Emotion and Meaning in Music

Emotion and
Meaning
in Music
Emotion and
Meaning
in Music
By LEONARD B. MEYER
1211
Library of Congress CatalogCard Number:56-9130
THE UNIVERSITY OF CHICAGOPRESS. CHICAGO 60637
The University of Chicago Press, Ltd., London
1956 by The University of Chicago. All rights reserved.

Published 1956. Printed in the United States of America
00 99 98 97 96 95 94 93 92 91 16 l7 I8 I9 20 21
ISBN: 0-226-52139-7 paperbound!

Library of Congress Catalog Card Number: 56-9130
To the memory ofmy father
ARTHUR S. MEYER
His life was gentle,and theelements

So mixedin him that Naturemight standup
And say to all the world, 'I`his was a man.
Preface
The diversity and complexity of twentieth-centurymodes of

thought, togetherwith the clear and pressing needfor a more
sensitive andcomprehensive understanding of how the exchange
of attitudes, information, and ideas takesplace, has made the
analvsis ofmeanings and an examination of the processes by
which
they are communicated animportant focusof interest for many
nominally disparatefields of inquiry. Philosophy,psychology, so-
ciology, andanthropology, toname some of thosemost directlyin-
volved, haveall becomeconcerned withthe problemof meaning:
the variety of meanings, their significanceand epistemological
status, their interrelationships, and manner of communication.
Other fields,such aseconomics, political
science, various
branches
of the humanities, and even the natural sciences, have likewise
directed attentionto theseproblems.
The problemof musical meaning andits communicationis of
particular interestfor severalreasons. Not
only doesmusic useno
linguistic signsbut, on one levelat least,it operatesas a closed
system, thatis, it employs nosigns or symbols referringto the
non-musical world of objects, concepts, and human desires. Thus
the meaningswhich it imparts differ in important ways from
those conveyedby literature, painting, biology,or physics. Un-
like a closed, non-referential
mathematical system,music issaid to
communicate emotional and aestheticmeanings aswell as purely
intellectual ones.This puzzling combination ofabstractness with
concrete emotional and aesthetic experience can, if understood
correctly, perhapsyield useful insights into more generalprob-
vii
viii Preface
lems of meaning andcommunication, especially
those involving
aesthetic experience.
However, beforethe relationshipof music to other kinds of
meaning andother modesof communicationcan be considered,
a detailedexamination ofthe meanings of musicand theprocesses
by which they are communicated must be made.Thus although
it is hoped thatthe relevanceof this study to the largerproblems
of meaningand communication will be apparent, thesematters are
not explicitlyconsidered. Noattempt, forexample, ismade todeal
with the general logicalphilosophical status of music-to decide
whether musicis a language orwhether musicalstimuli aresigns
or symbols.
The relationshipbetween musicand other realms of aesthetic
experience is likewise left for the reader to determine. Where
reference hasbeen madeto othermodes ofaesthetic experience, it
has beendone in order to clarify or bring into sharper reliefsome
point in connection withmusical processes rather thanto establish
a general aesthetic system. On the other hand, one can hardly
fail to become awareof the striking similarityof someaspects of
musical experience to other types ofaesthetic experience,
particu-
larly thoseevoked byliterature.
The subjectof the present study,though perhapsof more than
passing interestfrom the general viewpointsdiscussed above, is
of vital and paramountimportance inthe field of musicitself.
For if the aesthetics and criticism of music are ever to move
out of the realmsof whim, fancy, andprejudice, andif the analy-
sis of music is ever to go beyonddescription whichemploys a
special jargon,then someaccount of the meaning,content, and
communication ofmusic moreadequate thanat presentavailable
must be given. As I. A. Richards putsit, The two pillars upon
which a theory of criticism mustrest arean accountof valueand
an account of communication 1-and included in an account of
1. I. A. Richards,Principles ofLiterary Criticism New York: Har-

court Brace & Co., 1928!, p. 25. Although value judgments are un-
avoidably impliedthroughout, thepresent studyis primarily concerned
with presentingan accountof meaningand communication.
Preface ix
communication isobviously an account of t_he meanings
com-
municated.
Meaning andcommunication cannot be separatedfrom the cul-
tural contextin which they arise.Apart from the socialsituation
there can be neither meaning nor communication. An understand-
ing of the culturaland stylisticpresuppositions of
a pieceof music
is absolutelyessential tothe analysisof its meaning. It should,
however, benoted that the converseof this proposition is also
true: namely,that an understanding of the generalnature ofmusi-
cal meaningand its communication isessential toan adequate
analysis ofstyle andhence tothe studyof music history andthe
investigations ofcomparative musicology as well.
The arguments and debatesof aestheticians,

the experiments and
theories of psychologists, andthe speculationsof musicologists
and composersstill continue and are ample indicationthat the
problems ofmusical meaningand communication are with us to-
day. In fact, the inclusion of music aspart of liberal education,
the unpatronizingand seriousconsideration given to non-Western
music, andthe attemptsto includet_he artof musicin studiesdeal-
ing with cultural historyhave madethe problemseven morepress-
ing. It is becauseof theseneeds, aswell as the more speciiically
musical onesmentioned earlier,that the author hasthe temerity
to attemptanother study in thisfield.
The book is divided into three main parts. Chapter i considers,
Hrst, the nature of emotional and intellectual meanings, their in-
terrelationship, andthe conditionswhich give rise to them, and,
second, howin general these conditionsare fulfilled in the re-
sponse tomusical stimuli.Chapters ii-v are devotedto a fairly
detailed examination of the social and psychological conditions
under which meaning arisesand communicationtakes placein
response to
music. Andchapters vi-viipresent evidence
of various
kinds, taken from several cultures and several cultural levels, to
support thecentral hypothesisof the study.
Because thisstudy drawsso freelyupon work in many diverse
fields, itis perhapsimportant toemphasize that
the basictheoreti-
x Preface
cal formulationsadvanced in it were derived from a study of
music ratherthan, for instance, froma study of aestheticsor psy-
chology. Otherfields often furnished excitingand encouraging
confirmation forconclusions originallyreached througha careful
consideration of
music andmusical processes. Fields outsidemusic
have alsoserved to refine conceptsor have led to more general
formulations. Butmusic wasthroughout thecontrolling guidein
the formulationof the theory presented here.
The debt which this book owes to other scholars is both so
manifest andso vastthat only a few of the most importantones
can bementioned. Inthe fieldof philosophythe workof HenryD.
Aiken, ]ohn Dewey, SusanneLanger, and George Mead has
been a source ofinsight and understanding. Inthe field of psy-
chology have
I obviouslyleaned heavilyupon theworks ofK. Koffka,
]. T. MacCurdy, and ]ames Mursell. Though contributing little or
nothing to the theoretical formulations made,the work of musicians
and musicologists,particularly those working in comparative musi-
cology, has
been animportant source
for mostof the evidence pre-
sented inthe later portions of this book.
Throughout thepreparation andwriting of this book, I have
received valuableadvice andencouragement from my colleagues
and students. In particular I am indebted to Grosvenor Cooper
for his sympathetic understanding
of the viewpoint of this study
and his many excellentsuggestions; toCharles Morris for his
cogent criticisms and his precise analysisof many of the problems
discussed inthe course of this work; to Knox C. Hill, who helped
me to edit and cut the text; and to Otto Gombosi, whogave so
freely ofhis wisdomand erudition.
Last but as tradition hath it! by no means least,
I wish to ac-
knowledge thedebt I owe to my wife. For it was shewho en-
couraged me when I was depressed; prodded me when I was
lazy; ran the household so that I had a maximum of peace and
quiet; and at the same timemanaged toput up with my many
moods andperversities.
Table of Contents
I. THEORY .
II. EXPECTATION AND LEARNING _
III. PRINCIPLES OF PATTERN PERCEPTION: THE LAW OF Coon

CONTINUATION
IV. PRINCIPLES OF PATTERN PERCEPTION: COMPLETION AND

CLOSURE .
V. PRINCIPLES OF PATTERN PERCEPTION: THE WEAIIENING OF

SHAPE _
VI. THE EVIDENCE: DEVIATION IN PERFORMANCE AND TONAL

ORGANIZATION
VII. THE EVIDENCE: SIMULTANEOUS ANDSUCCESSIVE DEVIATION
VIII. NOTE ON IMAGE PROCESSES,CONNOTATIONS, ANDMOODS .
NOTES _
INDEX
xi
I
Theory
Past Positionsas to the Nature of

Musical Experience
Composers and performers of all cultures, theorists of diverse
schools andstyles, aestheticians
and criticsof many different per-
suasions areall agreed that music has meaningand that t_his mean-
ing is somehow communicatedto both participants and listeners.
This much, at least, we may take for granted. But what constitutes
musical meaningand by what processesit is communicated has
been thesubject ofnumerous and
often heateddebates.
The Hrst main difference of opinion exists between those who
insist that musical meaning lies exclusively within the context of
the work itself, in the perceptionof the relationships setforth
within the musical work of art, and those who contend that, in
addition to these abstract, intellectual meanings, music also com-
municates meaningswhich in some way refer to the extramusical
world of concepts, actions,emotional states,and character. Let us
call the former group the absolutists and the latter group the
referentialists.
In spite of the persistent wrangling of these two groups, it seems
obvious thatabsolute meanings
and referentialmeanings arenot
mutually exclusive: that they can and do coexist in one and the
same piece of music, just as they do in a poem or a painting. In
short, the arguments are the result of a tendency toward philo-
sophical monism rather than a product of any logical opposition
between types of meaning.
Because this study deals primarily with those meanings which
1
2 Emotion and Meaning in Music
lie within the closed context of the musical work itself, it is neces-
sary to emphasize that
the prominence
given to this aspectof
musical meaning
does notimply thatother kindsof meaningdo not
exist or are not important.
On thecontrary, themusical theoryand practiceof manydiffer-
ent culturesin many different epochsindicates thatmusic canand
does conveyreferential meaning. The musicalcosmologies of the
Orient in which tempi,pitches, rhythms, and modesare linkedto
and expressconcepts, emotions, and moral qualities; themusical
symbolisms depicting actions, character
and emotion,utilized by
many VVestern composers since the Middle Ages; and evidence
furnished bytesting listeners
who havelearned tounderstand West-
ern music-all these indicate that music can communicate refer-
ential meanings.
Some of those who have doubted that referential meanings are
real have based theirskepticism uponthe fact that suchmean-
ings are not natural and universal.Of course, such meanings
depend uponlearning. Butso, too,do purely musical meanings-
a fact that will become veryclear inthe courseof this study.
Others have found the fact that referential meanings are not
specific intheir denotationa great difliculty in granting statusto
such meanings. Yet such precision isnot a characteristic ofthe
non-musical arts either. Themany levelsof connotationplay avital
role in our understandingof the meanings communicated by the
literary and plastic arts.
Both the importance ofsuch referentialmusical meanings
and
the difficulties encountered in attempting to base an adequate
aesthetic upon them are discussed in chapter viii. For the present
we must set them aside and simply state that it is not this aspect
of meaningwhich will primarily concernus in the courseof this
study. For an adequate analysis of the problems involved in the
meaning and communication of the referential content of music
would require a separate study of its own.
Let us now make a secondpoint clear, namely, that the dis-
tinction just drawn between absolute and referential meanings is
not the same as the distinction between the aesthetic positions
which are commonly calledformalist and expressionist. Both
Theory 3
the formalist and the expressionist maybe absolutists; that is, both
may seethe meaningof music as being essentially intramusical
non-referential!; but the formalist would contend that the mean-
ing ofmusic liesin the perception and understanding ofthe musical
relationships setforth in the workof art and thatmeaning inmusic
is primarily intellectual, whilethe expressionistwould arguethat
these same relationships are in somesense capable of excitingfeel-
ings and emotions in the listener.
This point is important because t_heexpressionist position has
often beenconfused withthat of the referentialist.For although
ahnost all referentialists areexpressionists, believing
that music
communicates emotional meanings, notall expressionists
are refer-
entialists. Thus
when formalists,
such as
Hanslick or
Stravinsky,
reacting against
what they feel to be an overemphasis upon
refer-
ential meaning,have deniedthe possibility or relevanceof any emo-
tional responseto music, they have adopted an untenable position
partly because
they haveconfused expressionism
and referentialism.
One might, in other words, divide expressionists intotwo groups:
absolute expressionists
and referentialexpressionists. The
former
group believethat expressive
emotional meaningsarise inresponse
to music and that these exist without reference to the extramusical
world of concepts, actions,
and humanemotional states,
while the
latter group would assert that emotional expression is dependent
upon anunderstanding of
the referentialcontent ofmusic.
THE PRESENT POSITION AND CRITICISM OF

PAST ASSUMPTIONS
The presentstudy isconcerned withan examination and analysis
of thoseaspects ofmeaning whichresult fromt.he understanding of
and responseto relationships
inherent inthe musicalprogress rather
than with any relationshipsbetween themusical organization and
t_he extramusical
world of concepts, actions,
characters, andsitua-
tions. The position adoptedadmits both formalist and absolute
expressionist viewpoints.
For though the referentialexpressionists
and theformalists areconcerned withgenuinely different
aspects of
musical experience,
t.he absolute
expressionists and
the formalistsare
actually considering
the samemusical processes and similarhuman
experiences from
different, but not incompatible,viewpoints see
p. 39!.
Broadly speaking, then, thepresent investigation
seeks topresent
an analysisof musicalmeaning andexperience inwhich both the
expressionist andthe formalistpositions willbe accountedfor and
in which the relationshipbetween themwill becomeclear.
Past accounts given bythe proponents of eachof thesepositions
have sufferedfrom certain important weaknesses. The chief diili-
culty of those whohave adoptedthe absolutistexpressionist position
is thatthey havebeen unableto accountfor the processes by which
perceived sound patterns become experienced asfeelings andemo-
tions. Infact, strangeas it may seem, they havegenerally avoided
any discussionof emotional responses whatsoever. These short-
comings have led to a generallack ofprecision bothin their account
of musicalexperience and in their discussions of
musical perception.
But, at least, theexpressionists haverecognized theexistence of
problems intheir position.The formalists,on the other hand,have
either foundno problemsto recognizeor have simply turnedthe
other way, seeking to divert attention from their difficulties by
attacking referentialism whenever possible. Yet the formalists are
faced witha problemvery similarto thatconfronting theexpression-
ists: namely,the difficulty and necessity of explainingthe manner
in which an abstract, non-referential succession of tones becomes
meaningful. Infailing toexplain inwhat sense
such musical
patterns
can be said to have meaning,they have also found themselves
unable to show the relation of musical meaningto meaning in
general.
Finally, thisfailure toexplain theprocesses by
which feelingsare
aroused andmeanings communicated has preventedboth groups
from seeing that theirpositions shouldmake themallies ratherthan
opponents. Forthe samemusical processes
and similar psychological
behavior giverise to both types of meaning; and both must be
analyzed ifthe varietymade possibleby this aspect ofmusical ex-
perience isto be understood.
Readers familiar
with paststudies inthe aesthetics
and psychology
of musicwill perhapsnote thatmuch of the earlierwork in these
fields isnot discussed
in this study andthat manytraditional prob-
Theory 5
lems are ignored. This neglect stemsfrom the conviction that the
assumptions and orientation ofthis literaturehave provedsterile
and are today untenable.Since thisliterature hasbeen explicitly
and cogentlycriticized by such writersas Cazden, F arnsworth,2
and Langer, only a brief commenton theseearlier assumptions
seems necessary
here, inthe hopethat theposition ofthis bookwill
thereby be clarijied.
The psychology of musichas, sinceits beginnings,been plagued
by threeinterrelated errors:hedonism, atomism,and universalism.
Hedonism isthe confusionof aestheticexperience withthe sensu-
ously pleasing.As SusanneLanger writes:
Helmholtz, Wundt, Stumpf, and other psychologists. . . based their
inquiries onthe assumptionthat musicwas aform of pleasurable sensa-
tion .... This gaverise to an aestheticbased onliking and disliking,
a huntfor a sensationist definition
of beauty.... But beyond adescrip-
tion of tested pleasure-displeasure
reactions tosimple soundsor elemen-
tary soundcomplexes .. . this approachhas nottaken us. . . .*
The attempt to explain and understand music as a successionof
separable, discretesounds andsound complexesis the error of atom-
ism. Even the meager achievement which Mrs. Langer allows to
studies ofthis kindmust bestill furtherdepreciated. For
the tested
pleasure-displeasure reactions
are notwhat mostof thepsychologists
tacitly assumed them to be: they are notuniversals goodfor all
times and all places! but products of learning and experience.
This is the third error, the error of universalism: the belief that
the responsesobtained by experiment orotherwise areuniversal,
natural, andnecessary. This
universalist approachis alsorelated to
the time-honored
search fora physical,quasi-acoustical explanation
of musicalexperience-the attempt,that is, to accountfor musical
communication in terms of vibrations, ratios of intervals, and the
like.
These sameerrors have also plagued music theory. Attempts to
explain the effect of the minor mode of Western music, to cite but
One example,
in termsof consonance and dissonanceor in terms of
the harmonicseries have
resulted inuncontrolled speculations
and
untenable theories. Even those not thus haunted by the ghost of
Pythagoras have
contributed littleto our understanding ofmusical
meaning and its communication. For, onthe whole,music theorists
have concerned themselves with the grammarand syntaxof music
rather thanwith its meaning orthe affectiveexperiences to
which it
gives rise.
Today weare, I think, able to take a somewhatmore enlight-
ened viewof thesematters. Theeasy access which almostall indi-
viduals haveto greatmusic makes it quite apparent thata Beethoven
symphony isnot a kind of musical banana split, a matter ofpurely
sensuous enjoyment. The work of the Gestalt psychologists has
shown beyonda doubt that understanding is not a matterof per-
ceiving singlestimuli, or simple soundcombinations inisolation,
but israther amatter ofgrouping stimuliinto patternsand relating
these patternsto oneanother. And Hnally, thestudies ofcomparative
musicologists, bringingto our attention themusic ofother cultures,
have madeus increasinglyaware thatthe particular organization
developed in Western musicis not universal, natural, or God-given.
Evidence as to the Nature and Existence of

the Emotional Response to Music
Any discussion
of the emotional response
to musicis facedat the
very outsetwith the fact that very little is known about this response
and its relation to the stimulus. Evidence that it exists at all is based
largely uponthe introspective
reports oflisteners andthe testimony
of composers, performers, andcritics. Otherevidence ofthe exist-
ence ofemotional responsesto musicis basedupon thebehavior of
performers and audiences and upon thephysiological changes that
accompany musicalperception. Although the volume and intercul-
tural characterof this evidence compels
us to believe thatan emo-
tional responseto musicdoes takeplace, it tells us almost nothing
about thenature of the responseor about the causalconnection
between themusical stimulusand the affective response it evokes
in listeners.
SUBJECTIVE EVIDENCE
From Plato down to the most recent discussions of aesthetics and
the meaningof music,philosophers and
critics have,with few ex-
Theory 7
ceptions, affirmedtheir belief in the ability of music to evoke
emotional responses in listeners. Most of the treatiseson musi-
cal composition and perfonnance stress theimportance ofthe com-
munication of feeling and emotion. Composershave demon-
strated in their writings and by the expressionmarks used in
their musicalscores theirfaith in the affectivepower ofmusic. And
finally, listeners,past and present, havereported with remark-
able consistency that music does arouse feelings andemotions in
them.
The first di$culty with this evidenceis that, taken at its face
value, withoutbenefit ofa generaltheory ofemotions as a basisfor
interpretation, ityields noprecise knowledge
of the stimulus which
created theemotional response. Because musicflows throughtime,
listeners and critics have generally been unable to pinpoint the
particular musical
process which
evoked the affective response
which
they describe.They havebeen prone,therefore, tocharacterize a
whole passage,section, orcomposition. Insuch cases the response
must havebeen madeto those elements ofthe musical organization
which tend to be constant, e.g.,tempo, generalrange, dynamiclevel,
instrumentation, and texture. What these elements characterize are
those aspects
of mental life which are alsorelatively stableand
persistent, namely,moods andassociations, ratherthan the changing
and developingaffective responses
with which this study is con-
cerned.
Much confusion has resulted from the failure to distinguish be-
tween emotionfelt or affect! and mood. Fewpsychologists dealing
with music have beenas accurateon this point as Weld, who notes
that: The emotional experiences
which ourobservers reported
are
to be characterized ratheras moodsthan asemotions in the ordinary
Sense of
the term.... The emotion istemporary andevanescent;
the moodis relativelypermanent and
stable. As a matter offact,
most of the supposedstudies of emotion in music are actually con-
cerned with mood and association.
Taken at face value the introspective data under consideration
not onlyfail to provide accurate
knowledge ofthe stimulus music!
but theycannot even
furnish clearand unequivocal
information
about theresponses reported.
For severalreasons the
verbalizations
of emotions,
particularly thoseevoked bymusic, areusually decep-
tive and misleading.
Emotions arenamed anddistinguished from
one anotherlargely
in terms of the external circumstancesin which the responsetakes
place. Since,
aside fromthe oftenfortuitous associations which may
be aroused,music presents no externalcircumstances, descriptions
of emotionsfelt while listening tomusic areusually apocryphaland
misleading. Ifthey areto be used atall, theymust beanalyzed and
considered inthe light of a general theoryof therelation ofmusical
stimuli to emotional responses.
Second, a clear distinction must be maintained between the
emotions feltby the composer, listener,or critic-the emotional
response itself-andthe emotionalstates denoted by differentas-
pects ofthe musicalstimulus. Thedepiction of musical moods in
conjunction withconventional melodicor harmonicformulas, per-
haps specilied by the presence of a text, can become signs which
designate humanemotional states see pp.267 f.!. Motivesof griefor
joy, angeror despair,found in the worksof baroquecomposers or
the affectiveand moral qualities attributed to special modes orrcigas
in Arabian or Indian music are examples of such conventional de-
notative signs.And it may well be that when a listener reports that
he felt this or that emotion, he is describing the emotion which
he believesthe passageis supposedto indicate, not anything which
he himself has experienced.
Finally, even where the report given is of a genuine emotional
experience, it is liable to become garbled and perverted in the
process ofverbalization. For emotional statesare much more subtle
and varied than are the few crude and standardized words which
we use to denote them.
In this connection it would seem that many of the introspections
supplied bysubjects inthe studies
made byVernon Lee,C. S.Myers,
Max Schoen,and otherscontain alarge amountof what psychia-
trists calldistortion. For example, when
a subjectin anexperiment
by Myers reports that while listening to a particular musical selec-
tion shehad a restful feelingthroughout . . . like one of going
downstream whileswimming, she is obviously translating un-
speakable feelings
into symbolicform. The interpretation ofsuch
Theory 9
symbols is the task of the psychiatrist, not the music critic. To the
music critic such introspectionsshow only that some response, not
necessarily aspecifically musical one, waspresent. For it is always
possible that the thoughts and reveries thus revealed are without
any relation to musical experience. The musical stimuli may have
functioned merelyas akind of catalytic agent,
enabling theresponse
to take place butplaying nocontrolling partin shapingor deter-
mining the experience andfiguring nowhere in the end result, ex-
cept perhaps
negatively see
chap. viii!.
OBJECTIVE EVIDENCE: BEHAVIOR

The responsesof listeners can also be observed and studied ob-
jectively. Two general categories of observable responses canbe
distinguished: a! those responseswhich take the form of overt
changes in
behavior and9! those responses which
take theform of
less readily observable physiological changes. Suchobjective evi-
dence, thoughit undoubtedlyavoids thedifliculties ofthe verbaliza-
tion of subjective feelings
and emotions,presents otherdifliculties
no lessperplexing.
In the first place, emotional responsesneed not result in overt,
observable behavior.
As HenryAiken pointsout," oneof the special
characteristics of
our responses to aestheticobjects isthe very fact
that, dueto ourbeliefs asto the nature ofaesthetic experience,we
tend to suppress overt behavior. Furthermore, as an important
adjunct tothis point,it should be notedthat emotion-feltor affect
is most intense preciselyin those cases wherefeeling does not result
in or take the form of overt behavior or mental fantasy see p. 14!.
This isclear assoon asone considers
the tendencyof humanbeings
to work off or relieve emotionaltension in physical effortand
bodily behavior.In short,the absenceof overtemotional behavior,
particularly in response to aesthetic stimuli, is no indication as to
either thepresence or
force ofemotional responses.
However, evenwhere overt behavior is present, its interpretation
is difficult and problematical.When, on the one hand, overtbe-
havior isthe productof particularlypowerful emotional
tensions,
it tendsto be diffuse, generalized,
or chaotic.Extreme conflict,for
example, mayresult in either motionlessrigidity or frenzied activity;
weeping may
be theproduct ofeither profoundgrief orextreme joy.
Unless wehave accurateknowledge ofthe stimulus situation before-
hand, suchbehavior cantell us little or nothing as to the significance
of theresponse or
of its relation tothe stimulus.
On the other hand, when emotional behavior does become differ-
entiated it tends to be standardized-to become part of more gen-
eral patternsof social behavior. Thusalthough thephilosophical
aspect ofthe stimulussituation, thefact that an aestheticobject is
being considered,tends towardthe suppression of overt behavior,
the social aspect of the stimulus situation permitsand at times
indeed encourages certain standardizedtypes ofemotional behavior.
This isapparent inthe conductof performersand audiences alike.
The jazzperformer andhis audience, for example,have onemode
of socially sanctioned emotional behavior; the concert performer
and his audience have another. The difference between the two is
more a matter of conventionally determinedbehavior patterns than
it is a matter of musical differences seep. 21!.
Such behaviormust be regarded atleast in part as a meansof
communication rather than as a set of natural, reflex reactions. It
indicates and designates not only appropriate mental sets but also
the proper i.e., socially acceptable! modes of response.Once this
sort ofbehavior becomeshabitual, andit doesso veryearly inlife,
then it may beactivated bythe socialaspects ofthe stimulussitua-
tion alone, without regard for the stimulus itself. In short, given no
theory asto the relation of musical stimulito affectiveresponses,
observed behavior can providelittle information as to either the
nature ofthe stimulus,the significanceof the response, or
the rela-
tion betweenthem. Forconduct whichmight to an observer appear
to indicatethe presence of an emotional responsemight in point of
fact be the resultof the subjects day.dreams, hisobservation and
imitation of the behavior of others, or his beliefs as to the kind of
behavior appropriate
and expected
in the given socialsituation.
OBJECTIVE EVIDENCE! PHY SIOLOCICAL RESPONSES

On the physiological levelmusic evokesdefinite andimpressive
responses. has
It a marked effect on pulse,respiration andexternal
Theory 11
blood pressure.... [It] delays theonset ofmuscular fatigue. . .
[and] has a markedeffect uponthe psychogalvanic reflex .... 8
In spite of the fact that these changes are the very oneswhich
normally accompany emotional experience,
the significanceof these
data is not completely clear. Two principal difliculties are involved.
To beginwith, norelation canbe foundbetween thecharacter or
pattern ofthe musicalselection evokingthe response and the par-
ticular physiological
changes which take place.These changes ap-
pear to be completelyindependent ofany particularstyle, form,
medium, orgeneral character. The sameresponses willtake place
whether themusic isfast orslow, excitingor soothing,instrumental
or vocal, classical orjazz.
Because tonal stimulation is a constant factor of all musical
stimuli, Mursell is led to conclude that the power of tone as such
must bethe causeof the physiological changes
observed.
There is, however, another constant involved in the perception
of music; namely, the mental attitude of the audience. The listener
brings tothe actof perceptiondefinite beliefs
in the affective power
of music. Even before the first sound is heard, these beliefs activate
dispositions to respond in an emotional way, bringing expectant
ideo-motor setsinto play. And it seems morereasonable to suppose
that the physiological changesobserved area responseto the listen-
ers mental set rather than to assume that tone as such can, in some
mysterious and unexplained way, bring these changes about di-
rectly. For while the relationship between mental sets and physio-
logical changes
has beendemonstrated beyond
doubt, theeffect of
tone as such hasnot seepp. 74f. !.
This doesnot imply that the presence ofa physiological environ-
ment, whichis a necessary conditionfor the arousal ofemotion, is
not a significant fact.The existenceof this necessary condition
increases thelikelihood that emotional responsesdo take place-a
fact which some criticshave soughtto deny. W`hat thisanalysis
indicates is that not only are these physiological adjustments pre-
emotional, as
Mursell wouldadmit, butthey arealso pre-musical.
Furthermore, eventhe conclusionsjust reachedabout thesignifi-
cance ofthe physiological
data areprobably anexaggeration, not
if
12 Emotion and Meaningin Music
from apsychological point of view,at least,from a logical one.For
such adjustments not only accompany affective responses, but they
are alsoconcomitants of clearly non-emotional responses.
In the light of present knowledge it seems clear that though
physiological adjustments are probablynecessary adjuncts of affec-
tive responsesthey cannotbe shownto besufHcient causes for such
responses and have, infact, beenable tothrow verylittle light upon
the relationshipbetween affective responses and the stimuli which
produce them. The situationis conciselysummarized byRapaport:
a! On the basisof thematerial surveyed nothing canbe definitelystated
as tothe relationto emotion felt" of physiological processes
concomitant
with emotions.Proof hasnot beenoffered toshow that the usually de-
scribed physiologicalprocesses are always presentwhen emotionis felt.
b! Nothing is knownabout thephysiological processes underlying emo-
tional experience.However, sufhcientproof has been adduced that
neither thejames-Lange theory nor the hypothalmic theoryexplains the
origin of emotion felt. c! The investigations intothe physiologyand
the neural correlates of emotional expression are of importance; their
relation tothe psychicprocess designated as emotion felt is the crucial
point of every theoryof emotions.However, theknowledge concerning
this relationis so scant thatinvestigations into
the influenceof emotions
on otherphysiological processes will have to be based ratheron what is
known aboutthe psychologyof emotions.
There is one basicproblem with all the objective data discussed:
namely, thateven whenaffective experiences
result in objective
adjustments, whetherbehavioral or physiological, what can be ob-
served isnot the emotion-felt, theaffect, butonly its adjuncts and
concomitants, which in the case of behavior tend to become stand-
ardized andin the case ofphysiological changes
are not specific to
emotion. What we wish to consider, however, is that which is most
vital andessential inemotional experience:
the feeling-toneaccom-
panying emotionalexperience, thatis, the affect.
Here weface a dilemma. Onthe one hand, the response with
which weare concernedis profoundlyand permanentlysubjective
and hence
of necessity
concealed from
the scrutinyof eventhe most
scrupulous observers;and, on the other hand, we have found that
the subjective
data available,taken bythemselves, provide
no defi-
nite and unequivocal information about the musical stimulus, the
Theory 18
affective response,
or therelation between
them. ThisdifHculty can
be resolved only if the subjective data available, including the re-
sponses ofthereaders and the authorof thisstudy, canbe examined,
sifted, andstudied in the light of a general hypothesisas to the
nature ofaffective experience and theprocesses by which musical
stimuli mightarouse such experience.
Such a hypothesis isprovided by the psychologicaltheory of
emotions. Foralthough muchwork undoubtedly remains to be done
in the Held ofemotional theory,
there appearsto be general agree-
ment amongpsychologists and psychiatrists atleast asto the con-
ditions under which emotional responses arise
and asto the relation-
ship betweenthe affectivestimulus andthe affectiveresponse.
The PsychologicalTheory of Emotions

Since the physiological changeswhich accompany emotional ex-
perience, whatevertheir importance,do not provide a basis for
differentiating affectivefrom non-affectivestates, the differentia
must besought inthe realmof mentalactivity.
However, notall mental responses are affective. We speak of
dispassionate observation, calm deliberation,and cool calculation.
These are non-emotional states of mind.
If we then askwhat distinguishes
non-emotional states
from emo-
tional ones, it is clear that the difference does not lie in the stimulus
alone. The same stimulusmay excite emotion in one personbut not
in another.Nor doesthe diiierencelie in the respondingindividual.
The sameindividual mayrespond emotionally to a given stimulus
in one situation but not in another. The difference lies in the rela-
tionship between
the stimulusand theresponding individual.
This relationship must iirst of all be such that the stimulus pro-
duces atendency inthe organism to think.or actin a particular way.
An object or situation which evokesno tendency,to which the
organism isindifferent, canonly resultin a non-emotional state of
mind.
But evenwhen atendency isaroused, emotion may notresult. If,
for example,a habitualsmoker wants a cigaretteand, reachinginto
his pocket,finds one,there will be no affective response.
If the
tendency issatisEed without
delay, noemotional response
will take
place. If,
however, the
man finds
no cigarette
in hispocket, discovers
that there are none in the house, and then remembers that the
stores areclosed andhe cannotpurchase any, he will very likely
begin torespond inan emotionalway. Hewill feel restless, excited,
then irritated, and Bnally angry.
This bringsus tothe centralthesis ofthe psychological
theory of
emotions. Namely:Emotion oraffect isaroused whena tendencyto
respond isarrested orinhibited.
SUPPORTING THEORIES
In 1894]ohn Deweyset forth what hassince become
known as
the conflict theory of emotions.
In an article entitledThe Conflict Theory ofEmotion, 1 Angier
shows thatthis generalposition hasbeen adopted, in more or less
modified form,by many psychologists ofwidely differentview-
points. Forinstance, the behaviorists, who
emphasize theexcitement
and confusionwhich disruptbehavior asimportant characteristics
of emotionalconduct, wouldseem tobe describing objectively what
others view as the result of inner conflict. But the diiiiculty with
examining emotions
from the point of view of behaviorism isthat,
as wehave seen,
emotion maybe feltwithout becoming manifest as
overt behavior.
MacCurdy, whoseown attitude is psychoanalytical, points out
that it is preciselywhen instinctivereactions are
stimulated thatdo
not gain expression eitherin conduct, emotional expression,or fan-
tasy, thataffect ismost intense.
It is the preventionof theexpression
of instinct either in behavior or conscious thought that leads to
intense affect.In otherwords theenergy ofthe organism, activating
an instinctprocess, must be blockedby repression before poignant
feeling is excited. 13MacCurdys analysis involves threeseparate
phases: a! the arousalof nervousenergy inconnection withthe
instinct ortendency; 14 b! the propensity for this energyto become
manifest asbehavior or conscious thought once the tendency is
blocked; and o! the manifestation ofthe energyas emotion-felt
or affectif behavior and conscious thought arealso inhibited.Of
Theory 15
course, ifthe stimulation is sopowerful that the total'energy cannot
be absorbed by eitherbehavior oraffect alone,both will result.
It is obvious thata shift of emphasishas takenplace inthe state-
ment ofthe theoryof emotions.Dewey andhis followerstended to
stress theconflict oropposition oftendencies as being thecause of
emotional response. MacCurdy andmost ofthe morerecent workers
in the field believethat it is theblocking orinhibiting of a tendency
which arousesaffect. Actuallythe conceptof conflict through the
opposition of
simultaneously aroused
conflicting tendencies
may be
regarded as
a specialand morecomplicated caseof the arrest of
tendency.
This pointwas madein Paulhansbrilliant work, which in 1887,
almost tenyears beforeDeweys formulation,set forth a highly
sophisticated theory
of emotions.If we ascend inthe hierarchyof
human needs and dealwith desiresof a higher order,we still find
that theyonly giverise toaffective phenomena
when thetendency
awakened undergoesinhibition. 1
However, morecomplex phenomena are possibleas theresult of
the simultaneousor almostsimultaneous cominginto play of sys-
tems which tend toward opposite or different actions and which
cannot bothculminate inaction atthe sametime; alwaysprovided
that thepsychical systems brought intoplay donot differtoo widely
in intensity.... 17 Such a situationresults, according
to Paulhan,
in anemotion or affect characterized
by confusionand lackof clarity.
In otherwords, inone case a tendencyis inhibitednot by another
opposed tendency
but simplyby the fact that for somereason,
whether physical or mental,it cannotreach completion.
This is the
situation ofthe inveteratesmoker inthe examplegiven earlier.In
the other case two tendencies which cannot both reach fruition at
the sametime arebrought intoplay almostsimultaneously. Ifthey
are aboutequal instrength, each tendency willblock thecompletion
of the other. Theresult isnot onlyaffect, asa productof inhibition,
but doubt, confusion, anduncertainty as well.
These latterconcomitants ofconflict areof importancebecause
they maythemselves become the basisfor further tendencies. For
to the human mind such states of doubt and confusion are abhorrent;
and, whenconfronted with them, the mind attemptsto resolve them
into clarity and certainty,even if this meansabandoning allother
previously activatedtendencies.
Thus confusion and lack of clarity, growing out of conflicting
tendencies, may
themselves become
stimuli producingfurther tend-
encies-tendencies toward clarification-which may become inde-
pendent of the originally conflicting tendencies.
Such tendencies
need not be definite in the sense that the ultimate resolution of the
doubt andconfusion isspecified. Some resolution ofthe confusion
may bemore importantthan thisor thatparticular solution, assum-
ing that the final result is not in conflict withother aspects
of the
stimulus situation or other mental sets.
Furthermore, it should be noted that uncertainty and lack of
clarity may be products not only of conflicting tendencies but also
of a situation which itself is structurally confused and ambiguous.
This is of capital importance becauseit indicates that a situation
which is structurally weak and doubtful in organization may directly
create tendenciestoward clarification. Delay in such a generalized
tendency toward clarification may also give rise to affect.
Although the main tenets of the psychological theoryof emotions
have been widely accepted, there have, needless tosay, been criti-
cisms of the theory. In the main these have come from those who
have sought, as yet without success, to account for, describe, and
distinguish emotions in purely physiological terms. The theory of
emotions, it is objected, does not tell us what an emotion is; it does
not tell us precisely
what takesplace inthe bodyto makeus feel.
This objection, though valid, is irrelevant for our purposes. For
just asthe physicistlong definedmagnetism in
terms ofthe lawsof
its operation and was able to deal with the phenomena without
knowing thenature ofthe magneticstates so, too, thepsychologist
can defineemotion in terms of the laws governing itsoperation,
without stipulatingprecisely what,in physiologicalterms, consti-
tutes feeling-what makes affectfelt.
THE DIFFERENTIATION OF AFFECT

Thus farwe haveconsidered emotionas thoughit were a general,
undifferentiated response,
a feeling whose characterand quality
Theory 17
were always more or less the same. While there is a good deal of
evidence for
this view,it is nevertheless clear
that incommon speech
and everyday experience we do recognize a variety of emotional
states-love, fear,anger, jealousy,
and thelike. The whole problem
of whether undifferentiated feelings,
affects perse, exist,of their
relation todifferentiated emotional
experience, and of the basis for
such differentiation
is of importance inthe presentstudy. Forwhile
music theorists and aestheticianshave found it difficult to explain
how musicdesignates particular
emotions, theyhave foundit almost
impossible toaccount for the existenceof less specific affective
experience.
Were the evidence to show that each affector type of affect had
its ownpeculiar physiological
composition, then
obviously undiffer-
entiated feelingwould be out of the question.However, Wood-
worths summaryof the work in this Heldmakes itclear that this
is not the case.
The evidencein the case ofemotional affective! behavior the
term which will henceforth be used to designate the overt and ob-
servable aspects of emotional conduct! is more complex. Much
emotional behavior,though habitual and henceseemingly automatic
and natural, is actually learned. Becausethis aspect of behavior
serves in the main as a means of communication, it will be called
designative denotative! behavior. To this category belong most
of the postural sets,
facial expressions,
and motorresponses accom-
panying emotional behavior. Thoughdesignative behavior
is defi-
nitely and clearly differentiated,the differentiation is not a necessary
one andindicates nothingas to the possibledifferentiation ofthe
affect itself.
Other aspectsof affective behavior, suchas skeletaland muscular
adjustments, havebeen said to be automatic, natural concomitants
of theaffective response.
These willbe calledemotional reactions.
Supposing that such automatic reactions do exist, a fact that has
been debated,
it has not beendefinitely shown
that they are differ-
entiated asbetween typesof affective experience.
However, even if it were demonstrated that emotional reactions
were differentiated,
this wouldnot necessarily
prove oreven indicate
that the affects whichthey accompanyare alsodifferentiated. For
the reactionis aresponse made to the total emotion-provoking
situ-
ation andnot necessarilya productof affectitself. In other words,
it may well be that such automatic behavioris called forth by
the peculiarnature of the objectivesituation ratherthan by the
operation ofthe law of affectitself. Werethis the case, sucha re-
action would be independentof affect and might indeed take
place, asdoes designative emotional behavior, in the absence of
affect.
The suppositions that behaviorreactions are essentially undiffer-
entiated, becoming characteristic onlyin certainstimulus situations,
and that affect itselfis basicallyundifferentiated aregiven added
plausibility whenone considers the following:
a! The more intense emotional behavioris, andpresumably there-
fore the more intense the affective stimulation, the less the control
exerted bythe egoover behaviorand the greater theprobability
that the behavior is automatic and natural.
19! The more intenseaffective behavior is, the less differentiated
such behaviortends tobe. In general, thetotal inhibitionof power-
ful tendenciesproduces diffuseand characterless activity. For ex-
ample, extreme conflict mayresult ineither complete immobility or
in frenziedactivity, whileweeping mayaccompany deepest grief,
tremendous joy, or probablyany particularlyintense emotion.
c! Thus the more automatic affective behavior is, the less differ-
entiated it tends to be.
It seems reasonable then to conclude that automatic reflex reac-
tions notonly failto providereasons for
believing thataffect itselfis
differentiated butthe evidenceseems topoint to just the opposite
conclusion.
Finally, ourown introspectiveexperience and
the reportsof the
experiences of others testify to the existence of undifferentiated
emotions. Itis affectas suchwhich Cassireris discussingwhen he
writes that Art gives us the motions of the human soul in all their
depth andvariety. Butthe form,the measure and rhythm,of these
motions isnot comparable to any single stateof emotion.What we
feel in art is not a simple orsingle emotionalquality. It is the dy-
namic processof life itself. 2
The conclusion that affect itself is undifferentiated does not mean
Theory
that affectiveexperience is a kind of disembodiedgenerality.+or
the affectiveexperience, asdistinguished from affectperse,includes
an awareness and cognitionof a stimulussituationwhich alma s
involvesparticularrespondingindividualsandspecificstimuli.
Not only do we becomeawareof andknowour own emotion<in
termsof a particularstimulussituationbut we interpretandcharac
terizethe behaviorof othersin theseterms."Whenan organismjn
a situationwhich resultsin a disturbedor wrought-upcondition7
thenthe situationPlusthe reactiongivesusthe nameor word which
characterizes the wholeas a specificemotion.The reactionitself is
not sufficientto differentiatethe emotion,the characterof the sigua
tion is involved in this differentiation." "
Thus while affectsand emotionsare in themselves undiffereut;
ated,affectiveexperience
is differentiatedbecause
it involvesaware
ness
andcognition
ofa stimulus
situation
whichitselfisnecessarily
differentiated.The affectivestatesfor which we have namesare
groupedandnamedbecause of similaritiesof the stimulussituation
not becausethe affectsof differentgroupsareper sedifferent,L>ve
and fear are not differentaffects,but they are differentaffective
experiences.
Awareness of thenatureof the stimulussituationalsoseems tq be
the real basisfor thedistinctionwhich Hebb drawsbetween"pleas
ant" and "unpleasant" emotions.Accordingto the presentanalysis
there areno pleasantor unpleasantemotions.Thereareonlypl<as
ant or unpleasant emotionalexperiences. This is of importancein
understanding thedistinctionmadeby Hebb.
Accordingto Hebb,the differencebetweenpleasantandunpltas
ant emotionslies in the fact that pleasantemotions or, in aur
terminology,pleasantemotionalexperiences!are alwaysresol>ed
They depend"on first arousingapprehension, then dispellingit" 22
But werethisactuallythecasewe couldonlyknowwhetheran ego
tion were pleasantor unpleasantafter it was over.Yet, surely,we
knowmorethanthiswhilewe areexperiencing affect.Thepleasant
nessof an emotionseems to lie not somuchin the fact of resolutIon
itself asin thebeliefin resolution the knowledge,whethertrut or
false,that therewill be a resolution.It is not, as Hebb seem~ to
assert when he cites as pleasurable the "mildly frustrating or g e
mildly fear-provoking,23 thecontrol actuallyexercised over
a situa-
tion which distinguishes pleasantfrom unpleasantemotions. Itis
the control which is believed to exist over the situation.
The sensationof falling through space,unconditioned byany
belief or knowledge asto the ultimate outcome,will, for instance,
arouse highlyunpleasant emotions. Yet a similar fall experienced
as a parachute jumpin an amusement park may, becauseof our
belief inthe presence
of controland inthe natureof the resolution,
prove most pleasurable.
The foregoinganalysis isof genuineimportance inthe present
study becauseit explainsand accounts for the existence and
nature
of the intangible, non-referentialaffective statesexperienced in
response tomusic. For in so far as the stimulus situation, the music,
is non-referential in the sense that it pictures, describes, or sym-
bolizes noneof the actions, persons,
passions, and
concepts ordi-
narily associated with human experience!, there is no reason to
expect thatour emotional
experience of
it shouldbe referential.The
affective experience
made inresponse to
music isspecific anddiffer-
entiated, but it is so in terms of the musical stimulus situation rather
than in terms of extramusical stimuli.
In the light of this discussionit is evident that, though it is wrong
to assert, as some have done, that emotions exist which are sui
generis musicalor aesthetic, it is possible to contend that there are
emotional experiences
which areso. By the sametoken, however,
any numberof emotionalexperiences canbe groupedtogether so
long astheir stimulus situations arein somerespects similar.Musical
affective experiences,for example, might be differentiated into
operatic, orchestral,
baroque, andso forth.But the most significant
distinction would still lie in the fact that musical stimuli, and hence
musical affective experiences, arenon-referential.
EMOTIONAL DESIGNATION
Although emotionalbehavior is frequently characterlessand
diffuse, often it is differentiated andintelligible. Even without
knowledge ofthe stimulussituation, motorbehavior, facialexpres-
sion, toneof voice, and mannerof speakingcan tell us not only
that anindividual isresponding inan emotional way but also some-
Theory 21
thing of the characterof his feelings or,more accurately,
of the
character ofhis adectiveexperience.
Differentiated behavior, as we have seen, is not an automatic or a
necessary concomitant
of affectitself oreven ofaffective experience.
The more automatic behavior is, the less likely it is to be diHeren-
tiated. Differentiation
involves control,
and controlimplies purpose.
The purpose of emotionallydifferentiated behavior
is communica-
tion. The individual responding, having anaffective experienceor
simulating one,seeks to make othersaware ofhis experiencethrough
a seriesof non-verbalbehavioral signs. Because the gestures and
signs whichdifferentiate suchbehavior arepurposeful, thismode
of behaviorwill be called emotionaldesignation ordesignative
behavior. 2
Such signsnot only act as cues forappropriate behavior in the
social situationbut areprobably, atleast inpart, aimedat making
other individualsrespond inan empathetic way. Asthe sayinggoes:
Misery lovescompany. Andso doother emotional states. Notonly
do wedislike physical isolation, butwe wantto shareour emotional
life with others. And,indeed, suchsharing does take place.For an
observer, recallinga situationin his own experiencesimilar to the
one signiiied by the behavior of another, may respond to the re-
membered situation in an affective way.Though designativeaffec-
tive behaviormay, throughconstant use, become habitual
and auto-
matic sothat it is almostinvariably calledup aspart of the total
emotional response,it is not basicallya necessary
concomitant of
the
response butone broughtinto play as aresult ofa desireto com-
municate.
Designative behavioris differentiated largely by custom and
tradition. It varies from culture to culture and among different
groups within a single culture. 'This does not mean that there are
no features of such behavior which are natural and widespread.In
all probabilitythere are.However, threepoints shouldbe kept in
mind: ! There is no real evidence to show that there is only one
Single naturalmode of behavior relevant to a given stimulus situa-
tion. When alternative modes of behavior are possible, cultural
Selection probably determines thecomposition ofany particular
pattern of affective designation.
! Whatever natural tendencies
toward aparticular patternof behaviorexist maybe alteredor sup-
pressed bythe demandsof the larger behaviorpatterns of the
culture. ! Even where natural behavior is retained in the pattern
of emotionalbehavior, it inevitably becomes
codilied and stand-
ardized for the sake of more eflicient communication.
Above all, we understand and make appropriate responses to
designative behavior as a total behaviorpattern, notjust to some
features ofit, whethernatural orotherwise. As
a totalpattern desig-
native behavior is a cultural phenomenon, not a natural one. It is,
in the Hnal analysis,learned.
This is important as it takes most of the sting out of the criticism
that music which attempts to designate emotional states depends
for its effect uponthe learningof conventionalsigns andsymbols.
For this fact is not peculiar to music but is characteristic ofall
emotional designation. If one excludes suchdesignation asa legiti-
mate means of musicalcommunication, one must bythe sametoken
exclude itas ameans ofhuman communication in general.
The Theory of Emotions Related to

Musical Experience
AN ASSUMPTION
An examinationof the psychological theoryof emotionswas made
because theevidence furnished by the introspections of musicians,
aestheticians, and
listeners and
the objectivedata gathered
from the
observation ofbehavior and the study of the physiological responses
to musical stimuli did not yield reliable information about the musi-
cal stimulus or the affective responsesmade to it. Implicit in this
examination wasan assumptionwhich must now be made explicit:
Though the stimulus situation may vary indefinitely, the conditions
which evoke affect are general and are, therefore, applicable to
music. In other words, it was assumed that the law of affect, which
states that emotion is evoked when a tendency to respond is in-
hibited, is a generalproposition relevant
to human psychology in
all realms of experience.
This assumption
does not,however, imply
or stipulatethat musical
Theory 23
affective experiences
are thesame asthe affectiveexperiences made
in responseto other stimulus situations.
Musical experiencediffers
from non-musical or, morespecifically, non-aesthetic
experience in
three important ways.
First, as we haveseen, affectiveexperience includesan awareness
and knowledge
of thestimulus situation.
This beingso, theaffective
experience of music will differ from other types of aEective ex-
perience, particularly
in sofar asmusical stimuli are non-referential.
Second, ineveryday experience the tensionscreated bythe in-
hibition of tendencies oftengo unresolved. They aremerely dissi-
pated inthe pressof irrelevantevents. Inthis sense daily experience
is meaningless and accidental.In art inhibition oftendency becomes
meaningful because the relationshipbetween thetendency andits
necessary resolutionis made explicit and apparent. Tendenciesdo
not simply cease toexist: they are resolved,they conclude.
Third, in life the factors whichkeep atendency fromreaching
completion maybe different in kind from those which activated the
tendency inthe Hrstplace. Thestimulus activating
a tendencymay,
for example,be a physical orpsychic needof the organism, while
the inhibiting factors maysimply be a seriesof external circum-
stances whichkeep the organism from satisfying the need. This is
the situation in the case of the habitual smoker who can End no
cigarette. Or the situation may be reversed; that is, a tendency
activated byan externalstimulus maybe inhibitedby the psychic
processes ofthe organism.
Furthermore, in everyday experiencethe resolutions of the ten-
sions brought
into playby inhibitionmay beirrelevant tothe tend-
encies themselves.Tensions arising from psychic needs may be
worked off in sheerphysical activitywhich iswithout meaningful
relation tothe originalstimulus orto thetendency itself.
In music, on the other hand, the same stimulus, the music, acti-
vates tendencies,inhibits them, and provides meaningful and rele-
vant resolutions.
TENDENCY AND EXPECTATION IN MUSIC

The assumption that the same basicpsychological processes
un-
derlie allaffective responses,
whether thestimulus bemusical orof
some other kind, has been implicit in much musical theory and in
the speculations of many aestheticians. Butthis doesnot in itself
increase our understanding of
the natureof musicalexperience and
of the musical processeswhich formit. It does notexplain thena-
ture of the relationshipswhich exist between the stimulus, the
listeners perceptionsand mentalprocesses, and his responses.To
do thisit is necessary to
demonstrate preciselyhow musicalstimuli
do, in fact, arouse and inhibit tendencies andthereby give rise to
emotions.
What is meant by a tendencyto respond?A tendency or, as
MacCurdy usesthe term, an instinct is a pattern reactionthat
operates, or tends to operate, when activated, in an automatic
way. 2 A pattern reaction consists
of a set or series ofregularly
coincident mentalor motorresponses which,once broughtinto play
as part of the response toa given stimulus, followa previously
ordered course,unless inhibited or blocked in some way. The order
established bya pattern reaction isboth temporaland structural;
that is, the seriesinvolves notonly therelation ofthe partsof the
total pattern to each other but also their timing. Thus a seriesmay
be disturbed either becausethe successionof the parts of the pat-
tern isupset orbecause the
timing of the seriesis upsetor both.
The term tendency, as used in this study,comprises allauto-
matic responsepatterns, whether natural or learned. Sincehabit is
a mechanismof action,physiologically ingrained,
which operates
spontaneously andautomatically, 3 the term tendency also in-
cludes habitreactions and,
inevitably, acquired
concepts and
mean-
ings. 31
The tendency to respondmay beeither conscious or unconscious.
If the pattern reactionruns its normal courseto completion,then
the whole process maybe completelyunconscious. Countless re-
action patterns,of which the respondingindividual isunaware, are
initiated and completed each hour. The more automatic behavior
becomes, theless consciousit is. The tendencyto respondbecomes
conscious whereinhibition ofsome sort is present,when thenormal
course ofthe reactionpattern is disturbed orits final completion
is inhibited. Such conscious and self-conscious tendencies are often
thought ofand referredto asexpectations.
Theory 25
In a broader sense all tendencies, even those which never reach
the levelof consciousness,
are expectations.
For sincea tendencyis
a kindof chainreaction inwhich apresent stimulus
leads througha
series ofadjustments toa more or less specified consequent,
the
consequent isalways impliedin the tendency, once the tendency
has beenbrought intoplay. Thuswhile ourconscious minds do not
actively expecta consequentunless thepattern reactionis disturbed,
our habits and tendenciesare expectantin the sense thateach seeks
out or expects the consequents relevantand appropriate to itself.
Though he may never become aware of his expectations as he
reaches in his pocket for a pack of cigarettes, the behavior of the
habitual smokershows thathe doesexpect or,perhaps moreac-
curately, his habits expectfor him.
If tendencies are pattern reactions that are expectantin the broad
sense, includingunconscious aswell as conscious anticipations,then
it is not difficult to see how music is able to evoke tendencies. For
it hasbeen generally
acknowledged that
music arouses
expectations,
some consciousand others unconscious, whichmay or may not be
directly andimmediately satisfied.
. . _ the pleasure .. . arises fromthe perceptionof the artists play
with forms and conventions which are ingrained ashabits of perception
both in the artist and his audience. Without such habits . . . there
would be no awareness whatever of the artists fulfillment of and subtle
departures from established forms .... But the pleasure which we
derive from style is not an intellectual interestin detecting similarities
and differences,but an immediate aesthetic delight in perception which
results from the arousaland suspensionor fulfillment of expectations
which arethe productsof manyprevious encounters with works of art.
EXPECTATION, SUSPENSE, AND THE UNEXPECTED
Sometimes avery speciiicconsequent isexpected. InWestern
music ofthe eighteenthcentury, forexample, weexpect aspecific
chord, namely,
the tonic C major!, to follow this sequence
of
-lr
EXAMPLE 1
harmonies see Example 1!. Furthermore, t_heconsequent chordis
expected toarrive ata particulartime, i.e.,on theHrst beatof the
next measure.
Cf course,the consequent which is actually forthcoming,
though
it must be possiblewithin the style, neednot bethe onewhich was
specifically expected.Nor is it necessarythat theconsequent arrive
at the expected time. It may arrive toosoon orit may be delayed.
But no matter whichof theseforms theconsequent actually takes,
the crucialpoint tobe notedis thatthe ultimateand particulareffect
of the total patternis clearly conditioned bythe specificityof the
original expectation.
At other times expectationis more general; thatis, though our
expectations may be deHnite,
in the sense of
being marked, they are
non-specific, inthat we are not sure preciselyhow they will be
fulfilled. The antecedent stimulussituation maybe suchthat several
consequents may
be almostequally probable. For instance,after a
melodic fragmenthas beenrepeated several times, we begin to
expect achange and
also thecompletion ofthe fragment.A change
is expectedbecause webelieve that the composeris not so illogical
as torepeat thefigure indefinitelyand because
we lookforward to
the completion of the incomplete figure. But precisely what the
change will be or how the completion will be accomplishedcannot
perhaps be anticipated. The introductions to many movements
written in the eighteenth or nineteenth centuries createexpectation
in this way, e.g.,the openingmeasures of
Beethovens NinthSym-
phony or the opening measures ofthe March to the Gallows from
Berliozs SymphoniaF antastique.
Expectation may also result because the stimulus situation is
doubtful or ambiguous. If the musical patterns are less clear than
expected, ifthere isconfusion as to therelationship between melody
and accompaniment, or if our expectationsare continuallymistaken
or inhibited, then doubtand uncertaintyas to the generalsignifi-
cance, function,and outcome of the passage willresult. Aswe have
already seensee pp. 15 ff.!, the mind rejects andreacts against
such uncomfortable states and,if they are morethan momentary,
looks forwardto and expects areturn tothe certaintyof regularity
and clarity.This is particularly strikingin the responses made to
Theory 27
works of art where,because ofa firm belief in the purposefulness
and integrity of the artist, we expect thatorder will in the end
triumph, andprecision willreplace ambiguity.
However, the manner in which clarification and order will be
restored maynot be predicted or envisaged. Expectation is not
specific; thestate isone of suspense. In fact, if doubt anduncer-
tainty arestrong enough,almost anyresolution, withinthe realmof
probability, whichreturns usto certaintywill be acceptable, though
no doubtsome resolutions will, given the style,seem morenatural
than others.
The inclusionof suspense arising outof uncertaintymay, atfirst
sight, appearto be an extensionand amplificationof the concept
of arrestand inhibitionof a tendency. Butwhen thematter iscon-
sidered morecarefully, itwill be seen thatevery inhibitionor delay
creates uncertainty
or suspense, if only briefly, becausein the mo-
ment of delay we become aware of the possibility of alternative
modes of continuation. The difference is one of scale and duration,
not of kind. Both arouse uncertainties
and anxietiesas to coming
events.
Suspense is essentially aproduct of ignorance asto the future
course ofevents. This
ignorance may arise either
because the present
course ofevents, thoughin a sense understandablein itself,presents
several alternativeand equallyprobable consequents or because
the presentcourse ofevents isitself sounusual andupsetting that,
since it cannot be understood, nopredictions as to the future can
be made.
From the outset ignorancearouses strongmental tendencies to-
ward clarificationwhich are immediately affective. If ignorance
persists inspite ofall, thenthe individualis throwninto a state of
doubt and uncertainty see pp. 15-16!. He commencesto sensehis
lack of control overthe situation,his inabilityto act on the basis
of the knowledge whichhe supposed that he possessed. In short,
he beginsto feel apprehensive, even fearful, thoughthere is no
object for his fear. Ignorance andits concomitantfeelings ofim-
potence breed apprehension andanxiety, evenin music. But igno-
rance alsogives riseto more sanguine feelings;
for since the outcome
cannot beenvisaged, itmay bepleasant. These
feelings arethem-
selves tendencies the avoidanceof painful apprehension and the
expectation of
a propitiousconclusion! which become focused upon
an expectedresolution ofthe unpleasantstimulus situation.
The longerdoubt anduncertainty persist, the greaterthe feeling
of suspensewill tend to be_.The stimulussituation creatingdoubt
and uncertaintymust, of course, beprogressively intensified if
suspense is
to be maintained orincreased. Foras we become ac-
customed toa givenstimulus situation,even anunpleasant one, its
effectiveness tendsto diminish. Moreover, without a change in the
stimulus situation in the direction of complication and uncertainty,
those vitalanticipatory feelingsthat a break mustcome, thatdoubt
and perplexitymust giveway to knowledge! whichmake usexpect
both apprehensively and hopefully!would belost.
The greaterthe buildup of suspense, of tension,the greaterthe
emotional release upon resolution.This observationpoints up the
fact that in aestheticexperience emotional pattern mustbe con-
sidered notonly in terms of tension itselfbut alsoin terms of the
progression from tension torelease. And the experience of suspense
is aesthetically
valueless unlessit is followed bya releasewhich is
understandable inthe given context.
Musical experiences
of suspense
are verysimilar tothose experi-
enced inreal life. Both in life and in music the emotionsthus arising
have essentially
the samestimulus situation:the situationof igno-
rance, the awareness ofthe i.ndividuals impotence and inability to
act where the future course of events is unknown. Because these
musical experiences are so very similar to those existing in the
drama andin life itself, theyare oftenfelt to be particularlypower-
ful and effective.
Musical suspense seems tohave directanalogies inexperience in
general; itmakes us feel somethingof theinsignilicance and power-
lessness of man in the faceof the inscrutable workings of destiny.
The low, foreboding nimbleof distant thunder on an oppressive
summer afternoon, its growing intensity asit approaches,the cre-
scendo ofthe graduallyrising wind,the ominousdarkening ofthe
sky, all give riseto an emotional experience in which expectation
is fraughtwith powerfuluncertainty-the primordialand poignant
uncertainty of human existencein the face of the inexorable forces
Theory 29
of nature. With mixed feelings of hope and apprehension in the
presence of
the unknown,
we anxiously
await thebreaking ofthe
storm, thediscovery ofwhat unrelentingfate hasdecreed.
Similarly inmusic thestate ofsuspense involves
an awareness
of
the powerlessness
of manin the face ofthe unknown.
What is expected in this state of suspensemay not be specified,
but this does not mean that any consequent is possible. Our ex-
pectations areinevitably circumscribedby the possibilities and
probabilities ofthe styleof the composition inquestion. Thecon-
sequent must,given the circumstances, be possible within what
Aiken hascalled an ordering system of beliefsand attitudes.33
Although the consequent in any musical sequence must,in this
sense, bepossible, it may neverthelessbe unexpected. But the un-
expected shouldnot be confused with the surprising. For when ex-
pectation isaroused, theunexpected isalways consideredto be a
possibility, and,
though itremains theless expected
of severalalter-
natives, it is not a complete surprise. Conditions of active expecta-
tion especially general expectation and suspense! are not the
most favorable
to surprise.For thelistener ison guard,anticipating
a new and possibly unexpected consequent.Surprise is most in-
tense where no special
expectation is active, where, because there
has
been noinhibition ofa tendency,continuity isexpected.
As soonas theunexpected, or for thatmatter thesurprising, isex-
perienced, thelistener attempts to fit it into the generalsystem of
beliefs relevant
to the style ofthe work.This requiresa very rapid
re-evaluation of either the stimulus situation itself or its cause-
the eventsantecedent tothe stimulus.Or it may requirea review
of thewhole system
of beliefsthat thelistener supposed
appropriate
and relevant to the work. If this mental synthesis doesnot take
place immediately,
three thingsmay happen:! The mind may
suspend judgment,so to speak, trustingthat what follows will clarify
the meaningof the unexpected consequent.
! If no clarification
takes place,the mindmay rejectthe wholestimulus andirritation
will set in. ! The unexpectedconsequent maybe seenas a pur-
poseful blunder. W'hether the listener respondsin the first or third
manner will depend partly upon the character ofthe piece, its mood
or designative content. The third response might well be made to
music whose character was comic or satirical. Beckmessers music
in Wagners Die Meistersingerwould probablyelicit this type of
interpretive understanding.In a piece whosecharacter admitted
no suchpurposeful blunders, the secondresponse wouldprobably
be elicited.
CONSCIOUS AND UNCONSCIOUS EXPECTATIONS

In the light of these observationsit is clear thatan expectation
is
not a blind, unthinkingconditioned reflex. Expectation frequently
involves ahigh orderof mental activity. Thefulfilment of a habit
response, in art aswell asin daily life, requiresjudgment andcog-
nition both of the stimulus itself and of the situation in which it
acts. Thestimulus asa physicalthing becomesa stimulusin the
world of behavior onlyin sofar asthe mindof theperceiver isable
to relate it, on the one hand, to the habit responses whichthe per-
ceiver hasdeveloped and,on the other hand, to the particular
stimulus situation. This is clear as soon as one considers that the
same physical stimulus maycall forth different tendenciesin differ-
ent stylisticcontexts orin different situations withinone andthe
same stylistic context. For example, a modal cadential progression
will arouseone setof expectationsin the musical styleof the six-
teenth centuryand quite another in the style of the nineteenth
century. Likewise the same musical progressionwill evoke one set
of expectations at the beginning of a piece and another at the
end.
Expectation then is a product of the habit responses developed
in connection with particular musical styles and of the modes of
human perception,
cognition, andresponse-the psychological
laws
of mental life.
The mental activity involved in the perception of and response
to music need not, however, be conscious. . . . the intellectual
satisfaction whichthe listener derives fromcontinually following
and anticipatingthe composersintentions-now, to see hisexpec-
tations fulfilled,and now,to see himself agreeably mistaken .. .
this intellectualflux andreflux, thisperpetual givingand receiving
takes place unconsciously, andwith the rapidity of lightning
flashes. 3'So long as expectations are satisBedwithout delay,so
Theory 31
long as tendencies areuninhibited, thoughintelligence isclearly
and necessarilyinvolved in the perceptionand understandingof
the stimulussituation, theresponse willprobably remainuncon-
scious.
Mental activitytends to become conscious
when reflectionand
deliberation are involved in the completion of the response pattern,
that is, when automaticbehavior isdisturbed because a tendency
has beeninhibited. Impulsionforever boostedon its forward way
would run its course thoughtless, and dead to emotion .... The
only way it can become awareof its nature andits goal is by
obstacles surmountedand means employed. 38
But even when a habit responseis inhibited, conscious awareness
of the mental activityinvolved in the perceptionof and response
to the stimulus situationis by no meansinevitable. Intellectual
experience the conscious awareness of ones own expectations or,
objectively, of the tendencies of the music!, as distinguished from
intellectual activity,
is largelya productof thelisteners ownattitude
toward his responses and hence toward the stimuli and mental
activities which bring them into existence. That is to say, some
listeners, whetherbecause oftraining or natural psychologicalin-
clination, aredisposed to
rationalize their
responses, make
to experi-
ence self-conscious;
others arenot sodisposed. Ifintellectual activ-
ity is allowed to remain unconscious,then the mental tensions and
the deliberations
involved whena tendencyis inhibitedare experi-
enced asfeeling or affect ratherthan asconscious cognitionsee
pp. 38 f. !.
Having shownthat musicarouses tendencies
and thusfulfils the
conditions necessary
for the arousal ofaffect seep. 22! and having
demonstrated how this is accomplished, we can nowstate oneof
the basichypotheses of this study.Namely: Affector emotion-felt
is arousedwhen anexpectation-a tendency to respond-activated
by the musical stimulus situation, istemporarily inhibitedor per-
manently blocked.
As notedearlier seepp. 22-23! in musicalexperience the
same
stimulus, the music, activates tendencies, inhibits them, and pro-
vides meaningfuland relevantresolutions for
them. Thisis of par-
ticular importance
from a methodological standpoint.
For it means
that grantedlisteners who
have developed
reaction patterns
appro-
priate to the work in question,the structure of the affective response
to a piece ofmusic canbe studiedby examiningthe musicitself.
Once those sound successions
common to a culture, a style, or a
particular workhave beenascertained, then,
if the customary suc-
cession ispresented andcompleted withoutdelay, it can be as-
sumed that, since no tendency would have been inhibited, the
listener would not respond in an affective way. If, on the other
hand, the sound succession fails to follow its customary course,or if
it involves obscurity or ambiguity, then it can be assumedthat the
listeners tendencieswould be inhibited or otherwise upsetand that
the tensionsarising inthis processwould beexperienced as
affect,
provided that they were not rationalized as consciousintellectual
experience.
In other words, thecustomary orexpected progression
of sounds
can be considered asa norm, which from a stylistic point of view
it is; and alteration in the expected progressioncan be considered as
a deviation. Hence deviations can be regarded as emotional or
affective stimuli.
The importanceof this objective point of view of musicalex-
perience is clear. It means that once the norms of a style have been
ascertained, the study and analysis of the affective content of a
particular work in that style can be made without continual and
explicit reference to the responses ofthe listener or critic. That is,
subjective content
can bediscussed objectively.
The Meaningof Music

THE PROBLEM OF MEANING IN MUSIC
The meaningof musichas oflate beenthe subjectof muchcon-
fused argumentand controversy.The controversyhas stemmed
largely from disagreements asto what music communicates,while
the confusionhas resultedfor the most partfrom a lack of clarity
as to the natureand definitionof meaningitself.
The debates as to what music communicates have centered around
the questionof whethermusic candesignate, depict,
or otherwise
communicate referential
concepts, images,
experiences, and emo-
Theory 33
tional states.This isthe old argument between the absolutistsand
the referentialistssee pp.1 f.!.
Because ithas notappeared problematical to them, the referen-
tialists havenot asa rule explicitly consideredthe problemof mu-
sical meaning. Musical meaning according tothe referentialists
lies
in the relationship between a musicalsymbol orsign andthe extra-
musical thing which it designates.
Since ourconcern inthis study is not primarily with the refer-
ential meaningof music, suffice it to say that the disagreement
between thereferentialists andthe absolutistsis, as was pointed
out atthe beginning of thischapter, theresult ofa tendencytoward
philosophical monism rather thanthe resultof any logical inc0m_
patibility. Bothdesignative and non-designative meaningsarise out
of musicalexperience, just as they do in other typesof aesthetic
experience.
The absolutistshave contendedthat the meaning ofmusic lies
specifically, and
some wouldassert exclusively,
in the musical prog-
esses themselves.For themmusical meaningis non-designative.
But
in what sense theseprocesses aremeaningful, in what sense a suc-
cession or sequence ofnon-referential musical stimuli can be said
to give rise tomeaning, they have beenunable tostate witheither
clarity or precision. Theyhave alsofailed to relate musicalmean-
ing to other kindsof meaning-to meaning ingeneral. Thisfailure
has ledsome critics to assertthat musicalmeaning isa thing apart,
different insome unexplained
way fromall otherkinds ofmeaning.
This is simply an evasion of the real issue. For it is obvious that if
the termmeaning is to haveany signification
at all as appliedto
music, thenit musthave thesame signihcation
as whenapplied to
other kinds of experience.
Without reviewingall the untenable positionsto which writers
have tenaciouslyadhered, it seems fairto say that much of the
confusion and uncertainty as to the nature of non-referential mu-
sical meaninghas resultedfrom two fallacics. Onthe one hand,
there hasbeen a tendency to locate meaningexclusively inone
aspect of the communicative process; onthe other hand, there has
been a propensity toregard all meanings arising
in human com-
munication as designative, as involving symbolism of some sort.
Since these
difficulties canbe bestresolved inthe light of a gen-
eral delinition of meaning, let us begin with such a definition:
. . . anything acquires meaning if it is connectedwith, or indi-
cates, orrefers to,something beyond itself, sothat its full nature
points toand isrevealed inthat connection.4
Meaning isthus nota propertyof things.It cannotbe locatedin
the stimulus alone. The same stimulus may have many different
meanings. Toa geologista largerock mayindicate thatat onetime
a glacierbegan torecede ata givenspot; toa farmerthe samerock
may pointto the necessity ofhaving theHeld cleared for plowing;
and to the sculptor the rock may indicate the possibility of artistic
creation. A rock, a word, or motion in and of itself, merely as a
stimulus, is meaningless.
Thus it is pointlessto askwhat the intrinsic meaningof a single
tone or a seriesof tonesis. Purelyas physicalexistences they are
meaningless. They become meaningful only in so far as theypoint
to, indicate,or imply something beyond themselves.
Nor can meaning be located exclusivelyin the objects, events,or
experiences whichthe stimulus indicates, refersto, or implies. The
meaning ofthe rockis the product ofthe relationshipbetween the
stimulus andthe thingit points to or indicates.
Though theperception ofa relationshipcan only arise asthe
result of some individualsmental behavior,the relationshipitself
is not to be located in the mind of the perceiver. Themeanings
observed are not subjective.
Thus therelationships existing
between
the tones themselves or those existing between the tones and the
things theydesignate orconnote, though a productof cultural ex-
perience, arereal connections
existing objectively
in culture. They
are notarbitrary connections
imposed bythe capricious mind ofthe
particular listener.
Meaning, then,
is not in either the stimulus,or what it points to,
or the observer. Rather it arises out of what both Cohen and Mead
have called the triadic relationship between! an object or
stimulus; ! that to which the stimulus points-thatwhich is its
consequent; and! the conscious observer.
Discussions of the meaningof musichave alsobeen muddledby
the failureto stateexplicitly whatmusical stimuliindicate orpoint
Theory 35
to. A stimulus mayindicate eventsor consequences which are
different fromitself in kind, as when a word designatesor points
to an object oraction whichis not itself aword. Ora stimulusmay
indicate or imply eventsor consequences which are of the same
kind as the stimulusitself, as when a dim light on the eastern
horizon heralds the comingof day. Here boththe antecedentstim-
ulus andthe consequent event arenatural phenomena. The former
type of meaning maybe calleddesignative, the latter embodied.
Because mostof the meanings which arise in human communica-
tion are of the designative type,employing linguisticsigns or the
iconic signsof the plastic arts, numerous criticshave failed to
realize thatthis isnot necessarily
or exclusivelythe case.This mis-
take hasled evenavowed absolutists to allow designation toslip in
through thesecret doorof semanticchicanery.
But evenmore importantthan designativemeaning iswhat we
have called embodied meaning. From this point of view what a
musical stimulus
or a series ofstimuli indicateand pointto are not
extramusical conceptsand objects but other musical events which
are aboutto happen.That is, one musicalevent be it a tone, a
phrase, ora wholesection! hasmeaning becauseit points to and
makes us expect anothermusical event. This is what music means
from the viewpoint ofthe absolutist.
MUSIC AND MEANING

