Infinite Regress from the Real: The Butterfly Boom and Schizophonic Mimesis

In 1940, long after cut-paper animation had ceased to be a mainstay of Japanese cinema, professional dentist and amateur animator Arai Wagorō produced an intricate, twelve-minute masterpiece of the genre. Two facts about the single 332-meter-reel film make it a case study of the peculiar relationship between media and law that continues to shape the Japanese cultural industry today.5 First, Arai’s Madame Butterfly’s Fantasy (Ochōfujin no gensō) features the singing and compositions of the preeminent twentieth-century prima donna of Puccini’s opera, Miura Tamaki. Second, the film depicts the denouement of the famous opera in black-and-white silhouette with a round window (marumado) at its center. It is my contention that these two aesthetic choices (one aural, the other visual) are the result of long cultural and legal histories of media and mimesis, representation and mimicry. Together, they demonstrate what happens when cultural products are copied and remediated in other forms: occasionally a new version copies so well that, despite the new media contexts, it fades into the cultural background of other, older versions; in other cases, occasionally a new version stands out because of its difference. How can a given musical performance or recording harmonize with the historical cultural context and how can it stand out? Are there times when a copied song can do both? When does faithful mimicry seem powerful and even original? What inspired Miura Tamaki, already a global star, to help with the production of the short film? Why did she compose her own abridged Japanese version of the opera rather than simply sing the aria she had sung over two thousand times in at least fifty countries?

The simple answer carried in newspapers at the time is that Miura took pity on the amateur animator Arai, whose film comprising over eighteen thousand meticulously composed frames had incurred fees from the German copyright lawyer Wilhelm Plage, who claimed to be representing the estate of Puccini.6 The proposed fees were so high that the film could never have been screened were it to include music from what Arai called “the real thing” (honmono); so Miura stepped in. But this simple story does not provide the context, with which newspaper readers at the time would have been familiar, that explains why Miura would be predisposed to help out anyone against Plage. The explanation of Miura’s motivation cuts to the heart of the Japanese broadcast and recording industry in the late 1920s and 1930s, when the law and industry were beginning to come to terms with the new disembodying aural media of records and broadcast radio. The determining aspects of the new media and copyright contexts of Miura’s work include the 1920s to 1930s boom in music inspired by Madama Butterfly, the rise of a global recording industry, and the impact on Japanese copyright law of the Rome Convention of 1928, which updated the Berne Copyright Convention laws to account for new media.

Prior to the phonograph record, the first true medium for the circulation of music was the globalization of Western sheet music notation in the nineteenth century. Published in Osaka and Leipzig in 1891 and 1894 respectively, two scores of Japanese folk music compiled by musicologists Y. Nagai, K. Kobatake, and Rudolph Dittrich would become sources for Puccini’s operatic copies.7 Scholars Hara Kunio and Arthur Groos, among others, document how the opera musically transposes several melodies without attribution directly from Japanese Popular Music: A Collection of the Popular Music of Japan Rendered in to the Staff Notation (1891) and Six Japanese Popular Songs (1894).8 The Berne Convention of 1886 covered sheet music in the form of such written notation, so it would have been possible for such recorders of music to bring suit against Puccini. However, there were relatively few cases in Japan that dealt exclusively with scores prior to mechanical methods for recording music, and there was doubt about the ability to copyright folk songs, long considered to be in the public domain.9 The new medium of the record also provides a direct connection between Puccini and Japanese music. And further, because of the capital investments involved and the perceived lack of special knowledge involved in spotting copies (the idea that anyone could hear and recognize a musical copy, whereas only those who had studied music could understand a copied score), records were more likely grounds for copyright suits.

Puccini’s own record collection indicates a direct connection to the nascent Japanese record industry. Reminiscing many years later about her visit to Puccini’s house, prima donna Miura Tamaki recalled this connection:

Puccini pointed to the piano, “Madame Miura, fifteen years ago I composed Madama Butterfly at that piano. Back then I heard Japanese music from Ōyama Hisako, the wife of the counsel to the ambassador from Japan in Italy. She gave me records of Japanese music. With those as a foundation, I aesthetically wove in the melodies of ‘Takai yama kara,’ ‘Genroku hanami odori,’ ‘Kimigayo,’ ‘Miyasan miyasan,’ and ‘Echigojishi,’ as I composed Madama Butterfly.”10

Miura’s account of the diplomat’s wife, who not only played records but also performed on the koto for Puccini, is bolstered by evidence from Puccini’s extant record collection and his record orders.11 Puccini’s admission that he employed themes without changing them significantly should not be shocking. At the time, such borrowing from what were perceived as vernacular sources would not be considered worthy of a lawsuit or even something to hide.

After recounting these cases and remembering the fact that Puccini copied from multiple sources, it is not the place here to trace Puccini’s sources or to label him a plagiarist, copyright infringer, or derivative artist but rather to recognize the fact that, in Japan’s first major record copyright suit, even the same person performing the same song was not deemed to violate copyright.12 In addition, we need to remember that all art takes place within conditions of small changes, sampling, repetition with differences—in other words, through being mimetic not only of reality but also to other arts. As one of the most studied, commodified, global cultural productions of the twentieth century, Madama Butterfly exemplifies this dynamic. The fact that music can be smoothly adapted and adopted is characteristic of the schizophonic mimesis prevalent in the age of recorded music, but such smoothness is in the ear of the beholder. If we know nothing about the putative original sonic culture of a series of sounds, such appropriations might seem smooth. If we know enough, such appropriations are never imperceptible or smooth. In other words, our distance from the source of music stands in direct relation to the ability to consider an adaptation or adoption legitimate.

