Editorial: A Hitch Hiker's Guide to Semiotic Music Analysis Author(s): Jonathan Dunsby Source: Music Analysis, Vol. 1, No.

3 (Oct., 1982), pp. 235-242 Published by: Blackwell Publishing Stable URL: http://www.jstor.org/stable/854177 Accessed: 21/06/2010 16:54
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=black. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Blackwell Publishing is collaborating with JSTOR to digitize, preserve and extend access to Music Analysis.

http://www.jstor.org

EDITORIAL

A HITCH HIKER'S GUIDE TO SEMIOTICMUSIC ANALYSIS

Introducingmusic semioticsinto analysisand theory is the work of a firebrand.*The musicallyrighteousareinflamedby a new idea that is neitherin originnor even intentiondirectlymusical,and they demandquite properly of any new conceptualtrend (or, in the demystifyingwords of the semiologist GeorgesMounin, of any new 'topic of conversation') that it should interest them by implyingwhat the musicalreturnsmight be: why should analysts take a ride with semiosis?Jean-Jacques Nattiez'swork has been sensitiveto the question,idiosyncratically, but with the appropriate passionand musical conscience: it is up to othersto makesurethathis Fondements d'une semioZogie de la musique (Paris:Union Generale/1W18, 1975)can appearas the 'Foundations' needed for informed debate by the English-speaking community. Nattiez'swritingsin the 1970'sdealtmagisterially with the historyof ideasin ourcentury,andhis discussions touchedon a comprehensive rangeof musical issues. In response, however, there was often an element of doubt. The novelty of music semioticsremindedsome that we cannot sidestepauthoritarianstructuresentirely;we cannot entirelyevade the call for a discipline, and it was the possibilityof such an evasionwhich seemed good cause for suspicionabout this kind of musicalspeculation.One wantedto know what the workof music-semiotic analysiswas supposedto achieve.It was possible to see the shapeof a semioticframework for musicalstudies, even to see the tracesof conventional musicaldisciplinesin this new garb,but manyaskedwhat was the questionmusic semioticswantedto answer? The need for somekind of guidederivesmorethananythingfromthe sheer scope of linguistics,anthropology, and all the other 'topics of conversation' which shaped,and have been shapedby, Americansemioticsand European semiology.In the sixteenthand seventeenth centuries,with a rhetorical model to hand, there was a consolingexternalfactorfor composersand theorists, with convenientformulations thatgenuinelyaidedmusic-making andmusical thinking.With the Enlightenment, the inter-disciplinary field became,ironically enough, more obscure. In our own century,it beganto be asked, in a way that attracted popularattention,whatit meansin fact for language itself to be a 'language'.This was an exciting field in which musicianscould be
* This essay is based on a paper given at City University, Londonl May 1981.

(C)MUSICANALYSIS1: 3, 1982

235

EDITORIAL

seducedout of any sense of responsibility to theirfamilydiscipline.Linguistics often led musicalthinkinginto blind alleys;those who turned to other models, in philosophyor psychology,managedonly a little better. Then, in the 1970's, linguisticsitself betrayedthe musicianswho had amusedthemselves in its service.Linguisticswas undergoing a processof deconstruction: even the true amateurcould hardlymiss the fact. In case it should be missed, the arch professional, Roland Barthes, announced it with sweepingclarityin 1977. 'In short',he said, 'eitherdue to an excessiveascesisor excessivehunger,whetherfamishedor replete,linguistics is deconstructing itself. It is this deconstruction of linguisticsthat I, for my part, call semiology' ('Lecture',TheOxford Literary Review,1979, p. 38). The same process as this retreat into the study of signs semiology or semiotics has been detectablein musicalthinking. Nicholas Ruwet's concern with musical signs was indeed 'famished'.He avowed and later disavowedwith equalvivacity-a certainmethodological premise.Nattiezwidened the grip of this skeletalmusicaldiscourse.In his studiesboth of Syrinx and 'Density21.5' he developedrelatively fleshlessformsof enquiry,not only seekingto expressthe most neutralkinds of articulatory pictureof a piece of music, but also investigating the uses of a descriptive,distributional account of musical information.In contrastto the ascetic in semiotic analysis, the 'replete' is to be found everywhere,appropriately enough, and the more repleteit is, the moreit threatens the possibilityof a methodological practice. Frits Noske's 'musico-dramatic sign' is an example, 'a musical unit which stresses,clarifies,invalidates,contradictsor suppliesan elementof the libretto. The sign is sematically interpretable and disclosesdramatic truth'(The Signifier and TheSignified, The Hague, Nijhoff, 1977, p. 316). To Ruwet, such a liberalconcept of the sign has been too often a dubioushallmarkof musicalthinking:'Onedistrustslike the plaguethe "ideological" presuppositions underlyingtraditional music theories, just as one does the recourseto musicians' intuitions in short,one adoptsthe idealof the tabula rasa' (Musique enffeu, No. 17, p. 13). It is easyto discernthe deconstruction of analytical ideologyin such trends. Our music facultiesreel under the pressureof this deconstruction. One person'sterminology is another's nonsense,and, whatis not the samebut what could well be thoughtto guarantee the valueof music semiotics' 'neutral'stance when the musicologyabout today comes to be written,one person'snonsenseis another's music. The harshdebatebetween musicologyand theorypales beside the true casualtiesand potential,in the creativeworld, of this ferment. The best brief responsehere to this pressureis to move directlyinto the studyof some music. The perspectives involvedmay be quitefamiliar.First, segmentation: RomanJakobsonmade suggestiveanalogies betweenlanguage andmusic on the basisthat both displaytemporal extension whetheror not we regardthem as similar systems of communication in other ways-and Nattiezexplained why this impliesthatmusicis segmental.We expectto find a boundary,a zero degreeof segmentation beyondwhieh the music does not

