Integral Volume 16/17, 2002/2003 Articles Matthew BaileyShea Wagner's Loosely Knit Sentences and the Drama of Musical Form 1 Mitchell Turner Interval-ClassExchanges in a Two-Dimensional Pitch-Class Space 35 Brian Alegant Inside the Cadenza of Schoenberg'sPiano Concerto 67 Stephen Slottow Carl Ruggles'sCadential Complex 103 Mark Sallmen Composition with a Single Row Form? Webern's "Schatzerlklein," Op. 18, No. 1 139 Daphne Leong Kaleidoscopic Symmetries:Time and Pitch Relations in Cordon Nancarrow's Tango? 187 Reviews Forum: AnalysingMusicalMultimedia^by Nicholas Cook Scott D. Lipscomb Daphne Leong LawrenceM. Zbikowski 225 227 237 25 1 ClassNotesfor AdvancedAtonal Theory^ by Robert D. Morris Robert W. Peck Reply by Robert D. Morris 269 285 Contributors 286 Review Forum Analysing Musical Multimedia by Nicholas Cook. Oxford: Clarendon Press; New York: Oxford University Press, 1998. Why a whole forum of reviews on Cook's AnalysingMusical Multimedia* I hope readers find that the diverse contributions by Scott Lipscomb, Daphne Leong, and Lawrence Zbikowski justify the forum almost entirely on their own. Then there is the book. It inaugurates its own sub-discipline which reaches into several disciplines besides mainstream music theory, such as music perception and cognition, human physiology, cognitive science, film studies, cultural history- -even television marketing. Of course the book invites reaction from so many angles. From the perspective of music perception and cognition research, several issues arise. To some extent these spring naturally from the perceptual interaction between aural and visual sensory modalities, which is inherent to multimedia. Yet they also stem from the particular orientations Cook chooses in addressing his topic. For instance, how relevant is synaesthesia to analyzing multimedia? What is the significance of a theory devoted to analyzing music-derived multimedia, as opposed to theatrical films, in which music is secondary? How does Cook's book spur empirical scientific work on music perception and cognition, and stimulate interdisciplinarydialogue? Lipscomb's contribution to the forum considers these issues, also bringing to bear his experience teaching and applying Cook's theory in a classroom context: a course he has taught on Multimedia Cognition. Cook's topic raises issues that rarely,if ever, arise when music is considered alone. So it provides a fresh context to examine the application of pre-existing theories, such as those of musical rhythm and grouping. For instance, what does it mean for one medium, say a visual one, to model a particular "hearing" of a musical passage?How is such a model evaluated?And how does the artistic merit of the multimedia piece influence the process and 226 Integral result? These are among the issues examined in Leong's contribution, which is probably the first in our field to evaluate a multimedia analysis on grounds of rhythmic accuracy. It opens doors to future rhythmic analysisof film. Multimedia also provokes inquiries into musical meaning. Schoenberg envisioned Die gluckliche Hand from the start as multimedia, combining music not just with text, but also elaborate instructions for stage action and color lighting. How do we interpret the meaning(s) from such a work? How do recent developments in the musical application of conceptual blending theory offer ways to streamline and formalize Cook's approach to Die gluckliche Hand's "lighting crescendo"? Zbikowski explores these concerns. He then re-analyzes the "lighting crescendo" using the technology of conceptual integration networks (CINs). His analysis points to the psychological tumult of the opera's protagonist, as a meaning conjured by the multimedia experience. Finally, I suspect the relevance for our discipline of analyzing musical multimedia is not yet fully appreciated. Music theorists already routinely lavish analytical attention on a large body of classical music works, operas and ballets- from Monteverdi and Striggio's Orfeo to Stravinsky and Balanchine's Agont and beyond- which are actually instances of musical multimedia, and therefore fall within the scope of Cook's theory.1 Furthermore, multimedia permeates many aspects of our culture- both high and low - more now than ever. The climate of technological change in the early 21st century forecasts a rise in multimedia, and our access to it. There is good reason for the sub-discipline of analyzing musical multimedia to grow. Joshua Mailman Reviews Editor In fact, aspects of Cook's theory have already been applied to opera in Philip Rupprecht's Britten s Musical Language (New York: Cambridge University Press, 2001). Modeling Multimedia Cognition: A Review of Nicholas Cook's Analysing Musical Multimedia Scott D. Lipscomb It is perhaps surprising that a text published over five years ago should warrant a series of reviews in a respected scholarly journal. This occurrence suggests at least two very important outcomes. First, "multimedia"- specifically, the role that music plays within such a context- is finally being given the attention it deserves as a sociologically relevant artifact of contemporary culture, thus worthy of discussion in scholarly music journals. Second, Nicholas Cook's Analysing Musical Multimedia has made a significant contribution to this dialogue in its emphasis, as evident from the book's title, upon the musical component of this multimodal experience. Cook's text does not serve, however, to initiate the analysis of musical multimedia. In fact, the practice of combining music and drama dates back millennia to the Greek dramas of Aeschylus, Euripides, and Sophocles and can be traced throughout the evolution of Western civilization, as represented in the Medieval sacred drama, courtly displays of the Renaissance, Baroque opera, Wagner's Gesamtkunstwerk,and the development of sound film in the 20thcentury. Beginning in the 1950s, with interest intensifying during the most recent two decades, music researchers and psychologists have begun to investigate empirically the relationship between hearing and seeing... sound and image. As I have noted elsewhere,1 in the field of perceptual psychology, interaction between the aural and visual sensory modalities is welldocumented.2 Empirical studies investigating the intermodal relationship in more ecologically valid contexts was initiated in the middle of the 20th century, but did not begin to attract significant attention until the late 1980s. Using a simple "drop the needle" technique, Tannenbaum discovered that music does influence Lipscomb, in press. Sec, for example, McGurk and MacDonald 1976, Radeau and Bertelson 1974, and Staal and Donderi 1983. 228 Integral verbal ratings collected from participants following a dramatic presentation, whether live on stage, in a studio-taped version, or in a video recording of the live performance.3 Using an industrial safety film depicting three accidents, Thayer and Levenson found that skin conductance level, a physiological measure, varied significantly between two conditions, one using a series of mildly active major seventh chords ("documentary music") and one using a repetitive figure based on diminished seventh chords incorporating harsh timbres ("horror music").4 Marshall and Cohen, a study that Cook uses as an empirical basis for his own model of musical multimedia and its cognition, found that the information provided by a musical soundtrack significantly affected judgments of personality attributes assigned by subjects to each of three geometric shapes presented as "characters" in the film.5 Based on the results of this investigation, the authors proposed a paradigm to explain the interaction of musical sound and geometric shapes in motion entitled the "Congruence-Associationist model." They assumed that, in the perception of a composite A-V presentation, separate judgments were made on each of three semantic dimensions (i.e. Evaluative, Potency, and Activity)6 for the music and the film, suggesting that these evaluations were then compared for congruence at a higher level of processing. Since the publication of Cook's text in 1998, Annabel Cohen has gone on to expand the model, significantly clarifying the multi-level relationships that occur between sensory modalities.7 Cook's text is divided into two parts. The first half of the book provides a foundation for the theoretical framework proposed by the author. The remainder of the text consists of three analytical case studies or exemplars to which this specific framework is applied. As a general outline, this organizational structure is extremely clear and provides the reader a functional and concise method of analysis with examples of its practical application. A Tannenbaum 1956. Thayer and Levenson 1984. Marshall and Cohen 1988. For a detailed discussion of these studies and other related work, see lipscomb 1995 or lipscomb and Kendall 1994. Osgood, Sud and Tannenbaum 1957. 7 Cohen 2001. Review Forum: Lipscomb on Cook 229 detailed analysis of the actual content, naturally, reveals numerous possibilities for further discussion or debate, a continuing process for which Cook has provided an excellent starting point. I have had the opportunity to use Analysing Musical Multimedia as a textbook in a graduate-level selected topics course on "Multimedia Cognition" (MUSJTHRY 335-0) at Northwestern University. Cook's text, Michel Chion's excellent Audio-vision? and a course reader including a variety of theoretical and empirical works related to the multimedia experience provided an excellent triumvirate upon which to build knowledge and facilitate discussion about the multi-modal experience.9 Given my own background and experience, the present review of Cook's text will represent a dual perspective: that of a music/multimedia researcherand a university professor. The opening section of Part I introduces many of the concepts that Cook deals with in the following chapters. To demonstrate the manner in which music can influence (or determine) the meaning of a sequence of visual images, several highly creative commercials are deconstructed according their content, both visual and auditory. In my opinion, this is one of the most valuable sections of the text, clearly demonstrating the extremely important role that music plays in this context and hinting at issues of congruence between the audio-visual (A-V) components... an element that will come to play a defining role in Cook's paradigm. I found myself at times, about the selected frustrated, reading examples- completely unfamiliar to me and wanting desperately to view the commercials so that I could experience the A-V combinations described for myself, affording a basis for critical analysis and debate. Perhaps selecting exemplars that are more readily available would have served the audience better or, ideally, making these commercials available on a DVD, either accompanying the text or available separately as a "companion." Instead, at the outset, the reader is placed in a position where one must simply trust the author's description and analysis of the 8 Chion 1990. The complete course syllabus, including a list of literature contained in the course reader, can be found on the "syllabi"page of the present author's web site: http://faculty-web.at.northwcstern.edu/music/lipscomb/. 