Developed to name the appropriation by major music industry figures like Herbie Hancock and Madonna of ethnographic and musicological recordings of Indigenous peoples to their own monetary and cultural gain, schizophonic mimesis occurs precisely at the intersection between cultural production and copyright law. But Feld’s lament about the practice sidesteps what is often taken to be the latent critique in his development of the notion. It is claimed that the negative concept of schizophonic mimesis (pathologized as it is) overlooks what is ubiquitous in cultural appropriation and creativity. Indeed, as problematic as they may be, Herbie Hancock’s controversial claim to Feld that his unacknowledged borrowing from anthropologist Colin Turnbull’s recordings of hindewhu music is a “brother thing” and copyright law’s inability to assign jurisdiction over folk music both acknowledge that community and individual ownership of culture differ. Arguments centered on identity suggest property is communal or folk and thus cannot be stolen.

Much of the “schizo” quality of some forms of phonic mimesis has to do with how jarring or smooth the fit is between musical types. The examples Feld provides are quite jarring. But Roger Caillois’s notion of a tendency toward evolutionary and legendary “psychasthenic” camouflage as mimetic mimicry would seem to indicate that our ability to distinguish between Italian operatic music and Japanese folk song in Madama Butterfly might be equivalent to the degree to which the opera can be called mimetic: for some, it might seem like an operatic mode representing the truth about Japan; for others, it might seem like schizophonic mimesis through its pastiche that mimes several genres of music. When Schafer coined the term schizophonic in 1969, he did so to name the unease of encountering sounds split from their origin.13 But this unease indicates that the person feeling it already held beliefs about the correct assignment of origins. Such beliefs connote their own logics of authority to seek and recover origins. As vibrations moving through a medium outside of the putative soundmakers, however, sounds are always already removed from their origin. Voices splitting from bodies happens even prior to the advent of recording media and is natural because voices travel through the medium of air. In other words, schizophonia suffers from its presumption of a unified subject—here presented as a mythic moment when a voice and body were one. A schizoasthenic approach, however, takes the split (or schizo) as given and shows how one can overtly perform in multiple registers to fit in with a variegated background.

When we think of the orientalism of the Western male creators of Madama Butterfly—including Pierre Loti, John Long, David Belasco, and Puccini—we must recognize that these are but individual names in a longer chain of signification through which the repetition of mimetic representation and mimicry plays the high stakes game of identity, identification, difference, and differentiation. If none of their orientalist representations are proper or appropriate to the infinite reality of the real Japan, we should not be surprised—no mirrors (as finite distillations) can ever perfectly reflect that which they purport to represent. But that such imperfect reflections would affect Japan itself is a function of power inequalities. Further, that both records and legal conceptions of copyright would also have iterations in Japan is only natural. The foreign and new in Japan would be mediated through the known and close at hand, just as the Japanese music had been filtered through the operatic form by Puccini.14 This is the situation in which Miura was enabled to create through reappropriation.

By 1940, when Miura recorded her own composition for the short animated film, she had already made at least nine recordings of music from the opera between 1917 and 1937 for the Columbia, Victor, and Nipponophone record labels in Japan.15 These best-selling records (sales were in the hundreds of thousands) and more than a dozen performances of songs from the opera on the national radio network JOAK between 1932 and 1939 made Miura Tamaki a household name in Japan by the 1930s. Along with covers by other Japanese divas, they also made Puccini’s Madama Butterfly familiar to the Japanese public.

As a result, Madama Butterfly became something of a musical subculture in and of itself in Japan, birthing a spate of spin-off records. Between 1933 and 1939, at least ten different popular songs with lyrics or motifs derived from the opera were recorded in Japan. With titles like “Mourning Butterfly” (“Nageki no chō”), “Ms. Butterfly of Nagasaki” (“Nagasaki no Ochōsan”), and “Cio Cio’s Love” (“Koi no Chōchō”), song lyrics evoked the opera, transposing its situations into modern contemporary life with jazzy and even boogie-woogie popular melodies rather than those of the more foreign-sounding opera and situations deemed too last century. “Ms. Butterfly of Nagasaki” (1939), sung by Watanabe Hamako, even picked up on the melody of Puccini’s “Un Bel di, Vedremo” transposing a riff from the aria into the middle of the otherwise popular-style song.16 It is important to note in this context that Puccini’s aria itself lifts a piece of the traditional Japanese folk song “Jizuki-Uta,” most likely from Dittrich’s score.17 On the one hand, we can distinguish the riffs; on the other hand, the context is so new and different it seems not to matter legally. Identifying origins in this dizzying array of sampling is almost beside the point of such schizoasthenic mimesis. The more important and more historically accurate point is that the mediation, the circulation of records themselves, fundamentally transformed the music on them so much so as to be thought outside the realm of legal adjudication.

In May 1929, after a long tour of European and American stages playing Madame Butterfly in Puccini’s opera, Miura returned to Japan to much fanfare. Though she had recorded tunes from Puccini’s opera in Italian with Japanese record labels on previous returns home from her world tours, this time Miura recorded in Japanese.18 Although print translations of the opera had been circulating for at least a decade, it was the combination of translation and recorded music that led to the viral spread of derivative recordings in the late 1920s through the mid-1930s. The lyrics for the Victor release had been translated into Japanese by Senō Kōyō.19 This was an attempt to repeat the commercial success that Miura enjoyed in July 1922 when, in a banner month for the record industry, her Italian-language recordings of Puccini’s opera with Nipponophone sold over eighty thousand records.20 And yet the translation and domestication of Madama Butterfly was a new beginning. It was after the success of this opera in translation record that the sales of Madama Butterfly–derivative works flourished.

