Hearing Transcendence: Distorted Iconism in Tōru Takemitsu’s Film Music

Abstract

Tōru Takemitsu (1930–96), dean of Japanese composers, wrote scores for 100 films. In many cases, he faced the task of giving presence to an unseen, suprahuman force or being. This article deploys musical semiotics and topic theory to examine a gesture I call “distorted iconism,” in which an aural resemblance is altered so as to suggest transcendence. I analyze distorted iconism in Takemitsu’s film work, looking initially at Kwaidan, Hi-Matsuri, and Woman in the Dunes. In these films, Takemitsu needed to indicate that what we see is insufficient, and used such devices as delayed sonic cues, distorted instrumental sound, and strongly marked static music to create a sacred space in which the unseen and transcendent manifests. I use these examples to set up a close investigation of Ran, in which Takemitsu effects complex layers of meaning that both enrich and subtly contradict the director’s vision.

Within the Western art music tradition, representation of transcendence entails a basic paradox. The transcendent is commonly held to be ineffable and music nonrepresentational, yet since the fifteenth century music has often been granted superior access to and the ability to represent transcendence. The many arguments to this effect do not generalize: they depend on the nature, social status, and philosophical understanding of music in a given place and time.

To judge from the last century of religious studies scholarship, this long-standing connection might well not exist. Neither Beethoven’s Missa Solemnis nor Stravinsky’s Rite of Spring, for example, has occasioned significant comment within this discipline, as though such music had no bearing on the imagination of religion. Setting aside the ideological blinders at work, this situation likely arises from technical ignorance: without the music-theoretical training to engage in serious analysis, scholars outside musicology find it difficult to engage with such data.¹

In the present article, I draw on musical semiotics to sketch a bridge over this gap. Focusing on film music composed by Tōru Takemitsu (1930–96), I examine the means by which he used music to manifest the transcendent in a number of major films; for accessibility’s sake, I emphasize Akira Kurosawa’s well-known film Ran (1985).² I propose a semiotic device I call “distorted iconism,” which has a long history in Western art, and consider how it operates in Takemitsu’s scores. This analytical process enables scholars outside musicology to address musical data rigorously, thus positioning semiotics (as always) at the intersection of multiple disciplines. I also hope to suggest that analogous methods might open conversations between religious studies and the many disciplines in which semiotics has a stronger theoretical history.

Music and Semiotics

The academic study of music, founded in the same moment as Wilhelm Dilthey’s Geistestwissenschaften and originally conceived as akin to the study of literature, visual art, and religion, has tended toward isolation. Guido Adler’s expansive manifesto (Mugglestone and Adler 1981) laid a groundwork that maps onto many humanistic fields, but subsequent history has not seen his hopes realized. In part, musicology’s ever-growing technical sophistication, concomitant with the collapse of common education in music theory, has made communication with nonmusicologists difficult. In addition, the positivistic drive toward scientific rhetoric in the 1950s and ’60s created a technical elitism: the “serious” scholar need not communicate with the music-theoretically illiterate. And, too, the normative judgment of professional musicologists through at least the 1960s held Western art music to be unaffected by social phenomena; indeed, the fledgling field of popular music studies had to contend with a scholarly presumption that popular music’s obvious social embeddedness was precisely evidence of its triviality (e.g., Shepherd 2012, 240–41).

The New Musicology movement demonstrated many scholars’ frustration with this situation. Susan McClary noted with venom that by the 1970s, musicology was sick indeed: like autism sufferers, “[such musicology] ‘cannot read the expressions on people’s faces or the intentions in their voices’; [is] ‘generally blind to social meanings’ and create[s] instead [its] own networks of correlations ‘that bring to mind the elaborate pseudoscientific systems of numerology and astrology.’” McClary holds out hope that another approach might “allow us to make sense of music [and] part company with those in the discipline who prefer to identify with … Asperger-syndrome autistics” (McClary 2001, 326–28; quoting Sacks 2001, 4).³

While the New Musicologists clarified serious failings of the discipline, hope for solutions has come from a different quarter—that of music semiotics. Beginning with Leonard Ratner’s Classic Music (1980), followed by the work of his students Wye Allanbrook and V. Kofi Agawu, the approach known as “topic theory” leans on Peircean semiotics to integrate historical, aesthetic, and dramatic meaning with technical or “structural” analysis. More recently, the late Raymond Monelle and especially Robert S. Hatten have sought to reformulate topic theory on a more rigorous basis, Monelle in a historical vein and Hatten in a strictly Peircean one. This approach has the immense merit of taking historical context seriously without by that token discarding the enormous achievements of theory and analysis. Given the vibrancy of these debates, as well as the relative accessibility of its products to nonspecialists, such musical semiotics represents the best hope of forging connections between musicology and other humanistic disciplines.

Topic theory began, as Ratner’s title indicates, with late eighteenth-century classical, or “classic,” music. In essence, a topic (referring to Aristotelian topoi) is a small musical unit carrying a stock meaning known to competent listeners. For example, chorale textures carry churchly connotations, horn calls reference the pastoral. By deploying, manipulating, and juxtaposing such topics, classic composers expressed semantic meaning, even in works without overt programmatic content. As topic theory has developed, scholars have explored the history of particular topics (Monelle 2000, 2006), their deployment expressing or clarifying underlying meanings (Allanbrook 1983), and the theoretical implications of these signs. Agawu has drawn a valuable if problematic distinction between “extroversive” topics, which refer extramusically, and “introversive” topics, whose meaning manifests principally in intramusical connections (Agawu 1991, 2009; Hatten 1994, 2004).

Unfortunately, it remains uncertain how far this approach can extend historically. Once one moves beyond the common-practice period (after about 1920), the principles of topic theory seem to break down: in the absence of normative rules and meanings, topics easily become little more than clichés. Yet because these principles have been solidly referred to broader semiotic theory, one would expect that translation ought to be possible, and it should open avenues into other sorts of music. Philip Tagg (2003, 2009) has made valuable moves with popular music, but the larger semiotic problem remains.

The present article takes a deliberately quixotic approach. I focus on a late twentieth-century composer who was neither wedded nor opposed to tonality or to any particular successor thereto, and I look principally at film music, whose concerns and structures differ radically from the classic era. I aim to explicate semiotically how Takemitsu’s film music manifests—and not merely depicts—transcendence: how gods, demons, elemental forces, and haunting memories take shape on-screen through the medium of music. In many films, of course, visual representations of such phenomena receive musical underscoring that gives both texture and emotional emphasis, but these Takemitsu scores depart sharply from this expectation: the music “shows” what is not seen.

Tōru Takemitsu

Takemitsu was the most prominent and influential Japanese composer of the twentieth century. It is usual to mention a lack of formal training, but given his close early encounters with distinguished composers, this amounts to unnecessary mythmaking. His early Requiem for Strings (1957) was judged by Igor Stravinsky a masterpiece, and with such works as November Steps (1967), Quatrain (1975), and A Flock Descends into a Pentagonal Garden (1977) he gained and held international prominence. His compositional style was remarkably varied, but it is usual to align him with the influences of Claude Debussy, Erik Satie, Olivier Messaien, and John Cage; more recent critics acknowledge the importance of the Second Viennese School. Stylistically a Western composer, Takemitsu worked within a range of tonal and posttonal harmonic frameworks rooted in the Western tradition. Yet he never quite escaped being a Japanese composer: he is over-strongly associated with traditional Japanese instruments and aesthetics, as with the biwa and shakuhachi in November Steps. Like most avant-garde artists of his circle (especially the Jikken Kobo [experimental workshop] group), he grew up linking premodern Japanese aesthetic notions with wartime militaristic nationalism, and he was equally concerned that the insertion of traditional instruments into a Western orchestral context would inevitably lead to tokenism, colonialist trivialization, or even support for right-wing nihonjinron (“Japaneseness”) essentialism. Ultimately, as I discuss later, he found his own idiosyncratic ways of making use of selected traditional musical elements, but Takemitsu was never a “national” composer in the senses that Wagner, Dvořák, Smetana, or even Bartók might have accepted. Simply, he never slipped into self-orientalizing, nor catered to easy commercialism (see Burt 2001).