Embodied musical
meaning is,in short,a productof expectation.
If, on the basis of past experience, a present stimulus leads us to
expect amore orless definiteconsequent musical
event, thenthat
stimulus has meaning.
From this it follows that a stimulus or gesture which does not
point to or arouseexpectations ofa subsequent musical eventor
consequent is meaningless. Because
expectation is
largely aproduct
of stylistic experience, music
in a style with which we are totally
unfamiliar is meaningless.
However, oncethe aesthetic attitude has been brought into play,
very few gestures actuallyappear to be meaningless
so long as
the listener has someexperience with the style of the work in ques-
tion. For so long as a stimulus is possible within any known style,
the listenerwill do his bestto relateit to the style,to understand
its meaning.
In and of themselves,for example, the opening chords of Bee-
thovens ThirdSymphony have no particularmusical stylistictend-
ency. They establish nopattern of motion, arouseno tensions
toward a particular fulfilment.Yet as part of the total aesthetic
cultural act of attention they are meaningful. Forsince theyare
the first chords ofa piece,we not only expectmore musicbut our
expectations are circumscribed by the limitationsof the style which
we believethe pieceto be in andby the psychological demand for
a morepalpable patternsee chapsii-v!.
Thus thephrase pastexperience, used in thedefinition ofmean-
ing givenabove, mustbe understoodin a broad sense. It includes
the immediatepast ofthe particularstimulus orgesture; thatwhich
has alreadytaken placein this particular work to condition the
listeners opinionof the stimulus andhence hisexpectations as to
the impending,consequent event. In the example given above, the
past wassilence. Butthis fact of the past isjust aspotent in con-
ditioning expectation as awhole section of pastevents. Thephrase
past experiencealso refersto the more remote, but everpresent,
past experience of similarmusical stimuliand similarmusical situa-
tions in other works. That is it refers to those past experiences
which constituteour sense and knowledge of style.The phrasealso
comprehends the dispositions and beliefs whichthe listenerbrings
to the musical experiencesee pp. 73 if.! as well as the laws of
mental behavior which governhis organization of stimuli into pat-
terns andthe expectations aroused onthe basisof thosepattems
see chaps.and iv!.
The words consequent musicalevent must be understoodto
include: ! those consequents which are envisaged orexpected;
! the events which do, infact, follow thestimulus, whether they
were the ones envisagedor not; and ! the more distant ramifica-
tions or events which,because thetotal seriesof gesturesis pre-
sumed tobe causallyconnected, are considered as being thelater
consequences of the stimulusin question.Seen inthis light, the
meaning ofthe stimulusis not confined toor limitedby the initial
triadic relationship
out of which it arises. As
the laterstages ofthe
Theory 37
musical process
establish newrelationships with
the stimulus,
new
meanings arise.
These later
meanings coexistin memorywith the
earlier ones
and, combining
with them,
constitute the
meaning of
the work as a total experience.
In this development three
stages ofmeaning maybe distin-
guished.
Hypothetical meanings
are thosewhich arisedin-ing theact
of expectation.
Since what
is envisaged
is aprodnot of
the Prob-
ability relationships
which existas partof style gee nn 45 Hi
54 ff.!, and sincethese probability
relationships always
involve
the possibility of alternative consequences, given
a stimnlns in-
variably givesrise to several alternative
hypothetical 1'neaningS_
One consequent may, ofcourse, be so muchmore Probable than
any otherthat thelistener, thoughaware ofthe possibility
of less
likely consequences,
is reallyset andready only
for themost Prob-
able. Insuch acase hypothetical meaning iswithout ambiguity
In othercases several
consequents may be almost
equally Probable
and, since the listener is in doubt as to which alternative will
actually materialize,
meaning is
ambiguous, though
not necessarily
less forceful
and markedsee pp.51 ff.!.
Though theconsequent which
is actuallyforthcoming must
be
possible withinthe style,it may or maynot be one of those which
was mostprobable. Or
it mayarrive only
after adelay ora deceptive
diversion through
alternative consequences.
But Whether ont exPee_
tations areconfirmed ornot, a new stageof meaningis feaelied when
the consequentbecomes actualizedas aconcrete musicalevent,
Evident meanings are those which are attributed to the ante-
cedent gesture
when theconsequent becomes
a physico-psychic
fact and when the relationship between
the antecedentand oon-
sequent isperceived. Sincethe consequent
of a stimulus itselfbe-
comes a stimulus withconsequents, evidentmeaning alsoinolndes
the laterstages of
musical developmentwhich arepresumed to bg
the productsof achain ofcausality. Thus
in thefollowing seqnenoe,
where astimulus S! leads toa consequentC!, which is also a
stimulus that
indicates and
is actualized
in furtheroonseqnents,
S, ...... C,S2 ...... CZS3 ...... etc,
evident meaningarises notonly out of the relationship between
S, andC, but also outof the relationships between S1 andall sub-
sequent consequences, in so far as these areconsidered toissue
from Sl.It is also importantto realizethat the motion S,...... C,
may itselfbecome agesture thatgives riseto envisagedand actual
consequents and hence becomes a termor gestureon anotherlevel
of triadic relationships. Inother words,both evident and hypo-
thetical meanings come intobeing andexist onseveral architectonic
levels.
Evident meaningis colored and conditioned by hypothetical
meaning. Forthe actualrelationship between the gestureand its
consequent is always considered in the light of the expectedrela-
tionship. Ina sensethe listenereven revises his opinionof the hy-
pothetical meaningwhen the stimulus doesnot move to the ex-
pected consequent.
Determinate meaningsare thosemeanings whicharise outof
the relationshipsexisting betweenhypothetical meaning,evident
meaning, and the laterstages ofthe musicaldevelopment. In other
words, determinate meaning arises only after the experience of the
work is timeless inmemory, onlywhen allthe meanings which the
stimulus has had inthe particularexperience are realized andtheir
relationships toone anothercomprehended as fully as possible.
TI-IE OBIECTIFICATION OF MEANING

A distinctionmust bedrawn betweenthe understandingof mu-
sical meaningwhich involvesthe awareness
of the tendencies, re-
sistances, tensions, and fuliilments embodied in a work and the
self-conscious objectiflcation
of that meaning in the mind of the
individual listener.The formermay besaid toinvolve ameaningful
experience, the latter involvesknowing whatthat meaningis, con-
sidering it as anobjective thingin consciousness.
The operationof intelligencein listening to music need never
become self-conscious. We are continually behavingin an intelli-
gent way,comprehending meanings and acting upon our percep-
tions, cognitions,and evaluationswithout evermaking themeanings
themselves the objects ofour scrutiny-without ever becoming self-
conscious about what experience means. What Bertrand Russell
says
Theory 39
about understanding
language alsoapplies tothe understanding of
music: Understanding
language is. . . like understanding cricket;
it is a matterof habitsacquired inoneself and
rightly presumed
in others. 47
Meanings becomeobjectiiied only under conditions of self-
consciousness and when reflectiontakes place,Une attains self-
consciousness only as hetakes, orEnds himselfstimulated totake,
the attitudeof the other. B Though trainingmay makefor a gen_
erally self-conscious
attitude, oneis stimulatedto takethe attitude
of the other when the normal habits of response aredisturbed in
some way;when oneis driven to ask ones self;What doesthis
mean, whatis the intention ofthis passage? Reflection islikewise
brought intoplay wheresome tendency is delayed,some patternof
habitual behaviordisturbed. Solong asbehavior isautomatic and
habitual thereis no urge forit to become self~conscious,though it
may become so. If meaning isto becomeobjectified atall, it will as
a rule become so when difficulties are encountered that make nor-
mal, automaticbehavior impossible.In other words, givena mind
disposed toward objectification, meaning
will becomethe focusof
attention, anobject ofconscious consideration,
when atendency or
habit reaction is delayed or inhibited.
MEANING AND AFFECT

It thus appears thatthe sameprocesses which
were saidto give
rise toaffect are
now saidto giverise tothe objectification
of em-
bodied meaning.
But this is a dilemma onlyso longas thetraditional dichotomy
between reason and emotionand theparent polaritybetween mind
and bodyare adopted.Once it is recognizedthat affectiveexperi-
ence isjust as dependent uponintelligent Cognitionas conscious
intellection, thatboth involve perception, takingaccount of, en-
visaging, andso forth, then thinking and feeling need not be
viewed as polar oppositesbut as different manifestationsof a single
psychological process.
There is no diametricopposition, noinseparable gulf,between
the affectiveand theintellectual responses
made tomusic. Though
they are psychologically differentiated
as responses,
both depend
upon the same perceptive processes, the
same stylistichabits, the
same modes of mentalorganization; andthe samemusical processes
give riseto and shape bothtypes ofexperience. Seen in this light,
the formalists conception ofmusical experience and the expres-
sionists conception of it appear ascomplementary rather than con-
tradictory positions.They are considering notdifferent processes
but differentways ofexperiencing thesame process.
VVhether a piece ofmusic givesrise toaffective experience or to
intellectual experience depends uponthe dispositionand training
of the listener. Tosome mindsthe disembodied feeling of affective
experience is uncanny andunpleasant and a processof rationaliza-
tion is undertaken inwhich themusical processes are objectifiedas
conscious meaning. Belief alsoprobably playsan importantrole in
determining thecharacter ofthe response. Those whohave been
taught to believe that musical experience is primarily emotional
and who are thereforedisposed torespond affectivelywill prob-
ably do so. Thoselisteners whohave learnedto understandmusic
in technicalterms willtend tomake musical processes an object of
conscious consideration.This probably accounts for the fact that
most trainedcritics andaestheticians favor
the formalistposition.
Thus whilethe trainedmusician consciously
waits for the expected
resolution ofa dominantseventh chord
the untrained,but practiced,
listener feelsthe delay as affect.
MUSIC AND COMMUNICATION

Meanings and affects may, however, arise without communica-
tion taking place. Individual A observesanother individual B wink
and interpretsthe wink as afriendly gesture.It has meaning forA
who observes it. But if the wink was not intentional-if, for in-
stance, B simply has a nervous tic-then no communication has
taken place,for to B the act had no meaning.Communication, as
Mead haspointed out,takes placeonly wherethe gesturemade
has thesame meaning for theindividual whomakes itthat it has for
the individualwho responds to it.
It is this internalizationof gestures,
what Meadcalls taking the
attitude of the other 5 the audience!, which enables the creative
artist, the composer, tocommunicate withlisteners. It is becausethe
Theory 41
composer isalso alistener thathe is able tocontrol hisinspiration
with reference to the listener." For instance, the composer knows
how the listener will respond to a deceptive cadence and controls
the laterstages ofthe composition
with referenceto that supposed
response. The performer toois continuallytaking the attitude of
the other-of the listener.As Leopold Mozart puts it, the performer
must play everything insuch away thathe will himself bemoved
by it. 52
It is precisely because
he iscontinually takingthe attitudeof the
listener thatthe composer becomes aware and conscious of his own
self, hisego, inthe processof creation.In this process ofdifferen-
tiation betweenhimself ascomposer andhimself asaudience, the
composer becomes self-conscious and
objective.
But though the listenerparticipates inthe musicalprocess, as-
suming therole whichthe composer envisaged for him, andthough
he must,in somesense, create his ownexperience, yet he neednot
take the attitude of the composer in order to do so. He need not
ask: How will someoneelse respondto this stimulus? Noris he
obliged toobjectify hisown responses,
to ask,How am I respond-
ing? Unlike the composer,the listener may and frequently does
lose himself in the music; and, in following and responding to
the soundgestures madeby the composer, the listener maybecome
oblivious ofhis ownego, whichhas literallybecome one with that
of the music.
We must,then, bewary of easy andhigh-sounding statements
to
the effect that we cannot understand a work of art without, to a
certain degree,
repeating andreconstructing the creative process
by
which it has comeinto being. 5* Certainly
the listenermust respond
to the work of art as the artist intended, and the listeners experi-
ence of the work must be similar to that which the composer en-
visaged forhim. Butthis is a differentthing fromexperiencing the
creative processwhich broughtit into being.
However, thelistener maytake theattitude ofthe composer. He
may be self-conscious inthe act of listening. Those trainedin music,
and perhaps those trained in the other artsas well,tend, becauseof
the criticalattitudes whichthey havedeveloped inconnection with
their ownartistic efforts,to becomeself-conscious and objective in
all their aesthetic experiences.
And it is no doubt partly for this
reason that,as notedabove, trainedmusicians tendto objectify
meaning, toconsider itas anobject ofconscious cognition
see also
p. 70n. 24! .
Finally, andperhaps most
important ofall, this analysis ofcom-
munication emphasizes
the absolutenecessity of
a commonuniverse
of discourse in art. For without a set of gestures common to the
social group,
and withoutcommon habit
responses those
to gestures,
no communication whatsoever would be possible. Communication
depends upon, presupposes, andarises outof the universe ofdis-
course whichin the aesthetics ofmusic iscalled style.
Il
Expectation and
Learning
In the preceding chapter

the inhibitionof a tendency torespond
or, on the consciouslevel, the frustration of expectation wasfound
to bethe basisof theaffective and
the intellectualaesthetic response
to music. If this hypothesis iscorrect, then an analysisof the process
of expectation is clearlya prerequisite
for theunderstanding of how
musical meaning, whether aflectiveor aesthetic,arises inany par-
ticular instance.Such ananalysis isalso necessary if the evidence
used in support ofthe hypothesis,evidence whichrelates speciilc
musical processes to stipulationsof affectivity and aestheticpleas-
ure, isto be interpreted ina meaningfulway.
A general distinction must be drawn at the outset betweenthose

expectations thatarise out of the nature of human mental processes
-the modesin whichthe mindperceives, groups, and organizes the
data presentedby thesenses-and thoseexpectations thatare based
upon learningin the broadest sense
of the term. In the actualper-
ception of music there is, of course, an intimate and subtle inter-
action betweenthe two types of expectation.
Paradoxical thoughit may seem, theexpectations based
upon
learning are,
in a sense, prior
to thenatural modes
of thought.For
We perceiveand think in terms of a specific musicallanguage just
as wethink in terms ofa speciHcvocabulary andgrammar; andthe
possibilities presented
to usby a particular musical
vocabulary and
grammar condition the operation of our mental processes and
hence ofthe expectations
which areentertained on
the basisof
43
those processes. The mind,for example,expects structural gaps to
be filled;but what constitutes such a gapdepends upon what con-
stitutes completeness within a particular musicalstyle system.Mu-
sical language, like verballanguage, isheuristic inthe sensethat
its forms predetermine forus certain modes of observation and
interpretation. 1Thus theexpectations which result fromthe nature
of humanmental processes are alwaysconditioned bythe possibil-
ities andprobabilities inherent in the materials andtheir organiza-
tion aspresented ina particularmusical style.
In this chapter therelationship between expectation andlearn-
ing will be examined. The mannerin which the mind groups and
organizes thedata presented to it by the senses, the structure of
the thinking process asconditioned bythe learnedresponse se-
quences, and the manner in whichthis process gives riseto expecta-
tion will be thesubject ofchapters iii,iv, andv.
The study of expectationwhich follows makes nopretense to
completeness: Hrst, because a complete andsystematic study of the
process ofexpectation would be aformidable task, requiring asep-
arate monograph of its own; 2second, because a detailed account
of expectation would haveto bepreceded bya greatdeal ofexperi-
mental workin the Held ofpattern andfigure perception in music;
and third,because such a studywould entaila detaileddescription
and sensitiveappreciation ofthe stylisticcontext withinwhich the
process ofexpectation was being studied.
This necessityfor stylistic understanding hasdetermined the
choice of examples inthe following chapters. In order not to further
complicate thealready difHcultand delicatetask of discussing ex-
pectation, noattempt ismade inthis andthe followingthree chap-
ters toprove thatthe processes examined do,
in fact, have affective
aesthetic meaning; that is, no commentaries
from outsidesources,
from composers,critics, theorists, and the like, as to the affective
aesthetic nature of the various examplesare introduced. Since the
general reader is more likely to have developed sensitive habit re-
sponses tothe musicof WesternEurope ofthe pastthree hundred
years thanto anyother partof the literature ofmusic, theexamples
in thesechapters have been chosen from the music ofthis period.
In chaptersvi andvii, wherecomments on the examples by com-
Expectation andLearning 45
posers, perfonners,
theorists, andcritics areintroduced inevidence,
both the examples and the commentaries have beentaken from a
wide varietyof cultures,styles, andepochs.
Style: Formal Considerations

Musical stylesare moreor less complex systems of sound rela-
tionships understood and usedin commonby a group of individ-
uals. The relationships obtaining within such a style system are
such that: a! only some sounds or unitary sound combinations
are possible;b! those sounds possible within the system maybe
plurisituational withindened limits; c! the sounds possible
within
the system can becombined only in certainways toform compound
terms; d! the conditions stated in a!, b!, and ! are subject
to the probability relationships
obtaining withinthe system;3 e!
the probabilityrelationships prevailing
within thesystem are a func-
tion of context withina particularwork aswell aswithin the style
system generally. The occurrence of any sound orgroup ofsounds,
simultaneously orin sequence,will be more or less probablede-
pending uponthe structureof the system and the contextin which
the sounds occur.
SOUND TERMS AND SOUND STIMULI

A soundor group of sounds whether simultaneous, successive,
or both! that indicate,imply, or lead thelistener toexpect amore
or lessprobable consequent event area musicalgesture orsound
term within a particularstyle system. The actualphysical stimulus
which is the necessary but not suflicient conditionfor the sound
term will be called the sound stimulus. The same sound stimulus
may giverise to different soundterms in different stylesystems or
within oneand thesame system. This is analogous tothe fact that
the same word sound stimulus! may have different meanings
may become different sound terms, implying different conse-
quents! indifferent languages
or within one andthe samelanguage.
The word gauche, for example, hasdifferent, thoughrelated,
meanings inEnglish and French, while words such as "cross,"
"ground," orinterest have diHerent meanings within one and the
same language. In other words, asound stimulus does notbecome
a soundterm untilit becomes realized aspart of a systemof sound
relationships anduntil its particular functionwithin that system is
made apparent.
On theother hand,although itis clearthat asound stimulus can-
not becomea soundterm apartfrom the context ofa particular
style system,
it must be remembered that, sincethe listeneris part
of a culture that he takes for granted, a single isolated sound
stimulus will tend to be interpretedas part of the prevalent style
system ofthe culture,i.e., asa soundterm. Thusa dominantseventh
chord, for example, even though not incorporated intoa specific
context, is for the Western listener still a sound term, since the
sound stimulus is heard within the prevalent style of Western
music.
As we shall see,almost all studies in comparative musicology
emphasize that the samesound stimulusoften hasdifferent mean-
ings, is a different sound term,in different musical culturesand
styles andthat seemingsimilarities areoften verydeceptive. Fox
Strangways, forinstance, points out that a pieceof Indian music
which sounds to Westernears asthough itwere inC majoractually
has quite a different tonic and, consequently, quite a different
group of tendencies andprobability relationshipsfor the knowl-
edgeable Hindu listener#
Within a single cultureor even within one piece of music the
same sound stimulus maygive riseto severaldifferent sound
terms.
This iseasily seen
in the tonal systemof Westernmusic ofthe past
two hundredyears. Froma harmonicpoint of view, for example, a
chord sound stimulus! mayhave differentfunctions in different
keys. A chord which is a tonic in one key which bearscertain
more or less definiteprobability relationshipsto other harmonic
possibilities! maybe a dominant in another key,and so forth.
Within one and thesame tonalitya particularsound stimulus may
give riseto a sound termat onetime and not at another. Forwhether
a soundstimulus becomes a soundterm dependsupon itsfunction
in theparticular passage.
At onetime thesound stimulus
may imply
and indicate consequents andbe consideredas being structural,
as beinga soundterm; at another timethe samestimulus, though
it is part of a soundterm which has implications,is not itself a
sound term-doesnot in and ofitself giverise tomeaning.
Since musical structures are architectonic, a particular sound
stimulus whichwas considered to be a soundterm or musical ges-
ture on one architectonic level will, when considered as part of a
larger moreextended sound term, nolonger functionor be under-
stood asa soundterm in its own right. In other words,the sound
stimulus whichwas formerlya soundterm canalso beviewed as
a part of a larger structurein which it does not form independent
probability relationswith other sound terms.In short, the same
sound stimulus may bea soundterm onone architectonic level and
not on another.
The various levels of architectonic signification are, of course,
interdependent. ]ust as therecan beno chapterswithout meaning-
ful relationshipsbetween paragraphs, or paragraphs without mean-
ingful relationshipsbetween sentences, so the significance ofthe
longer partsof amusical workdepends upon the existenceof mean-
ingful relationshipsbetween theshorter ones.There couldbe no
musical sections if one period didnot in some wayimply andlead
to consequentperiods, andthere could be no periods if the phrases
which form them did not follow one another in an understandable
and meaningfulway. Theexistence and
coherence of
higher, more
extensive architectoniclevels isdependent uponthe meaningfulrela-
tionships established on lower architectonic levels. This is not to
say thatthe meaningof higher architectonic levels
is merely the
sum of the meaningsof the sound termsincluded in them, any
more thana chapteris the sum of the paragraphs,sentences, and
words contained in it.
While the meaning ofa musicalwork as a whole, as a single
sound term,is not simply the sum of the meaningsof its parts,
neither isthe entiremeaning ofthe work solely thatof its highest
architectonic level. The lower levels are both means to an end and
ends inthemselves. Theentire meaningof a work, asdistinguished
from themeaning ofthe workas asingle sound term, includesboth
the meaningsof the several parts
and themeaning ofthe work as
a single sound term or gesture. Both must be considere_din any
analysis of
meaning.
As observedin chapteri, meaningis notstatic andimmutable but
an evolving,changing attributionof a gesture orsound term.The
meaning ofsound termson all architectonic levels,
even thehighest
one, exhibitsgrowth and change. Andthe entire comprehensive
meaning ofa givenmusical workincludes thehypothetical, evident,
and detenninatemeanings of the multitudeof soundterms thatare
contained in it as well as the relationships existing between these
sound terms.
Often thehypothetical meaning of a sound termis verydifferent
from its evident meaning, and its evident meaningis conditioned
and modifiedby this difference. TheC-minor Fugue fromthe first
book of the Well-tempered Claoier furnishes an excellent instance
of sucha changein the meaning ofa soundterm measures 9-11,
Example 2!. In measure9 a sequence throughthe cycleof fifths
Ix *.
anvil A171
:uri 111 lg i-1-;.; l- T- rl 1 !I 5 -
` ._
,Q1'' l Lit! YT
':-| l 1 il; l..lY' ' i3 _ll ii
_
F=I .ll -Y I ilrqni il-__-.l
_ U
l151-
il-il'lll1|$- riiiil: 11-4r-11-l921I!.| r "r1=F=:-_-nz
YTYI _ 2 =_4 r ' |v92i=;
vA.92|1_11_-II; L14|.v'._ll!,92`l-IJ
;ll_lI-"$_II ;41 I'l1lZ*1_1l.A Tl Ld r!lI_; --Q
rnuzlginng ri
`
EXAMPLE 2
with imitationsbetween the soprano andalto isbegun. The sequence

continues throughmeasure 10and apparentlyinto measure11,
where the motive markedx is at first understood aspart of the
sequence; that is, we suppose thatthe sopranowill move to D in
the followingbar. However,once thewhole ofmeasure 11 and the
beginning of measure 12have been heard, we realize that the
hypothetical meaning attributed tox waswrong, thatit is not really
part of the sequencebut the beginning of the fugue theme, in
short, that its evident meaning is quite different from its hypo-
thetical meaning. It is very clear that Bach intends us to make this
mistake. For he could easily havemade it clear that the fugue
subject begins
at this point by stopping thesequential progression
in the left hand at the beginning of measure 11.Notice that om'
cognition of the evidentmeaning includesour conceptionof the
hypothetical meaning; the sound term is not only evidently the
beginning ofthe fuguesubject, butit is the beginningabout which
we wereoriginally mistaken. Furthermore, itis notonly ouropinion
of thesignificance of x thatis revisedin measure11 butour opinion
as to the significance of the whole episode, which nowappears to
have thismusical pun as oneof its meanings.
The factthat aswe listento musicwe areconstantly revising our
opinions ofwhat hashappened inthe pastin the light of present
events isimportant because it meansthat weare continually altering
our expectations. It means,furthermore, thatrepetition, thoughit
may existphysically, neverexists psychologically. Thus, thoughit
may seema truism, it is of some moment to recognize thatthe
repetition, say,
of theexposition section of a sonata-form movement
or that of the first-theme groupin the recapitulation hasquite a
different meaningfrom that communicated bythe original state-
ment.
It is also importantto realize that the more completea series
becomes, themore specificbecome the hypothetical meanings
attributed toparts ofthe series.
The relationshipsobtaining between
two tonesprovide thelistener withless basisfor specificexpecta-
tion thanthe relationships between five,six, orten tones.Similarly
the repetitionor seemingrepetition ofa part arouses more specilic
expectations than the firststatement ofthe part.The lesscomplete
the part,the moreprobable thatwe shallhave torevise ouropinion
of someor all of its terms. Toput it another way,the lesscomplete
the part,the weakerthe probabilityrelations between those terms
already established and anyfuture parts.
Here perhapsan illustrativeanalogy mightbe helpful. Suppose
that weare presented with thenumber series
2. 3. 5
The continuationof the series isin doubt. It might continue with
the number10 if the serieswere arrivedat throughover-all sum-
mation, orit might continue withthe number8. In the latter case
the seriesmight continuein at least two ways, dependingupon
whether thenumber 8was obtainedby progressivelyaugmenting
the amount of increase between successivenumbers + 1 : 3,
3 -|- 2 : 5, 5 -|- 3 : 8!, in which case the next number would be
twelve 8+ 4 : 12!, or whetherthe serieswas obtained
by adding
its two Bnal numbers -|- 3 : 5, 3 -}- 5 : 8!, in which case the
next term would be 13 -1- 8: 13!. As the series unfolds, our
expectations as to subsequent terms become more andmore specifdc.
This is exactly whathappens asa musicalsound termunfolds.
It follows from thisthat, sincedepartures fromor delaysin the
normally expected course ofmusical events will be most eHective
where that course is most specificallyand preciselyenvisaged,
deviations willbe mosteffective where the patternis mostcomplete.
And presuming that suchaffective deviantswould occurwhere they
would bemost effective, we shouldexpect tofind themwhere the
pattern ismost complete. This expectation is borneout bythe prac-
tice of musicians. Observe especially thatembellishments are best
applied tothose places where amelody istaking shape, as itwere, or
where its partial, if not complete,meaning orsense has been re-
vealed. Hence with regardto the latter case,they arefound chiefly
at half or full closes, caesurae,
and formate. Sachsattributes the
fact that a new tone generally ventures toappear onlytoward the
end of the phrase,when thenucleus hasbeen wellestablished, 7
to the power oftradition. Butthe explanationwould alsoseem to
lie in the fact that suchnew tones,which arepalpably deviations,
delaying thearrival ofexpected, traditional consequents, are prob-
ably introducedfor the sake of expression andaffect. Theyare
brought in at the end of the phrase,when it has alreadytaken
shape, because it is at this point, wherethe subsequent terms of
the seriesare mostspecifically envisaged, that they will have the
greatest effect.
Thus theeffect ofany particulardeviant isa functionof its posi-
tion in the series.
A deviantwhich mighthave onlya slighteffect at
the beginningof a series, whereexpectation entertains a greater
number of alternatives ofapproximately equalprobability, may
have apowerful effecttoward theend ofthe series, where expecta-
tion is more particularand wherethe probabilityof expectationis
liable tobe greater.Of course,if a series isbeing repeated,
then any
point in the serieswill arousedefinite expectations based uponthe
earlier version of the series; and a variation in the series will be
most effective.
AMBIGUITY
A soundterm canhave differentmeanings atdifferent times,but
this does not prove that the term, or the hypotheticalmeaning
which it Hrst has,is ambiguous.For ambiguityis a state ofmind
in the listener, notsimply a case ofdouble meanings. If we are
certain in our mindsas to the meaningof a sound termwhen it
first appears,then it is not ambiguous atthat time.And if we are
not in doubt when the same sound term is understood in a new
way, whenwe knowits evidentmeaning, thenit is still not ambig-
uous. Thusin the Bach Fugue,discussed previously,we are at Hrst
quite surethat the motive soundedon E-flat is a continuation of
the establishedsequence; butas soonas theexpected sequence is
not forthcoming, we revise our opinion and are certain, as we hear
more music,
that whatwe heardand arehearing isthe fuguesubject
itself rather than a fragment of it.
But even a sound term which does imply several alternative
modes ofcontinuation mayseem clearand unambiguous. For the
expected consequents
need notbe mutuallyexclusive. Theymay be
realized successively.
Often awell-shaped melody,
for instance,im-
plies several alternative goals.And the realization of one mode of
continuation doesnot precludethe subsequentrealization of an-
other seep. 100!. VVhat isimportant is that the implications be
definite and clear.
There are, however, soundterms that are decidedly ambiguous?
Ambiguity arises
either becausethe progressions involved ina pas-
sage areso consistently
irregular andunexpected thatthe listener
begins todoubt therelevance and efficacy ofhis ownexpectations
or becausethe shapesof the sound termsare soweak anduniform
that there is only a minimal basis for expectation. Thefeeling is one
of suspenseand ambiguity.Both theseaspects ofambiguity are
more fully discussed inchapter v.
Ambiguity isimportant becauseit givesrise toparticularly strong
tensions andpowerful expectations. For the human mind, ever
searching for
the certaintyand controlwhich comes with the ability
to envisageand predict,avoids andabhors suchdoubtful andcon-
fused states and expects subsequent clarification see Dp. 16, 26!.
There would seem to be various degrees of ambiguity. A sound
stimulus becomesa sound term by entering into probability rela-
tionships withother soundterms withinthe style.These probability
relationships are of differentdegrees. Forexample, itis quite prob-
able that the tone which comesafter an upward skipof a minor
sixth will descend, whilethe probabilityof which tone will come
after a skip of a majorthird is more doubtful.The moreequal the
probability ofdifferent alternativeconsequents, themore likelythat
the musicalprogression willseem ambiguous.
The factthat aswe listento musicwe not only interpretpresent
stimuli on the basisof past events butalso viewpast eventsand
expect futureones onthe basisof present stimuli meansthat a
process atfirst felt to be ambiguous may later be seen asless so.
Similarly processes at first considered unambiguous may later be
seen asinvolving or leading toward ambiguity. In other words,
ambiguity dependsupon the structural architectonicviewpoint
taken towardthe stimulusseries inquestion. Apassage orsection
which on the level of the phrase or period appearsto the mind as
ambiguous anddoubtful will, as a rule, seemunambiguous when
considered fromthe viewpoint of the total section.To put the
matter paradoxically:
the unambiguous meaning ofthe wholemay
be aproduct ofthe ambiguityof the part.
STATIC VERSUS DYNAMIC CONCEPT IONS OF

MUSICAL PROCESS
The precedingdiscussion points
up the dangers ofconcentrating
too much attention upon the structure of the musical work as a
single soundterm interpretedas a stable whole.The disciplesof
Schenker have not beensu$ciently awareof thisdanger. Toomuch
emphasis uponthe highestarchitectonic levelnot only tends to
minimize theimportance ofmeanings asthey ariseand evolveon
other architectonic
levels butit also leads toa staticinterpretation
of the musical process.
The very conception ofchord prolongation, so importantto
their view of musical growth, is a semantic confession that the
musical processis, in spite of their explicit statements to the con-
trary, basicallyseen asstatic. If what occursafter a given struc-
tural harmonyprolongs it, the implication is that the music is
heard more in relation to the past than in relation to what is still
to come.And while the pastof any sound termis of great irnpor-
tance, it is of importance mainlybecause ourexpectations asto
impending events are based
upon ourexperience andremembrance
of the past.
If, for example, theintroduction to the first movement ofBee-
thovens Piano Sonata, Op. 81a isregarded as an extendedprolonga-
tion of the openingE-flat majorharmony, themain pointof the in-
troduction is,it seemsto me,missed. Forthe passage is heardjust
the otherway around:the openingprogression leads us to expect
a cadence in E-flat, and the whole introduction consists, in a sense,
in delayingthe arrivalof sucha cadenceuntil after the allegrohas
already begun. The meaningof the passage and its affectivepower
derives fromthis inhibited tendency towarda perfect cadence in
E-Hat. Allthis is missed ifthe introductionis considered as a pro-
longation ofE-flat major.At best, we understandthe introduction
as aprolongation onlyafter it is finished.
Felix Salzers condemnation ofthe concept of modulation is
symptomatic ofthis essentially static viewof musicalmeaning. Itis
true that,when weconsider theevident anddeterminate meaning
of the whole work,modulations can be regardedas passinginten-
sifications ofthe main key. But this view ignores thatthe entire
meaning ofthe workincludes themeanings ofthe severalparts and
the various architectonic levels.
While we are experiencing music, we hear modulations and
changes of
key; weexperience shifts
in tonalcenter. Merely
because
some of these changescannot be directly and immediately related
to the key of the work as a whole does not mean that they are not
felt and lack signification;nor doesit mean that they cannot be
understood. Toextend ananalogy, borrowed from Salzerhimself,
harmonic excursions can be understood injust the same wayas
departures fromthe straightnarrative linein a novel-the compli-
cations of the plot-can be understood.
Only when the piece of music is complete, when it is timeless in
memory, doesSalzers pictureof music exist. Andeven thenthe
picture isincomplete, since
it ignoresthe experience
of the work in
time, which is part of our picture of the work as a whole.
Theories of music which imply that melodic similarity results in
musical unityof necessity
adopt amore orless mechanistic
concep-
tion of what constitutes aesthetic unity. Unity is not a matter of
employing asingle tonalityor a single melodickernel asthe basis
of all the themes of a piece.
What is required if the elementsof a work of art are to be compounded
into an aesthetic wholeis the presence ofan orderingsystem ofbeliefs
and attitudeswhich makethem mutuallyrelevant to one another;and,
conversely, thematerials handled in a work of art and the emotionswhich
they express,may vary indehnitely without endangering theintegrity
of the whole solong asthey areheld togetherby a controlling system
of
expectations.
The criticismsof thedisciples ofSchenker should
by nomeans be
understood asa wholesalecondemnation. Themethod and many of
the conceptswhich Schenkerand othershave developedcan be
a greatvalue inthe analysis of music,and theirinfluence upon
this
study is obvious. Thecriticisms aredirected merelyagainst those
aspects ofthe theorythat tendto treat a musicalcomposition as a
thing insteadof as a process which givesrise to a dynamicexperi-
ence.
PROBABILITY
We havestated thatstyles inmusic arebasically complex systems
of probability relationships inwhich the meaning ofany term or
series ofterms depends upon itsrelationships withall other terms
possible withinthe stylesystem. Aglance atalmost anybook on
the theory of music_ whether Zarlinos or Rameaus! or the exami-
nation of any discussion or descriptionof style whether oriental,
occidental orprimitive! will indicate, eitherdirectly orby implica-
tion, tl1atthis is the case.For example,the followingtable only
the beginning of whichis cited! given byWalter Piston 11 isactually
nothing morethan a statement ofthe systemof probabilitywhich
we knowas tonalharmony:
TABLE OF USUAL ROOT PROCRESSIONS
I is followed byIV or V, sometimes by VI, less oftenII or III.
II is followed byV, sometimes VI, lessoften I, III, or IV.
III is followed byVI, sometimes IV, lessoften II or V.
IV is followed byV, sometimes I or II, lessoften III or II.
V isfollowed byI, sometimes VI or IV, lessoften III or II.
Laws of melodic progression,
such asthe Lipps-Meyerlaw, are
essentially statements
of probabilityrelationships stated
in a quasi-
mathematical formulation relevant to particular style systems.
Stylistic descriptions
are alsoexpositions ofthe probabilityrela-
tions thatprevail withinthe systemunder investigation.When, for
example, FoxStrangways givesthe scale of Rag Pilu 12as in
Example 3,he is indicating by his notation certain probability
EXAMPLE 3
relationships withinthe tonalmaterials available. He is telling us,

for example,that C is likely to be the final tone, thatthe melody
will tend to center about the tone E-flat, that the tones D-flat, E,
F-sharp, A-flatand B-naturalwill be tones whichtend to move
toward themore stabletones inthe system,and soforth. Therela-
tionships obtainingwithin this particular part of the total style
system arefurther speciHed by written exposition: . . Pilu, for
instance, has an E and an E-flat with a D and an F on either side of
them; butin a given passage either E or E-flat will occur, but not
both as a rule. 1 Notice, too, that certain temporalrelationships
are alsoimplicit in this material;that is, that the tones writtenas
underlined wholetones arelikely to be sustainedlonger thanthe
tones notso markedand thatthose notunderlined arelikely to be
held longerthan thosewritten asquarter notes,etc.
Statistical style studies, such as those made by Frances Dens-
more in her work on American Indian music, also indicate that
probability is one of the centralfacts of style. Tableslisting the
number ofascending and descending intervals,the numberof acci-
dentals, or the number of times a certain interval is employed in
ascent ordescent, forinstance, are
all statements
about probability.
This conclusion
is emphasized by the fact thatthe figuresare given
in percentages.
The difficultywith statisticalstyle studies
is threefold.First, there
are, aswe shallsee inchapter iii,certain naturalprobabilities, such
By permission ofThe ClarendonPress, Oxford.
56 Emotion and Meaning in M usio
as the fact that a processonce establishedtends to continue in the
same manner, which need not becomemusically actualizedin a
majority ofcases inorder tobecome probable within a style system.
Although this might be overcome bypositing certain natural
probabilities, itis alwayspossible thatwhat isnatural, evenin this
psychological sense, may becomeculturally overlaid and hence
inoperative. Statisticscannot tellus whetheror not this is the case.
Second, instyles whichare not static theprobability relationsare
constantly changing, albeit slowly and subtly.This is simply an-
other wayof sayingthat, in a sense, each particularpiece isalso a
particular stylesystem. Third,it is clear thatone stylesystem may
presume aknowledge ofother styleswhich do not becomeovertly
realized ina statisticalsense. Thus, although thefull cadenceand
diatonic melodic motion arenot prevalent in thestyle ofWagner, for
example, Wagnersstyle nevertheless presupposes these as basic
norms. It seems tothis writer that in stylistic studyand analysis
there is no substitute for a sensitive responseto the style. This
can beachieved onlythrough practicein listening and betterstill
in performance.
Finally, wemay notethat in many theoreticalsystems the
impor-
tance ofprobability relationships
is madeclear inthe wayin which
the tonesof the system arenamed. Thusthe normative tones, those
toward whichother toneswill probablymove, havebeen giventhe
basic names,
while the other toneshave beengiven namesrelated
to these, often in terms of their probable motions. In Western
music, forexample, thestable tones
are namedthe tonic,mediant,
and dominant,while the subsidiary tones
are namedin relationto
these, forinstance, leading
tone leadingto the tonic! and super-
tonic. In the music of China non-structural tones take the name of
the structuraltone to which they move togetherwith the word
pin, meaning on the way to or becoming Probability relation-
ships arelikewise implicit in the names givento severalof the
structural tones
in the theory ofHindu music,e.g., amsa,
samuadi,
etc.
FORM, PROBABILITY, AND EXPECFATION

The architectonic nature of most larger musical structures has
been mentioned.Although the probability 'relationshipsof the
smaller unitsare alsoappropriate tothe organization
of larger
structures, it is clear that the larger groups and sections exhibit
certain specialmodes of organization andcombination, certain
special probability
relationships, which
exist inaddition to,though
not in conflict with, the probability relationships ofthe smaller
parts. Inother words,
forms are
special aspects
of style,alternative
probability groups,
each of
which exhibits
its ownspecial probability
relationships within
the totalstylistic context.
Like theperception of
and response to the more generallycontinuous aspects of style,the
understanding of fonn is learned, notinnate.
The conceptof a form involvesabstraction andgeneralization.
Our feelingof whata sonataform ora themeand variationsis does
not derivefrom our experience ofthis or that particularsonata or
theme andvariations butfrom ourexperience of a hostof worksin
such forms. Out ofthis experience the classconcepts which we label
as thisor thatform aredeveloped. The genesis ofclass conceptsin
which forms, under the influence of mental tendencies, become
normalized will be further discussed inchapter iii. Here it is sufli-
cient topoint outthat oncea workis recognized
as beinga typefor
which an abstract, nonnative class concepthas been evolved, then
that ideal type becomesthe basisfor expectations.
At Hrstglance theformation ofclass concepts
seems todistinguish
the probability relationships developed
in connectionwith form
from those established in connection with the more continuous
aspects ofstyle discussed earlier. Inthe caseof styleit might seem
as thoughhabit responses and probabilities are establishedby exact
repetition, whilein the case ofform exactrepetition isunknown.
However, thedevelopment ofall stylistic response sequences in-
volves abstraction;and everyoccurrence of a giventonal sequence
or rhythmicsuccession aisparticular,for it becomes significant and
meaningful only in its context, whichis of necessity particular.
Thus
our conceptionof a plagal cadence is just as muchan abstraction
as isour conception of, for instance, aconcerto grosso.
We have,let ussay, aconcept ofwhat afugue is.The conceptis
not ofthis orthat particularfugue butis basedupon ourexperience
of amultitude offugues. Aswe listento a particular fuguewe com-
pare itsspecial progress with the progress expected on thebasis of
Our normalized concept ofa fugue.Those progressions which seem
58 Emotion and Meaning in M usic
irregular andunexpected relative
to the generalized fugue
of our
imagination are
then thedeviants thedelays and
resistances! which
arouse the affective aesthetic response.
Such idealtypes are not, however,Bxed andrigid. They are
flexible fromtwo separatepoints of view. Our class concept of a
form isconstantly being modified bynew experiences of that form.
Each time,for example,we heara new work that can berelated
to our concept ofsonata formor perceivea work already heard
from a new point of view, our generalizedconception ofsonata
form ismodified, ifonly slightly.It is partly thiscontinual modifica-
tion of formal conceptionsand, incidentally,general stylisticones
as well! which enables us torehear awork manytimes. Foras the
norm with which we compare theparticular haschanged since a
previous hearing,the expectationswhich are entertained onthe
basis ofthe normwill alsohave changed, and thenew hearingwill
involve newperceptions andnew meanings.Norms, furthermore,
are flexible,in the sense thateach ofthe variouspossible antecedents
usually hasseveral alternative consequents, someof whichmay be
equally normative,i.e., equallyprobable. Ofcourse, as
noted earlier,
as thework progresses the alternativesbecome fewerand the se-
quence becomes more determined.
Not only are thereclass concepts of forms in generalbut these
concepts are always modified by a particular style.That is,we not
only havean abstractconception offugue in general butwe also
have anideal typefugue inthe styleof Bachas distinguished from
one byBrahms orHindemith. Awhole hierarchyof formsis main-
tained inthe mind,from thegeneralizations resulting from several
performances ofthe samework and those arisingfrom stylistic
experience to those based on theconcept ofform in general.
Thus itis notonly importantto know,in a general way, what the
style ofa pieceof musicis sothat the responses brought into play
will be the relevantones, butit is also importantto know what
formal procedures are beingemployed. Forour opinionsas toform
modify andcondition ourexpectations. We bring differentsets of
expectations to a Schubert impromptu than to a sonata movement
by thesame composer. Moreover, as noted above, nominally similar
forms whichdiffer in style areoften quitedifferent inform aswell.
Hence formis alwaysspecified withreference tostyle, justas style
should beparticularized withreference toform. The experienced
listener will, for example,bring a very different set of habit re-
sponses intoplay if he is about to hear a sonata movement by
Stravinsky from those which will be activated if he is about to hear
a sonataby Schubert.This doesnot meanthat the experience of
the Schubertsonata does not play a part in the perception ofthe
one by Stravinsky. Inso far as the general conceptof sonatais
brought tobear onthe listenersexperience, itis clearthat having
heard asonata bySchubert does influence ourperception ofStra-
vinsky. Likewise
our experience of sonatasby Stravinskyor another
modern composer, by modifying our classconcept ofsonata, will
influence, thoughto a lesser extent,our experienceof Schuberts
sonatas.
Furthermore, information about theform and style of a work is
important because, as weshall seelater inthis chapter,it conditions
not only what we look for, and hencewhat we perceive, butalso
the speedof our perceptions and our responses.
Of course,we neednot be told what we are going to hear. An
experienced listenercan placea work as to form and style onthe
basis ofmusical clues,such asharmony, melody,texture, instru-
mental style,and thelike. Noris it necessary that
we shouldbe able
to namethe composer or the style. Whatis vital is that we recog-
nize, inthe senseof bringingappropriate habitresponses into play,
the styleand form early enoughin the course oflistening sothat
important initialrelationships are not missed.
A distinction was drawn earlier between active and latent expec-
tation, andactive expectation was foundto be a productof a delay
or deviationin the normal sequence of events.It would seem that
the situationwith regard to form is somewhatmore complex.In
fonn we are, ina sense,constantly expecting. Under certaincondi-
tions we expect change,under otherscontinuity, andunder still
others repetition;until, finally, we expect the conclusionof the
piece. Thus in a very general way expectation is always ahead of
the music, creating a background of diffuse tension against which
particular delays articulate the affective curve and create meaning.
Formal expectationis constantly active on several architectonic
levels asa sortof generalized
aesthetic tension
which isshaped and
particularized in the course of listening.
Revision ofopinion, stressed earlier in the discussionof prob-
ability, is also importantin the perception ofform. Here,too, the
listener oftenfinds it necessary to revise hisopinions ofthe sig-
nificance ofwhat haspassed andhis expectations of what is still
to comein the light of an unexpected present. Thusthe meaning
and signiiicanceof the slow introductionto a sonata formmove-
ment will depend inpart uponlater developments which maytake
place in the allegro.The significanceof the slow introductionto
Beethovens Piano Sonata, Op.111 is quite diiferentfrom that of
the introductionto his String Quartet,Op. 130.The Sonatacreates
strong tension and suspenserelative tothe impendingallegro which,
because ofwhat we know about sonata form in the classical style,
is expected.The Quartet creates muchless tensionbut serves as a
source for many later developments aswell as a factor in the artic-
ulation ofevents within
the allegroproper. These
differences become
clearer andmore specificas eachwork unfolds.
Style and Social Process

Musical meaningand significance, like other kinds of significant
gestures and symbols, arise out of and presuppose the socialproc-
esses ofexperience which constitute themusical universes of dis-
course. Theperception ofand response to the probability relation-
ships obtaining within anystyle system are notnaive reflexreactions.
Nor are the probabilityrelationships universals having somekind
of natural, physical meaning. The responseto music as well as
its perception depend uponlearned habit responses. Thestyle
systems to which theseresponses are made are,in the last analysis,
artificial constructs
developed by musicians within a specificculture.
The veryfact thatthere aremany differentmusical stylesystems,
both in different culturesand evenwithin a single culture,demon-
strates thatstyles areconstructed by musicians ina particulartime
and placeand thatthey arenot basedupon universal, natural rela-
tionships inherent in the tonal materialitself. Andif the experience
of musicis not based uponnatural, universal
responses, must
it be
based upon responses whichare acquiredthrough leaming.
LEARNING AND STYLE

The normsand deviantsof a style uponwhich expectationand
consequently meaning are basedare to be found in the habit re-
sponses of
listeners whohave learnedto understandthese relation-
ships alsosee p.83!.
We speakof "traditions,"styles of art, meanings and soon, asif these
things hada kind of independentreality of their ownwhich areeternally
attached toworks ofart. But traditions andmeanings are kept aliveonly
through thedispositions and habits whichform the subjective contexts
of
countless individuals
.... There can beno aestheticresponse whatever
apart fromthe responses of individualmen whichgives itmeaning.
These dispositions
and habitsare learnedby constantpractice in
listening andperforming, practicewhich should,and usuallydoes,
begin in early childhood. Objective knowledgeand conceptual
understanding do not providethe automatic,instinctive perceptions
and responses which will enable the listener to understand the
swift, subtle,changeable course of the musical stream.To para-
phrase Bertrand Russell seep. 39!: Understanding music is not a
matter ofdictionary definitions,
of knowingthis, that,or the other
rule of musical syntaxand grammar,rather it is a matter ofhabits
correctly acquiredin ones self andproperly presumed in the par-
ticular work.
It is not enough,for example,for the listener to know that in
Western musicof the past threehundred yearsa particularsound
term, the dominant seventh chord, creates an expectation that an-
other particularsound term,the tonic chord, will be forthcoming.
The expectation must havethe statusof an instinctive mentaland
motor response,a felt urgency, beforeits meaningcan betruly com-
prehended. The story ofthe youngcomposer who got outof hisbed
and ran to the piano to resolve adominant seventhchord which
someone else had left unresolved isa good instance ofthis power
of felt urgency-of ingrainedhabit.
I emphatically repeat, writesHugo Riemann, that practice and
good will are requiredfor the understanding ofa greatand com-
plicated musicalwork of art. 17This practiceis both mental and
motor. The relation between thinking and motor responseswill be
discussed insome detaillater in this chapter.The distinctionbe-
tween mental habits and motor habits is a diflicult one; however,
both playan importantpart in the learningof musicalstyles. There
is probably a time in the development ofchildren whenmotor
learning playsa particularlyimportant rolein the development of
response patterns.
And, hence,early instructionin musicalperform-
ance isimportant, notonly because of the immediate pleasure in
performance which it gives,but alsobecause itinstills intothe child
the properhabit responses, which are the life stream ofmusical
perception and communication.
THE PLURALITY OF STYLES

As Russellobserves, not
only musthabits beproperly acquired in
us but they mustalso beproperly presumedin others;that is, our
trained habits of discrimination and responsemust be relevant to
the particularstyle of music tobe heard.For the habits acquired
are notuniversal butare acquiredin connectionwith a particular
style andare relevantto thatparticular style.
Music isnot a universal language.The languages and dialects
of musicare many.They varyfrom cultureto culture,from epoch
to epochwithin the same culture, and evenwithin a single epoch
and culture.An Americanmust learnto understandjapanese music
just ashe mustlearn to understand the spoken languageof japan.
An individualfamiliar withthe traditionof modernEuropean music
must practiceplaying andlistening tothe musicof theMiddle Ages
just as he must practice readingand speakingthe languageof
Chaucer. Even within oneand thesame cultureand epochit is the
exception rather than the rule whena musicalstyle isunderstood
by all members of the culture. Witness the fact that in our own
culture the devotees ofserious music have great difliculty in
understanding themeaning andsignificance ofjazz andvice versa.
Yet, while recognizing thediversity of musical languages,
we
must alsoadmit thatthese languageshave importantcharacteristics
in common. The most important of these, and the one to which
Expectation and Learning 63
least attention has been paid, is the syntactical nature of different
musical styles.The organizationof sound terms into a systemof
probability relationships,the limitations imposed uponthe com-
bining of sounds, and so forth are all common characteristics of
musical language. It is to thesethat comparativemusicology must
turn if it is to makefurther progressin studyingthe musicof differ-
ent cultures.In this respect musical languages arelike spokenor
written languages which alsoexhibit commonstructural principles.
But differentmusical languages may alsohave certainsounds in
common. Certain musical relationships appear tobe well-nighuni-
versal. Inalmost allcultures, forexample, theoctave andthe fifth
or fourth are treated as stable, focal tones toward which other
terms ofthe systemtend tomove. Similarlymany systems have or-
ganized tonal progressions, scales,though therelationships between
these sound stimuli will vary greatlyfrom systemto system.
In so far as different styleshave traitsin common,the listener
familiar with the musicof onecan perhapsget the gist of music
to whichhe is not accustomed to respond;just asone canat times
get the drift of a play or poemheard in a foreignlanguage that
has some words incommon withones nativetongue. Itis important,
however, tonote thatthe unpracticedlistener isalso verylikely to
make mistakes by readinginto orientalor primitive music irnplica-
tions relevantonly to the stylesystem ofrecent Western music.
Because harmonies
are usedconstantly inour music, they haveper-
meated our musical consciousness to such an extent that the Western
listener bynecessity experiences
music asharmonic-whether harmonies
are actuallypresent, aremerely implied as in the folk-songsof Western
Europe fromthe last few centuries!,or are entirely missing,as in most
Primitive music.Only by' prolonged trainingand familiarity is the in-
vestigator ableto acquire the ability to experiencemonolinear musicas
such. Harmonichabits condition not only our mode of experiencing
music, butalso thenature ofour musicalconcepts.
In general it seemswise andprudent to treat all aspects ofa

style system
as learnedand culturallydetermined. First,
because it
seems likely that even the so-called natural stylistic traits are
actually learned,just as certain phonemes
are commonto a lan-
guage familybut arenevertheless learned.
And second,
because the
distinction between natural and learned characteristics is unneces-
sary. If the naturaltraits persistin a given stylesystem, theycan
be studied as though they were learned, culturallydetermined
elements just as easilyas theycan beas naturalones. Whileif nat-
ural traitsare notoperative withinthe givenstyle, thenthey need
not be considered, except perhaps froma genetic point of view,
i.e., wemay askwhy theyare notoperative."
PATTERNS OF STYLE CHANGE

Thus far we have been dealinglargely with style systems,by
which term something analogousto language has been meant.
Where style systems are similar in important ways, we may say
that theybelong tothe samestyle-system family, just asthe Indo-
European languages have certainbasic traitsin commonbecause
they stemfrom acommon rootlanguage. By style, asdistinguished
from stylesystem, ismeant themore particularvariants andmodi-
Hcations ofa stylesystem made at differentepochs withina culture
or by different composerswithin the same epoch. Thus Bachand
Beethoven represent diHerent styleswithin a single stylesystem,
while Mozartand Machautemploy differentstyle systems.
Styles and style systems are not permanent, Hxed, and rigid.
Within cultureswhich do not imposestrong socialsanctions upon
art, changesin style have beenthe rule rather thanthe exception.
One style gradually replaces another, attains its own particular
fruition, declinesand is replaced byanother style.The processis
gradual and,since not all aspectsof the system arenecessarily
changed, itis oftenimpossible tomark offthe historicallimits of a
style. We must be content to point out its ultimate fruition and its
general limits.This hasalso beenthe case,though lessfrequently,
with stylesystems.
It has been customary to relatesuch changes to social,political,
and cultural changes-to explain the history of styles and style
systems interms of general, non-musical history. No doubt such
extrastylistic events
are of great importance as necessary causes in
the historyof styleand stylesystems. This appears tobe particularly
true in the caseof the radical changes which occurwhen onestyle
system replaces another, e.g.,the stylisticcultural changeswhich
took place during the period of Western historyknown as the
Renaissance.
Yet the explanations furnished by referenceto political, social,
and culturalhistory tellonly partof the story. Forstylistic changes
and developments are continuallytaking plapewhich appearto be
largely independent of suchextramusical events. Although animpor-
tant interaction
takes placebetween the political, social,
and intellec-
tual forcesat work in a given epoch, on theone hand,and stylistic
developments, on the other, there is also a strong tendencyfor a
style todevelop inits own way. If this is the case,then thecauses
of thesechanges must be lookedfor in the natureof aestheticexperi-
ence, sinceboth for the composerand listenerstyle is simply the
vehicle for such an experience.
A discussion of the causes of such purely aesthetic stylistic de-
velopment isimportant, notonly aspart of a generaldiscussion of
style, but also because the hypothesisof this study derivesaddi-
tional weightand supportfrom the fact that it is able to account
for processes which haveas a rule beendescribed ratherthan ex-
plained. Toput it in anotherway, oneof thelogical consequences
of
the presenthypothesis wouldbe that a tendency toward intra-
aesthetic change would bethe rule,a deductionwhich isconfirmed
by the facts ofmusic history.For in any stylethe deviantsas well
as thenorms areHnite innumber; andit is both possibleand likely
that a deviant through constant employmentmay become so Hxed,
so common in its recurrence inparticular situations,
that the prob-
ability relationshipsof the system become modified bythis recur-
rence. Consequently a soundterm whichwas oncea deBnitedeviant
may become more orless normative within the style andthus lose
its potentialfor expression.
In other words, deviation, originating as expression, mayafter a
time becomenormative, and when this occurs it is necessaryeither
to invent new deviations for the sake of aesthetic effector to point
up those already in use. This means thatonce a style is established
there is a constant tendency toward the addition of new deviants
and toward pointing up, through emphasis or exaggeration, those
deviants already
present. Inshort, thenature ofaesthetic communi-
cation tends
to makefor the eventual destructionof anygiven style.
This processof stylistic genesis can
be seennot only in the his-
tory of Western music but alsoin muchoriental andprimitive mu-
sic.
In Westernmusic wemay takeas anexample the changing useof
the vibratoin stringplaying. Originallyin theeighteenth century
the
vibrato wasan expressive device whose use wasconfined tospecific
passages. Gradually it becamea fairly constant featurein string
playing, thuslosing someof its expressive effect. At present the
ordinary vibratois a norm of string playingfrom which there are
two typesof deviation:first, theuse ofan unusuallyrapid, andsome-
times wide, vibrato and, second, the use of no vibrato at all. It is
particularly interesting
to notethat thislatter alternativeis becom-
ing more and more prevalent inthe rendition of expressivepas-
sages. Several contemporary scores specifically stipulateno vi-
brato, e.g., Bart6ks String Quartet No. 4, third movement, or
Bergs Violin Concerto. What was once an affective aesthetic
deviant has,through constant employment, become normative, and
what was once considerednormative hasbecome avaluable expres-
sive device.
We cansee asimilar changeof function in the employment of
modal cadenceswhich, though normative in the Middle Ages and
Renaissance, become expressive deviantsin the style of some com-
posers ofthe late nineteenth and twentieth centuries. Similarly the
authentic cadence, a norm in classical and early romantic music,
sometimes appears to be a deviant in the style of the late nineteenth
century. Thereis a striking example
of this in Ein H eldenleben
by
Strauss. _lust
before number77 Eulenberg,miniature score!there
is aperfectly regular
cadential progression,
II-I -V,in E-flatmajor,
which in a piece written a hundred years earlier would lead us to
expect the tonic chord. Here, however,it leads us to expect almost
anything but the tonic; and when the tonic does come, it is def-
initely felt to be a deviant.
From Herzogsdescription ofthe developmentof Pueblo mu-
sical style,it seemsclear thatthe sameprocess takes
place inprimi-
tive music: the deviantsbecome normative within the style and
provide the basis for further deviation.
If oneof thetwo sectionsis apentachord-which oftenresults fromthe
extension ofa tetrachord-this wider sectionis frequently found in the
lower position.... On the fringes of such sections decorative tones
appear; in time thesebecome standardized and strengthened,and this
new growthfinally resultsin extendedforms ....
Tonal growthhas progressed to such a degreeof saturationthat the

original structure-probablypentatonic-often becomesgrown overand
obscured. Sharp accentuation andother featuresof the singing technic
give riseto a greater numberof secondarytones whichin turn provide
material forfurther melodicgrowth.
A similar development seems to havetaken placein the case of
Byzantine melodicstyle. At first deviationand expressionwas a
matter of combining briefmelodic formulasin different and sur-
prising ways,thus producingnew hymnmelodies. However,
The immensenumber of hymns introducedinto the service madeit
necessary forthe ecclesiastical
authorities toprohibit the addition ofnew
hymns tothe repertory,and the artistic activityof the monks fromthat
time onwardswas concentratedupon the embellishment ofthe music,
which, in the followingcenturies, andeven afterthe fall of the Empire,
became increasingly rich and elaborate, untilthe originally simple struc-
ture of Byzantine melodieswas transformedinto an ornamented style
and thewords ofthe textmade unrecognizable by extendedcoloraturas.
Here we have anexcellent exampleof the relation between socio-
political conditionsand stylistic development. Forthe pressure
exerted bythe authoritiesof the Byzantine Church,though it in-
fluenced thecourse of stylistic development,
did so largely in a
negative way; certain possibilities of deviation were excluded, but
there was no stipulation as to the future course of stylistic change.
This is particularly interesting
because under rather similarcondi-
tions the composers ofthe WesternChurch eventuallyturned to
other methods of deviation,e.g., thevertical embellishment called
organum.
The fact that the socio-cultural situation in which an art flour-
ishes limits,at leastin a negative way,the modesof deviationis
perhaps mostclearly seenin the case offolk music. Because the
true tradition of folk music is aural rather than written, deviation
is a matter ofimprovisations made
upon a learned basicstructure
and shape.Sometimes this shape maybe purely melodic, whilein
others it is harmonic as well:
Hot jazz melody isimprovisatory, butits structureis held to a coherent

formal pattern which rest1'ainsit from complete chaos.This coherent
pattern isprovided bythe harmonicsequences of the underlyingaccom-
paniment .... It is the simpleharmonic phrase. . . that provides the
unifying principle in hot jazz improvisation.... This phrase is re-
peated overand over again, with occasional interpolations, perhaps, of
other similarchordic sequences,
forming asort of ostinato on which the
melodic andrhythmic variationsare built. At each variation of the har-
monic phrasea new melodic andrhythmic superstructure is improvised
by thehot player.
In the case offolk music,including jazz,the basic,normative pat-

terns arefixed bycustom andtradition, butthe degreeand manner
of deviationmay change,bringing newstyles intoexistence. Thus,
for example, Dixie Landjazz andBebop areboth basedupon essen-
tially thesame basicpattern, buttheir mannerand styleof deviation
differ.
Suppose that
a devicewhich wasonce adeviant ina givenstyle
becomes iixed in its relationships and constant in use. Does this
mean thatit necessarilyceases to be aestheticallyeffective, thatit
becomes anorm? Theanswer appears to be negative. Thougha
deviation mayno longeractually functionto inhibit a tendency,it
may still function expressivelyas a sign. Whethera deviationbe-
comes anorm ora signof expression would seemto dependlargely
upon the context in which it is employed.If it is associatedin
practice withreal deviants,it will probably continueto functionin
an affectiveway. If, on theother hand,it becomesassociated with
clearly normativeprogressions, then it will tend to become norma-
tive within the style.
Even wherea deviantdoes notbecome anexpressive signit need
not necessarily
become anorm. Ifthe expressive
value ofa relation-
ship becomes
weakened throughstandardization, several
alternatives
present themselves:
! The degree ofdeviation canat timesbe in-
creased as, for example, it was in the elaboration of coloratura
passages in
late Byzantinemelodies. ! New deviantdevices can
be introducedinto the style asalternatives, weakeningthe prob-
ability relationships
between the
former deviantand itsconsequents.
That is, if A to D a former deviant! is becoming a normative
probable! relationship,the introductionof D1, as an alternative,
will of necessity weaken
the probabilitythat Awill be followed byD
and hencerenew, asit were, the deviantquality of D. ! New
deviants canbe used to replacethose which are becomingnor-
malized. Theintroduction ofmodal relationships into the harmonic
style ofthe latenineteenth century
was aninstance ofthis. ! Old
relationships canbe revitalizedthrough changesin other aspects
of styleand throughnew anddifferent usesfor Hxedrelationships.
Harmonic styleunderwent sucha revitalization in the second halfof
the eighteenthcentury. The essential structureof the harmonic
scheme which flourished duringthe later baroque wasmaintained,
but its use inthe organization
of thetotal structureof thework was
new.
Several instances in which norms became deviants have been
noted. Actually,however, thisis neithera necessary
nor a common
occurrence. If norms do become deviants,the change of function
does notas a rule take place immediately
but rather after a con-
siderable lapse
of time and theestablishment of
a newstyle system.
STYLE CHANGES AND THE COMPOSER

Styles aremade, modified, and developed by composers
and per-
formers, bothas individualsand asgroups. Thetendency toward
stylistic changeresults notonly fromthe musiciansconscious aes-
thetic intent but also from the fact that the composerand per-
former, by their very nature ascreators andmakers, regardthe
traditions andstyles whichthey inheritfrom their predecessors as
a challenge-as a more or lessfixed, recalcitrantmaterial, whose
resistance to
change and
modification the
true artistdelights in
overcoming andconquering. Stravinsky, for example,writes that
as I am bynature alwaystempted byanything needing prolonged
effort, andprone topersist inovercoming difficulties
_ . . the pros-
pect . . . greatly attracted
me. 2In his experimenting and playing
with his inheritance theartist often taxes hisown ingenuityand
imagination tothe utmostand, likea virtuosoon thehigh wire,tries
to seehow farhe cango in creating new aesthetic problems,
or how
he cansolve oldproblems innew andsignificant ways. How far,in
short, hecan deviatewithout losinghis aesthetic balance.
The creationand overcoming of difficulties,an apparentlyintel-
lectual process,
and themodification and remolding ofstyle forthe
sake ofaffective aesthetic
responses are but two facets ofa single
process. Andonce againwe arereminded ofthe groundlessness of
the traditional dichotomy betweenemotions andintellect.
This analysisof the relation between artistic creationand devia-
tion explainsin part the process of stylisticchange. Italso enables
us to introduce asevidence inthe ensuingchapters thestatements
of composers, performers, andcritics referringto the creative ex-
perience ofthe artistrather thanto affectiveexperience itself. For
if the conquest ofdifficulties andthe affectiveaesthetic processes
can usually 2* beequated, then a passage designated by a writeras
involving the delight in conquering difliculties can alsobe con-
sidered aspotentially affectiveor aesthetic. It must then beshown
that thepassage in question does, in fact,involve delayin expecta-
tion or, in other words, deviation.
The relationof artisticcreation toplay mustbe mentioned.
Many
references aremade in the literature of music to the playfulness of
a particularpassage orto the delight takenby musiciansin play.
It seemsvery probable that thistoo isa wayof referringto thecon-
quest ofself-imposed difhculties. Karl Groosfrequently emphasizes
that this is an essential feature
of all play; that,in his own words,
play leadsup from what is easy tomore difliculttasks, since only
deliberate conquest can producethe feeling of pleasurein suc-
cess. 2
This delightin the conquest ofdifficulties, inaesthetic play
with
the recalcitrantmass oftraditional materials, is importantin per-
formance as well ascomposition, particularlywhere theperformer
is freeto, or supposed to,
improvise upon either awritten score,as
in theseventeenth and eighteenth centuriesof theWestern tradition,
or upon a tune handed down by oral tradition, as in much of the
music ofthe Orientand infolk music.But eventhe performerwith
a fullynotated score,
if he is trulycreative, is
engaged in
this process
of deviationfor thesake ofexpressive play see p.199!.
CYCLIC CHANGE AND STYLE

Implicit in this wholediscussion the
is conclusion t_hat the
process
of stylistic change isa cyclic one. Theremust first be a period
during whichnorms andtheir relateddeviants become established.
Tl1is cannotbe accomplishedby theoristsor by decree; thenew
norms andtheir related deviants mustgradually becomepart of
the habit responses ofcomposers, performers, and listenersalike.
Such aperiod isusually markedby a plurality of styles. Thesitua-
tion will tend to be uncertainand ambiguous. There will be con-
servatives whoadhere to the old style and there will be avant-
gardists whoare attemptingto build the new style. Bothgroups
will be very conscious of technique,and the partisans ofthe new
style willbe especially conservative, in
the sensethat theywill tend
to imposevery strict limitations uponthemselves. Cultural tension
and conflictwill give rise to schools, pamphleteers,
and apologists.
The increasedconcern withmusic theory,with the grammar and
syntax of style, will produce a host of theoretical andaesthetic
treatises.
Following thisthere isa periodin which the new style becomes
established and accepted. There is a tendency towardsingleness of
style. In contrast tothe first period, whichwas largelyconcerned
with the establishment of norms, wenow findan equalconcern for
deviant andnorm. Themusical situation is relativelystable andall
energy isturned tothe productionof music.Theoretical andparti-
san writingsabout musicbecome infrequent.
In the course oftime, however,some ofthe deviationsdeveloped
become almost clichs, andcomposers search for new means ofex-
pression andnew difficulties to conquer..The whole system of
probabilities graduallybreaks downunder the weight of an in-
creased number and degreeof deviationsand theend of the style
IS in view.
Such acyclical viewof theprocess of
music historyseems, whether
we like it or not, to be a part of the facts of aesthetic process. This
process doesnot, of course, resultin any rigid determinism. For the
rate ofchange, the
kind of change, and
even thefact of change, all
are conditionedby the social, political,and cultural climate in which
the processmust takeplace. Andthough theseextramusical factors
may, andoften do,obscure thecyclical processes which markthe
genesis of
style systems
and idiomsyet thetendency toward cyclical
change seems to beconfirmed bythe factsof musichistory.
The probabilitiesof style and form,the normsupon which ex-

pectations rest,
differ fromculture toculture andstyle tostyle. What
remains constant in the flux of music historyis not any particular
organization ofthe materialsof sound.The patternsof style are
fixed by neither Godnor naturebut are made, modified,and dis-
carded bymusicians. What remains constantis thenature ofhuman
responses and the principles of pattern perception, theways in
which themind, operatingwithin theframework ofa learnedstyle,
selects andorganizes the sense datapresented to it. But before these
perceptual processes are broughtinto play,before themusic begins
to sound,the listenerprepares toattend.
The Preparatory Set

Like otherintentional activities
listening tomusic ispreceded by
a number of mental and physical adjustments, performed con-
sciously orunconsciously, which serve tofacilitate and condition
the subsequentresponses made to the expected stimulus.These
adjustments are known asa preparatory set. The specific adjust-
ments made are products of ! the listenersbeliefs aboutaesthetic
experience in general andmusical experience in particular, ! the
experience and knowledge previously acquired inlistening to and
studying aboutmusic, and! information gathered onthe partic-
ular occasionin question.
AESTHETIC BELIEF
The listenerbrings tomusic notonly specificallymusical experi-
ences, associations,
and dispositionsbut also important beliefs as to
the natureand significance
of aestheticexperience ingeneral and
the expectedmusical experiencein particular.
The beliefthat we are dealingwith an aesthetic object
leads to
what Henry Aiken has called the idea of framing, that is, the
belief thatan aesthetic
object isa specialkind of stimulus towhich
we do not respondby overt action. Thefact that the response to
aesthetic experience
is not overt has,as wehave alreadyseen, very
important consequences in conditioningour responses; for the re-
pression ofovert behavioris a vital factor in the development of
affect.
The ideaof framingdoes not, however, detract from thefeeling of
reality whichis so important in aesthetic experience.The mecha-
nism ofdenial canoperate; afirm beliefin the reality of play can
coexist witha certaintythat it is play only. Here lie the roots of
aesthetic illusion.2' Furthermore,the ability of the mind to be-
lieve, to enter into the special nature of the aesthetic situation, in
part makespossible thefact that a singlework canbe heardmany
times. Forhere, too,the mechanismof denial operates insuch away
that thelistener holdshis knowledgeof the final aestheticoutcome
in suspense and believesin the reality of all the expectations, sur-
prises, anddelays setforth in the work,even thoughhe mayhave
experienced them in an earlier hearing.
Nor shouldthe role played bythe beliefin the seriousness, sig-
nificance, andpower of aesthetic experience be overlooked. For
the attentiongiven toa workof art is a direct productof the belief
in thesignificance andvitality of aesthetic experience.
And attention
not onlyfocuses our minds uponthe musicalwork but also modifies
perception itself,since when the organismis active, at a high
degree ofvigilance . . . it will produce goodarticulation; whenit
is passive,in a low state of vigilance, it will produce uniformity. 29
It seemsquite probable,moreover, thatit is the belief in the
power and importance of aesthetic experience,the belief that we
are going to have such an experience, that is responsible for the
fact, notedearlier p. 11!, that tone as such hasa very powerful
emotional influence.It setsup organicconditions whichare involved
in strongfeeling .... 3 It is very doubtfulwhether anindividual
engaged inthe choresof everydaylife will respond inthis wayto
the toneof a violin playedby a child practicinghis scalesor to the
sound ofthe chimesof a particular radiostation. Thechanges in
pulse, respiration,metabolism, andpsychogalvanic reflex, which
Mursell attributesto tone as such, do not appear toaccompany
all acts of attention, though attention is an important factor in their
arousal. Rather believing in the aestheticaffective significanceof
musical experience, we expectto havesuch anexperience, and our
bodies, responding to this mental set,prepare themselves for the
experience. This supposition issupported byevidence indicating
that the act of attention, ofwhich listeningto music is a special
type, isoften accompanied by physicaladjustments, including those
of the central nervous system. Thereis alsoevidence thataffect is
related to motor attitudes which, as will be shown below, form an
important partof the total preparatoryset.
The situationis further complicated bythe fact that the belief
that we are aboutto have an experiencemay itself give rise to
special tensions
which arerelieved onlywhen themusic beginsto
sound andthe morespecifically aesthetic tensions begin
to prevail.
The atmosphere of the concert hall,hushed andquiet beforethe
music starts,
is chargedwith thetension ofexpectancy. The behavior
of the audience isusually anindication ofthis tension.They are
not calm and relaxed but strained and excited, their mental tensions
often Endingrelief in bodily behavior,e.g., coughing,whispering,
and so forth.
BELIEF AND THE PRESUMPTION OF LOGIC

Related tothe belief in the power andsignificance ofaesthetic
experience isthe belief in the seriousness, purposefulness, and
logic of the creativeartist and the work he produces.The pre-
sumption thatnothing inart happens without areason andthat any
given causeshould besufHcient and necessary forwhat takesplace
is a fundamental condition for the experience of
art. Thoughseem-
ing accidentis a delight, webelieve thatreal accidentis foreignto
good art.Without this basic belief the listener would have no reason
for suspendingjudgment, revisingopinion, andsearching forre-
lationships; thedivergent, theless probable,the ambiguouswould
have no meaning. Therewould be no progression,only change.
Without faithin the purposefulness and rationality of art, listeners
would abandon their attempts to understand, to reconcile deviants
to what has gonebefore, or to look for their raison dtre in what is
still to come.
76 Emotion and Meaninzgin Music
The termserious asapplied toart doesnot, then,mean heavy
or world-shakingas opposed to comic or light but rather that the
relationships set
forth inthe artwork aresigniicant, logicalrelation-
ships andhence tobe takenseriously. To
put it paradoxically, a
rollicking rondoby Haydn is lesscapricious, ismore serious,than
some ofthe statelysymphonies by Mahiler.
Because ofthe tremendous importance ofbelief in the response
to art, the mostdevastating criticism
that can be leveledagainst a
work isnot thatit is crude or g but that it is not aestheti-
cally purposeful
and meaningful
Statements that
compositionsthe
in
twelve-tone techniqueare conceivedWithin an essentially mathe-
matical framework,implying that they are not honestly felt or
aesthetically conceived
by the composer, havedone moreto make
the music of this school unpopular and hated than all the accusa-
tions of cacophony and ugliness puttogether. It seems probable
that audiencesobject tothe dissonancein this music, notbecause
it is unpleasant, but
because they believe thatit is the productof
calculation ratherthan an aesthetic azifective conception. These
criticisms haveweakened belief in the logic and seriousness of the
music, andlisteners haveconsequently abandoned their attempts
to understand. The power of most journalistic criticism derives
not somuch fromits abilityto influencefudgment asfrom its power
to enhance or weaken belief.
Much of the informationsupplied in. the programnotes for a
symphony concert,the popular biograp-hies ofcomposers, orthe
run-of-the-mill music appreciation course is aimed, albeit uncon-
sciously, primarilyat enhancingbelief. The story ofthe composers
life and hard times, the circumstances under whicha particular
composition waswritten, the testimonials tothe greatnessof the
work to be heard,and soforth do not help us to appreciate to
understand! thework directly,only ourown proper habit responses
can dothis, ratherthey aidappreciation bystrengthening belief
and
creating awilling attitude see pp.61-62! .
]ust ascriticism canenhance beliefand hencethe dispositionto
respond! throughpraise or negate belief and the disposition to
respond! throughblame, sotoo the composition ofthe audience
and its attitude toward the performers and the compositions to be
heard canplay an important partin coloringbelief. A half-empty
concert hall with an unenthusiastic audience or even a full hall with
an inattentiveaudience willtend to minimize beliefand probably
the responsesof agood manymembers of the audience,
while a full
house with a devoted audience will tend to enhance belief.
Obviously fashions, right opinion, as setby the social group
which constitutesa particular segment ofthe total audience, also
influence beliefin important ways. Andit would seem thatsuch
socially determinedbeliefs and tastes arebecoming increasingly
effective inconditioning theresponses of
what David Riesman has
called theother directed man of our society.
Learned Habitsand the Preparatory Set

T'he preparatory sets whicharise asa resultof our beliefs asto
the natureof musicalexperience are not specificto any particular
musical styleor form. The dispositionto respondis general,i.e.,
mental atdtudesand bodily tensions whicharise are relevant to
musical experience per se.
Together with these generalbeliefs aboutaesthetic experience
the practicedlistener alsobrings withhim, aswe haveseen, awealth
of more specific dispositions or ideo-motorsets basedupon past
experience in listening and knowledge acquired either systematically
or by chance. Once the listenerknows, eitherprecisely orin terms
of generalstyle characteristics, what kind of musiche is going to
hear, thisinformation conditions his perceptions,
modifies hisopin-
ion of what isheard, andqualifies hislater responses.
The informationwhich bringsthe preparatory sets intoplay need
not be verbal. It may consistof visual signs, suchas thepresence
of a particular instrumentalgroup, or the gesturesof performers,
the kind of audience, and so forth.
THE INFLUENCE OF KNOWLEDGE AND EXPERIENCE

