metaphor and music

C H A P T E R 28

Metaphor and Music

Lawrence M. Zbikowski

The music in Example 1 is a shorthand ver-sion of the string parts and solo bass melodyfrom the opening of the fourth movement ofJohann Sebastian Bach’s 1714 cantata for thefirst Sunday of Advent, “Nun komm der Hei-den Heiland.” The music in the string partsfor the passage might be described in one oftwo contrasting ways. The first is more col-orful and more accessible: “The bass obsti-nately plods along throughout the passage;the chords above are either sour and bit-ing dissonances (as in the first half of mea-sure 1, and all of measures 2–3) or sweet butdark consonances (as in the second half ofmeasure 1, or at the end of the excerpt).The mostly murky sounds of the pizzicatostrings, together with the slow tempo, makethis a brooding, melancholy piece.” The sec-ond description is drier, and makes more useof technical jargon: “Above an ostinato tonicbass Bach sounds first a dominant-seventhchord (in the opening portion of measure 1),and then a leading-tone chord with seventh(in measures 2–3). He provides momentaryrelease for the tension created by these dis-sonances through the introduction of thetonic chord in the second half of measure

1 and then again at the conclusion of thepassage.”

In analyses of how language is used tocharacterize music, the first description istypically characterized as metaphorical, thesecond as literal. The metaphors in the firstdescription are readily apparent:

� the “plodding bass” is nothing morethan a repeated note plucked by thecellos

� the consonant harmonies of measures 1

and 4 and dissonant harmonies of mea-sures 1–3 are sounds, and so cannot tastelike anything, sweet or sour

� the dissonances in measures 1–3 are sim-ply a consequence of F� and D� soundingagainst the E in the bass – no mechanismsfor biting are in evidence

� as products of the resonance of a soundingmedium the E minor chords of measures1 and 4 have no particular reflectance andso can be neither dark nor light

� even if we grant that what is meant by thecharacterization of the piece as “broodingand melancholy” is that it is expressive ofthese feelings, how can a simple sequence

502

METAPHOR AND MUSIC 503

Example 1. Measures 1–4 of the fourth movement (Recitativo) from J. S. Bach’s cantata “Nun kommder Heiden Heiland” (BWV 61).

of sounds – which is not in any way sen-tient – express anything?

It would appear that none of the thingspicked out by the first description couldin fact be found in the music – thus themetaphoricality of the description.

The second description does seem tobe more literal. The definition of an osti-nato is, indeed, a repeated pattern of notes.The pitches B-D�-F�-A are those of thedominant-seventh chord of E minor, andD�-F�-A-C are those of the leading-tonechord with seventh. With mention of ten-sion and release in the second sentence weare, however, on somewhat shakier ground,for the dissonant chords in the passagerequire no more or less tension for theirproduction than do the consonant chords.One might argue that such chords give allknowledgeable listeners the sensation of ten-sion and release, but this begs the question:the description is supposedly about a givenmusical passage, not a listener’s reaction tothe passage. One solution would be to elim-inate the second sentence from the descrip-tion, but we would then be left with littlemore than a narrativized rendering of themusical terms that could be applied to thispassage. It is the second sentence, with itsmetaphorical evocation of tension andrelease, that gives some indication of howthe music sounds, rather than what musicalelements it comprises. Metaphor seems tobe an inescapable part of musical descrip-tions that aspire to more than a rehearsal ofdefined terms.

There are, of course, many areas ofhuman experience about which discourse

is resolutely metaphorical, emotions beinga prominent example. What makes musicspecial is its relationship to language. Bothmusic and language, for instance, are uniqueto the human species, both unfold over time,both have syntactic properties, and bothmake use of sound. Indeed, the notion thatmusic is a language is the basis for someof the most prevalent metaphors used todescribe music. But music is also not likelanguage in at least one important respect:aside from a limited number of exceptionalcases when music mimics natural sounds,music makes no reference to the outsideworld. Music does make reference to – orperhaps embody – the interior world of emo-tions or physiological states, but it is just thisworld that typically escapes the grasp of non-metaphorical language.

Given this situation, one could conceiv-ably trace connections between the phe-nomenon of metaphor and the culturalpractice of music back to the earliest writtenrecords. (There is, for instance, a brief dis-cussion of metaphors used by Aristoxenus,a fourth-century BCE writer on music, inthe introduction to Zbikowski, 2002 .) Thefocus in this chapter, however, is on workthat has contributed directly to discussionsabout metaphor and music and on theo-retical frameworks for understanding howthe domain of music correlates with otherconceptual domains, including that of lan-guage. Most of this work dates from thepast 50 years, and encompasses a range ofdisciplines, including philosophy, semiotics,cognitive science, and the critical and ana-lytical study of music. It should be notedthat scholars of music are often divided

504 LAWRENCE M. ZBIKOWSKI

into three subdisciplines, reflecting differentintellectual commitments. Although musi-cology is the most comprehensive term, itis currently used to refer to those whoseresearch concentrates on music viewed as ahistorical practice; ethnomusicology, by con-trast, tends to focus on the social and cul-tural contexts of musical practice, with anemphasis on non-western music; music the-ory is concerned, by and large, with develop-ing systematic perspectives on musical orga-nization and on close readings of individualmusical works based on these perspectives.

The first part of this chapter is givenover to a historical and conceptual surveyof music and metaphor (moving through arange of disciplines, including the three sub-disciplines of music scholarship), organizedaround some of the topics adumbrated in thediscussion of my opening example. Theseinclude the status of knowledge about music,the nature of musical semiotics, the rela-tionship of music to other aspects of humanexperience (and in particular the expressionof emotions), and music as a manifestationof human cognitive capacities. The secondpart of the chapter will return to the musicof example 1 and explore a theoretical frame-work for analyzing how the domain of musiccorrelates with other conceptual domains,including that of language.

Research on Metaphor and Music

Music and Knowledge

Perhaps the first extended discussion thatconnected music with metaphor appearedin the philosopher Nelson Goodman’s Lan-guages of Art (1968/1976). Goodman wasinterested in developing a theory of symbolsthat could apply to works of art as well as tonatural language. According to this theory,a painting is a symbol (if of a rather specialsort); so are a sequence of musical sounds ora sculpture. One distinctive feature of artis-tic symbols is that they are typically regardedas expressive: a gray-toned painting with asomber theme is thus described as “sad,” aswould be a lugubrious melody in a minorkey. For Goodman, such sadness is not an

attribute of the symbol proper but is insteadfigurative or metaphorical: in describing thepainting or the melody as “sad,” we transfera system of concepts from its typical realm(the emotional states associated with sen-tient beings) into a new realm (colors andshapes on a canvas or a sequence of sonicevents; Goodman, 1976, 72). The expressiv-ity of an art work is, in consequence, not anattribute of the work as such but is simplyattributed to the work.

Goodman’s account of the expressivity ofworks of art – and in particular, music –was met with two sorts of challenges. Thefirst, and most straightforward, came fromphilosophers who argued that the expressivecharacter of a work is basic to it: expressiv-ity is an ineliminable property of the musicalwork. When Goodman relegated the sadnessof a melody to the domain of the metaphor-ical, he simply missed the point, since thepurpose of the melody was to be expressiveof some emotion (Budd, 1989; Davies, 1994 ,150–166).

The second challenge to Goodman camefrom the philosopher Roger Scruton, whonoted that Goodman’s approach made noplace for human understanding – indeed, forGoodman artistic symbols and the expres-sive values attributed to them are com-pletely independent of human cognition(Scruton, 1974 , 222). Scruton’s aesthetictheory, as a whole, sought to place worksof art in the intentional realm; in subse-quent work on music this strategy led Scru-ton to argue that to hear various sounds asmusic (as opposed to unconnected if pleas-ant noises) requires construing such soundsin terms of concepts taken from some otherdomain. To take a simple example, whenthe bass sings the first three notes of mea-sure three in Example 1 – C4-A3 -F�

3 –we typically describe them as descending.1

This descent is, however, an illusion: notonly does the singer remain where he is,but there is nothing in a scientific accountof the sounds themselves that supports thenotion that they descend. From Scruton’sperspective, this “illusion” is key to under-standing the sequence of notes as music: the“motion” that we ascribe to the sequence


of notes sung by the bass is a consequenceof our framing their succession in terms ofthe motion of physical objects through spacefrom low to high. This sort of metaphor-ical transfer – taking concepts from onedomain (such as that of movement, or space,or concrete objects) and applying them toanother – is essential to hearing sounds asmusic. “If we take away the metaphors ofmovement, of space, of chords as objects,of melodies as advancing and retreating, asmoving up and down – if we take thosemetaphors away, nothing of music remains,but only sound” (Scruton, 1983 , 106).

