Language and other communication systems

A widely recognized problem with the term 'language' A more useful approach to language, an
is the great range of its application. This word has by most modern linguists, is to identify
prompted innumerable definitions. Some focus on the properties that are thought to be its essen
general concept of 'language', some on the more specific characteristics. The aim is to determine \
notion of 'a language'. Some draw attention to the formal as a human language, as opposed to some
features of phonology (or graphology), grammar, and of communication. Two main kinds of
semantics (Parts HI-VI). Some emphasize the range of been used. One focuses upon identifying
functions that language performs (Parrs I, II). Some stress structural properties of language, and
the differences berween language and other forms of in Part 1ll (§§13-15). The other is to co
human, animal, or machine communication (see below). with non-human forms of communi
Some point to the similarities. At one extreme, there are other forms of human communication.
definitions that are highly technical in character; at the
other, there are extremely general statements, reflecting Design features of communica:
the way in which the notion has been applied figuratively The most widely acknowledged comp
to all forms of human behaviour, such as the 'language' has been that proposed by the American Ii ~
of music, cookery, or the cinema. F. Hockett (1916-2000), who used a zool :0-

Most textbooks in the subject avoid the problem, of enquiry to identify the main points 0 - ~'-__

preferring to characterize the notion of language rather berween language and other systems of conu==0:::1
Charles F Hackett especially those found in animals. His set
than define it. They recognize that the question of
identifying an individual language has no single, simple features of communication using spoken I _
answer, because formal and social criteria are often in as follows:
conflict (§47). Similarly, they note the correspondingly Auditory-vocal channel Sound is
complex problems that arise when arternpcing to mouth and ear, as opposed to a visual, ra
construct a definition of language in general that makes means (pp. 425-7).
a precise and comprehensive statement about formal and Broadcast transmission and directio
functional universal properties. The set of definitions A signal can be heard by any auditory
given below exemplifies the way different writers have earshot, and the source can be located ••c;,...,,...--._
attempted to tackle the problem, and illustrates some of direction-finding abiliry (p. 148).
the difficulties involved. There seems little to be gained by Rapid fading Auditory signals are
trying to summarize the content of the present volume in do not await the hearer's convenience
a single sentence - unless it is the banal observation that tracks, or writing, §3l).
'language' is what this encyclopedia is about! Interchangeability Speakers of a

LANGUAGE DEFINITIONS
'Language is a purely human and non-instinctive method
of communicating ideas, emotions and desires by means
of voluntarily produced symbols.' (Sapir, 1921.)
'A language is a system of arbitrary vocal symbols by
means of which the members of a society interact in
erms of their total culture.'
(Trager, 1949.)
anguage is 'a set (finite or infinite) of sentences, each
" e n length and constructed out of a finite set of
ements'.
msky,1957.)
e IS 'the institution whereby humans
- ~:cae and interact with each other by means of
- .sed oral-auditory arbitrary symbols'. (Hall,
A dictionary definition
1. the words, their pronunciation, and the methods
of combining them used and understood by a
considerable community and established by long
usage.
2a. audible, articulate, meaningful sound as produced by
the action of the vocal organs.
2b. a systematic means of communicating ideas or
feelings by the use of conventionalized signs, sounds,
gestures, or marks having understood meanings.
2c. an artificially constructed primarily formal system of
signs and symbols (as symbolic logic) including rules
for the formation of admissible expressions and for
their transformation.
2d. the means by which animals communicate or are
thought to communicate with each other.
3. the faculty of verbal expression and tile
in human intercourse ... significant co -
4. a special manner or use of expression
Third New International Didionary, 19E
And a comment
'The question 'What is language?" is com
and, some would say, hardly less profoure
is life?", the presuppositions of which circu
unify the biological sciences ... it is not so ~
question itself as the particular interpretat
biologist puts upon it and the unravelling
detailed implications within some curre
theoretical framework that nourish the b
day speculations and research. So it is for -
relation to the question 'What is language
p.L)
 64 • LANGUAGE AND OTHER COMMUNICATION SYSTEMS 421

uce any linguistic message they can understand

re the differing courtship behaviour of males and
....-- .... ...
es in several species).
, ," ,
udback
~""'-hi'ing
Speakers hear and can reflect upon
that they say (unlike the visual displays
,
I
used in animal courtship, which are not visible I
I
- e displayer). I
'ization The sound waves of speech have no I
I
'on other than to signal meaning (unlike the
I
e panting of dogs, which has a biological I
se), I

ticity The elements of the signal convey

~. g through their stable association with real- \
\
iruations (unlike dog panting, which does not
. a dog is hot; it is 'part of' being hot).
\
,
trartness There is no dependence of the element
ignal on the nature of the reality to which
(unlike the speed of bee 'dancing', which
. reflects the distance of the nectar from the
guage' of bees
Productivity There is an infinite capacity co 0;: '-e ~OS! closely
teness Speech uses a small set of sound elements and understand meaning, by usinz old sentence ~.,,:""'~:~ 'oons of animal
dearly contrast with each other (unlike elements to produce new sentences unlike =._ .•...."""''''''.IS me 'dance'
~_~·.u;;'"and other emotional noises, where there limited, fixed set of calls used by animals). a honey bee when
ntinuous scales of variation in strength). Traditional transmission Language is transmitted from which
ement It is possible to talk about events remote one generation to the next primarily by a process of eel:''';. CJr:'CSe ~ ormation
ce or time from the situation of the speaker teaching and learning (unlike the bee' ability co e Ild amount of
e most animal cries, which reflect immediate communicate the source of nectar, which is passed iiXl:::it io (:JS(lJ'>'ffled. Several
• pi! ern have
nmental stimuli). on genetically).
• e round
Duality of patterning The sounds of language have
- ity of the 13 design features to six systems of t; used when
no intrinsic meaning, but combine in different \'<l~
-====;:;:x)n (after Hockett. 1960, pp. /0-1/). The music column e ISdose 0 the
to form elements (such as words) mat do convey
western music Since the time of Bach. A question mark -c, • e Je'i' es in circles
meaning (unlike animal calls, which cannot be
a it is unclear or unknown whether a system has a , and right In
3Jrure. A blank space indicates that a feature cannot be analysed into two such levels of structure . _~ dance' (above,
e: oecouse other information is lacking. - ,U5C()' en me source is
iffliJ'{, e bee moves in
Bee Stickleback Western Gibbon Language , line while wagging her
dancing courtship meadowlark song calls en from side to side, then
~uditory no no yes yes yes • rns to her starting point. The
srraight line points in the direction
yes yes yes of the food, the liveliness of the
yes
00 and dance indicates how rich a source
_ reception it is, and the tempo of the dance
_::'1g yes yes, repeated es es provides information about its
distance. For example, in one
eability limited no ? yes es ?
study, an experimental feeding
~::DOck no yes yes es es dish 330 metres from the hive
in part yes? yes es es was indicated by 15 complete
runs through the pattern in 30
yes no in part? yes es no (in general)
seconds, whereas when the
ess no if sema ntic, yes yes yes dish was moved to 700 metres
ess no ? yes yes in part distance, only 11 runs were
carried out in that time. No other
e=en: yes, always no yes, often
animal communication system
yes no no yes yes seems able to provide such a
_ tJansmission probably not no? yes yes quantity of precise information
~:patterning - except human language. (After
no no yes
van Frisch, 1962.)
 G GE A D COMMUNICATION

Chimp communication A quite different way of proceeding was __ ~

The formal and functional complexity of language is in the case of a 5-year-old chimpanzee cal1ec.

such a distinctive human trait that many scholars think a research programme that began in 1954
the designation homo loquens ('speaking man') to be a Premack, 1983). She (and, later, several others
better way of identifying the species than any other single a form of written language - ro arrange and
criterion that has been suggested (such as rool using) (p. vertical sequences of plastic rokens on a mag
301). This is not to disregard the complex patterns that Each roken represented a word, e.g. small bl ~
have been observed in the natural communicative systems apple, small pink square = banana. In due course .•...•~--.

of birds, insects, apes, and other animals (the subject was able to teach Sarah ro respond correctly to ~_~~
names semantic sequences (e.g. 'give Mary apple' .
paruons) matter of the field of zoosemioticsi. But no animal system
remotely compares with the level of sophistication found in number of more abstract notions, such as 'same' _
~ Jr G, Greg, Roger, Linn,
rs G. Susan, Washoe human language. The evolutionary gap is very wide. Only and 'if/then' (e.g. ? apple different banana).
Pronouns the experiments in teaching language ro chimpanzees have Chimp language research attracted
me, we, you suggested that this gap may be somewhat narrower than media publicity in its early years, with repo
Common nouns has traditionally been assumed. on the implications of the work. What "
baby dirty nut Early experiments ro teach chimpanzees to say if they could use language? What woul
bath drink pants of the human race? Would they claim civil
communicate with their voices failed because of the
bed flower pencil
insufficiencies of the animals' vocal organs (p. 300). speculations were wholly premature, given
berry food purse
However, when attempts were made to communicate with findings of the research ro date. These fin
bird fruit ride
blanket gun shoe them using the hands, by teaching a selection of signs from any case controversial, receiving a range
book hammer smoke American Sign Language (ASL, see Part VI), dramatic extending from rota] support to total anripa
brush hat spoon of interpretations seems possible. It is evident ;_ ~....:...._
progress was claimed. The first subject was a female
bug Ice swallow
chimpanzee named Washoe, whose training began in 1966 can learn to imitate signs, combine them in
car key sweet
when she was less than a year old. It rook her just over four and use them in different contexts, but the ..-..__
cereal leaf tree
chair listen water years ro acquire 132 ASL signs, many of which bore striking of this behaviour is less clear. Many scholars
cheese lollipop window similarities to the general word meanings observed in child the chimps' behaviour can be explained as a _
clothes look wiper imitation ability rather than as evidence for c~.~,--
language acquisition (Part VII). She also began to put signs
comb man woman linguistic processing, and they argue the n
together ro express a small set of meaning relations, which
cow meat
resembled some of the early sentences of young children, accounts to be provided of chimp behaviour
Possessives training methods used, in order to evaluate
mine, yours such as want berry, time drink, there shoe (Gardner &
Gardner, 1975). Washoe died in 2007 aged 42. being made about learning. More sysrerna
Traits
Since then, several other chimpanzees (and also begun ro be collected, but it will be some time
funny, good, hungry, stupid
gorillas) have acquired a vocabulary of signs, and alternative questions can be resolved.
Colours
black, white, green, red teaching procedures have been tried. For example, in the
NOUNS
Temporal case of the chimps Moja and Pili, sign language teaching
time
Negative
began soon after birth) and training was carried out by
native signers. Both chimps began to sign when they were
I ~