ON PERCEPTION
What we know and hence expect influences what we perceive,
that is, the wayin which the mind groups andorganizes thesense
data presentedto it. Knowledge asto the style and form brings
about anincreased clearness and acuteness in perception;for at-
tention, addingenergy tothe particularHeld part,will increaseits
articulation, ifit is not articulatedas well as it might be. 3 This
direction of attention toward a particular aspect of the musical
structure and texture is also important because where the center
of ourinterest lies,
there, ceteris
paribus, afigure islikely to arise. 8'
Thus, for
instance, aif piece
were known
to bebuilt upona ground
bass, attention,
focusing uponthis aspectof the musical structure,
would tendto bring the bass
out, eventhough other voices might
tend toobscure its
progress. Similarly
if we know thata particular
movementaistheme and variations,
we areintent onfollowing t_he
theme, and
hence those variations which
in thequality ofthe Hgure
has beenmuch weakenedwill seem better structuredthan they
might seemotherwise. Froma negativepoint of view this search
attitude isimportant because
small differences,
which may
be very
important in the understanding
of a work, may pass unnoticed
if
one isnot setto perceive
them. Itis oftenthe preparatory
set which
brings thisreadiness to
perceive intoplay.
Knowledge and
experience often
color or modify ouropinion
about whatis heard.If, for example, we
see alarge orchestra on the
concert stage,
we immediately become aware of its potential sound.
Consequently opening
an solo
for asingle instrument,
e.g., the
flute
solo atthe beginningof DebussysAfternoon ofa Faun,will have
quite a different effect,will be heard differently,than it would be
were it the openingmusic of a sonatafor unaccompaniedflute.
Furthermore, our expectations what
of will follow, partlybased
upon ourbelief thatthe musicians
are notgathered upon
the stage
by chance,
are colored by thepresence the
of orchestra;
the longer
the solopassage continues,
the strongeris the presumption that
the
orchestra will enter.
To takeanother example,
if oneis listeningto a bell tollingthe
hour andknows, say,
that it is tenoclock, thenthe tenthstroke will
probably be perceived as accented andlonger than the others, al-
though, inpoint of fact, allstrokes were
equal inintensity and
duration. In like manner Ortmanns experimentsshow that what is
considered to
be the end ofa melodyor rhythm depends notonly
upon certainnatural tendencies
of closinginflection andupon
Expectation and Learning 79
cadential formulas
learned byexperience but
also uponwhich tone
in the series-fourth, fifth, seventh, etc.-the subjects were told
would be the final one.
The practicedlistener canrecognize t.he
style andoften theform
of a composition without being giveninformation beforehand. But
even forhim knowledgewhich bringsthe preparatoryset intoplay
is sometimesimportant because it conditionsnot only what is per-
ceived butalso thespeed ofperception and hence ofresponse. An
expected stimulus will be perceived andunderstood more rapidly
than would otherwise be the case.
Motor Attitudes and Motor Responses

Like otheracts ofattention, listening
to musicis accompanied by
physiological and motor adjustments. The physiologicalchanges, as
we haveseen, appear to be products ofthe belief and expectation
that weare goingto havean affective aesthetic experience.
They are
quite generaland areprobably notdifferentiated as between differ-
ent kindsof aestheticobjects. Nordo theyundergo changes that can
be tracedto changesin the stimulus conditions-themusic. Motor
attitudes andresponses involve the voluntarymuscle systems, and,
aside froma generaltensing ofthe muscles related toall feelingsof
effort, ofwhich listeningto musicis aspecial kind,they aremore or
less specific
to particularstyles andforms andtend to change with
changes in the stimulusconditions.
Anticipatory motorattitudes formpart of the preparatoryset.
They arebrought intoplay onthe basisof: a! information asto
composer, style,or form which leads the listener to expect a repeti-
tion ofpast motorexperiences evoked
by theparticular typeof work;
b! program notes orother statementsas totempo, volume, mode,
mood, andso forth that supplyinformation asto the appropriate
motor attitude;and c! visual cluesprovided byperformers inthe
form of gestures and postures, which
lead thelistener toassume a
like attitude, though theseneed notbe manifestin the listeners
overt behavior.Whether basedupon experienceor current clues,
the listenersanticipatory motorbehavior willbe differentif he is
about tohear aStrauss waltz from whatit will be if he expectsto
hear a Bach cantataor a Schoenberg string quartet. Suchadjust-
ments may also bemade toa particularmovement of a workor even
special partswithin a given movement. The motorpreparation for
the hearingof a minuet or scherzo ofa classicalsymphony will
usually bevery different,whether weknow theparticular workor
not, fromthat assumed toward theplaying ofthe slowmovement or
the finale.
Motor attitudesnot onlyform partof thepreparatory set but also
play a part in the perceptionand response sequences made to the
changing progress of the musical form.Changes inrhythm, dy-
namics, tempo, and thelike will bring aboutappropriate changes in
motor attitude. For this reason thepresent discussion of motor
attitudes isnot confinedto their function inthe preparatoryset.
The importance of thelisteners motorbehavior hasbeen implied
or directly stated bycomposers and psychologists alike.
C. P. E.
Bach, for example, tellsus that:
A musician cannot move others unless he too is moved . . . for the re-
vealing ofhis humourwill stimulate a like humour in the listener....
Those whomaintain thatall of this canbe accomplished without gesture
will retract their words when, owing to their own insensibility, they
find themselves obliged tosit like a statuebefore theirinstrument. Ugly
grimaces are, of course,inappropriate and harmful; butfitting expressions
help thelistener tounderstand ourmeaning.
Although espousing a very different aestheticposition, Stravinsky
also emphasizes t_he importanceof motoradjustments int.he under-
standing ofmusic. Thesight ofthe gestures and movements of the
various partsof the body producingthe music is fundamentally
necessary itif is to begrasped inall its fullness. 1
At first sight there would appear to be a distinction between a
response tothe gesture or motor behavior of a performer and a re-
sponse toones ownaural experience.
In point of fact, however, the
distinction is apparent rather than real. For the motor behavior of
the performer, in so far as it is related to the musical continuum at
all, arises out of his own musical perceptions and is therefore be-
havior thatthe listenermight haveperformed directly.That is,the
empathetic response to anothersbehavior, which
is itself a response
to a stimulus perceivedby both persons, generallyserves to initiate
or enforcebehavior thatmight havetaken place
as adirect response
to the stimulus. Thatthe players gestures must
be made only in
response tothe music is also stressed byStravinsky, who observes
that only if the players movements are evokedsolely by the ex-
igencies ofthe music will they facilitate ones auditory percep-
tions. 2
Although motorattitudes bothanticipate andaccompany the re-
sponse tomusic, theprecise roleplayed bymotor behaviorin the
perception andunderstanding ofmusic is both problematicand
complex.
On the one hand, it seems clear that almost all motor behavior is
basically aproduct ofmental activityrather thana kind of direct
response made to the stimulus assuch. Foraside fromthe obvious
fact that muscles cannotperceive, thatthere seems
to be no direct
path from the receptorsto the voluntary musclesystems, motor
responses arenot, as a rule, made to separate, discretesounds but
to patternsand groupings of sounds.The moreorder andregularity
the mind is ableto imposeupon thestimuli presented to it by the
senses, themore likely it is that motor behavior will arise. Such
grouping andpatterning of sounds ispatently a result of mental
activity.
In the Held ofrhythmic experience,
where motorresponses have
been mostsystematically studiedand their importance mostem-
phatically stressed,
]ames Mursell, after a careful and thorough
review of the literature, while admitting the importance of motor
behavior, decidesthat the ultimate foundationof rhythm is to be
found in mental activity.4 CurtSachs, writing
from a very differ-
ent viewpoint,arrives atthe sameconclusion, quotingBrelet tothe
effect that:Rhythm comesfrom themind notthe body.
On the other hand, the facts indicate that somehow motor be-
havior doesplay animportant partin facilitatingand enforcingthe
musical aesthetic
experience. Howthis takesplace neednot detain
us here. However, it does seemsignificant to recognize that motor
behavior often
plays animportant partin makingthe listeneraware,
whether consciously
or unconsciously,
of the structure andprogress
of the music. Some listeners become aware of the tendencies of
music partly in terms of their own bodily behavior. Such listeners
might be said to objectify and give concrete reference to music, to
perceive it through their own motor responses. Andperhaps this in
part accounts
for the emphasis which
has beenplaced uponmotor
responses.
What doesseem clear is thatsince motorbehavior isa productof
and runsconcurrent withmental behavior,it requires no special,
independent analysis; for the experience ofmotor attitudesis not
structurally differentiatedin any way from the mental responses
made inlistening butrather exhibitsa oneto one correspondence
with them.
One point,however, remains. It has been fairly well established
that a regular, periodicmotor pattern,once begunby and in con-
gruence witha mentally perceived pattern,tends to continue, to
perpetuate itself.
Does thismean thatmotor behaviorcan become
independent ofthe perceptionof new stimuli? In concrete terms:
Will the motor responsemade to, say, a rhythm in triple meter
continue andpersist afterthe meterhas changed to four-four?The
answer would seem to be that it can and often will.
Yet, evenhere, the separation between mental and motor re-
sponses, ifone exists,creates nogreat difliculties.For, aswe shall
see in the following chapter, thetendency of a motor action to
perpetuate itselfhas its mental counterpart in the Gestalt con-
cepts ofthe lawsof continuityand completion,which recognize a
similar tendency in the habits ofthe mind see pp.92 f.!. The ques-
tion whichmight beraised, andone thatwe will not attemptto an-
swer, isthis: Towhat extentare thelaws ofcontinuity andcompletion
themselves product
a of the tendencyof voluntarymotor behavior
to perpetuate itself and follow the line of least resistance?In other
words, isthe tendencyof the eye tocontinue itsmotion ina given
way or the mental ear to continue its motions to some extenta
product of the natural tendencies ofmotor behavior?
In conclusion we may say that there appears to be nothing
autonomous and independent about the motor response tomusic.
Everything whichoccurs asa motorresponse can be accounted for
in termsof mentalactivity and,since theconverse of
this isnot true,
music isbest examined in termsof mentalbehavior. Wedo notby
this statementintend tominimize theimportance ofmotor responses.
Their ability to give force and urgency tomusical experience is
evidently ofgreat importance.
III
Principles ofPattern Perception:

The Law
of Good Continuation
General Considerations
Our wholeintelligent process seems tolie in the attentionwhich isselec-
tive of certain typesof stimuli. Other stimuliwhich are bombarding the
system are in somefashion shunted off .... Our attentionis an organiz-
ing processas well as a selective one .... The organism goes out and
determines what it is going torespond to,and organizes that world.1
The organizationwhich the mind imposesupon the separate
stimuli whichare constantlybombarding thesystem isnot anac-
cidental oran arbitraryone. Themind inits selectionand organiza-
tion of discrete stimuliinto iiguresand groupingsappears toobey
certain generallaws. Thesenot only account, inpart, for the way
in which the mind organizes musical stimuli but also explainhow
some ofthe expectations which themind entertainson thebasis of
such groupings arise.
LEARNING AND PERCEPTION

Many ofthese mental laws, formulated
upon awealth ofempirical
data, wereiirst discernedby a group of psychologists who
later
became known as the Gestalt school and whose theories were in-
corporated into a system now known as Gestalt psychology. It is
important todistinguish between the experimental
Endings made in
connection withGestalt theoryand the theory itself, because the
distinction makes clear that it is possible toaccept theempirical
data, thelaws, discovered by Gestaltpsychologists without adopt-
ing thehypothetical explanations furnished bythe theory.
83
For Gestalttheory in reacting against the sensationist concept of
perception and the association theory oflearning leaned too far in
another direction.It attributed ahnost allgrouping tothe spon-
taneous organization of simpleshapes andtended tominimize or
deny the role of learning in the perceptionand organizationof
figures. Sincethe presentanalysis ofexpectation hascontinually
stressed theimportance oflearning inthe selectionand organiza-
tion of sense data,it is necessary to emphasize thatit is employing
Gestalt terminology and utilizing the data supplied byits experi-
ments but that it is not adopting its theoretical explanation of
perception.
This isnot theplace fora critiqueof Gestalttheory. Inhis book,
The Organizationof Behavior,Hebb examines the Gestalttheory
of learningin somedetail andadvances convincing evidence ofits
shortcomings. Heshows that animal experiments and human
clinical dataalike indicatethat the perception ofsimple diagrams
as distinctivewholes isnot immediatelygiven butslowly acquired
through learning.2 Accordingto Hebb,the fundamentaldifliculty
with configurationtheory, broadlyspeaking, isthat it leaves too
little roomfor thefactor ofexperience. 3
It is possible that
the lawsof themind mayin somecircumstances
be independentof cultural conditioning. Wherehuman communica-
tion is involved, however,
though thelaws stilloperate, theydo so
within a socio-cultural context
where attitude,belief, andlearning
qualify theiroperation. Thatthis is so caneasily beseen fromthe
following example. The symbols
RS ET EL T
appear atfirst tobe discrete,individual stimuli.If so instructed, the
mind cangroup thesesymbols, butit does so with difliculty and
the resultis somewhatarbitrary. Ifthese stimuliare arranged thus:
TT RL SE E
the similarity and symmetryof the grouping appearimmediately
LT,,RLS,92Q,!. The
grouping could
be changed
if the factor of
proximity were altered:
T TR L SE E
though evenhere similaritywill play some partin the organization
of the patterns sothat TR may beseen asa subgroupof TTR and
The Law of Good Continuation 85
SE asa subgroupof SEEwith L as anisolated middleterm. In all
these casesthe natural laws of grouping arefunctioning, though
even herethe ability to discriminateeasily between
the symbolsig
probably aproduct oflearning.
Notice, however, that these same
LETTERS
immediately forma convincingand satisfactory Gestalt, whichhas

as itsbasis oforganization nota naturalmode ofgrouping butone
leamed throughexperience. Were it not for the fact that this is 3
word in our languageand ourbeliefs asto the nature ofletters in
general, ourgrouping ofthese stimuliwould be quite different.In
other words, though, aswe shallsee, themind organizes and groups
the stimuliit perceivesinto thesimplest possible shapes or the most
satisfactory and complete figurespossible, what is, in fact, themost
satisfactory organizationin any given caseis a product ofcultural
experience.
DIFFICULTIES IN THE APPLICATION OF GESTALT CONCEPTS

The vitalrole occupiedby learningin conditioningthe operation
of Gestaltlaws andconcepts indicates at the outset thatany gener-
alized Gestaltaccount ofmusical perception is out of the question,
Each stylesystem andstyle will form figuresin a different way,
depending upon the melodicmaterials drawn upon, theirinterrela-
tionships, thenorms ofrhythmic organization, the attitudestoward
texture, and so forth.
For instance,
in a culture orstyle system
where thetonal materials
are arrangedin two disjunct tetrachords:
,E F ALB CE
the fragmentin Example4 wouldbe groupeddiiferently fromwhat
it wouldbe wereit interpretedwithin thecontext ofthe minormode
of Westerntonal music#Furthermore, justas tonal relations will
always bemodified byrhythmic structureso basicstructural tonal
re-gli
EXAMPLE 4
groupings play
an important
part inrhythmic perception.
The fact
that wehear theopening notes of theFinale ofMozarts Symphony
No. 40as aunitary, upbeatgroup ispartly a result of our having
learned toregard thetriad asa singlefigure. Yeteven thisstatement
must beconditional, forthe unity of the triad or any othertradi-
tionally developed tonal setdepends for its unityupon otherfactors,
such as rhythm, tempo, instrumentation, and so forth.
Nor doesit seemthat, evenwithin the confined limitsof a par-
ticular style,a preciseand systematic account ofmusical perception
solely inGestalt terms is possible.Even givenadditional empirical
data aboutaural perception, certain basicdifiiculties inthe applica-
tion of Gestalt principlesto any specific musical process would still
remain.
'I`hese difliculties
do notderive fromany basicweakness in Gestalt
laws perse butfrom thefact thatthe number,interdependence, and
subtlety ofthe variablesinvolved in musical perception make the
establishment a ofsystemof analyticalrules ofthumb impossible.
Although thereis ample reason forbelieving thatthe laws de-
veloped byGestalt psychologists, largely in connection withvisual
experience, are applicable in a generalway to aural perception,
they cannotbe madethe basisof a thoroughgoing system for the
analysis ofmusical perception and experience. This perception,as
we havealready afHrmed, must depend on thesensitive responses
of
experienced listeners.Nor isthe development of thepresent analysis
contingent upon the discoveryof further laws of aural perception.
The laws already establishedcan lead us to, and form the basisfor,
a generalunderstanding of
the naturalmodes ofexpectation as
they
function within the cultural context.
Basic Concepts and Formulations

THE LAW OF PRAGNANZ
The fundamentalaxiom ofGestalt theory is the law of Pragnanz,
which states
that psychologicalorganization willalways beas good
as theprevailing conditions
allow. Inthis definitionthe termgood
is undefined.It embracessuch properties as regularity,symmetry,
simplicity and others which we shall meet in the course of our
discussion. It is of utmostimportance torealize thatthis lawdoes
not meanthat psychological organization will
always besatisfactory.
On the contrary, in many instancesthe figures perceived, the rela-
tion of figure toground, orthe relationof figuresto eachother will
be less than satisfactory, either in and of themselves or in relation
to thestylistic context in whichthey appear or both.
It is this lackof satisfactionwith the psychological organization
that givesrise towhat wehave referredto asthe naturalmodes of
expectation. For the mindis constantly striving towardcompleteness
and stabilityof shapes. This tendency of themind towardregularity
and simplicityof organizationis shown,among otherthings, bythe
fact thata systemleft to itself will, in its approach toa timeinde-
pendent state, lose asymmetries and becomemore regular.7 The
mind whenleft to operate onits own, as it does in the caseof
remembered patterns and organizations, will improve those figures
which arenot as good as they mightbe. This tendency ofthe
mind toimprove thepsychological organization, to discriminatebe-
tween satisfactory patterns andthose whichrequire improvement,
has beenconfirmed bystriking empiricalevidence.
The converse of thisis alsotrue: Goodorganization, stable shapes,
will resistchange and will tend to remainconstant even in changes
of thestimulus conditions. For example, a themeor motivewhich is
well formedwill be perceived as an identity,as thesame theme, in
spite ofchanges in instrumentation, range, dynamics, orharmoniza-
tion. A thing isa particularlywell integrated part of the totalfield.
The strongerits integration,the strongerthe forceswhich hold it
together, themore constantwill it be in changes ofstimulation.
. . . The betterthe psychological organization, the less likelyis it
that expectation will be aroused.
THINKING, MEMORY, AND EXPECTATION

Without thoughtand memorythere couldbe no musical experi-
ence. Because they are the foundationfor expectation,an under-
standing ofthe wayin which thought andmemory operate throws
light both upon themechanism of expectation itselfand uponthe
relation of prior experienceto expectation.
Max Wertheimer,one of the most important membersof the
Gestalt school,
describes the
thinking process
in the following way:
Generally speaking, there isfirst a situation
Sl, the situation inwhich the actual thoughtprocess starts, and then,
after anumber ofsteps,
S2, in which theprocess ends, the problemis solved.
Let us consider thenature of situation 1 and situation2 by comparing
them, andlet us consider whatgoes onbetween, howand why. Clearly
the process is atransition, achange fromS1 intoS2. S1 as compared to S2,
is structurallyincomplete, involves a gapor a st1'uctural trouble,whereas
S2 isin theserespects structurally better, thegap isfilled adequately,the
structural troublehad disappeared; it is sensibly completeas againstSl.
When theproblem isrealized, S1 contains structuralstrains andstresses
that areresolved inSz. Thethesis isthat the very characterof the steps,
of the operations, ofthe changesbetween S1and S2springs fromthe
nature of the vectorsset up in these structural troublesin the direction
of helping the situation,of straighteningit out structurally.
. . . often the process
does notstart with S1 andend with S2, butthat in
S1 already
is partof a development; that,
moreover, S2,the very solution,
does notrepresent anend but that by its very nature leadsto further
dynamic consequences.
The relationof this description ofthinking andproblem solvingto
the aestheticprocess isself-evident. Problem solving andthe aes-
thetic processare essentially
one andthe samething exceptfor this
proviso: thatin aestheticthinking, therelationship betweenstruc-
tural troubles and their resolutions are intelligible and resolvable.
It is also evident that thinking, the overcoming of difficulties, and
expectation are one process.Finally, theplace ofaffect inthis proc-
ess isnot diflicult to discover.If expectationresults from,say, a
definite structuralgap, the delay in completion of the thought
process will result in affect unlessthe processis rationalized on the
conscious level.Alternatively a confused anddoubtful situation,
resulting ina generalizedexpectation ofclarification, willgive rise
to feelingsof suspenseand affectiveresponses.
Expectation depends in very important waysupon memoryproc-
esses. Aswe listen to a particular musicalwork we organize our
experience and hence ourexpectations both in termsof the past of
that particularwork, whichbegins afterthe Hrststimulus hasbeen
heard andis consequently past, and in termsof our memories of
earlier relevant
musical experiences.
Memory isnot simplya recordingmachine inwhich pastevents
and experiences
are storedin an immutable way.
The tracesleft in
the memory
by experience
are constantly
changing. These
changes
may be grouped intothree classes:
normalization, emphasizing
or
pointing, andautonomous changes.
Normalizing occurswhen the reproductions [ofremembered figures
made bya subject]approach successively a familiarform; pointing,when
a particular feature of the pattern, which strikesthe observeras such
when heperceives it,becomes more and moreexaggerated; autonomous
changes, lastly,are suchas derivefrom neitherof the two other sources
but are inherent in the trace pattern itself, a result of its own intrinsic
stresses.
All such changes takeplace in the rememberingof music. Since,

however, musicis an art which is essentiallywithout external
referents, these
processes occurwithin a more orless closed system,
and memoryoperates either between differentmusical experiences
or betweendifferent partsof the same experience.
The fact, already noted,that a system leftto itself will, in its
approach toa time independent state,lose asymmetriesand become
more regular12 applies alike to parts of a pieceand to the whole
musical work.The law of Pragnanzfunctions withinthe memory
process, which tends tocomplete what was incomplete, to regularize
what wasirregular, andso forth.Moreover, those shapes which are
not wellHgured and which thememory isunable tostraighten out,
complete, or make symmetrical will tend to beforgotten.
In other words, anunstable memory trace will Hrst tendtoward
stability; and,if this is impossible,it will tend to disintegrate. This
aspect of the memory process would seem to account for the fact
that thewell-organized processes
of themesand melodies are better
remembered than the moreor lessirregular partsof a musical work;
e.g., themelodies ofthe Hrstor secondtheme groupof a sonata
exposition arebetter rememberedthan the irregular and often
quasi-chaotic processes
of the development section.
As wehave seen, even thoseHgures whichare remembered are
changed inthe process.This meansthat we tend to remember
themes as being simplerthan theyreally areand thatwe remember
forms asideal types rather thanas particularthings. Foras rec-
ollections of similar types,whether of parts or wholes, become
regularized inthe memory,they tend to be grouped intoclasses,
t.hus formingthe normswhich are the basisof stylistic perception
and expectation.In other words, tracesproduced by similar ex-
citations donot remainindependent ofeach otherbut form larger
trace systemswhich influence newly formed traces in definite
ways. 1*This process of normalizationis connected to what Koffka
has calledthe averagingeffect-an effect whereby similarfigure
processes become formed intoa singleclass concept. Of course,such
concepts ortrace systemsare not really averages;but . . we
possess within our storeof tracesmany systems which, througha
process ofcondensation and assimilation formthe basisfor 'class'
perceptions, forthe normal and unusualf1
Normalization plays an importantpart infacilitating therehearing
of musicalcompositions. As noted earlier,the normsdeveloped in
the memoryare notrigidly fixed but changewith the addition of
each newmemory trace; to the extent thatthe normshave changed,
a rehearingof awork isa newhearing, yielding new insights.Those
factors which are the immediate cause of affect and aesthetic re-
sponse, the
deviations, are
the veryones thateither become regular-
ized andaveraged orforgotten. Forthis reason
they tendto surprise
us, to remain deviantseven after many hearingsof a work.
It may, at this point, beobjected thatafter all we do not really
forget the asymmetrical partsof a piece of music. We can, for
example, recognize a symphonyby hearinga few measures ofits
development section, presumably theportion of the piece which
will most likely be forgotten or changed inthe memoryprocess.
The objectionis not cogent because there isa greatdifference be-
tween recognitionand recall. For instance,we may be able to
recognize aportion of a musicalwork, evento the point of recog-
nizing eachtone asit is presented, withoutrecalling it,in the sense
of beingable topredict thesuccession of tones correctly. But since
the ability to envisageaccurately, topredict, is just what distin-
guishes normsfrom deviants, the deviants would still seem unusual.
The veryfact thatwhen thedeviants arepresented we say toour-
selves Oh, yes, nowI remember is a clear indicationthat they
were notreally, oronly vaguely,expected.
Memory tendsto improve shapes inthe directionof regularity,
symmetry, and
completion. Butthis doesnot meanthat completed
tasks or shapes are better remembered than ones which are not
complete. Whereincomplete Hgures
set up real tensionstoward
completion, where shapes are well articulated so that the manner
of their completion isclear, theyare betterremembered than
com-
plete ones."
Finally, memory,activated byknowledge andinformation, isa
force inorganizing musical
experience, for
it bringssearch attitudes
into play: . . . a true search attitude does much more than estab-
lish the communication between trace andprocess, itdetermines
new processesnot only through suchcommunication butalso by
prescribing whatthe resultof the communication is
to be: the new
process isto be organized in terms of the trace. 17
The Principlesof Pattern Perception

In the statement ofthe law of Pragnanzto the effect thatpsy-
chological organization will always be as good as the prevailing
conditions allow see p.86 f.!, the term good was leftundeBned.
It is to the conditions making for satisfactoryand, equallyimpor-
tant, unsatisfactorypsychological organization that we now turn.
These conditions, which are at bottom but corollaries of the axio-
matic lawof Pragnanz,
are expressed
in a set oflaws andprinciples
formulated byGestalt psychologists
on the basis ofempirical evi-
dence.
Although thepsychological organization is alwaysas goodas pos-
sible, thisdoes notstipulate thatthe organization
is alwaysas good
as the mind would wish. It is this dissatisfaction with the psycho-
logical organization
which givesrise to expectation and
perception
of deviation.Good continuationand goodshape [are] powerful
organizing factors,
and both[are] in the truesense understandable:
a line carries its own law within itself, and so does a shaped area
or volume. Violations of this law due to external forces are felt as
violations; they conflict with our feelings of the fit, hurt our sense
of beauty. 1
Actually they"hurt oursense ofbeauty onlywhen theirfunction
and significance cannot beunderstood. Ajagged line understood
merely asline may be unpleasant;for if it cannot be related to
other aspectsof experience,
its irregularitywill seempointless. Con-
sequently thetensions aroused in perceivingit as pattern will seem
meaningless and unpleasant. But the same line placedin an aesthetic
context, whereits perceptionis understoodas part of a total ex-
perience andwhere belieftends tocreate adisposition torespond,
will seem exciting and significant. Similarlya dissonanceor an
ambiguous progression which might be unpleasantwhen heardin
isolation maybe beautifulwithin apiece ofmusic whereits relation-
ship to past eventsand impendingresolutions isunderstandable.
The lawsand principlesthat follow are closelyinterrelated, and
their functionsoften overlap.A violationof the law of completion,
for example,almost alwaysinvolves disturbances in the factor of
good continuation, though thereverse ofthis isnot necessarily true.
Because ofthe interactionthat takesplace, thelaws andprinciples
discussed below must be treated as convenient distinctions between
various facets
of mentalorganization ratherthan asclearly separable
mental functions.
For this reason norigid and systematic compart-
mentalization of
the discussionhas beenattempted.
THE LAW OF GOOD CONTINUATION

A shape or pattern will, other things beingequal, tend to be
continued inits initial mode of operation. Thusto the factor of
good continuation in purely spatial organization
there corresponds
the factorof thesmooth curve of motionand continuous velocity in
spatio-temporal organization.2 Amongother thingsthis law helps
to accountfor our being ableto hear separate, discrete stimuli as
continuous motionsand shapes.
Actually, ofcourse, aline or motion doesnot perpetuateitself. It
is only a seriesof lifelessstimuli. Whathappens isthat the percep-
tion of a line or motion initiates a mental process, and it is this
mental process which, followingthe mentalline of least resistance,
tends tobe perpetuated and continued.This is important, notonly
because we shall, forconvenience sake,
often speakof a process as
continuing itself,
but alsobecause itemphasizes that
a lineor motion
is actually a processof the mind rather than a thing. Sincethe
complexity ofa motionoften makesit difiicult to decidewhat con-
stitutes continuation and whether it has been disturbed or not, it is
the processas determiningthe motion both from the standpoint
of the perceivers mindand fromthat of the composers
technique!
which we must examine.
Process continuationis the norm of musical progression,and dis-
turbances incontinuation arepoints of deviation. Thesedisturbances
in the process of
continuation may
be oftwo kinds: a! gaps in the
process inwhich a process istemporarily halted and then continued
again, andb! changesin process,in whichthere isusually, though
not necessarily,
a breakin lineand one
manner of
progression takes
the place of another. Both kinds of disturbancesmay occur to-
gether, aswhen a process change takes placeafter a halt in the
progress ofthe music.
The motion by which one processchanges toanother will be re-
ferred to as processreversal or simply asreversal. Since proc-
esses may be moreand lesssimilar, itfollows thatreversals may
be
more and less drastic. For example, during a modulation several
types of sequences may be used, one replacingthe other. Each
change willconstitute aslight reversal
and willbe apoint oftension.
However, thepoint at which the modulation process is replaced,
say, bya steadystatement onthe dominant,will constitutea major
change in process.
Continuation mustbe carefully distinguished fromrepetition.
Continuation always implies changewithin a continuous process,
not mererepetition. Andwhile continuationappears tobe a normal
mode of operation, repetitionis so only up to the point at which
saturation sets
in seepp. 185f.!. However,even ourexpectations as
to continuationare tosome extentsubject toour expectations as to
change andlogic; thatis, weexpect continuation only solong asit
appears significantand meaningfulin the sense thatit can be un-
derstood asmotion towarda goal. If meaningbecomes obscured,
then change will be expected.
MELODIC CONTINUITY
Chopins Prelude,Op. 28,No. 2 presents aclear exampleof the
establishment ofa process,its continuation,a disturbance,and.
finally, there-establishment aofvariationof theoriginal process.
The
melodic phraseExample 5! consists oftwo similarmotives joined
Lento we
:fn
'- _ iv- 77 ; ii . ia
'o 00 ~f ~-
f
EXAMPLE 5
by the fact thatthe sametone beginsthe second fragment asended

the first.The firstand secondphrases are similarly linkedby a com-
mon tone,though thesecond phrase is displacedby an octave Ex-
ample 6!.This process of conjunctionby commontone establishes a
Phrase I, Phrase 2.
lf 4x A_ f "1 X 92
Y_ l Il
|l.__l1-Q_|lF1lT|1
92l|_-111--ll H
o t - #4
EXAMPLE 6
strong forcetoward continuation.

We expectthe nextphrase, even if
it involvesnew melodicmaterials, tobegin withsuch atone conjunc-
tion. But this does not take place. The continuity is broken in
measure 14 by the entrance of the A where F-sharp would have
been theexpected toneExample 7!.
UA I __brea92l~92
EXAMPLE 7
The forceof thisbreak isnot completely apparent untilthe motive

is completedon theF-natural, sincethe A to E progression might
simply betaken tobe a repetition ofthe endof the second phrase.
The F-naturalenforces theeffect of the breakbecause, following
the E,it introducesthe firsthalf stepin themelody. Afterthis break
in continuity,the originalprocess ofprogression bytone conjunc-
tion is re-established and, with some modifications as to motivic
order, continues until the Hnal cadential formula is reached.
The melodic break which occurs in measures 14through 16 is
paralleled bya breakand change in harmonic process, but
with
this difference:
the harmonicchange conclusive,
is the
in sense
that
the oldprocessnot
is re-established
as was the case
with themelody.
Without presentinga detailedharmonic analysis
of thePrelude, it
is clear that the harmonic motion of the iirst sixteen measures
might be symbolized as
in Example8.
:F
ia;
E : _ ` I f 3iaI f ' 1_1{r==iYi'i=1~lìl192
Q xiii __.- '=r_
:= - : i=;:, :i1=-,_. _,,. _ , 1av.: f
:Sis 2 ,_
2""""""'"if"'1'~;; = = -. -~ -- ` 11-53'/'.iZI=
U; ,_ , ;'l xi.:ls92-L-l1392ii
ro -_eat ' I LQJ -- ----I-'I'-I'--_ _
G:YI .... ---- If Y.--- 'I----.,
D:Y! ..... YI -.If-.Y Loltered!
a:lI_1-_h,--.Z ZITI: Y
EXAMPLE 8
Although theopening phrase is originallyheard in E miner, for

the sakeof simplicity it has been symbolized in terms of G, to
which it moves. The
second phrase,
which seems
to be exactly
parallel tothe first,leads usto expecta D major chordat its con-
clusion. Butthis hannony
never materializes.
Instead there
is an
irregular resolutionto an altered chord,whose rootmust be con-
sidered asbeing D altered to D-sharp. Herethe processchanges
and the change is,so to speak, suspended during the progressive
alteration ofthis chorduntil the augmented sixth
chord in the last
half of measure 14 is reached.The irregularand indecisivechar-
acter ofthe harmonic
motion gives
rise tofeelings of
ambiguity and
uncertainty, whichare resolvedby the relatively clearand regular
cadence from measure 14 to 15, in which the augmented sixth
chord still an alterationof the harmony whichshould havebeen
D major! moves to the tonic 6/4 in A minor.
One of the most striking things about this progression isthat
had thesequence continued in the regular manner,with an altera-
tion to minor at the end of the phrase, thesame harmonicspot
might have
been reached
with onlyminimal deviation;
that is,be-
ginning with the secondphrase, wewould havehad the following
progression:
D: IV!-VI-If-V-I
A: IV!-VI-I2-V-I aminor!
It seemsperfectly clearthat any technical explanationsof meas-
ures 12to 16 purely in terms of harmonic goalsand modulations
must beinadequate, since the samegoal couldhave beenreached
in a much moreregular way.The explanation lies inthe importance
of doubt and uncertaintyin the shaping ofaesthetic affective ex-
perience.
It is important to realize that certainty anddoubt are relative
terms. Thebeginning phrases of this Prelude areonly relativelycer-
tain, particularly if we consider thestylistic contextin which it
should beheard andits relationto the preludes whichprecede and
follow it. The constant use of non-harmonic tones and added sixths,
etc., in the accompaniment Hgure togetherwith the over-all sub-
dominant progressionC to D to A! produce a feeling of indef-
inite tension,a kind of relative uncertainty, fromthe very begin-
ning. The relative uncertaintymoves tomuch morestriking and
forceful uncertainty in measure 10, where it is, in turn, resolved to
a relativecertainty inthe arrivalof theA minor6/4 harmony.Com-
plete certainty,toward whichthe pieceprogresses from the begin-
ning, isachieved only with the final cadence,
the proprietyof which
is apparentfrom thisanalysis.
Intensity ismaintained tothe endby the delay in the resolution
of the 6/4 chordon A minor. Noticethat while,from onepoint of
view, the harmonic process is discontinued,from anotherpoint of
view, the whole motion from measure 12 to the end is at least
similar tothe originalprocess, though
much prolonged:
a: IVj#-IVj#-
VI, omitted!
-I prolonged!
of-V-V
V-V-I
ii
This exampleis interestingfrom severalpoints of view. First, it

does not have a beginning in the sense that there is no statement
but only a motionand aconclusion. Inthis respectit is reminiscent
of Wertheimersdescription ofthe thinkingprocesses in which we
have ...... S, ...... S2 ...._. ; that is, the openingphrase Sl!
is alreadypart of a process,though in this casethe final cadence
does representa final solution. Beginningduring a process also
contributes to the general aura of vagueness thatpervades the
whole Prelude. Second, thispiece illustrates the difference between
discontinuity through
a changeof processin the harmonic motion!
and discontinuitythrough adelay andbreak in a process in the
melodic motion!which is subsequently resumed. Finally, the ex-
ample isnoteworthy becausethe reversalof processand thebreak
in melodic continuity which constitute the climax do not occur as
the result of the typically risingprogression buttake place,so to
speak, inthe courseof a gradually descending progression.
Some aspectsof a process may

exhibit continuitywhile othersdo
not. Thesequential opening
phrase ofthe Liebestod of Wagners
Tristan undIsolde Example9! establishesa processwhich leads
us to expect a definite continuation. Up to measure 5 the clarinet
and voicepresent essentially
the sameline. But at measure5 the
sequential process
is brokenin the vocal partwhile theinstruments
continue thesequence.
Sehr missig
beginnend
Voice
Bass cloringf
:r::f'x:==*" '-: 1:+=:.i = a:
clarinet!
EXAMPLE 9
For the sake ofclarity, letus turn our attentionHrst tothe vocal
part, thento theinstrumental line,
and, finally,
to therelation be-
tween the two.
Although thesequential process
is brokenin the vocal lineat
measure 5,
since weexpect theline to begin onA andmove toD as
the hornpart does,
the over-allline Example10!, whichbegan
with the A-Hat inthe firstmeasure and
moved upwardthrough the
B-flat toC-flat andthen C-sharp,
is notbroken. Thetones whichwe
expected are presented, but
not inthe orderexpected on the basis
of theestablished sequential
process. Onthe otherhand, whilethe
break inprocess represents
no basicbreak inthe line,it doesherald
a delayin the over-all line,
which doesnot continue its upward
surge untilthe Hnal
measure, when the motionto theupper A-Hat,
a naturalpoint of completion, isreached.
l;S'|1 ii
|;n4_1;lI*ili1i _ 11 1 'll 1 - 1 'lrlgl
_ ,iii
,_1.1__1; __gii_ _ 1|11r1;!-I
$=92i1$| _ U 2-lilii-I
l'1.l;l.L-l'Y;l_

:n|. QT fm-z.| __
I! 'H | II !_- I I 0
fvlnf-.|.-X
l.l'__2' _I |- -92.|w'-'--"----_jf-Z2-._'.
7-i-tlxalul-Q-,_ I -|-- _ *_*
92| i 1I 'I m 2 - '-
@
EXAMPLE 10
Notice thatthis breakin the melodic sequence is accompanied by

discontinuity ofrhythm, whichis all the morestriking becauseof
the continuousness of theinstrumental line.Actually thevocal part
in measures5 and6 issyncopated against the instrumentalline and,
as weshall seepresently pp.113 f.!, involves importantrhythmic
changes. Thesechanges anddeviations from the instrumental
rhythm shouldnot beconfused withthe slightlydelayed entrances
in measures2 and 3. For though the latter have an expressiveeffect
by delaying and disturbingthe process,they work within that
process, whilethe rhythmicchanges inmeasures and
5 6 are both
more striking and more important.
._ 1
Ivlu. |I'92..- 'I i 7_ ,_ E _ _ -l .
r|N92;92_|;1x.r|1192.;111-|
|r_:A~11|I-| _lui |lT'`_i"H_92l._
111l|ll1u'4.;l _ |-1; ii?-
i'-i ' 1
EXAMPLE 11
The instrinnental line presentsa strong contrast to the vocal line.

While the vocal line involves abreak in process andrhythm at
measure 5, the instrumental line persists with single-mindedness
Example 11! on its sequential way as the whole motion is accel-
erated. The second measure of the motive is now omitted so that
t.he motion
of anascending fourth
followed bydescending half
steps
is heard in each successive measure.
The interesting thing about the
process isthat, thoughpart of the melodyis omitted,the intervallic
progression upward is notaltered, forthe newmotive always begins
on the same toneas thesecond halfof the motive does Example 12!.
As aconsequence this
of modificationthe process
continues basically
as beforebut at a morerapid rate.
EXAMPLE 12
The acceleration is again increased in measure 7,where the en-

trances of the motive occur at t.he half measure. However, t_his in-
creased rate
of progression
also marksthe endof the sequence. For
the interval A-flat to D-flat in measure 7 does not continue the se-
quence but instead serves
to reunite the vocal and instrumental
lines. Both the point at which the break takes place the A-flat!
and the melodic motionof the union measure 7! are foreshadowed
in the previous measures.
The importantbreaking point,the A-flat,
is impliedby themotion fromD in measure 5!through theC and
B-flat in measure 6!.The melodicmotion isimplicit in the vocal
line of measure 6. That is, if the vocal line of measure 6 had moved
upward in fourths, it would have had the same melodiccontour
as both the instrumental and vocal line have in measure 7. The
unification of the instrumental and vocal lines is emphasized and
articulated bya moredecisive harmonic
motion. Andboth melodic
and instrumentallines moveupward to the high A-flat, toward which
they have been tendingfrom the beginning.
Here, then, we have the simultaneous occurrence of ! a break
in process in the vocal line! from the point of view of sequential
progression but only adelay fromthe pointof view of over-allmo-
tion; ! a continuous process in the instrumental line! whose
mode ofprogression altered
is indetail butnot inbasic motion;and,
finally, ! the stabilizationof both processes through their re-
uniHcation and through their motion to a point of relative repose.
The relationshipbetween theprogress ofthe vocal line and the
instrumental lineis particularlyinteresting. Forthe feelingof delay
in the vocal lineis intensifiedby the accelerated motion of the in-
strumental sequence and viceversa.
The melodyunder consideration presents an interesting example
of alternativepossibilities ofmelodic continuationsee p.51!. The
opening of the two-measuremotive createsa structural gap by
skipping from E-flat to A-flat. The ensuing descending motion
A-Hat, G,C-Hat, F!begins toH11 this gap, establishinga continuity
process whichleads thelistener toexpect areturn to the opening
E-flat see Example 13!.But the introduction ofa rising progres-
sion C-flat, A-flat, B-Hat!during the course ofthe descentgives
rise toaltemative expectations. The tendencies inherent inthis new
process aresupported bythe rising melodic linefrom measureto
measure see Example 10!.Not only is thearrival ofthe low E-flat
delayed infavor of this risingprogression, but from measure5 to
measure 8the ascending motion is itself delayed.And the arrival
of both low andhigh E-Hatsmeasures 7 and 8!, thus makea sig-
nificant contributionto the feeling of resolution whichprevails at
the endof the whole period.
_c 1 -l
| ._'vii
92 3 i _ ilnlri 1
'l'l_._Ql@|i
.li:1Y'1l_l U-1 llil
1 rl 1.
EXAMPLE 13
Notice that just ast.he pointof processdisjunction andthe be-

ginning of delay in the large melodic shapewere markedby an
unexpected tonalsequence andbolder temporalgaps, so in the
smaller motive
delay isaccompanied by
a lessregular sequence
of tones and a slight rhythmic disturbance. Theparallelism be-
tween themotivic construction and theover-all phraseorganization
is anotherinstance ofthe architectonicnature ofmusical processes.
Since theaffective qualityof the whole isconditioned bythat of
its parts, the smallerdelays incontinuity mustalso be examined.
These takethe form of thosedevices thatare generallydiscussed
under the subject of omamentation. Inthe exampleunder con-
sideration, theappoggiaturas actto delay the normal continuation of
the quarter-notemotion Example14!. That is, we expect themo-
tion initiated by the skip of a fourth-for instance, fromE-Hat to
A-flat in the firstpresentation of
the motive-tocontinue onthe next
beat. Variousalternatives are possible withinthe stylesystem. The
motion fromthe fourthmight, for example, move as partof a triad
on up to the C; it might movestepwise afterthe fourth, as in the
melody whichbegins theslow movement of Beethovens Symphony
No. 2;or it might beginto closethe structuralgap bydescending at
once, asin the Faith motive from Wagners Parsifal see Prelude,
measures 44f.!.However, whatdoes happen is that the tendency
toward furthermotion isdelayed bythe repetitionof theA-Hat, and
the effectof this delay isheightened because
the repeatedtone is
a dissonance.
_+
1|'l
EXAMPLE 14
The affectivecharacter ofthe appoggiatura and otherornaments

as wellas theaffective character
of chromaticism,which alsoplays
an importantrole in the effect of this passage, willbe discussed
more fullyin chaptersv andvi. At this pointwe mustbe contentto
remark that the expressiveaffective nature both of ornaments and
chromaticism isclearly indicatedby evidencefrom Western,non-
Western, and
primitive sources.
Continuity always operates within a particular cultural stylistic

context, whichnot only conditions expectationsabout suchobviously
cultural factors as harmony but often plays an important part in
determining our expectations of melodic continuation. A common
example of this is the melodic progression upwardthrough the
triad, asin Example15. Thistypical patternforms thebasis ofsuch
102 Emotion and Meaningin M usio
diverse melodies as the main theme of the Erst movement of Schu-
manns Piano
Concerto, the
theme ofthe slowmovement ofHaydns
Symphony No.97 in C Major, and thetune Three Blind Mice.
l; f& "5
- 1' l92
EXAMPLE 15
RHYTHMIC CONTINUATION
Thus farcontinuity hasbeen discussedlargely interms ofspatio-
melodic andharmonic processes. The vital and everpresent factor
of temporal organization hasreceived little attention. In view of
the numerous and well-known d'iculties involved and the incom-
plete stateof our knowledge ofthe subject,it would indeed be
pleasant toignore thefactor of temporal organization altogether.
It is, however, sovital in achieving anddisturbing continuitythat
we must at leastmake sometentative observations on this aspect
of the musical process.
Distinctions mustfirst of al] be made betweenpulse, meter,and
rhythm.
1. The perception ofpulse involvesan objectiveor subjective
division of time into regularly recurring,equally accentedbeats.
The ticks
of ametronome aorwatchare pulses
or beats.
Such equal
pulses willnot giverise toan impression
of eitherrhythm or meter
unless, ofcourse, themind of the listenerimposes somesort of
differentiation upon
the separatebeats.
Though afeeling ofpulse isnecessary ifan impressionof meter
is to arise andis generallypresent inthe perception
of rhythm,
pulse can and doesexist alonewithout creatingeither meter or
rhythm. In order for this to occur, there must be no differentiation
of thebeats withrespect toaccentuation. They
must beequal. In
fact, whatare later
to betermed incomplete
rhythms areactually
a seriesof pulsessee pp.144 f.!.
2. The perception ofmeter involvesan awareness of the regular
recurrence ofaccented and unaccented beats.The necessary condi-
tion for metric organizationis the differentiation of pulses into
accented andunaccented. There must be a feeling of the basic
beat if the feelingof meteris to arise. Butthis pulseof beat need
not beactually heard. It may be carriedin the mind of the listener
see pp. 118-19, 242!.Because theimpression ofrhythm depends
not onlyupon theexistence of accented and non-accented beats but
also uponthe groupingof thosebeats, metercan, in a sense,exist
alone withoutany impression of rhythm. For wherethe listeneris
unable togroup theunaccented pulses in a definitive way-where
rhythm isambiguous-the impression may merelybe oneof strong
beats andweak beats following oneanother witha givenfrequency.
An instanceof this type of organization isdiscussed inchapter v
see pp.147 f.!.
3. The perception ofrhythm involvesa mentalgrouping of one
or more unaccented beats in relation to an accented beat. These
groupings may,of course,be more and lessclear; andwithin any
given meterthey mayvary indefinitely.They mayin this sense be
more and less regular.Furthermore, theaccents ofthe rhythmic
group, thoughgenerally supporting
the metricorganization, may
at
times conflict with that organization.
In referringto thepatterns whichresult fromsuch groupings,
we
shall use the terminologytraditionally associatedwith prosody:
iamb V-!, anapest VV-!, trochee -V!, dactyl -VV!,
and amphibrachV-V!. Finally it should be notedthat rhythm
can exist alone without pulse or meter as it does in plain chant, the
rhapsodic fantasias
of many diEerent cultures,
or recitativo secco.
The basicdifference betweenpulse, onthe onehand, andmeter
and rhythm, on the other, lies in the fact that the latter modes of
mental organization involve the differentiation of beats into ac-
cented and unaccented, while the former does not. This makes
some sort of definition of accent desirable.
Basically anything is accented when it is marked for conscious-
ness in some way.Such mentalmarking may be the result of differ-
ences in intensity, duration, melodic structure, harmonic progres-
sion, instrumentation,or any other mode of articulation which can
differentiate onestimulus or group of stimuli from others. Even
silence, arest, maybe accented,as isthe casein the second measure
of the fifth movementof BeethovensString Quartetin C-sharp
Minor seep. 149!.In other instances tone
a orgroup oftones may
appear to be accented,not becauseof any particular distinction
which it possesses per se, but because apreviously established
grouping tendsto perpetuateitself, makingthis type of organiza-
tion the simplest one.
Accent should not be confused with stress. As observed earlier,
silence maybe accented;the literature of music is replete with
examples ofpianissimo accents.
Stress isthe dynamicemphasis of
either an accented or an unaccented tone. VVhere an accented tone
is stressed,the stressmay changethe rhythmic grouping or may
help to clarify anotherwise ambiguous
rhythmic organization,
but
it does not create the accent. Nor does stress placedupon an un-
accented beat alter therhythmic grouping.Such abeat is still per-
ceived asunaccented, not only because of the tendency ofa given
grouping toperpetuate itself, but alsobecause, as we shallsee, the
placement in the temporal organization ofan unaccentedbeat,
whether stressed or not, is physicallydifferent fromwhat it would
be were it really an accent.Thus asa rule an offbeatsforzando or
forte should be classed as a stress rather than as an accent. A
familiar instance of such an offbeat stress is the fortissimo sur-
prise in the second
movement ofHaydns Surprise Symphony. It
is clear in this case that the strong stress doesnot affect the basic
rhythmic structure,though it obviously modifies the characterof
the theme.Indeed, theeffect of the fortissimostress ispartly the
product of our knowingthat it is not the real accent. VVhileit is
important to distinguish betweenaccent andstress, it should be
noted that there are instances where such an offbeat sforzando or
forte shouldbe treatedas an accent alteringthe rhythmic group-
ing. The performer mustbe awareof this possibility, forhis deci-
sion asto the significance ofsuch toneswill literally determine his
placement ofthe beats,his performance of the passage.
Rhythm and meter, though obviously intimately interrelated, are
nevertheless independent variables. This will be evident as soon
as one considers thatseveral differentrhythmic groupingscan
arise within the samemetric organization,as the Examples 16-20
show.
The Law of Good Continuation
a! Iambic grouping.-
Allegro molto
e vivoce
nf
ri..ii11 1- ' 1
11 x |.ln
-" '- - |-4.
EXAMPLE 16
BEETHOVEN, FIRST
SYMPHoNY, MINUETTO
b! Anapestiogrouping.-
I . .' - 0' ` Xhg T
111_*l-1_11111-_jI.llj1l1-$1.1
1 li 1. 1-11 il,-
|92
`Q .--_-_I_- -Q1'1___2_-1l1_1__--1
:1-ml 1111-_1_1j_--__l'_ ____-Q-Q
!- il ll
92.
-4 ga
*wsu -_vyd '*_,92|I92-I -_9292
-L *vv -9 J1 Q ._
EXAMPLE 17
B1zE'r, CARM1-:N, Ac'r III, ENTRAC'I'E
c! Trochaic grouping.-
Menuetto
I _.ill-_-I----_-Qi
1 -__"2-Z_-- ="
; - .J f 1
e- "4r"- r"-1n'-1
EXAMPLE 18
MOZART, STRING
QUART1-:'r INA MAJOR KA64!, M|NUE|~r0
d! Daotylio grouping.-
Molto vivace
YCiI_111llQ121-.
I92_ 'l-
J _i I xuilqiiiiiii
_Z l
** 92 _ hl92J P sl MJQ'._92SJJ
EXAMPLE 19
BEETHOVEN, NINTHSYMPHONY, Scmnnzg
e! Amphibraohgrouping.-
Allegro molto
_;. 1_
r|.11_r';11;:-r.i;1l1..1 1-1 _! 11
| _:vin
92 IZ' ;Axiii!:
I 1 11 _ I_ S
3
S-I '* 92l"92I hd i 92 I V ig5
EXAMPLE 20
HAYDN, SURPRISE SYMPHONY, MINUEITO
In like manner thesame rhythmic grouping mayarise outof differ-
ent metric organizations. Forinstance, aniambic grouping can
exist induple aswell asin triple meter, asis thecase inthe opening
theme of the first movement ofBrahms FourthSymphony; like-
wise an anapest grouping can arisein duple meter, asit does in
the Cavotte en Rondeau from Bach's Partita No. 3 for unaccom-
panied violin;a trochaicpattern canalso arisein duple meter, asit
does inthe tuneAu Clair de la Lune; and an amphibrachrhythm
can existin duple meter, asis the case inthe openingtheme of
Mendelssohns Violin Concerto.
However, some
rhythmic groupings
tend tobe incompatiblewith
particular metric schemes. Duplemeters, for instance, do not easily
give rise to dactylicpatterns, whiletrochaic patternsare diflicult
to achievein triple meters. Tounderstand thisincompatibility of
rhythmic groupings with metric structure, itis necessary
to discuss
some ofthe waysin which the mind organizes accented and un-
accented beatsinto cohesive groups.
A seriesof stimuliof equalintensity andequal durationcreate no
impression of rhythm unless the mind imposes its own differentia-
tion on them. Someof the stimuli mustbe perceivedas accented
relative toothers. Whenthe mindimposes its grouping uponsuch
a seriesnot onlyis thevividness ofthe beatssubjectively conditioned
by the mind butso alsois their position inthe series.
If a series ofequidistant soundslike the ticks of a metronomeare heard
the observermay phrasethem into iambs, trochees, or dactyls, as he
chooses. To the ear,the choiceseems toaffect merelythe vividnessof the
beats. Butif the observer tapsa key at eachbeat of the metronome,it
will be found thatin the case ofthe iambhe alwaysmisplaces thebeats;
often the same thinghappens inthe caseof the trochee andthe dactyl.
Because ofthe influenceof the rhythmic process, the observerdoes not
hear the beats wherethey actually occur. In the caseof the iamb he
always hearsthe beatwhich he makes subordinate nearer theaccented
beat thanit actuallyis in the objectiveclicking ofthe metronome.
These conclusionswere conHrmed by the studies of Woodrow,
who found that when time intervalsare equaland every second
sound is accented, therhythm will appear tobe trochaic. If intervals
are equaland everythird soundis accented,
the rhythmwill appear
as a dactyl. Thusthe trocheeand dactylmay begrouped together
in the sense thatboth are primarily productsof intensity differ-
ences rather than durational differences.
]ust the opposite isthe casewith iambic and anapestic rhythms.
They arebasically products of durationdifferences. Ifwe startwith
a trochaic rhythm and gradually increasethe interval after the
louder sound,we arriveat an iambic rhythm.Similarly if we begin
with a dactylic rhythmand graduallylengthen theinterval afterthe
louder sound,the rhythm tends to become ananapest. Thusthe
greater therelative durationof one tone or beat of a group, the
greater thetendency forit to complete thegroup; whilethe greater
the relativeintensity ofa beat, the greaterthe tendencyfor it to
begin the rhythmic group.In other words, durationaldifferences
tend to result in end-accented rhythms,and intensitydifferences
tend toresult inbeginning-accented rhythms.
This analysisexplains whytrochaic rhythmsdo not easily arise
in triple meter. Forthe two-unittrochaic rhythm -V! can arise
in a three-beat meteronly if one of the rhythmic units is two beats
long J J. And the difficulty with this pattern liesin the fact that
such temporally
differentiated groups
tend to
become end-accented.
The only way in which sucha normallyend-accented pattern can
become trochaicbeginning accented!is to place a strong stress
upon theaccent sothat intensitydifferences outweigh,as it were,
durational ones.
This is precisely whattakes placein Example18.
Without theforte-piano onthe first beat of measures 1
and 8, the
rhythmic grouping
would have
been amphibrach
Q J,I l J } . But
the stresson the first note binds the C-sharp tothe A, and the
rhythmthought
is and
of played
as atrochee AQ|A J} - If is
;_I I
important to emphasize thatthe performers conception of the
rhythm literally influences hisplacement ofthe beats.That is, in
interpreting thispattern astrochaic, heactually places
the C-sharp
nearer tothe A than hewould had the groupbeen considered to
be iambic.
A similar situation prevailsin the case ofdactylic groupingsin
duple meter.In orderfor adactylic rhythmto existin a duple meter,
one of the two metric pulsesmust be divided to form the third
impulse necessary in the dactyl group: -VV. Thus the pattern
accented beatin amphibrach rhythms see Example 20!. For
though aclear temporaldifferentiation mayleave nodoubt about
there beingan anacrusis to the group, theaccent mustbe heavily
stressed sothat the afterbeat doesnot becomegrouped with the
anacrusis, that
is, sothat theamphibrach organization
~/"~.,,~"-/,
does not
become an
anapest one
9292_,-jv,v'jV- .
These observationsas to the modesof mentalgrouping arenot,
however, absolute laws. Theiroperation isconditioned and
modified
by the organization of
the otherelements ofthe musicalstructure-
melody, harmony,instrumentation, andthe like. This is simply
illustrated byan analysisof Example22. In the themeof the first
movement ofMozarts PianoSonata inA Major the secondhalf of
each ofthe first two measuresa! is clearly trochaic,despite the
normally end-accented iambic! durational differentiation. This
grouping isthe resultboth of the absenceof any prior anacrustic
organization andof the disjunct motionbetween measures; that is,
the skipfrom Eto Btends tomake themotives discrete,isolating the
rhythmic groups.
Andante grozioso
. 0-
_'I JXLQH- . _b.
Yi si rt.: iiir _ HiI
I..11'!Qi1QlTl.l1.ll will 13.11 11' FC-Q- _-
lt.; Qi;-.ana
.ii--_11Q1..1._l viii
Pi ivm VA yr
EXAMPLE 22
The relationshipbetween melodic and rhythmicorganization can

easily beseen ifthe second measure ischanged insuch away that
the melodicmotion between measures isconjunct, asis the case in
part b of Example22. Nowthe finaleighth noteof thefirst measure
is clearly heard asan upbeatto the second measure, making the
rhythm iambicacross thebar.
But even without thesemajor alterations,the groupingcan be
changed ifan upbeat,an eighthnote E,is placedbefore theC-sharp
in the Brst measureof thetheme. Ifthe readersings thetheme with
such anupbeat, hewill find that the last eighthnote of the Hrst
measure now tends to seem anacrustic.In short, earlier rhythmic
groupings influencelater ones;or, to put it in another way, an
established rhythmicprocess tendsto perpetuateitself. Equally
important isthe factthat futureorganization also inHuences group-
ing. Thusthe performerwill play the Hrstmeasures of the Mozart
theme in such away that its trochaicpattern isclear becausehe
knows whatthe organization of the two-measure group is.
It is, then, thetotal dispositionof all the musicalmaterials that
determines what the rhythmicgrouping willbe. Thisis anotherway
of sayingthat theentire musical pattern willtend tobe perceivedin
the simplestand mostsatisfactory terms. For this reason rhythmic
organization isnot merelya matter of duration and accentbut a
matter of these elements in relation to all other aspectsof the
pattern organization.However, whilethe mental organization of
the musicalstimuli will be as good as possible underthe given
circumstances,willit not necessarily be as satisfactory
as thelistener
might wish. Often therhythmic organizationis discontinuous,in-
complete, orambiguous.
HIGHER LEVELS OF RHYTHMIC ORGANIZATION

In Example 23 an important rhythmic change takesplace in
measure 3,even thoughthe quarter-note
motion continues
as before.
The alteredgrouping isbased uponthe changein melodicprocess,
a more active harmonicbass line,and the altered phrasing.The
second beatof themeasure isno longergrouped withthe first beat
as partof an amphibrach butbecomes partof the anacrusis tothe
Allegro Molto
.I - v!
sk; WNW; 92JJ*92a -I

EXAMPLE 23
next measure.
This reversal
of the rhytlun isparticularly striking
because thenew group enters beforethe old one hashad a chance
to completeitself. It is so accented forconsciousness that
one is
tempted to analyze the
tones Dand Cas atrochaic subgroup of the
larger iambwhich ends on theA ratherthan asan anapest. What
is crucialhere both
for theperformer andthe criticis that,though
we baseour interpretation of what the rhythmshould beupon
the availableinformation supplied
by the score, theinterpretation
itself changesdepending onwhere weplace thebeats. Inthis in-
stance, forexample, the G in measures or1 2 will actuallybe closer
to the B which precedes itthan theD in measure 3will be to the
F -sharpwhich precedes it.
Although thereversal ofthe rhythmic process isundoubtedly a
disturbance inthe continuity,from anotherpoint ofview it is appar-
ent that this very reversal weldsthe final six beats counting the
rest! of the phraseinto a single group,as opposed to the first six
beats whichfonn two clearly definedpatterns. Thisunity arises
partly because the finalgroup involvesno repetitionof similarparts
and partlybecause, in a sense, the toneD servesas apivot belong-
ing both to the preceding amphibrach and to the ensuinganapest.
Its anacrustic function is not immediatelyapparent, thoughthe
rhythmic displacementwhich the performers interpretationwill
force uponthe temporalrelationships willbe sensed, and thetotal
group will, in the end, appearas constitutingthe upbeat to the
final A.
It is this creationof a larger rhythmicunit that gives thetotal
phrase itsover-all rhythmic form. Forjust asa seriesof beatswhich
are equalboth in accent andduration will not giverise to an im-
pression ofrhythm exceptin so far as the mind imposes itsown
arbitrary differentiation
upon thestimuli! so,too, thesmaller rhyth-
mic groups will not give rise to larger patterns unlessdifferentiation
of accentor durationis present.Thus in this examplethe group-
ing might be symbolizedas A-A-B or, in terms of duration, as
3-3-6. In rhythmic terms this isnothing butan anapestgrouping.
The function of the pivot tone in joining two separablegroups
together iseven clearer
in Example16, wherethe repeatedE in the
third measure can be interpreted both as a sort of afterbeat in the
iambic rhythm and aspart of an anapestfoot forming the upbeat to
the final C. Here again the construction of a differentiated Enal
group gives rise to an anapestphrase rhythm. This pattern of con-
struction canalso beseen inExample 17.
The constructionof the Mozart exampleis quite different see
Example 24!.There eachtwo-measure group exhibits arhythm of
its own, but the whole consistsof a series of such rhythms rather
than a more compactover-all grouping.The rhythm of the Hrst
two-measure phases is notabsolutely certainbut thedynamics would
lead oneto feel that theyexhibit a trochaic grouping. The second
group of four measures is quite clearly iambic.If this analysis is
correct, thenthere isa mild rhythmic reversalbeginning withthe
fifth measure. It is mild because the nature of the reversal is not
apparent untilit is practically over-until we arriveat measure6-
*- `. P -' `/ F 92./ "" sq
, sg - *T ._
92|7.i
1 *ii
1111
, *i if5 ,'
_D3 _ q`,-!
r", "I 9,
EXAMPLE 24
and because the openingphrase isnot emphaticallytrochaic. The

regularity of the motivic construction andthe lack of any strong
rhythmic breakcontribute muchto the gracefulness ofthe total
period. Perhaps the mostunusual andaffective aspect of the whole
example isthe openingrhythm, whosepiquant character is largely
a product of the fact that a trochaicrhythm is, as we have seen,
forced upona naturallyiambic temporalrelationship.
How far one cango in discussing form in termsof rhythm is an
open question.It is partly a matter of definition. If, for example,
one adoptsa motortheory of rhythm, thenit seemsobvious that
one canseldom includeas a rhythmic groupanything largerthan
the phraseand sometimes not that.It is also partlya matterof the
temporal limitsof the psychological response to larger groupings.
Certainly Sachss argument against the conceptof form as atype of
rhythm, on the groundsthat most formal plansdo not consist of
units that are multiplesof two, is no argument atall. For, as we
have seen, it is just becausenot all motives and phrases are of equal
duration andaccent thatwe canspeak ofthe rhythmof phrasesor
larger unitsat all. Nor shouldany discussion of form in rhythmic
terms be taken to exclude otherviewpoints.
Our conceptionof rhythmic process results not only from the
immediate organizationof melody,harmony, dynamics, texture, and
so forthbut alsofrom their past organization within the particular
piece beingheard. Thepassage fromTristan alreadyconsidered
pp. 97ff.! furnishesan interestingillustration ofthe influenceof
prior melodicorganization uponrhythmic processes. The melodic
reversal discussed earlier is accompanied bya striking rhythmic
reversal, whichwe cannow examinein more detail seeExample
25!. Eachof theopening measures establishesclear
a iambicrhythm
with trochaic subgroups. Thisorganization issupported by the
phrasing inthe clarinets,the hannonicmotion, andby the rhythm
of the text itself. Notice that the main rhythmic accentalways
_:_I 1_ _ i U
YC192l ____ ` Y
AQ; l'Ql 1 -" III?!1 1.-11
Y .1 il
_ -if- 92J__ i sl ,L ,f 'JA
Ty; -J.92/ "L
EXAMPLE 25
occurs onthe topnote ofthe ascending melodic line,after orbefore

a skipof a fourth. Whenthis top tone appears, even thoughout of
its expectedorder, itis givenan importantaccent partlybecause of
its kinshipwith earlieraccented tonessee Example 26, measure5!.
Since itis theHrst toneof a group, it becomes the accented portion
of a trochaic group;and this change froman iambicrhythm to a
trochaic oneconstitutes arhythmic reversalwhich in conjunction
with the melodic changes is a powerful affectiveforce.
reversal!
Q
.I Q
_'Q'|. _ |lll.1-I1 I 1.Y'Q"lll1ll1 _
-I |11 llT11Ll.L'-`:lTY'1l
'IY922QQZf*__1'-ID_l*'iI'l I
I JI i Q- _ `-1|'?- l.l11Il`Y*Q.l'l 1'1|i1'Y`l-I
ll LQ11 71121-I lI_ _ILl ll
Q T12 1 Xi-l
I l-- ' lf - if ` _.J %
EXAMPLE 26
The analysiswould beincomplete, however, if we failed to rec-

ognize thatthis changeof processis not confined tothe rhythmic
organization. Theplacement of
the mainaccent ofthe groupon the
second beat of the measure creates a syncopation in which the
metric organization
is alsodisturbed seeExample 27!.While the
instruments continue the original metric scheme, if anything en-
forcing it through theaccelerated rate
of sequentialprogress, the
rl Q 1
I I
Ivi-I
I191131
YL | nm- r ii.Q.|ilz.;1;
:vi ll-I ln.; I|.-1_ .
, ,-I iuvtii
U ill#
I
nvnvl
Y-llN"'
Il-dl' I;1
I;r:.T 111
|;11
9292l' :r him; 1 1; 1
ll ` '
EXAMPLE 27
voice partplaces itsprimary accenton the second beat
of the bar
and its secondary accenton the fourth beat of the bar so that, in
effect, thewhole voicepart is syncopated against
the instrumental
parts.
This crossrhythm isresolved inmeasure 7, but rhythmicstability
is not achieved untilmeasure 8.And hereagain wesee theinflu-
ence ofprior rhythmic-melodic groupings; forthe upwardskip of a
fourth maintainsits original anacrustic effect,and the series of
fourths in the seventhand eighthmeasures are without a strong
downbeat Example28!. Even in the instrumental partswhere
suv `__3___;
i? ." -_ J
EXAMPLE 28
downbeats doappear, theyare obscuredby the over-all seriesof

anacruses, which reach areal downbeatonly at the final A-Hat.
That theaccented placement of theA-flat isno accidentcan beseen
if the melodic motionof measure 8 is compared withthat of meas-
ure 2.The comparison makes itclear thatmeasure 7 is a variant of
the firstmeasure ofthe two-measure motive usedat thebeginning,
but it is soarranged that
it achievesno decisivedownbeat within the
bar. In other words,we againflnd an example ofthe weldingto-
gether ofparts whichwere formerlydivided sothat 92__:u__1
`_,-`,- be-
comes -
,~.~./V ,.
This unificationof the final two measures has
consequences in
organizing thestructure ofthe whole period of eight measures.
Notice thatthe reversalin measure5 not only changes
the rhythm
within the measure but also that between measures; i.e., that the
accented measure in the opening phrase
is the second ofthe group,
while in bars 5 and 6 the first measure ofthe groupis accented
Example 29!.As aresult ofthis thesecond phrase does noteasily
1:34,
rjxiii 7I |; int.
_ ;| '' _
._ -
LV ig ,' {1 iM i: :
r*- "'.f* 92-/,
EXAMPLE 29
split itself into segments,
for our attention iscarried, soto speak,
from the downbeat in measure 5 to the final downbeat in measure 8.
In short,the totalperiod mightbe schematized
something like
this:
measures measures measures
1and2 3and,4 5and6 7and8
1, _uv _u_ N/92Q -_,
iamb iamb trochee anapest
METRIC CONTINUATION
Meter isa productof thedivision ofa giventime spaninto parts
of equal duration but unequal accentuation. The metric group is
measured and numbered interms ofthe equalbeats fromone accent
to thenext. Thusif the beat issymbolized bythe composeras being
a quarter note and,counting theaccent whichbegins thegroup,
there are four beats before the next accent, then the meter is said
to be4/4, meaningthat thereare fourquarter notes
to a metric unit.
As wehave seen,the metricgroup doesnot determinewhat the
rhythm is to be. The samemetric unit may be the basisfor various
different rhythms.In other words, althoughthe relationshipof
unaccented to accented beat is not fixed, there must be accent and
release if there is to be meter at all. What is Hxed about metric
organization isthe numberof beats,not their disposition.
On theother hand,this doesnot meanthat rhythmand meterare
completely independent of oneanother. Achange whichalters the
position of an accentin relation to other accents will obviously
affect bothrhythm andmeter.
Since the beats which measure the meter may themselves be
divided intoequal parts,some ofwhich will be accented,
it follows
that mostcompositions present
a hierarchyof meters.For example,
the beatsof a 4/4 grouping might be divided as in Example 30.
JJ JJ
U -'T Eli?
uucgrirumtnm
EXAMPLE-30
Needless tosay, otherpermutations andcombinations are
possible
on all metric levels.
We areinclined tothink of there beingonly onemeter-the one
designated inthe timesignature. Forpurposes ofdiscussion letus
call this metric level the primary or fundamental meter. Metric
levels whichare includedwithin the primary meterwill be called
inferior metriclevels anddesignated as secondary, tertiary,
and so
forth, in order of decreasing length. Those metriclevels in which
the primaryrhythm isitself includedwill be referred toas superior
metric levelsand will be designatedas secondary, tertiary, andso
forth, in order ofincreasing metric length.
Of course,some timesignatures dospecify theorganization of
the inferiormetric levelsand othersimply whatthe organizationis
to be.Thus atime signature of 3/4 specifies that
the primarymetric
level isto bein threesand implies,but onlyimplies, thatthe second-
ary metric level is going to be a division into two. On the other
hand, asignature of9/8 specifies that boththe primaryand second-
ary metriclevels areto be divided into three. However, composers
have beenmore andmore inclinedto treat the organizationof the
secondary and tertiary inferiormetric levelswith greatfreedom. A
change from this typeof metricorganization 2jj J] D to this one
i is acommon occurrence
in music
of the
past two
hundred yearsand ismade withoutany changeof time signature.
Even lessseldom does the composer make anystipulation asto the
organization ofthe superiormetric levels.One examplethat comes
to mind occurs in the Scherzoof BeethovensNinth Symphony
measures 160-260!.
This somewhat casual treatmentof changeson theinferior metric
levels isa resultof thepeculiar predominancegiven tothe primary
metric levelin the style ofthe pasttwo hundredyears. Thishas not
always been the case.During thelater MiddleAges andthe Renais-
sance therewas exceptin dance music or music influencedby
dance rhythms! no all-embracing, predominant metric beat. In
the polyphonic music of the time each voicetended to have its own
metric organization and the relationship of the severalmetric levels
within each voice, both to each other and to the metric structure of
the othervoices andthe whole,was animportant facetof the style
see Example99, p. 244!. The importance ofthe organizationof
the inferior metrical levelsis shown by the fact that they were
specified indetail bythe timesignatures of
the period.In this music
there is no basic, over-all meter. Each metric line and metric level
proceeds, soto speak,on itsown onequal termswith otheraspects
of the metrical organization;
that is, the severalmetrical organiza-
tions arenot generallyspeaking subsumed under anysingle superior
beat. Forthis reasonit would seem thatattempts torespond tothis
music in a motor way aremisplaced, formotor behaviordepends
upon andrequires asingle basicbeat towhich allothers are referred
and under which they are subsumed.
The rise of tonal harmony, necessitating a coincidencein the
vertical organizationof texture,the emergence of the homophonic
style, andthe increasingimportance ofdance stylemusic with its
emphasis on motor patterns, all madefor the predominance of what
we havecalled theprimary metriclevel. Duringthe seventeenth,
eighteenth, andnineteenth centuries the primary metric level be-
came thealmost exclusive focus of metrical attention.Changes on
other metriclevels, because they couldbe andwere referredto the
regularity ofthe primarylevel, wentmore orless unnoticed, becom-
ing relatively unimportant inthe organizationand articulationof
the rhythmicprocess. In short, metriccontinuation became identified
almost exclusivelywith the primary metriclevel.
Continuation withinthe style under consideration involves not
only the persistence of the motionof the meter but also its unity
through dominance. Because of this, disturbancein metriccontinua-
tion may take three forms: ! it may involve an over-all changein
meter; ! it may take theform of misplacement in
one partof the
meter sothat althoughthe numberof beatsis not disturbed their
placement is; in other words, there may be syncopation; and ! it
may involve the oppositionof simultaneousmeters.
An excellentexample ofa temporarymetric change,the hemiola
rhythm, whichis commonin thebaroque style,
is tobe foundin the
final movement of Handels Concerto Grosso No. 4. In order to
understand the function and effect of the metric disturbance in
measures 97
and 98,it is necessary to
note thatthere issome doubt
as tohow thepreceding measures 85-96! will continue. Thedoubt
arises because the sequence
that recurshere previouslyled to sev-
eral differentconsequents measures 5-8, 23-27,and 27-30! and
the listener is uncertain about the outcome in this case. From meas-
ure 90 on the listener begins
to expecta strongcadential progres-
sion. This expectation isintensified and colored by the doubts which
arise asa resultof thecontinued repetition
of the short motiveand
the essentially static harmonic structure of these measures.
In measure 96 the situation becomes clear and certain, and the
listener eagerlyawaits thecadence whichhe now knows will be
in the tonic, A minor. But instead of giving us a regular rhythmic
structure aswell as the expectedharmonic progressionsee Ex-
ample Sl!, Handel heightensour now definitive expectations by
delaying andprolonging thecadence and by disturbingthe meter,
thus intensifyingthe motionfrom tensionto rest.
I JL" If-
f`
: =.._.! iz
f.Q`_"`
E?" E:
' I
I' -
4
is
1'
EXAMPLE 31
In thesemeasures we have aclear metricgrouping intwos. But

the feelingfor thebasic triplemeter isnot lost.It continuesas part
of our mental set and our motor response so that the new metric
group, conflictingwith the old, intensifiesthe drive toward resolu-
tion and unification. The resolution is inevitable, in the sense that
any conflictof meters,barring furtherchanges, must
finally reach
agreement ona superiormetric level. That is, groups of threes
against groupsof twosmust eventuallymeet-have coincidingdown-
beats everysixth beat-asthey dohere.
The disturbanceof the metric organizationnot only acts asan
intensification it also actsto recondition the metric scheme present
throughout the movement. It makes the meter seemfresh and new
when it once againmoves with its usual regularity. This process of
reconditioning isparticularly commonwhere triple meter is used
because, sincetriple meter does not generally admit of secondary
120 Emotionand Meaningin Music
causehe wishesto enforceand guaranteethe syncopationof the
tonesB and C-sharpin the secondmeasure.In other words, the
meter is not irregular,but the metersof the accompaniment
and
melodydo not coincide;the melodyis syncopatedagainstthe ac-
companiment.
EXAMPLE 33
With this clueto Stravinsky'smodeof working,the metricscheme

of the wholepassage is seenas a syncopation with minor irregular-
ities rather than as a series of irregular meters.This is clear
if the passageis rewrittenas in Example34. Noticethat there are
two degreesof syncopation. TheBrstoccursin thethird bar.Herethe
syncopationis at the half bar,i.e.,the primaryaccentin the melody
EXAMPLE 34
coincideswith the secondary accentof the accompaniment. The sec-

ond syncopationbeginsat the spotmarked y!. Herethe mainbeat
of the melodyfails to coincidewith either accentin the accompani-
ment.The syncopationof the uppervoiceis resolvedat x! through
the prolongationof the endof the melodicfragmentin the previous
measure.
The pointsof metricdisturbance occur not at the seemingpoints
of disjuncture,i.e., wherethey appearin notation,but at the begin-
ning ofeach ofthe syncopations. In other words, themelody enters
earlier thanexpected atx! and later thanexpected aty! and z!.
The effectof syncopationin this passage issomewhat problem-
atical. Cn the one hand, it seems clear that within the context of this
particular workand style,syncopation does
not per se representa
deviation nor, as such, is it normative. Rather deviation and in-
tensification areaccomplished through changes inthe placementof
the syncopationrelative tothe basicbeat. Onthe otherhand, the
effect ofsyncopation isthere aspart of the basicprimary beat.The
desire ofthe listenerfor simplicityand unityof beatgives thispas-
sage itsspecial feeling
of urgency-of striving forthe pointof metric
coincidence, which is reached at No. 9.
The questionultimately isa matterof processcontinuation. Syn-
copation becomes an affectiveforce, asdistinguished froma force
toward characterization, when it creates a disturbance in an on-
going process.
When thesyncopation isitself theprocess, then
qua
syncopation nolonger functionsas a powerful affectivemodus
operandi.
The samething is true of polymeters. Inthe secondmovement
of the Ravel StringQuartet Example35!, for instance, apoly-
metric structurewithin the bar is present fromthe outset,and its
normative natureis shownby the fact that it characterizesall of
Assez vii
il 1_ _-I Q]
______El E-I
_rx nr _
EXAMPLE 35
the main melodic ideas-the stable musical formations. From a

rhythmic point of view, tension is produced largelythrough the
weakening of the rhythmic pattern, as at No. 17 in the Durand
miniature score,
by creatinggaps inthe temporalcontinuum, asis
Reprint
permission by
granted
Durand
et Cie,
Paris, France,
copyright
owners; Elkan-VogelCo., Inc.,Phi adelphia,Pa., agent.
the casejust beforeNo. 15; or, lessdecisively, byincreasing t_he
rhythmic complications somewhat, as before No.17.
In order for a polymetric structure to be felt as such, eachpart
of the structure mustbe preciselyand definitelyarticulated. Ifone
part of the structureis souniform inits rhythmas topresent onlya
vaguely articulatedmeter, thenit will probably beperceived asa
ground to a single line. This is what occurs aftermeasure 70in
Debussys Nocturne, F tes Example 36!.That is,the tendency to
hear arelatively uniformpart ofa textureas beingas uniformas pos-
sible sothat otherfigures willappear moreclearly shapedsee pp.
186 f.!weakens our feeling forthe meterof thestring accompaniment
in thissection almostto vanishingpoint. Thuswhat wehear isa Hgure
in the flutes, oboes,and clarinetsagainst ageneral background of
strings, horns,and lowerWoodwinds.
l5'Fi l'|i ' __ |

as- 1 5Y 2 _- - s=!a
one
1 '={=EEE_1='-{==5f:lF=|-=
"='i
:::i:f - gT 'i __ _ l-:
am* _f }v _
EXAMPLE 36
However, wherethe introductionof polyrhythm interrupts the

rhythmic process,
it mayplay avital role in articulatingthe musical
form and, consequently, in
shaping theemotional response to the
music. This does not mean to imply that the metric structure of a
theme, suchas thatof the Ravel Quartet,plays nopart in creating
musical responses.
It doesplay sucha part,but, inso faras theproc-
ess proceedsin a regular manner,the part is that of establishing
the basicmetric processto which other laterprocesses will be re-
ferred. It also creates what, for want of a better term, we shall call
character or spirit With the formerwe are directly concerned
in this study; with the latter we are concerned only in so far as
spirit or character may effect expectation see p. 30!.
Reprint permission granted
by ]ean ]obert, Paris,France, copyrightowner;
Elkan-Vogel Co.,Inc., Philadelphia,Pa., agent.
Real polymetersare notas common in musicof theclassical pe-
riod asare otherforms ofmetric disturbance.
But whenthey are
used, theyserve avery clearpurpose-they actas deviants.The
next example is taken from the Hrst movement of Mozarts Piano
Sonata in
F Major 332!. It includesalmost the entire bridge
section between
the Hrstand second themes inthe exposition see
Example 37!.The rhythmin theleft handis consistently
in triple
|I Df nl
[1 -MJ fml-I-J' f'1'T92Ql_d"'-T
. /
r*-- --- -=--4-#EEE:=" *- =:::=.;.:: _=:::;;::..-=r:=:
'T I 2 ,_ V'
|92--I m-
ilQYYi!1'L!.Ll-_'Tl--llll _ -l _
_-Yu---1 l--_--
IQ.-| f -K--22--Q92l1--I"
1Jm?_11E!i;i|!7;;!i1 11 -_ I---1-!
l**=-*il
V
A _ *!1'lY-I--4'1"-;-'Y
-'Znl--n] 5 Q fl | |l I-
llulj---l|T IJ YH--!l-I ITT- _ _ *I ll_ _--'I
-I-QI'-_
ll..lm--L|.*_2Il_l--_-..l||L__"f'1-ll!Il$l]_-
92l'l_'- _-1!'f3!l.- "Tl--=!-1"'--l"-'--
_* J __ I I
|9292'fl-ljl!dh'_-2lA
1l_.1_-IQ", H I __ '4 -_ l
_Z-V jl_ I! Q
IQJ---"-LQ
in-111|-1_1L_ _X 12--mi--2-2
l_-1?-1:1----1
-;12192: l .A_4ll lji._Y1l_I-11-----