Scruton took pains here and in later workto emphasize the disjunction between theproperties of sounds and the properties ofmusic, for this disjunction pointed directlyto the intentionality of art works like music:

There lies, in our most basic apprehensionof music, a complex system of metaphor,which is the true description of no materialfact, not even a fact about sounds, judgedas secondary objects. The metaphor cannotbe eliminated from the description of music,because it defines the intentional object ofthe musical experience. Take the metaphoraway, and you cease to describe the expe-rience of music. (Scruton, 1997, 92 )

For those who placed little trust inmetaphor as a tool for discovering the essen-tial properties of music, however, Scru-ton’s account of musical understanding wasjust as flawed as Goodman’s, if in a dif-ferent way. Where Goodman isolated thefact of music from its expressivity, Scrutonfailed to explain how metaphorical state-ments connected with musical facts (Budd,1985 , 2003). Raising a similar objection,the music theorist Naomi Cumming notedthat the sharp distinction between literaland figurative language that Scruton drewwas ultimately untenable when the range oflanguage used to describe music was consid-ered (Cumming, 1994 , 2000, 49–51).

This difficulty points to a problem com-mon in philosophical writings about music,which tend to treat language as the goldstandard for conceptualization and gram-mar (see, for instance, Dempster, 1998).

Against such a standard music comes offrather poorly, for it cannot supply the fac-tuality that is believed to mark language. Asomewhat different, albeit related, problemstems from the special status granted instru-mental music by some nineteenth-centurythinkers. As Lydia Goehr has observed,within German Romanticism “‘The purelymusical’ . . . served as a general metaphorfor all that was unknowable by ordinarycognitive or rational means” (Goehr, 1998,18). For writers who adopt this perspec-tive the impenetrability of music is its rai-son d’etre: “the musical mystery is not ‘whatcannot be spoken of,’ the untellable, butthe ineffable” (Jankelevitch, 2003 , 72 ; seealso Charles, 1995). The metaphors used todescribe music – especially to the extent thatthe mechanisms behind these metaphorsremain unexamined – are thus symptomaticof music’s ineffability. A final complicationis that accounts of metaphor grounded in thephilosophy of language may simply be inad-equate for music, as can be seen in StevenKrantz’s application of Max Black’s theoryof metaphor to music (Krantz, 1987), and asis demonstrated in Leo Treitler’s critique ofGoodman (Treitler, 1997).

Musical Semiotics

The somewhat uncomfortable relationshipbetween language and music evident inphilosophical treatments of music andmetaphor is also apparent in efforts to adaptsemiotic theory to music. Hints that suchan adaptation might be possible can be seenas early as Ferdinand de Saussure’s Coursein General Linguistics, where, summing upthe possibility of separating out the ele-ments of language for analysis, Saussurecomments, “Similarly, a musical series do, re,mi can be treated only as a concrete seriesin time, but if I select one of its irreducibleelements, I can study it in the abstract”(Saussure, 1959, 40). A half-century afterSaussure the Belgian linguist Nicolas Ruwet,in what proved to be an influential essay,adopted a similar perspective for detailedanalyses of four melodies from the middleages (Ruwet, 1966; reprinted in Ruwet, 1972 ;


Example 2 . Two descending melodies from Cooke 1959 (from Cooke’s Ex. 58b, p. 134).

for a discussion see Powers, 1980, 10–22).Limitations of the approach, however, soonbecame evident, especially where meaningwas concerned. As the music theorist KofiAgawu has observed, while it is the case thatthe basic units of language have a more orless fixed lexical meaning, the basic units ofmusic most typically do not (Agawu, 1999,144). Related to this, the symbolic structureof language consists of a dense network ofmutually interrelated symbols which typi-cally share little if anything with the things towhich they refer (Deacon, 1997, chapter 3 ;2003). The symbolic structure of music hasnothing like this level of complexity: therelationships into which symbols enter aretypically more local, and there tends notto be the sort of abstract reference typi-cal of linguistic symbols (see, however, theanalyses in Agawu, 1991). Where the per-spectives of semiotic theory have reapedthe most benefit has not been in show-ing how music replicates the features oflanguage but through explorations of howmeaning specific to music – and in somecases beyond the capacities of language – ispossible.

Deryck Cooke, in The Language of Music,proposed that certain types of musical mate-rials (with a special focus on the intervalsthat occur between the notes of a melody)were expressive of certain types of emotions.For instance, Cooke proposed that a descentfrom the fifth note of a minor scale throughthe first, of the sort shown in the melodiesof Example 2 , expresses “acceptance of,or yielding to grief, discouragement anddepression; passive suffering, and the despairconnected with death” (Cooke, 1959, 133).Cooke’s account of the vocabulary of musicis much more complex than suggested bythis example (for instance, the descent fromthe fifth through the first note of the scalecould be filled in with the fourth and second

notes of the scale, or embellished in variousways) and much more detailed. Indeed, thevery specificity Cooke offered may have toldagainst him, for this provided fuel for crit-ics who argued that musical meaning wasmuch more various than Cooke seemed tomaintain. Nonetheless, Cooke’s basic idea –that musical meaning is tied up with theexpression of emotion – is not only broadlyaccepted among musicians (as suggested bythe commentary in Agawu, 1999, and theessays in Juslin and Sloboda, 2001) but hasrecently been the focus of further work. JanBroeckx, for instance, has argued that musi-cal meaning is a consequence of the directrepresentation of emotion through musicalfigures (Broeckx, 1997). While we can cer-tainly describe these emotions through lan-guage (thus giving rise to the metaphori-cal descriptions of music’s expressivity) suchdescriptions do not create the meaning thatthe emotions have. Broeckx, however, doesnot develop his methodology further, andjust how musical figures express emotionsremains obscure. Hallgjerd Aksnes, for herpart, has made use of current work in cogni-tive science and metaphor theory to bringclarity to this perspective, proposing thatthe emotions summoned by passages in themusic of the Norwegian composer GeirrTveitt can be grounded in embodied expe-rience (Aksnes, 2002 , chapter 8). Addi-tional methodological support can be foundin the composer and semiologist DavidLidov’s work, recently brought togetherin Lidov (2005). Although metaphor the-ory does not figure large in Lidov’s theoryof musical signification, one can find theintegration of an approach sympathetic toCooke with a general theory of semiotics.The result is a theory of musical mean-ing based on correlations between emotions,physical gestures, and sequences of musicalevents.


Example 3 . The melody of Schumann’s “Traumerei” combined with the melody ofDvorak’s “Humoresque” (adapted from Example 4 from Karbusicky 1987, 436).

Another perspective on musical mean-ing and its relationship to metaphor wasprovided by the musicologist VladimirKarbusicky. Working from a thoroughknowledge of the history of semioticapproaches developed by German musi-cologists under the influence of ErnstCassirer, Karbusicky made a strenuous argu-ment against using semiotic theories for-mulated for language to explain musicalmeaning: “The popular definition of music asa kind of language or as auditive communica-tion, which has all too often been taken forgranted even in scientific essays, is nothingmore than a metaphor” (Karbusicky, 1987,431). For Karbusicky, “thought in musicoccurs primarily in asemantical shapes andformulas” (433); any attempt to interpretthese shapes and formulas through languageor linguistic theory would ultimately fail tocapture the substance of musical thought.Karbusicky was, however, willing to enter-tain the notion that there might be purelymusical metaphors (as distinct from linguis-tic metaphors used to describe music). Theexample Karbusicky chose to illustrate thisidea was inspired by a cabaret pianist whoput together the melody of Robert Schu-mann’s “Traumerei” (from Kinderszenenop. 15 , no. 7) with the melody of AntoninDvorak’s “Humoresque” in G-flat major (op.101 no. 7; here transposed to F major);see Example 3 . (The title “humoresque”makes reference to a term used for lit-erary sketches by German writers during

the early nineteenth century. Applied tomusical works, it often indicates short occa-sional pieces with a relaxed – but notnecessarily humorous – character.) Theopening sections of both Schumann’s“Traumerei” and Dvorak’s “Humoresque”are eight measures long, and for the mostpart Karbusicky’s example moves back andforth between the two pieces: measure 2

of Example 3 is measure 2 of the “Hu-moresque”; measures 3–4 of Example 3 aremeasures 3–4 of “Traumerei.” The excep-tion occurs in the second phrase (measures5–9): while measure 5 replicates measure 5

of the “Humoresque” and measure 6 repli-cates measure 6 from “Traumerei,” measure7 and the first half of measure 8 are drawnfrom measures 6–7 of the “Humoresque.”This change results in an added measure,with the latter half of measure 8 and allof measure 9 of Example 3 drawn frommeasures 7–8 of “Traumerei.” This minormodification notwithstanding, the free inter-changeability of musical materials evidentin Example 3 is important for the pointKarbusicky wishes to make, for it indicatesbasic structural similarities between the twomelodies that supports their meaningfulcombination. A closer look at the music ofExample 3 suggests that the materials of“Traumerei” – and the image of childhooddreaming that they are meant to evoke –control the musical discourse. These mate-rials frame the beginning and ending of thefirst phrase (measures 1–4), and, in doing so,


help to define its tonal structure. Althoughthe “Humoresque” melody attempts to takecontrol in the second phrase – shoving asidethe reprise of the opening of “Traumerei” inmeasure 5 , and running on for nearly twomeasures in measures 7–8 – the melody of“Traumerei” ultimately wins the day to con-clude the second phrase.