Unlike many, perhaps most, contemporary composers, Takemitsu never held—and indeed refused—an academic post, and he made most of his money through commissions, especially for film scores, of which he ultimately wrote some hundred. Yet only rarely does one suspect that he composed for money alone. Not every score is equally good, but the quality of a Takemitsu score generally balances that of the film: where we find Takemitsu’s music mediocre, the film is likely frankly bad.⁴ By contrast, the number of brilliant scores for equally brilliant movies is extraordinary, including many of the best films by most of the great Japanese directors of his day, particularly Masahiro Shinoda, Hiroshi Teshigahara, Masaki Kobayashi, Nagisa Ōshima, and Hideo Onchi.

From an analytical standpoint, Takemitsu’s work for film is frustratingly varied. He has no consistent, identifiable style, the way Max Steiner, Miklós Rósza, and John Williams do. One can only weakly point to a few reasonably common characteristics, particularly a tendency to set a relatively small amount of a given film.⁵ Otherwise, what regularity one finds in Takemitsu’s film music primarily resides at the semiotic level, as well as that of practical approach.

Takemitsu’s approach to film composition differed remarkably from industry norms. As a rule, a composer comes to a film well into the process, when a good deal has been shot. Commonly, the director shows rushes, discusses what sort of sound the film ought to have, and presents a general outline for the composer’s work. The composer writes music in rough form, focusing on particulate themes (leitmotifs) to underscore characters, themes, scene types, environments. Once the film is edited, the composer produces a full score. The leitmotifs receive full expression, interweave to reflect screen action, and in sections requiring extended musical accompaniment are linked by incidental music, all precisely timed to on-screen cues. The music is rehearsed and performed with the conductor—often the composer herself—simultaneously watching the film in order to ensure a smooth correlation.

Directorial involvement varies considerably. Composers usually audition pieces of the work to avoid last-minute overhauls. Many directors are present for the recording process, and they may ask for changes: at this cue, could we have a stronger bass, and perhaps the music might otherwise be a little thinner? A successful film composer quickly masters this system, so that in a matter of minutes a page of the score can be edited. Some changes are also made electronically at the mixing board. In sum, the normal process of successful film composition in the late twentieth century sees the composer as a subcontractor, like a cinematographer, who brings specialist knowledge to the task of producing whatever the director imagines.

Takemitsu, however, generally insisted on early involvement. He visited sets regularly and discussed with the director not only the music but many other elements: appearance, concept, script. His attitude was that of an active, creative participant whose contributions might occur at any level. Directors who trusted him fully, like Shinoda and Teshigahara, granted him considerable power. For example, when filming Rikyu (1989), Teshigahara asked Takemitsu to oversee decisions on some script changes.⁶ Amazingly, Takemitsu even told Shinoda to reedit the final eponymous scene of Double Suicide (1969), cutting more than half (Zwerin 2007). Yet when it came to composing, Takemitsu preferred to work alone. He recorded his score with whatever instrumental ensemble was appropriate, and the tape was then mixed with the director present. Thus Takemitsu’s working methods constitute the opposite of most mainstream film music composers’, and this radically different practice often clarifies the scores.

Takemitsu, Kurosawa, and Ran

Kurosawa apparently thought about music in terms of great Romantic warhorses, and knew little more. He regularly showed rushes, to composer, cast, and crew alike, accompanied by recordings, and often insisted that composers essentially rewrite these classics. For Red Beard (1965), Kurosawa told Masaru Satō to rewrite Haydn’s Symphony no. 9 (“Surprise”) to accompany the snowfall scene. Teruyo Nogami, Kurosawa’s continuity assistant, later assistant producer and production manager, recalls the incident: “‘Why not just go ahead and use the Haydn?’ suggested Satō, smiling stiffly. ‘You want to know why? Because everyone in the audience already has a different image of this music, and that would get in the way. I tell you what you do: write me something better than the Haydn.’ … In the end, Satō produced a sort of musical paraphrase of the Haydn. When Kurosawa heard it, he said dismissively, ‘Hmph. It’s no different from the Haydn’” (Nogami 2006; Tessier and Takemitsu 2005, 19–20).

This approach inevitably led to crises. In Kagemusha (1980), there was “the scene of Takeda’s army retreating along the Sanshu Kaido. … Kurosawa chose ‘Solveig’s Song’ from Grieg’s Peer Gynt Suites.” After seeing the rough cut with this music, Satō telephoned Nogami to bow out. Years later, he commented, “There was too great a gap between Kurosawa’s goals and what I was thinking. … To come up with something that not just resembled an impossibly famous piece, but surpassed it, was beyond me” (Nogami 2006, 197–98).

Takemitsu had worked with Kurosawa before, and the potential for friction was apparent. When Kurosawa showed rough cuts of Dodes’kaden (1970) against the L’Arlesienne Suite, Takemitsu remarked, “That’s nice, you should probably hire that guy—his name is Bizet, lives in France, but I think he’s dead” and walked off the set. Kurosawa begged him to return, and he produced a striking but uneven score.⁷

Kurosawa began serious work on Ran in 1976, and he secured Takemitsu’s involvement immediately. Kurosawa wrote the script while listening to Takemitsu’s November Steps, and “when Takemitsu heard that, he would urge Kurosawa with delight, ‘Hurry up and make the film, won’t you?’” (Nogami 2006, 201).⁸ They agreed on a score dependent on tategoe, a “shrill-voice” chant style adapted from Nō apparently without instrumentation.⁹ Unfortunately, production was delayed: Tōhō Studios refused to spend the estimated $5 million, which would then have been the most expensive Japanese film made, especially as Dodes’kaden had flopped at the box office, and not until the success of Kagemusha could Kurosawa secure funds from French producer Serge Silberman.¹⁰

By that time, Kurosawa had a new idea: he wanted Gustav Mahler—only “bigger!”¹¹ Takemitsu, though furious, eventually agreed: “well … we did fight a lot about this. … But, finally, what could I do? The decision belongs to the director” (Zwerin 2007; ellipses in original).

As usual, Takemitsu spent a good deal of time visiting the set and watching filming and apparently consulted regularly. Kurosawa also sent him notes, the contents of which are unknown but prompted Takemitsu to threaten resignation (Nogami 2006, 203). That situation smoothed out, and Takemitsu composed and recorded the score. But the battles were not over:

Scene number 36, in front of the great gate. Already driven out of the first castle, Hidetora … is rejected anew by Jiro in the second castle, and leaves through the gate. … Behind Hidetora, the great doors creak shut. At that moment, Hidetora’s hauteur crumbles, his shoulders sag, and he reels. As he stumbles, there is a shrill blast from a flute … and an immediate cut to the blazing sun. Beneath the flute music sounds a rumble of low tympani.¹²

Kurosawa said, “Can’t you increase the low notes? Make it hit you in the pit of the stomach. It’s got to have more weight.”