"
••-
can't, enough, no about 3 months old, and had over a dozen signs by the age
SARAH MARY PAUL
Imperative of 6 months - a marked contrast with Washoe, who had
gimme, help only 2 signs after 6 months of training. ~
Appetitive CHOCOLATE APPLE BANANA APRICOT
please, want VERBS
Quantitative
hurry, more
Verbs
b e, catch, cry, go, hug, open,
peekaboo, smile, tickle
IS

CONCEPTS/CONDITIONALS
~
GIVE TAKE

*
INSERT

Locatives SAME
up, there
. 01sequences
( C3
NAMEOF COLOROF
:: ::shoe Food fruit
ADJEaIVES (COLORS)
Time drink
Good me
'Peony nose touch' Peony, one of the 'second generation'
of chimps trained in the Premack study, carries out this instruction, RED YELLOW BROWN

which her trainer has placed on the magnetic board (From Premack Sarah's symbols Symbols used in communicating
& Premack, 1983, p. 29.) and the other chimps. (From Premack & Premack, 1983. z _
 64 • LANGUAGE ADO E

IOTIC5 derivative use of the visual mode that resulted in

developmenr of written language. Further writing-
e can also be studied as part of a much wider
based codes, such as semaphore and morse, would
of enquiry: semiology, or semiotics - a subject also be included here. Non-linguistic forms of visual
"'-' •..••owes a great deal to US philosopher Charles Sanders
communication include the systems of facial expression
1839-1914), as well as to the work of Ferdinand
and bodily gesture, which are the subject matter of
. """-""Ire (p. 431). This field investigates the structure
kinesics (p. 426).
possible sign systems, and the role these play in
. we create and perceive patterns (or 'meanings') Tactile
ulrural behaviour. The subject is all-inclusive, Tactile communication has very limited linguistic
re, dealing with patterned human communication function, apart from its use in deaf-blind communication
modes (sound, sight, touch, smell, and taste) and and in various secret codes based on spoken or written
nrexts (e.g. dance, film, politics, eating, clothing). language (p. 60). Its main uses are non-linguistic, in the
ject matter of the present book would form but a form of the various ways in which bodily contact and
'on of any proposed encyclopedia of semiotics. physical distance between people can signal contrasts of
meaning - the subject matter of proxemics (p. 425).
: .orv-vocal
"ram below shows the relationship between The communicative use of the visual and tactile modes
_:_":.'_;'"' as identified in Parts III-V1, and other aspects is often referred to as 'non-verbal communication',
an communication. The structured use of the especially in academic discussion. In everyday terms, it is
J -uocal mode, or channel (p. 424), results in the area of 'body language'.
nmary manifestation of language: speech. But
zuistic uses of the vocal tract are also possible:
Olfactory and gustatory
There seems to be little active r Ie for the olfactory
gical reflexes, such as coughing and snoring;

_.-.w _
and gustatory modes in human communication (a
effects, such as whistling; and the communication
marked contrast with the important use of these senses
ciry, in the form of voice qualiry (§6). The
for communicative purposes in the animal kingdom).
~--PCn""enral aspects of vocal expression (§29) are
However, they do play an important part in our reception
• included within the study of language, though
of information about the outside world (e.g. in smelling Other modes? This is
cult to draw a clear-cut boundary line between
and tasting food). The communicative use of body odour the pictographic message
of these effects (those placed under the heading
seems to have a mainly sexual role in human sociery; bur transmitted into space by the
anguage', such as giggling and whispering) and Arecibo radio telescope in
there are several anecdotes of its use in other domains.
mat clearly fall outside language. Puerto Rico in 1974. The signal
One linguist even claimed to be able to tell when his
was aimed at the cluster of
informants (p. 434) were under strain (and perhaps 300,000 stars, known as M 13, In
ual mode is used for a variery of purposes, some therefore were being less reliable) by the different body the Hercules constellation.
ic, some not. The primary way in which visual odour they exuded! The message consists of a
have a linguistic use is in the various deaf sign series of radio pulses which can
<Yes(Part V1). In addition, there is the historically be arranged into a pictogram.
includes data on the chemica
basis of Iife on earth, the huma
SEMIOTICS
form, and the solar system.
assumes, of course, tha the
communicative system olt e
receiving species is capa e
responding to the same
contrasts as are diso ec
the pictogram (sha
Audirory-Vocal Visual Tactile Olfacrory Gustarory
etc.). If the en rec

/1\/1\
the signal happens •
commumcarUve'svs'E!::
on, sa hea·-
wtll have w"",o""t'-"",, !'--.d
Physiological Musical Voice Sign Writing Kinetics Deaf- Secret Proxemics
vocal reflexes effects qualities languages I \ Blind codes
• language
Codes
'Body language'
(non-verbal communication)
 I'l GUAGE AND COMMUNICATION

Auditory-vocal effects hut. Without saying a word he dumped his

The main systems of communication using the auditory- ground. Eusebio looked the load over, wen
vocal channel have been described elsewhere in this returned with some money, and paid the
volume (Parr IV). However, from time to time linguists The man turned and left. Not a word had
have reporred types of auditory communication that fall They had talked, bargained over the price.
ourside the normal use of the vocal apparatus - notably, an agreement satisfactory to both parties -
the whistled speech of several rural populations. This is whistles as a medium of communication. C
found in some Central and South American tribes, as p.280)
well as in the occasional European community (e.g. in This conversation took place berw
Turkey and the Canary Islands, based on Turkish and speakers, members of a tribe that lives in
Spanish respectively). the state of Oaxaca, Mexico. The whistled ~" __ ~_"""I
closely correspond to patterns of spoken
Whistled speech has been shown by having the whistlers
Eusebio Marrinez was observed one day standing in front tunes into speech. It is thus quite unlike the =::::::.==:IIIi
of his hut, whistling to a man a considerable distance whistling patterns used as attention sig
away. The man was passing on the trail below, going to whistling') in Euro-American culture. For
market to sell a load of corn leaves which he was carrying. the following sequence of whistled uneran
The man answered Eusebio's whistle with a whistle. The tones are classified from 1 (high) to 4 (I
Men of the Libinzo tribe interchange was repeated several times with different between tones are marked by a dash .
(Democratic Republic of
whistles. Finally the man turned around, retraced his meanings are signalled, as the following ...,,"=",.....--
Congo) making Mokoto drums
steps a shorr way and came up the footpath to Eusebio's Mazareco shows:
1,1,3,3,2,4 bme' C?al J~ kr-C? ar-ve
DRUM SIGNALLING 'What did you bring there?'
1,4,1,1 C?al na' hme=ni'
In several parts of the world - notably Africa, the The drummer, an official of the town's law-
'It is a load of corn.'
Americas, and the Pacific - drums, gongs, horns, enforcing authority, controls the way meetings
and other musical instruments have been used to take place, using special signals to do such things 1,3,3,4,3 hnd' tr- ?mr koaii- ?nr
simulate selected features of speech (primarily, as call for order, summon people, and end the 'Well where are you going with it?'
tones and rhythms). In Africa, drums are the usual meeting. These signals consist mainly of fixed 3,2,4,2,3,4 tel na' nko' tr-vhr koa'
instruments involved, and quite elaborate systems formulae, with a few variations and additions. The 'I am taking it to Tenango.'
of communication have developed. Jabo rarely use these drums for communicating 3,3,3,3,2,3,2-4,3 ?a'-tr-1mr ka' tc
One system, used among the Jabo tribe of with other villages (unlike the drum signalling 'Are you going to sell it then?'
Eastern Liberia, makes use of a wooden signal found in many other parts of Africa). 2,3,3,2,2-3 tr-vhr ka' tll na2-3
'drum' (actually, more like a bell, as it has no skin The words and syllables of Jabo are tonal 'I am going to sell it.'
covering) - a hollowed tree trunk, often over 2 (§29): there are four basic tones, which are often
1,1,3,2,4,4,2,3,1-3,4 hOItb]' f?ar- 1'1.
metres in length. This has a longitudinal slit with linked by glides, and these interact with aspects
?i4_ta4 tll na'-nail'-3-vi4
lips varying in thickness, thus allowing several of the vowel and consonant system. There is also
'How much will you take then? Sell it
different tones to be produced. Two straight sticks considerable variation in the length of these tonal
are used for beating, and further tonal variations contrasts, which accounts for several of the drum 4-3,4,3,3,3,2,4 ka4-3tfJl/ k1oa' nka'
can be made by altering the way these sticks hit patterns used. Some examples of these signals, 'I will take $2.50 a box.'
the drum. Other types of drum are also used for with a transcription in Jabo, are given below. (Cowan, 1948,pp. 284-5)
differen purposes (such as dancing). (From Herzog, 1945.) The whistled tunes are based on the pa:-
and rhythm used in the spoken lang
-,)))) 2. ) )) ) 7) ..t ..t ..t ) 7 convey precise distinctions. With very Ii
• :11
each 'syllable' of whistle corresponds to
'F'-4 wil ~192 ba2 di22 leI ba2 po2le2 kpe2Jel
speech. Ambiguity is uncommon, because
'Grtttings!' 'Come ye quick! Put ye your effort there!'
the conversation is usually something
situation of the speakers. However, it i
both speakers to use the same musical
confusion may arise.
ve the noise. Speak ye one by one!' (Played in the men's assembly Whistled dialogues tend to contain a
CLS<:us:s.ioo rhrearens to get out or hand.)
of exchanges, and the utterances are shorr.
commonly heard when people are at a distarce _
other (e.g. when working the land), bur th
found in a variety of informal settings. Al
'To collect fines-hunger 15 raging-we are going to collect fines (indeed)' (Played are able to understand whistled speech, :_
before the assembly sends out. group to collect fines imposed by the courr.)
used only by and between males.
 64 • LANGUAGE AND OTHER COMMUNICATION SYSTEMS 425