In his analysis Karbusicky proposed thatboth melodies carry a basic semantic chargethat might be described as “nostalgia, senti-ment” (keeping in mind that each melodyprojects this charge in a different way).Musical metaphor, as Karbusicky conceivedit, comes about because the meaning of“Traumerei” is changed when the unfold-ing of its languid melody is interrupted bythe sprightly gestures of the “Humoresque.”The resulting modification of the senti-ment of “Traumerei” – pushing it towardcheerfulness – is a consequence of boththe introduction of the contrasting seman-tic content of the “Humoresque” and thecommon structural features of the twomelodies, a commonality that supports asso-ciating the sentiment of “Traumerei” withthat of the “Humoresque” (Karbusicky,1987, 436–437).

The basic idea behind Karbusicky’s no-tion of purely musical metaphor, in whichdisparate musical materials are broughttogether to generate new meaning, canalso be seen in Robert Hatten’s work onmusical meaning (Hatten, 1994 , chapter 7;1995). Hatten, for his part, proposes thatthe correlations between musical materi-als and meaning must be established priorto their being brought together to createnew meaning. This process occurs not inthe manner of Karbusicky’s rather excep-tional example (which relies on structuralsimilarities between the two melodies tosupport their combination) but instead takesadvantage of what Hatten calls functionallocations, which can be thought of as impor-tant structural moments within a musicalwork (such as the reprise of a significanttheme). A functional location may bea consequence of syntactic expectationsset up within a particular piece, or mayreflect stylistic formal schemas common

to any number of pieces. When musicalmaterials with markedly different meaningsare subjected to the syntactic pressures thatcharacterize such locations, new meaningemerges. Hatten’s general term for this pro-cess is musical troping – metaphor is just onetype of musical trope that may result. (Fora similar perspective, but framed relative tothe work of Roman Jakobson and JacquesDerrida, see Ayrey, 1994 .) As Hatten read-ily admits, his approach bears more in com-mon with poetic than with linguistic theory(Hatten, 2004 , 297, n. 1); such a perspec-tive is well suited to the interpretive chal-lenges presented by the music of Mozart,Beethoven, and Schubert on which Hattenhas focused.

Relationships between similar but struc-turally (or conceptually) distinct musicalentities have long been recognized by musi-cians, although such relationships are of asort closer to the pragmatic ones that under-lie Karbusicky’s analysis that they are tothe poetic ones invoked by Hatten. Indeed,one could argue that teaching students howto identify and exploit such relationshipsis one of the cornerstones of music ped-agogy. It is perhaps for this very reasonthat the framework provided by theoriesof metaphor or analogy has not generallybeen used to characterize such relationships.When such frameworks have been applied torelationships among musical materials, it hasbeen as part of a more comprehensive studyof correlations between music and othermedia (Kielian-Gilbert, 1990) or to char-acterize how idealized musical constructsrelate to actual musical practice (Dubiel,1990, 327; Perlman, 2004 , chaps. 6, 8).Again, the topic of metaphor more typi-cally crops up when the issue is how thingsthat are musical relate to things that are notmusical.

Music and Other Aspectsof Human Experience

The issue of music’s connection with otheraspects of human experience emerged withforce in music scholarship in the periodafter World War II when, as a consequence


of developments in music compositionbegun a generation before and in keepingwith the climate of aggressive positivism thatinformed a broad range of humanistic stud-ies, there arose the idea that the analysisof music could proceed along the lines ofscientific inquiry (see, for instance, Babbitt,1972/1961). The analyses produced wouldfocus solely on matters of musical struc-ture, on the assumption that a comprehen-sive account of this structure would explaineverything of importance about music. Mat-ters such as what music expressed wouldeither be answered by such an accountor regarded as beyond analysis. This per-spective was troubling to some; in 1960,Donald Ferguson proposed that “scientific”music theory, in fact, could not provide anadequate account of musical expression. Asa corrective he offered a careful and thor-ough consideration of the basis of expres-sivity in music. Although the approach hadmuch in common with that of Cooke (as dis-cussed in the appendix to Ferguson, 1960),Ferguson was adamant that expressivity inmusic had to be connected with humanexperience. It was because the expressiveelements in music were connected to emo-tional experience that music could serve asa metaphor for the significance of experi-ence (Ferguson, 1960, ix, 185). (For a sim-ilar perspective, but one more thoroughlygrounded in semiotic theory and moresystematically presented, see Coker, 1972 ,chapter 10.)

Metaphor served as a powerful image forFerguson but not as an explicit part of hisresponse to analytical practices that ignoredmusic’s expressivity. For Marion Guck, itwas metaphor itself that suggested an alter-native to positivistic descriptions of musicalstructure. Early in her career Guck becameinterested in the communicative potentialof metaphorical language about music, hav-ing noticed that some of her students pre-ferred such language over the formalisticaccounts of musical structure that werecommonly the focus of instruction in musicanalysis. Through a series of analytical exer-cises she and her students explored theuse of such language and its relationship to

traditional structuralist approaches. She con-cluded that metaphorical language could putstudents more directly in touch with thoseaspects of music upon which traditional ana-lytical techniques were focused, and addrichness to their understanding of thoseaspects.

If perceived musical structure is indivisi-ble from physical and emotional response,then metaphors may offer an embryonicstructural interpretation reinforced by –explained through – physical-emotionalresponses. If a structural interpretation isnot understood by itself, experiencing theresponses may be another avenue to under-standing the structure. Equally, metaphorsoffer a physical-emotional experience rein-forced by – explained through – an embry-onic structural interpretation. (Guck, 1981,42 )

Guck eventually came to argue that thewhole of analytical discourse was rooted inmetaphor (Guck, 1991), although she alsostrove to connect metaphorical discoursewith the “scientific” approach adopted bymany music analysts. She concluded that,while the claims for a scientific languageabout music could not be sustained, state-ments about music – whether such state-ments made use of explicit metaphors orwhether they were restricted to less colorfultechnical descriptions – could be organizedinto consistent and coherent systems cor-related with intersubjectively apprehendedmusical events, and thus able to approachthe methodology of scientific inquiry (Guck,1994).

Although the Anglo-American approachdominated research in music theory andanalysis in the latter half of the twenti-eth century, scientism of the sort to whichFerguson and Guck (among many others)took exception never grabbed hold as firmlyon the English side of the equation. When, in1990, Nicholas Cook described music anal-ysis as metaphorical he was attempting toplace it within the context of aesthetic andpsychological approaches to music that heviewed not only as viable but as offeringkey insights into music as a cultural prod-uct. Analysis, from this perspective, is a way


of imagining music – a metaphor for musi-cal experience rather than any sort of lit-eral record of that experience – much as forScruton musical understanding itself wasfundamentally imaginative (Cook, 1990,10–43).

As a whole, Anglo-American musicol-ogy (as distinct from music theory) was lessin thrall to the scientific paradigm duringthe post–World War II period than weremusic theory and analysis (although somemusicologists aspired to a commensuratepositivism; see Kerman, 1985 , chapter 2).On the rare occasion when metaphor rose tothe surface of musicological inquiry it wasin the context of an over-arching patternof thought that shaped ideas about music.Thus Ruth Solie, in her study of melody,proposed to explore the metaphoric lan-guage used to characterize melody to betterunderstand how earlier periods conceivedmelody.