MUSIC ANALYSIS1: 3, 1982

236

EDITORIAL

authorize us to proceed.Music-semiotic analysistakingthis leadwouldentail two concepts, the concept of minimal segmentaton and the concept of musicalmeaning.If a relationship is to be discovered betweenour description of segmentation and our intuitionof meaning,we may want to characterize the result by giving it a special semiotic name. Meaningarticulated in this specialmannerwill be called 'signification.' The second perspectiveis derived mainly from work by David Lidov. Lidov wants to make us awareof the discoverieswhich can emergefrom a special confrontation,between traditionalanalyticalproceduresand procedureswhichhavebeen, if not inventedby, then at leastappropriated by music semioucs.The followingis an exemplary case(adapted fromLidov, 'Nattiez's Semioticsof Music', TheCanadian 3rournal of Research in Semioties, Vol. 5, No. 1, p. 40):

Bb.

vi

111

vi
F:

(1l)
V I IV
w

Bb
I

111

vi

Eb

ii,

F-

Bi. V F: I

I Eb: V

N I

vii

MUSICANALYSIS1: 3, 1982

237

Bb:

(I

IV

vi

II)

Eb:

iii

EDITORIAL

Paradigmatic scheme for theharmonic progression: A

I I *

C
1 |

B
.

Bb: Bb:

I vi III vi (II) I V III vi

F: V I IV V I Bb: V I IV V I Bb: I IV ii V (I) F : VI Eb: V I

Bb: (I IV) Bb: vii I IV V I I IV V I

HeretheBachsongO7esulein Sussis analysed conventionally, accounting for the unitsof harmonic structure and theirsequence. This account is sifted througha grid familiarfrom Ruwet'sanalyses,lookingfor the longest repeated units of harmonic progression and arranging themin as many categories, or paradigms, as areneeded.Noticethatthereareno remainders here.Threesegmental types,designated A, C andB appear to represent the entiresong.Lidov's analysis whichin the fullaccount involves moresensitive complexitiesefficientlyexposesa criticalsynthesis.The harmonic analysis of the Bach song derivesfromone kind of musictheoryand its presentation in Levi-Straussian paradigms appealsto quite another.The semioticquestionhere is clear enough.We can acceptthat the Roman numerals indicate unitsof harmonic progression andwe knowthatsuchunits couldbe arranged in anynumber of different strings, so whatis theparticular arrangement whichcharacterizes this piece,whichleadsto ourcomprehensionof it?Jakobson madea goodruleof thumb whichoughtto be as truefor music as for language: Structural analysis,he insisted,must have both descriptive andintuitive adequacy. I believe thatLidovhasperformed wellon thatbasis. Withthesetwoperspectives in mind,we moveto somewell-known music (seebelow). Thewholepiece,Brahms's Intermezzo in B minor, Op. 119,No. 1, hasbeenanalysed in various ways:forinstance, witha voice-leading graph in Salzer'sStructural Hearing; also from the point of view of historical moment,it mighteven be a kind of geneticstructuralism, in an articleby Brian Newbould, whereBrahms's reliance on harmonic progression through cycles of descending,diatonicfifths is considered to be the sign of a Chaconne-type compositional procedure (The Music Review, Vol. 38, No. 1). Takinga more 'famished' premise,it is not difficultto imaginethat the smallest unit in suchmusicis the note, moreexactly the pitch-class attack. Thisin itselfimplies twoanalytical operations, theelimination of thevariables duration andregister. Theseoperations donotnecessarily haveanequal status froman interpretive pointof view:in otherwords,I do not supposethat duration andregister areprecisely of equalsignificance in the articulation of pitch structure. Fromthepointof viewof method, however, eliminating these
MUSICANALYSIS1:

3, 1982

238

EDITORIAL

variables is an explicitstep,wherethereis no placeforinterpretation: it is a formof analysis whereit is not possible to makea mistake (except trivially). The pitch-class attacksmay also be countedwithouterrorand the results arranged arithmetically. If theseunitsare signsin any sense,perhaps they signify moreif therearerelatively moreof them,andlessif there arerelatively fewer.
Brahms, AdagiO
_

Intermezzo,

Op. 119,No. 1 bs 1-16

lXd"4

JA:;

F""fl<

<D

'

MUSICANALYSIS1: 3, 1982

239

EDITORIAL

So far, the methodseemsa fool's errand,'unmusical' and of doubtful potential musical value.Intuitive adequacy, though,is easyto add. Fig. 1
Figure 1 (a) pitchclasses: F# numberof attacks; 28 (= 163) (b) degreestatusin B rxiinor: D major: F# rxiinor: (c) Bb major: D M T ? C# 24 A 20 D 19 B 19 E 16 G 15 A# 7 G# 6 E# 4 C 3 D# 2

St L D ?

(L) D M L

M T Sm M

T Sm Sd ?

Sd St (L) ?

Sm Sd ? Sm

L ? ? T

(Sm) ? ? ? St L ? D

? ? ? St

? ? (Sm) Sd

(T = tonic, St = supertonic, M = mediant,Sd = subdoniinant, D = doniinant, Sm = flat submediant in niinoror submediant in major,(Sm) = sharpsubmediant in rxiinor, L = leadingnote, (L) = flat seventh degreein niinor)

showsan interpretation of the initialresults in termsof traditional harmonic theory: A gridis applied to the countof pitch-class attacks set out in line (a). Belowthe pitch-class namesis a classification of eachaccording to its degree status(whichin minorraisesless than'neutral' issues,ignored here, aboutthe sixthandseventh degrees) in oneof threetonallevels,B minor,D major andF# minor.It is a largestepfromthe logicof signsystems, herea simplearithmetical logic,to the intuitive choiceof harmonic classifiers. The results canbe tested,however, by the application of counterintuitive classification,forinstance by applying Bb major as in line(c). Thiskindof classification canbe completed by examining thedistributional pattern whichwould arise if allkeys,major andminor, wereapplied to line(a);butit suffices here to observe theleft-hand bunching of important degrees according to whatcan be considered, informally, the mostpertinent grids,in the (b) lines. Incasethisseems readily acceptable, I stress thattheanalysis suggests there is a specialrelationship betweenthe quantity of pitch-class attackand the quality of harmonic effectin themusic.Thisrelativity, thatis, formsa tonal sign,perhaps specific to thepiece,perhaps not theinvestigation of either is demanded by Nattiez withhisnotion of 'seriation', andin thelatter casethere would be somechallenge to conventional viewsabout theworkings of tonality. Lidov'sapproach suggestsa confrontation of different analytical methods, and his Bachanalysis in particular showsthat an injection of traditional intuition canwellbe thefirststep.In theBrahms analysis so farI havemoved fromdistribution to interpretation, so the resultsshouldnow be testedby reversing thatprocedure. Thefirststep,then,is interpretive orintuitive: it is to identifya melody.Fig. 2 showsthe resultsof this new approach to the music.Therelative quantities of pitch-class attack in themelody (thetoppart of the music)are presented in line (a). Evenallowing for the deliberately weighted presentation (takingthe orderof pitch-classes fromFig. 1 rather thanfromline(a)of Fig. 2) thisdistribution is evidently akinto thatof line(a) in Fig. 1. Thedistribution in Fig. 2 is notindependent of thatin Fig. 1 since
240
MUSICANALYSIS] : 3, 1982

EDITORIAL

thepitch-classes of themelody werecounted in withthetotals firstexamined. However, a newkindof gridwillbe ableto separate thisstatistical interaction. In Fig. 1, thegridwasfrankly intuitive, classification according to traditional harmonic concepts. Thegridapplied in Fig. 2 is thatof periodic emphasis and seemsto me ratherless interpretive and rathermore,in Nattiez'ssense, neutral.Successively larger metrical groupings areusedto eliminate metricallysubsidiary pitchclasses, in a self-limiting procedure which endsatline(f) withthe beginning of the sixteen-bar period,a melodic Ft:
Figure2 distribution of pitch-class attacks (a) in melody: (=38) (b) first beats: (c) first beatsof 1,3,5etc.: F: 8 3 3 C: 7 3 3 1 0 0 A 5 1 1 0 0 0 D 4 1 0 0 0 0 B 5 1 0 0 0 0 E 2 2 0 0 0
0