Senō’s Japanese translation of the aria is remarkably accurate in its rough match of the meter of the original, such that Miura’s performative talent for subtly stretching a few syllables here and there to fit the tune is hardly perceptible. The inevitable iterative differences are more noticeable between the translation and the translated. For instance, Madame Butterfly’s American paramour Pinkerton climbs the hill not like “a little speck” (un picciol punto) because he is distant, as in the original Italian, but like “a doll” (ningyō) in Senō’s translation. As a result, the Western male is more than trivialized, almost castrated here. And Butterfly’s teasing by staying hidden so as “to not die” a little (Un po’ per non morire) becomes teasing “so as not to draw attention” (ki no nukenu yō ni). So the association of Japanese culture with the exoticized death of prideful suicide foreshadowed in the Italian disappears in Senō’s translation. The “little dear wife, blossom of orange” (Piccina mogliettina Olezzo di verbena) becomes simply the more generic “my pretty wife, beautiful flower” (Kawaii watashi no tsuma yo utsukushii hana yo).21 Here, the specific floral reference in the original Italian is flattened in the Japanese version, perhaps to downplay the specific aesthetic connotations of white verbena in Japanese poetics, which would point toward death.

Differing from Senō’s work, Miura’s own translations, which contributed to something of a rebirth for her celebrity in the mid-1930s, add a tinge of nationalism. Her translation retains the “my pretty wife, beautiful flower” but adds to it a flavor of the tradition of ancient works using such poetic archaisms as unaji, which appears in both the Nihon shoki (Chronicles of Japan) and the Manyōshū (Poetry Collection of Ten Thousand Leaves) to mean “sea way,” in referring to Pinkerton’s return from overseas.22 In her 1935 Japanese translation of “To Die with Honor” (“Con Onor Muore”), Miura replaces the specificity of the lily and rose flower references (“Fior di giglio e di rosa”) in favor of the more generic “Oh my Boy, more beautiful than flowers” (hana yori mo kirei na utsukushii bōya yo).23 That the local Japanese translations of the original soften some of the exoticism and fetishization of the “foreign” Japaneseness should not be surprising. These Japanese versions shift the emphasis from an infatuation with the exotic other to the tragedy of an impossible love, bringing the story more into tune with the trope of tension between social duties (giri) and individual passions (ninjō) popular on stage since at least the days of dramaturgist Chikamatsu Monzaemon (1653–1725).

All of this copying of Madama Butterfly offered ripe ground for a copyright lawyer to go fishing for lawsuits, and Plage brought the music industry to a near standstill with what was dubbed in the papers at the time the “Plage Whirlwind” (Purage senpū). His pursuit of copyright might be seen as legally correct and just were it not for these facts about the musical origins of Madama Butterfly and the unequal role of Japan in the global order. In terms of geopolitical considerations, the case of Madama Butterfly on Japanese records exemplifies an openness to globalization and international relations as well as a reaction against unequal terms. When confronted with an obstacle, culture production can be ramped up by the enforcement of copyright law rather than simply restricted. For a time when foreign songs were not allowed on records or radio, derivative works were produced that echoed their sounds, content, and aura.

Two Mediums: The Prima Donna and the Copyright Hound

New media produce a demand for material content. To no small degree, much of the “new” content begins as a remediation of material available on older media, so it is no surprise that when the Japanese record industry and radio broadcasting corporations were hungry for material in the 1920s and 1930s, they turned to print media and musical scores. In Japan, print infringements of copyright were rarely adjudicated in the courts or even by capitalist middlemen, not because the law had no such protections (it did) but because particular factors impeded enforcement of the law. These factors included the linguistic inabilities of would-be enforcers, the limited purview of potential litigants, and the crafty hidden work of publishers and translators. However, with the new media of records, broadcasting, and film, culture did not have to be translated to be transmitted and understood as a copy, and corporate producers could hear and see infringements and so sent copyright lawyers around the world in search of new revenue (particularly after the Rome Convention, which solidified music rights in new media). The case of Madama Butterfly in Japan flitted around the problems of who could stand for whom and who could possess what culture, revealing the crises of new media to be a problem of the original medium—of people. This section examines both Plage and Miura as media that transmitted or conjured two realms for a Japanese audience—those of the law and those of opera.

Of Screens and Windows: Open Locks, Paper Walls, and Silhouettes

If Miura’s soundtrack for Arai’s film stands out as a copy that is different enough to have skirted copyright, the visual aspects of the film are in some sense more consonant with a history of visualizations of Madame Butterfly. In its stunning silhouettes, Arai’s short cut-paper animation from 1940 echoes images of the Madame present onstage and in print almost since her inception. Two important elements of the animation’s context are present in John Luther Long’s 1898 short story “Madame Butterfly”: popular Western misperceptions of Japanese marriage and the typical architecture of Japanese homes in the late nineteenth century, which included paper windows. For Long, the illogic of locking a house with paper walls is something of a joke and metaphor:

Some clever Japanese artisans then made the paper walls of the pretty house eye-proof, and, with their own adaptations of American hardware, the openings cunningly lockable. The rest was Japanese.

Madame Butterfly laughed, and asked him why he had gone to all that trouble—in Japan!