For example, if your are dealing with an“organic structure” or a “melodic curve”or a “universal language,” what sorts ofbehavior will you expect to observe from it,and therefore make note of? What charac-teristics will you perceive in “the embodiedwill to motion” that you might not see ina “pitch-time trajectory” or in a “stochas-tic process with sequential dependencies” –notwithstanding the fact that all threephrases refer to the same melody? (Solie,1977, 9; see also Solie, 1980)

Some 20 years later Bennett Zon used asimilar approach in his exploration of con-ceptual models used by nineteenth-centuryBritish musicologists, but focused on themetaphorical templates provided by art,religion, and science. These templates pro-vided British musicologists with alternativesto straightforward chronological narratives;Herbert Spencer, for instance, writing in1857, adopted the framework of evolution-ary theory to explain the development ofmusic, and the colloquy that arose aroundthis proposal had a significant impact on thecourse taken by British musicology in thelatter half of the nineteenth century (Zon,2000, 120–125).

Music, Metaphor, and Cognitive Science

The perspective that guided Solie’s work,focusing as much on the language used todescribe music as the music itself, was alsoone that came to prominence in the fieldof ethnomusicology around the same time.Steven Feld, who had a long-standing inter-est in how language was used to describemusic, noted this trend in a 1981 essaythat also made an important contributionto the study of music and metaphor. Draw-ing on the work of Lakoff and Johnson(1980), David Rumelhart (1979/1993), andRobert Verbrugge (1979), Feld argued thatthe metaphorical descriptions used by theKaluli of Papua New Guinea were a reflec-tion of key aspects of their everyday expe-rience. The Kaluli describe melodic inter-vals – whether in their own music or in themusic of others – with the same terms theyuse to characterize features of waterfalls. Forinstance, in the language of the Kaluli sameans “waterfall,” and a mogan is a still orlightly swirling waterpool; sa-mogan is theflow of a waterfall into a level waterpoolbeneath it. Sa-mogan is also used to describea melodic line that descends to a repeatednote, the contour of which replicates that ofa waterfall flowing into a pool (Feld, 1981,30–31; see also Feld, 1982). The system ofmetaphorical relationships upon which suchcharacterizations draw offers a rich descrip-tion of musical events, but one that also hasits limitations: for example, the Kaluli do nothave specific names for ascending intervals,which nonetheless do occur in their music.

Feld’s work pointed toward a newapproach to metaphor and music thatwas based on two important assumptions.The first was that metaphor was not sim-ply a literary device but was instead abasic structure of understanding (Lakoff,1993). The second was that music consti-tuted a conceptual domain that was, insome measure, independent of language. Asa consequence of these two assumptionsmetaphorical descriptions of music cameto be regarded as capable of providing keyinsights into how the understanding of musicwas structured. A notion closely associated


with the contemporary theory of metaphor,and that would prove important for musicscholars, was that of an image schema(Johnson, 1987). Image schemas provided atheoretical basis for metaphorical descrip-tions of music grounded in embodied expe-rience, an approach that fit with many ana-lysts’ intuitions about the nature of musi-cal knowledge and that offered a way tomove beyond – or add another dimensionto – the abstract formalisms prominent inmuch music-theoretical work. Subsequentto a special session at the 1996 annual meet-ing of the Society for Music Theory, an issueof the journal Theory and Practice was givenover to connections between music the-ory and embodied knowledge, and includedarticles by Janna Saslaw on force dynam-ics in the theoretical writings of HeinrichSchenker and Arnold Schoenberg (Saslaw,1997–1998), Candace Brower on embodiedschemas in Edgard Varese’s Density 2 1.5 forsolo flute (Brower, 1997–1998), and SteveLarson on how the understanding of tonalmelodies is shaped by experience with theforces of gravity, magnetism, and inertia(Larson, 1997–1998).

The cognitive perspective on metaphorand music was, in some instances, partof a broader perspective on the cognitivecapacities that shape humans’ understand-ing of music (Spitzer, 2004 ; Zbikowski,1991, 1998, 2002) but was often employedin one of two more restricted ways. First,metaphor theory was brought to bear onrecognized but not clearly understood con-ceptual models within music theory, includ-ing those pertaining to musical invari-ance (Saslaw & Walsh, 1996), modulationtheory (Saslaw, 1996), hierarchical struc-tures in music (Zbikowski, 1997), andhistorical conceptions of tonal organiza-tion (Gur, 2008). Second, metaphor the-ory provided a way into novel reperto-ries, including heavy metal (Walser, 1991),musical multimedia (Cook, 1998, chap-ter 3), the music of the Grateful Dead(O’Donnell, 1999), the music of Neil Young(Echard, 1999, 2005 , chapter 4), JavaneseGamelan (Perlman, 2004 , chapter 6), andfilm music (Chattah, 2006).

Recent research on metaphor and musicthat embodies a cognitive perspective hascoalesced around a somewhat broader setof issues, in many cases offering alternativesto previous approaches. Prominent here iswork on musical meaning that takes as itsstarting point the assumption that mean-ing is grounded in embodied experience(Aksnes, 2002 ; Borgo, 2004 ; Chuck, 2004 ;Cox, 2001; Johnson, 1997–1998; Walker,2000); an account of the ontology of themusical work framed around the metaphor-ical notion of a musical object (Butterfield,2002); and explorations of the bases for andapplications of ideas about musical motionand musical space (Adlington, 2003 ; Cox,1999; Johnson & Larson, 2003 ; Johnson,2007; chapter 11; Spitzer, 2003).

The broad-based approach advocatedby Zbikowski (2002 , chapter 2), whichadopts a generalized view of metaphor asa kind of cross-domain mapping and pro-poses that music represents a conceptualdomain that can be drawn into such map-pings, has recently been extended to cor-relations between patterns in Azerbaijanicarpet weaving and musical practice (Nar-oditskaya, 2005) and to theoretical workon conceptual blending and music. Prelim-inary work on conceptual blends in whichmusic occupies one of the input spaceswas focused on the possibilities for mean-ing construction created by the correlationof text and music in nineteenth-centuryart songs (Zbikowski, 1999, 2002 , chapter6) but has since been applied to analysesof the nature of musical meaning (Cook,2001), analyses of film music, opera, andmusical multimedia (Johnson, 2004 ; Sayrs,2003 ; Zbikowski, 2002–2003), the analysisof Gyorgy Ligeti’s Lontano (Bauer, 2004),the role of the arabesque in the music ofRavel (Bhogal, 2007), and to the construc-tion of musical meaning as a whole (Chuck,2004).

Although most applications of the con-temporary theory of metaphor to music havebeen broadly theoretical, recent empiricalstudies by Zohar Eitan and his colleagueshave begun to show in greater detail howmetaphor structures our understanding of

Lawrence Zbikowski

Highlight


music (Eitan & Granot, 2006), and how themetaphors used to characterize musical rela-tionships reflect the influence of culuture(Eitan & Timmers, 2006). These studies sug-gest not only ways to study how metaphorstructures our understanding of music, butalso ways to investigate how metaphor-ical processes operate in nonlinguisticdomains.

Metaphor and the Analysis of Music

Although the preceding section provided ahistorical and conceptual context for thequestions asked at the opening of this chap-ter, it did not provide a methodology foranswering these questions. The purpose ofthis section is to present such a methodology,which takes as its point of departure the con-temporary theory of metaphor (as character-ized by Lakoff, 1993) and the compositionaltechnique of text painting.

The basic idea of text painting is simpleenough. When a particularly strong or com-pelling image occurs in the text for a musicalwork, the composer writes the accompany-ing music to suggest, or “paint,” the image.Thus, if the text mentions a galloping horse,the music coincident with the text mightimitate the sound and action of a horse pro-ceeding at full speed. While there are lim-its to what can be represented in this way,composers have found the means to por-tray descents from heaven, rippling streams,spinning wheels, physical trembling, sexualclimax, and a host of other vibrant images(Macy, 1996; Zbikowski, 2002 , chapter 2).

The example of text painting I want toconsider here involves the portrayal of theact of knocking on a door. It comes tothe fore near the beginning of the move-ment that provided the music for Example 1,which was from Bach’s Advent cantata “Nunkomm der Heiden Heiland” (BWV 61). Eachof the three preceding movements of thecantata explores an aspect of the Adventtheme. The first movement is an overturewhose text is taken from a chorale byMartin Luther: “Come now, Savior of thegentiles, known to be the child of a

Virgin, the whole world marvels that Godshould have ordained such a birth for Him.”The text for the second-movement recita-tive is by the Hamburg poet, theologian,and pastor Erdmann Neumeister and speaksof the wonder of God made incarnate. Thethird movement, an aria with a text also byNeumeister, returns to the summons statedby the overture: “Come, Jesu, come to Thychurch and grant a blessed New Year!” Butwith the fourth movement Christ is sud-denly before us, speaking words from thethird chapter of Revelation: “Behold, I standat the door, and knock. If any man hearmy voice, and open the door, I will comein to him, and will sup with him, andhe with me.” Bach sets this passage as anaccompanied recitative for baritone, withthe strings playing pizzicato throughout; thescore for the entire movement is given inExample 4 .