Takemitsu said nothing. He sat so silently that it was as if he had not heard Kurosawa’s words. He disliked the tympani to begin with, and had introduced it here only reluctantly, at Kurosawa’s request. … They replayed the scene with the low tones higher in volume.

“No good. You can’t hear them. Make it stronger, make it grab you.”

Afterward I [Nogami] heard from one of the assistant recording engineers that Takemitsu’s shoulders were shaking with the effort to control his anger. Unaware, Kurosawa issued a request directly to Seihachi Ando, who was manipulating the music track: “Slow the tape down, would you? I want the sound good and deep.” …

Takemitsu finally found his voice. “Kurosawa-san! You can cut and paste my music. You can use it as you please. But I want my name off the credits. That’s all. I quit. I’m leaving!”

With that, he gathered up his sheets of music and his briefcase and walked out. (Nogami 2006, 204–5)

Somehow, over the next two days, producer Masato Hara made peace, and Takemitsu’s name remains on Ran.

This revealing story requires context. Given his methods and the film’s prehistory, Takemitsu’s anger is unsurprising, and certainly Kurosawa was notoriously controlling, a classic auteur. Yet what Kurosawa did is normal in both Hollywood and Japan: directors request changes large and small, and composers usually implement them. One can sympathize with Takemitsu, but the intensity of his fury should strike us as peculiar.

In the end, Takemitsu won two awards for Ran,¹³ but some years later he commented, “Overall, I still have this feeling of … ‘Oh, if only he’d left more up to me …’ But seeing it now … I guess it’s fine the way it is [iijanai]” (Zwerin 2007; ellipses in original). Donald Richie, who was close to Takemitsu and deeply moved by this music, found Takemitsu somewhat dismissive, yet he also notes that Takemitsu’s gentlemanly humility and constant self-doubt require caution in interpreting self-deprecatory remarks. Indeed, Richie believes Takemitsu felt he “won,” and in fact “got away with” writing more or less what he wanted.¹⁴

Interesting though this narrative is, my aim is to examine the semiotics of Takemitsu’s film music, particularly his methods for representing transcendence. Ran may seem a peculiar example, as the transcendent does not manifest overtly in the film. I shall argue, however, that Takemitsu composed a score somewhat at odds with what Kurosawa intended, reshaping Ran through musical means, and inflecting the film to emphasize a subtle and complex relationship between human and transcendent spheres. In order to develop this claim, I begin by explicating the semiotic device I label “distorted iconism,” and show how Takemitsu used it, in different forms, in several important films.

Distorted Iconism

Under Peirce’s tripartite division of sign types, an icon relies for referential power on resemblance, as with a more or less realistic drawing, and thus depends on rules or norms of a given medium. For example, Renaissance painting deploys rules of perspective that do not obtain in every mode of representational painting. The power of reference in complex icons thus presumes competent use, manipulation, and interpretation of such rules.

What I call “distorted iconism” is simply a deliberate, more or less subtle violation of such rules by competent artists. This technique is therefore intrinsically syntactical: in order for a distorted icon to manifest paradigmatically, it must gesture syntactically to rules otherwise obeyed within the same work. Furthermore, this device requires “markedness” (Hatten 2004): the distorted icon must stand out against the rule-determined background.

Distorted iconism is not limited to contemporary music, nor indeed to music itself, and has long been used in the West to suggest or evoke a sense of otherworldiness: transcendence, divinity, the uncanny, or marvelous. Theologically, the Scholastics distinguished a wonder, which appears extraordinary to those ignorant of its cause, from a miracle, which violates the laws of nature and thus requires divine intervention (e.g., Aquinas, Summa contra Gentiles 3.101). Since divine manifestation necessitates a violation of natural law, artists in many media have represented transcendence by carefully and precisely violating norms—artistic “laws.” For example, Leonardo da Vinci composed his Virgin and Child with St. Anne (1508) according to normative Renaissance triangular construction: the three figures and the background strongly indicate the triangle, supporting the calm, harmonious mood. Yet the triangle is sharply violated: Jesus’s head and hands extend beyond the boundary. This is distorted iconism: normative rules of representation are emphasized, yet violated. Why? This is not just some kid playing—this is Jesus, who cannot be constrained by human rules. By manipulating the normative in order to violate it, Leonardo conjures a powerful hint of transcendence.

Turning to music, the final Andante of Mozart’s 1792 Magic Flute, act 1, no. 5 (Quintet) gives a strong example. In this number, Prince Tamino and the comic “natural man” Papageno receive magical instruments (the eponymous flute and a set of bells, respectively) and some sage advice from the Three Ladies, servants of the Queen of the Night (who later turns out to be a villain), and accept the quest to rescue Princess Pamina from the sorcerer Sarastro (who later turns out to be a good guy). Finally ready to depart, the men ask how to find Sarastro’s castle. Suddenly the tempo slows to a grave andante, orchestration nearly vanishes, and the Ladies deliver a brief chorale about the Three Boys, wise spirits who will guide our heroes.

This chorale opens with a remarkable gesture: a soprano skip down into a perfect fifth. This is a serious violation of basic voice-leading principles: one can never move into a perfect fifth or octave with a skip in an upper voice. Brahms sketched a lengthy essay about this principle; Beethoven sought to formulate these rules as simply as possible; Schoenberg joked that occasionally Mozart allowed himself internal fifths only acceptable because he was Mozart—but he did not have these skips in mind (e.g., Mast et al. 1980; Benjamin 1981).

I read this as distorted iconism. Mozart so strictly observes the rules, even in the incredible density of a vocal quintet in orchestral setting with every voice truly independent, that this sudden violation is striking. The passage is clearly marked in tempo and texture, but only an expert paying close attention would detect this violation: it sounds “right” and yet is “wrong.”

The ultimate point of voice-leading rules is simply to maintain independent voices. Mozart’s skips, however, collapse the voices—deliberately. Why? Because the Three Boys are not three characters, but rather one transcendent voice, and they therefore blur together. When they themselves appear, at the opening of Act 1, no. 8, they immediately sing sixteen consecutive parallel sixths—yet another voice-leading violation. The distorted icon here suggests that the Three Boys are not human, nor bound by human rules, but are manifestations of something transcendent.

One could also read this device as indexical, using distortion to call attention to the “othering,” that is, to the ways in which the work otherwise represents the transcendent. By this reading, the Mozart voice-leading distortions index the transcendent by highlighting what is already taking place in the libretto and staging; similarly, Leonardo’s triangle breaking indexes the transcendent by pointing to a child who the viewer knows, from other dimensions of the work, is Jesus. On such a view, what I have called distorted iconism is a misnomer and, indeed, perhaps an impossibility.¹⁵

I grant the legitimacy of such a reading, but I would argue that there is a genuine choice here (which may suggest some weaknesses in the Peircean view of syntagmatic relations). If we read Mozart’s voice leading as indexical, the sign functions by manifesting ambiguity in a context that is otherwise less so. Yet nothing in the libretto explains who the Three Boys are: they are simply “wise spirits,” of unknown origin, and throughout the opera they appear without warning or explanation. The distorted voice leading only indexes transcendence and “othering” insofar as one chooses to interpret the Boys in this way; in many stagings, where the Boys are given some kind of apparent context or meaning that is not alien or transcendent, the index shifts.

