my games have
proven enj oyable and/or instructive, then I will be satisfied that the effort that
went into the creation of this book was more than worthwhile.

Evans, l eft, and Gross, 1 9 5 7, wolfing down a meal afte r a hard d ay of blitz and cards.

SILMAN: Ron, I know you and Pal are very close. When did you first meet?
GRO S S : When Benko first came to the U . S . , he was floating about, staying
anywhere he could while he figured out how to survive . He ended up in
Cleveland for a while with his father. At the end of 1 95 7 they had a big
tournament in Dallas. Though there wasn't a place for him in that event, he
decided to take a chance, leave the security of Cleveland, and went there
anyway. Remember that he didn't speak any English at all, and nobody re
ally knew how strong he was.
It has to be remembered that, unlike Szabo who was a communist guy
that always towed the party line, Benko never bought in to that garbage and
was always his own man. After the war, he was the second best player in
Hungary, behind S zabo. Actually, he was quickly catching up to S zabo, but
people in the U . S . just didn't know who Pal Benko was.
In early 1 95 7 grandmaster Larry Evans found himself with some money
to invest, so he bought a motel in Compton. I lived in that same area of Los
Angeles, and Larry and I became friends. Evans, who was playing in Dallas,
sent me some clippings from the event - that was the first time I had ever
heard of Benko.
After the tournament in Dallas, Evans drove both Benko and Bent Larsen
back with him and put them up in his motel. The occupancy rate at his
motel was fifty percent, so they were able to stay there comfortably for free.
SILMAN: Benko insists that Evans made him pay to stay there, but Evans says
he didn't.
GRO S S : Well, now I'm not sure what the truth is. It was a long time ago.
4 13

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

SILMAN: How was Benko doing financially?

GRO S S : When Benko got to Dallas he had $2 00. He earned $500 for the match
414

with Smith that was held concurrently with the Dallas tournament. When
he arrived in LA (after spending some money in Texas and during the trip to
California), he had a grand total of $600 as a nest egg- that was everything
he had in the world !
So he was forced to live like a mouse, but he had been through so much
poverty and hardship in Hungary that he faced his situation with a lot of
courage. However, like so many from his time, and so many who experi
enced bitter poverty when they were young, he still lives very modestly to
this day, even though he's actually done very well for himself.
SILMAN: Ron, what did Benko do in California?
GRO S S : Benko and Larsen were killing time in Compton- they both had to
wait a few weeks for other events to begin (Benko was waiting for the Inter
zonal, while Larsen was waiting to leave for a tournament in Mar del Plata),
so they decided that this would be an ideal place to rest up and experience
California firsthand. Compton was a really nice place in those days, though
now it's changed radically and is close to being the murder capital of the
world.
My father owned a big, old-fashioned drug store that I worked at, and
Benko would go with me and check out everything. Benko wanted to see
what Americans did in their spare time, he really wanted to fit in, and he
was very cunous.
I was a student at Compton College, but I'd go on deliveries for my father
(I'd take prescriptions to people) and take Pal along. I'd show him various
restaurants, the cheapest places to eat; in fact, after the deliveries were com
pleted he would often come to my apartment and eat there.
SILMAN: Did you, Pal, Larry, and Bent all get along?
GRO S S : I went to Evans' motel everyday, so I began to take both visiting grand
masters everywhere. I introduced them to all my friends, they were hosted
at clubs, had dinners, and were honored. However, Larsen got the most out
of this simply because Pal didn't speak much English. Naturally, we all be
came very close friends.
Larsen, who was a sensation at that time due to his great result at the
Moscow Olympiad, was a tremendous storyteller, and he also spoke thirteen
languages fluently. He was a bit like Smyslov in that he was shy and got
embarrassed easily. But he was very gracious and made many nice contacts.
Benko and I hung out quite a bit. He showed me a lot about chess in gen
eral, and we played endless blitz games. In fact, all four of us played con
stantly. Benko did the best. Evans was close on his heels, and Larsen was a
distant third. Evans and Benko pounded me mercilessly, but I won several
games against Larsen.

With Ron at the FI D E World

Cham p i o n ship in Las Vegas

( 1 999 ) .

We played s o much five-minute that w e broke the clock. Since w e didn't

want to stop, we got an egg timer, put lipstick on the end of it, divided the
sand equally in half, and turned it over each time someone moved- the person
who ran out of sand first was the loser.
SILMAN: Ron, I'm a bit surprised that someone as incredibly strong as Larsen
would do so poorly at blitz.
GROSS: It was all about style. Benko and Evans played very correctly, but Larsen
was a wild man in these sessions ! He sacrificed everything and lost a lot of
games as a result. However, even though he wasn't doing well at blitz, he
was a real gentleman. He showed us lots of new ideas in the openings. He
showed me an important theoretical novelty that he had prepared as an im
provement on a Korchnoi game from the Russian Championship. Bent never
got to use that new idea, but I was able to win a game with it ten years later!
I loved Larsen. He had a great sense of humor, would happily talk to
people when he gave a simultaneous exhibition, would always j oke around,
tell incredible stories, and was also classy and distinguished. He was just a
wonderful man. His only problem was a rigid set of morals. He was terri
fied of women, though he was highly attracted to them. He was very shy in
this respect. Odd for a man who was so handsome and who was such an
incredible speaker.
When we were all hanging out in Compton, Evans loved playing the devil's
advocate by talking about women and making Larsen blush. Larsen is a real
games player, and he likes to win at chess, hearts . . everything. But Benko
was better than the rest of us at blitz, and Larsen was the best hearts player, so
Evans tried to do anything he could to get the upper hand in any other area.
For example, Evans loved to talk about going to brothels in Mexico. Benko
refused because he didn't want to spend the money, while Larsen was simply
horrified and didn't want any part of it! Benko always gauged his activities
according to expense, Larsen was very straight-laced, and Evans looked at
everything and anything as an adventure . . . he loved to experience things

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

firsthand : beautiful girlfriends, gambling, always making sure that he lived

life to the fullest.
416

SILMAN: Actually, Benko mentioned this to me. He said he refused because it

was against his principles. In fact, he told me that Evans once offered to
treat him to this kind of "service, " but Benko wasn't interested - he just didn't
think it was right.
GRO S S : That's probably true. I know a lot of well-known American players who
have been with prostitutes, but Benko never showed any interest whatsoever.
SILMAN: Ron, do you stay in touch with everyone?
GRO S S : Benko and I stay in touch, and I've remained close to Evans. For some
reason, Larsen has distanced himself from me.
SILMAN: Why is that?
GRO S S : I'm not really sure, but we did go through a couple "episodes." The
first occurred in ' 5 7 , when we were living together (he eventually left Evans'
motel and moved into my apartment with me). I was playing at the local chess
club and faced Irving Rivise, a strong California master. Rivise had beaten me
in our four previous encounters, so I really wanted to win this one. I played a
great game and was winning easily. Larsen was watching and seemed to be
enjoying the game. We adjourned and Larsen congratulated me on a nice
effort. Sadly, instead of sealing a move that would have won immediately, I
instead sealed a losing blunder (we had just gone through a time trouble situ
ation and I sealed my move rather quickly) ! He told me that I was losing after
that move but I refused to believe it. So we went home and he set it up on the
board and showed me that I was toast. He began laughing and telling me
what an idiot I was for not following his advice (he had told me that, after
time trouble, you should get up, calm yourself down, and relax.). I was really
angry with myself, but on seeing how much he was enjoying the situation, I
jokingly said, "Hell, I think I'll go back and change my sealed move . " You see,
the sealed move was left hanging on the wall at the club in its envelope.
Well, Larsen jumped all over me. He freaked out! It was like he pulled out
a FIDE badge. He began ranting about being a FIDE grandmaster. How he
had to uphold FIDE rules. He went on and on, clearly thinking that I really
intended to cheat and change the move.
I tried to convince him that I wasn't serious, but we went to bed with a
sour taste in our mouths. The next day I met with Rivise and told him the
whole story, how I had sealed an idiot move, that I was resigning the game,
and how Larsen thought I was serious about sneaking into the club and chang
ing the sealed move . Rivise thought the whole thing was hilarious. Unfortu
nately, Larsen thought I was serious about my "plot, " and he wanted Rivise's
phone number so he could warn him against my dark plan. I gave him the
number (after telling him that I had already resigned the game), and he ac
tually called to make sure that I had indeed given up !

INTERVIEW WITH N ATIONAL M ASTER RONALD GROSS

The second "blip" was in 1 96 6 , when Larsen came to town for the
Piatigorsky tournament. I told him that I had been looking at some chess
with Fischer, but that Bobby didn't want a second. Then I asked Larsen if
he wanted a second, and said that I'd be happy to help in any way I could.
He got insulted. I guess he thought that Fischer had refused help, so it would
make him look bad if he accepted
my o ffe r. He actu ally s a i d , " I
would never take a second ! " Of
course, I wouldn't really be able to
help him - I wasn't close to him
in strength and he always had to
show me what was going on. But
my offer was made in innocence,
I just liked him and wanted to help
an old friend .
In 1 968 he returned to play in
the U . S . Open in Aspen, Colo
rado. He told me that he liked to
come to the U . S . and beat up on
American masters. He said, and I
quote, "I enjoy that very much. "
Larsen loved having a small night

Larsen, in his p ri m e , battling Spassky at Santa

Mo n ica, 1 9 66.

cap, then sitting back and talking

about chess and/or politics (he always knew a lot about world politics) . One
night Browne, Tarjan, and I went to his room and we got into a long argu
ment about the necessity of NATO keeping an eye on things after the war.
He was completely against NATO . I knew a lot about NATO at that time,
and it kind of irritated him because I was making good points and holding
my own in the conversation. Finally he decided that it was late and that he
wanted to sleep, so we all left.
Next morning, I went to breakfast and all the newspapers had these huge
headlines about Russia invading Czechoslovakia. In the meantime, Larsen
was still sound asleep (he always slept late) . Anyway, I rushed upstairs with a
couple of these newspapers and pounded on his door. Eventually he answered,
visibly upset that I'd wake him up.
"Gross, what are you doing, waking me up at this hour? "
I guess I was overexcited, because I stuck both papers into his face and
said, "Remember what we were talking about last night? Well, look at these
headlines ! "
He was totally shaken. In fact, his face lost all its color. He opened the
blinds to let in light, told me that he had to read about it, and asked me to
leave so he could contemplate the whole situation. Ever since the '68 U . S .
Open, he's been cold and unpredictable in his behavior towards m e . At times
he's been friendly, and other times he's been rude or oddly distant.

417

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

SILMAN: We seem to have forgotten about Benko, so let's get back to him.
Did you see Pal when he played in the 1 963 Piatigorsky Cup?
418

GRO S S : Pal always stayed with me whenever he came to Los Angeles. For the
Piatigorsky event, he arrived three weeks before it began. I remember us
preparing a line for Keres, who was a very difficult opponent for Benko.
The variation was: l . e4 e6 2 . d4 d5 3 .Nd2 e5 4.dxe5 dxe4 5 .Nxe4 Qxd 1 +
6.Kxd l . Later we found a bust to i t (Fischer thought that the position after
6.Kxd 1 was probably in White's favor, but he insisted that 5 . Qe2 ! simply
refuted the whole line.), but to my knowledge, this refutation still hasn't
been published. Anyway, in the second half of the tournament Benko was
Black against Keres. Pal played the French, Keres played 3 .Nd2 and, Pal
chickened out and played 3 . . .c5 . Benko lost this game and was very upset
afterwards for not giving it a try.
We actually did a lot of opening work before the Piatigorsky - he was
very serious about preparing for this event. He prepared a specific line in
the Griinfeld against Petrosian. During this game, Frank Sinatra sat next to
me in the audience, took a long look at the position, and said, "The champ's
going to win."
I told him that Benko was fine, and that we had prepared the very position
on the demo-board. Unfortunately, Sinatra proved to be right and Pal lost
the game.
Benko had a difficult time in this tournament. In fact, the very first round
set the tone: Pal was playing Sammy Reshevsky, a man who had no person
ality and no sense of humor. Sammy wasn't well liked by the other players.
The assistant tournament director who was in charge of that game was Jack
Moskowitz (an old time friend of Sammy's), and Benko was winning or draw
ing the game. Reshevsky's flag went down on the 3 8th move and the in
structions were very clear: it was not up to the players to claim, it was up to
the director. I saw the flag drop, but Benko was low on time and wasn't looking
at the clock. Then I looked at Jack's face and watched him stare directly into
the clock. He saw that Reshevsky had forfeited! Incredibly, he didn't say any
thing. After the 3 8th move was played by Benko, I still thought he'd notice
Reshevsky's fallen flag on the 3 9th (Reshevsky was very aware that his flag
had fallen !), but Jack still kept quiet.
After the game was played (and Pal managed to lose due to time pressure
errors), I told Benko and he was outraged. He asked me to bring this up to
Jack, and Jack looked a bit flustered and said that he didn't notice it, that he
wasn't sure if it fell or not; of course, I knew this wasn't true.
The Piatigorsky Cup was located in the Ambassador Hotel, and it was an
incredible event. A great location, the best players in the world under one
roof (Keres was very quiet and would vanish right away, he wasn't social at
all. Petrosian was the father confessor to many players - they would rush
up to him after a game and ask him what he thought.), Gregor Piatigorsky
and Koltanowski telling endless stories to anyone who would listen (with

INTERVIEW WITH N ATION AL MASTER RONALD GROSS

the exception of Larsen, these two were the greatest story tellers I've ever
seen) . . . an amazing tournament.
Afterwards, Gregor Piatigorsky gave Benko a signed album as a gift, and
Benko gave it to me as a thank-you for helping him prepare.
SILMAN: During the many times Benko stayed with you, did you ever see him
composing problems?
GRO S S : He took problem composing seriously and often worked on them.
However, one stands out in my mind. I call it, The Great Marijuana Prob
lem. He had been agonizing over a specific problem for several days and just
couldn't make it work. I saw that he needed to relax so I lit up a j oint and
told him to take a toke. He didn't really know what it was and sort of freaked
out when it hit him. I managed to calm him down, though, and soon he
became very intense and turned his attention to the problem. He stayed up
all night and managed to finish it. It turned out to be one of his most suc
cessful helpmate problems, and players such as Botvinnik, Keres and Geller
were not able to solve it. Perhaps it can only be solved if you're stoned .
SILMAN: It's clear from your Piatigorsky story that you're not fond o f Reshevsky.

A young Fischer is on the far left, wh i l e Gross is i n the front row with shades, a striped s h i rt
and white sh oes. Ron and Bo bby were longti m e fri ends.

419

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

GRO S S : He was not a pleasant guy to be around. 1 remember that he once

went to a chess club to give a lecture and simultaneous exhibition. Many
people came just to hear him speak. When it began, he said, "I don't want
420

anyone to move until 1 get to your board, and no passes ! " That was it; that
was the lecture! He had a personality like a dishpan.
On another occasion, Tarjan, his girlfriend Sharon, and 1 were driving back
to Los Angeles from Lone Pine. Reshevsky needed to get to a North Holly
wood Synagogue before sundown and Sharon, who had often helped Sammy
find kosher food before this, promised that we would get him there on time.
Actually, we didn't have

clue where this place was, but the drive was

going to take a few hours so we stuck Reshevsky in the back, lit up a joint,
and enjoyed ourselves as we tooled down the highway. Unfortunately, we all
got a major case of the munchies about halfway to Los Angeles, so we stopped
at a market and began eating everything in sight.
Naturally, Reshevsky began to panic. Soon everyone noticed a joint that
was sitting around and we asked him about the drug culture . We went on
and on and he realized that we were about to offer him a hit. So 1 told him
that it would be a wonderful time for him to try it and added that it tasted
just like nuts. He said in a mean voice, "I don't like nuts ! " Then he sat there
in a bad mood until we got him to where he wanted to go.

SILMAN: Larry, do you have any recollections about you, Larsen, Benko, and
Gross coming together?
EVANS : I owned a motel in Compton, California that had about twenty units,
all separate bungalows. I think it cost around $60,000. In 1 95 7 , we all drove
there from Dallas in a car that Art Zeller and I arranged to deliver for some
agency to Los Angeles. Benko, Larsen, Art Zeller, and myself were the oc
cupants, and the trip took several days . When we arrived, Benko was imme
diately smitten with my girlfriend Clementine, an aspiring actress who man
aged the motel when I was gone . Both Larsen and Benko rented a room
from me for about a month and we usually all played hearts at night.
SILMAN: Did you actually charge them rent?
EVANS: The going rate was $2 5 a week, but I can't recall ever collecting any
rent from either Benko or Larsen.
SILMAN: Ron talked a lot about long, grueling blitz sessions.
EVANS: I remember playing hearts with Benko more than I remember playing
blitz, but I have no reason to dispute Gross' account. We all got a laugh out
of Benko when he got caught with the queen of spades and accused us of
ganging up on him !
SILMAN: Ron also mentioned you in relation to some trips to Tijuana. Any
comments on these "adventures? "
EVANS: None. They happened and I won't deny i t a s I don't believe i n censor
ship!
42 1

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

SILMAN: Back to chess ! Pal often

talked of you as a difficult oppo
nent. Why was that?
422

EVANS : I beat him the first time we

played and it seemed to leave an
indelible impression on him. The
only time he beat me was at the
1 974 U . S . Championship, but I
won our last game at Lone Pine

With Evans, 1 9 6 5 .

1 97 5 with a nice attack. I prob

(Photo Edward Laske courtesy USCF.)

ably did well against Benko because I had a knack for exploiting his time pressure.
SILMAN: You're famous for being a chess materialist. What did you think of
the Benko Gambit when he first started playing it? Also, did Fischer ever
share an opinion about it?
EVANS : I never discussed the Benko Gambit with Fischer, but always felt com
fortable a pawn up on the White side . I once gave Benko a chance to play it
against me in the U. S . Championship but he didn't avail himself of the op
portunity. I had prepared for that game and my idea was to quickly fianchetto
White's dark-squared Bishop in order to neutralize Black's counterplay on
the long a 1 -hS diagonal.
SILMAN: Both Benko and Gross enjoy talking about the ladies. Top sports fig
ures always have women clamoring after them, but professional chessplayers
don't seem to be as appealing. Did you ever find groupies in international
tournaments?
EVANS : No. The awful thing about chess is that there were no groupies. I hung
out with Fischer a lot in the Buenos Aires tournament and I introduced him
to a young lady there. It was the first time he ever got laid. Later I asked
him how he liked girls. "I'd rather play chess, " he said.
SILMAN: What are your impressions
about the other American grand
masters that you had to cross
swords with ? L e t 's start with
Reshevsky.
EVANS : Reshevsky was usually aloof
from the youn ger p layers and
closer to Horowitz and other
members of his own generation.
Essentially, he was a loner who felt
that God had ordained him to be
the Messiah of chess.

Fou rteen year- o l d Evans p l aying

Ameri can c h ess l egend Weaver Adams.

(Photo courtesy USCF.)

INTERVIEW WITH GRANDMASTER L ARRY EVANS

One problem with dealing with Sammy was his wife, who was a pest that
interfered in his games. At the U. S . Open in 1 9 5 5 I was analyzing some
game with Donald Byrne at the far end of the tournament hall while one of
Sammy's games was in progress. His wife came over and swept the pieces
from our board. "Stop analyzing my husband's game ! " she shrieked.
I recall that whenever he won a game on time
forfeit and it looked like a dispute was brewing,
he would simply get up and walk away in case
the decision was reversed. This happened in his
famous game with D enker where the referee
made a mistake and reversed the clocks, forfeit
ing Denker instead of Reshevsky.
When Sammy overstepped on time against me
at Buenos Aires 1 960 in a position which would
have been very difficult for me to win, I took a
page out of his book and vanished while he was
protesting. The forfeit stood. It took place in the
first round and I remember thinking this deci
sive result sent a message to the Russian players
there : the Americans played as individuals, not
as a team. S a mmy later ti e d fo r fi rst with
Korchnoi.
SILMAN: The Byrne brothers?

Sammy Reshevsky. Is th i s the face

of the " Messiah of chess"?

(Photo courtesy U5CF,)

EVANS : Robert Byrne was a strong player, very se

rious about things in general . He was already a
master when I got into chess and he won our first few tournament games
until I evened the score later on.
Donald Byrne was more lighthearted than his brother and had a good sense
of humor. I always thought he was more talented than Robert, especially in
speed chess, but his tournament results weren't as good.
SILMAN: Lombardy and Bisguier?
EVANS : Lombardy was very tough to beat. Mter I won the U. S . Champion
ship in 1 962 , we played a ten-game unofficial match that I won narrowly by
5 1'2 -4 1'2 after losing the first game. He made enemies at the board because
competition seemed to bring out something ugly in him. He got sore at the
church because they wouldn't use him as a goodwill ambassador by letting
him go to tournaments whenever he wanted.
I still remember him playing roulette during the National Open at a Reno
casino while smoking a cigar and enj oying himself immensely. Somehow it
seemed unpriestly. I always wondered what drove him to the church .
Bisguier was a great natural talent who, like Capablanca, was too lazy to
study openings. He was five years older than me and was already a master

423

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

when I got into chess. We later became good friends. He beat me at the
start and lost to me at the end of our careers.
424

SILMAN: Do you know much about Raymond Weinstein?

EVANS : I didn't know Ray very well, but remember some incident at the chess
board where he lost his temper for no apparent reason. Perhaps this was a
harbinger of things to come. He killed a man with a razor and spent the rest
of his life in an insane asylum.
SILMAN: When one thinks of American chess, Reuben Fine always comes to
mind. Did you ever play him?
EVANS : He and Reshevsky were two icons when I was a kid. We drew three
tournament games when I was very young. Fine lost his taste for chess after
turning to psychology. Unlike Reshevsky, he was interested in bettering the
plight of chess pros and fought for things on principle.

I n a 1 9 5 7 exh i b it i o n , Evans shows

h i s fon d ness for the ladies by
d e m o n strating the q u e e n i n g of a
pawn in an u n u sually lovely fas h i o n .

(Photo Sherman Taylor, courtesy USCF.)

SILMAN: \Vhat's the story with Fischer's "shuffle chess? "

BENKO : Fischer shuffle chess was actually an idea of mine over thirty years ago,
though Fischer made changes to my original idea. I never liked theory and
thought that this would enable players to show their skills and not their memo
ries. Today, with computers and databases, this is more pertinent than ever.
Of course, Fischer used to be the greatest theoretician in the world, and
it's funny that he's endorsing this version of chess so he can avoid theory. I
mentioned this to him . . . told him that he used the system when it suited
him, but now he wants to bypass the question of theory because he doesn't
study anymore.
SILMAN: The game where Fischer beat you with a Rf6 sacrifice has been pub
lished everywhere. You were unrecognizable in that game. \Vhat happened?
BENKO : That was the U . S. Championship where Fischer got a perfect score .
He played Saidy in the last game . Tony sealed a bad move in a drawn posi
tion and lost.
Everyone thinks that this Rf6 game against me was something special, but
I don't know what's so great about it. I was exhausted for this game. I was up
all night necking in a car with a young lady . . . kissing and kissing. But it
didn't go beyond that, so the combination of no sleep and frustration led to
me losing badly to Bobby.
SILMAN: It seems as if you had a new girlfriend every tournament.
BENKO : Look, in those days international tournaments often took a month.
Was I supposed to be without feminine company for that whole time? Man

425

PAL BENKO: MY LIFE, GAMES AND CO M POSITIONS

does not live by bread and chess alone, so if a pretty young lady wanted to
spend time with me, I was all for it.
426

SILMAN: I know you and Bobby are great friends, but not many friends would
have given up their spot to an Interzonal. Why did you do it?
BENKO : When I gave my place to Fischer for the Interzonal in Palma de
Mallorca in 1 970, I was sure Bobby would advance to the Candidates Matches
and beat the Russians. My own career was nearing its end, so why shouldn't
I give Fischer a chance to embrace his fate? I never had any doubts about his
success, and he didn't disappoint me.
SILMAN: Does Fischer still look at chess games?
BENKO : Fischer enjoys looking at games, though not as systematically as he
used to. We also play blitz. When he was younger, he wasn't that good a
blitz player. Now he's great at blitz.
SILMAN: It's well known that Fischer is anti-Semitic. However, don't you find
it odd that most of Fischer's friends are Jewish or have Jewish blood?
BENKO : Yes, most of them are Jewish, but that's because a large percentage of
players are Jewish. He doesn't really pay attention to individual Jews, his
look at it is much more expansive and impersonal. If he personalized it, put
a face on his "enemy, " he wouldn't be able to hold onto his delusion. For
example, Hollywood is virtually run by Jews, but Bobby loves to watch movies.
However, when it's over and the credits are running, Bobby will point to all
the Jewish names and say, "See! AJew . . . another Jew! And another! " Names
without faces again.
J;t's really sad, because Bobby has lost a lot of friends because of this. For
example, Edmondson, who was president of the USCF at one point, used to
go with Fischer to international tournaments and sit by the door. If some
one entered and made noise, he'd push him out of the room and close the
door in his face. For some reason, though, Bobby began to call Edmondson
a " CIA rat." I don't know exactly why. People try to be nice to him and he
treats them badly in return, but I don't think he really knows what he's
doing. Bobby is just oblivious.
SILMAN: Do you feel Fischer was the greatest player of all time?
BENKO : Yes I do. Fischer was the best opening player of his day, and he was very
sharp, very exact. Tactically he was good, but nothing like Tal - he would only
go into something that he could calculate. All in all, though, it's hard to beat
someone who had the will to win of an Alekhine and the crystal clear style of a
Capablanca, mixed with flawless openings and superb endgame play.
Another thing that made Fischer excel was his focus -he almost never
allowed anything to interfere with his concentration. I think he learned his
lesson in 1 960, in Argentina. He did horribly there because he got caught

INTERVIEW WITH GRANDMASTER P AL BENKO

up in women and sex. Every night he went out, and he only returned very
late. A player like Evans wouldn't be affected by this schedule, but it de
stroyed Bobby's game. Afterwards, Fischer said he'd never mix women and
chess together, and he kept that promise.
SILMAN: Fischer often complained about fixed games and prearranged draws.
How do you feel about this?
BENKO : I see nothing wrong with prearranged draws, it's a part of the game.
But losing on purpose is another matter. This is simply stealing, and it's com
pletely unacceptable. Actually, Fischer's complaints about people making draws
should have made him happy, because while they drew, he kept trying to
win every game, and this should have turned out to his advantage.
I've always been impressed by players who go all out to win. Larsen comes
to mind -a friendly man that I always liked and respected. He was a great
tournament player, had terrific self-confidence (as Fischer did), always play
ing every game to win, and was a fine chess psychologist.
I remember that Larsen had already qualified in the Amsterdam Inter
zonal, and he was playing Smyslov in the last round. Smyslov, who needed a
draw to quali fy, had White and approached Bent before the game and of
fered to make a quick draw. Larsen instantly refused ! This was shocking. By
the way, Larsen almost won this game, though it ended up as a draw anyway.
Of course, I've seen this kind of courage on a few other occasions . In
Hastings, Tal offered Timman a last-round draw that would give them both
equal first, and Timman refused. Also Seirawan did this at the Olympiad. I
was the team captain and he played second board. In our match against Rus
sia, Tal offered a draw to Yaz and he approached me and said, "Tal offered
me a draw but I think I'm better so I'm going to refuse." And he won a very
nice game. Impressive, because Tal had a big name while Vasser was young,
but he had faith in himself, which you need if you want to succeed.
SILMAN: Do you have any experiences in prearranged draws? One personal
fiasco comes to mind where I thought an opponent (a friend of mine) and I
had agreed to draw, but then I noticed him setting traps and trying to win . I
offered a draw on move twenty, but he said, "It's too early. Let's play for a
while so it looks good for the audience . "
B y move fifty, I had a winning Rook and pawn endgame. Then h e offered
a draw, which I took as a matter of honor, though it made me look like an
imbecile to those watching. We weren't friends anymore after that.
BENKO: (laughs). Actually, your experience isn't uncommon. I remember leading
a tournament in Milwaukee (in the '60s) by half a point and only one guy
could catch up. So this guy in second offered to make a draw before the
game, saying we'd play a few moves and then agree. Naturally, I had no prob
lem with this.
So the game started, we played a few moves, and he hung a piece ! So now
I'm a piece up and in a very bad spot. If I agree to a draw, I'll look like a

427

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

complete idiot. So I spent my time desperately looking for a way to hang

the piece back in a semi-convincing manner!
I heard that Capablanca once agreed to make a draw before a game. Capa
428

won the game and his opponent complained afterwards. Capablanca said,
"How can I make a draw with you if you play so badly? "
I once had a similar piece of confusion with Fischer in a U . S . Champion
ship. There were two rounds to go. I was Black and Fischer would win the
title by splitting the point -nobody could catch him. I approached him be
fore the game and offered peace, and he made some mumbling noises that I
took as an agreement.
So we played and I got an excellent position in a Maroczy Bind, Gurgenidze
system, Accelerated Dragon. I always did well on the Black side of this, against
Korchnoi and other good players-I never lost a game with it. So I was bet
ter and I saw him trying to make strong moves, with no draw offer coming,
and I thought, The son of a bitch is trying to beat me! Perhaps he can hold
it, but he can't do better than equality. Now I had a problem: I could play on
and perhaps win, but then I risked going against an agreement that I thought
we had made. But if I played without verve, the position might turn against
me, so I really needed to know where I stood !
I looked at him and said, " D o you want a draw o r not? " H e agreed and I
went on, "You were trying to beat me ! " "Of course ! " "I thought we had an
agreement before the game? " He replied, "You said that, I didn't! "
Another incident with Fischer occurred earlier, in the Stockholm Inter
zonal. The position was microscopically better for him, but still a complete
draw and I was happy to split the point. However, I saw that he was playing
for tricks so I picked my piece up, waved it towards the square that would
have trapped me, shook my head to show him I saw his silly traps, and then
made the correct move. Only then did he agree to a draw.
Sometimes the draw offer doesn't even come from the players ! Tal and I
played in a Zonal tournament in Czechoslovakia where we were approached
by the organizers to make a draw in our individual game. So you see, draws
fall from trees like ripe fruit, it's just no big deal.
Of course, cheating does occur at times, and in many different forms. I
remember having to judge the brilliancy prize during some tournament. I
had it down to two wonderful games, and I gave the prize to one of them.
When it was announced, everyone began laughing and whistling, and I asked
why. Some players told me that the whole game was made up beforehand,
and I replied that, if I had been aware of this, I would have awarded the
prize to the other game. He said, "Don't worry, the other game was also
made up ! "
SILMAN: The modern players always like to say that they are much stronger
than the players of old. I brought this up to Anand and asked him if he had
studied the classics, meaning the games of Steinitz and Lasker. Vishy said,
"Oh yes ! I once made a real study of Larsen's games. "

INTERVIEW WITH GR ANDMASTER P AL BENKO

Do you think that there is a huge difference in strength between players

thirty or forty years ago and now?
BENKO : No, I don't think so. Of course, they are more booked up due to the
use of computers, but aside from that I don't see any big change in strength.
Yes, there are more good players today, but the best, like Botvinnik and
Smyslov, would still be great today. Don't forget that Tal was doing very
well right up until his death a few years ago, and Korchnoi is still kicking
the youngsters around.
SILMAN: How about a best ten list.
BENKO : I'd have to think about this. I wouldn't want to make a big mistake
like Fischer did on his top-ten list when he left out Lasker. Now he tells me
that Lasker was a truly great player. He's changed his tune.
SILMAN: You are known as an endgame expert. How did you become adept at
this phase of the game?
BENKO : I've always felt at home in the endgame . Somehow I'm comfortable
there, and I quickly realized that most other players didn't handle endgames
very well. I suppose my studying Capablanca's games helped a lot. World
Champions tend to be very strong endgame players, though Tal was an
exception.
I have always loved the endgame since this phase of the game doesn't bow
to fashion like openings do. I realized how vitally important the endgame
was when I lost a Knight-and-pawn ending to the great Savielly Tartakower
during my first international tournament in Budapest 1 948 . Now, when I
analyze an endgame, I search for the basic ideas and seek to understand the
position rather than to memorize every possible sub-variation. Unfortunately,
a game of chess begins with the opening and finishes with an endgame. I
would rather have it the other way round.
SILMAN: You are also famous for hanging Rooks and getting into constant time
pressure. Do you have any comments on these two painful afflictions?
BENKO : I began hanging Rooks early in my career, and have hung in excess of
twenty Rooks over the years. I guess it's because they don't move for a long
time; you just forget about them until an endgame is reached!
I remember my game (a Benko Gambit!) against Hort. I was doing very
well when he offered to exchange a pair of Rooks. For some inexplicable rea
son, I just ignored him and left my Rook sitting there without protection. My
girlfriend came up to me and said, "They made a mistake on the demo board
and forgot to trade Rooks. " This problem came back to haunt me again and
again. I wasn't even discriminating -I'd hang Rooks versus anyone at any
time !
Time pressure, though, has always been my worst enemy, right from the
beginning of my career.

429

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

SILMAN: At some point you must have tried to cure your time pressure
problem.
430

BENKO : I did, but I'd get so enraptured by the position's possibilities that I'd
forget all about my resolutions and let the time tick away. Early in my ca
reer, I played very well in time pressure. However, for some reason that I
can't quite grasp, my play became very unsteady when the flag got within a
couple minutes of falling. In fact, I even began to worry about getting into
time pressure during the game, which made me move slower since I wasn't
paying full attention to the actual position. In the end, the actual fear of
time pressure ultimately became a self-fulfilling prophecy.
SILMAN: When did you get into chess compositions?
BENKO: I started creating problems when I was young, though I didn't do many
endgame compositions until I was older. I ultimately decided that endgame
compositions were much closer to realistic chess, that the positions were
very similar to an over the board situation (unlike other compositions, that
would never occur in a game). Also, computers can solve normal mate prob
lems in a second or two. Endgame problems, though, are often too difficult
for even the strongest machine.
The prizes for compositions are very small amounts of money, so we do it
for love, not wealth . I can't describe the satisfaction I get from the creation
of a well-balanced composition !
About fifteen years ago, David Levy asked me to have a little contest with
one of his programs solving, first, three-move mates, and then four-move
mates. He set up a position, and as my eyes were focusing on the board,
before I really even looked at the position, the computer lit up (zoom !) and
gave the solution, virtually in less than a second. This happened over and
over for every three-move problem.
I was getting very discouraged, but the computer wasn't very good at four
movers, and I beat it. Now, of course, much longer problems are easy for
any of the commercial programs.
SILMAN: Who was the best problem solver you've ever met?
BENKO : Nunn was the best solver I've seen, and Suttles was also good at solv
ing problems- even blindfold. I just told him the position and he'd easily
solve it.
SILMAN: You talked about various players in the section on annotated games,
but there are some people who weren't addressed. First, let's start with
Addison.
BENKO : I remember being Addison's second at the Palma de Mallorca Inter
zonal. His one ambition was to finish ahead of Reshevsky, but he failed, ending
up half a point behind Reshevsky at the end. We were looking at his ad-

INTERVIEW WITH GRANDMASTER P AL BENKO

journment against Portisch and he complained, saying that I was finding all
the good moves for his opponent!
As it turned out, his wife only agreed to marry him if he quit chess. After
this, Addison vanished from the chess scene.
SILMAN: How about Reti? Like you, he was a very strong player who also loved
compositions.
BENKO : He was one of the very best endgame composers, and his interest in
the subj ect was similar to mine in that he enjoyed positions that could occur
in real games. I remember being told how Reti was playing in a tournament
and a friend approached him and asked how he was doing in his game. Reti
said, "Never mind that. However, this is really interesting ! " Then he would
begin to discuss his latest endgame composition, and it was obvious that this
was a much greater passion for him than over-the-board play.
SILMAN: Did you have any interaction with Spassky?
BENKO : Not much. In Spassky's prime, he was a great player. Later in life , he
became very lazy. He told me, "Fischer made me a millionaire, so why bother
working? "
SILMAN: There was always discussion about Bronstein losing on purpose to
Botvinnik. What do you think about this?
BENKO : I don't believe he purposely lost to Botvinnik. Brontstein said, "How
can one beat Botvinnik . . . he was like a god ! " I'm sure he felt some pressure
by the government, but I can't imagine him tossing the match on purpose.
He always grabbed me and said, "Let's go to the corner and talk! " He was
afraid of being overheard; in fact, he had a lot of fear bottled up inside. Some
times he would say strange things. He once told me (in Monte Carlo) "What
is the problem with the Czech people? So many Russian soldiers died in
their country trying to make them free, yet they constantly put us Russians
down. Why do they want us to leave? We should stay there forever! " He
was always saying anti-communist stuff, he was clearly afraid of the system,
yet he would occasionally say some odd, even stupid, things.
SILMAN: Have you ever read Fine's book on chess psychology? What are your
impressions of him?
BENKO : He was an exceptionally strong player in his prime, but he quit the
game and never reached his highest potential. He became a pure Freudian,
and his book about chess psychology made very little sense. In that book, he
obsessed on the Queen, saying how people that swapped Queens hated women
. . . things like that. However, in Hungarian the word for this piece has no
female connotations -it's not a Queen! So this completely refutes his view
about all chess players being anti-woman due to their treatment of this one
piece !

43 1

PAL BENKO: M Y LIFE, GA M ES AND COM POSITIONS

His early books were very impressive. He told me that he wrote his endgame
book, Basic Chess Endings, in just three months, which in the days before
computers was amazing.
432

SILMAN: I've read many seedy tales of Matulovic's behavior. Care to shed any
light on this?
BENKO : Most of those tales are probably true! Matulovic would always play to
the bitter end so the game would look better on paper. Also, if he adjourned,
even if he was in a resignable position, he would entertain the hope that you
would oversleep and forfeit, or perhaps have an accident.
SILMAN: Larry Evans ?
BENKO : He was a difficult opponent for me. He somehow always got me i n
time pressure and won several games. Mter that I calmed down and w e drew
many games, though I beat him once during this streak of split points.
SILMAN: In the preface to one of your games, you mentioned playing Najdorf
blitz. You have any more memories about this sort of thing?
BENKO: The first time I played Naj dorf in Budapest he offered me 1 0- 1 money
odds at seven-minute chess. He was a much better blitz player than I was at
this time limit (I'd beat him at five), but I sat down and won the first two
games - then it was hopeless for him since I was already twenty up !
SILMAN: Walter Browne?
BENKO : Browne had good opening preparation . . . he was easy to prepare for,
but he knew his stuff very well. He's a tactician, a calculator, and this was the
cause of his endless time pressure.
I played Browne in Wijk aan Zee and we adjourned. He told me, "Why
don't we analyze and if it's a win for you, I'll resign ? " We started to analyze
and I kept winning every line. He'd try another defense, and I'd refute that
too. Finally I had to stop because I realized that he was just picking my brain.
Fortunately, I won the game. Afterwards, he chose me for his second in the
Canary Islands. Perhaps he liked the way I analyzed.

with, I" me <>y th" my pCNon.1 int""ction with Pal Benko h"

been limited. Over the years I have been at many tournaments that he took part
in, but have played him only a few times and alas, seldom conversed with him.
As an admirer of his chess, however, I have always been struck by Benko's cre

ativity, most obviously in the area of endgames and studies, but less famously in
other aspects of his game. I will concentrate here upon Benko's extensive contri
butions to opening theory.
As a young player, I always watched Benko's games to see what he was doing
in the opening. He is not widely known as an opening theoretician, but in my
opinion he should be. As we shall see, Benko has repeatedly come up with new
moves and ideas in both rare and cutting-edge openings. Unlike most leading
U . S . players in the 1 95 0s to 1 970s, he was not afraid to try systems that he hadn't
played before and was quite unpredictable in the early phase of the game. When
I looked in my two million game database at a breakdown of openings by ECO
code for Benko's games, I found it remarkable how few codes were left empty.
That is, Benko has at some point played almost every established chess opening.
Furthermore, those database games constitute only a small sampling of his over
all praxis, which includes a great number of unsaved games from U . S . Swiss sys
tem tournaments.
One obvious clue as to Benko's importance in opening theory is that several
opening systems are named after him, most famously the B enko Gambit ( l . d4
Nf6 2 .c4 c5 3 . d5 b 5 ) , but also the Benko Variation of the Main Line King's
Indian Defense ( l 1 .g4 after 9 .Ne l Nd7 1 O. f3 f5) , the " Benko System" (a com
bination of Nf3 , g3 , and c4) , and the B enko Variation of the Sicilian Defense
( 1 . e4 c5 2 .Nf3 Nc6 3 . d4 cxd4 4.Nxd4 Nf6 5 .Nc3 d6 6 . B c4 Qb6). As we shall
see, there are a number of other variations and moves that we take for granted
435

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

today, of which B enko was either the d i scoverer or one of the le a d in g

developers.
In this survey, I therefore want to take a wide-ranging look at Benko's opening
436

ideas, particularly in the openings that he played regularly. I will do this o n an

opening-by-opening basis, beginning with White, and in general but not very
strictly following the order given by the well-known InformatoriECO opening
code (AOO through E99) . This mainly applies to opening complexes rather than
specific code numbers. Thus I look at the English Opening first (A1 O-A3 9, al
though not in that order), but reserve the Benko Gambit (A5 7 -A5 9) to combine
with 1 .d4 openings in DOO- E99. Sections will be explicitly named to reflect their
contents, with Benko's openings as Black in a completely separate second part.
My general approach has been to imbed all or most games with a particular
variation into a main game. I have tried as much as possible to make that main
game one that is included elsewhere in this book (citing it as " Game X") , with
other Benko games from other sources analyzed in the notes. In this way it
should be easier to find the theory related to a game in the main text. Some
times, of course, none of the book's annotated games cover the variation under
consideration.
What do we gather about Benko's style from this investigation? It is impos
sible to characterize a player's style without oversimplifying, but one will see
that Benko tends to play positional openings (with notable exceptions, particu
larly in his earlier games) . He often trades Queens at an early stage so as to
achieve small advantages in complex, Queenless middlegames (a specialty of his),
or just to retain double-edged prospects with no particular advantage. These
Queenless positions are by no means dull, and are highly recommended to any
student of the game . But Benko is also happy to enter into complications: many
of the positional openings he plays do not involve an immediate confrontation
of forces (Scheveningen Sicilians, English Openings, or even Benko Gambits,
for example), so piece exchanges may not occur at all and extremely complex
positional maneuvering results. What we don't see much of are openings char
acterized by early tactics or forcing main lines such as Poisoned Pawn Naj dorf
Sicilians or Exchange Griinfelds with innovations on move 2 3 . To be sure, I will
extend the analysis of some of the examples I use well into the middlegame, but
because the key ideas of the opening often manifest often themselves only at
that stage.
Before entering into this rather lengthy survey, allow me to tell a story that I
feel is revealing about Pal Benko. I was once in a tournament with mostly titled
players, and for perhaps the only time in my life I was not only alone in first
place going into the last round, but a full point ahead of the field and thus need
ing only a draw to finish clear first. The game became very complicated and my
opponent, a well-known international player, offered me a draw. We were both
in some time trouble, and as I recall I had something like five minutes left on my
clock to make a good many moves, perhaps 1 0 or so. Being young and foolish, I
considered playing for a win (I was in fact worse, as it turned out) .

INTROD U CTION TO OPENING SURVEY

So I used up all of my time except for about 45 seconds, and then accepted the
draw. To my astonishment, my opponent said, "I didn't offer a draw" and tried
to continue the game. When I protested, he repeated himself. In this moment of
sheer panic, Benko, on the adjacent board and right in the middle of his usual
very serious time trouble, took the time to turn to my opponent and say "You
offered the draw, " settling the issue. Not only had Benko broken his concentration to support me but he had also given up his own chance for a first-place tie.
A trivial story, one might say, and Benko himself has probably forgotten it.
But I think that it illustrates his integrity, a trait perhaps less common among
top players than might be supposed. Reflecting upon the hypothetical reactions
by players that I have known over the years, I think that in the same situation a
great many of them would have been indifferent, saying nothing, or even se
cretly pleased that a competitor was embroiled in difficulties. Sadly, another group
would have done some silent calculations (being chessplayers) and self-interest
edly decided against interfering. I don't think that remaining silent would ever
have occurred to Benko, whose commitment to a fair result reflects his overall
character.
Let's move on to the openings.

437

T H E E N G LI S H O P E N I N G
When I wrote my first chess book, a four-volume work on the English Opening,
I already had a notion of Benko's facility in that opening by having watched him
play 1 .c4 in tournaments. In the course of my research for that series, I was so
impressed with his numerous ideas (with both White and Black) that I men
tioned him in my Introduction as one of the most important contributors ever
to the theory of l .c4 systems . Thus, while the following section is quite long, it
only represents a limited selection of Benko's games and innovations in the
English Opening.

Sym m etrical English: l .c4 c5

The first games w e shall look a t feature Benko o n the white side o f the Sym
metrical English ( l .c4 c5). This is probably the most difficult defense for the
1 .c4 player to drum up chances against, although Black will often unbalance the
position to avoid a nagging disadvantage. Let's begin by looking at some games
in variations with high degrees of symmetry that have generally been considered
drawish, much more so than 1 .c4 e 5 or other 1 .c4 lines. Benko was nevertheless
consistently able to conjure up complex problems that tested the skills of his
opponents. Even top players weren't always up to solving them.

Benko - Trosclair, New York, 1 984 is such a purely symmetrical variation,

particularly known for its drawish tendencies. Benko demonstrates that this is
not the case : l .c4 Nf6 2 .Nc3 c5 3 .g3 g6 4.Bg2 Bg7 5.e3 0-0 6.Nge2 e6 7.Nf4
The first attempt to bother Black, but the game soon reverts to normal chan
nels. 7 ...Nc6 8.0-0 Ne7 9.d4 cxd4 1 0.exd4 d5 l 1 .cxd5 Nfxd5 1 2 .Nfxd5 Nxd5
We have arrived at a position that has been the occasion of numerous grandmas
ter draws, either immediately or, after 1 3 .Nxd5 exd 5 , within the next few moves.
Benko tries to spice things up : 1 3 .Qb3 ! An obscure move at the time; today, it is
still used to generate winning chances.

439

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

440

1 3 . . . Bxd4?! This is a natural move, and for some time it was the most important
response. The first thing to note here is that Benko himself had this position
with Black versus Bolbochan in Havana 1 966 and played 1 3 . . . Qb6, but without
achieving equality. That game went 1 4.NxdS exdS l S . Qxb6 ( I S . BxdS Qxb 3
1 6.Bxb3 Bxd4, 112_112, was Botvinnik-Smyslov, Moscow 1 9S 7 , an early game with
this line) I S . . . axb6 1 6.BxdS Rd8 ? ! 1 7 .Re l ! RaS ! ? 1 8 .Bb3 Bxd4, and now White
would have had a very nice position after 1 9.Be3 ! . It is therefore logical that
Benko now wanted to try his hand with White.
This is not an easy position, and theory demonstrates that there are many ways
to go wrong. In my opinion, 1 3 . . . Ne7 ! ? (intending . . . Nc6 or . . . NfS) is probably
best, despite its passive appearance, for example, 1 4.dS ( 1 4.Rd l Nc6 I S .Be3 Nxd4!
1 6.Bxd4 Bxd4 1 7 .Ne2 eS

=,

Pfleger-Matanovic, Hamburg 1 965) 14 . . . exdS I S .BgS

h6 1 6.Bxe7 Qxe7 1 7 .NxdS QeS 1 8 .Rab l Re8, Forintos-Sapi, Magyarorszag 1 969.

Naturally there is room for further exploration here. 14.Bh6 Bg7 14 . . . Re8
I S . Rad l Bxc3 1 6.bxc3 Qb6 1 7 .Qc4! Qc6 1 8 . Qe2 ! led to a winning attack in
Cardoso-E. Torre, Manila 1 9 7 3 , a game that inspired great interest in 1 3 .Qb3 .
For example, Torre himself later played 1 7 . Qa4 Qc6 1 8 .Qe4 as White, with per
haps an even bigger advantage. 1 5 .Bxg7 Kxg7 1 6.Bxd5! A new move, and con
siderably better than 1 6. Rfd l Qb6 1 7 .NxdS Qxb3 1 8 .axb3 exdS 1 9. RxdS Re8, or
here, 1 9 . . . a6 2 0.b4 Rb8 from Smyslov-Petrosian, Moscow 1 974. Benko's game
notes mention Black's successful defense in U. Andersson-Gheorghiu, Moscow
1 98 1 after 1 6.NxdS exdS 1 7 .BxdS as ! . Thus the formerly approved move 1 3 . . . Bxd4
is only cast into serious doubt by Benko's own strategy. 1 6 . . . exd5 1 7.Nxd5

BENKO AS WH ITE

White is surprisingly much better here, the dominating Knight and Black's
weakened dark squares causing the second player real difficulty, especially as the
natural 1 7 . . . Be6 runs into 1 8.Qc3 + f6 1 9 .Nc7 . See Benko's detailed notes in Game
1 3 4. The move 1 6.Bxd5 ! requires a great deal of positional insight, and this game
is very much worth playing over.
Benko's mastery of apparently innocuous symmetrical positions was effective
even against his famous opponent in Benko-Geller, Wijk aan Zee 1 969: I .NB

cS 2 .c4 g6 3 . g3 Bg7 4.Bg2 Nc6 S .Nc3 e6 6.0-0 Nge7 7.e3 0-0 8.d4 cxd4
9.Nxd4 dS Shades of the last game. 1 0.cxdS exdS l O . . . Nxd5 is held to be equal.
I I .Qb3 Benko mentions that Botvinnik called this a new move, "although the
position must have occurred in Soviet chess practice. " I l .Qb3 had been played
at least once, as seen in the note to Black's 1 2 th, but the game before you perma
nently changed the perception of l O . . . exd5 as a "drawing line. " 1 1 . . .Nxd4 Benko
gives 1 1 . . . Bxd4 1 2 . exd4 Nxd4 1 3 . Q d l with an edge. 1 2 .exd4 NfS 1 2 . . . Nc6
1 3 . Qxd5 ! ? ( 1 3 .Be3 ! Nxd4 1 4. Bxd4 Bxd4 1 5 .Nxd5 gives White better prospects)
1 3 . . .Bxd4 1 4. Qxd8 Rxd8 1 5 .Bg5 f6 1 6.Bf4 Bf5 was fine for Black in Doda-Donner,
Havana 1 96 5 . 1 3 .QxdS QxdS 1 4.NxdS! Benko mentions 14.Bxd5 Nxd4 1 5 . Bf4,
but after 1 5 . . . Nc6, this doesn't seem as clearly advantageous as the text. 14 ...Nxd4

1 5 .Bg5!

Posing Black all kinds of difficulties. The opening has been a great success for
White. 1 5 . . .Bh3!? 1 6.Rad l Bxg2 1 7.Kxg2 , with a nice advantage. See Game
93 for Benko's notes, and for the impressive conclusion.
The game Benko-Gilden, U.S. Ch. (Chicago) 1 974 features a line that was
to become ultra-theoretical over the years. Benko positionally outplayed his op
ponent in the opening and won a nice ending: l .NB c5 2.c4 b6 3 .g3 Bb7 4.Bg2

g6 5.0-0 Bg7 6.Nc3 Nf6 7.d4 cxd4 8.Nxd4 Benko quotes 8 . Qxd4 Nc6 9.Qf4
Rc8 l O.Rd l Nh5 I l .Qe3 Nb4! ? 1 2 .Rb l ! Rxc4 1 3 .Ne5 "with wild complications, "
Panno-Ljubojevic, Petropolis 1 97 3 . This was later repeated and shown to favor
White. Today, the dominant move at the top level is 8 . . . d6, as played, e.g., in the
recent Kramnik-Kasparov Braingames World Championship match. 8 . . . Bxg2

9.Kxg2

44 1

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

442

...

Qc8 9 . . . 0-0 1 O. e4 Qc7 I l .b 3 Nxe4! 1 2 .Nxe4 Qe5 , as in Karpov-Kasparov,

World Ch. ( 1 3 ) 1 984- 5 , leads to theoretical equality. This is probably the most
direct route if Black knows the theory. 1 0.b3 Qb7+ l 1 .Kgl!?

An independent move that has, surprisingly, only been played five times that I
know of. By contrast, 1 1 .f3 has a long history; Benko quotes the game Capablanca
Botvinnik, Nottingham 1 9 3 6 : 1 1 . . . d 5 (Benko gives this an " ! " . These days,
1 1 . . .0-0 1 2 . e4 d6 with a Hedgehog is considered equal. The queen is well-placed
on b7 to support . . . a6 and . . . b 5 .) 1 2 .cxd5 Nxd5 1 3 .Nxd5 Qxd5

1 4.Bb2 0-0 1 5 . Q d 3 Rd8 1 6. Rfd l Nd7 , assessing this as good for Black. But it
was later discovered that 1 4 Be3 ! was better for White, just as in the game here
(and a bit more so). Thus, although his assessment of 1 1 .f3 is not correct, Benko's
basic idea in the game is justified by later theory.

BENKO AS W H ITE

1 1 ...d5? Benko queries this move and suggests castling with . . . d5 to come,
since White cannot play e4. This is true, although later theory about Hedgehog
positions suggests that White doesn't really want to play e4 in a position like
this anyway. Therefore Black might try 1 1 . . .0-0 1 2 . e 3 ! ? RdS ( 1 2 . . . d5 1 3 .cxd5
Nxd5 1 4.Qf3 RdS 1 5 . Rd 1 , with a slight edge) 1 3 .Ba3 d6 (worse is 1 3 . . . d5 1 4.cxd5
Nxd5 1 5 .Nxd5 Rxd5 1 6 . Q f3 ) 1 4.Qe2 Nbd 7 , and if White has any advantage, it's
very small. 1 2 .cxd5 Nxd5 1 3 .Nxd5 Qxd5 1 4.Be3! An idea that Benko came up
with over the board. He correctly avoids 1 4.Bb2 0-0 and . . . RdS . 1 4.Be 3 ! had
been played once in Kottnauer-Bondarevsky, Moscow 1 946, but was not fol
lowed up accurately. 14 ... Nc6 1 5 .Nxc6 Qxc6 1 6.Rc 1

1 6 ... Qe6 Benko suggests 1 6 . . . Qb7 , when 1 7 .Qd3 0-0 l S.Rfd l still threatens Qd7.
Then the Kottnauer-Bondarevsky game went lS ... h 5 , when 1 9.Qd7 would have
still maintained the advantage. But perhaps l S . . . RfcS 1 9.RxcS+ RxcS 2 0 . Qd7 Qa6
was worth trying (2 1 .QdS+ BfS). 1 7 .Qd3 h5?! Benko explains that Black didn't
like 1 7 . . . 0-0 l S . Rfd 1 , but this should still be tried, e.g., l S . . . Bf6 1 9.Qd7 Qe5
2 0 . Rc7 a6. 1 8.Qb5+ Kf8 1 9.Rfd 1 h4 20.Qd5! Qxd5 2 1 .Rxd5, and the ending
favored White considerably-see Game 1 2 2 .
We shall see Benko happily exchange Queens in any number o f openings. It is
important to note that he is not running away from complications to get to an
endgame, but rather entering into a Queenless middlega m e . Queenless
middlegames require a special skill, including the ability to distinguish between
promising transitions into the endgame and ones that go nowhere. Benko was as
adept at this type of position as anyone in his day. Benko-Balogh, Hungarian

Ch. (Budapest) 1 947 is an example: l .Nf3 Nf6 2 .g3 b6 3 .Bg2 Bb7 4.c4 e6
5 .0-0 Be7 6.b3 0-0 7.Bb2 In his early days, Benko played quite a few of these
double fianchetto games. His virtuosity with g3/b 3 systems is illustrated in the
classic victory with the Reti Opening versus Horowitz from New York 1 965,
which you can see in Game S5. 7 . . . c5 8.d4 cxd4 9.Qxd4 Nc6 1O.Qd2 Typical
Benko: a simple move that to this day is neglected by theory. Another endgame
follows, and it may be equal, but White has all the (nagging) pressure. 1O . . . d5

1 1 .cxd5 Nxd5 1 2 .Rd 1 Bf6 1 3 .Na3! Bxb2 1 4.Qxb2

443

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

444

14 ... Qf6! ? As a general principle, one should avoid the exchange of Queens against
Benko! Better was 14 . . . Qe7. 1 5.Qxf6 Nxf6 16.Ne5 Nxe5 1 7.Bxb7 RabS I S .Ba6!

RfdS 19.Nb5 Rxd 1 + 20.Rxd l Nc6 2 1 .a3 ! with a virtually decisive advantage,
since Rc 1 (or Rd6) comes next. A deceptively simple performance.
A good example of Benko's proficiency in Symmetrical English positions is his
opening as White versus Donald Byrne in Vrsac 1 969: 1 .c4 Nf6 2 .NfJ g6

3 .b3 Bg7 4.Bb2 0-0 5 .g3 c5 6.Bg2 Nc6 7.0-0 RbS ? ! S.Nc3 a6 9.d4 d6 Benko
suggests 9 . . . cxd4. 1 0.dxc5 dxc5

I 1 .Na4! A surprising leap to the side of the board, which not only threatens
the c-pawn, but opens up the powerful Bishop on b2 . 1 1 .. .Nd7 B enko prefers
1 1 . . .b6, but then Black's awkward queenside can be exploited by 1 2 .Ne5 . 1 2 .Bxg7

Kxg7 1 3 .Qd2 e6? As B enko says in his notes, this weakens d6. S till, something
like 1 3 . . . Qc7 1 4. Rad l Rd8 1 5 . Qb2 + Kg8 1 6 .Ng5 ! , with moves like Bd5 and Ne4
in mind, will prove awkward for the second player. 1 4.Rfdl Qe7 1 5 .Ng5 , head
ing for d6 via e4. See how Benko converted this clear advantage in Game 92 .
The Symmetrical English positions with Bishops on g2 and g7, Knights on
c3 1f3 and c61f6, and pawns on d3 and d6, are notoriously dull. Yet here again,
B enko inj ected life into the positions that arose. In the notes to the following
game, Benko-Bilek, Hungarian Ch. 1 957, you will find a number of examples
of this, and the main game itself almost defines the term "positional mastery" !

1 .NB Nf6 2 .g3 g6 3 .Bg2 Bg7 4.0-0 0-0 5.d3 d6 6.c4 c5 7.Nc3 Nc6

BENKO AS W H ITE

445

8.a3 A Symmetrical English has arisen with Black having no theoretical problems.
But in practice, Benko could be deadly in such positions. A great example is his
imaginative opening versus Averbach in Game 2 6 . 8 ... Bd7 Here S . . . a5 9.Rb 1 NeS
transposes into Benko-Zuckerman, Atlantic Open 1 967. Of course, this is per
fectly equal; but within a few moves, an original Benko idea throws his opponent
completely off: 1 O.Bd2 Nc7 1 1 .Na4! RbS? ! ( 1 l . . .Ra6 - Benko) 1 2 .Nb6 Bf5 1 3 .Bc3

and White maintained a bind on the queenside.

Another sterling example: S . . . h6 9 . Rb 1 as 1 0.Ne 1 ! (intending Nc2 and b4)
1 O . . . Be6 1 1 .Nd5 Kh7 1 2 .Nc2 Ra7 1 3 .Be3 ! (to threaten b4) 1 3 . . . Ng4 1 4.Bd2 Nd4
1 5 .Nc3 ! ? Bd7 1 6.Nxd4! cxd4 1 7 .Nb5 Bxb 5 l S .cxb 5 , with an ideal position for
doubling or tripling on the c-file: l S . . . Nf6 1 9. Qc2 Nd7 2 0 . Rfe l Nc5 2 1 .b4 axb4
2 2 .axb4! ? (2 2 .Bxb4 b6 2 3 .a4 intending Ra 1 and as) 22 . . . Ne6 2 3 .Bd5 ! Qb6 24.Qc4
Raa S ? (2 4 . . . NdS was necessary, although White can build up at his leisure by
Rc2 or play 2 5 .Ra 1 ) 2 5 .Bxe6 fxe6 26. Qxe6 Qxb5 2 7 .Rc7 RaeS 2 S .Ra 1 Qh5 2 9 . Ra 5 !
b 5 , and now all White had t o d o was play something like 3 0.f3 , but h e threw all
his hard work away by 3 0. Rxe7 ? Rxe7 3 1 . Qxe7 Rxf2 ! 3 2 .Kxf2 Qxh 2 + 3 3 .Kfl
Qh 1 + with a perpetual, Benko-Rejfir, Moscow 1 9 5 6 . 9.Rb l Ne8 1 O .Bd2 as

l 1 .Qc 1 By comparison with the Zuckerman game above, Black has played the
useful . . . Bd7 in place of . . . Nc7 , so the Na4 idea is useless. Instead, B enko wants
to exchange Black's g7-Bishop. 1 l . . . Nc7 1 2 .Bh6 Rb8 1 3 .Bxg7 Kxg7 It's still
hard to believe that White has anything. The move b4 has been prevented, but
Black is ready for . . . b 5 . 14.NbS ! ? Just to create problems. Black can ignore the
Knight without disadvantage; still, White might be able to get something from

PAL BENKO: MY LIFE, GAMES AND COMPO S I TIONS

Qc3 +, e 3 , and d4, so Black tries to resolve things straightaway: 14 ...Nxb5 1 5 .cxb5

Nd4 ! ? I S . . . Na7 1 6 .a4 eS ! ? would be fine. 1 6.Nxd4 exd4 1 7.a4 Re8 At some
point hereabouts, the opening is formally over, but the game retains its funda446

mental characteristics almost to the end. 1 8.Qd2 b6? Now Black will be pas
sively placed after b4. 1 8 . . . Qb6 was better. 19.b4 e5 20.bxa5 bxa5 2 1 .Rfc1 Qb6

22 .Rxe8 ! Rxe8 2 3 .Rc 1 Calm and collected : White has a protected passed pawn,
and the pawn on as has to be tended. 2 3 " .Re5 2 3 . . . Rd8 24.BdS h6 2 S .e3 , or
even 2 S .Qc2 Rc8 2 6 . Qxc8 ! Bxc8 2 7 .Rxc8 intending things like Rc6-a6 and pushing
the b-pawn. 24.Rxe5 ! dxe5 2 5 .Qa2 ! It's impressive to see how Benko works
with such simple ideas. Now QdS-b7 is a real threat. 25 . . . Be6 26.Bd5 Kf6

Now it looks like Black has gotten his King close enough to defend, but . . . 27.Qe4!

g5 ! A good try. Otherwise, for example, 2 7 . . . Ke7 2 8 .Bxe6 ! fxe6 2 9 . Qc 1 ! Kf6

3 0. Qh6 would win straightaway. 28.Bxe6 fxe6 29.Qc 1 ! The Queen's moves have
a subtle geometry worthy of Benko's studies. Now it can fight against the King
almost by itself, since Black's Queen is tied to the b-pawn: 29 . . . h6 3 0.h4 Qd6

3 1 .hxg5 + hxg5 3 2 .Qb l Qb6 3 3 .Qf1 e4 34.Qh3 Qd8 3 5 .Qh6+ Ke5 36.Qg7+
Kd6 3 7.dxe4 g4 3 8.Qb7, and White won easily. This is a case of winning posi
tional play that extended from the opening all the way to the endgame .

Benko-Seirawan, Lone Pine 1 978, Game 1 3 1 in this book, sees Benko quickly
entering into and exploiting another Queenless middlegame : l .Nf3 Nf6 2 .e4

e5 3 .Nc3 d5 4.exd5 Nxd5 5 .e4 Benko won a couple of nice games with this
idea of Nimzowitsch. 5 . . . Nxc3 6.dxc3 ! ? Traditionally the most important move
here has been 6.bxc3 , when 6 . . . g6 7 . d4 is a Griinfeld Defense. 6 ... Qxd 1 + 7.Kxd l

BENKO AS W H ITE

A typical opening result for Benko: the Queens come off and White has the bet
ter practical chances with a slight theoretical edge. Among other contests, Euwe
Palmason, Munich 1 9 5 8 and Miles-Tal, Las Palmas 1 97 7 preceded this game, as
did two games by Andersson (versus Diesen and Timman); but Benko was quick
to notice the latter ventures and refine the ideas. 7 . . . Nc6 8.Be3 e6 The best
move was 8 . . . b6 ! , with White only slightly better. 9.Kc2 b6? ! Benko: "This looks
solid, but the newly weakened light-squares on a6 and c6 will haunt Black for a
long time. " Instead, 9 . . . Bd7 1 0 .Be2 Be7 1 1 .Rhd 1 0-0-0 1 2 .Rd2 f6 1 3 .Rad 1 was a
little better for White in Cvetkovic-Palatnik, USSR-Yugoslavia 1 976 (played before
this game) . 1 0.BbS Bd7 l 1 .a4! a6 1 2 .Be2 Now Black has surprising difficulties
due to his weaknesses.
A much sharper version of the same opening arose in Benko-Peters, Lone

Pine 1 978:
l .c4 Nf6 2 .Nc3 cS 3 .NB dS 4.cxdS NxdS S .e4 Nb4 At the time, this was a
major theoretical line, being explored again after a long absence from interna
tional play. 6.Bc4 Nd3 + 7.Ke2 Nf4+ 8.Kfl Be6?

Here Benko showed for the first time why this move is a mistake (8 . . . Ne6 is
correct) : 9.BbS + ! Bd7 9 ... Nc6? 1 0.d4, when the fact that both d 5 and Bxf4 are
threatened is fatal for Black. 1 O.d4 cxd4 In this and the next few notes, we see
Benko's line confirmed by later games. Here 1 O . . . Ne6 1 1 . d5 Nc7 1 2 .Bxd7+ Nxd7
1 3 .Bf4 is also great for White, Engelbrecht-Molineus, Germany 1 99 5 . I 1 .Nxd4

Ng6 ! 1 1 . . . e 5 1 2 .Bxf4 exf4 1 3 .Qh5 ! is hard to meet, e.g., 1 3 . . . Bxb 5 + 1 4. Qxb 5 +

Nd7 ( 1 4 . . . Qd7 1 5 . Q e 5 +) 1 5 .Rd 1 Q b 6 1 6.Nd5 Qxb 5 + 1 7 .Nxb5 with a winning
advantage, Chytilek-Langner, corr. 1 99 7 . 1 2 .Be3 e6 1 3 .Qb3 Be7 1 4.Rd l Qc8

I S .Rc 1 with a substantial advantage. White played g3 , Kg2 , and Rhd 1 , when his
control of more space and his better-placed pieces (note the awkward Knight on
g6, for example) made it hard for Black to unwind.
We now look at Benko-Reshevsky, u. s . Ch. (New York) 1 968, an opening
with a surprising turn of events. Benko doesn't know theory, unintentionally plays
what is now the main line, and then finds a clever sacrifice that is still crucial:

l .d4 Nf6 2.c4 cS 3 .NB cxd4 4.Nxd4 Nc6 S .Nc3 e6 6.g3 Qb6 7.Nb3

447

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

448

Benko's original note here (from an old Chess Life magazine) reads: " ? ! The usual
move is 7 .Nc2 . I must admit that I wasn't sure which Knight move was correct
and I chose the inferior one. The purpose of 7 .Nc2 is to avoid the pin that Black
now applies and to be able to defend the c-pawn with Ne3 if necessary. Even so,
7 .Nb3 is not so easy to refute. 7 . . . NeS looks dangerous but 8 . e4 Nfg4 9. Qe2
is safe . "
This i s a n interesting note for more than one reason. First, 7 .Nb3 h a s been
the main move in this incredibly popular position for almost 30 years, but wasn't
appreciated at the time of this game. Today, by contrast, Hansen's recent book
on the Symmetrical English treats 7 .Nb 3 exhaustively, devotes a bit more than a
page to 7 .NdbS , and doesn't mention 7 .Nc2 at all ! My own (originally 1 980)
book gives just one game with 7 .Nc2 dS ! . Older books claim that 7 .Nb3 was
introduced (or given its boost, more likely), in Taimanov-Geller, USSR Ch. 1 9S S .
Schwarz's English Opening book ( 1 96 3 ) quotes only that game while devoting
the rest of his space to 7 .Ndb S . And Taimanov himself switched back to 7 .Nc2
versus Korchnoi in 1 967 ! Thus there does seem to be a large gap of time before
7 .Nb3 caught on. Regardless of his later assessment, Benko's "mistaken" em
ployment of 7 .Nb3 here is one of the very earliest appearing in books and data
bases, and his choice shows typically good instincts. 7 . . . Bh4 7 . . . NeS 8 . e4 Bb4
9 . Qe2 is the modern main line. Benko is of course right that 7 .Nb3 is not easy
to refute ! 8.Bg2 d5 9.cxd5 9.0-0 dxc4 1 O. B e 3 has also been played. With his
move, Benko prepares to sacrifice a pawn. 9 . . .Nxd5 1 0.0-0! Nxc3 Even this
move goes unmentioned in Hansen and ECO; the older theory disappears! l 1 .hxc3

Be7 Way too risky is 1 1 . . .Bxc3 1 2 .Be3 . 1 2 .Be3 Qa6 1 2 . . . Qc7 has been the more
common move, although 1 3 .Nd4 Bd7 14.Rb l ( 1 4.NbS Qb8 I S .Rb l is also thought
to be good for White) 1 4 . . . a6 I S .Nxc6! Bxc6 1 6.Bxc6+ Qxc6 1 7 . Rb6, with the
better game was Gulko-Sokolov, Parnu 1 97 7 , "planning Qa4, Rib l , " according
to my own book.

BENKO AS W H ITE

449

It seems this game pretty much put 1 2 . . . Qa6 out of commission- the Queen
move seems to have been tried only once more and ended up losing in the same
manner! 1 3 .Bc5 0-0 1 3 . . . Bxc5 1 4.Nxc5 allows Qd6 next. But now 1 4 . Bxc6 Bxc5
gives White nothing. 1 4.Qd3 ! A very nice move. Just when Black seemed to be
getting out, Benko espies a very favorable ending. 14 ... Qxd3 1 5.exd3 Rd8 1 5 . . . Re8
1 6. Rab l a5 1 7 .Bxe7 Rxe7 1 8 .d4! Rc7 1 9 .Nc5 Ra7 2 0 .Rb5 Kf8 2 1 . Rfb i com
pletely tied Black down in the later game Galic-Bogdanovic, Saraj evo 1 974.
Reshevsky's move isn't any better. 1 6.Bxc6 ! Bxc5 1 7.Nxc5 bxc6 1 8.4!

And White has a decisive advantage based upon his magnificent Knight versus
the miserable Bishop on c8. See Game 86 for the conclusion.

English Opening with l .c4 e S

The English Opening with l .c4 e 5 was nowhere near a s developed during
Benko's peak international playing years as it is now. As White, Benko helped to
solve certain important positions (his examples are still those cited), and at the
same time he failed to establish that other, experimental, moves gave White enough
positive chances. In many lines, we can follow the actual development of En
glish Opening theory through his games.
To this day, what we call the Closed English (1 .c4 e5 2 .Nc3 Nc6 3 .g3 g6

4.Bg2 Bg7) is popular and closely contested up through the grandmaster level.

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

From the opening of Benko-Emma, Skopj e 1 972, one can learn pretty much
everything that is known about an older but still important variation: l .c4 Nc6

2 .g3 e5 3 .Bg2 g6 4.Nc3 Bg7 5.e3 d6 6.Nge2 Be6 ! ? 7.d4 Remarkably, the
450

natural move 7 .NdS is no longer mentioned by either Hansen or ECO (the pri
mary sources on l .c4 eS), although at the time of my book I devoted a whole
column to it. Why this is remarkable is that against all conventional moves, White
almost certainly has the advantage, since Black has a difficult time dealing with
the Knight on d S . So (and one would think that a player of either color would
want to know this), the only real reason not to play 7 .NdS is 7 . . . Nce7 ! :

Benko is also involved i n the theory of this move, although without success:
S . d4 (S .Nxe7 Nxe7 9.Bxb7 RbS , and White must either return the pawn or sub
j ect himself to a dangerous counterattack, e.g. 1 0.Bg2 Bxc4 I 1 . Qa4+ BbS 1 2 .Qxa7
0-0 1 3 .Nc3 Bd3 1 4. B e4 Bxe4 I S .Nxe4 dS 1 6 .Nc3 d4, and Black is obviously
better) S . . . c6 9.Nxe7 Nxe7 1 0. d S Bg4 1 1 .f3 ? ! ( 1 l .h 3

Hort; after 1 1 . . . Bxe2

1 2 . Qxe2 cxdS 1 3 .cxdS 0-0 1 4.0-0 fS , Black threatens . . . e4, but I S . e4 fxe4 1 6.Bxe4
NfS intending . . . Qb6, . . . Nd4 and . . . RacS is at least equal.) 1 1 . . .Bd7 1 2 .Nc3 cxdS
1 3 .cxdS b S ! , with advantage for Black, B enko-Hort, Venice 1 969. Hort was also
successful with 7 . . . N ce 7 ! versus Panno in Palma de Mallorca 1 969 and in later
contests. It was actually played once before by Levenfish in 1 949, but it's un
likely the players knew about it! 7 . . .exd4 Benko cites 7 . . . Bxc4 S . d S (In my En
glish 1 . . . e S book, I also give a wild and unclear line beginning with S .Qa4! ? )
S . . . Bxe2 9.Qxe2 N b S 1 O.QbS+ N d 7 I 1 . Qxb7 R b S of Larsen-Suttles, but a s the
game notes point out, after 1 2 . Qxa7 ! White has a pawn as well as his other ad
vantages- this is too much for Black to expect from his position.
Other moves that are played here include 1 1 . . . QbS 1 2 . QxbS + RxbS 1 3 .e4, and
although Black has nagging play with a combination of . . . NcS and . . . fS , it's hard
to argue against the two Bishops or, for that matter, the holes on Black's queenside.
Finally, the best move is probably 1 1 . . .Ne7 which prepares . . . fS/. . . fxe4/ . . . NfS - d4
in some positions; again, White's position is preferable, but at least this might
make life interesting. 8.Nxd4

BENKO AS W H ITE

45 1

8 ...Nxd4 8 . . . Bd7 (given as equal in old analysis by Schwarz) 9.0-0 Nge7 1 0.Nde2 !
gave White a nice edge due to his control over d5 in Mikenas-Podgaets, USSR
Ch. 1 97 0 . 9.exd4 Ne7 ! ? B enko quotes Quinteros-Hort, Vincovci 1 97 0 , which
went 9 . . . Qd7 1 0 .0-0 Ne7 1 1 .Re 1 0-0 1 2 . d 5 Bf5 1 3 .h3 ! , with a clear advantage.

1 0.d5 ! As Benko points out, 1 O.Bxb7 O-O! 1 1 .Bxa8 Qxa8 1 2 . d 5 Bh3 would be
extremely difficult in practice. 1 0 ... Bd7 1 1 .0-0 0-0 1 2 .Ne4!