G 4 3 0 0 0 O

A: 0 0 0 0 0 O

G: 1 1 0 0 0 O

E: 2 2 1 0 0 O

C D: 0 0 0 0 0 0

(d) of bs 1, 5, 9, 13:3 (e)ofbsl,9: 2 (f)ofb. 1: 1

0 0 0 0 O O

It appears from the left-hand bunching in Fig. 2 that the relativities of pitch-class attack operate in two ways.First,the texture as a wholeandthe melodyas a specialstructural layerbothdisplaya hierarchy of pitch-class attack whichrepresents something likethe harmonic emphasis anyinformed listener wouldpredict. Butthere is alsoanequivalence between thishierarchy andthe hierarchy of metrical emphasis. Simply, whatappear to be the most important harmonic notesarealsothemostimportant notesmetrically in the melody.I assume thatthe periodic metreof nineteenth-century musicmust playan important rolein the structural synthesis of a tonalpiece,but one wouldhardlyexpectwhatFig. 2 shows:thatperiodic metreenables us to identify unitsof the melodyalone,deprived of registral andof actual durationalvalue,representing harmonic valuesof the entiremusic. This analysis,obviously put forward as an exemplar and not as a new contribution to ourenjoyment of Brahms, canbe calleda classicexample of the best-known aspects of semiotic procedure in musicanalysis. It concerns units segmented in a consistent way. It uses someanalysis whichcan be applied quitemechanically, without interpretation, andwhich is self-limiting. It establishes a consistent, strictly patterned formof relationality fortheunits andit passesthemthrough grids,or 'codes',to investigate theirsignifying potential. Theunitsandtheirarithmetic values mightbeconsidered signifiers andsignifieds forming twelvetonalsigns.Theremustalsobe a rolefor the 'interpretant' in the relationship betweenunitsand function: the musicis
MUSIC ANALYSIS1: 3, 1982

241

EDITORIAL

nothing if not perceived. Thatrolewouldbe clearin the rejection of righthandsignsforleft-hand signs.Thisneedbethecaseonlyif a particular 'code', thatof major/minor tonality, is in play.Perhaps we do not actually haveto hearthismusicin B minor withits related tonalareas. If theBrahms werean ethnomusicological object,for whichwe hadno culturally authorized code, the analysis couldjustas wellsuggest thatthe right-hand bunching of questionmarks andof voidentries in thetwofigures identifies thesignification of the music,identifses whatit smeans' in the senseoutlined above. Nevertheless, eventhis'famished', relatively neutral kindof result suggests whatseemsan intriguing question.Schoenberg, as is well known,had a theoryaboutthe evening-out of the numberof pitch-class attacksin his composing, a theory, thatis, about theavoidance of tonicrepetitions andtheir associated functions. It is a theory whichhasbeensomewhat undervalued in its implication for the studyof nineteenth-century music,wherethe roleof pitchquantity in pitchstructure is notexactly a burning issue.Could Schoenberghavebeenmoreor less rightandmostaccounts of tonality, especially Schenker's, moreor less wrong,or makinga compositional rather thana perceptual point:thatis, to reverse theusual,indeed myusual reading of this matter, couldit be thatSchoenberg hinted at ansesthesic' account of tonality, moreanalytical in its concern withhowmanynotesthe listener dealswith, andthat Schenker's was a spoietic' one, morecompositional in its concern with how voice-leading hierarchies arise to give shapeto pitch material (especially: howtheyarosein Beethoven's sketching)? Sixteen barsof music cannot answer sucha question. WhatI wantto indicate is thatthis typeof question maynotbe asked in thefirstplace if wearenotprepared to challenge inherited disciplines, albeitchallenge themfromwithin. Myideaof a 'guide' in thisfieldsprings fromsomerecent English writing whichmixessciencefictionand offbeathumour, in a waythathas indeed givenit a popular cultstatusas a genuine 'topicof conversation'. The Hitch Hiker'sGuideto the Galaxy, aboutwhichDouglas Adams writes,has two remarkable characteristics. First,it cantellyoueverything youneedto know aboutanything. Second, everything it tellsyou-and it is all true is rather odd. Semiotics is not farfromthis, withits limitless fieldof enquiry andthe strange aspects of anysocialactivity it canso easilyuncover. If we bringto bear a point of view conceived in abstraction on objectsconceivedand performed in practice, it is onlyto be expected thattheresult willjar.Almost by definition, a musicsemiotics cannot share thepureintuitive sympathies of eithersemiotics on the one handor music-making on the other(an idea pursued relentlessly in Reinhard Schneider's Semiotik derMusik, Munich, Fink, 1980).Naturally, then, we haveto thinkin termsof a disciplinary deconstruction. The answers, one mightsay, areall around in our concert hallsand musiclibraries. Musicsemiotics maymakesomecontribution to formulating the questions so manyanalysts haveaboutthem. Jonathan Dunsby
242
MUSICANALYSIS1: 3, 1982