Reading the same distortion as iconic, as I do, depends on the iconic nature of the Ratnerian topic. So long as the sonic resemblance is close enough to hear, a composer can present a march at a very wide range of tempi, modes (major/minor, etc.), timbres, and so on. The funereal marches in Beethoven’s Symphony no. 3 (“Eroica”) and Mahler’s Symphony no. 1 are both instantly recognizable as funeral marches, but they take on radically different meanings not only contextually but within their own formations. As we shall see, the latter example is certainly distorted—one does not expect to hear a loud street band in the middle of a funeral march—yet it does not clearly point to some less ambiguous musical sign; indeed, the very juxtaposition of these two icons creates more ambiguity rather than less.

Taking the Mozart voice leading as iconic prompts different questions from those raised by the indexical view. Rather than ask to what other sign the distortion points in order to grasp how the transcendent is highlighted, the iconic reading asks what is transcendent and how it manifests. Both are legitimate readings, as I have said, but I find the latter—“distorted iconism”—more helpful in interpreting musical data. Most particularly, this interpretive approach refuses to presume in advance that musical signification is necessarily subordinate or supplementary to some other mode of semiosis.

Takemitsu’s use of distorted iconism in film music required him to find syntactical and paradigmatic norms with clear meanings—topics, in short. These he manipulated to create meaning and in some instances, through distortion, to evoke transcendence. If we read purely indexically, we assume from the outset that the music supports the other dimensions of the film, as indeed is usually done in film music. But a central part of Takemitsu’s film work is the way in which it goes well beyond any such “underscoring” expectation. As I will argue in reference to Ran in particular, sometimes his music actually alters the content of a film so dramatically that its indexical function comes near to collapse.

Composing for the Films

In their seminal book Composing for the Films, Theodor Adorno and Hanns Eisler explicated Hollywood film music techniques and challenged them at every level. They argued that the film music style arising largely from Max Steiner was aesthetically trivial, oppressed musicians and composers, and supported the industry’s attempts to make audiences into uncritical and alienated sheep. Whether one is satisfied with this argument or otherwise—and Adorno’s broader formulation in Dialectic of Enlightenment (Horkheimer and Adorno 2007) is disturbingly persuasive—Adorno and Eisler’s list of film-musical techniques remains remarkably comprehensive (Adorno and Eisler 2007). After a brief explication, I will turn to Takemitsu’s manipulation of these norms in constructing distorted icons.

The dominant model relies on the leitmotif. Unlike the far more sophisticated expressions of this technique in Wagnerian opera, Hollywood leitmotifs are brief tunes, sonorities, and/or rhythms attached to characters, locations, themes, objects. The composer combines these to map the film, such that an argument between A and B about object C uses elements of the three relevant motifs.

Leitmotifs and their intersections depend on several techniques, notably melody, illustration, setting, and cliché. Film music is usually not only tonal but strongly melodic, with hummable tunes, locking film music to what Adorno and Eisler consider the more trivial Romantic repertoire—Tchaikovsky and Rimsky-Korsakov, not Beethoven and Brahms. In addition, film music commonly illustrates screen action through iconic imitation; at its extreme, one finds “Mickey-Mousing,” in which music literally imitates gross action. Film music may employ geographical and historical references to establish setting, reinforcing musical stereotypes to assist immersion; one example is Maurice Jarre’s score for Lawrence of Arabia, which insistently repeats devices that Western listeners associate with Arabic music. Structurally, movie music is usually unobtrusive, hiding in the sonic background; where music comes forward, it reinforces the strongest clichés by presenting known motivic material in an emotionally obvious manner, emphasizing the expectation of music as incidental. In general, these devices construct and manipulate leitmotifs as topics: viewers must know a range of musical clichés, recognize their standard meanings, and accept their subordinate, underscoring function. For Adorno and Eisler, the method acts to produce and reproduce passive audiences, and furthermore impoverishes “serious” music by limiting it to emotional experience.

While Adorno and Eisler propose some alternatives, their principal concern is a devastating critique of the system. They only indirectly recognize the possibility of manipulation, of working with film-musical norms to achieve something at odds with expectation. On the whole, they would likely resist my elision of film-musical clichés with classic topics, just as they dismiss the filmic leitmotif as a bastard imitation of Wagner. I argue, however, that it was through exactly this kind of deliberate manipulation, even inversion of film-musical topics, that Takemitsu composed some of his greatest scores.

Takemitsu and Distorted Iconism

Since Takemitsu did not normally rely on leitmotifs or melody, his film music always deviates from norms. One might also assume that, because Takemitsu was an avant-garde composer working with “art film” directors, he would disregard conventions as a matter of course. Yet there is an important difference between disregarding norms and actually violating them. Distorted iconism depends on a norm’s being in place and accepted: only against the norm can the icon be marked. And while Takemitsu wrote some experimental scores that bear little clear relationship to the conventional system, he usually drew on the topics and syntax of mainstream film to achieve his compositional ends.¹⁶

In most films, Takemitsu largely disregards illustration, in the sense that his music does not overtly reference action. Where he does do so, however, he does it strongly and consistently, thereby establishing a norm. Having positioned the viewer to expect relatively straightforward, if occasional, illustration, Takemitsu may then violate it sharply, producing distorted iconism. A clear example is Kwaidan (1964, directed by Masaki Kobayashi), which presents four horror vignettes based on stories by Lafcadio Hearn. All four are of considerable musical interest, but for present purposes I focus on the second, “The Snow Woman” (Yuki Onna).

Caught in a sudden blizzard, two woodcutters stumble on an empty hut, collapse inside, and, exhausted, fall asleep. The young apprentice wakes to see a beautiful and terrifying woman, dressed in cold bluish-white kimono. She breathes icy mist on the older woodcutter, and he dies, frozen. The Snow Woman turns to the apprentice but smiles. She had planned to kill him too, she says, but he is so young and handsome that she will let him live—but if he ever tells anyone, she will return and kill him. Back in his village, the apprentice soon meets a lovely young woman. They fall in love, marry, and have a baby. One night, sitting by the firepit, he is struck by an odd resemblance and tells the story of the Snow Woman. Before his eyes, she is revealed: his wife is the Snow Woman herself. Furious at his broken promise, she nevertheless decides, for the baby’s sake, not to kill him. She walks into the snow and vanishes forever.

Takemitsu’s music for this tale uses two sonic distortion devices. When the Snow Woman departs the snowbound forest hut, the door slams closed behind her—and the crash sounds nearly a second after the visual cue. (Takemitsu controlled all sound for Kwaidan.) More remarkable, however, is the storm’s howling wind. Takemitsu composed an intricate wash of high woodwinds, principally various flutes, then electronically bent the music up a few pitches. The effect is extraordinary: something seems wrong with the sound, which wavers disturbingly between “music” and “noise.” Although one might call the music illustrative, its strong distortion and categorical uncertainty suggest something quite other: what is “illustrated” is unseen, in that this storm takes on an alien, threatening quality not clearly expressed visually until the Snow Woman herself manifests in the hut—at which point there is almost total silence.

Takemitsu’s Kwaidan score powerfully demonstrates his use of distorted iconism. The music is strongly marked, preceded and followed by extended silence. Disregarding the boundary between “noise” and “music,” Takemitsu composes recognizable topics: the howling wind and slamming door are readily comprehensible. Having done this and marked the music aurally, he then systematically distorts it, lending eerie presence to the supernatural.