230 Integral existing interrelationships. Despite this criticism, the clarity of descriptions and select captured still images make the author's intended points effectively. Particularly important in these introductory pages is Cook's insistence that music be considered a communicative medium, extending beyond mere effectinto the realm of meaning. This is an important distinction, though not novel,10 since the model of multimedia toward which he is leading the readerwill require that the meaning attained by each modality be compared for similarity and/or difference. Equally important is his distinction between connotative and denotative meaning, characteristics clearly differentiated in a musical context within the work of Susanne Langer.11 Drawing upon information in Joseph Kerman's monograph on opera,12 Cook suggests that, within this specific musical context, "the identification of word with denotation and music with connotation suggest [a] kind of layered, noncompetitive relationship" (119). To accomplish its connotative task, according to Cook: Musicalstylesand genresofferunsurpassed for communicating opportunities complex social or attitudinal messages practically instantaneously; one or two notes in a distinctive musical style are sufficient to target a specific social and demographic group and to associate a whole nexus of social and cultural values with a product (16-17). With these basic concepts clearly delineated and the multimedia artifact as an object of study, the foundation for Cook's framework has been laid. At this point in the text, I was confused to find myself thrown into a discourse concerning synaesthesia. Though perhaps a topic worthy of brief mention within a book on multimedia, the amount of verbiage devoted to this phenomenon, affecting such a small percentage of the population, seems to imply an importance that is hugely disproportionate to its actual impact upon the typical multimedia experience. Its relevance might be more marked were there a consistent A-V relationship from one synesthete to another. See, for example, Campbell and Heller 1980 or Kendall and Carterctte 1990. 11 Langer 1942. 12 Kerman 1956. ReviewForum:Lipscombon Cook 231 This is not the case, however,and the fact that colorsperceivedin the music listening experiencevary greatly between individuals affectedby this highlyuncommonperceptualanomalysuggeststhat this is not an appropriatebasisupon which to build an overarching theory of multimedia perception. This is, of course, the same conclusion to which Cook comes prior to formulatinghis own model, makingthe guidedtour throughsynaesthesiaand associated theorists- fascinating as it is at times- seem an unnecessary detour. The manyfascinatingmultimediaworksupon whichCook focuses in this section (Messiaen's Couleursde la Citi celeste, and Schoenberg'sDie glilcklicheHand)could Scriabin'sPrometheus, easily have been made relevantbased on aestheticvalue, without the need for a long-windeddiscussionaboutsynaesthesia.It is with the introductionof a metaphor-basedmodel that Cook returnsto what will truly become useful in the analysis of cross-modal relationships. The present author found the discussion about "recordsleeves"laterin this samechapterto be of little relevanceto the primarythesisof the text andwould havelikedto haveseen this space allotted to more meaningful and relevant subject matter relatingto the truemultimediaexperience. In the second chapter, Cook's critical analysis of several important models of cross-modal relationships (Kandinsky, Eisensten,and Eisler)puts the readerright back on trackin the process of consideringthe interrelationshipof the auditoryand visual perceptualmodalities. Supplementedwith commentsmade by esteemedfilm composerBernardHerrmannand the resultsof empirical researchinto the relationship,13the author carefully builds a case for considerationof the multimedia context as a metaphoricalrelationship,based on "enablingsimilarity"and the resulting"transferof attributes"(70).u Cook also identifiesthe - of "emergentproperties" presence- and stressesthe importance in the multimediacontext.Such attributesaresaid to be negotiated between the interacting media within a specific context and "...cannot be subsumed within a model based on the simple mixing or averagingof the propertiesof each individualmedium" thatofMarshall andCohen1988;discussed Primarily previously. 14AfterMarks andLakoff andJohnson 1980. 1978, 232 Integral (69). Just prior to the presentation of his own model, Cook summarizeshis perspective concisely in the following way: . . .whatever music's contribution to cross-media interaction, what is involved is a dynamic process: the reciprocal transfer of attributes that gives rise to a meaning constructed,not just reproduced,by multimedia (97, emphasis added). This emphasis on an emergent meaning that is constructed as a result of the interaction between the various components of a multimedia work is a significant contribution to the study of multimedia. Cook's own paradigm ("Models of Multimedia") is carefully delineated in the final chapter of Part I. Here, the author sets out his approach to the study of multimedia, interrelationshipsbetween the various component media, and potential source(s) of the resulting meaning(s). At its most fundamental level, this model consists of two steps: a similarity test and a difference test. Space limitations for the present review preclude a full description of the model. Its essence, however, involves determining whether component media are communicating the same basic meaning via different perceptual modalities or whether these constituent elements consist of varying messages. In the latter case, the listenerviewer's cognition is a more complex interpretive process. If the media are considered to be communicating the same message, the relationship is said to be conformant (Lakoff and Johnson's "consistent"). At the other extreme, if the media communicate in a manner such that the meaning of each contradicts that of the other, the relationship is said to be one of contest. In the middle ground between these two polar extremes of a continuum exists a complementaryrelationship, in which the relationship is neither consistent nor contradictory. In selecting the identifiers used to describe these models, Cook consciously decided to coin a new set of terms instead of using the common terms already used frequently to describe these relationships (consistent, coherent, and contradictory). Upon initial contact with Cook's model, I considered this a major weakness, incorporating - what I considered at the time- an unnecessary level of interpretation and resulting in needless complexity. However, as I continued to use these concepts in a classroom context and to participate in Review Forum: Lipscomb on Cook 233 animated discussions about the roles of the various media in a multimedia context by all involved, I found that having such were clearly terms - once their meanings reserved understood- actually served to facilitate the resulting discussions and enhanced the ability to readily distinguish a variety of meaningful interrelationships. At this point, with the primary intent of Part I of Cook's text clearly accomplished, it was time to enter the realm of analysis, using specific examples from the vast repertoire available. My own purpose in this review is not to agree or disagree with the specific application of Cook's model to the analyses presented in Part II. Other reviewers in the present volume will take the opportunity to do so. I do, however, wish to take issue in a very general way with the examples selected by Cook for the purpose of demonstrating the appropriateness and functionality of his models. As a musicologist, Cook has chosen explicitly to focus on multimedia examples in which the music plays a primary role (i.e., "musical multimedia"). As a result, each of the three selected examples (the video for Madonna's "MaterialGirl," the "Rite of Spring sequence" from Fantasia, and "Armide" from Arid) represents a multimedia context in which the music predates the accompanying visual component and dominates the multi-modal texture. Quite the opposite of instances in which the visual image is autonomous (a situation dealt with by the author at length in Part I of the text), the chosen excerpts focus solely on a relationship at the opposite extreme of the spectrum, rather than providing a variety of media types and representativeinterrelationships. In cinema, arguablythe most sociologically significant form of multimedia at present- and, admittedly, the present author's primary area of research interest- the sequence of events involved in production is quite the opposite. Typically, though exceptions to this rule certainly exist, the film composer is given a finished product for which she is asked- within a phenomenally short period of time- to produce a musical score for the purpose of enhancing the dramatic narrative. It would seem appropriate to have included at least one excerpt from a feature film in the set of examples for analysis, given the significance of this artform as evidenced by box office receipts. This is not intended to denigrate the selections of the three very interesting pieces analyzed, each useful in its own right and quite 234 Integral different one from another. I question only whether- other than the music video- they represent types of multimedia that are exemplary to the extent that the analytical method applied to them can be shown to be appropriate for other similar examples of multimedia that occupy a position of high sociological significance within our culture. I wonder if the selection of such materials doesn't run counter to the author's stated intent to "contribute to the current reformulation of music theory in a manner that loosens the grip on it of the ideology of musical autonomy" (vi). Selecting these specific types of multimedia, intentionally or unintentionally, raises the musical component to the position of most significant feature, upon which all others are based and/or to which they relate specifically. Though perhaps no longer "musically autonomous," in the sense meant by Peter Kivy (according to Cook's own reference), these chosen works represent- at best- music-centric multimedia examples. To what extent does a model formulated for the analysis of such specialized examples generalize to multimedia artifacts in which the roles of individual components share more equally in the emergent meaning of the piece? Despite the minor critiques offered in these paragraphs, I found AnalysingMusical Multimedia to be a highly informative and stimulating read. The clarity with which Cook expresses his wellinformed ideas is exemplary, as is the manner in which he introduces formative concepts that support the basis for his proposed model of analysis. Though this text does not provide the definitive guide- Cook certainly does not presume to make this claim- to understanding or analyzing multimedia, it certainly takes admirable strides in that direction. I found that the book served my educational objectives extremely well in the context of the previously referenced "Multimedia Cognition" course. It provided an interesting and highly useful counterpoint to Chion's Audiovision and the additional selected readings intended to augment understanding of aesthetics in general and inform students regarding the results of empirical research investigating the multimedia context specifically. Students responded well to the manner in which the material was introduced and developed; this communicated to me that Cook's proposed model facilitated their understanding of the interrelationships between various media and Review Forum: Lipscomb on Cook 235 enhanced their ability to communicate about these matters clearly and concisely. As