Here is the simple but ingenious positional idea that discouraged Black from us
ing this line in practical play; the Ne4 idea also applies to a number of similar
positions with this pawn structure. White makes it awkward for Black to better his
piece position, while .. .f5 will always be badly weakening. 12 . . . h6 On 1 2 . .. Nf5 ,
1 3 .Bg5 obviously favors White. 1 3 .Bd2 f5 Not 1 3 . . . Bxb2 1 4.Bxh6, and versus the
natural 1 3 . . . Re8, White might play 1 4.Bc3 Bxc3 1 5 .Nxc3 Nf5 1 6 .Ne4! - the
Knight returns to threaten g4 and Qd4 or Qb3 -c3 . 14.Nc3, and Black's kingside
and e6 are seriously weakened. See Game 1 09 for the continuation. Such ideas as
1 O.d5 and 1 2 .Ne4 look simple, but they bespeak the play of a positional master.
Playing with an early g3 (prior to Nc3 ) versus 1 . . . e 5 was a Benko specialty
which even led to the name "Benko System" being applied to the move l .g3
itself. I discuss the Benko System on page 464 and this is not a case of it. Never
theless, he played a number of interesting games with the idea. See, for example,
the opening of Benko-Tal, World Student Team Ch. (Reykjavik) 1 957: l .c4

Nf6 2 .g3 These days White plays 2 .g3 fairly routinely, but in 1 9 5 7 it was con
sidered harmless due to an early . . . c6, as in this game. 2 . . . e5 3 .Bg2 c6 4.Nf3 ! ?

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

452

Aiming for something between a 2 .c3 Sicilian Reversed and an Alekhine De

fense Reversed. Black has found solutions to this approach (which costs White
some time), so 4.d4 is most popular today. 4 ... e4 5 .Nd4 d5 6.exd5 Qxd5 ! 7.Ne2

Qh5 8.h3 To prevent . . . Bh3 , which has at least equalized in several games. 8 ... Be5
Both sides are very likely just winging it here. Over a decade later, Black estab
lished that 8 . . . Qg6 9.Nc3 Bd6 was a good setup, when 1 O.Ne3 0-0 I l . Qc2 Re8
protects Black's strongpoint in the center. Tal's 8 . . . Bc5 is not bad, but the Bishop
can either be hit by b4 or in some cases lose a tempo when White's Queen comes
to c2 . 9.Nc3 Qf5 ? !

9 . . . Qg6 still seems best. Tal probably wanted to threaten f2 i n order t o avoid
ideas like 1 O .b4 with ideas of a quick fianchetto or further harassment of Black's
Bishop. 1 0.Ne3 ! Bxe3 1 1 .dxe3 0-0 1 2 .Qe2 Re8 1 3 .f4! ? A bold but probably
incorrect decision. One feels that simply 1 3 .0-0 would have kept an advantage.
Perhaps Benko was banking upon Tal 's unwillingness to exchange Queens.

13 ... Na6 ! ? There is nothing at all wrong with this move, but now Benko gets
the positional game he wants. Could Tal have tried 1 3 . . . exf3 1 4. Qxf5 fxg2 ! ?
( l 4. . . Bxf5 1 5 . exf3 Bd3 or 1 5 . . .Na6 is also o f interest) 1 5 .Qf3 gxh l =Q+ 1 6 . Qxh l
Bf5 followed by . . . Na6-c5 ? This seems roughly equal, and i n any case not much
fun for White to play. 14.g4 Now White's structure makes a better impression.

14 . . .Nh4 1 5 .Qa4 Qe5 1 6.a3 Nhd5 1 7 .Nd l ! ? Nd7 Here 1 7 . . . b 5 ! was better.
1 8.Bxe4 Qe7 1 9.Bf3 Ne5 20.Qe2 White holds some advantage, and in a battle
between young players on their respective student teams, the more experienced
Benko slowly outplayed his soon-to-be-famous opponent. For the tragic last few
moves of this contest, however, see the comments before Game 44.

BENKO AS W H ITE

Benko played many games with similar setups. I will cite several in the middle
of the following game, Benko-Zuidema, Belgrade 1 964: l .g3 Few grandmas
ters used this flexible move at the time. The possibilities for transposition abound;
here the game becomes an English opening. 1 . ..e5 2 .c4 Benko entered this kind
of English opening often before it became popular. He lost a few key games
with it, e.g., to Tal and Korchnoi; but those were not due to the opening.

As so

often with Benko's openings, the system can hardly be considered advantageous,
but he got out of that era's book and immediately tested his opponent. 2 . . . c6
For positional gems against 2 . . . d6 and 3 . . . fS , see Benko-Formanek and Benko
Bisguier, Games 84 and S l . The latter game, featuring the modern 6.Nh3 and
the counterattack with 8 . cS ! , is another example of Benko's facility with English
Opening positions.
Finally, Benko outplayed the formidable Larsen with the same system as fol
lows: 2 . . . d6 3 .Nc3 fS 4.Bg2 Nf6 S . d 3 Be7 6.NB 0-0 7 .0-0 Kh8 8.M! as 9.bS
Nbd7 1 0.Ba3 Qe8 1 1 .Rc l Rb8 1 2 . e 3 ! Nb6

1 3 .cS ! dxcS 1 4.Nxe S , and White was much better; see Benko's notes in Game
7 0 . 3 .Nf3 3 . Bg2 Nf6 4.Nf3 e4 S .Nd4 would transpose into the above-analyzed
game Benko-Tal from 1 9S 7 . 3 . . . e4 4.Nd4 d5 5 .d3 ! ?

A move hardly ever seen. Amazingly, there are only a few games with i t in my
database, and the two Benko games here are still the only ones quoted by theory!

As usual, he was trying to pose problems at the earliest stage, and had some
success in doing so. Again, S . cxdS QxdS 6 .Nc2 (6 . e 3 might be met by 6 . . . Nf6
7 .Nc3 Q e S 8 . d 3 B c S ) 6 . . . Nf6 7 . Nc3 QhS transposes to Benko-Tal above .

45 3

PAL BENKO: M Y LIFE, GAMES AND COMPOSITIONS

S . . . Bb4+ ! ? There is more than one acceptable equalizer here, in my opinion,

but none of them lead to dull simplicity. For example, S . . .dxc4 hasn't been played,
to my knowledge, but looks a reasonable solution, intending 6 .dxc4 Nf6 7 .Nc3
454

(7 .Bg2 Bb4+ S .Bd2 ! ?) 7 . . . Bb4 S. Bg2 Na6 9.0-0 Nc5 . Notice that no one ven
tured S . . . cS against Benko, presumably for fear of overextending his pawns. A
possible continuation would be 6.NbS a6 7 .NSc3 dxc4 S .dxc4 Qxd 1 + 9.Kxd 1 ,
and White's control of d S and quick development give him a slight edge. A straight
forward solution is 5 . . . exd3 6 . Qxd3 dxc4 7 . Qxc4 Nf6 S .Bg2 Nbd 7 , intending
9.0-0 (9.Qc3 ! is probably more accurate) 9 . . . NeS , but there is plenty of play in
the position.
But the most obvious and logical move is S . . . Nf6. Then, five years after the
first B enko-Tal encounter, Benko invented a line against the Riga giant which to
this day is the prime theoretical example: S . . .Nf6 6.Bg2 BcS (Here 6 . . . exd3 7 . Qxd3
Na6 was another solution in Sehner-Miethke, corr 1 994) 7 .Nb3 Bb4+ S.Bd2 Bxd2 +
9. Qxd2 dxc4 (or 9 . . . exd 3 ! ? intending 1 O. Qxd3 dxc4 1 l . Qxd S + - 1 1 . Qxc4 0-0 is
equal - 1 1 . . .KxdS 1 2 .NaS Kc7

=;

but White might try 1 0. cxdS) 1 O. dxc4 Qe7

1 1 .Nc3 0-0 1 2 .0-0 (That interest still exists in this line is illustrated by 1 2 . QgS
ReS 1 3 .Rd 1 Nbd7 1 4. Rd4! ? of Hulak-Savchenko, Ohrid 2 00 1 !) 12 . . . e 3 ! (Else
White is just better due to his rapid development and Black's problems with his
e-pawn.) 1 3 . Qxe3 Qxe3 1 4.fxe3 Ng4 l S .Nd 1 ReS 1 6.NcS ! Nd7 1 7 .Ne4 Ndf6
l S .Nd6 Re6 1 9 .cS

and White's development gives him some advantage after the e-pawn falls, Benko
Tal, Curaao 1 962 . 6.Bd2 Qb6 7.Bxb4 Qxb4+ 8.Qd2 Qxd2 + 9.Nxd2 exd3

1 O.e3

BENKO AS W H ITE

In this position, White has a small pull in the kind of Queenless middlegame
that Benko excels in. 1 0. e4 ! ? would have been an active move, but the text is
more subtle: Black's queen Bishop is slightly worse than White's, and White can
wait to play cxd 5 , since . . . dxc4 Iets a Knight in to attack the weakened d6 square.
This isn't much, but Black has to be careful. 10 . . . Nf6 I 1 .Bxd3 0-0 1 2 .cxdS

NxdS 1 2 . . . cxd5 B .Re l ! prevents . . . Nc6 and gives White easy play against the
IQP and down the c-file. ECO gives the position after 1 2 . . . Nxd5 as equal, and
that's close to being true, but see how Benko cashes in on his slight lead in de
velopment and his extra center pawn (a la the Sicilian) : 1 3 .Re l I think that
B .O-O-O ! ? is obj ectively more accurate, as Black can now equalize. 1 3 . . . Nd7

1 4.Be4 ! ? An exotic maneuver designed to lure the d7-Knight to f6 (instead of

e5), and then expand by e4. It successfully confuses his opponent. 1 4. e4 is premature due to 14 . . . Nb4 1 5 .Bb l Ne5 . 14 ... N7f6 1 S .Bb l Bd7 ? ! Instead, 1 5 . . . Bh3 !
1 6.Nc4 would be double-edged. 1 6.e4 Ne7 1 7.0 ! as I S.Kf2 Now White's position is very harmonious, for example, Rhd l and Nc4-d6 is in the air. In a few
moves, Benko gains a clear advantage: I S . . . b6 1 9.a4 cS 20.NbS Nc6 2 1 .Nc4
RabS 22 .Rbd l Be6 2 3 .Rd6, and White, with his superior central and queenside
position, went on to win.
Another difficult opening system, this time aimed against the King's Indian
Defense formation, involves the so-called Botvinnik System. Benko was instru
mental in its development, as we will see in the notes to Benko-Panno, Mos

cow 1 956: l .c4 Nf6 2 .Nc3 g6 3 .e4 eS 4.g3 Bg7 5 .Bg2 In a similar c4/e4 setup
against Steiner in Budapest 1 948, Benko showed the advantages of playing Nge2
and 0-0 before d3 . One positive is that White can sometimes play d4 in one step
(as we will see), or in other cases, the move f4 can threaten f5 before Black has a
chance to meet it by, say, . . . Nf6-e8 and . . . f5 . Despite this, the major sources now
tend to have every c4/e4 variation transpose from 5 . d 3 . Benko-Steiner went l .c4
e5 2 .Nc3 g6 3 .g3 Bg7 4.Bg2 d6 5 . e4 Ne7 6 .Nge2 Be6 7 . d4! (This is based upon
7 . . . Bxc4 8 . Qa4+) 7 . . . Qc8 8 . 0-0 0-0 9 . d 5 Bd7 (9 . . . Bh3 1 O .Be3 Bxg2 I l .Kxg2 f5
1 2 .f3 and besides greater space, White has much the better Bishop) 1 O.Be3 b6
1 1 .b4! .

The opening is over and White is obviously better, threatening c 5 .

455

PAL BENKO: MY LIFE, GAM ES AND COMPOSITIONS

Another example was Benko-Popov, Reggio Emilia 1 97 0 : 1 .c4 eS 2 .Nc3 Nc6

3 .g3 g6 4.Bg2 Bg7 S .e4 (With this order, Black can play an immediate S . . . Nd4,
but White meets this with just 6.Nce2 ! , since Nxd4/ . . . exd4 gives White a known
45 6

positional advantage.) S . . . d6 6.Nge2 Be6 7 .NdS Nge7 (7 . . . BxdS S.cxdS Nce7 9.d4!)
S . d 3 0-0 9.0-0 fS 1 O. Be 3 Qd7 1 1 . Q d2 RaeS ! ? (This tends to waste time. Now
White does well to play 1 2 . Rc l and b4-bS .) 1 2 .Rae 1 ! ? Rf7 1 3 .b 3 (or 1 3 .b4)
1 3 . . . RefS 1 4.f3 !

A very modern approach which has been found to secure a slight advantage.
Now Black has a hard time finding a plan: 14 . . . KhS ( 1 4 . . . Nd4 l S .Bxd4! exd4
1 6.Nef4! Bh6 1 7 .Qb2) l S .Bh6 Bxh6 1 6. Qxh6 NgS 1 7 . Q d2 fxe4 l S .dxe4 Bh3
1 9.Qc3 ! , and White, eyeing the cS break, was better. Back to Benko-Panno: 5 .. d6
.

6.Nge2

Botvinnik brought the e4/c4 structure to general attention by proving that White's
hole on d4 is counter-balanced by other advantages, for example, Black's diffi
culty in playing . . . dS and the risks of .. . fS , which can be answered by exfS and
either Qd2 and Bh6 or f4, in both cases with positional pressure. Better still,
White has attacking plans on both sides of the board: f4-fS and b4-bS . 6 . . . 0-0

7.0-0 Be6 8 .d3 c6 A double-edged idea: Black would like to enforce . . . d S , but
he foregoes the natural developing move . . . Nc6. Furthermore, White can now
consider defending his c-pawn and playing for d4. 9.h3 ! ? A remarkable move,
allowing Black to gain a tempo in support of his . . . dS plan. White would like to
play f4 without being harassed by . . . Ng4 and . . . Qb6+. The move h 3 is very com
mon in such positions if . . . Nc6 is in, but it takes some good calculation and posi-

BENKO AS W H ITE

tional sense to make it work once . . . c6 and . . . Be6 have been played. Another
attractive option was 9.b3 (threatening d4), since 9 . . . d 5 ? fails after 1 O. exd5 cxd5
I I .Bg5 ! (when 1 1 . . . dxc4 1 2 . Bxb7 Nbd7 1 3 .dxc4! wins a pawn with no counterplay
for Black). 9 . . . Qd7 1 O.Kh2 d5 ? ! Surprisingly (and as Benko had anticipated),
this natural move is a mistake. However, White also looks better after 1O . . . Na6
1 l .f4! , for example, 1 1 . . .Rad8? ! 1 2 .f5 ! gxf5 1 3 .Bg5 ! with a clear advantage. I l .exd5

cxd5 1 2 .d4!

The most aggressive move, since 1 2 .Qb3 e4! is unclear. 12 ... e4 12 . . . dxc4? 1 3 .dxe5
Ne8 1 4. f4 is depressing for Black; and he is also worse after 1 2 . . . exd4 1 3 .Nxd4
dxc4 1 4.Nxe6 fxe6 1 5 .Qe2 ! ? ( 1 5 .Bf4 is also interesting) 1 5 . . . Rc8 ( 1 5 . . . Qc8 1 6. Re l
Re8 1 7 .Nb5 !) 1 6. Rd l (or 1 6.Re l ) 1 6 . . .Q f7 1 7 .Rd6 ! . 1 3 .cxd5 Bxd5 14.Nxd5 Qxd5

1 5 .Nc3 Qf5

Or 1 5 . . . Qc6 1 6.f3 ! ? exf3 1 7 .Bxf3 , for example, 1 7 . . . Qd7 ( 1 7 . . . Qb6 1 8 .Na4) 1 8 .Bg5
h6 1 9.Bxf6 Bxf6 2 0 .Bg4! Qxd4 2 1 . Qf3 , and White has too many threats. 1 6.Qb3
The opening is over and Black has been positionally outplayed. Here 1 6.f3 ! was
also strong. 16 ...Nc6 1 7.Be3 Rad8 1 8 .Rad l Rd7 1 9.Kgl ! Threatening g4 with
out having to worry about . . . Nxg4. 1 9 . . . h5 20.d5 ! Ne7, and at this point simply
2 1 .f3 ! would have won outright: 2 1 . . . exf3 2 2 .Rxf3 Qe5 2 3 .Bd4, etc. Unfortu
nately, Benko only drew, and this excellent game was consigned to obscurity.
The next two games involve . . . Bc5 systems for Black, with Benko adding a
little twist to the opening move order. In Benko-Consulich, Venice 1 974, there
results merely a transposition: l .c4 e5 2 .g3 Nf6 3 .Bg2 Nc6 4.NB

457

PAL BENKO: M Y LIFE, GA M ES AND CO M POSITI ONS

458

Surprisingly, this interesting "trick" (to avoid . . . Bb4 lines by delaying Nc3 ) doesn't
appear in the major sources like Hansen, ECO, and even my own 1 979 work. In
theory's defense, the alternative 4.Nc3 Bb4 (what Benko says he was avoiding)
5 .Nd 5 ! has been considered better for White -perhaps Benko didn't trust the
lines after 5 . . . Bc5 , although most sources then give 6 . e 3 as yielding White a small
advantage . Interestingly, 4.NB has gradually caught on during the 1 980s and
1 990s. Gelfand, for example, has used it several times. 4 . . . Bc5 I can't find many
games predating Benko's use of 4.NB (although Euwe played it in 1 92 6) . One of
them went 4 . . . d5 5 .cxd 5 Nxd5 (an idea used fairly often in recent times), when
6 . 0-0 Be7 7 . d4! ? (7 .Nc3 transposes into a normal Dragon Reversed) 7 . . . e4 8 .Ne5
f5 9.Nxc6 bxc6 1 O.Nc3 0-0 1 l . f3 ! ? exB 1 2 . BxB with perhaps a slight edge, Zita
Hromadka, Prague 1 943 . There is of course newer theory on this line. 5 .d3 0-0

6.0-0 d6 7.Nc3 Bd7 ? ! 8.e3 , and we have one of the admittedly rather slow
versions of l .c4 e5 2 .Nc3 Nf6 3 .NB Nc6 4.g3 B c 5 , with White effectively a
tempo up due to . . . B d 7 . His best plan is probably b 3 , Bb2 , and an eventual break
by d4, perhaps preceded by a3/ . . . a 5 . See Game 1 2 3 for the game continuation.
The next game, Benko-Rossolimo, Novi Sad, 1 972, has also disappeared
from theory. In my 1 979 book, I called Benko's performance "ingenious and worthy
of study. " I still think so: l .g3 e5 2.c4 Nf6 3 .Bg2 Nc6 4.NB Bc5 5 .Nc3 d6
6.d3 a5 7.0-0 h6 8.e3 0-0 9.b3 Bg4 1 O.h3 Bf5 I 1 .Bb2 Re8 1 2 .a3

A Karpovian, prophylactic opening by White, which is almost like a Hedgehog

in the latent power it confers upon White's position. Notice how slowly Benko
builds up his position, priming for an eventual d4. He has prevented every Black

BEN KO AS W H ITE

move that would improve his position, so Rossolimo's apparently well-posted

pieces are doing little. 12 . . .Nh7 If 12 . . . Qd7 1 3 .Kh2 RabS 1 4. Rc 1 , and White is
ready for Nd5 . 1 3 .NdS NgS 1 4.NxgS hxgS I S .Kh2 Qd7 1 6.Qd2 Ba7
459

1 7.g4! Now, after so much preparation, Benko grabs space and then cracks open
the position for his Bishops: 17 ... Be6 1 8 .b4 Bb8 l S . . . axb4 1 9.axb4 exposes Black
to doubling on the a-file. 1 9.bS Nd8 20.Khl ! ? 2 0. a4 and 2 1 . Rc 1 was perhaps
more accurate, but White heads for the attack: 20 . . . BxdS 2 1 .cxdS ! ? Introduc
ing a bold attack based upon activating the two Bishops. 2 1 . . .QxbS 2 2 .f4 gxf4

2 3 .exf4 f6 24.Rab l Qa4 2 S .gS ! Ba7? Now Benko's usual time trouble was al
ready present, or he would have found 2 6. gxf6 ! intending 26 . . . gxf6 2 7 .fxe5 , and
White is just winning with ideas like Qh6, Be4, Rg 1 etc. A very aesthetic game
that probably escaped its deserved audience because of poor play at the end .
Benko didn't play a great number of games on the white side of the Reversed
Dragon, but it brought out his understanding of Sicilian structures . Benko
Diickstein, Sarajevo 1 967 illustrates an advanced opening strategy, well before
theory became very developed in this line. See also the game in the note to White's
Sth : l .g3 eS 2 .c4 Nf6 3 .Nc3 dS 4.cxdS NxdS S.Bg2 Nb6 6.NfJ Nc6 7.0-0 Be7

8.a3 A very pleasing and smooth Benko win in the same opening went S.d3 0-0
9.Be3 f5 1 O . Q c 1 Bf6 1 1 .Bg5 ! ? Bxg5 ? ! (later simply 1 1 . . .h6 and 1 1 . . .Be6 were
played) 1 2 .Nxg5 Nd4 1 3 .NB NxB + 1 4.BxB c6 1 5 .b4 Nd5 ? ! ( 1 5 . . .Be6 1 6.b5 cxb5)
1 6. b 5 Ne7 1 7 . Rb 1 ! Rf6 l S .Qa3 ! with typical Benko queenside pressure and a
significant advantage, Benko-Palme, Bad Gastein 1 945. 8 . . . 0-0 9.b4 Be6 1 0.d3

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

B enko points out the mistake 1 0.bS Nd4 1 1 .NxeS ? ? Bb3 1 2 .Qe l Nc2 . 10 . . . f5

I 1 .Bb2 The same goes for 1 1 .bS Nd4 1 2 .NxeS Bf6 1 3 .f4 Nb3 . 1 1 . . .Bf6 1 2 .Nd2 !
460

See Benko's notes on this move, now the normal one, in Game 66.

As so often,

B enko's game is the very first over-the-board game with 1 2 .Nd2 among the mul
titude in my databases and even in my older books. There is a short draw in a
Swedish championship game and a Finnish correspondence game from 1 96 2 ,
the latter very poorly played. To show how long it took before 1 2 .Nd2 entered
the mainstream, I suggested it myself in my 1 979 1 . . . eS English book, but couldn't
find any examples (not knowing this game) . When I played it myself versus
Sosonko at Lone Pine 1 980, thirteen years after B enko, I thought that it was an
innovation ! 1 2 . . . Qe8 Later Hubner found 1 2 . . . Bf7 ! ? 1 3 .Nb 3 e4! with great com
plications, but even there White may be able to keep a slight edge. 1 3 .Nb3 Rd8

14.Ne5 Be8 1 5 .Qb3 + ! ? The game notes point out the even better move I S .NbS ! ,
but White can be very happy with the check a s well. 1 5 . . . Kh 8 1 6.Nb 5 ! with a
clear advantage. A perfect exploitation of White's space and flexibility on the
queenside.

English Openi ng: Black Does Without I

. . .

cS or I

. . .

eS

Finally, we look at a few English and c4-related openings in which Black does
without 1 . . .cS or 1 . . . e S . In Benko-Pribyl, Majdanpek 1 976, there is another
early Queen exchange: l .e4 Nf6 2 .NB g6 3 .Nc3 d5 4.exd5 Nxd5 5 .e4 Nxc3

6.dxc3 Qxd 1 + 7.Kxd l

BENKO AS W H ITE

A handy anti-Griinfeld weapon (notice its similarity to Benko-Seirawan above)

that has typical Benko features: good central control, a centralized King (heading
for c2 or e2), and a little space. Khalifman recommends this Queenless middlegame
in his 2 0 0 1 book Openingfor White According to Kramnik, 1.Nf3. The line had just
become fashionable at the time of this game and Benko doubtless was aware of
the trend. His 9 th move, however, is an original idea. 7 ... Bg7 Certainly natural;
but today other moves are preferred, e.g., 7 .. .f6 (to play . . . eS and keep the bishop
on the f8-a3 diagonal, 7 . . . Bg4, and 7 . . . Nd7 . 8.Be3 Nd7 9.Nd4 ! ? A novelty. The
conventional plan is 9.Nd2 intending f3 and Nc4 with pressure. But Benko's move
is also valid and makes it hard for Black to develop without making some conces
sion. 9 . . . a6 Else NbS is a real problem. 1 0.f3 0-0 I 1 .Kc2

1 1 .. .Rd8 1 1 . . .eS 1 2 .Nb3 b6 allows 1 3 .a4! Bb7 ( 1 3 . . . aS ? ! 1 4.Bb S !) 14.aS, and Black
isn't coordinating very well. 1 2 .Be2 NfS 1 3 .Rhd l Bd7 Passive, but 1 3 . . . Ne6
1 4. Bc4 Nxd4+ I S . cxd4 intending Rac 1 and Kb l gives White a straightforward
central advantage. 14.a4 e5 1 5 .Nb3 Benko's opening innovation has resulted in
a white advantage. The game went 1 5 . . . Ne6? 1 6.Na5 !

1 6 . . . Bc8 1 7.Bc4 Rxdl 1 8 .Rxd l Kf8 1 9.Bxe6 fxe6 20.Bg5 Kf7 2 1 .Bh4 ! , with
a winning position, since . . . Bf6 will be answered by Bg3 and at the least Black's
e S -pawn will fall.
White's 8th move in Benko-Krnic, Sombor 1 97 6 should be given more atten
tion: l .c4 Nf6 2 .Nc3 d5 3 .cxd5 Nxd5 4.g3 g6 5 .Bg2 Nb6 6.d3 Bg7 7.h4 ! ?
An unusual but not unheard-of move (Suba even used t o play i t o n move 6 ) that
tries to weaken Black's kingside. It is now out of favor. 7 . . . h6 8.Bd2 !

46 1

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

462

A move that might bring attention back to the moribund 7 .h4. Benko points out
in his notes that this move (unknown to theory, whereas there are many examples
of 7 .h4 and 8 . B e 3 ) makes it possible to answer . . . e S , . . . Nc6-d4 with Nf3 , Nxd4.
Another point is that in lines with 8 . B e 3 , . . . NdS will sometimes gain a tempo on
a Bishop and force an undesirable exchange on d S . S . . . Ne6 9.Qc1 Nd4 1 0 .NO
e6 1 1 .Ne4 NdS 1 2 .Rh 1 Qh6 1 3 .Nxd4 Bxd4 1 4.h4. In this position, I prefer
White with his queenside advantage. So 8 .Bd2 is at least noteworthy, and as the
game goes, Black falls into a positionally inferior game without making any ob
vious mistakes - see Game 1 2 7 .
Benko liked to get two Bishops in positions where there were n o violent ex
plosions in the center, so that he could build up at his leisure to exploit the long
term advantage of the bishop pair. In Benko-Brasket, Lone Pine 1 976, the
line l .e4 Nf6 2 .Nc3 e6 3 .Nfl Bh4 4.Qe2 gave him the opportunity to aim for
this. 4 . . . eS 4 . . . 0-0 S . a3 Bxc3 6 . Qxc3 d6 7 . e 3 eS 8 . d 3 Bg4 9.Be2 was Benko
Guillermo Garda New York 1 98 8 - see Game 1 3 7 for some fine positional play.

S . a3 BaS 6.e3 Ne6 7.Be2 0-0 S.O-O d6 9.d4 Bxc3 1 0.Qxc3 as 1 1 .h3 Qe7
1 2 .Bh2

Now White is fully developed and can start to expand. 12 ... h6 1 3 .dS! NdS 14.Nd2

Bh7 l S .Bfl eS Otherwise White starts to open lines, but now Black's Bishop
and his Knight on d8 are pretty awful. 1 6.h4! Nd7 1 7 .Nb3 axh4 l S. axh4 ReS

1 9.Ra7, and White was breaking through.

BENKO AS W H ITE

F LAN K O P E N I N G S W I T H T H E b-PAW N : l .b3 and l .b4

For several years Benko experimented with b-pawn ideas, including 1 .b 3 and
1 .b4. But his favorite was to combine c4 and b4 in a sort of English/Polish opening, as in the game Benko-Vukcevich, U.S. Open 1 969: l .NfJ Nf6 2 .c4 g6
2 . . . d6 3 .b4! ? is featured in Game 1 02 , Benko-Kavalek, U . S . Open 1 97 0 . There
Benko plays like a raving hypermodern, permanently refusing to move his center pawns past the third rank while slowly increasing pressure on Kavalek's center and achieving a substantial advantage from nowhere. 3 .h4

Benko fiddled around with this system in quite a number of games (sixteen that
I have found), with just average results. Although it shows his creative and some
times anti-theoretical bent, there is nothing exceptional about the setup he nor
mally preferred : e 3 , Be2 , 0-0, with Nc3 and/or d3 . Rather than an opening ad
vantage, he would aim for a game without immediate tactics that left both sides
dependent on their own resources. L .Bg7 4.Bh2 0-0 5.e3 d6 6.Be2 e5

7.0-0 Even the great Petrosian was a little flustered by b4 in Benko-Petrosian,

Palma de Mallorca 1 968, in which he tried to answer 7 . d 3 by 7 . . . aS 8 . a 3 Ne8 ? !
9.Nc3 Nc6 1 O.Qb3 ! ? (or 1 O.bS) 1 O . . . Be6 1 1 .0-0 g S 1 2 .bS with a spatial advan
tage. A quick draw ensued, although Benko was still somewhat better at the end.

7 . . ReS S.Nc3 Nbd7 9.d3 The standard setup. 9 ... e4? ! Impatient. 9 . . . c6 looks
better, e.g., 1 O.Qb3 as l 1 .a3 Nb6, preparing . . . d S . 1 O.Nxe4 Nxe4 I 1 .Bxg7 Nxfl
.

1 2 .Rxfl Kxg7 1 3 .Qd2 . Here White is slightly better due to his central pawn

463

PAL BE NKO: MY LIFE, GAM ES AND COMPOSITIONS

mass and some pressure against Black's weakened kingside. The game went

1 3 . . . NeS 14.Nd4 c6 I S .Rafl QgS 1 6.Rf4! Bh3 1 7 .Rlf2 Bg4? I S.Bxg4 Nxg4
1 9.Rxf7+ KgS 20.R2f4 Rxe3 2 1 .NB . Now White is winning, in view of 2 1 . . .RxB
464

2 2 .Rf8+ ! Kg7 2 3 .R4f7 + Rxf7 24.Rxf7+ Kh6 2 S .Rxh 7 + .

B E N KO SYSTE M : l .g3
As Benko points out, calling l .g3 "The Benko System" is just silly. To me, the
only setup we might call a B enko System would be one with NB , g3 and then c4
against . . . d S , in which White omits or delays d4. In addition, the game has to be
in some sense different from a simple Reti System. Benko tried a number of
versions of such setups, including the game Benko-Bisguier, U.S. Ch. (New

York) 1 966: l .g3 dS 2 .Bg2 Nf6 3 .NB b6 4.c4 e6 S.cxdS exdS 6.0-0 Bd6
7.d4 ! Only now, since it makes the combination of . . . Bd6 and . . . b6 look suspi
cious. 7 . . . 0-0 S.Nc3 c6 9.Ne S ! Bb7 1 0.BgS !

White has achieved a unique and favorable position, of which I have only found
two other examples in various databases. His well-placed pieces prevent . . . cS
and prepare e4. For Benko's notes on both the opening and the rest of this game,
see Game 62 .

Benko-Keres, Curaao 1 962 shows the "Benko System" at its best versus
the legendary Estonian: l .Nf3 dS 2 . g3 Bg4 3 .Bg2 Nd7 4.0-0 c6 S.d3 eS 6.h3

BhS 7.c4

BENKO AS W H ITE

Once again, we have the unusual combination of Nf3 , g3 , and c4 versus a . . . dS

setup. 7 . . . dxc4 8.dxc4 Ngf6 9.Be3 White prevents a possible ... BcS . Even so,
the chances are about even. 9 ... Qc7 The similar 9 ... Be7 1 0.Nc3 0-0 I l .a3 a s
1 2 .Nh4! Bg6 1 3 .Nxg6 hxg6 1 4.Qb3 Q c 7 I S .Na4, almost mimicking Benko's game,
was played 30 years later in Pigusov-Kuczynski, Cappelle la Grande 1 992 ! 10.Nc3

Bb4?! 1 1 .Qb3 a5 1 2 .Na4! Be7 1 3 .Nh4! 0-0 1 4.g4 Bg6 1 5 .Nxg6 hxg6 1 6.Rfd l
Rab8 1 7.c5, and White had space and a bind. See Game 42 .

KI N G 'S I N D IAN ATTAC K

Any discussion of Pal Benko's openings has to include the King's Indian Attack,
an opening of which he was arguably the leading practicioner for many years.
What we have here is not so much a case of specific innovations, but of Benko's
strategies and their execution. For those who wish to play the KIA, I recom
mend playing through the overview in this game, paying attention to the games
in the notes and referring to others throughout this book. Our main game is

Benko-Uhlmann, Stockholm Izt 1 962 : l .e4 e6 2 .d3 d5 3 .Nd2 Nf6 4.g3 c5

5.Bg2 Nc6 6.NgB Be7 7.0-0 0-0 8.Re 1 B enko's delayed-eS idea (used against
Uhlmann here) was prevented by Schweber in the same tournament, but to Black's
greater detriment: 8.c3 Qc7 9 . Qe2 b6 l O. Re l (now White is ready for Nfl and
Bf4) lO . . . dxe4 I l . dxe4 e S ? ! 1 2 .Nfl h6 1 3 .Ne3 (White has the standard advan
tage of a King's Indian formation: an outpost on dS with no corresponding out
post for the opponent. What's more, Black's committal . . . h6 means that a white
Knight on fS will have greater effect than usual) 1 3 . . . Rd8 1 4.NM! Be6 I S .NhfS
Kh7 1 6. g4 Bf8 1 7 .NdS ! , and White already had a decisive advantage. 8 . . . Qc7
8 . . . b6 is slow, e.g., 9.eS Nd7 l O.c3 Rb8 1 1 .Nfl b S (an admission that Black hasn't
succeeded in making 8 . . . b6 useful) 1 2 .114 c4 1 3 . d4 b4 1 4. Qc2 Re8 I S .NgS Nf8
1 6.Ne3 ! Ba6 1 7 .f4 bxc3 1 8 . bxc3 h6 1 9.Nf3 QaS 2 0 . fS , and White is much bet
ter, Benko-Bielicki, Mar del Plata 1 96 5 .

9.Qe2 As w e shall see, this move contains a special idea. The conventional treat
ment, which Benko employed more than once, is seen in Game 1 03 versus Csom:
9 . e S Nd7 1 0. Qe2 bS 1 1 .Nfl as 1 2 . Bf4 Ba6 1 3 .114 b4 1 4.Ne3 Nb6 I S .Ng4 Qa7

465

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

(As Benko points out, Black wants the Queen to be out of the line of fire in the
case of possible sacrifices on d5 or f6.) 1 6.h5 Rfc8 (For an example of an f4-f5
plan, Benko cites yet another Stockholm 1 962 Interzonal game, Benko-Bisguier)
466

1 7 .h6 g6 1 8 .c3 bxc3 1 9.bxc3 Nd7 2 0.Bg5 , dominating the dark squares, Benko
Csom, Palma de Mallorca 1 97 1 . Superb notes are to be found with the complete
game. 9 . . . b5 1 0.Nfl Ba6

1 1 .Bf4! Pal Benko has to be considered one of the most important, if not the
most important, developers and players of the King's Indian Attack. Including
games with the KIA versus a French setup (but not those against the Caro-Kann),
I find him involved in 3 7 King's Indian Attacks against international competi
tion. These games are in databases that are quite incomplete for his peak KIA
period ( 1 960s and early 1 97 0s) and include very few U. S . Swiss system events.
Of the thousands of King's Indian Attacks in Mega base 2 00 1 (expanded by vari0us other databases), the game before you is the only one with this exact move
order. More importantly, none of the games with 9. Qe2 (or other alternatives to
9.e5) even contain the creative idea of Nfl and Bf4 before the automatic e 5 .
Benko's point i s that both moves are played i n the e 5 systems (i. e . , Bf4 i s played
after 9 . e 5 or 9. Qe2 and 1 0.e5), so why not play them here with the gain of a
tempo (Bf4 attacks the Queen)? 9. Qe2 does allow ideas associated with . . . dxe4,
or perhaps . . . Nd4, although in practice these have not worked out very happily
for Black. So to this day, it may well be that 9 . Qe2 and 1 0 . Nfl is an underrated
option. 1 1 . . .Qb6 1 2 .e5 Nd7 1 3 .h4

BENKO AS W H ITE

Now we have a standard KIA position in which White's attack is one move fur
ther along than usual. White's idea is h5 -h6 and in many cases Ne3 -g4 or N l h2 g4. Compare Benko-Csom, Palma d e Mallorca 1 97 1 , a s alluded to earlier. I n the
notes to that game, Benko mentions Fischer's comment "Today I play like Benko ! "
In fact, this KIA attacking scheme is often called "Fischer's," despite his sparse
use of it and his play along the same lines that Benko (and others) had previously
employed many times. For the continuation of Benko-Uhlmann after 1 3 .h4, see
Game 39. This is an extremely instructive contest that should not be missed by
anyone who plays the KIA.
The game Benko-Pachman, Varna 1 962 is a sterling example of B enko's in
ventiveness : l .e4 c5 2 .NB e6 3 .d3 Nc6 4.g3 g6 5.d4!

" ! " for novelty! This is the earliest game in my databases with 5 . d4; even if it had
been tried before in some forgotten game, Benko brought 5 .d4 to international
notice. Soon after this game other high-level players were trying it out, includ
ing Tal himself. 5 . . . cxd4 This seems natural, but in my book Play the French I
give what I believe to be the solution to 5 . d4, namely 5 . . . d 5 ! (as I played against
Fedorowicz in New York 1 97 7) . Most of the details are in that book. 5 . . . cxd4
gives White what he wants, but is naturally playable. 6.Nxd4 d5 A move that
has been played several times since this game. It is not bad, but White will retain
some edge. For a complete analysis of 6 . . . Bg7, see the game notes. By the way,
6 . . . a6 has been played a lot, notably by Hiibner, Tal, and Kramnik! So at the
very least it deserves consideration. 7.Bb5 7 . Bg2 Bg7 8.Nxc6 bxc6 is reasonably
solid for Black. 7 ... Bd7 8.exd5 Nxd4! This seems to be best, given Black's weak
nesses (and poor record) after 8 . . . exd 5 . 9.Qxd4 Bxb5 10.Nc3 ! And not 1 O. Qxh8 ? ?
Qxd5 1 1 . Rg i Q f3 wins. 1 0 . . . Qf6 1 1 .Nxb5 Qxd4 1 2 .Nxd4 0-0-0 1 3 .Be3 At
tractive, but one wonders if something like 1 3 .Nb5 ! ? a6 ( 1 3 . . . Rxd 5 1 4.c4 Re5 +
1 5 .Kfl with Bf4 next is uncomfortable for Black) 1 4.Nc3 exd5 1 5 .0-0 might just
favor White, who has ideas like Bg5 and Ne2 -d4. 1 3 . . . Rxd5

467

PAL BENKO: M Y LIFE, GAM ES AND CO M POSITIONS

468

1 4.0-0-0 Interestingly, the game S . Kagan-Webb, Hastings 1 97 7 took almost

the same course: 1 4.c4 Rd7 1 5 .0-0-0 a6? ? ( 1 5 . . . Bc5 ! is unclear) 1 6.Nxe6! etc.

14 . . . a6? ? Benko suggests 14 . . . Nf6, which is still a bit better for White. 1 5 .Nxe6 !
fxe6 Or 1 5 . . . Rxd 1 + 1 6. Rxd 1 fxe6 1 7 .Bd4. 1 6.Rxd5 exd5 1 7.Bd4 and wins- see
Game 47 .

KI N G PAW N : l .e4
Keeping in mind that Benko played 1 .e4 in a relatively small percentage of his
games (about 1 8 % , whereas with Black he had to face it about half the time), his
interpretation of it was typically creative, involving many new ideas and novel
ties. Benko tended to play 1 .e4 more in the beginning and middle years of his
career, which would seem to indicate a gradual shift to more positional open
ings. Nevertheless, we will see that against many defenses, Benko finds a solid
but pointed system by which to annoy Black, avoiding the wide-open positions
that sometimes result from e-pawn openings.

P i rc Defense
We begin with a look at the Pirc D efense (one of Benko's frequent openings as
Black, by the way). Here is the opening setup that he used in what are two of his
best-known games, wins versus Fischer and Tal. After his successes in these games,
this opening was called the "Benko System" by various annotators, but is in fact
just a Pirc Defense by transposition. The variation is illustrated by Benko-Tal,

Cural;; a o (C andidates) 1 9 62 : l . g3 g6 2 . B g2 B g7 3 . d4 d6 4 . e 4 Nf6

5 .Ne2 0-0 6.0-0 6.Nbc3 is the most accurate way to enter this system, in view
of 6 . . . e5 7 .h3 (7 .Be3 Ng4; 7 .0-0 Nc6 ! ) 7 . . . Nc6 8 .Be3 ; compare the next note.

6 . . . Nbd7 White's order allows 6 . . . e 5 intending 7 .Nbc3 Nc6 ! . 7.Nbc3 Now we

have a position from the Pirc, i . e . , l . e4 d6 2 . d4 Nf6 3 .Nc3 g6 4 . g 3 B g 7
5 . Bg2 0-0 6.Nge2 Nbd7 7 .0-0. 7 . . . c6 These days Black tends t o play 7 . . . c 5 , be
cause the text is cramped. 8.a4

BENKO AS W H ITE

469

It's interesting that in the books, White has almost always played h3 by this point,
or at the very least within the next few moves. Benko has avoided making that
move (usually played to prepare Be3) because he intends to play b3 instead. 8 ... a5 ! ?
This typical King's Indian move stops White from advancing to a5 and secures
the c 5 -square for Black's Knight. Instead, 8 . . . e5 9.a 5 ! exd4 1 O.Nxd4 was Benko
Fischer, Curaao 1 962 , in which White had a nice advantage and later won (see
Game 4 1 ) . 9.b3 Even in this position White has almost never played b 3 and Ba3
without h3 first. Benko gains an important tempo thereby. 9 . . . Re8 1 0.Ba3

Now the position is unique in chess history, as far as I know! At any rate, White
is better, because . . . e5 will weaken d6, but otherwise Black has trouble getting
his pieces out. Benko excelled in such prophylactic maneuvers. 10 . . . Qc7 I I .Qd2

e5 1 2 .Rad l exd4 1 3 .Nxd4 Nc5 1 4.8

PA L BENKO: MY LIFE, GAMES AND COMPOSITIONS

Here White is clearly better: not only is d6 weak, but Black has none of the
typical King's Indian counterplay versus White's rock-solid center. Benko sys
tematically increased the pressure by Rfe 1 , f4, and then BD ! and Qg2 ! , a ma470

neuver that is another original feature of this game. To see the continuation, go
to Game 44.
The game Benko-Jansa, Siegen 1 970 illustrates another of Benko's approaches
to the Pirc: l .Nf3 g6 2 . e4 Bg7 3 . d4 d6 4.c3 Nf6 S .Bd3 0-0 6.0-0

This setup was a B enko specialty. He played the same position with Be2 (after
Nbd2) more often, but later decided that Bd3 was better. B ecause Benko devel
oped so much of the theory on the Bd3 lines, I will concentrate upon some of
his innovative games with it.
Of course, White can achieve the key position via Bd3 or Nbd2 on any of
moves 3 through 6 . B enko brought this system to the attention of modern play
ers at the Siegen Olympiad in 1 97 0 (including this game), and it took off from
there. Of course these moves are too natural not to have been played many times
before; but of the relatively few precedents, I have found games by only a hand
ful of top players, e.g., Marshall (in 1 92 4), Stahlberg ( 1 940) , Ragozin ( 1 944),
and Petrosian ( 1 946) . Interestingly, Fischer played it in blitz versus Ivkov in 1 970.
We will talk elsewhere about how many Benko openings were taken up by Fischer
in more serious contexts. 6 . . .Nc6 This and 6 . . . c5 are considered the main lines.
A strategically and tactically attractive game by Benko against 6 . . . c5 went 7 . dxc5
dxc5 8 . Qe2 Nc6 9.Nbd2 Qc7 l O. Re 1 b6 1 1 .Nc4! Bg4! ( 1 1 . . .b 5 1 2 .Ne3 c4 1 3 .Bc2
and a4 or Nd5 causes some problem) 1 2 .a4 a6 1 3 .Bc2 b 5 1 4.Ne3 BxD 1 5 . Qxf3
e6 (about equal) 1 6. Qe2 c4 1 7 .g3 Nd7 1 8 . f4 Nc5 1 9.Ng4!

BENKO AS W H ITE

(White is conjuring up something out of nothing!) 1 9 . . . f5 2 0 .Nfl RfdS (2 0 . . . Na5 ! ?)

2 1 .Be3 BfS 2 2 .g4! fxg4 2 3 .axb5 axb5 24.RxaS RxaS 2 5 .Nxg4 RdS 26.f5 ! exf5 2 7 .exf5
Nd3 2 S .Rfl , and White's attack is getting very serious indeed; he won shortly in
47 1

Benko-Saverymuttu, London 1 9 7 3 .
The alternative 6 . . . Nbd7 7 .Nbd2 c6 is solid but passive. Benko came up with a
creative prophylactic measure aimed to discourage . . . e 5 in Benko-Schmidt, Lin
coln 1 9 7 5 : S . b 3 ! Qc7 9.a4 e 5 (9 . . . a5 1 O.Ba3 ! ) 1 0.Ba3 ReS l 1 . dxe 5 ? ! (But there's
no need to let up the pressure yet; better is I 1 .Re l ! or I 1 .Qc2 .) 1 1 . . .Nxe5 1 2 .Nxe5
dxe 5 , with equality. 7.Nbd2 7 .b4! ? has also been played, to discourage . . . e 5 . 7 ... eS

8.dxeS NxeS S ... dxe5 9.Nc4 Nh5 ! is also okay. As so often, Benko is trying to
achieve a strategically interesting position from the opening, not a forced win.

9.NxeS dxeS 1 0.Ne4

A standard c3lNbd2lBd3 scenario. White isn't better, strictly speaking, but he

stands very solidly and has a plan involving b4 and a4 to create trouble on the
queenside. 1 0 . . . Nd7 Benko faced what is probably the best move versus Mednis
in El Paso 1 97 3 : 1 O .. .Nh5 ! l 1 .g3 Be6 1 2 . Qe2 Qd7 ( l 2 . . . QeS occurred in a later
game, when White played b4 and a4 to establish a queenside advantage) 1 3 .Rd l
a5 ! (preventing a 5 i n certain lines) 1 4.Ne3 Nf6 1 5 .Nc4 Nh5

=.

I l .b4 Benko:

" Gaining queenside space and taking the c5 -square away from Black's Knight. "

1 1 . ..Re8 1 2 .Be2 Qe7 1 3 .Be3 Nb6? ! Benko prefers 1 3 . . . NfS , although it's hard
not to like White after something like 1 4.a4 Be6 1 5 .Na5 ! . 1 4.NaS, with nag
ging positional pressure. For the continuation, see Game 1 0 1 .
Benko often played the Pirc as Black and therefore knew a variety of systems
against it. He tended to choose very solid variations and, as Benko-Ostojic,

Wijk aan Zee 1 969 shows, he was extremely familiar with the Classical Main
Line: l .Nf3 g6 2 .d4 Bg7 3 .e4 d6 4.Nc3 Nf6 S .Be2 0-0 6.0-0 Bg4 For 6 . . . c6
7 . a4 a5 S.Be3 Ng4 of Benko-R. Chavez, Sao Paulo 1 97 7 , see Game 1 2 9 . 7.Be3

Ne6 8.Qd2

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

472

S ... e5 In his notes to this game, Benko mentions his own invention, played against
Ivkov in Caracas 1 97 2 : 8 . . . Re8 ! ? with the idea 9 . Rad 1 eS 1 O .dS BxB 1 1 . BxB
Nd4 1 2 .Bxd4 exd4 1 3 . Qxd4 Nxe4. I don't have that game, but a later correspon
dence game Pesolano-Piccinali, 1 9 8 1 went 1 2 .Be2 ! Qd7 ! ? 1 3 .B Nxe2 + 1 4.Nxe2 ,
threatening c4-cS with an advantage for White (compare the King's Indian De
fense). Something like that probably accounts for the almost exclusive use of
8 ... eS today. 9.d5 After 9 . dxe S dxeS 1 0. Rad 1 , which as Benko says, "was all the
rage" at that time, he came up with the solution that to this day is important:
1 O . . . Qc8 !

I remember this move well from that time, because it spoiled part of my reper
toire! The stem game for this line was Hug-Benko, Sao Paulo 1 97 3 : 1 1 .h3 ( l 1 .BgS
BxB 1 2 .BxB Nd4 1 3 .Be2 c6 equalized handily in Weinstein-Benko, Chicago
1 974; 1 1 . Q c l is considered best by theory, but Black doesn't have to work very
hard for equality.) 1 1 . . . Rd8 1 2 .Q c l BxB 1 3 .BxB Nd4 1 4.Bxd4 exd4 1 S .NdS NxdS
1 6 . exdS , with a dead equal game due to the opposite-colored Bishops. 9 . . .Ne7

1O.a4 The idea of this move is to prevent an early . . . b S . In this case, Benko
didn't invent it himself - he was following Geller-Vasiukov, Kislovodsk 1 96 8 .
Although 1 O .a4 was the main move at the time and is played upon occasion to
this day, it actually doesn't appear in Nunn and McNab's The Ultimate Pire ! They
give only the current favorite 1 O. Rad l . l O . . . NeS Benko talks about his win over
S eirawan following 1 0 . . . Nd7 , a move which is also frowned upon now. The best
response to 1 0 .a4 is probably 1 0 . . . B d 7 , considering . . .Ng4 and preserving the
idea of . . . NhS . The latter plan was employed by Vasiukov in the game men-

BENKO AS W H ITE

tioned in the last note. 1 1 .a5 Grabbing space on the queenside. White has come
out of the opening with an edge. Nunn and McNab give only 1 1 .Ng5 , which
has ideas of f4 and also appears to lead to an advantage. 1 1 . .. a6 1 2 .Ra3 ! ? Bd7

1 3 .g3 Perhaps thinking about Nh4 and f4? 13 ... f5 ! ? 1 4.Ng5 h6 1 5 .Ne6 Bxe6
1 6.dxe6, and Black's weak light squares combined with White's activity gave Benko
excellent chances-see Game 8 7 .

French Defense
Benko consistently employed the King's Indian Attack against the French De
fense (via l .e4 e6 2 .d 3 ) ; please refer to that King's Indian Attack section above .
However, he could invent important new theory in variations that he seldom
played. The game Benko-German, Stockholm (Interzonal) 1 962 introduces
a formation that was subsequently used in hundreds of top-level encounters: 1 .e4

e6 2.d4 d5 3 .Nd2 Nf6 4.e5 Nfd7 5 .c3 c5 6.f4 Nc6 7.NdB Qh6 8.g3 ! ?

Benko:

"As far as I know, I introduced this setup in international tournaments in

the Dublin Zonal event in 1 9 5 7 against L. Schmid. Now 8 . . . cxd4 9.cxd4 Bb4+
can be met by 1 O .Kf2 f6 1 1 .Kg2 . "
The 1 9 5 7 game Benko refers t o went 7 . . . Q a 5 8 Kf2 f6 9 g3 . That i s indeed the
first g3 , Kf2 -g2 idea given in Megabase 2 00 1 versus Black's formation. The con
cept is remarkable in that White makes eleven moves with only one piece devel
oped! The game before you is one of the earliest examples of the currently popular
7 . . . Qb6 8 g3 plan, just a year after the now well-known contest Portisch-Tal,
Oberhaus en 1 96 1 . Benko-German was played in the seventh round at Stockholm,
and in round twenty, super-GM Leonid Stein, perhaps impressed, also played
the line successfully.
Finally, in Benko-Foguelman, Amsterdam 1 964, Benko used the same g3 idea
(also followed by Kf2 -g2 ) , but with . . . cxd4 and cxd4 included, again one of the
earliest games with that particular move order. It is fair to say that although
Benko was not the first to play the exact order before you, he was the leader in
establishing this now-routine maneuver in the Tarrasch French. 8 . . . f6

As Benko

says, the key idea is 8 . . . cxd4 9.cxd4 Bb4+ 1 O.Kf2 f6 1 1 .Kg2 . Today, the move
1 O . . . g5 ! ? is by far the most popular move with that particular order. 9.Bd3 Later,
9 . Bh3 became the main line. 9 ... cxd4 1 O.cxd4 Bh4+ ? ! This assists White in his

473

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

plan, I give a lot of analysis on 1 O" ,Be7 in the first edition of my Play the French
book, and it is still a fully satisfactory move. I 1 .Kfl f5 ? ! But " . f6-fS makes no
sense. Having committed to " . f6, Black should try 1 1 . , ,0-0 1 2 .Kg2 Rf7 ! ? intend474

ing " .NfS , " .Bd7 etc. But White still has obvious attacking chances. 1 2 .Ne2 Be7

1 3 .h3 NfS 1 4.g4 g6 1 5 .Nc3 Bd7 1 6.Be3 . Now White has a super-solid center
and much more space. See how Benko converted these advantages in Game 3 S .
Another early French Defense, Benko-Ogaba, Hungarian Ch. 1 950, involves
the irregular but still respectable Guimard Variation. Here's a brief look at some
opening ideas pertinent to this contest: l .e4 e6 2 .d4 d5 3 .Nd2 Nc6 First Black
hits d4, and after e S , he will aim for " . e S or, after moving the knight, " .c S . 4.NgfJ

Nf6 5.e5 Nd7 6.Bd3 A logical move that players of White might want to add to
their arsenal. Now Black switches to the " .cS plan: 6 . . . Nb4 7.Be2 c5 S.c3 Nc6
9.0-0 9.Bd3 would transpose into a well-known line beginning 3 .Nd2 Nf6 4.eS
Nfd7 S .Bd3 cS 6.c3 Nc6 7 .Ngf3 ! ? , with both sides having lost two tempi. That
line is considered dynamically balanced but is certainly interesting, perhaps even
more so than the text.

9 . . . Qb6 ? ! A key decision. The Queen doesn't really stand very well on b6 unless
it can pressure d4, which is not the case here. Black should save this valuable
tempo, keep the pawn on cS to meet Nb3 with " . c4, and leave b6 open for a
Knight in some cases. 9 " . cxd4 1 0.cxd4 f6 has been played, when 1 1 .exf6 can
theoretically be met by either recapture. But Black needn't give White an out
post on eS so quickly, and two other moves deserve strong consideration:
(a) 9" .Be7 1 O .Nb3 ( 1 O.Bd3 can now be met by 1 0" .Qb6! with extremely un
comfortable pressure on d4, since 1 1 .Nb3 fails to 1 1 . . .c4 - compare the game;
also l O".aS l 1 .a4 gS ! ? is very interesting, resembling other lines in the French.)
1 O".c4 1 1 .Nbd2 b S , with queenside prospects. White has some trouble untangling,
so logical is 1 2 .b 3 , when 1 2 " .RbS 1 3 .a4 b4 is promising, or 1 2 " . QaS 1 3 .Bb2 RbS;
(b) 9 " .aS 1 0. a4 cxd4 is more conservative : I l . cxd4 Be7 (or 1 1 . . .Nb4) 1 2 .Nb3
Nb4 1 3 .Bf4 0-0 1 4. Rc l b6 I S . Q d2 Ba6 1 6. Bxa6 Rxa6, and Black has gotten rid
of his bad Bishop in exchange for White's good one. White has space, but Black
can penetrate or simplify on the queenside. 1 0.Nb3 ! cxd4 1 2 . a4 Be7 1 3 .Bg5 ! .
White was better and went o n to win nicely - see Game 1 4 for notes.

BE NKO AS W H ITE

Double King Pawn

Benko tended to play irregularly with White against double king pawn openings ( l .e4 e5), with his usual variety of opening schemes. For example, looking
through one database I see him playing two Four Knights Games, two Ponzianis,
one Philidor Reversed, six Scotch Games, two Vienna Games (with 3 .g3), and
seven Spanish Games in lines with varying degrees of irregularity! Most of these
games are of little theoretical significance (as far as I can see), but Benko

Rossolimo, Atlanta (U.S. Open) 1 967 is an example of his general philosophy

with such openings: l .e4 e5 2 .NB Nc6 3 .Bb5 a6 4.Ba4 Nf6 5 .0-0 Be7 6.Bxc6

dxc6 7.d3 This is the Delayed Deferred Exchange Variation, a perfect weapon
for Benko, who can maneuver to his heart's content without being disturbed by
nasty pawn breaks by his opponent. 7 . . . Nd7 8.Nbd2 Bf6 9.Nc4 0-0 1 0.b3 Re8

I l .Bb2 c5

1 2 .h3 1 2 .a4 is now played a lot. As is usually the case in this survey, you are
looking at a game from the early days of this opening's revival. We have one of
those positions in which both sides tend to take a long time to execute a plan. A
"normal" idea for White consists of Nh2 and f4 at some point, perhaps with
Ng4, but this has to be well timed. 12 . . . b5 1 2 . . . a5 1 3 .Nh2 Nf8 1 4.Ng4 Ng6
1 5 . Qf3 Ra6 was about equal in Taimanov-Gligoric, US SR 1 96 5 ; 1 2 . . . g6 1 3 .a4
Bg7 is a popular plan, after which White again plays 1 4.Nh2 and contemplates
f4. 1 3 .Ne3 Nb6 Not much better was 1 3 . . . Nb8 1 4. a4 b4 1 5 .Nd5 Nc6 1 6 .Kh 1
Be6 1 7 .Nxf6+ Qxf6 1 8 .Nxe 5 ! Nxe5 1 9.f4 (revealing the point of 1 6.Kh 1 ) in Beijar
Lind, corr. 1 99 3 ; nor was 1 3 . . . Nf8 1 4.Nd5 Ng6 1 5 . Q c 1 Bb7 1 6.Nxf6+ Qxf6
1 7 .Nxe5 ! any good in Vogt-Romanishin, Altensteig 1 992 . A logical and reason
able alternative is 1 3 . . . g6. 1 4.a4!

475

PAL BENKO: MY LIFE, GAM ES AND COMPOSITIONS

476

1 4 . . . bxa4! ? Black weakens himself on the queenside, isolating his two c-pawns.
But 14 . . . Bb7 was bad after 1 5 .Ng4. Instead, the modest 14 ... Rb8 was only some
what better for White. 1 5 .bxa4 a5 Now 1 5 . . . Bd7 ! ? might be met by 1 6. c4, e.g.,
16 ... Nxa4 1 7 .Rxa4 Bxa4 1 8 . Qxa4 Qxd3 1 9.Nd5 ! . 1 6.Bc3 ! B enko makes it look
easy. The a-pawn is almost impossible to defend. 16 . . . Bd7 1 7.Qd2 c4! A good
try in a bad position. 1 7 . . .Nxa4 1 8 .Bxa5 Nb6 1 9 . Rfb l , and the pressure mounts.

1 8.Bxa5 cxd3 1 9.cxd3 Nxa4 20.Nd5 Nb6 2 1 .Bxb6 RxaI 22 .RxaI cxb6 2 3 .Ra6.
Benko has his usual positional pressure, which led to a win. Apart from taking
on b6, he has the idea of Qa2 and Ra8 .

Caro-Kan n
Benko dabbled in the Caro-Kann himself, and he tried out different systems
against it as White. Perhaps because the main line gave him some trouble as Black,
he used it himself on several occasions, as illustrated by Benko-Kagan, Winnipe g

1 967: l . e4 c6 2.d4 d5 3 .Nc3 dxe4 4.Nxe4 Bf5 5 .Ng3 Bg6 6.h4 In a couple of
other games, Benko did without this standard move, preferring to play Nf3 and
Bd3 straightaway. This goes along with his general tendency to avoid potential
weaknesses (e.g., a pawn on h5), although ultimately the h4-h5 maneuver prob
ably gives White the best chances in this line. 6 ... h6 7.h5 Bh7 8.Nf3 Nd7 9.Bd3

Bxd3 1 0.Qxd3 Qc7 I 1 .Bd2 e6 1 2 .0-0-0 Ngf6 1 3 .Ne4 0-0-0 1 4.g3

This move occurred at least as early as Geller-Foguelman, Santiago (Chile) 1 96 5 .

Nevertheless, it's interesting how long the players followed a t least one branch

BE NKO AS W H ITE

of modern theory, and how accurately the opening was played without known
examples as precedent. 14 ... Nxe4 1 S .Qxe4 Nf6 1 5 . . . Bd6 at once has done well,
when 1 6 .Kb 1 Nf6 doesn't let White play his order in the game before us (i. e . , in
which he skips Kb 1 ) . The alternative 1 6 .c4 c5 1 7 .Bc3 cxd4 1 8 .Bxd4 Nf6 1 9 . Qe2
Qa5 has consistently equalized for Black. 1 6.Qe2 Bd6 1 7.e4 eS 1 S.Bc3 exd4

1 9.Nxd4 a6 20.NB Rd7 2 0 . . . Qc5 has also equalized. Apart from ideas like . . . Nxh5
and . . . Ng4, the point is 2 1 .Ne5 Bxe 5 2 2 .Bxe5 Ng4! All of this only became clear
in the 1 990s, however. 2 1 .Kb 1 After 2 1 .Ne5 Bxe5 2 2 .Bxe 5 , 22 . . . Qa5 ! is preferred today, with numerous draws resulting. 2 1 . . .RhdS 22.a3 , and instead of
22 . . . Qc6 n .Rh4! (see the very impressive Game 69), 22 . . . Qc5 ! intending . . . Qf5
is fully equal; on n .Rh4, n . . . Be7 2 4.Rxd7 Nxd7 is simple enough. Some of this
theory is included just for the reader's benefit. But it also puts into perspective
how very relevant a game from 1 967 can still be.

S i c i l ian Defense
Benko is a great connoisseur of the Sicilian Defense as Black, so it is no sur
prise that he knew how to play the other side. As White, B enko tended to avoid
the most fashionable and critical Sicilian Defense theory, but sometimes ended
up creating it anyway! A good example of this is Game 8 3 , Benko-R. Byrne,
U.S. 1 965, a Rauzer Sicilian: l .NB eS 2 . e4 Transposing to l . e4 c5 2 .NB .

2 ... d6 3 .d4 cxd4 4.Nxd4 Nf6 S .Nc3 Ne6 6.BgS Bd7 7.Qd2 ReS S.O-O-O Nxd4
9.Qxd4 QaS 1 O.Bd2

B enko: "I decided to avoid it [the variation starting with 1 O .f4] and came up with
this new move. At least Byrne had to stop and think. "
His over-the-board discovery of 1 O.Bd2 is noteworthy. Black's variation with
6 . . . Bd7 and without . . . e6 was still experimental at the time. In fact, I have found
only a few other games played in the position with 9 . . . Qa5 before 1 96 8 , and
none with 1 0.Bd2 until 1 97 0 . One should always be aware that databases are
very limited in their coverage of events from earlier decades (Who knows, for
example, how much the Soviet players knew about such a move ? ). Nevertheless,
it is clear that Benko was either the originator of this line in international play
or at the very least that he shifted attention to it. Today 1 O.Bd2 is almost the
only move played, for example, there are 1 95 games with 1 0 .Bd2 in the database

477

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

I am using. 1 0 ... a6 I I .Bc4 Not often used, but with results and theory in White's
favor. It may even be the reason that 6 . . . Bd7 isn't seen that much anymore com
pared to . . . e6, especially at the top. 1 1 . . .QcS Benko mentions 1 1 . . . e 5 1 2 .Qd3
478

Qc5 1 3 .Bb3 Qxf2 , when 1 4.Bg5 Be7 1 5 .Bxf6 Qxf6 1 6 .Nd5 looks dangerous.

1 2 .QxcS RxcS 1 3 .Bb3 g6 14.Be3 Some other orders have been used over the
next few moves, but I like Benko's choice .

1 4 . . . Rc8 Benko queries this and suggests 1 4 . . . Rc6, when 1 5 .f3 Bg7 still leaves
White several ways to search for advantage, e.g., 1 6 .g4 ( 1 6.Nd5 Nxd5 1 7 .Bxd5
Rc8 1 8 .c3 ; 1 6 .a4! ? 0-0 1 7 .a5) 16 . . . e6 ( 1 6 . . . 0-0? 1 7 . e 5 ) 1 7 .a4 ( 1 7 .Na4 ! ? ) intend
ing 1 7 . . . Ke7 1 8 .a 5 , and White is better, with ideas like Na4-b6. I S .Bd4 Bg7

1 6 .NdS NbS The only option was the anti-positional 1 6 . . . e 5 . 1 7.Bxg7 Nxg7
1 8.eS ! . Now White has a nice advantage.
We know that Benko is primarily a positional player who often likes a bit of early
simplification, but occasionally he just lets loose and attacks. Benko-Simms, At

lanta (U.S. Open) 1 967 is an example of sustained initiative in a standard opening:

1 .e4 cS 2 .NB Nc6 3 .d4 cxd4 4.Nxd4 Nf6 S.Nc3 d6 6.BgS e6 7.Qd2 Be7
8.0-0-0 Nxd4 9.Qxd4 0-0 1 O.Bc4 QaS l 1 .f4 I 1 . Qd2 is a solid answer, avoiding
the tempo loss by . . . e5 and thinking about Kb 1 and Nd 5 . The text is more aggressive:

1 1 . ..a6? 1 1 . . .h6 1 2 .Bh4 ( 1 2 .Bxf6 Bxf6 1 3 . Qxd6 Rd8 or 1 3 . . . Bxc 3 ) 1 2 . . . e 5 ! has

done well. But here 1 2 .h4! ? is very interesting, and if 1 2 . . . e 5 , 1 3 . Qgl ! ? with a
promising attack. Finally, 1 1 . . . Bd 7 is the theoretical move, when 1 2 .e5 dxe5 1 3 .fxe5
Bc6 1 4.Bd2 ! Nd7 1 5 .Nd5 1eads to a small white advantage. Over the years, White
has done well from that position, but with accurate play Black will equalize. 1 2 .eS!

BENKO AS W H ITE

dxe5 1 3 .fxe5 Rd8 1 3 . . . Nd5 1 4.Bxe7 Nxe7 1 5 .Bd3 ! ? , and Black was hard pressed
to defend his kingside in Adler-Bannik, USSR 1 97 8 . 14.Qf4 Nd5 1 5 .Bxe7 Nxe7

1 6.Rxd8+ Qxd8 1 7.Rfl Qf8 1 8.Ne4 b5 1 9.Bd3 White has all his pieces aimed
at the kingside and Black's position is critical. The opening is over, but let's enjoy the game : 19 ... Nd5 20.Qg3 Ra7 2 1 .Ng5 ! ? Or 2 1 . Qh4 g6 2 2 .g4 ! ? Rd7 2 3 .g5
Bb7 24.RB threatening Rh3 . 2 1 . . .h6 2 2 .Ne4 Rc7 2 3 .Nd6 Now White threatens Bg6. His position is undoubtedly winning, but Black's next allows a deadly
blow: 23 . . . Kh8? 24.Qg6 ! f5 Or 24 . . . fxg6 2 5 . Rxf8+ Kh7 2 6 . Rxc8 . 2 5 .exf6 gxf6

26.Nxc8 Qxc8 27.Qxh6+, with a clearly won game.

Benko-Mednis, New York (U.S. Ch.) 1 9 6 1 is another Rauzer example, and
a clear example of inventing theory. While both sides are just doing their best in
a strange and tremendously complex position, two correspondence players 3 5
years later decide to follow almost the entire course of the game: l .e4 c5 2 .Nfl

Nc6 3 .d4 cxd4 4.Nxd4 Nf6 5 .Nc3 d6 6.Bg5 e6 7.Qd2 a6 8.0-0-0 Bd7 9.f4
Ideas with B are very popular right now, and 9.f3 is one way of getting to some
of the main lines. 9 . . . Be7 1 0.Nfl b5 I l .e5

B enko talks about avoiding the "mainstream theory" of I I .Bxf6; but even at the
time, I l . e 5 had become popular, e.g., in 1 95 7 and 1 9 5 8 Gligoric, Larsen and
other well known players were already making use of I I . e 5 . It's also true that
I I .Bxf6 seems to have been a major choice before that, but the top players took
Black, e.g., Polugayevsky, Geller and Korchnoi all defended the position after
1 1 .Bxf6 gxf6 in 1 954. Be that as it may, I I .e 5 is still critical and unclear. 1 1 . . . b4

1 2 .exf6 bxc3 1 3 .Qxc3 gxf6 1 4.Bh4 d5

479

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

1 5 .a3 ? ! Unusual, and probably not best. Benko: "This is a cautious move that
aims to control the b4-square . It also avoids heavily analyzed lines like 1 5 .Kb 1
Nb4." In fact, 1 5 .Kb l is still the main line today, with White holding a small but
480

definite edge after 1 5 . . . Nb4 1 6.Nd4, so the vast majority of games go 1 5 . . . Rc8
(when no one as White risks 1 6 .Bxa6 Rc7), leading to play that has been con
tested in hundreds of games, normally assessed as slightly better for White, al
though some disagree and think that Black is equal. So this is the point at which
Benko really deviates. 1 5 . a3 is probably not one of his better innovations (see
the next note) , but it's enough to create problems and a fascinating gam e .

15 . . . Na5 ! ? Benko suggests that "the immediate 1 5 . . . Rc8 might have been more
accurate, " and indeed, it's hard to see how White will reorganize in the face of
. . . Nb4 or . . . Na 5 . This position has been played just a few times, with results
favoring Black, e.g., 1 5 . . . Rc8 ! 1 6.Qd2 ( 1 6.Qe l Qc7 1 7 . Qd2 Na5 1 8 .Kb l Rb8
gave Black very active play in Santacruz-Sinulingga, Thessaloniki 1 984; 1 6. Q e 3
Qc7 1 7 .Bd3 Na5 1 8 .Nd4 Qb6 is also a little awkward, Ueter-Schoen, Germany
1 998) 1 6 . . . Na5 1 7 .Nd4 Qb6 1 8 . Q d 3 Nc4 1 9.Qb3 Qc7 ( 1 9 . . . Qxb 3 2 0.Nxb 3 Ne3
2 1 .Rd2 Nxfl 2 2 . Rxfl Rg8 is also slightly in Black's favor) 2 0 . Qg3 Kf8 2 1 . Bxc4
Qxc4 2 2 . Q d 3 Qxd 3 2 3 .Rxd3 Rg8 Pustina-Holm, Varna 1 962 , again uncomfort
able for White. While 1 5 .a3 is playable, the fact that Black won each of these
games is not encouraging. 1 6.Be l ! Driving back Black's Knight and giving White
time to activate his pieces. 1 6 . . . Nb7 1 7.Nd4 Another idea was to activate the
relatively useless h4-Bishop by 1 7 .Bf2 , e.g., 1 7 . . . Rc8 1 8 .Qd2 Ba4 1 9.Bd3 Nc5
2 0.Bxc5 and 2 1 .Kb l . 17 . . . Ne5 1 8 .Nb3 Ne4 1 9.Qf3 ! f5 20.g4!

White's position is still not inspiring, but at least he is finding counterplay; as

Benko says, he is undermining the enemy Knight. 20 ... fxg4 2 1 .Qxg4 Qb6 22 .Bd3

Re8 2 3 .Kb l Ba4 24.Bxe4 dxe4 2 5 .Bh4 Benko talks about the option 2 5 .Qg3 ,
which doesn't look very promising but Black has some weaknesses that compen
sate for his activity, e.g., 2 5 . . . Bxb3 (2 5 . . . e3 ! ? ; 2 5 . . . Bf6 2 6 .Bf2) 2 6 . cxb3 (2 6 . Qxb3
Qxb 3 2 7 .cxb3 Rg8) 2 6 . . . e3 2 7 .Bc3 ! with the idea 2 7 . . . Rxc3 ? 2 8 .Qg7 . 2 5 . . . Bxh4

26.Qxh4 Qe7 Benko says that Black should play 2 6 . . . Bxb3 2 7 .cxb 3 Qxb3 , when
White would have had compensation after 2 8 . Rc 1 (or 2 8 . Rhg l ) . What is amaz
ing here (and why this discussion is in some sense still about the "opening") is
that at this point, a game Husak-Llorens Riera, corr 1 996 (!) went 26 . . . Rxc2 ! ?
2 7 .Kxc2 Bxb 3 + 2 8 .Kb l Bxd l 2 9 . Rxd l Rg8 3 0. Qxh7 Rg l 3 1 .Qh8+ Ke7 3 2 . Qh4+

BENKO AS W H ITE

Ke8, draw! I wonder if they knew about the original game (which was hardly
flawless and a difficult game to follow), or if this is just coincidence? Anyway,
the rest of this contest can be found in Game 3 7 . A remarkable performance
considering the undeveloped state of theory at the time.
While on the subject of Sicilians, it's worth it to note how Benko, as always,
jumped around from system to system. For example, versus the Najdorf, he played
both 6.Bg5 and 6.Bc4 several times each, and looked thoroughly modern in his
treatment of 6.g3 . Benko-Fischer, BlediZagreblBelgrade (Candidates) 1 959
is a game of little value to Sicilian theory, but an illustration of Benko's seem
ingly effortless ability to gain the advantage through innocuous-appearing ma
neuvers: l .e4 c5 2 .NB d6 3 .d4 exd4 4.Nxd4 Nf6 5 .Nc3 a6 6.Bc4 Nbd7 7.a4

g6 S.O-O Bg7 9.Bg5 0-0 1 O.Qd2 Ne5 1 1 .8 Bd7 1 2 .a5 ReS 1 3 .b3 ! Ne6 1 4.Be3
Qc7 1 5 .Nde2 !

Here White is a little better: 1 5 ... Bc6 1 6.Ra2 1 6 .Bb6 was also good. 1 6 . . . Nd7

1 7.Nd5 QdS I S.Nec3 Nc7 1 9.Nb4 Ne5 2 0.Be2 QeS, and now White could
have shown his undisputed mastery of the position by 2 1 .f4! Nd7 2 2 .Nxc6 bxc6
2 3 .Rd l , with a clear advantage. The game was eventually drawn.
To close out the Sicilian, the reader might like some additional theory on the
very practical l .e4 c5 2 .NB d6 3 .d4 cxd4 4.Qxd4 ! ? system that B enko sets
forth so attractively in Benko - Harrison, Australian Open (Ballarat) 1 9S 5
(Game 1 3 5) .