The use of sonic absence and presence for marking prompts Takemitsu to violate the expectation of musical unobtrusiveness in many films. A remarkable example is Hi-Matsuri (The fire festival [1985], directed by Mitsuo Yanagimachi). Although Takemitsu composed only some eight minutes of music, they have an unforgettable effect.

The beautiful, remote village of Nishigima is divided between forest and ocean factions, each worshipping its own goddess. The fishermen, interested in economic development and modernization, support a project to build a marine park. The woodsmen, led by the brutal Tatsuo, continue ritual woodcutting practices and oppose the park. Tatsuo takes savage delight in bloody hunts (including of sacred monkeys) and unrestrained carnality and indeed believes he has a private relationship with the forest goddess. When the fish in the marine nursery are poisoned, suspicion focuses on Tatsuo. After a season of torrential rain, Tatsuo believes he encounters the goddess in her forest. Embracing the sodden trees, he affirms, “I understand.” The rain stops. Tatsuo violently interferes with the local fire festival (hi-matsuri), then returns home, murders his family, and shoots himself.

Takemitsu’s music appears only during the rain-soaked epiphany. The hauntingly beautiful washes of not-quite-tonal music, the limpid high bells—all stand in stark contrast to Tatsuo’s savage, animalistic ways and give no warning of impending violence. The music is maximally obtrusive, utterly dominant. Framed by ordinary “noise,” the music demarcates a sacred space and time seemingly divorced from the ordinary world.

If the stark contrast between absence and obtrusive presence marks the music powerfully, it is not at all clear how to interpret it. Two possibilities appear: first, the music might indicate a manifestation of the goddess herself, in keeping with the musical workings we have seen in Kwaidan; second, the music might manifest the interiority of the perceiving character, Tatsuo. With the latter device, Takemitsu often uses music to reveal the inner life of an otherwise inarticulate character: in Dodes’kaden, the retarded Rokku-chan’s love of streetcars becomes powerful rather than laughable through the medium of Takemitsu’s jaunty score.

In Hi-Matsuri, however, Tatsuo’s brutality seems to reflect a fundamentally simple, animalistic interior. His relationship with the goddess is unabashedly carnal: “Only I can make the goddess feel like a woman,” he boasts. Yet the music in the forest does not suggest carnal or sexual topics. It is alien, not only to the film’s absence but also to recognizable musical norms. We hear a wash of fascinating sound without apparent development or movement. Static, the music reveals without acting. Tatsuo encounters something genuinely other, transcendent, nonhuman—and this leads him to seek resolution in violent inhumanity.

Hi-Matsuri’s score raises two important issues: perspective and uncertain reference. On the one hand, Takemitsu uses music to conjure perspectives not necessarily visually present, as also with Dodes’kaden. On the other, the referential and indeed representational dimension—music manifesting a transcendent being, for instance—at times functions ambiguously. Whose perspective? What manifests?

Woman in the Dunes (1964, directed by Hiroshi Teshigahara) is one of three collaborations by Teshigahara, Takemitsu, and the author Kōbō Abe. The absurdist plot, a sort of modernist Sisyphus, concerns an entomologist trapped with a young woman in a sand pit. Villagers lower food and water in return for buckets of sand, which eternally slides back down into the pit. Takemitsu’s music appears at numerous instances, perhaps most memorably when the man and woman first have sex. The music begins with a crackling, clearly referencing the endless sand. As the sounds of the couple fade, strings enter in a peculiar, sliding figure. The music swells climactically, and the camera turns away to the sliding sand, finally cutting to a bird in an empty sky as the music vanishes.

Teshigahara explained that the collaborators understood from the outset that the film, based on Abe’s novel, faced a serious technical difficulty. While it might appear that there are only two significant characters, there is a third—the sand itself—and Takemitsu had primary responsibility for giving it presence through music (Zwerin 2007). In the sex scene, the music initially seems merely erotically illustrative. As the scene develops, however, the camera drifts while the music swells, suggesting an alien perspective.

This music deploys many of the devices already discussed. As in Kwaidan, the strings are electronically bent upwards several semitones, and the resultant eerie sliding, heard against sliding sands, complements the overtly illustrative microphone crackle. Apart from the “noise” topic, the sliding has a sinuous, liquid eroticism that references the couple’s actions. The iconic distortion here effects a perspective, as the sand becomes a voyeur. Yet this voyeurism points to another device in many of Takemitsu’s best films: if the sand watches the couple, so does the audience. In essence, Takemitsu’s music creates a role for the audience, and precipitates us into it. The same effect has its greatest realization in Shinoda’s Double Suicide, where the silent, black-clad kuroko puppeteers who manipulate the actors are simultaneously witnesses and accomplices. In this film, Takemitsu’s music voices their inarticulate perspective, and in the process forces viewers to accept a measure of social responsibility for the tragedy.

Takemitsu’s Aesthetics

I have enumerated several techniques of distorted iconism. Takemitsu draws on extroversive topics in the form of both “noise” and easily comprehended musical types, as well as introversive topics in the form of echoes and subtle visual references that link musical and visual elements syntagmatically. With these established, he commonly marks moments of distortion by making them overwhelmingly dominant and bracketing with complete silence. Within each musical moment, the distortions themselves come in many forms, most obviously electronic manipulation of time or pitch. He also often makes visual reference ambiguous or unobvious, suggesting meanings or presences that contradict the apparent visual surface. This last device, by creating alternative perspectives distinct from those of the characters, in some cases draws viewers into an active role, and in others constitutes wholly distinct entities.

Many of these devices also inform Takemitsu’s concert works. While a full investigation is beyond this article’s scope, I will briefly locate these semiotic gestures in Takemitsu’s idiosyncratic aesthetic thought.

Beginning during his “middle period” of avant-garde modernism, Takemitsu became interested in Japanese garden design, and wrote a number of works explicitly modeled on, or constructed around, the strolling, meandering process of garden viewing, as in Arc for piano and orchestra (1963) (Burt 2001, 103). By the early 1970s, he had reinterpreted his garden imagery away from synchronic, vertical relationships among time sequences: “the metaphor is now related to the horizontal aspect of the music, the actual contents of the piece as it unfolds in time” (Burt 2001, 168). This latter formulation would come to full ruition in A Flock Descends into the Pentagonal Garden for orchestra (1977), and it would continue to inform the composer’s thinking as late as the 1991 Fantasma/Cantos for clarinet and orchestra: “[In this work], which the composer claimed was inspired by viewing a Japanese ‘tour’ garden, the ‘formal’ interpretation of the ‘garden’ metaphor becomes the dominant model: it is as if the soloist, and vicariously the listener, were wandering through a series of fixed musical ‘objects,’ some of which may be viewed in a different light, and some of which may be revisited—as, typically, at the end of the work, in which via a literal repetition of material the music usually arrives back where it began” (Burt 2001, 168–69).

Crucial to this metaphor is the minimal interaction or intersection between positions, viewer/soloist and garden objects/instrumental grouping. The viewer encounters objects and clusters in the garden, and the relationship is one of contemplation, meditation, perhaps empathy—but not dialogue or mutual engagement. Although Takemitsu carried this metaphor into his late-period reconsideration of syntactical relationships in music, he did not return to a traditionally Western-Germanic dialectical model—from which he had always been somewhat distanced by his fascination with French Impressionism.

