Chinese Ears, Delicate or Dull? Toward a Decolonial Comparativism

In September 1725, the Comédie-Italienne in Paris put on “one of the most exceptional novelties” involving two Indigenous dancers from Louisiana, one dressed as a chief wielding a calumet, the other a soldier carrying a bow and quiver.¹ Legend has it that in the audience was Jean-Philippe Rameau, who published his impression as the keyboard rondeau “Les sauvages.”² Later adapted as the final entrée of his opéra-ballet Les Indes galantes (1736), the piece continued to evoke the Amerindians until the entrée’s last Opéra billing in 1773, outliving both Rameau and New France.³

Rameau would probably have been surprised to learn that his rondeau would eventually reach the “Indies” on a different side of the globe: the Qing Empire (1636–1912), which ruled Manchuria, Mongolia, China proper, and Tibet at the time and was marching its army into Central Asia.⁴ The Qing debut of Rameau’s rondeau took place in the early 1750s at the house of the French Jesuits, located at the Church of the Savior in Beijing’s segregated Manchu quarter.⁵ Performing it was a Jesuit missionary from Toulon, Jean-Joseph-Marie Amiot (1718–93). Sent to Beijing in 1749 and remaining there until his death, Amiot was arguably the most authoritative source of information on China for late eighteenth-century European thinkers, and the letters, translations, and studies he dispatched to Paris laid a foundation for European sinology. Of the several manuscripts he drafted on Chinese music, Mémoire sur la musique des Chinois (1779) was the first monograph to be published on the topic in Europe.⁶ In his preface to the Mémoire, Amiot recalled that on first arriving in Beijing, he had entertained some Qing scholar-officials by playing various “sonates” and “airs,” including “Les sauvages.”⁷ Socializing with the Qing’s elites had long been an evangelical strategy of the Jesuits, who hoped to win the tolerance (if not the conversion) of these local powerbrokers by impressing them with European science and culture.⁸ Yet the response to Amiot’s musical showcase was apathetic. One listener remarked, as paraphrased by Amiot, “that [French] airs [are] not at all made for [Chinese] ears, nor [Chinese] ears for [French] airs.”⁹ Another added, as he quotes directly, that “the airs of [Chinese] music … go straight from the ear to the heart, and from the heart to the soul … but those you have just played do not have this effect on us.”¹⁰ Amiot remembered this encounter as a turning point: it was from this moment that he gave up on proselytizing for European music and began a study of Chinese music that would culminate decades later in the Mémoire. Driving this change was a new conviction about the importance of relativism: “We must … become accustomed to the ideas of the Chinese,” he writes in the preface, “we must attune ourselves to them, so to speak, if we want to hear them.”¹¹

There is reason to suspect that Amiot fabricated this uncorroborated story, especially given that it would not be the only untruth he told concerning the history of the Mémoire. At the end of the preface, he listed the titles of sixty-nine Chinese books he claimed to have read;¹² in his autograph manuscript of the work, these titles are presented in Chinese characters (see figure 1).¹³ According to Amiot, “All these books were carefully collected during the Ming dynasty [1368–1644/62]” and “abridged and published during the Wanli reign [1572–1620]” in an anthology entitled Lüshu cankao (Books on music theory for reference).¹⁴ Such an anthology never existed. No one seems to have noticed that Amiot simply copied these titles from the “Lüshu cankao”—in effect, the bibliography—of the music theory treatise Lülü jingyi (Essential meaning of pitch pipes, 1596) by Zhu Zaiyu.¹⁵ As shown in figure 2, Zhu listed the same sixty-nine books in his bibliography in the same order and with the same annotations and mise-en-page (ten titles per row) as would Amiot in his Mémoire manuscript two centuries later.¹⁶ Because Amiot’s proficiency in Chinese left little room for genuine miscomprehension, there is no explanation other than that he plagiarized Zhu’s bibliography in order to demonstrate his own immersion in Chinese sources, having criticized other European writers for lacking such insider knowledge earlier in the preface.¹⁷

Amiot’s appropriation of Chinese voices and sources in order to advance himself in the European Republic of Letters certainly amounted to the same kind of Orientalism as more explicit Enlightenment ventriloquizations of Others, such as Montesquieu’s Lettres persanes (1721). Yet such a critique is not what I am pursuing here. As Kate van Orden and Erith Jaffe-Berg have shown, applying the postcolonial framework to the early modern Mediterranean and Eurasia risks assuming a then nonexistent Western hegemony while erasing non-Western agencies.¹⁸ Instead of measuring Amiot’s power vis-à-vis that of those Qing officials to whom he gave voice, I seek to reflect on the sensory orientation of his formulations. Notice the verbal puns in the phrases quoted above: in promising to “hear” the Chinese, Amiot showed that he understood cross-cultural exchange as an auricular process. He even reached this understanding, reportedly, by “attuning” himself to the words of his interlocutors, those who had reduced the difference between Chinese and French musics to a difference between their ears in the first place.

But how did Chinese and French ears differ? Amiot does not say in his Mémoire, though he pursued the auricular comparison on two other occasions—with diametrically opposite conclusions. The first was in the earliest extant piece of writing on music that he sent to France, “De la musique moderne des Chinois” (ca. 1754).¹⁹ According to Rameau, who discusses it in his Code de musique pratique (1760), the autograph manuscript of this treatise was publicly available at the Académie des Inscriptions in Paris at some point during the 1750s, though it was never published.²⁰ Much of “De la musique moderne” is copied from Lülü zhengyi (Orthodox meaning of pitch pipes, 1714), the Qing’s official treatise on tuning and organology. Toward its end, Amiot recounts as he would again twenty years later his failed musical evangelism—except that here he specifies what it was in European music that supposedly offended the Chinese: harmony. As if describing a skirmish in the ongoing Querelle des Bouffons, Amiot uses Rameau’s concept of basse fondamentale or triadic root progression to characterize European harmony,²¹ while the alleged complaints from his interlocutors reek of Rousseauism: the Chinese dismissed harmony as a noisy artifice that upsets the natural beauty of melody.²² To such nonsense Amiot had no intention of “attuning” himself. The Chinese were unable to appreciate the subtlety of European harmony, “having only their stupid or dull organs by which to judge their sensations.”²³ He attributes the defective “organes auditifs” of the Chinese to climate and recent history. When the Qing conquered China in the 1640s, Amiot explains, they forced all Chinese men to adopt the Manchu hairstyle, which entailed shaving off all but a braid of hair at the back of the head, leaving the ears exposed to the harsh northern Chinese climate.²⁴ On the basis of his daily readings of the thermometer, barometer, and hydrometer installed at the Jesuit house—data shipped to Paris and now held at the Bibliothèque nationale de France²⁵—Amiot observed seasonal fluctuations in temperature and humidity to be more extreme in Beijing than in Europe.²⁶ Oscillating between wintry dryness and summer humidity, Chinese ears became desensitized, ergo “stupid or dull.”