As indicated, he became a leading expert i n this variation: l .e4 c 5

2 .NB d 6 3 .d4 cxd4 4.Qxd4 Nc6 The most frequent alternative is 4 . . . a 6 5 .Be3
Nf6 6.Nc3 Nc6 (Benko-Szekely, Sombor 1 97 6 went 6 ... e 5 7 . Qa4+ ! ? Nbd7 8.Bc4,
an extremely interesting concept for White and apparently strong. Theory gives
only 7 .Qd2) 7 . Qb6 ! ? Once again, theory concerns itself with 7 . Qd2 here. 7 . Qb6
is of course quintessential Benko, in that he seeks a Queenless middlegame :
7 . . . Qxb6 8 .Bxb6

48 1

PA L BENKO: MY LIFE, GAMES AND COMPOSITIONS

482

8 . . . g6 (8 . . . e6 9.0-0-0) 9.0-0-0 (9.NdS has enjoyed some success here) 9 . . . Bh6+

(9 . . . Bg7 is another story-then 1 0 .eS ! ? is interesting and very unclear) 1 O.Kb l
0-0 1 1 .Nd4. This is White's most ambitious idea, when Black has several moves:
(a) 1 1 . . .Nd7 1 2 .NdS Nxb6 1 3 .Nxb6 Rb8 ( 1 3 . . . Ra7 1 4.c3 , and for the time be
ing, the Rook on a 7 is an irritation to Black) 1 4.Nxc6 bxc6 l S .Nxc8 Rfxc8 1 6.Bxa6
Rc7 1 7 .c3 , Yandemirov-Kabanov, St Petersburg 2 000; and although the oppo
site colored Bishops are a factor, White is still a solid pawn up;
(b) l l . . .NeS is logical, thinking about both ... Ned7 and ... Neg4. Probably 1 0.Be2
is best, but two games have gone 1 2 .h3 Ned7 1 3 .BaS b6 1 4.Bb4 Bb7

=;

(c) 1 1 . . .Bd7 1 2 .f3 NeS 1 3 .Nb 3 Bg7 ? ( 1 3 . . .Be6 ! ? 1 4.NaS Rfc8; Benko gives
1 3 . . .Bc6 1 4.NaS) 1 4.f4 Nc6 l S .Be2 Bg4 1 6 .Bxg4 Nxg4 1 7 .NdS Nf6 1 8 .Rhe 1
"and White's superiority wasn't in doubt," Benko-Vukcevich, U . S . Telephone
Match 1 97 8 . S .BbS Bd7 6.Bxc6 Bxc6 7.Nc3 Nf6 S.BgS e6 9.0-0-0 In Benko
G h i t e s c u , R e g g i o E m i l i a 1 9 7 0 , Whi t e p l ayed m o r e s a fely: 9 . 0 - 0 B e 7
1 0.Rad 1 0-0 1 1 . Rfe l a 6 1 2 .Re 3 h 6 1 3 .Bh4 Rc8 1 4.h3 b S l S .a3 , which i s prob
ably about equal. 9 . . . Be7 1 0.Rhe l 0-0 1 1 .Kb l QaS 1 2 .Qd2

12 . . . KhS As pointed out in the notes to the game, several Queen retreats and
two Rook moves are also played. Some examples, including one from Benko's
practice:
(a) 12 . . . Rfd8 probably comes up a bit short after 1 3 .NdS Qxd2 1 4.Nxe7+ Kf8
l S .Nxd2 ( 1 S .Rxd2 Kxe7 1 6 .Nd4 h6 ! ? 1 7 . Bxf6+ gxf6 1 8 .f4 gave White a small
advantage in M. Popovic-Rajna, Budapest 1 989.) 1 5 . . . Kxe 7, and now 1 6 .f4 is the
most ambitious try, when 1 6 . . . h6 1 7 .Bh4 eS 1 8 .g3 ! has the ideas of both Nf3 and

BENKO AS WH ITE

Nc4, and leaves White better (weaker seems 1 8 .f5 g5 ! 1 9.fxg6 fxg6 2 0 . Bxf6+ ! ?
Kxf6 2 1 .Nc4 Ke6);
(b) 12 . . . Qb6 (avoiding Nd5) is very common here, for example, I 3 .Nd4 Rfd8
1 4.f3 Be8 1 5 .g4 Rac8 1 6 .Be3 Qa6 1 7 . Qe2 Qxe2 1 8 .Ndxe2 b6 1 9.Nd4 a6 2 0 . a 3 ,
liz _ liz , Benko-Andresen, Augsburg 1 99 1 ;

(c) 1 2 . . . Rfc8! is an improved version of 1 2 . . . Rfd8 in several respects, e.g., 1 3 .Nd5

Qxd2 1 4.Nxe7 + Kf8 1 5 . Rxd2 Kxe7 1 6 .Nd4 h6 1 7 . B h4 g5 1 8 .Nxc6+ Rxc6=,
Paveliev-Gorkavij , Novgorod 1 999. 1 3 .Nd4 Rac8 ! ? The moves 1 3 . . . Rfe 8 ,
1 3 . . . Rfd8, and 1 3 . . . Rfc8 have all been played here with varying success. I suspect
that one of the latter two is better than the text move . 1 4.f4 Theory was rather
undeveloped here, and this may even have been a novelty. At least one earlier
game had seen 1 4.f3 . 14 . . . h6? ! In some ways this just encourages White's attack,
but Black has also struggled here after 14 . . . Rfd8 1 5 .g4! , e.g., 1 5 . . . b 5 ! ? (Trying
for active play, in view of 1 5 . . . h6 1 6.h4! , but this doesn't seem to pan out. So
1 5 . . . a6 should be tried, with a limited White advantage after 1 6 .h4 b5 or 1 6 .f5 !
Bd7) 1 6 .Ncxb 5 ! Qb6 ( l 6 . . . Qxd2 1 7 .Rxd2 Be8 1 8 .Nc3 h6 1 9.Bxf6 Bxf6 2 0 .Red l
gave no real compensation in Ciric-Hartkopf, Baden-Baden 1 98 8) 1 7 .Nxc6 Rxc6
1 8.Nd4 Rb8 1 9.c3 Rc7 2 0.Bxf6 ! Bxf6 2 1 .g5 with White well on top, Innala-Lassila,
Finland 1 99 5 . 1 5 .h4!

1 5 ... Nxe4? Black had a similar disaster on his hands after 1 5 . . . hxg5 1 6.hxg5 Nxe4
1 7 . Qd3 Bxg5 ( 1 7 . . . Nxc 3 + 1 8 .bxc3 Kg8 1 9. Rh l f5 2 0 . g6 Bg5 2 1 .fxg5) 1 8 .Nxe4
Bxe4 1 9.Rxe4 Bh6 2 0.g4 f5 2 1 . Rxe6 Bxf4 2 2 .Nxf5 , 1 -0, Tal-R. Byrne, Biel 1 97 6 .
The best move is doubtless 1 5 . . . Rfe 8 , when more proof that Black's life is diffi
cult was shown by 1 6 .Bxf6 Bxf6 1 7 .g4 ! , and White had quite a strong attack in
Logunov-Kotsur, Novokuznetsk 1 999. 1 6.Rxe4! hxg5 1 7.Nxc6 Rxc6 1 8.hxg5
Qf5 Here the notes to Game 1 3 5 reveal how desperate Black's position is. I don't
think that there is a defense here, for example, 1 8 . . . g6 ? 1 9 . Qd4+ ! Kg8 2 0. Rh l e 5
2 1 . fx e 5 dxe 5 2 2 .Reh4 f5 2 3 . Q d 7 , 1 -0 . Yandemirov-Predein, Podolsk 1 99 3 .

483

PA L BENKO: MY LIFE, GAMES AND COMPOSITIONS

d-PAWN O P E N I N G S
484

By my calculation, Benko either played l . d4 or transposed into l . d4 systems in

about 48 % of his games. When one factors in the 24% English openings and
some remaining l .Nf3 games (e . g . , certain Reti Openings and King's Indian
Attacks) , he may be said to have played "positional" openings in over three
fourths of his games with White . Of course, each of these first three moves may
in fact lead to attacking and tactical opening play; but Benko tended to avoid
such lines, and naturally many of his setups with l . e4 should be described as
"positional" as well.

Bogo- I nd ian
In many of his 1 .d4 games, as we shall see, Benko was on the cutting edge of
theory, often being involved in one of the first games being played with a certain
variation, or even inventing entirely new lines. This is illustrated by our first
game, Benko-Flesch, Belgrade, 1 964, in which he revived an extremely rare
and forgotten idea. After this seminal game and the one mentioned in the note
to 7 . . . d6 (also by Benko), the idea disappeared from practice again, only to be
noticed and played a number of times over 20 years later: l .d4 Nf6 2 .c4 e6

3 .Nfl Benko also played 3 .g3 here, entering a Catalan after 3 . . . d 5 . 3 . . . Bh4+
The Bogo-Indian. For 3 . . . b6, see the next game. 4.Bd2 Qe7 A variation of the
Bogo-Indian that has had a lot to do with the recent revival of that opening.

5.g3 Nc6 6.Bg2 Bxd2 + 7.Nbxd2 d6 Black plans . . . e 5 , when after d5 he will
lose space but have the better Bishop. In a later game versus Platanov at Wijk
aan Zee 1 970, Benko used a variant of the key idea under consideration: 7 . . . 0-0
8.Re l ! ? d6 9.Nf1 ! e5 1 0 .Ne3 e4 1 1 .Nd2 Nxd4 1 2 .Nxe4 Nf5 1 3 .Nd5 Nxd5 1 4.cxd 5 ,
with pressure down the c-file. See Game 96. 8.Nf1 !

A remarkable notion. White understands that after 8 . 0-0 e 5 , he won't have a

particularly good square for his Knight on d2 . Now, however, he can place it on
e 3 , where after . . . e 5 , it will control the crucial d5 and f5 squares. This costs time
and delays development, but Benko believes that Black can do nothing too dis
ruptive in the meantime. 8 . 0-0 9.Ne3 Ne4 1 0.0-0 ! ? The only other game I
..

BENKO AS W H ITE

can find for the Nfl idea went 1 O. d 5 Nd8 1 1 .Nd4 Nc5 ( l 1 . . .f5 1 2 .Qc2) 1 2 .b4
Na6 1 3 .a3 e5 1 4.Ndf5 Qf6 1 5 .Be4 with a significant advantage, Trifunovic-Keres,
Prague 1 9 3 7 . 1 O . . . Nd8 ? ! A preemptive move against d 5 . Thus Black prepares to
play l 1 . . .f5 , and on 1 2 .d5 he can respond by 1 2 . . . e 5 . This plan is awfully slow,
but 1O . . . Bd7 1 1 .Nd2 ( l l .b4! ?) l 1 . . .Nxd2 1 2 . Qxd2 gives White space and the
possibility of expanding on the queenside via b4 or on the kingside via f4-f5 .

1 1 .Qc2 f5 1 2 .d5 Ng5 ! ? Now 1 2 . . . e5 1 3 .Nh4! wins a pawn . 1 3 .dxe6 NxB +

1 4.Bxf3 Bxe6 1 5 .Nd5 Qf7 1 6.Nf4!

1 6 . . . c6 The pawn can't be taken, since 1 6 . . . Bxc4? 1 7 .Rfe l ! Be6 1 8 . Qxc7 destroys
Black's position. 1 7.Rad l g5 And here 1 7 . . . Bxc4 1 8 .b3 Be6 1 9.Rxd6 g5 2 0.Nxe6
Nxe6 2 1 .Rfd 1 still doesn't free Black's game, but it does create weaknesses in his
own camp . 1 8.Nxe6 Qxe6 1 9.Rd4. White has pressure on d6 and active pieces.
Benko went on to win.

Queen's I nd ian Defense

As in many other openings, Benko helped to establish the initial theory on

one of the main lines of the Queen's Indian Defense: l .d4 Nf6 2 .c4 e6 3 .NB
Benko's usual preference, as played in the previous game. He did allow the Nimzo
Indian (3 .Nc3 Bb4) early in his career, but very seldom after the mid- 1 95 0s.

3 ... b6 4.g3 Bb7 5 .Bg2 Be7 6.Nc3 Ne4 7.Bd2

Presumably this whole setup, which was to become a main line, had not previ
ously appealed to White because it allowed Black to win the Bishop pair. Games
7 1 and 57 in this book (versus Matanovic and Parma) feature Benko at the fore-

485

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

486

front of QID theory, and the notes to those games cover the ideas and variations
well. I just want to add some examples from Benko's extensive experience with
the 6.Nc3 and 7 .Bd2 system to complement those two games. I have cited no
less than ten games of his with 7 .Bd2 here ! 7 ...dS This move was played repeat
edly in the early days, and in fact Benko's own games still constitute the core of
theory on 7 . . . d S . But he has also faced no less than four other instmctive moves,
many in the early days of the variation:
(a) To begin with, 7 . . . fS S.dS Bf6 9.Qc2 Na6 is the Matanovic game mentioned
above (current theory gives the nod to 9 . . . Qe7). But Benko was also involved
with Black's attempt to make the line 9 . . . 0-0 ! ? 1 0.Nxe4 fxe4 1 1 . Qxe4 exdS work,
since as early as 1 964 he found the positional nicety 1 2 . Qc2 ! Nc6? (avoiding
1 2 . . . ReS 1 3 .O-O! with the idea 1 3 . . .Rxe2 1 4.cxd S , with some advantage in view of
14 . . . Bxd S ? ? l S .Qd3 ) 1 3 .cxdS Nd4 1 4.Nxd4 Bxd4 1 S .e3 Qf6? ! 1 6.Bb4! Bxb2 1 7 .Rb 1
BeS l S .BxfS Bc3 + 1 9 .Kfl RxfS 2 0.Be4 ! , and Black lacked real compensation in
Benko-Milic, Belgrade 1 964.
(b) Another early game went 7 . . . Nxd2 S.Qxd2 dS 9.cxdS exdS 1 0.0-0 ( l O.NeS ! ? )
1 0 . . . Nd7 1 1 .Rfd 1 0-0 1 2 .NeS c 6 1 3 .Nd3 ReS 1 4.b4 with unclear play, Benko
Sherwin, New York 1 961 .
(c) Double-edged play resulted from 7 . . . d6 S .O-O Nd7 9.dS ! ? (9.Nxe4 Bxe4
1 0.Bc3 0-0 I l .Bh3 ! ? might be of interest) 9 . . . Nxd2 1 O.Nxd2 eS 1 1 .b4 as 1 2 .a3
0-0 1 3 .Nb 3 in Benko-Pomar, Torremolinos 1 97 3 .
(d) Lastly, 7 . . . 0-0 S.dS Nxd2 9.Qxd2 Bf6 1 O.0-0 c S 1 1 .Ne l ! threatening 1 2 .dxe6
and preparing moves like Nd3 , e4 and b4, was Benko-Sahovic, Lone Pine 1 97 5.
8.NeS Benko played this in four games. He also tried S.Nxe4! ? dxe4 9 .NeS f6
1 0.Qa4+ c6! ? ( l O . . . KfS ! I l .Ng4 Qxd4 1 2 .Bc3 Qd7 looks better) I l .Ng4 Qxd4
1 2 .Bc3 bS 1 3 .Qc2 against Kuijpers at Tel Aviv 1 964, when it seems that 1 3 . . .Qxc4
should have been played, with a sort of dynamic equality after 1 4.Ne3 . Of course
there are various options here . 8...0-0

The alternative is S . . . Nxd2 9.Qxd2 c6 1 O .e4 ! ? dxc4 ( 1 0 . . .f6 l 1 .Nf3 dxc4 1 2 .Bh3
Qd7 1 3 . Qe2 ! bS 1 4.dS ! cxdS l S .exdS BxdS 1 6.NxdS QxdS 1 7 .Bxe6 etc.) 1 1 . Nxc4
Nd7 1 2 .0-0 ( 1 2 .0-0-0 ! ? ) 1 2 . . . Ba6 1 3 .b3 Bxc4 1 4.bxc4 NeS ! l S .Rfd l Nxc4 1 6. Qe2
bS 1 7 .a4 Bb4 l S .Na2 , with a position that was wild and unclear in Benko
Rossolimo, New York 1 966. 9.Nxe4! 9.cxdS was played versus Parma in Game
S 7 . After 9 . . . Nxc3 1 0 .Bxc3 exdS I l .Qa4, Parma played 1 1 . . . Bd6, while Benko

BENKO AS WHITE

took a quick draw following 1 1 . . . Qd6 1 2 .0-0 RdS 1 3 .Rad l versus Karpov in
Caracas 1 970 ( 1 3 . . . Nd7 looks fine for Black). 9 ... dxe4 1O.Qc2 f5 Correctly avoid
ing 10 . . . Qxd4? 1 1 .Bc3 QdS 1 2 .Bxe4. 1 1 .Bc3 Bf6 1 2 .Rd 1 Or 1 2 .0-0-0; White is
surely better in this position. 12 ... Qe7 1 3 .0-0 a5 14.f4 This is Benko-Rossolimo,
New York 1 96 5 . White has a nice advantage .

Budapest Gam bit

Benko dealt with the Budapest in typically restrained positional style, as in
Benko-Ault, Boston (U.S. Open) 1 964 and in the note to move 4: l .d4 Nf6
2 . c4 e5 3 .dxe5 Ng4 4.Nf3 Benko-Ragozin, Budapest (! ) 1 949 (see Game 1 0)
went 4.Bf4 Nc6 S Nf3 Bb4+ 6.Nbd2 Qe7 7 .a3 Bxd2 + S. Qxd2 NgxeS 9NxeS
NxeS (White is employing an antidote to the Budapest that is very popular these
days. It was used throughout the 1 9 3 0s and 1 940s, but then faded from view for
a while) 1 O.e3 ( 1 0.cS ! - Benko) 1 O . . . d6 I l .Be2 Bd7 1 2 .0-0 Bc6 1 3 .Rac 1 0-0. Benko:
"White has a slight advantage because of his Bishop pair. " 4 . . . Nc6 5 .Bg5 !

This i s the earliest case I have found of this simple move (excepting a game that
continued S . . . Be7 6.Bf4 BcS, transposing back to a normal line). I'm sure that
there are earlier examples, but it's likely that Benko came up with S .BgS over the
board. At first, it may see odd to bring out more black pieces, but in return White
secures a basic positional advantage with his control of d S : 5 . . . Be7 6.Bxe7 Qxe7
7.Nc3 Ncxe5 Lalic analyzes 7 . . . Qb4 S.Qb3 Qxb3 9.axb3 NgxeS l O NxeS NxeS
I I .NbS , and White is much better (he wins a pawn). The main move is 7 . . . NgxeS .
Then S.NdS QdS (S . . . Nxf3 + 9.gxf3 favors White) 9.NxeS NxeS 1 O . Qd4, and
White has space and a positional advantage. 8.Nd5

487

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

S ... QcS 8 . . . Qd8 9.NxeS NxeS 1 O.Qd4 transposes to the note to 7 . . . NgxeS above.

488

9.e3 NxB + 1 0.gxB ! This was queried by Benko himself in one source, but ap
pears to be very strong. 10 . . . Nf6 I 1 .Nxf6+ gxf6 1 2 .Bd3 ! ? Benko has won the
opening battle, and while this move is not bad, 1 2 . Qa4! seems to give White a
clear advantage. 1 2 .QdS ! ? also deserved consideration. 1 2 ... d6 1 3 .Qb3 ! Bd7
14.Qxb7 ! ? Leading to wild complications; safer was 1 4.0-0-0 Bc6 I S .Be4 with
some advantage . 1 4 ... Bc6 I S .Qb3 BxB 1 6.Rgl Ke7 ! 1 7.Rg3 QaS+ I S.KfI
RabS 1 9.Qc2 QhS 2 0.cS !, with an unclear attack on Black's King. 2 0 . . . dxcS
would be met by 2 1 .Rc 1 .

T he Slav Defense
Moving on to some Queen's Gambits, the following Slav Defence from Benko
Sigurjonsson, Caracas 1 970 is a good example of how Benko exploits a harm
less but irregular opening: l .NB dS 2 .d4 Nf6 3 .c4 c6 4.e3 e6 S .Bd3 Nbd7
6.Nbd2 ! " ! " for surprise value rather than theoretical superiority. This variation
of the Slav was popular in the 1 92 0s and 1 93 0s, with games featuring some of
the best players in the game. Today it has mostly lost its appeal, but is solid and
certainly playable. 6 ... cS ? ! This move has done badly, primarily because, as Benko
says, it wastes a tempo with . . . c6-cS . Slav players may be interested in some theory
here. Of the normal responses to 6.Nbd2 is 6. . . Be7, which has worked out well
enough over the years, but the move that most aggressively counters 6.Nbd2 is
6. . . Bd6, aiming for . . . e S . Then fairly typical play goes 7 .0-0 (7 .e4 e S ! 8.cxdS cxdS
=. For example, 9.exdS exd4 1 0. Qe2 + Qe7 1 1 .0-0 Qxe2 1 2 .Bxe2 NxdS 1 3 .Nxd4
0-0 1 4.Nc4 BcS I S .Nb3 Be7 with no problems, Mikhalchishin-Kosic, Yugosla
via 1 994.) 7 . . . 0-0 8 . e4 e S ! 9.cxdS cxdS 1 O.exdS exd4.

A well-tested position in which Black has sufficient play, for example, I I .Nc4!?
( l 1 .Nxd4 N e S 1 2 .Ne4 Nxd3 1 3 . Qxd3 Nxe4 1 4. Qxe4 Re8 I S .Qd3 ReS ! , Collas
Becerra, Linares 1 997; I 1 .Ne4 Nxe4 1 2 .Bxe4 NcS ! 1 3 . Qxd4 Nxe4 1 4.Qxe4 Re8
I S .Qd4 BfS ) 1 1 . . .NcS 1 2 .Nxd6 Qxd6 1 3 .Bc4 Bg4 1 4.Qxd4 Bxf3 I S .gxf3 Ne6
1 6. Qh4 NxdS 1 7 .Rd l Nec7 = Salov-Anand, Linares 1 992 . White has the Bishop
pair but at the cost of structural weaknesses on the kingside. 7.b3 This is to
avoid the isolated pawn after . . . cxd4 and . . . dxc4- see Benko's notes. 7 ... Be7 Black
must be careful to avoid something like 7 . . . cxd4 8 . exd4 dxc4 9.bxc4 e S ? (9 . . . Bd6

BENKO AS WHITE

leaves White with some advantage) l O.dxe5 Nxe5 1 1 .Nxe5 Qd4. This is awful
for Black, although it's still a little unclear after either of Benko's suggestions
1 2 .0-0 and 1 2 . Qa4+. However, 1 2 .Qe2 ! is simply winning, due to 1 2 . . . Qxa l
1 3 .Nb3 Qc3 + ( 1 3 . . . Bh4+ 1 4.Kd 1 Qc3 1 5 . Bb2 ) 1 4. B d 2 Qb2 1 5 .Ng6+ etc.
8.0-0 0-0 9.Bb2 b6 10.Qe2 Bb7 Again, please see Benko's game notes for a
discussion of these moves. We are already out of theory; a few more moves are
needed to demonstrate White's opportunities: 1 1 .Rfd l ! ReS Stepping out of
the shadow of the d 1 -Rook by 1 1 . .. Qc7 is also possible, when 1 2 .Rac 1 Rad8
1 3 .cxd5 exd5 1 4.e4! ensures White the better game. 1 2 . Rac 1 Re7 ! ? 1 3 .dxeS !
NxeS 1 4.Be2 QaS I S .NgS !. A surprisingly effective attacking idea that nets
immediate dividends; see Game 94.

Q ueen's Gam bit Tarrasc h Defense

White does nothing theoretically significant in Benko-Pomar, Las Palmas
1 972, but he exploits his opponent's slight inaccuracies in an instructive iso
lated queen's pawn position: l .e4 Nf6 2 .Nfl e6 3 .g3 dS 4.Bg2 eS S .O-O Ne6
6.exdS exdS 7.d4 After skirting through several possible openings, a Queen's
Gambit Tarrasch Defense finally appears on the board. 7 . . . Be6 A rare move .
Usual is 7 . . . B e 7 . S.b3 Benko liked this modest fianchetto, and he played the
analogous . . . b6 when confronted with a revers ed Tarrasch Defense. S e e
Burger-Benko, U. S . Ch. , Game 89. S . . . B e 7 9.Bb2 0 - 0 1 O.Nc3 Ne4 ! ? Ambi
tious, but probably playable . Black can also consider simpl y lO . . . Rc8 . 1 1 .dxeS
BxeS 1 2 .e3

1 2 ... Be7 ? ! Black intends to play ... Bf6 and neutralize White's Bishop on the
long diagonal. Now 1 3 .Nxe4 tends to free Black's pieces, and the classic block
ade of the queen pawn by 1 3 .Nb5 Bf6 1 4.Nbd4 is only superficially attractive
after 1 4 . . . Qa5 and the transfer of Rooks to the center ( . . . Bg4 is also an idea).
But Benko finds a precise response, and with hindsight, better looks 1 2 . . . Re8
or 1 2 .. .f6, the latter move covering e5 and blocking off the b2-Bishop (as in the
Keres-Parma line of the English Opening). 1 3 .Ne2 ! This subtle move is supe
rior to the more conventional 1 3 .Nb5 in the line 1 3 .Ne2 Bf6 1 4.Nfd4! , and
White can play Nf4 next with the serious threat of Bxe4. 13 ... ReS 1 4.Nf4. Sud
denly White is considerably better. He will play his Queen to e2 and a Rook to

489

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

d 1 , tying Black to defense of the d-pawn. Benko explains the further course of
the contest in Game 1 07 .
490

Semi-Tarrasch Q ueen's Gam bit

The line arising in Benko- Lounek, Vienna, 1 947 has been the subject of many
famous top-level encounters. The game is brilliantly examined in Game 6; so here
I will provide some theoretical background: l .Nf3 d5 2 .d4 Nf6 3 .c4 e6 4.Nc3
c5 5.cxd5 Nxd5 6.e4 Benko also experimented with 6.g3 , for example, against
Korchnoi at Curaao 1 962 (see Game 43 ) and Donner at Wijk aan Zee 1 970 (see
Game 9 5 ). In both of these games, Queens were exchanged early and Benko again
showed his remarkable strength in Queenless middlegames. 6 ...Nxc3 7.bxc3 cxd4
8.cxd4 Bb4+ 9.Bd2 Bxd2 + Exchanging Queens is definitely a poor idea after
9 . . . Qa5 1 O.Rb l Bxd2 + 1 1 . Qxd2 Qxd2+ 1 2 .Kxd2 , and White has better develop
ment, the center, and an active King. 1 O.Qxd2 0-0 I l .Bc4 Nc6 1 2 .0-0 b6

Over the years, several very strong players have tried 1 2 . . . Qd6 here (preventing
Qf4) and have enjoyed some success with it. 1 3 .Rad l ! As Benko points out, this
game (in 1 947 ) predated the famous games in the next notes by over 20 years !
Of course, his game with this line is very obscure and there was at least one and
probably more players of White who had played 1 3 .Rad 1 before, but in any case
we see Benko anticipating later theory in a very important line. 13 ... Bb7 A "schol
arly" aside here: database sources give Polugayevsky-Tal from the 1 969 USSR
Championship as continuing 1 3 . . .Bb7 1 4.Rfe 1 Na5 1 5 .Bd3 RcS . I think, how
ever, that Benko is right in asserting that 1 3 . . . Na5 came first, because Polugayevsky
himself gave that order in his Grandmaster Preparation. A brilliant game resulted
after 1 4.Bd3 Bb7 1 5 .Rfe 1 RcS 1 6.d5 ! ! exd5 1 7 .e5 Nc4 l S .Qf4 Nb2 1 9 .Bxh7 + ! ,
and Polugayevsky won one o f the finest games o f his career. 14.Qf4 Spassky
Petrosian, Moscow 1 969 went 1 4.Rfe 1 RcS I 5 .d5 ! exd5 I 6.Bxd5 Na5 1 7 .Qf4,
with a space advantage and open lines.

BENKO AS WHITE

49 1

1 4 . . . Rc8 ! ? Benko doesn't comment upon this move, but the many later contests
continued 14 . . . Qf6! with great success. Thus 1 4.Rfe 1 is considered more accu
rate than 1 4.Qf4. 1 5 .d5 exd5 1 6.Bxd5 Qe7 The only precursor to this game
that! can find went 1 6. . . Qc7 1 7 .Qg4 ( 1 7 .Qh4 ! ? ) 1 7 . . . Ba6 1 8 .Rfe 1 Ne5 1 9.Nxe5
Qxe 5 , with at best a minor edge for White due to his space, Reinhardt-Tautvaisas,
Buenos Aires 1 9 3 9. 1 7.Ng5 Ne5 1 8.Bxb7 Ng6 1 9.Qf5 Qxb7 20.Rd7 Qa6
2 1 .Rxf7 Qxa2 , and after this series of more or less forced moves, Black could
have held on with accurate play.

Queen's Gam bit Declined

In the traditional QGD, Benko was an early adherent of the 5 .Bf4 system that
later became quite popular. His opening against Lengyel from Venice 1 969
shows how easily Black can slip into positional difficulties: l .c4 e6 2 .NB d5
3 .d4 Nf6 4.Nc3 Be7 5 .Bf4 0-0 6.e3 Nbd7

7.cxd5 The main move these days is 7 . Qc2 . But the game Benko-Medina Garcia,

Palma de Mallorca 1 968 , quoted in Benko's notes, is still cited today: 7. a 3 c5 ! ?

(also possible i s just 7 . . . dxc4 8 .Bxc4 a6, e.g., 9.Qc2 c 5 1 0 .dxc5 Bxc5 l 1 .b4 Be7=,
Loebler-Singer, Austria 1 99 5 ) 8.cxd5 exd5 9.Be2 a6 1 O.dxc5 Nxc5 1 1 .0-0 Be6
1 2 .Be5, with a slight advantage .
7.cxd5 is a thematic response to Black's . . . Nbd7 because in the Exchange Varia
tion structure that arises, White's Bishop is well posted on f4. Indeed, White
often voluntarily loses a tempo by Bg5-f4 in the main lines of the Exchange Varia
tion of the Bg5 Queen's Gambit Declined. 7 . . .Nxd5 7 . . . exd5 8 . Qc2 would illus-

PAL BENKO: MY LIFE. GAMES AND COMPOSITIONS

492

trate the point in the last note. 8.Nxd5 exd5 9.Bd3 Nf6 Benko mentions 9 . . . Bb4+
1 0.Nd2 , after which White can continue with moves like 0-0 and Qc2 , although
this may not achieve much. He could also play 1 0.Ke2 ! , when I think that White
has a small but definite advantage. In a recent theoretical article on 6 .. Nbd7 ,
Zoran Ilic gives one fragment and suggests that 9 . . . Bb4+ equalizes, but h e sup
plies only one unconvincing example. 1 0.Qc2 c6 1 l .0-0 Bg4 Benko calls this
"an inaccuracy, " although after his 1 1 . . . h6, 1 2 .Ne5 would similarly grant White
nice prospects, whereas the continuation 1 1 . . .a5 1 2 .h3 keeps White's options open
and maintains some pressure. 1 2 .Ne5 Be6 l 3 .b4!

A very modern idea that comes up in a number of contemporary Queen's Gambit

lines, launching the traditional minority attack a tempo. The only issue is whether
Black can accept the pawn offer. l 3 ... Rc8? ! Conceding White's tempo gain. In
stead, 1 3 . . . Bxb4 1 4.Rfhl Qe7 1 5 .Qb3 still ought to favor White somewhat, e.g.,
1 5 ... c5 ( 1 5 ... a5 1 6.a3 Bd6 1 7 . Qxb7) 1 6.a3 ! ( 1 6.dxc5 Bxc5 1 7 .Qxb7 Qxb7 1 8.Rxb7
Nh5 =) 1 6. . . Ba5 1 7 .dxc5 Qxc5 1 8.Qxb7, with some edge in a complicated position. 14.Rab l h6 1 5.Rfe l Now White has a clear advantage. See Game 9 l . An
extremely instructive opening for players on either side of the Queen's Gambit.

Grunfeld Defense
Benko tried several systems against the Griinfeld Defense, but either by nor
mal means or by transposition, he most often ended up in a g3 system (fifteen of
them in my database). His first match game versus Reshevsky in New York
1 960 is still theoretically important: l .d4 Nf6 2 .c4 g6 3 .g3 Bg7 4.Bg2 d5 5.cxd5
Nxd5 6.Nf3 0-0 7.0-0 Nb6 8.Nc3 Nc6 9.d5 Nb4 1 O.e4 c6 I l .Qb3

BENKO AS WHITE

Benko says that he "dreamed up this move over the board" in preference to the
"more natural" 1 1 .a 3 . Actually, although there had not been many games with
9 . . . Nb4 up to this point of time; 1 1 .Qb3 had been played more often than 1 1 .a3 .
Interestingly, 1 1 .Qb3 has continued to score well over the years, but in the moderate revival of 9 . . . Nb4 in 1 999-2000, White has chosen 1 1 .a3 Na6 1 2 .bxc6 in
every game, with indifferent results ! Thus 1 1 .Qb3 is still critical to theory.
1 1 . . .Nd3 Oddly enough, the retreat to a6 isn't bad and has achieved fair results.
1 2 .Be3 c5 Benko's 1 2 . . . Nxb2 ! ? 1 3 .Qxb2 Na4 1 4.Nxa4 Bxb2 l S .Nxb2 hasn't been
tried. 1 3 .Ne l c4 14.Qc2 Nxe l 1 5 .Rfxe l , and as Benko shows in Game 3 5 ,
White i s better.
Another anti-Griinfeld system employed by Benko was 4.Nf3 Bg7 S . BgS . As so
often, this is an essentially harmless system, but at the time there was little theory
associated with it. Moreover, several key variations lead to Queenless middlegames
(surprise ! ). This occurred in Benko-Pinter, Szolnok 1 97 5 : l .c4 g6 2 .d4 Nf6
3 .Nc3 d5 4.NB Bg7 5 .Bg5 Ne4 Benko also met the respectable move S . . . dxc4
well before its theory was established, and outplayed his opponent: 6.e4 cS 7 .dS
bS 8.eS b4 9.exf6 exf6 1 0.Qe2 + Kf8 1 1 . Be3 bxc3 1 2 .BxcS+ Kg8 1 3 .bxc3 hS ? !
( 1 3 . . . Nd7 ! is equal after 1 4.Be 7- 1 4.Qxc4 Bb7- 1 4 . . . Qe8 l S . Bb4 Bb7 1 6.0-0-0
Nb6, Flear-Kouatly, Brussels 1 986) 1 4.Qxc4 Ba6 l S .Qb3 Nd7 1 6. Be 3 , and Black
has no compensation, Benko-Hartston, Hastings 1 974. 6.Bh4

6 . . . c5 The modern solution is 6. . . Nxc3 7 . bxc3 dxc4, when White has to choose
between a speculative gambit or misplacing his Queen: 8.e3 (8.Qa4+ Qd7 ! 9. Qxc4
b6! , with the idea . . . Ba6) 8 . . . bS 9.a4 c6 1 0. Be2 a6 1 1 .Nd2 0-0 1 2 . Bf3 Ra7 , and
Black was still a pawn up in Sorin-Kasparov, Buenos Aires (simul) 1 997. 7.cxd5
Nxc3 8.bxc3 Qxd5 9.e3 cxd4 1 0.cxd4 Nc6 1 1 .Be2 e 5 ? The sequence 1 1 . . . 0-0
1 2 .0-0 eS is now preferred, although it is still by no means easy for Black after
1 3 .dxeS . 1 2 .dxe5 Qa5+ 1 3 .Qd2 Qxd2 + 14.Kxd2 Nxe5 15 .Rab l 0-0 1 6.Nd4!

49 3

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

494

Benko arrived at this position twice and won both games. Black will have great
difficulties developing his pieces in view of Rhc 1 and Rc7 , among other ideas.
16 ... h6 ! ? This game predates the normally cited Gheorghiu-Jansa, Skopje 1 97 6.
There Black played 1 6. . . Nc6 1 7 .Nxc6 bxc6 1 8 .Rhc 1 Be6 1 9.Bc4 ! , which looks
pretty awful for him in view of his weak pawns combined with White's open files
and centralized King. Benko's first game with this line went 1 6. . . a6 (first played
by Enklaar versus Uhlmann in 1 97 2 ; Benko seems to have been only the second
player to have played the whole line up to 1 6.Nd4) 1 7 .Rhc 1 Re8 1 8 .f4! Nd7
1 9.BB with an obvious advantage, Benko-Rosino, Venice 1 974. 1 7.f4!? gS l S.fxgS
Ng6 1 9.Bg3 hxgS 20.Bc7 ! , and White, preventing . . . Rd8 , retains the advan
tage, as seen in Game 1 2 5 .

King's Indian Defense

Versus the King's Indian Defense, Benko played g3 systems, 4.Bg5 , the Averbach
S ystem (5 .Be2 and 6.Bg5 ), orders with 4.e4 and 5 .Ne2 , h3 systems, the Samisch
Variation, and the Four Pawns Attack! An incredible assortment. In addition to
all those, he came up with a new and important scheme in the Main Line ("Clas
sical") KID: l .c4 g6 2 .d4 Bg7 3 .Nc3 d6 4.NfJ Nf6 S .e4 0-0 6.Be2 eS 7.0-0
Nc6 S.dS Ne7 9.Ne 1 Nd7 1 0.fJ fS The most-played and best-analyzed posi
tion of the King's Indian. After countless master games and analysis, Benko be
gan to use a paradoxical and unique new idea: 1 1 .g4!

BENKO AS WHITE

" ! " for the move's originality and boldness. Benko explains the point of this move
in the well-annotated Game 3 2 . What he doesn't mention is the reason that no
one had come up with I I .g4 before, i.e., that it moves a pawn in front of one's
own King on the very side of the board in which Black has his attacking chances !
Such an idea flouts classical chess principles and simply wouldn't occur to most
players. Since Benko helped to introduce and then developed the key ideas of
l 1 .g4, this is usually called the "Benko Variation. " Today, l 1 .g4 is still played
with varying results-I have 468 games with it in my database, many from the
last few years. True, Black has come up with plans to prevent the paralysis of his
attack that can follow White's moves h4 and Ng2 . Therefore this is no longer
feared as a refutation of Black's play, i.e., theory considers that both sides have
satisfactory play. It is worth mentioning, however, that Benko's idea has been
given an extra dimension via the order 1 0.Nd3 f5 I 1 .Bd2 Nf6 1 2 . f3 f4 1 3 .g4! ? ,
which, a s Nunn and B urgess say, i s "a critical test of Black's whole system. " Thus
the g4 concept has an ongoing vitality. 1 1 . . .f4 Certainly not the best idea, as
Benko shows. Led by John Nunn and others, players and theoreticians have turned
to 1 1 . . . Kh8 as the most reliable solution. Here I should note two historically
significant games. Portisch-Dely, Budapest 1 9 5 6 was apparently the first l 1 .g4
game by a major player (with Benko's 1 95 8 game mentioned below probably the
second). That game went 1 1 . . .h6 1 2 .h4 Kh7 ! ? (there are later games with 1 2 . . . Nf6)
1 3 .Be3 ? ( l 3 .Nd 3 prepares to protect g4 and h3 by Nf2 ) 1 3 . . . fxg4 1 4.fxg4 Rxfl +
1 5 .Bxfl Nf6 1 6.Be2 Qd7 ! 1 7 . g5 Qh3 , with a good game. Notice that this poor
start for White didn't divert Benko from investigating and playing the line himself. A still earlier example (although stemming from the order 1 O.g4 f5 1 1 . 8)
was again of Hungarian origin: 1 1 . . . fxg4 (not mentioned among the eight 0)
moves in Nunn and Burgess) 1 2 .fxg4 Rxfl + 1 3 .Bxfl (or 1 3 .Kxfl intending 1 3 . . . Nf6
1 4.g5 ) 1 3 . . . Nf6 1 4.h3 ( 1 4.g5 and 1 4.Be2 100k good) 1 4 . . . c6 1 5 .Nc2 cxd 5 1 6.cxd5
Bd7 1 7 .Be3 , and White had some edge in Sandor-Gereben, Budapest 1 95 2 . 1 2 .h4
as Benko's first game with his system went 1 2 . . . c5 1 3 .a3 Kh8 1 4.Bd2 Ng8 1 5 .Ng2
B f6 1 6.Be l , and Black was at a loss for a plan in Benko-Pachman, Portoroz 1 95 8 ,
although the game was eventually drawn. 1 3 .Ng2 NcS 14.Bd2 Kh 8 I S .Be l !
Bd7 1 6.Bf2 . We are following Benko-Eliskases, Buenos Aires 1 960. Benko's
notes to this struggle are terrific- see Game 3 2 .
Benko enjoyed playing g3 versus the King's Indian, but did not produce any
important opening novelties that I know of. However, his treatment of the . . . Nbd7
defense to g3 in Game 63 versus Medina is well worth seeing-the concept of
1 3 .c5 ! is characteristic of a variety of such positions.
In a rare Samisch Variation (5 .8), Benko filched an important point from Fischer
in the 1 958 Portoroz Interzonal after the following interesting move order:
l .d4 Nf6 2 .c4 g6 3 .Nc3 Bg7 4.e4 d6 5.3 e S ? ! Benko had quite a bit of experi
ence from this position. He discussed it from the Black side with Bronstein in
1 9 5 8 and again versus Wade in 1 960 (a game he lost). In a game versus Tatai
(presented below), he showed his facility with the white pieces. My own feeling

495

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

is that S . . . eS is unnecessary and probably premature (S . . . O-O is a normal Samisch

Variation). ECO seems to agree , giving White an edge in all variations .
6.Nge2 0-0 7.Bg5 !
496

7 . . . exd4 Benko's suggested pawn sacrifice 7 . . . Nc6 S. dS Nd4 ! ? 9.Nxd4 exd4

1 O.Qxd4 h6 1 1 . Bxf6 Bxf6 led to a small white advantage after 1 2 . Qd2 c6 1 3 .Bd3
Qb6 1 4.0-0-0 Bg7 I S .Kb l in Timoshenko-Yrjola, Helsinki 1 986. The other game
featured in this book, Benko-Tatai, Malaga 1 969 (Game 8 8 ), went 7 . . . c6 8 . Qd2
(S .dS is also supposed to confer an edge) 8 . . . QaS 9.dS cxdS l O.NxdS (" Surprise ! "
Benko says, although can w e really b e surprised that h e chooses to enter yet
another Queenless middlegame? ) 1 O . . . Qxd2 + I l . Kxd2 NxdS 1 2 .cxdS f6 1 3 .Be3 ,
with advantage . 8.Nxd4 Nc6 9.Nc2 Be6 9 . . . h6, as Glek once played, looks like
a better move, although 1 O.Bh4 is still lightly better for White. The move . . . Be6
is not useful. 1 0.Be2 h6 1 1 .Bh4, and White was better, even without Fischer's
mistake 1 1 . . .gS ? See Game 2 S for Benko's excellent notes.

o.

e m;ght well "y th" Benko raHed to moh Sup,,-GM ""u, ooly
because his results with Black were slightly subpar. Indeed, this situation is re
flected by Benko's own choice of games for this book, which are heavily skewed
towards White. When one looks at his results, his winning percentage with White
is well above what would be expected from his overall record, and his score with
Black is correspondingly disappointing. My own observation, indicated often in
what follows, is that Benko tended to be satisfied to accept draws as Black, and
that he neglected opportunities to play on versus strong players when he had
achieved an advantage. On occasion, he may also have played less ambitiously
than possible in the opening against lower-rated opponents, leading to a higher
than-normal drawing ratio.
What we have just said makes it all the more remarkable that Benko's opening
creativity as Black was probably more impressive than as White. We will en
counter fresh ideas in just about every major defense that he played, many of
which confounded the world's very top players. Even limiting oneself to exami
nation of the Benko Gambit and Sicilian Defense sections below, one could ar
gue that Benko's most significant contributions to opening theory were while
playing Black. But he also discovered moves and schemes in a variety of other
defenses, especially but not limited to the English Opening, Pirc Defense, Griinfeld
Defense, and King's Indian Defense. Consequently, this section is longer than
the first, and I think that you will be surprised to see his leadership in the devel
opment of many opening schemes that we take for granted today.

497

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

THE ENGLISH OPENING

498

We begin with the l .c4 e5 English Opening. Benko had little trouble with the
main line (2 .Nc3 Nf6 3 .NB Nc6 4.g3), comfortably drawing a number of strong
players. The following game has several examples in its notes that are not terri
bly exciting, but instructive for the calm neutralization of White's plans: l .c4 e5
2 .Nc3 Nf6 3 .Nf3 Nc6 4.g3 Bb4 5 .Bg2 5 .Nd5 was particularly popular at the
time of these games, because the main line with 5 .Bg2 0-0 6.0-0 Bxc3 or 6. . . e4
was doing so well for Black, and thus White wanted another weapon. After 5 .Nd5,
B enko was willing to give up the Bishop pair for effective development and cen
tral control: 5 . . . e4

and now:
(a) 6.Nxb4 Nxb4 7 .Nd4 0-0 (Fairly recently, the move 7 .. .Ng4 ! ? has been dis
covered, threatening . . . Qf6. That seems to equalize.) S.Nc2 Nxc2 + 9.Qxc2 d5 !
1 0.Bg2 ! ? ( 1 O .cxd5 Qxd5 ! I l .Bg2 , and Black has more than enough for a pawn
with the moves . . . RacS and . . . e3 threatened. Here 1 1 . Qxc7 is met by 1 1 . . . Bg4! .)
1 0 . . . dxc4 ( 1 0 . . . d4! ?) 1 1 . Qxc4 Be6 1 2 . Qc2 ReS 1 3 .b3 ( 1 3 .0-0 Bg4!) 1 3 . . .Bg4 1 4.Bb2
Re6, liz_liz, Smyslov-Benko, Wijk aan Zee 1 97 2 . White has two Bishops, but
lacks space and can't castle. Probably "equal" is a legitimate assessment.
(b) 6.Nh4 Bc5 7 .Bg2 d6 S . O-O Be6 9.d3 Nxd5 1 0.cxd5 Bxd5 I l .dxe4 (This was
all book at the time, but later I l .Nf5 ! ? with the idea 1 1 . . .0-0 1 2 .d4 Bb6 1 3 .Ne3 !
was discovered, with advantage to White. So Black is best off playing 1 1 . . . Qf6,
which has been satisfactory in a few tests.) I l . ..Be6 1 2 .Bd2 0-0 1 3 .Bc3 Qd7 1 4.Qd2
a5 ! ? , liz_liz, Christiansen-Benko, Greenville 1 9S0. A move ago Black could have
played for more, but now 1 5 .Bxg7 Kxg7 1 6.Qg5+ draws . 5 . . . 0-0 6.Nd5 6.0-0
Bxc3 7. bxc3 is the main 1 . . . e 5 line of the English Opening: 7 . . . ReS (7 . . . e4 is now
more common) S . d 3 e4 9.Ng5 exd3 1 O.exd3 d6! ?

BENKO AS BLACK

499

(This kind of position has always been considered equal, but here Black should
be a little careful not to get tied down. An easier solution is 1 O . . . b6.) 1 1 .Rb l !
Ne5 (or 11 . . .h6 12 .Ne4 Nxe4 1 3 . Bxe4 Ne5 1 4.f4! Nxc4 1 5 .f5 ! with attack,
Mecking-Tan, Petropolis 1 97 3 ) 1 2 . f4 Ned7 ! ? 1 3 .Ba3 RbS 1 4.Qd2 b6 1 5 .Rbe l
Bb7 1 6.RxeS+ QxeS, 112_112, Ree-Benko, Wijk aan Zee 197 2 . I think White should
play on, since the Rook is badly misplaced after 1 7 .Bxb7 Rxb7 I S .Re l followed
by Qg2 -c6. On the other hand, the Bishop on a3 is itself poorly placed, so what
advantage exists is limited. 6 . . . Re8 7.0-0 Again, Black has a nice space advan
tage to compensate for White's Bishop pair after 7 .Nxb4 Nxb4; for example,
s.o-o e4 9.Nd4 d5 ! 1 O.Nc2 a5 = . 7 . . . Bc5 8.d3 According to modern theory, S.e3
is the best chance, and achieves a small advantage. 8 . . . Nxd5 9.cxd5 Nd4

This maneuver has held up very well over the years. 1 0.Nd2 To play e3 and
chase Black back, but in the meantime White really only has one effective piece
out. 10 . . . d6 l 1 .e3 Nf5 1 2 .Nc4 Ne7 A modest move, preparing . . . c6 and free
ing the cS bishop . 1 3 .Bd2 c6 1 4.b4 Bb6 1 5 .dxc6 bxc6 1 6.b5 ! ? 1 6.Nxb6 Qxb6
1 7 .Re I intending Qa4 in certain cases keeps some pressure on Black, but not
much. 16 ... Bc7 ! 1 7.bxc6 d5 18.Na3 Ba6, 112_ 112, Gheorghiu-Benko, Las Palmas
1972 . Black's positional combination has actually netted him the advantage after
being slightly worse for some time. A possible continuation would be 1 9. Qc2
Bd6! 2 0.Nb l RcS , with advantage.
In the game Steinmeyer-Benko, Omaha 1 959, Benko demonstrates how Black
can play positionally against the 4.rl4 line, which has recently become popular
again: l .c4 e5 2 .Nc3 Nf6 3 .NfJ Nc6 4.d4 Some English Opening players who

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

500

are tired of the maneuvering after 4.g3 have turned to this older, more open
variation. 4 ... e4 Much more common but not necessarily better is 4 . . . exd4 S .Nxd4
Bb4. 5 .Nd2 Bb4 This is a very reasonable setup for Black, and one requiring
little theoretical knowledge. 6.e3 Bxc3 The knight should be captured now, be
fore White plays Qc2 . 7.bxc3 0-0

S.Ba3 S.Be2 is more common, but as one can imagine, Black's e4 strongpoint in

conjunction with White's doubled c-pawns should ensure him good play. One idea
that I presented in my 1 979 book but I've never seen again is S . . . ReS (S . . . Qe7 is rare
but also equal.) 9.0-0 d6 1 O.f3 BfS ! ? ( l O . . . exB I 1 .BxB Rxe3 1 2 .Nb3 Rxc3 ! 1 3 .BgS
Ne7 is an older solution.) l 1 .fxe4 Nxe4! intending 1 2 .RxfS ? ( 1 2 .Nxe4 Bxe4 is slightly
better for Black.) 1 2 . . . Nxc3 1 3 .Qf1 Nxe2 + 1 4.Qxe2 Nxd4. S ... ReS 9.Qc2 d6

1 0.c5 ? ! Best seems 1 O.Be2 BfS I 1 .Rb l b6 1 2 . 0-0 Qd7 =. 1 0 . . . d5 I l .h3 Perhaps
White had intended I I .c4, but Black can utilize dS and develop smoothly there
after, for example, 1 1 . . .Be6 1 2 .Rb l RbS 1 3 .Be2 dxc4 1 4.Nxc4 QdS ! I S .0-0 QgS .
1 1 . . .Be6 1 2 .Be2 Qd7 1 3 .g4? ! Awful looking, but White needs a plan and is
understandably afraid of 1 3 .0-0 Bxh3 1 4. gxh3 Qxh3 with the idea . . . Re6. 1 3 ... KhS.
White was threatening to win a piece by gS . With that threat out of the way, it is
obvious that Black has gotten the better game out of the opening.

Continuing with 1 . . . eS games, Gasic-Benko, Sarajevo 1 970 clarifies the as

sessment of a natural attempt by White that has never been adequately analyzed:
l .c4 e5 2 .g3 c6 3 .NfJ e4 4.Nd4 One of those sidelines that pops up from time
to time. 4 . . . d5 5 . cxd5 Qxd5 6.e3 Nf6 7.Nc3 Qe5 S.Bg2 Na6 !

BENKO AS BLACK

50 1

This move is not mentioned in the main theoretical books and I Ow) thought
that I had first discovered and suggested it myself, being unaware of this game.
Remarkably, 8 ... Na6! has still only been played three times that I can find, with
Benko's use of it predating the next one by twenty-two years ! At any rate, it is
the best move and makes it slightly difficult for White even to equalize. 9.d3
Bg4 1 0.f3 1O.Qd2 O-O-O ! 1 0 ... exf3 1 1 .Nxf3 Qc7 1 2 .0-0 Rd8 ! ? 12 ... 0-0-0 ! looks
risky at first, but Black's King is secure and he would gain time in bringing his
Rooks to the center. 1 3 .d4 Be7. Now Black has the better game-White's cen
tral structure is suspect, in view of 14.e4? ? Rxd4! .
Another historically important line appears in Bukal-Benko, Sarajevo 1 970.
Black demonstrates how to exploit White's natural but flawed opening moves,
which have now disappeared from praxis: l .g3 This move will transpose to an
English Opening as the game goes. 1. . . d5 2 .Bg2 e5 A sort of "Modern Attack"
( 1. . .g6 with colors reversed) occurred in Durao-Benko, Malaga 1970. I really only
mention it because it again reveals Benko's feeling about Queenless middlegames:
2 ... c6 3 .d3 Nf6 4.e4 dxe4 5 .dxe4 Qxd 1+ 6.Kxd 1 e5. Here Benko comments: "Evalu
ating this position, it can be stated that Black has a tiny advantage. White's King
can no longer castle, further loss of time seems inevitable and the Bishop on g2
doesn't have a great future. " He admits that White will have to play badly to be in
danger of losing, but obviously feels that with all the pieces still on the board there
will be plenty of play. 3 .c4 c6 4.cxd5 cxd5 5.Qb3 Nf6 6.Nc3 Again citing Benko:
"This position, now well-known, is usually arrived at from the following move
order: l .c4 e5 2 .Nc3 Nf6 3 .g3 c6 4.Bg2 d5 5 .cxd5 cxd5 6.Qb 3 . " 6 ... Nc6!

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

7.d3 The consistent move is 7 .Nxd 5 . The game notes cite the 1 968 correspon-

502

dence game Abramov-Kuuksmaa in the line 7 ... Nd4 8.Nxf6+ Qxf6 (In fact, 8 . . . gxf6!
is almost certainly better according to modern theory, intending . . . Qc7, . . . Be6,
and . . . Rc8 .) 9.Qd 1 ? (9.Qd3 ! Bf5 1 0.Be4 is double-edged and considered dynami
cally equal.) 9 . . . Bf5 1 O.d3 Rc8 1 1 .Kf1 Qa6! , a pretty move that practically wins
due to the idea of . . . Nc2 and . . . Qxa2 and sacrificial threat on e2 . An earlier game,
Nazzari-Perez, Mar del Plata, saw White play better after 1 1 . Rb 1 ! Bb4+
1 2 .Kfl 0-0 B .Nf3 , although the resulting position was obviously very difficult
to defend . 7 ... d4 8.Nb l Bb4+ 9.Bd2 as ! 1 0.a3 Or 1 0.Bxc6+ bxc6 1 1 . Bxb4 axb4
1 2 . Qxb4 Qd5 . 1 O ... Be6 I 1 .Qa4 Be7 1 2 .Bc1 0-0 With . . . Nd7-c5 and . . . a4 in
mind. The opening has ended with a large advantage for Black. See Game 1 00
for its brutal course.

Mikenas Variation
Feldman-Benko, Budapest 1 945 is a very early and theoretically important
game in the development of Nimzo-English theory, i.e., l .c4 Nf6 2 .Nc3 e6: l .c4
Nf6 2 .Nc3 e6 3 .e4 This is the MikenasVariation, still one of the main lines of
the English Opening. 3 ... d5 Benko's old notes mention the line L. c5 4.e5 Ng8
5 .f4 as a possible disincentive to L.c5; today this is no longer played due to
5 . . . Nc6 6.Nf3 d6, when White is struggling in the center. 4.cxd5 exd5 5.e5
Benko says that this move is "refuted" by the game, and in the sense of it being a
practical winning attempt, he's certainly right. Instead, he suggests 5 .exd 5 ; how
ever, Black is already better after 5 . . . Nxd5 6.d4 Bb4 7 .Bd2 0-0, so this is no im
provement. 5 . . . Ne4!

Rummaging through my own dusty treatise on 1 . . . Nf6, I discovered a game

"Mueller-Kolman, 1 9 3 2 " with this move. Bogoljubov also played 5 . . . Ne4 in 1 93 3 ,
but there were only a few other minor contests with it until the game before
you, and only one in which White accepted the pawn sacrifice. In any case, Benko
probably wasn't aware of any precedents, and this game has to this day discour
aged White from trying to win a pawn as he does in the game. For the record,
although theory on 5 . . .Nfd7 is no longer in the books, I believe that I showed it
to be equal after 6.d4 c 5 . 6.Nxe4? Benko demonstrates this to be risky at best.
Today the normal moves are 6.Nf3 and 6.d4, both apparently leading to equal
ity. 6 ... dxe4 7.Qa4+ Nc6 8.Qxe4 Qd4!

BENKO AS BLACK

503

The most direct and pointed move; but in fact, 8 . . . Be6 was played successfully in
Carls-Helling, Bad Pyrmollt 1 93 3 and then again in Kuppe-Unzicker, Oldenburg
1 949. So one already wonders about White's undeveloped position and poor pawn
structure. 9.Qxd4 Nxd4 1 0.Bd3 A later example was 1 O.Kd 1 Bf5 1 1 .d3 0-0-0
1 2 .Be3 Bc5 of Drimer-Podgaets, Budapest 1 961 . This is playable for White, but
still depressing. 1 0 . . . Be6 Calm development, indicating Black's self-confidence.
I 1 .Be4 My 1 9 79 book has 1 1 .Ne2 here, leading to equality after 1 1 . . .0-0-0
1 2 . Be4 (see the game notes for 1 2 .Nf4) 12 ... Bd5 ! ? Thus, on White's 1 2 th below,
he should consider Ne2 , transposing to that line. 1 1 . . .0-0-0 1 2 .f4? ! Continuing
with the thought of the last note, 1 2 .Ne2 might be answered by 1 2 . . . Bc5 = (or
1 2 . . . Bd5). The alternative 1 2 .Kd 1 Bd5 1 3 .Bxd5 Rxd5 1 4.f4 g5 transposes to the
game. 12 ... Bd5 1 3 .Bxd5 Rxd5 1 4.Kd l g5 ! ? Balogh in the bulletin gave 1 4 . . . Bc5 ,
which is probably better for Black. 1 5 .fxg5 Bb4! 1 6 .Nh3 ? ! , and here the notes
to the game correctly indicate that 1 6. . . Rxe5 ! 1 7 .Rfl Rhe8 favors the second
player. That is enough to define the opening. Thorough notes to the whole con
test can be found in Game 1 .

FLANK OPENINGS
Moving on to a non-English flank opening, Barcza-Benko, Budapest 1 946
saw White slip up in the first few moves of a b3 system. This opening is well
covered in the game notes, so I will just add a few comments: l .Nf3 d5 2 .b3 c5
3 .Bb2 Benko: "It seems that White's third move is already a mistake ! " 3 ... f6 !

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

4.d4 4.c4 d4 S .d 3 eS 6. e3 Ne7 ! 7 .Be2 Nec6 was Petrosian-Fischer, Buenos Aires

504

Candidates Match, 1 97 1 when most commentators think that Black is already

better. The key point here is that once again, Fischer is following Benko's lead
with 3 . ..f6, whether or not he was aware of it. Although seemingly natural, this
move hadn't been played much before Fischer's game and was by no means gen
erally known as being so effective. Benko's was one of the first high-level uses of
3 . . . f6. Nevertheless, a big-time precedent existed, i.e., Lisitsin-Botvinnik, Mos
cow 1 944, played only two years before. That game went 4.e3 eS S .BbS+ Nc6
6.0-0 Bd6 7 .Be2 Nge7 8.d3 Be6, and Botvinnik obviously enjoyed a large advan
tage. 4 . . .cxd4 5.Qxd4 e5 6.Qd2 White is at a loss for good squares. 6.Q d 1 Nc6
7 . e 3 Bb4+! is still awkward. 6 . . .Nc6 7.e3 Bb4! S.c3 Ba5 9.b4 Bb6 Here Black
was already much better; see the extensive notes in Game 3 .
In Kalme-Benko, New York U.S. Ch. 1 960, Benko uses one of the most
effective anti-Reti systems: l .NB Nf6 2.g3 d5 3 .Bg2 c6 4.0-0 Bg4 Today masters
everywhere play this setup, bringing the Bishop outside of the pawn chain. 5.c4
A "Benko Opening"! As we discuss elsewhere, the most legitimate use of this
(dubious) appellation involves an early Nf3 , g3 , and c4, without transposition to
a normal Reti Opening. Such is the case here. 5 . . . e6 6.Qb3 ! ? Qb6 Benko natu
rally has no fear of trading Queens. The resulting Queenless endgame offers
many chances and indeed, he outplays his opponent over the next few moves:
7.Qxb6 axb6 S.cxd5 exd5 9.d3 Be7 1 0.Nc3 b5! I l .a4? !

1 1 . ..Nbd7 ! ? Looking for more, Black foregoes 1 1 . . . Bxf3 ! 1 2 .exf3 (l2 .Bxf3 d4

1 3 .Ne4 Nxe4 1 4.Bxe4 bxa4) 1 2 . . . d4 1 3 .Ne2 Rxa4 1 4.Rxa4 bxa4 l S .Nxd4 0-0,
with the kind of edge that one would normally expect Benko to convert. 1 2 .b3
An alternative was connecting Rooks by 1 2 .Bd2 . 1 2 ...Nc5 1 3 .Rb l ! So as to capture
on b 7 in reaction to any Black capture on a4. Instead, 1 3 .Nd4 bxa4 1 4.bxa4
(1 4.Nxa4? ? b5) 14 . . . 0-0 would have been a little awkward for White. After 1 3 .Rb 1 ,
he has equality. The game ended tragically (but, alas, not entirely atypically) when
Benko's flag fell in a dead equal position.

Modern Defense Against c4

Benko has long been a fan of the Modem Defense ( 1 . . .g6), playing some key
games that influenced its early development. This opening is conceptually di-

BENKO AS BLACK

vided into systems with White having played c4 at some point (as he does in our
example here), and setups without that move (see the games in the next section).
Since there are three orders by which White can get to a c4-based Modern (via
1 .c4, 1 .d4, and 1 .e4), it is not strictly a 1 .e4 opening (nor does ECO assign it as
such); hence its placement here.
In the following game, Benko gladly enters into what is now the best-known
Queenless middle game that arises from the Modern with c4: l .e4 g6 2 .d4 Bg7
3 .c4 d6 4.Nc3 e5 Of course, there are many setups here involving moves like
. . . Nc6, . . . Nbd7, . . . c6 with . . . a6, and the like. Benko played 4 . . . Nd7 5 .Be3 e5 6.Nge2
Nh6! ? in a game versus Roeder in Augsburg 1 99 1 . The experiment didn't work
out badly after 7 . f3 f5 8.d5 0-0 9.Qd2 Nf7

1 O.g3 ( l 0.exf5 gxf5 1 1 .0-0-0 a6! ?) 1 0 . . .Nf6 ( l 0 . . . a5 1 1 .Bg2 Nc5 looks fairly solid)
1 1 .Bg2 c5 ! ? 1 2 .0-0 ( l 2 .dxc6! bxc6 1 3 .0-0 should keep some edge.) 12 . . . Bd7 1 3 .a4
Qe7 1 4.Kh 1 h5 ! 1 5 .b3 h4 1 6.gxh4 f4 1 7 .Bf2 Nh5 , with serious kingside pressure
via . . . Bf6, . . . Kg7 , . . . Rh8 etc. S.dxe5 dxe5 6.Qxd8+ Kxd8 7.Bg5+ 7 . f4 is consid
ered the most dangerous move versus Black's setup, but the text move is also
popular, developing rapidly. 7 . . . f6 8.0-0-0+ Nd7 9.Be3 Bh6

One of Black's main ideas is to exchange his bad Bishop for White's good one.
1 0.Bxh6 Nxh6 1 1 .h4! ? A move that hopes to put pressure on the kingside, but
also wastes time. 1 1 . . . c6 ! Here is Black's key move, after which the King goes to
e7 or c7 without being harassed by Nd5 + . A Knight will try to get to the hole on
d4. 1 2 .h5 Ke7 1 3 .hxg6 hxg6

505

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

506

1 4.NO White's dilemma with his bad Bishop is illustrated by 1 4.Be2 (versus an
eventual . . . Ng4) 1 4 . . . Nc5 1 5 .f3 Be6 1 6.Nh3 a s , and Black's pieces are more effective. 14 ... NfS 1 5 .Nh4 Be6 1 6.Be2 Rd8 1 7.g3 Finally White is ready for the
positive idea of f4. 1 7 ... Rxd l + 1 8.Nxd l Ntl 1 9.Ne3 Ke8 ! ? To prevent forks
on g6 or f5 . 1 9 . . . Ng5 might be played immediately, thinking about . . . Bd7 or
. . . Bf7 and . . . Nge6-d4. Black tries the same idea next move. 2 0.Rd l Ng5 2 1 .0
Btl 2 2 . c 5 ! Finally freeing the f1 Bishop ! 22 . . . Nge6 2 2 . . . Bxa2 2 3 .Bc4 Bxc4
2 4.Nxc4 Nf7 2 5 .Na5 regains the pawn with a good game. 2 3 .Nc4 Nd4 24.Nd6+
Ke7 2 5 .Bd3 Bxa2 26.Nxb7 Bb3 I like the look of Black's game here, although
White kept an approximate balance and achieved the draw in Janosevic-Benko,
Majdanpek 1 976.

l.e4 OPENINGS
This highlight of this section is Benko's beloved Sicilian Defence, which accounts
for a numerical majority of his l .e4 games as Black. But he by no means limited
himself to that response. A close look at the sections below (especially on the
Pirc Defense) reveals how extensive his contributions have been in other e-pawn
openings as well.

Modern Defense
The Modern Defense was just coming into fashion in the late '60s, and Benko
jumped right into playing this anti-theoretical opening, one that had the refreshing
quality of allowing for entirely original play. My database contains quite a few
non-transpositional Modern Defenses (i .e., without . . . Nf6) by Benko from 1 966
to 1 970, and less frequent use later. But here's an interesting game from 1 9 7 5 in
what can be considered the main line of the Modern Defense (i.e., without the
move c4 by White) : 1 .e4 g6 2 .d4 d6 3 .Nc3 Bg7

BENKO AS BLACK

507

4.Nf3 4.Be3 c6 5 . Qd2 is a common choice today. In R. Byrne-Benko, Sarajevo

1 967 , Black essayed 5 . . . b5 ! ? 6. Bd3 Nd7 7. Nge2 Nb6! ? This is perhaps too
hypermodern (primarily due to the queenside weaknesses more than his lagging
development), but White did not react aggressively and Black got enough play
to achieve a quick draw.
For 4.Be2 c6! ? 5 .ND Bg4, see Mestrovic-Benko, Sarajevo 1 967 , Game 67 .
4 . . . a6! ? 5 .Be2 b5 6.0-0 Bb7 7.Re l Nd7 Avoiding 7 . . . Nf6 S.e5 Nfd7 (S . . . dxe5
9.Nxe5 intending BD) 9.e6! fXe6 1 O.Ng5 NfS 1 1 .Bg4 with advantage, e.g., 1 1 . . .b4
1 2 .Bxe6! bxc3 1 3 .Qg4! etc. 8.Bg5 c5 Benko's play is very similar to that of some
leading players in recent Modern Defense games. Now White allows Black to
enter into a sort of Sicilian Defense.

9.a4 Maintaining the center by 9.d5 doesn't seem right after 9 . . . Qa5 ! (threaten

ing . . . b4) 1 O.Qd2 h6 1 1 .Be3 Ngf6, threatening . . . b4 and planning . . . Ng4. In this
line, 1 O.a3 Ngf6 ( l O . . . Bxc3 ! ?) 1 1 .Bfl 0-0 is very comfortable for Black. 9 . . . h6
1 O.Bh4 cxd4 I l .Nd5 ! ? Trying to exploit the pinned e-pawn. 1 1 . Nxd4 Qb6 is
again effective for Black. In general, his fianchettoed Bishops have considerable
influence in this position. 1 1 . . .bxa4! Black doesn't want to lose time. Ideas such
as 1 1 . . .g5 1 2 .Bg3 d3 1 3 .Bxd3 Bxd5 1 4.exd5 Bxb2 are much too risky - White
could even sacrifice the Exchange by 1 5 .axb5 ! . 1 2 .Nxd4 Ngf6 1 2 . . . Nc5 1 3 .Bc4
RcS would continue to neglect development of the kingside, but might be in the
spirit of Black's setup anyway. Benko's move is more conservative. 1 3 .BO This
is Browne-Benko, U.S.A. 1 975; now Benko played 1 3 . . . Nc5 , allowing the dou
bling of his f-pawns (the game was quickly drawn); but he might well have been

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

able to risk 13 . . . eS ! ? , e.g., 14.Rxa4! ? (14.Ne2 ? ? gS 1S .Bg3 g4) 14 . . . exd4 1 S .Nxf6+

Bxf6 16.Bxf6 (16. eS Bxh4) 16. . . Qxf6 17 .eS dxeS 18 .Bxb7 Ra7 . In any case, the
opening was a success for Black.
508

Pirc Defense
The Pirc Defense ( l .e4 d6 2 .d4 Nf6) can be reached by a variety of orders.
Apart from his favorite Sicilian Defense, Benko chose the Pirc as often as any
other defense to l .e4, about as much as 1. ..eS and somewhat more often than
the Caro-Kann . I would say that Benko made more important contributions to
Pirc theory than in those two openings, and much of what he played is still con
sidered optimal today. This is exemplified by the game before you, Janosevic
Benko, Belgrade 1 964, as well as in our next few games: l .e4 g6 2.d4 Bg7
3 .Nc3 d6 4.f4 Nf6 S .NB 0-0 6.Bd3 Na6

A remarkable move for the time, played at least twice by Benko in 1 964 (I can
see no earlier examples) . It is a striking example of his originality in the opening.
When the game was played, it was thought that keeping in touch with the center
(and particularly d4) by 6. . . Nc6 or 6. . . Bg4 was essential. Developing away from
the center, apparently with greater vulnerability to white pawn advances such as
eS and d S , only caught on much later. To quote Nunn and Crouch: "From small
beginnings in the 1970s (!), 6. . . Na6 has risen to be one of Black's main lines
against the Austrian Attack. " In fact, when you look at their and other Pirc books,
the earliest games come from the mid-to-late '70s, over a decade after Benko's
use. A search of databases finds only 10 or so mostly obscure games from the
1960s, all two years or more after Benko's games from 1 964. For example, Bent
Larsen, always open to unorthodox ideas, tried 6. . . Na6 as far back as 1 966.
Black's idea with 6. . . Na6 is to enforce . . . cS . He is willing to live with an ap
parently shattered pawn structure and the weakness of c6 after Bxa6 because he
gains dynamic play with the two Bishops and b-file. The truly courageous fea
ture of 6. . . Na6, however, is that it places a Knight on the rim and does nothing
immediate in the center against White's eS (as 6. . . cS does) . On the other hand,
Black retains the option of . . . Bg4 and can wait for White to commit himself.
7.0-0 In the same year, Benko encountered the critical line 7 .eS dxeS (7 . . . Nd7
is now the established move) 8 . fxe S NdS 9.Nxd S Qxd S , Bisguier-Benko, New

BENKO AS BLACK

York (Zonal) 1964- another foray into experimental waters. White may have a
slight advantage, but Black has reasonable counterplay. 7 . . . c5 8.e5 8. dS is now
the main line, but Black obtains dynamic counterplay and is holding his own
theoretically. An early example from B enko's practice went 8 . . . Nc7 9 . a4 a6
1O .Qe l Bd7 ! ? ( l 0 . . . Rb8 ! with the idea 11.aS bS 12 . axb6 Rxb6 is more active)
11.aS BbS 12 . Qh4 Bxd3 13 .cxd 3 e6 14. fS ! , with a dangerous attack which won
out, although at this point Black can defend with accurate play, ZuckermanBenko, New York 1968. 8 . . . Ng4 ! This move has scored 11112 out of 13 in my
database ! White's problem is that his center is falling apart, and . . . cxd4 will be
the answer to almost any move, e.g., 9.Bxa6 or 9.exd6. Likewise for kicking the
Knight: 9.h3 cxd4

1 0.Ne4 Not 1O. Nxd4? dxe S , and yet 10 .hxg4? ! dxc3 11.bxc3 Qb6+ 12 . Kh l Bxg4

gives Black an extra pawn and far superior position. 1 0 . . . Ne3 ! I 1 .Bxe3 dxe3
1 2 .Qe2 Qb6. Here Black is obviously much better because, among other things,
he has the two Bishops and an extra pawn.
Another game featuring the popular Austrian Attack (4.f4) is Letzelter-Benko,
Monte Carlo 1 968, in which Black neutralizes White's attack and finds a way
to successfully utilize his own positional advantages: l .e4 g6 2 .d4 Bg7 3 .Nc3
d6 4.f4 Nf6 5.Nfl c5 We saw 5 . . 0-0 6.Bd3 Na6 in Bisguier-Benko above. 6.dxc5
And 6.BbS+ is discussed in the next game. 6 . . . Qa5 7.Bd3 Qxc5 A very early
game with this formation. Yet both Szabo and Pirc had used it the year before,
and Benko undoubtedly noticed those Hungarian games. 8.Qe2
.

509

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

SI0

8 . . . Nc6 S . . . O-O 9.Be3 QaS 1 0.0-0 Bg4 1 1 .Rad 1 Nc6 was the (more accurate)
order of the Spassky-Fischer, Reykjavik 1 972 World Ch. game. Benko and Fischer
worked together, and it's remarkable how often Fischer played a specific opening first tried by Benko. The King's Indian Attack is an obvious example, but I
have pointed out others in various openings. 9.Be3 Qa5 1 0.0-0 0-0 1 0 . . . Bg4
first is better, as was later worked out. 1 1.Kh 1 1 1 .h3 ! is the reason . . . Bg4 was
supposed to have been played first. According to Nunn and McNab, 1 1 . . .Bd7
1 2 .a3 "makes it difficult for Black to create any counterplay. " F. Olafsson-Benko,
Wijk aan Zee 1 969 continued from that point with 12 ... RfcS 1 3 . Qf2

1 3 . . .BeS ! (to protect f7 and clear d 7 . Black may be worse, but a simplistic attack
based upon fS now looks less likely to succeed.) 1 4.fS NeS ? ( 1 4 . . . Nd7 ! is more
logical; e7 can be defended by a queen on dS, and a Knight will occupy eS next.)
1 S . NxeS dxeS 1 6 .Rad 1 e6 ! ? (Something must be done in the face of g4-gS .) 1 7 .fxe6
fxe6 1 S .Qh4! NhS 1 9. Ne2 (or 1 9 .Be2 Nf4 2 0 .Bg4! with the idea 20 . . . hS 2 1 .Rxf4)
1 9 . . . BbS 2 0.Qe7 Bxd3 2 1 .Rxd3 RfS 2 2 . Qxe6+ KhS 2 3 .Rfd 1 RacS 24.Qb3 (24.RdS !)
24 . . . Qa6, drawn. White is still better, of course. 1 1 . . .Bd7 1 1 . . . Bg4 is still preferable, but the text is not bad. 1 2 .Nd4 This doesn't look right. More aggressive
was 1 2 .Bc4. 12 . . . Qh5 !