“To keep out those who are out, and in those who are in,” he replied, with an amorous threat in her direction.88

Many of the orientalist stereotypes of the era are raised and some subverted in the story. Even though Madame Butterfly is clearly duped by Pinkerton, the American sailor, the tragedy for the reader remains that Butterfly, Pinkerton, and their son Trouble are all unwitting victims: Butterfly of Pinkerton’s carefree privilege and her misunderstanding of his feelings for her, Pinkerton of his carefree privilege and misunderstandings of Japanese culture, and Trouble of his mother’s tragic love. Butterfly’s Japaneseness is conveyed by rendering the dialogue in her voice in a drawl resembling Black minstrelsy. On such gestures of verisimilitude by Long, Arthur Groos writes, “This attempt to confirm verbally the Japanese local colour is anything but realistic.”89 Long’s story is mimetic not to reality so much as to preexisting stereotypical representations of otherness and Japan. Mimesis, to be effective, must seem like that which we already know from other mediated experiences. To seem realistic, or as Groos puts it, to construct “a convincing oriental ambience,” the representation must copy enough to make its consumers swallow any differences between their expectations and the version presented.90 In that sense, the Black minstrelsy evident in the depictions of Japan represented as the other to the white male American in this construct of Japan at the turn of the century.

Japanese houses, and by proxy Japanese loves, Pinkerton assumes, are free, open, and flimsy. There are sinister overtones in the controlling American husband’s attempts to establish ownership and property rights to a woman who can, perhaps, give herself but who cannot be contained. If the tension in this scene is between American and Japanese culture, property rights and human agency, the attempt to control and the inability to do so, this dynamic repeats wider problems of cultural control happening outside the realms of fiction between privileged Western cultural logics and Japanese modernization. The space where this tension exerts itself is particularly located within or behind the Japanese fusuma (movable partitions, variously doors or walls) made of shōji (opaque paper glued to a wooden lattice), the opacity of which makes it a screen to vision but also a screen upon which shadows may be cast (thus giving not an eyeful but at least an alluring sense of what is going on within). The tissue paper walls transform or render the three-dimensional space of the room into a two-dimensional image. In that sense, tissue paper walls are a Japanese medium. Though this screen as shadow-bearer is part of a long visual tradition in Japanese art, in Madama Butterfly it is a Japanese medium constructed through the eyes of the Western male gaze.

Butterfly’s room is, therefore, a female space limned by the opaque walls of male desire that act as a medium similar to Irigaray’s version of the Platonic cave. The missing piece in Plato’s story of the cave, according to Irigaray, is one of origins: Do the cave walls reflect a projection from outside, or is the outside a reflection of what goes on in the cave? This veiled question is the question of the feminine: Is the feminine the origin for the masculine, or is it its shadow? The Japanese home itself is thus doubly feminized by the male gaze as a womb, the movable and removable, opaque and pierceable walls that are the screen upon which images are cast or upon which dreams have always already been cast. Madame Butterfly’s media are those of Irigaray’s theorizations of womb and speculum.91

There is something more specific to this Japanese medium that grows through successive iterations of Madame Butterfly through the novel, drama, and opera: this medium particularly associated with the Madame as the meme became an international phenomenon. Among the hundreds of architectural points and home furnishing details included in art critic Moriguchi Tari’s 1922 book Bunka-teki jūtaku no kenkyū (Research in Cultural Homes, written with Hayashi Itoko) is the following curious entry:

Figure 158 looks as though it is a Japanese style room because it contains a window made by westerners imitating the Japanese taste [Nihon shumi o manete]. This sort of thing is made expressly for facing the garden from the dining room. They call it the “Madame Butterfly window” [ochōfujin no mado, glossed as madamu batafurai uindō]. It seems they think of the well-known Madame Butterfly as a representation of Japanese taste. It combines the tokonoma [display alcove] and shōji [paper and wooden partition commonly used as a door or window cover]. Euro-Americans say the meeting of the vertical and horizontal lines of the shōji gives the sense of a very simple harmony [hijō ni kansona kaichō o kanji saseru]. However, the sliding door is not paper, but glass.92

A Japanese-reading audience is introduced to the “Madame Butterfly window” as an architectural feature of a Western house. It is something both of a boon to the nation and a curiosity to have a thing so obviously related to one’s own culture adopted elsewhere; mimicry is, as the old adage goes, the sincerest form of flattery. But there is a strange sense of loss and displacement accompanying such mimicry. The other thing is not the same. The Madame Butterfly window connotes Japanese style or “taste,” while transforming it into something different. Such cultural appropriation requires translation and explanation for a Japanese audience. Within that definition is an implied critique—Madame Butterfly is not a representative of Japanese taste and the proof of that lies in the window that is not Japanese but perhaps could be labeled “Japanesey.” The Madame Butterfly window may have the shape of paper shōji, but it is glass, making it fundamentally different from anything in the Japanese home.

Reimagined with glass in place of paper, the Butterfly partition is revealed as a medium par excellence. Glass is so closely related with the term media that windows are a standard metaphor for the concept of media.93 A window is a border or threshold that simultaneously separates and connects two environments. Glass allows light to pass through, even as it filters, distorts, and bends its rays. Different styles of glass render different mediating affordances.