The second comparison came decades later in a letter of July 26, 1780, to his patron, French minister Henri Bertin.²⁷ In its last paragraph—censored by Bertin prior to publication (see figure 3)²⁸—Amiot writes scornfully, “the auditory nerve of European ears is to that of the ears of the Chinese, Egyptians, Greeks, and others of all peoples of the universe, both ancient and modern, what a belt strap is to the chanterelle of a violin.”²⁹ With tables turned, Amiot tells Bertin that “your Europeans” enjoy harmony not because of auricular refinement but the opposite.³⁰ “On the basis of several experiments I had performed on corps sonores, I imagined that in a simple melody, each note … was always accompanied by its natural harmony”;³¹ unlike the Chinese and other peoples, Amiot argues, modern Europeans cannot hear these natural harmonics and must remedy their defective nerves with artificial harmony in their music. As if changing sides in the Querelle des Bouffons, Amiot’s new assessment parallels Rousseau’s Essai sur l’origine des langues (1781), which dismisses harmony as a “supplement” needed only by modern Europeans to compensate for their distinct lack of sensibility.³² And yet, unlike Rousseau’s Essai and Amiot’s own thoughts of decades earlier, the 1780 letter to Bertin takes a hard physiological turn: “I had attributed the disagreeable impression experienced by [Chinese] ears on hearing our harmonies to their interior configuration. I imagined that the cavity of their ears might be parabolic or elliptical in shape, and that the sounding together of several different tones would thus produce not a harmonious sound but a sort of rumbling.”³³ Here, neither climate nor history plays a role in creating harmony-hungry European ears or harmony-hating Chinese ones: the physical difference between the structures of their inner ears seems as innate as that between belt straps and chanterelles.

This article wrestles with the dramatic flip-flop between these two comparisons by Amiot of European and Chinese ears. Not, however, in an attempt to explain it away or really to explain it at all. I do not seek to probe into Amiot’s sources or cross-examine any contacts who could have prompted him to change sides. Such heavy-handed forensics are not entirely appropriate for what were merely cursory remarks by a writer many of whose more thought-out volumes still sit unvisited in acid-free boxes or 35 mm film reels. Besides, my interest lies not in what Amiot wrote about Chinese ears, much of which can be dismissed as racist and inaccurate, but in how he wrote about them—that is, comparatively, a methodology that has been underexplored in anglophone musicologies in recent decades out of concern for difference and emic narration. And rather than extracting any methods of comparison from Amiot’s jottings, I reckon it more fruitful to proceed differently: why not write with the Jesuit and his auricular comparisons, instead of writing about them?

In what follows, I take Amiot’s shifting comparisons of European and Chinese ears as a heuristic—even better, a cue—and perform in lockstep a comparison of my own. Because I am moving in response to Amiot’s footsteps rather than merely retracing them, the places where I land will often leave him out of view. Yet my comparison follows and extends Amiot’s in juxtaposing the two ends of Eurasia in the late eighteenth century with respect to hearing, despite the geopolitical and disciplinary distance still separating them today. Just as Amiot did not shy away from conjuring disproportionately larger questions on universal history and the very possibility of comparison in his paltry musings, I too springboard off those musings to stage parallels between Europe and the Qing as two centers of empire building. These parallels show that an extractivist ideology of listening and voice operated at both ends of Eurasia. Through analyzing this shared ideology, I hope to illustrate the decolonizing and deimperializing potentials of comparativism, specifically with respect to what the xwélméxw/Stó:lō scholar Dylan Robinson describes as “hungry listening” in the context of settler colonialism: a positionality of listening that, akin to colonial resource extraction, treats “Indigenous culture as a resource that is there to be mined.”³⁴

My comparison begins by tracing Amiot’s jottings on ears to similar remarks by Benjamin Franklin, who was connected with the Jesuit through Henri Bertin. This connection prompts me to contextualize Amiot’s discussions of Chinese hearing within contemporary European discourses on song and language. Here, the writings of Johann Gottfried Herder (1744–1803) are particularly revealing. Though Herder concurred with Amiot’s later assessment of Chinese ears as distinctly sharp, his disparagement of China specifically on account of this sharp hearing surprisingly resembles his denigration of Deaf people.³⁵ This apparent contradiction lays bare the dialectics of Herder’s phonocentrism—his famous notions that hearing is “the true gateway to the soul” and that “speech alone has rendered man human.”³⁶ The voice (“phono-”) privileged in this phonocentrism is neither a sonorous, bodily voice nor a silent, disembodied one. Rather, it is a voice rooted in the body yet constantly removing itself—indeed, extracting itself—from that body while moving toward transcendence. The dialectics of such a “self-extracting voice,” as I term it, underpinned the pluralist and progressive visions of the post-Thermidorian French elites and of Georg Wilhelm Friedrich Hegel (1770–1831). In the liberal society they envisioned in contrast to the Reign of Terror and the alleged culture of conformity in China, recognition of one’s subjecthood would depend on opening oneself up to the state’s extractive ears in order to be heard. Alphabetic writing was deemed indispensable for such a society, because its phonographic or “voice-writing” letters allowed the state to extract voices from its subjects and to compel them to speak to and only to itself.

The second part of my comparison pivots to China. As well as elucidating Europe’s extractive phonocentrism, contextualization of the shift in Amiot’s thinking on Chinese ears within European discourses reveals a concurrent shift in his understanding of the ontology of Chinese music. This latter shift saw Amiot toward the end of the 1770s emphasizing popular songs, particularly the qupai, or stock melodies, used in Chinese theater, as a defining characteristic of Chinese music. Taking my cue from Amiot’s discussions of Chinese theatrical songs, I begin by examining the use of qupai in Qing court ritual theater. As many scholars have noted, incorporating Kunqu theater—popular among the Chinese gentry and the scholar-official and merchant classes—into court rituals was a political act on the part of the Qing’s Manchu elites, who ruled China as an ethnic minority. What has not been noted is a seismic change concerning qupai in eighteenth-century court theater. This change not only broke with the pervasive use of these stock melodies according to Kunqu convention but transformed the phenomenology of listening in the theater, revealing that an extractive phonocentrism also underpinned Qing court theater and the Qing imperium at large. Eighteenth-century Qing-imperial ethnographies of songs and languages further illustrate this phonocentric ideology at work. As my analysis of these ethnographies shows, subjecthood in the Qing Empire depended on being heard as a voice addressing the transcendent imperial ear. And the empire also used phonography as a technology by which to extract voices from its constituencies and consolidate its multiethnic realm.

The phono-extractivism shared by the two ends of Eurasia allows me to reflect on comparativism in the concluding section of the article. Through cross-referencing the two cases of hegemonic voice extraction, I argue that comparativism, insofar as it creates incommensurable spaces between different centers of extractive listening, is an indispensable method of decolonization and deimperialization. After building my case for comparativism by engaging with Indigenous studies and postcolonial studies, I conclude by returning to Amiot, whose comparisons of ears initially inspired my comparison of phonocentrism. Drawing on Dylan Robinson’s “apposite methodology,” which treats writing as a strategy for sharing space, I argue that “writing with” the experience of navigating the margins between different empires, such as Amiot’s, may subject those imperial extractions to a kind of comparison that brings them to a halt.

On June 2, 1765, Benjamin Franklin wrote to the Scottish philosopher Henry Home:

Some thirteen years later, on May 26, 1778, two months after France declared war on Britain, Franklin was invited by the French minister Henri Bertin to his estate north of Paris, together with John Adams.³⁸ When there, Franklin almost certainly discussed his theory of the inner ear with Bertin, who must have relayed it in a letter to Amiot. Only thus could the Jesuit—who had by then been in Beijing for almost three decades—have written in the censored part of his abovementioned letter to Bertin of July 26, 1780,

The apparent loss of Bertin’s letter makes it difficult to assess whether Amiot took inspiration from Franklin in theorizing why the Chinese can hear harmony in monophony whereas Europeans cannot. Even so, regardless of any actual transfer of ideas, Franklin’s and Amiot’s auricular models present many parallels regarding what was a prevalent topic in late eighteenth-century European discourse on music—what may be retroactively referred to as “folksongs.” In his letter to Henry Home partially quoted above, Franklin sought to explain why “a plain old Scotch tune” delights everyone, while music “in the modern taste” triggers “no signs of pleasure” in concert halls.⁴⁰ The ear’s ability to retain notes, he maintained, provides the answer. Because “the construction of the old Scotch tunes … is that almost every succeeding emphatic note is a third, a fifth, an octave, or in short some note that is in concord with the preceding note,” the successive notes of each melody are stacked and transformed into consonant harmonies in the inner ear.⁴¹ This natural “union of melody and harmony” convinced Franklin that “the Scotch tunes have lived so long, and will probably live forever.”⁴²