Bach’s text painting is centered on thewords “und klopfe an” – that is, “and knock.”Bach uses three compositional techniquesto paint this activity. First, he summonsthe repetitions we associate with the actof knocking by repeating the words, andby using three notes to set the first sylla-ble of the initial “klopfe” (a device called amelisma). Second, he uses staccato marks onthe three notes of the melisma, which placesilences between these notes; these silencesare similar to those that fall between knockson a door. Third, he sets the words witha broken chord (or arpeggio). This placesa kind of distance between each successivenote but also allows us to hear all as belong-ing to a single connected gesture.

The conventional explanation for whytext painting works relies on the idea ofmimesis: the image of knocking is sum-moned by Bach’s setting of “und klopfean” because the music imitates the soundof knocking. While this is partially truefor Bach’s text painting, there are certainthings that are not quite right. Knocksare usually unpitched, but Bach gives usdifferent pitches for each blow; knockingis not usually accompanied, but here wehave pizzicato strings pulsing in the back-ground. A few writers have gone so far as to


Example 4. Score for the fourth movement (Recitativo) from J. S. Bach’s cantata “Nun komm derHeiden Heiland” (BWV 61).

interpret the steady plucking of the orches-tra in this movement as a further embodi-ment of knocking, but this seems somethingof a stretch. Not only are the attack points

too widely spaced to sound much like knock-ing but the effect is far too persistent, morelike Edgar Allen Poe’s telltale heart than asummons from the Savior.


Example 4. (cont.)

In fact, text painting is not a matter ofsimple mimesis, in which music, through itsresemblance to a natural sound, representsthat sound, but of a more complex processthrough which music represents the image-

schematic structure of some event or situ-ation. This sort of representation is some-what like the iconicity of rhetorical figuresdiscussed by Mark Turner (1998). Turnernoted that the form of a rhetorical figure


is sometimes matched to the meaning thespeaker wishes to convey, connecting theimage-schematic structure of the form withthe image-schematic structure of the mean-ing. Thus, a rhetorical figure based on rep-etition, such as anaphora (which involvesthe repetition of the same word or group ofwords at the beginning of successive clauses,sentences, or lines), can be used to summonthe image of repeated blows, as in an exam-ple attributed to Longinus: “By his man-ner, his looks, his voice, when he strikesyou with insult, when he strikes you like anenemy, when he strikes you with his knuck-les, when he strikes you like a slave.” Theefficacy of such a connection is straightfor-ward enough – Turner remarks, “Involvingmembers of the audience in the imageschema of the iconic form automaticallyinvolves them in the basic structure of themeaning, thus moving them part way towardaccepting the whole” (1998, 50–51). In asimilar way, Bach’s recitative embodies theimage-schematic structure of the act ofknocking at the very moment when knock-ing is mentioned in the text. Bach’s musicthus moves the listener part of the waytoward understanding the force of Christ’sact of knocking: where previous movementsin the cantata have summoned Christ, Christis now summoning us.

More generally, the connection of musicto text in instances such as this relieson structural correlations between thetwo domains. The specific correlations arebetween image-schematic structures. In thepresent example, the text calls up the famil-iar situation of a person standing before adoor with the intent of communicating withpeople on the other side of the door. Ascene of this sort typically involves knock-ing on the door to establish communication;knocking, in turn, is accomplished througha series of regularly spaced physical gesturesthat yield a sequence of unpitched soundsof short duration. The conceptual domainset up by the text thus includes the image-schematic structure associated with the actof knocking. The bass melody in the openingmeasures of this movement does not sum-mon anything as specific as does the text,

but it nonetheless participates in establish-ing a conceptual domain structured in partby image schemata. Features of this con-ceptual domain include the steadily puls-ing strings which contrast with the flowingmelody of the bass voice, the dissonancesthat occur against the pedal E3 in the accom-paniment, and the E minor tonality that isprojected. The projection of any tonality isa process that unfolds over time – an impor-tant part of that process in the case at handis the bass melody, which is restricted to justthose pitches that are necessary for defin-ing E minor. In the course of this melody thedistinctive melodic gesture that occupies thebeginning of measure 3 stands out: it intro-duces the largest leap thus far (the minorseventh from D3 to C4), the only melisma,and concludes with another minor-seventhleap (A3 to B2). The passage ends with theshortest notated durations of the passage(the sixteenth notes at the end of measure 3)which serve to further set this measure offfrom the rest. The image-schematic struc-ture that is relevant here is of a series of dis-crete events that are evenly spaced and thatstand out from their surroundings; this thencorrelates with the image-schematic struc-ture of the conceptual domain set up by thewords to produce an instance of text paint-ing.

One question raised by this example iswhether the connection between the image-schematic structure of knocking and theimage-schematic structure of this musicalpassage is necessary. The answer is a qualified“no.” It might indeed be possible, withoutthe text, to make a connection between theact of knocking and the music of measure 3 .There is enough urgency in the music Bachwrites – an urgency that includes both themore rapid durations at the end of the mea-sure and the much smaller registral spacethey inhabit (contracting from C4 to B2 inthe first part of the measure to F�

3 to B4 atits end) – that knocking seems a fairly gooddescription for the music. But the musiccould be described in other ways as well:the whole of the melody in these openingmeasures could be characterized in terms ofthe imagined movements of an actor on a


stage who, after a series of relatively con-strained gestures (in measures 1–2), suddenlygestures in an exaggerated, expansive way(at the beginning of measure 3) before cor-recting her excess at the end of the passage.This characterization would draw on themetaphor of musical “space” (and its atten-dant image-schematic structure) and tendto emphasize the way the pitches of themelody are disposed within this space morethan their rhythmic features. The charac-terization would also provide a perspectivethat encompasses the whole of the passagerather than focusing on measure 3 . The con-text provided by the text from Revelations isthus key to hearing the music of measure 3

as a representation of knocking and not as arepresentation of something else. There are,nonetheless, limits to how the passage canbe characterized – were someone to describethe passage as an energetic portrayal ofreligious bliss we would wonder if they hadlistened to the same music as we had, sincethere is almost nothing in the musical eventsof measures 1–4 – or the image-schematicstructures through which we might orga-nize our understanding of these events – thatwould support such a characterization.

Another important factor that shapes cor-relations between music and other domainsis cultural knowledge. Describing musicalpitches in terms of their disposition in space(with one pitch “higher” or “lower” thananother) has been a commonplace in west-ern traditions since at least the Middle Ages(Cox, 1999; Duchez, 1979; Zbikowski, 2002 ,chapter 2). Other descriptions are, how-ever, possible: in Bali and Java, for instance,pitches are conceived of not as “high” and“low” but as “small” and “large” (Zanten,1986, 85), a conception that reflects accu-rately the norms of acoustic production –small things typically vibrate more rapidlythan large things. Thus we would not expectmembers of a culture that did not practiceknocking as a way of announcing an arrival orthe initiation of communication to make theconnection between the music of measure 3

and the words “und klopfe an” (translated,of course, into the appropriate language forcommunication).

Text painting is, admittedly, a some-what rarified compositional technique. Itnonetheless points to the basis for metaphor-ical descriptions of music and gives somesense of how the conceptual domain ofmusic might participate in metaphoricalmappings. When we describe a musical pas-sage as “obstinately plodding” or a chord as“sour and biting” we are making connec-tions between one domain of experience(having to do with the ways bodies canmove through space, the sense of taste, orthe physical actions accomplished by teeth)and the domain of music. The domain ofmusic includes various musical events aswell as ways of understanding their rela-tionships to one another; these relationshipsare in part structured by image schemata.Just how this is accomplished is still beingexplored empirically, but one of the best the-oretical accounts is provided by LawrenceBarsalou’s theory of perceptual symbol sys-tems (Barsalou, 1999). According to thistheory, sequences of musical events pro-duce brain maps that can be correlated withbrain maps produced by other modalities(including vision, taste, and proprioception);these correlations then operate as symbolsto form the basis for conceptual knowl-edge. The array of perceptual symbols (orimage schemata) that may be used tostructure a given relationship is potentiallyquite extensive; cultural knowledge pro-vides one constraint on which structures arechosen.