In Takemitsu’s thinking, the spatial and temporal often manifest in terms of the aesthetic concept ma, loosely speaking “space, time, pause.” While ma permeates various modes of Japanese aesthetic production, Takemitsu’s understanding is distinctively his own:

The sounds of a single stroke of biwa plectrum or a single breath through the shakuhachi can so transport our reason because they are of extreme complexity; they are already complete in themselves. Just one such sound can be complete in itself, for its complexity lies in the formulation of ma, an unquantifiable metaphysical space (duration) of dynamically tensed absence of sound. For example, in the performance of Nō, the ma of sound and silence does not have an organic relation for the purpose of artistic expression. Rather, these two elements contrast sharply with one another in an immaterial balance. Japanese sensibility has produced the unique concept of ma in response to hearing these highly complex sounds, each one of which can be appreciated as individual and complete in itself. … It is through being freed from fixity of expressive meaning, through becoming all the more complexly refined, that sounds can be equal with nothingness, like the natural sounds which come from a grove of bamboo. (Takemitsu 1994)

This correlation of nothingness with the natural even leads to meaninglessness: “I feel that the final goal that Japanese sounds strive to reach is ’nothingness,’ which is by no means the same as that of ’liberation.’ On the contrary, it seems to be in opposition… [S]ounds in traditional Japanese music seem to reject the scale to which they belong. Having been polished, each sound, standing outside the scale, becomes meaningless. As a result, the totality of sound becomes itself equivalent to nothing. The component sounds approach a condition indistinguishable from the naturally occurring sounds that are individual and do not cohere” (Takemitsu et al. 1992, 46).

Takemitsu’s remarks are extremely suggestive in light of his filmic distorted iconism. The extreme markedness isolates each musical moment, makes it uniquely itself. Each is thus an object of contemplation, but one that potentially seeks meaninglessness, the point at which sounds become “indistinguishable from the naturally occurring sounds.” The garden metaphor clarifies the importance of perspective: each object contemplated, each contemplating being, is equally isolated.

Consider Hi-Matsuri again. Tatsuo encounters something in the forest: it is “individual and [does] not cohere,” as though “from a grove of bamboo.” It is “meaningless.” If Tatsuo says “I understand,” what does he understand? What Takemitsu often sees as a human desire for control and refinement drives Tatsuo to integrate this experience into a totality—but that is precisely what it is not. The sound of the forest and the silence of Tatsuo’s life (or perhaps the reverse) have no “organic relation” to one another. Just so, the viewer watching the film unfold is driven to seek explanations, to “make sense”—but for Takemitsu, this may miss the point.

All these techniques and aesthetic issues operate synthetically in Ran, to which I return for the remainder of the present article. In short, I will argue that distorted iconism in Takemitsu’s score dramatically alters the film’s meaning. Specifically, while the film overtly references Buddhism and divinities rather sporadically, Takemitsu’s score transmutes the film into a sophisticated meditation on human-divine relations, on the nature of human and divine vision.

Interpreting Ran Musically

The battle for the third castle is colossal. Blood runs in streams. Armored men impaled by thickets of arrows crawl in agony and collapse. Weeping women slash their throats. A fleeing warrior is ridden down by an army of horsemen. The predominant colors are the yellow and red of the attackers’ flags, the black of the defenders’ armor, streaked everywhere with lurid crimson and a hazy blue reflected from the sky. Smoke billows from the castle, set afire by waves of blazing arrows. It is a scene from hell, only too believably human.

Across it all we hear nothing but music. Instead of screams and shots, an agonizing wash of Mahler-like orchestral sound bathes our ears, as the warriors bathe in blood, lending the scene an even greater horror.

One commander rides into the ruined courtyard. He gazes with grim satisfaction on the burning keep. The music rises to a thin, high note in the strings. Then a shot rings out, blood leaps from his back, and he falls. At once, the music ceases, and the thunderous noise of terrific violence plunges the screen into chaos.

For Ran, Takemitsu clearly did not write the music he originally intended, nor did Kurosawa get quite what he wanted. Yet the music for Ran is an extraordinary achievement. A quarter-century after its release, Ran has achieved recognition as a masterpiece, despite an initially tepid reception, and there can be no question that Takemitsu’s score plays a central role. But with Takemitsu in disagreement with Kurosawa—unlike his smooth collaborations with Kobayashi, Shinoda, or Teshigahara—interpretation must move cautiously.

Initial access is provided by the solo flute, the “shrill blast” that occasioned the ultimate battle with Kurosawa. This flute is the nōkan, an instrument used almost exclusively in Nō drama. It appears, generally briefly, some nine times during the film, once diegetically (played by the blinded Tsurumaru).¹⁷ Here Takemitsu and Kurosawa agreed: the film contains numerous explicit allusions to Nō, as in the way Hidetora’s makeup gradually transforms into a living expression of an old man mask (kojō).

The nōkan’s use in Ran is unique in having received the only extended music-theoretical analysis of any part of Takemitsu’s mainstream film music known to me (though see Koozin 2010a and 2010b). In her analysis, Tomoko Deguchi (2006, 2010) refers to the Nō aesthetic triplicity in which a full program or a single play follows a progressive structure: jo (beginning, slow), ha (development, faster), kyū (climax, fast). This dictates rhythm and intensity in each small piece of a play, each act, and across the whole. In a full program, the God play is jo, beginning the event in a stately and minimally dramatic fashion; the Woman play (ha) expresses the height of mysterious (yugen) power, when maximal energy is developed but remains coiled up like a spring; and the Demon play (kyū) releases this energy in a burst of excitement. Deguchi argues that the nōkan in Takemitsu’s score indicates each of the three acts and also a three-step progression within each. Yet while I find Deguchi’s argument valuable, she does not take up the relationship between the nōkan and the dominant Mahlerian music, and her interpretation is weakened by uncritical generalizations about Japanese aesthetics; she also makes little of the fact that the nōkan music is structured quite unlike what one hears in Nō: that is, it is a distorted topic with ambiguous reference.

Preliminarily, I suggest interpreting the orchestral music in terms of Nō. In encoding Nō by essentially Western musical means, Takemitsu drew on both the compositional resources he favored and the qualities he discerned in Mahler. Most important, Takemitsu’s refusal to grant Kurosawa control indicates that despite initial distaste for this score conception, he nevertheless composed a work to which he felt committed.¹⁸

At the heart of classical Nō aesthetics lies a remarkable conception: the principal performer, who takes the central shite role, chants and dances not within the role but in a sense behind it. The crucial issue is the relationship between performer and audience. In the most classical view, the actor manipulates the shite as a hollow marionette. Through his “intensity of mind,” the actor draws the meditative audience member into the hollowness, so that the audience experiences the shite’s transformations personally. Because the transformations in question, in the most typical Nō dramas, concern letting go of desire and moving beyond worldly things—processes framed explicitly in Buddhist terms by the second or side character (waki)—the audience member is drawn along the path toward release. The actor does not so much perform the shite role as enact a Bodhisattva, a transcendent being guiding humans on the path to enlightenment.¹⁹ Speaking of “warrior” plays (shura mono), Zvika Serper writes:

Most of them are losing-battle Noh (make shura nō) that depict the agony of the defeated warrior, usually of the Heike [clan] … [M]ost of them [have] the two-act pattern in which the waki, usually a priest, meets an incarnation of the warrior soul in the first act; then … the ghost of the warrior appears in order to reach salvation, through which its soul will cease to reincarnate and merge at last with the cosmos. The second act climaxes in the ghost telling the story of its battle and death, through song (mainly by the chorus) and dance, in a part called kuse. (Serper 2005, 325)

Similarly, in “woman” plays (katsura mono), one major type has the lovelorn main character achieve the love she desires through her fidelity beyond death, while the other major type shows the rejected main character coming to terms with the agony of her loneliness. One could say with some justice that such Nō plays—and warrior and woman plays are commonly held to be the epitome of the form—demonstrate the transformative power of letting go desire, and thus of accepting this illusory world for the vale of tears it really is.²⁰

In this light, Ran is a sort of anti-Nō. The shite, Hidetora, achieves no transformation or release. Indeed, his death manifests a flat rejection of letting go: having regained some semblance of sanity through reconciliation with his third son Saburo, he begins plans for a warm paternal relationship—plans brutally cut short by an assassin’s bullet. Returned to gibbering incoherence beside Saburo’s corpse, Hidetora simply dies, giving in to the horror. Hidetora’s death rejects reality: utterly bound by his desires, all fundamentally personal and indeed selfish, he cannot accept this final end of hope. One can imagine few scenarios in which Hidetora’s death would so dramatically oppose Nō’s Buddhist principles.

Moreover, Nō’s ritual-performative dimensions seem entirely absent, again raising the question of an “anti-Nō.” One has little sense that the audience should enter the film empathetically, and indeed a number of factors appear contrived to prevent this, notably the pitiless and distant camerawork, the “Olympian view that holds little sympathy for most of the main characters” (Wilmington 2005, 6; also Canby 1986; Hoile 1987, 29; Richie 2002, 7). The style of speech and motion is not realistic, but it hardly evokes the ritualistic manner of Nō. Hidetora is particularly distant: unlike a traditional shite, he seems steadily to decline in charisma and presence, until by the end he is merely pitiful. And, of course, there is no actor to manipulate the marionette, nor any direct connection with the audience: this is not a live performance.

Kurosawa may well have intended to construct “anti-Nō.” His work includes a number of such critical inversions, such as Yojimbo (1961), with its wicked parody of chanbara samurai flicks. Yet Deguchi’s work suggests that Takemitsu set out to invoke and manipulate Nō dramaturgy through music, without apparent inversion.²¹ Whether or not Kurosawa sanctioned or even recognized what Takemitsu was doing, I suggest that the music supplies the missing audience and actor. By treating both Nō and Mahler topically, then distorting their apparent meanings and functions, Takemitsu reinterprets Ran.

The crucial music here is the Mahlerian pastiche, that which has the least obvious screen reference. The nōkan appears diegetically once, overtly references Nō, and inevitably “sounds Japanese”: in Adorno and Eisler’s terms, it has a clear “setting” function. Less obviously, the nōkan operates structurally to mark jo, ha, and kyū phases, as Deguchi showed. Yet this music also provides a sonic contrast against which the Mahlerian orchestral mass becomes subtly marked, just as, most dramatically, the abrupt gunshot marks the preceding music as much as the succeeding chaotic noise.

The orchestra appears somewhat sporadically, in units of varying length, and provides a meditative perspective on the events. The distant cinematography supports Takemitsu’s nondialectical musical viewer, who perceives and considers but does not interact with the garden. Given Takemitsu’s film tendencies, this already hints that the music not only guides audience contemplation of the events but also constructs our perspective in terms of an essential but invisible role in the drama:

Kyoami: Are there no gods, no Buddha? If you exist, hear my words: You’re all cruel and fickle pranksters! Are you so bored up in heaven that you must crush us like worms? Do you think it’s fun to see men weep, you bastards?

Tango: Stop it! Do not curse the gods! It is they who weep. In every age they’ve watched us tread the path of evil, unable to live without killing each other. They can’t save us from ourselves. Stop your crying! Such is the way of the world. Men live not for joy but for sorrow, not for peace but for suffering.²²

In a way reminiscent of Woman in the Dunes or Double Suicide, Takemitsu’s music draws the audience into the role of the weeping, impotent gods.

Partial confirmation of this hypothesis comes with the first appearance of the colossal battle music, when Hidetora confronts Tsurumaru’s sister Sué at the second castle. Sué is a devout Buddhist, and the first inklings of the battle music come when Hidetora opens the screens of her prayer cell and the camera focuses on a scroll of the Buddha. High on the castle walls, Hidetora demands to know how Sué can affect not to hate him, the one who burned her parents’ castle and slaughtered its inhabitants. As the music swells, Sué explains that she takes comfort in the mercy of the Buddha. “There is no Buddha in our world,” declares Hidetora. “We live in a dark age from which the Buddha’s guardians have been routed by the Asuras.” The music ceases with the arrival of Jiro, Hidetora’s second treacherous son, to whom the castle belongs.

Topically, it seems peculiar indeed that this traditional religious element in Ran should be matched not with the nōkan but rather with Mahler. The notion of distorted iconism is helpful here, precisely because of the ambiguity of reference at work. I have argued that this music functions to construct of a distant (and weeping) transcendent viewer, but the depth and richness of the music strongly suggests that there is more at stake.

Importantly, Takemitsu was highly adept at pastiche, able to reproduce and to adapt himself thoroughly to almost any musical style, an ability that clearly contributed to his constantly being called a “genius” by nonmusicians who met him.²³ When one considers Takemitsu’s appropriation, in his concert music, of Debussy, Satie, and occasionally Messaien, it becomes apparent that this gift was as much analytical as imitative, and afforded him access to a wealth of possibilities. Agawu has argued that this relationship between composition and analysis is fundamental to both procedures, such that “analysis is most productively understood and practiced as a mode of performance and as a mode of composition” (2004, 273). On this basis, I suggest that Takemitsu’s re-composition of Mahler reflects a sophisticated interpretation of the Austrian master, not a mere emulation. That is, we must hear in this music a reimagining, one that not only sounds like Mahler but actually does musically what he does.

In his scintillating book on Mahler, Theodor Adorno analyzes the composer’s lack of acceptance (up to the 1960s) into the mainstream canon. For Adorno, the listener’s unease arises from Mahler’s “brokenness,” his shattering of the late-Romantic sweep of symphonic grandeur by insertion of unintegrated material. Bits of popular music, imitated natural sounds, or antiformal elements manifest without preparation, then vanish. Symphony no. 1, for example, “hurls unmediated contrasts to the point of ambivalence between mourning and mockery. The potpourri of the third movement admits defeat by the world’s course, which it despairs to overcome, and coordinates irreconcilables” (Adorno 1992, 52). Adorno here refers particularly to rehearsal number 6 in the third movement, when Klezmer-style band music in the woodwinds and then brass shatters the funeral dirge, with an alternating “oom-pah” from the cymbals and bass drum, only to die away after number 9. The first entry, marked “With parody, not dragging,” has the bassoon and clarinet tune “emerging clearly.” One cannot avoid the viciousness of this parodic element, the way it undercuts the funereal.