Although his 1780 letter to Bertin did not discuss Chinese music per se, Amiot’s later writings saw a similar shift in favor of “plain old” songs. Up to the completion of his Mémoire in 1776, Amiot had largely been studying Chinese music, particularly lülü, or music theory, in dialogue with Rameau and Pierre-Joseph Roussier, author of the Mémoire sur la musique des anciens (1770) and editor of Amiot’s Mémoire.⁴³ For Rameau, the pentatonicism of Chinese music confirmed the universality of European harmony founded upon the “triple geometric progressions” (1:3:9:27:81).⁴⁴ Roussier expanded Rameau’s theory into a universal genealogy of musical scales descending from Noah through the Egyptians, Chinese, and Greeks to modern Europeans.⁴⁵ Increasingly sinophile, Amiot rejected such Eurocentrism and credited the Chinese with inventing everything, from Pythagorean tuning to ballooning.⁴⁶ But he too sought after a universal music history, albeit one centered on China rather than Europe. His Chinese sources—studies on tuning and organology in the context of Confucian ritual music—were always deemed commensurable with European ones.

This assumed commensurability appeared less confident by the end of the 1770s, as Amiot began to portray China as distinctly dominated by popular songs, unlike Europe. To respond to the questions raised by Roussier in editing his Mémoire, Amiot sent a “Supplément” in 1779, attaching to it a collection of forty-one songs, “Divertissements chinois.”⁴⁷ The collection drew not from the ya (elegant) or highbrow music of Confucian rituals, the repertoire that engendered Chinese music theory, but from the more su (popular) or lowbrow songs of Chinese theater, supposedly “the finest musical airs that have been sung in this very ancient and very large empire … for many centuries.”⁴⁸ These were not his first dispatches of “airs chinois,” but, differently from his practice in the 1750s, Amiot did not transcribe their melodies into standard Western notation or eliminate their lyrics, often the only thing “notated” in Chinese song prints.⁴⁹ Concepts and tools for writing down European music could no longer do Chinese songs justice.

To many in the Republic of Letters, Franklin’s Scotch tunes and Amiot’s airs chinois would have sounded the same. As Thomas Irvine shows, both were heard as the songs of ancient, closer-to-nature musical Others, thanks particularly to their purported pentatonicism, in distinction to European art music.⁵⁰ Little might seem to have changed since the mid-century, when Rousseau’s Essai sketched a similar history of the human voice as deteriorating from ancient non-European naturalness to modern European artificiality.⁵¹ Nonetheless, by the time Amiot was vouching for Chinese ears and airs, the tide was turning in Europe, especially in the German reception of French scholarship. According to Johann Nikolaus Forkel (1749–1818), for example, writing in the 1780s, Chinese music—his main source for which was Amiot’s writings—still represented an aesthetic of “edle Einfalt” (noble simplicity). Yet his attitude toward this aesthetic was much more ambivalent, given the increasingly prevalent view that the historical development of music from ancient simplicity to modern complexity represented not degeneration, as Rousseau and Amiot might have argued, but progress.⁵² According to Irvine, this cooling of sinophilie reflected a growingly triumphant and industrialized European self-image, which was contrasted to the steadily colonized rest of the world.⁵³

As Gary Tomlinson argues, this new version of Eurocentrism did not merely flip the script on Enlightenment primitivism/Orientalism: it also heralded a narrative of incommensurability between European “music” and non-European “songs” that would later define the musicology vs. ethnomusicology divide. What characterized this incommensurability, according to Tomlinson, was neither the denial of coevality (a common mid-century trope) nor musical and physiological difference per se (Amiot in 1779 and 1780) but rather abstraction. European (instrumental) music was characterized by a progressive abstraction or autonomy from linguistic meaning, social functions, and even historical time itself, while non-European songs, lacking such abstraction, appeared forever submerged in the here and now.⁵⁴ Tomlinson also draws on Forkel, specifically his Allgemeine Geschichte (1788), to exemplify this incommensurability narrative. Whereas Rousseau had used alphabetic writing, harmony, and musical notation alike to pinpoint modern European degeneration,⁵⁵ Forkel argued that only civilizations reaching the highest level of abstraction were capable of using conventional alphabetic letters to represent conventional speech sounds. Those of a “less developed mode of abstraction”—ancient Egyptians, modern Chinese, and much of the non-European world—could use only nonalphabetic pictographs to mimic the appearance of objects.⁵⁶

In light of this European music vs. non-European song dichotomy, it is easy to attribute the colonialist baggage of modern musicologies to the privileging of “music” over “songs,” abstraction over contextualization, and writing and notation over voice and gesture. This verdict would mean rooting out musical autonomy as a Eurocentric ideology and embracing the orality and embodiment of “song, [a] universal corollary of the human propensity toward language,” as the true universal for a global music history.⁵⁷ Yet it would be mistaken to assume that Eurocentrism or any other forms of hegemony can arise only from privileging the autonomous and abstract (“music”) over the material and bodily (“songs” or “sounds”). In fact, late eighteenth-century European narratives of incommensurability privileged abstraction not as the opposite or to the exclusion of the body, but as a dialectical process that both presupposes and outgrows the body. I refer to this process as “extraction,” to emphasize not a static attribute but a dynamic action of “drawing away from,” reflecting the etymology of “abstraction.” Colonizing the extraction from the bodily to the spiritual meant that Europe—excluding its fringes relegated to “the folk”—no longer had to choose between ancient simplicity or modern sophistication. The ruling and lettered classes of Europe could monopolize for themselves both body and mind, both primitivism and development, and both nature and industry, leaving no room for the savages or the Orientals except the impossible assimilation through extraction.

Such extractivism is best illustrated by an echo of Amiot and Franklin in the works of Herder. Though more polemical than representative of his time, Herder inadvertently revealed the dialectics of extraction in relation to the voice through his writings on China and Volkslieder. Regarding the Chinese, Herder concurred with Amiot’s 1780 diagnosis—with a twist. “All accounts agree,” he writes in his Ideen zur Philosophie der Geschichte der Menschheit (1784–91), “that the mungal nations [Mongolische Völkerschaften] … are distinguished by an acuteness of hearing.”⁵⁸ As proof, he uses a trope dating back to Athanasius Kircher (1602–80) to characterize the tonal Chinese language as song-like.⁵⁹ “The auditory organs of a mungal alone could [form] a language out of three hundred and thirty syllables, distinguished … by five or more accents,” whereas “a European ear … can with utmost difficulty … accustom [itself] to this forced syllabic music [hervorgezwungene Silbenmusik].”⁶⁰

This “forced syllabic music” shows that, for Herder, the sharp Chinese ears are nothing to admire. Instead, he uses hearing—the first specific trait analyzed in his portrayal of China—to challenge the Enlightenment fantasy of China as a paragon of virtuous governance. The chapter on China in the Ideen begins with a litany of astronomical numbers: “25200000 husbandmen paying taxes,” “3158 stone bridges,” “2796 temples,” “10809 ancient edifices,” and so on.⁶¹ These culminate in the thousands of syllables of their language and the “eighty thousand compound characters” of their script.⁶² To Herder, such sophistication reveals an unfavorable “genetic character” of the Chinese: “acuteness of the organ of sense, and uninventive ignorance of mind.”⁶³ In particular, the “learned industry” required to distinguish their numerous speech sounds and written characters bespeaks an “artificial mode of thinking.”⁶⁴ “As the Chinese are immoderately fond of … the neatly painted lines of their intricate characters, and the jingle of fine sentences, the cast of their minds resembles … these characters and clink of their syllables,” with “so little relation to nature, so little feeling of internal satisfaction, beauty, or worth.”⁶⁵