One possible source for Moriguchi’s description of the Madame Butterfly window is a June 1915 House and Garden article about the Frank A. Pattison House in Colonia, New Jersey.94

The small telephone room has a high-backed settle with double casements above it, while the garden window of the dining-room is a very interesting “Madame Butterfly” window with sliding sashes and a low platform. Here is a window, suggested by Puccini’s opera, full of Japanese tradition, charmingly picturesque, and yet in harmony with the simplicity and the decorative interrelationships of vertical and horizontal lines that make up the panel-like treatment of the walls.95

The window has little to do with the content of the famed opera; the nomenclature seems meant only to capture an idea of Japanese style. Though labeled as part of “Japanese tradition” in the magazine, the fact that this style window must be introduced suggests it is new (not part of the American tradition yet) for the House and Garden readership. Further, the proximity of its mention with that of a small telephone room suggests something of the modern technical and design innovation of the window style for the U.S. home.

That this window should have been remade to be Japanesey, and rebranded with a cliché name for Japanese taste, is of interest here because it suggests a kind of reverse mimesis—one iteration of a new real. Madame Butterfly, the fictional character (who was by the 1920s more typically heard on record and radio rather than the supposedly unmediated realm of the opera house), has come to mean something in our tangible world. Rather than the representation of, say, the realities of Japan mattering through the medium of Madame Butterfly, what matters or comes to the fore in the example of the Madame Butterfly window is that the medium itself takes up space, becomes part of the built environment, part of our material reality. Madama Butterfly may not represent Japanese taste, but the Madame Butterfly window is a version of the Western taste for Japan made manifest for a suburban New Jersey home.

How does the association of Madame Butterfly with a window work? Melissa Eriko Poulsen argues in her study of the mixed-race identity of Butterfly and Pinkerton’s son, Trouble, that there was a clear association among nineteenth-century orientalists between objects and people:

The desirable Japanese aesthetic—as demonstrated by Madame Butterfly—just as often included Japanese people as it did decorative objects. As Lafcadio Hearn wrote about purchasing Japanese objects, “although you may not, perhaps, confess the fact to yourself, what you really want to buy is not the contents of a shop; you want the shop and the shopkeeper, and streets of shops with their draperies and their habitants, the whole city and the bay and the mountains begirdling it. . . . all Japan, in very truth.”96

As with Hearn’s summation of the desire to possess the Japanese merchandise, the Japanese merchant, and all of Japan, so too, perhaps, the inclusion of the Japanese-style window in an American house displaces desire for the Japanese (displayed by Pinkerton’s “amorous threat”) with the Japanesey. In addition to this logic of association of object with fictional person and displacement of desire for people to desire for objects, such taste for the new Japanese medium likely also came to be named after the ill-fated title character because of a series of associations that became magnified over several iterations or remediations of the story.

The global popularity of the opera alone (besides the short story, play, and various and sundry offshoots and modernizations) has created the myth that from the first performance of Puccini’s work at Teatro alla Scala in Milan in 1904, it has been staged every night since somewhere in the world. There have been numerous versions of the opera, many of which pick up on the association of the character with the window and screen. Gradually, the Japanese window/screen as introduced in successive versions of Madama Butterfly had a profound-enough effect on non-Japanese audiences such that an association grew, imprinting that particular style window with the name.

Though the window and screen play small roles in textual versions of Madama Butterfly, something significant happened with the visual representations of the story. It seems unlikely that Madame Chrysanthème by Pierre Loti in 1887, the short story by John Luther Long titled “Madame Butterfly” published in 1898, or Belasco’s play from 1900, for that matter, alone or in combination had made the window so potently associated with the fictional character to have crept into mainstream architectural consciousness. Although windows appear in Loti’s story, of the five mentioned, only two have people staring through them, and none depicts Chrysanthème herself using the medium.97 Though a window has a minor role in the text of Long’s story (the consul looks out of one), shōji paper walls appear several times (including the scene of configuring locks on them). In one important scene, Long depicts Madame Butterfly poking three holes in the shōji in order to espy Pinkerton. The function of the translucent tissue-paper partition as an opaque demarcation of space is violated to transform it into a makeshift transparent window.98 Belasco’s play Madame Butterfly repeats this scene and also uses the shōji as a screen that the characters hide behind: Pinkerton to hide his presence from Butterfly, and her later to hide her suicide from her son. Shōji is used in the plot as that which one can hide behind as well as see through.

The constraints and possibilities of theatrical staging may have contributed to the association of the screen with Madame Butterfly. Screens were a relatively cheap and easy way to evoke Japan visually, and their presence is more significant in Belasco’s play than in Long’s short story. What does it mean to hide when the hiding is done potentially in plain sight onstage for the audience? It would be a striking visual effect that would have less impact on readers of a print story. Puccini’s opera retains the scene of making holes in the shōji and, confirming their use as a window, adds the maid Suzuki looking through what the script calls the “shosi” at the garden. While these various scenes are important plot points, none of them closely associates the character of Madame Butterfly with the window.

Rather, it seems that the woman and the window became one through successive staged and visual depictions of Madame Butterfly. When Long’s short story, which originally appeared in the Century Illustrated Magazine in January 1898, was expanded as a book published by Grosset & Dunlap in 1903, the volume included several photo illustrations. These visual additions made Butterfly’s architectural environment tangible for the reading audience. Consider that all but one of the sixteen “photo illustrations” include wood and paper partitions as convenient studio backdrops behind the white models posing as characters in the story; in seven of those, Butterfly herself is depicted in kimono in front of the screen. This preponderance of images depicting Butterfly and her portals for an audience with scant experience of Japan associated the character with the Japanese-style shōji window.