1l_'l 1. _ 4| ll Im_-I1-!|TL|_J-_J .
1_1-192ll11Fl_;l.|'1Y'!Y
IJ
UI! it QI ll llll1 1- 1 1'_1l N
l92l|I1_l-lll_ -_Q_ f_- -
"J--' Il _lll_lTIZ1._1_1'C]1ll-11!!!1_m92.hll1-=l- Q _:l
-i 92 .--.... _..-___
|`92||'jl
-I-" I1 -|._-K2-D-2-2-----_--I
-I |! _ -'|H"-
11 1-'l -'--I-92 :_
_IIII
J-I-Q!-Q-|_2--.-|"
ll-_--I |'l _ _"-_-'7 ll Q Q --_-r-!'|
AI I'|'v"
_
EXAMPLE 37
meter, whilethe meterof themelody inthe righthand atfirst seems

to waver between twos and threes. The measures numbered 1 and 2
and 5 and 6 seem to be made of three times two. At the very least,
the metric structure is ambiguous. Measures
3 and 4 and 7 and 8 are
apparently intriple meter.From measure 9 throughmeasure 14 the
meter of the right hand isclearly duple,while the left hand con-
tinues in threes. Notice that this specification of the polyrhythmic
structure isparalleled bya more decisive harmonic motion: both
begin a new processthat delaysthe arrival of a solid C in the
soprano, whichthe motion from the B-flat measures3 and 4!
through theA to A-Hat measures 7 and8! leadsus toexpect. The
importance ofthis delay is demonstratedby the _emphasis the G
receives when it finally does arrive.Notice, too,that the definition
of thepolymeter coincides with the definition ofthe harmonicproc-
ess andthat bothare, soto speak,resolved together in measure15.
Any discussion of thesuperior metriclevels isfaced withniuner-
ous difHculties and problems. The difficultiesare thoseof space,the
fact thatthe largermetric structures cover alarge portionof a com-
position. Theproblems can only betouched upon.The crucialone
is, asmentioned earlierp. 112!, that of the spanof attentionand,
concurrently, therelation ofmetric perceptionto motor habits. If
the response to the metrical scheme can extendbeyond thephrase,
how far can it extend andwith what sort of accuracy? Howare
sections which are dovetailedor overlappedto be interpreted met-
rically, etc.?
In view of theseproblems we shall merelypoint outthat weprob-
ably dofeel phrases partly inmetrical termsand thatlarger metrical
schemes willusually demonstrate regularity andirregularity justas
the metrical scheme ofthe basicbeat does.If we turn again to
Example 37,we can see thisplainly. If we count each baras a
quarter note,then themetric organization of thepassage including
the finalthree measures which arenot given! might beschematized
as in Example 38. There seemsto be little doubt but that measures
9-14, comingas theydo afterthe regularsuccession four-measure
of
phrases inthe first-themegroup andthe two four-measure phrases
which beginthis section,create anappreciable disturbance
on the
superior metriclevel.
Ju Imeasure
...u JJ.
.1.|.92.;4;;J.|_4.|J.JJ.1.
3- 3-I
+l+I!++l+I!+ 2 + 2 + 2_ !+ |+|+2!
mmf I-4. 5-8. 9---14 . 15-18.
EXAMPLE 38
FURTHER CONSIDERATIONS
Before closingthis section,there are several furtheraspects of
continuity processes
that needconsideration. Inthe examples thus
far examined, disturbances in continuity have, on the whole, re-
sulted fromeither a reversal ofan establishedmusical process,
a
delay in the process,or both. But continuity can also be disturbed
by anticipation.Syncopation is,
in a sense, simply
a rhythmicantici-
pation inwhich anaccent inone ofthe partsenters toosoon. This
is the case inthe examplecited earlierfrom StravinskysHzlstoire
du Soldat,where therepetition ofthe motivebegins onebeat too
soon andcontinues inthis way see pp.120 f.
The second movement of Beethovens Third Symphony Example
39! furnishesan excellentexample ofthe disturbanceof continuity
by anticipation. Here the B-flat in measure 20 disturbs the estab-
lished sequence
by enteringone beat earlier thanexpected. This
92 c--- _ -1 A _I
.:.... "l ` _"_ -------
92/
f-;..-;; 1 r= ==='F=;:==
'viii
'42 A '-r'- - EEEEE===:=FEEE.?E:=2:=,,==:;. "-ff'r :-
EXAMPLE 39
anticipation isimportant, notonly becauseof the immediate affec-

tive aestheticconsequences which it has, but alsobecause italters
the rhythmicand melodicstructure ofthe last two measures. The
rhythm groupsof the opening fourmeasures are amphibrachs. But
the displacement of theB-flat togetherwith the occurrence of
a rest
where that tone should have been breaks the motive in two so that
the detachedB-flat isno longerfelt as the downbeatof an amphi-
brach rhythm. As a consequence therhythm of measures 21and 22
is reversed,and thesemeasures are perceived asbeing anacrucial
to the final G in measure23. Furthermore,the displacementof the
B-flat enablesthat partof themotive whichwas formerlythe begin-
ing to come atthe end,that is,the toneC, whichis a continuation
of the over-all motion from A-flat to B-flat, has a new melodic-
rhythmic functionwithin themotivic structure.
Although continuity and its disturbance have been treated
largely in terms of melodic and rhythmic processes,it is obvious
that the law of good continuationis relevant and applicableto other
musical processesas well-e.g., harmony, instrumentation, texture,
form, andso forth.Any aspectof the musical progress
governed by
probability relationships, whether these relationships are products
of learning or the result of relationships createdwithin the context
of the particular work, establishes preferredmodes of continuation.
A detailed treatment of all these aspects of the musical materials
is beyondthe scope of thisstudy. Somehave already
been described
in the discussions melodic
of andrhythmic continuation.
Others will
be examinedwhen wedeal with further aspectsof the perceptual
process. Andit is hoped thatthe readerwill, on the basisof the
general viewpointadopted inthis chaptertogether withhis own
musical experience,be ableto elaborateupon andsystematize the
observations made here.
These various
aspects ofprocess continuationare not,of course,
compartrnentalized and
separated from one anotherin the mind of
either thecomposer or
the listener.As a rule melody,rhythm, har-
mony, andso forthare perceivedtogether asa singleunitary proc-
ess. Thus in the Chopin Prelude already discussed pp. 93-97!
melodic, harmonic, and rhythmic processes werefound to be con-
gruent: theyall actedtogether increating anddisturbing continuity.
Though thisis generallythe case,it is not alwaysso. Sometimes
one aspectof the musical organization will continuean established
process whileother aspects of the organization which were at first
congruent withthat processbecome changed in importantways. We
saw this to be the case in our examination of the Liebestod where,
it will be recalled,the total process established
by the voice and
instruments together is continued by the instruments but is broken
in the vocal part.
In the first movement
of BeethovensString Quartetin B-Hat,Op.
130, sucha bifurcationof processcontinuation takes place at the
beginning ofthe recapitulation.
It is particularly interesting
because
of its effect upon the form of the whole movement.
All that need concern us aboutthe expositionsection isthe fact
that thereis a general congruenceof harmonicand otherprocesses
and that from a tonal point of view there is a motion from B-flat to
G-Hat. Anenharmonic change at the beginning ofthe development
section measure97! establishesthe key of D major. At measure
104 anew processis established, which is regular andvery well
organized from the point of view of Bguration. The harmonic mo-
tion of this processconsists ina regular progression through
the
circle of fifths, from D to G to C minor and so forth. Because this is
a time-honored norm of harmonic progressionin the classical style,
it establishes a specially powerful tendency toward continuation.
The otherprocesses established
continue alongwith the harmonic
movement until the tonality of B-Hat major the tonic! is reached,
At this point the bifurcation occurs.For the melodic, rhythmic,
textural, andother processes begin to recapitulate thematerials of
the expositionsection while the harmonic motion continueson
through thecycle offifths, fromB-flat toE-Hat, skirting
A-Hat, tgD-
Hat. This is schematizedin Example 40. In other words,while
thematically therecapitulation begins when the earlier B-flat in
Exposition Development
Recapitulation S:;1:u,es
Melodic V measures I-93!
Tlmeasures
I ij' 94-132i
measures 132-312!
214`34!
-!
processes
I 1Group
I-szrucfoup
n Group maz-59!
92 I Group
II i "
@s ml- E f` rA
Ea Fr
Harmonic , p g H, L4
processes Exposition Development Recapitulation
MGGSUISS I-93! measures 94~l7! and Coda
measures I74-
234!
EXAMPLE 40
measure 182!
is reached,
harmonically or
tonally itis notsecurely
established until
the second
group ispresented inthe tonicat meas-
ure 174.The thematic
recapitulation unsteady,
is not
only tonally,
but alsomelodically, because
the orderof theparts isconstantly be-
ing shiftedaround. Although
there isa kind of recapitulationof the
Hrst group,
the basicreversal, the
change from
an ongoing,
develop-
ing process
to arelatively stable
one, does
not take
place, as
is usually
the case,at thepoint wherethe Hrstgroup isrepeated butrather at
the pointwhere thesecond group is heardin the tonic.
IV

Completion and Closure
General Considerations
To assertthat incompletenessgives riseto expectationsof com-
pleteness istantamount totautology. For things seemto be in-
complete onlybecause we entertain definitefeelings, latentexpec-
tations, asto what constitutes completeness in a given stimulus
situation.
Our opinionor feelingas tothe completeness of a given stimulus
is a product of the natural modes ofmental organization. These
function bothwithin theframework ofwhat isgiven inthe styleand
within the sound terms established inthe particularwork. In other
words, themind, governedby the law of Pragnanz, iscontinually
striving for completeness, stability,
and rest.But what represents
completeness will vary fromstyle tostyle andfrom pieceto piece.
For instance,to the listener practicedonly in the music of the
eighteenth andnineteenth centuries, the cadentialformula which
closes and completes manypieces written during the Renaissance
will seemto be a semicadence.
It will be felt to be incomplete, to
lack finality. However, toa listener who understandsthe Renaissance
style, the same cadencewill seem to be a Hnal, satisfactory con-
clusion. Similarly the practiced listener will not feel a senseof real
completeness and conclusion after
only theminuet partof a minuet
movement ofa classical symphony has been heard, though thesame
series ofsound termswill appear as finalwhen theyare repeated
later. He expects the
trio and minuetto repetitionto follow, partly,
128
Completion andClosure 129
because ofa desirefor repetitionand, partly,and thisis important,
because heknows, thoughperhaps onlyunconsciously, what con-
stitutes completeness
for this kind of movement inthis particular
style.
Our senseof completeness or incompleteness is alsoa productof
those patterns
or soundterms whichbecome established as moreor
less Hxed,given partsof a particular work.That is, once a sound
term hasbeen establishedas acoherent, though not necessarilyas a
complete orclosed unit,then partof the series taken by itself will,
generally speaking,
seem tobe incomplete,particularly if the frag-
ment occursin the earlier partsof the total work.Thus repetitions
of thebeginning ofa well-shaped theme already heard severaltimes
will arouseexpectations that
the themewill be completed asit has
been in the past.
Even this seemingly plausible statement issubject toimportant
qualiiications. Particularly
in the later stagesof a work, part of a
sound term or an abstraction of its essential motion or its motives
may cometo standfor andrepresent the whole soundterm. In such
a case,the repetitionof a fragment ofa largerpart maynot be felt
to constituteincompleteness but, on thecontrary, betaken asa sign
of closure,making thewhole workseem complete and stable.
This observation illustrates thedifliculties involvedin setting up
any accuratedistinctions between the variouslaws of perception.
For our feeling of expectation where a normal stylistic process is
broken offor wherean established
theme isonly partiallypresented
is as much a result of the desire for good continuation as it is a
result of our desirefor completionand closure.Indeed, it would
appear reasonable to considerthe law of completionas a corollary
of the law of continuation, sinceall incompletenessis, in some sense,
a lack of good continuation and since that which is complete must
have been well continued.
Completeness and
closure arepossible onlybecause themotions
presented in music are processes involving relationships between
antecedents and consequents. Completionis possible only where
there is shape andpattern. Repetitionin itself does notmake for
completeness andclosure, nordoes change in itself.For completion
is not simply cessation-silence. It involves conclusion-almost in
the syllogisticsense that
the conclusionor completionis implicit in
the premises,in the earlier phases
of the musical motion.It is for
this reason that, granted a sensitivity to the style in question, it is
not difficult to know when a sound termis incomplete,partially
complete, orfinally closed.
Furthermore, completenessand closure exhibit the same archi-
tectonic order as the music itself. That is, what is felt to be a com-
pleted process
on onelevel mayappear tobe incomplete
on ahigher
architectonic level.
This is important because
the mind of the lis-
tener is able to take certain patterns as units, to close them out, so
to speak, and take them as given.
This analysismakes itclear thatneither continuitynor complete-
ness createshape orpattern. Rather they areproducts ofshape and
pattern. Theproblem ofwhat makesa shapeor patternis an ex-
tremely intricateand diflicult subject, whichcan be investigated
only superficiallyin this study seechap. v!.
Two types of incompleteness can be distinguished: ! those
which arisein the course ofthe pattern because something was
left out or skippedover; and! those in which the figure,though
complete sofar as it goes, simply is not felt to have reached a
satisfactory conclusion,
is not finished. Thefirst typeof incomplete-
ness maybe saidto be a productof a structural gap, the second
type, aproduct ofa delayin the need anddesire forclosure.
STRUCTURAL CAPS
A structural gap occurswhere somethingis felt to be left out.
The hiatuscreated bysuch agap neednot disturbthe processof
continuity. Infact, it might bebetter to consider such
a breakas a
disturbance incontinuity ratherthan as a structuralgap. For in-
stance, thehiatus createdby the rests in measures 45and 46 of the
Minuetto fromHaydns Symphony No. 104in D Major Example41!
can beconsidered as
a structuralgap onlyin a Pickwickian sense.
For theterm gap implies thepossibility ofsubsequent completion.
:fr .'1rt; 1 1 i1i|f ;3"_ ' ll g
7.;7=E========='f'=iE:=5 ==E.? ==EEEEEE
'f
EXAMPLE 41
And though a temporalprocess canbe brokenoff and then con-
tinued again, the subsequentcontinuation doesnot in any real
sense fillthe gapthus created.
Such abreak inprocess mayarouse
the keenest expectation, asindeed it does in this instance; but ex-
pectation is satisfied after the rests, not because something which
was missing is introduced, but because that which was interrupted
is begunagain.
It should be notedthat thoughthis is not properlyspeaking an
instance ofa structuralgap, it is an example oflack of closure. For
the breakin processis effectivebecause, when it occurs,the figure
is patently incomplete, notclosed. Hadthe hiatus occurred two
measures later,after theprogression to D major,it would have been
a weak,insipid anticlimax.
The tendencyfor structuralgaps tobe completedis mostclearly
seen in the case of melodic structure. It is to this aspect of the
musical organization
that we shall devotemost ofthe ensuingdis-
cussion.
In the music ofa culture the tonal materials givenin the style
system establish
a norm of melodic completeness. That is, in any
tonal system
there isa normalrepertory oftones whichmark oHthe
distance betweentones ofequivalence orduplication, usuallythe
octave. The total complement of such tones constitutes complete-
ness forthe system.When thepracticed orcultivated listenerbe-
comes aware that oneof thesesteps hasbeen passed over left out!
he expects,
albeit unconsciously,
that themissing tone will be forth-
coming laterin the series. Heexpects, inshort, thatthe structural
gaps createdby sucha skipwill eventuallybe filledin.
This tendencyfor structuralgaps tobe completedand filledin is
evident in the musicof many different countries.But a large number
of thesecrooked forms of Hindu Rags consistedin creating gaps in
ascent tobe filledup in descent, orvice versa.This notonly avoids
anticlirnax, but includes climax. To pass over a note immediately
creates a desire for it, and it then becomes a fit note to bear the
climax. 1 Studying theintervallic motions of extremelydiverse types
of music, Watt found that the larger the interval the larger the
skip!, the more likely that it will be followed by contrary motion
motion which will fill in the tones passedover!. He also notesthat
the studiesof Turkish music madeby Hornbostel and Abraham and
the study of Swedishfolk tunes made by Fox Strangways
reveal
similar tendencies? In their article on Muhammedan Music in
Groves Dictionary,Lachmann andStrangways note that after a
third, ret1u'nis usuallymade toone of the noteswhich havebeen
leapt over. 3 In our own culture the rule of counterpoint which
states thatafter askip themelody shouldmove bystepwise motion
in the opposite directionis simply an applicationof the law of
completion toa particularpractice!
Of course,different cultures as arule havedifferent stylesystems,
different ways of organizingmusical space. In onesystem the normal
repertory oftones maybe five,in anotherseven, and in still another
only three;and the distances between the tonescomprised inthe
system maybe equal or unequal.For this reason, anintervallic
distance whichwould constitutea skip or gapin onesystem might
not beone inanother system. In a style systemin whichthe musical
space betweenidentical tones,the octave,is divided into seven
steps, askip of a third will probably be perceivedas a structural
gap. Butin a tonal system in which the octaveis dividedinto only
Eve stepsand in which one of the normal distancesis a third, such
an intervalwill probablynot beconsidered as
being astructural gap.
The preceding statements wereintentionally conditional. The cul-
tural criterionof completeness
is by no meansabsolute. We are able
to evaluatecompleteness asidefrom purelycultural facts.For if, as
is the case inmost tonalsystems, thedistances betweentones arenot
equal, themind will assume thesmaller distanceas astandard and
accordingly judgethe larger distances ashaving gapswhich re-
quire completion.In short, a series which is diilerentiated into
larger andsmaller distances
will, so to speak,have structuralgaps
built IIIIO it.
According tothe presentanalysis, therewould bea tendencyfor
such aseries ofunequally spaced steps tobecome Hlledin. This is,
in fact, what hashappened tomost unequallyspaced scales. There
is a strong tendencyfor tonal systems tobecome morecomplete.
Most scales the abstracted linearization of tonal materials! have
developed inthe directionof closure,toward theelimination ofgaps
in their structure.
The Chinese,for instance,have introducedpin tonesinto their
Completion and Closure 133
essentially pentatonic
tonal system,thus filling in the open thirds
of the pentatonic scale.
Other folk cultures andprimitive cultures
have done likewise.
It has no doubt been noted. . . that the Quechuamusician oftenfills

up one of the minor triads of his pentatonic scalewith an extra tone
which he uses in an ornamentalcapacity-usually as a passing tone
between two more importantmelodic notes .... Often a kena-player
will take a well known pure pentatonic melodyand ornamentit ad
libitum with these extra tones.
In his book, ATheory ofEvolving Tonolityj]oseph Yasser

traces
the growthand development of tonal systems, attempting
to show
that this development follows certain characteristic mathematical
patterns, buthe doesnot attemptto accountfor this process. The
present analysiswould seemto presentan explanationfor the grad-
ual increasein the tonal materialscomprising theoctave. Forthe
new tones,introduced at first for the sakeof both completion and,as
we shallsee later,expression, eventually become fixed parts of the
tonal repertory-normsof the tonal system.If this new repertory
of tonesalso exhibitsinequalities ofdistance, then new filler tones
will be required, sincethe systemwill still be felt to havestructural
gaps.
But thereis anotherpossible coursewhich thedevelopment ofa
tonal systemmay follow,and thisraises questions as to the neces-
sity forthe mathematicalevolution prescribed
by Yasser.This isthe
possibility ofequal temperament.The ]avanese,for instance,elimi-
nate structuralgaps fromtheir music,not byadding newfiller tones,
but by making alltones equally distant.
It is surely worthyof noteonce againto seethis tendencyto equidis-
tance at work in Malay music;the sametendency whichonce before,
i.e. in Further India, had alreadyturned the same basicscale into an
equidistant sequence of tones.In the latter case,however, thiswas not
done byinserting twonew tones, but by dividing upthe octaveinto seven
equal intervals of approximately 171C, each.
A third instance ofthis tendencyto equidistancemay beobserved _. .
in a number of slendro scales."
A seven-tone
tempered scale
can, according
to Tracey,also befound
in Portuguese East Africa. This inclination toward temperament
also appearsto have been felt in the Near East.
Thus thetendency toward equal temperament and thepropensity
to addnew tonesto a scale withunequal distances both seem,from
this point of view, to be products ofa moregeneral psychological
need forstructural completeness-for the eliminationof structural
gaps not only in the melodic line of the individual piece but also in
the tonal system itself.This, to borrow a term from Kunst, is the
tendency towardequidistance.
If a general tendencytoward equidistance does infact exist,and
there seem to be reasonable grounds for believingthat it does, and
if the continued development of tonalsystems, thegradual accretion
of new tones whichfill structural gaps, isa manifestationof this
tendency, then Yassers suppositionof a kind of inevitable growth
up to the limits of the physio-psychological abilityof the human
ear andmind! would be opento question.For the process ofac-
cretion wouldlogically cease once theneed forcompletion was ful-
filled in equal temperament. An importantproviso must,however,
be made:Should allthe temperedtones ofa systembecome struc-
tural points norms!, thennew tones, tendency tones,
would haveto
be introduced into the system sothat meaningful relationships
would existwithin thetonal system.
A structuralgap, then,creates atendency towardfilling in. And
if this tendency is delayed, if the completionof the pattern is
blocked, affect or the objectiiication of meaning will probably
follow.
If, for instance, we
compare the
opening theme
of BachsBranden-
Allegro
1' I ,. `
flI ...-I!"
Q-1-_Ll _ _1_1I-I Y11i_T
71--
'_-l I --11ll I
EXAMPLE 42
burg Concerto No. 5 with a part of the aria Che faro senza Euri-
dicel from Glucks Orfeoed Euridioewe findthat theopening notes
_ _; -
EXAMPLE 43
of both themes createthe samestructural gaps. In the Bach theme,
however, thefilling in of thesestructural gapsbegins immediately;
and, partlyfor this reason, this
theme, thoughdefinite andforceful
in both mood and character, is not itself affective. The affective
experience ofthis movement is rathera productof the progression
of the larger parts. But the fragment from Orfeo is, Hanslicks
derision notwithstanding,notably affective. The poignancyof this
passage isdue in part to the tensions which arise, the tendencies
which areinhibited, because there isa delayin the filling in of the
structural gapscreated bythe openingmotive. Andit is only after
all the skipped toneshave beenpresented thatthese tensions sub-
side andthe melodyconcludes.
Notice thatdisturbances in process continuation also playan im-
portant partin creatingthe affectivequality of this passage.
For the
repetition ofthe C after thetriadic motivecreates abreak inboth
melodic and rhythmic processes. That is, we expect the triadic
motion of the openingmotive tocontinue. Thepowerful effectof
the high E is partly a result of the fact that it was unconsciously
expected at the beginningof thefirst completemeasure. The eighth-
note motion is also expected to continue, and for this reason the
quarter noteC becomes
a particularlyeffective appoggiatura.
SATURATION
The principleof saturationis related,on theone hand,to thelaws
of goodcontinuation andcompletion and, on theother, tothe beliefs
which thelistener entertains
as tothe natureof aestheticexperience.
Since themeaning ofany soundterm is a functionof its relation-
ships toother consequent terms whichit indicates,our normalex-
pectation isof progressivechange andgrowth. A figure which is
repeated over and overagain arouses a strongexpectation ofchange
both becausecontinuation is inhibited and because thefigure is not
allowed to reach completion. Our expectation of change and our
concomitant willingness to go along with the composer in this ap-
parently meaningless
repetition arealso productsof our beliefs in
the purposefulness
of art and theserious intentions,
the integrity,of
the composer.We believethat he will bring about a change.
A particularly clear exampleof the arousal of expectation through
saturation isto be found in the Hrstmovement measures16-26! of
Beethovens SymphonyNo. 6, where the same motive is repeated
ten timeswith only minimal dynamicand orchestralchanges see
Example 44; also seethe developmentsection ofthis Symphony and
that of Beethovens Symphony No. 8!. The useof saturationis com-
mon particularlyin theslow introductions
and development sections
of eighteenth-and nineteenth-century works wherethere is expec-
tation of a returnto a theme alreadyheard. Forat suchplaces the
fragment need only berepeated oneor twotimes inorder toachieve
the desired effect.
l' .1;0`__ Wf '' 'W

YSYHZYTTY11
|a;"r-"- -.L|_1l ==|v1'v'i: glrgirglrgl
lll,92.lTl17A1Tl1Yl1fl_fAnrgnrglrgl _
s_ 0- ' g
EXAMPLE 44
Koffka observesthat there is an intimate relationshipbetween

saturation andemotion: Saturationis emotionalbehavior. Itsanaly-
sis revealedan interplay of forces leading to increasing tension
within the Ego system .... 11 Althoughit was found that,gener-
ally speaking,these tensions
could berelieved byvarying thetasks
being performed, it would seem thatin the case ofmusical percep-
tion tensions may arise even though variation is present. That is,
even with variation in harmony or range or dynamics avery marked
figure may,in the proper context,give rise to saturationif it is
repeated with great enoughinsistence.
It is necessary onceagain to emphasize theimportance of context
upon theseprocesses. repeated
A patternat theend ofa workneed
not give rise to saturation, sinceat this point the listener under-
stands, orthinks heunderstands, the significance ofthe repetition:
that is, because thisis the end of the piece, lack of forward motion,
a composedfermata, isexpected anddesirable. Thusthe law of
saturation isconditional: Ina situationwhere repetitionis not nor-
mal and understandable, the longer a pattern or process persists,
the strongerthe expectation of change.
In the last analysiseverything depends upon ourability to com-
prehend the signiHcance ofthe repetition. If it is understandable,
then therewill be no saturation.
For instance,a repeatedpattern or
figure may,because ofits positionin the work andits relationship
to other patterns, be understood asan ostinato. If it is to be heard
in this manner, the composer mustso articulate the texture that the
listener graspsthe significanceof the figure as an ostinato,He
must asa rule make evidentthe fact that the repeated pattern is a
ground, e.g.,it must be morehomogeneous than the otherpatterns
and will, as a general rule,surround theother figurestemporally,
begin beforethey do. An ostinatomust alsohave aless palpable,
well-figured shape than theother figuresof the work so that it is
clear that it is not the chief pattern.In other words, it must not
develop tendencies of its own, asdid the figure inExample 44. This
relative weakness of figure is what distinguishes an ostinato, such
as theone employed by Debussyin his Piano Prelude Des Passur
la Neige, from a groundbass. Fora ground bass develops tend-
encies ofits own; it has shape, andit is a shapingand controlling
Trista etlent
'' ' ' _1 1
_. -1- -s-s ,I "':.'i*."

'LSTZSSSZ3===S===L"==Z-==='=.====Z===E?E
-_ PH - ._ ._ .. _
EXAMPLE 45
force in the articulationof the musical progressand growth.That

is, theground bass changes its meaning asit formsnew andchang-
ing relationshipswith theother partsof the texture. Itis not simply
a given static entity.
It should,however, beobserved thatan ostinato-like
pattern may
at timesseem ambiguous, though it need neverdo so.The listener
understands that the function of the pattern is to establish a con-
tinuous and repeated ground against which other more clearly
articulated figures
are tobe projected.If no other moresubstantial
Bgures areforthcoming, thenthe listenermay beginto entertain
doubts as to whether the motive in question really is an ostinato.
Then asthe repeatedfigure becomes
the centerof attention,the
listener will begin to expect changesto take place. And if the figure
remains stable or is subjected to only minor variations, saturation
Reprint permission grantedb Durand et Cie, Paris, France,copyright
owners; Elkan-Vogel
Co., Inc.,
Philadelphia,
agent.
Pa.,
may setin. This would seemto be what takesplace inthe opening
section, Prludea la nuit, of Ravels RapsodieEspagnole, where
the ostinatoBgure isdominant evenduring the brief appearances
of the theme.
MELODIC COMPLETENESS AND CLOSURE

One ofthe mostpowerful andpersuasive forces conditioning and
controlling thesense ofcompleteness which a melodicpattern gives
is the tonal organizationor scaleof the culture. It establishes a
system of order, a system of expectancies, asystem of tonal-
ity .... 1 Since theimportance andfunctioning of tonality has
been discussed in chapterii and will be the subjectof further dis-
cussion inchapter vi, its overwhelmingsignificance need only be
recalled at this point: Tonality is probably the most important
single facetof style, the sinequa non of even the mostprimitive
musical organization.
It is important alsoto recallthat the norms ofstyle, oftonality,
important asthey are,can bealtered throughtraining andknowl-
edge. For the individual musical work also establishesnorms which
condition ourfeelings andopinions asto completeness and closure,
and thesenorms ofthe individualwork maybe unusualwithin the
style. Acomposer may, for instance,by dint of repetitionwithin a
work, establish as Hnal, as representingclosure, a melodic pattern
which might not normally seem Hnishedor final. The listeners sense
of closure and finality can, as Farnsworth has shown, be altered
through training. W. V. Bingham 15 has shownthat opinionsas to
the finality of a series oftones canbe influencedby information
stipulating thenumber of tones tobe heard in the series. If,for
instance, listeners were told that the fifth tone of the series was to
be the last, it appeared topossess finality
a which it did not as a
rule have otherwise. Thisfurnishes experimental evidence, ifany
was needed, for the statement that what we know, either because
we havebeen told or have learned through practice and experience,
iniiuences our
judgment ofwhat weperceive andhence ourfeeling
of completenessand ourexpectation based
upon thatfeeling.
Our sense of completeness is directly related to our ability to
understand themeaning ofa particularpattern. A stimulus series
which developsno process,awakens notendencies, will,if it be-
comes thefocus of attention, alwaysappear to be incomplete.A
sustained tone,
for instance,will give rise toexpectations of
change,
unless, of course, it is not the main object of attention or can be
understood moreeasily inanother way,e.g., asa pedalpoint.
Moreover, thedirection in which the tone maybe expectedto
move is also related to our sense ofcompleteness. The direction of
expectation, whether to a tone belowthe Hrsttone of a melodyor
above it, depends largelyon the absolute pitchof the latter. In
general, thelower this is the less will the melodybe expectedto
descend onthe next following tone; while correspondinglythe
higher it is the less willthe melodybe expectedto riseon thenext
tone. 1Actually thesituation issomewhat more complicated than
this. Forour expectations depend uponour awareness of potential
range aswell. That is, a tone whichwould be high for one voice
or instrument,and would consequently beexpected todescend,
might be low for another voiceor instrument and would conse-
quently beexpected torise.
Although notdirectly relatedto the Gestalt principlesof pattern
perception, oursense ofclosure isin part a productof the general
configuration ofrelaxation andquiescence. Melodically speaking,
relaxation is associated with the decline in tension which is effected
when pitches
are lower-whena progressiondescends atits close."
Curiously enough,however, thelowering of the dynamiclevel,
which is also a normal concomitant of relaxation, does not neces-
sarily accompany closure. Thereason forthis would appear tolie
in thefact thatthe dynamiclevel isintimately associated
with those
aspects ofthe musicalmaterials whichtend to designate thechar-
acter of the piece. More particularly, the dynamic level would seem
to be linked to tempo and general range.These elementsare them-
selves interassociatedso that fast tempi, loud dynamics, and high
ranges but not necessarily ascending melodic lines! are usual to-
gether, whileslow tempi,soft dynamics,and low ranges arealso
usual together.Of course, there are pieces that get softer at the end
as well as onesthat become slower; and both of these changesplay
a part in making closure moreemphatic. But,on the whole, the
slowing down which brings a piece of music to its close is not a
slowing downin t_he physical tempo but a slowing downof the rate
of musicalprocess. Thatis, thoughthe tonesmay flyby with great
rapidity, themelodic, harmonic, and rhythmicchanges which create
a sense of tendencyare broughtto a standstill. Themusic nolonger
progresses; marks
it time;it is static.
In any particular musicalwork certainmelodic patterns because

of their palpable andcohesive shapes become established in the
mind of the listeneras given,axiomatic sound tenns. Theyset the
mode ofcontinuation, completion, and closurewhich arethe norms
of the particular work.And if part of such apattern isintroduced,
it will arouse definite
expectations as
to the manner ofcontinuation
and completion.
The followingmelody isestablished assuch anaxiomatic sound
term in the iirst movement ofHindemiths Symphony,Mathis der
Maler Example46!. This is not the placeto makean exhaustive
:2:= '- -L- &- --4:=:H

aa:"7"---:=--=======?'.= ====E
_ . El J it si
fi I' s
_ _,_,_.__,._
__ .!=_=.=s _;=!!.__
r,.:x::====.==::==::===:===::::=:=====:
""' " ' _'_' "7" " ` """'
.rl-!='.__;._r 3 e*
. '..== F F Li
EXAMPLE 46
analysis ofthis melody,particularly sinceHindemith himselfdeals

with it in somedetail. However, one pointwith which Hindemith
fails todeal inhis own
analysis must
be discussed,
for it plays an
important part
in theorganization the
of melody
and, therefore,
in
conditioning subsequent
expectations and
feelings of
completion.
This isthe factthat theover-all motion
of themelody istriadic,
moving asindicated inanalysis Aof Example46! from the initial
Copyright
1934 by
B. Schott's
Soehne, Mainz;
used by
permissionAs-
of
G, throughB andD, up to F-sharp. Thefirst motionis up to the D;
but this tone is, so to speak, leftdangling in mid-air. It does not
really moveto anythingwhen it is first heard. Ratherthe B which
precedes itmoves downward to the C in measure 48see analysis
B!. This is of special interestin the present discussion since this
effect creates
a deHnitesense ofincompleteness. We feel andexpect
that we must return to the D and move in a coherent manner from it.
Before thehigher D is pickedup in measure 50, the intervallic
and rhythmicmotion whichfirst ledto it is re-established
in a lower
range measures47-48!. The whole motion from measures48-49 is,
in a sense, nothing
but a composed upbeat to the higher D. Notice
in this connection thatthe important tones in this sequence, the
B, D and A-sharp enharmonically, B-Hat!, are alsothe vital tones
in the figure in which the D first occurs measure 43!. The retu1'n
to the high D leads toa stepwisemotion to the F-sharpand then
down throughthe Dto B.
In makinga furtherabstraction ofthe essentialmelodic motionof
this melody,it would seem thatthe G in measure49 playsa much
more importantrole thanit would appear tohave atfirst. It serves
to tie together theHrst andsecond phrases and by doing so to
provide a long delayedresolution ofthe motion from B to A in
measures 41-42. From thispoint of view, andit is certainly notthe
only one possible, theover-all schemeof the melody mightlook
something likeExample 47.
..:._;..;,_..-LL-Qiiu._ .rv-.-1;n
1
ml - g j 43-49p g gg,
EXAMPLE 47
In this analysis thehiatus createdwith the break in measure 42

involves bothincompletion ofthe motion down to the G, which
arrives only in measure49, and a halt in the upward motion begun
in the first four measures. Whatreally happens is that the signifi-
cance of the D remains to be established. This is accomplished by
its upwardmotion tothe F-sharp,which marksthe reversalof proc-
ess inthe melody as a whole. That is, it not only literally marks the
first majordescent inthe over-allline but it is also precededand
marked bya significantharmonic clarification, a fact which Hinde-
mith notesin his own analysis.
In terms both of its motive and its over-all triadic motion, this
melody providesone ofthe normativesound terms in the first move-
ment of Mathis derMaler. Its analysis furnishes
a basisfor under-
standing howsubsequent deviationsarouse expectation.
In thepassage that
follows the
melody given
above, the
first part
of thetheme ispresented in
an incomplete
form. It therefore arouses
, ' .
QQTEEE::E5==:=:.:::;'====::-5gi
- -L s i
____
,_ _, e---- ----- ---- --------- - ~---------- -~----1
I
EXAMPLE 48
our expectations
both becausewe havea standardof completeness,
we knowhow it should continue,
and because the structural
gaps
remain unfilled
until thewhole melody
is heardbeginning at
meas-
ure 63.But we do havea very strong sense
of directionand con-
tinuity once
the firsttwo fragments
have been
heard. We
expect the
rising motionto continuebecause itis incompleteand becauseit
has beenestablishedaasmodeof continuation in themelody itself.
Notice, incidentally,
the factthat themotion fromC toD toF-sharp,
which wassomewhat veiled in the original melody,
is heremade
very clear and forceful.
The arrivalof theF-sharp in
measure 60 has adecisive effect
upon
our expectations
for tworeasons. First,
it crystallizes
and fixesthe
basis ofthe ascendingprogression, making
clear thefact thatthe
sequence isrising throughthe major triad, and second, itmarks
the onset of saturation.
Up to measure 60
we expectthat theprogression will
rise, butwe
do notknow precisely
what formthe ascending
progression will
follow. Oncethe F-sharpis reached,this is made clear.And it is
this knowledge
which enables
the composer
to repeat
the motive
on
the F-sharp.For had it been repeated sooner, it would not have
appeared asso stronga delayin the line, andit is partly thisfeeling
of delay that contributesto the effect of saturation.
Actually theeffect ofsaturation iscumulative. Eventhough the
motive isat first stated ondifferent degrees
of the scale, itis felt as
repeated. Andthe powerfuleffect ofthe exactrepetition onthe F-
sharp is,in part, a productof the earlier statementsof the motive.
The effectof saturationis alsodependent uponthe fact of incom-
pleteness. Hindemith emphasizes this by makingthese laterversions
of the motive even shorter than the earlier ones. This also has the
important effect of increasingthe psychological tempo. Forour sense
of timing is guidedby the time spanbetween theentrances ofthe
motive, whichat Hrstare tenbeats apart,and laterbecome onlysix
beats apart,and, finally,only five beats apart.This hurryingplus
the repetitionof the F-sharp motivewhich heightensour expecta-
tion bothbecause itrepresents delay
a inthe process
of continuation
and because it becomesa signthat the main event,which wehave
been awaiting,is aboutto take place! build up tension,which is
climactically released at the arrival of the high A and a complete,
though somewhat varied, statementof thewhole melody.
RHYTHMIC COMPLETENESS AND CLOSURE

The difference between meter and rhythmwas discussedin chap-
ter iii. The distinctionis importantin connectionwith an examina-
tion of rhythmic completeness. For rhythm may be incomplete
though meterremains intact; and, conversely,
rhythm maybe com-
plete withoutthere beingany meter.
The completeness of a rhythm dependsupon the apprehension
of a relationship between accented andunaccented parts of a co-
hesive group. The sequence of rhythmicfeet mayvary considerably
so long as the relationships within the various feet are clear. 'I`hat
is, trocheesmay followiambs, iambsmay follow anapests, and so
forth without creating incompleteness,
though continuitywill, of
course, bedisturbed insuch asequence. The
accented or
unaccented
parts of a rhythmic grouping may be implied, imagined by the
listener, rather than stated, but not both. However, whether the
relationship isexplicit orimplied, it must existif there is to be any
sense ofrhythm. Aseries ofunaccented beats will seemincomplete
and give rise to expectations ofcompleteness, expectationsof an
accent. Conversely a seriesof accentswill appear incomplete and
give riseto expectationsof a return to an understandable
relation-
ship between accent andrelease. One might wellcontend thatthere
is no difference between a series of accents and a series of beats
which areunaccented, since
we canknow eitherone onlyin terms
of theother. Both
are, ina sense,
only pulses.
Finally, itshould be
noted that, since rhythmsform an architectonic hierarchy,
a pas-
sage whichis rhythmically incomplete ona higher architectonic
level maybe completeon alower one.However, thereverse ofthis
is not true; for if the lower level is incomplete, fails to differentiate
beats into accented andunaccented, then the higher level is de-
prived of the basisfor its organization.
The apprehension of meterdepends upon the establishment of a
regularly recurringaccented beat. The dispositionand numberof
unaccented beats in sucha metricgroup isa matterof indifference.
The temporal recurrence of the main beat is all important. Hence
a meteris incompletewhen themain beatof the group ismissing.
Since thelistener expectsan objectivestimulus to correspond to
his inner beat, metriccompletion cannot be subjective.
It shouldbe notedthat, althoughmetric incompletenessof neces-
sity involvesmetric discontinuity-forthe absenceof a beat where
one isexpected creates awareness of a break-metric discontinuity
need notinvolve incompleteness. For discontinuitymay arisebe-
cause thebeat, thougharriving atthe propertime, isno longerfelt
to be accented, as,for instance,in the bridge passagemeasures
23-25! from the secondmovement ofMozarts ]upiter Symphony
K. 551!, where there are obvious metrical disturbances but no
metric incompleteness.
One otheraspect ofthe problemmust beconsidered. Though the
perception ofmeter doesnot dependupon theexistence ofa fixed
relationship between
accented andunaccented beats
it doesdepend
upon there being a difference between the several beats within a
metric group. That is, some beatsmust be accented, othersnot. For
this reasona seriesof equally accented pulses
will no more give
rise to an impression of meter than it will give an impression of
rhythm. Sucha serieswill be ambiguous; andits significancewill lie
in the very factof ambiguity:the factthat the series indicates and
points to,leads thelistener toexpect, clarification
and normality.
Since these matters canbest bediscussed in the lightof a specific
example, letus turn to an examination ofa portion of the fifth
movement of Beethovens String Quartet inC-Sharp Minor, Op. 131.
The main theme of this movementis regular and symmetrical.
Each ofits two four-measure phrases is composedof two comple-
mentary motives see Example49!. The Hrst islabeled a,the second,
b. Three levels of rhythmic organizationcan be distinguished in
each ofthe phrases. Level 1,the rhythmwithin thebar, istrochaic.
This is true for every measureof the theme. Level 2, the level of
the halfphrase, isiambic. Aniambic rhythmcontinues on
this level
throughout theperiod. Level3, comprehending the wholephrase,
is also iambic; that is, the second two measures are accented rela-
tive tothe Iirsttwo because
of thechange ofharmony andthe more
decisive motion both within and between the several voices. The
pattern of the consequent
phrase isthe sameas that of the ante-
cedent phrase.
Presto 1. 'I-;_`_1 s. '___

lu .
-'nl I IIII'I'ITI . . ; _ |l |'II.- I im'- 92C IQ
'IL
|[._-_UIII CI I I III|vÌI-I
I I Iil Il TI I_
1-I-ll-Ii -"II
92'___I'y"II'-II I I92 |l IuI---I-IIII-I--Ill
}A
~ `` ~ : 3-2
'IRT
lb _|
f`|n--H-LII I I III Il I'I II-IIUIIIIIIIII' I--l-I-_II
III I- II I-I-Il!-I I II-=-II-
PI
' I- I | |I
I. it . r_ 94-'_"||_ `/1r_ gi Etc'
2. -92'/ _IJ ._ HI/ -.fl j| |
3. * |{ Vi Hi ,Y x? rf'
EXAMPLE 49
After this basic rhythmicstructure hasbeen wellestablished in

the mind of the listener byrepetition bothwithin the phrase and
the totalperiod, Beethovendetaches motive a from the phraseand
bases six-measure
a sequence
upon it measures 19-24, Example 50!.
This change makes therhythm on level 3 incomplete. For motive a
now hasno accentedmotive tomove to.Considered on
this rhyth-
mic level, the sequence measures 19-24! is a series of anacruses.
It is incomplete, andthis incompleteness,
creating tendencyand
expectation, givesrise to affect and aesthetic signiiication.
_ .._.
YL1lJ1'Z1'l2.1-1 III
. .____
' ~r-
,-,1
"f'i"'a"""; ' """ "
u92.iva.:n.: ii i
xuvllil nliin-nu:
|192 hint:-:nl1n1__11_~;'
an/v L11:1 I1 lil In l_1 lx I"
:_A :_ 11 li rlnilullH-i-1*-"Q 11
Y PCR. ZiZI1I1Q11IQl111QQ11f1I1C C1111
I.. c:;L._i. ETC-
2. r e"" i ng 1 5
3. ;.../ 92.... ;J
EXAMPLE 50
Since it is the relationship between motives within the phrase

which is changed inthe constructionof this sequence and not the
relationships withinthe motive,the completenessof the rhythm of
the first two levels remains intact. This accounts for the fact that,
though thereis evident incompleteness, there
is no discontinuity,
no break in the basic motion.
At measure 25 a somewhat varied version of the second half of
the motiveis separated out, and,if we include theslight variations
to which this fragmentis subjected,it becomesthe basisfor the
organization ofthe following twelve measuresmeasures 25-36,
Example 51!.Rhythmic level2, which depended forits existence
upon the differentiation between, and a juxtaposition of,an un-
stressed Hrstmeasure anda stressedsecond measure, is destroyed
by the repetition of this fragment. That is, since we are presented
with a series ofprecisely equivalent
measures, itis impossibleto
differentiate theminto accentand release;and, consequently,on
level 2 the rhythm is now felt to be incomplete. This is, of course,
precisely whattook placein measures
19-24 onrhythmic level3.
Molto poco
odogio
__ .H . hl n hl @
===- =Hs!z;z.e=;.-fis
-; - . -
%;?"%/=~?'`tc.
EXAMPLE 51
The variationin the half notefragment isa necessity

if rhythmic
suspense is
to be maintained. Forhad measure25 begunon the
accent onboth the second andthird rhythmic levels, anda new
rhythmic process
would havebegun.
This omission of the first beat of the measure creates metric in-
completeness as well as rhythmic incompleteness.
For since the
listeners sense
of the metric pulse,once it has beenestablished,
tends topersist, thesuppression of
the downbeatis felt as agap or
incompleteness. Notice that Beethovenhas beencareful to make
sure thatno displacementand reconstitutionof the metric beat takes
place. Forinstance, hadthe Hgurebeen written as a syncopation
with halfnotes beginning on thesecond and fourth beat,the listener
might eventuallyhave heard the meteras complete, though itwould
be displacedrelative to its originalposition. Asit is, the listeners
sense ofthe original metric organizationis supportedby the fact
that the unaccented portion of the meter comeswhere it formerly
did, and this helpsto emphasizethat the accented portionof the
metric group is missing.
In spite of the patent incompleteness
of the meter andrhythm
on thesecond andthird levels,some feelingof rhythmic organiza-
tion remains. For on the first level a differentiation between accent
and releasestill exists.However, theomission ofthe first quarter
note of the measurehas broughtabout a change inthe rhythm.
That is,the fragmentwhich wasformerly trochaic as in measure
20! is now iambic. This reversal creates a certain amount of in-
decision inthe feelingfor therhythmic organization.
This feelingis
subjected to further strain when the tempo changesat measure33.
In fact, our feeling for the relationship betweenthe metric organiza-
tion built up and maintained within the listener and the objective
metric organization
is all but destroyedat this point. Andour feel-
ing for the rhythmsuffers thesame fatewith the change intempo
in measure 36.
The progressive incompleteness ofthe rhythmic and metric or-
ganization from measure 25through measure36 not only creates a
desire for a return to the well-deiined structures of the theme itself
but it also makesthe listener feel that further disintegration is im-
possible. Consequentlythe return to a regular, uninterrupted series
of pulsesand the use of new material, based on the first measure of
the motive, are at first understood to constitute this expected re-
turn. Butthis understandingsoon proves to be illusory. Forthough
there iscontinuity ofpulsation anda clearrhythm shouldemerge,
there is no differentiation into stressed beats and unstressed ones.
Since thesuccessive but overlapping entrances of the motive on
the samepitch level provide no basis for a distinction between
accented andunaccented beats, these measures 8-41, Example
52! are perceived asa seriesof equal pulsations. Inspite of the
regular pulsation,
there isno meter,for thereis nodifferentiation. If
there isany dillerentiationat all, it exists on the subprirnary level
as a trochaic rhythmof quarternotes. Butit is doubtful, in view
of therapid tempo,whether thissuccession of quarter notesis per-
ceived asa rhythmicgrouping atall.
Tempol _ _ __
.__._.-__
- :|, ...!
! rw: _ _ _?_ l.92.-id
, 1 il " 4"T"" lf;iii; 4
lr'.=.f==EEE!'!!._ . ._.. ._.._ iE
,;:-:~--.~f~:-'~-:~,f@~ -.
3_ if g 1g '
EXAMPLE 52
Even thereturn in measure 41

to the general rangeused in the
beginning ofthe movement, though itcertainly whets
our expecta-
tions, does
not clarifythe situation
completely, since
overlapping still
takes place.
In fact,since itappears that
the rebuilding
of therhythm
is supposedto be gradual, itseems important
that nospecial accent
be givento theHrst beat
of thisgroup. TheD-sharp inthe firstviolin
measure 41! should probablybe underplayedrather than the
other wayaround. More
precisely: itis clearthat regularityis
definitely establishedonly after the fermatabetween measures 44
and 45.If tensionis tobe maintained, then thewhole rhythmof the
phrase should,
as it were, beheld back.This canbe achieved
by
making therhythm inmeasure 41 indecisive andambiguous. Ifthis
is thecase, then
the rhythmof level1 becomes established in
meas-
ure 42,and therhythm of level 2becomes established in measures
43 and44. However, since therhythm ofmeasure 41 was ambiguous
and unformed,measures 41and 42fail to set up a rhythm on level 2;
and, consequently,these fourmeasures areunable togive riseto a
feeling ofrhythm onlevel 3.In short,the significanceand eflicacy
of this period measures37-44! lie in the fact that while it re-
establishes theoriginal rhythmicorganization, itdoes not do so
completely. Thelistener remainsin a state of suspense untilthe
theme is presented asa total unit.
This analysisalso increasesour understandingof the over-all
rhythm of the first section of this movement measure 66! and en-
ables usto understandbetter its psychological effect
and intellec-
tual significance.
For it is clearthat theincompleteness rhythm
of in
measures 19-44creates suspense. The measuresare an anticipatory
upbeat, an anacrusis, tothe complete thematic statementwhich be-
gins at measure 45 and extends through measure 66. Seen in this
light, thesignificance and
importance ofthe twointroductory meas-
ures alsobecome clear.For theynot onlyprovide abrief buteffective
anacrusis to the whole first statement of the main idea measures
3-18!, but they also establish thebasic iambicrhythm which is
continued throughoutthe section.For even though the second
measure isa rest,it is accented-strongly marked
for consciousness
by the mind of the listener.The over-all rhythm of this section
might be schematized asfollows:
measures: 1-2 3-18 19-44 45-66.

92"' me-"T . 14
HARMONTC COMPLETENESS AND CLOSURE

Rhythmic organization
does notoperate asan isolatedindepend-
ent variable without reference to other aspects of the total musical
organization. Theformulation andoperation of the rhythmic or-
ganization areconditioned andmodified bythe melodicstructure,
the dispositionof the texture, and,in Western music at least, by
the harmonicorganization. Inthe fifth movement ofthe C-Sharp
Minor Quartet the relationshipbetween harmoniccompleteness
and rhythmiccompleteness very
is clear.For this reason itmight
be well to discussthe questionof harmonic completeness with
reference to this example.
A feelingof harmoniccompleteness arises
when themusic returns
to theharmonic basefrom whichit beganor movesto onewhich was
in someway implicit in the opening materials.In this movement
see Example 49! boththe openingperiod andits repetition,though
rhythmically complete,are harmonicallyincomplete because they
end not on the tonic but on the dominant of the submediant, the
chord builton thesixth degreeof the scale.
This harmonicincompleteness leads the listenerto expect that
this main theme will eventually closeon the tonic. In measures
10-24 theharmonic progression from VI to IV to V in E major, a
typical cadentialformula, togetherwith the incompleteness of the
rhythm activatestrong expectations of a close. Butthese aredis-
appointed. Forin measures25-28 the dominant harmonyis not
resolved to the tonic but rather is weakenedby the omission of the
all-important leadingtone and continues onlyby implication. In
measure 29this dominant feeling is not so much resolvedas simply
replaced byanother incomplete
triad whosesignificance isambig-
uous. Thesubsequent harmonicchanges inthis passageare like-
wise vagueand indeterminate.This lack of decisivenessis important
because it enables the incompleteness ofthe rhythmic and metric
organizations tomake themselves
felt. And the converseof this
is also true: the rhythmic incompletenessmakes the harmonic mo-
tion seemparticularly vapidand ambiguous.
One mightsay that,though harmonic
motion continues,
harmonic
progression is
suspended. From measure 25
through measure41 the
harmony changes,but it does not appear to have direction. This
permits therhythmic structureto be progressively weakened with-
out destroyingthe listenerssense oftonality. The listener knows
what keyhe is in. Thisis provedby the fact that,although various
preparations are made for a cadence in G-sharp minor measures
42-44!, the listener never really expects them to materialize; and,
consequently, the
motion to E major in measure45 comesas no
surprise.
The effectof this whole sectionis clearly dependent uponthe
intimate co-operation of rhythmicand harmonicorganization. The
hannonic incompleteness establishes the
forces towardcompletion,
and thetendencies thus established giveforce andurgency tothe
rhythmic delay.Because this delay takesplace atthe beginningof
the movement, Beethoven does not wish to destroyor weakenthe
tonal orientationof the listener. Indeed,
had thefeeling ofE major
been destroyed, the wholeraison dtreof the passage would have
been lost,for its purpose isprecisely todelay the cadence inthe
tonic E major. For this reason Beethoven avoided extensive chro-
maticism andthe useof very irregular chordprogressions, turning
rather torhythm tocreate thefeeling ofsuspense which he wanted.
Here, again,we seethe complementary relationship betweenhar-
mony andrhythm. For the rhythmic incompleteness necessary to
create delayand suspensewas possible only because the harmonic
motion was weak and indecisive.
THE PRINCIPLE OF RETURN

The differenttonal systems
upon whichthe musicof variouscul-
tures hasbeen built, whether theyare primarily melodic orboth
melodic and harmonic, are all basically special, though vitally im-
portant, instances of what has been called the law of return, the
law that,other thingsbeing equal,it is better toreturn toany start-
ing point whatsoever than not to return. 2 This law apparently
influences thestructure ofprimitive melodiesbuilt upon only two
or threetones aswell as the organizationof the complex musical
structures of art music. Of course, once a tonal style has been
established, has
become partof the habit responses of the cultural
group, theterm return need not be taken literally; that is, the
opening materials may indicatewhat theBnal toneof a piece isto
be withoutexplicitly presenting
it in the openingmoments.
The law of return depends forits operationupon recurrence,
a formof repetitionwhich mustbe distinguished from reiteration.
Recurrence is repetition which takes place after there has been a
departure from whatever hasbeen establishedas given in the par-
ticular piece. There can be a return to a pattern only after there has
been something
different whichwas understood
as adeparture from
the pattern.Because thereis departureand return,recurrence al-
ways involvesa delayof expectation
and subsequent fulfilment.
Reiteration, whether exact or varied, is the successiverepetition
of a given soundterm which,even if it is very extensive, is never-
theless perceivedas aunit. Reiteration does notnecessarily giverise
to expectations of further repetition. Onthe contrary,if repetition
is fairlyexact andpersistent, changerather thanfurther repetitionis
expected, i.e.,saturation setsin.
Reiteration isthe basisof what may be called theprinciple of
successive comparison. A given pattern establishesan intra-opus
norm, a base for expectation withinthe particular piece. Sub-
sequent deviations from the pattern, occurringin repetitions,give
rise to affective or aesthetic responses
because theyfunction to
arrest orinhibit the tendency towardprecise repetition.There are
no long-rangedelays that give rise to suspense.Expectation is
latent, andaffect tendsto be ephemeral. Thecomposer plays, as
it were, with the listeners latent expectations of precision and
regularity.
The operationof thelaw of return maybe contrasted with that of
the principleof successive
comparison bynoting that in the latter
case tensionarises outof deviationsinvolved in the repetitionof
the soundterm, out of the fact of repetition, while in the former case
tensions arise
because repetition is expectedbut is not asyet forth-
coming. The recurrence itselfrepresents, nottension, butthe relaxa-
tion phaseof the total motion.It createsclosure anda feeling of
completeness. Notice, too, that while the principle of successive
comparison tends to emphasize the differencesbetween thepattern
and itsrepetition, thelaw of return tendsto emphasize the similar-
ity between the pattern and its recurrence. This does not mean, of
course, that differences will not be noticed. VVhenever there is
repetition thereis boundto be successive comparison; and, though
diHerences will tend tobe minimizedin the case ofreturn, theywill
nevertheless be aesthetically effective.
In spiteof thesesignificant differences,it is not alwayspossible to
designate a passage as involving onlyreturn or only reiteration.In
the firstplace, someaspects of a givensound termmay returnwhile
others do not. For more than a melodic or harmonic norm is estab-
lished asgiven in any particularpiece. Normsof tempo,dynamics,
texture, inst1'umentation,and goodshape arealso established; and
the recurrenceof these,either singlyor in combination, maygive
rise to feelings of partial return. In the development ofMozarts
String Quartetin D Major K. 575!, for instance, anew themeis
introduced. Thistheme creates a feelingof partial return, notbe-
cause itsmelody issimilar tothat ofthe openingtheme orbecause it
is in the samekey asthe openingtheme, butbecause itstexture is
like that of the opening. Furthermore, since it is precededby less
palpable patterns, it appearsto be well shaped,like the opening
theme. To take another example, the Hrst fugue of the Well-
tempered Claoier
clearly involvesmelodic reiteration,but its har-
monic articulation is a result of the operation of the law of return.
Thus thereare degreesof return both becauserecurrences may
be only partially similar and becausesome aspects
of the sound
term may recur while others do not.
The law of return appears to operate most effectively where the
given soundterm is left incomplete.Since the sound termis a
Gestalt whichsets upforces towarda particularkind of closure, the
only wayit can be closedis by repeating itwith a new andmore
Hnal ending.A corollary of this would seemto be the fact that a
varied recurrenceof a well-shaped butincomplete termwhich has
developed strongtendencies toward completion willseem more like
a return than will a somewhat more exact recurrence of a closed
sound term.In theformer casethe changes
are understoodas arising
out ofthe necessity
for closure.Conversely the
more closed
a sound
term is, the more its recurrenceis likely to be exact or almost exact.
The listenersopinion ofthe significance
of any given repetition,
and hence his impression of completeness,is partly a result of the
experience ofthat aspectof style which is called form.Repetition
has onemeaning ina movementwhich is believed tobe a fugue,
another in one which is believed to be a theme and variations, and
still another in one believed to be a sonata form. Moreover, the
same generalscheme of repetition may have different signil-ications
in different contexts. For instance, in a piece known to be a fugue,
the imitative entrances of the fugal exposition arouseexpectations
of continuity and continuation. 'I`he repetitions do not give rise to
saturation. However,
a similarseries ofirnitative entrances
gives rise
to differentexpectations itif formspart of the developmentsection
of a sonata-form movement, as, for instance, in the first movement of
Beethovens PianoSonata, Op.111. Hererepetition and the texture
to whichit givesrise arenot thenorm; continuityof processis not
expected, atleast for any periodof time; and repetitionmay well
give riseto saturationif it continues forany lengthof time.In both
cases notonly are different expectationsaroused but different stand-
ards ofcompleteness are brought intoplay.
Fortunately themodes ofrepetition andthe law of return are
among thebest understoodprinciples in music. Thereforerather
than expatiate in generalupon theseprinciples, letus returnto the
examination ofa particularpiece ofmusic. Sincethe openingsec-
tion of the fifth movement ofBeethovens String Quartet, Op.131,
furnishes an excellent example of theoperation ofthe lawof return,
we onceagain turnto this music ratherthan to another example
whose explication would involvea fairly lengthy preliminaryanaly-
sis.
After thetwo introductorymeasures already discussed, the
section
opens withan eight-measure theme. Thetheme ispatently incom-
plete bothmelodically andharmonically. Butthe listenersattitude
at the end ofit is neutral; thatis, thoughhe knowsthat the music
will continue, he does not know whether the theme will be com-
pleted bya consequent phrase orwill simply be repeated.Observe
that hadthe consequent or concludingphrase been presented imme-
diately afterthe antecedent,as indeedit easilycould havebeen, no
feeling ofreturn wouldhave beenengendered. In other words,had
Beethoven written the followingsee Example 53!, thesecond period
would haveconstituted avaried andconcluding reiteration of the
first phrase.There wouldhave beenno feeling of strong tension
toward closure,no suspense,and it would have received little atten-
tion. It would havebeen apert themewith a rather witty conclu-
sion, asort of OHenry ending.There wouldclearly havebeen no
feeling of return.
What Beethoven
actually does,
however, to
is repeatthe opening
phrase in a higher register. Oncethis has taken place,further
reiteration isimpossible. Another
repetition ofthis phrase,even in
its concluding form measures 45-54!, would have resulted either
in anticlirnax or in saturation. In order to create a forceful desire for
return, it is necessary both to depart fromthe regularityand con-
ciseness of the themeand, atthe sametime, tocreate unambiguous
expectations. Beethoven at oneand thesame timebegins tocreate
rhythmic incompleteness and movestoward a harmonic position
whose incompleteness will be more patent,
i.e., thedominant meas-
ures 19-24,Example 50!.In the following measurethe harmonic
incompleteness as is, wehave seen,heightened byincreasing rhyth-
mic incompletenessExample 51!. In fact, both grow together.
Presto e , 92 _,
.I _' ,
ll__'I
-W _-v'l-__--Z-
-_ __'___l_
_-'____'_U"_
lY ___TL_ _ - 'i- _ _ll
__ #_ _
__J_.__
|9292
_I l _J
|"|.____l_ _ __I
_I ___ _ I _'___ ' ____
_ll_l__|I'l___-__-____--_'__ l_
__Z_"____l_l____l__ll.
-_.____l_l___l_92!'L__ _ _'_________'_'_ __ ___
_ _ _l|"'l
ll92.lI I-lJ.'.__"Y_
_ _'_____ __ ___ _____-l____-.___
I | I ____
@ Af;
.,, _lJ
_'--"
U-B'
1414 414144
__ 1:1 l _:___
I l
4 _'
|t_1;l_
ll.._._
AAAIJT_tit ilitidd-|_._ll'1t _I
9292a'_ti' lJl__. - l ll i|Il _ I r1.l
Il :_lt_
J I l lii
if '92/l___l1_11I11__'l
YL flYr_|1r All I l I l 9292.ll11tt,|i ltlliiii
|l_.92!lD I f'1l1111l1$._
pgixgg IIYltl I5.4
rl! IlU11lZl _1lilI11;u1l;ClZ'1H|U
'1l
I I T|1_rv'1 l_|l1?| $1l?$l-iilldiblliiilil -1 -Q1
1
rl If
EXAMPLE 53
Notice that every feature of the music is changed in order to

heighten ourdesire for return: rhythm and harmonyare incom-
plete and,in the end, quiteambiguous, whereas earlier theywere
very clearand wellformed; melodicmotion, whichin the opening
phrase was strong andwell patterned,is alsodestroyed; thetempo
too is changed; and,finally, in measures 87-41Example 52! the
distinction betweenfigure and ground is almost completelyanni-
hilated.
But at the very moment whenrhythm, harmony,texture, and
even melodyin the sense ofpattern seemall but destroyed, the
little figure which opensthe movementand the first phraseraises
our hopesand redirectsour expectations
of completionand return.
Now we are certain as to what is coming. Expectation is definite
and powerful,a fact which makesthe delaycreated bythe ritard
and fermata all the more forceful. Then in measure 45 there is a
return tothe openingtheme, which is nowcompleted; and the com-
pletion isemphasized by the measures which follow. The satisfac-
tion createdby this return is not merelya function of the incom-
pleteness of the openingmeasures but also aproduct ofthe creation
of further incompleteness inthe middle part measures19-44!.
What weexperience at measure 45 is notsimply thereiteration of
a themeheard earlier.We havereturned toa melodywhich almost
had beendestroyed, toa texturewhich is familiar, to a harmonic
motion whichis certain and regular,to a tempo whichhad been
altered, andto a palpable unifiedGestalt which had been dis-
rupted. Indeed,the feelingof return, the pleasureof resolution,is
so strongthat the rather piquantflavor of the additionalclosing
measures 1-54! goes almostunnoticed unless it is called to our
attention. At all events, the effect of these measures would have
been much more noticeable had this version of the theme irnme-
diately followedthe openingphrase, as
in the construction given
in
Example 53.
V

The Weakeningof Shape
Throughout thetwo precedingchapters wehave beenconcerned

either directlyand explicitlyor indirectly and by implication with
the problemof shapeand its articulation. Generallyspeaking the
discussion has involved anexamination of the wayin which expec-
tations are aroused and tendencies inhibited when well-articulated
Egures orprocesses are disturbed ordelayed insome manner. In-
deed, suchdelays and irregularities aremost effectiveprecisely
when patternsand shapesare distinctand tangible;for it is then
that expectationsof continuationand closureare mostclear and
unambiguous.
The Nature of Shape

The apprehension of a series ofphysically discrete
stimuli ascon-
stituting apattern or shape results from the ability of the human
mind to relate the constituent partsof the stimulus or stimulus series
to one another inan intelligibleand meaningfulway. For an im-
pression ofshape toarise anorder mustbe perceivedin which the
individual stimulibecome partsof a larger structureand perform
distinguishable functions within that structure. A shape or pattern,
then, isa soundterm, asdefined inchapter ii,and it is meaningful
and significant
because its
consequents canbe envisaged with some
degree of probability?
One of the absoluteand necessary conditions forthe apprehen-
sion of shape, forthe perceptionof any relationships atall, no
157
matter whatthe style,is the existence ofboth similaritiesand dif-
ferences amongthe severalstimuli whichconstitute theseries under
consideration.
If the stimuli comprisingthe seriescannot beperceived as being
similar in any respectwhatsoever, then they will fail to cohere, to
form a group or unit, and will be perceived asseparate, isolated,
and discretesounds, signifyingnothing. Sincecontrast andcom-
parison canexist onlywhere thereis similarityor equalityof some
sort, themental impression created bysuch aseries willbe oneof
dispersion, notdisparity; of diffusion, notdivergence; ofnovelty,
not variety.
Complete similarity, proximity, andequality ofstimulation, on
the
other hand, will createan undifferentiated
homogeneity out of which
no relationships can arisebecause thereare noseparable, individual
identities tobe contrasted,compared, orotherwise related. There
will be coexistence and constancy, butnot connection;uniformity
and union,but not unity. In short, bothtotal segregation and total
uniformity will produce sensation,but neitherwill be apprehended
as pattern or shape.
Because from both atemporal anda musico-spatial
point of view
the factorswhich in large measureare responsible
for segregation
and uniformity-factorssuch assimilarity, proximity,and equality
of stimulation-admitof varyingdegrees or
gradations, theappraisal
of shapeis a relative one,depending upon
the generallevel of dif-
ferentiation and homogeneity prevalent in a particular musical
style. Thatis, in a stylein whichthe members of the stimulus series
are not differentiated orsegregated from one anotherin a partic-
ularly emphaticor marked manner, modestdifferences though
necessarily appreciable ones! willgive riseto well defined, palpable
shapes. While in a style marked by radicalsegregation and extreme
differentiation, thesesame modest differences willproduce onlyuni-
formity. Totake thereverse case: a stimulusseries which appears to
be well articulated andclearly organizedin a piece wherediffer-
ences arevery marked might well seem chaoticand incoherentin a
piece inwhich thelevel of uniformity washigher.
These observationscall attentionto anothernecessary condition
for theapprehension of
shape and pattern. Namely,it is not enough
The Weakeningof Shape 159
that diHerentiationand unificationsimply exist. The articulation
must be sufficiently marked and salient relative to the context in
which it appears tobe noticed.The gradualaccretion ofrelatively
slight differences,which inthe givenstylistic contextare subliminal,
will not give riseto animpression ofshape. Forinstance, inequality
of stimulationwill tend to producesegregation only provided that
the inequalityentails abruptchange. 2Thus a mode of differen-
tiation which would have been incisive in a context of moderate
similarity, proximity,and equality,might be subliminal in a style
where these factors were more extreme. These remarks are also
applicable tothe perceptionof unification:if the factors making
for
unification aresubliminal withinthe context,then noimpression of
shape will arise.
The factthat theapprehension of shape isa functionof thenorms
of articulationestablished withina givenstyle doesnot meanthat
any stimulation can createan impression of shape.As wehave seen,
neither totalsegregation nor total uniformity-no matter whatthe
style-will create an impression of shape.But evenbetween these
extremes, shape does notseem tobe completely relative. Some
series
of stimulido, infact, seem better shapedthan others,irrespective of
style. Indeed,were thisnot thecase, allstyles wouldseem aesthet-
ically andpsychologically equivalent-which they do not.
The basisfor this seemingly absolute standard ofarticulation and
differentiation isto be found in the fact that our comprehensive,
over-all stylisticexperience, ourexperience ofBach aswell as of
Bartok, of Mozart aswell as of Monteverdi,gives riseto an all-
embracing normof pattern and shape,just as our total stylistic
experience becomes the foundationfor comprehensivenorms of
consonance and dissonance and for an over-all norm of texture. And
the peremptorypresence ofthis generalstandard enables us, even
compels us,to compare albeit unconsciously! the qualityof Schu-
berts patternswith thoseof Schoenberg and thoseof Debussywith
those ofDufay.
However, withinthe contextof a particular work in a specific
style theevaluation andappraisal ofshape isrelative. Thisrelativity
is a product of the psychological
demand fordistinct, substantial,
good shape: it is a result of the operation of the law of Pragnanz.
If there is anyarticulation ofthe stimulusgroup intosimilarity and
difference, proximityand remoteness, equality anddisparity ofstim-
ulation, thenthe mind will probably find somebasis no matter how
feeble orfarfetched itmay appearto be! for relatingthe stimulito
one anotherin an orderly andmeaningful way-for apprehending
the stimulusgroup asa shapeor pattern.Indeed, evenwhere no
objective diiferentiation
exists, themind will, if at all possible,im-
pose suchdifferentiation. But neither the differentiation of the
stimulus series nor the existence of unification between its several
parts aresufficient conditions
for the impression ofshape. Forthe
parts of t_he series
must be related to one anotherin a consistent and
orderly manner.They are so related and orderedby the social
phenomenon we call style. And if we do not understand a given
style, if we lack the properhabit responses, we will either fail to
apprehend shape or, if we apprehend it in terms ofanother style,
we
will fail to comprehendit. That is, the impression ofshape may
arise because two stylesystems have sound stimuliin common,but
such animpression willbe incorrectlyinterpreted and,in point of
fact, notreally comprehended at all.
The Weakeningof Sluzpe