:-ile effects THE AMPLIFIED HAND-SHAKE

communicative use of touching behaviour,
In a culture where handshaking is a
~ .cs, has in recent years attracted a great
normal formality, extra warmth can
of research by psychologists, sociologists, and
be expressed only by extra activity,
pologists. A very wide range of activities is
such as increased firmness, longer
, as is suggested by this small selection of terms duration, and more vigorous vertical
. g bodily contact: movements. The second hand may
race lay on (hands) punch also be brought into play, as shown
de link (arms) shake (hands) in the diagrams, which illustrate
nudge slap increasing warmth: (a) hand clasping,
(b) arm clasping, (c) shoulder
pat spank
clasping, and (d) shoulder embracing.
pinch tickle
(From Morris, 1977, p. 93.)
communicative value of tactile activities is
:airly clear within a culture, as they comprise
uncomfortable when others invade their 'body space',
: the most primitive kinds of social interaction
by moving too close to them (e.g. in a queue, ourside a Distance zones
of the activities are found berween animals).
cinema, on a beach). Any cultural variations can easily lead An American study suggests that
ress such 'meanings' as affection, aggression
to conflict and misinterpretations. Latin Americans, for there may be four proximity
. and pretend), sexual attraction, greeting and zones when people interact:
example, prefer to stand much closer to each other than
o congratulation, gratitude, and the signalling
North Europeans, so that when the former and the latter • Intimate Less than 45 cm,
n. They operate within a complex system of
converse, there may be a problem. The present author used for intimate relationships.
.nts: some of the acts tend to be found only • Personal Between 45 cm
recalls one such conflict during a conversation with a
notably, sexual touching); some are specialized and 1.3 metres, for reasonably
student from Brazil, who came and stood before him at
~ e.g. the tactile activities carried on by doctors, close relationships.
some 45 ern distance - a normal interaction distance for
_----"'_"'. dressers, or tailors); and some are restricted to • Social consultative Between
her, but much too close for him. He instinctively moved
ernonies (e.g. weddings, graduation, healing). 3 and 4 metres, for more
back to the distance he found most comfortable - nearer 1
a subjective impression about how these impersonal relationships.
metre. However, as he did so, the student moved forward, • Public Above 4 metres,
. place, and what they mean. But there are
unconsciously maintaining her own norm. He retreated for public figures and public
ences in behaviour berween individuals and
further, not wishing to be so close to the student. After occasions .
.[ is not easy to make accurate generalizations
both had circled the desk several times, he capitulated, (Hall, 1959.)
as a whole.
and asked her to sit down! The rules of Indian caste (p. 38)
':::cult to study tactile activity in an objective
illustrate the point even more
problem is how to obtain clear recordings
TADOMA COMMUNICATION precisely. According to tradition
-- participants are unaware of the observer
Tadoma is a method of tactile speech communication that has in one part of India, members
.- the behaviour is being filmed). There are
evolved between people who are both deaf and blind. Speech is of each caste may not approach
_ ••••• ....,;:;...o.u. ed accounts of the range of communicative
perceived by placing a hand against the face of the speaker and each other within the following
- a society, and of the factors governing their distances:
monitoring the articulatory movements involved. Usually, the
- nt, however, that some societies are much
thumb or index finger is used to sense the movements of the lips, Brahmins - Nayars: 2 metres
-- of touching than others, so much so that Nayars - Iravans: 8 metres
and the other fingers fan out over the side of the face and neck.
__ ~ has been proposed berween 'contact' and Devised in Norway in the 1890s, it got its name from its first use Iravans - Cherumans: 10 metres
societies - those that favour touching (such in the USA with two deaf-blind children, Tad Chapman and Oma Cherumans - Nayadis: 20 metres
Latin Americans), and those that avoid it Simpson (Vivian, 1966). The rules, which are still
- Europeans and Indians). In one study Several other tactile methods of communication are used with followed in some areas, work
. g together in cafes, it was found that in disabled people, such as braille (p. 290). It is also possible to in an additive way: thus, a
.: e people touched each other on average 'translate' such codes as morse and finger spelling (p. 235) into Nayadi may not come closer
our; in Paris it was 110 times an hour; tactile form. to a Brahmin than 40 metres.
don there was no touching at all (Jourard, (Argyle, 1975.)

e people stand from each other, and the

their bodies when interacting, are other
of proxemic behaviour. There are
iry and orientation within a culture that
-===~ .nformation about the social relationship
icipants. A common research procedure
- e point at which people are made to feel
 I· LA GUAGE AND COMMUNICATION

Visua I effects EYEBROW FLASHING

De earned out
'...c-':o..J"~;5'UW~
The field of non-verbal visual communication, kinesics,
-~ ~- - 0' e hand
can be broken down into several components: facial
r down. People used
expression, eye contact, gesture, and body posture.
er could interpret
Each component performs a variety of functions.
ner 0 mean 'Go away'!
- e chart shows the preferred Movements of the face and body can give clues to a
pa em in countries between person's personality and emotional state. The face, in
Britain and North Africa. (After particular, signals a wide range of emotions, such as
Morris et 01., 1978) fear, happiness, sadness, anger, surprise, interest, and When people greet each other at a distance, wish I =
they are ready to make social contact, they raise the' ~
disgust, many of the expressions varying in meaning
with a rapid movement, keeping them raised for a
from culture to culture. In addition, the face and
of a second. The behaviour has been noted in ma :
body send signals about the way a social interaction is
world, and is considered universal (though some CL
proceeding: patterns of eye contact show who is talking
it, e.g. the Japanese, who consider it indecent). We ~'::
to whom; facial expression provides feedback to the aware that we use this signal, but it evokes a strong'
speaker, expressing such meanings as puzzlement or in a greeting situation, and is often reciprocated. To -:
disbelief; and body posture conveys a person's attitude eyebrow flash from someone we do not know is u~
towards the interaction (e.g. relaxation, interest, embarrassing, or even threatening. (After Eibl-Eibes:_
boredom). Several kinds of social context are associated The pictures show an eyebrow flash made by a Ya
with specific facial or body behaviours (e.g. waving Indian (above) and a Huli (Papuan) Indian (belo
while taking leave). Ritual or official occasions are often
primarily marked by such factors as kneeling, standing,
bowing, or blessing.
Visual effects interact very specifically with speech.
Gestures and head movements tend to coincide with
points of emphasis. Hand movements in particular
Being humble
can be used to add visual meaning to what has been
Points of similarity as well as
said (,drawing pictures in the air'). Patterns of gaze
difference can be seen in the BODY TRANSCRIPTION
distinguish the participants in a conversation: a listener
expression of an attitude among
various cultures. In one early looks at a speaker nearly twice as often as the speaker Some of the symbols, or kinegrophs, which have ~
study, the communication of looks at the listener. They also assist in marking the order to transcribe the various movements of face =.-
humility was found to make use structure of a conversation (§20): for example, speakers Different sets of symbols have been devised for d--
of such body postures as the tend to look up towards the ends of their utterances, of the body: head, face, trunk, shoulder/arm/wrist. -
following: thus giving their listeners a cue that an opportunity to hip/leg/ankle, foot activity, and neck. The symbols -
• Join hands over head and the set for facial activities. (From Birdwhistell, 1952.
speak is approaching.
bow (China). Several visual effects may well be universal, but
- ~ - Blank-faced ~9>
• Extend or lower arms the focus of interest in recent years has been on the
(Europe). Single raised brow eo9
cultural differences that can be observed in face and
• Stretch arms towards person ( ~ indicates brow
body movements. Some societies use many gestures -e 09-
and strike them together raised)
and facial expressions (e.g. Italian); others use very few
(Congo).
(e.g. Japanese). Moreover, a visual effect may seem to Lowered brow "<I) <J)" Glc=
• Crouch (Fiji, Tahiti).
• Crawl and shuffle forward; be shared between societies, but in fact conveys very ?)
\I
Medial brow
walk on all fours (Dahomey). different meaning. Thus, in France, using a finger to contraction F\ Poc
pull down the eyelid means that the speaker is aware of
• Bend body downward ..
..:::.... Medial brow nods
(Samoa). something going on, whereas in Italy the same gesture Oe
• Permit someone to place a means that the listener must become aware. Cultural
,00 on one's head (Fundah, Raised brows 't>:d Ioo;
variations in visual effects are among the first things a
onga). foreigner notices, but it can be very difficult working 00 Wide eyed EfE Sq -::
• Prostrate oneself, face down
out what they mean, and even more difficult deciding
:Jo esia). -0 Wink @ Ope-
whether one is permitted to use them.
extend right arm, then
0> (D
Sidewise look S@L SIO\
down, up to head, and
era (Turkey, Persia). 0)1> Focus on auditor Q@L
esel on the back,
-0 side and slap ~ao Stare co
:-e -~ (Ba okas). ex:> Lip -
<§><§> Rolled eyes
- S.!2)
 64 • LANGUAGE ADO

anguage' sports players or officials can signal me state of play. 0-

gestural systems have evolved ro facilitate intention ro act in a certain way .

.inication in parricular situations, They are often entertainment a group of performers can coordinate their
ro as 'sign languages', bur few have developed any activities, such as acrobats, musicians.
of structural complexiry or communicative range, theatreslcinemas ushers can signal the number and
therefore important ro distinguish them from location of seats.
~!mage proper' - the natural signing behaviour of casinos officials can reporr on the state of play, or indicate
•. Parr VI). Several mighr properly be described as problems that mighr affect the participants in a game.
languages' (p. 58). sales/auctions auctioneers can convey the rype and amounr
=y pans of the world, such as India, Thailand, of selling and buying.
~, panromime and dance have come ro use aviation marshaLling ground staff can send information The bodily represen
systems of symbolic hand gesrures in association about the position of an aircraft, the state of its engines, the sound a, expfesSl::lg ex
_. ~,..,,~Iexpressions and body movements. The events of and its desired position. meaning of astonsh
deeper meaning, and the emotional states of the radio/television direction producers and direcrors can signal wonder, as recom e
_ ••••••••••.• .> may all be conveyed in this way. For example, ro performers the amount of time available, instructions eurythmy (Steiner, 1931, p -
. drata Ndtya-ldstra ('principles of dramatic art'), about level of loudness or speed of speaking, and This approach, develo
the founder of anthroposo
mury BC manual of Hindu dance, there are over informarion about faults and corrections.
Rudolf Steiner (JB61-1925),
""-'...'~",,".ue patterns for the hands (mudrtis). diving divers can communicate depth, direction,
aimed to promote a dose
- us or quasi-religious groups and secret societies time, and me nature of any difficulties rhey have
harmony between the sounds
develop rirual signing systems so that members encounrered. of speech and patterns of body
ognize and communicate with each other, Such truck driving drivers can exchange courtesy signals, give movement. Eurythmy was seen
are used in Freemasonry, practised by some 6 information about the state of the road, or show they as 'visible speech', With the body
o people mainly in the USA and Britain, and in are in trouble. reflecting in its physical shape
of me secret societies of me Far East, such as the heavy equipment drivers people conrrolling cranes, hoists, the forms of sounds as they are
clery. and other equipment can signal the direction and articulated. The different sounds
extent of movement. are interpreted symbolically
monastic orders developed signing systems
(§30). e.g. U IS seen as the
complexity; especially if their members were fire service can send directions about che
fire-officers
expression of something which
. silence, as in the case of the Trappist monks, supply of water, water pressures, and the use of
chills or stiffens, and this is
____ ••.•..rnenr of the medievalCistercian order. equipment. shown in the body by a pressing
_ ing systems are Found in a wide range of bookmaking bookies send signals abouc che number of a together of the arms and legs.
race or horse, and irs price (see below). According to Steiner, 'The entire
noisy conditions environmental noise may make verbal universe is expressed when the
communication impossible (e.g. rhe srock exchange) whole alphabet is repeated from
and a signing system may result. beginning to end.'