Here is Benko heading for his usual Queenless middlegame! He is doubtless at

tracted by the Sicilian pawn structure. Satisfactory but perhaps drawish would have
been 1 2 . . . Nxd4 1 3 .Bxd4 Qb4 1 4.Bxf6 Bxf6 I S .NdS Qxb2 1 6.Rab l Qxa2 1 7 .Rxb7.
1 3 .Qxh5 Nxh5 14.Nxc6 Upon 1 4. Nde2 , 1 2 . . . Nb4 wins the two Bishops, which
can be quite significant in this kind of ending. 14 ...bxc6 1 5 .Be2 ! ? rub8 ! ? 1 6.Rab 1

BENKO AS BLACK

Rb7 1 7.BxhS ? White can still equalize with 1 7 .Ba6. 1 7 ... gxhS 1 8.Nd l ? ! A mis-

take, but 1 8 .Bd2 f5 and the Bishops are coming to life. 18 ... Rb4! Winning a pawn.
1 9.Nf2 Bxb2 ! ? 20.Nd3 ? Rxe4 2 1 .Rxb2 Rxe3 , and Black won easily.
Finally, we see that Benko indeed helped to develop the theory of every im
portant Austrian Attack line, including this still-critical one played in Tringov
Benko, Sarajevo 1 967: l .e4 g6 2 .d4 Bg7 3 .Nc3 d6 4.4 Nf6 S .NO cS 6.BbS+
The main option to 6.dxc5 . 6 ... Bd7 7.eS Ng4 8.e6 BxbS 9.exf7+ Kd7 1 O.NxbS
QaS+ I 1 .Nc3 cxd4 1 2 .Nxd4 Bxd4

As in so many openings, Benko was way ahead of the curve for Pirc Defense

theory. In my database, this capture was first played by Benko and Suttles in
1 966 (see the next note). It didn't become popular for several years, and went on
to become one of the main lines of the entire Pirc Defense ! 1 3 .Qxg4+ Now
considered fairly harmless. Zuckerman-Benko, New York 1 966 had seen 1 3 .Qxd4
Nc6 1 4.Qd2 ! ? (1 4.Qc4! replaced this and other retreats such as 1 4.Qd l and 1 4.Qe4
in the mid- 1 970s) 1 4...Rhf8 1 5 .h3 (Another example that has disappeared from
theory. Nunn and McNab give only analysis with 1 5 .Nd5 Qxd2 + 1 6.Bxd2 Rxf7
1 7 .h3 Nf6 =) 1 5 . . .Nf6 1 6.b3 Rxf7 1 7 .g4? h5 1 8 .g5 Ne8 19 .Bb2 Ng7 ! 2 0 .0-0-0
Ne6, and Black was already winning material although, as was his wont, Benko
lost the thread in severe time trouble and only drew. 1 3 . . . QfS 1 4.QO Nc6

I S .Ne2 ! ? A risky move that indirectly attacks b7 via the threat of Nxd4. The

problem is that White will have difficulties castling on either side of the board.
A safer way of protecting c2 is 1 5 .Ne4, when Nunn and McNab give 1 5 . . . Qe6

51 1

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

512

1 6.Be3 ( 1 6.c3 dS) 1 6. . . Bxe3 1 7 . Qxe3 Kc7 I S .Kf2 RafS =. I S . . . Qe6 I S . . . Bb6! was
better, preventing castling and keeping the pressure on. The text allows a trick.
1 6.c3 ? ! My computer correctly suggests 1 6.fS ! gxfS ( 1 6. . . QxfS ? 1 7 . QxfS + gxfS
I S .Bh6!) 1 7.Rfl . Black is in no real danger after something like 1 7 . . . BeS I S .QxfS
QxfS 1 9.RxfS Ke6, but he would have no winning chances. 1 6 . . . B h 6 1 7.Bd2
RafS 1 7 . . . RhfS , anticipating l S .b3 as ! ? , would keep the Queen's Rook in posi
tion to attack White's king after 0-0-0 . The text is also fine. 1 8 . h 3 Preparing to
castle queenside, but also creating weaknesses. 1 8 . . . Rxf7 1 9.0-0-0 hS ! ? Prob
ably thinking about the Benko-like . . . Qg4. 20.h3 ? ! h4 Both fixing White's pawns
and preparing his 2 3 rd move. The opening is over and Black has obvious posi
tional advantages, as was demonstrated by 2 1 .Rhe l Qf6 22 .Be3 Bxe 3 + 2 3 .Qxe3
RhS ! 24.Kh 2 RfS 2 S .RfI

Now White is tied down to defense, but he probably isn't expecting what hap
pens next: 2S ... gS ! Typical Benko ! By hook or crook, he gets the Queens off!
26.fxgS Something like 2 6.g4 hxg3 2 7 .Qxg3 gxf4 2 S . Qf3 is no better, because
the pawns roll after 2 S . . . e S . 26 . . . QeS ! 2 7.QxeS NxeS 2 8 .RxfS RxfS Black has
anticipated this position in which he will gain the seventh rank and successfully
attack White's kingside pawns, as well as having centralized pieces. 29.Rd4 Rf2
3 0.Re4 Ng6, and with . . . dS and . . . Rg2 threatening, Black had a winning advan
tage. This was the consistent result of his opening and early middlegame strat
egy, so I have shown you most of the game.
I already explained in the previous chapter (Benko as White) that in my opin
ion B enko was the first to neutralize one of White's most irritating weapons in
the Pirc. Rather than remove it from that section (it seemed to belong there),
I'll offer up a sparely commented reminder of the game Hug-Benko, Sao
Paulo 1 973 : l .e4 d6 2 .d4 g6 3 .Nf3 Bg7 4.Be2 Nf6 S .Nc3 0-0 6.0-0 Bg4
7.Be3 Nc6 This is the most popular defense to the "Classical" Pirc (Nc3 , Nf3 ,
and Be2). 8.Qd2 eS 9.dxeS For some time, this was thought to lead to a small
but definite White advantage in the ending ... 9 . . . dxeS 1 0.Rad l Qc8 ! But that
was after 1 0 . . . Qxd2 I l .Rxd2 . Benko's simple but easy-to-overlook move, pre
paring . . . RdS with things like . . . Bxf3 and . . . Nd4 to follow, achieves complete
equality. He gets no credit in the books. I I .h3 I I .BgS Bxf3 1 2 . Bxf3 N d4 = , was
Weinstein-Benko, USA 1 974. 1 1 . . .Rd8 1 2 .Qc1 Bxf3 1 3 .Bxf3 Nd4 1 4.Bxd4

BENKO AS BLACK

exd4 1 5 .Nd5 Nxd5 1 6.exd5 Qf5, and the players agreed to a draw shortly

thereafter.

Caro-Kann Defense
In addition to the Sicilian and Pirc Defenses, Benko has often played the Caro
Kann, a defense befitting his positional style. In my database (severely limited,
remember, by the fact that so many Benko games in the U. S . were never saved),
Benko plays 2 7 Caro-Kanns and as always, he jumps from system to system.
After 3 .Nc3 , for example, he played 3 . . . g6 before it became popular. Similarly,
after 3 . . . dxe4 4.Nxe4, Benko had many games with the now fashionable 4 . . . Nd7,
going as far back as 1 963 . Here are examples of the latter move, including this
one versus Suetin at Bad Worishofen 1 992 : 1 .e4 c6 2.d4 d5 3 .Nc3 dxe4 4.Nxe4
Nd7 5.Bc4 Ngf6 6.Ng5 Still the main move today. 6 . . . e6 7.Qe2 Nb6 8.Bb3 A
frequent choice by White, although 8.Bd3 is the current preference. 8 ... h6 9.N5f3
c5 A certain neglect of history accompanies McDonald's designation in 2 000 of
9 . . . aS as "Karpov's move" and citing the "stem game" as Karpov-Kasparov, Linares
1 994. In fact, both Smyslov and Bronstein played it as early as 1 961 , with over
2 00 games (many by top grandmasters) preceding the K-K game. Benko himself
used 9 . . . aS versus de la Heras at Buenos Aires 1 965 : 1 0.a3 a4 I l .Ba2 Be 7 ( l l . . .cS
has been successful here) 1 2 .Be3 ( l 2 .Bd2 cS 1 3 .dxcS Nbd7 1 4.NeS NxcS =,
Marsalek-Smyslov, Oberhausen 1 961 was perhaps the top-level introduction of
9 . . . a S .) 1 2 . . . NbdS 1 3 .Bd2 bS 1 4.NeS Qb6 I S .Ngf3 Ba6 1 6.0-0 cS =.

1 0 .Bf4 Spassky-Benko, Amsterdam 1 964 saw Black equalize in the opening:

1 0. B e 3 Qc7 I l .NeS Bd6 12 .Ngf3 0-0 1 3 .0-0 Nbd S 1 4.c3 b6 I S . Rad l Bb7
1 6.Bc l ! ? Rad8 with no problems, although Benko later allowed the future world
champion too many opportunities and lost. 10 . . . a6 ! ? Preparing a very aggres
sive and in some ways atypical Benko pawn sacrifice: 1 1 .0-0-0 c4! 1 2 . Bxc4
Nxc4 13 .Qxc4 b5 14.Qc7 After 1 4.Qd3 ? ! Bb7, Black dominates dS and the
light squares, possesses the Bishop pair, and has the c-file to pressure White's
King. 1 4 . . . Qxc7 1 5 .Bxc7 Bb7

513

PAL BENKO: MY L IFE, GAMES AND COMPOSITIONS

51 4

What a surprise: a positional Queenless middlegame! Black has the advantages

just mentioned, and White has a pawn. But White also has the problem of de
veloping his king's Knight without allowing . . . Bxf3 . 1 6.Ne l ! ? 1 6.Ne2 Rc8 1 7 .Be5
Be7 still leaves White tangled up. 1 6 . . . RcS 1 7.Bg3 Be7 l S.Ngf3 0-0 1 9 .Ne5
RfdS 20.f3 To activate the e 1 -Knight. 20 . . . Nd5 2 1 .Bfl Bg5 + 22 .Kb l f6 After
all of White's efforts, his only good piece is driven away. 2 3 .N5d3 b4! ? Another
method was 2 3 . . . Be3 24.Bxe3 Nxe3 2 5 .Rd2 Rxd4, with at least equality.
With n . . . b4, Black has prevented 2 4.Nc 5 ? due to 2 4 . . . Rxc5 ! 2 5 .dxc5 Nc3 +
2 6.bxc3 Rxd 1 + 2 7 .Kb2 bxc 3 + 2 8 .Kxc3 Bf4 with a devastating attack. As the game
went, White managed to hang on with good defensive moves and achieve the
draw, but this was a fine active effort against a strong player.
Benko played the main line Caro-Kann with 4 . . . Bf5 upon occasion, mostly
unambitiously, content to draw against top grandmasters (An exception was
Bisguier-Benko, New York (4) 1 964, for which see Game 5 3 .) . Here are a few
quick snippets from the main line, all ending in draws : l .e4 c6 2 .d4 d5 3 .Nc3
dxe4 4.Nxe4 Bf5 5 .Ng3 Bg6 6.Nf3 6.h4 h6 7 .Nf3 Nd7 8 .h5 Bh7 9.Bd3 Bxd3
1 0. Qxd3 Qc7 1 1 .Bd2 Ngf6 1 2 .0-0-0 e6 1 3 .Ne4 0-0-0 1 4.g3 Nxe4 1 5 . Qxe4 Bd6,
Gheorghiu-Benko, Palma de Mallorca 1 968. See my notes regarding this open
ing variation in Game 69, which is placed in the section on Benko's openings as
White. 6 . . . Nd7 7.Bd3 e6 S.O-O Ngf6 9.Re l

9 . . . Be7 1 0.b3 1 O.c4 0-0 1 1 .b3 Qa5 1 2 .Bxg6 hxg6 1 3 .Qe2 b5 1 4.Bd2 Qa6 1 5 .c5

Qb7 1 6 .h4 Rfe8 1 7 .Bg5 Bf8 1 8 .Rac 1 Nd5 , with a permanent outpost and in-

BENKO AS BLACK

tending . . . f6 and . . . e 5 , although White should still have some advantage, Evans
Benko, U.S. Ch. 1 96 1 . 1O ... Bxd3 I 1 .Qxd3 0-0 1 2 .Bb2 ReS 1 3 .c4 Qc7 1 4.Rad l
RadS 1 5 .Qc2 BfS White may have a litle something here, but Benko frees his
game and achieves a quick draw: 1 6.h3 h6 1 7.Nfl c5 I S.Ne3 cxd4 1 9.Bxd4 a6
20.Qb2 b6 2 1 .b4 a5, liz _ liz , Keres-Benko, Wijk aan Zee 1969.

French Defense
The French Defense wasn't very nice to Benko, although he used it from time
to time throughout his career. In the Tarrasch Variation (3 .Nd2), he lost games
to both Tal and Keres, and versus 3 .Nc3 he lost to the same players ! His other
results were solid, but not inspiring. Stein-Benko, Stockholm Interzonal 1 962
is an amusing game in which Benko takes his positional style into hypermodern
territory strongly reminiscent of Nimzowitsch, Petrosian, and Bronstein. I quote
most of the game because the opening strategies extend well into the game: l .e4
e6 2 .d4 d5 3 .Nc3 Nc6 ! ?

A favorite o f the German player Reefschlaeger-I once wrote an article about

this variation and found it to be surprisingly playable . 4.NfJ Nf6 5.e5 5 . exd5
exd5 6.Bb5 was Fischer's solution in a famous Candidates Match win versus
Petrosian, but the game was easy to improve upon and I don't consider this a
critical test. 5 .Bd3 and 5 .Bg5 are the two other challenging continuations. 5 ...Ne4
6.Bd3 Nxc3 ! ? I think that 6 . . . Bb4, as Ulf Andersson once played, is a better
move. But that is open to dispute. 7.bxc3 Be7 S.h4 h6 Black's last two moves
are probably necessary, to prevent Ng5 and Qg4 or Qh5 . 9.h5 Na5 ! Showing
good instincts. Developing by . . . b6 or . . . Bd7 is delayed to clamp down on c4.
1 0.Nd2 A very natural response, both contesting c4 and preparing Qg4. Rather
that block off the c1 Bishop, however, White might consider l O .Rh3 first, to
massage the kingside after Rg3 . Still, the move . . . Bf8 would batten down the
hatches and White's plan would not self-evident. 1 0 . . . c5 I 1 .Qg4 And here
I I .O-O ! ? would be a paradoxical move, intending the simple f4/g41f5 idea. 1 1 ...BfS

515

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

51 6

Backwards, ever backwards ! Black has only one piece out, and it's a Knight on
the edge of the board! 1 2 .Rh3 ! ? This all seems so active, but soon White lacks a
plan. Maybe 1 2 .Ba3 or 1 2 .dxc5 could be tried, in the latter case to open lines.
12 ... c4! A move that is very often bad in the Winawer, since Black no longer has
queenside play via moves like . . . Rc8 , . . . cxd4, . . .Nc4 and the like. But see the note
to Black's 1 4th . This may be the best move of the game. n .Be2 Bd7 1 4.a4 To
prevent . . . Ba4 and in some cases prepare Ba3 . 1 4 . . . Nc6

Another retreat. At first sight, . . . c4 doomed Black's d7-Bishop to permanent pas

sivity; but notice how (by foreclosing White's dxc5) it cuts off all of White's de
veloped pieces from the queenside. In the meantime, White is not well placed to
enforce a kingside pawn advance by f4-f5 or g4-g5 . In fact, White is beginning
to regret foregoing 0-0. 1 5 .Rf3 Qa5 1 6 .Qf4 Nd8 These retreats look silly, but
Black has ideas of . . . Bxa4 or . . . b5 and is starting to get the advantage! Ingenious
play. 1 7.Rg3 b5 ! ? 1 7 . . . Bxa4 ( ! , I think) looks fully playable, since tactical ideas
like 1 8 .Nb 3 Qb5 and 1 8 .Bxc4 fail. A possible follow-up might be . . . Qb5-d7 and
eventual advance of the queenside pawns. Benko has another reasonable idea, to
advance by . . . b4, although of course that will also allow White's kingside pieces
to finally coordinate with his queenside. 1 8 .Qe3 Nc6 A tacit draw offer in view
of 1 9.Qf4 Nd8 etc.; Black is understandably afraid to open the position without
supporting pieces. 1 9.f4? ! Maybe he should take the draw! Benko's coming breakthrough will justify his entire opening strategy. 1 9 . . . Rb8 20.Kf2 Another try
was 2 0.Bd l b4 2 1 .cxb4 Nxb4 2 2 .Kfl , about equal. 20 . . . b4 2 1 .Bb2 Qc7 22.cxb4?
Better 2 2 .Kg l . 22 . . .Nxb4 Now White has major problems. 2 3 .Qc3

BENKO AS BLACK

51 7

2 3 . . . Nc6 Probably 2 3 . . . Qa5 ! was still better, preventing Ba3 . Benko seems to

have a considerable advantage at this point, but he has trouble converting. Time
pressure? 24.Ba3 Bxa3 25 .Rxa3 Qb6 Hitting the d-pawn and contemplating
. . . Qb2 . 26.NB 0-0 After all that time ! 2 7.Rh3 ! And not 2 7 .Qe3 ? Ne7 . 27 . . . Qb2
2 S.Rhl Preventing . . . Qc l . Stein manages to resist successfully for the moment,
but Black is still in control. 28 . . . Qxc3 29.Rxc3 Rb2 30.Ra l rubS 3 1 .Ke3 Na5
3 2 .Ne l ! 3 2 .Nd2 RSb4 keeps the pressure on. Now White can play R3a3 and c3
in some lines. 32 . . . Rb l H .Rca3 Rxa l 3 4.Rxa l Rb4 3 5 .Kd2 Kf8 ! ? Benko's
easiest course was to simply grab the a-pawn. After 3 5 . . . Rxa4, he was probably
afraid of 3 6.Rb l , but 3 6 . . . KfS ! 3 7 .RbS+ Ke7 is good, because 3 S.RgS Bb5 ! threatens
. . . c 3 + and poses White too many problems. 3 6 .Kc3 Rb8? A mistake near the
time control. Still winning a pawn was 3 6 . . . Nc6 ! with the idea 3 7 .a5 Rb5 . Fol
lowing 36 . . . RbS , both sides made mistakes in the next few moves heading for
the time control. This intriguing game was drawn on move 42 in an obscure
position, perhaps still favoring Benko.

Dou ble King Pawn

Benko played double king pawn openings ( 1 .e4 e5) fairly often, but mostly in
the earlier days of his career. In my database, 1 . . . e5 appears roughly as often as
the Pirc and 1 . . .c6 (S-9% of the time), although I suspect that the latter two
moves would be better represented if the selection included more Swiss System
tournaments. For the record, non-transpositional games with 1 . . .g6 accounted
for another 4. 5 % of Benko's games versus l .e4, and he very rarely dabbled in
other replies, so you can see what a dominant factor the Sicilian Defense was in
his repertoire with Black.
Benko liked the Petroff Defense ( l .e4 e5 2 .NB Nf6), which he used to draw
various highly ranked players. More often, however, he played 2 . . . Nc6. He handled
the less mainstream openings such as the Four Knights and irregular Ruy Lopez
variations well, but in fairly standard fashion. Interestingly, he played several
early games with the Berlin Defense (2 .NB Nc6 3 . Bb5 Nf6), now all the rage.
However, the one time that his opponent played 4.0-0 Nxe4 5 .d4 Nd6 6.Bxc6
bxc6 7 . dxe5 Nf5 S . QxdS+, as in several games of the 2 000 Kramnik-Kasparov
World Ch match, Benko lost. This was against Panno in the important Portoroz

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

51 8

Interzonal of 1 9S9 ; and it may have put him off 3 . . . Nf6 . His real 1 . . . eS Achilles
heel was the main line Ruy Lopez (3 .BbS a6 4.Ba4 Nf6 S .O-O Be7 6.Re l bS),
with which he lost to Smyslov, Fischer, Gligoric, and Keres. An exception was
the following contest, E Olafsson-Benko, BledlZagreblBelgrade 1 9S9 (Can
didates), which anticipated modern theory: l .e4 eS 2 .Nfl Nc6 3 .BbS a6 4.Ba4
Nf6 S .O-O Be7 6.Re l bS 7.Bb3 0-0 S.c3 d6 9.h3 Bb7 Actually, this is the
wrong move order if one wants to play . . . Nb8, as Benko does next. Normal is
9 . . . Nb8 . In this game, White doesn't exploit the inversion of move order and
goes into the main line . See the note to move eleven. 1 0.d4

1 0 . . . NbS ? ! This unnatural but logical retreat, which should have been played
via 9 . . . Nb8 1 O.d4 Nbd7 when . . . Bb7 (see the Keres-Benko game below) trans
poses in this game to the popular "Breyer Defense," named after the creative
Hungarian player and writer of the early part of the twentieth century. This
variation was revived, fairly modestly, in the early and mid-S Os, with players such
as Spassky, Petrosian, and Portisch (another Hungarian!) leading the way. In the
late 1 960s and 1 970s it was probably the main line of the Ruy Lopez. The idea is
that Black's Knight is poorly placed on c6, blocking . . . cS and masking the b7Bishop. Moving the Knight to as achieves the same thing, but at the cost of
placing it on an awkward square. Instead, the Knight will go to d7 to support eS
and keep the Bishop on b7 unobstructed. 1 1 .Nbd2 ? ! Now we're back in the
main line. Keres punished B enko for his choice of move order in the same 1 9S9
Candidates tournament: l 1 .dxe S ! dxe S ? (Oddly enough, this seems t o be a mis
take; better is 1 1 . . .Nxe4 1 2 .BdS ! ? or 1 2 .Bf4 ! , although White has a serious ad
vantage.) 1 2 . Qxd8 Bxd8 1 3 .NxeS Nxe4 1 4.Be3 !

BENKO AS BLACK

( 14.Bc2 is also awkward for Black, but 1 4.Be3 is even better, covering c5 .) 1 4 . . . Bf6
( 1 4 . . . Nd6? 1 5 .Bc5; 1 4 . . . c5 1 5 .f3 c4 1 6.Bxc4!) 1 5 .Ng4 Nd7 ( 1 5 . . . Nc6 1 6.Nd2 !
Nxd2 1 7 .Nxf6+) 1 6 .Nd2 Nxd2 1 7 .Bxd2 RfeS (What else? 1 7 . . . c5 l S .Nxf6+ Nxf6
1 9.Re7) 1 8 .Bf4 Rxe l + ( I S . . . c5 1 9.RxeS+ RxeS 2 0 . Rd l .) 1 9.Rxe 1 RcS 2 0 . Bc2 g6? ?
(2 0 . . . h5 2 1 .Bf5 ! hxg4 2 2 .Bxd7 nets a pawn and the game.) 2 1 .Rd 1 , 1 -0. An em
barrassing defeat for Benko, but in retrospect the opening was pretty awful after
1 0 . . . NbS . 1 1 . . .Nbd7 1 2 .Bc2 Re8 l 3 .Nfl BfS 1 4.Ng3 g6

There had been just a few games from this position in the two preceding years,
and only one of note - Sanguinetti-Panno, Buenos Aires 1 968. Subsequently this
became a very important position for Ruy Lopez theory. 15.h4 ! ? Today, moves
such as 1 5 .a4 and 1 5 .b3 are played instead. 1 5 . . . d5 ! ? A radical solution; 1 5 . . . Bg7
1 6.h5 d5 ! is more accurate, as played by Tukmakov almost 2 0 years later. 1 5 . . . c5
is also equal. 16.dxe5 Here 1 6 .Bg5 h6 1 7 .Bxf6 Qxf6 equalizes. The real problem with 1 5 . . . d5 is 1 6.Nxe 5 ! , although it was very hard to assess in advance. One
line goes 1 6 . . . Nxe5 1 7 . dxe5 Rxe5 ( 1 7 . . . Nxe4 l S .Nxe4 dxe4 1 9 . QxdS RaxdS
2 0 .Bg5 !) l S .Bg5 ( l S.f4? ReS 1 9 .e5 Nd7) l S . . . h6 1 9 .Bf4 Re6 2 0 .e5 Nd7 2 1 .Qg4
Bg7 (apparently equalizing) 2 2 .h5 ! g5 2 3 .Nf5 ! , with a clear advantage in view of
Nxg7 or Nd4 and e6. 16 ... dxe4 1 7.Nxe4 Nxe4 1 8.Bxe4 Bxe4 1 9.Rxe4 Nxe5
Now Black stands very well indeed. He retained his advantage after 20.Bg5 Or
2 0.Rd4 Nxf3 + 2 1 . Qxf3 Re 1 + . 20 ... Qxd 1 + 2 1 .Rxd l Nxf3 +, and went on to win
a protracted struggle.

Sicilian Defense
Benko's favorite defense to 1 .e4 was 1 . . . c5 , and he remained loyal to it throughout
his career. As might be expected, he explored a wide variety of Sicilian Defense
variations including the Scheveningen, Rauzer, Kan, Taimanov, and Accelerated
Fianchetto (via the English Opening). In this section, we will delve into the theory
of the Sicilian Defense as B enko played it, keeping in mind that we can cover
only a small fraction of his games with it and even less of the massive published
theory on each variation.
To launch our discussion of 1 .e4 c 5 , let's start with the Sozin Variation, and
look at a few games with a move that is named after Benko himself. I will orga
nize the material around the game Cardoso-Benko, Portoroz (Izt) 1 95 8 : l .e4
c5 2 .Nf3 Nc6 3 .d4 cxd4 4.Nxd4 Nf6 5 .Nc3 d6 6.Bc4 Qb6

51 9

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

520

This is Benko's own system against the dangerous 6.Bc4. Among his numerous
contributions to Sicilian theory, 6 . . . Qb6 has to be one of the most important.
Particularly in the 1 970s, quite a few years after Benko's advocacy began, 6 . . . Qb6
caught on like wildfire, and it has been employed regularly by top players ever
since. Benko's simple idea is that after 7 .Nb3 and Black's later Queen retreat to
c7, Black has spent two moves developing his Queen to c7; but White's Knight
isn't well-placed on b3 , and if it returns to d4 at some point, White has actually
lost a tempo on a normal line ! As far as I know, Benko first played 6 . . . Qb6 against
Honfi in 1 95 7 ; regardless, his many games with the move earned it the name
"Benko's Variation. " I will try to show, with brief comments, a series of Benko
games in this line that are representative of his play. 7.Nb3 Here are few other
examples from other Benko games: 7 .Nde2 (7 .Ndb5 used to be played a lot, but
is to be well-answered by 7 . . . a6 8.Be3 Qa5 9.Nd4 and now 9 . . . Ne 5 , as played by
Kramnik among others, seems to be scoring well, whereas the solid 9 . . . e6 is also
fine.) 7 . . . e6 8 .0-0. A common starting position. I won't pretend to cover theory
exhaustively, but hope to illustrate characteristic ideas:
(a) What is possibly the 6 . . . Qb6 stem game in international play went 8 . . . a6
(Interestingly, Benko made a point of avoiding this move in later 7 .Nde2 games.)
9.h3 Be7 1 O .Be3 ! Qc7 ( 1 O . . . Qxb2 I l .a3 !) I l .Bb3 b5 1 2 .Ng3 Bb7 = , Honfi-Benko,
Budapest 1 95 7 . Black has perfectly good play typical of the Sicilian, and the Knight
on g3 is not well placed.
(b) 8 . . . Be7 9.Bb3 0-0

Here are some of Benko's experiences with this position:

BENKO AS BLACK

(b l ) 1 O.Qd3 NaS ? ! ( l 0 . . . Bd7 ! I l .Be3 Qc7 intends 1 2 .NbS ? ! Qb8 1 3 .Rad l ? ?

NeS) 1 1 .Be3 Qc7 1 2 .Nb S ! Qb8 1 3 .Bf4! Rd8 1 4.Rad l Ne8 I S .eS ! ? d S 1 6.c3 Nxb 3
1 7 .axb 3 , and White is for choice, Ljubojevic-Benko, New York 1 98 5 ;
(b2 ) 1 O.Ng3 NaS 1 1 .Kh I B d 7 1 2 .BgS h 6 1 3 .Bxf6 Bxf6 1 4.NhS B e 7 ( 1 4 . . . Bxc3
I S .bxc3 QcS ! looks good for Black, e.g., 1 6.f4 BbS !) I S .f4 Nxb3 1 6. Qg4 g6
1 7 .axb3 Kh7 1 8 .Ng3 fS ! , opening lines for Black's Bishop pair, Renner-Benko,
Augsburg 1 986;
(b3) 1 O.Kh I NaS 1 1 .BgS QcS ! (hitting gS) 1 2 .f4 ( 1 2 .Be3 Qc7 reroutes the Queen
with an extra tempo in some lines due to BgS -e3 .) 12 ... bS with full equality, Fischer
Benko, Bled/Zagreb/Belgrade 1 9S9 (Candidates) . Black need only have played
. . . Nxb 3 and . . . Bb7 next, but got too fancy and fell for an eS trick. 7 . . . e6 8.0-0
Another example was 8.Be3 Qc7 9.f4 Be7 Browne-Benko, Aspen 1 968, when
Black had transposed into a well-known variation of the Scheveningen Sicilian.
In that line, 9 . . . a6 1 O.Bd3 bS is a viable alternative examined in ECO. 8 . . . Be7

9.Be3
(a) Parma-Benko, Sarajevo 1 970 saw 9.Bd3 a6 1 0.a4 NaS I l .Be3 Qc7 1 2 .NxaS
QxaS 1 3 .Bd2 Qc7 1 4.aS Bd7 I S .Qe2 Bc6, with equality.
(b) 9.BgS 0-0 1 O .Bxf6 ! ? Bxf6 I l .Qxd6 Rd8 1 2 .QcS ? ! (It's probably better to
avoid these complex Queenless positions against Benko !) 1 2 . . . QxcS 1 3 .NxcS b6
gives Black the Bishop pair in return for giving up a pawn: 1 4.N d3 ( 1 4.Nb3 is
the best that White can do, but he still has serious problems, e.g., 14 . . . Bb7 I S .f4
Bxc3 1 6.bxc3 Rac8 1 7 .Bd3 e S ! 1 8 .fS ! ? Nb8 1 9.c4 Nd7 , and the c-pawn is too
weak) 14 ... Bb 7, and now Black clearly has more than enough compensation, Hoyos
Millan-Benko, New York 1 99 1 ;
(c) 9.Kh l a6 1 0.f4 Qc7 I l .Bd3 b S is another known Scheveningen position
with Black nicely placed, Klundt-Benko, Augsburg 1 986. 9 ... Qc7 1 O.f4 0-0 Later
it was found that 1 O . . . a6 l 1 .a4 b6 1 2 .Bd3 Bb7 is probably better. 1 1 .g4? ! This
overextends. Best was I 1 .Bd3 . l 1 . . .dS ! 1 2 .Bd3 Or 1 2 .exd S Nb4! . 12 . . . dxe4
1 3 .Nxe4 NdS 14.Bd2 as ! Also good is winning the bishop pair by 14 . . . Ncb4.
l S .c3 Not much better is I S .c4 Ndb4 1 6.Be2 a4 ! . l S . . . Rd8 1 6.gS ? ! a4 1 7 .Nc 1
Qb6+ 1 8.Khl Ne3 1 9.Bxe3 Qxe3 20.Qf3 What else? But we all know by now
about taking the Queens off in a Benko middlegame . 20 . . . Qxf3 + 2 1 .Rxf3 b6
22 .Bc2 Bb7, and White can't defend the long diagonal. Black won easily.

52 1

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

522

Before discovering 6 . . . Qb6, Benko played 6 . . . e6 versus the Sozin with some
success. As pointed out in B enko's notes to the game Szabo-Benko, Hungarian
Ch. 1 95 1 , it was only after the Velimirovic Attack (7.Be3 and 8 . Qe2) began to
win so many games that Benko began to search for an alternative. Since some
top players now feel that there are adequate answers to the Velimirovic, these
earlier Benko contests might be of renewed interest: l .e4 c5 2 .NB Nc6 3 .d4
cxd4 4.Nxd4 Nf6 5 .Nc3 d6 6.Bc4 e6 7.0-0

7 ... Be7 7 . . . a6 8.Bb3 Qc7 is a type of Bc4 Naj dorf that looks a little slow for
Black. Nevertheless, Tapaszto-Benko from the 1 954 Hungarian Ch. proceeded
interestingly with 9.Be3 b5 1 O.a3 ? ! ( 1 0.Nxc6 ! Qxc6 1 1 .Re 1 100ks strong, for ex
ample, 1 l .. .Bb7 1 2 .Bd4! threatening Nd5 and/or Bxf6 with Qh5 to follow) 1 O . . .Na5
1 1 .f4 Be7

1 2 .g4? ! (but 1 2 .Ba2 Nc4 and 1 2 .f5 Nxb3 1 3 .cxb3 0-0 aren't terribly inspiring
either) 1 2 .. .Nxb3 1 3 .cxb3 Bb7 1 4.Rc 1 Qd7 1 5 . Qf3 Rc8 ! ? (This has a tactical
idea seen in the next note. The more straightforward 1 5 . . . d5 1 6.e5 Ne4 would
be very comfortable for Black) 1 6.f5 ( 1 6.g5 Rxc3 ! and . . . Nxe4 with more than
enough for the Exchange) 1 6 . . . e5 1 7 .Nde2 h6 1 8 .M and although 1 7 .. .Nh7 was
fine, this was a good moment for 1 8 . . . d5 ! 1 9.Nxd5 ( 1 9.exd5 Nxd5) 1 9 . . . Nxd5
2 0.Rfd 1 Rxc 1 2 1 .Nxc 1 Bxh4, etc. 8.Be3 0-0 9.Bb3 This is the old main line
Sozin Sicilian, still debated today. As one might imagine, the theory on it was
quite undeveloped in 1 95 1 . 9 ... Bd7 1 0.f4 Planning f5 . The modest 1 O.Qe2 is
also played. 10 . . . Nxd4 I 1 .Bxd4 Bc6

B ENKO AS BLACK

523

1 2 .Qd3 ? ! A move that exposes the Queen to . . . Bxe4 in some lines. As subse
quently confirmed, a more accurate move is 1 2 .Qe2 - see the next two notes.
1 2 . . . b5 Once again, we see a new solution by Benko that is still approved by
theory to this day. In this exact position, at least according to my database, this is
the first game in which . . . b5 was used. But Benko's complaint that Geller got
credit for it later is not that relevant, because Geller's 1 954 game as Black against
Nezhmetdinov went 1 2 .Qe2 b5 instead. After our game's 1 2 .Qd3 b5 1 3 .Nxb 5 ,
1 3 . . . Bxe4 wins the crucial e-pawn with tempo, a key difference. Even Fischer
felt that 1 2 .Qe2 b5 1 3 .Nxb5 was worth playing for White, and he won with it
versus Saidy in 1 9 5 7 . And as for the general plan of . . . b5 in such a position, it is
so obvious that really no player can claim it as "his" idea. 1 3 .Nxb5 ! ? See the last
note. Better according to theory is 1 3 .e5 dxe5 1 4.fxe5 Nd7 1 5 .Ne4 Bxe4 1 6.Qxe4
Nc5 , with equality.

1 3 . . . Bxe4 Notice that 1 3 . . . Bxb5 1 4.Qxb5 Nxe4 (the best line for Black if the
Queen were on e2) is double-edged after 1 5 .f5 . But eventually Black found that
1 5 . . . e5 1 6.Be3 Bg5 ! sufficed for a practical equality. 14.Qe2 Qd7 ? ! In order to
forestall c4, Chekov suggests 1 4 . . . a6 ! , e.g., 1 5 .Nc3 Bc6 1 6.Rad l Qc7, with easy
play. 1 5 .c4! d5 ? ! Benko: "This seemed good at the time, but it only took a couple
moves for me to realize that I'd done something wrong. Better was 1 5 . . . Bc6 . "
This i s true, although something like 1 5 . . . Bc6 1 6.Nc3 Q b 7 1 7 . Bc2 would also
keep some pressure on Black. 1 6.Rad l Qb7 1 7.cxd5 Bxd5 1 8.Bxd5 Nxd5 1 9.f5
The opening is over and "White has dangerous initiative, " as Chekov says. Al
though standing worse here, Benko managed to create some chaos and eventu
ally triumphed after a fascinating struggle. See Game 1 7 .

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

524

A Benko favorite in the Sicilian Defense Rauzer Attack is illustrated by his

game versus R. Byrne at New York (U. S . Ch.) 1 969: l .e4 c5 2 .Nf3 d6 3 .d4
cxd4 4.Nxd4 Nf6 5 .Nc3 Nc6 Often called the "Classical Sicilian," this has been
a great specialty of Benko's. It's worth noting how many and varied early Sicilian
move orders B enko employed. The order before us often transposes to a
Scheveningen Sicilian ( . . . d6 and . . . e6), but so can 2 .ND d6 3 .d4 cxd4 4.Nxd4
Nf6 S .Nc3 and now S . . . e6 or S . . . a6, both orders played by Benko. Another sequence he often used was 2 . . . e6 3 . d4 cxd4 4.Nxd4 and now 4 . . . a6 or 4 . . . Nc6,
both of which can go their own way into Taimanov or Kan Sicilians, or they can
again transpose to a Scheveningen. And there are further examples; this experi
mental attitude is in sharp contrast to today's Sicilian specialists. 6.Bg5 The Rauzer
Attack, still an extremely popular variation for White. 6 . . . Bd7

Benko helped to develop the theory of this move, not only here but also in simi
lar positions of the Sicilian. (Not surprisingly, Benko also used the standard 6 . . . e6
with reasonable success. A few examples will follow in the next game.) 7.Qd2
Snippets from two Benko games involving 7 .Bc4: 7 . . . QaS 8.Bxf6 gxf6 9.Bb3 (9.Nb3
QgS 1 O .g3 fS ! 1 l .f4 Qg6 1 2 .BdS Bg7 , with excellent play, Paoli-Benko, Reggio
Emilia 1 97 0) 9 . . . Bg7 1 0.0-0 0-0 1 1 .NdS Rae8 1 2 .c3 Nxd4 1 3 .cxd4 fS ! ? 1 4.exfS
BxfS I S . QhS e6 1 6 .g4? ! Qd2 ! ( l 6 . . . Bg6 ? ? 1 7 .Ne7+) 1 7 .gxfS exdS 1 8 .Rad l Qe2
with some edge, Huguet-Benko, Malaga 1 970. Benko had wonderful instincts in
the Sicilian and was more at home with it than any other Black opening. 7 ...Nxd4
7 . . . a6 8 . 0-0-0 bS is another order Benko experimented with, e.g., 9.a3 e6 1 0 .f4,

B ENKO AS B LACK

and we have a position that can easily transpose into a traditional Rauzer with
the passive a3 in, e.g., 1 O . . . Qb6 ( l 0 . . . Rc8 is less flexible but also playable: I l .ND ! ?
-[I l .Kb l ] 1 1 .. .b4 1 2 .axb4 Nxb4 1 3 .Bxf6 gxf6 1 4.Kb l QaS I S .Nd4 Bh6 ! ? 1 6 .Nb3
Qb6 1 7 .g3 Ke7 , with double-edged play, F. Anderson-Benko, Tel Aviv 1 964.)
I I .Bxf6 gxf6 12 .Nxc6 Bxc6 1 3 .fS hS 1 4.Qd4? (One doesn't want to volunteer for
a Queenless middlegame against Benko.) 1 4 . . . Bh6+ I S .Kb l Qxd4 1 6.Rxd4 Bf4
(Two Bishops, the lovely eS square, pressure down open files and versus the e
pawn- Black is arguably already winning!) 1 7 . Rd 1 BeS 1 8 .fxe6 fxe6 1 9.Be2 Ke7
2 0.BD h4 2 l .h3 Bxc3 22 .bxc3 Rac8, and White's position is hopeless, HartmetzBenko, Badenweiler 1 98 5 . 8.Qxd4 QaS 9.Bd2 Qc7 1 O.Bc4 1 O .NdS is discussed
in the game notes. lO ... e6 l l .Bb3 Be7 l 2 .f4? ! 0-0 1 3 .Rfl Benko points out
1 3 .0-0? dS ! . 1 3 ... bS ! . The opening has turned out well for Black- see Game 90
for Benko's skillful conclusion.
Benko very seldom limited himself to only one setup in a given variation, as
shown in the next game: l .e4 cS 2 .Nfl Nc6 3 .d4 cxd4 4.Nxd4 Nf6 S .Nc3 d6
6.BgS e6 The main line of the Rauzer. For many examples of Benko's successful use
of 6 . . . Bd7, see the previous game. His play with 6 . . . e6 was also impressive. 7.Qd2

7 ... Be7 7 . . . a6 8 .0-0-0 is a popular Sicilian position. Benko's early games are still

relevant for modern theory: 8 . . . Bd7 (8 . . . h6 9.Bf4 ! ? Bd7 1 O .Nxc6 Bxc6 1 1 .3 dS

1 2 . Qe 1 Bb4! 1 3 .a3 BaS 1 4.b4 Bb6 l S .exdS BxdS 1 6 .Nxd S , '12_ 112, Zuckerman
Benko, Chicago 1 974) 9.f4 h6 1 O.Bxf6 ( 1 0.Bh4 Nxe4 1 l .Qe 1 Nf6 1 2 .NfS QaS
1 3 .Nxd6+ Bxd6 1 4.Rxd6, Szily-Benko, Budapest 1 947 is an approximately equal
position that has occurred many times and was popular in the 1 9S 0s.) 1 O . . . Qxf6

525

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

526

I I .Nxc6 Bxc6 (This is also frequently playe d , and is again about equal)
1 2 .g3 0-0-0 1 3 .Qe3 Qe7 ! ? 1 4.Qa7 Qc7 1 5 .Bh3 Bd7 1 6.e5 Qc5 ! 1 7 . Qxc5 + dxc5 ,
and Black held his own in Dely-Benko, Hungary 1 954. S.O-O-O 0-0 9.f4 These
older positions with . . . 0-0 and without . . . a6 are considered better for White, but
they served Benko reasonably well before the theory developed. 9 ... Nxd4 1 0.Qxd4

1 0 . . . h6 I 1 .Bh4 Qa5 1 2 .e5 1 2 .Bc4! creates the most problems for Black, as be

came clear over the years. Thus we have seen the second player turning to other
move orders . Rossetto-Benko, Portoroz 1 9 5 8 saw 1 2 .Be2 ? ! e 5 ! 1 3 .fxe5 dxe5
1 4.Qd3 Be6 1 5 .Qb5 Qxb5 1 6.Bxb5 Rfd8, and Black's pieces were better placed,
although the game was drawn. 1 2 . . .dxe5 1 3 .Qxe5 Qxe5 1 4.fxe5 Nd5 1 5 .Bxe7
Nxe7 1 6.Bd3 An extremely popular position in the 1 950s and early 1 960s. 16 ... b6
1 7.Be4 RbS I S .Rhe l Bb7 1 9.Rd7 Bxe4 20.Rxe4 Nc6 2 1 .Nh5 RfdS 22 .Rc7
RbcS =, according to ECO. 2 3 .RxcS RxcS 24.Rc4 Ne7 2 5 .RxcS+ NxcS, and
Black had no problems, Gligoric-Benko, Yugoslavia (Candidates) 1 959. This
is the game usually cited as the model for Black to equalize versus 1 2 . e 5 .
The games that are discussed i n the following example are Sicilian Defense
Maroczy Bind Variations. For Benko, this always occurred via the English open
ing- he didn't allow it versus any Sicilian move order, for example, l .e4 c5 2 .NB
Nc6 3 .d4 cxd4 4.Nxd4 g6 5 .c4 Bg7 etc. Nevertheless, the games contribute to
Sicilian theory. Benko had a good record versus elite opposition in this varia
tion, achieving a lot of draws with some of them ending in superior positions for
him. The examples presented illustrate some major themes and would be useful
for an Accelerated Dragon player to study: l .c4 c5 2 .NB Nf6 An alternate move
order leading to the Maroczy Bind is 2 . . . g6 3 . e4 Nc6 4.d4 cxd4 5 .Nxd4 Nf6
6 .Nc3 , and here B enko held his own against some real heavyweights using
Gurgenidze's system 6 . . .Nxd4 7 . Qxd4 d6 8 .Be2 Bg7

B ENKO AS B LACK

5 27

9.Be3 (9.0-0 0-0 1 O.Qe3 Be6 1 1 .Rb 1 a6 1 2 .Bd2 b5 1 3 .cxb5 axb5 1 4. Bxb5 Bxa2
1 5 .Nxa2 Rxa2 1 6.Bc4 Ra8 1 7 .b4 d5 1 8 . exd5 Nxd5 with a slight advantage, also
ending in a draw, Korchnoi-Benko, Curac;ao 1 962) 9 . . . 0-0 1 O. Qd2 Be6 1 1 .B Qa5
1 2 .Nb5 Qxd2 + 1 3 . Bxd2 (The next year Botvinnik switched to 1 3 .Kxd2 versus
Matulovic and got nothing there either.) 1 3 . . . a6 1 4.Nc7 Rac8 1 5 .Nxe6 fxe6 1 6.Rc l
Nd7 1 7 .Rc2 h 5 , drawn, Botvinnik-Benko, Wijk aan Zee 1 969. 3 .Nc3 g6 4.d4
exd4 S .Nxd4 Ne6 6.Ne2 A fairly conservative system that is popular among
solid positional players and carries a touch of poison. 6 . . . Bg7 7.e4 0-0 7 . . . d6
8 . Rb 1 ! ? Nd7 9 . Qd2 0-0 1 O.b4 Nb6 1 1 .Ne3 Be6 1 2 .f4 Rc8 1 3 .Ncd5 f5 1 4.Bd3
Nd4, and Black had an edge (eventually drawn) in Larsen-Benko, Boston 1 970.
S .Be2 d 6 9.0-0 Nd7

1 0.Bd2 This is the most important move, just as in the reversed position for
Black versus the English Opening (Rubinstein Variation) when he plays . . . Bd7.
Another good example was 1 O.Kh 1 Nc5 1 1 .B f5 1 2 .exf5 Bxf5 1 3 .Ne3 Nd4! (very
modern) 1 4.Nxf5 Nxf5 1 5 .Bg5 Qd7 1 6 .Rc l Ne6 1 7 .Bd2 Bd4 ! , with control of
the weakened dark squares, Korchnoi-Benko, Palma de Mallorca 1 968 (also even
tually drawn). l O . . .NeS 1 1 . h4 Ne6 1 2 .Rh l as 1 3 .a3 1 3 .b5 Ncd4 1 4.b6 ! ? was
played by Salov versus Adams at Dos Hermanas 1 99 5 , with even chances result
ing. 1 3 . . . axh4 1 4.axh4 Ned4 I S .Nxd4 Nxd4 1 6.Bd3

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

528

1 6 . . . Ra3 !, I12 - 112, Vaganian-Benko, Sao Paulo 1977. Amazingly, this is all cur

rent theory! Indeed, a game Kaidanov-Khasin, Belgrade 1 9S5 went 1 6 ... Bd7 1 7 .Re l
Ra3 I S .Bfl ReS ? ! 1 9.Re3 , and White was well o n top, although Nielson and
Hansen recommended I S . . . e6 instead. Later, in Linares 1 993 , Illescas followed
the same line against Ljubojevic but deviated by 1 7 .Qc l . The game went 1 7 . . . e6 !
I S .Bh6 Bxh6 1 9. Qxh6 Qf6, and Black was fine.
The interesting part is that in his notes, Illescas recommends Benko's move
1 6 . . . Ra3 instead, assessing this as slightly better for Black! I think that it is prob
ably only equal, if only because White has the option of 1 7 .Qe I Nb3 ! I S . Qxa3
Nxd2 with space versus the two Bishops. Still, this is a nice testament to Benko's
strength, and again, the game anticipates later theory by fifteen to twenty years.
Benko loved the Scheveningen Variation of the Sicilian D efense, and espe
cially the queenside minority attack, which in his games always seemed to
outrace White's kingside attack. Here are a number of examples: l .e4 cS 2 .NB
Nc6 B enko often used a Kan/Paulsen move order to set up similar queenside
themes, e.g., 2 . . . e6 3 .d4 cxd4 4.Nxd4 a6 S .Nc3 Qc7 6.Be2 Nf6 7 .0-0 Be7 S .Nb3
bS 9.Bf3 Nc6 1 0.g3 ! ? 0-0 I l .Bf4 d6 1 2 .Bg2 Bb7 1 3 .g4 b4 1 4.gS NeS I S .Ne2 as
(Already Black's attack is far more advanced than White's; Benko proceeds to
drive White away, create weaknesses, and then mop up on the queenside.)
1 6 .Ng3 a4 1 7 .Nd2 Nd4 I S .Re I a3 ! 1 9 . b 3 RcS 2 0 .Nc4 Nb S ! 2 1 . Q g4 Nc3
2 2 . Rfe l Nxa2 2 3 .Ra l Nc3 24.Re 3 e S 2 S .NhS exf4 2 6.Rh3 Nxe4, Conrady
B enko, Dublin 1 95 7 . 3 .d4 cxd4 4.Nxd4 Nf6 S .Nc3 d6 6.Be2 e6 7.0-0 In the
following game (as in our main game) Benko avoids the natural but fairly use
less . . . RacS: 7 .Be3 Be7 S .O-O 0-0 9.f4 Bd7 1 0.Nb3 a6 1 1 .Bf3 RbS ! (removing
the Rook from the influence of the f3 -Bishop, and leaving room for the king's
Rook on cS!) 1 2 .Qe l b S 1 3 .Rd l b4 (Twenty-five years later, Gheorghiu played
the same line versus R. Cardoso in Manila 1 9 7 3 , and used the direct minority
attack: 1 3 , . . Qc7 1 4. g4 b4 I S .Ne2 as 1 6 .gS NeS , with the kind of queenside
prospects B enko also liked.) 1 4.Ne2 eS I S .fS

B ENKO AS B LACK

529

1 5 . . . Na5 ! (This time Benko wants to play . . . Bb5 and . . . Nc4, although . . . a5 -a4
was naturally still valid.) 1 6.Nxa5 Qxa5 1 7 .g4 Rfc8 ! 1 8 .g5 Ne8 1 9.Rd2 Qxa2
2 0.Ng3 BfS 2 1 .Nh5 Qxb2 2 2 . Qg3 Rd , and White's attack is at a standstill, whereas
Black has two extra pawns and ideas like . . . b 3 , Shipman-Benko, U.S.A. 1 95 8 .
7 . . .B e 7 Here 7 . . . a 6 8.a4 B e 7 9.Kh 1 0-0 1 O .f4 Qc7 was the actual order o f Reyes
Benko below. 8.Khl 0-0 9.f4 Qc7 1 O.Bfl Rd8 ! ? A Benko trademark, espe
cially in his early years. Often this Rook goes to e8 to protect e 7 and open the e
file in case of . . . d5 and exd51 . . . exd 5 . On d8, Black prepares . . . e6-e5 followed by
. . . d 5 . 1 1 .Nb3 a6 1 2 .a4 b6 1 3 .Be3 Rb8

Notice how Black foregoes . . . Bb7, a natural move that many players have played
on this or the next move of exactly this position. By leaving the Bishop on c8, he
gives e6 extra support and, as we shall see next, keeps the b-file uncluttered.
1 4.Qe 1 Many years later, the game Schammo-Wirthensohn, Luxemburg 1 9 8 1
followed Benko's plan after 1 4.Rf2 Na5 ! 1 5 .Qd3 Nc4 1 6.Bc l b5 1 7 .axb5 axb 5 ,
and the queenside attack was well underway: 1 8.Nd4 b 4 1 9.Nd 1 e5 ! 2 0.b3 (2 0.Nb3
exf4 2 1 .Bxf4 Nd7 ! and . . . Nde5 next) 2 0 . . .Na 3 2 1 .Bxa3 bxa3 2 2 .Ne2 d5 ! (With
all of White's pieces awkwardly defending on the first 3 ranks, Black switches to
the center.) 2 3 .exd5 Bc5 ! ? (2 3 . . . exf4 leaves White with nothing to do) 2 4.Rfl
exf4 2 5 .c4? ! Qe5 ! 2 6 . Ra2 Bf5 2 7 .Qd2 Bb 1 , and Black was winning the Exchange
and then more on the queenside. 14 ... Na5 !

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

5 30

From the game notes: "Now 1 5 .Nxa5 bxa5 gives Black a strong initiative thanks
to the open b- and c-files. At the time, . . . Na5 was a new idea . . . However, today
(in many different forms and positions) it's accepted as a normal maneuver. "
1 5 .Rdl Nc4 1 6.Bcl bS 1 7.axbS axbS 1 8.Nd4 Or 1 8 .Ne2 e5 1 9 . Qg3 Bb7 2 0.Nc3
b4 2 1 .Nd5 Bxd5 2 2 .exd5 e4 (isolating and winning the d-pawn) 2 3 .Be2 Nb6
2 4.Nd4 Nfxd 5 , L. Reyes-Benko, Lugano 1 968. 1 8 ... b4 1 9.Nce2 eS ! 2 0.Nb3
dS ! . Note the central pawn breaks following the retreat of White's pieces into
passivity. This is the same theme we saw in the note to move 1 4. For the rest of
this dynamic Sicilian attack, see Game 7, Foltys-Benko, Budapest, 1 948.
It's worth mentioning the following game, not so much for its theoretical value
as for its philosophy: l .e4 cS 2 .Nf3 d6 3 .d4 cxd4 4.Nxd4 Nf6 S .Nc3 Nc6 6.f4
e6 7.Be3 Be7 8.Qf3 eS 9.NfS ? As pointed out in the notes to this game, 9.Nxc6
bxc6 1 0.f5 is White's best line. 9 . . . BxfS to.exfS

1 0 ... Nd4! Many players would be tempted to launch an attack or at least gain

some traditional Sicilian advantages here by 1 O . . . Qa5 1 1 .0-0-0 ( l 1 .fxe5 dxe5

1 2 .Bb5 Rc8 1 3 .0-0 0-0) 1 1 . . .0-0 1 2 .Kb l Rac8 1 3 .g4 Nb4! 1 4.a3 ( 1 4.Bd2 Rxc3
1 5 .Bxc3 Qxa2 + 1 6 .Kc 1 Nbd5) 1 4 ... Rxc3 1 5 .bxc3 Nbd5 etc. Of course, White
can play solidly and achieve normal middlegame play. But 1 0 . . . Nd4! moves the
game onto Benko's turf: I l .Bxd4 exd4 1 2 .NbS QaS+ n .c3 dxc3 1 4.Qxc3 Qxc3 +
I S .Nxc3 O-O-O! These moves and the rest of the contest are covered in the
game comments. What's significant is Benko's willingness to take the Queens
off for little apparent advantage in a position where one might think he could

BENKO AS BLACK

attack or exert pressure. Here's his comment: "Black has a slight but definite
advantage due to the weakness of White's doubled pawns. But winning the game
is far from easy because of the presence of opposite-colored Bishops. " In other
words, Benko has tremendous confidence in his ability to convert such a position, even against first-rate opposition. H e was indeed u p t o the task i n Lom
bardy-Benko, Seattle 1 966
Game 6 1 .
-

An irregular system for White meets with a creative response in Bisguier

Benko, New York 1 966: l .e4 e5 2 .Nf3 d6 3 .c3 Nf6 4.Bd3 In an ironic twist

(see the note to 4 . . . Bg4), Benko derides this move, saying: "Partly as the result
of this game, I suggest that the 'Hamburger Variation' be renamed the ' Chopped
Meat Variation. '"

4 . . . Bg4 Amazingly, in Kopec's entire video o n this Bd3 system (which he calls

"The Kopec System"), he doesn't mention this order! Kopec gives games with
2 . . . d6 3 .Bd3 Nc6 4.c3 Bg4 5 .Bc2 d 5 , but by that time White can try 6.d3 and his
e-pawn doesn't hang. Benko has indeed found an effective remedy because, by
delaying . . . Nc6, he gives himself time for . . . d5 before White can get organized
see the next note.
Only the year before, Benko as White (!) played the " Chopped Meat" himself
and ran up against 4 . . . Nc6 : 5 .Bc2 e 5 ! ? (5 . . . Bg4 6.d3 intending Nbd2 and h3 is
not as easy for Black as in our main game) 6.0-0 Be7 7 .h3 0-0 8.Re 1 (8 .d4) 8 .. .Nd7
9.d4 cxd4 1 O.cxd4 Bf6 l 1 .Bd , and White had a modest advantage, Benko-Tringov,
Havana 1 966. 5 .Be2 Benko: "The immediate 5 .h3 seems better. " Indeed, 5 .h3
was played against Benko later, by Camara in Sao Paulo 1 97 7 : 5 ... BxB (5 ... Bh5
has been played in several games. One example: 6.Bc2 Nc6 7.d3 e6 8 .Nbd2 d5
9.Qe2 Be7 1 0.Nfl b 5 l 1 .Ng3 Bg6 1 2 .0-0 0-0 1 3 .Nh4 d4 1 4.Nxg6 hxg6 1 5 .Q d 1
Rc8 1 6 .Ne2 dxc3 1 7 .bxc3 b4! , Boric-Ermenkov, Pula 1 99 1 ) 6. QxB Nc6

531

PAL BENKO: M Y LIFE, GAMES AND COMPOSITIONS

532

7 .BbS (Wasting time; one high-class example with 7 .0-0 went 7 . . . g6 S.Re l Bg7
9.Bfl 0-0 l O.a4 a6 1 1 .Na3 RbS l 2 . Qd l Nd7 1 3 .Nc2 NaS ! = , Anand-Van Wely,
Tilburg 1 992 .) 7 . . . RcS S .O-O e6 9.d4? ! Qb6 l O.Bxc6+ Qxc6, and Black had supe
rior development. S . . . dS S . . . g6 has also done well, simply playing for dark-square
control. Then the most famous Bd3/c3 game ever played went 6.h3 Bxf3 7 . Qxf3
Bg7 S .O-O 0-0 9.Qd l Nc6 l O.d3 ( l O.d4 cxd4 1 1 .cxd4 Qb6 or . . . RcS; Black is
well-developed with dark-square pressure) 1 0 . . . RbS 1 1 .Nd2 bS l 2 . a 3 Nd7
( 1 2 ... Qc7) 1 3 .Nf3 NdeS ? ! 1 4.NxeS BxeS I S .Be3 Qb6 1 6 .Kh l Bg7 1 7 .f4 as, and
from this roughly equal position, Black (Bosboom) actually beat Kasparov in Wijk
aan Zee ! ! (Okay, it was a blitz game).
Instead of 6.h3 , K. Georgiev-Kindermann, Thessaloniki 1 9S5 continued 6.d3
Bxf3 ! ? 7 . Qxf3 Nc6 S . O-O Bg7 9.Be3 0-0 l O .Nd2 ( lO . d4? cxd4 I l .cxd4 Qb6)
lO . . . RcS 1 1 .Rad i b S ! 1 2 .a3 as 1 3 . d4? ! cxd4 1 4.cxd4 Nd7 I S .Nb3 a4 1 6 .Na l Nb6
1 7 .b 3 b4! I S .bxa4 bxa3 1 9.Qe2 NaS ! , and Black, ready for .. .Nbc4, was consid
erably better. 6.eS Benko: "Naturally, 6.d3 dxe4 7 . dxe4 Qxd 1 + is not appetizing
for White. " But at least he would retain some sort of balance, e.g., 6.d3 dxe4
7 . dxe4 Qxd 1 + S.Kxd l Nc6 ! ? (S . . . Bxf3 + 9.gxf3 eS ! ?) 9.Ke2 eS l O.h3 Be6 followed
by . . . Be7 and . . . 0-0. Black is somewhat better, since White is behind in develop
ment and can"t get a piece to dS whereas Black has ideas like . . . bS -b4, depend
ing where the b l -Knight goes. 6 . . . Nfd7 7.d4 e6

Benko's note says it all: "Finally arriving at a French Defense that is favorable
for Black. It's well known, of course, that Black's troubles in the French stem
from his 'problem' light-squared Bishop, which is usually difficult to develop

BENKO AS BLACK

effectively. Here the Bishop has already found a good home. " 8 .Nbd2 8.Be3
Qb6 9.b3 cxd4 1 O.cxd4 Bb4+ 1 1 .Kfl Nc6 wasn't much of an improvement in
Rellstab-Kottnauer, Helsinki 1 9 5 2 . 8 ... exd4 9.exd4 Ne6 1 0.h3 BhS 1 1 .g4 Bg6
1 2 .Bxg6 hxg6 1 2 . . . fxg6 ! followed by . . . Be7 and . . . 0-0 is a typical French idea to
exert pressure down the f-file. O f course, 1 2 . . . hxg6 i s also good. 1 3 .Nb3 as 1 4.a4
Re8, and the opening has concluded to Black's advantage. This was BisguierBenko, New York 1 967; see Game 64.

l.d4 O PENIN GS
It should come as no surprise that Benko tried a wide variety of defenses versus
l .d4. But here we see a bit more specialization than in the other parts of this
survey. For a period of many years, he didn't deviate from the Benko Gambit,
and I will devote a large section to it next. At other times, the Griinfeld Defense
was a mainstay of his repertoire, and he employed the King's Indian Defense
regularly throughout his career, showing particular creativity in his handling of
it. Not surprisingly, he used other defenses (notably l .d4 d5), but the three open
ings above account for the bulk of his practice and will be treated with the most
depth here.

T he Benko Gam bit

In general, the task of designating an opening with a particular player's name
tends to be unresolvable and perhaps not very interesting. But it's hard to imag
ine any modern mainstream opening to which one player has clearer naming
rights than the Benko Gambit, i.e., 1 .d4 Nf6 2 .c4 c5 3 .d5 b 5 . Of course, histori
ans can make the case that almost any opening and/or something similar to it
was tried before the player whose name it is associated with. Some trivial ex
amples are the Alekhine Defense, the Pirc Defense, the Najdorf Variation, the
Tarrasch French, and Queen's Gambit lines associated with names such as Lasker,
Tarrasch, and Capablanca. There are many more such examples. But who has
played the opening first is not an important criterion, as Benko himself points
out, citing Alekhine's comments at the beginning of the section on the Benko
Gambit. In some of Europe the Benko Gambit is still called the "Volga Gam
bit, " because a few Volga masters played it some years back. Or at least they
played something similar, involving the sacrifice of Black's b-pawn in the same
structure, the precise position varying. Their games are not a part of modern
theory at all, as far as I can tell. More importantly, their games didn't inspire or
even interest international players, as evidenced by the fact that when Benko
began to play the 3 . . . b5 gambit in 1 967, it was met with almost total skepticism
and didn't catch on until Benko had won a multitude of games with the gambit
and published articles about it. Those articles certainly inspired leading Ameri
can players such as Fedorowicz, Browne, and Alburt to become Benko Gambit
advocates and cement its respectability once and for all.

533

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

5 34

On the same subj ect, Steffen Pedersen says: "The Benko Gambit . . . has been
around in some form or another since the 1 92 0s . . . ". Naturally, "in one form or
another," so has just about any opening, and "similarity" has not generally been
a criterion in naming openings. No one seems to complain about the name Najdorf
Sicilian, for example, and yet the Scheveningen Sicilian is certainly very similar
to the Najdorf, as are other Sicilian variations (with exceptionally similar struc
tures, themes and ideas) . Certainly the "Taimanov Sicilian" goes back to struc
tures and even piece development that were seen in the games of Anderssen and
Paulsen. And so forth. The reality is that we should name openings after those
who: (a) popularized them among strong players; (b) developed their theory be
fore or more extensively than others; and, as Alekhine says in Benko's introduc
tory citation, (c) "gave the move real significance. " Benko very clearly meets all
three criteria.
Vukic-Benko, Sarajevo 1 967 was the first game that attracted top players'
attention to the sequence l .d4 Nf6 2 .c4 c5 3 .d5 b 5 , although Benko had already
used it often in less significant U. S . tournaments. As it was, the gambit took
several years and a great number of games by Benko to overcome the natural
and negative response that most players initially gave it. In brief, this response
was: "What? Give away a pawn as Black on the third move, for no attack whatso
ever and without creating any enemy weaknesses? Ridiculous ! " I freely admit
that I felt the same way when I first saw it, and was sure that a way would be
found to secure White a safe advantage. But the Benko Gambit continues to
thrive today. The hidden key is that White's queenside pawns are immobile and
subject to attack while his opponent develops more freely and actively.
Remarkably, just about all of Benko's early games are still relevant for theory.
Since the Benko Gambit is so integral to any exposition of his play, I will analyze
the theory behind most of the games in this section in unusual detail. Vukic
B enko won't get such a meticulous treatment, leading as it does into the next
few games; and yet it remains an exemplary game for the variation involved, a g3
Fianchetto System: l .d4 Nf6 2 .c4 c5 3 .d5 b5 4.cxb5 a6 Notice the difference
between this and l .d4 Nf6 2 .c4 c5 3 .d5 g6 4.Nf3 b5 etc., where the Knight is
already committed to f3 . 5.bxa6 Bxa6 6.Nc3 d6 7.Nf3 g6 8.g3 See the discus
sion of g3 systems, and many games with it, in the next few games. 8 ... Bg7 9.Bg2
0-0 1 0.0-0 Nhd7 I 1 .Qc2

BENKO AS BLACK

Still the most popular move today. White would like to play Rd l and e4, or
perhaps Rb I , b 3 , and Bb2 . 1 1 . ..Qb6 As Benko points out, there are alternatives
such as 1 1 . . .QaS (the most popular move today; see the next two games) and
1 1 . . .Qc7 . 1 1 . . . Qb6 tries to exert pressure down the b-file, to leave c7 open for
the standard idea . . . Nf6-e8-c7 , and t o avoid the loss of tempo by Bd2 that . . . Q a S
allows. 1 2 .Rd l 1 2 .Rb l , once thought t o be good, can run into 1 2 . . . Qb7 ! , intending moves like . . . Nb6 and . . . Bc4 with threats against White's d-pawn.
1 2 . . . Rfb8 1 3 .Rb l Ne8

We haven't left modern mainstream theory! 1 3" .Ne8 intends . . . Nc7-bS , further
pressuring White's queenside. It is given " ! ? " by Pedersen, but it's the only move
that he mentions. 1 3" .Ng4 is a bit more popular than 1 3" .Ne8, also opening the
g7-Bishop's diagonal and preparing . . . NdeS . Both moves have their points. 1 4.BgS
By transposition, Csom-Gligoric, Ljubjana Portoroz 1 97 3 went 1 4.Nd2 Nc7 I S .h3
NbS 1 6.Nc4 Qa7, leading to an "unclear" position (Pedersen). 14 ... Qd8 14 . . . Kf8
was played in two other games, with even results. l S .Bfl h6 1 6.Bd2 Nc7 1 7.b3
Nb6 1 8.e4 Bxfl 1 9.Rxfl Amazingly, we are only now leaving current theory,
although this was the first international test of the gambit! 1 9.Kxfl Qd7 2 0.Kg2
was played in a 1 986 correspondence game Bernhardsson-Blodig. Black won with
a plan similar to Benko's, by playing . . . e6 and exploiting White's light squares,
but at this point it wasn't clear who had the advantage. 19 ... Qd7 20.Rfe l Kh7
2 1 .Kg2 e6 A standard plan once the g2 -Bishop has been traded. Nevertheless,
as Benko mentions, there are alternatives involving . . . c4 and .. .fS . 2 2 .dxe6 Nxe6
2 3 .Ne2 dS. The opening can be considered over. Black has achieved counterplay
with an important central break, but White is still a pawn up. This is hard to
assess. See Game 72 for the continuation.
The g3 systems were extremely common in the first days of the Benko Gam
bit and are still very important. White creates no weaknesses and, unlike sys
tems with e4 or f4, he gets castled without much trouble. Thus playing against
g3 provides a real test of Black's positional compensation. The good news for
him is that the Bishop takes two moves to get to a square (g2) where it is blocked
by its own d-pawn. Following the Vukic-Benko game just seen, systems with g3
became popular and took on many forms. Since Black has no targets and cannot
launch an immediate attack, these fianchetto systems show clearly how, without

535

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

536

hurried or forcing moves, Black generates counterplay in a typical Benko Gam

bit. The next game, Aspler-Benko, Vancouver 1 97 1 , contains various examples
played by Benko in its subvariations, and also a number of White alternatives to
the current main line : l .d4 Nf6 2 .c4 c5 3 .d5 b5 4.cxb5 a 6 5.bxa6 Bxa6 6.Nc3
B enko was faced with his own gambit in Benko-Berry, Canada 1 97 1 : 6.g3 d6
7 .Bg2 g6 8 .Bd2 Bg7 9.Bc3 ! ? (an interesting choice) 9 . . . 0-0 1 0.Nh3 Nbd7 1 1 . 0-0
Nb6 1 2 .Nf4 with complications, for which see Game 74. Remarkably, I find no
reference to this idea in the two main recent Benko Gambit books, and only one
example in ECO. The latter cites analysis by Ruzele: I 1 .Nf4 Nb6 1 2 .h4 Ra7 !
1 3 .Qc2 Bb7 1 4.e4 Na4 1 5 .0-0 Qb6, with compensation. This whole line with
Bd2 -c3 has been passed over by theory, even though Benko himself chose to
play it as White. 6 . . . d6 B enko has some notes on these moves in the Games
Section. 7.Nf3 Here are three more B enko games, all without NB , after 7 .g3 g6

8.Bg2 (8 .Nh3 ! ? Bg7 9.Nf4 Nbd7 1 O.h4 O-O? ! - Benko suggests the safer 1 0 . . . h6
or 1 0 . . . h5 - 1 1 .h5 N e5 1 2 .hxg6 hxg6, Kovacs-Benko, Debrecen 1 9 7 5 , Game 8 1 .
This is given thorough treatment in the game notes) 8 . . . Bg7 9.Nh3 (9.Qa4+ Nbd7
1 0.Qh4 h6 Barlay-Benko, U. S . Open 1 969; White's Queen has been misplaced
at the cost of tempi and he stands worse.) 9 . . . Nbd7 1 0.0-0 0-0 I 1 .Qc2 Qa5 1 2 .Bg5
Rfb8 1 3 .Rfe l Ra7 1 4.Rab l Rab7 1 5 .Bd2 c4! ? (I'm not sure that this is best; an
alternative is 1 5 ... Qc7.) 1 6.b4 cxb3 1 7.Rxb3 Bc4!? ( 1 7 ... Rxb3 1 8.axb3 Qa3) 1 8.Rxb7
Rxb7 1 9.Ng5 Qa6 2 0. Re I Rb8 , Denker-Benko, Lincoln 1 969. Black has com
pensation, but this position is hard to assess. The game itself was drawn. 7 . . . g6
8.g3 Bg7

BENKO AS BLACK

9.Bg2 The fianchetto variation is solid but also has its disadvantages. As men
tioned above, White's light-squared Bishop is often not active on g2 because his
own center pawns limit its scope. 9.Bh3 was played at least twice by Hort in
games that Benko expresses interest in: 9 . . . 0-0 ! ? (This "natural answer .. .is probably not best" -Benko. Presumably he wants to get on with 9 . . . Nbd7 1 0.0-0
Nb6 immediately, contemplating ... Bc8 and recapture with the Queen in the
case of Bxc8 .) 1 0 .0-0 Qb6 (Benko also refers to Hort-E. Meyer, U.S .A. 1 974:
1O ... Nbd7 1 1 .Qc2 Qa5 1 2 .Bd2 Rib8 1 3 .b3 ? ! - [ 1 3 .Rab 1 ] 1 3 . . . Qb4! 1 4. Rab 1 c4!
1 5 .Rfc 1 Nc5 1 6.Na4 cxb3 1 7 .axb 3 Qe4 1 8 .Nxc5 dxc5 1 9.Bfl Qxc2 drawn, in
view of 20.Rxc2 c4 = . ) 1 1 .Rb 1 Bc8 ! ? ( 1 1 . . .Nbd7 followed by . . . Rib8, . . . Qd8 ! ? ,
and . . . Nb6 i n some order was perhaps a better way to prepare . . . Bc8 and possibly . . . Bd7 if White retreats his Bishop.) 1 2 .Bxc8 Rxc8 1 3 . e4 Nbd7 1 4.h3 h6 ! ?
1 5 .Bd2 , and Black was having diffficulty improving his position in Hort-Diesen,
Hastings 1 974. 9 ... Nbd7 10.0-0 0-0 Benko's own recommendation of 1 0 . . . Nb6
first is now a main line, but this position is also critical and is sometimes reached
by different orders. I I .Re l 1 1 .Qc2 , the main line of the Fianchetto Variation, is
considered in the next two games. 11...Qb6 12 .e4? Better is 1 2 .h3 . 12 . . . Ng4!
13 .Qc2 rub8

14.h3 Gordon-Benko, U.S .A. 1 976 (Game 82) saw 1 4.Rb 1 Nge5 ( 1 4 . . . c4 was

also very promising, according to the game notes.) with advantage for Black.
14 . . . Nge5 15 .Nxe5 Nxe5, with a very promising position. See Game 7 7 .
Hort-Benko, U.S.A., 1974 follows the same line u p to move 1 1 . Qc2 : l .d4
Nf6 2 .c4 c5 3 .d5 b5 4.cxb5 a6 5 .bxa6 Bxa6 As Benko notes, 5 . . . g6 is more

accurate. This is not only because it discourages the double fianchetto with b 3 ,
but because there are lines in which Black wants to play . . . Qa5 first and then
. . . Bxa6, in order to respond to White's moves e4 (or e 3 ) and Bxa6 by recapturing
with his Queen on a6. 6.Nc3 d6 7.NfJ g6 8.g3 Bg7 9.Bg2 Nbd7 10.0-0 0-0
I 1.Qc2 Qa5 Rather than 1 1 . . .Qb6, as we saw in Vukic-Benko. 12 .Rd l

537

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

538

Now White wants to play b 3 , and will often play Rb I to prepare it. He also has
the idea 1 3 .Bd2 , and if 1 3 .. .Nc4, 1 4.Be l . 12 . . Rib S This is the normal move,
which is still doing well enough, although the continuations seem to have Black
only holding on without many positive prospects. Also, White's idea in next game
may discourage some. For those who don't like this move, 1 2 . . . Ng4 (preventing
b3 ) has a decent reputation (it was played by Adams, for example). Benko him
self suggests and analyzes a third option, 1 2 . . . Nb6 ! ? :
.

I n very limited practice this hasn't done well, and I have to admit that I a m skep
tical about at least one response. Still, there is no refutation, and the lines don't
look that bad. Anyway, here are some ideas: 1 3 .Rb l (the standard idea, which
Benko doesn't mention. He gives 1 3 .b3 Nbxd5 ! , correctly, and 1 3 .Nd2 c4, which
I'm not sure will equalize after 1 4.Nfl , e.g., 1 4 . . . Bb7 1 5 .Ne3 , with either b4 or
first Bd2 and then b4 in the works. This is not set in stone, however.) 1 3 . . . Nc4
(Maybe 1 3 . . . Bc4, although even 1 4. e4 intending b3 and Bd2 should keep some
edge, e.g., 14 . . .Ng4 1 5 .h 3 Ne5 1 6.Nxe5 Bxe5 1 7 .a3 , with ideas like b4 or Bd2
and b 3 ; not appropriate here is 1 3 ... c4? ! 1 4. e4.) 1 4.Nd2 Nd7 , which has led to
White wins after both 1 5 .Bh3 and 1 5 .Nxc4. Nevertheless, one could argue with
assessment-by-result, understanding that Black still has his basic Benko Gambit
advantages here, and proceeding on the belief that these will provide compensa
tion. 1 3 .Bd2 For 1 3 .Rb l , see the next game. The move 1 3 .b3 should be taken
very seriously as a potential threat to 1 2 . . . Rfb8, for example, 1 3 . . . Ng4 1 4.Bb2
( 1 4.Bd2 intending 14 ... c4 1 5 .h3 is also possible and may be better.)