From stills of the New York production of Belasco’s play in 1900 and 1906, it is clear that shōji partitions (along with kimonos and flowers) played a major visual role in the staging of those productions.99 Puccini, who saw Belasco’s play in its London staging, drew not only on the plot but also on some of its visual motifs for his opera. These were amplified in the marketing materials for it. Although the original poster by Italian graphic designer Adolfo Hohenstein for the debut performance simply depicts Madame reaching from within a darkened interior for Trouble, who sits blindfolded and basked in light with a door (possibly a fusuma but no window or shōji) between them, the enduring images of the opera came from Hohenstein’s disciple. Soon after, Leopoldo Metlicovitz, who had worked under Hohenstein for G. Ricordi (the rights holders to Puccini’s opera), designed what would become the iconic poster and series of postcards, which prominently feature several windows and screens. Metlicovitz’s classic poster shows Butterfly waiting for Pinkerton in three-quarter view from behind (the position from which the ornate hairstyle of Japanese entertainers was often displayed in Edo-period woodblock prints) gazing out of a balcony window with a lattice similar to shōji. And in Metlicovitz’s series of twelve postcards depicting key scenes from the opera, nine depict the title character and shōji or windows. One striking image shows Pinkerton and Butterfly in amorous embrace silhouetted on a shōji in a marumado (or yoshinomado, a circular window that often looks out onto a garden in Japanese architecture).100 The power of these two images is attested to by the imprint they left behind, having been copied in many visual renderings of Madame Butterfly over the ensuing four decades.

These myriad depictions of a woman waiting by or in front of her lattice-covered window more and more transition to shadow images of the woman behind her screen in later versions. The marumado and silhouette from Metlicovitz’s postcard contributed to the association of Butterfly with a screen. The Mary Pickford film has two marumado. One, shown early in the film, is a small, round window covered with shōji in Butterfly’s dressing room. The second appears as an architectural feature in the garden and has simple bamboo latticework in place of paper. A brief kiss scene depicts Pinkerton and Butterfly embracing in silhouette not behind a shōji screen but in front of a bright, moonlit garden pond not far from the round window.

The 1932 Sylvia Sidney and Cary Grant film version solidified the linking of the images. When Pinkerton and his friend Lieutenant Barton (played by Charlie Ruggles) first arrive at the teahouse, geisha dance with stark shadows projected behind them on shōji. In the following scene prior to the one in which Pinkerton and Butterfly meet face-to-face, Pinkerton is “introduced” to her as she dances behind a shōji and he espies her shadow. As the scene begins, we see Butterfly nonchalantly dancing for herself (rather than for an audience). The film then cuts to her shadows dancing on the screen, then to Pinkerton as he examines a shamisen with the dancing shadow behind him. As he puts down the shamisen, he notices the movement and becomes enthralled with the shadow dance. After a big Cary Grant smile, we get a silver screen full of the shōji shadow dancing. Pinkerton then stands to go open the partitions and meet the real Butterfly for the first time. Later, as she watches Pinkerton’s boat leave the harbor through binoculars, she gazes out of a marumado partially covered by a round shōji screen, and she is in the same spot again months later, now with Trouble, waiting through the night for his return. Finally, in the climactic scene, Madame Butterfly kneels in front of the shōji screen holding her father’s knife, handed to her in the previous scene by her son. When she unsheathes the knife to reveal the blade, etched with the following inscription in Sinitic prosody “為名譽不生 為名譽之死,” the characters dissolve before our eyes through the wonders of animation and become English: “To die with honor when one can no longer live with honor.”101 Visually, Butterfly and the window or screen become one over the decades-long series of depictions.

The visual trope of Butterfly beside or in front of the window/screen suggests that together they act as mirrors of Western male desire, as well as provide the possibility of the tain of the mirror to invigorate and turn that image to other uses. The 1940 cut-paper animated short film is part of this network of visual representations and associations that we might label “the Madame Butterfly window nexus.” The film pays homage to and innovates on these representations, picking up on the marumado used in Metlicovitz’s postcard and the 1915 Pickford film, as well as drawing on the silhouette dance of the 1932 film. Indeed, the decision to use cut-paper techniques to create a silhouette film (kage’e eiga) in the first place seems as much an homage to Lotte Reiniger (the German cut-paper animation master whose orientalist films circulated widely in Japan) as reference to the 1932 scene of Sylvia Sidney dancing behind shōji to Cary Grant’s delight.102 The film does so at a time of highly strained U.S.–Japan relations and growing Japanese nationalism. As such, the short animation turns the story of a product of Western imperialism into an undoing of it.

The story of Pinkerton locking away his butterfly bride is a story of possession—albeit temporary. The Japanese marriage depicted in the story is a mockery of the Western institution precisely because it gestures toward longevity even as it is clear that it cannot last. The story of Pinkerton attempting to lock Butterfly away behind tissue paper walls when he wants her only temporarily seems like a parable for cultural appropriation. It seems perverse to think about Long’s appropriation of Japanese culture on par with Pinkerton’s roguery toward Butterfly only if we fail to understand Marxist expropriation, Lott’s “cultural robbery,” or Walter Benn Michaels’s notion of “cultural genocide” as enabling the privileged “to treat something that didn’t belong to them as if it did.”103 However perverse and misplaced questions of the propriety of culture or cultural ownership are, when culture is appropriated because of the identities attached to it, rather than solely on aesthetic grounds (and the possibility of any appropriation solely on aesthetic grounds too must be suspect), there is a sense of loss from the pilfered culture even as the spread of the culture itself may be a source of pride and even as there is a sense of expansion and inclusiveness within the culture of the pilferer/plunderer. Appreciation and copying of the art of the other is inevitable. It is what human communication does. But not all appropriation is equal. It may also be done well or done poorly. It may be done out of reverence and homage as well as out of mockery and ignorance.