TALK
- guing sights at dog
rses in Britain is the
SIgning used to circulate
- e way bets are being
as an agent for a group
5 right hand on shoulder
6 sign 5 then 1
7 sign 5 then 2
8 sign 5 then 3
9 sign 5 then 4

~ '-+
__ ~ ~- have bought his 'twist 10 clap hands
., dogs or horses are given £5 right hand held up, palm outwards,
•c : se on the official fingers spread
No bet. 'I don't want it' Horse number two Each movement outwards
set of tick-tack signs is £ 10 both arms held up with fingers spread from the crossed position
titying the amount £50 clenched fists held together denotes £ 100

- g number, and the £100 left hand held up with fingers spread
only those who have £500 hands outline a circle
s card will be able £1,000 hands play imaginary piano
~ -!>ef refers to. ('grand piano' = 'a grand' = £1,000)

Some signs for odds

Evens arms held in front, moving up and
down
11/10 hands together, forming a pyramid Nine to four against Evens
I want to pay to lose
6/4 one right finger in the left ear-hole
 Linguistics

Language has been an objecr of fascinarion and a subjecr saw the reality of a name to lie in its formal p-
of serious enquiry for over 2,000 years. Often, the shape, irs relarionship to the real world being
observarions have been subjecrive and anecdoral, as people and indirect: 'no name exists by nature, b
reflecred on such topics as rhe narure of meaning, ideals becoming a symbol'.
of correcrness, and the origins oflanguage (§§ 1,49). Bur These firsr ideas developed inro [\"0

from the earliesr periods, rhere has also been an objecrive philosophical thought, which have since
approach, with scholars invesrigaring aspecrs of grammar, conventionalist and naturalistic. Modern linf
vocabulary, and pronunciarion in a derailed and organized poinred out that, in rheir extreme form
way. Ar the end of the 18rh century, the subjecr attracted view is valid (p. 105). However, various m
an increasing number of specialisrs (§50), so much so intermediate positions were also argued a
that it rapidly became possible to see the emergence of a much of the debate inspiring a profound inr
new field of scienrific research with language analysis as Greek language.
irs focus. This approach, firsr known as philology, dealt Another theoretical question was discussed
exclusively wirh the historical developmenr of language. Greece: whether regularity (analogy) or irre __ ~~
Plato (c. 427-c. 347 BC) In the presenr cenrury, the subjecr has broadened ro (anomaly) was a better explanation for the
Luna marble head, is: century include the whole range of subjecr marrer represenred in facts of Greek. In the former view, language w
AD rhis book, and it is now generally called linguistics (or be essenrially regular, displaying symmetries in
linguistic science). Linguisrics roday is a widely pracrised paradigms, and meanings. In the latter, arten
academic discipline, wirh several domains of applicarion focused on the many exceprions to these rules,
(p.432). existence of irregular verbs or the lack of corr
between gender and sex (p. 97). Modern lin
not oppose the two principles in this way: lane
EARLY HISTORY analysed wirh reference to both rules and excep
A religious or philosophical awareness oflanguage can be aim being ro understand the relationship be
found in many early civilizarions (p. 404). In parricular, two rather than ro deny the importance of ei
several of the important issues of language analysis were The historical significance of the debate is the
addressed by the grammarians and philosophers of it provided for detailed srudies of Greek
ancienr Greece, Rome, and India. grammar.
In the 3rd century Be, the Sroics establi
The Greeks formally the basic grammatical notions that ha
The earliesr surviving linguistic debate is found In the via Latin, become traditional in western rhougr
pages of Plaro (c. 427-347 BC). Cratylus is a dialogue grouped words into parrs of speech, org .
about the origins of language and the nature of meaning variant forms inro paradigms, and devised
- first between Socrares and Hermogenes, then between rhem (e.g. the cases of the noun). Dionysius Thra
Socrates and Cratylus. Hermogenes holds the view that BC) wrore the first formal grammar of Greek -
Aristotle (384-322 Be) language originated as a product of convenrion, so that that became a srandard for over 1,000 years.
National Archaeological the relationship between words and things is arbitrary: The focus throughout the period was
Museum, Athens, Greece 'for nothing has its name by nature, but only by usage on the written language. The word grammar
and cusrom'. Crarylus holds the opposire position, that grammatike) in fact originally meant 'the art of
language came inro being naturally, and therefore an Some attention was paid ro basic norions co
inrrinsic relarionship exisrs between words and rhings: the articulation of speech, and accenr marks w~_,_~
'there is a correcrness of name exisring by nature for to wriring as a guide to pronunciarion. Bur
everyrhing: a name is not simply that which a number inreresrs were in the fields of grammar and e .
of people jointly agree ro call a rhing.' The debate is rarher rhan phonerics. A doctrine of correc
conrinued at length, bur no firm conclusion is reached. stylistic excellence emerged: linguisric srandar
The latter posirion is more fully presenred, with divine set by comparison with the language of the
origin being invoked in supporr: 'a power grearer than writers (e.g. Homer). And as spoken Greek (
that of man assigned the firsr names to things, so that increasingly diverged from the lirerary standard.
they must of necessity be in a correct state.' By contrast, find the first argumenrs about the undesirable
Aristotle (384-322 BC) in his essay De interpretatione linguistic change (§1): the language had ro be p_
(,On interpretation') supporred the former viewpoint. He from corruption.
 65 • LINGUISTICS 429

Romans their belief was that cerrain relisious ceremonies, [Q be

writers largely followed Greek precedents and successful, needed [Q reproduce accuratelv" the oriainal
b

uced a speculative approach to language. On the form of these texts. Change VI no corruprion, as in
e, in their descriptive work on Latin, they used Greece, but profanarion. reral ancillary disciplines

. categories and terminology with little change. (Vedanga, 'limbs of the Vedas' • • 0" honerics,

er, the most influential work of the Roman period etymology, grammar, and metri . grew -0 overcome
-ed to be an exception to this trend: the codification this problem.
..atin grammar by Marcus Terentius Varro (116-27 Their solution was [Q establish the -

under the headings of erymology, morphology, language clearly and systematically and rh
• "'ntax. De lingua latina (,On the Latin language') an authoritative text. The earliest evi
," ed of 26 books, though less than a quarter of of this feat is the work carried our by the g~ranlffia...;'t.::l
survive. Varro's work takes into account several Panini, sometime between the 7th and - th cen

ences between Latin and Greek (e.g. the absence of in the form of a set of 4,000 aphoristic starernen
efinite article in the former). He also held the view as stitras (rhreads'). TheA$tadhyaYI (,Eight books' .
ich is remarkably modern) that language is first and mainly with rules of word formation, are compo
ost a social phenomenon with a communicative such a condensed sryle that they have required extensiv
se; only secondarily is it a tool for logical and commentary, and a major descriptive tradition has in
ophical enquiry. been established. The work is remarkable for its detailed
Especially towards the end of the millennium, several phonetic descriptions: for example, places of articulation
ors wrote major works in the fields of grammar are clearly described, the concept of voicing is introduced
- rhetoric (§12) - notably, Cicero (106-43 BC) on and the influence of sounds on each other in connected
e, and Quintilian (Lst century AD) on usage and speech is recognized (the notion of sandhi). Several
lie speaking. Julius Caesar wrote on grammatical concepts of modern linguistics derive from this tradition.
=:rlariry - it is said, while crossing the Alps on a
racy campaign. Aelius Donatus (4th century AD)
re a Latin grammar (Ars maior) that was used right
- the Middle Ages, its populariry evidenced by the
that it was the first to be printed in wooden rype,
had a shorter edition for children (the Ars minor).
the 6th century, Priscian's Institutiones grammaticae
~matical categories') was another influential work
continued to be used during the Middle Ages: it
rains 18 books, and remains the most complete BOOK SlXTIt.
_ ammar of the age that we have.
CIIAP1'BB FIIUI'l\
The main result of the Roman period was a model of
zmmarical description that was handed down through
y writers in Europe, and that ultimately became
t~'t' ~.~ •••:,.,~,
"'" II~ .~tft"'" ~"'~ '" ~ ~ I
basis of language teaching in the Middle Ages and IIQIIPII •••_.R4d\ •••ltUl~ ~ l1l1nt tift "tAb"'" •~ fiIItt
Renaissance. In due course, this model became the .iftI"RR_ •"" .41(~44(""~ IIq""'i~ l: ~I II
.tiona!' approach to grammar, which continues to 1. .In the room of the first portion, containing a
cise its influence on the teaching of English and
"Ilgio vowel, there are two.
er modern languages (§§ 1,62).
Upto SBtta VL I. 12 inclusive it is tG be understood that for the fi
syllahle two ale to be made, i. e, the first syllable is to be reduplicated,
e Indians an adhiklra sttta: ait the three ords via ~:, and \: and lI1AfII are
.rring the above period, techniques of minute descriptive fad in the subsequeat sBtras Ilpto VI. r. 12, before the rule of Sampr;aa~1&
ysis were being devised by Indian linguists, which begins. Thus SQtra VI. t. 8 says· when fiIlt. follows, of an unredupJica
d have been of great influence had these descriptions •.oot ". The sense of this sBtra is incomplete, unless we spy
ched the western world (something that did not take words of this sBtra, when it will read thus: ••when ~ folio
ce until the 19th century). The motivation for the ia the room of the first portion, containing a single vowe
cated verbal root". Tl]us from "IIf1l:-1f1II1Tf( (1fPl+atl!.IV. 3- z-W+1111f+lr-
'an work was quite different from the speculative
~+1IWIt+" VII. 4- 59-"C1IP1R "he awoke"). So a
rrers that attracted Greek and Roman thinkers tbe roots ~. t and • • The reduplication takes plsee
, ough they did not ignore those topics). The Hindu affix that follows. Thus the affixes of the Perfect T'-- "••...••
iests were aware that their language had diverged from tive (VI('. the Intensive (q). the ,.. vikara,a,
mar of their o/desr sacred rexr.$.; roeT1:das 4'tJ{) in rP redu. o/ka/hn.
tb. pronunciation and grammar. An irnporranr parr of
An excerpt from the AffrJdhyoyl 7!ofls/oled 1Il."() E :1
430 PART I· l GUAGE AND COMMUNICATION