a music researcher, I find that my primary remaining concern with the text echoes that previously stated by a colleague and friend. In her review of the same text, Annabel Cohen identifies the author's "unwillingness to endorse the cognitive psychology experimental approach."15She goes on to state that many of the ideas presented in the text afford a perfect opportunity to be tested experimentally, specifically mentioning issues related to conscious attention, cross-modal figure-ground relationships, the effect of music on perceived synchrony, the effect of synchrony on awareness of the music, and the effect of music on the perceived quality of activity. Many of these topics have alreadybeen broached in empirical work investigating the multimedia experience. In agreement, I would argue that experimental researchin general and the cognitive approach specifically offer the perfect tools with which to further revise and develop Cook's set of models. Looking to the future, I see Cook's text as a musicological "statement"made to the interdisciplinary academic community at large to which the community of music cognition researchers can respond with an appropriate "answer." If I had but one wish, I would ask that this scholarly "dance"might proceed through numerous iterations, in a way that will afford an opportunity for dialogue and discussion across extant disciplinary boundaries, bringing us closer to an understanding of the processes inherent in the multimedia experience through the systematic investigation of the intriguing relationships proposed by Cook, supplemented by researchalready carried out, and clarified by researchyet to come. After reading this text and formulating its many testable hypotheses, a research agenda could be set that would occupy the next two decades... at least. I hope Cook's text and others like it will stimulate others to join in the search. 15Cohen 1999:258. 236 Integral References Campbell, W. and Heller, J. 1980. MAnOrientation for Considering Models of Musical Behavior." In Handbook of Music Psychology.Edited by D. Hodges, 29-35. Lawrence, KS: National Association for Music Therapy. Chion, M. 1990. Audio-vision: Sound on Screen. Translated by C. Gorbman. New York: Columbia University Press. Cohen, A.J. 1999. "Musicology Alone?" (a review of Nicholas Cook's Analysing Musical Multimedia). Music Perception 17:247-260. Cohen, A.J. 2001. "Music as a Source of Emotion in Film." In Music and Emotion: Theory and Research.Edited by P.N. Juslin and JA. Sloboda, 249272. Oxford, UK: Oxford University Press. Kendall, R.A. and Carterette, E.C. 1990. "The Communication of Musical Expression." Music Perception8/2: 129-164. Kerman, Joseph. 1956. Opera as Drama, 1* ed. New York: Knopf. Lakoff, G. and Johnson, M. 1980. Metaphors We Live By. Chicago, IL: University of Chicago Press. Longer, S.K. 1942. Philosophy in a New Key:A Study in the Symbolismof Reason, Rite, and Art. Cambridge, MA: Harvard University Press. Lipscomb, S.D. in press. "The Perception of Audio- visual Composites: Accent Structure Alignment of Simple Stimuli." Selected Reportsin Ethnomusicology 12. Lipscomb, S.D. 1995. "Cognition of Musical and Visual Accent Structure Alignment in Film and Animation." Unpublished doctoral dissertation, University of California, Los Angeles. Lipscomb, S.D. and Kendall, R.A. 1994. "Perceptual Judgment of the Relationship Between Musical and Visual Components in Film." Psychomusicology13/1: 60-98. Marks, L.E. 1978. The Unity of the Senses: Interrelations Among the Modalities. New York: Academic Press. Marshall, S.K. and Cohen, A.J. 1988. "Effects of Musical Soundtracks on Attitudes Toward Animated Geometric Figures."Music Perception6: 95-1 12. McGurk, H. and MacDonald, J. 1976. Nature 264: 746-748. "Hearing Lips and Seeing Voices." Osgood, C.E., Suci, G.J., and Tannenbaum, P.H. 1957. The Measurement of Meaning. Urbana: University of Illinois Press. Radeau, M. and Bertelson, P. 1974. "The After-effects of Ventriloquism." QuarterlyJournal of ExperimentalPsychology26: 63-7 1. Staal, H.E. and Donderi, D.C. 1983. "The Effect of Sound on Visual Apparent Movement." AmericanJournal of Psychology96: 95- 105. Tannenbaum, P.H. 1956. "Music Background in the Judgment of Stage and Television Drama." Audio-Visual CommunicationsReview 4: 92-101. Thaycr, J.F. and Levenson, R.W. 1984. "Effects of Music on Psychophysiological 3: 44-54. Responses to a Stressful Film." Psychomusicology Fantasia's Rite of Spring as Multimedia: A Critique of Nicholas Cook's Analysis* Daphne Leong Nicholas Cook's AnalysingMusical Multimedia undertakes the lofty task of creating "a generalized theoretical framework for the analysis of multimedia" (v), one that moves from music to other media (vi). The first half of the book sets out this theoretical framework followed, in the second half, by three analytical case studies: Madonna's "Material Girl," the Rite of Spring sequence from Fantasia, and Godard's sequence in Aria based on Lully's Armide. Cook states that his case studies "do not illustrate the analytical approach in any literal way, but rather attempt to embed its results within the context of broader critical readings" (ix-x). One can therefore read his case studies as a presentation of theory, analysis, and criticism. The present essay focuses on Cook's treatmenttheoretic, analytic, and critical- of Fantasia's Rite of Spring sequence. The sequence exemplifies what Cook calls "music film": "a genre which begins with music, but in which the relationships between sound and image are not fixed and immutable but variable and contextual, and in which dominance is only one of a range of possibilities" (214). Cook further proposes viewing the Rite of Spring sequence "as the construction of a fundamentally new experience, one whose limits are set not by Stravinsky nor even by but by anybody who watches - and listens Disney..., to- 'Fantasia'" (214). This proposition, as fleshed out by Cook in the chapter, implies among other things that (a) Fantasia's Rite is not defined by the music from which it originated, (b) the combination of visuals and music in Fantasia's Rite creates a new entity, and (c) this new entity is worthy of the analytical attention that Cook devotes to it. " Nicholas Cook, Disney *s Dream: The Rite of Spring Sequence from 'Fantasia'," Chapter 5 in Analysing Musical Multimedia (Oxford: Oxford University Press, 2000 reprint of 1998 edition): 174-214. 238 Intigral I will qualify all three claims and, along the way, question certain of Cook's analytical observationsand methodologies. The claim that the Rite of Spring sequence provides a "fundamentally new experience" derives from a central thesis of Cook's book. Simply put, meaning emergesfrom the combination of disparate media (115). Attributes are transferred from one medium to another; the resulting combination is qualitatively different from its constituents (84). l In the case of Fantasia's Ritey Cook counters the assumption that the music is primary, and the visuals merely "the projection through ancillary media of an originary meaning" (214). Ironically, his argument rests in part on similarities between Fantasia and the original ballet production of The Rite of Spring. Pointing to the close collaboration between Nikolai Roerich, Stravinsky, and Vaslav Nijinsky during the ballet's genesis, and to the tight interlacing of ethnography, music, and dance during Stravinsky's compositional process, Cook argues for the intrinsic multimedia nature of Rite, and, by implication, for the validity of its new multimedia incarnation in Fantasia (198ff.).2 He defends "Disney's realization of the music as the story of life" as "an alternative metaphor to that of the pagan celebration of spring" (206), and compares Disney's visuals to Stravinsky'schoreographic annotations.3 I find Cook's placement of Fantasia's Rite of Spring in the context of the original Stravinsky-Nijinsky choreography particularly his comparison of Rites rhythmic structure, Stravinsky's choreographic annotations, and Fantasia's visualization of the score on several levels- fascinating, and one of the strongest contributions of his analysis. Yet Cook's contention that Fantasia's Rite of Spring sequence constitutes a new entity fails to persuade. Indeed, negative musiccritical reaction to the film focused on the lopsidededness of its Lawrence Zbikowski's review-essay in this volume describes the theory of "conceptual blending," which explores and formalizes this notion of emergent concepts. Others have presented evidence for the intertwining of ethnography, music, and dance in the creation of Rite. See, for example, Taruskin 1984, also Taruskin 1995 and Pasler 1986. As transcribed in Stravinsky 1969. Review Forum: Leong on Cook 239 music and animation, on the failure of its two media to cohere. According to critic Olin Downes, "Stravinsky's'Sacre' is a piece of music almost as difficult as that of Bach to visualize in any way that corresponds to the inherent quality of the score;" this in contrast to Fokine's choreography of Rimsky-Korsakov's Schihe*razade> which "magnificently companioned the music. The point here is that while the music was not slavishly followed, it was represented essentially by a companion creation of a parallel character which completed and did not belie the nature of the score."4 The notion of image and sound as a united whole has a long history in film theory. Rudolf Arnheim, for example, exploring questions of multimedia in 1938, emphasized both difference at a surface and unity on a deeper level: "...elements conform to each other in such a way as to create the unity of the whole, but their separatenessremains evident." "...a combination of media that has no unity will appear intolerable."5 A history of this "need for unity, totality, continuity, fusion of some form among disparate elements" is traced by Scott Paulin, who states that "in addition to the need to create the impression of internal unity within both the imagetrack and the soundtrack separately, the two tracks must also cohere so as to invite perception as a unified whole. Sound and image must bear some relation of appropriateness or 'realness' to each other...."6 Fantasia's Rite of Spring sequence presents a wide chasm between music and image. Disney's animation and Stravinsky's score (or Stokowski's soundtrack) simply do not match well in terms of artistic quality and depth. Their juxtaposition creates, not a "fundamentally new experience," as Cook would have it, but an uneasy amalgamation of cartoon and sound. For those acquainted with Stravinsky's score, at least, the music looms large over the supposedly new creation that is Fantasia. A slight aside is in order here about the music in Fantasia's Rite. The soundtrack of Fantasia's Rite departs from Stravinsky's intentions in several obvious respects. As Cook states, Disney's team cuts sections of the score, reorders it, and reorchestratesit in 4 Downes 1940. 5 Arnheim 1957: 207, 201. 6 Paulin2000: 63-64. 240 Integral parts.7 Conductor Leopold Stokowski and the Philadelphia OrchestradepartsignificantlyfromStravinsky's tempoindications. I now turn to some detailsof Cook's analysis. My discussion focuses on Cook's applicationof three theoreticalconcepts: the associationof accentuationwith coincidence of visual and aural "cuts,"Pietervan den Toorn's Type I and II rhythmicstructures, and Andrew Imbrie's "conservative"and "radical"hearings of metricstructure. (1) Accentuationand coincidenceof visualandauralcuts Cook's Figure 5.2 diagramsthe Rite of Springsequencefrom 104 to 117, showing blocks of musicalmaterial,largergroupsof these blocks, and placeswherevisualcuts align with beginningsof musicalblocks. u...