The fact that the appraisal of shape variesfrom style to style, and
even fromwork to work within a singlestyle, doesnot meanthat
all patternsor shapesin a particular work will be perceived as
equally good.Some shapes will at times appearto be significantly
stronger, moreclearly articulated,than others.
The words good or strong, poor or weak as used in the
discussion of
shape arenot to be construed as implyingvalue judg-
ments. 0nthe contrary,weak, ambiguous shapes may, as hasbeen
mentioned fromtime to time throughoutthis study,perform aval-
uable andvital functionin creatingand moldingthe affectiveaes-
thetic musicalexperience. Forthe lackof distinct, tangible shapes
and ofwell-articulated modesof progressionis capableof arousing
powerful desiresfor, andexpectations of,
clarification andimprove-
ment.
This aspect of musical structure and expression isone which has
unfortunately received but scant attention from music theorists,
aestheticians, and critics, who,enticed andbefuddled bythe hon-
orific garlandsso plausiblybestowed uponregularity, symmetry,
and pristine clarity, have continually construedinevitability to
mean unequivocal progression. These scholars, intheir intricateand
ingenious attempts to deriveall the melodic materials of a musical
work from a single motivic germ or to subsume allharmonic
changes underthe structureof a single cadence,have with Tal-
mudic tenacityconfused both unification anduniformity withunity.
Yet thefact ofthe matteris thatsome ofthe greatestmusic isgreat
precisely because the composerhas not feared to let his music
tremble on the brink of chaos,thus inspiring the listeners awe,
apprehension, and anxiety and,at the same time,exciting hisemo-
tions and his intellect.
A stimulusseries, then,
is wellshaped when its progress,
its artic-
ulation intophases ofactivity andphases ofrest, itsmodes ofcon-
tinuation, its manner of completion and closure, and even its
temporary disturbances and irregularities
are intelligibleto theprac-
ticed listenerand enablehim toenvisage with some degree of speci-
Bcity andaccuracy whatthe later stages ofthe particularmusical
process willbe. Because good shapeis intelligible in this sense, it
creates psychological
a atmosphere
of certainty,security, andpatent
purpose, inwhich thelistener feelsa senseof controland poweras
well asa senseof specifictendency anddefinite direction.
Because, as has already
been observed, the apprehension of shape
is in part dependent upon the existence of a balanced relationship
between uniformity and differentiation,
the impressionof shapecan
be weakened by the exaggeration or intensification ofeither force
within the context ofa givenmusical work.Shape may,from this
point of view, be regarded asa kind of stylistic mean lying be-
tween the extremes of chaotic overdifferentiation and primordial
homogeneity. Thus while a balanced relationship, as established
within the style, producesan impressionof well-organized patterns,
exaggerated differentiationgives rise to disorganized ones, and in-
tensified uniformity results in unorganized ones.
It is extremely difficult,if not impossible, toweaken theimpression
of shapethrough exaggerating
differences without
destroying shape
completely-without producingmeaningless chaos. The causesof
this diflicultylie in the characteristics
of the musical stimulithem-
selves, inthe capabilities of thehuman ear,and inthe natureof the
human mind.
Because musicalstimuli are necessarily similar to each other in
many respects, it is almost impossible to isolatetones throughdis-
similarity. Thevery fact that musicalmaterials areas a rule tonal
means thatthey have such physicalcharacteristics as frequency,
amplitude, and timbre in common. Itis likewisediflicult to achieve
an impression of tonal segregation non-proximity!. For the mind
can withthe aidof the overtone series relate pitchesin widely sep-
arated registers.
Furthermore, diflerencesin pitch, volume, andduration arelim-
ited bythe physiologicalabilities ofthe humanear. Foronly certain
pitches canbe usedin music,are audibleto humanears, andthose
must be of a certain volume and duration if they are to be per-
ceived. Thus the numberof stimuliand theirquality is limited.
Most importantof all is the fact that if segregationor differen-
tiation, whetherspatial or temporal, isnot so extreme asto pre-
clude all possibility of pattern apprehension, then some sort of
shapes willprobably arise.For the human mind,operating under
the aegisof the law of Pragnanz, willgo to almost anylengths to
avoid the doubts andanxieties whichare createdby ambiguity.
That is, the mind will tend to apprehenda group of stimuli as a
pattern orshape ifthere isany possible way of relating thestimuli
to one another.
Indeed, if weak shapeshave resultedlargely from exaggerated
segregation ofstimuli e.g., extreme distancebetween pitches, ex-
cessive temporaldisjunction, orboth!, thenthe mindwill prefer to
regard thestimulus seriesas beingincomplete see pp.130 f.!;and
since furtherdevelopments will be expectedto completewhat was
incomplete orto effect some othersort of clarification, thefeeling
of mentalcontrol willnot belost. Oruncertainty anddoubt maybe
avoided if the series can be or is understood as having a special
designative meaning.
For example,if segregation non-proximity!,
disparity non-similarity!, and inequality of stimulation can be
understood as
a humorous or grotesquedistortion ofsome normative
type of pattern construction,then the doubt and anxiety which
might otherwisehave arisencan berationalized andremoved.
Though exaggerated differentiation wouldtheoretically provide
a methodof weakening shape, itis extremely
difficult to achieve and
hence extremelyrare. Indeed, this author has beenable to discover
no clear-cutexample which
is notprogramatic incharacter.
Uniformity
Uniformity arises
where theseveral elements of musicalorganiza-
tion-pitch progression,rhythm, harmony, and timbre-combineto
create equality,similarity, andequiproximity ofstimulation onthe
same architectonic level.
In terms of pitch progression, uniformity is produced when a
series oftones, whetherconjunct ordisjunct, areequidistant from
one anotherand aredifferentiated inno otherway; e.g.,rhythmically
or harmonically.For it is only by virtue of the existence ofdiffer-
ence of distance that the mind is able to establish ordered relation-
ships between successive tones.If all the membersof a series are
equally distant,none will be markedfor consciousness, and hence
there will be no focal pointsin the series withreference towhich
organization cantake place and shapecan arise.
Psychologically speaking,what happenshere, and this applies
with equalforce to rhythmic andharmonic uniformity,is that the
uniform series,lacking internalarticulation, establishes
no pointsof
activity and rest. Consequently the listeneris unable to envisage
the ultimategoal, thelogical stopping place ofthe series.He is able
to apprehendthe modeof continuationbut is unable to compre-
hend thebasis forcompletion and closure. The series is,so tospeak,
frictionless andout of control. Doubtand uncertaintyarise notbe-
cause thelistener doesnot know what the next stimulus is going to
be-he knows, for instance, if the series is a chromatic scale that the
next tonewill probably be a half step away from the one being
heard at the moment-but because heis unable to envisage where
the serieswill end, to controlits destinymentally.
Of course,if equalstimulation isonly momentary, if the ambiguity
is resolvedalmost immediately,then doubt, uncertainty, and anxiety
will be transient andevanescent. But when uniformitypersists for
a relatively long time, the serieswill have a cumulative effect,
arousing strong desires fora return to more intelligible and con-
trollable shapeand amore certainand securepsychological atmos-
phere.
Because chromatic and whole-tonescales andaugmented and
diminished triadsall involve intervallic equidistance, they create
uniformity andproduce ambiguity. And it is no accident thatsuch
weakly shaped, ambiguous serieshave tendedto becomeidentified
with affectivityand haveso oftenbeen used to expressintense emo-
tion, apprehension,and anxietysee pp.218 f.!.
The followingfragment fromLiszts PianoSonata isa particularly
interesting example of uniformity of pitch succession because it is
doubly chromatic;that is, chromaticism existsboth within the
groups ofsixteenth notes and betweenthem Example54!. Theo-
1--w
:ax-_-'=~" "i" `*` r ""' ""` "" "'' *"'

EXAMPLE 54
retically thisseries couldhave endedanywhere andclosed inany

key at all. For instance, had motive x, which breaks the series,
occurred afterthe first group in the secondmeasure, thefinal key
would havebeen eitherB majoror C-sharpminor. Orhad theseries
continued untilthe end of the measure, the final key would have
been Fminor or A-flat major.This dualityis possiblebecause the
cadential motivex! is itself somewhatambiguous. Forthe final
tone Amight havebeen thetonic ofA majoras wellas thethird of
F-sharp minor.'I'his, incidentally,is entirely appropriate sincethe
passage quoted comes atthe conclusionof a section whichcon-
sistently employs,
both melodicallyand harmonically, a diminished
seventh chordthat might have beenin either A major or F-sharp
mmor.
Uniformity ofpitch succession
must besupported byuniformity
of rhythmif an impression ofhomogeneity andweakened shape is
to arise.Conversely rhythmic
uniformity, producedby equality of
accent andduration, mustbe complemented by uniformity in other
The Weakening of Shape 165
aspects ofthe series-suchas pitch succession, harmonic
progres-
sion, and timbre-if the over-all impression is to be one of homo-
geneity. Thus a majorscale ora cadentialchord progression,though
realized intones ofequal duration,will not seem weaklyformed or
ambiguous because their articulationliterally createsaccentuation,
even whereno changein physical intensity is made.
Uniformity likearticulation may
be architectonic
in nature.It may
be establishedon all levels, asin the example fromthe Liszt Sonata,
or it may existon somelevels butnot onothers. Inthe lattercase it
is generallyone ofthe higherlevels, though
not the highest, which
is uniform. Furthermore, on some levelsall aspects of the stimulus
series maybe uniform, while on other levelssome aspectsmay
exhibit uniformityand othersmay not; for instance,rhythm may
be uniform though pitch succession isnot.
In Example55 from the first movement ofBruckners Seventh
Symphony, the stimulus seriesis wellarticulated onthe lowestlevel
but isweakly shaped on thehigher architectoniclevel. Onthe lowest
level the articulation ofthe rhythm into an anapest footis clear
and unambiguous. But since the seriesconsists onlyof anapests,
the rhythmof thenext higherlevel isuniform. It is a series ofbeats
1;:.?'b 'l T I; - _-7_ _ _ T-- -'

nhl ...nv _Q ' ii--?"" _ _*_ _
- Cl' A 3.1b&'-F b' 1-1 g, bib!,
SS pe'r-fect mf d`_ min- whole
4th step ished 4th step
EXAMPLE 55
of equalduration andapparently equal accentuationf The same

is trueof pitchsuccession. the
On primary architectonic level
a well-
shaped motive consisting of
a fourth,followed bya half step, fol-
lowed by a minorthird arises.
But onthe second level pitchprogres-
sion isuniform-merely asuccession minor
of thirds see analysis
a!.
Or the series might
be analyzedas aseries ofaugmented fourthsa
minor third apart.
Notice, however,that the primary groupsare not all precisely
equivalent inpitch succession. For thethird groupin eachmeasure
is slightlydifferent in organization fromthe other two see analy-
sis b!. However, because of the general uniformityof pitch suc-
cession and temporal durationas wellas because of its inconspicu-
ous placementin the series, this slight diEerentiationbetween
groups, though important infacilitating thechanges which bring the
series toits conclusion,does notupset or weaken theprevailing
feeling ofhomogeneity. In short, thedifferences are subliminal. The
listener becomes consciously aware of the differentiation, ifat all,
only in retrospect, onlyafter thepassage is over andthe existence
of thedifference has been pointedup by the partit playsin moving
the line smoothly toits conclusion.
Harmonic uniformitymay arise: ! as a result of equality of
interval within the vertical harmonic organization itself; ! be-
cause, thoughthe vertical organization isdifferentiated, allthe
harmonies inthe seriesare preciselyalike; ! where, thoughdif-
ferentiation between successive harmoniesoccurs and a process is
developed, the
progressions making
up the series areall equal;and
! where any or all of these modesof achievinguniformity are
combined.
As wasthe casewith pitch succession, ifthe vertical organiza-
tion is undifferentiated as to intervallic distance, then there can be
no focal point aroundwhich organizationcan take place-out of
which shapecan arise.Thus, for instance, adiminished seventh
chord, anaugmented triad, or aharmonic structure
built uponequal
fourths will have noroot andno shapeand henceno tendency.In
many contextsthe root is implied or the series iscontrolled in
some other way. Thisis thecase withthe followingseries ofdimin-
ished seventhchords occurring near the end of the last movement
of BrahmsSecond Piano
Sonata Example
56!. The ambiguity of
F.
F .l1H - HI'
Q ''-'_| If1_1-1
~ HH -_Ill
nlrrgslrnfi ll ll .=r_;.:r_;v
ll 1
.,
1` ". 'I Vz . =~ '!' =-E
1 `1
'-` . | . ;_r I gl 1
"_ EE 2:1
EXAMPLE 56
intensified bythe mannerof chord construction; thatis, the outer
voices arevery consonant
and tendto makethe internaldifferences
seem minimal. On the other hand, there is some melodic articulation
in the series, thoughthis is too minimizedby the fact that the two
halves ofthe seriesare in this respectequivalent.
Finally, it should benoted thatin the Debussy example and to
some extentin the Brahms example as well ambiguity is also a
product ofrhythmic uniformity.To put it somewhatdifferently, the
equality ofrhythmic stimulationallows theuniformity of the other
elements to take their full effect.
On the higher architectoniclevels harmonicmelodic uniformity
may result,not from the fact that the chords themselves are con-
structed ofequal intervalsor from the fact that all the chordsof
the series are of the same construction, but because the successive
groups, whetherprimarily melodicor both melodic andharmonic,
are equivalent.Uniformity here is a matter of continuous and
equivalent process.
What iscommonly called a sequence is a type
of uniformity.
One of the mostcommon typesof melodic-harmonicsequences
is one that movesharmonically throughthe cycle of fifths, either
descending asin A1 in Example 59! or ascending asin A3!.
I. 2. 3.
A.
:I il wi' 11 be i t
Implied root
progressions
1. 2.
B|
22| 1:1 it .
EXAMPLE 59
Uniformity results,not onlyfrom thefact that every otherchord is

the dominant of the chord which follows it, but also from the
fact that one of the voices follows a chromatic line while the
other voicemoves inminor thirdsand halfsteps. Otherversions of
such asequence of dominants arepossible; as, for instance,in A2,
where theseries followsnot the cycle offifths but a chromatically
descending pattern,or in BI, where the implied root progression
moves byminor thirdsand fifthsand theguiding or conjunct line
moves bywhole tones!
Any seriesin which a continuousprocess isdeveloped willtend
to be ambiguous because the point at which the processwill be
broken or the series concluded is in doubt. This is the case in the
following continuous
schemes Example
60!. In the first of these
I 1. I zI 3.
~ I -4. 5. I II Is.I I
EXAMPLE 60
CI!, tonalharmony is employed and, though thesteps ofthe con-

junct lineare notall equal,the groupsare equal:
the conjunct line
moves by half step,
whole step, half step,
etc. Because
of thisequal-
ity of groups, the
series couldcontinue indefinitely
in a frictionless
way; andthere isno wayof knowingat whatpoint theseries will
be articulatedand reach its conclusion.Many other such series
might beconstructed. For
instance, inC4, C5,and C6there areno
functional, classicalharmonic relationships, but the sequences are,
nevertheless, uniform
in their mode ofprogression.
Such equalthough harmonically non-functional passagesare not
uncommon incontemporary music. The followingexcerpt fromthe
Finale of Bart6ks Fifth String Quartet is given as an example
Example 61!.
In this instance we
Hnd notonly uniformity
in the
succession two-measure
of groups
which descend in whole-tonesbut
also uniformitywithin the groups. Fora glanceat the pitch succes-
sion of the lower voices makes it clear that the lower voices are mov-
ing throughthe augmented triad seeanalysis ofreduction given in
Example 61!. In additionto thisuniformity ofpitch succession,
there
is alsouniformity ofrhythm andmotive between
measures both
in
upper and lower voices.
_'ll -ID ~ --_ I b

.--liiinillg-,!
-. I ._ Il _ _'Y-_ _
|l_Y1H -.l_-'11
9292l'lfI2H_1 Q-_lllj-I _Y"'| 92'| IQ- Il* 'I _
.' - ll |I 1|
cu .
|`B1!--l;-_-1_1-HIl1-
1llY1I1"'l 2--1
_--_--K--TI-Y _ll |=2]!__!|'F-_.h|_-
Yl m -1___'f1T|'i-i4
2 2-_'| ll"|-if-_-filil ILO 2-|__2"if
_1 _--
gf
E
EXAMPLE 61
Many differentforms of continuous uniformityare possible.For

instance, thesuspension technique
used in fourth-species counter-
point ora variant
of it,a series
of prepared
appoggiaturas, may
give
rise tocontinuous processes,particularly wherethe controllingline
is tonallyambiguous. Although in the example givenhere theinter-
val pulling the upper tone down,causing theseries tomove, is
either amajor orminor seventh,this neednot be the case.Theo-
retically sucha seriescould be constructed atany interval.As a
matter offact, someof the sequences given in Example59 A2 and
B2! arealso forms
of fourthspecies suspension!
counterpoint.
bf- ~- -.._., ,_______
2:--ld " Ef tr kr F pf '... I'{} -
EXAMPLE 62
Continuous sequential processes neednot, of course, seemam-

biguous. If
the series
is tonal or differentiated
in someother way!
and wellarticulated asto rhythm and melody,both the successive
steps ofthe sequenceand theultimate pointof its completion may
seem tobe predictable.But whereother aspects of the series tend
toward uniformitysuch a series maygive rise to uncertaintyand
doubt, eventhough it is not completely uniform.For oncesuch a
continuous sequentialprocess isestablished, itdevelops aninner
momentum anddrive which enable it to veer very easily in the
Copyright1936 byUniversal-Edition. Copyright
assigned 1939
to Boosey
&
Hawkes, Ltd.Used bypermission.
direction ofuniformity or away fromthe envisagedpoint of com-
pletion. Forthis reason
irregularities insuch processes
will not,
as arule, be apprehended as
articulating theprocess butwill ap-
pear as
disturbances and
deviations from
the process
which addto
the insecurityand uncertaintyinherent in the fact of continuous
process itself.
Where_ the
process isregular anduniform, thelistener
knows or believes that he knows what the next stimulus will be but
is in doubt as to where the process will be broken. Where such
seemingly continuous
processes are
irregular, thelistener isable to
envisage neither
where theprocess willbreak norwhat thenext step
in the process will be.
The statement that the listener does not know or is doubtful as to
where suchcontinuous processes
will end or breakmust beunder-
stood in a conditional sense. Whatthe listener really doubts is the
manner inwhich a sometimes onlydimly envisagedgoal will be
reached. Hehas ageneral feelingas to what thefinal goal of the
series is,but he is uncertainas tohow thepresent process
will get
him there and what detours and obstacles will be encountered en
route. Andit is only atthe pointof reversal, the pointat whichthe
process isbroken andanother modeof continuationtakes itsplace,
that thelistener finallyis ableto envisage
his goalwith any degree
of security.It is thus the point of reversal ofprocess whichcon-
stitutes theclimax andturning point of the passage, thepoint at
which doubtand anxietyare replacedby morecertain anticipation.
These facts of process continuation are well understood by the
practiced listener.
Because of
this, sequential
process willgenerally
tend to raise doubts in the listeners mind, will seem more or less
ambiguous. The
controlling factorfor the listener iscontext. Ifthe
sequence occurs
in a generally controlledcontext, forinstance, as
part of or in connection with an important and well-articulated
theme, thenit would probably giverise tominimal tensionand un-
certainty. But if the continuous processoccurs in connection with
other typesof uniformity-in a passagewhich is clearly develop-
mental-it might well create powerful doubts and anxieties, even
though it apparently isquite tonaland wellarticulated; forin this
case thelistener isaware thatsuch passages are unpredictable,that
the seriesmay veerwith greatease from its presentcourse.
Finally, it is importantto makeclear thatit is the hypothetical
meaning ofthe severalsuccessive stimuli which is ambiguous, not
necessarily their evident meaning.
A seriesis ambiguousbecause the
listener isunable toenvisage withany reliableor decisivedegree of
probability whatthe future terms will be. Indeed,because ofthe
listeners attitude toward suchpassages, almost
all soundterms tend
to seemequivocal. However, once theconsequent term arrives, the
listener maywell be able to understand itsrelationships to the terms
which havepreceded it.Yet eventhis understanding is conditional.
For thedeterminate meaning of anygiven soundterm canonly be
known whenthe whole passage has been completedand a clear
point ofstability andorientation has
been reached.
The developmentsection measures72-86! of the Finale of
Haydns PianoSonata inA-F latMajor furnishesan excellentex-
ample ofuniformity of rhythm andpitch succession coupled with
continuous sequential processes Example 63!.
-;_'|.1-_._*1Y1111Uifiijflm!1lHK-ll-l.:
I1 II _|1r1r1r1|1-iiiiuixiih l
I7l'9292|Ql
Nv1.i---Il'
ll '--J-J--12-1-I II
.JaI--1-1-l11__llIQlQ__-__--_-_A1.l-AQ-
-'I ll ll ll l_r1|111l1!
~~-1I____:__- 1l:::!:::_|!!!l!g-_|_|:4_.|;-
--_ _-_ -Il --I
;|lI.nvt' | 'l LU i;|-us-l-i1"7|is_1-1111iiii `
UQ.92|T1'i1U'il1il_l~ll1.l__'VlI
l._|'1l1iQ 11';l:'i_1111
U I 1 _1 iw! I; S 1
YT 1
11 lll1l'Y'i_
Y- 1; 1
1'I| _II
I.-I ll I 11- ill ___ ____ ____
'7'l"l7`lV`__lIl11|-T1__l1.___Il--..ll.;______.l_.Q..;__;;l
1.l9211l1;.l_AZ Llhll ll IQ Illlf'_V`l1l1Ill"lflllll-Il_l_YlY'-_III
I92Q'lL7l
..`J_T'1ll1|1l1i11i1L1LA1I' I l;LLl1l'Y'|rI|
Z 1lZY'i"Z 1I1l17'i'lY`l1"1i111l'LLL1l_11Tl || |_`,_g|.| |
|`g ; 11
1l||T11111Tilliiiililill I l
7|9292ll_ilv11lilC11Zll_1i11_I
11| I 13.111111 Qxlllv-_xl | U|
H5 b
EXAMPLE 63
In the Hrst place,a glanceat the example showsthat there is

general rhythmicuniformity. At the beginningthe rhythmicstruc-
ture within the measure is somewhat problematical. Because the
passage begins with an anacrusis, thefirst rhythmic group is per-
ceived asan amphibrach,and this feeling of amphibrach rhythm
tends toperpetuate itself.
In orderfor this to takeplace, thesecond
beat of each measure must functionboth asthe end of one group
and thebeginning ofthe next see Example 64!. This impression of
the dovetailingof groupsis strengthened by the fact that the Hnal
tone of each measureis a preparation for the first beat of the fol-
lowing measure and alsoa resolutionof thefirst beatof themeasure
in which it appears.The rhythmic articulation within measures
could beclarified werethe performerto makethe final eighth note
of each measure an anacrusis to the first beat of the next measure
V-V-v-v
1411
EXAMPLE 64
by playingit slightlynearer tothe downbeatand accenting it some-

what, asin Example65. However,since sucha performance would
tend to weaken theeffect of the passage,
this would seem aques-
tionable interpretation.
Indeed, anyinterpretive accentuationon
the partof theperformer isunnecessary and uncalled for.The pas-
sage shouldbe playedquite uniformly.Incidentally thisis an ex-
ample ofmeter withoutrhythm. Thebeats areclearly differentiated
into accentedand unaccentedbut grouping is inappropriateand
undesirable.
nfl-|i1!r1 , - " _
I'A%F=J= '=1"= i===|I=|l-i'
lil-_J&21_;gZ;1;3!~';'-11;3w:ll1L.!: 1
sa * ,92./J,92./ * 92/,,92-/_
EXAMPLE 65
After measure7 as numbered inExample 63!this overlapping

of rhythmicgroups ends
and afairly cleartrochaic rhythm,implicit
in the lower voice from measure 3, emergeswithin each measure.
But there is still uniformity between measures; henceno impression
of rhythmarises onthe higherarchitectonic level.Incidentally it is
just atthis point,where therhythm withinthe barbecomes clarified,
that the melodic and harmonic structure of the series becomes most
uniform, counteractingthe clarificationof rhythmic shape.
The line of pitch succession, though not completely uniform,
could hardlybe termedwell articulatedor distinctlyshaped, since
it lacks those changesin pitch direction andthose differences in
accent and duration which create well-defined substantial shapes.
Nor isthe scaledescribed bythe outervoices unambiguous. At first
it might be interpretedas thescale ofF minor,later asC minor,or,
even, asan altered scale in B-flat. And just as it seems fairly clear
that thescale isthat ofE-flat major,which itreally is,the uncertainty
of chromatic uniformity is introduced.
The absence of a well-shaped melodicpattern is important. For
were rhythm and melody distinctly shaped,the attention of the
listener would be directed to these and the feeling of security and
control foundin suchclearly articulatedshapes wouldconsiderably
diminish thesuspense and uncertainty engendered
by the uniform-
ity of the otheraspects ofthe musicalprocess. This
occurs attimes
in music of the nineteenth century,where ambiguousharmonic
progressions bear such distinctand palpablethemes thatthe un-
certainty of the harmonic continuum is more or less nullified be-
cause thelistener focuseshis attention upon the expressive and
well-shaped tunes.
The listenersdoubts andanxieties are
partly aproduct ofhis own
stylistic experience,
his awareness
that thisis part of a development
section andthat passagesin suchsections tendto be unpredictable,
deceptive, and irregular. And the general uniformity coupled with
the presenceof obviouslysequential processesconfirms thelisteners
attitudes andaccentuates hisfeeling oftension.
At the outset ofthis passagea combinationof suspension tech-
nique and harmonic sequence create amomentum ofprocess in
which whatarticulation thereis, is more thancounterbalanced by
the everpresent possibility
of equiprobable alternative consequents.
A schematizationof thesemeasures measures 1-6! is given inEx-
ample 66together withsome ofthe alternativeconsequents which
might havebeen forthcoming.For instance,the sequences might
have continued unvaried asin a!, moving throughE-flat majorand
D-Hat majorto a cadence inF minor,the keyin which the passage
begins. Suchfalse modulations are not uncommon. Theprocess
might as in 19! have beenbroken by using an augmented sixth
chord, which would have moved to the dominant of C minor, and
after thisnew modusoperandi theharmony mighthave progressed
through B-flatto thetonic A-flatmajor. Orthe suspension
technique
might havecontinued foranother measureas in 0!, reachingthe
dominant E-flatearlier thanis actuallythe case.Finally, had the
original modusoperandi beenperpetuated stillfurther as in cl!,
the samepoint would have been reached that is actually achieved
through the introduction of chromaticism.
-:...,, r-". "_ '" .. -.-
_ - C.-` ___ _ == -v-'__h_i_

aswf tisuw-3 uv wr ns 8 r=+=w_. u = ,1-
EXAMPLE 66
This lastobservation points

up thefact thatneither theintroduc-
tion of chromaticism in
measure 7 nor thepassage taken as awhole
can beunderstood and accounted forsolely onthe basisof the ne-
cessities of modulation and the desire to return to the tonic. Had
Haydns purposein writing this passage been merelyto return to
A-flat major,then itwould seemthat hewent toa lot of unnecessary
trouble. Foras everystudent ofelementary harmony knows, there
is no trick to modulating fromF minor to A-flat major. It can be
done with a couple of diatonic chords.
The introduction of chromaticism in the upper voice serves sev-
eral different though complementary purposes. First, it creates
enough varietywithin the continuous processof harmonicsequence
to precludetedium. Second, it allowsthe sequenceto becomemore
uniform, avoiding the more apparent articulation such as takes
place in measure 5.And third, it allows the basicprocess tocontinue
without creating a stylistically unacceptable whole-tonescale in the
upper voice,which wouldhave beenthe casehad the suspension
technique beencontinued see Example 67!.
_ whole-tone
scale.
fs 92 ,"" _ -I
fs 9. 9.
EXAMPLE 67
But these ends could have been achieved with other means. It is
to meaningand contentthat we must turn if the reasons forthis
change are to befully comprehended. By emphasizing the uniform-
ity of pitch progressionin theupper voiceand ofharmonic sequence
progression throughout, chromaticism and regularity intensifythe
inner impulsionand drive of the series, weakening the listeners
sense ofcontrol, hisability to envisage thepoint at which theseries
will break and articulationwill take place. While in processthe
chromaticism and potential uniformityof chord progression make
the passageseem almostformless and,in terms of more remote
tendencies, incomprehensible.
Yet thepassage really
is notas uniformas hasbeen impliedin the
foregoing discussion. Though theouter voices are uniformand regu-
lar, the inner voicemoves ina somewhatdeceptive manner. The
irregularity of this voice not only produces acertain amountof
variety, thoughthis is minimized bythe morepatent uniformityof
the outervoices, but,by effectingchanges anddeceptive progres-
sions in the harmonicprogression, italso intensifiesthe listeners
sense ofdoubt andinsecurity. Forthe listeneris now able to en-
visage neitherthe terminal point of the seriesnor even the next
term of the series. This is illustrated in Example 68, in which the
sequence, together
with its presumed line
of harmonicroot progres-
sion, isschematized as
it would have beenhad it been completely
regular.
In this version ofthe sequence the rootsof the chords, beginning
in measure 6, areall a fifth apart.From measure 7 throughmeasure
9 each chord is the dominant of the one which follows it, so that the
progression is,
in a sense, completely
uniform evenbetween chords;
that is, as A is to B, so B is to C, and so forth. Between measures
there isprecise equivalence
of process.
Here eachsuccessive chord
_ I A !.".|. " I5 _
-0- |64-ng.1|
: :_ 3_ Q-
s' .C.'H|-#|._}hi'u_
Implied root
progressions
EXAMPLE 68
of the sequence is
predictable, butthe terminalpoint of the series
is not.
In Haydnsvariant ofthis regular,uniform progression,measures
8 and9 containirregular, deceptiveresolutions Example 69!. Be-
cause of its place in the sequence, thediminished triad in the
beginning ofmeasure 8sounds asthough it were thedominant of
A-Hat; butthis chordmoves deceptively to F minor. Sincethe am-
biguous diminished triad hasalready beenresolved intwo different
ways, thelistener isahnost completelyundecided whenit appears
again in measure 9whether it will move to G-flat major in the
manner ofmeasure 7!,to E-flat minor in the mannerof meas1ue
8!,
or, perhaps,even toC minor.In short,he isnow in doubt notonly
as to the terminalpoint of the seriesbut alsoas to the following
term of the series.
ll7*H;bT Tb; bf 5 T
i sa' - . *P* 9.
EXAMPLE 69
The insistencein the foregoing discussion

upon the importance
of uniformity and ambiguityin arousinguncertainty andanxiety
does notimply that this passageis chaoticor unstructured.If the
main outlines Example 70! of the series areabstracted from
the vitally important ornaments
which tendto obscurethem, it be-
-:,,., g 9 ,|1;1v92|UY4tl;=il
1 i:=.- 4 -fy* ;H g iw--t- _
55 5 a u ' |ii!ii - ' li
EXAMPLE 70
comes apparent
that theseries follows
a definiteover-all plan,
in
which the outer voices, moving in parallel tenths, delineate a de-
scending scalein the key of E-Hat major and minor!. But this
structure becomes clear and is understoodonly after it has been
completed andthe serieshas beenterminated. Onlywhen t.hese-
quence istimeless inmemory canthe relationshipof its parts toone
another and to the total series be comprehended.
The terminal part of this series,its return to more decisive, clear-
cut shapetakes place in measure10, where
the uppervoice changes
the directionof its motion fromdescending toascending, and
the
lower voice moves by skips of a fourth from F to B-flat to E-flat,
creating an incisive accent and motion in the bass.
When measure 11 isreached Example
71!, the listener is,so to
speak, outof thewoods; heknows where
he isgoing andenvisages
:ill _ V- :_--~ -I1----1--ll---1---_1-

|E175 - E I' : ::
EXAMPLE 71
the goalwith clarityand certainty.He expectsa returnto thetonic,

which isnecessary fora feelingof completeness,and tomore distinct
and palpableshapes, inparticular tothe first theme ofthe move-
ment. Thoughexpectation isnow veryspecific, itis not to be ful-
filled straightaway.
There isa waitingperiod, adelay ondominant
harmony, inwhich themind isable toorient itself,evaluating what
has takenplace andadjusting itselfto whatis now impinging.
M ininuzl Diferenoes
Uniformity isa mentalfact, nota physicalone. Notall differentia-
tion necessarily creates an impression of good shape. The differ-
entiation mustbe decisiveenough andsalient enough,relative to
the particularcontext inwhich it appears, toplay an appreciable
role in articulating shape. Some instancesof subliminal differentia-
tion have already been cited, e.g., in connection with the Bruckner
and Debussyexamples. Thepresent discussion
will, for the most
part, be concerned with certain aspectsof harmonic differentiation,
with theuniformity thatarises through
the accretionof smalldiffer-
ences.
Harmonic uniformity may arise because the difference between
harmonic terms
is notgreat enough
to createa senseof progression.
That is, a change may appear to be a variation of an already pre-
sented termrather thana motion to a new term. There may be
alteration, but not succession.
Although thecriteria determining what constitutes harmonic pro-
gression-unambiguous chord succession-become established as
part of the listenersstylistic responses,the stylisticnorms ofwell-
articulated succession are themselves, in part at least,products of
the needfor suflicientlymarked differentiation between harmonic
terms. It is partly because ofthe necessityfor appreciablediffer-
ences between successive harmonic terms thatthe harmonicpro-
gressions that are leastambiguous are those whichinvolve changes
of at least twotones between chords. Thisis madeamply clearby
an examination, for instance, of Pistons Table of Usual Root Pro-
gressions seep. 54!. The mostdecisive progression
of all in the
delineation of tonality, the one from the subdominant to the domi-
nant to the tonic, is so devised that the Hrst two triads have no tones
in common and that each of these triadshas only one tone in com-
mon withthe tonicto whichthey resolve.Other progressions might
be devisedinvolving complete tonal change,but all of them would
of necessity
employ one chord havingtwo tonesin commonwith the
tonic, thusweakening thetotal effectof progression.Further evi-
dence ofthe importanceof markedand appreciable pitch changes
between harmonic structures isfurnished bythe fact that when
chords have their root and third in common,they tendto serveas
substitutes for one another in terms of function. The submediant
VI! may serve as
a substitutefor thetonic I!, the supertonicII!
may serveas a substitute for the subdominant IV!, and the medi-
ant III! may serveas asubstitute forthe dominantV!.
Harmonic articulation
is not merely amatter ofpitch diEerences
between successive
chords. Thedegree ofarticulation alsodepends
upon the manner in which the chords are constructed whether
there is general uniformityof constructionor not!, the way in
which thevoices pitches!move fromchord tochord whetherby
conjunct or disjunct motion!, and the rhythmic articulation of the
harmonic changes
in question.For instance,a progression
from the
tonic chord to that built on the sixth degree of the scale can be
accomplished bymoving only one tone as in Ia in Example 72!, in
Which casethe feeling of progressionis minimal; or it can be
18Q Emotion and Meaning in Music
achievedusing disjunct motion in the outer voices as in Ib!, in
which case the sense of harmonic progression is substantially
b.
EXAMPLE 72
strengthened.
Likewise,variety of chordconstructionmay createa
sense of well-defined motion, as is the case in IIb as contrasted with
IIa. It is clear, from this point of view, that the apprehension of
harmonicprogressionis partly a function of melodic articulation.
Hencethe continuedprohibitionof parallelfifths and octaves,even
after linear independence wasno longera prime concernof musi-
cal style.
Decisive,clearlyarticulatedharmonicprogressionis not, of course,
necessarilya desideratum. Vagueand ambiguousprogressions, cre-
ated by minimal differentiation,may play an important role in
creating tensions,uncertainties,and expectation.
This is the casein the introductory measuresof the secondmove-
ment of Berlioz'sSymphonicFantastique Example 73!. Before
discussingthe nature and effect of minimal differentiationin this
passage, it shouldbe noted that theseare able to take their full
effect becausethe passage measures1-29! is quite uniform as
to pitch succession,
texture, and instrumentationand quite inde-
cisiveand ambiguousas to rhythm.The openingmeasuresof the
movementestablishthe generalpattern,which continuesthrough
the first twenty-nine measures:
Allegro
manontro
EXAMPLE 73
The triadic figure in the cellosand basses in the third and fourt_h
measures and the arpeggioin the fifth measurein the harp could
hardly be called well shaped sincethey involveno real melodic
progression. The continuous tremolo in the strings tendsto accentu-
ate thegeneral uniformityof instrumentalcolor andtexture. What
rhythmic articulation there is,is aproduct ofharmonic change; and,
since, aswe shallsee, thisis minimal,so isthe feelingof rhythmic
accentuation. Only toward theend ofthe passage where thetriadic
figure recursat two measure intervals and the harmonic changes,
though stillambiguous, are more markeddoes thereseem tobe an
intensification ofrhythm, creatinga feeling of increased excite-
ment, quickening tempo, andthe approach of a resolution toclarity
and certainty.
The harmonicambiguity ofthis passage
is theproduct ofminimal
differentiation between successive vertical structures. This is made
very clearby theabsuact ofthe harmonicmotion ofthe firstthirty-
three measures of the introduction Example74!. Observe,for in-
Jlsl measure
r _ @n 6
~s= T T?'IFa"{i ? !` f3~' i
' `Q'1 _1 . b '. uf.
*f Q.
MH _ -_ TofY EIS Y
EXAMPLE 74
stance, ! that upto measure13 onlyone toneis changedin each

successive harmony; ! that the slightdisjunct motionin the lower
voice the skips fromA to F to A and so forth!, though theydo
articulate themotion somewhat, are minimizedby the static quality
of thebasic bass line, whichmoves onlyfrom Ato A-sharp;and !
that stylisticallythe harmonicrelationships established by this par-
ticular seriesgive riseto no clear-cut probabilityrelationships, pro-
vide nobasis forenvisaging the nature ofthe nextterm in the series.
Even the motion from measure 12to measure13, thefirst stylistically
unambiguous progression,
involves aslittle motion and changeas
possible. Only
after measure
19 doesthe degree of harmonicchange
increase, i.e.,two tonesare changedin eachsucceeding chord and
once frommeasures 21 to 22! three. Butthese progressions, though
they aremore marked and createa sense of increased motion, areno
less ambiguous. For not only doesthis seriesfail to establish stylis-
tically unambiguousharmonic relationships but just at this point
other aspects of the musical structurebecome increasingly uniform
see brackets in Example74!. Thatis, thechords ingeneral move in
parallel motion;the outer voices movechromatically in octaves,
tending toobscure what inner articulationthere is;and themeasures
are motivicallyand rhythmicallyprecisely equivalent to oneanother.
Though thelistener isnot able to envisagethe nextterm in the
series withany greaterspeciBcity thanbefore, theincreased har-
monic motionin thesemeasures measures 20-26! doesperform an
important function.It creates a feeling of intensified activity, a
sense ofthe approach to, andimpending arrivalof, a decisive point
of structuralarticulation. Thisfeeling isheightened inseveral ways.
First of all, moremarked andnoticeable harmonic changes together
with moreregular motivicrepetition bringabout anincreased sense
of accentuationfrom measure19 on.Both because of the general
uniformity of the passageand becauseof the listeners awareness
that this is not the real substance of the movement, this repetition
of the motive inthe bassgives riseto thoseexpectations of
change
which wehave called saturation. The gradually growingcrescendo
also intensifiesthe sensethat important musical events,the arrival
at regular progressions andwell-shaped themes,are in the offing.
These expectationsreceive finalconErmation just
at theend ofthese
measures, wherethe entrance of the Woodwinds measure28! acts
as a sign of the comingchange. Butit is only when the six-four
chord so striking because of its brilliantly majorsound, itsopen
spacing, andits full orchestral setting!arrives thatthe listeneris
really sure of his orientation and certain that his expectations will
not be disappointed.
Before concludingthis discussionof minimal differences, it should
be pointedout that the distinctionbetween uniformityand differ-
entiation is not alwaysas markedand obviousas it has beenin most
of the examples citedabove. Uncertaintymay arisebecause the
listener is unable to decide whether the stimulus series is thematic
or not. For instance a sound term that is neither markedly uniform
nor clearlyarticulated maymake thelistener uncertainas to how
to interpretthe series
as awhole. Hemay notbe surewhether heis
hearing thereal themeor anintroduction involving
thematic antici-
pation. _
The openingmeasures ofthe secondmovement ofSchuberts
Seventh Symphony in C Major Example75! provide an instance
of this equivocal typeof articulation.Here the upper voicesare,
generally speaking,
quite uniform
and homogeneous.
There islittle
melodic motionand whatthere isnot onlymoves bysemitones but
Andante con
moto.
_ EXAMPLE
75 l l
is confined to the inner voices where its effect is minimized. Al-
though stylisticallythe harmonic progressions are unambiguous,
there is only a limited feelingof harmonicprogression because of
the minimal linear changesand becausethe changesconsist only
of an alteration oftonic and dominant harmony.What rhythmic
articulation thereis inthe uppervoices isproduced bythe harmonic
fluctuations, andthese couldhardly be said to produce astrongly
articulated rhythmic structure.
Taken by themselves these aspects ofthe stimulusseries would
probably leadthe listenerto concludethat this was definitelyan
introduction, an accompaniment figure over which a melody was
still to appear. Butthe presence of a rather well-shaped
melody in
the bass precludes such
a definite,unequivocal interpretation.
Rhyth-
mically this lower voice is very distinctly articulated. And even
though the pitch successions
seem somewhat
static becausethe
melody continually moves through the tonic triad, the passage asa
whole might seem thematic,were it not for the listeners awareness
that in music of this style the melody is usually,though not in-
variably, in one of the upper voices.
The equivocalbalance betweenthe forces of uniformity and
those ofdifferentiation, togetherwith the listeners stylisticexperi-
ence, createuncertainty asto whetherthe seriesis to be understood
as anintroductory prestateof thetheme orwhether itis thetheme
itself. Theexpectations of
clarification towhich thisambiguity gives
rise are satisfied in measure 8, where the entrance of what is ob-
viously thereal themeenables thelistener tounderstand, inretro-
spect, thesignificance of
the openingmeasures. He now knowsthat
the passage was anintroduction inwhich somefeatures ofthe main
melody wereanticipated. Thuswhile the hypothetical meaning of
the passage is ambiguous,
its evidentmeaning isclear andunequivo-
cal.
The equivocalcharacter ofsuch passagesis perhapsbest demon-
strated byshowing thatother equallyambivalent series
may inother
cases bethe real thematic substanceof a piece. The opening meas-
ures of Mozarts Piano Concerto in D Minor K. 466!, for instance,
are justas doubtfulin characteras thosein the second movement
of Schuberts Seventh Symphony. Rhythmically and melodically
they aremore uniform,more like an introductionor even like an
accompaniment over which a melody ortheme mightbe expected
to appear Example 76!.Yet, asthe passage gradually evolves and
the seriesbecomes progressively more differentiated,the listener
realizes thatthis isthe mainmaterial ofthe openingsection ofthe
Allegro
e '-__ i::==;22=EL==;'=:=="..::
=LLi{FTJIL _T@Tj=E 5_*=_ E' 1 g _
92/ 92_/
EXAMPLE 76
movement. Inshort, thoughthe hypotheticalmeaning ofthe open-

ing of the MozartConcerto isin a very generalway similarto that
of the Schubert movement,
the evidentmeaning isvery different.
The mental demand for good shapeis, of course, conditionedand
modiiied by the stylistic experience ofthe listener and his awareness
of the kind of music towhich he is listening.For instance,while
listening tothe development
section ofthe HaydnSonata discussed
above or the introduction to the second movement of the Berlioz
Symphony, thepracticed listeneris awarethat thesepassages are
not aswell shapedas theyshould berelative tothe stylisticnorm-
that theyare not,in a sense, the main event.
But in other musicalstylistic situations
his attitudetoward weakly
defined shape may bequite different,and thisdifference inattitude
will modifyhis responses. For example,the free,rhapsodic, toccata-
like piecesof the baroque period,though perhapsexhibiting in-
telligible shapewhen viewedas awhole after the piecehas been
concluded!, often appear tobe diffuseand amorphous when viewed
in terms of their parts and their succession. Though this lack of
well-defined, articulate patterns doesarouse thedesire for, and
expectation of,more distinctand substantialshapes, thepracticed
listener whounderstands the nature ofsuch freeand effusivepieces
tempers hispsychological desire for bettershape withthe knowledge
that such preludes will eventually leadeither to more defined
shapes withinthe pieceor to another piecewith a more palpable
pattern. Sucha listener,following musicalprogress as best hecan,
awaits theoutcome, ifnot with perfect equanimity, at leastwithout
the intense impatience andanxiety that similar passagesmight
arouse inother stylisticcontexts.
Texture
Texture hasto do with the ways in which themind groupscon-
current musicalstimuli into simultaneous Hgures, a figure and ac-
companiment ground!, and soforth. Like other musicalprocesses
textural organization,
or the lack of it, maygive riseto expectation.
According toKoffka, if conditions aresuch asto producesegre-
gation oflarger andsmaller units,the smallerwill, ceterisparibus,
become figure;the largerground. 6Does thismean thatin a poly-
phonic piecesmaller fragmentswill necessarilybe perceived as
Bgures andthe longerparts asground? Forinstance, willa well-
articulated themepresented inaugmentation become a ground on
which shorterthough equallywell-articulated motives will appear?
The answeris t.hatsuch adistribution oftexture isnot necessarily
apprehended interms of a figure-ground distribution but can quite
readily be perceived as the co-existence of several independent,
well-articulated figures.
Koffka tendsto emphasizethe necessityof a figure-ground dis-
tribution because he is primarily concerned
with visual experience.
Figures seen in visualspace always appear tobe projectedagainst
or framedby a ground ofsome sortjust asthese wordsare seenon
a ground,the whitepaper onwhich theyare printed.It is difficult,
if not impossible, evento imagine a visual figure without also
imagining themore continuous, homogeneous ground against which
it appears.But in aural space, in music,there isno givenground;
there is no necessary, continuous stimulation, against which all
figures mustbe perceived.The only thing that is continuousin
aural experienceis unorganized,timeless silence-theabsence of
any stimulationwhatsoever."
Due to the absenceof a necessary, givenground in aural experi-
ence, themind ofthe listeneris ableto organizethe datapresented
to it by thesenses in several different
ways. Themusical Held can be
perceived as containing: ! a singlefigure withoutany groundat
all, as, for instance,in a piece for solo flute; ! several figures
without anyground, asin a polyphonic composition in which the
several partsare clearly segregated andare equally, or almost
equally, wellshaped; ! one or sometimes more than oneHgure
accompanied by a ground, as in a typical homophonic texture of
the eighteenth or nineteenthcenturies; ! a groundalone, asin the
introduction toa musicalwork-a song, for instance-where the
melody orfigure isobviously stillto come;or ! a superimposition
of small motives which are similar but not exactly alike and which
have littlereal independence
of motion,as inso-called heterophonic
textures.
Which of these various organizations or combinations of them
the mindimposes upon
the sensory
materials presented
to it depends
upon thepsychological demand
for goodshape uponthe operation
of the law of Priignanz! and upon the attitude and expectations of
the experiencedand practiced listener.
The psychologicaldemand forgood shapenot only impels the
mind to articulate the stimuli presentedto it into well-shaped
figures andpatterns butalso makesthe total set of relationships
resulting from such articulation as simple and distinct as possible.
For instance, if the over-all articulation is simpler when a piece for
a single instrument isunderstood asimplying severallines or
voices, thenthis modeof organizationis theone thatwill probably
appear; whileif the Hnal resultof articulationis moredistinct and
the patternsperceived ofbetter shapewhen thestimuli areappre-
hended in terms of a Hgure-grounddistribution, then this mode of
organization willprobably emerge.
Furthermore, whichever textural organization
emerges, itwill be
made toappear assimple anddistinct aspossible. Themind will
tend toimprove theover-all articulation
even ifthis meansweaken-
ing someof theindividual shapes. For example,if the Held is most
easily perceivedin terms of a Hgure-ground distribution in which
one partis wellshaped and the othertends towarduhiformity, then
the uniHedpart of the Heldwill seem as uniformand ashomoge-
neous as prevailing conditions allow. If, however, it is easiest to
apprehend thestimulus Heldin termsof a single shape
or in terms
of severalmore or less equalshapes, thenthe mind will tend to
improve boththe individualshapes andthe relationshipsbetween
shapes asmuch aspossible.
The mind of the listener which produces suchorganization is not
a kind of neutral, disinterested tabula rasa. The stimulus Held is
organized, partlyat least, on the basis of past experience-the
learned habitsof discriminationand perception of thelistener. Such
learning, bydirecting thelisteners attentionto certainparts of the
total Held, conditions what is looked for and expected and hence
modiHes whatis perceived;where the center of our interest lies
there, ceteris
paribus, aHgure islikely toarise. 8Learning also
tends
to influencethe qualityof the Hgures whicharise, because atten-
tion, adding energy to the particular Held part, will increase its
articulation, if it is not articulated as well as it might be. In other
words, the practiced listener has learned to direct his attention in
particular ways,depending uponthe stylisticcircumstances; hence
he not only tendsto improve articulation in general buttends to
favor certaintypes of organizations over
others in a given set of
stylistic circumstances.1Thus, in perceiving music which he sup-
poses tobe polyphonic,the practiced listener will tend to emphasize
the equalityof moreor lessequally well-articulated
Hgures; while
in attending to music which is presumed tobe homophonic,the
listener will favor one strand of the texture over the others, and
these latteraspects ofthe texturewill appear more uniformthan
might otherwisebe the case.
Whether agiven texturalorganization isfelt to be satisfactory
depends uponthe speciHcdisposition ofthe variousparts within
the totalperceptual field,
the normsestablished within the particular
work, andthe listenersstylistic habitsof perception.
There aretextures inwhich theorganization of
the fieldinto
co-existing figures,
into figure and ground,or into some othertype
of relationshipis clear, complete, andnormative withinthe style
and in which changesin texture are immediatelyunderstandable.
Such textures will not in themselvesbecome thebasis forexpecta-
tion.
There are also textural situations in which the nature and organ-
ization ofthetexture, though clear andimmediately understandable,
is nevertheless felt to be incomplete.By way of illustration, three
different incompletetextural situations may bedistinguished.
First, textureplays a part in determining thelisteners senseof
formal completeness. For, sincethe law of return see p. 151! ap-
plies tothe organization of the total musicalstructure and,at times,
even tothe recurrence of separate aspects of
that organizationsuch
as melody,tonality, ortexture!, a change fromone typeof textural
organization toanother orchanges within one genushomophonic,
polyphonic, etc.!will activate expectations of return to previously
established modes of organization.Even thisgeneral statement is
subject toqualification. Forthe expectation of returndepends upon
the listenersunderstanding of the passagein question.For instance,
in an opera or oratorio the introduction of new charactersin a scene
may bringabout changes in texture-what was ahomophonic aria
may becomea polyphonicquartet; butbecause these changes are
understood interms of the stage action, theywill probably not
arouse expectationsof returnto theoriginal textureof the number.
Texture does not asa ruleact asan independent variable. Changes
in texture are usuallymade in conjunction withchanges inother
aspects ofthe musicalorganization. And it is not onlyan established
texture which will be expected to return but rather the whole
complex ofthe stimulusseries, ofwhich texturewas butone aspect.
However, melody,
tonality, instrumentation,
and soforth may vary
indeHnitely while the basic textural organization remains constant.
Because ofthe desire to organize the textural Held as it was
organized whenan establishedmelodic harmonic rhythmic pattern
Hrst appeared,expectation will be particularly active and intense in
cases inwhich thesalient shaping forces ofa workremain constant
while texturechanges, e.g., where atheme ormelody already estab-
lished asa normin one textural organization is repeatedor recurs
in different textural setting.Moreover, suchdeviant texturesoften
tend to create uncertainty and tensionin the mind of the listener.
This isparticularly evident in caseswhere atheme whichoriginally
appeared as part of a homophonictexture issubsequently used in
conjunction eitherwith other equally well-articulatedshapes or
in imitative counterpoint with itself. Here there is often a conHict,
though onlya minorone, between the subjectiveorganization which
the listenerattempts toimpose and the objectivefacts ofthe textural
distribution. In such casesthe introduction of additional Hgures
which are well shaped appears asan intrusion, as a disturbance of
what was supposed tobe a homophonic texture.
Second, asense ofincompleteness of texture mayarise as the
result ofabnormally wide distances betweenthe partsof thetextural
Held. That is, whether the Held is organizedinto equally well-
articulated Hgures or into a Hgure-ground distribution, theseveral
parts of the texturemay be so widely separated inmusical space
that theyare expected to cometogether orto beHlled in by other
stimuli. Such abnormally widespacings may be consideredas a
special caseof structuralgaps. InBergs LyricSuite Example77!
the very wide spacingbetween the Hrst violin and the cello is
Hrst graduallyHlled in by the other instruments, and then all the
parts movetoward acommon meeting point in the middle range.
Notice, incidentally,
that thetendency ofthe outervoices tomove is
not a result ofdissonance-at least,not at Hrst. Thefeeling of ten-
sion toward a middle range is almost entirely textural. The entire
Hfth movement of theLyric Suitemay betaken asan illustrationof
the fact that thoughgenerally speakingtexture doesnot act as an
independent variable, it does at times become a very important
factor in creating motion
and in shaping musical
experience.
The Weakening
of Shape 191
is understood
as simply
constituting
change
a materials,
of begin-
the
ning of a new section.
In afugue, for
example,
single
a type
of texture
generally persists
throughout the
entire piece,
though there
will bevariations within
this general
genre. Because continuity of
texture is
expected ain
fugue, the
significanceaofdecisivetextural change
may notbe
immediately apparent.
If themeaning such
of change
a not
is clear,
doubts may
arise inthe listeners
mind about the composers
inten-
tions andabout the
relevance ofhis ownexpectations, which were
brought into
play partlyon thebasis ofhis beliefas tothe nature
of thework being
heard. other
In words,
the change
createsfeel-
a
ing ofuncertainty because
it weakens
the listeners
ability toen-
visage thefuture courseof the music.
The lastmovement ofHandels ConcertoGrosso No.2 furnishes
an excellent
example of
such aninterruption of
a texturewhich is
expected be
to continuous.
The movementbegins like
a normal
fugue Example
79!, andthe listener
expects the
texture to
con-
Allegro
1 manon troppo
___- ...-ull YAiiiil1i1Yl llg4 lllln ||lll___ L_:-I
f'_i1l_2i`l1 it
ffl
| ;; _!lLz'I_!1.| ral I1__Y'll |14 1
-gl -_ '-- in I Lal
921'l1 l i
EXAMPLE 79
tinue with the customary modifications and variations until the

conclusion isreached. However,at measure27 Example 80! a
=. J" i;l l _mitt-Q
f.5EiiE1:1::===E
. -. ..1;- .
=~=-======Q:===:===55555
= :===- _
-/ 1/
E
EXAMPLE 80
radical changeof texturetakes place.The organization becomes as

homophonic as possible. Because these changes
constitute adramatic
deviation fromstylistic norms,they raisedoubts andanxieties both
as to the significanceof the textural discontinuityand as to the
relationship betweenthe partsof the movement. Only gradually as
the accompaniment becomes moreactive melodically,as though
influenced bythe fugalsections withwhich it alternates, doesthe
listener beginto understandthat thehomophonic texturewill even-
tually becomepart of the generalpolyphony ofthe fugue.The real
meaning ofthe texturaldisturbance becomesapparent inmeasures
66 Example81!, 74,and 82,where thetwo melodicideas arecom-
bined.
1 Q mm iff-!
~..5 ==EEEEE fi` 'iiiii
Il I
EXAMPLE 81
The powerfuleffect ofthis changeof textureis particularlyinter-

esting since,in a sense, thehomophonic texture is simplerthan the
polyphonic one;that is,it involvesless expenditure of mentaleffort
and would probably in other circumstances seem ratherrelaxed
and calm.The crucialfactor creatingtension inthis exampleis the
1isteners difficulty
in understandingthe significance of the textural
changes. This is shownby the fact that homophony coming at the
conclusion of a fugue, where it is immediately understandable,
creates nodoubt or uncertainty in the mind of the listener.
The precedingdiscussion was
concerned withtextural organiza-
tions which were, taken separately and individually, unambiguous.
Ambiguity arose, if at all, becausethe relationshipbetween succes-
sive textureseach ofwhich wasclear initself !was notimmediately
understandable. Only where thechange oftexture involvedno con-
comitant changes in the other aspectsof the stimulus seriescould
the texturalorganization itself be saidto havebeen ambiguous.
Ambiguity may,however, arisebecause theorganization ofthe
field is itself unclear.This mayoccur eitherbecause the progressive
weakening ofthe figuresobscures previously
a clear
organization or
because thenumber, diversity,and placementof the several parts
of the texture obscure the definition and articulation of the individ-
ual constituents of the texture.
If the number anddiversity of the individual elements ina texture
are so great as to obscure one another or if their placement is such
that they cannot beperceived asclearly separatedentities, then
ambiguity mayarise. But, in a sense, itis notreally thetexture which
is ambiguous,rather thenumber of elements and their placement
make the individual shapesappear unclearand ambiguous.Such
textures arenot common.For the most partthey occurin program
music wherethe resultantobscurity hasa more or less obvious
designative meaning and createslittle uncertainty or tension.An
instance ofthis type of ambiguity occurs inStrauss AlsoSprcwh
Zarathustra, where a fuguewritten in the lowestregister creates a
muddy, indecipherable texture, whichsupposedly depicts academic
obscurity andbefuddlement.
A uniform texture doesnot necessarily appear ambiguous. It may
simply be understood asconstituting a ground for some sortof
theme whichis still to appear.For this reason uniformity of texture
will seem ambiguous onlyunder specialconditions: i.e.,where it
persists fora seeminglyintolerable lengthof time and hencegives
rise to saturation orwhere apreviously clearorganization becomes
progressively weakened either because the figureslose theirdefini-
tion or because theground becomes more distinctlyarticulated or
both.
The openingmeasures of Beethovens NinthSymphony furnish a
striking examplein which a progressiveweakening oftexture, to-
gether with both harmonic and motivic incompleteness andambigu-
ity, create powerful expectationswhose inhibition and ultimate
resolution intoa clearlydefined theme
produce apowerful affective
experience. In the sixteen measures of introduction Example 82!
the structuralgaps arenot onlymelodic, theyare harmonicas well.
For within the cultural context in which this work is heard, the
complete triadis thenorm, andthe openfifths presented
throughout
the introduction are felt to involve incompleteness. Indeed, this
Allegro, ma
non troppo,
un pocomaestoso.
s
-- --
EXAMPLE 82
incompleteness is signified by the very fact that such Hfthsare
commonly referred to as empty. Moreover, the harmonicincom-
pleteness creates
ambiguity asto mode;the listeneris in doubt asto
what the completion willbe. Thus,the powerfuleffect of the first
theme is,at leastin part,due tothe factthat it presents acompleted
triad andin thisway removes the previousambiguity andcompletes
what wasclearly incomplete.
But the forceful impression of the first theme is even more the
product ofthe factthat it is a distinct, substantial
shape. Thewhole
introduction isbuilt upon a motive which is obviously psycholog-
ically unsatisfactory. It is not merelyincomplete, itlacks direction
and coherence. The listenersenses that it must have meaning,but
he hasno ideaof what that meaningis sincehe can envisage its
conscquents only in the vaguest terms.Because the motive estab-
lishes noprogressive motion, either melodically
or harmonically,the
listener haslittle feeling for where this passageis leading; he is
merely awarethat it seems topresage some momentous, fateful
event. l
In a sense, theonly really satisfactory, unambiguous aspect of

measures 1-12 is therelationship betweenthe figureand theground.
All the factors makingfor a clearly articulatedrelationship are
present: notonly isthe groundmuch moreuniform thanthe figure,
but it begins beforethe Hgureis introduced,thus surroundingit in
a temporal sense. But this clarity also givesway to ambiguity
Example 83!. For as the point of culmination approaches, the
figure becomes progressively weaker-more like the ground. First,
the rhythmicarticulation ofthe motiveis lostin therising crescendo
of volumeand thepsychological accelerando measures 13-14, Ex-
ample 83!.Then itsintervallic definitionbecomes obscured in the
general octavemotion; and,in the end, whenthe distinction between
l3 _92
_'_l 1wl- Z-1 l
-l _11;l',i_1jljj
~ I_ ' r'l _____ _
;_j111111
Fl
i
1
== =E "f= "=E= -==E?E
PP FOR.
'` '"EEEEEEEEFEEQEQEEEEEE
Y
5
,- . ., . r :K `' :En-'!_
:L is 8r .r rrrrrzr p _ . FFFEEE
'_
EXAMPLE 83
figure and ground is obliterated altogether,the texture becomes
completely ambiguous and the listener isuncertain asto what the
textural organization is.
The feelingthat a momentous event is impendingis heightened
by the crescendo which begins inmeasure 11 and bythe increased
psychological tempo at that point. For the figureinstead ofbeing
repeated everytwo beatsis now repeated onevery beat.And in
measure 13 it occurs at a still shorter time interval. It should also be
noted thatwhen themotive isstated insixteenth notesbeginning
in measure18!, themetric opposition
of twosagainst threes,
implicit
in the opening measures,
becomes actualized and tendsto increase
the generalambiguity ofthe wholetexture.
Although the entrance of the low D in the horns and bassoonin
measure 15is not strongly emphasized,not markedfor conscious-
ness, itplays animportant partin intensifyingexpectation atthe
very endof the passage. Since
it createsthe firstharmonic motion
in the whole introduction, the listener senses that this is it, that
at last he will know and understand what heretofore had been only
a powerfulpremonition.
The sense of therelentless power of inexorable
fate whichcharac-
terizes themain themeof this first movement is a result notonly of
the elemental force of the theme itself and of the ambiguities and
expectations excitedby the introduction but also of the particular
manner in which the introduction leads into the theme. Unlike most
introductions seeExample 74!this onedoes notconclude witha
dominant preparation,
a waiting period, in which the listener is
given an opportunity to orient himself to what has passedand pre-
pare for what is to come.Instead the music moveson without pause,
without pity, to its stark andawful declaration.
As fliesto wantonboys arewe to the gods.
They kill us for their sport.
These compellinglines call attention to those inexplicableand

inescapable events
of our existence which the Greeksattributed to
fate, the Renaissance to fortune, and we ascribe to chance.
Chance isan uncompromisingfact which permeates allthe realms
of being: the physical,the biological,the social,the personal,and
the aesthetic.It is but chancethat there is a solar system;but chance
that life developed onthis planet; but the chance of a seriesof
fortuitous mutations that there is human life as we know it; but
chance thatany particularhuman consciousness comes intobeing;
and, onceliving, both past andpresent chance events continually
condition andmodify thecourse and tenor ofour lives.
Thus itis thatmusic, mirroringthe essentialshape andsubstance
of hmnanexperience, from time to time containssudden, shocking
clashes withunpredictable chance.Lesser composers tend toeschew
such harsh encounters with the unexpected,avoiding them by em-
ploying asingle-minded samenessof musicalmaterials orminimiz-
ing them by makinga fetish of well-oiled,smooth transitions.
But
the great masters havefaced fate boldly, and capricious clashes
with chance are presentin much of their Hnest music.
The pedantshave piouslyattempted toexplain awaythe inex-
plicable inorder tomake theiranalyses jibe
with their mechanistic
misconceptions what
of constitutesthe basisfor musicalunity, logic,
and inevitability. But chance will not be denied. And while such
encounters mustof coursebe possiblewithin the style employed,
we mustaccept thefact that one passagemay follow another not
because ofinescapable inner
necessity butmerely because chance
will have it so.
Nor do such seemingly
fortuitous encountersweaken ourfeeling
of musicalcoherence and
credibility. Rather,
since suchthings have
been experiencedin life as well as in art, these encounters often
strengthen ourimpression ofthe truth and realityof the aesthetic
experience. Initiallythey must be understoodas chance.But as
they becomepart of our experienceof the musical work,causing
us to revise ouropinion of what haspassed andconditioning our
expectations of
what isstill to come, theirsignificance emerges
and
we discerntheir influenceupon the shape andthe courseof the
musical process.This does not, however, mean that the tendencies
developed before chance intervened are obliterated and canceled
out. They persist; but their course and modeof fulfilment are neces-
sarily conditionedand modiHedby the unexpected intrusionof
chance, Itis pointless
to attemptwith tortuousargument toanalyze
away such unpredictable encounters.They can be truly understood
only by boldly facing the fundamental fact of chance.
VI
The Evidence: Deviation in Performance
and TonalUrganization
The Nature of the Evidence

The present
study seeks
to establishand explain
the generalcauses
and conditionsfor the affective aestheticresponse tomusic. It is thus
relevant that the basichypothesis adoptedis applicable to the music
of different cultures and various cultural levels. The evidence pre-
sented in support of the hypothesishas beenchosen frommany
different kindsof music: from folk music, primitivemusic, jazz,
oriental music, as well as from the music of Western Europe. The
very diversityof the musical stylesystems that
can beused asevi-
dence increasesthe probability that the hypothesis issound.
The evidence to be considered consists of: ! statements of
composers, performers,
theoreticians, andcompetent criticswhich
relate specific musical practices to affect or to aesthetic pleasure;
! statements which relate a specific musical passageto affect or
to aestheticpleasure; ! musical processes in Western music which
are by common consent
considered tobe affective,e.g., chromati-
cism; ! musical examples,taken from the familiar style of Western
music, where common habit responses will allow the assumption of
common understandingand interpretation.
All four types of evidence will be used in the case of the familiar
Western music. However, because it is impossible to assume a
common responseand because of the danger of reading Western
meanings and expectations into passages wherethey are not rele-
vant, only the Hrst and second type of evidence will be used in
197
discussing non-Western
music. In dealing withoriental, primitive,
folk music, or Western music before the seventeenth century, we
will rely upon thestated responses
of thosefamiliar with and sensi-
tive to the nuancesand tendencies
of the particular stylesin ques-
tion.
The statements of composers,performers, theoreticians,
and com-
petent critics are not, however, alwaysprecise andunequivocal.
Often theyrequire interpretationand explication.Before introduc-
ing theevidence itself,several problems
relating toits meaningand
interpretation must be considered.
Both the theoretical and critical literature about music abounds
with references to its emotional character. Innumerable works or
parts ofworks arecited asemotionally moving,exciting, and so
forth. Yet simply labelinga passage,a mode,or a riiga as having
emotional qualitytells usnothing aboutthe basis
for sucha response.
What in the passage or modeis aHective?Most references of this
sort do not refer to affect as we have beenusing the term; they
refer insteadto mood and associativeresponses. They are based
upon aresponse to those aspectsof musicalmaterials, such
as range,
tempo, dynamiclevel, which do not change inthe courseof an
excerpt. To be of value asevidence in the presentstudy, statements
about affectivitymust connect
emotion oraesthetic satisfaction
with
a specific musical processin the passage cited-a process in which
the relationship of deviation and norms to affective aesthetic re-
sponses canbe examinedand discussed.
It was argued in chapter i that the same stimulusconditions give
rise to both theaffective andthe aestheticintellectual response
to
music. Which one emergesdepends not upon the musical organiza-
tion whichbrings itinto beingbut uponthe disposition
and training
of the listener. Self-conscious minds tend to rationalize musical
processes by
bringing their observations tothe level of conscious
awareness. Theytend, that is, to respond on the aesthetic intellec-
tual level.Because most
of the evidence presented
here comes
from
persons with a technical interest in music-from persons who tend
to rationalize the musical process seep. 40!-many of their state-
ments connectthe musical processes withaesthetic satisfactionand
with affectonly byimplication.
Deviation in Performance and
Tonal Organization 199
Even thewritings ofcomposers and performers are
not alwaysas
explicit asone mightwish. Forthe composerand performertend to
view the musical processin a special way. For them aesthetic satis-
faction comes from theovercoming ofdifficulties seep. 69!. The
difficulties overcomeare the norms whichHx theregular andex-
pected. Thus
when reference
is made
to theexcitement and
pleasure
engendered by
technical difficulties,
the process
described is
usually
one of deviation from the normal and regular. This processhas, in
turn, beenviewed asan aspect of play,and thenumber ofreferences
to theplay factorin the performance and creation ofmusic isvery
striking.
Performance and Deviation

The musicalrelationships embodied in a score orhanded down
in anoral traditiondo notfix with rigid andinflexible precision
what
the performers actualization of the score or aural tradition is to be.
They areindications, moreor less specific, ofwhat the composer
intended and what tradition has established.The performer is not a
musical automaton
or akind of musico-mechanical medium
through
which a score or tradition is realized in sound. The performer is a
creator whobrings tolife, throughhis ownsensitivity offeeling and
imagination, therelationships presented in the musical scoreor
handed down in the aural tradition which he has learned.
The amount of freedom allowed to the performer in his creative
realization of a score or oral tradition varies from culture to culture
and within different epochsof the same culture.In some periods of
Western music composers have indicated in great detail exactly
how theywish their music tobe played,and theperformer isnot
supposed to add to or embellish with new patterns the notes that
the composerhas setdown. Thisdetailing of the scorereached
its most extreme stageat the end of the nineteenthcentury. Yet even
here, aswe shall see, theperformer doesplay a creative role.In
other stylesthe performeris given much morefreedom to add to and
embellish the schematic guide given by the composer or handed
down byaural tradition.This is the casein almostall non-Western
music, in folk music, and in much of the music of the West, e.g.,the
music ofthe seventeenth
and eighteenth centuries. Thus,
in a sense,
one mightsay thatqualitatively theperformers roleis alwaysthe
same; heis alwaysan active creator, shapingand moulding the
abstract schemefurnished him by the composer orby tradition.
Quantitatively his role varies. At times his task is limited to com-
municating themeaning latentin a relatively fixedset of musical
relationships; at
other times,
in othercultures, the
performer addsto,
alters, andmakes major modifications inthe materialswhich serve
hirn asa pointof departure.
Distinctions between deviations in pitch, rhythm, or tempo and
deviations which involve ornamental additions to the basic scheme
furnished bythe composeror by tradition involve differences in
degree ratherthan kind. The presentdiscussion arbitrarilylimits
deviations inperformance tothose whichinvolve slight modifica-
tions of the substantive
pattern furnishedby tradition or the com-
poser. Alterationsand additionsto that pattern will be discussed
under thetopic ofornamentation. Though such minordeviations in
pitch, rhythm,dynamics, andthe like exist in non-Western music,
they are more easilyobserved in Western music,where the scheme
of relationshipsis more definitely Hxedby traditions of intonation
and bythe scoreof thecomposer.
EXPRESSIVE PERFORMANCE IN THE WEST

Since itis impossibleto studybygone modes of musicalperform-
ance directly,evidence ofdeviations beforethe invention of the
phonograph must come fromthe writings of theoristsand critics.
These tendto dealwith deviationsin dynamics,tempo, andrhythm
explicitly andwith deviationsin pitch only byimplication.
C. P. E. Bach, for instance,maintains thatcertain purposeful
violations ofthe beatare often exceptionally beautiful.1 Leopold
Mozart notes that he who accompanies a true virtuoso must not
allow himself to be seduced intohesitating or hurrying by the
prolongations and anticipations ofthe notesthat the soloist knows
how to bring in so skillfullyand touchingly
.... 2 Muchthe same
relationship betweenmelody and accompaniment isimplied by
Chopin, whois reportedto have said thatthe singing hand may
deviate fromstrict time, but the accompanying hand must keep
O e. 3
Deviation in Performance and Tonal Organization 201
L. Mozart also prescribesa special mode of performance for
ghromatic notes,which, as we shall see, are themselves to be re-
garded as
deviants.
The vibrato also see p. 66! is essentially a deviation in pitch,
an oscillationabout abasic pitch.At times it is coupled withcre-
scendo and diminuendo effects. That the vibrato was considered
particularly expressive
may beseen fromCeminianfs discussion
of it:
When it [the vibrato] is long continued swellingthe soundby degrees,

drawing the bow nearerto the bridge, and ending verystrong it may
express majesty, dignity, etc. But making it shorter, lower and softer,
it may denote affliction,fear, etc.,.... And with regard to musical
performances, experience has shownthat the imagination ofthe heareris
in generalso muchat thedisposal ofthe master,that by the helpof varia-
tion, movements, intervals andmodulation hemay stampwhat impression
on themind hepleases.
These extraordinaryemotions areindeed most easily excitedwhen
accompanyd withwords; andI would besides advise as well the com-
poser asthe performer,who is ambitious toinspire hisaudience, tobe
first inspiredhimself; whichhe cannotfail to be if he chusesa work of
genius, ifhe makeshimself thoroughlyacquainted withall its beauties;
and if while his imagination iswarm andglowing hepours thesame ex-
alted spirit into his own performance!
The affectiveaesthetic valueof deviationsin the performance of
music is perhaps evenmore clearly illustrated by the criticisms
which chidethe performerfor merely playing thenotes or playing
mechanically
Employing moreaccurate techniques for the study of musical
performance, recentresearch hasshown that deviations from exact
pitch, tempoand rhythmare presentin mostmusical performances.
Carl Seashoreand his associates foundthat in contemporary per-
formance atleast, Theconventional musical score-the composers
documentation ofthe tonal sequences which he feelswill express
beauty, emotion,and meaning-is for the singer onlya schematic
reference aboutwhich he weaves, through continuous variations in
pitch, a nicely integrated melodic unit. 5
In part such deviationsmust be regarded asattempts to add
emphasis to the tendencies of tones whose motion is strongly di-
rected or to those which already function as deviants within the
tonal system
employed. Partly,
however, deviations
in pitch, rhythm,
and volumeappear tobe productsof the performers ownexpres-
sive intentions.
In music and speechpure tone, true pitch, exact intonation,perfect

harmony, rigid rhythm, eventouch and precise timeplay a relatively
small role.They aremainly pointsof orientationfor art and nature.The
unlimited resources for vocal and instrumental expression lie in artistic
deviation fromthe pure, the true, the exact,the perfect,the rigid, the
even, andthe precise.This deviationfrom theexact is,on the whole, the
medium forthe creationof the beautiful-for the conveying ofemotion.
The variation from the exact which is due to incapacity for rendering
the exactis, onthe whole,ugly. The artist whois to vary effectivelyfrom
the exact must know the exact and must have mastered its attainment
before hisemotion canexpress itselfadequately through
a sort of Hirta-
tion with it.
At Hrst glance Seashores viewpoint appearsto be quite similar

to the one adoptedin this study. Buton closerscrutiny several
im-
portant differencesare evident.
Although Seashore sees emotion and beauty notice that he as-
cribes thesame cause to both! arising fromdeviation, yethis basic
philosophical position
is in many respects diametrically opposedto
the one taken here. Seashore views the norms and deviants as abso-
lute and fixed. He believes thataesthetic qualitycan be quanti-
tatively measurediHowever, if one considersthe great variety of
style systemsand tonal organization which exist in various cultures
-tempered twelve-tonescales, tempered
five- or seven-tone scales,
and all the many untempered scales and modes-it seems very
evident that there are no absolute norms.
Moreover, although
we may know that a particulardeviation is
not accidental because it tends to be constantwithin a given
style!, the criterion of aesthetic deviation is itself culturally and
stylistically determined.
What is considered an
expressive deviation
in one style may be considered an abomination in another. The
wide pitch deviations in Hawaiian guitar playing, though perhaps
expressive tosome, aremost distasteful to those accustomed to the
more modest departures of European art music.
Many problems arise in connection with Seashores methodand
the datadeveloped byit. He himself admitsthat thereare devia-
tions whichresult fromcauses not relevant toexpressive deviation;
eg., the use of Pythagorean ornatural-scale intonation,the non-
linear relationship
of pitch and frequency,and therelation ofpitch
production tomotor behavior!How is one todetermine which
deviations are expressive and which are not? Only by a careful
study andanalysis ofthe generalplan of expression, both within
the stylein generaland withinthe particularpiece inquestion, and
by attemptingto correlateexpressive deviationwith the total affec-
tive aesthetic musical structure.
Because Seashore advances notheory andattempts noexplana-
tion of the relationshipbetween deviationand affectiveaesthetic
experience, hisviewpoint lacks substance andplausibility. He
demonstrates a correspondence but no causality. Furthermore, in
failing to explain oraccount forthe relationshipbetween deviation
and beauty in music, he alsofails to see thatdeviation in per-
formance isonly oneaspect ofwhat is actually amuch morecom-
prehensive and general principle.
EXPRESSIVE DEVIATION IN NON-WESTERN MUSIC