Herder’s critique of Chinese artificiality resembles Rousseau’s critique of Europe and civilization in general—except that for Herder it is paradoxically natural for the Chinese to be artificial. Note that he repeatedly calls them “mungals” (Mongolen), a demonym commonly used in race theories (cf. “Mongoloids”). While Herder’s legacy in scientific racism is too consequential to unpack here,⁶⁶ I note that labeling the Chinese “mungals” allows him to demonstrate the immutability of national character. A nation is forever tied to its place of origin, because it was there that its “living organic power” first expressed itself into the nation’s bodies, thinking, and customs.⁶⁷ Though they had long migrated south from Mongolia, their homeland according to Herder, “the Chinese … have kept themselves, like the [J]ews, unmixed with other people” and continue to behave like “wandering mungals.”⁶⁸ Their herd mentality from the steppes continues to give them a “talent of imitating [Nachahmung]” in following the orders of their parents and superiors and to deprive them of the nature-given “freedom of action.”⁶⁹

Since his unflattering account of the Chinese begins with their ears, Herder might seem prejudiced against hearing. Any reader familiar with Herder’s work knows this to be unlikely. In fact, Herder’s portrayal of the sharp-eared Chinese resembles his view of the diametrically opposite group, Deaf people. “[A deaf person] imitates whatever his eye sees,” remarks Herder, going on to refer to “more than one instance of a person born deaf and dumb, who murdered his brother in consequence of having seen a pig killed, and tore out his bowels with tranquil pleasure, merely in imitation [Nachahmung] of what he saw.”⁷⁰ Just as the Chinese can hear a profusion of syllables, only to mindlessly follow their herd leaders, so “they who are born deaf and dumb, though they may live long in a world of gestures … act analogously to what they see and do not understand.”⁷¹

Seemingly contradictory, Herder’s identical dismissals of the sharp-eared Chinese and Deaf people as unthinking imitators reflect a single dialectical understanding of hearing first articulated in his Abhandlung über den Ursprung der Sprache (1772) in rejection of French sensationism. Influenced by Lockean empiricism, mid-century philosophes such as Condillac and Rousseau traced all languages and knowledge to the sensations, arguing that the first humans communicated with a “language of actions,” using mimes, cries, and drawings.⁷² Many eighteenth-century French writers regarded sign languages and Chinese writing as remnants of this original language: both seemed to represent abstract concepts by breaking them down into sensible components imitable in pictures or gestures. Hence, unlike the arbitrary, convention-based signs of French (spoken syllables and written letters), sign languages and Chinese characters appeared to give every concept a natural signifier and definition.⁷³ They afforded the sensationist fantasy later pursued in the early French Revolution, a society whose language and laws all derive from natural sensations alone.⁷⁴

Herder counters that reason and language arose not from the sensations but from extraction from them. According to Nigel DeSouza, Herder rejects the sensationist theory that humans originally lived as animals until geographic changes activated their “potential of reflection”; for Herder, the human soul differs from those of animals not in its additional potential but in its qualitatively different organization.⁷⁵ Whereas each animal is embedded in one “circle” of life toward which all its senses are sharpened into instincts, humans “have senses for everything” and are therefore not embedded in any singular circle.⁷⁶ As soon as the first human came into existence, there was not only “a whole ocean of sensations which flood[ed] [their] soul” but also a “Besonnenheit,” or reflective awareness, that had always already been detaching itself—extracting itself—from the senses.⁷⁷ It is through this self-extracting awareness that the human soul had always used an internal mental language of “characteristic marks” to distinguish the sensations and reflect on them.⁷⁸ Later externalized into sensible language, these mental marks were destined to become sounds, because hearing is “the middle one of the human senses”: it is defined by neither nearness nor distance, neither obscurity nor brilliance, neither warmth nor cold, neither evanescence nor permanence, neither ineffability nor obviousness, neither innateness nor learnedness.⁷⁹ Detached, like the soul, from any particular sensuous circle or orientation, hearing is “the true gateway to the soul.”⁸⁰

Human Besonnenheit is therefore coterminous with what may be called the voice: internal reflections manifestable as sounds. “Speech alone has rendered man human,” argues Herder.⁸¹ “By speech alone the eye and ear, nay the feelings of all the senses, are united in one, and center in commanding thought.”⁸² Thus, though Herder’s theory of humanity is certainly “phonocentric,” the “phono-” here is neither a bodily voice of sensations, nor a disembodied voice deprived of sensations. Rather, it is a self-extracting voice: like Besonnenheit, such a voice is rooted in the senses yet is also constantly extracting itself from the senses, transcending and reflecting on them. It is always presupposing a sensuous body, only so that it can outgrow and negate such a body. Only thus could Herder deny free thought and action to both the Chinese and the Deaf: the Chinese are so sensuous in hearing and uttering their song-like language that the transcendence of their voice becomes neutralized, and the Deaf are deprived ab initio of the very sense that the voice must presuppose in order to negate.

This dialectical self-extraction is further manifest in Herder’s seemingly contradictory writings on “song” and “music.” Herder famously celebrated Volkslieder for embodying the “living customs” of the Volk and contrasted this popular orality with the learned sophistications of the cosmopolitan philosophes, whom he disdained.⁸³ Nonetheless, he simultaneously identified the telos of music not as its continuing immersion in communal song, but as its autonomy from it. “Music has long adhered closely to dance and song,” argues Herder in Kalligone (1800): “it was the celebration, the social context, the form, and the movement of the human performers that commanded music.”⁸⁴ Destiny dictates that music must “separate itself from its siblings, words and gestures, in order to establish itself as an art in and of itself.”⁸⁵ Therefore, what Herder longs for is a music that is embodied in the singing-dancing masses yet has always already been outgrowing them. Because song is the root of music and autonomy its destiny, true Volkslieder always germinate absolute music, just as true absolute music always presupposes authentic Volkslieder. Absolute music growing out of the singing-dancing Volk epitomizes the voice extracting itself from the senses.

This voice qua absolute music bore significant political affordance for the expanding European states and empires, most evidently in the parts of Hegel’s Enzyklopädie der philosophischen Wissenschaften im Grundrisse (1817) that are related to China. Unlike Herder, Hegel focuses not on the Chinese “forced syllabic music” but on their nonphonetic writing. This distinction proves to be null, because Hegel also conflates the Chinese and the Deaf. Like eighteenth-century observers of sign languages, Hegel understood Chinese characters to be either mimicking a sensible object via a picture, or representing an abstract concept via a compound that breaks the concept down into sensible components that can be mimicked in pictures.⁸⁶ Consider, for example, the Chinese character 男, which combines 田, a picture of paddy fields, and 力, muscle fibers, to mean “male.”⁸⁷ For Hegel, such a character cannot accommodate “the progress in the cultivation of our thoughts”⁸⁸: 男 can be used to signify “male” only if masculinity continues to be defined by agrarian labor. “A hieroglyphic writing would require a philosophy as stationary as is the civilization of the Chinese”⁸⁹—and, I would add, as uniform as is the Chinese, since in Hegel’s logic, “hieroglyphic writing” presumes a consensus on all definitions.