[T]here is a contrastbetweenthe groupsthat are characterized by cuts at the beginningsof blocks(I and III) and those that are not (II and IV); these coincidencesestablishwhat might be termedaudio-visualdownbeats.The resultis that groups I and III create the effect of being accented as comparedwith groups II and IV, giving rise to a kind of large-scaledownbeatafterbeatpattern;groupII constitutesa kind of prolongedafterbeat followingon the initialgroupI, while the moreextendedgroupIV follows on from the composite downbeatformed by the second groupI and groupIII"(184). This analysismissesseveralimportantfeatures. Cook's group IV segments into two parts,which I will call IVa (114) and IVb (115 to end of 116); the beginning of group IVb is clearly articulatedby a changein orchestrationat 115. Cook ignoresthis change, implicitlygroupingall of 114 to the first measureof 115 Cook suggests a rationale for Disney's reordering of the score (176-177); it is interesting to note that the reordering also parallels the palindromic plan of the animation. Disney orders parts of Stravinsky's score as follows: Introduction to Part I- first part of Part I ("Augurs of Spring," "Ritual of Abduction")- most of Part II (all but the "Sacrificial Dance") - last part of Part I ("Kiss and Dancing Out of the Earth")- Introduction to Part I, corresponding roughly to the visual plan space- earth- life- earth- space. (Cf. Cook Figure 5.5.) Both plans are palindromic. Review Forum: Leong on Cook 241 together, as "the six successive appearances of block V from 114, ...enlivened by means of a relatively autonomous visual structure" (184). The visual structure is actually not particularlyautonomous; it aligns with Stravinsky's alternating 5/4 - 6/4 bars (Cook's alternating el 0-1 2). As shown in my version (Example lb) of the end of Cook's Figure 5.2 (Example la), a visual cut that Cook misses at 114+2 completes a clear pattern of visual cutting at the beginning of every 5/4 measure (Cook's elO's). Examplela. Cook'sFig. 5.2 "AnalyticalOverviewof the Fight Sequence"(last line) (183). 114 115 116 IV 103 I I |elO 12 10 12 10 |1O |f6 el2 f4 e6 f6 g5 Examplelb. Revisionof Cook'sFig. 5.2 (last line). 114 115 116 IVa IVb I II |elO 12 |1O t 12 |10 I 10 |f6 el2 jf4 t e6 |f6 g5 t Furthermore, a closer look at group IVb shows that important visual changes occur at the beginning of every f block. At 115+1, as Cook notes, a visual cut marks the beginning of f6; but at 115+3 and 116+1, unmentioned by Cook, significant visual changes (Stegosaurus tail hits Tyrannosaurus, and Tyrannosaurus bites Stegosaurus and holds, notated in Example lb with dotted vertical lines) coincide with the beginnings of blocks f4 and f6 respectively. 242 Integral The latter two 'changes of scene/ while not cuts per se, accomplish functions analogous to cuts. By marking the beginning of each f block, they create the effect of an alternation of upbeats (e blocks) and downbeats (f blocks). Thus, even accepting Cook's assertion that coincidence of visual and aural cuts results in "accented" groups,8 a more consistent reading based on this criterion would be: I (accented) II (unaccented) - I and III (accented) - IVa (accented) - IVb (somewhat accented?). Cook, however, wishes to divide the passage into two parallel sections I-II and I/HI-IV; interpreting IV as unaccented supports the parallelism of his reading. Other features contribute to a sense of group IVa as unstressed- it has a lower dynamic level and lighter orchestration than preceding material- but any unstressed quality is not due to lack of coincidence between visual and auralcuts. (2) Pieter van den Toorn's Type I and II rhythmic structures In his monograph on the Rite of Spring, Pieter van den Toorn proposes two prototypical types of rhythmic structure. Type I consists of irregular or shifting meter, with alternations of contrasting material delimited by bar lines; all concurrent instrumental parts synchronize metrically. Type II displays foreground metric regularity (usually a steady meter), but superimposes two or more repeating motives whose periodicities differ from one another.9 Cook terms the opening of "Augursof Spring" from 13 to 22 Type I, arguing that, rather than Stravinsky'snotation in 2/4 meter with cross-metric accents, the passage could well be notated in changing meters (presumably 9/8 - 2/8 - 6/8 - 3/8 - ... ) (187188). Several objections can be raised to this interpretation. I would prefer a different term, since accentuation more precisely occurs at points in time, not over extended passages. Imbrie 1973: 52-54; Benjamin 1984: 379; and Lester 1986: 16 attribute accent to time points, rather than time-spans. Schachter 1987: 6; Hasty 1997: 16-17, 103-104; and Berry 1985: 30 attribute accent to events, but describe it as focused on a particular point within an accented event. Imbrie 1973: 54 suggests the term "weighting" for cases in which "an important downbeat accent ... impart[s] a generalized sense of greater heaviness to an entire rhythmic unit." See Leong 2000: 39 for further discussion of this issue. 9 Van den Toorn 1987: 97-1 14; see especially 99-100. Review Forum: Leong on Cook 243 Elsewhere, Stravinsky clearly emphasizes the significance of his notated meters, and the difference between his bar lines and accent markings;10 renotating these cross-accentsas changing meters alters their sensibility drastically. Furthermore, preceding, interspersed, and overlying material (12+8 to 12+9, 14 to 14+3, and 15+1 to u And 15+5 respectively) unambiguously articulates 2/4. subsequent material, at 16, displays characteristics of Type II structure: 2/4 in English horn and viola plays against an offset 2/4 in oboe and 3/8 in the lower strings. When a metric structure similar to that at 16 appears at 28, Cook calls it a "textbook example" of Type II (189). Thus Cook's labeling of the passage from 13-21 as Type I is questionable at best. (3) Andrew Imbrie's "conservative"versus "radical"hearings Cook's appropriation of Andrew Imbrie's terms "conservative" and "radical"suffers throughout from subtle misinterpretations of their meaning. According to Imbrie, a conservative listener maintains the established meter for as long as possible in the face of conflicting evidence; a radical listener "converts" quickly when presented with evidence of changing meter.12 According to Imbrie via van den Toorn via Cook, "'radical'readings [are] based purely on surface rhythms," while "'conservative' readings [emphasize] underlying metrical continuity" (187). The main difference between radical and conservative listeners, however, lies not in their compliance with surface rhythm versus metrical structureper se>but in the speed with which they adapt to the changing metric implications of surface rhythms. Two examples follow to illustrate the misunderstanding. The first continues our discussion of the passage from 13 to 22. In Stravinsky andCraft1959:21, Craftasks,"Canthesameeffectbe achieved by means of accents as by varying the meters? What arc bar lines?*, to which Stravinsky replies, "To the first question my answer is, up to a point, yes, but that point is the degree of real regularity in the music. The bar line is much, much more than a mere accent, and I don't believe that it can be simulated by an accent, at least not in my music." Van dan Toorn 1987: 69-70 describes 13 in terms of the continuation of the previously established 2/4 meter, its disorientation, and its reestablishment at 14. Imbrie 1973; see especially 65. The concept has been adopted by van den Toorn (1987: 67 ff.); and Lcrdahl and Jackendoff ( 1996: 22-25); among others. 244 Integral Beginning at 18, Stravinsky reinterprets the material beginning at 13: the eight measures of 18 repeat those of 13, but unlike the material at 13, which is followed by straightforward 2/4 meter at 14, that at 18 leads to 2/4 meter shifted by one eighth note at 19. Beginning at 19, the accented melodic entrances on the second eighth note of the measure, combined with the accompaniment accents and dynamic changes at the same metric position, make it extremely difficult for the listener to maintain the previouslyestablished notated 2/4 meter. Even the most "conservative" of listeners would be hard put to avoid converting to the shifted 2/4 meter articulatedby so many musical cues. Disney's visual cuts make the same shift. From 13 to the end of 18, the primary visual cuts, as Cook notes, coincide with the downbeat of each new block. (I prefer to say that the cuts align with the beginnings of musical blocks, which happen to occur on downbeats.) At 19, the musical blocks shift, at least aurally, to begin on the second eighth of the measure. Here Disney follows the aural cues, aligning visual cuts with the aural cuts on the second eighth of the measure. Cook calls the animation of this passage "predominantly 'radical'" (188), because of its alignment with surface patterning. This description is incorrect, because a radical reading implies shifting sooner rather than later, shifting on the basis of lesser rather than greater evidence. Disney's visual cuts shift to the second eighth note only when the musical evidence makes it difficult to avoid shifting; they follow the preponderance of musical cues, ratherthan anticipating them. Cook's analysis of the latter part of "Augurs of Spring," from 28 to its end, displays a similar misunderstanding of Imbrie's terms. Essentially, Cook argues that visual cutting rhythms serve first to reinforce four-bar periodicity, and then to play against it. His Figure 5.4 shows cutting rhythms in the passage, and, as far as I can tell, is inaccurate. Examples 2a and 2b show Cook's Figure 5.4 and my revision of it. The inaccuracies do not alter Cook's argument much, except that the "cut on the hyperdownbeat at 31+4" actually falls an eighth after the hyperdownbeat, coinciding with the syncopated horn entrance, and there is an additional coincidence of visual cut and hypermetric downbeat, at 32+4. At 31+4 the cuts do start out Review Forum: Leong on Cook 245 supporting the hypermeter (albeit one eighth late at the beginning), and then begin cutting against the music's four-measureunits. But one cannot, as Cook does, describe this process as a conservative hearing migrating to a radical one. A radicalhearing implies a shift in listening stance commensurate with changes in musical surface, from an established meter to a new meter, or from an established meter to changing motivic metric identities. After 32+4 the cuts do not follow musical metric, grouping, or motivic structure; they create their own largely independent rhythm. They cannot be interpretedas a visualization of a radicalhearing of the music. Example2a. Cook'sFig. 5*4 "CuttingRhythmsin The Augursof Spring'"(190). 28 29+1 30 31+4 8 8 15 15 14 6 9 33+2 6 7 35+2 36+2 6 7 5 3 4 3 Example2b. Revisionof Cook'sFig. 5*4. 