BENKO AS BLACK

539

1 4 . . . Bc8 ! ? ( 1 4 . . . Qb4? ! I S .h3 NgeS 1 6.Nd2 ! , threatening f4, van Wely-Kogan,

Antwerp 1 998) I S .Nd2 ( I S .Ne4 ! ? looks more active) I S . . . Nb6 1 6.h3 NeS , and
Black had decent play with the idea of . . . c4, Milos-van Riemsdijk, Sao Paulo
1 99 7 . But this all seems a little marginal to me. 1 3 . . .Ne8 ! 1 4.Bh3 ! ? Bc8 1 5 .Bel
Nc7. Here Benko stops-there is plenty of compensation for the sacrificed pawn.
The next few moves were 1 6.e3 Qa6 ! ? 1 7 .Bfl Qb7 1 8 .Rab l Nb6 1 9. e4 Bg4 ! ,
with typical Benko Gambit pressure.
Following the moves of the previous game (Hort-Benko) up to 1 2 . . . Rfb8 ,
Zaltsman-Benko, Lone Pine 1 98 1 was nineteen years ahead o f its time after 1 3 .Rh l
This has been the recent choice for White, instead of 1 3 .Bd2 or 1 3 .b3 . 1 3 . . . Nb6
1 4.b3 Ne8 1 5 .Bd2 Qa3 1 6.Bc1 Qa5

1 7.Bh 2 This variation is important for current Benko Gambit theory, and an
overview of its recent practice should be of interest. Notice of course that White
can repeat. Thus for guaranteed winning chances, Black has to look at his earlier
options. The game Avrukh-Ponomariov, Batumi 1 999 deviated at this late point
from Benko's game, but had strong similarities after 1 7 .Bd2 Qa3 1 8 . e4 ! ? Bc8 !
(preventing NbS , even after Bfl , because the a-pawn would fall) 1 9.h3 Bd7 2 0.Bel
QaS 2 1 .Bd2 Qa3 2 2 .Bel QaS 2 3 .Bb2 (Also Zaltsmann's solution ! Avrukh is again
reluctant to take a draw as White.) 2 3 . . .Nc7 2 4.a3 (A recent game went 24.Qel
Qa6 [24 . . .fS ! ?] 2 S .Re l Qc8 2 6.Kh2 Qf8 - [2 6 . . .fS ! ?] 2 7 .Ra l Ra7 2 8 .Nd l , 112 _ 112,
Kruppa-Malakhatko, Kyiv 2 00 1 . Black again follows with .. .fS , either before or
after . . . Bxb2.) 24 . . . Qa6 2 S .Bfl ? ! (This looks good, but takes support away from
d S . 2 S .Nd2 is better.) 2 S . . . Qc8 2 6.Kh2 fS !

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

540

2 7 .Ra l Qf8 2 8 .Rd2 fxe4 2 9 . Qxe4 Qf5 ! (A surprise ! 2 9 . . . e6 was another ap

proach.) 3 0 . Qxf5 Bxf5 3 1 .Re 1 Kf8 (Black wants more than 3 1 . . .Rxa3 3 2 .Bxa3
Bxc3 3 3 .Rxe7.) 3 2 .Ng5 h6 3 3 .Nge4 Nd7 (B . . . Bxe4 3 4.Rxe4 Bxc3 3 5 .Bxc3 Nbxd5
3 6 .Bb2 Rxb3 3 7 .Bc4 is difficult for both sides.) 3 4.b4? cxb4 3 5 .axb4 Rxb4 3 6 .Bg2
Nf6 ! ? 3 7 .Nxf6 Bxf6 3 8 .Rc 1 ? (Time pressure, but Black has a clear edge.) 3 8 . . . Nb5
3 9.g4 Bg5 .
Returning to Zaltsmann-Benko: 1 7 . . . Nc7 1 8 .e4 Nd7 ! ? 1 8 . . . Bc8 ! and . . . Bd7
was Ponomariov's maneuver in the same type of position. The move ... Bg4 can
also be useful . 1 9.Bh3 Bc8 20.Bfl

20 . . . Ba6 Black is content to repeat. Instead, although he would have wasted some

valuable time, Benko could also have tried 20 . . . Nb6 ! ? followed by . . . Bd7, mim
icking Ponomariov's idea. Then . . . f5 would be a factor. 2 1 .Bh3 Bc8 2 2 .Rdc 1
Nb6 Exchanging light-squared Bishops is generally useful to Black, but in this
case it's hard for him to make progress afterwards. 2 3 .Bxc8 Rxc8 24.Nd2 Qa6 ? !
I t may be time t o bail out by 2 4 . . . Bxc3 2 5 .Bxc3 Qxa2 . After White's next, it's
hard to justify the pawn deficit. 2 5 .a4! Qb7 26.Nc4 Nxc4 27.bxc4 Qb4 2 8.Qe2
Bxc3 29.Bxc3 Qxa4 3 0.e5. Black's position is unpleasant and it led to one of
Benko's few losses with his gambit. A remarkable anticipation of modern theory,
however.
The following game, I. Zaitsev-Benko, Szolnok 1 97 5 , is a classic struggle
between attack and counterattack in a still-critical variation. I have included quite
a bit of theory in the notes for the reader's sake: l .d4 Nf6 2 .c4 c5 3 .d5 b5

BENKO AS BLACK

4.cxbS a6 S .Nc3 This is now known as the Zaitsev Variation versus the Benko

Gambit. Its first appearance was in this game, appropriately enough, between
Zaitsev and Benko! S . . . axbS 6.e4 b4 7.NbS d6 White was threatening e 5 , and
7 . . . Nxe4? ? 8 . Qe2 is disastrous . 8.NO We should turn our attention to Benko's
thoughts about 8.Bf4 and the theory that surrounds it:

8.Bf4 is White's most popular move . In response, it turns out that Black has
several decent alternatives, but his most frequent and still satisfactory response
is Benko's own idea 8 . . . g5 ! . Neverthelss, it's very interesting to see what Benko
said at the time of the game about some alternatives that are now fairly respect
able, for example after 8 . . .Nxe4 9.Qe2 g5 , he anticipates the move that became
popular more than a decade later: 1 0.Be5 ! ? dxe5 (Recently, even the bizarre 10 .. .f6
has been achieving equality.) I I . Qxe4, which he naturally thought at the time to
favor White. Today, Black gets equal theoretical chances and often even better
practical ones out of I l . . .Bg7 ! 1 2 .d6 Ra5 ! , for example, 1 3 .Rd 1 0-0; Benko is
correct that 8 . . . Nbd7 9.Nf3 Nb6? l O .Rd ! (threatening I l .Rxc5 !) l O . . . Nxe4
I I .Bd3 favors White, but after years of analytical and practical input, it appears
that 9 . . . Nxe4! is satisfactory for Black- see 8.Nf3 Nbd7 9.Bf4 Nxe4! below.)
9.Bxg5 Nxe4 1 0 .Bf4 ( l 0.Bh4 Bg7 1 1 .Bd3 Nf6 1 2 .Nf3 Nbd7 1 3 .0-0 Bb7, with
strong pressure on the d-pawn, Halldorsson-Benko, Philadelphia 1 97 8) l O . . . Bg7
(Benko also mentions lO . . . Qa5, which is hotly contested to this day.) 1 1 .Qe2

541

PAL BENKO: MY LIF E, GAMES AND COMPOSITIONS

542

1 1 . . .Nf6 1 2 .Nxd6+ Kf8 1 3 .Nxc8 Qxc8 and Benko stated that he thought Black
was assuming the initiative. Theory recommends 1 4.Qf3 ( 1 4.d6 exd6 I S .Bxd6+
Kg8 is easy to play and probably better for Black) and now 14 . . . e6 (or perhaps
14 . . . QfS) is held to be equal.

8 ... Nbd7 ! ? Benko says " ? ! " too pessimistically. But he is right that 8 . . . g6 is the
most reliable move, one main line today proceeding 9 . Bc4 Bg7 1 0 .0-0 0-0
( 1 0 . . .Nxe4! ? 1 1 .Re i Nf6 1 2 .Qe2 Bb7 is praised by theory, but it doesn't con
sider 1 3 .Qxe7+ Qxe7 1 4.Nxd6+, with some problems for Black.) 1 1 .Re l Nbd7
1 2 .a3 ! ? Nb6 ! , and Black is holding his own. 9.Bf4 Nh5? Although one would
hardly dare play it over the board, Black should play 9 . . . Nxe4! here, a main line
of theory going 1 0. Qe2 fS 1 1 .NgS Ndf6 1 2 .f3 NxdS 1 3 .Nxe4! Nxf4 ! , and two
games have demonstrated that Black gets dynamic equality (but no more) after
either 1 4.Nexd6+ or 1 4.Nbxd6+ exd6 1 S .Nxd6+ Kd7 1 6.QbS+ Ke6. 1 0.Bg5 Nhf6
1 0 . . . Qb6 ? ! I l .Nd2 g6 1 2 .Nc4 Qb8 1 3 .a4 with a nice advantage was Rashkovsky
Zilberman, USSR 1 9 7 5 . 1 1 .Qe2 ? ! This leads to a game illustrating Benko's ge
nius in these positions. Correct was either taking the draw by I I .Bf4, or Zaitsev's
suggestion I I . e S , which theory considers clearly better for White. But Benko
quite correctly suggested (in 1 97 5 !) the line 1 1 . . . dxeS 1 2 .Qe2 RaS 1 3 .NxeS NxeS
1 4.QxeS Qb6 I S .Nc7+ (Some older theory, still quoted, gives I S .a4! bxa3 1 6.Rxa3
RxbS 1 7 .BxbS+ QxbS 1 8 .Ra8 Qb4+ 1 9.Ke2 , winning, but 1 8 ... Qb7 ! 1 9.Rb8 Nd7 !
turns this assessment around.) I S . . . Kd8 1 6.Bf4 ( 1 6.d6 exd6 - [or 1 6 . . . Qxd6 ! ?]
1 7 .Bxf6+ Kxc7 1 8 .QgS gxf6 1 9 .Qxf6 Qc6 ! 2 0 . Qxh8 Qe4+ is too strong.), and
here he gives the still-unanalyzed 1 6 . . . Nd7 !

BENKO AS BLACK

( 1 6 . . . Ra7 1 7 .Nb5 Rd7 l S .Bc4 is practically winning; 1 6 . . . Bd7 ! ? 1 7 .Rd l intend

ing Ne6+ is very messy.) 1 7 .Qe3 ( 1 7 .Qf5 Nf6 ! ? or Benko's 1 7 . . . Qf6) 1 7 . . . g5 l S. Bg3
Bg7 . Benko stops here. At first, 1 9.d6 ! looks killing, but the position is far from
clear after 1 9 . . . exd6! 2 0 .Nd5 ReS 2 1 . QxeS+ KxeS 2 2 .Nxb6 Nxb6 2 3 .0-0-0 Na4 ! ,
when Black's Bishops turn out to be stronger than they at first might seem. Thus
Benko's idea for Black is still alive and yet another example of highly relevant
older theory that has been lost in the flood. After 1 1 .Qe2 , threatening e 5 , Black's
game seems critical. But there is a resource: 1 1 .. .RaS ! !

Benko expresses his philosophy behind this beautiful move in his game notes.
Such . . . Ra5 moves later became common in the Benko Gambit, but at the time
this was a bombshell. 1 2 .eS It's hard to do without this move. Old Benko notes
gave 1 2 .Qc2 b3 " ! ", but 1 3 .Qxb3 Nxe4 1 4.Bf4 Ba6 1 5 .Nd2 ! leaves Black the
question of how to develop his kingside. Better is 1 2 . . . Ba6 ! 1 3 .a4 bxa3 1 4.Nxa3
Bxfl 1 5 .Kxfl g6 or 1 5 . . . QaS, with typical gambit pressure. 1 2 ... Ba6 ! 1 3 .exf6
BxbS 1 4.fxe7 Bxe7 I S .Bxe7 Qxe7 1 6.Qxe7+ Kxe7 1 7.BxbS RxbS, and Black
is clearly better. For the rest of this pretty game, see Game so.
With Pfleger-Benko, Skopje 1 972 and its associated games, we move into
more contemporary practice with the 5 .b6 System: l .d4 Nf6 2 .c4 cS 3 .dS bS
4.cxbS a6 S .b6

White returns the pawn for positional reasons, e.g., to occupy c4 with a Knight
or to play directly for e 5 . This is one of White's best options, although it didn't
really catch on until many years later. S . . . d6 5 . . . Qxb6 Ied to a poorly-played but

543

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

544

still instructive game in Mendoza-Benko, Malaga 1 969: 6.Nc3 g6 7 . e4 d6 S . Rb l

(S.a4; and S .NB are normal today, and both have loads of theory behind them.
S . Rb l prepares to develop the c 1 -Bishop.) S . . . Bg7 9. Qa4+ ! ? (9.NB 0-0 1 O.BgS ! ? ,
but this is nothing special) 9 . . . Nbd7 1 0 .Bd2 0-0 I 1 .NB RbS ? ! ( l 1 . . .Bb7 1 2 .Bd3
e6! is interesting) 1 2 .Bd3 c4? (Too ambitious; better is 12 ... e6) 1 3 .Be3 ! NcS
1 4.Qxc4 Nfd7 I S .0-0 as 1 6.Bd4? ? , 1/2_ 112 . White's final move, and the result, are
very strange. White would have been much better after 1 6.Na4! with a clear
pawn up and good position. But after 1 6. Bd4? ? , Black can actually achieve a large
advantage by 1 6 . . . Bxd4! 1 7 . Qxd4 ( 1 7 .Nxd4 Ba6 wins a piece) 1 7 . . . Nxd3 I S . Qxd3
Ba6 1 9. Qd2 Bxfl 2 0.Kxfl RfcS, and not only does White have only a pawn for
the Exchange, but Black's pieces are much more active and will soon infiltrate.
6.Nc3 Nbd7 7.NB Natural and sound, but today 7 . a4 is more popular. 7 ... g6
8.e4 Bg7 9.Be2 0-0 1 0.0-0 Nxb6 I 1 .Bf4 Bg4

1 2 .Nd2 B enko gives 1 2 .h3 BxB 1 3 .BxB Nfd7 ( 1 3 . . . NeS 1 4.Qb3 RbS I S .Be2

Nc7) 1 4.Be2 c4 intending . . . NcS . But whether Black should give up d4 and ex
pose his c-pawn in return for central pressure is at least open to question. A
plausible continuation might be I S .Be3 Qc7 1 6.Rc 1 intending 1 6 . . . NcS ? 1 7 .M ! .
But 1 3 . . . NeS intending . . .Nc7 should be fine, e.g., 1 4.Qb3 RbS I S .Be2 Nc7 .
A similar idea was seen i n Gamboa-Leko, Yopal 1 99 7 : 1 2 .a4 NeS ! 1 3 .Qd2
BxB 1 4.BxB Nc4 I S . Qc2 ( l S .Qe2 NeS = ) I S ... RbS 1 6.Rab l Nc7 , assessed as
equal by Pedersen. The natural 1 7 .Be2 can be met by 1 7 ... Nas ! ? , heading for
d4 via b3 . Then I S .b3 ( l S . Rfd l Nb 3 1 9. Bc4 Nd4 is fine for Black) I S . . . fS ! gives
active counterplay, since the move . . . Bxc3 is often in the air, e.g., after 1 9. exfS .
1 2 . . . Bxe2 1 3 .Qxe2

BENKO AS BLACK

1 3 . . . Qd7 ! As Benko points out, this clever move threatens to exchange pieces

(normally desirable in the Benko) by . . . Na4, and to play 1 4.a4 Qg4! Enforcing
the exchange of Queens. 1 5 .Qe3 Nh5 1 6.Bh6 Bxh6 1 7.Qxh6 Qf4! 1 8.Qxf4
Nxf4 19.b3 Rab8 20.g3 Nd3 "With a clear plus for Black," according to Benko,
which may be slightly exagerrated, but certainly Black is better. Nevertheless,
the game was quickly drawn after 2 1 .f4 f5 ! 22.a5 ? ! 2 2 .Rf3 Nb4 2 3 .Rd 1 is a
tougher defense. 22 . . . Nd7 2 3 .Rad l fxe4 24.Ncxe4 Nb4 ! ? Still better was
24 . . . RfS ! 2 S .Nc3 Nb4. 2 5 .NB h6? 2 S . . . RbS wins a pawn safely enough. 2 6.Nc3
Nc2 , Ih _ 1h . Black is obviously still better. Benko often seemed inclined to take
draws like this in superior positions, perhaps due to his habitual time trouble.
The opening theory behind Portisch-Benko, Palma de Mallorca 1 97 1 , and
the game itself, includes several Benko Gambit features we haven't run into yet.
With the S .e3 Variation, Portisch challenged the very soundness of 3 . . . b S : 1 .d4
Nf6 2 .c4 c5 3 .d5 b5 4.cxb5 a6 5.e3 The single most popular response to the
Benko Gambit for quite a long time, with its peak in the early-to mid- 1 980s. It
is still played regularly. 5 . . . g6 6.Nc3 d6

At first sight it is quite surprising that this move has completely fallen from fa
vor. Jacobs and Kinsman go so far as to give it a " ? ! " . The point is that White
shouldn't be allowed even a tempo to consolidate (see the note to move 8), so a
plan with the moves . . . Bb7, . . . e6, and . . . QaS in some order is now preferred. Of
course, the Benko games looked at in this game were all played from 1 968 to
1 97 1 , before theory had developed beyond infancy, much less become consoli
dated. And in any case most of the themes in these games are typical of other
S .e3 variations. 7.Nfl Here are some earlier games involving Benko:
a) 7 .Bc4 Bg7 8.Nge2 0-0 9.0-0 Nbd7 1 0.bxa6 Nb6 1 1 .BbS Bxa6 1 2 . Bc6 Rb8
1 3 .Rb l Ng4! intending . . . NeS and in some cases . . . Nc4, Blumin-Benko, Atlan
tic City 1 969;
b) 7 .bxa6 Bg7 (7 ... Bxa6 is an option, e.g., 8 . Qa4+ Nfd7 9.BbS Bg7 1 O.Nf3 0-0,
with fairly typical Benko play following ... Nb6 and . . . Nbd7.) 8.BbS + ! ? Nfd7 (1
don't like this much. 8 . . . Bd7 prepares for . . . Nxa6 and trading White's light-square
protector, while retaining pressure on e4 and dS .) 9.N ge2 ! 0-0 1 0.0-0 Nxa6 I l .e4
Nc7 1 2 .Bd3 NeS 1 3 .Bc2 Ba6 1 4.f4 Nc4 I S .Rf3 ! ? ( l S .b3 Nb6 1 6 . Bb2 100ks solid)
I S . . . e6! 1 6.dxe6 fxe6 1 7 .Rb 1 d S , and Black had gained a fair amount of play in

545

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

546

Kuijpers-Benko, Wijk aan Zee 1 970, although he later lost. 7 . . . Bg7 8.bxa6 ? ! A
key moment. It was later discovered that 8.a4! seriously limits Black's counterplay
while remaining a pawn ahead. 8 . . . 0-0 9.e4 Qa5 ! ? Benko expresses some skepticism about the necessity of this move, but it is the standard idea in such posi
tions. After . . . Bxa6 and Bxa6, Black can recapture with the Queen and either
prevent castling or exchange Queens after Qe2 . The latter happens in the game.
1 0.Nd2 Bxa6 I 1 .Bxa6 Qxa6 1 2 .Qe2 How else to get castled?

12 ... Nfd7 Benko points out in his notes that 12 . . . Nbd7 is also playable and in
deed, Farago-Benko, Vrnjacka Banja 1 970 saw that move. There followed 1 3 . Qxa6
Rxa6 1 4. Nc4 Rb 8 I S . f3 Ne8 1 6 .Kd2 Nc7 1 7 .Kd3 Ne S + 1 8 . Nxe S Bxe S
1 9.Kc2 fS ! , with unclear compensation (the game was drawn) . Benko apparently
felt, and it's easy to agree with him, that 1 2 . . . Nfd7 was attractive because it re
tained the idea of a Knight recapture on a6. 1 3 .a4! ? Benko cites 1 3 .Nc4 fS 1 4.f3 ?
fxe4 I S .fxe4 ( I S .Nxe4! ?) I S . . . Nb6 1 6.Nxb6 Qxb6, Kaufman-Benko, Aspen 1 968.
White cannot castle, whereas Black can play . . . Nd7-eS next. See Benko's notes
in Game 7 S . The alternative line 1 3 .Qxa6 Nxa6 1 4.0-0 Nb4 with control over
d3 and the b-file gives Black full compensation. 1 3 ... Qxe2 + 14.Kxe2 Na6 1 5.Rb l
f5 ! , Ih _ lh, Another case of Benko failing to pursue an advantage, a bad habit he
seemed to be attached to as Black. He mentions that Larsen among others thought
that "Black had more than enough compensation for the sacrificed pawn." In
deed, this type of ending is now notoriously good for Black, whose great activity
and open lines against White's weaknesses have led to win after win. Neverthe
less, games like this, against chess greats such as Portisch, were very influential
in converting many of the world's players to 3 . . . b S .

I n the early days, B enko twice faced a perfectly logical plan that has almost
disappeared from recent books, although ECO deals with it in some detail. White
manages to achieve e4 in one step without forfeiting castling, and Black must
play carefully to demonstrate full compensation. We follow Vranesic-Benko,
Toronto, 1 97 1 : l .d4 Nf6 2 .c4 c5 3 .d5 b5 4.cxb5 a6 5 .bxa6 Bxa6 6.Nc3 d6
7 .NO g6 8.Nd2 ! ?

BENKO AS BLACK

547

Here is White's idea: he will play e4 and then answer . . . Bxfl by Nxfl . 8 . . . Bg7
9.e4 O-O ! ? Benko had already tried 9 . . . Bxfl 1 0Nxfl versus Taimanov in Wijk
aan Zee 1 970. The negative result of that game may have influenced him to try
something else, but capturing on fl isn't bad.

The Taimanov game continued 10 . . . 0-0 ( 1 O . . . Qa5 was played by no less than
Kasparov versus Bareev in Linares 1 994: I l .Bd2 0-0 1 2 .Ne3 Nbd7 1 3 .0-0 Qa6!
1 4.Qc2 Ne5 1 5 .b 3 Nfd7 1 6.f4 Nd3 1 7 .a3 Bd4 I S .Kh I RfcS, with unclear pros
pects.) I l .Ne3 Na6 ! ? ( 1 1 . . .Nbd7 1 2 .0-0 Qb6 or 1 2 . . . Qa5 was later found to give
satisfactory play.) 1 2 .0-0 Nd7 1 3 .Qe2 Qc7 1 4.Bd2 ( 1 4.Nc4 is unpleasant for Black)
1 4 . . . Qb7 1 5 .Rab l Nc7 1 6.b3 e6 1 7 .a4! RfeS I S .Nc4 Qa6 1 9.QB , and White
was better. 1 0.Bxa6 Nxa6 1 1 .0-0 Nd7 1 2 .Nc4 A Benko game from the year
before had gone 1 2 .Qe2 Nb6 1 3 .NB Nc7 1 4.Rd l Na4 (A standard Benko Gambit
idea; once the c3 -Knight is exchanged, White's queenside becomes more vul
nerable.) 1 5 .Nxa4 Rxa4 1 6.b3 ! ? Bxa l 1 7 .bxa4 Bg7 I S . Qc2 Qd7 1 9.Bd2 RbS !
2 0 . Qc4 QcS 2 1 .Bg5 (2 1 .a 5 Qa6 ! ) 2 1 . . .Rb4 2 2 .Qc2 Rb2 2 3 . Qc4 Rb4, 112 _ 112,
Ghitescu-Benko, Siegen 1 970. 12 ... Nb6 1 3 .Ne3 Qd7 1 3 . . . Qc7 followed by . . . c4
and . . . Nc5 has also equalized. 14.a4 B enko discusses the good and bad points of
this move in his notes. The less committal 1 4.Bd2 Nc7 1 5 . Qc2 RfbS 1 6.Qd3
Na4 ! (that theme again) equalized in Tisdall-Gertarsson, Gausdal I 992 . 14 . . . Rfb 8
1 5 .Ra3 Benko prefers 1 5 .a5 Nb4 1 6 .Ra3 . 1 5 ... Bd4!, and Black may already be
better. See B enko's notes in Game 76.

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

548

B enko played a quite a number of games in what Pedersen calls the "Classical
Main Line", in which White forfeits castling by e4, allowing . . . Bxfl , and then
(normally) scurries over to g2 or h2 with his king. I'm not going to cover the
theory of this variation in great detail. For those interested, it is well worked out
in the books by Pedersen and by Jacobs and Kinsman (as well as in ECO) . But
here is a quick organization of some of Benko's games in this line, most of them
commented upon in this book: l . d4 Nf6 2 . c4 c5 3 .d5 b5 4.cxb5 a6 5 .bxa6
Bxa6 6.Nc3 d6 7.e4 7 .Nf3 g6 8 . e4 Bxfl 9.Kxfl Bg7 1 0 .h3 introduces the Kg l
h 2 version of White's idea, b y which he stays o ff the a8-h l diagonal (g2 can be a
dangerous spot if Black plays moves such as . . . Qc7-b7 and . . . e6).

1 0 . . . 0-0 1 1 . Kg 1 Na6 1 2 .Kh2 Qb6 1 3 .Qe2 Rfb8 , and Black had plenty of play in
Camara-Benko, Sao Paulo 1 97 3 - see Game 7 8 . 7 . . . Bxfl 8.Kxfl g6

9.g3 Another idea that used to be popular is 9.g4: 9 . . . Bg7 I O.Kg2 (Benko's 1 969
game versus Visier with the passive 1 0.f3 ? ! is given in Game 7 3 .) 10 . . . 0-0 1 1 .h3
(also slow) 1 1 . . . e6 ! ? ( 1 1 . . .Nbd7) 1 2 .dxe6 fxe6 1 3 .e5 ! Ne8 1 4. exd6 Nxd6 1 5 . Qe2 ?
( 1 5 .Nf3 ) 1 5 . . . Qd7 1 6.f4? Nc6 1 7 .Nf3 Nd4, and White's King was already very
exposed in the early Benko Gambit game Avram-Benko, USA 1 968.
Alternatively, 9.Nge2 Bg7 l O.g3 0-0 I l .Kg2 Qb6 1 2 .Rb l Na6 intending a quick
. . . e6 is seen in Game 7 3 . As indicated in the notes, an alternative is 1 1 . . .Nbd7
1 2 . Qc2 Ra6 1 3 . Rd l Qa8 1 4.b3 e6 ! W Schmidt-Conquest, Poznan 1 986. I like
this example because Black combines attack down the queenside files with pres
sure along the h l -a8 diagonal.

BENKO AS BLACK

9 ... Bg7 1 0.Kg2 0-0 I 1 .NB Nbd7

549

1 2 .Qe2 An unusual move. After the normal 1 2 .h3 ( 1 2 .Re l is probably the most
accurate move, as has only lately been established.), Benko's 1 2 . . . Nb6 is still con
sidered unorthodox, but has done quite well, as shown by Jacobs and Kinsman
in their book (without crediting Benko, of course) . Mter B .Re l Qd7, both the
Gligoric and Taylor games given in Game 79 are still relevant to theory. Benko's
notes to the Gligoric game are particularly insightful and revealing. 1 2 ... Qb6
1 3 .Rb l Rfe8 ! With this creative move, Benko prepares . . . c4 and . . . Nc5 . For
most players, 1 3 . . .Rfb8 would be automatic. 1 4.b3 Ng4 1 5 .h3 Nge5 1 6.Nxe5
Bxe5 1 7 .Bd2 e4! 1 8.Rhc 1 exb3 1 9.axb3 Qb4 20.Nd l Rxe2 , liz_ liz, Donner
Benko, Palma de Mallorea 1 97 1 . See Benko's comments in Game 79 regard
ing his small advantage in the final position. This is another example of his ten
dency to take draws as Black when he wouldn't do so as White. But it is also an
example of his extraordinary ability to find compensation, and often more, in
the opening that bears his name.

Q ueen's Gam bit Accepted

Naturally, any consideration of Benko defending against l .d4 should first re
fer to the B enko Gambit. But Benko wasn't the type of player to specialize in
one opening and never stray from it. Not only before but also after he "discov
ered" and played his gambit, Benko employed a remarkable range of other de
fenses. These included many Griinfeld and King's Indian defenses, but in addi
tion occasional forays into the Nimzo-Indian ind Queen's Indian defenses. With
the classical l .d4 d 5 , B enko played both the Slav Defense and Queen's Gambit
Declined. In the following game, Uhlmann-Benko, Buenos Aires 1 960, he
experiments with the Queen's Gambit Accepted, and ends up advancing the most
advanced theory that existed at the time: I .d4 d5 2 .e4 dxe4 3 .Nfl Nf6 4.e3 e6
5 .Bxe4 e5 6.0-0 a6 7.Qe2 b5 8.Bb3 Bb7 9.a4 Nbd7 All book so far. This
active move is the preferred solution for Black. 1 0.axb5 axb5 I 1 .Rxa8 Qxa8
1 2 .Nc3 b4 1 3 .Nb5 Qa5

PAL BENKO: M Y LIFE, GAMES AND COMPOSITIONS

550

Benko call this more aggressive than 1 3" .Qb8 but riskier. The verdict between
the two is unresolved to this day. Both 1 3" .Qa5 and 1 3 , . . Qb8 are still played,
1 3 , . .Qb8 more often, although Black has had clearly better results with 1 3 , . . Qa5 .
I think that the next note is critical in this regard. 1 4.e4 Benko mentions 1 4.Ne 5 ,

a very interesting and extremely rare move ( 1 4.e4 i s normal) . It i s even possible
that 1 4.Ne5 casts 1 3 , . .Qa5 into doubt. For example: 1 4, ..Nxe5 ( 1 4, . .Be7 ? ? 1 5.Nc4!;
1 4, . .Ba6 1 5 .Nc7+ ! and White must be somewhat better) 1 5 . dxe5 Nd7 , and now
a fascinating line is 1 6.Rd l , when 1 6, . .Nxe5 is given a " ? " by what little theory
there is due to 1 7 .Qd2 " ! " , but the ever alert computer finds the only move (and
refutation) right off: 1 7 , . .BB ! . Yeesh - don't you hate these things? Instead, White
should try the promising 1 7 .f4! with the idea f5 . I'm surprised that there hasn't
been more investigation of 1 4.Ne 5 . 1 4 . . . Be7 Capture of the e-pawn by 1 4, . . Bxe4
is punished by 1 5 .Ng5 according to ECO, but that would take a lot of proving. I
think that 1 5 .Bf4 is very strong instead. Similarly, ECO gives 1 4, . .Nxe4 1 5 .Ng5 ,
but once again I like 1 5 .Bf4 ! . In any case, it's hardly conceivable that Black would
capture the pawn.

BENKO AS BLACK

551

1 5 .e5 An example with 1 5 .d5 went 1 5 . . . exd5 1 6.exd5 0-0 1 7 .Rd l Ba6 1 8 . Bc4

Nb6 1 9.d6! Bd8 2 0.b3 Nxc4 2 1 .bxc4 Re8 , Ruzhele-Howell, Gelsenkirchen 1 99 5 ;

and now according t o ECO, 2 2 .Qd3 ! ? leads t o a small advantage for White, but
I'm doubtful that this is the case. 1 5 ... Ne4! This game might well have been
either the first or second with 1 5 . . . Ne4, and it's still considered equal. Writing
at the time, without the help of hindsight, Benko said "Better than 1 5 . . . Nd5
because 1 6.Bg5 would then be strong. " But in fact, 1 5 . . . Nd5 1 6 .Bg5 Bc6 ! has
fared very well, even better than 1 5 . . . Ne4. Probably the options 1 4.Ne5 and
1 5 .d5 have to be reexamined if White wants to get anything out of this line.
1 6.Bc2 ? ! 1 6.Rd l became the normal move later, with Black gaining equality in
most games. 16 ... Bc6. Now Black seems better. For the rest of this original game,
complete with other opening notes, see Game 34.

Semi-S lav Defense

As shown above with regard to the Queen's Gambit Accepted, Benko some
times entered "high theory" in openings he seldom played. The Queen's Gam
bit Semi-Slav Defense was a rare choice and played only in his early years as far
as I can make out. In any case, the following opening is conducted interestingly
by both sides and is still relevant: l .d4 d5 2 .c4 c6 3 .NB Nf6 4.Nc3 e6 5 .e3 In
Game 2 , Benko uses the "Botvinnik" System: 5 .Bg5 dxc4 6.e4 b5 7 .e5 h6 8. Bh4
g5 etc. I would only note in passing that, despite the note indicating earlier games
with this system, Botvinnik deserves its appellation because he was the first lead
ing player to employ it consistently: first in 1 94 1 , twice in 1 944, then in 1 945 , as
well as in two World Championship games (Bronstein 1 95 1 and Smyslov 1 954).
Benko's argument for the name "Benko Gambit" certainly applies to this case as
well. 5 ... Nbd7 6.Bd3 dxc4 7.Bxc4 b5 8.Bd3 Bb7

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

552

Benko says "At the time, this was a new idea," and I'm sure that it was from his
point of view. Interestingly, the Semi-Slav is one of the few modern dynamic
openings that already had a well-established body of theory, and S . . . Bb7 even
had its precedents in the late 1 9 3 0s, e.g., Kashdan played it. Wade (after whom
the variation is named) and Donner both used it in 1 950, a year before this game.
Today, it is an extremely popular alternative to S . . . a6. 9.0-0 9.e4 is most often
played, but the text has its supporters. 9 . . . b4 1 0.Ne4 c5 ! ? 1 0 . . . Be7 is the main
line here. Sometimes played is 1 O . . . Nxe4 1 1 . Bxe4 Be7, but White retains the
advantage. I I .N:xf6+ gxf6 1 2.Qe2 1 2 .e4 cxd4 1 3 .Nxd4 is the most frequent choice,
thought to give White a small advantage. 12 . . . Qb6

1 3 .a3 1 3 . Rd 1 might be answered by 1 3 . . . 0-0-0 1 4.a3 b3 ! ? , or here 14 . . . cxd4.

1 3 . . . Bd6 1 3" .cxd4 1 4.exd4 Bd5 is suggested by Pedersen, but White has a lot of

freedom here. 1 3 . . . RgS is unclear and hard to assess. 14.axb4 cxd4 1 5.exd4 Rg8
The opening is at an end, and Black seems to have enough active pieces to com
pensate for his weaknesses and lack of a shelter for his King. At this point, Benko
suggests 1 6.Be4. Instead, White blundered with 1 6.b5? and ran into the lovely
shot 1 6 . . . Qxd4! . See Game 1 6, Korody-Benko, Budapest 1 95 1 .

Q ueen's Gam bit Declined

The Queen's Gambit D eclined wasn't a real favorite of Benko's, but he used it
in some key contests, and insightfully played the variation that is now used by
most grandmasters. One example of this is Spassky-Benko, Winnipeg 1 967:

BENKO AS BLACK

l .d4 d5 2.c4 e6 3 .Nc3 Be7 4.NB Nf6 5 . Bg5 h6 6.Bh4 0-0 7.e3 b6

553

This is the Tartakower Defense, which served Benko well on the rare occasions
he employed it. When Benko first used it, of course, the setup had by no means
gained the popularity it did later. He succeeded in neutralizing some big-name
players with the Tartakower, for example, Benko drew Gligoric in 1 963 , Spassky
in 1 967, Larsen in 1 968, and Reshevsky in 1 969, with no losses that I can locate.
Given his difficulties with the black pieces versus elite players, he might have
done well to keep the Tartakower in his repertoire. S.Re l A couple of other
examples from Benko games:
(a) 8.Bd3 Bb7 9.0-0 Nbd7 1 O.Qe2 cS ( 1 O . . . Ne4 is also played) I l .Rae I dxc4
1 2 .Bxc4 Ne4 1 3 .Bg3 ( 1 3 .Bxe7 Qxe7 1 4.Nxe4 Bxe4 =; 1 3 Nxe4 Bxh4 1 4.Nd6 Bxf3 !
I S .Qxf3 cxd4 1 6. exd4 Qe7 is equal-among other factors, note the opposite
colored Bishops.) 1 3 . . .Nxg3 1 4.hxg3 cxd4 I S Nxd4 NeS 1 6 .Ba6 Bxa6 1 7 . Qxa6
BcS 1 8 .Rfd l Qc8 1 9 . Qe2 Qb7, '12_ '12, Reshevsky-Benko, Netanya 1 969.
(b) 8.cxdS NxdS 9.Bxe7 Qxe7 (For historical reasons, this line still uses up an
entire ECO code, yet it has faded in importance as Black has learned how to
equalize.) 1 0NxdS exdS I l .Re I Be6 1 2 .Bd3 (the move that to this day is consid
ered the most promising) 1 2 . . . cS 1 3 .dxcS bxcS 1 4.0-0

14 ... Rc8 ! ? (This move, unmentioned in ECO or Janjgava's recent book, has a re
spectable score in practice. Nevertheless, the main line with 14 . . . Nd7 I S .e4 d4 is
easier to play and doing excellently at present, in particular after 1 6.Bb l as ! , and
even 14 . . . aS is fully playable.) I S .e4 Qb7 ! ? ( 1 S . . . d4 isn't bad) 1 6.exdS BxdS 1 7.Re l ! ?
Nd7 ( 1 7 . . . Bxf3 1 8 .Qxf3 Qxf3 1 9.9xf3 will win a pawn) 1 8 .Ba6? ! ( 1 8.BfS Re8 !)

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

554

I S . . . Qxa6 1 9. Qxd5 Nb6 2 0.Qb3 c4 2 1 .Qc2 Qxa2 2 2 .Ra l Qb3 2 3 .Qxb3 cxb 3 ,
112 _ 112 Larsen-Benko, Aspen 1 965. Black has a n optical advantage after 24.Re3 Rc2
2 5 .Rxb3 Nc4 2 6.Rc3 ! Rxc3 2 7 .bxc3 , but White can just bring his King over to the
queenside. 8 . . . Bb7 9.Bxf6 Bxf6 1 0.exd5 exd5 I l .Be2 Qe7 1 2 .0-0 Nd7 Or
12 . . . RdS = . White's line is considered harmless. 1 3 .Qa4 a6 1 4.Bd3 e5

1 5 .Bb l ? ! White is wasting some time here. 1 5 ... Rfd8 1 6.Qe2 ? ! g6 1 7.Rfe l
Rae8 1 8.Red l ? White continues to mess around while Black expands. Better
was I S .Qd2 . 1 8 . . .b 5 ! 1 9.dxe5 Nxe5 20.Nd4 Ne4 2 1 .Qb3 Nxc3 22 .bxc3 Re4! ?
2 2 . . . Rc5 ! intending . . . RdcS and . . . Qc7 would have been very strong, probably
winning. 2 3 .Bd3 Ra4 24.Re2 Re8 2 5 .Re2 Qa3 ! ? Black should also consider
2 5 . . . Qc5 . 26.Rdc l Bxd4 27.exd4, 112 _ 112 . As noted before, Benko had a ten
dency to take draws as Black even if he was better; hence 2 6 . . . Bxd4. His ability
to outplay these elite players, however, demands our admiration.

T he Grunfeld Defense
The Griinfeld was a bread-and-butter d-pawn defense for Benko, especially
early on before the Benko Gambit came along. As I see it, Benko's Griinfeld
games are not full of sparkling innovations, but he plays solidly with confidence,
and can create play for himself. Here is a nice miniature in one of White's harm
less sidelines: l .d4 Nf6 2.e4 g6 3 .Nc3 d5 4.exd5 In another "safe" sideline, we
see the type of Griinfeld ending in which Benko excelled: 4.NB Bg7 5 .cxd5 Nxd5
6.Qb3 Nxc3 7 . bxc3 c5 S . e 3 0-0 9.Be2 Nc6 1 0.0-0

BENKO AS BLACK

1 O . . . Na5 1 1 .Qc2 (The preferred move is 1 1 . Qa3 , but this yields nothing after
1 1 . . .cxd4 1 2 .cxd4 Be6.) 1 1 . . .cxd4 1 2 .cxd4 Be6 (hitting c4) 1 3 .Rd 1 Qd7 1 4.Ne5
Rfc8! 1 5 .Qb2 Qa4 (Black's pressure on the light squares grows) 1 6.Qb5 Qxb5
1 7 .Bxb5 Nc4 1 8 .Bxc4 Bxc4 1 9.Nxc4 Rxc4 20.Bb2 Rc2 2 1 .Rab 1 b6, Kramer-Benko,
New York 1 96 1 , and Black has considerable pressure in the ending. He went on
to win. 4 ... Nxd5 5 .g3 Bg7 6.Bg2 Nb6 7.e3 0-0 8.Nge2 c6 9.0-0 e5 1 0.d5
cxd5 I l .Nxd5 Bg4! 1 2 .e4 Nc6 1 3 .h3 Be6 1 4.Be3 f5 ! 1 5 .Bxb6 axb6

Black has a very nice game and it's hard for White to find a move . But the end
ing is cute 1 6.Qb3 ? ? fxe4 1 7.Bxe4 Nd4 1 8 .Nxd4 exd4, 0- 1 . Soos-Benko,
Havana (Olympiad) 1 966. The move . . . Ra 5 will win a piece !
Here are a few examples of Benko playing against a system that he himself
used. We feature Petrosian-Benko, Curaao (Candidates) 1 962 : l .d4 Nf6
2 .c4 g6 3 .Nc3 d5 4.Nfl Bg7 5 .Bg5 Ne4 6.cxd5 Benko played 6.Bh4 here, as
discussed in our opening survey for White . 6 ... Nxg5 7.Nxg5 e6

8.Nfl 8.Qd2 h6 9.Nf3 exd5 1 O.e3 0-0 1 1 .b4 b6 ! ? (This move isn't mentioned in

ECO or in my database, but it doesn't seem bad.) 1 2 .Rc 1 Bb7 1 3 .Be2 a6 1 4.a4
c6 (so that b5 can be met with . . . c5) 1 5 .0-0 Nd7 1 6.Rfd 1 Re8 1 7 . Qc2 Rc8, with
a solid position (Black might follow up with . . . Nf8-e6), Saidy-Benko, New York
1 966. The game was drawn. 8 . . . exd5 9.e3 0-0 1 0.Bd3 b6 ! ? The same idea,
again unknown to theory. 1O . . . c6 is the normal move, as Benko himself played
against Saidy in New York 1 963 : 1 1 .h3 (" ? ! " ECO) 1 1 . . .Qe7 1 2 .0-0 Nd7 1 3 . Qc2
Nb6 ! ? 1 4.Na4 ! ? Nxa4 1 5 . Qxa4 f5 ! 1 6.Rfe 1 g5 ! (Now ECO says that Black is

555

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

556

clearly better; perhaps not, but at any rate, White will have a hard time on the
kingside.) 1 7 . Qc2 Kh8 1 8 .Nh2 Be6 1 9.Rab l as 2 0 .Rbe l Rf7 ! intending . . . Rg8 .
White can probably defend with accurate play, but in practice, that will be difficult. Indeed, Benko won with a very nice kingside attack involving a sacrifice of
the Exchange. 1 1 .0-0 eS 1 2 .Be2 Ne6 ! ? Perhaps better is 12 ... Be6 B .Re l Nd7
= . 1 3 .Re l exd4? A real concession: although Black retains his Bishop pair, he
cedes central squares and isolates his d-pawn. Instead, 1 3 . . . Be6 intending 1 4.dxc5
bxc5 1 5 .Na4 c4 was feasible due to the play down the open b-file and the strength
of the Bishop on g7 . 1 4.NbS ! Bb7 1 S .Nbxd4 Nxd4 1 6.Nxd4 Re8 1 7.Qa4 Rxe l
1 8 .Rxe l Qb8 1 9.93 Bxd4 20.Qxd4. This is pretty bad for Black, but Benko
ouplayed his famous opponent to weasel into a Bishop ending with only one
weakness. He managed to draw.

Nimzo-Indian Defense
It's surprising that B enko didn't use l .d4 Nf6 2 .c4 e6 more often, since it is
one of the most solid and positional of d-pawn defenses. He did play the Nimzo
Indian (3 .Nc3 Bb4) in about 4% of his database games versus l .d4, and the or
der 3 .Nfl d5 led to some Queen's Gambits Declined (he rarely played the Bogo
or Queen's Indian). In the Nimzo-Indian, Benko may have been put off by some
early defeats against lines such as 4.a3 and 4.Bg5 . He lost Nimzos to Petrosian
twice, for example, although in the game that follows it was not due to the open
ing. In fact, Petrosian-Benko, BledlZagreblBelgrade (Candidates) 1 9S9 shows
a modern positional method of playing against 4.Bg5 that only caught on many
years later: 1 .d4 Nf6 2.e4 e6 3 .Nc3 Bb4 4.BgS eS S .dS Bxe3 + ! 6.bxc3

6 . . . e S ! ? The right idea, but later it was decided that 6 . . . d6 and 7 . . . e5 was a better

order, with or without . . . h6 thrown in. 7.Re l ? ! Not a very useful move. Petrosian
is worried about 7 . . . Qa5, which can no longer be answered by 8 .Bxf6, but 7 .Nfl
d6 (7 . . . Qa5 8.Qc2 ! ? Ne4 9.Re l Nxg5 1 0.Nxg5 is awkward for Black.) 8 .Nd2 !
intending e3 and Bd3 is a known reorganization that probably favors White.
7 ... d6 8.e3 Nbd7 9.3 h6 1 0.Bh4 Now we're in a position that is a main line
today except that Re l is really a waste of time. Again, Benko demonstrates a
subtle understanding of a complex positional opening, in this case outmatching
Petrosian, who was a great genius of such positions. 1 0 ... gS Other good moves

BENKO AS BLACK

are 1 O . . . Qe7 and 1 O . . . e4. I l .Bg3 Qe7 1 2 .e4 Nf8 1 3 .Bf2 Ng6 1 4.Ne2 Bd7
I S .Rb l Rb8 ! In fact, Black could have played 14 . . . Kd8 and then . . . Kc7 j but it's
instructive that he resists . . . 0-0-0 and realizes that he needs the Rook for queenside
defense. 1 6.Be3

1 6 ... Kd8 ! 1 7.Qc 1 Kc7 1 8 .Qa3 Ra8 Now Black has consolidated and can slowly
organize an attack on the kingside. His own queenside is perfectly secure. This
. . . Kd8-c7 strategy in conjunction with . . . Bxc3 and . . . e5 was only rediscovered
many years later, and in my opinion this is Benko's most creative idea in the
Nimzo-Indian. The fact that he went on to lose the game (in fact blundering)
kept it from receiving the notice it deserved.

King's Indian Defense

Finally, we examine Benko's games with the King's Indian Defense ("KID"),
an aggressive and tactical opening that really doesn't seem to fit his positional
style. And yet he defended with the KID against 1 .d4 as often as he did with the
Griinfeld Defense. I think that Benko trusted the character of the tactics that
arise in this opening (perhaps due to his expertise with the King's Indian At
tack), and he also managed to find positional ways to proceed in several impor
tant variations.
We begin with a line of lesser importance, the Torre Attack, as played in
Trifunovic-Benko, Belgrade 1 964: l .d4 Nf6 2 .Nf3 g6 3 .BgS Bg7 4.Nbd2
cS A move that anticipates later theory. It provokes the following response, which
is only good for equality. S .Bxf6 Bxf6 6.Ne4 Bxd4 7.Nxd4 cxd4 8.Qxd4 0-0

557

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

558

With no weaknesses, Black has little to worry about. 9.c4 9.0-0-0 QaS (9 . . . Nc6
1 0.Qd2 dS ! ? with the idea 1 1 . QxdS Qc7 is also a promising line) 1 O.Kb l Nc6
I l .Qd2 Qxd2 1 2 .Rxd2 d6, and Black has the slightly more comfortable position.
9 ...Nc6 1 0.Qd2 d6 1 1 .e3 1 1 .Nc3 Be6 1 2 .e4 Qb6 was fine for Black in Spiridonov
Kasparov, Skara 1 980. 1 1 . ..Be6 1 2 .Be2 Qa5 ! ? Heading for another Queenless
middlegame ! Nevertheless, 1 2 . . . NeS ! was probably objectively better, for example,
1 3 .b3 ( 1 3 .Re l Rc8 1 4.b3 d S ; 1 3 . Qd4 QaS+ 1 4.Nd2 Nc6) 1 3 . . . dS 1 4.NgS dxc4
I S .Nxe6 fxe6 (threatening . . .Nd 3 +) 1 6. Qxd8 Rfxd8 1 7 .bxc4 Rac8 1 8 .Re l RcS !
with the idea of . . . b S or . . . Rdc8. This continuation would justify Black's opening
strategy and then some. 1 3 .0-0 Better 1 3 . Qxa S ! NxaS 1 4.b 3 , with a small ad
vantage. 1 3 . . . Qxd2 1 4.Nxd2 Rac8 1 5 .Rfc 1 1 S .b3 dS =. 1 5 . . . b6 ! ? I S . . . Rfd8 in
tending . . . dS would fully equalize. 16.NB Rc7 1 7.b3 Bg4 1 8.Nd4 Nxd4 1 9.Bxg4
Nc6 20.Rd l . White has a fairly standard slight advantage, but difficulties mak
ing progress. Benko was happy to maneuver for a while and he eventually won
when his opponent pushed too hard and created weaknesses.
In this old game, Platz-Benko, Budapest 1 95 2 , Benko demonstrates how, by
charting new paths, he was able to meet the g3 Fianchetto Variation of the King's
Indian with early positional pressure: l .d4 Nf6 2 . c4 g6 3 .g3 Bg7 4.Bg2 0-0
5 .Nc3 d6 6.NB Nc6 7.d5 Today 7 .0-0 ! a6 8.dS NaS 9 .Nd2 is considered the
most accurate order to get to the main line, but none of this was known at the
time. 7 . . . Na5 !

It's interesting that in the databases, the idea o f . . . Nc6-aS a s played here ex
ploded at about the time of this game. There were a few precedents, e.g., games
with obscure players as Black in 1 93 9 and 1 942 , and undoubtedly a few others,
although 6 . . . Nc6 was extremely rare until the 1 9 S0s. Moving to top-level play
ers, Borisenko tried . . . Nc6-aS in two games in 1 9S0, and in the same year the
idea was seen in the USSR Championship in Smyslov-Lublinsky. But it was in
1 9S 2 that numerous high-level games occurred, so it's hard to say whether Benko
inspired those efforts or if they occurred independently. See his comments in
Game 3 1 versus Fischer (Buenos Aires 1 960). In any case, Benko was once again
at the forefront of the development of a new and extremely important system.
Indeed, the ... NaS idea revived 6 ... Nc6 (now called the "Panno System", with
little justification), and it is now arguably the main line versus g3 in the King's

BENKO AS BLACK

Indian. I give the whole game because the finish relates directly to the opening.
8.Qd3 The preferred move in the first few years, since White didn't want to
expose himself to long-diagonal tactics after b 3 , or retreat by Nd2 . Benko later
lost a game after 8.b3 cS 9.Bb2 a6 1 O. Qc2 BfS ? ( 1 0 . . . Bd7 =) l 1 . e4! Nxe4 1 2 .g4!
Ng3 1 3 .gxfS Nxh l 1 4.Bxh l Qd7 I S .fxg6 hxg6, Pupols-Benko, Lincoln 1 969,
and now, although 1 6.Ne2 was fine and would have yielded an edge with accurate play, 1 6.Ne4! Bxb2 1 7 . Qxb2 was quite strong, for example, 1 7 . . . QfS ! ( 1 7 . . . bS
1 8 .Qc3 QfS 1 9.Nfd2 ! , and Black's queenside play is stalled whereas White's Queen
can run to the kingside to support an attack.) 1 8 .Nfd2 Qf4 ! ? 1 9 .0-0-0 favors
White in view of 1 9 .. .fS ( 1 9 . . . Qxh2 2 0 .b4! cxb4 2 1 . Qxb4) 2 0.Qc3 ! intending
20 . . . fxe4 2 1 .QxaS Qxf2 2 2 .Bxe4, with a big advantage.
The natural 8 .Nd2 cS 9.0-0 a6 transposes to the line which is played almost
exclusively today-see the note to 7 . d S . But with this order (7 .dS instead of 7 .00), Black can play 8 . . . c6 ! instead of 8 . . . cS, equalizing. All this was recognized
within a few years after this game. 8 ... c5 ! This looks self-evident to modern
eyes, but the inferior moves 8 . . . eS and 8 . . . b6 had been played in three earlier
games with this variation. 9.0-0 a6

1 0.Rb l Avoiding 1 0 .b3 ? BfS ! . 1 0 ... Qc7 1 O . . . Bd7 and 1 0 . . . bS ! ? l 1 .cxbS c4 are
more dynamic options. l 1 .b3 Bd7 1 2 .Bd2 Here 1 2 .Bb2 Rfb8 1 3 .Nd2 is about
equal. 12 . . . Rfb8 1 3 .a4 b 5 ! At least equalizing. Black doesn't even have to invest
a pawn to achieve this thematic break. 1 4.axb5 axb5 1 5 .cxb 5 ? A mistake. On
the other hand, I S .NxbS Rxb S ! 1 6.cxbS BfS not only regains the Exchange for
Black, but he also wins the d-pawn (or e-pawn, in the case of 1 7. e4). 1 5 . . . Nxb3 !
Platz probably missed this. Instead of losing the other b-pawn for nothing, he
vainly tries to sacrifice the Exchange: 1 6.Rxb 3 ? c4 1 7.Qb l cxb3 1 8.Qxb3 Qc5 !
1 9 .Rb l Ra3 2 0.Qb4 Qxb4 2 1 .Rxb4 Rxc3 2 2 .Bxc3 Nxd5, 0- 1 . A nice recom
mendation for . . . Nc6-aS !

In Bilek-Benko, Budapest 1 956 White plays a line of the Four Pawns At

tack that has only become popular in this decade, and Benko responds with a
thoroughly modern idea: l .c4 Nf6 2 .Nc3 g6 3 .e4 d6 4.d4 Bg7 5.f4 c5 6.dxc5
Qa5 7.Bd3 Qxc5

559

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

5 60

A standard opening position that only in the last few years has been considered a
serious winning try for White. 8.NB Nc6 9.Qe2 0-0 1 O.Be3 Qa5 1 1 .0-0 Nd7
1 2 .Rac 1 Nc5 1 3 .Bb l Bg4 1 4.Rfd l This remains a major book line today, al
though Black can also throw in . . . Rac8. 14 . . .Na4!

This is Kasparov's "new" idea from a very similar position in this variation. It
was also played recently by Aaron Summerscale and Tiger Hillarp Persson.
1 5 .Nxa4 Qxa4 1 6.Rd2 Qb4 The aforementioned Hillarp Persson game con
tinued with 16 . . . Rac8 instead. 1 7.Khl Rad8 1 8.b3 e5 1 9.Rcd l Nd4, l/z-Ih.
Black's strong Knight on d4 and poor position of the Bishop on b 1 yield ap
proximate equality. This game could have been played in 2 00 1 !
B enko played the King's Indian Defense a lot during the '40s, ' 5 0s, and early
'60s, but occasionally all the way up to the '80s. Here he chooses a particularly
sharp line versus the S amisch Variation: I .d4 Nf6 2 .c4 g6 3 .Nc3 Bg7 4.e4 d6
5.B 0-0 6.Nge2 Nc6 7.Be3 a6 8.Qd2 Bd7 ! ?

BENKO AS BLACK

56 1

Often played in the early days of the . . . Nc6 line. Today 8 . . . Rb8 or 8 . . . Re8 is
preferred. 9.Nc 1 In order to develop the King's Bishop. One reason that 8 . . . Bd7
is considered a bit slow is the dynamic 9.h4, but that is also complex and difficult
for both sides. 9 ... eS 10.Nb3 The other possibility is 1 O.d5 Nd4 1 1 .Nb3 Nxb3
1 2 .axb 3 . 1 0 ... exd4 I l .Nxd4 ReS 1 2 .Be2 NhS ! ?

I n his notes, Benko explains the virtues o f this move and h e i s o f course correct.
Nevertheless, White's position is solid and he has space. The objective assessment of 1 2 . . . Nh5 probably depends upon my suggestion in the next note .
1 3 .0-0 The notes to the game demonstrate how theory's 1 3 .Nxc6 is not so clear.
But 1 3 .Nc2 ! seems like a very good option since White protects e3 and prevents
exchanges or tactics on d4. Then we have a standard Maroczy Bind position in
which the moves . . . Nc6, . . . Bd7, and . . . Nh5 don't seem to coordinate. Black can
try 1 3 . . . Qh4+ 1 4.Bf2 Qf4, but White can avoid the exchange of Queens by 1 5 .Ne3 ,
when Ncd5 is a problem, as is 0-0-0 followed by a kingside pawn storm. Con
tinuing along this line, 1 5 . . . Qg5 is well answered by 1 6.g4! with the idea 16 . . . Nf4
1 7 .Ned5 Ng2 + 1 8 .Kd 1 Qxd2 + 1 9.Kxd2 and the g2 Knight is trapped, among
other problems. 1 3 ...Nf4! Benko's notes point out that 1 3 . . .Nxd4 1 4.Bxd4 Bxd4+
1 5 . Qxd4 Nf4 is also quite playable. 14.Bxf4 Nxd4 I S .BgS Nxe2 + 1 6.Nxe2
QbS ? ! . Preparing . . . b5 in some lines. Probably 1 6 . . .f6 1 7 .Be3 Be6 = was safer
however, because Benko points out that instead of 1 7 .Nc3 Be6 = , White could
now play 1 7 .c5 with some advantage. At any rate, the opening is now over, but
the game, Filip-Benko, Curaao (Candidates) 1 962 , is just beginning! See
Game 45 for its fascinating course.

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

5 62

Florian-Benko, Budapest 1 9S 1 is a fairly standard KID Exchange Variation,

but who would want to play against Benko in a Queenless middlegame? l .d4
Nf6 2 .c4 g6 3 .Nc3 Bg7 4.NB 0-0 S .e4 d6 6.Be2 eS 7.dxeS dxeS 8.Qxd8
Rxd8 9.BgS Re8 This is "book" so far. 1 0.0-0 A little slow. 1 0.0-0-0 and 1 0 .Nd5
are the critical moves. 10 ... c6 I 1 .Rad l Na6 1 2 .Rd6 Be6!

1 3 .NxeS ? ! The computer likes this, grabbing pawns, but 1 3 .Rfd 1 was better.
1 3 ... NcS 1 4.Bxf6 Bxf6 I S .f4 Rad8 1 6.Rfd l Rxd6 1 7.Rxd6 Bxe S ! 1 8 .fxeS Kf8
1 9 .b4 White is counting on the following tactic: 1 9 ... Nd7 2 0.Nd5? But other
wise . . . Nxe5 leaves Black with a positionally ideal game. 20 . . . cxd5 2 1 .cxd5 Or
2 1 .exd5 Bf5 2 2 .g4 Rxe 5 . 2 1 . . .Bg4! White must have missed this - easy to do
so! 2 2 .e6 The point is that after 2 2 .Bxg4 Nxe 5 , the move . . . Ke7 will trap the
Rook! 2 2 . ..Bxe2 2 3 .Rxd7 fxe6, and Black won shortly.

Our last game, van Scheltinga-Benko, Dublin (Zonal) 1 957, shows the other
side of the King's Indian-Benko attacks in the opening! It is fun to conclude
with such a venturesome effort: l .d4 Nf6 2.c4 d6 3 .Nc3 eS 4.NB Nbd7 5 .e4
g6 6.Be2 Bg7 7.0-0 0-0 8.d5 Nc5 A standard . . . Nbd7 King's Indian line has
arisen which can also occur via the "Petrosian System" in the Main Line (6 . . . e5
7 . d 5 Na6 8 . 0-0 Nc5). It is considered solid and equal. 9.Nd2

9 ... a5 1 0.Qc2 Bh6 1 O . . .Ne8 is also played, preparing .. .f5 . The idea of . . . Bh6 is

to exchange Black's bad Bishop for White's good one. I 1 .Nb3 Bxc 1 1 2 .Raxc 1
We're still in the (modern-day) books. 1 2 ...Nfd7 1 3 .Bg4 ! ?

BENKO AS BLACK

563

This odd-looking move aims to prevent . . .fS . Instead, 1 3 .Nd2 fS ( 1 3 . . .b6) 1 4.exfS
gxfS I S .f4 exf4 1 6.Rxf4 NeS is considered equal by theory. 1 3 . . . f5 ! Anyway!
14.exf5 Qg5 ! ? Admirably aggressive, if not very Benko-like ! Safer was 1 4 . . . Nxb3
I S .axb3 NcS (a position also reached by 1 4 . . . Nxb3 I S .axb3 fS) . Over a decade
later, this solution was used by Petrosian versus Bobotsov at Kapfenberg 1 970,
which continued 1 6 .Na4? ! ( 1 6.f4 Qh4 1 7.Bh3 BxfS 1 8 .BxfS RxfS is fine for Black)
1 6 . . . Na6 ! , and Black was already somewhat better.
Benko's 1 4 . . . QgS , even if second best, nevertheless required the courage to
sacrifice a few pawns for attack. 1 5 .Bh3 ? ! Trying to stay a pawn up, but I S .Be2 !
Nxb3 1 6.axb3 gxfS 1 7.f4! favors White to some extent, in view of 1 7 . . . exf4? 1 8 .NbS .
1 5 , . .Nxb3 1 6.axb3 Nc5 1 7.Nb5 Qe7

1 8.f4 ! ? White could temporarily win some pawns by 1 8.fxg6 Bxh3 1 9.9xh7+ Qxh7

2 0 . Qxh7+ Kxh7 2 1 . gxh 3 , but 2 1 . . .Nxb3 2 2 .Rc3 Nd2 is okay. Also interesting
would be 1 9 . . . Kh8 ! ? 2 0 .gxh3 Rf3 , with attack. 1 8,..Bxf5 1 9.Bxf5 Rxf5 ! 2 0.fxe5
Rxe5

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

5 64

2 1 .Nd4 Re3 Suddenly Black is better! 22 .Qd2 ? ! Rd3 2 3 .Qf4 Rf8 24.Qxf8+
Qxf8 2 5 .Rxf8+ Kxf8 26.Nb5 Rxb3 2 7.Nxc7 Rxb2 , and Black was winning easily.

(Photo by Demeter Balla.)

tte

d the 6"t ch,,, pcobl,m, ;n my Hf, ;n , "ud,nt m'gavn, 7AMZlnnk

(Our Flag) when I was about thirteen years old. I got interested and soon sent in
my first composition. As a beginner, I didn't know the special rules and require
ments that are so much a part of this chess "offshoot. " I recall submitting a two
mover in which the black Bishop was on h8 and the black pawn on g7 . The
columnist wrote to me and said that he couldn't accept it because the position
has to be legal- that it could really happen in a game. He sent some problems
to me from famous composers and I studied them. After that I tried again (see
problem 1 ) , and a year later one of my compositions actually won an award at
the yearly composers' tournament.
My composition "career" was put on hold when I got more involved with tour
nament chess - for a while I didn't have the time to create many new problems.
However, I still enjoyed solving them and, after the war, I won several solving
contests.
It was only many years later, in the United States, that I was able to let my
desire to compose run free. Already a seasoned composer, I started a column in
Chess Life that explored chess problems in several ways: well-known composi
tions were offered to the readers, my own work was published, and I explained
to the interested readership proper composition rules (this allowed them to bet
ter understand and appreciate the inner beauty of chess problems) .
Since you're about to be exposed to 3 00 problems, compositions, and puzzles,
many of them extremely subtle and difficult, I think it would be useful to list the
basic rules of problems here:

A problem should express one or more ideas known as

"themes. "

5 67

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

568

It must be original or at least an improvement on a known

theme (or a combination of several themes).

It must be sound- only the stated solution should work.

The following rules apply to the technical aspects:

Economy: Every piece must serve a purpose- the fewer

pieces you use, the better the composition.

Variations: The more "theme-related" variations you offer, the

better the composition.

Difficulty: The solver should be forced to work hard for the

solution, though this is not a top priority.

Naturally, anyone who aspires to be a great chess composer can't fall victim
to dogmatism. Chess problems display so much beauty that a dependence on
rigid rules would stifle their diversity. Nevertheless, the basic rules are impor
tant! The following anecdote about Wolfgang Amadeus Mozart sheds light on
this subject:
A young musician who longed to create beautiful music once asked Mozart
how to become a great composer. "It is first necessary to play an instrument
well, " said Mozart. "Of course, one must also learn the basic rules of composing
and be familiar with the work of all the great composers of the past. "
"But, you were only six when you began to write music! " interjected the young
mUSICIan.
Mozart replied: "That's true, but I never thought to ask anyone as you're do
ing now. "
Not everyone can be a Mozart, but it doesn't take any training or talent to
enjoy and appreciate his music. The same can be said of chess compositions.
Creating a top-notch problem is extremely difficult, but having fun with that
same composition is easy, relaxing and, hopefully, fulfilling.

I've never been particularly excited about this kind of problem. They always seemed
to be too easy to construct and too easy to solve. Perhaps it's precisely these facts
that have made two-movers the most popular kind of problem. In fact, there are
so many of them that almost every conceivable idea has been employed.
Though most players like problems with a minimal number of pieces, the short,
two-move solution often forces the composer to use a whole armada of men!
This adds to the complexity, inj ects a bit of confusion and enriches the whole
composition with many interesting variations.
Even though I'm not a fan of this problem "genre," I've still created quite a
few of them. Here I offer no less than forty-eight, all varying in degrees of diffi
culty. Most likely the highlight of this group is problem 7, where I further de
velop Sam Loyd's idea of the Organ Pipes (a formation where Black has Bishops
on c8 and f8 , and Rooks on d8 and e8. The idea is that each Bishop interferes
with each Rook and vice versa). Here I managed to double the Organ Pipe theme,
giving both sides the same setup of Bishops and Rooks.
I have to mention that the famous English two-mover specialist C. Mansfield
congratulated my effort. He wrote that he had also tried his hand at the Organ
Pipe and couldn't make it work. He added that besides me, the only other grand
master who could create good two-movers was the great Paul Keres !
Another two-mover I'm partial to is problem 1 0 , my Christmas Tree problem
(A glance will show you why it's called a Christmas Tree!). The Hungarian grand
master G. Paros told me that this problem is the best he's ever seen of its kind. I
recall starting to compose the Christmas Tree in the early evening and only fin
ishing when the morning's light began to pour through my window!
For comparison, I created problem 9 in a Hungarian composing contest (the
theme was mate by promotion). We were given a maximum of three hours to com
plete the job, and I managed to do it in two .
Two dozen of these two-movers were made especially for Chess Life magazine's
crossword puzzles. My friend Charles Schafer wrote the text and I, of course,
did the composing. I didn't care about originality or economy too much; instead
I concentrated on making them fit (a task that was far from easy!) and making
them as exciting and sharp as possible. The rest of these problems were written
for Blitz magazine, for Chess Life and for my own amusement.
A few Twins have been included in this two-move mate group. Each offshoot
of a Twin differs from the preceeding position by a change in the placement of
only one piece (or, on occasion, the addition or deletion of one piece).
Usually, when showing a Twin I will diagram the main position and then, un
der (b), will show the different piece placement that creates the Twin. Thus in
problem 3 0 (our first Twin), the original problem has a Knight on g3 . In the
Twin, we take the Knight off g3 and place it on d .
5 69

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

( 1 ) White mates in two.

(Zaszlonk, 1 942 .)
570

l .Nc4 ! , 1 -0.

(2) White mates in two.

(Commended, Magyar Sakkelet,
1 969.)
l .Kxd6 ! , 1 -0 . And not l .Qxa7 Bd3 ! .

(3) White mates in two.

(Commended, Pules, 1 97 2 .)
l .Qc6 ! , 1 -0.

(4) White mates in two.

(Chess Life, 1 972 .)
l .Re4 ! , 1 -0 . And not l .Bg2 Nc3 ! .

(5) White mates in two.

(Honorable Mention, Magyar Sakkelet,
1 97 3 .)
l .Qxe4! This leads to eight different promo
tions and mates. 1 -0.

TWO- MOV E MATES

(6) White mates in two.

l .Qb2 ! , 1 -0.

(7) White mates in two.

ORGAN PIPE.
(First Prize, Magyar Sakkelet, 1 974.)

l .Kgl ! , 1 -0. Lloyd's Organ Pipe theme doubled.

(8) White mates in two.

(Chess Life, 1 9 7 5 .)
1 3 .Qf4!, 1 -0.

(9) White mates in two.

(Chess Life, 1 9 7 5 .)
l .d7 ! , 1 -0.

( 1 0) White mates in two.

CHRISTMAS TREE.

(Chess Life, 1 97 5 .)
l .Qc5 ! , 1 -0.

571

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

(1 1 ) White mates in two.

572

If you place White's Queen on e l instead of

b3 , you get the same answer. Thus, this is a re
ciprocal theme problem. l .e4! , 1 -0.

( 1 2) White mates in two.

(Chess Life, 1 9 7 5 .)
l .d4! , 1 -0.

( 1 3 ) White mates in two.

(Chess Life, 1 9 7 5 .)
l .Qc2 ! , 1 -0.

( 1 4) White mates in two.

(Chess Life, 1 9 7 5 .)
l.B g l ! , 1 -0.

( 1 5) White mates in two.

(Chess Life, 1 976.)
l .Ra7 ! , 1 -0.

1WO- MOV E MATES

( 1 6) White mates in two.

(Chess Life, 1 976.)
573

l .Nf1 ! , 1 -0. This is actually a trick, because any

computer will tell you that there's an alternate
solution via l .e4 Kxf3 2 .0-0 mate. However, that
isn't correct since the beginning position could
not have been legally reached if White still had
the right to casde! In other words, the position
before our starting diagram likely had White's
Rook on f1 and Black's King on h2 or h3 . Then
Rh l + Kg2 takes us to the diagram, but of course,
casding would not be legal after l .e4 Kxf3 .

( 1 7) White mates in two.

(Chess Life, 1 976.)
l .Qe8 ! , 1 -0.

( 1 8) White mates in two.

(Chess Life, 1 976.)
l .Qxf4! , 1 -0.

( 1 9) White mates in two.

(Chess Life, 1 976.)
l .Qc1 ! , 1 -0.

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

574

(2 0) White mates in two.

(Chess Life, 1 976.)
l .Nb5 ! , 1 -0.

(2 1 ) White mates in two.

(Chess Life, 1 976.)
l .Nc7 ! , 1 -0.

(2 2) White mates in two.

(Chess Life, 1 976.)
l .Qe3 ! , 1 -0.

(2 3) White mates in two.

(Chess Life, 1 976.)
l .B g4!, 1 -0.

(24) White mates in two.

(Chess Life, 1 976.)
l .Qc 1 ! , 1 -0.

TWO- MOV E MATES

(2 5) White mates in two .

(Chess Life, 1 97 7 .)
l .Ne2 ! , 1 -0.

(2 6) White mates in two.

(Chess Life, 1 9 7 7 .)
l .Ne8 ! , 1 -0.

(2 7) White mates in two .

(Chess Life, 1 980.)
l .Qb4!, 1 -0.

(2 8) White mates in two.

(Chess Life, 1 980.)
l .Nd3 ! , 1 -0.

(2 9) White mates in two .

(Chess Life, 1 980.)
l .Ng3 ! , 1 -0.

575

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

(3 0) White mates in two.

576

TWIN.
(Honorable Mention, Magyar Sakkelet,

1 98 1 .)
1 .0-0-0 ! , 1 -0. It seems that White could have
also mated in two by l .Ng4 Kxf3 2 .0-0 mate.
However, this is a trick since the initial posi
tion isn't possible if White still retains the right
to castle kingside.
(b) Relocate Ng3 to c3 : l .Ng4 Kxf3 2.0-0 mate.

(3 1 ) White mates in two.

(Chess Lije, 1 98 1 )
l .Qe l ! , 1 -0.

(32) White mates in two.

(Chess Lije, 1 98 1 .)
l .d3 ! , 1 -0.

(3 3 ) White mates in two.

(Chess Life, 1 98 1 .)
l .Qb4!, 1 -0.

TWO- MOV E MATES

(34) White mates in two.

TWIN.

(Chess Life, 1 98 1 .)
l .Bb l ! , 1 -0.
(b) Remove b3 -pawn: l .Bb3 .

(3 5) White mates in two.

(Chess Life, 1 982 .)

l .Qa2 ! , 1 -0.

(3 6) White mates in two.

(Chess Life, 1 982 .)

l .Rg4!, 1 -0.

(3 7) White mates in two.

TWIN.

(Schach Ecke, 1 98 3 .)
l .Qhl ! , 1 -0.
(b) Ra7 now stands on b7 : l .Qal ! .

(3 8) White mates in two.

TWIN.

(Chess Life, 1 98 3 .)
l .Qb4 ! , 1 -0.
(b) Relocate the a7-pawn to a3 : l .Qa4 ! .

577

PAL BENKO: MY LIF E, GAMES AND COMPOSITIONS

578

(3 9) White mates in two.

(Chess Life, 1 984.)
l .Nb5 ! , 1 -0.

(40) White mates in two.

(Chess Life, 1 984.)
l .Bc l ! , 1 -0.

(4 1) White mates in two.

(Blitz, 1 99 5 .)
l .Rg7 ! , 1 -0.

(42) White mates in two.

(Blitz, 1 99 5 .)
l .Bb6! , 1 -0.

(43) White mates in two.

SET.

l .Qf5 ! , 1 -0. Incorrect is l .h6 KxB 2 .0-0 mate

because castling couldn't be legal based on the
initial position!
(2) Black to move: l . . . KxB

2.0-0 mate.

TWO- MOV E MATES

(44) White mates in two.

l .Re5 ! , 1 -0. As we've seen before, a solution that

calls for kingside castling (via l .e 3 KB 2 .0-0
mate) isn't possible because the initial position
can only be legal if White has lost the right to
castle.

(45) White mates in two.

TWIN.

l .Qal ! , 1 -0. This is known as a Queen four

corner theme.
(b) Bh2 relocates to e 5 : l .Qhl ! .

(46) White mates in two.

l .Qal ! , 1 -0. This demonstrates a Queen four

comer theme combined with a Knight wheel.

(47) White mates in two.

l .Bal ! , 1 -0.

(48) White mates in two.

l .Qd6+ ! , 1 -0.