Such expressivedeviations areby no means confinedto the tradi-
tion of Western music.Although someof the variability in the
intonation of primitive singers may be attributed to the lack of in-
strumental accompaniment and some, where instruments are present,
to peculiaritiesof their intonation, notall of the variability need
necessarily be ascribed tolack of training .... It has definite
limits, and changes inintonation oftenhave an expressive func-
tion ....
Primitive singers
also employ
vibrato asa methodof pitch devia-
tion.
One of the characteristics of Chippewa singing observed during this

study is that a vibrato, or wavering tone,is especiallypleasing to the
singers. Thisis difl-lcultfor them to acquire and is considered asign of
musical proficiency. The vibratomay seemto indicatean uncertainsense
of tone, but the singer whouses it is ready to approvethe songwhen
sung withcorrect intonation. He declares,however, thatthis is not good
singing. 1
Expressive fluctuations
in intonationare likewisecommon infolk
singing. Theintonation ofa traditionalfolk singeris notnecessarily
worse than ours, but it is often more flexible. A tone is often taken
somewhat higher or lower than might be expected,for expressive
and ornamentalpurposes. 11 Bartok notesthat the deviations of
pitch in folk musicsince they show acertain system and aresub-
consciously intentional, must not be consideredfaulty, off-pitch
singing. Thisis the essential difference
between theaccidental off-
pitch singingof urbanamateurs and the self-assured,
self-conscious,
decided performance of peasantsingers. 11
Deviation alsoplays an important role in the performance of
oriental music, and there also it is considered to be affective and
aesthetically pleasing.
In his book, TenRules forPlaying theLute,
Wu Chn at the turn of the thirteenthcentury writesthat if one
just playsthe musicas it is written, one will not be able to express
the sentimentsof the composer. 13 And in the sixteenthcentury
Yang Piao-cheng stresses thenecessity forinterpretation, connecting
this with an understanding of the tendency ofthe music though it
is doubtful if he is using the word tendency in the same sense as
it has been usedin this study!. When one knowsits meaning,one
understands its tendency; when one understandsits tendency, one
may truly! understand the music. Though the music be tech-
nically wellexecuted, ifits tendencyis notunderstood, what benefit
shall it give? It is nothingbut a big noise,that availsnothing. 1*
In a collection of]avanese poetry, the Tjentin,the following com-
ment on the playingof the rebab appears:Every now and then
there wasa little deviation fromthe correctpitch soas to enhance
the charm of the music. 1
Ornamentation
In a sense theexpressive deviations
discussed above
are merely
unclassified forms of ornamentation. Certainly the distinction be-
tween the expressive deviationsmade by performers of all cultures
and thedevices which
the variouscultures systematize
as ornaments
is very diflicult to draw. Deviations in pitch which in Western music
would be considered asexpressive deviationsare, for instance, in
the music of India often classified as ornaments. Even within a
single culturethe samedevice mayat onetime beclassified asan
ornament andat anothertime may simply bepart of the style of
the performance or maybe incorporatedinto thebody of the com-
posers score.
For example,
the vibrato,which wasonce classed
as
an ornament,became anaspect oftraditional stringperformance;
while theappoggiatura, once
an ornament,became part of the com-
posers basic
plan. Inthe finalanalysis, whether
an expressive
de-
vice is classiiied as an ornament, as an expressive deviant,or as a
compositional techniquedepends largely
upon theparticular theory
of musicin which it appearsand notupon itsbasic function.
Unfortunately thetenn ornament has takenon pejorativecon-
notations.
It will be found that mostof the words for ornament! whichimply for
us thenotion of something adventitious and luxurious,added toutilities
but not essential totheir efficacy,originally implieda completionor ful-
Hllment ofthe artifact or other object in question; thatto decoratean
object or person originallymeant to endow the object or person with
its or his necessary
accidents witha view to properoperation .... 1
We must revise our attitude toward ornamentation. Ornaments are
of the essence of music. Indeed, since music is architectonic, it is
possible toconsider eventhe largestsections ofa compositionas
being essentially
ornamental, thoughof coursethis involvesa special
use of the term. From this point of view the analytic method de-
veloped by Schenker partlyconsists inexhibiting thisprocess of
ornamentation on higher architectonic
levels."
Ornaments, then,must be considered asinseparable fromthe
structural tones
and basicplan whichthey ornamentand to which
they givemeaning. Theythemselves are likewise inseparable
from
and meaninglesswithout thebasic substantive
tones, harmonies,
and
rhythms whichthey ornament.From this point of view, the basic
structural framework of a passage orcomposition maybe considered
as anorm, andthe ornaments
which breathemeaning andfeeling
into this plan may be regarded as deviants.
A melody without ornament, says anIndian theoretical treatise,
is like a night without amoon, ariver withoutwater, avine with-
out flowers, or a woman without jewels. 1
ORNAMENTATION IN THE WEST

Because ornamentationor embellishment!, in the strict senseof
the word, became codified and systematized during the baroque
period of Western cultureand becauseof the concern of com-
posers andtheoreticians ofthat period with emotionalexpression,
the relationshipbetween ornamentation and affectiveaesthetic ex-
perience isclearer andmore explicitin the writings of this period
than in those of most other epochs. For instance, though Reeses
discussion ofornamentation inmusic ofthe MiddleAges makesit
clear thatthe processwas ofconsiderable importance, its relation
to affectiveaesthetic experience
is not made explicit.Similarly in
the Renaissance, though otheraspects ofthe processof deviation
can be related to aesthetic experience,embellishment is not gen-
erally discussedin detail nor is it, generally speaking, explicitly
correlated with aesthetic considerations.After the baroque period
as composer
and performerbegan tobe specialists,
composers de-
tailed their embellishments into the score, and ornamentation as a
separable resource
of musical expression allbut vanished from
musical practice.
The typicalbaroque attitudetoward embellishments is concisely
and forcefullystated byBlanchet, whowrites thatthey aregreatly
beiitted to move powerfully the soul; to deprive music of such orna-
ments would be depriving it of the most beautiful part of its
essence. 2
Ornamentation has also beenconnected withthe spirit
of play and, by implication, with the pleasure of the creative per-
former in surmounting difficultiesand overcoming resistances. To
this exactnessthey [theItalians] join all the embellishments anair
is capableof; theyrun a hundred sortof divisionsupon it; they in
a mannerplay with it .... 21 Although the elaborate embellish-
ments, coloraturas,and cadenzasinvented byperformers perhaps
involve acertain amountof exhibitionism, only aviewpoint saturated
with nineteenth-century prejudice would imply that this is their
basic raison dtre. Cadenzas and other embellishments have an
aesthetic function, delaying an expected resolution,deviating from
the normative melodic curve, or otherwise creating psychological
tension. i
Having thus established the
connection between ornamentation
and affectiveexpressive experience,
it remainsto show that orna-
mentation involvesthe inhibition of a tendency or habit. Embellish-
ment functions in two ways in creating inhibition or tension.
In the first place,since mostembellishments are not structural
tones, theydelay inhibit! the arrival of the expected andantic-
ipated structural
tone. As
C. P.E. Bachputs it:_ . . thesmall notes
rather thanthe principaltone arestruck withthe bassand theother
parts, 22and because
these ornamental
tones are,
as arule, dis-
sonant with the basstheir tendency is very clear andthe expectation
of consonanceis active. Bach himself notes that appoggiaturas
retard the chords whichare calledfor by the bass.23 Discussing
ornamentation in the Palestrinastyle, ]eppesennotes that the third
note of the cambiatais followedby the skip of the third upward,
followed inturn by a stepof the second downward, through which
the real note of resolution comes at last and is thus doubly de-
layed. 2*
In the second place,
many ornamentstend to create doubtand
uncertainty, however
momentary, as
to which tone isthe structural
or substantive one. Trills, turns, or even the very wide vibrato
tremolo effect
present insome types of primitivemusic seep. 203!
often obscurethe musicalpattern temporarily.In other cases mo-
mentary uncertainty arises asa result of the manner inwhich the
ornament isperformed. Thiswould appearto be true of the em-
bellishment whichCouperin callsthe accent.Couperins ornament
is derivedfrom lute technique, beingoriginally thestopping ofan
auxiliary noteon the lute without actually pluckingit .... The
effect . . . cannot really be translated into terms of the harpsi-
chord; the player approximates
it by playing the auxiliary note,
usually a tone above and in dotted rhythm, as faintly and vaguely
as possible. 25
The relationshipbetween embellishmentand uncertaintyis clearer
where an elaborate ornamentalmode of performance, usually im-
provisatory, is sustained for a considerable time. This is well illus-
trated in the following passage in which Francois Raguenet com-
pares the French and Italian styles of musical performance at the
beginning ofthe eighteenth
century.
The Frenchwould thinkthemselves undone if they offended inthe least
against therules; theyflatter, tickle,and courtthe earand arestill doubt-
ful of success, thougheverything bedone withan exactregularity. The
more hardyItalian changesthe tone and the mode withoutany awe or
hesitation; hemakes doubleor treble cadences ofseven or eight bars
together upontones weshould thinkincapable ofthe leastdivision. Hell
make aswelling ofso prodigious a lengththat theywho areunacquainted
with it can't choose but be offended at Hrst to see him so adventurous
but before he hasdone theyll think they cant sufficientlyadmire him.
Hell have passages ofsuch an extent as will perfectly confound his
auditors atfirst, andupon suchirregular notesas shallinstill a terror as
well as surprise intothe audience,who will immediately conclude that
the wholeconcert isdegenerating intoa dreadfuldissonance; and betray-
ing 'em by that means intoa concernfor the music, whichseems tobe
on the brink of ruin, he immediately reconciles em by such regular
cadences thateveryone issurprised to see harmonyrising again, in a
manner, outof discord itself and owing its greatest beautiesto those
irregularities whichseemed tothreaten it with destruction.
Here boththe relationshipbetween embellishment

and affective
aesthetic experience
and the relationship between
doubt and un-
certainty and aesthetic experienceare madeclear andexplicit.
These relationships, which are the very ones assertedby the
hypothesis ofthis study,are impliedin descriptions of the response
to music of widely diiferent styles.For instance,a similar type of
experience is reported by the Renaissancetheorist, Heinrich
Clarean, whodescribes the effect of]osquins De Profundis in the
following words: I wish every oneto observeclosely __ _with
how muchpassion and how muchmajesty thecomposer has given
us the opening words_ __ with astonishing and carefully studied
elegance, hehas thrownthe phraseinto violentdisorder, usurping
now theleap ofthe Lydian,now thatof the Ionian, untilat length,
by meansof thesebeautiful refinements, he glides _ __ from the
Dorian to the Phrygian.2' Spanningboth time and style, a very
similar experienceis describedin StanleyDances account of a per-
formance bySidney Bechet, the jazz clarinetist: His most daring
flights of improvisation may momentarily have made the listener
a little nervous, a little doubtful of the outcome, but all were accom-
plished with confident ease.2
ORNAMENTATION IN ORIENTAL MUSIC

When we turn from the art music of the West to the art music of
the Orient,to folk music, orto themusic ofprimitives, weencounter
musical styleswhich are largely improvisatory. Even wherenota-
tion is present, onlythe basicplan of a pieceis writtendown. The
realization ofthat plan is left to the creative abilityof the com-
poser-performer. What he adds to the basic plan,which he has
learned byear or studied inscore, mightall be said to lie in the
realm ofornamentation. The task ofthe Indonesian singer, incon-
trast tothat of the Europeanexecutant musician, is a creative one.
Each timea laguis sung,the songflowers again from thetraditional
melodic ground work, the unalterable melodic nucleus; often to
the delight of those who have learned to esteem the native . . .
style ofperformance. 2 Notice thatornamentation would seem to
be particularlyaffective insuch traditionalmusic sinceexpectation
is quite precise; thatis, the performance ofsuch musicinvolves a
kind of double deviation, a deviationboth fromthe generalstylistic
norms andfrom thenorms whichare specificto theparticular song
being embellished.
In India, too, the distinction between the performer and t_he
creative artist is non-existent.The raga is the melodic structure or
ground plan which the master Hrst of all communicates to the
pupil; and to sing is to improvise uponthe themethus deHned.3
However, since successive variations, whether in art music, folk
music, orprimitive music,made upona givenground planwill be
discussed inchapter vii,we will here discussornamentation only
in the more limited sense.
The importanceof embellishmentin the music ofthe Orient is
made apparentby the elaborate classification
of various types of
ornaments. As Sachs remarks:
So vital in East Asiatic Music is the delicate vacillation that dissolves the
rigidity of pentatonic scalesthat all possible artificeshave carefully
been classified,
named, and, by syllabicsymbols oftheir names,embodied
in notation ....
Few noteshe wouldleave clearand hard;mostly, thestring, afterpluck-

ing, is given additional tension, so that the tone goes up for a moment
or for good; orelse, thestopping fingerleaves thetone justplucked and
rubs alongthe stringwith a wiping noiserather thana melodiousglis-
sando. Suchcontinual wailingand sobbing,though certainlyagainst our
taste, isindispensable whenEast Asiaticmusic appeals to the heart.
The relationship between embellishment and affectis clearlyim-

plied inthis quotation.The relationshipis emphasized again andagain
by orientalmusicians and poets. Forinstance, the
]avanese poet of the
Tjentin collection writes thatThe wiramawhich foundits expression
in the wilet stroveafter touchingthe heart.... More and moreopu-
lently did the wilet entwine itself,striving afterthe awakeningof
emotion.2 Thewirama isthe tempoexpressed time in intervals;the
wilet denotes the timeintervals Hlled by amelody sung or playedon
the rehab a stringinstrument!, andit alsomeans the elaboration of
basic tones,i.e., embellishment. Another stanza of theTjentin poems
is interesting because it specifies thepresence ofexpectation in re-
sponse tothis music.The power of emotiongripped theheart; the
wilet intoxicatedthe senses . . . the gripping powerof the gending
[orchestral composition]
. . _ awakened a feelingof hankeringexpec-
tation, asof onelooking forward, full of longing.3
In books about Indian music and in the works of the theorists
themselves therelationship betweenornamentation andaffect is
made explicitor is implicit in the namesgiven tothe variousem-
bellishments. But in India the note and the microtonal graces
compose acloser unity,for the grace fullilsjust that function of
adding light and shadewhich in harmonized music is attainedby
varying degreesof assonance.
The Indiansong withoutgrace would
seem toIndian earsas baldas theEuropean artsong withoutthe
accompaniment presupposes.
it 3
Both the delaying function of embellishment and its expressive
effect havebeen discussed
by Strangways
in connectionwith Indian
music.
A musical purpose ofimportance isserved bythis Predominantand

its consonant.One of the most effective resources of melody is appog-
giatura, i.e.the delayingof a note, the raising ofpleasing expectations.
. . . The essential condition of appoggiatura is that the delaying note
should befelt to be onlytransitory andthe delayednote to be relatively
final or permanent. Thisis effected in our music by the harmony,ex-
pressed or implied .... But when there are no chords at all in question
the substantivenature ofthe delayednote hasto be established insome
other way.In Indian, as nodoubt inCreek andprobably inthe ecclesiasti-
cal melodies,
this isdone bythe Predominantama, mes, reciting note!;
notes receivethe forceof appoggiaturas
as theydelay this.
As wasthe casewith Westernmusic, sometypes of embellish-
ment donot somuch actto createa delayin the arrival of an ex-
pected substantive
note as
they doto create
uncertainty as
to which
tone isthe substantiveone; or,to put the matterin anotherway,
some embellishments delay thearrival of the substantive
tone asa
simple, clear,
definite pointin the musical structure.
The function of coloratura and embellishmentin Byzantine chant
has alreadybeen touchedupon in another connection see p.
67!. The importance of embellishments in ]ewis-h chant of Asia
and eastern
Europe ismade clearin the following:
. . . for the outpouringof the heart in prayer-texts musical
embellish-
ments offeradded powerof interpretation,and the solo-recitative insures
freedom fort_he expression
of risingdevotion.
The coloratura
in the Eastern Europeanchazzanuth is like the soul inthe
body; withoutit that chazzanuth loses
its vitality, its charm,its fascina-
tion.
Ceneralizing onthe subjectof oriental melody, Sachs writes that

Melody, in the Grient, hasalways meantone of those flexible
patterns thatthe Arabsfinally classified
as maqarnatand theHindu
as ragas,
which imposed upon thesinger theirspecific genera,
scales,
pitches, accents,
tempos, and moods, butgranted himfull personal
freedom for their elaboration. 3
FOLK MUSIC AND PRIMITIVE MUSIC

The line between folkmusic andthe improvisatorytypes of art
music seemsto be drawn onthe basisof musicaltheory ratherthan
musical practice. For the folk singer like the oriental creator-
performer orthe performerof the baroque periodin the Western
tradition alsoperforms hiscreative taskby ornamenting and varying
11 basicplan. Thechief difference between thefolk singersart and
that ofthe trainedmusician liesin the fact thatthe planof folk song
is handed down in an oral tradition and embellishments are not
Codified butare traditionalor inventedby the singer himself.
The primitive musician unhesitatingly alters the traditional materialhe
has inheritedfrom thousandsof unknown talents andgeniuses before
him .... He not only remouldsold ditties, but also weaves together
fresh combinations of moreor lessfamiliar phrases, which he calls ma.k~
ing new songs .... It would be diflicult to exaggeratethe extent to
which suchtraditional singers embellish so-called simple melodieswith
a regularriot of individualistic excrescencesand idiosyncrasies of every
kind, eachdetail ofwhich . . . is a precious manifestation of real artistic
personality.
Since folk material has no theoretical codification of ornaments
and notheoretical literature,
we shallhave torely for testimony as
to the relation of embellishment toaffect in folk music on the experts
in the Held offolk music.Writing of Yugoslav folkmusic George
Herzog says:
Musical rhythm is treated more briefly, since its patterns in Yugoslav
folk music are free,very rich, and evendiffuse; thebasic structuresare
overlaid bya freerubato performance and luxuriantornamentation ....
South Slavicmusic has an especialappeal. Thismay wellbe dueto the
contrast betweenthe essentialsimplicity of its basic materials andthe
pulsing quality of life achieved throughan abundanceof expressive
devices, includingthe ornamentation.
Bartok alsodiscusses certain aspects ofthe rhythmicstyle of these

songs andrelates therhythmic irregularitiesto expression. He says
that a characteristic featureof the Serbo-Croatian and Bulgarian
folk melodiesis that the line-, word-, and syllable-interruptions
effected byshorter orlonger rests[are] not for articulationssake,
but for decorative-one might almost say for expressive-pur-
poses. 2
Each folkculture has its ownbasic plansand itsown style.Within
the limits of these the singer embellishes, alters,and often distorts,
making theimpact ofhis owncreative personality felt and reveling
in whathas been called thejoy of being acause. Hetakes pleasure
in techniqueand itsmastery, and, in the course ofexhibiting it, he
will deliberatelyattempt thediflicult. _American
folk fiddle players,
for example, seek deliberately for certain wild notes.3 In The
Book ofAmerican Negro
Spirituals Iames
Weldon ]ohnson
mentions
the curious turns and twists and quavers and the intentional strik-
ing of certain notesjust a shade offthe key,with which the Negro
loves to embellish hissongs. 44And Metfesselrelates theseem-
bellishments toemotional expression, saying thatthe ornamentsare
excessive, butlit into the general emotional intoxicationof the
singer. 45
Real jazzis alsoa kind of folk music involvingthe improvisation
about and embellishment of a basic plan or structure. In their con-
certed improvisation
the players. . . worry and cajolethe rhythms
and phrasesof their solos, extemporizing here and there, intro-
ducing solobreaks andotherwise ornamenting the printedskeleton
that hasbeen providedfor their collective guidance.4 Theimpor-
tance of these improvisationsfrom the aesthetic artisticpoint of
view isstressed by
jelly Roll Morton:
You mayhave noticedthat in playin jazz the breaksare oneof the most
essential thingsthat youcan everdo in jazz. Withoutbreaks andwithout
clean breaks;without beautifulideas in breaks, youdon't needto even
think aboutdoin anythingelse .... Without a break youhave nothin;
even ifa toon havn a break init, it is alwaysnecessary to arrange some
kind of spot to make abreak because as I said before,you havent got
jazz .... A riff is a background-a riff is whatyou wouldcall a founda~
tion as like what you would walk on!, and a break is something that
you break.... 4'
The quotation is particularly interesting becauseit relates the

foundation or riff the norm! to the breaks the deviation!.
Rex Harrisstresses the emotional aspect of the jazz performance:
]azz . . . like all folk music,is emotionalin outlook.It is playin
from the heart .... Inspiration, spontaneity, exuberance, absence
of inhibition, all thesequalities areaxiomatic injazz. 4In short,
a popular song played as written is seldomhot; it is the series of
artistic distortions given to the tune that makes it jazz. 4
Much primitive music is intimately connected with ritualistic
observances and magic ceremonies,and it is probablyfor thisreason
that it often tendsto beless subject
to variationand omamentation.
The religioussanctions imposed upon primitive art and the com-
munal nature of the primitive ceremoniesprohibit creative deviation
on the part of individual performers. When, however,primitive art
becomes separated
from ritualand magic,we I-lnd
the same
tendency
towards ornamentation and embellishment that is usual in folk
music and art music.
The tendency toward elaboration of a basic plan,and elaboration
exhibiting thecreative abilitiesand tendenciesof the performer, is
found in almost allprimitive musicalstyles." In African music:
The individual performer hascomplete freedomfor his own individual
and inspiredvariation withinthese fixedlimits. Providedhe keepsto the
main mlesan Africanperformer mayintroduce asmany subtlevariations
as hepleases-and allconceived onthe spur of the moment. It is spon-
taneous music.You neverhear the same songdone exactlythe same
way.
Even wheremusic has an avowedlyutilitarian functionas inAfrican
drum language or in the ]abolanguage ofwestern Liberia,an aes-
thetic elementoften enters.Commenting uponthe way in which
a hornplayer callson anotherman bysounding hisname inmusic,
Herzog writes:The free variation oftempo andprosody injectsa
semi-aesthetic play-element.
But a definite departurefrom mere
speech representationtoward musicalornamentation is achieved in
the occasional transposition of the signal from its normal level to
the lower one .... 5 Helen H. Roberts also notes and comments
upon theaesthetic playelement inprimitive musicand embellish-
ment.
Tonality and Deviation

The term tonality refers to the relationships existing between
tones ortonal sphereswithin thecontext ofa particularstyle system
see pp.45 ff.!. As Strangways has put it: A tonic is a tendency
rather thana fact. 55 Thatis, someof the tones ofthe systemare
active. They tend to move toward the more stable points in the
system-the structuralor substantive tones.
But activity and restare relativeterms because
tonal systems
are
generally hierarchical:tones whichare activetendency toneson
one level may be focal substantive tones on another level and vice
versa. Thus in the major mode in Western music the tonic tone is
the tone of ultimate rest toward which all other tones tend to move.
On the next higherlevel the third and fifth of the scale,though
active melodictones relativeto the tonic, join the tonic as structural
tones; and all the other tones,whether diatonic or chromatic, tend
toward oneof these.Going stillfurther in the system,
the full com-
plement of
diatonic tones
are structural
focal points
relative tothe
chromatic notes
between them.
And, finally, as wehave seen,any
of thesetwelve chromaticnotes maybe taken as substantiverelative
to slight expressive deviations
from their normal pitches."
At the other end of the architectonic scale it should be noted that
tonality playsa partin the articulation ofmusical forms larger than
those ofthe phraseor melody.A musical section ofconsiderable
length maybe in a tonalsphere which, relative tothe tonalspheres
of othersections, is
structurally active,so thatthe wholesection may
be saidto tend toward anothersection whose tonal sphereis sub-
stantive. However, even within a tonal realm of a section that is,
relatively speaking,
at rest there aretensions andreleases which
function at the sectional level.
If the architectonic structure of tonal levels is viewed from the
smallest deviation to the larger ones, it might well be argued that
the wholestructure canbe understood as ahierarchy ofembellish-
ment. ]ustas theexpressive microtonaldeviations, whether
classified
as in oriental musicor freelyimprovised asin Westernmusic, can
be regardedas embellishments of the tones to which they are
applied, sochromatic tones can beregarded asthe embellishments
of diatonictones. Similarlythe auxiliarydiatonic tonescan be re-
garded as embellishments ofsubstantive diatonic tones, and so on,
even to the sections some of which would then be considered as
embellishments of others.
The essential point is that auxiliary tones, whether expressive
deviants on the lowest architectonic level or whole sections on the
highest level,are justas vitaland importantto musicalcommunica-
tion asare thesubstantive tones toward whichthey move.Indeed,
the importanceof suchtones isoften indicatedeither bythe names
given to them or by the restrictions governing their use.
The Iavanesescales, forexample, consistof one group of three
and one group of two principal tones, separatedby gaps of one
tone. The tones that have been missed . . . are, as a matter of fact,
not always completely eliminated;they may have a modest function
as secondarytones. 5'What is of interest is the names givento
these secondary tones. Theyare calledpamanis, which comes from
the word meaning sweet or harmonious, or they are called
senggol nyimpang. According toCoolsmas Dictionary of the Sun-
danese Language, the wordsenggol meansthe modulationof the
voice, trills,turns, etc.with whichthe singerornaments the tune, 58
while nyimpangcomes fromthe word meaning tostep aside,to
evade, todeviate. Inthis casethe relationshipbetween ornamental
and auxiliarytones isabsolutely clear.So, too,is the relationship
between theseveral names given to these auxiliarynotes.
The situation in Chineseand japanesemusic is quite similar.
Here the two tendencytones filling in the basically pentatonic
structure kept a transitional,auxiliary character and hadnot even
the privilegeof individualnames: theChinese called them by the
name ofthe notedirectly abovewith the epithet pin,which means
on the way to, becomingf 5 Thenames ofthese activetones in-
dicate theirtendency toward other tones,
while theiraffective power
is indicatedby the fact that they werenot permittedin palace or
temple musicbecause farfrom soothingthe passions[they] filled
the soul with sensual lust.
From allof thisit seemsclear thatcomparative musicologists
who
treat auxiliarynotes asunimportant andincidental havemisunder-
stood their function in the total musical process.The error is serious
because anunderstanding andadequate description of style de-
pends uponthe recognitionand examinationof the relationship
between thestructural tonesand the tendency tones belonging to
the style.
Such misconceptions occur in part becausemusicologists and
ethnologists have too oftenbeen concerned with the collection and
classification of
scales orwith simple-minded statistical compilations.
In the study offolk musicthere hasbeen perhaps a little too much
preoccupation withscales and intervals, whichare merelythe raw
material ofmelody, atthe expense of studyingtonality. 61Even this
admirable statement reveals a curious distortion of the facts, for
scales arenot the raw materialsof melodies.It is just the other
way round: melodies arethe materialsfrom which scales areab-
stracted. Scalesare simply the tones of melodiesarranged in a
linear orderof pitch succession. Theyare not usually presentin
the minds of the musicians, who think rather in terms of melodies
directly.
CHROMATICISM
Almost all of the tonal systems that have been used in music,
whether Western,Eastern, folk,or primitive, are essentiallyand
basically diatonic. Chromaticism is almost by definition an altera-
tion of, an interpolationin, or deviation from this basicdiatonic
organization.
However, notevery alterationof a diatonic organization results
in a feeling ofchromaticism, in
a senseof deviation.Where several
different andalternative modes of tonal organization arepossible
within a given musicalstyle orstyle systemor wheresuch modes
are themselves subject to transposition according to the rules of
operation prevalentin the style, thealteration ofone tonalgroup
may wellbe interpretedby the listener asconstituting achange or
a transpositionof moderather thanas beinga chromaticpassage.
Of course,it is possible thatthe alterationsinvolved insuch mode
changes ortranspositions may at first be interpretedas chromati-
cism andonly laterbe understood for what they reallyare. Inother
words, thehypothetical meaning and the evident meaningof the
alterations maybe different.
The interpretationof anyalteration inthe basicdiatonic structure
established at
the outsetof apiece willbe governed by theoperation
of thelaw of Pragnanz as it functionswithin the particular cultural
stylistic context;that is, the changewill be interpreted in the simplest
way possible within the style system.
If it is simplerto considerthe
alteration asa changeof mode,the listenerpracticed inthe style
will adopt this interpretation.While if it is easier toregard the
alteration asa deviationfrom a single mode,as achromatic modihca-
tion or deviation, this will be the interpretation made bythe listener.
0bviously the listeners understandingof such alterations ispartly
a function of their relative duration. If the alteration is only tempo-
rary, thenit will probably beunderstood as
a deviation.While if the
alteration ofa tonewithin thetonal structurepersists forsome time,
then the alteration will be understood as constituting a change of
mode or a transposition.
The distinctionbetween chromaticism as a temporary phenom-
enon andthe expressive pitch deviationsdiscussed aboveis not an
easy oneto draw. The microtonalintervals whichso oftenaccom-
pany the embellishments in oriental musiccould havebeen con-
sidered underthe subjectof chromaticism. Or, aspreviously stated,
chromaticism may be viewedas atype ofembellishment. Both types
of pitch alteration or interpolation delay,block, or inhibit the
arrival of the expected,normal diatonictones ofthe given mode.
Both performan expressive function. Thedifference wouldappear
to be one ofstandardization. The microtonal intervals
employed in
oriental music,though moreconsciously introduced and controlled
than thoseintroduced byperformers inthe Westor thoseused by
primitive andfolk musicians, are notthemselves really
standardized,
and theintervals assuch appear to havebeen givenno theoretical
codiiication.
It seemsunnecessary toprove that in Westernmusic chromati-
cism represents
a deviationfrom the normal diatonicmodes. Its
treatment both in theory and practice is ample indication of its
exceptional and
unusual character.
The specialnature ofchromati-
cism is evident whether we consider the construction of musical in-
struments, whichare allessentially builtto performdiatonic music;
the prescriptions
as tothe useof musicafalsa; theextreme emotions
which chromaticism has beenused to express; orthe rules which
Leopold Mozartand C.P. E. Bach setforth for the performance of
chromatic notes.
The affective aesthetic powerof chromaticismnot only arises
because chromatic alterations delayor block the expectedmotion
to thenormal diatonic tones butalso becauseuniformity ofprogres-
sion, ifpersistent, tends,
as wehave seensee pp.164 f.!,to create
ambiguity andhence affective tension. Moreover, ambiguity leads,
particularly inthe realmof harmonicprogression, to a generaltonal
instability.
A particularlyinteresting instance
of the coincidence ofambigu-
ity, affect,and a quasi-chromatic deviceis to be foundin the jazz
blue note. The aestheticaffect ofthis tone is generallyacknowl-
edged. Itsambiguity andrelationship to chromaticism aremade
evident in the following: The outstanding quality of the blue notes
is theirambiguity .... This harmonicsuccession strongly suggests
the ambiguityof the melodic notewhich, nowa loweredthird de-
gree of the scale,now a raised seconddegree, produces either a
leading toneto the major third,or a blue minor third within the
major scale.5 Finally, it is important to bear in mind that chro-
maticism, though considered here as an independent variable, is
in practiceused in connection withother supportingtypes of de-
viation; e.g.,rhythmic delaysand irregularities,delays inthe filling
of structural gaps, weakly articulated melodic structures, and the
various other means of affective aesthetic communication discussed
in earlier chapters.
Instances of the connection between chromaticism and emotional
communication are so common in the history of music since the
Renaissance thatany musicianor music lover can easily cite a host
of examples for himself.Only afew examples
will be referred toin
this study.
The most striking evidence of the connection between chro-
maticism and affect is to be found in music written to a text-music
in which the emotional nature of the music is, as it were, warranted
and specifiedby the text. Suchis the case withthe Renaissance
Netherlands motet and the Italian madrigal.
Chromaticism always represents the extraordinary .... Again andagain

we find chromatic treatment given to such highlyemotional concepts
as
crying, lamenting,mouming, moaning,inconsolability, shrouding ones
head, breakingdown, andso forth.
In the Italian madrigal the same concepts find expression through
the mediumof chromaticism.There theyrepresent manas entangledin
his earthlypassions, while
in the music ofthe Netherlandsthey symbolize
the devoutbeliever struggling with the burden ofsorrow whichGod has
laid uponhim to test hisfaith.
About fifty years later, just at the beginning of the baroque period,
we find this relationship between chromaticismand affective experi-
ence neatly illustrated in the book of Chromatic Tunes 606! by
john Daniel Example 84!. One might cite almost countless ex-
amples inbaroque music of the relationship between
chromaticism
and affect both in our senseof the term affect and in the eight-
eenth-century sense!. Two particularlyfamous pieces
illustrate this
relationship clearly:namely, Didos Lament from Purcells Dido
and Aeneas and the Crucihxus of Bachs Mass in B-Minor.
chro-ma-tic tunesmost like

my pas-sions
sound
-; 1
iliiii
._92-1-_l1KE.- --- IU1
uuliilvifri I1 . _` _ Si
1L__J|'I
rm; nuiwri-$11
rgiinunil l_ 1:-1
ilunii l 1 1 ''|l`1 I
EXAMPLE 84
Chromaticism inWestern musicis not exclusively oreven pre-

dominantly amelodic phenomenon; S it is also a harmonicphe-
nomenon. As such itis capableof arousingaffective aesthetic
experi-
ence, notonly because it may delay or alter theexpected diatonic
progressions whichare the norms of tonal harmony,but also be-
cause it tends to create ambiguity and uncertainty as to harmonic
direction. Chromaticpassages ofconsiderable duration, passages
which areoften modulatory, appear tobe ambiguous because they
obscure thefeeling oftonal center,because the
ultimate endof the
progression cannot be envisagedor becausemore thanone tonal
center is indicated seepp. 171ff.!. Such ambiguity createssus-
pense anduncertainty which, as wehave seen,are powerfulforces
in the shaping ofaffective experience.
This relationshipbetween suspense and uncertainty,on the one
hand, and chromatic progressionsand modulations,on the other, has
been wellunderstood and fully exploitedby operacomposers ofall
periods. Forinstance, ]ohnBrown, writingin the early eighteenth
century, observesin his Letters uponthe Poetryand Musio of the
Italian Operathat:
They [Italian composers] must, in the Hrst place,have observed, that all
those passages in which the mind of the speaker isagitated bya rapid
succession ofvarious emotions, are, fromtheir nature,incompatible with
any particularstrain, orlength of melody .... But, whilst the Italians
conceive suchpassages to be incompatiblewith that regularity of meas-
ure, andthat unity of strainwhich is essential toair, they felt, however,
that they were of all othersthe mostproper subjectfor musicalexpres-
sion: And,accordingly, both poet andmusician seem, by mutualconsent,
to havebestowed onsuch passages their chief study; andthe musician,
in particular,never failsto exerton themhis highest
and mostbrilliant
powers ....It is in thisspecies song
of that
the finest
effects of
the
chromatic, and,as faras oursystem ofmusical intervalsis susceptibleof
it, even of the enharmonic scale,
are peculiarlyfelt; and it is here also
that the powers ofmodulation aremost happily,because most properly,
employed .... 6
Donna Annasconfusion andagitation upondiscovering thebody
of her father providesa fairly typical instanceof the use of chro-
maticism tocreate afeeling ofuncertainty andsuspense Example
85!. Noticethat not only arethe progressions chromatic, butthey
= "11 -1'792IYi_l 1_ If 11.12 l: _;Z 1I"'1
' ' '_===I$:'.==l =%==f==';?==S=9=f=|E
f; Q"|5"'9 /r92l92l|| Pl'9
/'i quelvolto
=.i:E:=E';:-;..r'1. ;1=__,|*;:..i ~i=
l92i%=HZ=f '=ITl- ;i~-e=3l. ' =
;
_: _ __i_i___. to orchestra. _ 71;

L$?%':Ei E5 ' :` e =` 555
A_
tin -toe
co -per-to
del col-or
diamor-te
1 if.
r 'E _ " "_' , Y
Las
QI
rf -=:i:r =' E. rrpa a :r ;:
Ls=:::: E c : c ca 1:
EXAMPLE 85
are alsosequential, which

heightens theelement of
suspense, since
the ultimate
goal ofthe passage
is in doubt. Also
observe that
the
passage extensively
employs oneof themost effective
ornaments, the
appoggiatura.
In the inst1'u1nental music
of the classical period chromaticism is
employed with
great effectiveness
but not, generally speaking,
in
the construction of themes and melodies. These tend to be diatonic-
with, of course, some
notable exceptions.
Rather chromaticism
finds
its placein the bridge passages and development
sections where,
contrasting with the moreregular andnormal progressions
of the
theme groups, it combineswith othertypes ofdeviation tocreate
Suspense and uncertainty. In
this musicchromaticism becomesone
Of thebasic organizing
forces ofthe total form. Thelevel of chro-
maticism isno longerfairly constantthroughout apiece. Instead,
the
generally normative, diatonic motionof the theme groupis con-
trasted with the intensity of suspenseto which chromaticism makes
an important contribution.
Since thisis not a historyof chromaticism,it seemsunnecessary
to traceits progressup to the present.Suffice itto point out that in
the nineteenthcentury chromaticism becomes analmost indispen-
sable resource of composers, and one can find it at work on all
levels of the musicalorganization-in the melodies, theirharmoniza-
tion, and in the construction ofthe musicalperiods, sections,
and
total structures. Indeed, it can be argued that its extravagant and
prodigal useserved, inthe last analysis, toweaken anddestroy its
effectiveness because
it tended to becomenormative within the style.
THE MINOR MODE IN WESTERN MUSIC

The theoreticaland psychologicalbasis forthe affectivepower
of the minor modein Westernmusic haspuzzled andperplexed
so manyexcellent musicologists
and psychologists that it may seem
rash topropose another
answer here.
But thereare twogood reasons
for doing so. First, any theory which purports to explain the basis
for the affective responseto music must take account of and at-
tempt tounravel thismystery. Andsecond, the
fact thatthe hypothe-
sis under consideration leadsto a satisfactory solutionof the riddle
of theminor mode, without recourse
to anyof thePythagorean mys-
teries, isin itself strong evidence
in supportof the theory advanced
in this study.
The theoreticalattempts tosolve theenigma ofthe minor mode
have rangedall the way from Reimanns theory,which contends
that theminor triadis built upon theinverted overtone
structure of
the major triad, to that of a psychologist who considers that since
the minortriad isa lowering of themajor oneit clearly represents
a castrationcomplex andhence arousesfeelings ofanxiety." Theo-
rists with an acousticalbias have attempted to show that the minor
triad is more dissonantthan the major.
Each of these accounts,
aside frombeing fairly farfetched, in-
volves significantdifficulties. Reimannstheory, for instance, notonly
provides no explanation of the psychic mechanism whereby the
minor triadis sensedas theinversion ofthe majorbut alsofails to
relate such inversion to general aHectiveprocesses. Aboveall, such
theories seemimplausible because they ignore both the facts of
music historyand thoseof comparativemusicology.
The acousticiansfail to convince because, though it might be
possible to
relate dissonance to affectiveexperience seep. 229!,the
minor triad is not more acoustically dissonant than
the major one,
at leastnot at its closedposition Example86!.
2* _ III-n IEW W-E!1J392n||0
CITY!
__ E
ali* 3%
'2 2- E .
EXAMPLE 86
Eventually allthese theories

are facedwith thesingle overwhelm-
ing factthat theminor triadhas nopeculiarly affective
significance,
carries no sad connotations in non-Western music, in folk music, or
in the music ofprimitives." Thiswould seemto indicate,as Hein-
lein haspointed out," that like most musicalresponses the
one to
the minor mode is not natural or universal but learned.
This statementrequires some
qualification. Forthe minor mode
as we know it did not exist before the Renaissance and assumed
universal importance
only with the adventof baroquemusic. 7*It
would bea gravemistake, though one commonly made, toconfuse
the minor mode as it has existed since the Renaissance with those
church modeswhich happento have a minor third. Nor is it en-
tirely accurateto saythat theaffective response
to the minor mode
must of necessity bea learnedresponse merely because nosuch
response isattached tofolk musicor primitivemusic havinga minor
triad. For the minor mode in Western music is not the same mode
as the melodic constructs of non-Western music, folk music, or
primitive musicwhich happento employ a minor third abovethe
tonic or keynote.
Here, then,we cometo the great stumblingblock of almost all
theorists, musicologists,
and psychologists.
They haveregarded the
vertical organization of the minor triad as itself constituting the
essential andpeculiar featureof the minor mode.Yet the very
name, minor mode, indicates that what is involved is not a single,
isolated structure the triad! within the style system but rather a
special systemof tonal relationships withinthe style as a whole,
i.e., a mode.
Melodically the minor mode differs from most other modes in
that it is quasi-chromatic
and changeable,
appearing inseveral dif-
ferent versions,while other modes, whetherthe major mode of
Western music, the modes of oriental music, or those of folk music
and primitive music, areessentially diatonicand stable.This can be
illustrated simplyby comparingthe numberof tonesavailable in
most modes with those available in the minor mode. For instance,
the majormode ofWestern musicExample 87,A! uses the same
seven tones, whether ascendingor descending,as doesthe prirni-
tive African scale B! given by Hornbostel," whilethe Iapanese
Hirajoshi scale C! given by Sachs," contains onlyEve tones,
ascending ordescending. The minor mode,on the other hand,ap-
;Liu'A'
: llii I I .11 5l1|' |I~li92 l; c:
EXAMPLE 87
pears in three basicversions: themelodic minorascending Ex-

ample 88,A!, the natural minor descending B!, and the harmonic
minor as well as in various permutations andcombinations of
these. If the commonlyused phrygian lowered! secondof the
scale, atone imported from another mode, is included in the reper-
tory of tones furnishedby the variants ofthe modegiven in A and
B, all but two of thetwelve tonesof the chromatic scale
are present
and available in the minor mode C !.
A s. c c.
EXAMPLE 88
Sometimes melodies
in theminor mode are markedly chromatic.
The theme of BachsMusical Ogering,
for instance, employs eleven
tones ofthe chromatic
scale inwhat is,to all intents and
purposes,
Deviation inPerformance and
Tomzl Ofganization225
a strict
linear order
Example 89!.
In suchoases the
relationship
between the
minor mode
and theaffective quality
attributed to
it
is nothard toestablish. The
relationship between
uniformity,
ambiguity, and
chromaticismalready
has been disonssedobap-
in
ter v.!
EXAMPLE 89
But not all melodiesor evena majority of the melodies Written

in the minor modeemploy allof the tones aVai1ab1e_Nor are the
tones used
generally arranged
in linearscalar! soooessiorr
Though
in the minor mode, the opening melody of Bi-abniss Sonata in
D Minorfor Violinand Piano
Example 90!
involves Praotioally
no
chromaticism. Nevertheless,
the affective
quality ofthe minormode
makes itself felt even in such diatonic minor inelodies_ Thereare
Allegro.
~92' e I' """"~--====li'2i"}L.:=

f
EXAMPLE 90
several reasons
for this.
First, the
minor modeis always
Potentially
chromatic, and
the listener
practicedthe
in Percgptign
of andre-
sponsethis
to musicis well
aware the
of ever
Present Possibility
of
chromaticism. Second,
the tendencies
of tonesas they
approach
substantive tones
is stronger
in minorthan inrnaior_ For
the two
most important
substantive tones
each havean additional
leading
tone inthe minor;
i.e., the
fifth canbe approaolied
from half
a step
above andthe tonic
can beapproached throughthe Phrygian
seeond_
Furthermore,tendency
the the
of third
toward the
tgnic is
Stronger
in minorthan inmajor because
of itsproximity to
the seeond
and
hence the
to tonic.
The proximity
of these
tendency tones
to sub-
stantive tones
in theminor mode
makes delay
a intbe arrival
of a
substantive particularly
tone intensely
felt. Thirdwhile the
Prob-
ability ofa particular
successiontones
of isgreater in
minor asWe
approach substantive
tones,isit considerably
weaker we
as depart
from substantive tones. For the very fact that the minor mode pos-
sesses richer
a repertoryof tonesmeans thatthe probabilityof the
occurrence ofany particulargiven toneis weaker,the moreso be-
cause thetones inthe repertorytend to be usedas alternatives for
one another-i.e.,either B or B-Hat,A or A-Hat, etc.,may, for in-
stance, beused inthe key of C minor to follow somesubstantive
tone. In other words,the minor mode isby its very naturemore
ambiguous than modes witha morelimited repertoryof tones.
From a harmonic pointof view, the minor mode isboth more
ambiguous and less stable
than themajor mode.It is more ambig-
uous because the repertoryof possiblevertical combinations
is much
greater inminor thanin majorand, consequently, the probabilityof
any particularprogression of harmonies issmaller. Whilethe tonic
chord in the majormode can,with varying degrees ofprobability,
move toany oneof six triads Example91, A!, the tonicchord in
the minor mode canmove toany oneof at least thirteendifferent
triads B!, not countingthe chromatically altered chords
so common
in minor. Furthermore, inthe majormode onlyone of the triads,
the onebuilt uponthe seventhdegree ofthe scale marked with a
cross inExample 91!,is itself ambiguous. Butin the minor mode
four such ambiguous triads are possible.The diminishedor aug-
mented triads marked with a crossin Example 91! are ambiguous
A. B. ++ + 4-
Q| " I-
r vow; 1|-g
; :: -- if =" TT:--=
EXAMPLE 91
because theiruniform constructionenables themto imply several

equiprobable resolutions
see pp. 166 ff.!. In short, because their
uniform construction
leaves them
without aroot orfocal point,they
can easily swerve into new and sometimes remotetonal spheres.
This instabilityof thetriadic unit is complemented by the fact that
in its natural form,i.e., withoutthe raisedleading tone,the minor
mode asa total system lacks stability in that it tends to gravitate
toward thekey of the relativemajor. Andthis tendencyis perhaps
a sign of the basic harmonicnormality of the major mode.
This analysis also accountsfor the practice, common in recent
Western music,of movingfrom themajor modeto theminor mode
at thebeginning ofbridge passages,
episodes, development sections
see Examples 37 and63!. Similarlyit explainsthe practiceof em-
ploying theminor modefor introductionsto pieceswhich are in
major see Example 73!. For since the minor mode tends to be
chromatic and ambiguous, provides
it anatural andconvenient way
of movingfrom thestable processes of thetheme groups to themore
irregular anduncertain progressions of the moving passages.
The interrelationship
between affectivity,melodic andharmonic
chromaticism, and the minormode isnot simplya theoreticalor an
accidental correspondence. It is a historical fact. The connection
between them is notonly alogical onebut, asthe followingpassage
makes clear,a genetic one aswell.
The desirefor ever more vivid expression brought about increaseduse
of chromaticism;a trend away fromconstructivism toward free designs
dependent uponthe changingcontent of the text; and the suggestion
of dramatic recitative. Thelast of these, with its approachto monody
through the predominance ofthe top voice and, incidentally, to the
major-minor system, thus wenthand in hand with much the same kind
of chordal writing and lightly imitative counterpoint thatare found in
other musicof the period."
The minor mode is not only associated withintense feelingin
general but with the delineation of sadness, suffering,and anguish
in particular.This association,
which aswe haveseen isalso con-
nected withchromaticism ingeneral, appearsto arise out of two
different though related facts: ! States of calm contentment and
gentle joyare takento be the normalhuman emotional
states and
are henceassociated withthe more normative musical progressions,
i.e., thediatonic melodies
of the major modeand theregular pro-
gressions ofmajor harmony.Anguish, misery,and other extreme
states ofaffectivity aredeviants andbecome associated
with the
more forceful departures of chromaticism and its modal represent-
ative, i.e., the minor mode. ! Marked or complex chromatic mo-
tions common in the minor mode-melodic lines which move con-
junctly by semitones or
disjunctly byunusual skipsand uncommon
harmonic progressions-have tended tobe accompanied by tempi
which were slower than those which accompanied more diatonic
music. Thiswas, of course, particularlytrue of the earlier use of
chromaticism duringthe Renaissanceand thebaroque period.This
coincidence of chromaticism and its modal representative with
slower tempican beexplained atleast inpart ontechnical grounds.
For notonly arethe instruments constructed withthe diatonicnorm
in viewso thatit is more difficultto playchromatic passagesrapidly,
but musicaltraining, bothinstrumental andvocal, is based upon
the normality and simplicity of diatonic progression. Evenafter
three centuriesof the major-minor system Heinlein foundthat in a
random samplingof some twenty-Eve hundredcompositions for
beginners only 7 percent werein minor and almostall of these had
descriptive titlesof somesort. It is, he writes, a difficult matter
to obtaina composition in the minor modewritten for children that
does nothave atitle relatedto the weird, the mysterious, thesad
and the gloomy. 7
Thus the association between the minor mode and emotional
states depictingsadness and
suffering isa product of the deviant,
unstable character of the mode and of the association of sadness and
suffering withthe slowertempi that tend to accompany thechro-
maticism prevalentin the minor mode.Of course,there are nu-
merous exceptions to thisassociation, as
a glanceat theliterature of
the pastcentury shows.
Although theaffective qualitiesof the minor modedepend pri-
marily upon its quasi-chromaticmodal characteristics, the minor
triad as such seems to havea peculiarpower. Thisis particularly
striking wherethe samemelody isplayed firstin the major andthen
in the minor, as,for instance,in the middle sectionof Chopins
Waltz in A Minor, Op. 34,No. 2. But it is not actually thetriad
per sewhich hasthis affectivepower. Forit has been shownthat
the characterof the triad dependsupon and can be changed by
training andmusical context.Indeed, oncea modalnorm hasbeen
established withina given work, the reverse progression, from
minor to major, canalso createa powerfuleffect-as it does inthe
first movementof SchubertsString Quartetin A Minor. In so far
as the minor triad in and of itself is an affective force it is so because
it has become through
association the
sign andherald ofthe mode
as awhole andall that the modeimplies byway of chromaticism,
modulation, and suspense.
Finally, it is clear that none of the foregoing discussionis meant
to implythat all chromatic pieces
are in the minormode. Thecon-
nection iscommon butby no means necessary.
CONSONANCE AND DISSONANCE

The role of dissonancein arousingaffect or in depicting emo-
tional states is evident in the practice of composers and in the
writings of theorists and critics. VincenzoGalilei, for instance,
writes:
. . . In setting to music a sonnet, canzone, romanzo, madrigal,

or other
poem inwhich occursa linesaying, forexample:
Bitter heartand savage, and cruelwill,
which isthe first line of one ofthe sonnetsof Petrarchthey [composers]
have causedmany sevenths,fourths, seconds,and major sixths to be
sung betweenthe parts and by means of these havemade a rough,
harsh, andunpleasant sound in the ears ofthe listeners."
The specification
of emotionalstates indicated
in this quotation de-
pends notonly uponthe useof dissonance but also upon conven-
tional association.
Other writers treat dissonance in more general terms. C. P. E.
Bach, forexample, says
that . . . in general it can be said that
dissonances are
played loudly and consonances softly, sincethe
former rouse
our emotions
and thelatter quietthem. 8Dissonance
is alsoconsidered froma morepurely aestheticviewpoint. Zarlino,
for instance,
points outthat whilea composition
is madeup pri-
marily ofconsonances, dissonance
adds beautyand eleganceto the
work andmakes theconsonance which follows moreacceptable and
Sweet. Infact, if compositions were
made up entirely of conso-
UHIICCS, although
beautiful soundsand good effects wouldissue from
them, they
would stillbe somehow
imperfect, both
as soundand
Composition, seeing
that _ _ _they would lack the great elegance
that dissonance affords. 81
The aestheticaffective pleasure
of dissonance
in relationto con-
sonance is not conned to Western music but has been noted in
connection withoriental andprimitive musicas well.This is made
clear, forinstance, inthe observations
made byRear AdmiralDEn-
trecasteaux aboutmusic in the Friendly Islands:
. . . She, the Queen, was there giving a vocal concertin which the
Futtafaihe sangand beat time, which all the musicians followedwith
greatest exactness.
Some performedtheir part in it by accompanying,
with different modulations, thesimple melodyof the others. We now
and then remarked some discordant notes, with which, however, the
ear of these peopleseemed much
gratified.
Acousticians and psychologists from Pythagoras toRvcz have
attempted toexplain andaccount forthe phenomena of consonance
and dissonance on acousticalpsychological grounds,
but as yet no
tenable, unobjectionabletheory hasbeen advanced. For conso-
nance and dissonance arenot primarilyacoustical phenomena,
rather
they arehuman mentalphenomena and as suchthey dependfor
their definitionupon thepsychological lawsgoverning human per-
ception, uponthe contextin which the perceptionarises, andupon
the learned response patternswhich are part of this context. The
case iswell statedby Cazden:
Though mucheffort hasbeen wasted in philosophicalblind-alleys, studies
of the psychology ofmusical perception have producedimportant nega-
tive resultsregarding consonance and dissonance. The naive view that
by someoccult processmathematical ratiosare consciouslytransferred
to musicalperception hasbeen rejected.Fusion or unitariness oftonal
impression, hasbeen foundto produceno fixed order of preference for
intervals, withthe remarkableexception ofthe octave.It has been dis-
covered that individual judgmentsof consonancecan be enormously
modified by training. Perceptionsof consonanceby adult standards do
not seemgenerally validfor childrenbelow theage oftwelve orthirteen,
a strongindication thatthey arelearned responses.
In musical harmony thecritical determinantof consonanceand dis-
sonance isexpectation ofmovement .... A consonantinterval is one
which soundsstable andcomplete initself, which does notproduce a
feeling ofnecessary movement to othertones. Adissonant interval causes
a restlessexpectation ofresolution, ormovement toa consonant interval.
. . . Context is the determining factor.
The resolution of intervals does not have a natural basis; it is a com-
mon response
acquired byall individualswithin a culture-area.*
It would undoubtedly simplifymatters considerablyif one could
adopt sucha completelycultural theory of consonancewithout
reservations. Yet,
in spite of recognizingthe social and cultural
nature ofmusical experience and theimportance ofcontext inthe
perception ofand responseto consonanceand dissonance, view-
points suchas this, reacting to the absolutismprevalent fortwo
thousand years,
go toofar. Suchan exclusivelycultural positionis
not only faced with the remarkable fact that the octave is a focal
point inthe musicof all cultures butwith the tendency forthe Bfth
or fourth to become substantive tones and restful, consonant inter-
vals.
Forin this
admittedly
art there
social
is one
constant;
nature
the
of humanthinking, thetendency toorganize thestimuli presented
to themind by the sensesin the simplest possible way. Toput the
matter inGestalt terms:because ofthe wayin which sounds com-
bine, thelearned responses
of thelistener, andthe contextin which
an intervalappears, aconsonance forms a stable,total entity,while
a dissonance forms aless stable,
a lesssatisfactory, though not less
necessary, Gestalt.
On this basis it would be expected thatif the
octave isa simpler,more unifiedshape, thenit would naturally
tend tobe moreconstant inchanges ofstimulation thanother in-
tervals would.Less well-integrated intervallic shapes,such asthe
third, sixth or second, would, on the other hand, presumably be
more subjectto changesin the cultural environment.Thus oncewe
leave theoctave, andperhaps thefifth and fourth, culturalfactors
such asthe accidents attendant upon the construction of instru-
ments orthe discoveriesmade in playful artistic deviation! play an
increasingly important
role in the development
of the style system.
Why the octave is perceived asa particularly stable, well-shaped
Gestalt and
why thefifth andfourth tend,as demonstrated
by their
very frequent appearance in widely different cultures, to become
normative intervalsrequires further study. The important point here
is that the modes of human perception and intellection must be
taken intoaccount if an adequatetheory of consonance and
dis-
sonance is to be developed.
It is evident, no matter what theory we adopt, that consonance
represents theelement ofnormalcy andrepose, [dissonance]
the
no lessimportant elementof irregularity and disturbance.Dis-
sonances, inshort, aretendencies. Thisbeing the case, it is not
diflicult to see thatdissonance derives its affectivepower, its ele-
gance, as Zarlino putsit, from the factthat it is a deviant, delaying
the arrivalof an expected norm, the consonance appropriate inthe
particular stylistic, musical context.
VII
The Evidence: Simultaneous and Successive

Deviation
The musicalprocesses examined so far-expressivedeviations in

pitch andrhythm, ornamentation,delays inthe expectedtonal suc-
cession, chromaticism,
and soforth-have, generallyspeaking, been
confined toa singlemelodic lineor, asin the case ofharmony, to
a group of voiceswhich wererhythmically moreor less sirnulta-
neous. Moreover,in these processes thenorms involved have been,
on thewhole, thosewhich aregiven inthe cultureand stylerather
than thosewhich, thoughrooted intradition, areestablished by the
composer-performer within the contextof thework itself.This chap-
ter considers
the relationshipof simultaneousdeviation andsuc-
cessive deviationto affectiveaesthetic responses.
The termsimul-
taneous deviation denotes those musical situations in which two
or morevoices areplayed off against oneanother, eitherrhyth-
mically, melodically,or both. The term successive deviationde-
notes those musical situations in which a norm established within
the work itself e.g., a melodicor rhythmic pattern! is repeated
with either embellishments ormajor modifications.
Of course,suc-
cessive deviationmay also include simultaneousdeviation.
These processesare still, in a broad sense of the term, types of
ornamentation, i.e.,they are rhythmic and melodic embellishments
of abasic structural
plan. Althoughthey differin meansrather than
in kind from the processes already
discussed, itis clearthat simul-
taneous variation is more closely allied to what we normally call
ornamentation, while successive variationis more closely related to
what wenormally thinkof asform.
233
Simultaneous Deviation
Simultaneous deviationis very common in the Far East. In this
type of deviation, whichis oftena kind of heterophony,the norm
itself is presented together
with oneor moredeviants whichorna-
ment it, modify its melodic outlines,and play with its rhythmic
structure. Such heterophony iscommon inboth accompanied songs
and largeorchestra pieces.Both musiciansand musicalethnologists
point outthat thisinterplay betweena basicmelodic, rhythmicand
metric structure and its simultaneous elaborationin the accompany-
ing partsis animportant source
of aestheticpleasure.
From a rhythmic andmetric pointof view the elaboratingpart
plays aroundthe chief melodic voice,now anticipatingits notes,
now coincidingwith them, and nowfalling behind them. Thede-
viating voicebreaks upthe longertones intoornate filigreesand
generally elaborates
the rhythmand meter.Thus in]avanese poetry
we find the statementthat the rebab part was characterizedby the
peculiarity thatit now anticipated the
melody, nowcoincided again
with the other instruments. 1 Sachs notes that
. . . nearly all East Asiatic accompanimentdepends onshifted phrases,

on canonlikeanticipation and retardation. The singer displaysa rich
ornamental realization of somemelodic pattern,and the player, having
this samepattern in mind, givesthe singerall the freedom requiredand
carefully triesto follow. His notescome incorrect-though not pedanti-
cally precise-order, but are delayed whenthe voice unexpectedly re-
strains its ornaments andare aheadwhen the singer dwells upon a
phrase.
The basic metric structure is regular in most of this music. Kunst
refers to the quadratic structure underlying the singers orna-
mentation and Eta Harich-Schneider notes essentially the same
thing with regard to Iapanese court music? Example 92 is a very
simple illustration of this type of rhythmic and melodic displace-
ment!
In the course of such deviations clashes often occur between the
notes ofthe structural voice and those of the ornamenting voice. To
these dissonances must be added those which occur because of em-
Simultaneous and Successive Deviation 235
bellishments, the
slight pitchwaverings and
glissandi sodear tothe
style of the Orient, and those which result from outright modifica-
tions andchanges inthe tonesof themelody. Inthis lattercategory
fall the meri-kari of Iapanese court music, which Harich-Schneider
describes asthe lowering and raisingof the pitch observedin the
parts ofthe vocalist,flute, andoboe-like reed
while the rest of the
instruments proceed in the unaltered intervals indicated in the
partbooks.
Voice
_ ro a - ni Ki IIs zu
za a- -as s
Samnsen
fa I' 5
EXAMPLE 92
These waveringaltered tones,which sound so weird and unusual

to Westernears, areconsidered bythe Iapaneseto be particularly
expressive. Thisis what we should expect. For such deviations,
delaying thearrival of substantive tones
and inhibitingthe norma-
tive coincidence
of patterns,would, according
to thepresent analysis,
give rise to affect.
In the orchestras of ]apan and ]ava simultaneous deviation is
often veryelaborate andcomplex. In]ava, for example, one group
of instruments,the singlerow bonangs, carry thenuclear theme and
occasionally add a modestornament, while other, higherinstruments
paraphrase thisbasic theme:Now they anticipate it, now they
analyze itinto smallervalues orimitate it at theoctave. Thenagain
they syncopate it .... In Kunsts words theyplay around with
the theme.It is this playof simultaneous variations againstone an-
other which creates the motion from tension to release and affords
to the trained listenera high degree ofaesthetic pleasure.
In the Far East drumming generally serves toreinforce the impor-
tant beatsin anessentially four-square
meter. Butin India the drum
not onlyarticulates themeter ofthe singersmelody butfrequently
varies it by means of cross rhythms and cross meters.Fox Strang-
Reprintpermission granted
by The Musical Quarterlyand G.Schirmer, Inc.,
copyright owners.
Emotion and Meaning in Music
waysquotesoneIndian drummerassayingthat "The beat with the
left hand! is like the seamof my coat that must be there; the other
notes with the right hand! are like the embroidery I may put ac-
cordingto my own fancy overthe seam!"'
Cross-rhythmdrumming in India, in someparts of Indonesia,and
in Africa is often extremelycomplex.Its function is to add tension
and interest by creating subtle counter rhythms and metersto those
of the melodic line. Beginningwith a coincidenceof beats,the
drummerwill play in what we would call 3/8 time, while the
instrumentalistplays his melody,which is in 3/2 time, as, for in-
stance,
in
Example
93.' inaudible
probabl
regula crotchetbeats,pp!
EXAMPLE 93
Noticethat at the beginningthe melodicinstrumentplaysa measure

of 4+ 4+ 2+ 2 or 3 !< 4= 12,while the drum playsfour groups
of three,equallingtwelve.Theycomeout togetherat the beginning
of the third measure, at a point of metric coincidence or "conso-
nance,"as Sachsputs it. Later the drum plays in units of eight
+ 3+ 2!, while the melodycontinuesin twelve + 4+ 2+
2!. Here the metric consonancetakesplace only after two measures:
i.e., in the melody 2 !< 12 = 24, while in the drum 3 !< 8 = 24.
The crossingof the rhythm and meter servesnot only to create
tension within the line but alsoto articulate a simple type of over-all
strophicform. That is, the melodyremainsconstant;but the rhythm
and meterwhich accompanyeachstatementof the tune modify it,
giving rise to a seriesof variations.
This is a fairly simpleexampleof an extremelycomplexpractice
and theory.' Such crossingaffords great delight to the Indian audi-
ence, particularly when it involves intricate deviations from the
o The exampleis given by permissionof The ClarendonPress,Oxford, pub-
lishers.
standard patternor the introduction of new ones.In all this there is a
strong element
of play. The singer and drummerlike to play hide
and seek with each other; and the audience watch the contest with
amusement. 1
The sametype of contest isdescribed insome detailby Sargeant
and Lahiri:
It often happens amongIndian Musiciansthat a vina player and a

drummer willengage ina friendly contest tosee whichcan confusethe
other intolosing trackof the sam [thestructural beat].The vina player
will improvisea melody.The drummer,to whomthe melodyis of course
unfamiliar, mustfirst discoverwhere the sam is. Having discoveredit
he startshis drumming.For a while everythingruns smoothly.Then the
syncopation andcross rhythmbecome verycomplex. Thevina player
uses all sorts of ruses to disguise the sam .... The drummer will mean-
while seekto confusehis opponentby insistingon his cross rhythmsas
though theywere the true basicmeter, playingmeters ofseven, orfive
against thelatters four or three and soon. Eachone striveswith might
and mainto retain his equilibrium.Eventually oneor the other misses
the sam and is worsted.11
The relationshipof such contests tothe creationand conquestof

diiliculties, tothe processof play,is clear.And play is, aswe have
seen p. 70!, but one of the ways of looking at the affective
aesthetic musicalexperience. Thereare, moreover,obvious similari-
ties between
this accountof musicalexperience and
those givenby
Raguenet, Glarean,and Dance see p. 208!. Indeed Sargeant
and Lahiri themselves point out the resemblance betweenthe ex-
periences ofHindu music and jazz.
This typeof syncopation, which isvery muchlike that employed by]azz
players in their "breaks," except thatit is infinitely more complicated,
is an essential partof the technique ofgoth. The oina player delights
in apparentlylosing himselfin the most abstruse counter-rhythms, leaving
the listener with a sense of utter bewilderment, only to issue forth trium-
phantly at the sam again without a hairs breadth of inaccuracy, and
with a sparkle ofobvious satisfaction.The effect,to one who is accus-
tomed tothis idiom of expression,is that of being hurled throughchaos,
and then suddenly landingright side up on terra firma again with no
bones broken and afeeling ofintense relief.
Not all Indian music has a drum accompaniment. But even in
unaccompanied music an underlyingbeat, against
which thesinger
or instrumentalistperforms intricaterhythmic and metric devia-
tions, is presupposed. Thiscan be seen in an example taken from
Sargeants article see Example 94!. The astai or at home is the
main theme and the tons are short sectionsof varied rhythmic and
metric material which arealways ofa certainlength andmetric struc-
ture. Themetric structure or talaupon whichthe tanis built consists
in this case offour beatsarranged thus:1 - 2 - 3 - 0 not sounded!.
This metricstructure isrepeated fourtimes duringeach statement
of the astai andfive timesduring the particular tangiven in this
example. The all-important sam or structuralguide beatfalls onthe
second beat of the tala, onthe firstimportant beatof the astai, and
Astai
tg|qLg $0|9292gl g1 g _.4 li2 , .3 e

_p ___ Tanll'
0I , ,Z 3 p! 0 I ri, , 24 s
L ATT'
I Y'Y`YÌ_
I l l II I L afq lll
False Sam False Sam
,2 '3 ,g o1 ,,z 3 _'

Astai
1 ]g g ugReal Sam 0 il v ,gm

EXAMPLE 94
on thefirst beatof thetan. Inthis passage

the oinaplayer mustcon-
stantly keepthis basicunderlying organizationin mind while play-
ing metersof 3/8, 4/8, 5/8, and 4/4 against it.Similarly theaudi-
ence mustkeep trackof thetala. Noticethat in the tangiven inthis
example theperformer playswith our expectations byplaying a
false sam, a devicevery similarto the so-called false recapitula-
Reprintpermission granted
by The Musical Quarterlyand G.Schirmer, Inc.,
copyright owners.
tion in sonata form.What happensis that the performerplays the
melodic patternwhich indicatesa return to the astai butbecause
the patternoccurs onthe wrongbeat, on1 insteadof 2,which isthe
sam, the return is false. Only when the relationship between
the tala and the melody is correct doesthe astai really return. Thus
by raising our expectations,
disappointing them,and finally satis-
fying them,the composer-performer
molds anexciting andmoving
experience.
Not all oriental music, of course,has sucha complexand intricate
rhythmic metric organization. For instance, in the preludes, which
so oftenprecede more stable, traditional
forms, therhythmic-metric
structure isfree, almostrhapsodic. Thisis trueof thenetori of]apa-
nese courtmusic, theIndian cilpa,the Arabian taqsim, andthe
Indonesian bebuka.
Unfortunately little of the extensive researchdone in the Held of
primitive musicis of value forthis study.First, because
the primi-
tives themselves do not make musical creation a self-conscious
endeavor, theyhave neithera theory of music nor evena crude
aesthetic which might serve to connect their musical practices to
their responses.It seemsclear that on the most primitive level music
is, on the one hand so intimately connected with ritual and magic
that its aesthetic contentis severelyrestricted and,on the other
hand, thatit is so closelyassociated with
bodily effortthat its shape
and organizationare to a considerabledegree productsof the phys-
ical activities connected with ritual, labor, or expressivebehavior.
And second,because musicethnologists havetended to collect and
classify tunesand instruments,compile statistics,and concernthem-
selves withthe sociologyof primitivemusic, theaesthetic meaning,
if any,which musichas forthe primitivemusician orlistener, has
for
the most part been ignored.
However, African music, which can be called primitive only with
great misgivings, has reacheda very high state of development. It
is especiallycomplex and subtle in its rhythmic organization.
The very essence ofAfrican musicis to cross therhythms. Thisdoes
not mean syncopation. Onthe whole African music is not based on
syncopation. Toput it in its simplest terms,what we mean bycrossing
the rhythmsis that if, say,two Africansare drummingin triple time and
at the same speede.g. in 3/8 time where adotted crochetequals 140!
the main beat of the bar of No. 2 drum will fall either on the second
beat of the bar of No. 1 drum or on the third, but never on the first. This
is absolutelyfundamental toAfrican music.In actual practice, ina large
ensemble, some instruments andsingers do actually maketheir main-
beats coincide. But this is only duplication ofa rhythmand it is possible
only because some otherperformers arealready crossing the beat.
If we look atan exampleof this,the wholeprocess becomes
quite
clear andits complexities
apparent. In the following,three drums
-the Nzupe-nzupe,the Cibitiku,and theKambape-play anintro-
duction knownas sefa.The Nzupe-nzupeN! begins alone beating
a 3/8group, as in Example95.
NlR L R
EXAMPLE 95
Then theCibitiku C ! alsobeating a3/8 patterncomes inin sucha

way thatthe firstbeat ofhis patterncoincides withthe lastbeat of
the N drum Example 96!.
cm-34| LJ M
C!
RLR RLR R LR
EXAMPLE96
This formsthe foundationwhich runs not only through sefabut

through thefollowing songand danceas well.
The Kambape K! plays eighths whichare equal in value to
those playedby the other drums,but they form a group of 2/4
notes ratherthan oneof 3/8. This givesus thetotal patternfor the
introduction Example 97!. Notice that, although the main beats
of all three never coincide, the C and K drums do arrive at a coinci-
dence at every twelfth eighth note.
The sefacontinues untilthe leaderof the dance gives
a signalfor
the beginningof the song anddance. Onthe last playing of the
Reprint permission grantedby African Studies, Universityof the Wit-
watersrand, johannesburg,
Africa, copyrightowners.
N!
K!
2!
EXAMPLE 97
sefa,the K drum lengthenshis beatfrom a quarternoteto a dotted

quarter in order to preparefor the rhythm which he will play
during the song and dance.The song and danceenter in 12/8,
which is really composedof 8 ><3/16 = 24/16. We now have four
differentpatternsgoingon simultaneouslyin sucha way that the
main beatsof three of the Bgurescorrespondeverytwelfth eighth
note seeExample98!. Later in the songthe organizationbecomes
even more complexbecausethe rhythm of the K drum changes
when the chorus enters.
Sefa.J 140 Lasttime

N!
C!
K!
5ongandDance
Cantor:A Msolo, Mfu-muyanu ndani? Chorus:

Eeya
EXAMPLE 98
It is clear that the organizationof this and other African pieces

bearsa strongresemblance to certaintypesof Indian andIndonesian
music. We Bnd here the same movement from complication of
rhythm and meter to greater unity and coincidence that was found
in Indian and other oriental music. And it is interesting to note that
]ones, like Sachs, thoughevidently quite independently ofhim,
makes ananalogy between the roleof the temporal organization
in
African music and the role of harmony in Western music.
All this rhythm-crossing isthe spiceof life to the African. It is his real
harmony. Heis intoxicatedby this rhythmic harmony,or rhythmic po-
lyphony, justas wereact tochordal harmony. It is this remarkableinter-
play of main-beats thatcauses him irresistibly, whenhe hearsthe drums,
to start moving his feet, his arms, hiswhole body.This to him is real
music.
Hornbostel impliesmuch the same sortof thing when he writes

that: The combination of binary and ternary time is characteristic
of African metre in general. Theprincipal divisionis two-fold: a
period breaksup into question andanswer, tension and relaxation,
arsis and thesis .... 1
Richard Watermansdiscussion ofoff-beat rhythmsin African
music confirms directly orby implicationmany ofthe observations
made inthis study.Waterman begins by observingthat in order for
there to be off-beating, a seriesof regularly recurring pulses,a
normative beat, must become established in the mind of the listener.
This beat,which Watermancalls themetronome sense,is primar-
ily mental:The regular recurrence ofrhythmic awareness involves
cxpcctancy.17 The off-beats are
deviants fromthis normativepulse.
They mustbe irregularor elsethey toowill becomenormative; for
complete off-beatinghas thesame eHect as completelack of off-
beat patterns;it is in this sense meaningless.1 And, finally, the
following statement, which cannotfail to recall thoseof Raguenet,
Clarean, Dance, and Sargeant, relates theprocess ofoff-beating to
the overcomingof difficulties, to uncertainty, and through these
to affectiveaesthetic experience:The off-beat phrasing ofaccents,
then, mustthreaten, butnever completely destroy, theorientation of
the listenerssubjective metronome. 1
Simultaneous variation is not common in most folk music, which
is essentiallysoloistic. But in jazz it is of the essence. In
jazz we
find thesame use of abasic beatagainst whichcounter rhythms and
opposing metricpatterns areplayed aswe saw in Indian music. The
procedures ofjazz arevery similar to the technique ofgoth: the
main melody,the jazztune, corresponds to the astai, anda break
in jazz corresponds toa tan in the technique ofgoth. Thetune or
astai alternates
with the breaks or tons againstthe background of
a recurringpattern. In jazz thisbackground, which is both metric
and harmonic,is calleda riff In India this background, which is
metric, iscalled thetala or tintal. Aswas thecase withgath tech-
nique, suspense and uncertaintyplay an important partin the re-
sponse tojazz.
The break,then, isa temporarylapse fromthe rigorsof strict structure,
in which logic is momentarily suspended and improvisatorychaos reigns.
Its effectis to heighten theelement ofsuspense and unrest. Thelistener
is thrown for the moment onunmapped confusingground. Thebasic
rhythm ceases to offer its familiarthumping landmarks.
The solodangles
dizzily without hope of support, andthen, just as thelistener hasabout
abandoned hope of reorientinghimself, thefundamental rhythmresumes
its orderlyway, anda feelingof relief ensues.
In this process thefundamental rhythmis not really destroyed.The
perceptive listenerholds in his mind a continuationof its regular pulse
even thoughthe orchestrahas stoppedmarking it.... The situation
during the silent pulsesis one that challengesthe listener to hold his
bearings. Ifhe hasany sortof rhythmic sense hewill not be contentto
lose himself.
In spite of the similarity between

the processesdescribed inthis
quotation andthose discussed by Waterman,the procedures
of jazz
are somewhatdifferent from those existingin African music. In
jazz aclear andquite regularnorm the tune! is generally estab-
lished beforeany rhythmicelaboration takes place. Thismakes the
break a clear deviation from a norm. Thus the expectation of a
return toregularity isa basicorganizing principlein this music. In
this respectjazz is,in a sense, more
like Hindu music inwhich the
astai alternates
with the tan. Inmost Africanmusic tension operates
within the phrase and is the organizing, articulating principle of
the phrase. But tension does not perform an over-all organizing
function, save in the sense thatit creates a type of variation form.
In other words, the level of tension is relatively constant between
parts, thoughnot within them.
Simultaneous melodicdeviation ispresent in African music as it is
in japaneseand Iavanesemusic. TheChopi musiciansof Portuguese
East Africaplay aheterophonic type
of musicon xylophones
which
is very similar to that of ]ava. After the composer-performer has
developed atune andits accompaniment, the otherplayers joinin
with their own versionsof the tune. Hugh Tracey reported that he
heard at least four variations on the basic melody in the front rank
alone. 21 Hornbostel gives a particularly interesting example in
which the flute, whichperforms themain deviations, plays a fifth
higher thanthe fundamentalmelodic linewhich is carried bythe
chorus. Speaking of the flute part, Hornbostel notes
that: _ . . it
follows the vocal part only half way and even therenot strictly,
indulging in playful deviations;then insteadof going on to the
second half, it repeats the first one, thus transforming the tune into
an ostinato. 23 A wind instrument follows the tune in so far as its
restricted repertoryof tones will allow-performing slight devia-
tions either of necessityor design-while a horn plays adrone onthe
tonic, apparentlyemploying acounter rhythmin duplemeter against
the triple meter ofthe mainparts.
Simultaneous deviation in the form of cross meters has not been

common in Western music since the Renaissance, that is, until the
twentieth century.Because aharmony must,almost bydefinition,
be perceivedas a single entityrather than an interplaybetween
independent coexisting
entities, therise of tonal harmonymeant an
ever increasingemphasis uponvertical simultaneity,upon a co-
incidence of main beats.
However, before the rise of tonal harmony, there was a consider-
able amountof metricand rhythmiccrossing inWestern music. The
beginning ofConverts chanson,
Se Mieulx ne Vient Copenhagen
MS, Thott2918!, forinstance, provides
a rather simple illustration
of what was oftena verycomplex practicein the fifteenth century
Example 99!.2 The presenceof crossing-of the oppositionof
, L3 _. 3" .. Q aa.
Y we * a/1+ _' __ /v a/4 ._