The Middle Ages the role of innate ideas in the development of thought The 'first gramma .
Very lircle is known abour me development of linguistic and language. Such ideas provided the basis of certainty The Prose Edda is a ~
ideas in Europe during me 'Dark Ages', though it in knowledge, according to Cartesian philosophy, but century textbook on
is evident mat Latin, as me language of education, their existence was denied by philosophers (such as style and construcnor
Locke, Hume, and Berkeley) for whom knowledge by the Icelandic chie
provided a continuity of tradition between me classical
Sturluson. Appended -
and medieval periods. Medieval learning was founded derived from the way the mind operated upon external
manuscript are four -
on seven 'arts', of which three - grammar, dialectic, and sense impressions. The issue was to resurface in the 20th
on grammar, written -
rhetoric - formed one division, known as the trivium. century (p. 433). 12th century, the first
Grammar (mainly using Priscian and Donatus) was Several other important trends have been noted has attracted specia
seen as me foundation for the whole of learning. A during me l Zth and l Sth centuries: the breakdown of because of the origina
tradition of 'speculative' grammars developed in me Latin as a universal medium of communication, and thought. The authorsr _
13m and 14th centuries, in which grammatical notions its replacement by modern languages (§59); the many 'First Grammatical Ires
proposals for universal languages, shorthand systems, and unknown, but the wr:
were reinterpreted within me framework of scholastic
secret codes (§§33, 58); the beginnings of a systematic come to be known as
philosophy. The authors (the 'Modistae') looked [Q
approach to phonetics (§27); the development of grammarian'.
philosophy for the ultimate explanation of the rules of
'general' grammars, based on universal principles, such This early exercse
grammar. A famous quotation from the period states
spelling reform (p. L::
mat it is not the grammarian but 'the philosopher [who) as the l Zth-century grammar of Port Royal (§14); and
summarizes and ill
discovers grammar' (philosophus grammaticam inueniti. me major elaborations of traditional grammar in schools
principles needed
The differences between languages were thought ro be (§1). Then, as the 19th century approached, the first
the use of the latin ~
superficial, hiding the existence of a universal grammar statement about the historical relationship between for writing Old Ice/a--
(§14). Sanskrit, Greek, and Latin was made, ushering in the the only work of th :
The Middle Ages also saw the development of western science of comparative philology (§50). to draw attention to .
lexicography (§ 18) and progress in the field of translation, problems involved -
as Christian missionary activity increased. In the East, Latin letters to a ve:.;
language. It conta
Byzantine writers continued ro expound the ideas of the
acute phonetic 0
Greek authors. There was a strong tradition of Arabic
and, in its emphas:s
language work related ro the Qur' an (which was not to be
symbols to express
translated, P: 404). From around the Sth century, several contrasts, anticipates
major grammars and dictionaries were produced, as well of 20th-century p
as descriptive works on Arabic pronunciation. For a theory (§28).
long time, these remained unknown in Western Europe. A translation 0'
Opportunities for contact with the Greek, Arabic, and opening page (Haw
Hebrew linguistic traditions only came later, as a result / have written an a _-
of the Crusades. us Icelanders also, -
it might become eas_
The Renaissance and read ... / have -
The rediscovery of the classical world that came with Latin letters that
the 'revival of learning', as well as the discoveries of me our language we
New World, transformed the field of language study. rightly pronounced _
Missionary work produced a large quantity of linguistic some other letters --
needful to me, wP -
material, especially from the Far East. The Chinese
taken out that did
linguistic traditions were discovered. Arabic and Hebrew sounds of our lang; -
studies progressed, the latter especially in relation to of the consonants
me Bible. In the 16th century, several grammars of alphabet were rei -
exotic languages came to be written (e.g. Quechua in new ones added. -
1560). There was a more systematic study of European rejected, but a gooc
languages, especially of the Romance family. The first added, since our la-
greatest number 0
grammars of Italian and Spanish date from the 15th
century. Major dictionary projects were launched in
many languages. Academies came into being (p. 4). The
availability of printing led to me rapid dissemination of
ideas and materials.
As we approach modern times, fresh philosophical
es emerged. The l Sth century is characterized by the A page from a 14th-century manuscript of Snorri Sturluson's
" ems between 'rationalists' and 'empiricists' over Prose Edda
 65 • LINGUISTICS 431

TlETH-CENTURY LINGUISTICS SAUSSUREAN PRINCIPLES

growth of modern linguistics, from the end of the
Some of Saussure's most central possible to assess the state of the signifies', or 'sound image') and
century to the present day, has in large parr already ideas were expressed In the game by studying the position of signifie (the thing signified', or
summarized in earlier sections of this volume. form of pairs of concepts: the pieces on the board. 'concept'). This relationship of
:najoriry of the concepts used in the discussion of Longage vs langue vs parole signified to signifier Saussure
_ ze history, acquisition, structure, substance, and .•••e rna senses of the word calls a linguistic sign. The sign is
C
m from this perspective (reflecting the background a age' promp ed Saussure the basic unit of communication
author). However, there remain several loose ends ; e a reefold set of within a community: langue is
aistorical and theoretical nature that need to be A I B --e ich ere seen as a 'system of signs'.
together in this final part of the book. - - g. Langage

·Jpe and America

~
, speech esen
••• e- ••1",., ,..,',
~~
Symagmatic
~
o he + can + go
ain approaches to language study, one European,
American, unite to form the modern subject of Diachrony vs synchrony
ics. The first arises our of the aims and methods He sharply distngushed
historical (diachronic) and
-~ He run
rh-century comparative philology (§50), with its
I Sit
on written records, and its interest in historical non-historical ('synchronic') c, \ou see
. and interpretation. The beginning of the 20th approaches to language study. , ere. ere.
The former sees language as a
ry saw a sharp change of emphasis, with the
continually changing medium; Syntagma -cad associative
of the principles governing the structure of living
the latter sees it as a living whole, (or paradigmatic)
~ zes being introduced by the Genevan linguist
existing as a 'state' at a particular A sen enee IS a seqcerre 0'
::...:laI1dde Saussure (1857-1913). Saussure's early moment in time. In his diagram, signs, each sign contrib ng
was in philology, bur he is mainly remembered AB represents a synchronic 'axis something to the meaning 0
theoretical ideas, as summarized in the Cours of simultaneities' - a language the whole. When the signs
istique generate (,Course in general linguistics'), state at some point in time; CD is are seen as a linear sequence,
is widely held to be the foundation of the modern a diachronic 'axis of successions' the relationship between
. This book was in fact published posthumously - the historical path the language them is called syntagmatic,
6, and consists of a reconstruction by two of has travelled. as in She + can + go. When a
.:re's students of his lecture notes and other In this view, it is always sign that is present is seen as
necessary to carry out some contrasting with other signs in
degree of synchronic work Concept the language, the relationship
e second approach arose from the interests and
before making a diachronic is called associative (in later
cupations of American anthropologists, who were Sound image
study: before we can say how studies, paradigmatic), as in
rned to establish good descriptions of the American a language has changed from She vs He, can vs will,go vs run
languages and cultures before they disappeared. state X to state Y, we need to in the above sentence. These
ere were no written records to rely on, hence know something about X and Y. two dimensions of structure
analysis was ruled our. Also, these languages Correspondingly, a synchronic Signifiant vs signifie can be applied to phonology,
red very different kinds of structure from those analysis can be made without Saussure recognized two sides vocabulary, or any other aspect
rered in the European tradition. The approach referring to history. Saussure to the study of meaning, but of language. The result is a
efore to provide a careful account of the speech illustrates this point using an emphasized that the relationship conception of language as a vast
of the living languages. A pioneer in this field analogy with a game of chess: between the two is arbitrary (p. network of interrelated structures
if we walk into a room while a 428). His labels for the two sides and mutually defining entities - a
ranz Boas (1858-1942), who published the first
chess game is being played, it is are signifiant ('the thing that linguistic system.
of the Handbook of American Indian Languages
- 1. Ten years later, another anthropologically
book appeared: Language by Edward Sapir
1939). These works proved to be a formative
on the early development of linguistics in
The new direction is forcefully stated by Boas
_ 'we must insist that a command of the language
dispensable means of obtaining accurate and
~ -;:_
..•.•
knowledge, because much information can be
by listening to conversations of the natives and
g part in their daily life, which, to the observer
no command of the language, will remain
- inaccessible'.
Ferdinand de Saussure (1857-/9/3)
432 PART XI· LANGUAGE
Roman Jakobson (/896-
1982) Jakobson was one of the
founders of the Prague School
of linguistics and 0 major
contributor to many fields of
study in Slavonic languages
and general linguistics. In 1941
he moved to America, where
he held professorial posts
at Columbia and Harvard
between 1946 and 1967.
Louis Hjelmslev (/899-1965)
The leading theoretician
of the Copenhagen School
of linguistics, Hjelmslev
propounded a formal approach
to language study in the 1930s
known as glossematics.
AND COMMUNICATION
Later developments SCHOOLS OF THOUGHT
Both European and American approaches developed
Many different approaches to linguistics emerge<i -
rapidly. In Europe, Saussure's ideas were taken up by
oecaoes 01 t'ne \wentiet'n century, some 01 which -
several groups of scholars (especially in Switzerland,
great deal of support. The distinguishing features
Czechoslovakia, France, and Denmark),
of thought emerged based on Saussurean principles
and schools
approaches is outlined below. (For corresponding -=_.-~~~
within the field of generative linguistics, which has -
(notably, the Linguistic Circle of Prague, which was since the 1960s, see p. 433.)
founded in 1926). The field of phonology (§28) was the
Functional sentence perspective
first to develop, with later progress coming in such areas
An approach used by the Prague School of linguists
as grammar and style. Saussure's influence continues to
utterances in terms of their information content, and
be strong today, with his notion of a language 'system'
used in the Czech Republic and other countries of East~
becoming the foundation of much work in semiotics and The semantic contribution of each major element in a se
Structuralism (pp. 81,423). rated with respect to the 'dynamic' role it plays in co
In America, the development of detailed procedures
Dependency grammar
for the study of spoken language also led to progress in
phonetics and phonology, and especial attention was paid A type of formal grammar developed in the 19505
the French linguist Lucien Iesniere (1893-1954). "_
to the distinctive morphology and syntax (§16) of the
grammatical relationships by setting up 'depende _
American Indian languages. The first major statement
'valencies') between the elements of a constructior
synthesizing the theory and practice of linguistic analysis
was Language by Leonard Bloomfield (1887-1949), Tagmemics
A theory developed since the 1950s by the America-
which appeared in 1933. This book dominated linguistic
thinking for over 20 years, and stimulated many Pike (1912-2000), which focuses particularly on the - _
linguistic 'forms' and 'functions'. A central notion is .-
descriprive studies of grammar and phonology. In due
between the 'emic' units, which are functionally cor
course, the Bloomfieldian approach came to be called
language (such as phoneme and morpheme), and -_
'structuralist', because of the various kinds of technique
that give them physical shape (d. phonetics, §28).
it employed to identify and classify features of sentence
structure (in particular, the analysis of sentences into Stratificational grammar
their constituent parts, p. 100). It also represented a A theory devised by the American linguist S. M. Lar--
in the 1960s that views language as a system of r
behaviourist view of linguistics, notably in its approach
('strata') of structure.
to the study of meaning (p. 105). However, its appeal
diminished in the 1950s, when there was a sharp reacrion Systemic linguistics
against the limitations of structural linguistic methods, A theory developed since the 1960s by the British -
especially in the area of grammar (p.l 00). K. Halliday (b.1925) in which grammar is seen as a-
This extract from an obituary of Bloomfield, written 'systems' of interrelated contrasts; particular atten -
the semantic and pragmatic aspects of analysis (§§ - _
by Bernard Bloch in the journal Language (1949, p. 93),
to the way intonation is used in the expression 0 c.
summarizes rhis scholar's achievement:
There can be no doubt that Bloomfield's greatest John Rupert Fi
contribution to the study of language was to make a (left), Professor a r_
science of it. Others before him had worked scientifically Linguistics in the ~ - _
in linguistics; but no one had so uncompromisingly London from 194L-
rejected all prescientific methods, or had been so a key figure in the " .•".c:---
consistently careful, in writing about language, to use of British linguistics. -
terms that would imply no tacit reliance on factors beyond notion in his app«
the range of observation ... It was Bloomfield who taught patterns of languo<;=
us the necessity of speaking about language in the at a particular lev<
style that every scientist uses when he speaks about (§13) cannot be
the object of his research: impersonally, precisely, a single analytic
and in terms that assume no more than actual systems may need '-
observation discloses to him. at different places. -
Bloomfield's opposition to unscientific handle the range r
involved (an appaxx: j-'-
impressionism in language studies is nearly
polysystemicism).
summarized by the wry comment he made
on one occasion: 'If you want to compare two
languages, it helps to know one of them!'
Leonard Bloomfield
Chomsky ro make a conrriburion ro our understanding of the nature
In 1957, Avram Noam Chomsky (b.1928), Professor of of the human mind.
Linguistics at the Massachusetts Institute of Technology, The essence of the approach is summarized by
published Syntactic Structures, which proved ro be a turning Chomsky in a 1986 book (Knowledge of Language, p. xxvi)
point in 20th-century linguistics. In this and subsequent as providing an answer ro the question 'How comes it that
publications, he developed the conception of a generative human beings, whose contacts with the world are brief
rrammar (p. 101), which departed radically from the and personal and limited, are nevertheless able ro know as
suucturali m and behaviourism of the previous decades. much as they do know?' By studying the human language
Earlier analyses of sentences were shown ro be inadequate facu Iry, it should be possible ro show how a person
ill various respects, mainly because they failed ro take inro constructs a knowledge system out of everyday experience,
ccount the difference between 'surface' and 'deep' levels and thus move some way rowards solving this problem.
f grammatical structure. At a surface level, such sentences A major feature of Chomsky's approach was the