Opposing such hieroglyphic fossilization is the voice’s self-extraction, which according to Hegel comprises a double negation. Whereas hieroglyphs signify through sensuous mimesis, speech sounds negate such mimesis by being used as “intrinsically senseless externalities” that normatively have only conventional or arbitrary (as opposed to onomatopoeic) meaning.⁹⁰ But sounds can negate sensuous mimesis only because they also negate themselves: not only do sounds disappear the moment they are heard, but being used as arbitrary signs also renders all their sonorous resonances “external” or irrelevant to their signifying function.⁹¹ Hence the telos of the voice-in-speech—always already negating its own sounds, which have in turn been negating sensuous mimesis—is speech-without-sounds, or alphabetic writing. Having extracted itself from even sound, alphabetic writing allows people with different sensory experiences to use the same signs without having to agree over their meanings. Whereas 男 requires masculinity to be uniformly experienced as agrarian (田) labor (力), and “male” as a spoken word still requires the sharing of a sensuous experience of uttering and listening, “male” as a written placeholder allows debate as to what masculinity is without any shared empirical or epistemic grounds. Alphabetic writing accommodates vastly different opinions and experiences across time and space—hence progress and pluralism.⁹²

Idyllic as it might seem, Hegel’s alphabetist vision threatens its own undoing. If alphabetic writing is speech-without-sounds, what about “hieroglyphs” and sign languages, which, in Hegel’s understanding, also function without the mediation of sounds ? After all, Hegel himself argues, in arch-dialectical fashion, that “habit makes [alphabetic writing] a hieroglyphic script for us, so that in using it we need not have the mediation of sounds” (i.e., spell any letter aloud).⁹³ He even describes alphabetic writing in habitual use as “a deaf reading and a dumb writing.”⁹⁴ Thus, Hegel’s phonocentrism stipulates two types of hieroglyphs that are different in origin but indistinguishable in use. The alphabet, Chinese writing, and sign languages all appear to have negated the voice’s sonorous body by being “deaf and dumb”—except that only the alphabet has actually negated that body: the other two, purportedly signifying through silent pictographs and mimes, had never even used sounds to negate their sensuous mimesis and thus had no sonorous body to negate in the first place. The Chinese and Deaf people can feign or “act analogously” (Herder) to the “deafness and dumbness” of alphabetic writing, without ever undergoing the double negation of self-extraction. They must therefore be assimilated to the actual dialectics of self-extraction, to safeguard the phonocentric foundation of pluralism and progress.

Such assimilation was already exemplified by the French Republic after the Thermidorian Reaction (1794), decades before Hegel’s Enzyklopädie. Terrified by the Reign of Terror, post-Thermidorian elites yearned for a society of “unfettered public discussion and debates,” as Sophia Rosenfeld shows.⁹⁵ Abandoning the dream of a sensation-based natural language, they considered arbitrary linguistic conventions not the root of despotism but the cornerstone of a free marketplace of ideas. Like Hegel, they believed that definitions of words must be diverse and dynamic in an enlightened society, not unified or unchanging.

This liberal language politics was codified in the Thermidorian Constitution (1795), which barred all illiterates from voting.⁹⁶ Though it seems fortuitous that this unprecedented literacy test was alphabetic, Antoine Destutt de Tracy, a key member of the new Institut de France, argued that alphabetic writing was foundational for a republic. A “hieroglyphic” script like Chinese “prevents the nation’s entire populace from using the whole written corpus of its ideas,” since learning its numerous characters requires “the study of a life time,” feasible only for the elites.⁹⁷ In contrast, knowing a very small number of alphabetic characters is “so easy that, with good social organization, after very few years, there would be almost nobody in a civilized nation who would lack this advantage.”⁹⁸ Accordingly, the Constitution suspended the literacy requirement for nine years to allow for “a widespread program of public instruction and cultural didacticism,” as Andrew Jainchill shows.⁹⁹ Thus, alphabetic writing literally “gives voice to” the people. It does not so much allow diverse “voix” (voices) to constitute a republic as it enables the republic to create its own constituents of “voix”—“votes.” It does not so much allow people to discourse with the state as it compels them to proffer a voice to the state’s extractive, self-legitimizing audition.

What about those “almost nobodies” who could not extract a voice from themselves? Having once valued “hieroglyphic” Chinese philosophy for its rational stability, French cultural elites after the Thermidor had come to dismiss it as stagnant.¹⁰⁰ While they yielded no sovereignty over China, a parallel shift in attitude toward deafness brought concrete changes at home. By the late 1790s, sign languages were no longer regarded as natural models for ridding French of its arbitrariness, but as evidence that Deaf people were incommensurable with the social contract of which the conventional signs of speech and alphabet were deemed paradigmatic.¹⁰¹ Deaf people must therefore be commensurated,¹⁰² which entailed turning each of them into a lack—a lack of hearing and even a lack of life—to be remedied by the hearing world. Writing in 1799, Roch-Ambroise Cucurron Sicard, principal of the Institution nationale des sourds-muets, described a deaf person as “a statue all of whose senses it is necessary to open up and guide, one after another, to compensate for the one sense [i.e., hearing] that it unfortunately lacks.”¹⁰³ Without such “a beneficent hand,” a deaf person remains “an utter nonentity in society, a living automaton.”¹⁰⁴ Sicard’s portrayal of unassimilated Deaf people as miming machines resembled many French elite complaints throughout the Revolution against the mindless parroting of the words “liberté” and “égalité,” whether by the sansculottes or by the slaves of Haiti.¹⁰⁵ As voices using intrinsically senseless sounds to negate sensuous miming and using silent letters to negate such sounds became the foundations of a liberal society, those unable to thus extract a voice from themselves must be pried open and animated, like statues, by society’s benevolent ear.

Through Franklin’s model of the inner ear as explaining the alleged universal appeal of Scotch tunes, I have connected the hardening of Amiot’s theory of Sino-European auricular difference—from climate to physiology—to his growing appreciation of “plain old” songs as the essence of Chinese music. Bringing this ear-song entanglement to bear on Herder teases out the dialectics of a self-extracting voice, one further manifest as a foundation for pluralism and progress in Hegel’s alphabetism and post-Thermidorian language politics. From these reverberations, Amiot’s approach to airs chinois may not seem very different from contemporary European narratives about non-European songs. Both mid-century celebrations of their ancient naturalness and late-century dismissals of their lack of autonomy from the sensations could easily digest Amiot’s understanding of Chinese airs (and ears) in contradistinction to European art music. When commenting on the songs in his “Divertissements,” Amiot even remarked that “if [France] were inhabited by the Chinese, one would still be enthusiastically singing ‘Charmante Gabrielle’ and ‘Dupont mon ami.’”¹⁰⁶ Both Rousseau and Forkel, it might well seem, could have written this sentence.

Or could they? Unbeknownst to Amiot, the two songs he named were in fact being “enthusiastically sung” in Paris at the time. Sources from the 1770s show that “Charmante Gabrielle” and “Dupont mon ami” featured as “airs connus”—stock melodies to which vaudeville numbers were sung—in opéras-comiques.¹⁰⁷ While analogizing eighteenth-century Chinese songs to French songs that have been dated to the reign of Henri IV (1589–1610)¹⁰⁸ reeks of an Orientalist “the East never changes” narrative, Amiot’s comparison was exceptionally apt. As François Picard shows, all forty-one songs of his “Divertissements” were drawn from an enormous body of stock melodies known as qupai, on which songs were based in many Chinese theatrical traditions.¹⁰⁹ Most qupai are securely traceable to the fourteenth century, though Chinese scholars of Amiot’s time dated many of them to the eighth century on the basis of sources such as Cui Lingqin’s Jiaofang ji (Records of the music academy, before 741).