28 29+1 30 88 31 15 14 6 31+4 32+4 li£ 2£ 34+2 6651 7 7 3 3 In his description of methodology for the analysis of multimedia (133-146), Cook suggests experiencing each medium on its own, and comparing this effect to the medium's effect in the totality, or in pairs of media; when interpreting media pairs, he proposes inverting relations (reading from one to the other), reading for gaps, and using "distributionalanalysis." In this list of methodologies, Cook, though he mentions music-specific analytical methods such as Schenkerian analysis, makes no mention of visualspecific tools. This may be one result of his orientation from "music-to-other-media," his attempt to "extend the boundaries of 246 Integral " music theory to encompass... words and moving images... (vi). Nevertheless, a music theorist would be rightly dubious of a film theorist making analytical claims about precise alignments of musical metrical structure and visual cuts without recourse to the musical score; and Cook, though he makes frequent reference to the score of Rite, makes no reference to frame-by-frameanalysis or to other visual tools beyond simple viewing. Furthermore, in his discussion of visual rhythmic structure, Cook relies heavily upon cutting rhythms. He makes little mention of rhythms articulated by analogous changes in visual content.13 His discussion of the Tyrannosaurus/Stegosaurus fight scene, for example, as mentioned earlier, overlooks the rhythmic articulation created by visual changes (not cuts) in group IVb. > The Rite of Spring sequence features in Cook's Analysing Musical Multimedia as an illustration of his theory of multimedia. The connections between the theory and Cook's analysis are rather broadly drawn; some readers might wish to see more rigorous connections between Cook's explication of conformance, complementation, and contest, and his analytical chapter on Fantasia}* The Rite of Spring sequence does illustrate contest,on a deeper level than that envisioned by Cook. The two media involved Disney's animation and Stokowski's performance of Rite I am speaking of changes akin to cuts, without actual filmic cutting. Cook does discuss alignments of surface musical activity with visual gestures such as shooting stars, volcanic puffs of flame, and Tyrannosaurus snaps, as well as with kinesthetic motions such as swooping pterodactyls and "jogging dinosaurs" (182). Furthermore, I am referring primarily to intermediate levels of rhythmic structure; Cook does explore large-scale form created by the chronological narrative and its symmetrical plan space-earth-life-earth-space, by color associations, and by camera or diegetic motion (193-196). See Cook (98-106) for his presentation of these terms. In a nutshell, conformancerefers to media consistent with one another, without "differential elaboration;" complementationentails media similar to one another, yet differing in significant ways; contest "implies an element of collision or confrontation between the opposed terms" (102). Review Forum: Leong on Cook 247 of Spring- juxtapose "popular" art with "elite" art. The two contradictory views come to the fore in statements by Disney and Stravinskyrespectively: Stravinsky saw his Rite of Spring and said that that was what he had in mind all the time. None of that matters, I guess. This isn't a picture just for music lovers. People have to like it. They have to be entertained. We're selling entertainment and that's the thing I'm hoping Fantasia does- entertain. When Walt Disney used Le Sucredu printemps for Fantasia, he told me: "Think of the numbers of people who will now be able to hear your music." Well, the •numbers of people who consume music... is of no interest to me. The mass adds nothing to art. It cannot raise the level, and the artist who aims consciously at mass appeal can do so only by lowering his own level. The soul of each individual who listens to music is important to me, not the mass feeling of a group. Music cannot be helped by means of an increase of the quantity of listeners, be this increase affected by the film or any other medium. It can be helped only through an increase in the quality of listening, the quality of the individual soul. What is at issue, as emerges from Stravinsky'sstatement, is not so much "popular"versus "elite," as it is "the level" and "quality"of the artisticvision, its execution, and its reception. Cook bases his models of multimedia on George Lakoff and Mark Johnson's model of metaphor.17The defining feature is "a distinctive combination of similarity and difference" (98). Cook assumes a basic level of similarity which, if one is to follow his basis in metaphor theory, means that the constituent expressions must be close enough to "fit together," to form a metaphor. For some readers, the metaphoric link, so to speak, between Disney's animation and Stravinsky's score or Stokowski's performance may be quite tenuous. In his development of the concept of contest,Cook writes that "each medium strives to deconstruct the other, and so create space for itself. Any IMM [instance of multimedia]18in which... one or more of the constituent media has its own closure and autonomy is likely to be characterized by contest; IMMs that involve the 15 Walt Disney, quoted in Culhanc 1983: 29. 16 Stravinsky 1946: 35-36. 17 Lakoff and Johnson 1980. Cook abbreviates his term "instance of multimedia" as IMM (100). 248 Integral addition of a new medium to an existing production are a particularlyrich source of examples"(103). This description would seem to be aproposof Fantasia'sRite of Spring sequence. But Cook argues that the sequence's overall relationship of visuals and sound is one of conformance. He describes the close synchronization of image and music on the film's surface, the 'contrapuntal' relationships of cutting rhythms and hypermetrical patterns at intermediate levels, and the creation of "a single filmic gesture... which reaches from the opening space sequence right up to the dawn of life" (196) on a large scale. "The result of all this is that music and visualization stack up into a single hierarchy whose highest level is visual. And in this way, what might be called the background model of the Rite sequence from 'Fantasia'is an unambiguous conformance''(208). Elsewhere Cook shows how conformant relations at more surface levels can contribute to conflicting relations at a deeper level (181-182). Although Fantasia'sRite of Springsequence can be seen as conformant on its surface and even deeper levels, the combination of Disney's particular choice of animation with Stokowski's performance is ultimately conflicted, in terms of aesthetic caliber. Stravinsky describes the problem of surface conformance versus deeper compatibility as follows: The danger in the visualization of music on the screen- and a very real danger it is- is that the film has always tried to "describe" the music. That is absurd. When Balanchine did a choreography to my "Danses Concertantcs" (originally written as a piece of concert music) he approached the problem architecturallyand not descriptively. And his success was extraordinaryfor one great reason: he went to the roots of the musical form, of the jeu musical, and recreated it in forms of movements. Only if the films should ever adopt an attitude of this kind is it possible that a satisfying and interesting art form would result.19 And later in the same interview: ...my ideal is the chemical reaction, where a new entity, a third body, results from uniting two different but equally important elements, music and drama; it is not the chemical mixture where. . .nothing either new or creative [results]."20 19 Stravinsky 1946: 35. Ibid. 20 TL'j Review Forum: Leong on Cook 249 The Rite of Spring sequence in Fantasia forms not a "chemical reaction," but a "chemical mixture." For although, as Cook argues, Fantasia may be a "fundamentally new experience" that constructs new meaning through the combination of its constituent parts, it remains a combination and not a coherent whole; it remains aesthetically unsatisfying. And so, Cook's analysis, in its choice of IMM for analysis, must also be unsatisfying, in analyzing an IMM that conforms on many levels yet fails to cohere on the deepest, aesthetic level.21 References Arnhcim, Rudolf. 1957. "A New Laocoon: Artistic Composites and the Talking Film" (original in Italian in Bianco e Nero 1938). Translated in Arnhcim, Film as Art. Berkeley: University of California Press. Benjamin, William. 1984. "A Theory of Musical Meter." Music Perception 1/4: 355-413. Berry, Wallace. 1985. "Metric and Rhythmic Articulation in Music." Music TheorySpectrum7: 7-33. Culhane, John. 1983. Walt Disney'sFantasia. New York: Harry N. Abrams. Downes, Olin. 1940. "Disney's Experiment: Second Thoughts on 'Fantasia' and Its Visualization of Music." New YorkTimest 17 November. Hasty, Christopher. 1997. Meter as Rhythm. Oxford: Oxford University Press. Imbrie, Andrew. 1973. "'Extra' Measures and Metrical Ambiguity in Beethoven." In BeethovenStudies 1, ed. Alan Tyson. New York: Norton. Lakoff, George and Mark Johnson. 1980. Metaphors We Live By. Chicago: University of Chicago Press. Leong, Daphne. 2000. "A Theory of Time-Spaces For the Analysis of TwentiethCentury Music." Ph.D. diss., Eastman School of Music, University of Rochester. Cook explains in his Preface (x) that he chose case studies largely on the basis of general availability. Another fairly obtainable "music film," and one that, I think, would provide a much more satisfying "chemical reaction" for analysis, is Chuck Jones' What's Opera, Doc? Like Fantasias Rite, this short takes a "classical"and originally multimedia work, Wagner's Ring, as a point of departure. Unlike Fantasia, it makes no claim to fidelity to the original, but freely snips, arranges, and adds sound effects, voices, and lyrics. However, the quality of the animation and the creative vision of the director in this example, unlike in Fantasia's Rite, result in a truly new entity of image and sound. 250 Integral Lcrdahl, Fred and Ray Jackcndoff. 1996 reprint of 1983. A Generative Theoryof Tonal Music. Cambridge: MIT Press. Lester, Joel. 1986. The Rhythms of Tonal Music. Carbondale: Southern Illinois University Press. Pasler, Jann. 1986. "Music and Spectacle in Pctrushka and The Rite of Spring." In Confronting Stravinsky: Man, Musician, and Modernist, ed. Pasler. Berkeley. University of California Press. Paulin, Scott. 2000. "Richard Wagner and the Fantasy of Cinematic Unity: The Idea of the Gesamtkunstwerkin the History and Theory of Film Music." In Music and Cinema, ed. James Buhler, Caryl Flinn, and David Neumeyer. London: Wcsleyan University Press. Schachter, Carl. 1987. "Rhythm and Linear Analysis: Aspects of Meter." Music Forum VI: 1-60. Stravinsky, Igor. 1946. "Igor Stravinsky on Film Music, as told to Ingolf Dahl." Musical Digest 28/ 1 (September): 4-5, 35-36. . 1969. "The Stravinsky-Nijinsky Choreography." Appendix III in The Rite of Spring:Sketches1911-1913. [London]: Boosey and Hawkes. Stravinsky, Igor and Robert Craft. 1959. Conversations with Igor Stravinsky. Garden City, New York: Doubleday. Taruskin, Richard. 1984. "The Rite Revisited: The Idea and the Sources of its Scenario." In Music and Civilization: Essaysin Honor of Paul Henry Lang, ed. Edmond Strainchamps and Maria Rika Maniates. New York: W.W. Norton. . 1995. "A Myth of the Twentieth Century: The Rite of Spring, the Tradition of the New, and 'The Music Itself." Modernism/Modernity111: 126. van den Toorn, Pieter. 