John is eager to please and John is easy to please can be technical apparatus he devised ro make the notion of Noom Chomsky
malysed in an identical way; but from the point of view competence explicit - the system of rules and symbols
f their underlying meaning, the two sentences diverge: in that provides a formal representation of the underlying
:.he first, John wants ro please someone else; in the second, syntactic, semantic, and phonological structure of
someone else is involved in pleasing John. A major aim of senrences (aspects of this apparatus have been referred
~ nerarive grammar was ro provide a means of analysing ro in earlier sections, pp. 101, Ill, 170). A primary
ntences that rook account of this underlying level of norian - the transformational rule (p. 101) - led ro the
o;rrucrure. approach being commonly referred ro as transformational
To achieve this aim, Chomsky drew a fundamental grammar (or TG). Since the 1950s, much of linguistics
. tinction (similar ro Saussure's langue and parole) has been taken up with proposals ro develop the form of
tween a person's knowledge of the rules of a language generative grammars, and the original theory has been
d the actual use of that language in real situations. The reformulated several times. During the same period, also,
- t he referred ro as competence; the second as performance. there have been several proposals for alternative models of
..rnguistics, he argued, should be concerned with the grammatical analysis ro those expounded by Chomsky and
rudy of competence, and not restrict itself ro performance his associates, some of which have attracted considerable
- something that was characteristic of previous linguistic support. As a consequence, linguistic theory, the core of
dies in their reliance on samples (or 'corpora') of speech scientific language study, is now a lively and controversial
.g. in the form of a collection of tape recordings). Such field .
..amples were inadequate because they could provide only
SOME THEORETICAL CONSEQUENCES OF GENERATIVE THEORY
tiny fraction of the sentences it is possible ro say in a
guage; they also contained many non-Huencies, changes Since the 1960s, several fresh structure within a generative alternatives to an abstract,
-plan, and other errors of performance. Speakers use their theoretical approaches to grammar. Further levels of formal approach to grammar.
grammatical analysis have phrase structure are recognized, This particular theory adopts
:ompetence ro go far beyond the limitations of any corpus,
emerged, most of which can and distinguished using different a pragmatic view of language
. being able ro create and recognize novel sentences, and
be seen as a development of numbers of bar symbols. as social interaction (§21 a d
- identify performance errors. The description of the
Chomsky's proposals - or as a Montague grammar sets up 'functional' units 0 a
-ules governing the structure of this competence was thus
reaction against them. This approach derives from pragmatic and syntactic
e more important goal.
Case grammar the work of the American within sentence struaure.
Chomsky's proposals were inrended ro discover
The American linguist, Charles logician Richard Montague Realistic grammar
- e menral realities underlying the way people use Fillmore (b. 1929) devised a (1930-70), and is based on the Grammatical analyses Qv)rllr' b=
.anguage: competence is seen as an aspect of our general theory which focuses on the study of logical languages. A 'psychologically real', ac
. chological capaciry. Linguistics was thus envisaged as semantic roles (or 'cases') close correspondence is set up to this approach. For.:1-
mentalistic discipline - a view that contrasted with the played by elements of sentence between the categories of syntax patterns should be ~ ~
havioural bias of previous 20th-cenrury work in the structure. and semantics. the psychologJc2.lL:';c-~:::.:!i
underf e I!l' 19rns,xoe:;a."!:U
bject, and connected with the aims of several earlier Relational grammar Generalized phrase structure
such as (O;~:-::J!"E~~g:::'
guists (such as the Port-Royal grammarians, p. 86). It This approach views grammatical grammar
This theory does not recognize m
'as also argued that linguistics should not simply limit relations (e.g. 'subject', 'object')
(Self ro the description of competence. In the long as central rather than the the role of transformations In
formal categories (e.g. 'noun a generative grammar. Instead,
zerm, rhere was a sriIl more powerful rarget: ro provide a
phrase', 'verb phrase') of earlier it focuses on developing the
pammar capable of evaluaring the adequacy of differenr
geflerative theory (§16). pl!rase strudure dimension •
aunts of competence, and of gojng beyond {he study
X-bar (X) theory grammatical analysIS
f individual languages ro rhe nature of human language
The theory provides an Functional gra
a whole (by discovering 'linguistic universals' (§§14,
alternative account of phrase Several approac-
». In this way, ir was hoped, linguisrics would be able
434 PART XI· LANGUAGE AND COMMUNICATION

A cautionary tale OBTAINING LINGUISTIC DATA the speakers being aware of the fact - a proc
The or-ar;a eda dwe obtains very natural data, though ethical objecci
sta red do you say Many procedures are available for obtaining data
be anticipated. Alternatively, attempts can be -
age?' The Indian about a language. They range from a carefully planned
make the speaker forget about the recording,
ow' ,-,-"",,:or a ile. First he intensive field investigation in a foreign country to casual
keeping the audio recorder out of sight, or
ec C ill, then he looked out. introspection about one's mother tongue carried out in
microphones. A useful technique is to introduce
So is face lit up as if struck an armchair at home. that quickly involves the speaker, and stimulates ••. _
a sudden flash of inspiration.
language style (e.g. asking older informants to
e spo e very rapidly. If I had
Informants how times have changed in their locality).
been able to transcribe what he
In all cases, someone has to act as a source oflanguage data- An audio recording does not solve all the
said, it would have spread across
the page several times, I gulped an informant, or consultant. Informants are (ideally) native problems, however. Speech is often unclear or amz
and bravely started to write; but speakers of a language who provide utterances for analysis Where possible, therefore, the recording has
after a few syllables, I was already and other kinds of information about the language (e.g. supplemented by the observer's notes about -
hopelessly bogged down. 'How did translations, comments about correctness, or judgments verbal behaviour of the participants, and
you say that?' With his repetition on usage). Often, when studying their mother tongue, context in general. A facial expression, for ex
I added two more syllables, then linguists act as their own informants, judging the ambiguity, dramatically alter the meaning of what is said -
bogged down again, When I acceptability, or other properties of utterances against their Video recordings avoid these problems to a large
asked for the third repetition, the own intuitions. The convenience of this approach makes but even they have limitations (the camera can
informant began to waver and
it widely used, and it is considered a primary datum in the intrusive, and cannot be everywhere), and tr'aIl!~;:==-
finally to change his story, and so
generative approach to linguistics (p. 433). But a linguist's always benefit from any additional commentary -_-- •••
I had to give up entirely, To my
personal judgments are often uncertain, or disagree with by an observer (p. 241).
self-justifying and half self-accusing
'But that surely doesn't all mean the judgments of other linguists, at which point recourse is
just I run', he said, 'Why of course needed to more objective methods of enquiry, using non- Elicitation
not. It means I was sitting here linguists as informants. The latter procedure is unavoidable Linguists also make great use of structured
with you; then I looked out of the when working on foreign languages, or in such mother- in which they systematically ask their info
door and saw a deer, so I quickly tongue fields as child speech (§38) or language variation utterances that describe certain actions, ob
grabbed my spear and now I am (§§8-12). behaviours. With a bilingual informant, or -
running after it.' Then, almost Many factors must be considered when selecting the use of an interpreter, it is possible to use rra2:=::_
philosophically, he added to informants - whether one is working with single speakers techniques ('How do you say table in your 1 -
himself, 'Only a fool would run for
(a common situation when languages have not been 'What does gua mean?'). A large number of poin
nothing.' (Loewen, 1964, p. 189.)
described before), two people interacting, small groups, covered in a short time, using interview wor .
or large-scale samples. Age, sex, social background, and questionnaires. Often, the researcher wishes
A reverse lexicon
An extract from the Brown other aspects of identity are important, as these factors information about just a single variable, in will
University Corpus listing words in are known to influence the kind of language used (Part restricted set of questions may be used: a particular
reverse alphabetical order. II). The topic of the conversation and the characteristics of pronunciation, for example, can be elicited
REDEMPTION of the social setting (e.g. the level of formality) are the informant to say a restricted set of words. -=-
EXEMPTION also highly relevant, as are the personal qualities of the also several indirect methods of elicitation, such
GUMPTION informants (e.g, their fluency and consistency). For informants to fill the blanks in a substitution fr.
RESUMPTION larger studies, scrupulous attention has to be paid to the I - see a car), or feeding them the wrong stim
PRESUMPTION sampling theory employed. And in all cases decisions correction (Ts it possible to say I no can see?')
CONSUMPTION have to be made about the best investigative techniques
ASSUMPTION to use. Corpora
OPTION
A representative sample of language, compi!
ADOPTION
SORPTION Recording purpose of linguistic analysis, is known as a
ABSORPTION Today, data from an informant are usually audio recorded. corpus enables the linguist to make objective sr;c=:::II
ERUPTION This enables the linguist's claims about the language to about frequency of usage, and it provides acc...",,,,-~,,·.;:~-
I TERRUPTION be checked, and provides a way of making those claims for the use of different researchers. Its range an
CORRUPTION more accurate (,difficult' pieces of speech can be listened variable. Some corpora attempt to cover the
ors UPTION to repeatedly). But obtaining naturalistic, good-quality as a whole, taking extracts from many kinds
::SERTION data is never easy. People talk abnormally when they others are extremely selective, providing a UJO. _