The notion of music as an indepen-dent domain with its own properties andrelationships – properties and relationshipsthat language attempts to capture throughmetaphorical descriptions – invites twoextensions of the discussion of metaphor andmusic. First, mappings within music (of thesort discussed in Karbusicky, 1987; Kielian-Gilbert, 1990; Perlman, 2004) are a logicalentailment of this perspective, and a straight-forward example of such a mapping wouldbe between a theme and variations derivedfrom the theme. (For a rich considerationof this topic see Cone, 1987.) These map-pings may, however, be closer to those ofanalogy (and emphasize the alignment of


Example 5 . Basic conceptual integration network for Bach’s text painting of “undklopfe an.”

structural features) than to metaphor (withits emphasis on the construction of mean-ing through the correlation of rich networksof knowledge; cf. Gentner, Bowdle, Wolff, &Boronat, 2001). Second, concepts from themusical domain may combine with conceptsfrom another conceptual domain to create aconceptual blend (Fauconnier & Turner, thisvolume). Two interconnected examples ofconceptual blends are provided by the musicof Example 4 .

The first blend is produced by the textpainting that occurs in measure 3 ; a dia-gram of the conceptual integration networkfor the blend is shown in Example 5 . Thegeneric space for the blend focuses on phys-ical aspects of the act of knocking: therepeated actions that make up knocking, the

sharp disjunction between sound and silencethat results, and the way knocking breaksinto our attention. The text space is set up bythe semantic associations generated by thewords “und klopfe an,” which not only bringthe physical act to mind but also its typicalcontext: a summons of some sort (if only tocome to the door and open it). The musicspace is set up by Bach’s text painting: hisrepetitions of “und klopfe an,” the melismawith staccato articulation, and the arpeg-gio that provides a contextual frame for thenotes that set the words. In the blend, musi-cal and linguistic concepts combine to pro-vide a musical representation of someoneknocking on a door.

There is more, of course, to the passagefrom the third chapter of Revelation than


simply a description of someone knockingon a door. The image is central to the sum-mons with which this fragment is concerned:Christ is calling us to his church. A simple,everyday act is thus made to resonate witha much more profound meaning: answeringthis summons is but the beginning of a chainof entailments with profound consequencesfor all who are concerned. A glimpse at onemanifestation of this chain of entailmentsis provided by a second conceptual blend,which is set up over the course of the move-ment as a whole.

As noted in my discussion of the first fourmeasures of the movement, Bach makes useof some striking dissonances in the processof establishing E minor. These dissonances –which involve the dominant-seventh andleading-tone chords of E minor – resolve atmeasure 4 , coincident with the second iter-ation of “und klopfe an.” Christ’s knockingthus has a real and audible effect on themusic.

After the arrival on E minor in measure 4

the bass pedal is abandoned, and the musicmoves toward G major, which first appearsin measures 6 and 7 and then is confirmedwith the final cadence in measure 10. Themove toward G major is accompanied by achange in the vocal writing, which becomesmore lyrical. By the end of this short move-ment we understand that the opening por-tion, with its obstinate pedal tone and disso-nant harmonies, is meant to lead toward thisdenouement – it is something we have toleave in order to get to the safe haven of thefinal cadence. This journey is in fact preparedby Bach’s text painting, for it is the musicalmaterials associated with the setting of “undklopfe an” that push us away from the staticand dissonant opening materials toward theprogress and consonance represented by theG major music. Just as the image of knock-ing is crucial for the larger story told by thetext, the musical representation of knock-ing is crucial for the larger story told by themusic.

This leads us to the conceptual blendshown in Example 6, which takes in thewhole of the movement. The generic space

for the blend focuses on the notion ofredemption. Within the general context ofan Advent cantata, redemption is naturallyassociated with the act of freeing the believerfrom the consequences of sin. Within themore specific context of the Pietism thatinfluenced Bach’s interpretation of the text,redemption means something closer to itsetymological roots – that is, to buy some-thing back – and thus requires somethingof the believer as well. The text space pro-vides the basic elements of the story ofredemption: Christ knocking at the door;the believer opening the door; and the actof redemption itself, symbolized by Christ’sentrance and the shared meal. Within themusic space we get nothing quite as preciseas this – indeed, the musical events couldbe mapped onto a variety of stories or sit-uations – but we do get a tightly organizedsequence of events. This sequence involvesa number of musical elements and connectswith some of our ideas about redemption,including movement from a static and dis-sonant situation into a progressive and con-sonant one. In the blend, the narrative fromthe passage out of Revelation is compressedwith syntactic processes proper to music. Webegin in a static, dissonant environment withChrist announcing himself and then knock-ing at the door. This knock is a summonsto redemption, and the dissonant environ-ment starts to become more consonant. Bythe time the opening of the door is men-tioned (in measure 6, with the words “unddie Tur auftun”) we have entered the orbitof G major, and the remainder of the move-ment fills out the theme of redemption andaffirms G major, with one important excep-tion. With the singer’s very last word (“mir,”in measure 9), Bach returns momentarily toE minor, a move that casts a shadow over thescene and seems to suggest the sacrifices thatredemption requires. Although there is littledoubt of the promise of redemption at theconclusion of the movement, the musicalsyntax through which the story is told pointsto the struggles that are required to achieveredemption. Tonal closure is not assured,but must be won; redemption requires


Example 6. Conceptual integration network for the fourth movement from J. S.Bach’s cantata “Nun komm der Heiden Heiland.”

more than simply opening the door to theSavior.

Conclusion

Music is a rich and complex product of cul-ture – the brief examples discussed hereinclude music that is part of the ritualof religious service (Bach’s cantata), musicwith a programmatic title (Schumann’s“Traumerei”), light instrumental music fordiversion (Dvorak’s “Humoresque”), andeven music for the cabaret (which inspiredKarbusicky’s example). These possibilitiesbarely scratch the surface of musical expres-sion, which is manifested in all known

human cultures; includes music for ritual,dance, song, diversion, and a multitude ofother activities; and touches on the completerange of human emotion. The cultural prac-tice of music is also largely non-linguisticand non-referential, although both languageand reference can play a role in musicalpractice. Given the range of musical expres-sion and its independence from language, itis not surprising that language about musicis often metaphorical, nor that the topic ofmetaphor and music has been touched on bya wide range of scholarly disciplines.

Michael Tomasello (1999, chapter 5)recently proposed that one of the pri-mary functions of language is to manipu-late the attention of another person within

52 0 LAWRENCE M. ZBIKOWSKI

a shared referential frame. It could beargued that one of the primary functionsof music is to manipulate the emotions ofothers. Although this argument is hardlynew (see, for instance, Meyer, 1956), it hasoften been advanced within the relativelynarrow context of instrumental music pro-duced in western Europe during the lateeighteenth and nineteenth centuries. Theargument could easily be broadened throughthe recognition that music can also manipu-late the emotions through the way it shapesritual, dance, and the rendering of a text. If itis that case that language and music have dif-ferent functions within human culture – thatthey comprise different domains of expe-rience – it follows that mappings betweenthese domains would yield numerous possi-bilities for the sort of meaning constructionassociated with metaphor.

According to current theory, mappingsbetween language and music rely on image-schematic structures that are commonto the two domains. Music will tendto instantiate such structures dynamically,while language will call them up throughreference. When music summons knockingin the fourth movement of Bach’s “Nunkomm der Heiden Heiland” it does so byreplicating features of the act of knock-ing. The text, by contrast, simply refersto the act, relying on the listener to callup the dynamic schema once the referen-tial frame has been activated. Combinationsof music and text such as those createdthrough the compositional technique of textpainting thus represent a kind of laboratoryfor the study of image-schematic structure.Any schema thought to underlie mappingsbetween the two domains will have to berepresented in each, and the two differentmodes of activating schemas – dynamically,and through reference – will give furtherindications of their relevant properties.

Music, as an expressive medium dis-tinct from that of language, can also offerinteresting possibilities for thinking aboutmetaphorical processes. More purely “musi-cal” mappings, such as those between atheme and variation, appear to be closer

to analogy. Similar relationships betweensonic patterns can also be seen in proseand, more typically, metered poetry, sug-gesting an exploration of these instancesin terms of analogy as well as metaphor.The participation of music in conceptualintegration networks, such as the two dis-cussed in connection with the movementfrom Bach’s cantata, offers possibilities formeaning construction that blends conceptsfrom music and other domains. Conceptualblends that involve music and some otherdomain also provide an opportunity to studythe structural features of each domain, giventhe assumption that blends require a uni-form topography between the mental spacesinvolved in the conceptual integration net-work.

The question that has often been posed is,“Is music a language?” The composer DavidLidov (2005) proposed reversing the termswith his question, “Is language a music?”The exploration of metaphor and music hasmuch to say to both questions, as well asto the constituent features of both of theseuniquely human modes of expression.