There was yet another irony. While Amiot was unwittingly comparing the stock melodies of Chinese theater to those of opéra-comique, the use of the former was changing inside the Forbidden City. Never before noted by scholars of Chinese theater, this change reflected a fundamental shift in the phenomenology of listening in Qing court theater and the Qing imperium at large, one that also moved along the dialectics of the self-extracting voice. To trace these, I follow Amiot’s inadvertent lead on qupai, beginning with their typical use in this section before turning to their reform at the Qing court in the next.

Of all Chinese theatrical traditions that derived their melodies from qupai, Kunqu, named after its birthplace in Kunshan, was paradigmatic.¹¹⁰ Classicized by Chinese literati in the Yangtze River Delta during the sixteenth century, Kunqu was co-opted by the Qing’s Manchu ruling class, who relied on its audience base—the Chinese scholar-officials and gentry—for revenue and personnel.¹¹¹ Rather than “opera,” as it is often called, Kunqu is best described as dramatized song suites. Each scene, be it a stand-alone excerpt (zhezi xi) or part of a full-length drama, comprises a suite of songs based on different qupai or on a formulaic suite of qupai (taoqu), interlaced with spoken parts. Composing a drama, or “measuring the tune” (duqu), entailed choosing qupai for the desired musical-dramatic effects and “fitting lyrics” (tianci) into these qupai according to their established rhyming, metrical, and tonal (pronunciation) forms.¹¹² The most “composerly” a dramatist could be was to take a few lines from several different qupai and combine them into a “composite tune” (jiqu).¹¹³ A typical play text—often read as a work of literature—would not notate the melodies but only name the qupai for each song.

As dramatized qupai suites, Kunqu entails a particular phenomenological effect whenever the plot of a play stipulates a scene of singing that is also heard as singing by the characters themselves. In her analysis of Kunqu, Judith Zeitlin describes such scenes as “phenomenal” performances, a term coined by Carolyn Abbate to mean “a musical or vocal performance [in theater] that declares itself openly, singing that is heard by its singer, the auditors on stage, and understood as ‘music that they (too) hear’ by us, the theater audience.”¹¹⁴ Yet the pervasive use of qupai meant that “phenomenal” singing in Kunqu could never be “diegetic,” to borrow the conventional term used in film studies. Sound films create what Michel Chion calls a “false depth”: the apparent synchronies between certain sounds and their onscreen sources “swallow [those sounds] up” into the film’s diegesis.¹¹⁵ Just as “silent” films do not impart such a false depth in their soundtracks,¹¹⁶ so Kunqu does not afford phenomenological identification—what I describe as “diegetic”—between what the spectator hears and what the characters presumably hear in the world of the play. When a scene of phenomenal music making occurs on stage and the characters begin to sing, a spectator of Kunqu may “understand” what she hears to be “music that they (too) hear” in the play, but she cannot hear or imagine herself to be hearing exactly what the characters are hearing at that moment within the diegesis.

This characteristic of Kunqu is illustrated in Hong Sheng’s play The Palace of Lasting Life (Changsheng dian, first performed 1688), which Zeitlin dubs “an extreme case” of phenomenal singing.¹¹⁷ The play depicts the celebrated romance between Emperor Xuanzong and Noble Consort Yang of the Tang dynasty (618–907) at the time of the An–Shi Rebellion (755–63): as the two lovers flee the capital, Consort Yang is forced by mutinous guards to commit suicide; she travels through the underworld and ascends to the Moon Palace, where she is eventually reunited with the emperor. But the plot also dramatizes the history of a famous piece of music, “The Dance of Rainbow Skirt and Feather Coat” (“Nishang yuyi wu”), a faqu, or song-and-dance suite, recorded as having been performed at Emperor Xuanzong’s court.¹¹⁸ Consort Yang first hears and watches “The Dance of Rainbow Skirt” in her dream at the Moon Palace in scene 11, “Hearing the Music.”¹¹⁹ She notates it from memory in scene 12, “Fashioning the Score,” and performs it in scene 16, “Dancing on a Disc,” accompanied by the emperor’s Pear Garden Ensemble (Liyuan). She later revises her notation in scene 40, “Immortal Remembrance,” for the Moon Goddess, who has it performed in the final scene, scene 50, “Reunion in the Moon,” at the lovers’ reunion.

Unsurprisingly, the play features several phenomenal performances of “The Dance of Rainbow Skirt.” Yet its spectators never get to hear the piece itself, because all its performances involve generic qupai—and completely different ones at that, so that there is no musical overlap between phenomenal performances of the same piece.¹²⁰ Hong could easily have created melodic unity had he wished. The final performance of the “Dance” in scene 50, for example, uses a composite tune combining lines from twelve different qupai.¹²¹ The eighth is entitled “Dancing ‘The Dance of Rainbow Skirt’” (“Wu Nishang”), directly evoking the “Dance.” Both this qupai and the ninth one, “Imperial Birthday” (“Qianqiu sui”), have also been used in scene 16—except not for the phenomenal performance of the “Dance” in that act but for a non-phenomenal duet between the lovers.¹²² Even though there exists a qupai that, given its title, could easily masquerade as the “real” “Dance of Rainbow Skirt,” it was apparently neither artistically appealing nor conceptually viable for Hong to have set the “Dance” to any recurring melody. No matter how much the lyrics tease the spectator’s imagination by describing what the “Dance” sounds like—sometimes employing music-theoretical terms—the spectator cannot hear its various iterations in the same way as the characters in the play. All she can hear is nondescript qupai that could just as easily stand in for the “Dance” as for any other song, phenomenal or otherwise.

The inability of the spectator to hear what the characters hear may seem to be a defect: the spectator desires to hear “The Dance of Rainbow Skirt” as if she too were in the play’s world, only to be prevented from doing so by Hong’s use of generic qupai. Such a reading presumes that singing in Kunqu was predicated on a divide between a diegetic inside and a nondiegetic outside, and that phenomenal singing such as Hong’s setting of the “Dance” was predicated on a desire to overcome or negate this divide. In truth, neither such a divide nor its overcoming ever existed in Kunqu singing. Every song, regardless of its connection with scenes of singing within the plot, must be set to a generic qupai. The experience of hearing a phenomenal song diegetically, as if one were inside the play, was radically impossible—there was neither inside nor outside when it came to qupai.

The irrelevance of the diegetic/nondiegetic divide made singing a unique dimension of Kunqu. According to Ling Hon Lam, the sixteenth century marked a changing “spatiality of emotion” in Chinese theater. Before this time, theater was “akin to the Buddhist-Daoist ‘dreamscape’” in which the audience qua dreamer passes from one illusion to another.¹²³ After the sixteenth century, “an indelible distance” pushed the dreamer out, so that one could only “paus[e] in front of the dream,” “trying to sympathetically identify with it” yet unable to be immersed in it.¹²⁴ This desire to “[put] one’s body in the other’s situation” (sheshen chudi) presupposed a divide between the theater’s and the spectator’s worlds only for the desire itself to negate and bridge across it.¹²⁵

The Palace of Lasting Life, despite being an “extreme case” of phenomenal singing, shows this “indelible distance” to be radically porous to singing. Hong’s qupai setting of “The Dance of Rainbow Skirt” exemplified this porousness not only on a technical level but also on a social and historiographic level. It idealized the ability of theater to nullify the boundaries between ancient and modern, court and commoners, Chinese and foreign—thus laying bare Kunqu’s political affordances to the early Qing.¹²⁶ Insofar as it concerns the Tang-era “Dance of Rainbow Skirt,” The Palace of Lasting Life is a piece of theater about theater, because Hong’s play, even theater at large, is the “Dance.” To begin with, the Tang court was often posited as the origin of modern theater. By the eighteenth century, professional actors had been revering Emperor Xuanzong as a patron deity, and his Pear Garden Ensemble had become a metonym for theater at large.¹²⁷ Music theorists also attributed the musical modes used in Kunqu to Tang court banquet music (yanyue).¹²⁸ In glossing the many foreign-sounding names of these modes, they pointed to Central Asian influences epitomized by “The Dance of Rainbow Skirt,”¹²⁹ which was adapted from a certain “Music of the Brahmins” (“Poluomen yue”) from the Tarim Basin, whose trade routes extended to northern India.¹³⁰ As implied by Amiot’s “Divertissements,” the names of many qupai are also found in Tang-era sources.