I S:: "'10 know they are being recorded, and sound quality can be of material that deals only with a particular ••~~-~.
:~-IO
poor. A variety of recording procedures have thus been feature. The size of a corpus depends on
devised to minimize the effects of the 'observer's paradox' factors, such as the time available to collect ~
(how to observe the behaviour of people when they are and store the data: it can take up to several -
not being observed). Some recordings are made without provide an accurate transcription of a few
 65 • LINGUISTICS 435

OMPUTER CORPORA
~ standard' corpus is a large
:... ection of data available
r use by many researchers.
=-glish linguistics now has
- eral computer corpora, all in
- chine-readable form, and thus,
Jlincip/e, available anywhere in
rid.
The London-Lund Corpus of
Spoken English This corpus of
educated spoken British English
consists of the spoken material
collected as part of the Survey of
English Usage (see below right).
The data consist of 87 texts of
5,000 words each. It was
transferred to computer tape in
the 1970s at the University of
Lund, and is also partly available
in printed form. In addition to the
running text, a lexical concordance
has been compiled.
The Brown University Corpus ICAME There is now a clearing
of American English This corpus centre for storing and distributing
is drawn from USA printed sources information on corpus studies in
published in 1961. It comprises English: the International Computer
500 samples of about 2,000 words Archive of Modern English (lCAME),
each representing 15 main varieties based at Bergen University. Its
of the language. It is available aims are to compile an archive of
via electronic files, printout, and English-language material available
microfiche. Apart from the running for computer processing, and to
text, there are lexical concordances, ec: and distribute information
word frequency lists, and a reverse research that uses this material.
alphabetical lst

(p. 241). Sometimes a small sample of data will methods for eliciting judgments about sentences or
ough to decide a linguistic hypothesis; by contrast, the elements they contain. Informants can be asked to
ra in major research projects can total millions of identify errors, to rate the acceptability of sentences to
g words. An important principle is that all corpora, make judgments of perception or comprehension, and
rever their size, are inevitably limited in their coverage, to carry out a variety of analytical procedures.
ays need to be supplemented by data derived from
tuitions of native speakers of the language, through Reconstruction
introspection or experimentation. The limiting case of linguistic study, one might
imagine, is when no data are available at all - as in the
Origin in writing 1(0)
case of the historical study of language where written Pnn ed 46
ental techniques are widely used in linguistics, records are lacking. But it is possible to break through Learned arts 6
y in those fields that have been inBuenced by the even this apparent barrier, by using the 'reconstruction' Learned sciences (7,
InstructJona (6)
of sciences where experimentation is routine. techniques of comparative philology (§50). The forms
Press: genera news ( ,
ics (§24) is the subject most involved in this of Proto-Indo-European and other reconstructed Press: specific reporung (4)
ch, but experimental testing is also common in languages may be totally hypothetical in status, but Administrativejoffiaal (4)
other areas, such as child language acquisition (§38) they have nonetheless become a major field of linguistic Legal and statutory (3)
Persuasive writing (5)
ruage pathology (§46). In grammar and semantics, enqUiry. Prose fiction (7)
ental studies usually take the form of controlled Non·printed (36)
Continuous writing: imaginative (5),
informative (6)
GING A TEXT Social letters: intimate (6), equal
(3), distant (4)
-' the operations that a computer can between such superficially identical items CD that used as subordinator Non·socialletters: equal (4),
on a corpus are linguistically trivial, as bear (animal) and bear (action), or the NP proper noun distant (4)
-ney save an enormous amount many different syntactic functions of that. RA personal pronoun, subject Personal journals (4)
0=. e.g. listing of words in frequency Larger constructions (such as different RN personal pronoun, object As spoken (18)
Drama (4)
alphabetical order). More kinds of clause, p. 99) can also be tagged, VA + 0 main verb, past tense
Formal scripted oration (3)
g is the possibility of automatically to facilitate the retrieval of grammatical VA + G main verb, -ing form
Broadcast news (3)
- the structure of the corpus, from information. VA + N main verb, past participle Talks: informative (4), ma
tical, semantic, or phonological Two tagged sentences from the London- VB + 5 was form of to be (2)
_. ew (§26). This is the aim of Lund corpus are given (from Svartvik et 01., *VH + 0 contracted form of have, present Stories (2)
anent research programmes. 1982, p. 57). Abbreviations are as follows tense Origin in speech (l(0)
ep is to provide an automatic (other symbols refer to suprasegmental VA + 0 main verb, base form Monologue (24)
-- <agging' each word in the corpus features of pronunciation, §29): (See further, Svartvik & Quirk, 1980, from Prepared (but
that indicates its word class (6)
which the classification of Survey of English Spontaneous .oo;;:x: ,(!:.]
- enables the user to distinguish Usage texts (right) has been taken.) Spo ta eous COO:::e:1i!rtc
(t) - -
Ora .f-~
51. ~_
0101000563 B I<RA> II knew <VA + 0> that <CD> he <RA> was <VB + 5> evoming. <VA + G>

0101000564 B I've <RA*VH + 0> II heard <VA + N> Stan <NP> 'Carter <NP> m-ention <VA + 0> him. <RB>
436 PART XI • LA GUAGE AND COMMUNICATION
THE COMPUTER AGE
The use of the computer to foster corpus analysis (p.
435) is but one of many ways in which the study of
language is changing in response ~~ the availa~iliry. o~
computational techniques. In aUU\t\on to tne routme
use of computers in such areas as numerical counting,
statistical analysis, and pattern matching, linguistics
provides a range of opporrun ities for the manipulation of
non-numerical data, using natural language texts. Some
of these tasks are fairly elementary, such as indexing and
concordancing. Others are far more complex, such as the
computational perspective required to facilitate speech
recognition and synthesis (§26), machine translation (p.
as the processing of statistical data in author
(p. 70).
The field of LP emerged out of
translation in the 1950s (p. 360), and was I
influenced by work on artificial intelligence.
a beus on o.ev·\smg \me\\\%et\t 'P1og1'dmS ~ •
systems') which aimed to simulate aspects 0:
behaviour, such as the way people can infer
from what has been said, or use their knowledg
world to reach a conclusion (p. 361). A more spe
linguistic contribution has since been in eo
involving detailed syntactic and semantic
often on a much larger scale than hitherto,
a variery of linguistic theories. Most recently.•1-'_~ •••

"HELLO,"Robin typed on the
keyboard.
"HELlO, HOWAREYOUFEELING?"
replied ELIZA,the printed
words appearing instantly
on the VDU.
"I'M FEELINGTERRIBLE,"Robin
replied.
"WHATDOYOUMEANBYTHAT
PRECISELY?"
"I'M FEELINGDEPRESSED."
"WHYDOYOUTHINKYOUARE
DEPRESSED?"
"BECAUSEI'M LONELY."
"DO YOULIVEWITHYOURFAMILY?"
"NO."
"DO YOUHAVEANYFRIENDS?"
"NOT REALLY
"WHYISTHAT?"
Robin Dempsey paused for
some considerable time
before he typed in his reply:
"I DON'TKNOW."
"DO YOUREALLY NOTKNOW
ORAREYOUJUSTAVOIDINGTHE
QUESTION?"
Robin Dempsey flushed and
typed rapidly: "I USEDTOHAVE
FRIENDSONCE."
"WHENWASTHAT?"
"BEFOREIWASMARRIED."
"YOUDIDN'TTELLMEYOUWERE
lED."
"I' DMJRCED."
"lEU MEABOUTYOURDIVORCE."
n Dempsey typed
__ or fifteen minutes
opplng ...
'~.'-.c o.",,~e.Small World,
360), and language learning (p. 397). attention has been paid to the nature of discoun
Since the 1980s, the chief focus of computational sense of text beyond the sentence) and there
linguistic research has been in the area known as natural confrontation with the vast size of the lexicon.
language processing ( LP). Here, the aim is to devise large amounts of lexical data now available in
techniques which will automatically analyse large readable form from commercial dictionary pro
quantities of spoken (transcribed) or written text in Progress has been considerable, but
ways broadly parallel to what happens when humans programs are still experimental in character
carry out this task. LP deals with the computational dealing with restricted tasks in well-defined
processing of text - both its understanding and its There is still a long way to go before computer
generation - in natural human languages. It thus forms can get anywhere near the flexible and creati-
a major part of the domain of computational linguistics; of real conversation, with its often figurative 4~==
but it is not to be identified with it, as computers can and ill-formed construction. The technology is :
also be used for many other purposes in linguistics, such of the linguistics.
COMPUTER CONVERSATIONS
SHRDLU block which is taller than the one I program has merely ana lysed H's E: Earlier you said you
SHRDLU (a name based on am holding. OK. input sentence, and used this to having trouble with the c
the second half of the 12 most H: What does the box contain? randomly trigger a prompt from a H: That's right, and it has.~
frequent letters in the alphabet S: The blue pyramid and the blue small range of potentially relevant any better.
according to Morse, p. 90) was block. responses, often using various
E: What does that sugges:
written in 1972 by Terry Winograd. The second exchange identifies lexical relations - computer
H: That computers are
It shows some of the features a problem, as there were three relates to machine, for example.
E: How would you react
of conversation which can be pyramids in the scene, and it was 'Empty' utterances provoke
simulated on a computer. This not obvious which one H meant. machines took over?
equally 'empty' responses.
system operated in a very small The third exchange shows that H: I think it would be a d
The performance is
world - a box, a table, blocks, SHRDLU detected a possible E: Does anyone else thl
nonetheless convincing. as a
and pyramids - which could be ambiguity in the word it. simulation - and people have
would be a disaster?
manipulated according to the (After McTear, 1987.) sometimes reacted to it in H: Yes. Quite a lot of p
human speaker's (H) instructions. unexpected ways. The author E: Quite a lot of people.
The computer (S) could answer ELIZA H: That's right.
reports that his secretary once
questions about its actions and By contrast, Eliza, devised by E: Are you prepared to
asked him to leave the room so
could assimilate new information, Joseph Weizenbaum in 1966,
that she could talk with ELIZA in H: No. I have to go.
but it could not change the topic and named after the heroine of
private! And Robin Dempsey, in E: What do you really th
H: Pick up a red block. Pygmalion, was programmed David Lodge's Small World, uses me?
S: OK. to converse about any subject -
it as a confidante (see left). H: Not much.
H: Grasp the pyramid. albeit by pragmatically cheating.
S: I don't understand which The extract is from a mock E: Good day. What is your E: Why not much?
pyramid you mean. psychotherapeutic interview, in problem? H: Because you are just a -
H: Find a block which is taller than which the computer has the part H: I am having trouble with the computer program.
the one you are holding and put of the doctor. The system asks computer. E: Perhaps the real reasor
it in the box. apparently relevant questions, E: Do machines worry you? to talk about.
S: By 'it' I assume you mean the but this is a false impression. The H: This one does ... H: Well, you have a point
 5

E LO, DAVE ...