Note

1 The pitch designation I use is that of theAmerican Society of Acousticians: middle Cis C4 ; the B below middle C is B3 ; the octaveabove middle C is C5 .

References

Adlington, R. (2003). Moving beyond motion:Metaphors for changing sound. Journal of theRoyal Music Association, 12 8(2), 297–318.

Agawu, V. K. (1991). Playing with signs: A semioticinterpretation of classical music. Princeton, NJ:Princeton University Press.

Agawu, V. K. (1999). The challenge of semiotics.In N. Cook & M. Everist (Eds.), Rethinkingmusic (pp. 138–160). Oxford: Oxford Univer-sity Press.

Aksnes, H. (2002). Perspectives of musicalmeaning: A study based on selected works byGeirr Tveitt. Oslo: Unipub AS.

Ayrey, C. (1994). Debussy’s significantconnections: Metaphor and metonymy in

METAPHOR AND MUSIC 52 1

analytical method. In A. Pople (Ed.), Theory,analysis and meaning in music (pp. 127–15 1).Cambridge: Cambridge University Press.

Babbitt, M. (1972). Past and present concepts ofthe nature and limits of music. In B. Boretz& E. T. Cone (Eds.), Perspectives on contempo-rary music theory (pp. 3–9). New York: W. W.Norton.

Barsalou, L. W. (1999). Perceptual symbolsystems. Behavioral and Brain Sciences,2 2 (November), 577–609.

Bauer, A. (2004). “Tone-color, movement, chang-ing harmonic planes”: Cognition, constraints,and conceptual blends in modernist music.In A. Ashby (Ed.), The pleasure of modernistmusic: Listening, meaning, intention, ideology(pp. 121–152). Rochester, NY: University ofRochester Press.

Bhogal, G. (2007). Debussy’s Arabesque andRabel’s Daphnis et Chloe (1912 ). Twentieth-Century Music, 32 (2), 171–199

Borgo, D. (2004). The play of meaning and themeaning of play in jazz. Journal of Conscious-ness Studies, 11(3–4), 174–190.

Broeckx, J. L. (1997). Beyond metaphor: Musicalfigures or “Gestalte” as expressive icons. Jour-nal of New Music Research, 2 6, 266–276.

Brower, C. (1997–1998). Pathway, blockage, andcontainment in Density 2 1.5 . Theory and Prac-tice, 2 2 –2 3 , 35–54 .

Budd, M. (1985). Understanding music. Proceed-ings of the Aristotelian Society, 59(Suppl.), 233–248.

Budd, M. (1989). Music and the communicationof emotion. Journal of Aesthetics and Art Crit-icism, 47(2), 129–138.

Budd, M. (2003). Musical movement and aes-thetic metaphors. British Journal of Aesthetics,43(3), 209–223 .

Butterfield, M. (2002). The musical object revis-ited. Music Analysis, 2 1(3), 327–380.

Charles, D. (1995). Music and antimetaphor(to Eero Tarasti). In E. Tarasti (Ed.), Musi-cal signification: Essays in the semiotic theoryand analysis of music: Vol. 12 1. Approachesto semiotics (pp. 27–42). Berlin: Mouton deGruyter.

Chattah, J. R. 2006. Semiotics, pragmatics, andmetaphor in film music analysis. UnpublishedPhD dissertation, Florida State University.

Chuck, G. (2004). Musical meaning and cognitiveoperations of the embodied mind. UnpublishedPhD dissertation, University of Rochester,Eastman School of Music.

Coker, W. (1972). Music and meaning: A theoreti-cal introduction to musical aesthetics. New York:The Free Press.

Cone, E. T. (1987). On derivation: Syntax andrhetoric. Music Analysis, 6(3), 237–255 .

Cook, N. (1990). Music, imagination, and culture.Oxford: Clarendon Press.

Cook, N. (1998). Analysing musical multimedia.Oxford: Clarendon Press.

Cook, N. (2001). Theorizing musical meaning.Music Theory Spectrum, 2 3(2), 170–195 .

Cooke, D. (1959). The language of music. Oxford:Oxford University Press.

Cox, A. W. (1999). The metaphoric logic of musicalmotion and space. Unpublished PhD disserta-tion, University of Oregon.

Cox, A. W. (2001). The mimetic hypothesis andembodied musical meaning. Musicæ Scientiæ,5(2):195–212 .

Cumming, N. (1994). Metaphor in Roger Scru-ton’s aesthetics of music. In A. Pople (Ed.),Theory, analysis and meaning in music (pp.3–28). Cambridge: Cambridge UniversityPress.

Cumming, N. (2000). The sonic self: Musical sub-jectivity and signification. Advances in Semi-otics. Bloomington: Indiana University Press.

Davies, S. (1994). Musical meaning and expression.Ithaca, NY: Cornell University Press.

Deacon, T. W. (1997). The symbolic species: TheCo-evolution of language and the brain. NewYork: W. W. Norton.

Deacon, T. W. (2003). Universal grammar andsemiotic constraints. In M. H. Christiansen &S. Kirby (Eds.), Language evolution (pp. 111–139). Oxford: Oxford University Press.

Dempster, D. (1998). Is there even a grammar ofmusic? Musicæ Scientiæ, 2 (1), 55–65 .

Dubiel, J. (1990). “When you are a Beethoven”:Kinds of rules in Schenker’s Counterpoint.Journal of Music Theory, 34(2), 291–340.

Duchez, M.-E. (1979). La representation spatio-verticale du caractere musical grave-aigu etl’elaboration de la notion de hauteur de sondans la conscience musicale occidentale. Actamusicologica, 51(1), 54–73 .

Echard, W. (1999). An analysis of Neil Young’s“Powderfinger” based on Mark Johnson’simage schemata. Popular Music, 18(1), 133–144 .

Echard, W. (2005). Neil Young and the poetics ofenergy. Bloomington: Indiana University Press.

Eitan, Z., & R. Y. Granot. (2006). How musicmoves: Musical parameters and listeners’

Lawrence Zbikowski

Highlight


images of motion. Music Perception, 2 3(3),221–247

Eitan, Z., & R. Timmers. (2006). Beethoven’s lastpiano sonata and those who follow crocodiles:Cross-domain mappings of auditory pitch in amusical context. In M. Baroni, A. R. Addessi,R. Caterina, & M. Costa (Eds.) Proceedings ofthe 9th International Conference on Music Per-ception and Cognition, Bologna, Italy, August2 2 –2 6 2 006 (pp. 875–882). [n.p.]: The Soci-ety for Music Perception & Cognition andEuropean Society for the Cognitive Sciencesof Music.

Feld, S. (1981). Flow like a waterfall: Themetaphors of Kaluli musical theory. Yearbookfor Traditional Music, 13 , 22–47.

Feld, S. (1982). Sound and sentiment: Birds,weeping, poetics, and song in Kaluli expres-sion. Philadelphia: University of PennsylvaniaPress.

Ferguson, D. N. (1960). Music as metaphor: Theelements of expression. Minneapolis: Universityof Minnesota.

Gentner, D., Bowdle, B. F., Wolff, P., & Boronat,C. (2001). Metaphor is like analogy. In D.Gentner, K. J. Holyoak, & B. N. Kokinov(Eds.), The analogical mind: Perspectives fromcognitive science (pp. 199–253). Cambridge,MA: MIT Press.

Goehr, L. (1998). The quest for voice: Music, poli-tics, and the limits of philosophy. Berkeley: Uni-versity of California Press.

Goodman, N. (1976). Languages of art: Anapproach to a theory of symbols (2nd ed.).Indianapolis: Hackett Publishing.

Guck, M. A. (1981). Musical images as musicalthoughts: The contribution of metaphor toanalysis. In Theory Only, 5(5), 29–43 .

Guck, M. A. (1991). Two types of metaphorictransfer. In J. C. Kassler (Ed.), Metaphor: Amusical dimension: Vol. 1: Australian studies inthe history, philosophy and social studies of music(pp. 1–12). Sydney: Currency Press.

Guck, M. A. (1994). Rehabilitating the incorri-gible. In A. Pople (Ed.), Theory, analysis andmeaning in music (pp. 57–73). Cambridge:Cambridge University Press.

Gur, G. (2008). Body, forces, and paths:Metaphor and embodiment in Jean-PhilippeRameau’s conceptualization of tonal space.Music Theory Online 14(1), March. http://mto.societymusictheory.org/issues/mto.08.14 .1/mto.08.14 .1.gur.html.

Hatten, R. S. (1994). Musical meaning in Beet-hoven: Markedness, correlation, and inter-

pretation. Bloomington: Indiana UniversityPress.