Such cultural appropriation, mixing, and marriage done without respect sets the moral groundwork that enables violence. There should be no easy moral equivalence between injustice to cultural products and the human products of human reproduction. Appropriation of culture and its products play defining roles in how we understand such injustice, in how we come to know it. And, thus, culture needs to be thought together with injustice to human beings. And yet, to borrow or even appear to steal that which was never possible to be owned in the first place can be as liberating an act as a confounding one. Just as Pinkerton was wrong to think the Japanese were too free with their possessions and their morals, and therefore required foreign locks, we might also see the notion that the Japanese lacked a sense of copyright until Meiji modernization and the Berne Convention to be a modernizationist myth. In fact, the Japanese had locks to protect tangible property and laws to protect intellectual property long before Meiji.104 Of course, to assume either such complete difference of Japanese history or a universalism that repeats familiar structures is to play at a kind of orientalism long since critiqued. Between the particularism that legal copyright in Japan has an entirely unique history and the universalism that all creation is copying with a change lies the truth of the complex genesis of a Japanese Madame Butterfly.

Cultural appropriation is, of course, a logical or even natural outcome of the mimetic faculty. If in theory copying and appropriation are natural, it is not practiced on an even playing field. In practice, some cultures (such as the Japanese one in the late nineteenth and early twentieth centuries) are easier for some people (such as white European men) to appropriate than others. Though copyright may attempt to check what is natural about the production and circulation of cultures in the world or the play between media and content (that things are copied in the process), it does not in practice check all copiers equally. We see this borne out, for instance, in the ease with which Puccini copies Japanese folk songs and the problems some of Miura’s performances of Puccini incur. We can see the conflict between mimetic faculty and modern law design in the context of new media playing out in all its guises in this discussion of what might now be called media ecologies around the trope of Madame Butterfly.

The Madame Butterfly window is, thus, both a window and a screen. It is a more or less clear object that affords a view of something else on the other side, that connects two sides while demarcating their separation, that transmits light through degrees of transparency and opacity. But it is also a gridded divider that allows shadows to be cast upon it, bearing a copy, a silhouette of something standing somewhere else behind or in front of the division; it divides even as it connects. The window divides the private room of a Japanese woman from the public space of the harbor it overlooks, Pinkerton from Butterfly, East from West. What it connects, then, are body and background or figure and ground. The body is a medium, as is the environment behind it. In the Japanese architectural book, the Madame Butterfly window is both a false representation of a real Japanese architectural feature and a real Western thing to be described in the world. This tension between two modes of mimesis, one representational—having to do with a degree of fidelity and the question of accuracy of the copy—and the other ontological—marking the unavoidable, ongoing material presence of the medium—is the focus of this book. The Madame Butterfly window is but one example of how the Western desire for a traditional other occupies and transforms reality, creating a sense of a new real.

The space described by the paper walls imprisons Butterfly, and to some degree this architecture (and the infrastructures of racism and orientalism) produces her defining act—suicide. The control of the flimsy marriage and its insubstantial opaque walls can be likened to the control of copyright: it is legally binding but also bendable. Paradoxically, the right created to protect the widows of authors is both a bar to blatant copying and generative of new derivative material. In the story of Madama Butterfly, a mother “widows” her husband and the father of her child; in death, she becomes the posthumous author of a future for the child as yet untold.105 In this sense, Butterfly’s and Trouble’s identities are linked to what Michael Taussig might call the “selfing space” of the Japanese house.106 In contrast to Caillois’s notion of mimetic psychasthenia, where an individual mimes a background to fade into it, Taussig proposes that playful agency can grow in precisely this willful choosing of identity formation. Where Taussig proposes that identity is freeing (when a culture incorporates the cultural products of the other so smoothly that they are hardly recognizable as linked to the “original”), in the opera we have the inverse case where identity is a trap. Butterfly gives up herself to free Trouble of his Japanese association so he can go and live with his father as an American. In this sense, Trouble is freed (within the rhetoric of the story), but does it make sense to talk about Butterfly as free through death? As such, the space seems more in tune with Irigaray’s notion of Plato’s cave as a pure projection of masculinist thought that allows little room for play but out of which play can still make meaningful critique. Indeed, Irigaray’s work on mimesis is close to that of Caillois, who argues that the individual may be “tempted” by their surrounding spaces and eventually dissolve into the background. For Irigaray and Caillois, the temptation by space may be the threat of the dissolution of the individual subject. Butterfly dissolves into the space in order to individuate Trouble from it; in this sense, her room is a selfless-ing space. Fundamentally, this is the reason why, despite all statements of equivalency to the contrary, Miura Tamaki differs from the character of Madame Butterfly: where the fictional projection of Western male desire ends by relinquishing her life, Miura finds her own voice in the animated film because of the power accrued through performing the role over a lifetime.