579

When I first started composing chess problems, my interest was mainly directed
at compositions that were three moves deep or longer. One of them, number 5 3 ,
was the last problem I composed as a teenager. It was published many times be
cause its original setup appealed to problem-solvers and tournament players alike.
During the Lugano Olympiad, which Bobby Fischer attended as a spectator, I
made a bet with him that he couldn't solve it in thirty minutes. As time ran out,
he became irritated and demanded to see the answer. When I showed it to him,
he insisted that other solutions had to exist. Naturally, this led to another bet!
After more time passed, he was forced to settle both wagers. The teenage Pal
Benko never would have guessed how much mileage he was going to get out of
that little problem!
For those that find problem 53 difficult, let me give you a BIG hint (I'll put it
in the form of a riddle): Why did Bobby have so much trouble solving this prob
lem? Because he always played the Ruy Lopez instead of the Giuoco Piano.
I've found that this kind of problem (mate in three or more moves) is very
conducive to the creation of a Twin. Problem 5 2 is a Twin and problem 5 5 has
six Twin solutions. There are many other Twins, and also quite a few Excelsior
themes (a pawn standing on its original square that marches down the board and
promotes -problem 62 is an excellent example of a Twin Excelsior) .
I should add that the culmination of my work with Twins featured a problem
with 1 0 Twins (problem 82), and another that had a world record 40 Twins (prob
lem 8 3 ) ! I'm very proud of both of them.

(49) White mates in three.

l .Nd4 a3 2.Bb3 , 1 -0 .

(50) White mates in three.

(Third Prize, Zaszlonk 1 943 .)

l .Ra4! Now 1 . . .Ka6 2 .b4 Kb5 3 .Rxa5 is mate,

as is 1 . . .a6 2 .Rc4 a4 3 .Rc 5 . So Black might as
well take the Rook. 1 . ..Kxa4 2.Kc5 a6 3 .Nb6
mate.
58 1

PAL BENKO: MY LIFE, GAMES AND COMPOS ITIONS

582

(5 1 ) White mates in four.

(First-Second Prize, Zaszlonk 1 943 .)

l .Nd3 Be2 2 .Nh2 Bh5 3 .Kh3 This move is

the point of the problem, but the earlier moves
are crude. I later refined this concept in a three
move problem (see problem 5 5). 3 ... Kxc5 Of
course, L .Be2 allowed 4.Na4 mate, and L.Kxa5
is met by 4.Nc4 mate. 4.Na4 mate.

(52) White mates in four.

TWIN.
(Zaszlonk 1 943 and Chess Life 1 968.)

l .Bh4 Kc1 If 1 . ..Ka 1 , then 2 .Nd2 Kb2 3 .Ke3

Ka l 4.Bc3 is mate. 2.Nd2 Kh2 3 .Ke3 Kc1
4.Ba3 mate.
(b) Relocate the white King to e6: l .Kd5 Kat
2 .Na3 Kh2 3 .Kc4! Kxa3 Or 3 . . . Ka l 4.Bc3
mate. 4.Bc1 mate. This problem isn't easy to
solve!

(5 3 ) White mates in three.

(Chess Life 1 968.)

l .Bc4 Ke5 1 . . .Kf5 leaves White two roads to

Rome: 2 .Qf3 + and 2 .Qh5 +. 2.Qd5+ Kf6 3 .Qg5
mate. M. Grigoriev imitated this problem by
placing Black's King on h4. He published it in
64 in 1 98 2 , claiming a mate in two by l .Q d 5 .
Unfortunately, h e overlooked the cook l .Qd4+.

(54) White mates in three.

(Chess Life 1 968.)

l .Bh8! e3 2.KB ! Other mates are 2 . . . Kf5 3 .Nh4

mate and 2 . . . Bf6 3 .Nf4 mate. 2 ... Kxh5 3 .Bf7
mate. I like the economy of this problem.

THREE- ( AND MORE) MOV E MAT ES

(55) White mates in three.

TWIN: Six solutions.
(Magyar Sakkelet 1 969.)
This unusual problem features six Twin solu
tions with all possible mating batteries. l .Bgl
a5 2.Kf2 Or 2 .Rf2 Kd4 3 .Rf4 mate. 2 ... Kd4
3 .Kf3 mate.
(b) l .Rfl a5 2.Kf2 Kf5 3 .Ke3 mate.
(c) 1 .Rf8 a5 2.Bf7 Kf5 3 .Bd5 mate.
(d) 1 .Rf8 a5 2.Ndf7 Kf5 3 .Ng5 mate.
(e) l .Bg8 a5 2.Nhf7 Kd5 3 .Ng5 mate.

(56) White mates in eight.

(Magyar Sakkelet 1 972 .)

l .Na2 ! Kxa2 2.Be6+ Kal 3 .Ba2 ! Kxa2 4.Nf5

Kal 5 .Nd4 Ka2 6.Ne2 Kal 7.Ncl a2 8.Nb3
mate.

(57) White mates in seven.

(Chess Life 1 97 2 .)

l .Bg8 ! a3 2 .Nf7 Kxa2 3 .Ne5+ Kal 4.Ba2 !

Kxa2 5.Nd3 Kal 6.Ncl a2 7.Nb3 mate.

(5 8) White mates in six.

EXCELSIOR.
(Honorable Mention, Magyar Sakkelet
1 97 3 .)

l .b4 Kh8 2 .b 5 Kg 7 3 .b6 Kh8 4.b7 Kg 7

5.b8=B Kh8 6.Be5 mate. An Excelsior with
underpromotion!

583

PAL BENKO: MY LIFE, GAMES AND COMPOSITIONS

5 84

(59) White mates in six.

ANTI-EXCELSIOR.
(Magyar Sakkelet 1 97 3 .)

l .e4 KhS 2 .Ne5 Kg7 3 .Nf7

Kf8 Or 3 . . . Kf6

4 . h S = Q + Ke6 5 . Q e S + Kf6 6 . Q e 5 mat e .

4.hS=Q+ Ke7 5.QdS+ Ke6 6.Qd6 mate. The

previous problem was an Excelsior. This one,
the same problem with a minor adjustment, is
an anti-Excelsior.

(60) White mates in six.

EXCELSIOR.
(Die Schwalbe 1 97 3 .)

l .d4 KhS If 1 . . .KfS 2 .hS=Q+ Ke7 3 .QcS Kd6

4.g7 Kd5 5 .Qc5+ Ke4 6.Qe5 mate. 2.d5 Kg7
3 .d6 KhS 4.d7 Kg7 5.dS=B! KhS 6.Bf6 mate.
An Excelsior with only pawns.

(6 1 ) White mates in seven.

EXCELSIOR.
(Chess Life 1 9 7 3 .)

l .Qf6 Kh7 2.e4 KgS 3 .e5 Kh7 4.e6 KgS 5.e7

Kh7 6.eS=N! KgS 7.Q g7 mate. An ultra-min
iature Excelsior.

(62) White mates in six.

TWIN EXCELSIORS .
(Shahmat 1 97 3 .)

l .a4 KeS 2.a5 Kd7 3 .a6 KeS 4.a7 Kd7 5.aS=B

KeS 6.Bc6 mate.
(b) Relocate the a2 -pawn to g2 : l . g4 KeS 2 . g5

Kd7 3 . g6 KeS 4. g7 Kd7 5 .gS=N KdS 6.Nf6

mate. These two problems, variants of a shared
theme, are Twi n Excelsiors with different
underpromotions.

TH REE- ( AND MORE) MOV E MATES

(63 ) White mates in four.

(Chess Life 1 974.)

l .Rg8 ! Neither l .Rd8 Bd7 ! nor l .Rb8 Bb7 !

2 .Re8 Be4! allow a mate in four. 1 ... B g2 2.Kb3 !
BdS+ 3 .NxdS Kh l 4.Rg l mate.

(64) White mates in three.

TWIN.
( 1 974)

l .d8=B Kd6 2 .c8=R Ke6 3 .Rc6 mate.

(b) Relocate White's King to e3 : l .d8=Q KeS

2.c8=N! Ke6 3 .Qd6 mate.

(65) White mates in four.

(Second Prize, Vizngysc 1 976.)

1 .Nh8 Kxh8 Other moves don't help: 1 . . . Rxh2

2 .0-0 Rg2 + 3 .Kxg2 Kxh8 4.Rf8 mate; 1 . . .Rh7
2 .Rf1 ; 1 .. .Rh3 2 .Rgl + Rg3 3 .Rxg3 + Kxh8 4.Bb2
mate. 2.Rf1 Rg4 3 .Rf8+ Rg8 4.Bh2 mate.

(66) White mates in three.

(Third Honorable Mention, Deutsche
Schachzeitung 1 9 7 5 .)

l .g8=B Kg7 1 . . .Kg5 2 .h8=B ! Kh6 3 .Bf4 mate.

2.Be7 Kh8 Of course, 2 . . . Kh6 is met by 3 .h8=Q
mate. 3 .Bf6 mate.

(67) White mates in three.

(Shahmat 1 9 7 5 .)

l .e8=B KeS 2.f8=B ! A rare sight, White has

four Bishops, all working together perfectly.

2 ... Kxf6 3 .Bd4 mate.

585

PAL BENKO: M Y LIFE. GAMES AND COMPOSITIONS

(68) White mates in three.

TWIN .
586

(Third Honorable Mention, Magyar

Sakkelet 1 97 5 .)

l .h8=N! Ke6 2 .f8=N+ ! Kxf6 Or 2 . . . Kd6 3 .Rd7

mate. 3 .Nbd7 mate.
(b) Relocate b7-pawn to e7: l .e8=B Ke6 2.f8=B,
1 -0.
(69) White mates in four.
(Chess Life 1 97 7 .)

l .Kd7 Kb7 1 . . .Kxa7 2 .Kc6 2 .c8=R! Kh6 Or

2 . . . Kxa7 3 .Kc6 Ka6 4.Ra8 mate. 3 .a8=R! Kb7
4.Rch8 mate.

(70) White mates in six.

EXCELSIOR with TWIN: Four parts
and four promations.
(Magyar Sakkelet 1 97 8 .)

l . g4 Ke8 2 . g5 Kd7 3 . g6 Ke8 4. g 7 Kd7

5 .g8=N! Ke8 6.Nf6 mate.
(b) Relocate Kb7 to b 5 : l .a4 Ke8 2.a5 Kd7 3 .a6
Ke8 4.a7 Kd7 5.a8=B ! , 1 -0.
(c) Relocate e6-pawn to a3 : l .g4 Ke8 2. g5 Kd7
3 . g6 Ke8 4. g7 Kd7 5 . g8=R! , 1 -0 .
(d) Place White's King o n a6: l . g4 Ke8 2.g5
Kd7 3 . g6 Ke8 4.g7 Kd7 5.g8=Q!, 1 -0.
(7 1 ) White mates in three.
TWIN : Three parts.
(Chess Life 1 97 8 .)

l .Kc6 d4 2.Bh4! axh4 3 .Ra8 mate.

(b) Relocate Rb8 to a l : l .Kc6 d4 2.Kh7 ! Kh5
3 .a4 mate.
(c) Place White's Rook on h2 : l .Kc6 d4 2 .Kc5 !
Ka3 3 .Kh5 mate.

THR E E- (AN D MOR E) MOVE MATES

(72) White mates in five moves.

TWIN.
(Chess Life 1 979.)
l .Bb4 h4 2 .Be l Be3 3 .Rb l Bgi 4.Rd l Be3
S .Bf2 mate.

(b) Remove White's Bishop and place a Knight

on a3: l .Ra l ! h4 2 .Nh I Bd4 3 .Nc3 + Bgi
4.Nd l , when S .Nf2 mate can't be stopped.

(73) White mates in five.

TWIN.
(Magyar Sakkelet 1 980.)
l .e8=B dS 2 .BhS ! Bxf4 3 .Bg6+ Bh2 4.Bh7
Bg3 S.Bxe4 mate.

(b) Relocate White's e7-pawn to f7: l .f8=N!

dS 2.Nh7 Bxf4 3 .Nf6+ Bh2 4.NhS, when
S .Ng3 mate can't be prevented.

(74) White mates in three.

SWITCH-BACK.
(Chess Life 1 980.)
This problem shows a Switch-Back with limited
material.
l .Rb2 a6 Or 1 . . .Ka6 2 .Kc5 Ka5 3 .Ra2 mate.
2 .Rb l Ka4 3 .Ra l mate.

(75) White mates in three.

(Magyar Sakkelet 1 98 1 .)
l .RbbS ! d l =Q 2 .Nd7 ! Now 2 . . . Qxd5 3 .Rb6
mates, while 2 ... Qd4 3 .Nb8 is also mate. 2 ... Qg4
3 .RbcS mate.

5 87

PAL B E N KO: MY LIFE, GAM ES A N D CO M PO S ITIO N S

5 88

(76) White mates in five.

T WIN.
(Chess Life 1 982).
l .Rb l 5 2 .Bdl Be3 3 .Be2 + Bgl 4.Bfl Be3
5 .Bg2 mate.

(b) Turn White's Bc2 into a Nc2: l .Na3 5

2 .Nh l Bd4 3 .Nc3 + Bgl 4.Nd l , when 5 .Nf2
mate can't be stopped.
(77) White mates in four.
(Magyar Sakkvilag 1 984.)
l .Bb5 ! axb5 If 1 . . .fxe4 2.Be8 Nf6 3 .Bf7+ Ng8
4.Rxg8 mate. 2 .Bb6 b4 3 .Bd8 Nf6 4.Bxf6 mate.

(78) White mates in three.

(Blitz Chess 1 996).
l .Be5 Bd5 1 . . .Bxb3 2.Rxb3 Kc4 3 .Rc3 mate.
2 .Rc4+! Bxc4 3 .b4 mate.

(79) White mates in four.

( 1 998)
l .Ra2 l .b8=Q Rxh3 . 1 . . .Rxh3 2 .e8=R! Re3
3 .Rxe3 Kgl 4.Re l mate. T he A. Spethmanan
problem has more material and no Zugzwang.

(80) White mates in nine.

l .Rg6 Kh7 2 .Rg4 Kh6 3 .Kf2 Kh5 4.Kg3 Kh6
5 .Kh4 Kh7 6.Kh5 Kh8 7.Kg6 Kg8 8.Rf4 Kh8
9.Rf8 mate.

THR E E- (AN D MOR E) MOVE MATES

(81 ) White mates in five.

TWIN.
( 1 999)
l .Rh8 g5 2 .Nh7 Be5 If 2 . . . Bg3 then 3 .NxgS+
Bh2 4.Nh3 forces mate via S.Nf2. 3 .Nf6+ Bh2
4.Nh5 Be5 5 .Ng3 mate.

(b) Relocate Nf8 to 0: l .Rh7 ! g5 2 .Nh6 Bg3

3 .Nf5+ Bh4 4.Nxh4 gxh4 5.Rxh4 mate.

(82)

White mates in three.

TWINS: Ten parts!
(First Prize Magyar Sakkelet 1970.)

Each position differs from the preceeding by a

change in the placement of one piece. For easy
reference, the changed piece is underlined.
l .Re6 Kd4 2 .Nf7, 1 -0.

(b) Wh: Kb4, Re7, BfS, NgS. Bl: Kd4. l .Nf3 +

Kd5 2 .Kb5 , 1 -0.
(c) Wh: Kb4, Re7, Bg2, NgS. Bl: Kd4. l .Re 1
Kd3 2 .Be4+ , 1 -0.
(d) Wh: Kb4, Re7, Bg2, Ng6. Bl: Kd4. l .Re2
Kd3 2 .Nf4+, 1 -0.
(e) Wh: Kb4, Re7, Bg2, Nf6. Bl: Kd4. l .Re2
Kd3 2.Bfl , 1 -0.
(t) Wh: Kb4, Re7, Bg2, Ne3 . Bl: Kd4. 1 .Nc4
Kd3 2 .Re l , 1 -0.
(g) Wh: Kb4, Re7, Bb3 , Ne3 . Bl: Kd4. l .Nc4
Kd3 ( l . . . KdS 2.Bc2) 2 .Kc5, 1 -0.
(h) Wh: Kb4, Re7, BbS, Ne3 . Bl: Kd4. l .Nf5+
Kd5 2 .Ka5 , 1 -0.
(i) Wh: Kb4, Re7, BbS, Nd8. Bl: Kd4. l .Nf7
Kd5 2 .Re3 , 1 -0.
(j) Wh: Kb4, Re7, BbS, Nd8. Bl: KdS. l .Re6
Kd4 2 .Nc6+, 1-0.

589

PAL B E N KO: MY LIFE, GAM ES A N D CO M POSITIO N S

5 90

(83 ) White mates in three.

TWINS: Forty parts!
(First Prize, Magyar Sakkelet 19 71)
Each position differs from the preceeding by a
change in the placement of one piece. For easy
reference, the changed piece is underlined.
1 .Qb4 Kd5 2. B, 1-0 .

(b) Wh: Kg5, Qb3, Nf5, P-f2. Bl: Ke5 . 1 .Qb7

Ke6 2 . B , 1-0 .
(c) Wh: Kg5, Qda, Nf5, P-f2 . Bl: Ke5 . 1 .B Ke6
2 .Kg6, 1-0 .
(d) Wh: Kg5, Qd8, Nb8, P-f2 . Bl: Ke5 . 1 .B Ke6
2.Qd7+, 1-0 .
(e) Wh: Kg5, Qd8, Nc 1 , P-f2. Bl: Ke5 . 1 .Qd7
Ke4 2 .Nd3 , 1-0 .
(t) Wh: Kg5, Qd8, Nfl , P-f2 . Bl: Ke5 . 1 .Qd7
Ke4 2 .Nd2 +, 1-0 .
(g) Wh: Kg5, Qd8, Nfl , P-f2. Bl: Ke4. 1 .Ne3
Kf3 2 . Qd2 , 1-0 .
(h)Wh: Kg5, Qhl , Nfl , P-f2. Bl: Ke4. 1 .Ne3
Kf3 ( 1 . . .Kd4 2.Qc4+) 2 .Qa2 , 1-0 .
(i) Wh: Kg5, Qb3, Ng l , P-f2. Bl: Ke4. 1 .Qb5
Kd4 2 .Ne2 + , 1-0 .
U) Wh: Kg5, Qb3, Nf8, P-f2. Bl: Ke4. 1 .Nd7
Kd4 2 .Kg4, 1-0.
(k) Wh: Kd6, Qb3 , Nf8, P-f2. Bl: Ke4. 1 .Ke6
Kd4 2 .Nd7, 1-0.
(1) Wh: Kg4, Qb3 , Nf8, P-f2. Bl: Ke4. 1 .f4 Kd4
2 .Ne6+, 1-0 .
(m) Wh: Kg4, Qb3, Nb7, P-f2. Bl: Ke4. 1 .Qd l
Ke5 2.Qd6+, 1-0 .
(n) Wh: Kg4, Qb3 , Na3 , P-f2. Bl: Ke4. 1 .Nb5
Ke5 2 .Qf7, 1-0.

(0) Wh: Kg4, Qb3, Nh3 , P-f2 . Bl: Ke4. 1 .Nf4

Ke5 ( 1 . . .Kd4 2 .Ne6+) 2 . Qb6, 1-0 .
(p) Wh: Kg6, Qb3, Nh3 , P-f2. Bl: Ke4. 1 .Qdl
Ke5 2 .Ng5 , 1-0.
(q) Wh: Kg5, Qb3, Nh3 , P-f2. Bl: Ke4. 1 .Qd l
Ke5 2 .Nf4, 1-0 .
(r) Wh: Kg5, Qb3, Ng7, P-f2. Bl: Ke4. 1 .Kf6
Kd4 ( 1 . . . Kf4 2.Qe3+) 2 .Ne6+, 1-0 .

THREE- (AN D MOR E) MOVE MATES

(5) Wh: Kg5, Qb3, Ne8, P-f2. Bl: Ke4. l .Qdl

Ke5 2 .Ne7, 10.
(t) Wh: Kg5, Qb3, Ne8, P-f4. Bl: Ke4. l .Qc3
Kd5 2 .f5, 10.

(u) Wh: Kg5, Qb3 , Ne7, P-f4. Bl: Ke4. l .Kg4

Kd4 2 .Qe2 , 10.
(v) Wh: Kg5, Qb3, Nd l , P-f4. Bl: Ke4. l .Kg4
Kd4 2 .Qb5, 10.
(w) Wh: Kg5, Qb3, Nf5, P-f4. Bl: Ke4. l .Kg6
Kxf4 2 .Qe3+, 1-0.
(x) Wh: Kg6, Qb3 , Nd8, P-f4. Bl: Ke4. l .Qc3
Kxf4 ( 2 ...Kd5 3 .Qe2) 2 .Qe3 + , 1-0.
(y) Wh: Kg6, Qb3, Nc2, P-f4. Bl: Ke4. l .Kf6
Kxf4 2 .Qh3 , 1-0.
(z) Wh: Kg6, QdI, Nc2, P-f4. Bl: Ke4. l .Kg5
Kf3 2 .Ne 1 +, 1-0.
(aa) Wh: Kg6, Qd2, Ne l , P-f4. Bl: Ke4. l .Kh5
Kf5 2 .Qd6, 1-0.
(bb) Wh: Kg6, Qd2, Ne5, P-f4. Bl: Ke4. l .Kg7
Kf5 2 .Qd4, 1-0.
(cc) Wh: Ke2, Qd2, Ne5, P-f4. Bl: Ke4. l .Qc1
Kf5 ( 1 ...Kd5 2 .Qe6+) 2 .Qe4, 1-0.
(dd) Wh: Kd6, Qd2, Ne5, P-f4. Bl: Ke4. l .Kd7
Kf5 2 .Qb4, 1-0.
(ee)Wh: Kd6, 00, Ne5, P-f4. Bl: Ke4. l .Qg3
Kf5 2 .Kd5 , 1-0.
(f) Wh: Kd6, Qf2, Ne5, P-f4. Bl: Kf5. l .Kd5
Kf6 2 .Qa7, 1-0.
(gg) Wh: Kd6, Qgl, Ne5, P-f4. Bl: Kf5 . l .Nd3
Kf6 2 .Qg8, 1-0.
(hh) Wh: Kd6, Qg2, Nh7, P-f4. Bl: Kf5 . l .Kd5
Kxf4 2 .Kd4, 1-0.

(ii) Wh: Kd6, Qg2, Nd7, P-f4. Bl: Kf5 . l .Qg3

Ke4 2 .Ne5 + , 1-0.
(jj) Wh: Kh5, Qg2, Nd7, P-f4. Bl: Kf5 . l .Qe2
Kxf4 2 .Kh4, 1-0.

(kk) Wh: Kg5, Qg2, Ng6, P-f4. Bl: Kf5. l .Qe2

Kf6 2 .Qe7 +, 1-0.
(ll) Wh: Kfl , Qg2, Ng6, P-f4. Bl: Kf5 . l .Qe6
Kg4 2 .Qd5, 1-0.
(rnm) Wh: Kg8, Qg2, Ng6, P-f4. Bl: Kf5 . l .Qe2
Kf6 ( 1 ...Kxg6 2 .Qe5) 2 .Qe5+, 1-0.
(nn) Wh: Kg8, Q.bQ, Ng6, P-f4. Bl: Kf5 . l .Qe3
Kf6 ( 1 ...Kg4 2 .Kf7) 2 .Qe5+, 1-0.

59 1

If you're looking for a complicated explanation as to what a letter problem is,

allow me to ease your fears. These compositions are nothing more than prob
lems (usually mate in three, but I've also sprinkled in some mates in four, endgames
and helpmates) that form the shape of a letter or number. Thus, problem 84
looks like the letter "A," problem 85 is the letter "B," and so on. In fact, the
reader can find the whole alphabet represented here !
I originally got interested in this type of "visual" problem when I couldn't de
cide what to give my friends on special occasions. The ability to create chess
problem initials (for letterheads, cakes, etc.) or even whole messages (let's not
forget birth dates and monograms!) has given a lot of people pleasure over the
years (I've often needed several particular letters, so I created quite a few re
serves-even in twin form!).
My favorite letter-problem memory took place on Max Euwe's seventy-sev
enth birthday in Brazil. The highlight of the party was a fancy cake that was
topped with two frosting chessboards. The positions on the chessboards, made
up of chocolate pieces, featured Euwe's monogram (an "M-problem" and an "E
problem") in the form of two letter problems.
The crowd loved the effect, but Euwe was put on the spot when my condi
tions were announced: he had to solve both problems before anyone could eat
the cake. Max solved the "M" (problem 108) very quickly but, for whatever rea
son, he got stuck on the "E" (problem 90), which consists of two Queen sacri
fices. After a while I had to whisper the solutions so that he could avoid embar
rassment. Then we all devoured the cake!

(84) White mates in three.

Letter A
l .b7 Kc6 2 .Ka4 b3 3 .b8=N mate.

(85) White mates in four.

Letter B.
l .Ne2 Kh5 2 .Kh3 Kg5 3 .f4+ Kh5 4.Ng3 mate.

593

PAL B E N KO: MY L I F E, GAM ES A N D CO M POSITIO N S

5 94

(86) White mates in three.

Letter B.
l .b3 Nxc3 l...BbS 2.R6xbS+ NxbS 3.Rxa4 mate;
1 ...axb3 2.Ra4+ Kxa4 3.Rxa6 mate. 2. R4b 5 +
Bxb5 3 .b4 mate.

(87) White mates in three.

Letter C.
l .Bd4 Kg2 2 .Nf1 + Kf3 There's no salvation
in either 2 ... Kxfl 3.Rgl mate or 2 ... Kh3 3.Rg3
mate. 3 .Nd2 mate. The next problem is a twin
of this one: everything has been pushed one file
to the right, and moved one rank (towards the
White starting position) up the board.

(88) White mates in three.

Letter C.
l .Kh5 Or 1 ...Kg I 2.Rg6+ Kh l 3.Rfl mate.
1 .. .Khl 2.Rg3 Kxh2 3 .Kg4 mate.

(89) White mates in three.

Letter D .
l .Bxd4 Ka4 2 .Bc3 d4 3 .Ne5 mate.

(90) White mates in three.

Letter E.
l .Ne4 Ne6 If 1 ...cS then 2.QxcS+! NxcS 3.Nd4
mate. 2 .Qxa4+ Kxa4 3 .Bxe6 mate.

LETT ER PROBLEMS

(9 1 ) White mates in three.

Letter E.
l .Bgl Kg3 2 .Kxg6 Kf4 Of course, 2 . . . Kh4
3.Bf2 is mate. 3 .Bh2 mate. The next problem
is a twin of this one: everything has been moved
one file to the right.

(92) White mates in three.

Letter E.
l .Bfl hS 2.Bd3 h4 3 .BxfS mate.

(93) White mates in three.

Letter F.
l .Ng6 Kg4 2 .Nf2 + KgS 3 .ReS mate.

(94) White mates in three.

Letter F.
l .Nf3 + KfS 2 .BhS Ke4 3 .Bg6 mate.

(95) White mates in three.

Letter G.
l .NeS Kxe4 2 .e3 KdS 3 .Bxc6 mate.

5 95

PAL B E N KO: MY LIFE, GAM ES AN D CO M POSITIO N S

596

(96) White mates in three.

Letter H.
l .ReS Kxe7 2 .Re7 + Kb6 It's still mate after
2 . . . Kd6 3.Ne8 mate. 3 .NeS mate.

(97) White mates in three.

Letter H.
l .RdS Kxf7 L.Kxh6 2.ffi=Q+ Kg6 3.Qxf6 mate.
2 .Rd7+ Ke6 3 .NfB mate.

(98) White mates in three.

Letter I
l .QeS KgS If L.Kf7 2.Qe7+ Kg8 3.Qg7 mates.
2 .Ke4 Kg4 Or 2 ... Kf6 3.Qe7 mate. 3 .QgI mate.

(99) White mates in three.

Letter I.
l .Qb2 Kh6 2 . QhS+ KgS 3 . Qh4 mate.

( 1 00) White mates in four.

Letter I.
l .BfS KB 2 .Bf4 Kg2 3 .Bg4 Khl 4.Bf3 mate.

LETT ER PROBLEMS

( 1 0 1 ) White mates in three.

Letter ].
l .Ba4 Kd3 2 .Re! Kc3 3 .Re3 mate.

( 1 02) White mates in three.

Letter ].
l .BhS Kd4 2.Kh4 KdS 3.Rd2 mate.

( 1 03) White mates in three.

Letter K.
l .QfS Kc4 If 1 . . .Nb6, then 2.Qe5 ! forces mate
next move, while 1 . . .c2 2. Qxc2 also ends the
game in one more move 2.QdS+ Kh4 3 .Qd4
mate.

( 1 04) White mates in three.

Letter L.
l .Ra6 Kh3 2 .Na 1 + Kh2 Or 2 ... Kc4 3.Ra4 mate.
3 .Ra2 mate.

( 1 05) White mates in one.

Letter L.
l .Re4 mate. The next two problems show twins
in ascending order: first a mate in two, then a
mate in three.

5 97

PAL BEN KO: MY LI FE, GAM ES A N D CO M POSIT IO N S

59 8

( 1 06) White mates in two.

Letter L.
l .Ba4! Kxc3 2 .Rc2 mate.

( 1 07) White mates in three.

Letter L.
l .Ra2 Kb3 2 .NaS+ Kb4 3 .Ra4 mate.

( 1 08) White mates in three.

Letter M.
l .Nf2 Ke3 2 . KI l KxB 3 .d4 mate.

( 1 09) White mates in three.

Letter N.
l .Bc4 Bg4 1 ...Bg3 2.Bxe2 and 3.Rd 1 mate can't
be stopped. 2 .Rd 1 + Bxd l 3 .Nb l mate.

( 1 1 0) White mates in three.

Letter N.
l .Bd4 Bg3 Or 1 ...Bh3 2.Bxf2 BfS 3.Re 1 mate.
2 .Re 1 + Bxe 1 3 .Nc1 mate.

LETT ER PROBLEMS

( 1 1 1) White mates in three.

Letter O.
l .Bd5 cxd5 If 1 ...KbS then 2.Bxc6+ KaS 3.b4
is mate. 2 .Nc3 d4 3 .b4 mate.

( 1 1 2) White mates in three.

Letter O.
l .dS=R Kf6 2 .RgS Ke6 3 .Rg6 mate.

( 1 1 3) White mates in three.

Letter O.
l .Nab6 Kb5 2 .NcS+ Kc6 2...Kxa6 3.Rb6 mate.
3 .Ne7 mate.

( 1 1 4) White mates in three.

Letter P.
l .Re3 hxg5 2 .Be6+ Kxf4 3 .Nd5 mate.

( 1 1 5) White mates in three.

Letter P.
l .b7 Bd5 2 .bS=N BO Or 2...Bf7 3.Nc6 mate.
3 .Nc4 mate.

59 9

PAL B E N KO: MY LIFE, GAM ES A N D CO M POSITIO N S

600

( 1 1 6) White mates in three.

Letter Q.
l .Rb8 Nxb8 Both 1 ...NxcS+ 2.RxcS+ Ka6 3.Ra8
and 1 ...Nh4+ 2.Rcxh4 Ka6 3.Ra8 also lead to
the announced mate in three. 2 .Nc3 Nxc6
3 .Ra4 mate.

( 1 1 7) White mates in three.

Letter R.
l .Nc 1 Kc6 No better is 1 ...KaS 2.Kc4 b3
3.N1 xb3 mate. 2 .Kc4 b3 3 .d5 mate.

( 1 1 8) White mates in three.

Letter S.
l .Bh5 Kxh4 Or l ...Kh2 2.Qf3 Kg1 3.Qg2 mate.
2 . QB Kxg5 3 .Qg4 mate.

( 1 1 9) White mates in three.

Letter S.
1 .Kf3 Kxf6 2 .Nf5 Kg5 Or 2...KeS 3.Bd4 mate.
3 .Bh4 mate. I created two twins here: in the
next problem I push this exact position one file
to the right. In the one after that, I switched
the f2-Bishop and g3 Knight (thus having a
Knight on f2 and a Bishop on g3).

( 1 20) White mates in three.

Letter S.
l .Rf6+ Kxg4 2 .Bg5 Kh5 3 .BB mate.

LETT ER PROBLEMS

( 1 2 1 ) White mates in three.

Letter S.

60 1

l .Rc6! Kd4 2 .Ne6+ KdS 3 .Be4 mate.

( 1 22) White mates in three.

Letter T.
l .f8=Q Kxh7 2 .Bf6 ! gxf6 3 .BfS mate.

( 1 23) White mates in three.

Letter T.
l . Qa7 KxeS 2.Ke3 and now, no matter what
move Black chooses, White mates with 3.Qe7.

( 1 24) White mates in three.

Letter U.
l .Rb4+ Nxb4+ 2.Bxb4 NdS 3 .Nb2 mate.

( 1 25) White mates in three.

Letter V.
l .Rdb4 KdS 2 .Ke7 ! KxeS Or 2 . . Kc5 3.R6b5
mate. 3 .R6bS mate.
.

PAL B E N KO: MY LIFE, GAM ES A N D CO M POSITIO N S

602

( 1 26) White mates i n three.

Letter W.
l .Na l ! Kc3 2.Ke4 d3 3 .Be5 mate.

( 1 27) White mates in three.

Letter X.
l .f8=Q Kxh7 2 .Kg4 Kg6 3 . Qg8 mate.

( 1 28) White mates in three.

Letter Y.
l .Rg6 Ke5 Or 1 . . .Kf4 2.Rg4+ KeS 3.Re6 mate.
2 .Rde6+ Kf4 3 .Rg4 mate.

( 1 29) White mates in three.

Letter Z.
l .g7 Kh3 2.g8=B Kg4 3 .Be6 mate.

( 1 30) White mates in three.

Letter Z.
l .g7 Kh3 Or 1 . . .KfS 2.NgS+ Kg4 3.Rf4 mate.
2 .g8=B Kg4 3 .Be6 mate.

LETTER PROBLEMS

( 1 3 1 ) White to move and win.

Number One.
This position is far trickier than one might imag
ine. l .Kc5 Bg3 2 .Nc6 Bf2 + 3 .Ncd4! The only
way to win! The reasonable looking 3.Nbd4? ?
only leads to a draw after 3 . . . Ka6 ! 4.Nb4+ Kb7
S.NdS (S.Nc2 Bgl , ) S . . . Bgl 6.Kc4 Bf2 7.Kd3
Bgl 8.Ke4 Bf2, and White isn't getting any
where. 3 ... Bg1 4.Nd6+ Kb8 5.Kd5, and White
wins easily.
=

( 1 32) White mates in three.

Number One.
l .Rd8 Kf5 2 .Rg8 Ke5 3 .Rg5 mate.

( 1 33) White mates in three.

Number One.
l .Kfl KB 2 .Bf5 Kxe3 3 .Rd3 mate.

( 1 34) Helpmate in two.

SET.
Number Two.
White moves first, Black goes out of his way to
mate himself in two. l .f8=N e5 2 .Nfe6 mate.
(b) Black moves first and makes sure he mates
himself in two. l .e5 f8=B 2 .e4 Bd6 mate.

603

PAL BEN KO: MY LIFE, GAM ES AN D CO M POSITIO N S

604

( 1 3 5) White mates in three.

Number Two.
l .BfS Bg6 2 .Nd7+ Kf7 3 .Be6 mate.

( 1 3 6) White mates in three.

Number Three.
l .Nd2 NbaS If 1 . . .Nb4, White mates by 2.Be4+
Re5 3 .Qe3 mate. 2 .Qa4+ Ka6 Or 2 . . . Re5 3 .Ne4
mate. 3 .Bc4 mate.

( 1 3 7) White mates in three.

Number Four.
l .Kg4 f3 2 .KhS ! f2 Or 2 . . . dxe5 3 .Rd6 mate.
3 .Ng4 mate.

( 1 3 8) White mates in three.

Number Five.
l .Rf7 NcS Or 1 . . .Nd8 2.Ng5 + Kxd4 3 .Rf4
mate. 2 .dxcS d4 3 .NgS mate.

( 1 3 9) White mates in two.

Number Five.
l .Bf2 Nxf2 2.Rg3 mate.

LETT ER PRO B L E M S

( 1 40) White mates in three.

Number Six.
l .Ne3 Kxe3 2 .Rd6 Kf3 3 .Rd3 mate.

( 1 4 1 ) White mates in three.

Number Seven.
l .Re4 Kc5 Or l . . .Ne7 2.ReS+ NdS 3.RxdS
mate. 2 .Ka4 Kd5 2 . . .NbS 3. Qxc6 mate. 3 .Qe5
mate.

( 1 42) White mates in two.

Number Seven.
l .Qd6 Nc4+ 2 .Rxc4 mate.

( 1 43) White mates in three.

Number Seven.
l .Nc5 If Black captures the Knight, the cS
square won't be available to the black King.
1 .. .Kb6 Or l . . .BxcS 2.c8=Q Be3 3.Qb7 mate.
2 .Na6 Ka7 3 .c8=N mate.

( 1 44) White mates in three.

Number Eight.
l .g4 Re6 2 .Rf7 Rd6 3 .Rh5 mate.

605

PAL B E N KO: MY LIFE, GAMES AN D CO M POS ITIO N S

606

( 1 45) Helpmate in two, Bl moves,

Number Eight.
Black has two ways to help White mate him.
L .Re7 The other solution is 1 . . .Nf6+ 2 .Kf4
Rd5 3 .Ng5 mate. 2.Kd4 Nf6 3 .Ng7 mate.

( 1 46) White mates in two.

Number Nine.
l .Ne4! Bxe6 2 .RgS mate.

( 1 47) White mates in three.

Number Nine.
l .Ra4+ Kxa4 2.Rb3 bS 3 .BxbS mate.

( 1 48) White mates in three.

Number Nine.
l .NeS ! Kxe4 Or 1 . . .Ke3 2 .Ng6! Kxe4 3 .Re5
mate. 2 .RSd4+ Ke3 3 .Ng4 mate.

( 1 49) White mates in three.

Number Zero.
l .e8=R Kc6 2 .Rb8 Kd6 3 .Rb6 mate.

It took me quite a while to get seriously involved in endgame compositions. This

might seem surprising since I've always loved endgames. Furthermore, endgame
compositions teach ideas that can be used in real games.
I suppose my temporary aversion to this kind of composition occurred early: I
created my first one while still a beginner ( 1 5 1). My second endgame composi
tion occurred a bit later (see 2 2 1), but I remember being very proud of it (it
wasn't published until 1 997). Eventually, I found out that another person had
discovered the idea before me. This took away my enthusiasm- instead of com
ing up with an original piece of work, I'd done nothing more than improve on
someone else's concept.
This period of disenchantment ended at the 1 964 Belgrade tournament when
I discovered a problem-like possibility in my adjournment with Matanovic (see
game 56). Mter showing it to G. Ban, a columnist for the Hungarian Chess Maga
zine, he suggested that I enter the magazine's yearly endgame contest. I won the
second prize (see problem 1 52) but, after personally cooking (refuting) the first
prizewinner's entry, I managed to advance to the head of the contest. After that I
began to send endgame compositions to this contest almost every year, and usu
ally managed to take the first prize.
I felt at home in these artistic endgames- my analytical skill let me find com
plex but highly satisfying ideas and elaborate on them in detail. Most of my prob
lems were so complex that the answers were almost always given with the prob
lem. "No one can solve them anyway, " said Ban.
One of my compositions (2 06) was so difficult that nobody ever managed to
solve it. When one of the world's best problem solvers, Dr. John Nunn, failed to
figure out the answer, he lamented, "Why do you create such difficult endgames? "
Actually, I don't say to myself, "Today I'm going to create a really hard prob
lem! " I just look at an interesting position that's filled with fascinating themes,
carefully find the best attacking and defensive possibilities (Very carefully! I can't
stand the idea of one of my problems getting cooked.), check and double-check
every variation with painstaking thoroughness, and whatever results, be it easy
or insanely difficult, is the final product.
I should mention that problem 2 06 originally only got third prize in a tourna
ment, but I cooked the first- and second-place compositions and ended up tak
ing top honors myself!
Nonetheless, I eventually went out of my way to compose some lightweight
endgames (not tournament quality, but they gave great pleasure to the average
players who tried to solve them), and I even got into improving endgames cre
ated by famous composers (finally curing the wounds of my youth!).
Today, I love working on endgame compositions because this is the only seg
ment of the problem world that is still relatively safe from the invasive touch of
607

PAL B E N KO: MY L I F E. GAM ES AN D CO M POSITIO N S

computers. Those horrible bits of software might be able to find a three-move

mate in half a second, but compositions like 2 06 are still beyond them.
60 8
( 1 5 0) White to move and win.
l .g7+ ( l .Rh4? Rh7+!, ) 1 . . .KgS 2 .Rh3 RB
3 .Rg3 ! Rxf2 4.Ra3 Rh2 + S .Kg6 Ra2 6.RB
Ra6+ 7 .Rf6 Rh6 S.Kh6 RhS 9.RfB+ and wins.
=

( 1 5 1) White to move and win.

(Zaszlonk 1 944.)
l .Re6+! Black has to take the Rook, since 1 ... KdS

2 . Nxf7 + is obviously hopeless, as is 1 . . . fxe6

2 .Ng6+. 1. .. Kxe6 2 .dS+ Ke7 And not 2 . . . Kd6
3 .Nxf7+. 3 .d6+ KeS Poor Black isn't given a
moment's rest. Since both L .Kxd6 and L.KdS
were met by 4.Nxf7+, and since 3 . . . KfS is crushed
by 4.g6 ! , he must move his King to eS and hope for the best. 4.d7+ Ke7 S .dS=Q+
QxdS 6.Nc6+, 1 -0. White wins the Queen and the game. This was my first
endgame composition!

( 1 52) White to move and draw.

(First Prize, Magyar Sakkelet 1 967.)
The birth of this problem came about in an in
teresting way: while analyzing my game with
Matanovic from Belgrade 1 964 (see game 5 6 in
the games section), I was confronted with the
problem of not being able to control the a7-g1
diagonal. Accordingly, I looked at the possibility
of Black picking up the h-pawn and pushing his f-pawn (This leads to an inferior
form of position analyzed by E. del Rio more than 200 years ago.). Although many
masters considered the position to be lost for White, I discovered that White can
actually draw ifhe plays very exactly and originally. l .Bc7 l .Be5? Ra5 ! 2 .Bd4 (2 .Bc7
Rf5) 2 . . . Kg3 3 .Bxf2+ KB 1 . ..Rh2 L. Ra7 2 .Bb6 2.Bd6 Rc2 3 .BeS Rd2 4.Bf4
Re2 S.BhS ReS 6.Bg3 Kxg3 , 1/2_112. Stalemate! Thus, after 2 00 years, there is a
new contribution to the del Rio position. As a matter of fact, it was published as a
study in Magyar Sakkelet magazine in 1 967 and won a prize.
=.

E N DGAM ES

( I S 3 ) White to move and draw.

(First Prize, Magyar Sakkelet 1 970.)
609
This game-like position should prove instruc
tive to tournament players of all strengths.
White i s up an Exchange, but the black
queenside pawns are advanced and will be dan
gerous after White's b-pawn falls: l .Kh6 ! ! The
obvious l .Ka6 fails to 1 . . .Nxb2 2 .KaS a3 ! 3 .Kb4
a2 4.RaS Na4! ! S .Rxa4 b2 , and Black Queens. 1 . ..a3 ! ! The tempting 1 . . .Nxb2
no longer gets the job done: 2 .KcS Nd l 3 . Kb4 b2 4.Ka3 ! KxD S .Ka2 Ke2
(S . . . Kxg4 6.Kb l KD 7.RdS Ke2 S.ReS+ draws) 6.Kb l a3 7.Rb 3 . 2 .hxa3 Nb4! !
If the Knight is taken the b-pawn will Queen, but if it's left on the board the b
pawn will cost White his Rook. Things look grim for the first player, but White's
game is not entirely hopeless. First, let's take a look at the position after 3 .axb4
b2 4.bS b l =Q. Here S .RfS followed by RfS was suggested by Bobby Fischer.
However, Black can prevent White from setting up his fortress by S . . . Qg 1 + 6.Kb 7
hS ! 7 .gxhS QcS , when Black wins. Another try is S . RcS KxD 6.Rc6, when Black
must protect his last pawn since the Rook and b-pawn versus Black's King and
Queen (with no black pawns) is a draw. Black can win the position after 6.Rc6 by
defending h6 and then bringing his King over to the queenside (The variations
are long-winded and complicated, and won't be presented here.). 3 .Rd8 3 .axb4
b2 4.bS b l =Q. 3 . . . h2 4.Rd l Na2 ! Stronger than 4 . . . NdS+ S .KcS Nc3 , because
the text wins a tempo by the threat of . . . Nc l . S .Rh l Nc3 6.RhI Of course,
6.Rxb2 Na4+ wins the Rook. Now Black manages to squeeze one more tempo
from the white Rook. 6 . . . Kg2 ! 7.Re l Kf2 8.RhI Ke2 Threatening 9 . . . Nd l
1O.Rh2+ Nf2 , winning. 9.Rh2 + KxB I O.RhI Black can't get the Rook to abandon its control over the b l -square. Since 1 0 . . . Ke2 (threatening . . . Nd l ) gets nowhere due to I l .Rh2+, Black is left with two promising tries: 1O . . . b l =Q, as seen
in the game, and 1O . . . Kxg4. Neither move, though, is good enough to score the
full point. 1O . . . h l =Q+ The main alternative, as mentioned above, is: 1 0 . . . Kxg4
I l .KcS ! b l =Q (There is no profitable way to delay this promotion. If 1 1 . . .hS
1 2 .a4! draws, while 1 1 ...Kf3 1 2 .Kd4! b l =Q 1 3 .Rxb l Nxb 1 1 4.a4 Nd2 I S .aS ! Nb3+
1 6.KeS NxaS 1 7 .KfS also leads to a drawn result.) 1 2 .Rxb l Nxb l 1 3 .a4 Nd2
1 4.Kb4! Ne4 I S .aS Nd6 1 6.a6 NcS 1 7 .KcS hS I S.Kc6 h4 1 9.Kc7 ! (The race is
almost over, but White still has to be careful. The obvious 1 9.Kb7?? loses to
1 9 . . . Nd6+ 2 0.Kc6 h3 ! 2 l .a7 h2 2 2 .Kxd6 h l =Q 2 3 .Kc7 QaS. Mter the text move
[ 1 9.Kc7 !], 1 9 . . . h3 20.KxcS h2 2 l .KbS h l =Q 2 2 .a7, we reach a well-known drawn
position.) 1 9 . . . Na7 2 0.Kb6! (Again, 2 0.Kb7?? NbS 2 l .Kb6 h2 wins for Black.)
20 . . .NcS+ 2 l .Kc7 ! , and Black can't avoid the repetition or the draw in the note
to White's 1 9th move. 1 l .Rxh l Nxh l 1 2 . a4 Nd2 1 3 .KhS Nb3 14.Kh4! The
duel between the black Knight and the white King opens a new and far from
easy chapter. The following exclamation points for white King moves indicate

PAL B E N KO: MY LI FE, GAMES A N D CO M POSITIO N S

61 0

that they are the only moves to hold the draw. Here, for example, if 1 4.Kc4??
Nas+ I S .KbS Kxg4 1 6.KxaS hS wins for Black since he queens first and simulta
neously stops the white pawn from safely promoting. 14 ...Nd4 1 5.Kc5 ! Of course,
I S .aS?? loses to I S,. .Nc6+. 1 5 . . . Ke4! The best try. Bad is I S,. .Ne6+? I 6.Kb6.
The new danger is that Black might be able to stop White's a-pawn with his
King and win the g-pawn with his Knight. At the same time, White must not
allow Black to sacrifice his Knight for White's a-pawn. It seems too much to ask
the white King to deal with all of this, but there is still a narrow path across the
abyss. 1 6.Kc4! Avoiding I 6.aS?? Nbh. 16 . . . Nc6 1 7 .Kc5 ! Instead, I 7 .KbS??
loses to I 7,. .KdS I 8 .Kb6 ( 1 8.aS NxaS I 9.KxaS Ke4 wins for Black) I 8,. .NeS I 9.aS
(any King move is met by , . .Nxg4) I 9,. .Nc4+ followed by 2 0, . .NxaS . 17 ... Na5
An important sub-variation is I 7,. .Nb8 I 8 .aS KeS I 9.Kb6 Kd6 20.Kb7 Nd7
2 I .Kc8! (The White King goes on a long journey. Its goal: kill the remaining
black pawn!) 2 1 . . .NcS n.Kd8! Nd3 (or n, . .KeS 2 3 .Kf7 KdS 24.Kd8 ) 2 3 .a6!
Kc6 24.Ke7 NeS 2 S .Kf6 Nxg4+ 2 6.Kg6, when 2 7 .KhS will force a draw. 1 8.Kb5 !
Bad i s I 8 .Kb4?? Nb7! I 9.KbS KdS 20.Kb6 Nd6. The same goes for I 8 .Kb6??
Nb3 . 18 ... Nb3 1 9.Kb4 Nd2 ! After 1 9, . .Nd4 2 0.Kc4 Nc6, we reach a previous
position (meaning that Black hasn't made any progress). The text presents new
problems for White to solve. 20.Kc3 ! ! It seems strange, but this is the only move
that holds, as we can see from the following variations: 20.aS?? KdS 2 1 .KbS Kd6!
n.a6 Kc7 2 3 .KcS Ne4+ 24.Kd4 Nf6! 2 S .KeS Nxg4+ 2 6.KfS hS 2 7 .KgS Nf6! ,
and Black wins; 2 0.KbS KdS leads to the same variation; 2 0.KcS KeS 2 1 .Kc6
Nb3 n.KbS KdS 2 3 .Kb6 Kd6 24.KbS Nd4+ 2 S .Kb6 Nc6 2 6.Kb7 KcS , and Black
wins again. 20 . . .Nf3 2 1 .Kc4! Ke5 The last try. Instead, if 2 1 . . .NeS+, winning
the g-pawn, Black will be unable to catch the a-pawn. 22.a5 ! Kd6 22,. .Nh2 2 3 .a6.
2 3 .a6 Kc6 24.a7 ! Kb7 2 5 .Kd5! Nh2 It looks like White has finally run out of
miracles, since 2 6.Ke4? Nxg4 n.KcS hS 2 8 .KgS Nf6 wins (a similar variation
was given earlier). 26.g5 ! ! The last point. White barely manages to draw the
game. 26 . . . hxg5 2 7.Ke4 Kxa7 28.Kf5 g4 29.Kf4 Kb6 3 0.Kg3 , Y2- Y2 .
=

E N DGAMES

( 1 54) White to move and win.

(First Prize, Magyar Sakkelet 1 97 2 .)
l .Nb5 ! ! A real key move which, being surpris

ing, deserves two exclamation marks. It is in

deed surprising for it allows the black King to
escape from the mating net and, in addition, it
takes the Knight further from the threatening
pawn, apparently allowing it to promote. Is this
key move necessary? Wouldn't I .Ne2 lead to the same result? No, I .Ne2 fails to
1 . . .h2 2 .Nac3 b5 3 .Nd4 b4+! 4.Kxb4 Kb2 , and Black escapes since his King is out
of the trap. 1 ...KhI The best try. The tempting 1 . . .h2 loses beautifully after
2 .Nac3 h l =Q 3 .Nd4 Qc 1 + 4.Kb3 Qb2+ 5 .Kc4 b5+ 6.Kd3 Qxf2 7 .Nb3 + Kb2
8 .Nd l +. It's now clear that the black Queen was led into a trap. Due to all the
possible forks, she would have been lost whether she captured the f-pawn or not.
2 .Nac3 + Kc2 White wins after 2 ... Kc 1 2 .g4! . 3 .Nd4+ ! Kd3 ! The obvious 3 . . . Kxc3
runs into 4.Ne2+! (A very important move! The obvious 4.NB? fails to 4 . . . b 5 ! ,
when Black can draw.) 4 . . . Kc4 (4 . . . Kd2 fails to 5 .Ngl h 2 6.NB+, while 4 . . . Kd3
5 .Nf4+ is also easy for White.) 5 .g4 h2 6.Ng3 Kc3 7.Ka4 Kc4 8 .g5 b5+ 9.Ka5,
and White wins since he will Queen with check. 4.Nd l ! ! This is the only way to
win. White fails to achieve the victory after 4.f4 Kxc3 5 .Ne2+ Kd3 , and Black
wins because White no longer can play Nf4+. If White had tried 4.8 instead of
4.f4, then 4 . . . Kxc3 5 .Ne2+ Kd2 wins since 6.Ngl h2 7 .NB+ isn't possible. White
can try 4.NB, but he still fails to win the game: 4 . . . Kxc3 5 .Ka4 Kc4 6.g4 b5+
7.Ka5 b6+! 8.Ka6 b4 9.g5 Kd5 1O.g6 Ke6 Il .Ng5+ Kf6 1 2 .Nxh3 b3 1 3 .Nf4 b2
1 4.Nd5+ Kxg6 1 5 .Nc3 Kf5, with a draw. 4 . . . h2 5.f4 Kxd4 And not 5 . . . h l =N
6.f5 , winning for White. 6.Nf2 Kc3 6 . . . Ke3 7 .Ng4+ wins for White. 7.f5 b5
8.f6 b4+ 9.Ka2 Kc2 1 0.7 b3 + 1 l .Ka3 b2 1 2 .f8=Q b l =Q 1 3 .Qf5+, 1 -0.

( 1 5 5) White to move and win.

(Magyar Sakkelet 1 9 7 3 .)
l .a4 Not l .Kd4? Kf4 2 .a4 e5+ 3 .Kc3 e4 4.a5 e3
5 .a6 Kg3 ! 6.Kd3 Kf2 =. 1 . . .e5 2 .a5 e4 3 .Kd4
Kf4 4.a6 e3 5.Kd3 Kf3 6.a7 e2 7.a8=Q+,

1 -0.

611

PAL B E N KO: MY LI FE, GAM ES A N D CO M POSITIO N S

61 2

( l S 6) White to move and win.

(First Prize, Magyar Sakkelet 1 97 S .)
l .Bc4+ ! l .b7? Kxb7! 2 . B d S + KbS 3 .B e4 as
4.BxaS a4! =. 1 . . .Ka5 2 .h7 Rf8 2 . . . RbS? 3 .BbS .
3 .Bd3 RgS 4.h4+ Ka4 S .Bc2 + Ka3 6.hS Rf8
7.Bdl RgS S.Bg4 RhS 9.Kc6 Kb4 l O.Be2 ReS
l 1 .Kd7 Rf8 l 2 .Kc7 Rf7+ 1 3 .KhS, 1 -0.

( l S 7) White to move and win.

(First Prize, Magyar Sakkelet 1 9S2 .)
1 .e7 l .g7 Rg3 2 .e7 BhS 3 .Rg6! (This idea is
known as a "Novotny. ") L.Rxg6! (L. Bxg6
4.Kf2! wins) 4.eS=Q Rf6+ S .Ke l (S .Kg1 BxeS
6.gS=Q Rg6+ =) 5...BxeS 6.gS=Q Bd7! = . 1 .. .Re3
Or 1 . . .Rf3+ 2 .Ke 1 ! Re3 + 3 .Kxd 1 Rxe7 4.RcS!
(4.Rf6? ReS! S . g7 RgS 6.Rf7+ Kb6 7.Ke2 Kc6
S.Kf3 Kd6 9.Kf4 Ke6! =) 4 . . . Kb6 S . RgS Rg7 6.Ke2 Kc6 7.Kf3 Kd6 S.Kg4 Ke7
9.KhS Kf6 1 0.Kh6, and White wins. 2. g7 Bh3 3 .Rc7+ 3 .Re6? Rxe6 (L.Bxe6
4.Kf2 wins) 4.gS=Q Bc4+! S .Kf2 Re2+ 6.Kf3 BxgS 7 .Kxe2 Bf7 =. L.Ka63 . . .Kb6
4.Rc3! wins. 4.Kf2 Gaining a decisive tempo. 4 . . . Re4 S.Rc6+ KaS 6.Re6 Rxe6
7.gS=Q Rf6+ S.Kgl S.Ke3? BxgS 9.eS=Q Re6+. S . . . BxgS 9.eS=Q The threat is
QeS+ or Qe7+ if Black's King had moved to its second rank on move five. 9 ... Be6
lO.QdS+, 1 -0.

( I SS) White to move and win.

(Third Prize, Canadian Chess 1 9S0.)
l .Rh l + l .dS exdS 2 .Kc3 d4+! 3 .Kb3 a l =N+! =.
l . . . Kh2 2 . Ra l Kxa l 3 . Kc2 e S 4 . d S e4
S.d6 e3 6.d7 e2 7.dS=Q (and not 7 . dS=B??

e l =N+ =), 1 -0.

E N DGAMES

( I S 9) White to move and win.

(First Prize, Magyar Sakkelet 1 980.)
l .g4 Kh2 2 .Rf2 + not 2 .Ra l ? Kxa l 3 .Kc2 h6

4.Kc 1 as S .Kc2 a4 6.Kc 1 hS!, and Black draws

because the h-pawn promotes with check. White
has to work to make Black lose a tempo 2 . . . Kh3
2 . . . Kb l 3 .Kc3 ! a l =Q+ 4.Kb 3 . 3 .Rf6! ! as 4.Rfl
Kh2 S .Ra l ! Kxal 6.Kc2 h6 ! Otherwise g4-gS
makes it too easy. 7.Kc 1 hS ! ? 7 . . . a4 8.Kc2 a3 9.g3! hS 1 0.gS . 8.gS h4 9.g6 h3
1 O.g:xh3 a4 1 1 .g7 a3 1 2 .Kd2 Kh2 1 3 .g8=Q a l =Q 1 4.Qg7+ Ka2 l S .Qf7+ Kh2

I S ... Kb I 16.QfS+ Ka2 1 7 .QdS+ Kb I 1 8.Qe4+ Ka2 1 9.Qc4+. 16.Qf6+Ka2 17.Qe6+

Kh2 1 8.QeS+ Ka2 1 9.QdS+ Kh2 20.Qd4+ Ka2 2 1 .Qc4+ Kh2 2 2 . Qc2 mate.

( 1 60) White to move and win.

(Chess Life 1 980.)
l .Rh7 + ! Of course, l .Rxc6 as is an easy draw
for Black. 1 ...Kc3 Trying to promote the c-pawn
is no better: 1 . . .Ka3 2 .Kc7 cS 3 .Kd6! c4 4.Rc7!
Kb3 S .KdS as (S . . . c3 6.Kd4 c2 7 .Kd3 ) 6.Kd4,
and White wins. 2.Kc7 as 3 .Ra7 ! This gains a
tempo by pulling back the black King. 3 . . . Kh4

On 3 . . . cS, White must avoid both 4.Kd6? Kd4! and 4.Kb6? Kb4. However, 4.Kc6
c4 S .KcS Kb3 6.Kd4! c3 7 .Kd3 wins. 4.Kd6! ! To understand this move, one has
to see the consequences of 4.Kxc6? a4 S .KdS a3 6.Kd4 Kb3 7 .Kd3 Kb2! (7 . . . a2?
8.Rb7+ Ka3 9.Kc2! wins by forcing Black to promote to a Knight. This theme
will appear again!) 8.Kd2 a2 9.Rb7 + Ka l , and Black has saved himself due to the
threatened stalemate. 4 . . . a4 S .KeS a3 6.Kd4 Kh3 7.Kd3 Kh2 8.Kd2 a2 9.Rh7+
Ka3 9 ... Ka l 1 0.Kc2 and I 1 .Rd7 mates. 1 0.Kc2 a l =N+ l 1 .Kc3 Ka2 1 2 .Rh2 +
Ka3 1 3 .Rh6 Ka2 1 4.Rxc6, 1 -0. The win after 1 4 . . . Kb l l S .Rf6 is very easy.

( 1 6 1 ) White to move and win.

(Chess Life. 1 980 Version.)
l .Bc7+ KhS 1 . . .Ka4? 2 .Bd7+ Ka3 3 .Bd6 Qxd6
4.Nc4+. 2 .Bd7+ KcS 3 .Bd6+! Kxd6 4.c3 ! Qxc3
S .Ne4+ Kxd7 6.Nxc3 Ke6 7.Kh7 as 8.Kc6
a4 9.Nxa4 KfS 10 .Nc3 Kg4 1 1 .Ne2 , 1 -0.

61 3

PAL BEN KO: MY LIFE, GAM ES AN D CO M POSITIO N S

61 4

( 1 62) "White to move and draw.

(E. G. 1 9S0.)
l .Rd4! l .Kgl ? h2+ 2 .Kg2 (2 .Kh l Bg6! 3 .Rd4

Ke3) 2 . . . BhS 3 .RhS (3 .RfS BB+! 4.RxB h l =Q+!)

L.h l =Q+! S .Kxh I Kf2 wins for Black. l .. .BhS
2 .Rh4 Kf2 3 .Rg4 Mutual Zugzwang! 3 . . . g2 +
If L.Kf3 , "White must use some care: 4.RgS?
Bg4 S .RfS+ Ke2 6.RhS Kf2 7 .RfS+ Bf3+ wins
for Black. However, "White can defend with 4.Rh4! Kf2 S . Rg4 Kf3 6.Rh4 Bg4
7 .Kgl Be6 S.RhS, and Black can't make progress. 4.Kh2 Bxg4, 2-2.

( 1 6 3 ) "White to move and draw.

(Chess Life 1 9S 1 ) .
1 .a7 RhS + 1 . . . d4! ? 2 . exd4 Kc6 3 .d S +! Kb6
4.aS=N+ =. 2 .Kh7 RaS 3 .KxaS KeS 4.c3 Ke7
S . e4 dxe4 6.c4 e3 7.cS e2 S .c6 Kh6 9.c7
e l = Q I O . c S = Q Qe4+ 1 1 . Kh S Q e 7
1 2 .aS=N+, 2-2.

( 1 64) "White to move and win.

(First prize, Magyar Sakkelet 1 9S 1 .
l .Be6 l .Kf7? Re I 2 .Be6 Rf1 + 3 .KgS Rg l +
4.KhS Rg7! S . eS=Q Rh7+ 6.KgS RhS+. l .. .Rdl

1 . . .Kg7 2 .Kd7 Rd l + 3 .Kc6 Rc 1 + 4.KdS Rd l +

S .KeS Re l + 6.KfS Rf1+ 7.KgS Rgl + S.KhS, and
wins. 2 .hS ! 2 .Kf7? Rf1+, =. 2 ...Kg7 3 .h6+ Kh7
4.Bf5+ KgS 5.h7+ S .eS=R!? forces Black to find
S . . . Rf7 ! ! , which is the only way to draw.
S . . . Kg7 6.hS=Q+ KxhS 7.Kf7 Re I S.Be6
Rf1+ 9.Kg6, 1 -0.

E N DGAM ES

( 1 6S) White to move and win.

(First prize, Magyar Sakkelet 1 9S2 .)
I .Re2 ! ! The only good square for the Rook. If

l .Ra2 RbS+ 2 .KcS Rb I 3 .Kd4 Rb4+ 4.Ke3 Rxh4

S .Kf3 Rh3+ 6.Kf2 , Black draws by 6 . . . Ra3 ! . This
trick doesn't work with the Rook on e2 . 1 ...RbS+
1 . . .Kgl 2 .Nf3+ Kfl 3 .Rxh2 RbS+ 4.KcS RbS+
S .Kd4 Rb4+ 6.KdS RbS+ 7.Ke4 Rb4+ S.Nd4.
2.KeS Rb I Averting mate. 3 .Kd4 ! ! No win is possible until the white King gets
into the action. Worse is 3 .Nf3? Rb2 ! . Also no good is 3 .KdS? RbS+. 3 . . . Rb4+
4.Ke3 Rxh4 Or 4 . . . Rb3+ S .Kf4 (S .Kf2 Rb2 =) S . . . Kg 1 6.Rg2+ Kh 1 7.Rc2 ! ! Kg 1
S.Nf3+ Rxf3 + 9.Kxf3 h l =Q+ lO.Kg3 wins. S .KB Rb3 + 6.Kf2, 1 -0.
( 1 66) White to move and win.
(First Prize, Magyar Sakkelet 1 9S2 .)
l .Bg4! ! The only move ! The threat is 2 .RaS+.
If l .Be6? RfS 2 .KcS Rf2 3 .Kd6 KfS, we have a
theoretically drawn position after 4. RaS+ .
1 . ..Kf8 There's a trick here: 1 . . .0-0 is illegal
because it can be proved that Black moved his
Rook or King on his previous turn. Also, 1 ... RgS
is wrong because of 2 .RaS+ Kf7 3 .Be6+. Finally, 1 . ..RfS 2 .BhS+ KdS 3 .Bf7 fol
lowed by KcS and Kd6 also wins. 2 .BhS RgS White also wins after 2 . . . h6 3 .KcS
KgS 4.Kd6 Rh7 S .RaS+ Kg7 S .Ke6 RhS 6.Ra7+ followed by 7.Kf6. 3 .Rf7+ KeS
4.KeS KdS White wins after both 4 . . . RgS + S .RfS + and 4 . . . RfS S .Ra 7 + KdS 6.Bf7.
S.Kd6 KeS 6.Ke6 KdS 7.Rd7+ KeS S.Ra7 KbS 9.Rb7+ KaS 9 . . . KcS 1 0.BEl!
Rg2 1 1 .Be6+ KdS 12 .Kd6 Rd2 + 1 3 .BdS reaches a position analyzed by Philidor:
1 3 . . .KcS ( 1 3 . . . Rc2 14.Rxh7 Re2 reaches a position analyzed more than 200 years
ago! White wins after l S .Rg7 Re l 1 6.Rb7 Rc 1 1 7 .Bb3 ! Rc3 l S.Be6 Rd3+ 1 9.BdS
Rc3 2 0.Rd7+ KcS 2 1 .Rf7 KbS 22 .Rb7+ KcS 2 3 .Rb4! KdS 24.Bc4! .) 1 4.Ra7 Rb2
I S .Rf7 KbS 1 6.RfS+ Ka7 1 7 .RaS+ and I S.RbS+ when Black loses his Rook.IO.BB!
Rg6+ l O . . . RfS ( l O . . . RcS+ 1 1 .Kd6! hS 1 2 .Bc6 wins for White) 1 2 .Rb4 Ka7 (or
1 2 . . . Rf6+ 1 3 .Kc7+ Ka7 1 4.Bc6) 1 3 .Ra4+ KbS 1 4.Kd6 Rf6+ l S .Kd7.
Another key try for Black is 10 ... Rg3 , but White still wins after 1 1 .Be4!
( 1 l .Rb3 is bad: 1 1 . . .Rg6+ 1 2 .Kc7+ Ka7 1 3 .Bc6 Rg7+ 1 4.Bd7 Rg6, and we now
have the only theoretically drawn position in which the Kings oppose each
other face-to-face, like those in the Philidor position) 1 1 . . .Re3 ( 1 1 . . .Rg4 1 2 .Re7
KbS 1 3 .ReS+ Ka7 1 4.KcS RgS+ I S .BdS) 1 2 .BdS Rd3 1 3 .Re7 Rc3+ 1 4.Kb6+ KbS
I S .Be6 ! . 1 1 .KeS ! ! Here 1 1 .KbS is bad due to 1 1 . . . Rg3 ! 1 2 .Rf7 + ( 1 2 .BdS? Rb3 +
i s a draw) 1 2 . . . KbS 1 3 .Kb6 KcS 1 4.Bc6 Rd3 , and Black has reached the Szen
position, which is drawn . 1 1 . . .Ra6 I 2 .Be6 hS 1 3 .Kd6 h4 I4.Rd7+ KbS
I S .RdS+ Ka7 I 6.Ke7, 1 -0.

61 5

PAL B E N KO: MY LI FE, GAM ES AN D CO M POSITIO N S

61 6

( 1 67) White to move and win.

(First Place, Hungarian Ch., 1 9S3 .)
l .Kb6! Paradoxically, White blocks his passed

pawn, but this is the only way to try to lock out

the enemy Knight. Most players would think
that "passed pawns must be pushed" and play
l .b6, but Black can draw by l . . .Ng4! 2 .b7 NeS
3 .Kb6 Nd7+ 4.Kc7 NcS S .bS=Q Na6+. 1 . ..Ng4
l . ..KxfS 2 .Kc7 leads to the promotion of the b-pawn. 2 .Kc7 ! Ne3 If 2 . . . Nf6,
White wins by 3 .Kd6! Ne4+ 4.Kc6 Nd2 S .KdS Nb3 6.b6 NaS 7.Ke6. 3 .Kd7 !
NdS 3 . . .Nc4 4.Ke6. 4.Kd6 Nb6 S .Ke6! Kh6 6.f6 Kg6 ! 7.Ke7 The obvious
7 .f7 Kg7 S.Ke7 NdS + 9.KeS Nf6+ 1 0.KdS NdS leads to a draw by repetition.
7 . . . NdS + 8.Kd6 ! Nb6 9.Ke6 Kh7 l O.Ke7! White only draws after l O.f7? Kg7
1 1 .Ke7 NdS+.l0 . . . NdS + 1 1 .Kd6 Nb6 1 2 .Kc6 Nc4 1 3 .Kd7 Kg6 14.Ke7, 1 -0.
Black can't stop both white pawns.

( 1 6S) White to move and win.

(First Prize, Magyar Sakkelet 1 9S3 .)
l .RdS + ! The two obvious moves, l .RaS and

1 .KaS , both fail:

l .RaS? Kd3 ! ( l . . .Kc2? allows White to win,
as shown in the actual solution) 2 .Kb3 Kd4!
(2 . . . Kd2 ? fails, but it's complicated: 3 .Kc4! Kc2
4.Ra2 + Kb 1 S .Kb3 Kc l 6.Kc3 Kd 1 7.Kd3 Ke 1
S.RaS Kf2 9.Kd4, and White wins. One example is 9 . . . Ke2 1 0.KcS Kd3 1 1 .Kd6
Kc3 1 2 .Kc7 Kb4 1 3 .RhS as 1 4.KbS Ra6 l S .Kxb7 Rxb6+ 1 6.Kxb6 a4 1 7 .Rh4+
Kb3 l S.KbS a3 1 9.Rh 3 + Kb2 20.Kb4 a2 2 1 .Ra2 + Kb 1 2 2 .Kb3 a 1 =N+ 2 3 .Kc3)
3 .Kc2 Ke4 4.Kc3 Ke3 S .Kc4 Kd2 (S ... Ke4? 6.KcS KeS 7.NcS! RaS S.Kb6+) 6.KcS
Kc3 7.Kd6 Kb4 =.
l .KaS? Kc3 2 .RhS Kc2 3 .RdS Kc l ! (Black loses on either 3 . . . Kc3 ? 4.NdS+ or
3 . . . Kb2 4.Na4+ followed by Kb6. Also, 3 . . . Kb3 ? loses, but in a more instructive
way: 4.Nd7 bS S .Kb6 Rxd7 6.Rxd7 b4 7 . KaS ! Ka3 S.Rd4! b3 9 . Ra4+ Kb2
1 0.Kb4 Kc2 1 1 .Ka3 ! b2 1 2 .Rc4+ Kb 1 1 3 .Rb4) 4.RcS+ (Too hasty is 4.Nd7 bS
S .Kb6 Rxd7 6.Rxd7 b4 =) 4 . . . Kb 1 ! S . Rc7 Kb2 6.RcS Kb3 7 . RcS Ka2 S.RbS
(S.ReS Kb3 ! holds, and not S . . . Kb 1 ? 9.Kb4 ! . If S.RdS, then S . . . Kb 1 ! is the only
move) S . . . Ka 1 ! ! (The only move. Black's King can't move to the b-file because
of 9.NcS b6+ l O.Rxb6+.) 9.Kb4 (9.ReS Ka2 ! l O.Kb4 as+ 1 1 .Kc3 Ra6 1 2 .Re2 +
Kb 1 =) 9 . . . a S+! (If 9 . . . Kb2 , then l O.NcS! as+ 1 1 .Ka4 Ra6 1 2 .Rxb7+ Kc3 1 3 .Nb6
wins.) 1 0.Kb 3 Ra6 1 1 . Rxb7 a4+ ! 1 2 .Nxa4 Kb 1 =. 1 . . .Kc2 2 .RcS + ! These im
portant checks are the first steps in White's campaign to evict Black's King from

E N DGAM ES

the b-pawn's neighborhood. 2 . . .Kd3 Black loses after both 2 . . . Kb l 3 .Kb3 as

4.Rc2 a4+ S .Nxa4 Rxa4 6.Rh2 ! , and 2 . . . Kb2 3 .Na4+ Kb l 4.Kb 3 bS S .Nc3 +.
3 .KaS Kd4 3 ... Ke4 4.Rc8 KfS S .Rb8 ! KgS 6.Rd8 Kf6 7 .NdS + wins for White.
4.Rc 1 KeS Also hopeless is 4 . . . Kd3 S .Rc8 Kd4 6.Rd8+ KcS 7 . Rd l Kc6 8 . Rc 1 +
Kd6 9.Nc8+. S .Rdl Ke6 6.RdS Ke7 7.RgS Kf7 S.RbS ! After this, Black's last
chance to stop the Rd8 and Nd7 setup fails . S . . . Kg7 8 . . . Ke6 9 .Nc8 b6+
1 O.Rxb6+. 9.RdS Kf6 1 0.NdS+ Kf7 1 1 .Kb6, 1 -0.

( 1 69) White to move and win.

(Chess Life 1 983).
l .b7 Ne4! 1 . . .Nd7 2 .b8=N+ ! (not 2 .b8=Q??
Nb6+! 3 .Qxb6+ Kxb6 =) 2 ... Nxb8 3 .Kxb8 Kb6
4.Kc8 Kc6 S .Kd8 Kd6 6.Ke8 Ke6 8.Kf8 fS 9.gS
wins for White, as does 1 . . .Nc6 2 .b8=Q (not
2 .b8=N? + Kb6 =) 2 . . . Nxb8 3 .Kxb8 . 2 .gS ! Either promotion would lead to a draw. 2 . . . fxgS
3 .g4 Nb6+ 4.KbS NdS S.KeS Nb6+ 6.Ke7 NdS + 7.Kd6 Kxb7 S.KxdS Ke7
9.Ke6 KdS 1 0.KfS Ke7 l 1 .KxgS Kf7 1 2 .Kh6, 1 -0.

( 1 70) White to move and draw.

(Chess Life 1 98 3 .)
l .d6 e2 2.d7 Ke7 3 .BeS c 1 =Q 4.b6+ KdS S .b7
Qf4 The best attempt to corner the white King.
6.Kg6 This is the only escape. Black wins after

6.Kg7? Qd6! 7 .Kf7 Qb6 8 . Kf8 Qf6+ 9 . Kg8

Qg6+ 1 0.Kf8 Qb6 ! 1 1 .Kf7 Qd6 (by changing
the tempo, Black can get his King into the fight
due to Zugzwang) 1 2 .Kg7 Ke7 1 3 .Kg8 Qg6+. 6 . . . Ke7 7.KhS ! Qg3 S.Kh6 Kf7
9.KhS Not 9.d8=N+?? Kf6. 9 . . . Ke7 1 0.Kh6 Qh4+ 1 1 .Kg6 Qg4+ 1 2 .Kh6, Y2Y2. Black can't make progress since 1 2 . . . Kf7 1 3 .d8=N+ ! wins the Queen. Note
that 1 2 .Kh7?? wins for Black after 1 2 . . . QhS + and 1 3 . . . QgS +.