S. L3 j gg-E gg
. as _. =u+ .Jaw #mn ._ =w _
EXAMPLE 99
Simultaneous and Successioe Deoiation
6/8 and 3/4 metersand subsequently of 3 !< 2/4 against3/4 plus
6/8 is obviousin this example.Sinceunfortunatelythe significance
of such metric crossing was not discussed in the theoretical or
critical writings of the fifteenthcentury,we have nothingthat re-
lates these metric proceduresto affective aesthetic experience.But
one cannot help feeling that this music moves through increasing
complexityto a metric"consonance"
on the B andG.
We find only a few examplesof metric crossingbetweenthe
Renaissanceand the twentiethcentury see,for instance,Example
37!. This does not mean that the resources of rhythm were not
utilized during this period.Metric organizationand deviationtook
otherforms,e.g.,the delays,anticipations,
gaps,and incompleteness,
whether of the main melodic line or of the over-all vertical organiza-
tion, which werediscussed
at somelengthin chapteriv.
The twentieth century has from its very beginning sought and
found inspiration in the resourcesof rhythm and meter. This,
coupledwith a tendencytowarda morelinear style of composition,
hasled to morefrequentuseof crossmeters,as,for instancein Ex-
ample 100 from the "SecondaParte" of Bartok's Third String Quar-
EXAMPLE 100
tet. Here,as can clearlybe seenfrom the abstractof the rhyt.hmic

pattern given in Example 101, the 6/8 and 3/8 meters of the cello
Vla.
Vla.
Ve,
EXAMPLE 101
o Copyright 1929by Universal-Edition.Copyright assigned1939to Boosey5
Hawkes,Ltd. Usedby permission.
and viola are playedoff againstthe 3/4 meter of the first violin,
which is itself crossed by that of the second violin.
What interms ofcontemporary style in generaland thatof Bartok
in particularis the significance ofsuch rhythmiccrossing? Does it
represent a breathing in and out of the phrase? Dowe feel a
motion from tension toexpected release? In this case, atleast, it
appears thatthe rhythmiccrossing isa normof the movement; for
not only does themodus operandiillustrated abovecontinue for
some timebut rhythmiccrossing ispresent almost at the outset of
the movement, where Bartok indicates the coincidence of two sep-
arate meters
by hisnotation.
We cannotas yetgeneralize aboutthe useof simultaneous rhyth-
mic deviationin contemporarymusic. Thenorms ofthe new style
have notyet clearlyemerged; and
it is entirely possible
that when
they do, simultaneous metricdeviation will be found to be a norm
of thenew style.At presentwe candetermine theintended effect
of
cross rhythmsonly by the normativeor deviantcharacter ofother
devices used
concurrently withthem.
Successive Deviation
Every pieceof musicestablishes norms-the melodies, rhythmic
figures, instrumentalgroups, harmonic progressions, etc.,
created by
the composer within the specific stylistic
context-which arepecul-
iar to that particularwork. Suchintra-opus norms may embodythe
stylistic normsupon whichthey dependor theymay themselves be
deviations from those norms. The Tristan motive, for instance, is
replete with deviations fromthe diatonic norm which it presupposes
and involvessignificant modifications
of the normal simultaneity
of
harmonic progression.Thus the basic materialspresented by a par-
ticular work may at times be said to involve successivedeviation
from the outset, in the sensethat the materials embody deviations
from the norms of the stylistic universe of discourse whichis always
prior to any particular work.
Often the stylistic universe of discourse contains alternative
norms, as for example where several melodic or rhythmic modes
are more or less equally normative within a style. When this is the
case, theparticular modeto be used maybe rehearsed in a kind of
prelude inorder toestablish the particular normin the minds ofthe
composer and listener alike.This is one of the main functions of
the Indian iilpo, the Arabian maqm, the Indonesian bebuka, the
preludes to more sophisticatedprimitive music, and the toccatas,
fantasias, and introductions found in the music of the West. The im-
portance and
significance of
such preluding-a practice found in al-
most every
type ofmusic-lie in the factthat theysupport byimplica-
tion the hypothesis thatthe processof deviation from habit-expected
norms is one of the basic forces shaping and articulating musical
experience. For in all the statements about such preludes it is
made clear
that thepreludes serveto establishthe normswith which
the mainpiece will operate andfrom which it will, in one way or
another, deviate.The norms thus established facilitate the percep-
tion and response tolater deviations and are therefore a necessary
condition forthe arousalof affectand objectifiedmeaning.
That such introductions involve considerable melodicand rhyth-
mic freedomand that they do not as a rule present well-shaped,
substantive patterns leads one to speculate whether they do not
perform aformal function,arousing tension
and expectation.
There
are several
reasons forsupposing that
they doso. In the Hrstplace,
in the particular cultural context the audience knowsthat the lcipa,
netori, sefa,etc., is not the real piece but is only a precursor of the
main event,which more often than not is known to involve dancing,
singing, orboth. Andit is only naturalto supposethat thelisteners
look forwardto andexpect thereal piece.Second, as we haveseen,
a stimulus series which is not as well shaped asmight be expected
within the given stylisticcontext arouses
an expectationof better
articulation and more palpablepatterns.
Once such a prelude or introduction has been presented, the
norms of rhythm, melody, and harmony specific to the particular
work areusually presented.
These aregenerally speaking
not only
relatively well structured in and of themselves but seem especially
so becausethey have been precededby the more weakly articulated
patterns of the introduction. These normsthen becomethe basisfor
subsequent deviations.
Unfortunately both music theorists and ethnologists have mainly
been concernedwith stylistic descriptions andtabulations rather
than with aesthetic effect.
They havethus tendedto discussthe
process andsigniiicance of
successive deviationonly in the broadest
terms. Theysimply tellus thatthe folk singer, theprimitive musi-
cian, andthe orientalcomposer dovary theirmaterials inan elabo-
rate andmore or less consciousway, and they then describe and
illustrate the processes involved.
But of the affective aesthetic mean-
ing of these processes
to the artists wehear comparatively
little.
Almost all writings aboutoriental musicstress thepresence of
variation in this music and acknowledgeits importance in the sev-
eral stylesconsidered. Although the greatestvariations deviations!
from inter-opusnorms generally occur towardthe middleof a song
or instrumentalpiece, exact repetition evenat thebeginning orend
is almostunknown. Veryoften, moreover, one getsthe impression
that, asin folk music, thereal normis not actually presented
by the
composer-performer but rather exists in the minds and habit re-
sponses ofthe musiciansand the audience. In other words,the
opening motivesare themselvesvariations-deviations from a
cultural stylisticideal type.
The problemin dealingwith thismaterial isone ofinterpretation.
The factsare quite clear. Evena cursoryglance atthis ]apanese
song, given in Example102, reveals
the presenceof successivedevi-
ation variation! between thethree versesquoted.
J=so
A __ __ N ,, _
_= ._5=EEEEEEEE5:`T5EEEEE=E!-E='=Ei
Kato
- A"
_ lllff 1_-
EXAMPLE 102
Reprint permission granted

by the Harvard UniversityPress, publishers.
To Western ears it seems clear that such successive deviation-the
displacement of
the rhythmand meterand thevariation inthe tone
order-constitutes an affective aesthetic intensification; but in the ab-
sence ofprecise testimonyfrom those whose habit responses areat-
tuned to this style, such an interpretation might easily involve a
misunderstanding of the meaningof the passage.
Though itis difficult to find written evidenceto directly support
interpretations ofthe affectiveaesthetic significanceof suchdevia-
tions, itis possibleto utilize other indirect,corroborative evidence.
For differenttypes andlevels ofdeviation tendto appearin concert,
supporting oneanother inthe evocationof the affective aesthetic
experience. Expressive pitch deviations,for instance,are usually
employed in coincidence withmarked ornamentation, as in the
music of India. Discontinuity of melodic and rhythmic patterns and
the weakeningof shape,are usuallyaccompanied bychromatic se-
quences or
ambiguity, asin Westernmusic. Inlike mannerit seems
generally true that successivedeviation tends to involve increased
ornamentation, an intensification of simultaneous deviation, etc. And
it is reasonable to suppose that since successivedeviation is com-
monly associatedwith other types of deviation, which can be shown
to have an affective aesthetic function, successive deviation itself
performs asimilar function.
The process and effectof successive deviation isespecially clear
in themusic ofIndia. In the followingsouth Indianpiece Example
103!, thepallaoi the south Indianterm for astai! meaninggerm
or sprout consists ofa three-measure melody A!, a conjunctive
measure x!, and a repetition of the three-measuretune A'!.2 It is
clear that only slight modiications of the tune occur in the first total
repetition !; the most important of these, for present purposes,
being the fact that the tone D occurs in the first measureinstead of
in thesecond. But
in thesecond repetitionof thetune ! the open-
ing measureof the Erst variationa"! is expanded intoEve meas-
ures. This prolongation performs an aesthetic function, not only by
deviating from the inter-opus norm establishedin the earlier state-
ments of the pallaoi, but also because it serves to delay the arrival
of the expected substantivetone C, which is repeatedly implied by
the descending scales g-f-e and f-e-d. It also seems probablethat
thefirst,extended
halfof thepallaoishouldbeconsidered
astending
toward the unextended, normative second half, thus creating a
hierarchyof tendencies.
In thethird variation! thefirstmeasure
of the themeis expanded still further to ninemeasures!,
partly
through
theaddition
of anupbeat
whichdelays
thearrivalof the
opening
motive.Suchsuccessive
deviations
arerelatedby Strang-
waysto affectandaesthetic
play."
A' Al
Itttvocluction
%mmmm
EXAMPLE 103 4
In JewishandByzantine
chantof theNearEastsuccessive
devia-
tion consistsmore in the arrangementof brief melodicformulas,
and the ornamentation of these,than in variationof the type found
in Indian or Far Eastern music."
The examination of the structure of the melodies has made it clear that
theyare all built up of a limitednumberof formulas,shortgroupsof
notesthataresignificant of themode,the echosof themelody.
The fact that the vast numberof Byzantinemelodiescan be reduced
to a limited numberof archetypesmay lead to a wrongjudgmentof the
creativequalitiesof Byzantine
composers.... The melodicarchetypes
he had to use and combinewereto his mind the apechem<the echoof
o Bypermission
of TheClarendon
Press,Oxford.
the divine hymns. The work of the composerconsisted ingiving the
melodies anew frameby linkingthem together.
Elsewhere Welleszrelates this type of successive deviationto aes-
thetic pleasure.
The advantageof this kind of technique isobvious. Thecongregation
heard the well-known musicalphrases inevery new Sticheron, butar-
ranged in a different way, and connected bynew transitionalpassages.
They musthave takenpleasure inhearing musicalphrases whichwere
familiar but were linkedtogether inany unexpected way, just as amod-
ern audiencetakes pleasure in the recurrence ofthe themesin a move-
ment ofa symphony.
The pleasure
taken inthe adroitintroduction ofnew melodiesand
new motives seems to be more than supposition qn Wellesz part.
Idelsohn notes
that theprocess of
skilfully introducinga newmelody
to vary the chant is considered the highest art among Yemenite
singers.
This typeof composition
is characteristic
not onlyof NearEastern
chant but of a good deal of the folk music of all countries. In all
folksongs, inthe art of the German Meistersinger,
in Lut_herscho-
rales, in Calvins Psalter,and way back in Gregorian chant,the
mosaic isquite obvious.35
As wasthe casein oriental music, theartistic performanceof even
a singleverse ofa folk song, ina sense,involves successive
deviation.
For the folk song, as distinguishedfrom any particular realization
of it, is anideal typewhich exists
as anorm not an average!in the
minds of the singer and listener alike; and the folk singers per-
formance is, if not an embellishment, at least a deviation from this
ideal type.To put the mattersomewhat differently:
the normative
ideal type is never actually performed but exists only in the collec-
tive consciousness of the group, in tradition.
Yet the folk singer is often aware of what is tune and what is em-
bellishment. In an interesting article on folk song variants Helen
Roberts discussesthis whole process together with its aesthetic
affective significance.
Working in ]amaica, Roberts found that the folk singers them-
selves recognizethe differencebetween thefundamental substratum
of the oral material handed down to them and their own creative
actualization ofthat material.Parts of songs maybe repeatedad
libitum, and this processof freely repeating a part is given special
semantic recognition: it is called doubling. It is in doubling that
the most freedom of embellishment is permitted. The natives End
particular pleasure in these improvisations and report that such
repetitions mekit sweet. 36 Notonly is the singerpermitted free-
dom with respect to repetition, but he also has considerablelatitude
with regard to the order of the severalphrases ofthe song.One
singer toldRoberts: Youcan changeit aroun, you know,an sing
about the akee in de middle or at de en. 37
Embellishments also receive recognition as facets of artistic ex-
pression. Themore noteworthy deviations are referred to as flour-
ishes. Accordingto Roberts,the techniqueof embellishment,of
flourishing, has reached a level of consciousness which almost
amounts toextemporaneous composition.38
Evidence thatsuch songs
are indeeda kind of ideal type is fur-
nished bythe fact that the natives themselvesbelieve that the thread
of the song is all one needs to maintain the identity of a song. This
process ofvariation isoften morenoticeable inthe performance
of
instrumental music where the presence ofa text does not act either
as a restraining, conserving influence or as a kind of automatic
cause ofslight deviation.One fluteplayer heardby Robertswas a
particularly accomplishedmusician, andher reportof his perform-
ance ismost interesting.
No onewould havebeen morecapable ofplaying a part overexactly,

yet this he seldomseemed ableto do. The urge to embellishand play
with the tune was greater than any to reproduce it exactly. I would
whistle the two versionsto him and emphasizethe differences,which
he would readily perceive, but when asked which was correct he would
laughingly reply that it made no difference if one had the thread. He
said the flourishes werenot the tune proper,which always, or nearly
always! remained the same, and so it did within limits that are rather
difficult to define.
The relationship between deviation and aesthetic pleasure and

aesthetic play! is emphasizedthroughout this article. According to
Roberts, suchfolk musiciansweave invariations at every oppor-
tunity. Themore clever,the morepleasure theygive. 4Changes in
detail were welcomed with delight and it was in these that the
individual expressedhis own self. 41
Much ofthis materialduplicates what
has alreadybeen discussed
in connection with ornamentation. For in the case of folk material,
successive deviationis also successive embellishment.
By way of
illustration, Example104 presentsBart6ks transcriptionof two
verses ofa folk song togetherwith what he hasanalyzed tobe the
basic structure underlying the elaborately embellishedversion
which the singer actuallypresents. In this example,at least, it
seems quite
clear that
successive ornamentation
is nota directprod-
uct ofchanges text.
in The role oftext changes
in determining orna-
mentation isoften overemphasized
since itis almostas common to
embellish andvary two verses whichhave preciselythe sametext
J!
. SEE: _ .._...
:__ -_ -=_.:: ui ___.,1-__
fun- ,......f
!vl;i1$i1_:::!1!_:n1--11_:_llr_.-115.11
rmiix
|r.:11.~lL '11'|92r1:|::_|_luu
1.31 11111 rnniin-_|!111
'KU' 4 _ 4__TC rr92,r|1 A Ill g*_T A~r' - T `-
nrrlru
; 11
C_ ii z 92 I
92
II; 92 92
5122.
92&ll0|l l JJ
U |U u
.
8P
"l'lT1 'tri S!-'a 'O . ;QQ- u li"-1 g
TQ:-:_ì 1 -1-'f--#_ :|
l'l_-i;`_1iiiilllill111Il11iQl1_lTIi
-'lm-' "_" Y:-2:-|__:" tl;
--'Q-|-I--!-Z_:
,{;.92-1n|||u1;.li
t im,-cn: sl: 1 rr:-92.i1:sa-;192||-1 ru-:ru-I I'_l1'1`_
I-I 1-|11
`92_ v. 4 92 vA -1. --
..-1 2
_- ~--- -
J
EXAMPLE 104
as it is to vary theornaments applied

to verseswith differenttexts,
though, ofcourse, texturalchanges willintensify thetendency to
vary ornamentation.
Real jazz,as mostwriters have recognized, isa kind of folk music
involving bothsimultaneous andsuccessive improvisationupon a
basic groundplan. Thisground planis essentially
harmonic, though
the specific tune used as the basis for variation may also be an im-
portant departure point for embellishment anddeviation.
Hot jazz melody is improvisatory, but its structure is held to a coherent
formal pattern which restrainsit from complete chaos.This coherent
pattern isprovided bythe harmonicsequences of the underlyingaccom-
paniment .... It is the simple harmonic phrase. . . that provides
the unifying influence in hot jazz improvisation .... This phrase is
repeated overand over again, with occasional interpolations,perhaps,
Given by permissionof the Columbia UniversityPress.
of similarchordic sequences,
forming asort ofostinato on which melodic
and rhythmicvariations arebuilt .... At eachvariation ofthe harmonic
phrase anew melodicand rhythmicsuperstructure isimprovised bythe
hot player.
Here, too, the aesthetic effect of the music depends both upon the
listeners awarenessof the basic ground plan, which is the norm
from which deviations are made, and upon his ability to compare
the successive
variations as
they followone another.
It is rather difficult to establish a relationship between successive
deviation and affective responsesin the realm of primitive music
see p.239!. Yetmany observers
have notedthe factthat primitive
musicians doderive aesthetic pleasure from music and particularly
from the process ofvariation. Willard Rhodes, forinstance, dis-
cusses theaesthetic play
interest whichsingers ofdifferent American
Indian tribesexhibited ineach otherssongs andespecially inthe
same song performed bysingers ofdifferent tribes. The aesthetic
importance ofsuccessive deviationin primitive music isalso noted
by ]ones:
But it is quite wrong to think that the tunesare repeatedover andover
again withno variationat all. There isvariation: it is frequentand it is
subtle. Aslight changehere, anextra notethere, makeall the difference
to thosewho know what they are listeningto .... It is all a question
of nuance:and it is only the practicedEuropean listenerwho can per-
ceive and enjoy this cunning compoundof bold repetition and subtle
variation.
INDIRECT EVIDENCE
The evidenceadvanced in support of the hypothesis that musical
meaning, whether
affective oraesthetic, arises
when atendency to
respond is inhibited is not confined to chapters vi and vii. These
chapters presentwhat might be called direct evidence. This mate-
rial has consistently demonstratedthe connection between the inhi-
bition oftendencies deviation!
and theaffective aesthetic
response
And while this evidenceis not exhaustive, itis clearly representative
The centralthesis ofthis bookis also supported byless direct,
though not less convincing or important, considerations. First, the
very factthat it has beenable tofurnish abasis forthe analysisof
music of very different styles and different cultural levels is per-

suasive. For it indicates that the account presentedhas reached a
workable level of generality.
Second, thehypothesis advanced has providedthe basisfor a
reasonable and consistent account
of manyproblems heretofore un-
solved orignored. Forinstance, ithas ledto a new andfruitful use
of Gestaltconcepts inaesthetic analysis;it has accounted forthe
affectivity ofthe minormode inWestern music without recourseto
mathematical mysticism. Employing corollariesof the hypothesis
we have been able to understand and explain processes previously
merely described;for instance, the filling in of tonal systems the
tendency towardequidistance!, the use of ornaments, andthe intro-
duction of new tones at the end of a musical pattern, and the func-
tion of poorly definedpattern processes.
Finally it is important to emphasizethat a theory of music does
not exist in a kind of splendid, irrelevant isolation. If it is to be
fruitful, musictheory mustnot only be internallyconsistent butit
must also be consistent with and relevant to concepts andtheories
in other realms ofthought. Thusit is significant thatmany of the
concepts presented
in this book haveclear counterpartsin the theory
of gamesand in information theory.To cite only one instance of
this: it seems possibleto equatethe inhibitionof a tendency, which
of necessitygives riseto uncertaintyand anawareness of alternative
consequents, withthe conceptof entropyin informationtheory.
A theoryis valuable,not only for conclusionsit reachesand the
phenomena explains,
it butalso forthe questions
and discoveriesto
which it leads. If the ideas presented in this book can lead to new
questions andthrough them to new answers, if they can lead to re-
formulations ofold questionsand throughthese tonew methodolo-
gies, andif they can lead to a more fruitful analysis andcriticism of
music, this will be their best ultimate validation.
VIII
Note onImage Processes,

Connotations,
and Moods
Image Processesand Agectioe Experience

The affectiveexperiences thus
far discussedresult froma direct
interaction between a series of musical stimuli and an individual
who understands the style of the work being heard. Because the
forces shapingsuch anexperience are exclusively musical,
the form
of theaffective experience
will be similar tothe formof themusical
work which brought it into being.
Not all affective experiences
are as direct as this. Often music
arouses affect through the mediation of conscious connotation or
unconscious image processes. sight,
A asound, ora fragranceevokes
half-forgotten thoughts
of persons,places, andexperiences; stirs
up
dreams mixing memory with desire; or awakens consciousconno-
tations of referential things. These imaginings, whether conscious
or unconscious, are the stimuli to which the affective responseis
really made.In short, music maygive riseto imagesand trainsof
thought which, because of their relation to the inner life of the
particular individual,may eventuallyculminate inaffect.
But if such imageprocesses are
really unconscious,we cannever
know them.
. . . only feeling penetrates

into awareness,
a feeling aroused bysome-
thing of which the subject isquite ignorant.Self-conscious mindsseem
to havea repugnancefor suchisolated disembodied mental phenomena:
they arefelt to be morbidand eerie.Consequently process
a ofrationali-
zation is undertaken at once. W'hatever is in the focus of attention at the
moment when the affect arises is held to be the direct cause of it.1
256
Note on Image Processes,Connotations, Moods 257
Thus many affective experiencesattributed directly to musical stim-
uli mayin point of fact be theproducts ofunconscious image
proc-
esses. Becauseneither we nor the subject himself can know any-
thing about such unconscious
image processes
any discussionof
such anexperience isclearly impossible.
Often, however, image processesare conscious. The listener is
aware of the associationswhich he makes while listening. Conscious
image processes
may beeither private,relating onlyto the peculiar
experiences of
a particularindividual, orthey maybe collective,in
the sense that they are common to a whole group of individuals
within a culture. The image processes
of a whole community will be
referred to as connotations.
Private images,even whenthey are brought to consciousness
without psychicdistortion, areproblematical because it is almost
impossible totrace the relationships existing
either betweenthe
musical stimulus and the image processesaroused or between the
image processes and the resultant affect.The peculiar experience of
an individual may, for example, causea happy tune to be associ-
ated with images ofa sadoccasion.
Even where the original association appearsto be relevant and
appropriate to the character of the music being played, affective
experience may
be aresult ofthe privatemeaning whichthe image
has for the particular listener. For example, the image of a trium-
phal procession might within a given culture be relevant to the
character of a piece of music; but the association mightfor private
reasons arousefeelings of humiliation or defeat. Thus while the
image itself is relevant to the music, the significance which it has
for the particular individual is purely personal.
Image processes,whether private or collective, are tremendous
temptations toward extramusical diversion. For an image, even
though originally relevant to a particular passage, mayitself initiate
further image processes. Thedevelopment andproliferation of these
may, however, proceed without reference to the subsequent suc-
cessions ofmusical stimuli. That is, one image may follow another,
not because of the associations whichobtain between the images
and the progress of the music, but because of the associations in
the mind of the listener betweenthe imagesthemselves.
Neither the form nor the referential content of such experiences,
however affective
they maybe, haveany necessary
relationship to
the form and content of the musical work which presumably acti-
vated them.The real stimulus isnot the progressive unfolding
of
the musical structure but the subjective content of the listeners
mind.
Yet, in spite of the many and cogentobjections whichcan be
leveled againstthe relevanceof suchresponses, itseems probable
that consciousor unconsciousimage processesplay a role of great
importance inthe musicalaffective experiences
of many listeners.
Indeed, it is often diflicult for even the most disciplined and ex-
perienced listeners
to escapethe deepseated
power ofmemory over
affective experience.
It should be noted in this connection that not only do memories
frequently resultin affective experience butaffective experiences
themselves tend to evoke memories and arouse image processes
appropriate to the character of the affective experience, whether
sad orgay, nobleor tender,as determined
by the objective situation.
In other words, even the most purely musical affective experiences
may give rise to image processes
which, developing their own series
of associations,may become independent of the musical succession
itself.
Connotation
By connotations,as distinguishedfrom image processes, are
meant
those associations
which are shared incommon bya group of in-
dividuals within the culture. Connotations are the result of the
associations made
between some aspect ofthe musicalorganization
and extramusicalexperience. Since
they areinterpersonal, not
only
must themechanism of association be
common tothe givencultural
group, butthe conceptor imagemust havethe samesignificance for
all the members of the group. The concept must be one that is to
some extent standardized in cultural thinking; it must be a class
concept that has the same meaningfor, and produces the same atti-
tudes in,all the members ofthe group.In the West, for example,
death isusually depictedby slow tempi andlow ranges,while in
Note on Image Processes,
Connotations, Moods 259
certain Africantribes itis portrayedin frenziedmusical activity;yet
this results from difference in attitudes toward death rather than
from differencesin the associative processes
of the human mind. The
particular wayin which a connotationis realized or represented
in music cannot be understood apartfrom the beliefs and attitudes
of theculture inquestion.
Some connotations are entirelytraditional. Association
is by con-
tiguity; i.e.,some aspect
of themusical materials and theirorganiza-
tion becomeslinked, by dint of repetition, toa referentialimage.
Certain instruments beeorne associatedwith special concepts and
states ofmind, The organ, forexample, isassociated forWestern
listeners withthe churchand throughthis with piety andreligious
beliefs andattitudes, Thegong islinked bycontiguity tothe Orient
and Oftenconnotes themysterious andthe exotic. In fact, even
where this association does not seem intended, as in Varses I onisa-
tion, it tends tomodify ourresponse tothis music.Certain modes of
tonal organization may awakenconnotations. The pentatonic mode,
for example,is usedin the nineteenth century to representthings
pastoral. Certain intervals maybe usedto indicatespecial concepts
or statesof mind.For instance,the diminishedfifth wasclosely asso-
ciated with expressions ofgrief and anguish duringthe baroque
period. Orspecific tunes may beemployed toevoke concepts, mem-
ories, orimage processes, This isa frequentdevice inthe musicof
Charles Ives.
As a rule such associations are used in combination so that each
reinforces the other, If the eornposerwishes to evoke connotations
of pietyand thoseconnected withreligious beliefs,
he will not only
employ theappropriate instrumentbut he will also use techniques
of composition-modality,Polyphony, andso forth-that have the
same associations.
Notice that all these associations areintracultural. The gong will
not havea specialexotic meaning
for the oriental inwhose musicit
is common, though itmay haveother differentassociations for
him.
Nor will the pentatonicmode connotethings pastoralto peoples
who use this mode for all kinds of music, for cultivated art music
as well as for folk music.
Because suchassociations arecompletely cultural and in no sense
necessary, they are subjectto change.Old associationsdie andnew
ones come into being.In Westernmusic, forexample, theharp is
no longerassociated, as it was in the Middle Ages,with religious
subjects. Because of its use inFrench musicof the late nineteenth
century, itis muchmore likelyto beassociated with a certaintender
vagueness.
A particularepoch maydevelop quitean elaborate
system ofcon-
notations in which certain melodic, rhythmic, or harmonic practices
become signsof certain states ofmind or are usedto designate
specific emotional
states. Thecomposers ofthe baroqueperiod de-
veloped sucha systemof connotations.Other composers, notably
Wagner, have invented their own systemsof connotative symbols,
in which a specificmelody, notjust a more or less generalfigure,
indicates andsymbolizes specific
a idea,
concept, orindividual.
If our responses to
such specialsystems ofconnotative ordesig-
native symbolsare tobe reallyeffective, they
must become habitual
and automatic.This requirestime andrepeated encounters with a
given association.We do not need to learn that an oboe is tradi-
tionally a pastoral instrument.
By hearingit used in this context
time andtime again,by readingabout pipes
and shepherds in litera-
ture, and by seeing such instruments depicted in paintings of Pan
or Marsyas,we graduallybuild up a set of powerful associations.
Once such an association has become firmly established, our re-
sponse to
it will be justas directand forcefulas if the response
were
natural.
However important
associations made
by contiguitymay be,they
constitute but a small fraction of the total group of connotations
evoked bymusic. Most of the connotations whichmusic arousesare
based uponsimilarities whichexist betweenour experienceof the
materials ofmusic andtheir organization,
on theone hand,and our
experience ofthe non-musicalworld of concepts, images,
objects,
qualities, andstates ofmind, on the other.
There isa greatdeal of evidence, some of it intercultural, which
indicates thatour experienceof musicalstimuli is not a separate,
special categoryof experiencebut that it is continuous withand
similar to our experiencesof other kinds of stimuli.
Both music and life are experienced as dynamic processes of
growth and decay, activity and rest, tension and release. These
processes are
differentiated, not
only bythe courseand shapeof the
motions involvedin them, but also by the quality of the motion.
For instance,a motionmay be fast or slow, calmor violent, con-
tinuous orsporadic, preciselyarticulated orvague in outline. Al-
most all modes of experience, eventhose in which motion is not
directly involved,
are somehow
associated qualitatively
with activity.
Spring, revolution, darkness, thepyramids, a circle-each, depend-
ing uponour currentopinion ofit, is experienced as
having achar-
acteristic motion. If connotations are to be aroused at all, there will
be a tendency toassociate the
musical motionin questionwith a
referential concept
or imagethat is felt to exhibit a similar quality
of motion.
The unity of perceptual experience, regardlessof the particular
sense employed, is alsodemonstrated by
the fact that in experience
even singlemusical tones tend to become associatedwith qualities
generally attributedto non-auralmodes ofsense perception. This
tendency isapparent notonly inWestern culturebut in the cultures
of theOrient andin manyprimitive cultures.
In Westernculture, for
example, tones are characterized with respect to size large or
small!, color value light or dark!, position high or low!, and
tactile quality rough or smooth, piercing or round!. Furthermore,
it should be noted that thesequalities areinterassociated among
themselves; that is, volumeis associatedwith position e.g., alarge
object is generally associated with a low position!, and both of
these are associated with color.
Through such visual andtactile qualities,which arethemselves a
part of almost allreferential experience,
tones becomeassociated
with our experience ofthe world. Thus the associations, ifany,
evoked bya lowtone willbe limited,though notdefined, bythe fact
that in Western culturesuch tonesare generallyassociated with
dark colors,low position, large size,and slowermotion.
Often referentialexperiences are
themselves partly
aural. A city,
the wind,solitude, orthe expressions
of the human voice-all have
a peculiarquality of sound whichmusic canimitate with varying
success. Such imitation will tend to awaken connotations similar
in somerespects atleast to the experiences
which originallycon-
ditioned the musical organization.
To what extent the associations arisingfrom similarities between
our experienceof music and our experience ofthe non-musical
world are products of cultural conditioning and to what extent they
are insome sense
natural isdifficult tosay. Themany studies
made
by psychologists,although theypresent ampleevidence ofasso-
ciative consistencywithin Western culture, throw little light upon
the problem of the naturalness of these responses;for the subjects
in such experiments have,
almost withoutexception, already
been
saturated with the beliefs and attitudes of Western culture.
Evidence from primitive and non-Western culturesis not conclu-
sive. Frequently
the associations
formed areones whichappear nat-
ural to us. But sometimes a connotation strikes us as odd or unusual.
In the latter case, however, it must be remembered that the associ-
ation evoked by a given musicalpassage depends
upon theattitude
of the culture towardthe conceptas well as uponthe mechanism
of association.In other words, although in a given culture one atti-
tude toward an object or processwill usually be dominant, others
are possible.
For example,
although inour culturedeath isgenerally
considered to be a solemn, fearful, and majestic summoner, it has
also been viewed as an old friend or as the sardonic mocker of
human pretensions.And obviously each of these attitudeswould be-
come associated
with verydifferent typesof musicalpresentation.
This much, however, is clear: ! In most cultures there is a
powerful tendency
to associate
musical experience
with extramusical
experience. Themany musical cosmologies ofthe Orient, the prac-
tice of most primitivecultures, andthe writings and practicesof
many Westerncomposers arestriking evidence of this fact. ! No
particular connotation is an inevitable product of a given musical
organization, sincethe associationof a specific musicalorganization
with a particular referentialexperience depends
upon the beliefs
and attitudes of the culture toward the experience. However, once
the beliefsof theculture areunderstood, most
associations appear
to
possess acertain naturalnessbecause theexperiences associated
are
in somesense similar.! No matter how natural a connotation may
seem tobe, it undoubtedly acquires
force andimmediacy through
cultural experience.
Obviously acomplex and subtle connotationis notdefined byany
single elementof the sound organization. Taken individuallyany
one aspectof the musical organization is a necessary butby no
means asuflicient cause for defininga given connotation. Forin-
stance, whileit would not bepossible inWestern cultureto depict
the joysof youth in the lowest ranges of the bassoon, highranges
alone wouldnot assuresuch anassociation either. Other aspectsof
the musicalorganization, such as tempo,dynamics, rhythmic char-
acter, andtexture, wouldhave to play a part in delining sucha
connotation.
But thedegree ofspeciiicity attained
in association,
the degreeto
which a given musicaldisposition willevoke thesame orsimilar
connotations inall listeners within the cultural group, is not merely
t_he function of thenumber ofelements defining the connotation.All
the elementsof music are alwayspresent ifthere is any musicat
all. That is, thereis alwaystexture, whetherit be that of a single
melodic lineor that of a complex polyphonic web; thereis always
dynamic level,whether itbe thatof a striking fortissimoor that of
a mezzoforte.
The specihcityof a connotation dependsupon thedivergence of
the elements of soundfrom aneutral state.
A tempomay beneither
fast nor slow; a sound maybe neitherloud nor soft; a pitch may
seem neitherhigh nor low, relative either to over-all range or the
range ofa particularinstrument orvoice. Fromthe standpointof
connotation these
are neutralstates. Connotation
becomes speciiied
only if some ofthe eleinentsof sounddiverge fromsuch neutral
states.
The elements of sound are interdependent with respect to neu-
trality and divergence. For instance, changesin pitch are generally
accompanied bychanges indynamics, timbre,and sometimestempo.
The relationshipis physical as well as psychological.If a 33%, r.p.m.
phonograph recordis played at 78 r.p.m., pitch will get higher, dy-
namics louder, and timbre more piercing. Thus it is possible to
build one divergence upon another. For instance, if tempo is fast
and pitches are high, very soft dynamics will be experienced as a
divergence, notonly from the neutralstate of moderate loudness,
but also from the contingent neutrality in which a rapid tempo
and highpitches aregenerally accompanied by loud dynamics.
In general,the moremarkedly theelements ofa soundpattern
diverge fromneutrality themore likelythey areto evokeconnota-
tions andthe morespecific those connotations areliable tobe. Note
that this accounts forthe fact that manymusical worksarouse a
wide variety of connotations.For the connotations aroused by a
piece ofmusic which,on thewhole, employs normal ranges,moder-
ate tempi,and soforth will be determinedmore bythe disposition
and susceptibilityof the particular listenerthan by the natureof
the musicalorganization itself.
But evenwhere themost complex disposition ofthe musicalma-
terials and the most effective deviations are presented in a piece
of music,they functiononly asnecessary causes
for the particular
connotative experiencearoused.
In the first place,unlike literatureor the plastic arts,which gen-
erally speakingcannot beunderstood apartfrom the designative
symbols theyemploy, mostmusical experience is meaningfulwith-
out any reference tothe extramusicalworld. Whethera piece of
music arouses connotations depends to a great extentupon thedis-
position and training of the individual listener and upon the pres-
ence of cues, either musical or extramusical, which tend to activate
connotative responses.
In the second place, unlike verbal symbols or the iconic signs
used in the plastic arts, musical sounds are not, save in a few
isolated instances,
explicit intheir denotation.
They limit and define
the associations
possible but,in the absence ofeither a specific
musical symbolismsuch asWagners or a definite program furnished
by the composer, they cannot particularize connotation. The mu-
sical materials and their organization are the necessary causes for
a given connotation but, since no summation of necessary causes
can ever amount to a sufficient cause, the sufficient cause of any
connotation experiencedmust be supplied by the listener.
The fact that musiccannot specifyand particularizethe conno-
tations which it arouses has frequently been cited as a basic difli-
culty with any attemptto theorizeabout theconnotative meanings
of music.Yet fromone pointof view,this flexibility of connotation
is avirtue. Forit enablesmusic toexpress what
might becalled the
disembodied essence of myth,the essence
of experiences which are
central to and vital in human existence.
The humanmind hasan uncannypower of recognizing symbolic forms;

and mostreadily, ofcourse, willit seizeupon thosewhich arepresented
again andagain without aberration. Theeternal regularitiesof nature,
the heavenlymotions, thealternation ofnight andday onearth, thetides
of the ocean, arethe most insistent repetitiousforms outsideour own
behavior patterns
.... They are themost obviousmetaphors toconvey
the dawning concepts oflife-functions-birth, growth, decadence, and
death.
What music presents is not any given one of these metaphorical

events but rather that which is common to all of them, that which
enables themto become metaphors for one another. Music presents
a genericevent, aconnotative complex,which thenbecomes par-
ticularized in the experienceof the individual listener.
Music does not, for example, present the concept or image of
death itself. Rather it connotes that rich realm of experience in
which deathand darkness,night and cold, winter and sleepand
silence areall combinedand consolidated
into a single connotative
complex.
The interassociations
which giverise to such aconnotative com-
plex arefundamental inhuman experience. They arefound again
and again,not only in the myths and legends of many cultures, but
also in the several arts. For example, the connotative complex dis-
cussed aboveis made explicit in Shelleys Ode to the West Wind:
O thou,
Who chariotestto their dark wintrybed
The wingedseeds, wherethey lie cold and low
Each like a corpsewithin its grave ....
Connotative complexesmay be more and less specific.Additional
divergences intimbre, dynamic level, and so forth may help to limit
the qualityof the complex. Association
by contiguity or the imita-
tion of actual soundprocesses heardin the extramusical world may
also playa part in definingthe extentof connotation.Finally, con-
notation maybe specifiedby the presence ofa text, a plot, or a
program established by thecomposer.
Ultimately itis thelistener whomust makeconnotation concrete.
In so doing the listener may draw upon his stock of culturally
established images, including thosederived from literature and
mythology, orhe may relate the connotative complex to his own
particular andpeculiar experiences. But in either casethere is a
causal connectionbetween the musical materials and their organ-
ization and the connotations evoked. Had the musical organization
been different, the connotation would also have been different.
Mood
Since, however, connotations are not necessary concomitants of
musical experience see p. 246!, a potentially connotative passage
may fail to evoke any concrete images whatsoever. Instead the
listener may become aware of how the musical passage feels in
relation to his own designative emotional experiences and the ob-
served emotional behavior of others. The music may, in short, be
experienced asmood or sentiment. For not only are connotations
themselves intimately
associated with
moods, inthe sense
that youth
or spring, for instance, are traditionally considered to be times of
exuberant and
carefree gaiety,
but the same psychological
and mu-
sical processeswhich arouse specific connotationsalso evoke def-
inite, thoughperhaps less
specific, mood
responses.
In a discussion of the communication of moods and sentiments
two important considerations mustbe kept in mind.
1. The moods and sentiments with which music becomes asso-
ciated arenot thosenatural spontaneousemotional reactions,which,
as noted in chapter i, are often diffuse and characterless. Rather
music depicts
those' modes
of behavior,conventionalized for
the
sake ofmore efficientcommunication, whichwere calleddesigna-
tive emotionalbehavior. In Western culture,for example,grief is
communicated by a specialtype of behavior: physical
gestures and
motor behavior tend to minimal; facial expression reflectsthe cul-
tural picture of sorrow; the rangeof vocal expression isconfined
and often sporadic; weepingis customary;and dresstoo servesas a
behavioral sign.It is this special,culturally sanctionedpicture of
grief whichis communicated in Westernmusic. Butsuch designative
emotional behavior is not the only possible wayof denotinggrief.
Were the standardized expressionof grief in Western culture dif-
ferent, wereit, for instance, that
of anincessant and
violent wailing
and moaning,then the expression of grief in Western musicwould
be different.
This isimportant because
it allowsfor andaccounts forvariation
in mood expression betweenthe music of different cultures. That
is, differentcultures may
communicate moods
and sentiments
in
very differentways, notbecause thepsychological mechanism of
association is
different butbecause thebehavior patternsdenoting
mood and emotional states are different.
2. ]ust as communicative behavior tends to become conven-
tionalized for the sake of more eflicient communication, so the
musical communication of moods and sentiments tends to become
standardized. Thus
particular musicaldevices-melodic figure,har-
monic progressions,or rhythmic relationships-become formulas
which indicatea culturally codified moodor sentiment.For those
who arefamiliar with them, suchsigns maybe powerfulfactors in
conditioning responses.
Association by
contiguity playsa considerablerole in the musical
definition ofmood. Amelodic figure,a set of modal relationships,
or a harmonic progressionis experiencedtime and time againin
conjunction with texts, programs,or extramusicalexperiences which
either designate the mood directly or imply it. In oriental music,
for instance,a particularmode oreven aparticular pitchmay be-
come associated with a specific sentimentor humour as well as with
connotative concepts such as winter, night, and blackness. Once
such associationsbecome habitual, the presenceof the proper mu-
sical stimulus will, as a rule, automatically evoke the customary
mood response.In Western music of the baroqueperiod, to cite only
one example, melodic formulas, conventionalized for the sake of
communication, attain precision and force through contiguity with
texts and programs which Hx their meanings Withinthe culture and
style.
Mood association by similaritydepends upon a likenessbetween
the individualsexperiences of moods andhis experienceof music.
Emotional behavior is a kind of composite gesture,
a motionwhose
peculiar qualitiesare largelydefined interms of energy, direction,
tension, continuity, and so forth. Since music also involves motions
differentiated bythe samequalities, musicalmood gesturesmay
be similarto behavioralmood gestures. In fact, because moodsand
sentiments attain their most precise articulationthrough vocalin-
flection, it is possible for music to imitate the sounds of emotional
behavior withsome precision.
Finally, sincemotor behaviorplays a
considerable role in both designative emotional behavior and in
musical experience, a similarity between the motor behaviorof
designative gestures and that of musicalgestures willinforce the
feeling of similarity betweenthe two types of experience.
Like connotation, mood or sentiment depend for their definition
upon divergence.If the elements ofsound areneutral then the
mood characterization,if any, will depend largely upon the dis-
position of the individual listener. That is, there will be no con-
sistency in the responses of various listeners. But, and this is of
paramount importance, the factthat themood isindefinite does not
mean thataffect isnot aroused. For a lack of divergence inthe ele-
ments of sound doesnot preclude significant deviationin those
dynamic processes which formour affectiveresponses to music.
It was observed earlier that imageprocesses, whether conscious
or unconscious, and connotations often resultin affectiveexperience.
Whether mood responses can eventuate in affect isdoubtful. Merely
because themusical designation of a mood or sentiment iscom-
prehended bythe listenerdoes notmean thatthe listenerresponds
affectively. Itis perfectly possible tobe awareof the meaning of
behavior withoutresponding as though thebehavior wereour own.
But evenan empathetic response to the materialsdelineating mood
or sentimentdoes notrequire aresultant affectiveexperience. We
may sympathizewith the mood of another individual without hav-
ing an emotional experience
ourselves. Infact, althoughsuch em-
pathetic behaviormay createa psycho-physiological condition in
which affect is likely to arise, it is diflicult to see what direct causal
connection could
exist betweenmood andaffect. It appears more
likely thatmood eventuates
in affectonly throughthe mediationof
image processes
or connotations.
That is,a moodarouses image
proc-
esses alreadyassociated inthe experienceof the individual with the
particular moodresponse, and these image
processes are
the stimuli
which actuallygive riseto affect'
The Role of Mood and Connotation in

Afective Experience
Not only do moodand connotation frequently giverise to affect
but theyalso colorand modifythe affectiveexperiences evoked by
the musicalprocesses discussed in the preceding chapters.The con-
verse ofthis is also true;namely, thecharacter ofthe deviations
embodied ina particularwork playa partin conditioningour opin-
ion of what, in general terms,its designativecontent is.If, for in-
stance, wecompare thefirst themeof the rondo of Haydns Sym-
phony No. 102, in B-Hat, with the first theme of the rondo of Bee-
thovens Waldstein Sonata, itis clear that the designative char-
acter ofeach, theroguish andspirited playfulness of the Haydn and
the flowinglyricism ofthe Beethoven, is a product notonly of such
factors astempo, phrasing,accompaniment, melodic contour, and
so forth but also of the fact that the themeof the Haydn rondo
involves considerable irregularity, abruptness,and deception,while
the themeof the Beethoven rondo is quite regular andforthright.
Unce the listener becomes aware of this diHerence in character, he
is definitely prepared for different kindsof movements-for the witty
and highly sophisticated surprises
of the Haydn and the striking
but not unexpected contrastswhich mark the Beethoven.
It was stated in the first chapter that an affective experience is
differentiated and
characterized by
the stimulussituation inwhich
it occurs. Both the stimulus and the situation serve to differentiate
musical experiencefrom real-life experience.
Since musical affective stimuli are obviously different from the
referential stimuliof real life, there will always be a generic dif-
ference between musical affective experience and the experiences
of everyday life. From this point of view musical experience is
unique.
However, in so far as music is able, through connotation, mood,
or the use of a programor text, to designatesituations similarto
those existing in extramusical experience, such designations will
tend to color and modify our musical affective experience. Our ex-
perience ofa melodyfirst playedforte in a low range bya trom-
bone andthen immediatelyplayed pianissimo in a high rangeon
the violin is not only shaped by the expectations arousedby the
repetition, the changes indynamics, range,and instrumentation but
it is colored andqualified bythe changesperceived inthe desig-
native characterof the two passages.
Last but most important of all, because ourunderstanding of the
designative character
of a passage may
affect ourinterpretation and
evaluation of its function within the total work, such understanding
of designativemeaning mayplay an important rolein the actual
shaping ofour affectiveexperience.
Connotation, Mood, and Aesthetic Theory

Reacting againstthe strongemphasis placedupon mooddesig-
nation andconnotation innineteenth-century music,
many critics,
theorists, andpsychologists have
in recent years questionedthe
relevance andpertinence ofthe connotativeand moodresponses
made to music.
The attackupon suchreferential musical
experience has
focused
! upon the causal connection betweenthe musical stimulus and
the referentialresponse, ! upon theapparent disparitybetween
the responses of differentlisteners, and! upon the lackof speci-
ficity in the responses made. Thedifliculty, writesHanslick, isthat
there is no causal nexus between a musical composition and the
feelings it may excite,as the latter vary with our experience and
impressibility. 8
These objections
are, however,without merit.In the Hrst place,
all significant responses tomusic, the affective and aesthetic aswell
as thedesignative andconnotative, varywith our experience andim-
pressibility. Theresponse tostyle is a learnedresponse, and both
the appreciationof styleand theability to learn requireintelligence
and musicalsensitivity. In the secondplace, thoughthe causalnexus
between musicand referential experience isa necessarynot a suffi-
cient one,there isa causalnexus, asis evidencednot only by the
practice ofcomposers withina givenstyle butalso bythe responses
of listenerswho have learned to understand thestyle.
While Hanslicksattack isdirected largelyagainst thebelief that
music communicatesfeelings, C. C. Pratt denies the possibility of
connotation.
The shapingof these [tonal] mosaicsmay receiveimpetus from all

sorts ofobjects andideas andthe composer, in his innocence, maybelieve
that hehas embodied his non-musicalidea in sound. Anda goodlynum-
ber of still more innocent listenersmay persuadethemselves thatthey
comprehend thecomposers ideas.It requires, however, onlythe sim-
plest sortof experimentto demonstratethe utter lack of correspondence
between theidea whichthe composer may thinkhe hasrepresented and
the interpretationswhich an unselected groupof listeners, if asked to
do so, will furnish ....
The first difficulty with this argument lies in the phrase un-
selected groupof listeners. Of course,if they are unselected,if
they havenot learnedthe style,they will give very different re-
sponses. Andthis will also be true of their responsesto embodied
meaning. But connotations will vary even among those who do
have thesame culturalbackground andwho are acquainted with
the modesof associationestablished within the style. However, this
variation, thoughsignificant, isoften not as wide as it seems at
first glance. Because themodes of experience are continuous with
one anotherand becauseexperience itselfcan be expressed ina
wide variety of metaphors, a connotative complex which has the
same potential meaning for all listeners may be actualized differ-
ently in the experienceof each. In other words, while it is true that
on one level that of specificmeaning! the ideas entertainedby
various listeners
are patentlydifferent, onanother level the level
of symbolicand metaphoricalmeaning! the concepts entertained
by the various listeners
are verysimilar.
The difficulty with an aesthetic of music basedupon connotative
and mood responses isnot that the associationsbetween music and
referential experienceare fortuitous or that there is no causal con-
nection betweenmusic andfeelings. Thedifliculty is that, in the
absence of
a specificreferential framework,
there isno causalnexus
between successive connotations or moods. In literature or in life,
successive experiences
are apparentlycausally connected
by the
sequence of events which take place between them. A depressing
experience is
followed bya joyfulone, andthe changeis understood
in the light of the events connecting them. But though music can
present the experiences themselves,if only metaphorically, it can-
not stipulate the causal connection betweenthem. There is no
logical reason,either musical
or extramusical,
for anyparticular suc-
cession of connotations or moods.
Confirmation forthis argumentcan befound in the practiceof
composers ofinstrumental musicwho, realizingthat the difliculty
with referential music lies in the lack of a causal connection between
successive moods
or connotations,
have sought
to correctthis weak-
ness by using descriptiveprograms. Although a program does serve
to specifyconnotation, its
main functionis notto designatemood or
arouse connotation.Music can as a rule accomplishthis more effec-
tively thana programcan. Whatthe programdoes isto providethe
causal connection between the successive moods or connotations
presented in the music.
Seen inthis light,the programis not the merewhim of the com-
poser, anunnecessary and superfluous addition to meaningsalready
inherent in the music,nor is it an attempt to depict moodsand
connotations. Itsfunction is to connectthem. Thegreat disadvan-
tage of a programlies in the fact that it is a powerful temptation
toward extramusical diversion.
Notes
NOTES TO CHAPTER I
1. Norman Cazden, Musical Consonance andDissonance: A Cultural

Criterion, Iournal of Aesthetics, IV 945!, 3-11.
2. Paul R. Farnsworth,Sacred Cows in the Psychology ofMusic,
Journal of Aesthetics, VII 948!, 48-51.
3. SusanneK. Langer,Philosophy ina New Key New York: Mentor
Book Co., 1951!.
4. Ibid., p. 171.
5. H. P. Weld,An ExperimentalStudy inMusical Enjoyment,Amer-
ican Iournalof Psychology,
XXIII 912!, 283.
6. C. S. Myers,Individual Differencesin Listeningto Music, in The
Effects of Music, ed. Max Schoen New York: Harcourt, Brace & Co.,
1927!, p. 14.
7. See H. D. Aiken, The Aesthetic Relevance of Belief, Journal of
Aesthetics, IX 950!, 301-15.
8. ]amesL. Mursell,The Psychology of Music New York:W. W. Nor-
ton & Co., Inc.,1937!, pp.27-28.
9. Ibid., p. 37.
10. David Rapaport, Emotionsand Memory New York: Interna-
tional Universities Press Inc.,1950!, p. 21.
11. ]ohn Dewey, The Theory of Emotion, PsychologicalReview, I
894!, 553-69; II 895!, 13-32.
12. R. P. Angier,The Conflict Theory of Emotions, AmericanIour-
nal of Psychology, XXXIX927!, 390-401.
13. ]. T. MacCurdy,The Psychologyof Emotion New York:Harcourt,
Brace & Co., 1925!, p. 475.
14. For MacCurdy the term instinct includes learned habit re-
sponses. Seep. 24of thisstudy.
15. Notice that this analysis makes apparent thegreat significanceof
Aiken's contentionthat ourbeliefs asto thenature ofaesthetic experience
lead to the suppression of overt responses; forsuch inhibiting of overt
behavior tendsto intensifythe affectiveresponse.
16. F. Paulhan, TheLaws ofFeeling, trans.C. K. Ogden New York:
Harcourt, Brace & Co., 1930!, p. 19.
17. zbzd.,p. 123.
273
274 Notes to Pages 17-24
18. The term affective or emotional state will henceforth be used to
designate those
aspects ofemotional experience
which havebeen given
names andwhich are in one way or another fairly standardized ina
broad sense.
19. See Robert S. Woodworth, How Emotions Are Identified and
Classified, in Feelings andEmotions: TheWittenberg Symposium, ed.
M. L. Reymert Worcester, Mass.: ClarkUniversity Press, 1928!, p. 224.
20. Ernst Cassirer, AnEssay onMan: An Introduction to a Philoso-
phy ofHuman CultureNew York:Doubleday &Co., 1953!,p. 190.This
admirable statement like so many of its kind suffers atthe end from an
irritating vaguenessin which an intangiblethe dynamic process oflife
itself is substituted fora definiteaccount ofhow andwhy the emotions
of art are notcomparable toany singlestate ofemotion. It is for a solu-
tion to this problemthat we are searchingin the present discussion of
emotional differentiation.
21. C. Landis, Studies in Emotional Reactions: II, General Behavior
and FacialExpression, ]ournalof ComparativePsychology, IV924!,
496.
22. DonaldO. Hebb,The Organizationof Behavior New York: ]ohn
Wiley & Sons, 1952!,p. 258.
23. Ibid., p. 232.
24. This statement must be qualiiiedby the reservation thatin so far
as it can designateor representextramusical stimuli,
music canbe said
to evokesuch affectivestates asare normallyconnected withthe situa-
tions represented.
25. It is alsoclear thatsince theworld of emotions isnot composed of
a seriesof separatecompartments, given
a listenermay feelthat a purely
musical emotionis comparableor analogousto affects experienced in
real life.
26. The term emotional expression ismisleading inthat it implies
that suchbehavior isthe direct,necessary expression
of affect.
27. Thoughnot within the province of this study, it can, I believe,
be shown that similar aspects of experience are involved in musical and
other designation.Both, for example, utilizethe generality of motion
fast or slow, continuousor interrupted, smooth ordisjunct, intenseor
weak! in such designation. And musicaldesignation, though probably in
some respects natural, is,like designativebehavior, inthe last analysis a
product ofculture andlearning ratherthan aproduct ofnature. Seealso
pp. 261 f.
28. See]ohn Dewey,Art As Experience NewYork: Minton, Balch &
Co., 1934!, pp. 35, 56.
29. MacCurdy, op. cit., p. 556.
30. ]ohn Dewey, Intelligencein the Modern World, ed. Ratner
Modern Library [New York:Random House, 1939]!, p. 733.
Notes to Pages 24-35 275
31. MacCurdy,op. cit.,p. 556.
32. Aiken, op. cit., p. 313; also seeArthur D. Bissell, The Role of
Expectation inMusic New Haven: YaleUniversity Press, 1921!, p. vii;
and HugoRiemann, Catechism of MusicalAesthetics, trans.
H. Bewerung
London: Augener& Co.,n.d.!, p. 29.
33. Aiken,op. cit.,p. 305.
34. If this takesplace, thelistener mayshift his attention to another
aspect ofthe musicalmaterials, orhe may simply abandonthe attempt
to makesense ofthe musicaltogether.
35. Thus the designationof mood and character,whether accom-
plished in purely musicalterms orwith the aid of a programor text, is
important notonly for its own sake, asa sourceof enjoyment,but also
because, as part of the stimulussituation, it is necessaryfor the proper
understanding ofthe musicalprocesses in progress.
36. Both these aspects of the process ofexpectation arediscussed in
the following chapters, wheremuch of the preceding discussion is
treated in more detail.
37. Eduard Hanslick, The Beautiful in Music, trans. E. Cohen Lon-
don: Novello,Ewer & Co., 1901!, p. 135. The difficultywith this state-
ment is that Hanslick confuses intellectual satisfaction with intellectual
activity. For although intellectualactivity, in the senseof mental aware-
ness andcognition, may,as we shall see,be unconscious,intellectual
satisfaction impliesa self-conscious awareness ofthe activitytaking place.
38. Dewey,Art As Experience, p.59; alsosee Intelligence in the Mod-
ern World, pp. 755ff. Robert Penn Warrenwrites to much the same
effect: . . a poem,to be good, mustearn itself.It is a motiontoward a
point of rest, butif it is not a resistedmotion, itis a motion of no conse-
quence Pure and Impure Poetry, Kenyon Review, V [1943], 251!.
39. It is clear that the terms norm and deviation are being used in
a very broad andgeneral sense.
Deviation includes all delaysand inhibi-
tions whichgive riseto expectationwithin the context of the particular
style in question.
40. Morris R. Cohen,A Prefaceto Logic New York: Henry Holt &
Co., 1944!, p. 47.
41. SeeGeorge H.Mead, Mind, Self, andSociety Chicago:Univer-
sity of Chicago Press,1934!, p. 76.
42. Thus Pratt, while maintaining that the ideas aroused in association
with music have little to do with the intrinsic nature of musical sound
C. C. Pratt, Music and Meaning, Proceedings of the Music Teachers
National Association,
Series XXXVII[1942], p. 113!, doescontend that
music soundsthe wayemotions feel Ibid., p. 117! ;a statementwhich
seems tobe a disguised form of referentialism.
Incidentally though Pratt's first statement is undoubtedly true, its
implications arenot. For while our associations may
have nothingto do
with the intrinsic natureof sound,whatever thatmay be, they do have
something to do withour experience of sound.See pp.261 f. of this study.
43. The term stimulus as usedhere includesany tone or combina-
tion of tones whichare markedoff asa unitaryevent whichis relatedto
other musical events. Itis, touse Meadsterminology, amusical gesture.
Or, in the terminologyused in chap. ii of this study, it is a sound
term. In this sense, a singletone, aphrase, ora wholecomposition may
be considered to bea gesture,a stimulus,or asound term.In other words,
meaning must be considered as beingarchitectonic aswell asconsecutive.
44. Of co1use,may it havedesignative meaning. The moredifficult it is
to graspthe embodiedmeaning ofa work, the greaterthe tendencyto
search fordesignative meanings.
45. Notice,too, that the final tones ofa piece,conditioned byall that
has gonebefore, leadus to expect silence and that it is this expectation
which makesthem meaningful.
46. Notethat hypotheticalmeanings as well asthe otherkinds ofmean-
ing are architectonic. If we are set to listen to a Haydn rondo, then
the ideawe haveof Haydn rondos is,in a sense, thehypothetical mean-
ing of that particularrondo; it is what we envisageand what points to
the impingingstimulus.
47. BertrandRussell, Selected Papers ModernLibrary, [New York:
Random House, n.d.]!, p. 358.
48. Mead,op. cit.,p. 194.
49. Mead,op. cit., pp. 42-75.
50. Ibid., p. 47.
51. Of course, ifthe composeris developinga relativelynew style,as
many contemporary composers have tried to do, the imagined listener
may correspond to no listener whoactually exists.He is rather onewhom
the composerhopes tocreate ashis style becomes partof the general
style, partof the listening publicsstock ofhabit responses.
52. LeopoldMozart, Versucheiner griindlichenViolinschule, quoted
in Source Readings in Music History, ed. Oliver Strunk New York:
W. W. Norton & Co., Inc., 1950!, p. 602.
53. Generallyspeaking thisstudy is not concernedwith the creative
act but rather withthe experiencewhich theart work brings into being.
This aspectof the composer's creative
life is discussed becauseit clearly
concerns theproblem of aesthetic experience. Obviously many other
mental processes and attitudeswhich havenot been touched uponare
involved inthe actof composition.One ofthese isdealt with on pp.69 f.,
but mostof themare left to the speculations of others.
54. Cassirer,op. cit., p. 191.
NOTES TO CHAPTER II
1. Edward Sapir, Language, Encyclopedia of

the SocialSciences, IX
New York: Macmillan Co., 1934!, 157.
2. I know of only one study, that of Arthur D. Bissell The Role of
Expectations in Music [New Haven: Yale University Press, 1921]!,
which dealsexplicitly with the subjectof expectations,
and this study is
by no means exhaustive.
3. Similarly any compoundor serial term establishedin accordance
with c! in turn enters intoprobability relationships
with other terms of
the systemand othercompound terms.
4. A. H. Fox Strangways, TheMusic of Hindostan London: Oxford
University Press,1914!, p. 18.
5. See Felix Salzer, Structural Hearing New York: Charles Boni,
1952!, pp. 182-83.
6. C. P. E. Bach, Essayon the True Art of Playing Keyboard Instru-
ments, trans. William Mitchell New York: W. W. Norton & Co., Inc.,
1949!, p. 84. The relation of ornamentation todeviation and delay is
discussed in
chap. vi,p. 204of thisbook.
7. Curt Sachs, TheRise of Music in the Ancient World, East and West
New York: W. W. Norton & Co., Inc., 1943!, p. 37.
8. Hindemiths objection tothe conceptthat chordsare susceptible of
various interpretations is unfortunate;Brst, becausethe chordswhich he
cites arenot necessarily ambiguous and, second, becauseambiguity may
function asan importantaffective aesthetic device. SeePaul Hindemith,
Craft of Musical Composition, trans. Arthur Mendel New York: Asso-
ciated MusicPublishers Inc.,1942!, pp. 90 ff.
9. Salzer,op. cit., pp. 19-20.
10. Henry D. Aiken, The Aesthetic Relevance ofBelief, Iournal of
Aesthetics, IX 950!, 305-6. Aiken includes under the term belief the
habits, dispositions,
and attitudeswhich the listener andcomposer bring
to the work of art.
11. Walter Piston, Harmony New York: W. W. Norton & Co., Inc.,
1941!, p. 17.
12. Strangways, op. cit., p. 150.
13. Ibid., p. 128.
14. See, for instance, Frances Densmore,Pawnee Music Smithsonian
Institution, Bureauof American Ethnology, Bulletin93, [1929]!, pp.
120 ff.
15. As Ernest Nagelnotes: An estimate ofa probability which is
made simply on the basis of unanalyzed samplesor trials is not likely to
be a safe basis for prediction. If nothing is known concerning the
mechanism of a situation under investigation, the relative frequencies
obtained fromsamples maybe poor guides tothe characterof the in-
definitely largepopulation fromwhich they are drawn Ernest Nagel,
Principles ofthe Theory of Probability, International Encyclopediaof
Unified Science,Vol. I, No. 6 [Chicagoz Universityof Chicago Press,
1939], p. 59!.
16. HenryD. Aiken, The Concept of Relevancein Aesthetics, ]our-
nal of Aesthetics, VI 947-48!, 159.
17. Hugo Riemann, Catechism of Musical Aesthetics, trans.H. Be-
werung London:Augener &Co., n.d.!,p. 31.
18. GeorgeHerzog, A Comparison of Pueblo and Pima Musical
Styles, ]ournalof AmericanFolklore, XLIX,No. 194,286.
19. Many acoustical andpsychological theories have been advanced
to explain why certain sound complexes tend to become normativein
some stylesystems. But a satisfactoryexplanation cannot
be found either
in the physics ofsound orthe pleasure-displeasure reactions oflisteners.
It lies in the nature ofhuman mentalprocesses see p. 231!.
20. Herzog,op. cit., pp. 290, 308-9. The relationship betweenorna-
mental ordecorative tones and theaffective aesthetic
response isdiscussed
in chap. vi.
21. Egon Wellesz, AHistory of Byzantine Musicand Hymnography
London: OxfordUniversity Press,1949!, pp.207-8.
22. Winthrop Sargeant, Iazz:Hot and Hybrid New York: E. P.
Dutton & Co., 1946!, pp. 156-57.
23. Igor Stravinsky, Chronicleof My Life London: Victor Collancz
Ltd., 1936!, pp. 185-86.
24. As a rule these self-imposedresistances anddifEculties havebeen
limited to the exigenciesof the communicative process see pp. 40 f.!.
But at times thecomposers concern with the creation andthe conquest
of self-imposedtasks hasbeen in part intrapersonal.In the rhythmic
structure of much late gothic and early Renaissance music and in the
permutations ofthe tone-rowin sometwentieth-century music, the com-
poser oftenseems tomanipulate andplay with the musicalmaterial, not
for the sake ofany aestheticeffect it may haveupon the listener, but
rather forthe pleasurehe derivesfrom thefeeling ofpower andcommand
over the materials of his art.
Of course,such worksdo communicate on the interpersonal level.
But
the specialcraft secretsin which the composer has takenobvious delight
are notapparent tothe uninitiate.To appreciatethis aspectof the com-
position, thelistener mustactively takethe attitudeof the composer. He
must self-consciouslysolve theriddle of the music,enjoying theintricacies
of its masterful manipulations in a manner similar to that of the com-
oser.
P 25.
Karl Croos,
The Play
of Man,
trans.L.E.
Baldwin
New York:
D. Appleton & Co., 1901!, p. 8.
N otesto Pages 73-81 279
26. Evenif the act of listening isnot directlyintentional, inthe sense
of beingplanned beforehand, once thedecision tolisten, topay attention,
is made,preparatory sets will be brought intoplay.
27. Ernst Kris, Psychoanalytic Explorations inArt New York: Inter-
national UniversitiesPress, 1952!,p. 42.
28. This belief appearsto haveits rootsin the seriousness of play, on
the onehand, andin the significance and power attributedto art, through
its ancientrelation toritual, magic,and religion,on the other. In spite of
the secularizationof art,this feelingof its significant seriousness
remained.
29. K. Koffka, Principles of GestaltPsychology NewYork: Harcourt,
Brace &Co., 1935!,p. 173.
30. ]amesL. Mursell, The Psychologyof Music New York: W. W.
Norton & Co., Inc., 1937!, p. 37.
31. B. Pasquarelli, ExperimentalInvestigation of the Body-Mind
Continuum in Affective States, Iournal of Nervous and Mental Diseases,
CXIII 951!, 512.
32. Part of this behavior probablyresults fromthe fact that absolute
silence makes us uncomfortable,and, whenit is lasting, conveysto the
mind aspecial qualityof emotion Croos, op.cit., p. 21; alsosee Donald
O. Hebb,The Organization of Behavior[New York:]ohn Wiley & Sons,
Inc., 1952],p. 252!.
33. Notice,for example,that suchcommon criticisms of modernpaint-
ing as my little brother couldhave donethat are not primarily state-
ments of aesthetic value,though this is implied, but rather statements
about belief.
34. Mere numbers are not the criteria for all members of the listening
public. Thereare thosegroups whoprefer tobe exclusive,
who arein the
know asto whatthe right music is.Here belieffunctions ininverse ratio
to general opinion.
35. David Riesman, The
Lonely Crowd New York:Doubleday &Co.,
1953!.
36. Koffka,op. cit.,p. 206.
37. Ibid., p. 19.
38. Otto Ortmann, On the Melodic Relativity of Tones, Psycho-
logical Monographs, Vol. XXXV, No. 1 926!.
39. O. H. Mowrer,Preparatory SetExpectancy: FurtherEvidence of
Its CentralLocus, Iournal of ExperimentalPsychology, XXVIII941!,
116-33.
40. Bach, op. cit., p. 152. For the psychological viewpoint see P. E.
Vernon, The Apprehension and
Cognition ofMusic, Proceedingsof the
Music Association,LIX 933!, p. 66.
41. Stravinsky,op. cit.,p. 122.
42. 1bid.,pp. 122-23.
43. Mursell, op. cit., o. 162.
44. Curt Sachs, Rhythm
and Tempo New York:W. W. Norton &Co.,
Inc., 1953!, p. 38.
NOTES TO CHAPTER III
1. GeorgeH. Mead, Mind, Self, and Society Chicago: Universityof

Chicago Press, 1952!, p. 25.
2. Donald O. Hebb, The Organizationof Behavior New York: ]ohn
Wiley & Sons, Inc.,1952!, p. 35.
3. Ibid., p. 58.
4. This example issimilar to one quotedin Curt Sachs, TheRise of
Music in the Ancient World, East and West New York: W. W. Norton
& Co., Inc., 1943!, p. 125.
5. K. Koffka, Principlesof Gestalt Psychology NewYork: Harcourt,
Brace &Co., 1935!,p. 303, and chap.x.
6. 1bia.,p. 110.
7. Ibid., p. 109.
8. Ibid., p. 143.
9. Ibid., p. 305.
10. MaxWertheimer, Productive Thinking NewYork: Harper& Bros.,
1945!, pp.193, 197.
12. Ibid., p. 109.
13. Seeibid., p.507.
14. Ibid., p. 473.
15. Ibid., p. 476.
16. Ibid., pp. 335,338, 618-23.
17. Ibid., pp. 608-9.
18. Ibid., p. 175.
19. Theseobservations point up a basic fallacyin the work of Helm-
holtz, Wundt,and Stumpf,all of whom isolatesensation from its aesthetic
context and make pleasure-displeasure reactions thebasis for a psy-
chology ofmusic. On this point see Susanne K. Langer, Philosophy in
a New Key New York: MentorBooks, 1951!,p. 171.
20. Koflka, op. cit., pp. 302-3.
21. For other examples of this see S. Bach,Well-tempered Clavier,
Vol. I, Preludes 1 and 2.
22. R. B. Stetson,A Motor Theory of Rhythm andDiscrete Succes-
sion, Psychological Review, XII 905!, 250-70, 293-350.
23. 1bid.,p. 308.
24. H. Woodrow, A Quantitative Studyof Rhythm, Archives ofPsy-
chology, XIV 909!, 1-66.
25. This should bekept in mind because there isalways atremendous
temptation tospecify andclassify, toregard thegroupings asclear and
deHnite. Butthere aremany casesin which the real significance ofthe
rhythmic organization lies preciselyin its elusive ambiguity.
26. Curt Sachs, Rhythm and Tempo New York:W. W. Norton &Co.,
Inc., 1953!, pp. I6 f. Actually Sachs probably means meter;and if this
is the case, hisargument ismore plausible.However, evenhere it may
well be that theestablishment of a metricpulse onthe levelof the phrase
will persist,furnishing thecomposer withthe possibilityof creativedevia-
tion. Fora discussion of metersee pp.115 f.
27. Also see discussion of the fifth movementof BeethovensString
Quartet inC-Minor, pp.145 If.
28. Seealmost anypiece ofSchubert orBrahms, forexample.
29. This is worthy of note because itseems toindicate that even our
tendency torespond tomusic in a motor way is, in the last analysis,a
learned response. This is also indicatedby the experimental literature on
the subject see ]amesL. Mursell, The Psychology of Music [New York:
W. W. Norton &Co., Inc.,1937], pp.153-54!. Furthermore, it emphasizes
the culturalcharacter ofeven suchseemingly basic responses asthe motor
response and the necessityof bringingthe properattitudes intoplay.
30. The quotation isnot completein all parts butpresents theessential
elements necessary for thepresent analysis. It may be foundin the minia-
ture score published by & W. Chester,London! from two measures
before No. 8 to two measures after No. 9.
31. This sense ofprogressive and
continuing harmonicmotion ismade
possible becauseas Example40 shows!there isno realdominant prep-
aration, nopause onthe dominant;rather themusic rushes
headlong into
the melodic-rhythmicrecapitulation.
NOTES TO CHAPTER IV
1. A. H. Fox Strangways, The

Music of Hindostan London:Oxford
University Press, 1914!, p. 329.
2. H. Watt, Functions of the Sizeof Intervalin the Songs ofSchu-
bert and of the Chippewa and Teton Sioux Indians, British Iournal of
Psychology, XIV 923-24!, 370-86.
3. Grooes Dictionary of Music and Musicians, ed. H. D. Colles New
York: Macmillan Co., 1936!, III, 577.
4. See, for instance, Knud ]eppesen, Counterpoint, trans. Glen Haydon
New York: Prentice-Hall, Inc.,1939!, pp. 85-86.
5. Winthrop Sargeant, Typesof QuechuaMelody, MusicalQuarterly,
XX 934!, p. 239. The importanceof the process ofornamentation is
discussed in chap. vi of this book.
6. ]osephYasser, ATheory ofEvolving Tonality American Libraryof
Musicology, 1934!.
7. Kunst, Music in Iaoa The Hague: Martinus Nijhoff, 1949!, p. 53.
Actually it is the slndro scalewhich addednew tonesto achieveequi-
distance. The nine-tone scale first referred to is a variation of the seven-
toned plogscale. Butthe additionof thetwo toneswould nothave made
the plogequidistant; temperament was requiredfor this.
8. Hugh Tracey, ChopiMusicians: TheirMusic, Poetry,and Instru-
ments InternationalAfrican Institute; Oxford: OxfordUniversity Press,
1948!. It is possiblethat this scale isan import from India; that it is
derived fromthe scalereferred to in the quotation fromKunst.
9. Curt Sachs, TheRise of Music in the Ancient World, East and West
New York: W. W. Norton & Co., Inc.,1943!, pp. 283 f.
10. Eduard Hanslick, The Beautiful in Music, trans. C. Cohen Lon-
don: Novello, Ewer & Co.!, pp. 48-49. Hanslickscontention thatthis
melody mightjust aswell expressjoy is probably justified;but this does
not mean,as heimplies, thatthe passage is not affective. It only means
that sinceGluck knewthat hecould dependupon thetext to characterize
the affect,it was unnecessary for
the musicto do so.
11. K. Koffka, Principles
of GestaltPsychology NewYork: Harcourt,
Brace & Co., 1935!, p. 414.
12. In this connectionit is important torealize thatthe contextwhich
determines theeffect of repetition is cultural aswell as purely musical.
For instance, while exact repetition will generally be viewed as wasted
effort bythe Westerner,it may seem efficacious and productiveto people
of othercultures: To us, for whom timeis a motion ona space,unvary-
ing repetitionseems toscatter itsforce alonga row of units of that space,
and bewasted. Tothe Hopi,for whomtime isnot a motion buta getting
later of everything thathas everbeen done,unvarying repetitionis not
wasted butaccumulated. Itis storingup an invisible changethat holds
over into later events Benjamin Lee Whorf, Collected Papers on
Metalinguistics [Department of State,Washington, D.C.,1952], p. 39!.
13. ]ames L. Mursell, Psychology and the Problem of Scale, Musical
Quarterly, XXXII 946!, 568.
14. P. R. Farnsworth,The Effect of Repetitionson Ending Prefer-
ences inMelodies, AmericanIournal of Psychology, XXXVII926!,
116-22; and Ending Preferences in Two Musical Situations, ihid.,
237-40. It is alsoimportant to remember thata particular listener may
have learned to understand
several different
styles.
15. W. V. Bingham, Studies in Melody, Psychological Review:
Monograph, Supplement,L 910!.
16. Arthur D. Bissell,The Roleof Expectationin Music New Haven:
Yale UniversityPress, 1921!,p. 19.
17. See]ames L.Mursell, op.cit., p. 569; andBingham, op.cit., p. 86.
18. Paul Hindemith, Craft of Musical Composition, Vol. I, trans.
Arthur Mendel New York: Associated Music Publishers Inc., 1945!
pp. 220-23.
19. Note that this is the case partlybecause ofthe intra-opuspro-
cedures setup thus far. That is, it is partly a result of the fact that the
previous cadences in measures 10, 18,and 24have allbeen deceptive.
20. Bingham,op. cit.,p. 33.
21. Sachs,op. cit., pp. 30-44.
22. For a further discussion ofsuccessive comparison see chap.vii.
23. Obviouslythe fact that a greater timespan isinvolved inthe case
of return plays arole in minimizing differences; for details which might
be noticedif juxtaposedwill often be lostif there is considerable separa-
tion.
NOTES TO CHAPTER V
1. It is interestingto observethat theGestalt natureof melodicshapes

is confirmedby the experience ofmany collectorsof pritnitive and folk
music, whohave foundthat native singers oftenare unableto break off
in the middle of a song and then continue fromthe same place. The
singers haveto go back to the beginningof the song andsing the total
melody the whole shape!,which is not simply an aggregationof its
parts. See, for instance, C. Anderson, MaoriMusic with Its Polynesian
Background PolynesianSociety, Memoire10 [1932-34]!, p. 95; or
Helen H. Roberts, A Study ofFolk SongVariants, Iournal of American
Folklore, XXXVIII 925!, 60.
2. K. Koffka, Principlesof Gestalt Psychology NewYork: Harcourt,
Brace & Co., 1935!, p. 126.
3. Also see thediscussion ofrhythmic incompleteness pp. 143 f.
4. The voices insuch aseries maybe invertedso thatthe uppervoice
is the conjunct oneand the lower voicethe disjunct.
5. Of course, salient differentiation between
harmonies isonly a neces-
sary cause, not asufficient one, for the apprehension of
progression. Other
factors alsoplay an important part in the Hxing of tonal relationships;
e.g., thefact that the toneF in the subdominant harmony limitstonality
on the sharp side, while the tone B in the dominant harmonylimits
tonality onthe flat side.
7. Within the contextof a piece ofmusic silence may appearto form
a continuousground; for then certain attributes alreadyestablished as
given inthe work,such asthe meteror evena repeatedrhythmic group,
are continued subjectively in the mind of the listener, even in the absence
of any objective stimulation.
8. Ibicl., p. 197.
9. Ibid., p. 206.
10. The importance of such trained attention is illustrated by the
difficulty which unpracticed listenersusually havein following a pol-
yphonic texture or by the trouble which Europeantrained musicians
have in apprehending the intricate crossrhythms ofAfrican music see
p. 254!. This againcalls attentionto the powerful influencewhich early
musical practice,even of a rather elementary nature, has upon the re-
sponses ofthe listenerin later life; for it seemsvery probable that a
listeners apprehension of a polyphonic textureor of African crossrhythms
is greatlyfacilitated byhis ownmotor practice-that initially thesecom-
plexities are perceived in terms of the listener-performersown motor
responses.
11. The converse ofthis will also tendto activateexpectation, though
not as strongly; i.e.,textures whichare abnormallythick, closelyspaced,
will be expected tobecome thinner, to separate.
12. See the discussionon pp. 184 ff. of the opening measuresof
Mozarts Piano Concerto in D Minor.
13. That this constitutesa deviationis shownby the fact that of the
eight fuguesin Handel'stwelve Concertithis is the only one in which
the polyphonicweb isdecisively broken.
15. Notice that had this final texture beenpresented byitself, had
there beenno progressiveweakening ofthe figure, it might well have
been understood simply asa rather active ground.
NOTES TO CHAPTER VI
1. C. P. E. Bach, An Essay on the True Art of Playing Keyboard