The received historiography of theater thus stipulated a longue durée dissemination of songs from the cosmopolitan Tang court “among the commoners” (minjian). The Palace of Lasting Life dramatized this historiography. While the main plot follows the circular course of the “Dance” away from and back to the Moon Palace through the two lovers, a subplot maps the court-to-commoner trajectory through two minor characters, Li Guinian and Li Mo, both based on real musicians who served at Emperor Xuanzong’s court.¹³¹ In scene 14, “Stealing the Melody,” Li Guinian, head of the Pear Garden Ensemble, rehearses the “Dance” from Consort Yang’s notation, while the flutist Li Mo eavesdrops.¹³² Later, in scene 38, “Popular Ballad,” both musicians flee to Nanjing amid the rebellion. Li Mo recognizes Li Guinian performing a tanci, or “storytelling ballad,” at a temple fair before a commoner audience. Besides satiating their appetite for court gossip, Li’s ballad features extensive descriptions of the “Dance.” When the two musicians recognize each other, they reminisce about the “Dance” and vow to have it “propagated for thousands of years.”¹³³

Was “The Dance of Rainbow Skirt” “propagated for thousands of years,” even though it could never be heard diegetically in The Palace of Lasting Life ? The answer is revealed in the title of scene 38, “Popular Ballad.” As Zeitlin shows, tanci as a genre is not only more “popular” or lowbrow than court music but also anachronistic in the context of The Palace of Lasting Life, since the word “tanci” is not attested until the sixteenth century.¹³⁴ Furthermore, a merchant from the Shanxi-Shaanxi region in Li Guinian’s audience even compares his ballad to xidiao (Shanxi-Shaanxi tunes),¹³⁵ which also did not exist in the eighth century but proliferated during Hong’s lifetime.¹³⁶ Yet the merchant’s comparison makes sense in nonlinear time. The eighth-century Tang court was located in Shaanxi where the seventeenth-century merchant and xidiao originated. Nanjing, where the two Tang court musicians sought refuge, was located in the Yangtze River Delta, where Kunqu flourished, temple fairs regularly featured theaters, and many Shanxi-Shaanxi merchants arrived amid the seventeenth-century commercial boom. Through Li’s ballad, the eighth-century Tang court and the seventeenth-century Yangtze River Delta were folded into one another. The resonance the merchant hears between the modern songs of his hometown and the Tang musician’s ballad embodies not only the transmission of Tang court songs from past to present or from court to commoners, but also the irrelevance of these dichotomies to songs—the same irrelevance as that of the diegetic/nondiegetic boundary to qupai. Li Guinian’s time-warping ballad and his ensuing conversation with Li Mo are all set to a generic taoqu, or suite of qupai.

This boundless mobility of songs embodied by qupai was an ideological focus for the early Qing. Though the imperially commissioned Lülü zhengyi focuses almost exclusively on reforming tuning, its draft materials in the manuscript Putong Guji 15251 (ca. 1707) in the National Library of China indicate that it grew out of a larger project spearheaded by Yūn-c’y,¹³⁷ the third son of the Kangxi Emperor (r. 1661–1722). The manuscript’s first third on Western music theory (fols. 1r–51v) and its last third on tuning and organology (fols. 84r–127v) fed respectively into juan 5 and juan 1–4 of Lülü zhengyi.¹³⁸ And its middle third (fols. 52v–83v), a six-chapter commonplace book of quotations, commentaries, and diagrams, reveals that the tuning reform in Lülü zhengyi—which divided the octave fourteen-fold—ultimately served to rectify the use of modes in Kunqu.¹³⁹ The first three chapters (fols. 53r–60v) prepare for Yūn-c’y’s tuning research, while chapters 4 and 5 (fols. 61r–70v) discuss musical modes and their history since the Tang era, and chapter 6 (fols. 71r–80v) studies the musical notation used by professional practitioners and concludes with a modally ordered catalog of qupai.¹⁴⁰ These three latter chapters laid the foundation for two modally ordered qupai compendia edited by Yūn-c’y.¹⁴¹

Even before these compendia were published in 1715, juan 5 of Lülü zhengyi already revealed theater to be the ideological backdrop to Qing-imperial music theory. This juan discusses Western music, which the Kangxi Emperor and his eunuchs had been learning from Catholic missionaries. Nonetheless, its title, “Xieyun duqu,” “coordinating all the keys to compose drama,” refers to duqu: using qupai to compose drama. Accordingly, the juan’s preface incorporates its European content—Guidonian hexachords and staff notation—into the historiography of Chinese theater. It draws transhistorical parallels between Suzup (fl. 568), a pipa player from “Western Region” (xiyu), or Central Asia, and Tomás Pereira (1645–1708) and Teodorico Pedrini (1671–1746), two missionaries from “Western Ocean” (xiyang), or Europe.¹⁴² Suzup, who came from the Tarim Basin, helped sixth-century Chinese musicians to repair their ancient instruments and develop a system of modes.¹⁴³ This system, according to the preface, became “the general framework for modal transpositions and mutations in both elegant [ya] and popular [su] musics from the Tang and Song dynasties [960–1279] onward.”¹⁴⁴ Qing-era Chinese musicians no longer understood Suzup’s teachings, the preface observes, but the Western theory taught by the missionaries turned out to be identical to Suzup’s system as recorded in historical documents.¹⁴⁵ Just as Central Asian music once rescued Chinese music from centuries of deterioration, so Western music can in turn rescue Suzup’s system from millennia of subsequent misuse. Instead of situating them in a linear genealogy, the preface describes Chinese music, sixth-century Central Asian music, and eighteenth-century European music as mutually constitutive, effacing distinctions between ancients and moderns, inside and outside, centers and peripheries.

Such effacements were critical for the Qing. As a conquest regime that hailed from Manchuria in the 1640s, the empire’s rule of China ran against the Confucian “distinction of the Chinese from the barbarians” (huayi zhi bian), whereby China, the “middle kingdom,” should rule over the surrounding barbarians, including the Manchus, not the other way round.¹⁴⁶ By mapping the free circulation of music without distinguishing between middle/China (zhong) and outside/foreign (wai), or even between foreigners from different “Wests,” the preface to juan 5 of Lülü zhengyi turns China into a terra nullius where all musics can proliferate: “ever since Our Dynasty was established, all the Four Seas have been pacified, and more people are coming from far away yearning for edification.”¹⁴⁷ It was also because of this universally transformative “edification” that seventeenth-century Chinese literati valued theater. Writing to the emperor circa 1714 in response to the ongoing composition of Lülü zhengyi, Li Guangdi, the most influential Chinese scholar-official during the Kangxi reign and a key source for Amiot’s early writings,¹⁴⁸ encourages the emperor to embrace theater’s potential in edifying and transforming (jiaohua) the masses.¹⁴⁹ In contrast to the moral censure of the vulgarity of theater by the state-sanctioned school of Confucianism known as the “Learning of the Principle” (lixue), Li quotes Mencius (ca. 371–289 BCE) as saying, “modern music is just like ancient music.”¹⁵⁰ Just as the ancient sage kings used “elegant” music not to embellish their own courts but to move and edify the people, so modern theater has the same transformative power precisely through its boundary-precluding “popular” or lowbrow nature. Instead of reconstructing the ancients’ tuning and instruments, Li pushes the emperor to emulate their intent by adopting modern theater to “gently sway the common folks” and to “facilitate the return of the simple and the pure.”¹⁵¹ As Shi Fang shows, Li’s idealism was representative of writings on the theater of the time.¹⁵² Heedless of all ethnic, social, aesthetic, and historical boundaries, theater became a vision of universal transformation, where all thrive in song without distinction.