BOOKING A FLIGHT
an's] movement in the field of view must have triggered
GUS (the Genial Understander G Air California flight 310 that it infers Irorr --e f'~
5OOI'-X';:
~ ing in the unfathomable mind that was now ruling over
System) was a system devised leaves at 8.30 am and arrives at likely departure "~:_ ••- u-;;."'-'-
- -- p; for suddenly, HAL spoke. ' ... Too bad about Frank, isn't
by D. G. Bobrow and others 10.00 am. answered Gs queston 2-
in 1977 which used frames H: I'll take the first one .... the time of leaVIng .11
3, Bowman answered, after a long pause. 'It is.'
to structure the discourse. G: What date do you want to a straightforward way 'I il--
suppose you're pretty broken up about it?'
A frame is a knowledge return on? to leave at 8 0' clock'). Ana-
hat do you expect?'
structure which represents a H: On Friday in the evening ... possible ambiguity which GUS
-A.L processed this answer for ages of computer-time; it was a
situation which is predictable, GUS continually had to make resolves is the interpretation a
- e seconds before he continued: 'He was an excellent crew
or stereotyped. It consists of assumptions about what the 'on Friday', which must mean
-- ber.'
a series of slots, which in turn enquirer meant. based on the the Friday after the departure,
- ,people have an image of human-machine linguistic consist of a very restricted set knowledge about the world and not, for example, the Friday
::'action which has been permanently coloured by an of possible responses. This type after this conversation (as in
which had been built into the
nter with a talking science-fiction computer - such as of approach can be used when 'I want to go to San Diego on
system. The first exchange is
- ee Thought in Douglas Adams' The Hitch-hiker's GUide to a dialogue is fairly predictable, interpreted as a request for Friday'). Features such as these,
: Galaxy, or HAL (the Heuristically controlled ALgorithmic and there are few options to a reservation, even though which we take for granted in
puter) in Arthur C. Clarke's 2001. HAL not only carried out take - such as booking a flight everyday conversation, have to
no such request is made
"" control tasks on the space mission, he could also talk, think, through a travel agent, where directly (Will you make me be thought out and carefully
9a hize - and plot. The above extract from his conversation there are only so many slots programmed, if a human-
a reservation for San Diego
flight commander Dave Bowman is indistinguishable from (point of departure, date of on May 28?'). It also assumes machine interaction is to be
numan-human interaction. Even the processing delay is departure, method of travel. successful.
the request is for a flight, not
_ ropriate, as a response to such a curt rhetorical question. By etc.) and types of filler (1 to 31 a journey by rail, road, or sea. (After Bobrow et 01. 1977.)
"""parison, the conversational powers of 21st-century computer for date, 1 to 12 months, etc.). Then in the second exchange,
ems are still in the electronic stone age.
G: Hello. My name is Gus. I can
help you plan a simple trip by
air. Where do you want to go?
H: I want to go to San Diego on
May 28.
G: What time do you want to
leave?
H: ( must be in San Diego
before 10 am.
G: Would you like PSA flight 102
that arrives at 9.15?
H: What is the next flight?

PRIMITIVE DISCOURSE

Anyone who uses a card to obtain

money from a bank teller machine
is engaging in a primitive kind of
human-computer interaction. The
bank computer does most of the
'talking'; the recipient has only to
punch in some basic numerical
information and respond to the
prompts. Usually only the most
elementary information exchange
is present. Politeness is minimal -
perhaps a brief word of welcome
or the use of please. Only the
more advanced machines dare
to say such things as Good
morning, preferring to avoid the
complications of working out what
time of the day it actually is.
438 P • LA GUAGE AND COMMUNICATION

THE DOMAIN OF LINGUISTICS INTERDISCIPLINARY FIELDS

The development of linguistics, the science of language, Anthropological linguistics and behaviour, especially with refererxe
has been particularly marked in recent decades. There has The study of language variation and use in relation social interaction proceeds (§9).
to the cultural patterns and beliefs of the human Geographical linguistics
been an increased popular interest in the role oflanguage
race, as investigated using the theories and The study of the regional distribution
in relation to human beliefs and behaviour (§§1, 63), and methods of anthropology (§§ 2-5). and dialects, seen in relation to geogra:
an accompanying awareness of the need for a separate
Applied linguistics in the environment (§8).
academic discipline to deal adequately with the range The application of linguistic theories, methods, and Mathematical linguistics
and complexiry of linguistic phenomena. The universiry findings to the elucidation of language problems The study of the mathematical prope
teaching of linguistics emerged during the 1960s, and that have arisen in other domains. The term is language, using concepts from such .:>

especially used with reference to the field of foreign algebra, computer science, and sta .
since then, several branches of linguistic enquiry have
language learning and teaching (§62), but it applies
been established. Neurolinguistics
equally to several other fields, such as stylistics
The subject has now developed a clear identity - The study of the neurological basis 0
(§ 12), lexicography (§ 18), translation (§57), and
development and use in human bel
notwithstanding occasional uncertainry over its name language planning (§61), as well as to the clinical
of the brain's control over the processes
and coverage. Linguistics is the usual designation, with and educational fields below.
and understanding (§45).
linguistic science often used as a paraphrase. The field of Biological linguistics
Philosophical linguistics
The study of the biological conditions for language
phonetics (§27) is sometimes considered to be a separate The study of the role of language in- y-.-" _---''''''
development and use in human beings, with
discipline, because of its emphasis on the 'pre-linguistic' of philosophical concepts, and of the ---_.L-_"j
reference both to the history of language in the
status of linguistic theories, methods :
aspects of speech analysis; but it is more commonly human race and to child development (§§45, 49).
observations (§§5, 17).
included within the coverage of linguistics, as many see Clinical linguistics
Psycholinguistics
it as an indispensable foundation for language research. The application of linguistic theories and methods
The study of the relationship betwee.-
Also, the label for the person who practises general to the analysis of disorders of spoken, written, or
behaviour and the psychological pr
signed language (§46).
linguistics has caused some concern: 'Iinguisrician' is memory, attention) thought to und
Computational linguistics
sometimes used, but it is not popular among students Sociolinguistics
The study of language using the techniques and
of the subject, who normally refer to themselves as The study of the interaction betweer
concepts of computer science, especially with
'linguists'. There is thus occasional ambiguiry with the the structure and functioning of sooe:
reference to the problems posed by the fields of
60-3).
general use of the term 'linguist' meaning 'fluent in many machine translation (p. 360), information retrieval,
and artificial intelligence (§26). Statistical linguistics
languages'.
Educational linguistics The study of the statistical or quan:i:.:~·:;;:;~
Different dimensions of the subject can be of language (§15).
The application of linguistic theories and methods
distinguished, depending on the focus and interests of to the study of the teaching and learning of a Theolinguistics
the linguist. Diachronic (or historicaL) and synchronic language (especially a first language) in schools and The study of language used by bi
Linguistics have developed as a result of the distinction other educational settings (§§44, 62). theologians, and others involved n-

introduced by Saussure (p. 431): the former is the study Ethnolinguistics practice of religious belief (§63).
The study of language in relation to ethnic types
oflanguage change (§54); the latter the study oflanguage
states, regardless of their history. When linguists try to
ENVOI
establish general principles for the study of all languages,
they are said to be practising theoreticaL (or generaL) Linguistics has provided the responsibly would include the At the same'
conceptual framework within consideration of a large number the book to a
Linguistics. When they concen trare on esrablishi ng the facts
which this encyclopedia has of authors, including Ludwig it has been ~~'-
of a particular language system, they practise descriptive been written. Little reference has Wittgenstein, A. J. Ayer, and more numer
Linguistics. And when the focus is on the similarities and therefore been made to the other Gilbert Ryle - a major enterprise systematic ill
differences between languages, the subject is often referred academic traditions of language in its own right. Similarly, many and structure
study mentioned on this page. It eminent psychologists, such as would otherwse r--......- ---~
to as comparative (or typologicaL) Linguistics (§§ 14, 50).
has occasionally been possible A. R. Luria, L. S. Vygotsky, and case. It has e. IO..nC-. _-""
Linguistics shares with other sciences a concern
to acknowledge them; but there Jean Piaget, have written at to be more 0 -.-
to be objective, systematic, consistent, and explicit in has been no attempt to give length on language, and there is patterns of lar6~'cv'1-_
its account of language. Like other sciences, it aims to a systematic account, as they a long tradition of experimental and the vari
collect data, test hypotheses, devise models, and construct operate within a quite different psychological enquiry into functions, and
theories. Irs subject matter, however, is unique: at one intellectual perspective, and use linguistic behaviour, which would approaches an
radically different procedures of have to be carefully considered. have been de-. _
extreme, it overlaps with such 'hard' sciences as physics
study. Issues of similar standing would these matters. _ •..•~..:.~_
and anatomy; at the other, it involves such traditional
An integrated account of have to be faced in the language controversies
'arts' subjects as philosophy and literary criticism. the history of ideas in language historiography of other fields, them. A comnr-f.o:r-~
e field of linguistics includes both science and the scholarship is beyond the such as anthropology, sociology, linguistic thoc-
iries, and offers a breadth of coverage that, for scope of the present volume. In and mathematics. attention to ';;--,E m:::-.:==-
- - aspiring students of the subject, is the primary philosophy alone, for example, The absence of systematic all these acade::::r C~
there is a major academic reference to these other major traditions, IS ~.- ..,;;=, •••••• ~
- irs appeal.
tradition, focusing on the study of traditions is thus the main that will have
ordinary language, which to treat limitation of the present volume. other encydo -