61 7

PAL B E N KO: MY LI FE, GAM ES AN D CO M POSITIO N S

618

( 1 7 1 ) White to move and win.

(E.C. 1 984.)
l .Nc6 l . c4+? Kxc4 ( l . . . dxc 3 ? 2 .Nb S) 2 . d 3 +

Kxd 3 3 .Nc6 Kc4 4 . B g7 d 3 S .Ne S + Kc 3

=.

1 . ..Kxc6 2 .Bg7 KdS 3 .d3 a2 4.c4+ KcS S .Kb7

a l =Q 6.BfS mate. This, and problems 1 72 and

1 7 3 , are improvements over a study by Kubbel

(in his work, the white King stood on a6 and
the white Knight on b8). Many composers objected that the white Knight's only
role was that of a sacrificial lamb. In my version, the Knight plays an important
part in the variation to White's first move. In problem 1 72 I give the Knight an
even more important role, and in problem 1 7 3 I take the Knight away altogether.

( 1 72) White to move and win.

(E.C. 1 984.)
l .NfS + Ke6 2 .Nd4+ exd4 3 .Bg7 KdS 4.d3

And we've reached the position from the pre

vious problem. 4 . . . a2 S.c4+ KcS 6.Kb7 a l =Q
7.BfS mate.

( 1 7 3 ) White wins - Black to move.

(E.C. 1 984.)
1 ...aS 1 . ..d3 !? 2 .c4+! (2 .cxd3 or 2 .c3 allows 2 ... as

) 2 ... Kxc4 3 .Bb6! KbS 4.Bd4 as S .Kb 7 a4 6.Kc7

Kc4 7.Bc3 a3 8 .Kd6 a2 9.KeS Kb3 1 O.Kd4 Kc2
I l .Ke 3 , and White wins. 2 .d3 ! 2 .Bh4 d3 !
=

=.

2 . . . a4 3 . Bh4 a3 4.Bf6 a2 S.c4+ KcS 6.Kb7

a l =Q 7.Be7 mate.

( 1 74) White to move and win.

(Sakkelet 1 98 5 .)
l .g7 h2 2 .g8=Q Kh3 A rook-pawn on the sev
enth with the enemy King far from the action
usually means the defender can draw. 3 . Qa8
Kg3 4.Kb6 a3 4 . . . Kf2 fails to S .Qh l . S.KcS a2
6 . Kd4 a 1 = Q + 7 . Qxa l Kg2 8 . Qb2 + Kg l
9.Ke3 , 1-0.

E NDGAM ES

( 1 75) White to move and win.

(First prize, Sakkelet 1 98 5 .)
1 .Kf2 The only way. Only leading to a draw is
l .Bb7 a3 2 .a7 a2 3 .a8=Q Kb2 . 1 . . .Be4 2 .Bh7
a3 3 .a7 a2 4.a8=Q Kh2 5.Qh8! a l =Q 6.Bxd5+
Kc 1 7.Qf4+ Kd l 8.Qg4+ Kc 1 9.Qg5+ Kc2
1 0 . Bxe4+ Kc 3 ! 1 1 . Q c 5 + I l . Q e 5 + Kd2
1 2 .Qxa l ? is a stalemate. 1 l ... Kb3 1 2 .Bd5+ Kh2
1 3 .Qd4+ Kh l 14.Be4+ Ka2 1 5 . Qa4+ Kh2
1 6.Qh4+ Kc 1 1 7.Ke l ! , 1 -0.

( 1 76) White to move and win.

(Sakkelet 1 98 5 .)
1 .Kf2 Bhl 2 .Kgl 2 .Bb7? a3 3 .a7 a2 4.a8=Q
Kb l ! =. 2 . . . Bf3 3 .Bg4 Be4 4.Bf5 Bf3 5.Kf2
Bhl 6.Be4 Bxe4 7.Ke3 Bg2 8.Kd4, 1 -0.

( 1 7 7) White to move and win.

(Chess Life 1 986.)
White is the Exchange up, but if Black can cap
ture White's remaining pawn the draw will be
unavoidable. Hence, White must first save the
pawn and then go after Black's. l .Ke7 Ke5
2 .Rg4 2 .Rg5+ Kd4 3 .Rg4+ Kc5 , and White finds
himself unable to save his pawn. 2 . . . c5 The idea
here is to fix the pawn on c4. After either 2 . . .Nb2 or 2 . . . Nb6, White wins with
3 .c5. 3 .Rh4! White falls into a draw by repetition after 3 .Kd7? Kf5 ! 4.Rh4 Kg5
5 .Re4 Kf5 . Black's threat, of course, is . . . Nb6+. If White tries 6.Re6, Black will
win the pawn after 6 . . . Nb2 7.Rc6 Ke4 (and not 7 . . .Nxc4? 8.Rxc5 + Ne5 + 9.Kd6,
winning for White) 8.Rxc5 Kd4. 3 ...Nb6 Black can't play L.Kf5 because of 4.Kd6.
But there is a reasonable try with 3 . . . Nb2 , though that also fails after 4.Kd7 Nxc4
5 .Rxc4 Kd5 6.Rc2 c4 7.Rc 1 Kc5 8.Kc7 ! . 4.Kd8 ! White falls into a theoretical
draw after 4.Ke8? Kd6 5 .Kd8 Kc6 ! . 4 ... Nxc4 A final bid for counterplay. He
goes down very easily after 4 . . . Kd6? 5 .Rh6+. 5 .Rxc4 Kd5 6.Rc 1 c4 7.Kd7 Kc5
8.Kc7, 1 -0. Both 8 . . . Kb4 9.Kd6 and 8 . . . Kd4 9.Kb6 are easy for White.

61 9

PAL B E N KO: MY LIFE, GAM ES AN D CO M POSITIO N S

620

( 1 78) White to move and win.

(First prize, Sakkelet 1 986.)
l .e4+ This check is necessary. The position is

drawn after l .Rh4? Nb2 ! ( 1 . . .c4? 2 .RhS + Ke4

3 .Kd6) 2 .Kd7 c4 3 .RhS+ (3 .Kc7? Na4! 4.Rh3
Nxc3 is easy for Black) L.Ke4 4.Kc6 Nd l !
(There's no hope with 4 . . . Na4?) S .Rh4+ (S.Rh3
KfS ! , but not S ... Nxc3?) S ... Kd3 ! 6.Rd4+ Kxc3 !
=. 1 . .. Ke5 This move places authentic obstacles in White's winning path. On
1 . . .Kc6, White will eventually squeeze out the black King by 2 .Rf6+ Kc7 3 .Ke6
Kc6 4.Rf8. 2 .Rg4! ! White must drop a tempo horizontally along a rank. Also,
he must exercise considerable care because it's easy to go wrong with 2 . Rh4?
Nb2 3 .Kd7 Nxc4 4.Rxc4 KdS S .Rc2 ! c4 6.Rc 1 ! . This appears to be a straightfor
ward win using the Reti maneuver, but Black has a surprise defense after 2 .Rh4?
via 2 . . . Ne l ! ! 3 .Rh l (better than 3 .Kd7 Nf3 4.Rg4 KfS , when the threat of . ..NeS+
leads to an easy draw) L.Ng2 ! , and Black holds the position. On the other hand,
Blaek would still lose after L.Nf3? 4.Rd l , or with L.Nd3? 4.RhS+ Kd4 S . Rh4+
Kc3 6.Kd6. 2 . . . Nh2 3 .Rh4! 3 .Kd7 KfS 4.Rh4 KgS =. 3 . . .Nxe4 4.Rxe4 Kd5
5.Re l ! S . Rc2? would not be good enough to win. 5 . . . e4 6.Kd7 Ke5 7.Ke7, 1 -0.
We've reached a basic theoretical win.

( 1 79) White to move and draw.

(First prize, Sakkelet 1 986.)
l .Re8 ! White must immediately cover the
passed pawn. Harassing the black King doesn't
help: l .Re8+? Kd3 2 . Rd8+ Kc4 3 .Rc8+ ReS , and
White loses. Pushing the passed h-pawn also
leads to a loss, albeit in a more complicated fashion. After l .h6? , Black should not play 1 . . .RcS
since he will only draw after 2 .Re8+ Kf3 (Surprisingly, Black can't win by 2 . . . Kd3
3 .Rd8+ Kc4 4.h7 c2 S .h8=Q c l =Q+ 6.Kh2 ! because his Rook won't be able to
enter the fray with check.) 3 .Rf8+ Kg3 4.Rg8+ Kf4 S.Rf8+ KgS 6.Rf1 ! Kxh6 7.Kg2
KgS 8.Rh l Kf4 9.Kf2 . Instead, Black should answer l .h6? with 1 . . .RgS + ! . This
important interpolation leaves White defenseless: 2 .Kfl (or 2 .Kh2 c2 3 .Rc8 Kd2
4.Rd8+ Kc 1 S .Rd6 [White also goes down in flames after S .Rh8 Rg6 6.Kh3 Rd6]
S . . . RcS ! 6.Kg3 Kb l ! 7.Rb6+ Ka2 8.Ra6+ Kb3 ) 2 . . . RcS ! (Although White will
Queen first, Black will promote with check. In such positions, the first player
who gives check can generally deliver mate with his Queen and Rook.) 3 .Re8+
Kf3! (Black tosses away the win after L.Kf4? 4.Re 1 ! or L.Kd3? 4.Rd8+ Kc4
S .Ke2 !) 4.Rf8+ Kg4! (sidestepping 4 . . . Ke4? S .Re8+ KfS 6.Re l ! =) S . Rg8+ KfS

E N DGAM ES

6.h7 c2 7 .RfS+ Kg6. So many complications, and we're still on the first move!
1 .. .Kd3 ! It might seem that Black would fare better with 1 . . .Kd2 ! ? , because he
wins easily after 2 . RdS+ Ke2 ! (threatening . . . RgS + followed by . . . RxhS+) 3 .ReS+
Kd3 4.h6 c2 . However, White can save the game by meeting 1 . . .Kd2 !? with 2 .h6
c2 (Black also doesn't win after either 2 . . . RhS 3 .RdS+ Kc l 4.Rd6 c2 S .Kg2 ! or
2 . . . Rf7 3 .RdS+ Kc l 4.RcS! [White must avoid 4.Kg2? Rc7 ! S .Kg3 c2 6.Kg4 Kb2 ]
4 . . . c 2 S .Kg2 Rd7 6.Kg3 Kd2 7 .Kg4) 3 .h7 RfS!? 4.RxfS! c l =Q+ S . Rfl QcS (Black
must not fall for S . . . QcS+? 6.Rf2 + !) 6.Rf7! (Later, in the note to White's main
line 4th move, we will see why White loses after 6.Rf2 + Ke3 7.Rh2 .) when we
reach a very interesting position, known to be a theoretical draw. Black can nei
ther win one of the white pieces nor launch a meaningful attack on the white
King. 2 .h6 c2 We have already discussed the shortcomings of such moves as
2 . . . Rf7 and 2 . . . RhS . 3 .h7 Rf8! Black keeps his winning chances alive with this
timely resource. 4.Re7! Black wins after 4.Rc6? RdS S .Kf2 Kd2 6.Rd6+ Rxd6
7 .hS=Q c l =Q, since White can't give a reasonable check. It would also be a mis
take for White to play 4.RxfS? , because the stationing of the black King on d3
permits the second player, appropriately enough, a problem-like victory. Black's
winning line runs 4 . . . c l =Q+ S .Kg2 Qd2 + ! (The only good check since Black
achieves nothing on S . . . QgS+ [worse are S . . . Qb2 +? 6.Kh3 and S . . . Qc6+? 6.RB+!]
6.Kfl .) 6.Rf2 (Other moves let Black win the Rook or the pawn by a series of
checks.) 6 ... QgS+ 7 .Kh 1 ! (The pawn falls after 7 .Kf1 QhS ! .) 7 ... QdS+! (Black
must carefully avoid theoretical draws like 7 . . . QhS+? S.Rh2 .) S.Rg2 (on S.Kg 1
Qd4, Black will be able to force the win o f the pawn by . . . Qg7+) S . . . QaS! 9.Kh2
QbS+ 1 O.Kh 1 (White's position is hopeless after 1 0.Rg3 + Ke2) 1O . . . Ke3 ! , and
White is lost since the black King and Queen will combine to form a mating net
( 1 1 .Rh2 KB! or 1 1 .Rg3 + Kf2 1 2 .Rg2 + Kf3 ). 4 . . . RhS Black can't win by 4 . . . RdS,
since the Rook would be lost after 5 .Kf2 ! Kd2 (There is also no black win after
S . . . RhS 6.KB! Rxh7 7.Rxh7 c l =Q S.Rd7+.) 6.Rd7+! Rxd7 7 .hS=Q c l =Q (The
Rook also goes after 7 . . . Rf7+ S.Kg3 c l =Q 9.Qh6+ followed by either Qg6+ or
QhS+.) S.Qh6+ Kd 1 9.Qh 1 + ! Kc2 1 0 .Qc6+. If Black tries 4 . . . RbS, White draws
by S .Kf2! Kd2 6.Rd7+ Kc3 (6 . . . Kcl 7.Rb7 ! =) 7.Rc7+ Kb2 S.Rb7+! . 5.Rd7+ White
must meet the immediate threat of S . . . Rxh7. The moves that follow are forced.
5 . . .Ke2 6.Re7 Kd2 7.Rd7+ Kc 1 So far, so good. White has temporarily com
pelled Black to block his own pawn. Still, it's far from easy to see how White will
proceed. S.Rg7 ! ! All White's other tries fall short: I) S.Kf2 ? Kb2 9.Rb7+ Kc3
1 0.Rc7+ Kb3 ! (Our earlier analysis showed that 1 O . . . Kd3 ? 1 1 .Kf3 ! is a draw.)
1 1 .Rb7+ Ka4 1 2 .Ra7+ KbS 1 3 .Rc7 Rxh7 ! . II) S.Kg2? RbS 9.Rb7 Rxb7 1 O.hS=Q
Kb 1 ! . Although White has queened and is on the move, he still loses, e.g., 1 1 .QcS
Rg7+ 1 2 .Kh3 c l =Q 1 3 .QfS + Qc2 1 4.QbS+ Ka2 1 S .QaS+ Kb2 1 6.QeS+ Qc3 +.
III) S.Kh2? (a poor try) S . . . Kb2 9.Rb7+ Ka3 , and as soon as White plays Rc7 he
will be dispatched by . . . Rxh7+. IV) S.Rb7? (to prevent S . . . RbS) S . . . RdS 9.Kf2
Kd 1 ! 1 0.Rd7+ Rxd7 1 1 .hS=Q Rd2 +. Note how, in this last variation, Black's
9 . . . Kd 1 ! left the d2 -square open for his Rook so that White could not win it by
checking with the new Queen. S ... RbS We have already discussed how White

621

PAL B E N KO: MY LIFE, GAM ES A N D CO M POSITIO N S

622

would win the black Rook after S . . . RdS 9.Kf2! Kd2 1 O.Rd7+ ! Rxd7 1 1 .hS=Q
c l =Q 1 2 .Qh6+. 9.Kh2 Kb 1 Black does not win after 9 . . . Kd2 1 O.Rg2 + Kd3
1 1 .Rg3 + Kc4 1 2 .Rg4+, when White can meet 1 2 . . . Kb3 ? with B.RgS. 1 0.RgS
Rb2 1 1 .hS=Q c 1 =Q+ 1 2 .Rg2 Qf4+ 1 3 .Kh1 , Yz- Yz . Black has no more good
checks and must acquiesce to a draw (B . . . Rxg2) since 1 3 . . . Qfl +? ? even loses to
1 4.Rg l .

( I SO) White to move and win.

TWIN.
(Chess Life 1 9S6.)
l .Rg7+! l .RhS produces our next problem, with
Black having the move. The game would be
drawn after 1 . ..e5 2 .KfS Kg5 3 .Kf7 h4 4.Ke6
Kg4 5 .Kd5 h3 6.Ke4 Kg3 7 .Ke3 Kg2 S.Ke2 h2
9.RgS+ Kh3 1O.Kf2 h l =N+ 1 1 .KB e4+! 1 2 .Kxe4

Ng3 + B.KB Nh5 ! . 1 ...KfS 2 .Kh7 h4 3 .Kh6 h3 4.KhS Kf4 S .Kh4 h2 6.Rf7+
Ke3 7.Rf1 eS S.Kg3 e4 9.Kxh2 Kd2 1 O.Kg2 e3 1 1 .Kf3 e2 1 2 .Rf2, 1 -0.
(b) Relocate Rh7 to hS. 1 .Kf8 KgS 2.Kf7! White must preserve the e-pawn so
Black won't have any stalemate possibilities. 2 . . .h4 3 .Ke6 Kg4 4.KeS h3 S.Ke4
Kg3 6.Ke3 Kg2 7.Ke2 h2 S.RgS+ Kh3 9.Kf2 hl =N+ 1 0.Kf3 Kh2 1 1 .Rg2 +
Kh3 1 2 .Rg7 Kh2 1 3 .Rxe7, 1 -0. It's interesting to see that a small change led to

a completely different solution.

( 1 S 1 ) White to move and win.

(First prize, Sakkelet 1 9S7.)
White seems to be in trouble since he's a piece
down. He faces mating threats and his passed
pawns appear to be frozen in their tracks .
l . Q d l ! White threatens 2 . Q d S ! BxdS
3 .cxdS=N+ ! . 1 . . .QaS If Black tries 1 . . .QeS!?,
White must avoid 2 .QdS? since the position
would be a draw after 2 . . . BxdS 3 .cS=Q+ Kd5 4.b7 Qe7 ! 5 .bS=Q (White gets no
where with 5 .Qxf5 + Kc6 6.bS=Q f6+.) 5 . . . f6+ 6.Kg6 QeS+. However, White still
has a win by answering 1 . . .QeS with 2 .Qe2 + Be5 3 .Qxe5+ Kxe5 4.b7 Qc6 5 .cS=Q
Qg6+ 6.KhS Qxh6+ 7.KgS Qg6+ S.KfS Qh6+ 9.KeS QhS+ 1 O.Kd7, and Black
must resign. 2 . QdS ! 2 .Qb3 + Ke7 3 .Qe3+ Kd6 4.Qd3 + isn't nearly as strong.
2 . . . BxdS 3 .b7!

E N DGAM ES

A remarkable situation: the two pawns beat the

Queen and Bishop. 3 . . . Qxb7 4.cxd8=N+ Ke7
The game comes down to a race between the
passed pawns. Black's alternative is 4 . . . KdS
S .Nxb7 f4 6.NaS Ke4 (Black also gets caught
after 6 .. .f3 7 .Nb3 f2 8.Nd2 b4 9.Kg7 b3 1 O.h7
b2 I l .h8=Q fl =Q 1 2 . Qd8+ Kc6 1 3 .Qc8+ Kd6
14.Qb8+ and I S .Nxfl .) 7.Nb3 Ke3 8.Nd4! Kxd4
9 . Kg8, and the h-pawn queens with check.
5 .Nxb7 Kf8 Black tries to trap the white King
and then push his own passed pawns. 6.Nc5 f4 7.Nd7+ Ke7 8 .Kg7 f3 9.Nf6
Weak is 9.Nb6? due to 9 . . . Ke6. 9 ... f2 1 0.Nd5+ Ke6 1 1 .Ne3 , 1 -0.

( 1 82) White to move and win.

(Benko and E. J anosi, First Prize,
Ajedrez de Estilo 1 98 7 .)
LV l .Rf7? Kxa2 2 .Rf8 Kb3 3 .f7 Kc3 4.d4 Kd3
=. 1 .. .Ka3 ! 1 . . .Kxa2 2 .Kg3 gS 3 .Rh2 + followed
by 4.Rf2 . 2 .d4! 2 .Kg3 ? Rf1 ! 3 .d4 RfS ! =. 2 . . . g5

2 . . .RfS 3 .Kg3 ! Kb4 4.Kg4 (4.dS? KcS ! =) 4 . . . Ka4

S .a3 KbS (S . . . Kxa3 6.Rhh and Rf3) 6.RhS ! , and
wins. 3 .Kg3 Kb4 4.d5 ! 4.a4? KaS ! =. 4 . . . Kc5 4 . . . cxdS S .Rh4! wins. 5.dxc6 Kxc6
6.a4 6.Rg7? Kd6 7.a4 Ke7 =. 6 . . . Kc5 ! 6 . . . Kd6 7.aS. 7.Rg7 Kb6 7 . . . Kd6 8.aS
Ke7 9.a6 Kf8 1O.Rh7 wins for White due to 1 0 . . . Ra4 I l .a7 Rxa7 1 2 .Rh8+ Kxf7
1 3 .Rh7+. 8.Kg2 ! 8.aS+? Ka6 ! =; 8.Kh3 Ka6 9.Kh2? Rf3 1 O.Kg2 g4 8 . . . Ka5
8 . . . Ka6 9.Kh3 Kb6 1 O.Kg3 or 8 . . . Rg4+ 9.Kf3 Rf4+ 1 0.Kg3 Ka6 1 1 .aS . 9.Rxg5+
Kxa4 1 O.Rg4 Rxg4+ 1 1 .Kf3 Rgl 1 2 .Kf2 , 1 -0.
=.

( 1 8 3 ) White to move and win.

(Chess Life 1 98 7 .)
1 .Kb7 Kg4 ! ? 1 . . .g4 2 .Kc6 g3 3 .KdS g2 4.Ra l
Kg4 S .Ke4. 2 .Rf6 2 .Kc6? Kf3 3 .KdS g4 =.
2 ... Kh3 3 .Kc6 g4 4.Kd5 g3 5.Ke4 g2 6.Kf3 ,

1 -0.

623

PAL B E N KO: MY LIFE, GAM ES AN D CO M POSITIO N S

624

( 1 84) White to move and win.

(Second Prize, Magyar Sakkelet 1 988.)
l .b4 Ne2 1 . . .Ne4? 2 .KfS Nd2 3 .B g6 Nxf3
4.Ke4. 2 .b5 Nd4 3 .b6 Nc6 3 ... NxB? 4.b7 NeS+
S .KfS Nd7 6.Bg6 Kg8 7 .Be8 Nb8 8.Ke6 gS
9.BhS . 4.b7 Nd8 ! 4 ... Nb8 S .f4 and fS -f6, when
. . . gxf6 is met by Kh6. 5 .b8=B ! Ne6 S . . . Nc6
6.Bg3 ! ! (6.Bc7? Ne7+ 7.Kf7 NdS =) 6 . . . Ne7+
7.Kf7 Ng6 !? 8 .Bg8 NeS + 9.Ke8 ! , and wins. 6.Bd6 ! ! 6.BeS? Nf8+ 7.Kf7 Ng6 =,
or 6.Bg3? Nf4+ ! (6 .. .Nf8+? 7.Kf7 Nxh7 8.Bh4!) 7.KgS Ne6+! (7 . . . Ne2? 8.Bd3 !
Nxg3 9.Kg4 Nh 1 1 0.Be2) 8 .KfS Nd4+ 9.Ke4 Ne2 1O.Bg6 Nxg3 + 1 1 .Ke3 Nfl +
1 2 .Ke2 Nh2 =. 6 . . . Nf8+ 7.Kf7 Nxh7 8.Be7 ! , 1-0.

( 1 8S) White to move and draw.

(E. G. 1 988.)
l . Kd7 ! l .KfS ? Kc4 2 . Kg4 Kb 3 3 . c4 Kxc4
4.Kxh4 Kb3 S . g4 Kxb2 6.gS a3 7 .g6 a2 8 .g7
a 1 =Q 9 . g8=Q Qh 1 + 1 0 .KgS Q g 1 + . 1 . . .Kc4
2 .Kc7 Kb3 3 .Kb8 Kxb2 4.c4,

Ih _1/z .

( 1 86) White to move and draw.

(E. G. 1 989.)
White is light-years ahead in material, but his
King appears to be in a mating net. l .Nc3 ! Rf1 +
1 . . .Rxc3 2 .Kb 1 and 1 . . .Kxc3 2 .Ka2 are both safe
for White. 2 .Nh l Rf2 Threatening mate on h2 .
3 .Na3 ! The only way to save the game! The
tempting 3 .Nc3 Kxc3 4.Bg3 (4.Ba7 Re2 S .Kb l
Rb2 +, when both 6.Ka 1 RbS and 6.Kc 1 Ra2 win for Black) 4. . . Rf3 S .Be 1 + loses,
though just barely: S . . . Kb3 6.Kb 1 Rfl 7 .Kc 1 Rxe 1 + 8.Kd2 Rfl ! 9.Kd3 !? (9.gS
Kc4 1 0.Ke3 KdS locks out the white King from the kingside) 9 . . . Kb4! (9 . . . Rxf6?
1 0.Ke4 Kc4 1 1 .gS Rfl 1 2 .KeS Rg 1 1 3 .Kf6 KdS 1 4.g6 Kd6 l S .g7 is only a draw)
1 O.Ke4 ( 1 O.Kd4? Rxf6 1 1 .KeS Rg6 ! 1 2 .KfS Rg8 1 3 .gS KcS 1 4.g6 Kd6 l S .Kf6
Rf8+) 1 0 . . . KcS ! 1 1 .KeS Re 1 + 1 2 .KfS Kd6 1 3 .gS Rfl + 1 4.Kg6 Rg 1 ! ( l 4 . . . Ke6?
l S .Kg7 Rf2 1 6 .g6 ! Rxf6 1 7 .Kh7 Rfl 1 8.g7 Rh 1 + 1 9.Kg8 ! draws.) l S .KfS Kd7
1 6.g6 Ke8 . 3 . . . Kxa3 3 . . . Rxf6? 4.BeS makes things easy for White. 4.Kb l Rb2 +
5 .Kc 1 Rxb8 6.g5 Rf8 Black has managed to stop the white pawns from advanc
ing. Now the King race begins. 7.Kd2 ! Kb4 8.Ke3 Kc5 9.Kf4 Kd6 lO.g6 ! !

E N DGAM ES

White would lose after the obvious 1O.Kf5? RaS 1 1 .Kg6 ( 1 l .g6 Ra 5+ 1 2 .Kg4
Ke6) 1 1 . . .Ke6 1 2 .Kh6 Kf7. 10 . . . Ke6 White also saves the game after 1 0 . . . Rxf6+
I l .Kg5 Rfl 1 2 .g7 Ke7, when 1 3 .gS=N+ ! is a draw, and not 1 3 .gS=Q?? Rg l +,
and Black wins. 1 l .KgS RbS 1 2 .7 Ke7 1 3 .Kh6 An amazing position that was
already shown to be drawn by Cozio in 1 7 76. Black would win if we shifted the
position to the left by placing the pawns on e7 and f6. The reason for this will
soon become clear. l L .Kf8 1 4.Kh7 RaS I S .KhS Ke7+ 1 6.Kg7 RbS 1 7 .Kh7
Kf6 Black appears to be winning, but that assessment turns out to be superficial.
I S.fB=Q+ ! Rxf8 1 9.97 Rf7 20.KhS Rxg7, Yz - Yz .

( 1 S7) White to move and win.

(E. G. 1 9S9, after T. Gorgiev.)
l .Bf6+ l .Ra4 Kg7 2 .RaS (2 .Rd4 Nc6 3 .Rd7+
KfS 4.BxdS KeS =) 2 ... Kg6 3 .Bh4 Kh5 4.Bf6
Kg6 5 .Be5 Bb6 6.Ra6 Nd5 ! , 1 . . .Kh7 2 .Rg7+
Kh6 3 .Rf7 Kg6 3 . . . Nc6? 4.BxdS NxdS 5 .Rd7
Ne6 6.Rd6. 4.RfB f4 S .Ke l ! ! 5 .Kd l ? f3. S . . . 3
6.Kdl t2 7.Ke2 Nc6! ? S.BxdS Kg7 9.ReS Kf7
1 0.RhS Kg7 I 1 .Bf6 + ! Kxf6 1 2 .Rh6+, 1 -0.

( I SS) White to move and win.

(First Prize, Magyar Sakkelet 1 9S9.)
A position of deceptively simple appearance.
Isn't the white Knight trapped? l .KcS The
more natural l .Kc7? only draws after 1 . . .Ke7 !
2 .Kc6 Ke6, and White will soon lose his last
pawn. 1 . ..Kf8 Now 1 . . .Ke7 fails to 2 .Kc7 Ke6
3 . KdS! Kf5 4. Nf7 Kg6 5 . KeS Kg7 6 . Ke 7 .
2 .Ng6+ ! ! The point! The Knight escapes since 2 . . . hxg6 Ieads to a winning King
and pawn endgame for White: 3 .KdS Kf7 4.Kd7 KfS 5.Ke6 Kg7 6.Ke7 KgS 7.Kf6
Kh7 S.Kf7, etc. Instead of 2 .Ng6+, tempting is 2 .Kd7 Kg7 3 .Ke6 KxhS? 4.Kf7 .
However, after the correct 3 . . . h 6 4.g6 h 5 Black draws the game. 2 . . . Kf7 The
fight is not over, because White must still make another difficult choice. On
2 . . . hxg6, White wins by 3 .KdS Kf7 4.Kd7 KfS 5 .Ke6 Kg7 6.Ke7 KgS 7.Kf6 Kh7
S.Kf7 KhS 9.Kxg6. 3 .Nf4! ! We can appreciate this move only by examining the
plausible alternative 3 .Nh4. White wins easily after 3 . . . Ke6 4.KdS Ke5 5 .Ke7
Kf4 6.Kf6, but Black has a resource: 3 . . . KeS! ! . This position is worth careful
study. White has no time to correct his error with 4.Ng2 Ke7 5 .Nf4, as 5 . . . Kd6
6.Nh5 Ke6 holds the draw. Even after the text, though, Black can still make the
outcome close. L .h6 ! 4.g6+ Of course not 4.gxh6? KgS followed by . . . KgS-

625

PAL B E N KO: MY LIFE, GAM ES A N D CO M POSITIO N S

626

h7xh6, drawing. 4 ... Kf6 S .Kd7 hS 6.KeS Kg7 Or 6 ...h4 7.Kf8 h3 8.g7 h2 9.NhS+,
and 1 O.Ng3 . 7.Ke7 h4 S.Ke6 h3 9.KfS h2 lO.NhS + Kf8 1 1 .Ng3 , 1 -0. The
win is clear: 1 1 . . .Kg8 1 2 .Kf6 Kf8 1 3 .g7+ Kg8, and now both 1 4.Nh l and 1 4.NfS
end the game.

(1 89) White to move and win.

(Chess Life 1 990.)
l .Ke7 d2 1 . . .c3 2 .Kf7 c2 3 . Rc6 Kh7 O . . . d2
4.Rc3) 4.g4 gS S .Rc8 d2 6.Rxc2 . 2 .Rd6 c3 3 .Kf7
Kh7 4.g4 e2 S.gS ! S.Rxd2 ? c l =Q 6.Rh2+ Qh6
=. S . . . d l =Q 6.Rh6+ gxh6 7.g6+ KhS S.g7+
Kh7 9.gS=Q mate.

( 1 90) White to move and win.

(Special Prize, Magyar Sakkelet 1 990.)
1 .NO gxhS ! 1 . . .bxc6 2 .Nxh2 Kxh2 3 .hxg6 Bd4

4.b4 g l =Q S .Rxg l Kxg l 6.Kd6 or 1 . . .Qh 3 +

2 .Ke7 Qxf3 3 .Bxf3 gxhS 4.BxhS Kh2 S .Rxg l
Kxg l 6.Bf3. 2 .KdS ! ! Other moves don't d o the
trick: 2 .Bxb7? Qh3 + 3 .Kf6 Qxf3 + ! 4.Bxf3 h4
S .KgS h3 6.Kg4 h2 7.Bdl Bf2 ! (7 . . . BcS? 8.Kh3
gl=N+ 9.Kg3) 8.Kf4 (8.Kh3 gl=N+ =) 8 . . . Bg3 +!
9.Kg4 Bf2 1 O.KfS Bb6 I l .Ke4 Kg l =; 2 .Be4? Qh3 + 3 .Kf6 Qxf3+ 4.Bxf3 Kh2 !
S . Bxg2 Kxg2 6.KgS Be3+ 7 .KxhS Kf3 8.Kg6 Ke4 9.Rd l Bd4 1 O.b3 KdS 1 1 .Rc 1
bS 1 2 .Kf7 Kd6 ! 1 3 .Ke8 Bf6 = . 2 . . .bxe6+ 3 .Ke4! ! e S 3 . . .h 4 4.b4! . 4.KdS ! ! e4
S.Ke6 h4 6.Kb7 h3 7.KaS ! c3 S.bxc3 QbS+ 9.RxbS h2 l O.RhS, 1-0.

( 1 9 1 ) White to move and win.

(Inside Chess 1 990.)
l .Rb7+ KxdS 2 .Rxd7+ Rxd7 3 .NeS Nae6
4.Nxd7 Kxd7 S.eS Nd4 6.e6+ ! , V2-1h. I wrote

an article in Inside Ch ess magazine about

endgame compositions by World Champions.
A Lasker endgame, similar to this one, was
cooked since it seems that he wasn't familiar with
Troitzky's discoveries concerning two-Knights-and-pawn endings. This compo
sition is an improvement over Lasker's attempt.

E N DGAM ES

( 1 92) White to move and win.

(Inside Chess 1 990.)
l .Bh7 Rf8 This forces White to find a pretty
win. Other moves last a bit longer, but allow
White to win mundanely. For example, 1 . . .Rb8
2 .Qf7+ KhS 3 .Bg6 Rb6+ 4.Kd5 Rb5 + 5 .Ke4
Re5+ 6.Kd3 , etc. Finally, 1 . ..Rxh7 allows a quick
mate: 2 .Qf7+ KhS 3 .QfS mate. 2 .Qg6+ KhS
3 .BgS ! RxgS 4.Kf7 Rxg6 5.fxg6 c 1 =Q 6.g7+ Kh7 7.gS=Q mate.

( 1 93) White to move and win.

(Inside Chess 1 990.)
l .Rg7 Rb4+ 2 .Ke5 Rh4 3 .Kf6 RxhS 4.Ra7
KeS 4 . . . KxgS 5 .Kg6. 5 .Ke6 KdS 6.Kd6 KeS
6 . . .KeS 7.Nf6+ KfS S.Ke6. 7.Ne7+ KbS S.Ne6+
KeS 9.Re7 mate. Capablanca tried to create a

similar mate in a problem with seven pawns. Like

Lasker, his problem was cooked.

( 1 94) White to move and win.

(Inside Chess 1 990.)
l .Kgl e5 Both l . ..Rf4 2 .Rb5 ! and 1 . ..Rh6 2 . ReS
also win for White. 2 .Rf5 ! Rxf5 3 .b7, 1 -0 .
Alekhine tried t o express this same idea i n a
problem, but with more men and only two the
matic variations.

( 1 95) White to move and win.

(Inside Chess 1 990.)
l .Re5 ! l .Ke7? Rh7+ 2 .Kd6 Rxd7+ 3 .Kxd7 Kc5 !
1 . . .Rd2 2 .KeS ! 2 .Ke7? Kc7 3 .Rc5 + Kb6
4.Rc2 Rd4 2 . . . Ke7 2 . . . Ka5 3 .dS=Q+ RxdS+
= .

=.

4.KxdS Ka4 5 .Kc7 b4 6.Kb6 b3 7 . Rb5 ! Ka3

S.Kc5 b2 9 . Kc4 Ka2 1 0 . Ra 5 + . 3 .Re7 ! Kb6
4.dS=Q+ RxdS+ 5.KxdS Ke5 6.Rb7 b4 7.Ke7
Ke4 S.Kb6 b3 9.Ka5, 1 -0. The next endgame is similar to this but is shifted to

the right.

627

PAL B E N KO: MY LIFE, GAMES A N D CO M POSITIO N S

628

( 1 96) White to move and win.

(Inside Chess 1 990.)
l .Rf5 ! Re I 2 .Kf7 ! 2 .KfS? Kb5 ! =, and not
2 ... Kd7? 3 .Rf7! . 2 ... Kd7 3 .Rd5 + Ke6 4.Rd l
Re4 5 . eS=Q+ RxeS 6.KxeS e 4 7 .Ke7 Ke5
S.Ke6 c3 9.Ke5 Ke4 1 O.Ke4 e2 1 1 .Rc 1 Kc3
1 2 .Ke3 , 1 -0.

( 1 97) White to move and win.

(E. G. 1 99 1 .)
l .Bb l f4 1 . . . d4 2 . g6 d 3 3 .g7 a2 4. Bxa2 d2
5 .Bbh ! . 2 .Ke5 Kb3 3 .g6 f3 4.g7 f2 5 .gS=Q
f1=Q 6.Qxd5+ Kc3 7.Qd4+ Kb3 S.Qa4+ Kxa4
9.Be2 mate.

( 1 9S) White to move and win.

(E. G. 1 99 1 .)
l .R3b6+ Ka5 2.R6b7 a6 3 .Rb5+ Ka4 4.R5b6
a5 4 . . . Ka5 5 .Rxa6+ Kxa6 6.RaS+. 5.Rb4+ Ka3
6.Rd4 a4 6 . . . e l =Q 7 . Rb3 + Ka2 S.Ra4 mate.
7 . Rxa4+ Kxa4 S . RaS + Kb4 9 . Rxa l Ke3
I O.Re l , 1 -0 . This is an improvement on a

cooked endgame by T. Gorgiev.

( 1 99) White to move and win.

(First prize, Magyar Sakkelet 1 99 1 .)
l .g7 Rg4+ 2 .Kf3 ! 2 .Kf2 ? Kf6 wins for Black.
2 ... Kh6 3 .e7! 3 .Kxg4? d2 4.gS=Q d l =Q+ 5 .Kh4
Qh5+! 6.Kg3 Qf h =. 3 ... d2 4.Ke2 Rxg7 5.Ng6

Not 5 . eS=Q? Re7 + ! 6.Qxe7 d l =Q+ 7 . Kxd l ,

stalemate. If 5 .Nd5 ? , Black draws by 5 . . . RgS
6.Nf6 RbS 7 .eS=Q RxeS S.NxeS Kg6 1O.Nd6
Kf6 I l .Nb5 Ke6 1 2 .Nxa7 Kd7 =. 5 ...RgS 6.NfB Rg5 ! ? 7.Ng6 ! (And not 7.eS=Q?
Re5 + ! S.Qxe5 d l =Q+ =, or 7 .eS=R Rd5 S.Kd l Rd6 9.Re6+ Rxe6 1 O.Nxe6 Kg6 =
I 1 .Nd4 Kf6 1 2 .Nb5 Ke6 1 3 .Nxa7 Kd7 1 4.Nb5 Kc6 1 5 .Kxd2 Kb6 1 6.a7 Kb7),
1 -0.

E N DGAM ES

(2 00) White to move and draw.

(Szachista 1 99 1 .)

629

LNbS eS l . ..e6 2 .Nd4 eS 3 .Nb3 e4 4.Nd2+ Ke2

S .Nxe4 Nxe4 6 . Kh6 =. 2 .Nd6 Ka l 2 . . .Nc3
3 .Kh6 =. 3 .Ne4 e4 4.Nd6 c3 S .NbS e2 6.Nd4
c1N 7.Ne2 + Kb l 8.Na3 + Kal 9.Ne2 + ,

Y2 - Y2 .

(2 0 1 ) White to move and draw.

(Chess Life 1 99 1 .)
l . Bf8 ! gxh6 2 . B e 7 ! Kg7 2 . . . a l =Q 3 . B f6+
Qxf6 stalemate. 3 .BeS Kf7 L.a l =Q 4.Bd4+
Qxd4, stalemate . 4. Bd4 Ke6 S . Kh4 Kd S
6.Bb2 6 . B g7 Ke4 7 . Kg3 Ke 3 8 . Bxh6+ f4+ !
9.Bxf4+ Ke4, and wins . 6 . . . Ke4 7.Kg3 Ke3
8.Be 1 + Ke2 9.Bb2 ,

Y2 - Y2 .

(2 02) White to move and draw.

(Chess Life 1 99 1 ).
LRd6 Nc 1 1 . . .Nf4 2 .Kh3 h3 (2 . . . Kh l 3 .Rd4)
3 .Rf6 h2 4.Rh6 =. 2 .Ka3 h3 2 . . . hS 3 .Rd4! h3
4.RdS ! =, or 2 . . . Kh l 3 .Rd7 h3 4.Rh7+ Ke2
S . Re 7 + =. 3 .Rd7 h6 4.Rd6 hS S . RdS h4
6.Rd4 h2 7.Rxh4 d2 8.Rxh2 d l =Q 9.Ra2 +
Nxa2 , Y2 - Y2 .

(2 03) White to move and win.

(E. G. 1 99 1 .)
Le7 1 .Bg6? g3 ! 2 .c7 g2 3 .e8=Q g l =Q 4.Qh3 +
KgS ! =. 1 . . .b2 2 .Bg6 Kxg6 3 .e8=Q b l =Q
4.Qxg4+ Kh6 S .Qh4+ Kg6 6.Qh7+, 1 -0.

PAL B E N KO: MY LIFE, GAM ES AN D CO M POSITIO N S

630

(2 04) White to move and win.

( Ninth place, WCCT 1 99 1 .)
l .Be4 l .Kxh4? Kh2 =. L.h3 2 .Kh4! 2 .Kg4?
h2 3 .Bb 1 Bd4 or 3 .Ra3 BcS =. 2 ... h2 3 .Bbl Bf2+
L .Bd4 4.Bc2 + Bg1 S . Bd l Bf2 + 6.KhS (6.Kh3
gS=N+ 7.Kg4 Kg2 =) 6 ... Bd4 (6 ... Kg 1 7.Bf3 +
Be l S.Kg4 Kf2 9. Bxg2 Kxg2 1 O.Rxe 1 h l =Q
1 1 .Rxh l Kxh l 1 2 .Kf3 Kg l 1 3 .Ke4 Kf2 1 4.KdS
Ke3 l S .Kc6) 7 .Bxb 3 + Bg1 S.Bdl Bd4 9.Bf3+ Bg1 1 O.b4. 4.Kh3 Bgl 4 . . . g l =N+
S .Kg4. S.Kg3 Bf2 + 6.Kf3 6.Kf4? Be3+ 7.KfS Kg 1 =. 6 . . . Bgl 6 . . . g 1 =Q 7 .Be4
Qxa 1 (7 . . . Qe 1 S.Rxe 1 + Bxe 1 9.Ke2 +) S.Kxf2 mate; 6 . . . Bd4 7.Be4+ Bg1 S.Rfl
gxfl =Q+ 9.Kg3 + Qg2 + 1 0.Bxg2 mate. 7.Ke2 Bd4 8.Bd3 + Bgl 9.Rfl gxfl =Q+
l O.Kxfl , 1 -0. White will mate with Be4.

(2 05) White to move and win.

(Chess Life 1 99 1 .)
l .NfS+ Kh7 1 . . .Kg6 2 .m Kxf7 3 .Nd6+. 2 .1
Re6 2 . . .ReS 3 . fS=R! , and not 3 .Nh4? Rb 5 +

(L.Kg7? 4.fS=Q+ KxfS S .Ng6+) followed by

. . . Kg7 . 3 .ffi=N+ ! , 1 -0. White avoided 3 .fS=Q?
ReS+ 4.QxeS, stalemate. These ideas came from
a Troitzky composition that I cooked.

(2 06) White to move and draw.

(First Prize, Sakkelet 1 992 .)
l .BeS 1 .Kg2? Ng3 . L.Rd 2.Bd4 Rd l 3 .BeS
Bad is 3 .Bb6? Kb7 4.BcS Kc6 S .Ba7 Ra l 6.Bd4
Rb I 7.Kg2 KdS S.Ba7 Ke4 9.h4 KfS 1 0.hS Kg4
1 1 .h6 Rb7 1 2 .Bd4 Ng3 1 3 .Bg7 Re7 1 4.h7 Re2 +
l S .Kg1 Kf3, mating. Other moves that lose are
3 .BgI? Ng3 , and 3 .Be3? Rd3 . 3 . . . Kd7 4.Kg2
Ke6 S.Bb6 S .Ba7? Rb I 6.h4 KfS or S .h4? Kf7 6.hS RdS . S . . . KfS S . . . Rc 1 6.Be3
Rb I 7 .Bf4 =. 6.Be7 6.Ba7? Rb I 7 .BcS Kf4 S.BfS Ke3 9.h4 Ke2 1 0.hS Nf2 I l .h6
Nd3 1 2 .h7 Ng4+ 6 ... Ke4 7.Bh2 Kd3 8.h4 Ke2 S . . . Ke4 9.hS KfS 1 0.h6 Kg6
1 1 .Bf4 Re I 1 2 .Bh2 Kxh6 1 3 .Bg1 Ng3 1 4.Bf2 =. 9.Bgl Ng3 l O.Kxg3 Rxgl +
1 1 .Kf4 Kd3 1 2 .hS Kd4 1 3 .KfS KdS 1 4.h6 Rh l l S .Kg6 Ke6 1 6.Kg7 Rg1 +
1 7.Kf8, Ih_ lh ,

E N DGAM ES

(2 07) White to move and win.

(Chess Life 1 994.)
l .Nf7+ Kh7 2.Bh3 ! QbS 2 . . . Qd5 3 .Bg4 Qb5

4.Ng 5 + KhS 5 . a 3 Q d 3 6.Nf7 + Kh7 7 . B f3 ! .

3 .NgS+ Kh6 4.Bg4 Qxb4+ 4 ... Qd3 5 .Nf7+ Kh7
6.Bf3! and 7.Be4+. S.Kg8 Qxf4 6.Nf7+, 1 -0.

(2 0S) White to move and win.

(Special Prize, Magyar Sakkelet 1 994.)
l .h7 Rd8 2 .Rg8 Rd2 + 3 .Kg3 Bad is 3 .Kh3 ? ,
since 3 . . . Rd5 i s drawn. 3 . . .Rd3 + 4.Kg4 Rd4+
S.KgS White has to be careful not to allow Black
to give up his Rook for the white pawn. S ...RdS+

5 . . . a2 6.hS=Q a l =Q 7.RbS+ Kc3 S.Qh 3 + Rd3

9.RcS+ Kd2 1 0 .Qh2 + Ke3 1 1 .ReS+. 6.Kxg6
Rd6+ 7.KgS ! White has completed his first goal: capturing the g-pawn and open
ing the g-file. He has to avoid 7 .Kh5? since Black can at least draw with 7 . . . Rd l
S.Rg3 + Ka2 9.Kh4 RdS 1 0.RgS Rd l I l .Rg2 + Kb l 1 2 .Kh3 Rh l +, and after the
Rook trade, both sides queen. 7 . . . RdS + 8.Kg4 Rd4+ 9.Kh3 ! Rdl 1 0.Rg3 + !
Kb4 1 1 .Kh4 Rd8 1 2.Rg8 Rdl 1 3 .Rg4+ Kb S 14.KhS Rd8 l S.Rg8 Rdl 1 6.RgS+
Kb6 1 7 .Kh6 Rd8 1 8.Rg8 Rd l 1 9.Rg6+ Kb7 20.RgS ! 20.hS=Q? Rh 1 + 2 1 .Kg7

RxhS 2 2 .KxhS a5 2 3 .Kg7 a2 24.Rg l Kb6 2 5 .Kf6 Kb5 26.Ra l Kb4 2 7 .Rxa2 a4 =.
20 . . .Rd6+ 20 . . . RdS 2 1 .RgS Rd l (2 1 . . .a2 2 2 .hS=Q RxgS 2 3 .Qb2 + followed by
2 4.Qxa2 , with a theoretical win) 2 2 .hS=Q Rh l + 2 3 .Kg5 RxhS 24.RxhS Kb6
2 5 .Rh3 ! Kb5 26.Rxa3 a5 2 7 .Kf4 Kb4 2S.Ra l a4 2 9 .Ke3 a3 3 0.Kd2 Kb3 3 1 .Rb 1 + !
Ka2 3 2 .Rb4 Ka l 3 3 .Kc3 ! , and White wins. 2 1 .KhS ! Rd l 22 .Rg4 RdS + 2 2 . . . a2
2 3 .hS=Q Rh l + 24.Rh4 a l =Q 2 5 .Qh7+, and White will win the Rook with sub
sequent Queen checks. 2 3 .Kh4 Rd l 24.Rg3 Rd4+ 2 S .Kh3 Rdl 26.Rg2 Rd3 +
2 7.Kh2 Rd8 28.Rg8 Rd2 + 29.Kg3 Rd3 + 3 0.Kf4, 1 -0.

(2 09) White to move and win.

(E. G. 1 997.)
l .NfS + Kh7 1 . . .Kg6 2 .f7 Rc4+ 3 .Kd7 Kxf7
4.Nd6+. 2 .f7 ReS 2 . . . Re6 3 .fS=N+ ! wins for

White, but not 3 .fS=Q? ReS+! 4.QxeS, stalemate.

3 .Ne7 Kg7 Both 3 . . . Rc5 + 4.Kd7 Kg7 5 .KeS and
3 . . . Rb5 4.Ng6 ! (and not 4.fS=Q? RbS+ !) win for
White. 4.fS=Q+ KxfS S .Ng6+, 1 -0.

63 1

PAL B E N KO: MY LI FE, GAM ES A N D CO M POSITIO N S

(2 1 0) White to move and win.

(Second Prize, Magyar Sakkelet 1 994.)
632
l .NgS ! l .Nh2 ? c6 ! 2 .Bxc6 c3 3 .Ba4 c2 ! , forcing
a stalemate after 4. Bxc2 f1 =Q+ S .Nxfl . 1 ...ft =Q+
2 .BB 2 .Nf7+? Qxf7+ 3 .Kxf7 c6! 4.Bxc6 c3 S .Be4
c2 6.Bxc2 , stalemate. 2 . . . Qgl or 2 . . . Qd3 3 .Nf7+
Kh7 4 . B e4+ ! . 3 .Nf7 + Kh7 4 . B e 4 + ! Qg6
S .NgS + ! S .Bxg6+? Kxg6 6.Ne S + KfS 7 .Nxc4

Ke4, followed by . . . Kd3

S ...Kh6 6.Bxg6 Kxg6 7.Ne4 KfS S.Nc3 Kf6 9.KeS

Ke6 1 0.KdS Kd6 I 1 .KeS Ke6 1 2 .KbS, 1 -0. Not falling for 1 2 .e4? Kd6 =.
=.

(2 1 1) White to move and win.

(Chess Life 1 994.)
l .Re l Rb3 + 2 .Kg2 ! 2 .Kf4? Kd8 3 .e7+ Ke8 4.a4
Rb4+ 2 . . . Rb2+ 3 .Khl ! KdS 4.e7+ KeS S.a4
Rb4 6.aS RbS 7.a6 Rb6 S.a7 Ra6 9.Rgl ! Kxe7
10.RgS ! Rxa7 1 l .Rg7+, 1 -0.
=.

(2 1 2 ) White to move and draw.

(Third Prize, Magyar Sakkelet 1 996.)
l .Rg2 BbS 2 .Kb7 Rb 1 + 3 .KaS ! RbS 3 . . . Rh4
4.Re2 + Kfl S . Rf2 + Kg 1 6 . Ra2 Rb6 7 . Rg2 +
Kh 8 8 . Rb 2
4 . Re 2 + Kfl S . Rf2 + Kg l
6.Rg2 + 6.Rb2? a6 7 . Ra2 as. 6 . . . Khl 7.Rg 1 +
Kh2 S.Rg4! Rb6 8 . . . aS 9.Rb4! = ; 8 . . . a 6 9.Ra4
Rb6 1 0 . Rb4! B c 7 I 1 . Ra4 =. 9 . Rg2 + Kh 3
1 0.Rg4! Bg3 I 1 .Ra4 1 1 .Kxa7 Bf2 . 1 1 . . .a6 1 2 .Ka7, Y2 - Y2 .
= .

(2 1 3) White to move and draw.

(Chess Life 1 996.)
l .Kf4! Not l .KeS? Kc7 2 .cS dS ! 3 .cxd6+ Kd7,
and Black wins. 1 . . .Ke7 2 .Ke S ! 2 .KgS? Kd6
3 .Kg6 Ke S wins for Black. 2 . . . KeS 2 . . .Kd8
3 .Kd6 Ke8 4.cS Kd8 5 .f6! gxf6, stalemate. 3 .Kf4!
3 .Kd6? Kd8 4.cS Ke8 S .Kc7 Ke7 6.Kb6 d6!
7.Kxc6 dxcS 8.KxcS Kf6 wins for Black. 3 ... KdS
4.KgS Ke7 S .Kg6 Kf8 6.eS 6.f6? gxf6 7 .Kxf6 cS ! wins for Black. 6 . . . KgS 7.KhS !
Kf7 S.KgS Ke7 9.Kg6 ! , Y2 - Y2 . Black can't make progress.

E N DGAMES

(2 1 4) White to move and draw.

(Chess Life 1 996.)
l .Kg4! Kd7 2.Kf5 ! Kd8 2 ... Ke8 3 .Ke6 Kf8 4.dS
Ke8 S .g6 ! hxg6 is stalemate. 3 .Ke6 ! From this

point on, the solution differs from the previous

problem. Trying to follow the previous solution
leads to failure: 3 .Kg4? Ke8 4.KhS W S .Kh6
Kg8 6.dS Kh8 is only an illusion because there
is one file less at the edge. 3 ... Ke8 4.d5 Kf8 5.Kd7 Kfi 6.Kc6! The saving
point. 6 . . . e6 doesn't win because of 7.Kxd6 exdS 8.KxdS Kg6 9.Ke4. 6 ... e5 6 . . . e6
7.Kxd6 exdS 8 .KxdS Kg6 9.Ke4. 7.Kxd6 e4 8.Kc6! 8.Kc7? allows an eventual
winning ... QaS+: 8 ... e3 9.d6 e2 lO.d7 e l =Q I l .d8=Q QaS+ 1 2 .Kd7 Qxd8+ 1 3 .Kxd8
Kg6. 8 ... e3 9.d6, IIz -V2.

(2 1 S) Draw.
(Chess Life 1 996.)
Black to move can only draw (White to move
would draw easily by l .g4). L .h5 Other moves
also fail to win: 1 . . .Kc2 2 . g4! Kd l 3 .Kd3 ! (3.Ke4
Ke2 4.KdS KB S .h3 Kg3 wins for Black) 3 . . . Ke l
4.Ke3 Kfl S .KD Kgl 6.Kg3 h6 7 .Kh3 ! (7 .h3?
Kh l ! wins), and Black can't make progress be
cause 7 . . . Kf2 is a stalemate; 1 . . .Kc3 2 . g4 Kc4 3 .Ke4 KcS 4.Ke3 KdS S .Kd3 , and
White is safe since he owns the opposition; 1 . . .g4 2 .Kf4 (White has to force
. . . h7-hS otherwise he is lost: 2 .Kd2? Kb3 3 .Kd3 Kb4 4.Kd4 h6! S .KdS Kc3 6.Kd6
Kd4 7.Ke7 KeS 8.Kxf7 KxfS 9.Kg7 KgS l O.W fS , and Black wins.) 2 . . . hS 3 .Ke4
Kc3 4.KdS Kd3 S .Kd6 Ke4 6.Ke7 KxfS 7.Kxf7 KgS 8.Ke6 Kg6 9.KdS ! =. 2.Kd2 !
Kh3 3 .Kd3 Kh4 4.Kd4 g4 4 . . . h4 S . gxh4 =. 5 .Kd5 Kc3 6.Kd6 Kd4 7.Ke7 Ke5
8.Kf8 ! ! 8.Kxf7? KxfS is Zugzwang: 9.Ke7 KeS l O.W fS I l .Kg6 h4! , and wins.
8 ... Ke4 9.Kg7 Kxf5 1 0.Kxf7,

liz_ liz .

(2 1 6) White to move and win.

(Chess Life 1 997.)
l .g6 ! Kh2 2 .d7 a2 3 .d8=B! 3 . d8=Q? a l =Q
4.Qd4+ Kb3 ! . 3 ... Kh 1 4.Bf6 h3 5.g7 h2 6.Bxh2
Kxh2 7.g8=Q a l =Q 8.Qh8+ Kc 1 9.Qc7+ Kh2
1 0.Qh6+ Ka3 1 1 .Qa5+ Kh2 1 2 . Qh4+ Ka2
1 3 .Kc2 , 1 -0.

633

PAL B E N KO: MY L I F E, GAM ES AND CO M POSITIO N S

(2 1 7) White to move and win.

(E. G. 1 997.)

634

l .Rg6! as 2.Kc7 a4 3 .Kd6 a3 4.KeS a2 S.Ra6

Kg4 6.Ra4+ KgS 7.Ra7 ! Kg4 7 . . . h3 S.Rg7+
KhS 9.KfS ! . S.Ke4 Kg3 9.KfS ! The only way.

The tempting 9 . Ra h doesn 't win: 9 . . . Kg4

1 0.Rxa2 (no better is 1 0.Ke3 Kg3 1 1 .Rxa2 h3
1 2 .RaS Kg2 !) 10 ... h3 1 1 .Ke3 Kg3 1 2 .RaS Kg2 !
(and not 1 2 ... h2?? 1 3 .RgS+ Kh3 14.Kf2) 1 3 .RgS+ Kfl 14.RhS Kg2 . 9 ... h3 1 0.Ra3 +
Kh4 1 0 . . . Kg2 1 1 .Rxa2 + Kg1 1 2 .Kg4 h2 1 3 .Kg3 . I 1 .Rxa2 Kg3 1 2 .Ra3 + Kg2
1 3 .Kg4 h2 1 4.Ra2 + Kgl l S .Kg3 h l =N+ 1 6 .Kf3 , 1 -0.

(2 1 S) White to move and draw.

(First prize, Sakkelet 1 997.)
l .RdS+ l .Rg4? Rg2 2 .Rc4+ Kb7 3 .Rc l Rh2

4.Rg1 Bg2 S .Ke6 g6 6.KeS RhS + 7 .Kf4 RfS+

S.Ke3 ReS + 9.Kf4 Re2 , and Black wins. 1 . . .Kc7
2 .RgS Rg2 3 .Kf7 gS 4.Ke6 g4 S.Rg7+ Kc6
6.Rg6 g3 6 . . . KcS 7 .KfS g3 S.RgS =. 7.KfS+ Kd7
S.RgS Rf2 + 9.Kg4 g2 1 0.Kh3 RfS I 1 .Rg7+
Ke6 1 2 .Kh2 Kf6 1 3 .Rg3 RgS 1 4.Rf3 + KeS I S .Kgl Avoiding l S .Re 3 +? Kf4
1 6.Re 1 g l =Q+! 1 7 .Rxg1 RhS mate. I S . . . RfS 1 6 .Rg3 , l/z _ lh, We've reached a

positional draw.
(2 1 9) White to move and win.
(Blitz 1 997.)
l .Kc2 ! h3 2 .Bh l ! KfS 3 .Kd3 ! KeS 3 . . . Kf4
4.Kd4 cS+ S .Kd S . 4.Ke3 KdS S . Kf3 , 1 -0.

(2 2 0) White to move and draw.

(E. G. 1 997.)
l .Be6+ Kh4 l ...g4 2 .Bxg4+ =, or l ...Kg2 2 .Ne3 +
Kf2 3 .Nd 1 + ! Ke 1 4.Bg4 =. 2 .Nxd2 g4 3 .Nf3 + !
gxf3 4.Kd2 Kg3 S . Ke l f2 + 6 .Kxe2 Kg2
7.Bh3 + ! , l/z _ lh,

E N DGAM ES

(2 2 1 ) White to move and win.

(Chess Life 1 998.)
l .Na2 Ke2 2 .Nxa3 + 2 .Nbc3? Be5 3 .Nd5 Kb l !
4.Ndb4 Bd4! 5 .Bd6 Be5 =. 2 o o.Kh2 3 .Kg2 ! Bh4
4.Kh3 Bg5 5 .Kg4 Bh6 6.Kh5 Bg7 7.Kg6 BhS
S.Kh7 Bf6 9.Nb4 Kxa3 I O.Nd5+, 1 -0.

(2 2 2 ) White to move and win.

(E. G. 1 997.)
l .KdS h2 2. eS=Q h l =Q 3 .Qa4+ A series of
checks now forces the trade of Queens, taking
White into an easily won King and pawn
endgame. L.Kh6 4.Qa5 + Ke6 5.Qe5+ Kh7
6.Qh5+ Ka7 7.KeS Qe2 + S.Qe5 +, 1 -0.

(2 2 3 ) White to move and win.

(E. G. 1 998.)
l .Rg6+ KhI 2 .Ng3 + Kg2 3 .Nf1 + ! Kxfl L.Kh3

(L.Kh l 4.Bd l ; L.Kf3 4.Nxh2 + and 5 .Rb6)

4.Bd7+ Kh4 5 . Rg4+ Kh3 6.Kh6 4.Bh5+ Ke l
5 .Re6+ Kd I 6.Ba4+ Kc 1 7.Re6+ Kh I S.Be2 +
Kc 1 9.Bf5 + Kd l I O.Bg4+ Ke l 1 1 .Re6+ Kf1
1 2 .Bh3 + KgI 1 3 .Rg6+ Kh l 1 4.Bg2 + KgI
1 5 .Bxd5+ KfI 1 6.Be4+ Ke I 1 7.Re6+ Kd I I S .Bh3 + Kc 1 1 9.Re6+ KhI 20.Be2+
Kc 1 2 1 .Bf5 + Kd l 22 .Bg4+ Ke l 2 3 .Re6+ Kfl 24.Bh3 + KgI 2 5.Rg6+ Khl
26.Bd7 ! f1 =Q 27.Be6+ Qg2 2 S.Rxg2 h l =Q+ 29.Rg6+ Qe4 3 0.Bxe4 mate.

(2 24) White to move and win.

(Chess Life 1 998.)
l .e4+ ! Kf6 2 .KeS Ke5 3 .Ke7 a4 Black can

choose between several different Queen endings,

but White wins his opponent's Queen in every
case: L . b4 4.d6 b 3 5 . d 7 b2 6 . d8=Q b l =Q
7 . Qd6+ Kxe4 7 . Qg6+; 3 . . . c4 4.d6 c3 5 .d7 c2
6.d8=Q c l =Q 7.Qd5+ Kf4 8.Qf5+ Ke3 9.Qxg5+;
L.g4 4.d6 g3 5 .d7 g2 (5 . . .gxh2 6.d8=Q h l =Q 7.Qd6+ Kxe4 8.Qc6+) 6.d8=Q gl=Q
7.Qd5+ Kf4 8.Qf5+ Ke3 9.Qxc5+. 4.d6 a3 5.d7 a2 6.dS=Q al =Q 7.QhS+, 1 -0.

635

PAL B E N KO: MY LI FE, GAM ES A N D CO M POSITIO N S

636

(2 2 5) White to move and draw.

(Chess Life 1 995.)
l .Be4 NcS 2 .Bg6 2 .Bc2? Kd5 3 .Kf2 Kc4 4.Ke2
Nb3 5 .Kd l a4 wins for Black. 2 ... a4 3 .Kf2 ! a3
4.Bb l Ne4+ S .Ke3 Nc3 6.Ba2 ! 6.Kd2? Nxb l +
7.Kc2 Nd2 ! wins for Black. 6 . . .Nxa2 7.Kd4!
7 . Kd2 Kc5 S.Kc2 Kb4. 7 . . . Kc6 S.Kc4 Nc 1
9.Kc3 KbS l O.Kc2 a2 1 1 .Kb2 , 1/z- 1f2 .

(2 2 6) White t o move and win.

(Chess Life 1 995.)
l .b6 Kd6 l ...d3 2 .b7 dxe2 3 .c7 ! wins. 2 .b7! Kc7
3 .Kg2 ! eS 3 . . . KbS 4.Kg3 ! Kc7 5 .Kf3; 3 . . . e6

4.Kf3 e5 5 .Ke4 KbS 6.Kd3 ! takes us into the

main solution. 4.Kg3 ! KbS S.Kf3 Kc7 6.Ke4
KbS 7.Kd3 Kc7 S . Kc4 KbS 9.Kb S ! Ka7
l O . Kc S d3 1 1 . Kd6 dxe2 1 2 . Kc 7 e l = Q
1 3 .bS=Q+ Ka6 14.Qb6 mate.

(2 2 7) White to move and win.

(First/Second Prize, Magyar Sakkelet
2 000.)
White's King is also in danger, so there is no
time for a quiet move. l .RgS+ The chase be
gins! 1 . ..Kf8 The other possibility is 1 .. .KhS
2 .Qh l + Rh7 3 .Qa l ! Rg7 4.Rh5+ KgS 5 .QaS+
Kf7 6.Rf5+, and wins. 2 . Qb4+ Checking on f2
is wrong since 2 . . . Rf7 3 .Qc5+ Re7+ 4.Kf6 Qh7 saves Black. 2 . . . Re7+ 3 .Kf6 Now
3 . . .Qh7 4.RgS+ ! KxgS 5 .QbS+ mates. This is the first line where White sacri
fices his Rook, but there are more Rook sacrifices to come! 3 . . . Qe4 3 . . .Qh7
4.RgS+ KxgS 5 .QbS+. 4.RhS ! The Queen exchange was only good for a draw,
but this mate threat forces Black's hand. 4 . . . Qf3 + S .RfS Qe4! ? The best spot
for the Queen. In case of 6.Kg6+, the Rook gets pinned and Black escapes by
. . . KeS. 6.QbS+ ReS Now the Queen check leads nowhere, but a quiet move
keeps the attack going. 7.Qc7 ! Re7 A key point is demonstrated after 7 . . . Qe7+
S.Kg6+ KgS 9.Re5 ! ! , when White wins. S.QcS+ ReS 9.Kg6+ KgS l O.Qc7! Re7
And not 1 0 . . . Qe7 I l .Re5 ! ! , when White ends Black's resistance. 1 1 .QdS+ ReS
1 2 .Qh4! Qe7 1 2 . . . Re7 1 3 .Qf6 wins. 1 3 .Rf8+ ! ! , 1 -0. The final shot. Black can
accept the Rook sacrifice in three different ways, but all lead to mate.

E N DGAM ES

(2 2 8) White to move and win.

( 1 998)
1 .Rf7+ Kg6 2 .h7 Ra 1 + 3 .Kh8 Rh 1 + 4.Kc7
Rc 1 + S .Kd7 5 .Kd8? Ra l . S . . . Rd l + 6.Ke7
Re 1 + 7.Kf8 Ra l 8.Rg7+ Kf6 9.Ra7 ! , 1 -0.

(2 2 9) White to move and win.

1 .0-O-0 ! Rf8 2 .Kh l The movements of the
White King make a pleasing impression.
2 . . . Rhg8 3 .Kal Rf6 4.Rh l , 1 -0.

(2 3 0) White to move and draw.

TWIN (four positions).
( 1 999.)
l . Ne 6 + R3 g7 2 . Kh S Kh7 3 . Bxg7 Bxg7
4.NgS + Kh8 S.Nti+, Y2 - Y2 .

This twin features three other positions that all

lead to a drawn result:
(b) White King on e4: l .NhS+ R3g7 2 .KfS Kh7
3 .Nf6+ Kh6 4.Be 3 + RgS+ S .Ke4 Rg7 6.h4
Re7+ 7.Kd4!
=.

(c) White King on d 5 : l .Ne6+ R3 g7 2 .NgS !

Ba3 3 .Ne6 Bf8 4.NgS , with a positional draw.
(d) White Bishop on f6: l .NhS + R3 g7 2 .Kh3 !
Kh7 3 .Bxg7 Bxg7 4.Nf6 + ! Bxf6, stalemate.

(2 3 1 ) White to move and win.

(First Prize, Magyar Sakkelet 2 002 .)
l .Rd2 ! Rf8 2.Kc2 ! ! 2 .Ra2 ? Rf3+ 3 .Kd4 O-O!
2 ... Rhg8 3 .Kh2 h6 4.h3 hS S .h4 Rh8 6.Ka2
Rhg8 7.Ra7 Rf6 8.Rh2 , 1 -0.
=.

637

PAL B E N KO : MY LIFE, GAM ES A N D CO M POSITI O N S

63 8

(2 3 2 ) White to move and win.

(First Prize, Chess Clinic First Internet
Tournament 2 000. )
l .QeS+ Ke t ! If 1 . . .Kc2 , White makes use of
an important theme to win the game: 2 .Rc8+
Rc6 3 .Qd6 ! . 2 .Re8+ Re6 3 .Rb8! 3 .Qd6?? Qb l +
4.Ke3 Qb3 + refutes White's pin. Notice that
4 . . . Rxc6?? runs into S .Qd2 mate . 3 . . . Re4+
4.Kd3 Rd4+ ! S.Kc3 ! S .Qxd4?? Qxb8 =. S . . . Qd6! Black can also try S . . . Re4? !
6.Qxe4 Qc7+ 7 .Qc4 Qxb8 (Any check would have been answered with a discovered check.) 8.Qf1 mate. 6.Qe3 + ! The obvious 6.Qxd4? fails to 6 . . . Qg3 + 7 .Qd3
QeS+ 8 .Qd4 Qg3 +, and Black escapes with a perpetual check. 6 . . . Rd2 7.Rb4!
Qe6+ 7 . . . QdS 8.Qg l + Rd l 9.Rb l + ! Kxb l l O.Qb6+ and mates. 8.Re4 QdS 9.Re8!
But not 9.RcS? QxcS+ 1 O.QxcS Rc2 + with a draw. The position after 9.Rc8 ! is
mutual Zugzwang. Black has to move, therefore, he loses. But if White were to
move, then Black would escape with a draw (i.e., 1 O.Rc4 Kd l !). 9 . . . Qd7 10 .ReS
Qd6 I 1 .Kb3 + Kb l 1 1 . . .Kd l leads to the same reply. 1 2 .Re 1 + ! Kxe t 1 3 .Qe 1 +
RdI 1 4.Qc3 + Kb l l S .Qb2 mate.

(2 3 3) White to move and win.

( 1 999)
l .Qa1 + ! Kb6 2.Qb 1 + KaS 2 . . . Ka6 3 .Rh6+ Rf6
4.Qf1 +!; 2 . . . Kc6 3 .Rh6+ Rf6 4.Qe4+ Kb6 S .QfS ! .
3 . Qe 1 + Kb6 3 . . . Kb S 4.Q e 5 + . 4.Rh6+ Rf6
S.Qfl ! , 1 -0.

(2 3 4) White to move and win.

( Vergio 1 999.)
l .a4! l .b3? Kb2 2 .Rh l Kxa2 3 .Rh3 (3 .RhS b4!
=) 3 ... Kb2 ! 4.Kg7 a4 =. 1 . . .Kxb2 1 . . .h4 2 .Rh l
Kxb2 3 .RhS Ka3 4.RxaS b3 S .Kg7 b2 6.Rb S .
2 .Ra3 ! ! 2 . axbS? Kxa l 3 .b6 a4 4.b7 a3 S .b8=Q
a2 is a basic draw. 2 . . . Kxa3 2 . . . b4 3 .Rh3 b3
4.RhS Ka3 S .RxaS b2 6.Rb S . 3 . axbS , 1 -0.

A lot of tournament players don't like helpmates, but I fell in love with the con
cept at a young age and immediately began to compose them. In this genre, both
parties cooperate (help) to find the only possible mate by White in a given num
ber of moves. In a helpmate problem, Black always moves first and is always the
one who gets mated. Since each move offers a turn by Black and White, we get
the opposite of the normal notation procedure: Black's move is given first, fol
lowed by White's.
There are two exceptions: The first, a "Duplex," is a double-solution version
where Black moves first and gets mated in one solution while, from the same
position, White moves first and gets mated in the other solution; In the second,
a "Set," Black gets mated in both solutions, but White moves first in one while
Black moves first in the other.
In the case of Twins, all the usual helpmate rules apply (Black moves first in
both solutions and Black also gets mated in both).
Since helpmates see both sides working for the same goal (Black getting mated),
the composer doesn't have to worry about opponents or refutations based on
superior defense. Instead, he can concentrate on the position, bending it with
his will so that it conforms to whatever his imagination desires. This allows many
artistic ideas and stunning mates that could never occur in over-the-board chess.
Due to these factors, the solver has to drag his mind away from the concerns of
real chess (which often reminds me of wrestling) and, instead, let both sides en
gage in a form of ballet (male and female, two sides, Black and White, blending
together rhythmically to accomplish a mutual, artistic desire).
This doesn't mean that helpmates are easy to solve . Surprising sacrifices are
possible (as in problems 2 45 , 2 60 and 2 65), and 244 embraces a form of paradox
that left Botvinnik, Keres and Geller baffled. They were all forced to give up
after an hour of energetic searching.
I created a particularly tricky helpmate especially for Fischer (2 5 1). After thinking
for a while (yes, another bet was involved!), he announced that he had the solu
tion (his "solution" involved castling). However, I won the wager because cas
tling on either side wasn't legal-he had fallen for my subtle trap!
Let's see if you can succeed where so many world-class players have failed!
(2 3 5) Helpmate in two.
TWIN.
Don't forget, Black moves first! l .Bd4 Kd7
2 .Ba7 Be5 mate.
(b) Relocate Kb8 to g8 : l .Be4 Ke7 2.Bh7 Bd5
mate.
639

PAL B E N KO: MY LIFE, GAM ES A N D CO M POSITI O N S

640

(2 3 6) Helpmate in two.
SET.
(Strate Gem 2 002 .)
l .Ke5 Nf2 2 .Bf7 + Nxf7 mate.

(b) White moves first: l .Nf7 Nb6 2.Rd3 Ne2

3 .Nf2 mate.

(2 3 7) Helpmate in two.
(Honorable Mention, Zaszlonk 1 942 .)
l .Bxd7 (Don't forget that this is a Black move !

Thus, Black's Bishop on bS is taking White's

Queen on d7.) 1 . . . exd7 2 .Ke6 d8=N mate.
This was one of the first helpmates I ever did.
It was printed in a junior magazine when I was
only fourteen years old.

(2 3 8) Helpmate in two.
SET.
(Honorable Mention, Zaszlonk 1 942 .)
This is a set: White moves first in one solution
with Black getting mated, and Black moves first
in the other, with Black also getting mated. First
let's look at White moving first: l .Nc3 exf2
2 .Rd l mate.
(b) Black moves first from the diagram: l .exd2
Ne3 2 .Ne2 Nd3 mate. If you look at number
2 8 1 , you'll see how I (after the passing of many
years) couldn't resist making the whole prob
lem tighter and more economical.

(2 3 9) Helpmate in two.
SET.
l .Ne6 Re4+ 2 .Kd5 e4 mate.

(b) White moves first: l .Re5 Nd3 2 .e3 mate.

H ELPMATES

(240) Helpmate in two.

SET.
(Zaszlonk 1 943 .)
l .Bxe4+ Rg2 + 2 .KB Rg3 mate.

(b) White to move: l .Qxc2 KB 2 . Qe2 mate.

(24 1 ) Helpmate in six. Two solutions.