1987. Stravinskyand The Rite of Spring: The Beginning of a Musical Language. Berkeley, University of California Press. Music Theory, Multimedia, and the Construction of Meaning Lawrence M. Zbikowski In the summer of 1938, as the storm clouds of war were gathering across Europe, Sir Donald Tovey delivered a lecture to the British Academy entitled "The Main Stream of Music." The lecture is a curious affair, not the least because for Tovey the mainstream of music was a thoroughly Germanic one. While sensitive to the accomplishments of non-German composers in the sixteenth and seventeenth centuries, Tovey nonetheless believed there was a sea change in musical composition in the early eighteenth century: "With the advent of Bach, music became an art so congenial to all that is best in the Teutonic intellect that for the next two centuries there is no musical art-form in which German musicians have not produced the supreme masterpieces."1 And it was the genius of Bach that discovered resources within music which rendered the medium independent of other media. Tovey continues, "There can be no supreme musical art without the qualities of absolute music, whether the art be as compounded with other arts as Wagnerian opera or as exclusively musical as the string quartets of Beethoven." The mainstream of music, then, was one flooded by the works of German composers, works whose excellence relied on the purely musical. This conclusion caused Tovey some anxiety. Indeed, both his long-held belief that music could speak to a broad audience, and his tireless championing of British music, were challenged by a central corpus of thoroughly German works that required neither text nor program for their understanding. But a deeper source of his anxiety was a nagging suspicion that musicians were in danger of losing their way. Some pages later, after having drawn his survey to a close with a brief contemplation of Wagner's enormous operas, 1 Tovey1938:128. 252 Integral he writes aI can go no further. At the present day all musicians feel more or less at sea, and not all of us are good sailors."2 Sixty-five years later one can only look with envy on the navigation problem that confronted Tovey, for his mainstream is now regardedby most as but a tributary, if a significant one, to the vast body of music through which scholars must find their way. This challenge to navigation is, in less metaphorical terms, a challenge to musical analysis, for analysis is one of the fundamental ways musicians chart their course through challenging or unfamiliar repertoire. And one seldom finds a repertoirethat presents as many challenging or unfamiliar problems as does musical multimedia, for the various ways music can combine with words or images yield phenomena that are often beyond the reach of our usual analytical tools. Indeed, as Nicholas Cook argues in Analysing Musical Multimedia, confronting multimedia opens up basic issues within the theory and analysis of music, and suggests a thorough reevaluation of the entire enterprise. As Cook notes, "What begins as an analysis of musical multimedia, then, turns ineluctably into an analysisof analysis"(viii). The analysis of analysis begins not with the somewhat shopworn questions of what counts as analysis and why one should do it, but with the issue of musical meaning, for the assumption that music means somethingis basic to musical multimedia. This is not to say that musical meaning is theorized in any profound way by those who create musical multimedia, only that these practitioners realize that a television commercial or a film means something quite different when the music is taken away or substantially altered. Thus, while music often occupies a place well below the obvious story-line within these media, its contribution is not inconsequential- as Cook observes, "Music transfers its own attributes to the story-line and to the product; it creates coherence, making connections that are not there in the words or pictures; it even engenders meanings of its own" (20). This leads Cook to the somewhat startling conclusion that music in the abstract- Tovey's "absolutemusic"- doesn't have meaning. 2 Tovey1938:139. Review Forum: Zbikowski on Cook 253 What it has, rather, is a potential for the construction or negotiation of meaning in specific contexts. It is a bundle of generic attributes in search of an object. Or it might be described as a structured semantic space, a privileged site for the negotiation of meaning. And if, in the commercials, meaning emerges from the mutual interaction of music, words, and pictures, then, at the same time, it is meaning that forms the common currency among these elements- that makes the negotiation possible, so to speak. (23) Cook goes on to argue that the same holds true for the words and music in songs, and the words about music in analytical prose- in all cases, the meaning that is produced is a consequence of interactions between various media. Musical culture is, in consequence, irreducibly multimedia in nature (23). Analysis must perforce deal not only with the interaction between musical elements but also with the interactions between media, for these interactions are basic to the construction of meaning. The interactions between media that Cook sees as most important are oppositional in nature- what is significant is not how media are like one another, but how they are differentfrom one another. This sense of discrete media that in some way interact is, Cook argues, what separates the experience of multimedia from synaesthesia. At the same time, the most compelling examples of multimedia are not simply the consequence of the coincidence of two discrete forms of communication. What is required is a limited intersection of attributes between the constituent media- what Cook calls an enabling similarity- which allows the media to be brought together into the same conceptual domain so that their differences can be noted and thus made accessible for the construction of meaning.3 This notion of domains that are in some respectssimilar which are brought into a correlation that reveals their differences brings Cook to the theory of metaphor first proposed by George Lakoff and Mark Johnson in the early 1980s.4 In the following, I would like to explore the contemporary theory of metaphor in just a bit more detail than Cook is able to do in Analysing Musical Multimedia. Further developments of this theory offer ways to A similar perspective, developed from research in psychology, can be seen in Gentncr and Markman 1994 and 1997. 4 LakofFand Johnson 1980. 254 Integral streamline a few aspects of Cook's account of multimedia, and extensions to the theory offer a somewhat more systematic approach to the analysis of multimedia in particular, and music in general. The Contemporary Theory of Metaphor Lakoff and Johnson's point of departure was the proposal that metaphor was not simply a manifestation of the figural use of language to create colorful if imprecise images but reflected a basic structure of human understanding.5 For instance, in speaking about a person's romantic relationships we might use expressions such as aHe is known for his many rapid conquests" or "She is besieged by suitors." The linguistic metaphors central to these expressions are based on the conceptual metaphor LOVEIS WAR, which correlates the conceptual domain of romantic love with the conceptual domain of warfare. Once this correlation is active we can access concepts drawn from the domain of warfare ("rapid conquests," "besieged") to characterize aspects of individuals' romantic relationships. More generally, WAR serves as a source domain, providing a rich set of structures that we can map onto the target domain of LOVE.Thus "quickly bringing an enemy to defeat" is used to structure our understanding of a situation in which an individual is able to cause other individuals to direct their affections only to him, and to do so with little effort: "He is known for his many rapid conquests." One question raised by this approach to metaphor was of the ultimate grounding of the process of mapping structure from one domain onto another. Even if we grant that we understand a target domain (such as LOVE)in terms of a source domain (such as WAR), how is it that we understand the source domain in the first place? Mark Johnson answered this question by proposing that meaning was grounded in repeated patterns of bodily experience, which give Expanded versions of the discussion that follows, along with more extensive citations to recent work on metaphor theory, can be found in Zbikowski 1998 and Zbikowski 2002: 65-71. Review Forum: Zbikowski on Cook 255 rise to what he called image schemata.6 An image schema is a dynamic cognitive construct that functions somewhat like the abstract structure of an image and thereby connects together a vast range of different experiences that manifest this same recurring structure. Thus our understanding of a source domain like WARis grounded in image schemata such as BLOCKAGEand COUNTERFORCE; these, together with evaluative judgments such as "winning" and "losing," provide a rich conceptual structurewhich can then be mapped onto domains such as LOVE. Although the theory of image schemata provides a way to explain how cross-domain mapping is grounded, it does not explain why some mappings are more felicitous than others. For instance, we could map structure from the domain of WARonto the domain to produce statements like "The G4 of PITCH RELATIONSHIPS vanquished the FI4." But if we simply want to describe how one pitch relates to another this seems a bit much- we tend to prefer IN PHYSICAL SPACE: mapping from the domain of ORIENTATION "The G4 is higher than the ¥14." To account for why some metaphorical mappings are more effective than others, George Lakoff and Mark Turner proposed that such mappings are not about the impositionof the structure of the source domain on the target domain, but are instead about the establishment of correspondences between the two domains. These correspondences are not haphazard, but instead preserve the image-schematic structure latent in each domain. Lakoff and Turner formalized this perspective with the Invariance Principle, which Turner states as follows: "In metaphoric mapping, for those components of the source and target domains determined to be involved in the mapping, preserve the image-schematic structure of the target, and import as much image-schematic structure from the source as is consistent with that preservation."7 Our mapping of orientation in physical space onto pitch thus relies on correspondences between the image-schematic structure of components of the spatial and acoustical domains. Both space and the frequency spectrum are continua that can be divided into discontinuous elements. In the spatial domain, division of the continuum results in points; in the 6 Johnson1987. 7Turner1990: in original.SeealsoLakoff1990. 