Disembodied Ownership: Cultural Appropriation Redux

In many senses, this chapter has been simply a work of mimesis, more or less creatively copying and sampling from the work of scholars who came before. Just as many Japanese divas would take up the role of Madame Butterfly after Miura, so too for scholarly innovation. The chapter has directly engaged with issues of (inter)mediation, ownership, and cultural reappropriation that not only permeate the circulation of Madame Butterfly but constitute the very grounds for that circulation and globalization. This media history of Madama Butterfly in Japan provides an iteration of the kind of male-dominated and privileged situation and image of female and femininity that Irigaray critiques from within. At the center of Butterfly is a character who is at least triply othered by gender, race, and class (as an indentured Japanese woman) from the powerful white, male American Pinkerton, and yet Butterfly’s difference as such is a construct. In the early twentieth century, Japanese popular culture appropriated the Madama Butterfly story and domesticated Butterfly through the promotion and circulation of versions and spin-offs. Miura Tamaki, the prima donna at the center of this cultural boom, took on the role, inhabiting the overdetermined space constructed by the desires of the Western male ego and making the image of Asian femininity her own, echoing Irigaray’s problematic (yet powerful) position within philosophy. Within this framework, Miura’s work of performance, translation, and composition shows just how transformative and how limited such work of mimicry and masquerade can be. There is nothing unique about the way this media history around Madama Butterfly in Japan reflects, recasts, and refracts issues of mimesis back at us, but rather because of its particular moment in history, its particular set of orientalist gender dynamics, and its role in the metamorphosis of copyright, the case of Madama Butterfly in Japan helps us to understand what is at stake in media studies.

The vast number of studies of media ecology attest to an unchangeability of aspects of the object of study (that which stays the same, such as the title and characters) even as they purport to show how media have fundamentally transformed the object. The arguments of this chapter may mime media ecologies, perhaps to simply provide one more history of myriad cultural forms in a global story. This chapter repeats Caillois’s notion of mimicry, whereby the individual melds with its surroundings. And, thus, it becomes simply one more iterative study in a vast array of faulty media studies of mediation and remediation, but with a difference: it shows how mimicry and subsumption of identity itself can be a form of active being in the world. But if the practices of Irigaray and Miura truly garner agency through mimesis of dominant and even oppressive forms (and I think they do), then this chapter’s repetition of other scholarly efforts to track the wider media ecology of Butterfly should have worked to perform difference through similarity, explicating how the stories of the madame and the copyright hound, the window and the screen, are not simply another media ecology but perhaps a more narrow media entomology.

Madama Butterfly in Japan proves through its infinite mimicries once again (as if we needed it) that the ideas of an origin and essence to a cultural phenomenon are mythical at best. At the same time, the booming interest in the opera after Miura’s return to Japan and the subsequent recording and filmic multiplicity of versions show us how the mimetic faculty (as a compulsion and as innate mechanism) is itself tied to our fetish for and fear of new media. We can see this through navigating the odd relations of recording and broadcasting music, copyright laws, and the identities of human media like those of Plage and Miura; though this story is mainly about the mediation of music in the early twentieth century through translation, performance, recording, broadcast, filming, copyright law, and human beings, it is applicable to our contemporary concerns for digital rights management software and ownership of cultural heritage today.

Cultivated and culturally constructed in Western Europe, copyright law forced the definition not only of what constituted legal artistic borrowing and illegal copying but also of melody, media, and music itself. Indeed, the legal battles to define music through media in the wake of records and radio relegated performance to but one instantiation of “music.” Radio broadcast, musical scores, and records the law would define as on a continuum with, but somehow separate from, original performance.107 Though amended in 1971, 1999, and 2000, copyright law in Japan continues to position JASRAC as the primary intermediary between rights holders and users of cultural material. The 1971 revision of the law to account for magnetic audiotapes and record rental shops recognized “duplication rights,” a change that occasioned the development of karaoke and, later, of minidisc technologies.108 Yet were it not for the underlying centralized intermediary system, the trading of rights that enabled karaoke would not have had the content for duplication in the first place. In 1999 and 2000, copyright law in Japan was revised to reduce the power of JASRAC in the wake of the new media issues that the internet seemed to bring. Even though formally replacing intermediary rights with the Act on Management Business of Copyright and Neighboring Rights, the new law still allows a powerful place for JASRAC to continue; for instance, the rights-holding organization was integral in 2007 in working to get thirty thousand YouTube videos taken down for copyright violations.109 So the legacy of Plage and Miura continues in the Japanese mediascape today.

The continued importance of intermediation in Japan reveals that the rise of digital media and their inevitably ensuing remediation give rise to anxiety not only over a potential loss of capital accrual but also over the very thing/message/information/music being remediated. And this ontological anxiety over what constitutes content (that was a major part of early music copyright law) continues today with music for which an originally recorded performance putatively does not exist. Virtual music that is first and foremost mediated by computational machine (without a recording of a performance as such) seems at first blush to raise an unprecedented specter of music produced without human labor. But recent controversies over copyright of the simulated and synthesized human voices sold by Yamaha as their VOCALOID software continue to place old questions of proxy ownership, media, and person at the forefront of Japanese copyright disputes. Just because the voice seems truly disembodied, it does not mean that race and body are not issues in VOCALOID circulation. We find their return in VOCALOID singer profiles and, indeed, in the way copyright seeks to maintain a space for the software production company to reembody or reincorporate the virtual voice in a corporate legal financial space. In VOCALOID, we can see the legacy of a content–media nexus (one already long evident in the Butterfly media histories discussed here).

For profit in the capitalist cultural marketplace, this means that entropy reigns and copyright continually must play catch-up to order the fact that cultural products spin away from their putative producers but that production ultimately lies beyond the limited powers and agency of a particular producer. This means that cultural production runs ahead of capitalism at every turn because of the flight patterns of mimesis.110 Copyright is an interesting case of trying to reconnect what has been alienated and hidden—the labor of the artist or cultural producer. And yet, it runs counter to the nature of cultural production, which is necessarily to copy. The problem really is not that copyright generally seeks to remunerate the labor done but that it seeks to accrue capital over the long run of time for the labor done once long ago that was generally already remunerated.