(Honorable Mention, Magyar Sakkelet
1 970.)
l .Bc2 + Ke6 2 .Rf4 b4 3 .Kg7 bS 4.KfS b6
S.KeS b7 6.KdS bS=Q mate.

(b) l .Rc3 bxc3 2 .Rd4 cxd4 3 .BeS dxeS 4.Qf6+

exf6 S .Ng7+ fxg7 6.NhS gxhS=Q mate.

(242) Helpmate in seven.

TWIN.
(Third Prize, Magyar Sakkelet 1 969.)
l .Ne4+ Kd l 2 .Nc3 + dxc3 3 .B e l c4 4.h l =R
cS S.Rh2 c6 6.Rxa2 c7 7.Ra7 cS=Q mate.

(b) Remove the h4-Bishop from the board. The

solution is then drastically different: l .Kb7 a4
2 .Kc6 as 3 .KdS a6 4.Ke4 a7 S.KB as=R
6.Kg2 Ral 7.Kh l Kxf2 mate. These two solutions featured Excelsiors with Black

and White underpromotions. I found the fact that Black gets mated on as in one
and one hI in the other to be particularly pleasing.

(243) Helpmate in two.

SET and TWIN.
(First Prize, Tipografia 1 969.)
l .NxdS Ke2 2 .Ke4 Nd6 mate.

(b). The second half of the set (with White to

move) is l .Rd6 Ng4 2 .Nd4 mate.
(c) Black to move. Be3 relocated to f2 : l .RxdS+
Nd4+ 2 .KeS Bg3 mate.

64 1

PAL B E N KO: MY LIFE, GAM ES AN D COM POSITI O N S

642

(244) Helpmate in three.

(Berliner Morgenpost 1 970.)
l .Kb2 ! RdS 2 .Kc3 ReS + 3 .Kd4 Nb3 mate. If
it was White to move, there would be several
corner mates (even without White's Rook!) .
However, with Black to move the mate i n three
is only possible in the middle of the board. This
apparent lack of logic (everyone looked for a
corner mate) baffled many strong players.

(245) Helpmate in two.

TWIN.
(Schach Echo 1 970.)
l .Nxd6! Kd3 2 .KeS Bb2 mate.

(b) B e l now stands on fl : l .Qxd6+ ! NdS +

2 .Ke6 Bh3 mate. Creating a double Queen sacrifice like this is quite an achievement, yet the
judge in the Schach Echo helpmate contest didn't
appreciate it and only awarded the problem an honorable mention. Later, this
problem became one of the most popular examples of the bad judging that is so
common in publications.

(246) Helpmate in two.

SET.
(Honorable Mention, Magyar Sakkelet 1970.)
l .Nc3 Bb2 2 .QeS Re4 mate.

(b) White to move: l .Rfl Ke4 2.Rf4 mate. This

problem featured changing double-pin mates
with limited material.

(247) Helpmate in three. Two solutions.

(Gazeta Czestochowska 1 970.)
l .Bf8 Rb2 2 .BeS Rb7 3 .KdS 4.Rb6 mate.

(b) l .ReS Bhl 2 .BeS Rg2 3 .KdS Rg6 mate.

H E LPMATES

(248) Helpmate in four.

TWIN.
(Second Prize, Magyar Sakkelet 1 970.)
l .d6 Nfl 2 .Kd7 Kb3 3 .Ke6 Nd4+ 4.KeS Ba7

mate.
(b) Relocate the Bb8 to b7: l .d4 Be8 2 .KdS
Nd3 3 .Ke4 Kb2 4.dS Ba6 mate. This Twin
creates an ideal Chameleon Echo Mate- the
second position reproduces, or echoes, the first, while an echo where the black
King is on a square of the opposite color is called a Chameleon Echo.

(249) Helpmate in two.

SET: Four Black to move solutions.
(Schach Echo 1 970.)
l .d6 This is a set, with four solutions with Black
to move. 1 . . . Re4+ 2 .KdS Nc3 mate.

(b) l .Qf4 RdS + 2.Ke4 Nc3 mate.

(c) l .Bf7 RdS + 2.Ke6 Ne7 mate.
(d) l .Qg6 Re4+ 2 .KfS Nd6 mate. If it's White
to move, the solution is: l .Na7 d6 2 .Ne6 mate.

(2 5 0) Helpmate in two.
SET.
(Chess Life 1 97 1 .)
l .KxaS Ka3 2 . axb6 b4 mate. The solution for

the other half of the set (White to move) is:

l .Kb l Kb3 2 .RxbS mate.

643

PAL B E N KO: MY LIFE, GAM ES AN D COM POSITI O N S

(2 5 1 ) Helpmate in three. Two solutions.

(British Chess Magazine 1 97 1 .)
644
1 .Bxe2 Nxh3 2 .RD Rgl 3 .Qe4 Rg3 mate.

(b) 1 .Qxe2 + Nxe2 2 .Kd3 Nxa3 3 .Be4 Rdl

mate. I created this problem with the express
purpose of fooling B obby Fischer, just like
Loyd did to Steinitz. Bobby fell right into my
trap solutions: l .Qxe2+ Nxe2 2 .Bg4 Nc3 + 3 .Kf3
0-0 mate, and l .Bxe2 Nxe2 2 . Qc4 Nbc3 + 3 .Kd3
0-0-0 mate. Both solutions are illegal because
White's last move had to be with his King, thus
voiding any thought of castling!

(2 5 2 ) Helpmate in three.
TWIN.
(Honorary Mention, Magyar Sakkelet
1 97 3 .)
1 .0-0-0 a8=N 2.hl =B a7 3 .Bb7 Nb6 mate.

(b) Relocate Black's Ra8 to h8. The solution is:

1 .0-0 a8=Q 2 .h l =R Qe8 3 .RhS Qg6 mate. I
like the fact that the four promotions are com
bined with opposite-wing castling.

(2 5 3 ) Helpmate in two.
DUPLEX: 2 -2 solutions.
(Honorary Mention, Magyar Sakkelet
1 974.)
1 .KfS gS=N 2 .Kg6 Qh7 mate.

(b) Black moves first: 1 .ReS g8=Q 2.KfS Qf3

mate.
(c) White moves first and gets himself mated:
1 .g8=B RhS + 2 . Qh7 BeS mate.
(d) White moves first and gets himself mated:
1 .gS=R BeS+ 2.Qg7 RhS mate.

H ELPMATES

(2 54) Helpmate in two.

DUPLEX: 2 -2 solutions.
(Die Schwalbe 1 974.)
1 .Bg4 Qhl 2 .Bd7 Qa8 mate.

(b) Black moves first: 1 .BB QhS 2 .Bh7 Qe8

mate.
(c) White moves first and mates himself: l . Qg3
RcS+ 2 .Kh4 RhS mate.
(d) White moves first and mates himself: l .Kh6
Bd3 2.QgS Rh7 mate.

(2 5 5) Helpmate in three.
SET: Black moves, two solutions.
(Chess Life 1 974.)
1 .NcS Rf4 2 .Ke3 Nc7 3 .Nd3 NdS mate.

(b) 1 .Nc3 + Kc 1 2 .Nh l Kb2 3 .Nd2 NcS mate.

(c) White moves first and mates Black: l .Kc 1
Nc3 2 .NcS mate.

(2 56) Helpmate in two.

TWIN: Three parts
(Chess Life 1 974.)
1 .h2 Kc3 2 .h l =R Ra4 mate.

(b) Place a white Bishop on b4 instead of the

Rook: 1 .h2 Kh3 2 .h l =B Bc3 mate.
(c) Place a white Knight on b4 instead of the
Rook: 1 .h2 Kh3 2 .h l =N Nc2 mate.

(2 5 7) Helpmate in two.
DUPLEX.
(Chess Life 1 974).
1 .Rf7 Qhl 2 .Be7 Qh8 mate.

(b) White moves first and mates himself: l .Kh8

BaS 2 .Kc8 Rd8 mate.

645

PAL B E N KO: MY LI FE, GAM ES AN D COM POSITI O N S

646

(2 5 8) Helpmate in four.
TWIN.
(Honorary Mention, Magyar Sakkelet,
1 97 5 .)
l .Na8 Kf4 2 .Ne7 KfS 3 .Ke8 Kf6+ 4.Kd8 Kfi

mate.
(b) Relocate the Bg5 to e 5 : l .Ne4 Kf4 2 .NaS
KfS 3 .Ke8 Ke6 4.Nh7 Ke7 mate. A fairly rare
sight: Chameleon Echo mates with only five
pieces!
(2 59) Helpmate in two.
TWIN.
(Chess Life 1 97 5 .)
l .f1 =R h8=N 2.Rf3 Ng6 mate.

(b) Relocate Black's g3 -pawn to h2 : l .h l =B

h8=Q 2 .BfJ Qh2 mate. This twin offers quite
a few promotions in a very limited amount of
moves.

(2 60) Helpmate in three.

TWIN: Five parts.
(Second Prize, Magyar Sakkelet 1 976.)
l .KdS d7 2 .Bd6 d8=N 3 .KeS Bd4 mate.

(b) Relocate Bc5 to g7 : l .KxfS d7 2 .KxeS d8=Q

3 .S Qd4 mate.
(c) Relocate Bc5 to c3 . Naturally, White's c 3 pawn must be removed from the board: l .fxeS d 7 2.Kd4 d8=B 3 .e4 Bb6 mate.
(d) Relocate Bc5 to h4: l .KxS d7 2.KgS d8=R 3 .S Rg8 mate.
(e) Relocate Bc5 to d6. Naturally, White's d6-pawn must be removed from the
board: l .Ke3 Bxd6 2 .eS Rf2 3 .e4 BeS mate. Lots of things happened in this
multi-twin problem. Underpromotions, piece sacrifices and, in the final solu
tion, we even removed the promoting d6-pawn!

H ELPMATES

(2 6 1 ) Helpmate in two.
SET.
(Die Schwalbe 1 97 7 .)
l .a l =R c6 2 .RaS Nc7 mate.

(b) White to move: l .Nc7+ KaS 2 .Be l mate.

(2 62) Helpmate in two.

SET.
(Honorary Mention, Magyar Sakkelet
1 977 .)
l .exd6 Bf6+ 2 .KxdS Nf4 mate.

B. White to move: l .Bf6+ eS 2.dxe6 e.p. mate.

(2 63) Helpmate in three.

TWIN. : Six parts.
(Honorary Mention, Magyar Sakkelet
1 97 8 .)
l .KcS Nhl 2 .Kd4 c3 + 3 .Ke4 Nfl mate.

(b) Relocate black King to e6: l .d4 Nhl 2.KdS

c4+ 3 .Ke4 Nfl mate.

(c) Black King on e6, white King on e2 : l .Kd6

d3 2 .KcS Ne4+ 3 .Kd4 c3 mate.
(d) Black King on e6, white King on b 5 : l .e4
NfS 2 .KeS Ne7 3 .Kd4 Nc6 mate.
(e) Black King on e6, white King on h5: l .d4
Kg4 2 .KdS NhS 3 .Ke4 Nf6 mate.

(f) Black King on b6, white King on h5 : l .KcS

Kg4 2 .Kd4 NfS+ 3 .Ke4 f3 mate. A nice mix
of ideal mates !

647

PAL B E N KO: MY LIFE, GAMES A N D COM POSITI O N S

648

(2 64) Helpmate in two.

SET.
(Chess Life 1 978.)
l .Ra7 0-0-0 2.Re7 Qg8 mate.

(b) White to move: l .Qa7 0-0-0 2 .Rc 1 mate.

(2 65) Helpmate in two.

SET.
(Chess Life 1 97 8 .)
l .Kxg4! RhS 2 .KxhS Nf6 mate.

(b) White to move: l .Kh8 Kg6 2 .Ne7 mate.

(2 66) Helpmate in two. Four solutions.

(Chess Life 1 97 8 .)
l .QaS Qa4+ ! 2 .Kxa4 Nc3 mate.

(b) l .Qa8 Qa7 2 .Qc6 Na3 mate.

(c) l .Kc4 Kc 1 2.Kb3 Nd2 mate.
(d) l .KaS Na3 2 .QbS Nc4 mate.

(2 67) Helpmate in two.

SET.
(Chess Life 1 97 8 .)
l .Kg4 QxhS + ! 2 .KxhS Nf6 mate.

(b) White to move: 1 .Kf8 Kg6 2 .Qb l mate.

H E LPMATES

(2 68) Helpmate in six.

TWIN: Four parts.
(Honorary Mention, Magyar Sakkelet
1 979.)
l .e4 f4 2 .e3 fS 3 .e2 f6 4.e l =R f7 S .Re2 f8=Q
6.Rd2 Qf1 mate.

(b) Relocate white King to g2 : l .e4 f4 2 .e3 fS

3 .e2 f6 4.e 1 =B f7 S . Bd2 f8 = Q 6.Ke l Qf1

mate.
(c) White King on g2 , relocate black King to
h4: l .e4 f4 2.e3 fS 3 .e2 f6 4.e 1 =R f7 S .ReS
f8=Q 6.RhS Qf4 mate.
(d) White King on g2 , black King on h4, relo
cate white pawn to c2 : l .e4 c4 2 .e3 cS 3 .e2 c6
4.e l =B c7 S.Bd2 c8=Q 6.BgS Qh3 mate.

(2 69) Helpmate in two.

TWIN.
(Chess Life 1 979.)
l .b l =R d8=N 2.RbS Nc6 mate.

(b) Relocate the b2 -pawn to f2 : l .fl =B d8=Q

2 .Bb5 Qd2 mate.

(2 70) HELPMATE: in two.

TWIN: Four parts.
(Magyar Sakkelet 1 980.)
l .Kb2 h8=R 2 .Ral Rb8 mate.

(b) Relocate black King to h I : l .Kxh2 h8=B

2 .Rhl Be5 mate.
(c) Relocate black King to h8: l .Kg7 h8=N
2.Kf6 Bd4 mate.
(d) Relocate black King to a8: l .Rb8 h8=Q
2 .Rc8 Qxc8 mate.

64 9

PAL B E N KO: MY LIFE, GAM ES A N D COM POSITI O N S

650

(2 7 1 ) Helpmate in two.
TWIN: Four parts.
(Chess Life 1 98 1 .)
l .f1 =N dS=B 2 .Ne3 Bf6 mate.

(b) Relocate white Rook to c5 : l .f1 =N dS=N

2 .Ne3 Nde6 mate.
(c) Relocate white Rook to e 5 : l .f1 =B dS=N
2 .Bc4 Ndc6 mate.
(d) Relocate white Rook to e 3 : l .f1 =B dS=B
2 .Bc4 Bb6 mate.

(2 72) Helpmate in two.

TWIN.
(Chess Life 1 980.)
l .f1 =N dS=N 2 .Ne3 Nde6 mate.

(b) Relocate white Knight to g5 : l .f1 =B dS=B

2 .Bd3 Bb6 mate. The change in promotions
led to two ideal, miniature mates.

(2 7 3 ) Helpmate in two.
(Chess Life 1 98 1 .)
l .Kd7 Qfi + 2.Kc6 Rf6 mate.

(2 74) Helpmate in ten. Two solutions.

l .KdS Kfl 2 .Ke7 Ke l 3 .Ke6 Kd l 4.KfS Kc l
S .Kxf4 Kb2 6.KeS Kc3 7.KdS Kxd3 S.Kc6
Kc4 9.fS d4 l O.f4 dS mate.

(b) l .dS fS 2 .d6 f4 3 .Kd7 f3 4.Kc6 Kf2 S .KbS

Ke3 6.KxaS Kd4 7.Kxb6 KxdS S.KaS Kxd6
9.bS KcS l O.b4 axb4 mate.

H E LPMATES

(2 7 5 Helpmate in five.
TWIN: Six parts.
(Benko and A. Kalotay, 1 984.)
l .Kd2 Kf4 2.eS+ Kg3 3 .Ke3 Kg2 4.Rd4 Kfl
S.e4 BgS mate.

(b) Relocate Bishop to b7: l .Rfl KdS 2 .Kf4

Kd4 3 .KfS e4+ 4.Ke6 eS S .Rf7 Be8 mate.
(c) Relocate Bishop to e8: l .Kd2 e3 2.Kd3 Kf4
3 .eS+ Kg3 4.Ke4 Kf2 S.RdS Bg6 mate.
(d) Bishop still on e8, relocate Rook to 1 : 1 .Kf2
e3 2.KB Kd4 3 .eS+ Kc3 4.Ke4 Kd2 S .RfS
Be6 mate.

(e) Relocate Bishop to h7: l .Rd l KfS 2 .Kd4

Kf4 3 .KdS e4+ 4.Ke6 eS S.Rd7 Bg8 mate.
(f) Relocate Bishop to d6: 1 .Kf2 Kd4 2 .eS+ Kc3
3 .Ke3 Ke2 4.Rf4 Kd l S.e4 BeS mate.
A five-move helpmate with six ideal mates might
well be a record! This problem was made with
my friend, Andrew Kalotay.

(2 76) Helpmate in three. Two solutions.

(Magyar Sakkelet. 1 98 5 .)
l .Ke7 ! bxa3 2 .Rf2 Rxb6 3 .Rf8 Re6 mate.

(b) l .Rxb2 Kd l ! 2 .Rf2 Rxb6 3 .Rf8 Re6 mate.

(2 77) Helpmate in two. Two solutions.

(Special Prize, Magyar Sakkelet 1 998.)
l .Kd4 ReS 2 .Bd3 e3 mate.

(b) l .Be6+ Re4+ 2 .KdS e4 mate.

65 1

PAL B E N KO: MY LIFE, GAM ES A N D COM POSITI O N S

652

(2 78) Helpmate in four.

TWIN: Three parts.
(Benko and A. Kalotay, 1 989.)
White moves first in all three solutions. l .Re2 +
K.d3 2 .Re3 + Kd2 3 .Bf4 Kc 1 4.Re l mate.
(b) Relocate Bishop to a7: l .Rh2 Kd3 2.RhI
Kd2 3 .Bgl Kc 1 4.Be3 mate.
(c) Relocate white King to c2 . Bishop still on
original b8-square : l .Rd l KB 2 .Kd2 Kg2
3 .Ke l Khl 4.Kf2 mate. This problem and its
solutions were created with the help of Andrew
Kalotay.

(2 79) Helpmate in two.

(SE 1 998.)
l .Ne4 e8=B 2 .Ng3 Bh5 mate.

(2 80) Helpmate in two.

SET.
(Strate Gems 1 999.)
l .cxh l =B Nb3 2 .Bc2 Nc5 mate.

(b) White to move: l .Ba2 c 1 =N 2 .Bh l mate.

(2 8 1 ) Helpmate in two.
SET: Two Black moves solutions.
l .exf2 Nxc3 2 .fl =N Re2 mate.

(b) l .e2 Ne3 2.cxd2 Nd3 mate.

(c) White to move: l .Nxc3 exf2 2 .Rd l mate.

This section is filled with a potpourri of problems. Some of them, like the logic
problems, more or less speak for themselves. However, selfinates, helpstalemates
and series motifs need to be explained for the uninitiated.
Selfinate: White moves first and forces Black to checkmate him.
Helpstalemate: Black moves first and helps White stalemate him in a stated
number of moves.
Series-helpstalemate: Series themes are fairy chess problems where one side re
mains stationary all or nearly all of the time (in the game of fairy chess, the winner
is the first side to force the opponent to capture all of his men. If a capture is
possible, it must be made. In this form of chess, even the Kings can be captured).
In the case of a Series-helpstalemate, we get a mix of fairy chess rules (legal cap
tures must be made) and normal rules (mates and stalemates are in effect), though
everything is dependent on how long White must remain stationary.
Thus, a S eries-stalemate in six has Black moving first (as in the normal
helpstalemate) and taking six moves in a row. White can only make his first move
after Black's sixth move, and this forced White move should lead to a stalemate.
If all this seems confusing, simply play through a problem without trying to
solve it. In this way, the dynamics of the problem type will quickly become clear.

(2 82) S erial Helpstalemate . Draw in eight.

TWIN: Draw in nine moves.
(SE 1 993 .)
1 .b5, 2 .b4, 3 .b3 , 4.bxc2 , 5 .cxb l =N, 6.Nc3 ,
7.Nxe2, 8.Kd3 Qa4 stalemate.

(b) Relocate black pawn to g5 . Helpstalemate

in nine moves: 1 .g4, 2 .g3 , 3 .gxh2 , 4.hxgl =N,
5 .Nh3 , 6.Nxfl , 7.Ne4, 8.Nxd2 , 9.Ke3 Rh4

stalemate.

(2 8 3 ) Serial Helpstalemate. Draw in six.

TWIN: Draw in Seven Moves.
SE 1 99 3 .
1 .a3 , 2.axb2, 3 .bxcl =N, 4.Nxe2 , 5.Nf4, 6.Nh3
Nxh3 stalemate.

(b) Relocate B l a ck's Kin g to b 8 . S eri e s

helpstalemate i n seven: 1 .a3 , 2.axb2, 3 .bxcl =R,
4.Rxc2, 5.Rxa2 , 6.Rxd2 , 7.Rd7 Qxd7 stalemate.
65 3

PAL B E N KO : MY LI FE, GAM ES AN D CO M POSITI O N S

654

(2 84) Mate in two.

(Games 1 996.)
l .axb6 Obviously, Black's last move had to be
. . . b7-b 5 . 1 . . . cxb6+ 2 .Nxb6 mate.

(2 85) Selfmate in four.

TWIN: Five parts.
(Sakkelet 1 996.)
l .c8=R e4 2 .Rxc5 e3 3 .Rc1 e2 4.Re l g2 mate.

Here are the other four parts of this five-part

twin:
(b) Relocate c5-pawn to e6. l .c8=B e4 2.Bxe6
e3 3 .Bc4 e2 4.Bxe2 g2 mate.
(c) Relocate c7-pawn to d7. Black's pawn is still
on e6. l .d8=N e4 2 .Nxe6 e3 3 .Nd4 e2 4.Nxe2
g2 mate.
(d) Relocate the d7 -pawn to h 7 . l .b8=Q e4
2 .Qe5 e3 3 . Qxe3 e5 4.Qxe5 g2 mate.
(e) Relocate Black's e6-pawn to c6. White's pawn
still resides on b7. l .b8=B e4 2 .Ba7 e3 3 .Bxe3
c5 4.Bxc5 g2 mate. The late International
Master Navarovsky entered a thematic tourna
ment for pieceless selfmates and created a prob
lem with two promotions. I enlarged it to five
promotions.

(2 86) White (who moves first) Helpmates in six

via en passant discovered mate.
(Chess Life 1 982 .)
To solve this, White's last move (his sixth move!)
has to be an en passant capture leading to a dis
covered checkmate. l .e4 e5 2 .Qh5 Nc6 3 .g4
d6 4.g5 Kd7 5 .Bh3 + f5 6.gxf6 mate. The only
player that I know who solved this was the
Canadian grandmaster Duncan Suttles.

PUZZLES

(2 87) Black is mated. Is it legal?

(Games 1 997 .)
655

Black is mated, but is this position legal? Yes, it

is legal. In the starting position White's King
stood on f3, he had a pawn on g2 and his d6pawn stood on c 5 . Place black pawns on d7 and
f4. l .g4 fxg3 + 2 .Kxg3 + dS 3 .cxd6 mate.

(2 88) Helpstalemate in two.

TWIN plus HELPMATE TWIN.
(SE 1 993 .)
l .KdS e3 2.Ke4 QhS stalemate.

(b) Place White's c2 -pawn on c4. This gives us

a twin helpstalemate solution: l .Kb7 Qb3 +
2 .Ka8 Qb6 stalemate.
(c) Having the pawn on c4 also allows us to cre
ate a twin helpmate (in two) problem: l .Kb6
Qb3 + 2 .KaS QbS mate.
(d) A twin to this (helpmate in two) comes about
if we relocate Black's King to d6 (the pawn still
stands on c4): l .KeS d4+ 2 .Ke4 Qd3 mate.

(2 89) Helpstalemate in five.

Don't forget that Black moves first in helpmate
and helpstalemate problems. l .No f8 = Q
2 . Ne S Kxe S 3 . Na3 Q x a 3 4 . Rg l Qxb2
S .h l =B Qxa l stalemate !

(2 90) Serial helpstalemate. Draw in ten.

TWIN.
(SE I 993 .)
l .hS , 2 .h4, 3 .h3 , 4.hxg2 , S . gxh l =B, 6.Bf3 ,
7.Bxe2, 8.Bd3 , 9.Bxc2 , l O.Bb3 Qxb3 stalemate.

(b) Relocate black King to b4: l .hS, 2 .h4, 3 .h3 ,

4.hxg2 , S . gxh l = B , 6 . B d S , 7 . Bxa2 , 8 . Bc4,
9.Bxe2 , l O.BhS QxhS stalemate.

PAL B E N KO: MY LI FE, GAMES AN D CO M POSIT I O N S

(2 9 1 ) Neither side gets mated in seven moves.

(Chess Life 1 976.)
656

Remember: all you have to do is not allow ei

ther side to get mated in seven moves. 1 .f"8=B
h l =N 2.Bb4 axb4 3 .h7 b3 4.hS=R b2 S.RgS
g2 + 6.Rxg2 Ng3 + 7.Rxg3 + Kxg3 .

(2 92) Helpmate in four. White to move. White

is not allowed to move his pawns!
Remember: White's pawns are frozen in place!
1 .NO Kc6 2 .NeS + KbS 3 .Nc6 Ka4 4.Nc3

mate.

(2 93) White mates in two after taking back his

last move.
(Games 1 996.)
White takes back his last move and mates in two:
White takes back f6-7 and plays 1 .0 Kd6 2 .Bf4
mate.

(2 94) Serial helpstalemate. Draw in 8 . Only

Black moves!
1 .dS, 2 .d4, 3 .d3, 4.dxe2, S.exfl =N, 6.Nxd2 ,
7.Nfl , S.Nxh2 stalemate.

(2 95) Serial helpstalemate. Draw in 1 6. Only

Black moves!
1 .Kc4, 2 .dS, 3 .d4, 4.d3 , S . dxc2 , 6.cxb l =R,
7 . Rxa l , S.Rxa2 , 9.Rxb2 , I O.Rc2 , 1 1 .Kb3 ,
1 2 . Ka2 , 1 3 . Kb l , 1 4 . Ra2 , I S . Ra l , 1 6 .a2

stalemate.

PUZZLES

(2 96) White to move. How many moves to

mate?
White to move. What's the shortest route to
mate? The trick to this problem lies in the word
ing. The answer is l .Rg8 mate. You see, Black
could not have made a legal move on his last
turn, so White could not have finished his last
move. If the Rook started out on f7, the Rf7g7-h7 is a half move. However, if the Rook
started out on g 1 , then Rg7-g8 is only a sev
enth of a move! If you gave this to someone and
he failed to solve it, I recommend you tell him
the solution by phone or by e-mail!

(2 97) White to move. Neither side gets mated

in four. Two solutions.
(Chess Life 1 976.)
l .a6 fl =B Making a Rook or Queen would mate

the white King. Don't forget that we're trying

to avoid mating either side's King. 2 .a7 Bfxc4
3 .a8=B Ba6 4.Bh7+ Bxh7.

(b) l .a6 fl =N 2 .a7 Nd2 3 .a8=N Nxc4 4.Nb6+

Nxh6.

(2 98 ) Selfmate in three.
(Benko and G. Bakcsi, Sachova Skladba
1 994.)
l .Rc4! Kxc4 Black's other moves also lead to a
mate in three: 1 ) 1 . ..cxb6 2 .Rc5 + bxc5 3 .Qa4+
Kxa4 mate. 2) 1 . . .c6 2 . Rxc6 bxc6 3 .Qb4+ Kxb4
mate. 3) 1 .. .c5 2 .Rb4+ cxb4 3 .Qxb4+ Kxb4 mate.
2 .Qc5+ Kd3 3 .Qc3 + Kxc3 mate. I made only
two selfmates with co-authors, and I improved
the contents in both of them. Problem grand
master Bakcsi published something similar, but
I told him how it was possible to use fewer pieces
and how to add another thematic variation. This
was the result.

657

PAL BEN KO: MY LIFE. GAM ES AN D CO M POSITI O N S

(2 99) Black takes back last move. Helpmate in

one. Three solutions.
658

Take back King (on d2) takes the Knight on e l

and play l .Kc3 Qb4 mate.
(b) Take back pawn on d3 takes Rook on e2 and
play l .Kd l Qfl mate.
(c) Take back pawn on d3 takes Bishop on e2
and play l .d2 Qfl mate.

(3 00) Various puzzles within a puzzle.

(Chess Life 1 982 .)
I'll be asking you to use all these pieces on the
bottom half of the board (nothing on the fifth
rank or further!) to create the following situa
tions: 1) Place the four pieces so that White can
mate in eight different ways on the same col
ored squares: Wh: K-f2 ; Q-d 3 ; B-f4. BI: K-h l .
2 ) Place the four pieces so that the black King
can stand on any of six squares and still be in
mate: Wh: K-b 3 ; Q-f1 ; B-f4. Black's King would
be mated if it stood on a I , b l , d2 , d l , h I or h2 .
3 ) Place the four pieces so that White can make
fourteen different moves that place Black in
stalemate: Wh: K-f3 ; Q-d2 ; B-gl . BI: K-h l . 4)
There is one configuration (using all four pieces,
of course) that allows White to make sixteen
different stalemates. What is that position and
why isn't it acceptable ? : Wh: K-f3 ; Q-d4; B-h2 .
BI: K-h l . The position is illegal. 5) Place the
four pieces so that White can make twelve moves
that stalemate Black. However, this time Black's
King can't be placed on the edge of the board:
Wh: K-f3 ; Q-c 1 ; B-d4. Bl: K-d3 . 6) Place the
White pieces so that Black's King can be on any
of five squares (White's pieces must still remain
behind the fifth rank, but Black's King can now
be placed anywhere on the board) and still be
in stalemate: Wh: K-h4; Q-d3 ; B-d4. The Black
King can be on h6, f4, e l , c1 or a2 .

Tou rnaments
YEAR

EVENT

PLACE

---- -------------------_..._-

SCORE

1 948

Budapest

ninth

8 th - 6 t h (+6, -4, = 5 )

1 948

Bad Gastein

= second

Wh-4th (+ 1 1 , - 1 , =7)

1 948

Hungarian Championship

first

1 1 - 3 (+9, -2 , =4)

1 95 0

Hungarian Championship

third

1 3 -6 (+ 1 0, - 3 , =6)

1 95 1

Hungarian Championship

= sixth

1 2 t h - 8 t h (+ 1 0, -6, = 5 )

1 952

Budapest

tenth

8 th - 8 t h (+6, -6, = 5 )

1 95 4

Hungarian Championship

second

1 3 -4 (+ 1 1 , -2 , =4)

1955

Hungarian Championship

= third

1 2 th-6th (+9, - 3 , =7)

1 95 7

Dublin Zonal

= second

1 3 -4 (+ 1 1 , -2 , =4)

1 95 8

Portoroz Interzonal

= third

1 2 t h - 7 th (+7 , -2 , = 1 1 )

1 959

U . S . Championship

fourth

7-4 (+ 5 , -2 , =4)

1 959

Zagreb Candidates

eighth

8 - 2 0 (+ 3 , - 1 5 , = 1 0)

1 960

U . S . Championship

= eighth

4th-6th (+2 , -4, = 5 )

1 960

Buenos Aires

= eleventh

9- 1 0 (+6, - 7 , =6)

1 96 1

U . S . Championship

= fourth

6 th-4th (+4, -2 , = 5 )

1 962

U. S . Championship

= ninth

4th-6th (+ 1 , - 3 , =7)

1 962

S tockholm Interzonal

= sixth

1 3 th-8 th (+8, - 3 , = 1 1 )

1 96 3

U . S . Championship

third

7 -4 (+ 5 , -2 , =4)

1 96 3

Piatigorsky Cup

= seventh

5 th - 8 t h (+4, - 7 , = 3 )

1 964

Belgrade

= seventh

9 th - 7 th (+6, -4, =7)

1 96 5

U . S . Championship

= seventh

5 -6 (+ 3 , -4, =4)

1 966/67

U . S . Championship

= third

6-5 (+4, - 3 , =4)

1 967

Winnipeg

fifth

5 -4 (+2 , - 1 , =6)

1 967

Saraj evo

= third

1 0- 5 (+ 7 , -2 , =6)

1 968

Monte Carlo

= sixth

7 t/2 - 5 1h (+ 3 , - 1 , =9)

1 968

U.S. Championship

fourth

6 th-4th (+5 , - 3 , = 3 )

1 968

Palma de Mallorca

seventh

9 1 h - 7 1 h (+ 7 , - 5 , = 5 )

1 969

Wijk aan Zee

sixth

9-6 (+ 5 , -2 , =8)

659

PAL B E N KO : MY LIFE, GAM ES AN D CO M POSITI O N S

660

1 969
1 969
1 969
1 969
1 969
1 969
1 970
1 970
1 970
1 970
1 970
1 97 1
1 97 1
1 97 1
1 97 1
1 972
1 972
1 972
1 972
1 97 3
1 973
1 973
1 973
1 974
1 974
1 974
1 974
1 97417 5
1 97 5
1 97 5
1 976
1 977
1 978

Venice

= first
= second

Monte Carlo

eighth

Malaga

Netanya

= second

Vrsac

second

U . S . Championship

third

Wijk aan Zee

Caracas

= sixth
= first
= fifth
= fourth

Reggio Emilia

second

Malaga
Vrnj acka Banj a

= third
= sixth
= fifth

Netanya

fourth

U. S . Championship

fifth

Wijk a a n Zee

Hastings

= seventh
= fourth
= fifth
= second
= seventh
= second
= fifth
= eighth

Orense

second

U . S . Championship

Malaga
Saraj evo

Olot

Malaga
L a s Palmas
Orense
Vrnj acka Banj a
S a o Paulo
U. S . Championship

Hastings

= second
= sixth
= ninth

Torremolinos

seventh

U . S . Championship

fourteenth

Majdanpek

= tenth
= third
= ninth

Venice

S a o Paulo
U . S . Championship

Eight time U . S . Open Champion:

1 1 -3 (+8, -0, =6)

91/z-5 1/z (+5 , - 1 , =9)
5-6 (+2 , - 3 , =6)
8-5 (+4, - 1 , =8)
1 01/z-41/z (+6,-0, =9)
7-4 (+4, - 1 , =6)
8-7 (+4 , - 3 , =8)
1 1 1/z-3 1/z (+9, - 1 , =5)
8 1/z-61/z (+3 , - 1 , = 1 1 )
1 1 1/z-5 1/z (+8, -2 , =7)
1 01/z-41/z (+8, -2 , =5)
1 0-4 (+6, -0, =8)
5-4 (+3 , -2, =4)
8-7 (+2 , - 1 , = 1 2)
9-6 (+4, - 1 , = 1 0)
8-5 (+3 , -0, = 1 0)
71/z-7 1/z (+ 1 , - 1 , = 1 3)
61/z-41/z (+3 , - 1 , =7)
9-6 (+5 , -2, =8)
7 1/z-3 1/z (+5 , - 1 , =5)
8-7 (+3 , -2 , = 1 0)
8-4 (+5 , - 1 , =6)
61/z-5 1/z (+2 , - 1 , =9)
7 1/z-7 1/z (+3 , - 3 , =9)
8-3 (+5 , -0, =6)
8-5 (+3 , -0, = 1 0)
7 1/z-5 1/z (+4, -2, =7)
7-8 (+3 , -4, =8)
7-6 (+3 , -2 , =8)
5-8 (+0, - 3 , = 1 0)
6-7 (+ 1 , -2 , = 1 0)
8-5 (+4, - 1 , =8)
61/z-7 1/z (+ 1 , -2 , = 1 1)

1 96 1 , 1 964, 1 965, 1 966, 1 967, 1 969, 1 974,

and

1975.

Matches
1 960 vs. Samuel Reshevsky 41/z-5 1/z (2 wins, 3 losses, 5 draws)
1 962 three way Interzonal Playoff match vs. Stein & Gligoric 2 1 /z- 1 1 /z (2
Stein, a win and draw vs. Gligoric)

1 964 vs. Arthur Bisguier 41/z - 1 1/z (4 wins, 1

loss,

draw)

draws vs.

Abramov, Lev, 3 0 5 , 502

Berry, Jonathan, 2 3 8 , 5 3 6

Adams, Mickey, 5 2 7 , 5 3 8

Bielicki, Carlos, 46 5

Addison, William, 2 0 5 , 4 3 0

Bilek, Istvan, 7 2 , 444, 5 5 9

Alburt, Lev, 2 3 4, 2 5 2 , 5 3 3

Bisguier, Arthur, 2 5 , 2 6, 7 5 , 1 2 7 , 1 7 3 ,

Alekhine, Alexander, 1 9, 2 5 , 2 7 , 1 94, 1 9 5 ,

1 7 6, 1 7 7 , 1 7 9, 1 8 1 , 1 8 3 , 1 99, 2 06, 2 0 7 ,
2 1 1 , 2 2 0, 2 2 1 , 3 1 3 , 42 3 , 45 3 , 464, 466,

2 3 3 , 42 6, 5 3 4, 6 2 7

508, 509, 5 1 4, 5 3 1 , 5 3 3 , 660

Anand, sh 1 1 5 , 42 8 , 48 8 , 5 3 2

Blodig, Reinhard, 5 3 5

Anderssen, Adolf, 5 3 4
Andersson, Ulf, 6 3 , 440, 5 1 5

Blumin, Boris, 545

Andresen, Steffen, 48 3

Bobotsov, Milko, 5 6 3

Anderson, Frank, 5 2 5

Bogdanovic, Rajko, 2 1 3 , 449

Aspler, G , 2 4 5 , 2 5 8 , 5 3 6

Bogoljubov, Efim, 502

Atalik, Suat, 40 1

Bolbochan, Julio, 75, 440

Ault, Robin, 487

Boleslavsky, Isaak, 5 1

Averbach, Yury, 8 8 , 92 , 9 3 , 445

Bondarevsky, Igor, 443

Avram, Herbert, 2 3 6

Bordonada, Glenn, 3 99

Avrukh, Boris, 5 3 9

Boric, Muhamed, 5 3 1
Borisenko, Georgy, 5 5 8

Bacrot, Etienne, 9 5

Bosboom, Manuel, 5 3 2

Badals, Magem, 2 3 9

Botvinnik, Mikhail, 9 , 2 6 , 5 8 , 62 , 6 3 , 1 02 ,

Bakcsi , Gyorgy, 6 5 7

1 0 3 , 1 06, 1 2 5 , 1 2 7 , 1 4 5 , 1 8 9, 2 1 6, 2 89,

Balcerowski, Witold, 1 2 1

3 69, 4 1 9, 42 9, 43 1 , 440, 44 1 , 442 , 5 04,

527, 5 5 1 , 639

Balogh, Dr. Janos, 9, 48, 443

Ban, J eno, 7 2 , 607

Brasket, Curt, 3 8 3 , 462

Barbero, Gerardo, 3 62

Bronstein, David, 3 7 , 5 1 , 1 1 7 , 2 3 3, 2 7 5 ,
2 7 7 , 43 1 , 5 1 3 , 5 1 5 , 5 5 1

Barcza, Gedeon, 1 3 , 1 4, 7 1 , 7 2 , 5 0 3

Browne, Walter, 2 3 5 , 2 3 8 , 243 , 2 4 5 , 2 4 7 ,

Barczay, Laszlo, 3 1 5

248, 3 43 , 3 44, 4 1 7 , 43 2 , 5 0 7 , 5 2 1 , 5 3 3

Bareev, Evgeny, 1 43 , 2 3 6, 2 42
Barlay, Imre, 2 5 5 , 5 3 6

Bukal, Vladimir, 3 0 5 , 5 0 1

Basman, Michael, 3 54, 3 5 6

Bukic, Enver, 2 2 9

Bata, Dr. J , 2 1

Burger, Karl, 2 7 9, 489

Becerra, Julio, 488

Burgess, Graham, 49 5

Bellon, Juan, 3 3 1 , 3 3 2

Byrne, D onald, 2 8 5 , 2 8 7 , 42 2 , 42 3 , 444

Beni, Alfred, 7 3

Byrne, Robert"

1 1 9, 1 2 0, 1 2 1 , 1 74, 1 96,

Benko, Francisco, 70

2 00, 2 0 1 , 2 0 5 , 2 2 4, 2 6 2 , 2 6 5 , 2 8 1 , 2 8 7 ,

Benko, Gizella, 1 7 3 , 1 74, 2 6 1 , 3 6 1 , 3 62 ,

40 1 , 42 3 , 4 7 7 , 48 3 , 5 0 7 , 5 2 4

409
Benko, Palma, 3 6 1 , 3 6 2 , 409

Camara, Helder, 247, 5 3 1 , 548

Benko, David, 3 60, 3 6 1 , 409

Capablanca, Jose, 4, 1 9 5 , 3 5 6, 3 5 7 , 3 69,

42 3 , 42 6, 42 8 , 42 9, 442 , 5 3 3 , 62 7

Berger, Bela, 7 2

Cantwell, Richard . Dr, 9 7 , 1 2 6, 1 2 8 , 1 2 9 ,

Bernhardsson, Sten, 5 3 5

1 45 , 1 5 4, 1 5 5 , 2 00, 2 0 7

Bernstein, Ossip, 3 9 7

66 1

PAL B E N KO : MY LI FE, GAM ES AND COM POSITI O N S

Cardoso, Rodolfo, 3 99, 440, 5 1 9, 5 2 8

662

Evans, Larry, xix, 8 5 , 8 6 , 1 0 5 , 1 06, 2 70,

Carls, Carl, 5 0 3

2 79, 2 8 7 , 3 1 0, 3 66, 3 67 , 3 7 2 , 4 1 2 , 4 1 3 ,

Chavez, Romelio, 3 8 7 , 47 1

4 1 4, 4 1 5 , 4 1 6, 42 1 , 42 2 , 42 3 , 424, 43 2

Chekov, Valery, 5 2 3
Chellstorp, Craig, 3 3 8

Farago, Ivan, 5 46

Christiansen, Larry, 6, 498

Fedorowicz, ]ohn, 3 94, 467 , 5 3 3

Chytilek, Roman, 447

Feldman, Tibor, 6 , 9 , 502

Ciocaitea, Victor, 1 07 , 1 2 3

Fernandez, Antonio, 3 8 7

Ciric, Dragoljub, 402 , 48 3

Ferrer, ] ose, 1 1 9

Cobo, Arteaga, 1 66

Filip, Dr. Miroslav, 1 5 6, 5 6 1

Coli as, Didier, 488

Filippov, Valentin, 2 0 1

Commons, Kim, 3 6 2 , 3 6 3

Fine, Reuben, 3 7 8 , 424, 43 1

Conquest, Stuart, 2 3 6, 548

Fischer, Robert, xiiiI, xiv, 1 0, 1 3 , 5 5 , 8 7 ,

Conrady, Alphonse, 5 2 8

8 8 , 89, 90, 92 , 9 5 , 1 0 3 , 1 0 5 , 1 06, 1 0 7 ,

Corral, Diez del, 243

1 08, 1 09, 1 1 9, 1 2 4, 1 2 5 , 1 2 6, 1 2 7 , 1 3 2 ,

Cosulich, Roberto, 3 7 1 , 4 5 7

1 3 8 , 1 3 9, 1 4 1 , 1 42 , 1 46, 1 5 0, 1 5 1 , 1 5 3 ,

Cozio, Carlo, 62 5

1 5 4, 1 5 5 , 1 6 3 , 1 7 1 , 2 74, 2 7 7 , 2 8 7 , 2 8 8 ,

Cramling, Pia, 2 74

3 1 0, 3 48 , 4 1 7 , 4 1 8 , 42 2 , 42 5 , 42 6, 42 7 ,

Crouch, Colin, 508

42 8, 42 9, 43 1 , 467, 469, 470, 48 1 , 49 5 ,

Csom, Istvan, 3 1 2 , 46 5 , 466, 467 , 5 3 5

496, 5 04, 5 1 0, 5 1 5 , 5 1 8 , 5 2 1 , 5 5 8 , 5 8 1 ,

Cvetkovic, Srdj an, 3 9 2 , 447

Czerniak, Moshe, 3 0 3

6 3 9 , 644
Fitzgerald, Kent, 2 4 3
Fiear, Glenn, 493

Damj anovic, Mato, 3 8 7 , 402

Flesch, ]anos, 2 9 7 , 484

D arga, Klaus, 1 5 8 , 1 94

Flohr, S alo, 3 5 , 1 94

De Fotis, Greg, 2 4 1

Florian, Tibor, 562

Dely, Peter, 7 2 , 494, 5 2 6

Foguelman, Alberto, 2 2 4, 47 3 , 476

D enker, Arnold, 9, 42 3 , 5 3 6

Foldi, ]ozsef, 82

Diesen, Mark, 447 , 5 3 7

Foltys, Jan, 28, 1 94, 5 3 0

Djurhuus, Rune, 7 2

Forgacs, Gyula, 5 1

Dolmotov, Sergey, 1 69, 3 46

Forintos, Gyozo, 440

Dominguez, ]uan, 3 1 8 , 3 2 0

Formanek, Edward, 2 6 5 , 4 5 3

Donaldson, ] ohn, 2 1 6

Fritz, 2 8 0

Donner, ]an Hein, 6 3 , 2 5 0, 2 94, 2 9 5 , 2 9 7 ,

Furman, Semen, 2 0 3

490, 549, 5 5 2

Fiister, Geza, 3 5 , 67, 6 8 , 8 8

D oroskevic, Vladimir, 243

Drimer, Dolfi, 5 0 3

Galego, Luis, 2 1 1

Diickstein, Andreas, 2 1 6, 4 5 9

Galic, Zeljko, 449

Dunkelblum, Arthur, 7 2

Gallia, Karl, 2 1

Durao, ]aoquim, 3 00, 5 0 1

Gamboa, Anibal, 5 44

D zindzichashvili, Roman, 1 66, 2 1 6

Gaprindashvili, Nona, 3 72

Echeverria, Raoul, 1 2 1

Garda, Guillermo, 406, 407, 462

Edmondson, Edmund, 9 7 , 3 1 0, 42 6

Gasic, Bozidar, 500

Ehlvest, ] an, 3 3 8

Gausel, Einar, 72

Garda, Raimundo, 1 97

Eliskases, Erich, 1 1 0, 495

Gelfand, Boris, 4 5 8

Emma, ]aime, 3 2 5 , 450

Geller, Efim, 48, 5 6 , 5 8 , 59, 7 5 , 1 2 6, 1 2 8 ,

Engelbrecht, 447

1 42 , 2 1 4, 2 2 4, 2 8 8 , 2 89, 2 90 , 3 1 6, 3 6 2 ,

Enklaar, Bertus, 494

3 9 7 , 4 1 9, 44 1 , 448 , 472 , 476, 479, 6 3 9

Ermenkov, Evgeny, 5 3 1

Georgadze, Tamaz, 7 8 , 243

Euwe, Max, 1 0, 2 5 , 2 7 , 1 94, 3 5 7 , 5 9 3

Georgiev, Kiril, 5 3 2

I N D EX OF NAM ES

Gereben, Erno, 49 5

Howell , James, 5 5 1

Gergs, Wolfgang, 3 4 5

Hromadka, Karel, 4 5 8

German, Eugenio, 1 2 8 , 4 7 3

Hubner, Dr. Robert, 6 , 3 8 7 , 467

Gretarsson, Helgi, 547

Hug, Werner, 2 74, 472 , 5 1 2

Gheorghiu, Florin, 2 6 2 , 3 3 5 , 440, 494,

Huguet, B ernard, 5 2 4

499, 5 1 4, 5 2 8
Ghitescu, Teodor, 243 , 402 , 482 , 547

Hulak, Krunoslav, 4 5 4
Husak, Karel, 480

Gilden, Larry, 3 6 8 , 44 1
Giustolisi, Alberto, 82

Ilic, Zoran, 492

Glek, Igor, 1 6 1 , 496

Illescas, Miguel, 5 2 8

Gligoric, Svetozar, 2 6 , 2 8 , 80, 1 0 3 , 1 2 5 ,

1 3 5 , 1 3 6, 248, 2 49, 2 5 0 , 2 5 1 , 402 , 47 5 ,

Innala, Heikki, 48 3
Ivkov, Borislav, 1 2 3 , 243 , 470

5 1 8 , 5 2 6 , 5 3 5 , 5 49, 5 5 3 , 660
Goldin, Alexander, 2 3 6, 2 6 2

Jacobs, Byron, 545 , 548, 549

Goodman, D avid, 3 8 3

Jakobsen, Ole, 2 4 3

Gordon, Stephen, 2 46, 2 5 8 , 5 3 7

Jandemirov, 40 1

Gorgiev, Tigran, 62 5 , 6 2 8

Janj gava, Lasha, 5 5 3

Gorkavij, Vadim, 4 8 3

Janosevic, Dragoljub, 3 5 2 , 506, 508

Gorog, Friedrich, 86

J anosi, Ervin, 62 3

Grabarczyk, B ogdan, 2 6 5

Jansa, Vlastimil, 1 66, 3 0 7 , 470, 494

Greco, Gioacchino, 5 1
Grefe, John, 402

Kabanov, Nikolai, 482

Grigoriev, M, 5 8 2

Kagan, Shimon, 2 2 3 , 2 2 4, 468, 476

Grob, Henry, 3 1

Kalme, Charles, 5 04

Gross, Ronald, xix, 1 64, 4 1 2 , 4 1 3 , 4 1 4,

Kalotay, Andrew, 6 5 1 , 6 5 2

4 1 5 , 4 1 6, 4 1 7 , 4 1 8 , 4 1 8 , 42 0 , 42 1 , 42 2

Kamsky, Gata, 3 6 3

Griinfeld, Ernst, 1 0, 1 1 , 2 1

Kan, Ilia, 3 5

Griinfeld, Yehuda, 404, 40 5

Kane, George, 248, 3 44, 3 4 5

Gufeld, Edward, 2 0 3 , 2 5 1 , 2 8 1

Karlsson, Lara, 3 2 6

Gulko, Boris, 2 5 3 , 448

Karpov, Anatoly, 2 74, 3 7 3 , 442 , 4 5 8 , 487,

Hajton, Jozsef, 9

Kashdan, Isaac, 1 64, 5 5 2

Halldorsson, B, 2 5 2 , 5 4 1

Kasparov, Garry, 5 1 , 2 0 3 , 2 42 , 3 3 8 , 3 6 3 ,

513

Hansen, Carsten, 448 , 4 5 0 , 4 5 8

Harrison, Kevin, 40 1 , 48 1

44 1 , 442 , 49 3 , 5 1 3 , 5 1 7 , 5 3 2 , 547, 5 5 8 ,
560

Hartkopf, Uwe, 48 3

Kasper, Maurice, 1 2 7

Hartmetz, Juergen, 5 2 5

Kaufman, Lawrence, 2 4 1 , 3 5 7 , 5 46

Hartston, William, 3 4 5 , 493

Kavalek, Lubomir, 3 09, 463

Hazai, Laszlo, xii

Keene, Raymond, 3 09 , 3 1 3

Hearst, Elliot, 7 5 , 2 8 7

Keller, Hans, 2 1

Heinen, 2 0

Keres, Paul, 1 9, 5 8 , 7 5 , 7 6 , 8 8 , 1 0 3 , 1 2 6,

Helling, Karl, 5 0 3

1 2 7 , 1 2 8, 1 2 9, 1 3 2 , 1 44, 1 4 5 , 1 6 3 , 1 64,

Henneberger, Walter, 2 1 , 2 2

1 8 3 , 2 2 9, 4 1 8 , 4 1 9, 464, 5 1 5 , 5 1 8 , 5 69,

Henrichs, Thomas, 1 6 1

639

Holm, Sejer, 480

Khalifman, Alexander, 46 1

Honfi, Karoly, 7 2 , 5 2 0

Kholmov, Ratimir, 3 48

Horne, D ennis, 6

Khrushchev, Nikita, 1 64

Horowicz, Israel, 2 6 7 , 2 69 , 42 2 , 443

Kindermann, Stefan, 5 3 2

Hort, Vlastimil, 2 7 , 1 1 7 , 2 4 5 , 2 46, 2 6 2 ,

King, D anny, 3 2 6

3 2 7 , 450, 45 1 , 5 3 7 , 5 3 9
Howard, Jack, 2 6 5

Kinsman, Andrew, 545, 549

Kishnev, Sergey, 3 5 7

663

PAL B E N KO : MY LI FE, GAMES AN D COM POSITI O N S

664

Klein, Robert, 2 6 3

Levy, D avid, 4 3 0

Kluger, Guyla, 7 2 , 7 5 , 1 3 6

Liepnieks, Aleksandrs, 9 0 , 3 1 0

Klundt, Klaus-Peter, 5 2 1

Lilienthal, Andor, 3 2 , 3 7

Koblentz, Alexander, 1 5 3

Lisitsin, Georgy, 5 04

Kochyev, Alexander, 3 8 8

Ljubojevic, Ljubomir, 3 68 , 3 94, 44 1 , 5 2 1

Kolman, 502

Llorens, 480

Koltanowski, George, 1 99, 3 1 0, 4 1 8

Loyd, S am, 569, 644

Kopec, D anny, 3 7 8 , 5 3 1

Loebler, Heirno, 49 1

Korchnoi, Victor, 1 1 , 1 07 , 1 1 7 , 1 2 6, 1 4 5 ,

Logunov, Maxim, 48 3

1 46, 1 4 7 , 1 48 , 1 94, 2 04, 2 7 7 , 2 94, 3 4 5 ,

3 5 7 , 3 7 3 , 4 1 5 , 42 3 , 42 8 , 429, 448 , 45 3 ,
479, 490, 5 2 7

Lombardy, William, 8 5 , 1 2 1 , 1 5 3 , 1 96,

2 0 3 , 2 69 , 2 74, 3 1 0 , 42 3 , 5 3 1
Lounek, 2 5 , 1 94, 490

Korhonen, Miss, 3 7 5

Lublinsky, 5 5 8

Korody, K , 5 3 , 5 5 2

Lukacs, Peter, 6

Kosic, Dragan, 488

Lundin, Erik, 34, 2 3 3

Kotov, Alexander, 3 3 , 3 7 , 1 4 5
Kotsur, Pavel, 483

Maciej a, B arciomiej , 2 8 8

Kottnauer, Cenek, 63, 443 , 5 3 3

Macskasy, Elod Dr, 2 3 7 , 2 3 8

Kouaciy, B achar, 493

Maderna, Carlos, 7 5

Kovacs, Laszlo, 2 5 5 , 5 3 6

Madonna, xvi

Kramnik, Vladimir, 2 94, 44 1 , 467, 5 1 7 ,

Malakhatko, Vadim, 5 3 9

555

Malich, Burkhard, 3 2 9

Krasenkov, Michal, 7 2

Mansfield, C , 569

Krnic, Zdenko, 3 8 1 , 46 1

Marin, M, xii

Krogius, Nikolai, 1 47 , 1 94, 3 6 1 , 3 62

Mariotti, S ergio, 204

Kruppa, Yuri, 5 3 9

Markowski, Tomasz, 2 8 8

Kuczynski, Robert, 465

Maroczy, Geza, 9, 2 9 , 3 1

Kudrin, Sergey, 2 1 1

Marshall, Frank, 76, 470

Kuijpers, Franciscus, 1 90 , 486, 546

Marshall, George, 3 1

Kunte, Abhijit, 2 2 1

Martz, William, 3 3 7 , 3 3 8 , 3 48

Kuppe, Wilhelm, 6, 5 0 3

Matanovic, Aleksandar, 1 90, 2 2 9 , 2 3 1 ,

Kushnir, AlIa, 3 7 2 , 3 7 3 , 3 74
Kuuksmaa, 3 0 5 , 502

440, 486, 607

Matulovic, Milan, 432
McNab, Colin, 472 , 47 3 , 5 1 1

Lalic, Bogdan, 487

Mecking, Enrique, 499

Lane, Gary, 2 2 0

Medina, Antonio, 2 09 , 2 84, 49 1

Langner, D avid, 447

Mednis, Edmar, 1 2 3 , 1 2 4, 2 8 7 , 3 1 0 , 479

Larsen, Bent, xix, 5 1 , 86, 1 1 7 , 1 7 2 , 2 0 3 ,

Mendoza, Rafael, 5 44

2 1 4, 2 2 6, 2 2 7 , 2 2 9, 2 42 , 2 5 0 , 3 2 6, 3 6 3 ,

Menvielle, Lacourrelle, 3 1 8

3 99 , 4 1 3 , 4 1 4, 4 1 5 , 4 1 6, 4 1 7 , 4 1 9, 42 1 ,

Mestrovic, Zvonimir, 2 1 8 , 5 0 7

42 7 , 42 8 , 45 3 , 5 0 8 , 5 2 7 , 546, 5 5 3 , 5 54

Meyer, Eugene, 5 3 7

Lasker, Emmanuel, 76, 2 6 7 , 42 8 , 42 9,

5 3 3 , 626, 62 7
Lassila, Erkki, 483

Miagmasuren, Lhamsuren, 1 3 2
Mieses, Jacques, 5 1
Miethke, Erhardt, 454

Laurine, Leho, 1 94

Mikenas, Vladas, 3 2 6, 45 1

Lautier, Joel, 2 94

Mikhalchishin, Adrian, 488

Leif, Lund, 1 0

Milic, B orislav, 486

Lein, Anatoly, 3 62 , 3 8 3

Millan, 5 2 1

Leko, Peter, 9 5 , 5 44

Milner-Barry, Sir Philip, 2 2 1

Lengyel, Levente, 2 84, 49 1

Milos, Gilberto, 5 3 9

Letzelter, Jean, 5 09

Milutinevich, Z , 6 1

I N D EX O F NAMES

Minasian, Art, 1 4 3 , 2 1 1
Moler, Castro, 2 3 9

Petrosian, Tigran, 1 3 , 1 4, 2 6 , 2 7 , 5 8 , 59,

6 1 , 8 8 , 9 5 , 1 0 3 , 1 2 6, 1 2 7 , 1 2 8, 1 2 9 ,

Molineus, Stefan, 447

1 4 5 , 1 6 3 , 1 64, 1 7 2 , 2 2 6, 2 7 0, 2 7 7 , 3 48 ,

Molnar, Istvan, 5 1

3 7 3 , 3 99, 4 1 8 , 440, 46 3 , 470, 490, 5 04,

Moore, Garry, 88

5 1 5 , 5 1 8 , 5 5 5 , 5 5 6, 5 6 3

Morphy, Paul, 1 7 5

Petrosian, Vartan, 6 1

Moser, Klaus, 402

Petursson, Margeir, 3 6 3

Moskowitz, Jack, 4 1 8

Pfleger, Helmut, 2 4 1 , 440, 543

Morrison, Chris, 3 2 6

Philidor, Francois, 5 1

Motylev, Alexander, 2 2 4

Piatigorsky, Gregor, 1 6 3 , 4 1 8 , 4 1 9

Movay, M , xii

Piatigorsky, Jacqueline, 1 6 3 , 3 6 7

Mozart, Wolfgang Amadeus, 5 6 8

Piccinali, Carlos, 472

Mueller, Hans, 502

Pigusov, Evgeny, 465

Pilnik, Herman, 5 8

agy, Ferenc, 70, 7 1

Pinter, J ozsef, 3 7 6, 49 3

aj dorf, Miguel, 7 5 , 76, 96, 1 1 9, 1 6 5 ,

Pirc, Vasj a, 509

1 66, 43 2

Planinc, Albia, 3 2 3

aumkin, Igor, 2 5 1

Platonov, Igor, 2 9 6

avarovsky, Laszlo, 654

Platz, Hans, 5 8 , 1 07 , 5 5 8 , 5 5 9

azzari, R, 502

Podgaets, Mikhail, 3 2 6, 45 1 , 5 0 3

ikolic, Predrag, 72

Pogats, B , 7 2

imzowitsch, Aaron, 5 1 5

Polgar, Judit, xii, xiv, 9 5 , 40 1

ovarovaki, 82

Polgar, Klara, xii

unn, Dr. John, 2 3 8 , 472 , 47 3 , 495 , 5 0 8 ,

5 1 1 , 607

Polgar, Sofie, xii, xiv

Polgar, Susan, xii, xiii, xiv
Polugaevsky, Lev, 2 6 , 59, 7 6 , 3 3 5 , 479, 490

Ogaba, Dr, 47, 48, 474

Polzer, Hans, 3 1 , 3 2 , 5 9

O'Kelly, Alberic, 58, 59, 2 6 7 , 3 46

Pomar, Arturo, 3 2 1 , 3 3 4, 3 3 5 , 4 8 6 , 489

Olafsson, Fridrik, 9 5 , 98, 99, 1 02 , 1 0 3 ,

5 1 0, 5 1 8

Ponomariov, Ruslan, 95, 5 3 9 , 5 40

Popo Luben, 3 0 3 , 456

Olafsson, Helgi, 3 6 3

Popovic, Mladen, 482

Opocensky, Karel, 43 , 1 94

Portisch, Lajos, 72, 7 8 , 79, 80, 8 1 , 82 ,

Ostojic, Predrag, 2 74, 47 1

2 40 , 2 9 5 , 3 2 0, 3 3 8 , 3 6 3 , 3 7 3 , 43 1 , 49 5 ,
5 1 8 , 545 , 5 46

Pachman, Ludek, 2 8 , 1 1 0, 1 60, 2 40, 3 0 5 ,

3 2 6, 3 5 7 , 467 , 495
Padevsky, ikola, 3 3 5, 402

Predein, 483
Pribyl, Josef, 460
Prins, Lodewijk, 3 9

Palamik, Semon, 3 9 2 , 447

Psakhis, Lev, 3 8 8

Palme, Rudolf, 459

Pupols, Viktors, 5 5 9

Panno, Oscar, 1 68 , 3 3 8 , 3 6 8 , 44 1 , 450,

Pustina, Ylvi, 480

45 5 , 5 1 9

Pytel, Krzysztof, 3 5 6

Paoli, Enrico, 5 2 4
Parma, Bruno, 1 90 , 2 2 9, 3 1 6, 486, 5 2 1

Quinteros, Miguel, 3 2 7 , 3 3 2 , 45 1

Paros, Gyorgy, 569

Paulsen, Louis, S 34

Radulov, Ivan, 2 8 1

Paveliev, Alexey, 48 3

Ragozin, Viacheslav, 3 7 , 470, 487

Pedersen, Steffen, 5 3 4, 5 3 5 , 548

Raicevic, Vladimir, 3 9 3

Perez, 5 1 , 502

Raizman, Maurice, 72

Pesolano, orberto, 472

Rajna, George, 482

Peters , Jack, 3 90, 3 9 1 , 447

Rakic, Tomislav, 3 2 8

Petrosian, Rona, 6 1

Rashkovsky, ukhim, 2 2 1 , 542

665

PAL B E N KO : MY LIFE, GAM ES AN D COM POSITI O N S

666

Ravikumar, Vaidyanathan, 2 5 6

Schweber, Samuel, 465

Ree, Hans, 3 3 5 , 499

Sehner, Norbert, 454

Reefschlaeger, Helmut, 5 1 5

Seirawan, Yasser, 2 2 4, 2 7 5 , 3 8 3 , 3 9 1 , 3 98 ,

Reinhardt, Enrique, 49 1
Rejfir, Josef, 445

42 7 , 446, 46 1
S elensky, Mrs , 3 7 5

Rellstab, Ludwig, 5 3 3

Shaked, Tal, xvi

Renner, Christoph, 5 2 1

Sherwin, James, 486

Reshevsky, Sammy, xix, 8 5 , 9 7 , 1 06 , 1 1 7 ,

Shabalov, Alexander, 40 1

1 1 8 , 1 1 9, 1 2 1 , 1 69, 1 7 2 , 2 0 5 , 2 70, 2 7 1 ,

Shamkovich, Leonid, 3 8 3

3 1 0, 4 1 8 , 4 1 9, 42 0, 42 2 , 42 3 , 424, 4 3 0 ,

Shipman, Walter, 5 2 9

447 , 492 , 5 5 3 , 660

Sibarevic, Milenko, 2 2 9

Reti, Richard, 1 0, 2 6 7 , 43 1

Sigurj onsson, Gudmundur, IX , 2 9 1 , 488

Reyes, L, 2 9 , 5 3 0

Silman, Jeremy, xiv,

Ribli, Zoltan, 3 3 8
Rind, Bruce, 3 2 6

xv ,

xvi, xvii, 4 1 3 , 4 1 4,

4 1 5 , 4 1 6, 4 1 8 , 4 1 9, 42 1 , 42 2 , 42 3 , 424,
42 5 , 42 6, 42 7 , 42 8 , 42 9, 430, 43 1 , 4 3 2

Rio, E . del, 1 89

Simms, Gary, 4 7 8

Rivise, Irving, 4 1 6

Sinatra, Frank, 4 1 8

Roeder, Frank, 408 , 5 0 5

Sines, Josip, 3 7 1

Rogoff, Ken, 1 1 9

Singer, Richard, 49 1

Romanishin, Oleg, 475

Sinulingga, Monang, 480

Roos, Nancy, 1 7 7

Slapikas, Vytautas, 2 2 4

Rosino, Antonio, 494

Smejkal , Jan, 3 3 8

Rossetto, Hector, 5 2 6

Smith, Ken, 8 5 , 86

Rossolimo, Hector, 3 4 , 8 5 , 1 9 1 , 3 2 9, 4 5 8 ,

Smyslov, Mrs, 1 5 3

459, 47 5 , 486, 487

Smyslov, Vassily, 3 7 , 3 8 , 5 8 , 1 00, 1 0 1 ,

Ruban, Vadim, 2 2 0

1 0 2 , 1 0 3 , 3 99, 42 7 , 42 9 , 440, 498 , 5 1 3 ,

Rubinstein, Akiba, 1 9 5 , 3 9 7

5 1 8, 5 5 1 , 5 5 8

Rudahl, Sharon, 420

Sokolov, Sergey, 448

Ruzhele, 5 5 1

Soltis, Andrew, 3 3 1

Ryskin, Alexander, 2 0 1

Soos, Bela, 243 , 5 5 5

Sorin, 493

Sadler, Matthew, 2 1 1 , 2 2 0

Sosonko, Gennadi, 460

Sahovic, Dragutin, 486

Spann, Jerry, 2 8 7

Saidy, Anthony, 5 5 , 2 0 5 , 42 5 , 5 5 5

Spassky, Boris, 2 6 , 2 7 , 96, 1 0 5 , 1 2 1 , 1 7 1 ,

Saint Amant, Pierre, 43

Salov,Valery, 2 5 1 , 488, 5 2 7

2 7 0, 2 74, 3 1 0, 3 6 3 , 43 1 , 490, 5 1 0, 5 1 3 ,
5 1 8, 5 5 2 , 5 5 3

Sandor, Bela, 7 2 , 495

Spiridonov, Nikola, 5 5 8

Sanguinetti, Raul, 5 1 9

Stahlberg, Gideon, 64, 7 3 , 2 3 3 , 470

Santacruz, Cesar, 480

Stalin, Joseph, 70

Santasiere, Anthony, 3 3 1

Staunton, Howard, 43

S api, Laszlo, 440

Steibler brothers, the, 40 1

Savchenko, Stanislav, 454

Stein, Leonid, 96, 1 2 5 , 47 3 , 5 1 5 , 660

Saverymuttu, Seth, 3 1 6, 47 1

Steiner, Herman, 1 3 6, 45 5

Savon, Vladimir, 2 8 1

Steinitz, William, 42 8 , 644

Schafer, Charles, 569

Steinmeyer, Robert, 499

Schammo, Jean, 5 2 9

Stohl, Igor, 2 3 8

Schlosser, Philipp, 3 48

Stoltz, Goesta, 5 1 , 2 3 3

Schmid, Lothar, 1 2 9, 47 3

Suba, Mihai, 3 2 6

Schmidt, Wlodzimierz, 2 3 6 , 47 1 , 548

Suetin, Alexe 5 1 3

Schneider, Attila, 3 5 7

Susan the Orangutan, 3 3 1

Schwarz, Rolf, 448 , 45 1

Suttles, Duncan, 1 8 5 , 3 2 6, 4 3 0 , 5 1 1 , 654

I N D EX OF NAMES

Sveshnikov, Evgeny, 2 2 0

Vogt, Lothar, 4 7 5

Szekely, Peter, 40 1 , 48 1

Vranesic, Zvonko, 2 42 , 243 , 5 46

Szigeti , Miklos, 9

Vukcevich, Milan, 40 1 , 46 3 , 482

Szilagyi, Gyorgy, 1 5 , 67

Vukic, Milan, 2 3 3 , 2 3 4, 5 3 4, 5 3 5 , 5 3 7

Szabo, Laszlo, 1 0, 2 8 , 47, 5 5 , 5 6 , 5 7 , 70,

Vukovic, Vladimir, 7 6

1 89, 2 2 9 , 2 3 3 , 3 8 7 , 40 1 , 4 1 3 , 439, 509,

522
Szily, Dr. Jozsef, 7 2 , 5 2 5

Wade, Robert, 495

Watson, John, xix, 43 3 , 5 0 1
Webb, Simon, 468

Tagmann, 1 9, 2 0
Taimanov, Mark, 1 1 3 , 242 , 448 , 4 7 5 , 547
Tal, Mikhail, 2 6, 78, 88, 1 0 3 , 1 07 , 1 2 6,

Weinstein, Raymond, 64, 96, 97, 1 5 3 ,

1 99 , 424, 47 2 , 5 1 2
Wirthensohn, Heinz, 5 2 9

1 2 8 , 1 3 8 , 1 5 0 , 1 5 1 , 1 5 3 , 1 5 4, 1 5 5 , 2 0 3 ,

Wolff, Patrick, 3 2 6

2 0 7 , 2 7 7 , 2 8 1 , 3 3 2 , 3 99, 40 1 , 42 6, 42 7 ,

Wood, Baruch, 2 8

42 8 , 429, 447 , 45 1 , 45 2 , 45 3 , 454, 467,

468 , 48 3 , 490, 5 1 5

Yandemirov, Valeri, 482 , 48 3

Tapaszto, Laszlo, 5 2 2

Yanofsky, Daniel, 1 2 5

Tarjan, Jim, 2 3 8 , 3 94, 4 1 7 , 42 0

Yrjola, Jouni , 2 74, 496

Tan, 499

Yusupov, Arrur, 1 1 5 , 2 5 6, 3 4 5 , 3 46

Tartakower, Saviely, 6, 2 8 , 3 3 1 , 42 9, 5 5 3
Tarrasch, Seigbert, 46, 5 3 3

Zagoriansky, Evgen 3 6 3

Tatai, Stefano, 2 7 7 , 49 5 , 496

Zaitsev, Igor, 76, 1 94, 2 5 1 , 2 5 2 , 2 5 3 , 2 5 5 ,

Tautvaisas, Povilas, 49 1

540, 5 4 1 , 542

Taylor, Timothy, 2 5 0, 5 49

Zaltsman, Vitaly, 3 96, 5 3 9, 540

Thorold, Edmund, 46

Zeller, Art, 2 6 5 , 3 67 , 42 1

Timman, Jan, 3 4 5 , 3 46, 42 7 , 447

Zilberman, Nathan, 2 5 3 , 542

Timoshenko, Georgy, 496

Zita, Frantisek, 4 5 8

Tisdall, Jonathan, 547

Zuckerman, Bernard, 445 , 509, 5 1 1 , 5 2 5

Tomasits, 2 1

Zuidema, Coenraad, 4 5 3

Topalov, Veselin, 2 5 6

Zukertort, Johannes, 5 1

Torre , Eugenio, 3 99 , 440

Trifunovic, Petar, 5 5 7
Tringov, Georgi, 1 96, 2 1 4, 5 1 1 , 5 3 1
Troitzky, Alexei, 6 3 0
Trosclair, Wayne, 3 99, 4 3 9
Tseshkovsky, Vitaly, 2 5 2
Tukmakov, Vladimir, 2 5 1 , 3 9 7
Turi, 45
Uhlmann, Wolfgang, 1 1 5 , 1 1 7 , 1 3 1 , 1 3 2 ,
1 74, 3 1 2 , 46 5 , 467, 494, 549
Unzicker, Wolfgang, 6 , 2 7 , 5 0 3
Vaganian, Rafael, 48, 2 5 1 , 5 2 8
Van Riemsdijk, Herman, 3 48 , 5 3 9
Van Scheltinga, Theo, 1 0, 1 1 , 5 62
Van Wely, Loek, 5 3 2 , 5 3 9
Varga , Zoltan , 6
Vasiukov, Evgeni, 2 5 3 , 472
Vaulin, Alexander, 2 6 5
Velikzhan, 1 0 1
Visier, Segovia, 2 3 6, 548

667

Numbers refer to game. Numbers in bold indicate a game where Benko was
White.

Alekhine Defense 5
Benko Gambit 7 2 , 7 3 , 74, 7 5 , 76, 7 7 ,

7 8 , 79, 80, 8 1 , 82

Nimzovich - Larsen Opening 3

Orangutan Opening 1 02 , 1 1 1
Pirc Defense 4 1 , 44, 87, 1 0 1 , 1 04, 1 2 9

Benko System 42

Queen's Gambit Accepted 3 4

Benoni Defense 59, 1 1 7

Queen's Gambit D eclined 3 3 , 9 1 , 99,

B ogo-Indian Defense 96

133

Budapest Gambit 1 0

Queen's Indian Defense 5 7 , 7 1 , 1 2 4

Caro-Kann Defense 4, 5 3 , 69, 1 3 2

Reti 8 5 , 9 2

Catalan 62 , 46, 1 2 0

Ruy Lopez 3 0

Dutch Defense 1 2 , 1 06

Scotch Opening 1 5

English Opening 1 , 24, 26, 5 1 , 66, 70,

84, 86, 93 , 1 0 0, 1 09, 1 1 0, 1 1 2 ,

S emi-Slav Defense 2 , 1 6, 94, 1 1 9

Semi-Tarrasch Defense 6, 2 2 , 43 , 58,

95

1 1 3 , 1 2 2 , 1 2 3 , 1 2 7 , 1 2 8 , 1 3 1 , 1 3 4,
137

Sicilian Defense 7 , 9, 1 1 , 1 7 , 2 8 , 29,

3 7 , 47, 6 1 , 64, 65 , 6 8 , 8 3 , 90, 1 1 6 ,

French Defense 1 3 , 14, 3 8

Grunfeld Defense 20, 2 7 , 3 5 , 40, 1 1 5 ,

125

1 1 8, 1 3 5, 1 3 8
Slav Defense 1 9

King's Fianchetto 9 7

Tarrasch Defense 107

King's Indian Attack 1 8 , 39, 4 5 , 50, 5 2 ,

Tarrasch Defense Reversed 89

103

Vienna Game 54

King's Indian Defense 8 , 2 1 , 2 3 , 2 5 , 3 1 ,

3 2 , 3 6, 48, 49, 60, 63 , 88, 1 2 6

Martinovs ky System 1 3 6
Modern Defense 5 5 , 67 , 1 08

668

I"
J

9 781890 0850

545 00