254;emphasized 256 Integral acoustic domain, it results in pitches. The mapping thus allows us to import the concrete relationships through which we understand physical space into the domain of music and thereby provide a coherent account of relationships between musical pitches. In contrast, mapping from the domain of WARonto the domain of PITCHRELATIONSHIPS works less well because it does not preserve the image-schematic structure of the target domain (our sense that the frequency spectrum is a continuum is almost completely suppressed) and because it imports structure (notions based on BLOCKAGE and COUNTERFORCE) foreign to the target domain.8 According to the contemporary theory of metaphor, then, metaphor is a basic cognitive capacity that involves mapping structure from one domain onto another. This mapping is possible because there are aspects of the structure of each domain that are invariant- these are the enabling similarities that Cook suggests are a precondition for musical multimedia. Thus, in the case of Schoenberg's Die gliickliche Hand (discussed by Cook on pp. 4156), the "Lighting Crescendo" that occurs in bars 125-53 relies on shared structure between the music, lighting, and action on the stage. As the musical materials get louder and coalesce the lighting gradually goes from dull red through a variety of hues until it becomes a glaring yellow, and the central character moves from a portrayal of exhaustion through stages that lead to a portrayal of extreme tension. The basic structure that unites these three domains relies on the notion of gradually increasing energy. The instantiation of this structure in each domain makes it possible for the media to combine; because the structure is instantiated differently within each domain the result of the combination is w«/tfmedia. An increase in energy such as that portrayed by the actor might well be soundless, but here it is accompanied by a crescendo and the emergence of musical themes from an inchoate background; that same increase in energy might well play out within consistent lighting, but in Schoenberg's conception it begins in murky gloom and ends in the bright light of day. Note, however, that if our concern were tonal relationships as opposed to pitch relationships a mapping from the domain of WAR might be completely appropriate. See Burnham 1995, Chap.l. Review Forum: Zbikowski on Cook 257 The perspective provided by metaphor theory leads Cook, at the conclusion of the first part of his book, to propose three basic models for multimedia. The models are shown on Example 1, which places them along a continuum which focuses on the relative degree of similarity among the constituent media of an instance of multimedia, or IMM. Leftmost on the diagram is the conformance model, distinguished by the large number of similarities that obtain between the constituent media of an IMM. Differences between the media are thus relatively attenuated, and in extreme cases a conformance IMM might be taken as an instance of a single medium. On the right of the diagram is the contest model, in which similarity obtains at only the most abstractlevel. Differences between the media are thus profound, and in extreme cases a contest IMM will simply break apart into its constituent media. These media would still be coincident, but they would not yield an IMM. In between these two extremes is the complementation model: differences between the constituent media of the IMM are significant enough that the media can be readily distinguished from one another, but not so marked that the media seem to contradict each other. Example1. Cook'sthreemodelsof musicalmultimediasituatedalong a similarity-dissimilarity continuum. constituent media ^ similar .constituent media * dissimilar I somewhat different conformance media consistent with each other complementation media contrary,but not contradictory I different very contest media contradictory 258 Integral Cook is thus able to extract a promising analytical approach to musical multimedia from contemporary theories of metaphor. This approach allows him to characterize the conditions that will yield an instance of multimedia, and to develop a typology of such instances based on similarities between their constituent media. However, the approach also raises two problems, both of which stem from limitations of the contemporary theory of metaphor. First, the emphasis in metaphor theory has been on mapping structure from one domain onto another. While describing what music contributes to our understanding of the stage action in Schoenberg's Die gliickliche Hand is an important first step in understanding that particular IMM (since it allows us to specify what structures from the musical domain are mapped onto the domain of the stage action), it does not yield a description of the IMM itself. That is because we also want to know which structures from the domain of the stage action are mapped onto the musical domain. The same holds true for mappings between the domains set up by the music and the lighting, and between the domains set up by the stage action and the lighting. Second (and related), metaphor theory offers no account of the unique conceptual domain that some cross-domain mappings produce. While mappings between the domains of music, stage action, and lighting are important for the process of meaning construction initiated by bars 125-53 of Die gliickliche Handy the unique domain that these mappings produce- the instance of multimedia specific to this moment in Schoenberg's monodrama is what we are really interested in. But it is just this specification of elements and relations proper to the IMM that is lacking in accounts that focus only mapping structure from a source domain onto a target domain. These two limitations of the contemporary theory of metaphor- the difficulty of accounting for coordinate mappings between two or more domains, and acknowledgement that such mappings often yield new conceptual domains- gave rise to the theory of conceptual blending. In the following I outline basic features of this theory, and describe its application to the analysis of musical multimedia. ReviewForum:Zbikowskion Cook 259 ConceptualBlendingand MusicalMultimedia A conceptual blend begins with concepts drawn from two correlated domains. Consider, for instance, Marcel Proust's recollection of one feature of the springtime walks along the way"duringvisitsto Combray: uMe*s£glise We would leave town by the road which ran along the white fence of M. Swarm's park. Before reaching it we would be met on our way by the scent of his lilac-trees, come out to welcome strangers. From amid the fresh little green hearts of their foliage they raised inquisitively over the fence of the park their plumes of white or mauve blossoms, which glowed, even in the shade, with the sunlight in which they had bathed.9 Proust'sdescriptionrelieson conceptsdrawnfrom the domain of trees (including not only concepts associatedwith the scent, foliage, and blossoms of trees, but also with their shape and disposition)and from the domain of intelligentbeings (including concepts associatedwith welcoming strangers,being inquisitive, and bathing).These conceptsare then blendedtogetherto createa domain in which the lilac treesare more than alive- they are also intelligent and animate. Within this domain there are new structures that cannot be found in either of the two original domains. In the blended space, the lilacs send forth their scent, raisetheirfoliageinquisitivelyoverthe fence,andbathein sunlight. In orderto studyconceptualblendssuch as that representedby Proust'sdescription,the rhetoricianMarkTurnerand the linguist Gilles Fauconnierdevelopedthe notion of conceptualintegration networks (CINs).10 Each CIN consists of at least four circumscribedand transitorydomainscalledmentalspaces.Mental spacestemporarilyrecruitstructurefrom more-genericconceptual domainsin responseto immediatecircumstances and areconstantly 9Proust 1981:147-148. I provide an overview of work on conceptual blending, and its application to music, in Zbikowski 2002: 77-95. The most comprehensive study of conceptual blending as of this writing is Fauconnier and Turner 2002. Cook has also made use of blending theory; see Cook 2001. 260 Integral modified as our thought unfolds.11 For instance, Proust's description of the walk along M. Swann's park sets up two correlated mental spaces. The first is that of the lilac trees, the second that of intelligent beings. Features of these two spaces are combined in a third mental space, producing the intelligent and animate trees of Proust's description. Turner and Fauconnier use CINs to formalize the relationships between the mental spaces involved in a conceptual blend, to specify what aspects of the input spaces are imported into the blend, and to describe the emergent structure that results from the process of conceptual blending. The CIN for the conceptual blend summoned by Proust is diagrammed in Example 2. The network involves four interconnected mental spaces, which are shown as circles. Central to the network are two correlated input spaces, the "lilac trees" space and the "intelligent being" space. The solid double-headed arrow linking these two spaces indicates that elements within them serve as structural correlates: tree is correlated with being, giving off scent with animatey and shape of foliage with inquisitive nature. Guiding the process of mapping between these spaces is the generic space, which maps onto each of the input spaces and contains what they have in common: a living being that we come to know on account of certain important properties. Guided by the conceptual framework provided by the generic space, structure from each of the input spaces is projected into the fourth space, called the blend, which yields Proust's characterizationof M. Swann's lilac trees. The mapping is only partial, however, reflecting the limitations imposed by the generic space. Since the generic space is not concerned with incidental properties, such as the means by which living beings take nourishment, the sunlight essential for the trees' continued life is relegated to the role of a simple sensual pleasure within the blend- that is, Proust does not liken the sun in which the trees have bathed to a good meal. The theoryof mentalspacesis developedin Fauconnier 1994;Fauconnier andSweetser 1997;Fauconnier 1996;seealsoTurner1996. 261 Review Forum: Zbikowski on Cook Example2, CINfor Proust'scharacterizationofM. Swann's lilac trees. /f I \ Genericspace \ • livingbeing \ withdistinctive 1 I properties \^^/ Inputspace ^^^ / I I \ ^^*aS?£ ,/^m^ ^^^^^S.^P^SP*06 \ / Lilactrees Intelligentbeing \ • being • tree 1 ^^^^^^^^^^ - animate • givingoffscent ^T^™ I "™""™T^^ • inquisitive • shapeoffoliage nature I I V /^ ^\ / Blendedspace \ 1 characterize!I • Proust's I tionofM.Swann's I I V trees The dashed arrows linking the generic space to the input spaces, and the input spaces to the blended space, indicate the directions in which structure is projected: from the generic space to the input spaces, and from the input spaces to the blended space. The arrows are double-headed because, under certain circumstances, structure may also be projected from the blended space back into the input spaces, and from the input spaces back into the generic space. The double-headed arrows also serve as a reminder of the limitation of all of the diagrams of CINs I shall use: mental spaces are dynamic structures, as are the CINs that are built from them. What Example 2 represents is a sort of analytical snapshot of this particular network, framed with the intent of capturing its essential features, but making no claim to exhausting the possibilities for description. Hints about how the CIN and its 262 Integral spaces may develop can be gleaned from the diagram, but a full account would requirea series of such snapshots. Two important features of the process of conceptual blending are illustrated by this example. First, new structure emerges in the blend. For instance, the inquisitive nature typical of intelligent beings combines with the shape of the lilac trees' foliage to yield trees that reach out toward the visitor. Given the various combinations of concepts we can easily complete the picture Proust has sketched for us, and imagine engaging in a dialogue with the trees. And we can also use the blended concepts to elaborate the story, and imagine that the lilac trees are but one of a number of intelligent plants that populate M. Swarm's park. All of this structure is specific to the blended space, made possible once we have correlatedthe two input spaces.12 Second, blending allows Proust to tell a complex, multidimensional story- one that extends to the sights, sounds, and lived experience of springtime- in a highly compressed