Instruments, trans.
William Mitchell New York: W.W. Norton & Co.,
Inc., 1949!,p. 150;also see
Sol Babitz,A Problem of Rhythmin Baroque
Music, Musical Quarterly, XXXVIII 952!, 533-65.
2. Oliver Strunk, SourceHeadings in Music History New York:
W. W. Norton &Co., Inc.,1950!, p. 607.
3. Grove'sDictionary ofMusic andMusicians, ed.H. C. Colles New
York: Macmillan Co., 1936!, I, 635.
4. David D. Boyden,The Violin and Its Technique inthe 18th Cen-
tury, Musical Quarterly, XXXVI950!, 35-36.
5. Carl E. Seashore,
Objective Analysisof the Musical Performance,
Studies inthe Psychologyof Music, Vol. IV Iowa City: University of
Iowa Press, 1937!, p. 26.
6. Milton Metfessel, Phono-photographyin Folk Music Chapel Hill:
University of North CarolinaPress, 1928!,p. 11, 12; seethe introduc-
tion by Carl E. Seashore. Noticethat Seashores word flirtation comes
very closeto the idea of play mentionedearlier in this study. Also see
Raymond B.Stetson A Motor Theory of Rhythm and Discrete Suc-
cession, Psychological Review, XII [1905], 337!, who notes that aside
N otesto Pages 202-7 285
from such irregularities whichthe rhythm requires, thereare various
minor variationsfor the purpose ofexpression.
7. Carl E. Seashore,A Base for the Approach toQuantitative Studies
in the Aesthetics ofMusic, American ]ournal of Psychology, XXXIX
927!,141-44.
8. Carl E. Seashore,Psychology ofMusic New York: McGraw-Hill
Book Co.,Inc., 1938!,p. 212.
9. GeorgeHerzog, GeneralCharacteristics ofPrimitive Music ab-
stract! ,Bulletin ofthe AmericanMusicological Society,
VII 942!, 24.
10. Frances Densmore, Chippewa Music, Smithsonian Institution,
Bureau ofAmerican Ethnology, Bulletin 45[1910] ,! p. 4.
11. GeorgeHerzog, Folk Song, Dictionary of Folklore, Mythology
and Legend New York: Funk & Wagnalls Co.,1949!, p. 1041. Notice
the juxtapositionof the words expressiveand ornamental; the connection
is not accidental.
12. BelaBartok andAlbert B. Lord, Serbo-Croatian Folk Songs New
York: ColumbiaUniversity Press, 1951!, p. 4. In addition to the experi-
ence of such singing,our acceptanceof these particular deviationsin
this particularsituation isprobably abetted
by ourbelief in their purpose-
fulness.
13. R. H. van Gulik, TheLore of the ChineseLute: An Essay inChin
Ideology Tokyo:Sophia University, 1940!, p.75.
14. Ibid., p. 77.
15. ]app Kunst, Musicin Iaoa The Hague:Martinus Nijhoff,1949!,
p. 59.
16. A. K. Coomaraswamy, Figures of Speech orFigures of Thought
London: Luzac& Co., 1946!, p. 86.
17. Thus in the examples analyzedearlier in this study, e.g., the
Hindemith example,p. 140, the basicoutline givesthe structuraltones,
and the other tonesand progressions
may be considered tobe an orna-
mentation of this basic structural line.
18. Alain Danielou, Northern Indian Music London: Christopher
]ohnson, 1949!,p. 102.The quotationis from the NatyaShastra, acom-
pilation of theoretical writings,which has been variouslydated from
the second century B.c.to thefourth centuryA.D.
19. Gustave Reese, Music in the Middle Ages New York: W. W.
Norton & Co., Inc., 1940!, pp. 204 ff.
20. From Blanchet's Artdu Chant, quoted in Groves Dictionary,op.
cit., III, 769.
21. FrancoisRaguenet, AComparison Between the Frenchand Italian
Music anon. trans. ca.1709!, MusicalQuarterly, XXXII 946!, 429.
22. Bach,op. cit.,p. 84.
23. Ibid., p. 322.
24. Knud Ieppesen, Counterpoint,
trans. GlenHaydon New York:
Prentice-Hall Inc., 1939!, p. 147.
25. Wilfrid Mellers, FrancoisCouperin London:Dennis Dobson Ltd.,
1950!, p. 305. Similareffects arevery commonin oriental music. See,
for instance,]an La Rue, The Okinawan NotationSystem, Journalof
the AmericanMusicological Society, IV 951!, 30; or Danielou, op.cit.,
pp. 104ff.
26. Raguenet,op. cit., pp. 417-18.
27. This quotation fromHeinrich Glarean,Dodecachordon, Book III,
chap. xxiv,is givenin OliverStrunk, op.cit., pp.222-23.
28. Iazz Journal, ]anuary,1950, quotedin Rex Harris, Iazz Har-
mondsworth, Middlesex:Penguin Books,1952!, p. 349.
29. Kunst, op. cit., p. 401. _
30. A. K. Coomaraswamy,India Music, Musical Quarterly, III
917!, 165.
31. Sachs, The Rise of Music in the Ancient World, East and West
New York: W. W. Norton &Co., Inc.,1943!, pp.108, 143.For specific
illustrations ofsuch classifications
see Danielou,op. cit., pp. 104 ff.; and
La Rue, op. cit., pp. 31-32.
32. Kunst, op. cit., p. 276.
33. Ibicl., pp. 333-34.
34. Ibid., p. 277.
35. A. K. Coomaraswamy, The Danceof Siva New York: The Sun-
rise Turn Inc., 1924!, p. 76. For a very similar statement seeA. H. Fox
Strangways, The
Music ofHinclostan London:Oxford UniversityPress,
1914!, p. 182.
36. Up. cit., pp. 146-47.
37. Seeexamples, ibid.,
p. 188.
38. A. Z. Idelsohn,Iewish Music New York: Tudor PublishingCo.,
1948!, pp.98, 183.
39. Sachs, op. cit., p. 83.
40. PercyGrainger, TheImpress ofPersonality inUnwritten Music,
Musical Quarterly,I 915!, 422.
41. From the introductionto Bartok and Lord, op. cit., pp. xii-xiii.
The relationbetween abundant ornamentation andfree or rubato tempo
is by no meansunique. Writingof musicin the early baroque,Bukofzer
notes that the accumulation of intricate embellishments sostrongly
affected therhythm thatthe musiccould nolonger beperformed instrict
time Manfred F. Bukofzer, Music in the BaroqueEra [New York:
W. W. Norton & Co., Inc., 1947], p. 28!.
42. Bartok and Lord,op. cit., p. 74.
43. Phillips Barry, FolkMusic in America AmericanFolk SongPub-
lications, No. 4, WorksProgress Administration,
Federal TheatreProject
[New York: National ServiceBureau, 1939]!, p. 112. The phraseis a
quotation from
In articleby RobertW. Cordon,which appeared
in the
New York Times November 27, 1927.
44. Metfessel,op. cit., p. 21.
45. Ibid., p. 48.
46. Winthrop Sargeant, Iazz: Hot and Hybrid New York: E. P. Dut-
ton, 1946!,p. 24. Of course,there needbe noprinted guideso long
as theseveral players
know the tune andplan.
47. Harris, op. cit., pp. 62-63.
48. Ibicl., p. 155.
49. RichardA. Waterman, Hot Rhythm in Negro Music, Journal
of the American Musicological Society, I 948!, 31.
50. This is emphasizedby Willard Rhodes inhis article, Accultura-
tion in North AmericanIndian Music, Selected Papers of the XXIXth
International Congress of Americanists,ed. SolTax Chicago:University
of ChicagoPress, 1952!, pp. 127-32;see inparticular p.130.
51. Ornamentation might itself be saidto be a partial basis ofmusical
style. Withina culturalarea styleis not so mucha matterof fundamental
tunes but of differencesin ornamentation.Discussing themusic of the
Maoris of New Zealand,Anderson writes:One tribe might adopt a
melody fromanother tribe,and theoutline wouldprobably beunaltered,
but the adventitious ornamentation of the melody mightbe varied-in
fact, Sir Apirana Ngatahas told me that any Maori familiar with their
music couldtell from the ornamentationwhat tribe was responsible for
that particularform of the melody." C. Anderson, Maori Music with
Its PolynesianBackground PolynesianSociety, Memoire10 [1932-
34]!, p. 191.
52. A. M. ]ones, African Music, African Affairs, XLVIII 949!,
295.
53. GeorgeHerzog, Speech-Melodyand Primitive Music, Musical
Quarterly, XX 934!, 456.
54. Helen H. Roberts,Melodic Compositionand ScaleFoundations
in Primitive Music, Journalof AmericanAnthropology, XXXIV932!,
80.
56. The expectations aroused by the active toneson any given archi-
tectonic levelare notsolely theproduct ofthe functionsof the individual
tones. Forthe expectations engendered byany giventone are not only
a product of its function andposition in the tonal system butare also
a resultof the tones whichhave preceded it. It is likewiseclear thatthe
rhythmic placement of tonesaffects theirdegree ofactivity or rest.
57. Kunst, op. cit., p. 51.
58. Ibicl., p. 51, n. 3.
59. Sachs, op. cit., p. 134.
60. ylbicl.,p. 133.
61. Herzog,Folk Song, op. cit., p. 1043.
62. Indeed,the authorknows of no tonal system thatis not diatonic.
The twelve-tonesystem ofSchoenberg and his followersis no exception
to this rule becauseit is intentionally andexplicitly non-tonal.The tem-
pered five-toneslndro scaleof Malaya, a seemingexception, appears,
in origin at least,to have been diatonicand it seems possible that it is
still heardas such Sachs, op.cit., pp. 130-31!.
63. Suchembellishing pitch deviations arenot, it must beemphasized,
to be confused withthe microtonal elements, e.g.,the sruti in Indian
music, whichfind a place in the theoreticalfoundation forthe various
modes andragas ofIndian or Arabian music;see Strangways, op. cit.,
chap. iv.
64. Strunk,op. cit.,pp. 602f.; and Bach, op.cit., p. 163.
65. Otto Combosi, The Pedigree of the Blues, Proceedings ofthe
Music Teachers, National Association, Series XL946!, p. 385.
66. Edward E. Lowinsky, Secret ChromaticArt in the Netherlands
Motet, trans. Carl Buchman New York: Columbia UniversityPress,
1946!,p.79.
67. The example istaken fromGroves Dictionary,I, 645.
68. Indeed, it is important to realize that melodic chromaticism which
occurs overa static harmony andonly on unaccented beats, as, for in-
stance, in the opening melody of Debussys Afternoonof a Faun, is
really arapidly passing
effect which may haveonly aminimal embodied
meaning andgive riseto but little affectiveresponse sinceit inhibits or
delays noimportant tendencies. However, suchpassages may give rise
to important designative meanings and, asis pointed out in chap. viii,
may arouseaffect throughthem.
69. ]ohn Brown, Lettersupon the Poetry and Music of the Italian
Opera Edinburgh:Bell andBradfute, 1789!,pp. 12-16.
70. A. Montani, Psychoanalysisof Music, Psychoanalytic Review,
XXXII 945!, 225-27.
71. SeeHerzog, Folk Song, op. cit.; andBarry, op.cit., Introduction.
72. C. P. Heinlein,The Affective Characters ofthe Major and Minor
Modes in Music, Iournal of Comparative Psychology, VIII 928!,
101-42.
73. Nevertheless, the influence of the minor mode in Western music
and onWestern listeners has beena verypowerful one,as evena cursory
glance atthe literaturewill show. See, forexample, KateHevner, The
Affective Characterof the Major and Minor Modesin Music, American
Iournal of Psychology, XLVII935!, 103-18.
74. Bukofzer,op. cit.,p. 287.
75. Erich M. von Hornbostel, African Negro Music, Africa, I
928!,16.
76. Sachs,op. cit., p. 125.
77. Gustave Reese, Music in the Renaissance New York: W. W.
Norton &Co., Inc.,1954!, p. 400. It seems possiblethat the minor mode
is not only a deviation from the norm of the diatonic but also froma
deeply rootedfeeling formajor, whichis presentin the West. SeeSachs,
op. cit., pp. 2.95-311;also seeGeorge Herzog,Some Primitive Layers
in European Folk Music, Bulletin of the American Musicological So-
ciety, IX 947!, p. 13. If it is true that our modernmajor modeis a
primordial normin Western musical culture,then the minor mode is,
so tospeak, adoubt deviant:it deviatesboth fromthe interculturalnorm
of diatonicismand fromthe normof major,which is specific toEuropean
culture.
78. Heinlein,op. cit.,pp. 136-37.
79. VincenzoGalilei, Dialogo della musicaantica e della moderna,
quoted inStrunk, op.cit., pp. 315-16.
80. Bach,op. cit.,p. 163.
81. GioseffeZarlino, Istituzioniarmoniche, Book
III, quoted in Strunk,
op. cit., p. 232.
82. Anderson,op. cit.,p. 63.
83. This literature neednot be reviewed heresince it is admirably
summarized inan article by M. Guernesey, TheRole of Consonance
and Dissonance in Music, American Iournalof Psychology,XL 928!,
173-204.
84. Norman Cazden, Musical Consonance and Dissonance: A Cul-
tural Criterion, Iournal of Aesthetics, IV 945!, 4-5. This general
position is supported byseveral authors.Guernesey, op.cit., concludes
that consonance is an aesthetic description,totally dynamic in nature,
and is not a scientifically determinable
constant and that its perception
depends upontraining, environment,and musicalcontext. The impor-
tance ofcontext isalso emphasized by P. A. D. Gardner andR. W. Pick-
ford, Relation between Dissonanceand Context, Nature, CLII 943!,
p. 358. The importanceof cultural factors in the perception of con-
sonance isstressed inan article by R. W. Lundin, Toward a Cultural
Theory of Consonance, Iournalof Psychology, XXIII 947!, 45-49.
85. Willi Apel, Harvard Dictionary of Music Cambridge, Mass.:
Harvard UniversityPress, 1945!,p. 18.
86. Nor is it diflicult to account for the fact that the dissonance norm
has constantlyrisen in Western culture.For it seems likelythat when
a vertical combination of sound has been heard often enough as a unit,
it achieves the status of an independent, unifiedGestalt, completein
itself. It becomes anorm and ceases toperform its affective aesthetic
function adequately.Therefore, thecomposer, seeking
for aestheticeffect
and expressionand wishing to explore less commonpaths, will tend
to treat what wasformerly a deviant asa norm and usethat which was
formerly unusedor forbiddenas adeviant.
NOTES TO CHAPTER VII
1. ]app Kunst, Musicin Iava The Hague: Martinus Nijhoff, 1949!,

. 59.
P 2.Curt Sachs,
The Rise
of Music
in the
Ancient World,
East and
West New York: W.W. Norton& Co.,Inc., 1943!,p. 145.
3. Eta Harich-Schneider, The Present Condition of ]apanese Court
Music, MusicalQuarterly, XXXIX953!, 58.
4. E. Cunningham, The]apanese Ko-uta and Ha-uta, Musical Quar-
terly, XXXIV 948!, 71.
5. Harich-Schneider, op. cit., p. 53.
6. Kunst, op. cit.,p. 157.
7. A. H. Fox Strangways, The Music of Hindostan London:Oxford
University Press, 1914!, p. 226.
8. Ibid., p. 238.
9. Curt Sachs, Rhythmand Tempo New York:W. W. Norton 8:Co.,
Inc., 1953!, chap. vi.
11. Winthrop Sargeant and
Lahiri, A Study ofEast IndianRhythm,
Musical Quarterly, XVII 931!, 435-36.
12. Ibid., p. 434.
13. A. M. ]ones,African Music, African Affairs,XLVIII 949!, 294.
14. All of theseexamples are
taken fromA. M. ]ones, African Music:
The Mganda Dance, African Studies, IV 945!, 180-88; also see
A. M. ]ones, The Study of African Musical Rhythm, Bantu Studies,
XI 937!, 295-320; and R. Brandel, Music of the Giants and the
Pygmies ofthe BelgianCongo, Iournal of the American Musicological
Society, V952!, 16-28.
15. ]ones, African Music, op. cit., p. 294. Also seeA. M. ]ones,
African Music in Northern Rhodesia and Some Other Places Occa-
sional Papersof the Rhodes-Livingston Museum[Oxford University
Press, 1949]!,pp. 20,78.
16. Erich M. von Hornbostel, African Negro Music, Africa, I
928!, 52.
17. Richard Alan Waterman, African Influence on the Music of the
Americas, Selected Papers of the XXIXth International Congress of
Americanists, ed.
Sol Tax Chicago: Universityof ChicagoPress, 1952!,
p. 213.
18. Ihid., p. 213.
19. Ibid., p. 214.
20. Winthrop Sargeant, Jazz:
Hot and Hybrid New York: E. P. Dut-
ton, 1946!, pp. 238-39.
21. Hugh Tracey, ChopiMusicians: TheirMusic, Poetryand Instru-
ments, International
African Institute London: OxfordUniversity Press,
1948!, p. 101.
22. Hornbostel, op. cit., p. 48.
23. Ibid., p. 47.
24. I am indebtedto ProfessorScott Coldthwaite, not only for calling
my attentionto this example ofrhythmic crossing, but alsofor permitting
me to use histranscription ofthe chanson.For a much morecomplex
example from the earlyfifteenth centurysee A.T. Davidsonand W.Apel,
Historical Anthologyof Music Cambridge, Mass.:Harvard University
Press, 1949!,p. 51, No. 48a.
25. See Tracey, op. cit., p. 91.
26. Such free preludesmay eventuallybecome moreor less stand-
ardized withinthe style,may becomefairly fixedforms whichare them-
selves normsof the style.
27. Sachs,The Rise of Music in the Ancient World, pp. 191, 285,
and 290; Strangways, op. cit., p. 281; Harich-Schneider;
op. cit., p. 60;
]ones, African Music: TheMganda Dance,op. cit., p. 185; and Kunst,
op. cit., pp. 310-11.Although thesepreludes mayat timesalso serveas
display piecesin which the individual performer mayexhibit his in-
genuity and virtuosity in improvising uponthe mode,this is not their
basic aesthetic function.
28. I have quotedonly the beginnings, theverse parts,not the con-
necting materialof a song givenin Davidsonand Apel,op. cit., pp. 3-4.
Note the presence ofboth successiveand simultaneous deviation. Not
only are the lcotopart and voice part varied melodicallyand rhythmi-
cally from verse to verse, but within each verse the accompaniment
deviates fromthe song proper or vice versa.
29. Strangways, op. cit., pp. 282-83.
30. Ihid., chap. xi.
31. A. Z. Idelsohn, Iewish Music New York: Tudor Publishing Co.,
1948!, pp. 24-25.
32. Egon Wellesz, Wordsand Musicin ByzantineLiturgy, Musical
Quarterly, XXXIII 947!, 306-7.
33. Egon Wellesz, Byzantine
Music andHymnography London:Ox-
ford University Press, 1949!, p. 286.
34. Idelsohn, op. cit., p. 370.
35. Sachs,The Riseof Musicin the Ancient 92Vorld,
p. 84.
36. Helen H. Roberts, A Study of Folk Song Variants Based on Field
Work in ]amaica, Iournal of American Folklore, XXXVIII 925!, p.
155.
37. Ibid., p. 168.
38. Ibid., p. 158.
39. Ibicl.,p. 167.
40. Ibid., p. 167.
41. Ibid., p. 215.
42. Bela Bartok andAlbert B. Lord, Serbo-Croatian
Folk Songs New
York: ColumbiaUniversity Press,1951!, pp. 114-15.
43. Sargeant,Jazz: Hotand Hybrid,pp. 156-57.The useof a common
ground bass, a commonharmonic sequence, can alsobe foundin Western
art musicof the sixteenth andseventeenth centuries;
for the relationship
between thesebasses andthose found in hot jazz seeOtto Combosi, The
Pedigree ofthe Blues, Proceedings ofthe MusicTeachers National
As-
sociation, Series XL 946!, 382-89.
44. Willard Rhodes, Acculturation in North American Indian Music,
Selected Papers of the XXIXth International Congress ofAmericanists,
ed. SolTax Chicago:University ofChicago Press, 1952!, p. 130.
45. ]ones,African Music, op. cit., pp. 292-93; also seeTracey, op.
cit., pp. 91 f.
NOTES TO CHAPTER VIII
1. T. MacCurdy,The Psychology of Emotion New York: Harcourt,

Brace &Co., 1925!,p. 568.
2. One of the dangers ofdescriptive programnotes andpoetic criti-
cism is that they tend to initiate suchimage processes, which later de-
velop withoutreference tothe musicitself.
3. For a detaileddiscussion ofthe relationshipbetween memoryand
affect seeDavid Rapaport,Emotions andMemory New York: Interna-
tional UniversityPress, 1950!.
4. After citing someinteresting experiments dealing with the inter-
sensory character of experience, Koffka observes
that we must, inaccord-
ance with our theory,conclude thatperceptual space is one and that it
can be filled with objects ofdifferent sensemodalities . . . K. Koffka,
Principles of Gestalt Psychology[New York: International University
Press, 1952],p. 303!.
5. SusanneK. Langer,Philosophy ina New Key New York: Mentor
Books, 1951!,p. 155.
6. Probablyone ofthe fundamentalconditions forthe existenceof any
connotative complexis the presence ofa single-moodresponse which
is common to all of its components. Indeed, connotation and sentiment
are so inextricably united that every connotative experience is to some
extent a mood experience as well.
7. Becausethe chainof responses usually takes
place withsuch rapidity
that there is no noticeable delay
between theinitial perceptionof mood
and the final affectiveexperience, themediating stimulusis less likely
to be a connotative reference involving some sort of conscious cognition
or re-evaluation of the original stimulus, i.e., the music.
8. Eduard Hanslick, The Beautiful in Music, trans. Gustav Cohen
London: Novello, Ewer & Co., 1891!, p. 25.
9. C. C. Pratt, Structural vs. Expressive Form in Music, Iournal of
Psychology, V938!, p. 150.
10. The need forvariety isundoubtedly areason fora changeof mood
or connotation,
but it does notprovide arationale forthe orderof change.
However, once a particular succession hasbeen established in a com-
position, wemay expecta return of the succession or, if one of the parts
does return,we mayexpect theother partsto return in their established
order. Butthis doesnot accountfor the order ofthe originalsuccession.
11. It is, of course, possible
for a rationale ofmood and connotative
complex succession to becomea feature of a particular style,for a set
of mood orders tobe establishedas part of a tradition. The opera com-
posers, thelibrettists, andthe theoristsof the early eighteenthcentury
were oftenconcerned withthis very problem, andrules werelaid down
for the sequence of moods. Weresuch anorder tobecome areally well-
known featureof the style, it would presumably not only establish area-
sonable relationship between moodsbut would also haveimportant ef-
fects uponthe embodiedmeaning of musical experience.
Index
Accent: defined, 103 f.; distinguished Uncertainty!; see also Behavior,

from stress, 104; on rest, 149; stress emotional; Philosophy
of music
on, may alter grouping,104, 107f.; African music: deviation and orna-
see alsoBeat; Meter;Rhythm mentation in, 214; differences of,
Aesthetic belief; see Belief, aesthetic from jazz,243; heterophonyin, 244;
Aesthetic experience:
belief in special rhythmic structureof [Ex. 95-98],
nature of, 73 f. see also Belief, 239-42; tendency toward equidis-
aesthetic!; problematicalin primi- tance in, 133
tive music, 239; responsesnot overt Aiken, Henry D., on aesthetic beliefs,
in, 9, 74, 273 n. 15!; see also Af- 9, 73 f.
fective experience; Meaning; Play Ambiguity: affect and, 16 see also
Affect: differentiation of, 16-20; dis- Suspense; Uncertainty!; architec-
tinguished fromaffective experience, tonic level modifies view of, 52;
19; distinguished from mood, 7; chromaticism and, 164, 218 f.; ex-
meaninfg 39-40;
and, subjective
na-
ture o , 12; 256; suppressionof
pectation and, 26, 51, 171; of
formal function, 182-84; harmonic,
overt behavior intensifies, 9, 74; 95, 166, 179-82; melodic, 163-67;
tendency inhibition creates, 15-16; metric, 144 f.; of minor mode, 266;
see also Affective experience; Psy- mitigated by good shape,174; na-
chological theoryof emotions;Sus- ture of, 51-52; and probability, 52;
pense rhythmic seeRhythm!; of texture,
Affective experience: basis of, in 192-95; seealso Suspense;Uncer-
music, 31; belief in mental control tainty
modifies, 19 f.; characterized by Amphibrach rhythm;see Rhythm
stimulus situation,19; conquestof Anapest rhythm;see Rhythm
difficulties an aspect of, 70; dif- Anxiety; seeSuspense; Uncertainty
ferentiation of,in music theory, 17, Appoggiatura: affective function of,
20; expectationand, 23-25, 30-32 101, 210 f.; inhibits tendencies,
see also Expectation!; Hebb'sac- 100 f., 207; see also Ornamenta-
count of, criticized, 19; image pro- tion
cesses create,256-58; inhibition and Architectonic levels:ambiguity ondif-
conflict create, 15, 16 f.; intellectual ferent, 52;completeness and closure
and, complementary,39 see also on different, 130; deviation on dif-
Meaning!; MacCurdys analysis of, ferent, 215f.; meaningon different,
14 f.; mood and connotation color, interdependent, 47; melodic-har-
269 f. see also Connotation; monic uniformity on higher, 168;
Mood!; mood indirectly evokes, metric organization on different,
268; necessarilydifferentiated, 19; 115 f.; ornamentation on different,
non-referential character of, in 205; overemphasis on highest,52 f.;
music, 20;Paulhan's analysis
of, 15; rhythmic incompletenesson differ-
play and, 70 see also Play!; sul ent, 144-48; rhythmic organization
generals musical, 20; suspense on different, 110-15
creates, 27-29 see also Suspense; Art: academic, traditional, and deca-
295
296 Index
Art continued! Beethoven: PianoSonata, Op.53, iii,
dent, distinguished,71; criticismof, interaction of designative andem-
and belief, 76 bodied meaning in, 269; Piano
Association: by contiguity, 259; by Sonata, Op.81a, i, introduction dis-
continguity, changes in, 260; by cussed, 53;Piano Sonata,Op. 111,
contiguity, intracultural, 259; by i, effect of fugue in, 154; Piano
contiguity, rolein mood representa- Sonata, Op.111, i,introduction dis-
tion, 267; inclination toward, 262; cussed, 60;String Quartet,Op. 130,
partl aural, 261; by similarity, i, bifurcation of processin, 126 f.;
260 ill;
by similarity,
of concepts String Quartet,Op. 130,i, introduc-
with motions, 261, 274 n. 27!; by tion discussed,60; String Quartet,
similarity, learning modifies, 262f., Op. 131, v [Ex. 53], law of return
267, 270 f.; see also Connotation; in, 154 f.; String Quartet, Op. 131,
Image processes; Mood v, rest accented in, 103 f., 149;
Audience, inHuence of, on belief, 76 f. String Quartet,Op. 131,v [Ex. 49-
52], rhythmic organization of,145-
Bach, CarlPhilipp Emanuel:on affec- 49; andstyle change, 71; Symphony
tive power of dissonance, 229; devi- No. 1, iii [Ex. 16], rhythmic or-
ation related to affect by, 200; on ganization of, 105, 111; Symphony
motor behavior, 80; on ornaments, No. 3, i, opening chords discussed,
207; on placement ofornaments, 50 36; SymphonyNo. 3, ii [Ex. 39],
Bach, _lohann Sebastian: Brandenburg anticipation changes rhythm in, 125;
Concerto No. 5, i [Ex. 42], struc- Symphony No.6, i [Ex. 44], satura-
tural gapsin, 134; Mass inB Minor, tion in, 135 f.; Symphony No.9, i
Crucifixus, affect and chromati- [Ex. 82-83], weakening of texture
cism in, 220; MusicalOffering [Ex. in, 193-95; Symphony No. 9, ii
89], chromaticism in, 224 f.; Partita [Ex. 19], dactyl grouping in, 105;
No. 3, iii, anapest rhythm in, 106; Symphony No.9, ii, metric organi-
and stylechange, 71;Well-tempered zation of, 116
Clauier, Vol. I, i, reiteration and Behavior, designative, 17; see also
return in, 153; Well-tempered Emotional designation
Clavier, Vol.I, ii [Ex. 2], change of Behavior, emotional: automatic, not
meaning in,48 f. differentiated, 18;characterized by
Baroque music: chromaticism in, stimulus situation, 19; differentia-
219 f., 228; connotation in, 8, 267; tion of, learned, 17; interpretation
expressive performancein, 200 f.; of, problematical,9; standardization
harmonic practice in, 66; improvisa- of, 10, 21
tion in, 70, 206 f.; ornamentation Belief, aesthetic: Aikens discussion of,
in, 206-8; use of vibrato in, 66 9, 73 f.; dependence of saturation
Bartok: on ornamentation in folk on, 135; differentiates musical re-
music, 212,253; on pitch deviation sponses, 40; importance of, in am-
in folk music, 204; String Quartet biguity, 27; influences on, 75-77;
No. 3, ii [Ex. 100], metric crossing in integrity of composer, 75; and
in, 245 f.; String Quartet No.4, iii, suppression of overt responses, 9,
no vibrato" marked in, 66; String 74, 273 n. 15!; and mechanism of
Quartet No.5, v [Ex. 61], sequence denial, 74; part of preparatory set,
in, 169 f.; transcription of folk song 73; physiologicalchanges relatedto,
[Ex. 1041, 253 11, 74;and thepresumption oflogic,
Beat: accented, stress on,alters group- 75 f.; in seriousness of art, 74
ing, 104, 107 f.; unaccented, group- Berg: Lyric Suite, v [Ex. 77], spacing
ing alters placement of, 104, 108; of texturein, 189f.; Violin Concerto,
unaccented, stressdoes not make ac- no vibrato" marked in, 66
cented, 104 Berlioz, Symphonia Fantastique, ii
Index 297
[Ex. 73], minimal differences in, 30 f.; expectation and,31; objecti-
180 f. fication of meaning and,38 f.
Bizet, Carmen, Act III [Ex. 17], ana- Cohen, MorrisR., on meaning, 33f.
pest rhythm in, 105 Communication: dependenceof, on
Brahms: Piano Sonata No. 2, iv [Ex. style, 42;emotional behaviorstand-
56], uniformity in, 166; Rhapsody, ardized for,10, 21;role of composer,
Op. 119, No. 4 [Ex. 21], stress af- performer, andlistener in,40-42
fects rhythmin, 108;Symphony No. Completeness: architectonic nature of,
4, i, iambic rhythm in, 106; Violin 130; closure and, related to satura-
Sonata No. 3, i [Ex. 90], minor tion, 135; dependence of,on pat-
mode in, 225 tern, 129, 138; expectation and,
Bruckner, Symphony No. 7,i [Ex. 55], 128-30; goodcontinuation and,129;
uniformity in, 165 f. harmonic, 149-51; hypothetical
Byzantine music:style changes in, 67; meaning influencedby, 49; inter-
successive deviation in, 250 f. opus norms influence, 129; and
learning, 78,138; melodic,depend-
Cadences, changingcharacter of, 66 ence of, on tonality, 138; of meter,
Cassirer, Ernst, on emotion in art, 18 144; not entirely cultural, 132;
Cazden, Norman, on nature of con- placement ofdeviants influenced by,
sonance and dissonance, 230 f. 50; principleof return and, 151-56;
Chance, placeof, in music, 195f. register influences,139; of rhythm,
Chinese music: ailective deviations in, 143-49; styleinfluences, 128 f.; tex-
204, 216; tone names and prob- ture influences,187; understanding
ability, 56 influences, 138 f.; see also Closure;
Chopin: Prelude,Op. 28,ii [Ex. 5-8], Incompleteness
analysis of, 93-97; Waltz, Op. 34, Composer: andcommunication, 40f.,
ii, chan e of mode in, 228 278 n. 24!; conquestof difficulties
Chord prolongation,
concept
criti-
of, and, 69 f., 199; influence of extra-
cized, 53 stylistic forceson, 71;a listeneralso,
Chromaticism: affective nature of, 41, 276 n. 51!; style influencedby,
218-22; both melodic and har- 69-71
monic, 220; effect of, and Law of Connotation: aesthetic problem of,
Pragnanz, 217; in Haydn Sonata, 27 0-72;dependence of, on training,
175; of minor mode, 224 f.; and 264; distinguished from imageproc-
suspense inopera, 220; a type of esses, 258; habit makes effective,
deviation, 217f.; a type of orna- 260; and learning, 2, 21, 270 f.;
mentation, 218;a typeof uniformity, mechanism of see Association!;
164, 168;use of,in classicalperiod, mood and, 266 see also Mood!;
221 f. music necessary cause of,264, 271;
Class concepts:an aspectof style, 57; presents essence of myth, 265, 271;
and expectation, 57 f.; and form,57; relation ofprogram to,271 f.; repre-
formation of, 57, 89 f.; modification sentation of, depends on cultural
of, 58 f., 90; see also Form concepts, 258f., 262; special sys-
Classical period: function of poly- tems of, 8, 260; specificity of, and
meters in, 123; use of chromaticism divergence, 263; specified by lis-
in, 221 tener, 266;see alsoMeaning, desig-
Closure: dependenceof, on incom- native
pleteness, 155 f.; lack of, in Haydn Connotative complex, presents essence
Symphony, 130f.; melodic com- of myth, 265, 271
pleteness and, 138-39; recurrence Consonance: Cestalt approach to, 231;
and, 152; relaxation and, 139; see past accountsof, untenable,230 f.;
also Completeness;Incompleteness see also Dissonance
Cognition: conscious or unconscious, Contemporary music: metric organiza-
298 Index
Contemporary musiccontinued! 72; successive,246-54; a type of

tion of, 119 f., 121, 245 f.; uniform- tendency inhibition,32; vibrato an
ity in, 169 f.; use of vibrato in, 66 instance of, 66, 201; see also Devi-
Context: effect of repetition depend- ants; Ornamentation; Process con-
ent on, 126; effectof sequencede- tinuation; Tonalorganization
pendent on, 171; effectof stylistic, Dewey, ]ohn:on aestheticexperience,
on shape,185; expectation modified 31; emotionaltheory of, 14
by, 30; importance of, in evalua- Differences, minimal; see Minimal dif-
tion of shape, 91f.; importance of, ferences; Subliminal differences;
in saturation, 136, 282 n. 12!; Uniformity
meaning dependent on, 46 Difhculties, conquest of: composer
Continuation, Law of Good, 92 f.; see and, 69 f.; intrapersonal at times,
also Process continuation 278 n. 24!; performer and, 70; re-
Control, mental: belief in, modifies af- lated to aesthetic experience, 69 f.,
fective experience,19 f.; ignorance, 199; seealso Play
suspense, and, 27 f.; uniformity Dissonance: affective power of, 229 f.;
weakens sense of, 163, 169, 171, in EastAsiatic music,235; in primi-
174, 176;see alsoUncertainty tive music, 230; see also Consonance
Criticism, influence of, on belief, 76 Divergence, connotation and mood
speciied by,263, 268
Dactyl rhythm; see Rhythm Doubt; seeSuspense; Uncertainty
Daniel, ]ohn, Chromatic Tunes [Ex.
84], affective nature of chromaticism East Asiatic music: deviation in,
in, 219 235 f.; ornamentation and affective
Debussy: Afternoonof a Faun, effect experience in, 209 f.; see also Chi-
of flute solo in, 78; Des Pas sur la nese music; ]apanese music; ]ava-
Neige [Ex. 45], ostinato in, 137; nese music
Ftes [Ex. 36], uniformity weak- Embellishment; see Ornamentation
ens polymeter in, 122; Nuages Emotion-felt; see Affect; Affective ex-
[Ex. 58], uniformity of chords in, perience; Psychologicaltheory of
167 emotion
Densmore, Frances, use of statistics Emotional behavior; see Behavior,
by,55 emotional
Designative meaning;
see Meaning Emotional designation:
affect neednot
Designation, emotional; see Emotional involve, 282 n. 10!; as communica-
designation tion, 21; cultural character of, 21;
Determinate meaning;see Meaning mood and, 226 f.; need not involve
Deviants: effectof, dependenton posi- affect, 8;not peculiarto music, 22
tion, 50; exaggeration of,65; for- Emotional response to music, 6-13:
gotten or normalized by memory, objective behavioras evidence of,
90; function as signs,68; tendency 9 f.; physiological changes influ-
of, to become norms, 65, 68 f.; see enced bybelief, 11, 74; physiologi-
also Deviation; Norms cal responses, 10-13; subjective
Deviation: chromaticism a type of, evidence of, 6-9
217 f.; composer and, 69-71; expres- Emotional states,
musical depictionof,
sive function of, 200-205, 252; conventional, 8; see also Behavior,
form and, 57 f.; importance of, in emotional; Emotionaldesignation
African music, 242; in jazz, 243; Emotions: named and classified, how,
ornamentation a type of, 200, 204; 8; pleasantand unpleasant,concept
performance and, 199-204; related of, criticized, 19 f.; see also Affect;
to aHective experience, 32; Sea- Affective experience; Emotional
shores views on, 201; simultaneous, designation; Psychological
theory of
234-46; style change and, 65-69, emotion
Index 299
Equidistance, tendencytoward, 132- affects meter,122; dispositionof, in
34: and equal temperament, 133 f.; ostinato, 137; incompleteness of,
tonal evolution explained by, 133; 190; see also Texture
see alsoStructural gaps Folk music: deviation in, and aesthetic
Evidence: explanationof minor mode experience, 212f., 252; expressive
as, 222;explanation ofstyle change deviation in, 204; ornamentation in,
as, 65; explanation oftendency to- 211 f.; structural gapsfilled in, 132;
ward equidistanceas, 133f.; indi- style changesin, 67; successive de-
rect, 254f.; natureand problemsof, viation in, 252 f.; see also]azz
197-99 Forgetting, of weak patterns,89; see
Evident meaning;see Meaning also Memory
Examples, basis for choiceof, 44 f. Form: architectonic nature of, 56, 205,
Expectation: affectand, 31;ambiguity 215; classconcepts of,57 f.; expec-
and, 26 f., 51 f. see also Ambigu- tation and,59; experienceof, modi-
ity!; chance and, 196; class con- fies perception,59; principle of re-
cepts a basis for, 57 f.; cognition turn and, 152-56; probability and,
and, 30 f.; completenessand, 49f., 57; revision of opinion and under-
128-30 see also Completeness!; standing of, 49, 60; rhythm and,
conscious and unconscious, 25, 30- 112
32; context influences, 30, 185; con-
tinuation and see Processcontinua- Galilei, Vincenzo, on dissonance, 229
tion!; equiprobable consequents Geminiani, on affective nature of vi-
and, 26, 50; form and, 57-60; har- brato, 201
mony creates see Harmonic pro- Gestalt; seePattern perception;Shape
gression!; inter-opusnorms create, Gestalt psychology:concepts of, ap-
140-43; knowledgeinfluences, 58 f., plied to music, 85f.; conceptsoper-
78; and learning, 30,43 f., 61, 84; ate in cultural framework, 86; learn-
and meaning, 35-38; melody cre- ing theoriescriticized, 83f.; seealso
ates see Melody!; and memory, Completeness; Continuation, Law
87-90; metric disturbance creates, of Good; Pragnanz, Lawof; Shape;
118 see also Meter; Polymeter!; Texture
minimal differences create, 179-85;
natural vs. learned, 43f.; probabil-
Glarean, Heinrich,a]osquins
De Pro-
fundis describe by, 208
ity and, 54 f.; ranges giverise to, Gluck, Che faro senza Euridice" [Ex.
139; revisionof opinion influences, 43]: affectivity of, 134f.; Hanslicks
49, 52, 60; rhythmic basis for see discussion of, criticized, 282 n.
Rhythm!; saturation creates, 135- 10!; structural gaps in, 134 f.
38; sequences create see Se- Ground; seeFigure and ground; Tex-
quence!; soundterms and,45; and ture
style, 44, 61 see also Style!; sur-
prise and, 29; suspense and, 26-29 Habits: acquisition of, 61-64; child-
see also Suspense!; tendency hood learningof, 61; locus of tradi-
equated with, 23-25; texture cretion and style, 61; and preparatory
ates see Texture!; and the unex- set, 77-79;tendencies equated with,
pected, 29;uniformity of shape cre- 24 f.; see alsoLearning
ates see Uniformity!; unsatisfac- Handel: Concerto Grosso No. 2, iv
tory shapes create, 87,
91 seealso [Ex. 79-81], changes oftexture in,
Shape; Uniformity!; see also Sus- 191 f.; Concerto Grosso No. 4, iv
pense; Uncertainty [Ex. 31], hemiole rhythmin, 117
Hanslick, Eduard: criticism of Gluck
Fashions, beliefinfluenced by,77 specious, 282n. 10!; intellectual
Figure; seePattern perception;Shape satisfaction andcognition confused
Figure and ground: articulation of, by, 275 n. 37!; objections of,to
300 Index
Hanslick, Eduard continued! Iambic rhythm;
see Rhythm
referentialism discussed,
270; posi- Ideal types;
see Class
concepts
tion of, untenable, 3 Image processes:
affective experience
Harich-Schneider, Eta, on deviation and, 256 f.; connotation distin-
in ]apanesemusic, 234f. guished from,258; conscious
vs. un-
Harmonic organization:evolution of, conscious, 256;lead away from
affects meter, 116 f.; incomplete- music, 258; private vs. collective,
ness of,193 f.; uniformity of, 234 f. 257; relation of, to music problem-
Harmonic progression: basis for sense atical, 157f.; see also Association;
of, 179; function of ambiguous, 96, Connotation
175 f., 181 f.; incompleteness of, Incompletencss: harmonic,193; har-
149 f., 145; minimal differentiation monic andrhythmic, interdepend-
weakens, 181f.; parallel fifth pro- ent, 150f., 154; law of return and,
hibition and, 180; rhythmicincom- 153; melodic, 140-43 see also
pleteness and,
149, 151;and styl- Saturation; Structural
gaps!; metric,
istic understanding,179; uniformity 144, 147;rhythmic, 144-49; segre-
and, 168 f., 180 see also Uniform- gation ofstimuli and,
162 f.;spacing
ity!; see also Tonalorganization of texture creates, 189; of texture
Haydn: PianoSonata inA-Flat, _iii without figure,190; typesof, 130;
[Ex. 63-71], analysis ofuniformity sec alsoClosure; Completeness
in, 174-77; Surprise Symphony, Inhibition oftendency; seeTendency
ii, stress not an accent in, 104; Intellectual activity;see Cognition;
Surprise Sympl1O1y,
iii [Ex. 201, Thought process
amphibrach rhythmof, 105; Sur- Introductions and preludes: function
prise Symphony,iii [Ex. 23], higher of, ambiguous, 183; functions of
rhythmic levels in, 110 f.; Sym- different, 26, 53, 60, 116, 180f.;
phony No.102, iv,types ofmeaning minor mode in, 227; norms estab-
interact in,269; Symphony No. 104, lished in,247 f.;rhythm of,ambigu-
iii [Ex. 41], continuity disturbance ous, 239, 247; texture incomplete
in, 130 f. in, 190
Hebb, Donald O.: account of emotions
criticized, 19f.; on Cestalt learning Japanese music: embellishments in,
theory, 84 234 f.; simultaneous deviation in
Herzog, George:on deviationin Afri- [Ex. 92], 234-35; successive devia-
can music, 214; on ornamentation tion in [Ex. 102], 248 f.
in folk music, 212;on stylechanges Javanese music: affective function of
in primitive music, 66 deviation in,204, 210,216; expec-
Hindemith: Mathis der Maler, i [Ex. tation related to affect in, 210;
46-48], importance of inter-opus heterophony in,
235, 243;tendency
norms in, 140-43; objections of, to toward equidistance in, 133
ambiguity criticized,277 n. 8! ]azz: deviationin, 68, 213, 253f.; dif-
Hindu Indian! music: crossing of ferences of, from African music,
rhythm and meter in [Ex. 93], 243; importance of breaks in, 213;
235 f.; expectation in[Ex. 94], 237- rhythm of, 243; similarity of, to
39; ornamentation in, 209, 210; play Hindu music,237; stylechanges in,
element in, 237; probability illus- 68; uncertainty aroused by, 208,
trated by, 55, 56; similarity of, to 243; understandingof, 62
jazz, 237,243; structuralgaps Hlled ]elly Roll Morton; see Morton, ]elly
in, 131; successive deviationin [Ex. Roll
103], 249 f. ]ewish music: deviation in, 211; suc-
Hornbostel, E. M. von, on African cessive deviation in, 251
music, 244 ]ones, Rev. A. M., on African music,
Hypothetical meaning;see Meaning 214, 239 f., 242
Index 301
Knowledge andinformation, influence see alsoBelief, aesthetic;Learning;
of: on belief, 76f.; on expectation, Style!; designative referential!,
58 f., 78; on motor responses, 79- 1-4, 8, 33, 35, 270; designative,at-
81; on perception, 77-79; on sense tacks on,22, 270; designative, and
of completeness,
78 f., 138; onspeed interpretation of unexpected, 29 f.;
of response,
78 f.;
see also
Learningdesignative, modifiedby embodied
Koffka, K.: aesthetic viewpoint 0 , meaning, 269 f.; determinate, 38;
criticized, 91f.; relationshipof satu- embodied, 35; embodied, and abso-
ration to emotion discussed by, 136; lutist position,35; embodied,modi-
see alsoGestalt psychology fied by designative, 269f., 275 n.
35!; evident, 37; evident, colored
Langer, SusanneK.: on past psycho- by hypothetical,38; evident,implies
logical studies,5; tendency toward later events, 37 f.; evident, includes
symbolism describedby, 265 hypothetical, 48f.; and expectation,
Learning: completeness influenced by, 35 f.; expressionist position
as to, 2,
128 f., 131 f., connotation and mood 40; formalistposition asto, 2-4, 40,
influenced by, 2, 259, 262 f., 267, 270; hypothetical,37; hypothetical,
270 f.; designative behavior
a result ambiguity and, 37, 184 f.; hypo-
of, 21 f.; emotional behavior influ- thetical, and evident, different, 48,
enced by,10; expectation and, 43f.; 184; hypothetical,and evident, in
form and, 57 f.; Gestalt theory of, sequences, 172; hypothetical, influ-
criticized, 83 f.; motor responses enced by completeness, 49;hypo-
modified by, 281 n. 29!; pattern thetical, probability and, 37; iso-
perception influenced by, 83-85; lated sounds do not have, 34; locus
perception of texture modified by, of, 33 f.; objectification of, 38-40;
187, 190 f.; preparatory set influ- objectivity of, 34; problem of, in
enced by, 77-79; responseto non- music, 1-4, 32-35; referential see
Wcstern musicdepends on,46; role above underdcsignative!; on vari-
of memory in, 88-90; stylistic un- ous architectonic levels, 38, 47 f.
derstanding dependent on, 46, 61- Melody; see Completeness; Devia-
64, 128 f.; tendenciesmay involve, tion; Incompleteness; Ornamenta-
24; thinking influenced by,43; see tion; Process continuation; Satura-
also Belief, aesthetic; Habits; tion; Structural gaps; Uniformity
Knowledge andinformation; Style Memory: affect influences, 258; ex-
Lipps-Meyer law, as statement of pectation and, 87-91; incomplete-
probability, 55 ness influences, 91; patterns modi-
Listener: attitudes of, influence be- fied in, 90; rehearing music and,
lief, 75; beliefs of see Belief!; 90; search attitudes and, 91; think-
composer asa, 41; responses of,as ing and, 87 f.
evidence, 6-13; role of, 41 f. Mendelssohn: Italian Symphony, iv
Liszt, Piano Sonata [Ex. 54], melodic [Ex. 78], absence offigure in, cre-
uniformity in, 164 ates expectation, 190; Violin Con-
Logic, belief and presumption of, certo, i, amphibrach rhythm in, 106
75 f.; see alsoUnity Meter: accent necessary for, 144; am-
biguity of, 145; architectoniclevels
MacCurdy, T., on affective experi- of, 115f.; crossingof see Polyme-
ence, 14, 256 ters!; discontinuity of, 144; distin-
Mead, GeorgeH.: on communication, guished from rhythm, 102 f.; dis-
40 f.; on meaning, 35 turbed by hemiole, 117-18; dis-
Meaning: absolutistposition on, 1-4, turbed by syncopation, 113, 121
33, 35, 270; affective experience see also Syncopation!; expectation
and, 39f.; defined, 34; dependence and, 118; incompleteness of,144,
of, on stylistic experience,36, 40 147 f.; influenced by rise of har-
302 Index
Meter continued! Sonata K. 332!, i [Ex. 37], metric
mony, 117;nature of,115; problem organization in,123 f.; String Quar-
of, on higher levels,124; pulsenec- tet K. 464!, ii [Ex. 18], rhythm in-
essary for, 103; regularity a norm fluenced bystress, 107;String Quar-
of, 119; relation of, to mood and tet K. 464!, ii [Ex. 24], rhythmic
connotation, 122; relation of, to mo- reversal in, 111f.; String Quartet
tor behavior, 124; without rhythm, K. 575!, i, partial return in, 153;
173; style changes and,116 f.; see Symphony No. 41, ii, metric dis-
also Polymeters; Syncopation turbance in, 144
Middle Ages: cadences of,66; orna- Mursell, ]ames L.: physiological re-
mentation in, 206; use of meter in, sponse tomusic discussedby, 11,
116 74; on rhythmic experienceas pri-
Minimal differences: influence of, on marily mental,81
harmonic progression,178-85; see Musical event; see Soundterm
also Subliminal differences Musical experience: differentiated
Minor mode: affective powerof, ex- from other types, 23;past positions
plained, 222-29; association with as to, 1-6, 32f.; theory of emotions
sadness explained,227 f.; connec- related to, 22-23; see also Affective
tion of, with chromaticism, 224, experience; Connotation; Image
227; past explanations of,unsatis- processes; Meaning; Mood
factory, 5,222 f.;use of,in Western Musical gesture;see Soundterm
music, 223, 227 Musical processes: static vs. dynamic
Modulation: function of, 96, 175; conceptions of, 52-54; viewed as
Saltzer's view of, criticized, 53 norm-deviant relationship,32
Mood: affectiveexperience and, 268 f.;
connotation and, 266; designative Normalization, 89 f.; see also Class
behavior and, 266; distinguished concepts
from affect,7; expression of, in dif- Norms: class concepts as,57 f., 90;
ferent cultures,267; musicaldepic- expectation arousedby inter-opus,
tion of, standardized, 267; speciHed 129, 140-43; and habits, 61; and
by divergence, 268; see also Con- probability, 56;of shape,149; style
notation changes and,64-69, 72f.; types of
Morton, ]elly Roll, on importance of art and, 71; see also Deviants; De-
breaks injazz, 213 viation; Expectation
Motor behavior,79-82: importanceof,
62, 81 f.; influence of experience on, Ornamentation: affective function of,
79 f., 281 n. 29!; law of continua- 206-14; architectonic nature of, 205;
tion and, 82; mental activity and, distinguished fromexpressive devia-
81; modifiedby visualcues, 80;per- tion, 204; in folk and primitive
ception ofmeter and,124; rhythmic music, 211-14;importance of,205,
experience and,81; separateanaly- 211; not a type of exhibitionism,
sis of, not required,82 206; in oriental music, 209-11; play
Motor responses; see Motorbehavior and, 206;style and,287 n. 51!; in
Mozart, Leopold:deviation relatedto Western music, 206-9; see also Ap-
affect by, 200 f.; performers atti- poggiatura; Deviation
tude describedby, 41 Ostinato: possiblyambiguous, 137f.;
Mozart, WolfgangAmadeus: DonGio- saturation notaroused by,136 f.
vanni, Act I [Ex. 85], chromatic se-
quence in,221; PianoConcerto K. Parallel fifths, harmonic basis of pro-
466!, i [Ex. 76], uniformity vs.dif- hibition, 180
ferentiation in, 184; Piano Sonata Pattern perception: general laws of,
K. 331!, i [Ex. 22], influence of 73, 83 see also Closure; Complete-
melody on rhythm in, 109; Piano ness; Pragnanz,Law of; Process
Index 303
continuation; Shape!; knowledge in- Pratt, C. C., referentialism attacked

fluences, 77f., learning influences, by, 271, 276 n. 42!
83-85; selective nature of, 83 Preparatory set, 73-82; see also Belief,
Paulhan, F., on affective experience, aesthetic; Knowledgeand informa-
15 f. tion; Learning; Motor behavior
Performance: deviation and, 199-204 Primitive music:aesthetic meaning of,
see also Deviation!; influence of, problematical, 239; difficulty in
on rhythm, 104, 107,148 f., 172 f. comprehending, 63;dissonance in,
Performer: creative role of, 199 f.; and 230; expressive deviation in, 203;
listener, 41, 80 ornamentation in, 213 f.; ritual lim-
Philosophy of music: absolutistposi- its deviation in, 213; saturation ab-
tion, 1-4; absolutist position,di$- sent in,282 n. 12!; structural gaps
culties in, 4, 33; absolutist position, filled in, 133; style change in, 66;
and embodiedmeaning, 35;expres- see also African music
sionist position, 2-4; expressionist Probability: ambiguityand, 52; devia-
position, distinguishedfrom referen- tion and, 50; on different architec-
tialism, 3; expressionist and
formal- tonic levels, 57; a fundamental con-
ist positionscomplementary, 40; for- dition of style, 45, 54 f., 63; hypo-
malist position,2-4; formalist posi- thetical meaningand, 37;influenced
tion, criticized, 270 f.; formalist by series completeness, 49 f.;
position, difllculties in, 4; referen- learned, 60; minor mode and, 226;
tialist position, 1-4; referentialist sound terms and, 45; and statistics,
position, defended, 2, 33, 270 f.; see 35, 277 n. 15!; and suspensein
also Connotation;Meaning Haydn Sonata,174 f.
Pin tones: affective function of, 216; Process continuation and disturb-
fill in gaps, 132f.; names of, and ance!: alternative modes of, 51,
probability, 56 100, 174 f.; bifurcation of, 126 f.;
Pitch, tempo, and dynamics: diver- distinguished from repetition, 93;
gence of, and connotation,263 f.; disturbance, types
of, 93; harmonic,
influence of, on closure, 139; inter- 95-97, 166-70, 175-77 see also
relationship of,139, 263f. Harmonic progression!; influence of
Play: aestheticexperience viewedas, style on,101 f.;melodic, 93-95;97-
70, 199;and conquest of difficulties, 100, 125, 135, 140-42; metric, 113,
70; references to, in evidence, 284 115-24, 144 see also Meter!;
n. 6!, 206, 214, 237 f., 244, 252 motor behavior and, 82; normative
Polymeters: inAfrican music,239-42; nature of, 92 f.; rhythmic, 102-15
articulation of parts in, 122; effect see alsoRhythm!; saturationa dis-
of, 121, 245 f.; function of, in classi- turbance of, 135 f.; textural, 190-
cal period, 123; in Hindu music, 92 see also Texture!; uniformity
235-39; in jazz, 243; in recent music, of, 164-77 see also Uniformity!;
121 f., 245; in Renaissance, 116, various modes of, congruent,
244; see also Meter 125 f.
Pragnanz, Lawof, 86 f.: implications Program: functionof, 272; leads away
of, 87; influence of, on complete- from music, 272
ness, 128;influence of,on rhythmic Program notes,
influence of,on belief,
organization, 110;influence of, on 76
texture, 186-88; influences under- Psychological theoryof emotions,13-
standing of chromaticism, 217;op- 22: applicationof, to musical expe-
eration of, influenced by belief, rience, 23-25, 31; central thesis of,
91 f.; operation of,in memory,89 f.; 14; conceptsof inhibition and con-
principles ofpattern perceptionand, flict in, 14 f.; objections to, 16; see
91; shapeand, 159f.; weakeningof also Affect; Affective experience;
shape and,162 Behavior. emotional
304 Index
Psychology ofmusic, criticismof past 125; beginning-accented, 107; com-
assumptions of,5f. plete, basis of, 143-45; dactyl,
Pulse: in Beethoven Quartet,147 f.; 103 f.; 106, 108;dactyl, difficult in
importance of, in response to music, duple meter, 106 f.; dactyl, result
242; incomplete rhythm same as, of intensity differences, 106f.; dif-
102; nature of, 102; perception in ferent, in one meter, 104 f.; dis-
terms of basic, 108 tinguished from pulse and meter,
Purcell: Dido's Lament, affective na- 102 f.; effect of different meters on,
ture of chromaticism in, 220 106 f.; effect of durational differ-
ences on,107; effectof intensitydif-
Raguenet, Francois,deviation related ferences on,106-8; effect of prior
to affect,play, andsuspense by,
206, organization on,109, 112f.; effect
208 of stress on, 104, 107 f.; end-ac-
Ravel: RapsoclieEspagnole, ostinato cented, 107;expectation and,145-
ambiguous in, 137 f.; String Quar- 48, 242; form and, 112; freedom of,
tet, i [Ex. 35], polymeter in, 121 f. in rhapsodic preludes, 239, 247;
Recall, distinguished
from recognition, hemiole seeMeter!; higher archi-
90 tectonic levels of, 110-15, 145-49,
Recurrence, distinguished
from reitera- 165; iambic, 103, 105 f., 112, 113-
tion, 151 f.; see also Return 15, 145, 147; iambic, result of dura-
Reiteration: deviations
emphasized by, tional differences,107; incomplete,
152; distinguishedfrom recurrence, 143-49, 165;incomplete, andpulse,
151 f.; saturation and, 152 102; incomplete,on various archi-
Renaissance music: cadences in, 66, tectonic levels,145; influenced by
128; chromaticism and affect in, harmonic progression,149-51; in-
219; deviation and affect in, 208; fluenced byLaw of Priignanz, 110;
harmony influencesmeter in, 117; influenced b melody, 109; motor
minor mode developed in, 227 f.; behavior81;
and!:
nature
of, 104-8;
ornamentation in, 206; polymeters performers conceptionalters, 104,
in, 244 107, 110,148; pivot alters, 111;re-
Repetition: continuation not, 93; effect building of, in Beethoven Quartet,
of, modifiedby context,49, 153;re- 148; reversal of, 110, 113 f., 125,
currence and reiteration types of, 147; trochaic, 105, 106, 109, 112 f.,
151 f.; see also Return 145, 173;trochaic, difficultin triple
Return: deviations minimized in, 152; meter, 106 f.; trochaic, result of in-
expectation and, 152; incomplete- tensity differences,106; uniformity
ness emphasizes,
153; partial,152 f.; of, 165, 168 f., 173, 181 see also
principle of, 151-56; on various Uniformity!; see also Accent;
architectonic levels, 151; see also Meter; Pulse; Stress
Recurrence; Repetition Riemann, Hugo:explanation ofminor
Reversal: defined,93; degreesof, 93; mode, unsatisfactory, 222 f.; on
melodic, 97; rhythmic, 110, 112, understanding music,61 f.
113, 125; from uniform to formed Roberts, Helen H., on deviation in folk
process, 171, 178, 182; see also and primitive music, 214, 251-53
Process continuation Romantic period:cadence changes in,
Rhythm: ambiguous, 110, 147 f., 66; uniformity mitigated in, 174
172 f., 180; amphibrach, 103 f., Russell, Bertrand,on understanding,
106, 110, 125; amphibrach, accent 39, 61
stressed in, 108 f.; anapest, 103f.;
106, 107, 125, 165; anapest, on Sachs, Curt: on East Asiatic music,
higher architectonic levels, 110f.; 209 f., 234; on position of new
anapest, resultof durational differ- tones, 50;on rhythm as form, 112;
ences, 107; anticipation changes, on rhythmictension andrepose, 242
Index 305
Salzer, Felix, views on modulation 160, 184f., 178; unsatisfactory, 87,

criticized, 53 161; weak,and ambiguity see Am-
Sargeant, Winthrop,on uncertaintyin biguity!; weakened by segregation,
jazz, 243 158; weakenedby uniformity, 163-
Sargeant, W., and Lahari: Hindu 78 seealso Sequence; Uniformity!;
music compared to jazz by, 237; on weakening of,and affectiveexperi-
play element in Hindu music, 237 ence, 161,163 see also Suspense;
Saturation, 135-38; cultural attitudes Uncertainty!; weakening of, and
modify, 282 n. 12!; depends onbe- texture, 186 f., 194 f. see also Tex-
lief, 135; emotional nature of, 136; ture!
examples of, 135, 143, 182; expecta- Simultaneous deviation; see Deviation
tion and, 135 f.;influenced bycon- Sound, elements
of; see Pitch, tempo,
text, 136, 143; relation of, to com- and dynamics
pleteness, 135;understanding and, Sound stimulus: distinguished from
136 f. sound term,45 f.; meaning of, de-
Schenker, Heinrich:disciples of,criti- pendent oncontext, 45-47
cized, 52 f.; method of, and orna- Sound term: and completeness,129;
mentation, 205 defined, 45 f.; on different archi-
Schubert: StringQuartet in A-Minor, tectonic levels, 47; distinguished
minor mode in, 228; Symphony in from sound stimulus, 45; meaning
C, ii [Ex. 75], uniformity vs. dif- of, dependent on context, 46; and
ferentiation in, 183 f. probability, 46
Schumann, Robert,Humoreske, Op. Statistics, use of, difficult in style
20 [Ex. 57], uniformity in, 167 studies, 55-56
Seashore, Carl: deviation related to Strangways, A. H. Fox: embellishment
affect by,201 f.;views of,discussed, discussed by, 210 f.; on Hindu
202 f. drumming, 235-37; probability
Segregation ofstimuli, shapeaffected shown in rciga given by, 55; on
by, 158, 162 f. tonality, 46,214
Sequence: alternative continuations of, Strauss, Richard:Also Sprach Zara-
174-77; context modifies under- tlzustra, ambiguity of texture in,
standing of,171; differenttypes of, 193; Ein Heldenleben, cadence
168 f.;effect ofirregularities in,177; from, discussed, 66
expectation and, 170; mentalcontrol Stravinsky: onconquest ofdifficulties,
weakened by, 170 f.; reversal of, 69; formalistposition of, 3; Histoire
171, 178; uncertainty arousedby, du Soldat, March [Ex. 32-34],
170-71; uniformity and, 168f.; see metric organizationof, 119-21; on
also Uniformity motor behavior, 80
Shape: appraisalof, 158-60; expecta- Stress: onaccent altersgrouping, 104,
tion and, 161 see also Expecta- 107-9; distinguished from accent,
tion!; Gestalt nature of, 283 n. 1!; 104; onweak beatdoes notchange
good, andmental security,161; inrhythm, 104
fluence of memory on, 89-91; in- Structural gaps, 130-35; built into
telligibility of good, 161; melodic tonal systems, 132; distinguished
see Melody!; nature of, 157-60; from continuity disturbance, 130;
perception of,and experience,161, filling of, delayed, 135,142; filling
184 f. see also Learning; Pattern of, in different cultures, 44, 131-33;
perception!; perception of, and and tendencytoward equidistance,
Law of Pragnanz see Priignanz, 132-34 see also Equidistance,
Law of!; produced by similarity tendency toward!;in texture,189
and difference,157 f., 161 f.; rhyth- Style: changes, influence ofcomposers
mic see Rhythm!; stability of on, 69-71;changes, influence of de-
good, 87; style and perception of, viation on, 65-69; changes, influ-
306 Index
Style continued! response, 24 f.; inhibition of, basis
ence of extramusical events on, 65, of affect, 14, 31 see alsoAffect!;
67, 72f.; changes,intramusical char- inhibition of, and cognition,31; in-
acter of,65 f.; changes, patternsof, hibition of, and conflict, 15; inhibi-
64-69; changes in, cyclic, 72 f.; tion of, a deviation, 32 see also
class concepts and, 57f.; continuity Deviation!; inhibition of, and mean-
influenced by,101; expectationde- ing, 39; inhibition of, and suspense,
pends upon, 30; and form, 57-59 27; see also Expectation; Psycho-
see also Class concepts;Context; logical theory of emotions
Form!; formal structure of, 45-60, Texture, 185-95;ambi uity of, 192f.,
62 f.; formal structure of, leads to 195; changes
of, inflguence
expecta-
changes, 65-69;learning and, 60- tion, 189; completeness and,188;
64 see also Learning!; meaning continuity of, expected in fugue,
and, 34f., 42, 46; misunderstanding 191; effectof changesof, 189,190 f.,
of unfamiliar, 63; and motor be- 192; expectation created by, 185,
havior, 79f.; plurality of, 60, 62- 188, 190; as independentvariable,
64; probability and, 45, 54, 60; 188 f.; Law of Pragnanz influences
shape and,159 f., 184, 187f.; sta- perception of, 186 f.; meter influ-
tistical studies of, 55 f.; structural enced by, 122; organizationof, in-
gaps and,44, 132;syntactical nature fluenced b experience, 187,190 f.;
common to, 62 f. spacing189
og:193;
f., types
of,
Subliminal differences, 158 f., 165 f. 186; weakeningof, 193-95; see also
Successive comparison, principle of, Figure and ground
152; see also Deviation, successive Thought processes:described, 88;il-
Successive deviation; see Deviation lustrated in Chopin Prelude, 97;
Surprise: distinguished from unex- likened to affective and aesthetic
pected, 29; expectation and, 29; processes, 88;see alsoCognition
understanding of,29 f. Tonal organization: architectonic
Suspense: affectand, 15 f., 27; am- levels and, 214 f.; auxiliary tones
biguity and, 16, 26, 51; anxiety, important in, 215 f.; changes in,
ignorance, and uncertainty create, 66 f.; development of,132-34; de-
27-29, 161, 163 f., 171 f. see also viation and, 214-32; and expecta-
Uncertainty!; chromaticismand, in tion, 138, 214; and law of return,
opera, 220f.; expectation created 151; learned, 46, 60-64, 85; and
by, 28; expectation notspecific in, probability, 54-56; and sha e,
27; a kind of tendency inhibition, 163178
f., structural
f.; inaps
cgi-
27; minimal differences and, 182; ferent,f.;131
syntacticaf
nature of,
musical and extramusical, 28 f.; res- 45-51; seealso Style
olution of, and style, 29; rhythmic Tonality; seeTonal organization
incompleteness creates, 149; texture Tracey, Hugh,simultaneous deviation
creates, 193-95; uniformity and, described by, 244
163-78; seealso Ambiguity;Uncer- Trochaic rhythm;see Rhythm
tainty
Syncopation: changes
of, 120f.; effect Uncertainty anxiety!: and affective
of, in contemporary music, 120; experience, 15-16, 27-29, 96, 161,
melodic-rhythmic changes create, 163 f., 194 f., 208, 237, 242-43;
113 f.; and process continuation, ambiguity and, 51 see also Am-
121; rhythmic crossing not,239 f.; bi uity!; balance ofuniformity and
see also Meter; Rhythm digerentiation 183;
produces,
belief
and, 26 f.; chromaticism creates,
Tendency: consciousor unconscious, 164, 220;loss of control andignor-
24; defined, 24; expectations ance arouse, 27 f. see also Control,
equated with, 23-25, 30; a habit mental!; minimal differences create,
Index 307
181 f.; mitigated by good shape, 194 f.;total, effecton shape,158 f.;
167, 174; relativity of, 96; stylistic uncertainty arousedby see Sus-
experience influences, 174; and pense; Uncertainty!; whole-tone
tendency, 16; uniformity creates, scale atype of, 164
163 f., 171; weakenin of texture Unity musical!, basis of, 53 f., 161,
creates, 193-95;
see also
Suspense
196
Understanding, natureof, in music,
39, 61; also seeMeaning Vibrato: affective function of, 201;
Unexpected: and designative meaning, changing useof, 66
30; distinguishedfrom surprise,29;
evaluation of, 29; see also Devia- Wagner: Die Meistersinger, under-
tion; Expectation standing ofBeclcmessers music, 30;
Uniformity, 163-78: ambiguity and, music of,norms presupposed by, 56;
166 f.; architectonic character of, system of connotations developed
165, 168; balance of, with differen- by, 260;Tristan undIsolde, ActIII,
tiation creates uncertainty, 183f.; bifurcation of process in,100, 113,
basis of, 163; chromaticisma type 126; Tristan und Isolde, Act III
of, 164;effect of,influenced bycon- [Ex. 9-14], melodic structure of,
text, 171; effect of irregularity 97-101 ; Tristan und Isolde, Act III
within, 174 f., 176; elements cre- [Ex. 25-29], rhythmic-metric struc-
ating, complementary, 164 f.; ex- ture of, 112-15
pectation and, 163, 170-71, 174, Waterman, Richard Alan, on African
178, 182, 194-95; harmonic, 166- music, 242
83; harmonic, minimal differentia- Watt, H. ]., on structural gapsin dif-
tion creates, 178-82; harmonic, miti- ferent cultures, 131
gated bymelodic-rhythmic articula- Weld, H. P., mood response distin-
tion, 166-68; harmonic, other modes guished fromaffect by, 7
of uniformity intensify, 167f., 169, Wetheimer, Max,on thinking process,
180; harmonic, types of, 166-68; 87 f.
melodic, 163-65, 167 f., 169, 173 f.;
rhythmic, 164f., 168, 172 f.; sense Yasser, ]oseph,
theories of, and tend-
of control weakened by,161, 163, ency towardequidistance, 133
171, 176; sequence atype of, 168
see also Sequence!; suspension Zarlino, function of dissonance de-
technique atype of, 170; of texture, scribed by,229, 232

Emotion and Meaning in Music

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Emotion and Meaning in Music

Uploaded by

Copyright:

Available Formats

Emotion and

1956 by The University of Chicago. All rights reserved.

ISBN: 0-226-52139-7 paperbound!

His life was gentle,and theelements

The diversity and complexity of twentieth-centurymodes of

1. I. A. Richards,Principles ofLiterary Criticism New York: Har-

The arguments and debatesof aestheticians,

II. EXPECTATION AND LEARNING _

III. PRINCIPLES OF PATTERN PERCEPTION: THE LAW OF Coon

IV. PRINCIPLES OF PATTERN PERCEPTION: COMPLETION AND

V. PRINCIPLES OF PATTERN PERCEPTION: THE WEAIIENING OF

VI. THE EVIDENCE: DEVIATION IN PERFORMANCE AND TONAL

VII. THE EVIDENCE: SIMULTANEOUS ANDSUCCESSIVE DEVIATION

VIII. NOTE ON IMAGE PROCESSES,CONNOTATIONS, ANDMOODS .

Past Positionsas to the Nature of

THE PRESENT POSITION AND CRITICISM OF

Evidence as to the Nature and Existence of

OBJECTIVE EVIDENCE: BEHAVIOR

OBJECTIVE EVIDENCE! PHY SIOLOCICAL RESPONSES

The PsychologicalTheory of Emotions

THE DIFFERENTIATION OF AFFECT

The Theory of Emotions Related to

TENDENCY AND EXPECTATION IN MUSIC

CONSCIOUS AND UNCONSCIOUS EXPECTATIONS

The Meaningof Music

MUSIC AND MEANING

TI-IE OBIECTIFICATION OF MEANING

MEANING AND AFFECT

MUSIC AND COMMUNICATION

In the preceding chapter

A general distinction must be drawn at the outset betweenthose

Style: Formal Considerations

SOUND TERMS AND SOUND STIMULI

with imitationsbetween the soprano andalto isbegun. The sequence

STATIC VERSUS DYNAMIC CONCEPT IONS OF

relationships withinthe tonalmaterials available. He is telling us,

FORM, PROBABILITY, AND EXPECFATION

Style and Social Process

LEARNING AND STYLE

THE PLURALITY OF STYLES

In general it seemswise andprudent to treat all aspects ofa

PATTERNS OF STYLE CHANGE

Tonal growthhas progressed to such a degreeof saturationthat the

Hot jazz melody isimprovisatory, butits structureis held to a coherent

In the case offolk music,including jazz,the basic,normative pat-

STYLE CHANGES AND THE COMPOSER

CYCLIC CHANGE AND STYLE

The probabilitiesof style and form,the normsupon which ex-

The Preparatory Set

BELIEF AND THE PRESUMPTION OF LOGIC

Learned Habitsand the Preparatory Set

THE INFLUENCE OF KNOWLEDGE AND EXPERIENCE

Motor Attitudes and Motor Responses

Principles ofPattern Perception:

LEARNING AND PERCEPTION

immediately forma convincingand satisfactory Gestalt, whichhas

DIFFICULTIES IN THE APPLICATION OF GESTALT CONCEPTS

Basic Concepts and Formulations

THINKING, MEMORY, AND EXPECTATION

All such changes takeplace in the rememberingof music. Since,

The Principlesof Pattern Perception

THE LAW OF GOOD CONTINUATION

by the fact thatthe sametone beginsthe second fragment asended

strong forcetoward continuation.

:ax-_-'=~" "i" `` r ""' ""` "" "'' "'