Takemitsu emulated this movement in Ran for the funeral of Hidetora and Saburo at the conclusion of the film. It is the most overtly Mahlerian passage, a genuine pastiche.²⁴ Yet when the Klezmer ought to enter, we hear instead echoes of previous thematic material associated with the great battle and with Sué’s Buddhism, material Takemitsu does not present parodically. The march dies away to nothing as Saburo’s army processes, with the sunset-lit ruins of Azusa castle in the distance. The next cut leaps to the battlements, where Tsurumaru inches along, feeling with his stick, the protective scroll of the Buddha clutched in his hand. Inevitably, he moves too far, stumbles and nearly falls, and drops the scroll over the ramparts. And at this moment we hear the solo nōkan one last time, for some 40 seconds, before the blackout and rolling credits.

The logic of the passage denotes the nōkan as the unintegrated material, that typically Mahlerian insertion of sardonic honesty. Read thus, the nōkan stands for the critical, negating quality in Mahler’s most tragic music, the way “his music seeks less to express [consolation] than to console … [which at the same time] is always mingled in Mahler with a feeling of the futility of mere consolation.” Adorno continues, on the Kindertotenlieder: “The songs look on the dead as on children. The hope of the unrealized, which settles like a ray of holiness about those who die early, is not extinguished even for grownups. Mahler’s music brings food to the mouth that is no more, watches over the sleep of those who shall never wake. If each of the dead resembles one who was murdered by the living, that person is also like one they ought to have rescued” (1992, 29).

And he links this discussion of mourning, solace, holiness, and martyrdom to Mahler’s “ironic or parodistic element,” seen as one entailment of his “notion of a bridge between popular and art music” (Adorno 1992, 34).

Thus, if the sweeping, symphonic elements in Takemitsu’s score are those most obviously nodding to Mahler, it is their juxtaposition to the nōkan that best approximates his music. Without this, without the challenge of the unintegrated and stylistically inappropriate, the score would emulate Bruckner, “with whom, as if mere length were a qualitative category, [Mahler] is so heedlessly yoked together,” as Adorno nastily puts it (Adorno 1992, 32). It is the tragic beauty of those orchestral passages that require the cutting nōkan: “The prevalent ideology of the true, beautiful, and good, with which Mahler’s music first made common cause, is inverted into valid protest. Mahler’s humanity is a mass of the disinherited” (Adorno 1992, 46).

Such a hearing prompts interpretive uncertainty. If the orchestral passages construct the gods’ perspective, the cutting nōkan carries a double valence, at once Tsurumaru’s voice and the sign of Nō drama. If “it is [the gods] who weep,” as Tango would have it, we must also see these events from another perspective, that of the “mass of the disinherited,” from below. Does not Kyoami have a point in cursing us, asking, “Do you think it’s fun to see men weep?” True, the gods “can’t save us from ourselves,” but it is the human audience, we “bastards” who “live not for joy but for sorrow, not for peace but for suffering.”

Returning to the battle for the third castle, we now see that there are three sonic perspectives at work. The structural logic imposed by the nōkan functions to mark out moments for contemplation and offers a “valid protest” from “the disinherited,” the countless victims of Hidetora’s bottomless ruthlessness. The peculiar, un-Nō-like strains mark the way in which this voice is unintegrated: isolated from both an orchestra and a drama that could give it meaning, the nōkan merely protests, providing honest (and impotent) commentary.

The grand orchestral sweep, the audience of gods who look down in tears, turns out to be not only impotent but cowardly: when the gunshot rings out, the gods turn away in silence. Kyoami’s criticism gains redoubled force: the gods do not “think it’s fun to see men weep” and hide their eyes. Only we who have paid for our entertainment think it’s fun.

The final voice, of course, is noise. For Takemitsu, as we have seen, the juxtaposition of music, noise, and silence requires mutual contemplation, not integration. By manipulating presence and absence, music and noise, Takemitsu marks multiple perspectives clearly and forces the audience to interpret their respective meanings. But as with Hi-Matsuri, this interpretive process cannot result in coherent explanation: to impose an orderly, consistent meaning would deny the validity of the protest.

Before moving to conclusions, let me note that I do not think this music-centered reading of Ran fits with Kurosawa’s own vision. To my eye, Kurosawa hoped to make a modernist tragic allegory. Drawing on King Lear among other sources, reflecting on the horrific destruction at Hiroshima and Nagasaki, he sought to use Nō to construct a grandly appalling tragedy. On the whole, he succeeded. But Takemitsu’s music makes Ran quite a different film, because so many elements are inflected or distorted to alternative ends. Kurosawa wanted Bruckner—“Mahler, only bigger”—and got Mahler: a biting, sardonic challenge to the audience. Takemitsu’s music asks us to take responsibility for a world in which meaningless horror gets blamed on gods, only victims have the honesty to protest, and we take voyeuristic pleasure in violence.

Conclusion

I have tried to demonstrate the utility of “distorted iconism” as a semiotic category, a technical procedure that often acts to represent transcendence and appears in Western art from Leonardo to Mozart to Takemitsu. In music, distorted iconism depends on competent manipulation of topics, gestures, and markedness and thus has a clear analytical foundation in topic theory.

At the same time, I would argue that analysis of distorted iconism has some practical advantages. Topic-founded musical semiotics emphasizes the relatively stable, rule-bound world of classic music, and not unreasonably moves analytically from scores to hearing to history. But in cases where the scores are unavailable, or the rules are unknown or of limited distribution, such a method has nowhere to begin. By isolating a complex semiotic process, I have tried to demonstrate an alternative that begins with hearing and moves outward to whatever sources are available or appropriate.

Topic theory seeks to integrate technical (structural) analysis of pitches, progressions, and forms with historical dimensions. It has produced readings that respect both fine compositional detail and cultural meaning, chipping away at the seemingly impenetrable wall isolating musicology from other disciplines. In order to continue this progress, however, music scholars must go beyond speaking to other humanists about matters of mutual interest: the discipline must also listen. That is, the study of music must consider how historical listeners—including composers—drew on extramusical concerns to shape the music of their time.

Distorted iconism, I suggest, arises as a solution to a seemingly intractable compositional problem. How does one represent the ineffable using a medium held to be nonrepresentational? By drawing on theological and philosophical conceptions, artists shape large-scale structural solutions in the form of distorted iconisms, then turn to their preferred compositional methods for the working out. Yet this simple imagining has enormous implications. On the one hand, the extramusical resources in question are often vague, principally representative of a composer’s social circle. On the other, the effectiveness of distorted iconism depends on its being recognized, even subliminally, such that composers deploying it successfully must by that very token grant musical force to religious expression. What might appear a minor compositional notion thus turns out to be deeply rooted in historical processes of imagining religion.

The total corpus of Takemitsu’s film music constitutes a grand achievement in a somewhat ill-regarded and under-studied art. His best works are very much his: one should not always insist that the director alone get credit, perhaps with the screenwriter and an actor or two. As with the Teshigahara-Abe collaborations or Shinoda’s Double Suicide, the best of these films find Takemitsu in a truly collaborative role. And, as I have argued for Ran, sometimes the director got rather more than he bargained for. By isolating a semiotic procedure, I have tried to show how Takemitsu manipulated film-musical expectation, visual representation, and sonic norms to represent transcendence.

There are many other devices to be found, many more films to examine. With Takemitsu alone, both Double Suicide and Ōshima’s Gishiki would repay intensive analysis, which would also necessarily draw in dimensions of Japanese modernity to which the transcendent is entirely peripheral. Takemitsu was also the first to use traditional Japanese instruments in a film score, Kobayashi’s Seppuku (1962): how should this topic be interpreted? The present article has only opened some conversations, to which scholars in many disciplines might contribute.