version that focuses on a single human-like form.13 Proust's blend takes a season that stretches over weeks if not months, that is manifested in a tremendous outburst of renewal and change, and compresses it into a single encounter with an imaginaryindividual. Evidence suggests that conceptual blending is pervasive in human understanding, and that blends can be much more complex than the one exemplified by Proust's vignette. There is, accordingly, much more to the theory of conceptual blending than what I have sketched here. Nonetheless, only two further aspects of this theory need concern us here: conceptual blends with multiple input spaces, and the technical resources offered by blending theory. Although introductions to conceptual blending often concentrate on the four-space model illustrated in Example 2, conceptual blends typically involve more than two input spaces, yielding CINs with five or more spaces. The situation is well Further discussion of the three operations important to conceptual blending that I have outlined here- composition, completion, and elaboration- can be found in Fauconnicr and Turner 2002: 48-49 and Zbikowski 2002: 77-95. "Story" as I use it here refers to the rich, embodied parabolic structures that Mark Turner has argued are essential to human thought; see Turner 1996. Review Forum: Zbikowski on Cook 263 illustrated by the lighting crescendo from Die gliickliche Hand. Here at least four input spaces are involved: one is set up by the music, another by the lighting, a third by the actions that take place on stage, and the final space by the dramatic story told by the text that is sung.14 The blend that results is, generically, an instance of multimedia, but more specifically it is a compelling portrayalof the moment when a dream turns into a nightmare. The essential features of the spaces proper to the blend are diagrammed in Example 3. The music space is occupied with a gradual assemblage of diverse musical fragments, sounding across a large orchestra, which eventually coalesce into a significant and inexorable force. The lighting space is built around the gradual transition from murky gloom to bright daylight noted above. The stage directions set up a space in which the central character, who moments before had created finished jewels with a single blow of a hammer, now finds himself in the thrall of incomprehensible and terrible powers. And within the dramatic space set up by the text there is a sense of anticipation: the man's "Das kann man einfacher!" ("That can be done more simply!," spoken upon seeing workers laboriously making jewels) and "So schafft man Schmuck!" ("This is the way to make jewels!,"spoken after the hammer blow) have taken him far from the misery he was in at the opening of the drama. Nonetheless, his inability to completely throw off that misery- his clothes are still ragged, and he still seems to wander rootlessly- suggests that this distance cannot be maintained. Elements from each of the four spaces are projected into the blended space, guided by the central image of a tempest (which is how Schoenberg refers to the cataclysm). From the mental space set up by the music are projected the gradual assembly of disparate sonic elements and a dynamic sweep that comprises both an increase in timbral resources and an increase in volume. There is a Above I discuss only three correlated domains within Die gliickliche Hand, in line with Cook's discussion on pp. 41-56; here I have added a fourth. And Schoenberg at least suggests that there might be additional spaces: he calls for a wind machine at m. 129 (that is, a sound generating device separate from the instruments), but he specifies that the sound of the wind should not cover the sound of the orchestra. It is, however, a bit unclear whether the sound effect would prompt the construction of yet a further space (for quasi-natural sounds) or whether it would be absorbed into the overall sounds of the orchestra. 264 Integral Example3. CINfor bars125-53 ofSchoenbergs Die gliicklicheHand. / I Generic space • tempest \ ) / •/ Z • [ I I I \ \ Music \ Lighting • gradualtransition \ from dulland ^ "* II to murky bright / I I \ \ • gradualassembly \ I of materials towarda forceful I c/imox ^7 ^"^ >/ x ^*^ \ Stage directions \ \ / • character,lately I \ \ I w triumphant, enacts I \ " I onsetof terror \ j Si \ S t• ^^^ ^>p •/ ^S?s. /^ Blend ^\ #####X • ^ I \ \ \ \ / aMM) \»**** somewhatorderly *\ dreamchanging I intocAoo/ic / J nightmare Drama \ • transitionfrom I \ effortlesscreation I I \ I ofjewels to I complete V^ •••• x^*^-ine»ectuality y - ^^ Review Forum: Zbikowski on Cook 265 similar dynamic sweep projected from the space set up by the lighting, but it is the inverse of what we might usually expect: the tempest begins in murky darkness and climaxes in the bright light of day. The increase in intensity is similar to what happens as a tempest gathers power, but the final result- glaring yellow light- is not strongly correlated with our impressions of storms. Rather, visual impressions of a body reacting to a tempest are projected into the blend from the space set up by the stage directions, as is the sense of anticipation from the dramatic space. The instance of multimedia that results from combining these various projections within the blended space is reminiscent of a tempest, but not one that comes from without: although the noise and tumult of the music and the sense of dramatic anticipation that derives from the stage directions and dramatic situation would support this, the receding darkness and restrained (if nonetheless tense) gestures by the actor speak of internal tumult. This is a tempest that arises from within, an impression that conforms with Schoenberg's stage directions: "Der Mann hat dieses Crescendo des Lichts und des Sturmes so darzustellen, als ginge beides von ihm aus" ("During the crescendo of lights and storm the man reacts as though both emanated from him").15 The second aspect of blending relevant for the analysis of musical multimedia concerns patterns within CINs. Although networks can take diverse forms, and can- as the foregoing discussion illustrated- involve multiple input spaces, most blends can be characterizedin terms of one of four types of CINs: simplex networks, mirror networks, single-scope networks, and doublescope networks.16 Most IMMs can be described in terms of the latter three. For instance, among the three IMMs represented in Example 1, the conformance model can be characterized as a I should note that the analysis I offer here is intended to illustrate a conceptual blend that involves multiple input spaces. Cook discusses evidence that Schoenberg did not regard all of these "input domains'*as equivalent, suggesting a differential projection from the various source domains into the blend. Such differential projection is well recognized within blending theory; see Fauconnier and Turner 2002. Each of the four main kinds of CINs are discussed in detail in Fauconnier and Turner 2002: 119-135. 266 Integral mirror network: because the input spaces share a common organizing frame they "mirror" one another, and differences between the inputs are greatly attenuated. The contest model, on the other hand, is a classic double-scope network: the input spaces have incommensurate organizing frames, and differences between the inputs are consequently accentuated. Finally, the complementation model can be characterized as a single-scope network: because the organizing frame for one of the inputs typically dominates the IMM, differences between the inputs are somewhat muted but still noticeable. In general, the technology of conceptual blending- which gives ways to describe what materials are projected into the blend (or IMM) and what the relationships are between the various mental spaces within the network- makes it possible to give detailed characterizations of the similarities and differences between the constituent media of an IMM. Conceptual integration networks, then, offer a way to simplify and make consistent technical descriptions of relationships between the constituent media of an IMM, and to allow for comparisons between IMMs and other conceptual blends. Conclusion Although Tovey's survey of the mainstream of music left him feeling somewhat at sea, the waters he navigated were nonetheless familiar ones, and ran deep with absolute music. Today the notion of absolute music has at best heuristic value; at worst, it obscures the entire enterprise of the scholarly study of music, for it strips musical practice of the cultural and social context that makes music meaningful. This perspective compounds the problems modern scholars face in navigating the vast body of music before them, for success requires that they look beyond "the music itself." As Cook shows, analyzing musical multimedia is a kind of laboratory for studying these problems because it pushes the issue of musical meaning, and the role of analysis in revealing that meaning, to the foreground. If music indeed contributes meaning to a television commercial or a film, what is the nature of this meaning, how can we describe it, and how does it connect with the sequence of sound phenomena we recognize as properly musical? Although theories of Review Forum: Zbikowski on Cook 267 cross-domain mapping and conceptual blending cannot answer all the questions we would ask about musical meaning, they do suggest that these issues are not unique to musical understanding, and they provide methodologies for exploring the relationship between music and other media. We may, with Tovey, still be more or less at sea, but the firmament may not be quite as distant as we once thought. References Burnham, S. 1995. Beethoven Hero. Princeton, New Jersey: Princeton University Press. Cook, N. 2001. "Theorizing Musical Meaning." Music Theory Spectrum 23/2 (Fall): 170-95. Fauconnier, G. 1994. Mental Spaces:Aspects of Meaning Construction in Natural Language. 2d ed. Foreword by G. Lakoff and £. Sweetser. Cambridge: Cambridge University Press. . 1997. Mappings in Thought and Language. Cambridge: Cambridge University Press. Fauconnier, G., and £. Sweetser, eds. 1996. Spaces, Worlds, and Grammar. Chicago: University of Chicago Press. Fauconnier, G., and M. Turner. 2002. The Way We Think: Conceptual Blending and the Mind's Hidden Complexities.New York: Basic Books. Gentner, D., and A. B. Markman. 1994. "StructuralAlignment in Comparison: No Difference Without Similarity." PsychologicalScience 5/3 (May): 152-58. . 1997. "Structure Mapping in Analogy and Similarity." American Psychologist52/1 (January):45-56. Johnson, M. 1987. The Body in the Mind: The Bodily Basis of Meaning, Imagination, and Reason.Chicago: University of Chicago Press. Lakoff, G. 1990. "The Invariance Hypothesis: Is Abstract Reason Based on Imageschcmas?" CognitiveLinguistics 1/1: 39-74. Lakoff, G., and M. Johnson. 1980. Metaphors We Live By. Chicago: University of Chicago Press. Proust, M. 1981. Remembranceof Things Past, Vol. 1, Swanns Way. Trans. C. K. Scott Moncrieff and T. Kilmartin. New York:Vintage Books. Tovcy, D. F. 1938. "The Main Stream of Music." Proceedings of the British Academy 24: 113-41. Turner, M. 1990. "Aspects of the Invariance Hypothesis.' Cognitive Linguistics 1/2: 247-55. . 1996. The LiteraryMind. New York: Oxford University Press. 268 Integral Zbikowski, L. M. 1998. "Metaphor and Music Theory: Reflections from Cognitive Science." Music TheoryOnline All (January). . 2002. Conceptualizing Music: Cognitive Structure, Theory, and Analysis.New York: Oxford University Press.