Hatten, R. S. (1995). Metaphor in music. In E.Tarasti (Ed.), Musical signification: Essays inthe semiotic theory and analysis of music: Vol.12 1. Approaches to semiotics (pp. 373–391).Berlin: Mouton de Gruyter.

Hatten, R. S. (2004). Interpreting musical gestures,topics, and tropes: Mozart, Beethoven, Schubert.Bloomington: Indiana University Press.

Jankelevitch, V. 2003 . Music and the ineffable (C.Abbate, Trans.). Princeton, NJ: Princeton Uni-versity Press.

Johnson, M. L. (1987). The body in the mind: Thebodily basis of meaning, imagination, and rea-son. Chicago: University of Chicago Press.

Johnson, M. L. (1997–1998). Embodied musicalmeaning. Theory and Practice, 2 2 –2 3 , 95–102 .

Johnson, M. L. (2007). The meaning of the body:Aesthetics of human understanding. Chicago:University of Chicago Press.

Johnson, M. L., & S. Larson. (2003). “Somethingin the way she moves”: Metaphors of musi-cal motion. Metaphor and Symbol, 18(2), 63–84 .

Johnson, S. R. (2004). Hearing the unvoiceable:Writer’s block in Benjamin Britten’s “Death inVenice.” Unpublished PhD dissertation, Uni-versity of Wisconsin–Madison.

Juslin, P. N., & Sloboda, J. A. (Eds.). (2001). Musicand emotion: Theory and research. New York:Oxford University Press.

Karbusicky, V. (1987). “Signification” in music:A metaphor? In T. A. Sebeok & J. Umiker-Sebeok (Eds.), The semiotic web 1986 (pp. 430–444). Berlin: Mouton de Gruyter.

Kerman, J. (1985). Contemplating music: Chal-lenges to musicology. Cambridge, MA: HarvardUniversity Press.

Kielian-Gilbert, M. (1990). Interpreting musicalanalogy: From rhetorical device to perceptualprocess. Music Perception, 8(1), 63–94 .

Krantz, S. C. (1987, June). Metaphor in music.The Journal of Aesthetics and Criticism, 45(4),351–360.

Lakoff, G. (1993). The contemporary theory ofmetaphor. In A. Ortony (Ed.), Metaphor andthought (pp. 202–251). Cambridge: CambridgeUniversity Press.

Lakoff, G., & Johnson, M. L. (1980). Metaphorswe live by. Chicago: University of ChicagoPress.

Larson, S. (1997–1998). Musical forces andmelodic patterns. Theory and Practice, 2 2 –2 3 ,55–71.

Lawrence Zbikowski

Highlight

Lawrence Zbikowski

Highlight

METAPHOR AND MUSIC 52 3

Lidov, D. (2005). Is language a music? Writings onmusical form and signification. (Musical mean-ing and interpretation). Bloomington: IndianaUniversity Press.

Macy, L. (1996). Speaking of sex: Metaphor andperformance in the Italian madrigal. The Jour-nal of Musicology, 14(1), 1–34 .

Meyer, L. (1956). Emotion and meaning in music.Chicago: University of Chicago Press.

Naroditskaya, I. (2005). Azerbaijani mugham andcarpet: Cross-domain mapping. Ethnomusicol-ogy Forum, 14(1), 25–55 .

O’Donnell, S. (1999). Space, motion, and othermusical metaphors. In R. G. Weiner (Ed.), Per-spectives on the Grateful Dead: Critical writ-ings (Foreword by R. G. Adams; Afterwordby S. Silberman; pp. 127–135). Westport, CT:Greenwood Press.

Perlman, M. (2004). Unplayed melodies: Javanesegamelan and the genesis of music theory.Berkeley: University of California Press.

Powers, H. (1980). Language models and musicalanalysis. Ethnomusicology, 2 4(1), 1–60.

Rumelhart, D. E. (1993). Some problems with thenotion of literal meanings. In A. Ortony (Ed.),Metaphor and thought (pp. 71–82). Cambridge:Cambridge University Press.

Ruwet, N. (1966). Methodes d’analyse en musi-cologie. Revue Belge de Musicologies, 2 0, 65–90.

Ruwet, N. (1972). Langage, musique, poesie. Paris:Editions du Seuil.

Saslaw, J. K. (1996). Forces, containers, and paths:The role of body-derived image schemas in theconceptualization of music. Journal of MusicTheory, 40(2), 217–243 .

Saslaw, J. K. (1997–1998). Life forces: Concep-tual structures in Schenker’s Free compositionand Schoenberg’s The musical idea. Theory andPractice, 2 2 –2 3 , 17–33 .

Saslaw, J. K., & Walsh, J. P. (1996). Musical invari-ance as a cognitive structure: “Multiple mean-ing” in the early nineteenth century. In I. Bent(Ed.), Music theory in the age of Romanticism(pp. 211–232). Cambridge: Cambridge Univer-sity Press.

Saussure, F. D. (1959). Course in general linguistics.(C. Rally, A. Sechehaye, & A. Reidlinger, Eds.;W. Baskin, Trans.). New York: PhilosophicalLibrary.

Sayrs, E. 2003 . Narrative, metaphor, and con-ceptual blending in “The Hanging Tree.”Music Theory Online, 9(1), March. http://www.societymusictheory.org/mto/issues/mto.03 .9.1/mto.03 .9.1.sayrs.html.

Scruton, R. (1974). Art and imagination: A studyin the philosophy of mind. London: Methuen &Co.

Scruton, R. (1983). Understanding music. Ratio,2 5(2), 97–120.

Scruton, R. (1997). The aesthetics of music.Oxford: Oxford University Press.

Solie, R. A. (1977). Metaphor and model in theanalysis of melody. Unpublished PhD disserta-tion, University of Chicago.

Solie, R. A. (1980). The living work: Organicismand musical analysis. 19th-Century Music, 4(2),147–156.

Spitzer, M. (2003). The metaphor of musicalspace. Musicæ Scientiæ, 7(1), 101–120.

Spitzer, M. (2004). Metaphor and musicalthought. Chicago: University of Chicago Press.

Tomasello, M. (1999). The cultural origins ofhuman cognition. Cambridge, MA: HarvardUniversity Press.

Treitler, L. (1997). Language and the interpreta-tion of music. In J. Robinson (Ed.), Music andmeaning (pp. 23–56). Ithaca, NY: Cornell Uni-versity Press.

Turner, M. (1998). Figure. In A. N. Katz, C.Cacciari, R. W. Gibbs, Jr., & M. Turner (Eds.),Figurative language and thought. Counterpoints:Cognition, memory, and language (pp.44–87). New York: Oxford UniversityPress.

Verbrugge, R. R. (1979). The primacy ofmetaphor in development. In E. Winner &H. Gardner (Eds.), Fact, fiction, and fantasyin childhood: Vol. 6. New directions for childdevelopment (pp. 77–84). San Francisco, CA:Jossey-Bass.

Walker, M. E. (2000). Movement and metaphor:Towards an embodied theory of music cogni-tion and hermeneutics. Bulletin of the Coun-cil for Research in Music Education, 145(Sum-mer), 27–42 .

Walser, R. (1991). The body in music: Epistemol-ogy and musical semiotics. College Music Sym-posium, 31, 117–126.

Zanten, W. V. (1986). The tone material of theKacapi in Tembang Sunda in West Java. Eth-nomusicology, 30(1), 84–112 .

Zbikowski, L. M. (1991). Large-scale rhythmand systems of grouping. Unpublished PhDdissertation, Yale University, New Haven,CT.

Zbikowski, L. M. (1997). Conceptual models andcross-domain mapping: New perspectives ontheories of music and hierarchy. Journal ofMusic Theory, 41(2), 11–43 .


Zbikowski, L. M. (1998). Metaphor and musictheory: Reflections from cognitive sci-ence. Music Theory Online, 4(1), January.(http://smt.vesb.edu/mto/issues/mto.98.4 .1/toc.4 .1.html).

Zbikowski, L. M. (1999, October). The blos-soms of ‘Trockne Blumen’: Music and text inthe early nineteenth century. Music Analysis,18(3), 307–345 .

Zbikowski, L. M. (2002–2003). Music theory,multimedia, and the construction of meaning.Integral, 16/17 , 251–68.

Zbikowski, L. M. (2002). Conceptualizing music:Cognitive structure, theory, and analysis. NewYork: Oxford University Press.

Zon, B. 2000. Music and metaphor in nineteenth-century British musicology. Aldershot, England:Ashgate Publishing.

Lawrence Zbikowski

Highlight

metaphor and music

Documents