Until such universally transformative singing outlived its imperial utility.

In the same way that Herder’s “middle sense” extracts itself from the bodily senses and reflects on them from a higher plane, a transcendent ear at the late eighteenth-century Qing court fractured the undifferentiated space of qupai only to tie all its pieces back together again. The earliest indication of this development occurred a few months after Amiot wrote his letter to Bertin in July 1780, on September 13, the Qianlong Emperor’s seventieth birthday. According to Bak Jiwon, a member of the Joseon dynasty delegation, the court celebrated the occasion at the Mountain Resort in Jehol, where theatrical performances lasted seven full days.¹⁵³ Among the eighty plays whose titles Bak recorded in his diary was one named Sixty-Four Dancers at Shun’s Courtyard (Bayi wu Yuting, 1780 or before). As Ye Xiaoqing shows, this play belonged to a repertoire that emerged at the Qing court known as chengying xi (ritual plays).¹⁵⁴ While theater had been part of the Forbidden City soundscape since the Ming, its ritual use was first codified by the Qianlong Emperor in 1742.¹⁵⁵ In charge of the project were the emperor’s uncle Yūn-lu, who had been apprenticed to Yūn-c’y before the latter’s fall from grace, and the Chinese official and theater aficionado Zhang Zhao.¹⁵⁶ Together, they established the ritual play genre by revising and/or composing plays to be performed at every court ritual. Each play was tailored to a type of recurring occasion, such as the celebration of the emperor’s birthday, though its lyrics were adjusted in each performance to reflect the actual circumstances, such as the age of the emperor or recent events. Notably, the plots of many ritual plays feature phenomenal performances of music that would have been heard in real life during the very rituals they accompanied. Following standard Kunqu practice, such phenomenal songs were still based on nondescript qupai in almost all ritual plays whose original composition can be dated to the Qianlong and Jiaqing reigns (1735–96 and 1796–1820).¹⁵⁷

But then there was Sixty-Four Dancers. This play was tailored for imperial birthdays and military triumphs, particularly conquests in Xinjiang, a region incorporated into the Qing during the Qianlong reign.¹⁵⁸ Its plot revolves around the ancient emperor Shun (? ca. 2294–2184 BCE), who instituted the oldest court music, known as Dashao, according to Confucian legends. In the play, the immortalized Shun orders a performance of Dashao in his courtyard to celebrate the Qing. This phenomenal performance is not based on any qupai, unlike all other songs in the play, as is evident from the lack of qupai indications in the playscripts and from the song’s distinctive form (see figure 4).¹⁵⁹ Whereas all qupai consist of lines of varying metrical lengths—hence the nickname “long and short lines” (changduan ju)—and typically feature extrametrical words (chenzi), this song consists of only four-syllable lines, rarely used after the fifth century except in archaist contexts.¹⁶⁰ And whereas all qupai in Kunqu are melismatically embellished to mimic the pronunciation tones of words,¹⁶¹ this song is strictly syllabic, to judge by the surviving notation.

The same metrical and melodic characteristics are found in at least two other ritual plays from the Qianlong-Jiaqing era, where, like the Dashao in Sixty-Four Dancers, a non-qupai phenomenal song evokes a supposedly ancient piece of music. In Lush and Peaceful Harvest (Fengsui gubao, 1796), peasants hunting game for the emperor’s birthday sing a song, “Illustrating the ‘Local Songs of Bin’” (“Binfeng huitu ge”).¹⁶² It evokes the “local songs” (feng) in Shijing (Canon of songs), the Confucian book of lyrics that had supposedly been collected from peasants living in the vassal states of the Western Zhou (ca. eleventh century BCE–771 BCE), including Bin.¹⁶³Fragrance of the Fairyland Spreading throughout the World (Yaolin xiang shijie, 1780 or before), also recorded in Bak’s diary, opens with immortal fledglings roaming a fairyland.¹⁶⁴ Their song, “Drink every ten steps, peck every five steps [十步一飲/五部一啄] … retreat with the sun, rise with the sun [日落而息/日出而作],”¹⁶⁵ reads almost identically to the ancient peasant song reputedly heard by Emperor Yao (r. 2333–2234 BCE) on an inspection tour: “Rise with the sun, retreat with the sun [日出而作/日落而息]; drink by digging wells, eat by farming the land [鑿井而飲/耕田而食].”¹⁶⁶

That these three non-qupai songs were possible reflected a radical transformation of Kunqu singing at the court. To wit, their form resembles that of the deliberately archaic Confucian ritual music used at court, which the Qianlong Emperor codified as strict one-note-per-syllable melodies and quadrisyllabic meter in 1746 (see figure 5).¹⁶⁷ Beginning with this ritual music codification, the emperor pursued a lifelong campaign against basing ancient songs on qupai, culminating in the treatise Shijing yuepu quanshu (Complete music for the canon of songs, 1788), in which each ancient lyric from the Confucian Shijing is provided with a newly composed melody (see figure 6).¹⁶⁸ In his preface, the emperor reveals that he has commissioned the project in order to rebuke Zhu Zaiyu’s melismatic qupai settings of these lyrics¹⁶⁹—the very settings that Amiot copied into his 1780s treatises on Chinese dance.¹⁷⁰

The emperor’s insistence that ancient and ancient-style songs should sound different from modern theatrical qupai—which likely trickled down into ritual plays—might suggest a sonic historicism. Certainly, the eighteenth-century “evidential learning” (kaozheng) movement, which used philology to critique received interpretations of Confucian classics, emphasized that the pronunciation of Chinese characters had been changing, and hence that reconstructing the meaning of ancient texts demanded first reconstructing their ancient sounds.¹⁷¹ As I show elsewhere, Shijing lyrics provided this linguistic restorationism with phonemic data in the form of rhymes.¹⁷² Yet Shijing yuepu quanshu bucked this historicist trend. In choosing the appropriate musical modes in which to compose melodies, it followed the interpretations of the lyrics by Zhu Xi, a principal progenitor of “Learning of the Principle.” Famously, Zhu Xi’s interpretations rejected the xiaoxu (short prefaces) that accompanied each lyric, brief exegeses first written down during the third century BCE together with what would become the orthodox version of Shijing.¹⁷³ Whereas Zhu Xi read the lyrics on their own as timeless moral lessons, Qing evidential scholars upheld the xiaoxu as proof of the lyrics’ original meaning and criticized Zhu Xi for ignoring them.¹⁷⁴

Though the Qianlong Emperor employed many evidential scholars, the compilers and composers of Shijing yuepu quanshu resembled Zhu Xi in showing little interest in restoring these lyrics to historical sound or meaning. In dictating they be sung to syllabic melodies, the preface and fanli (editorial principles) of Shijing yuepu quanshu marshal no philological glosses. Instead, they repeatedly complain that setting these lyrics to melismatic qupai would “pull the ancient into the popular” and “pull the elegant into the popular.”¹⁷⁵ By abstracting the temporal “ancient” into the aesthetic, class-conscious “elegant,” Shijing yuepu quanshu strove not for a historical, sensuously experienceable antiquity, but for one decontextualized from time and place. Whereas evidential language reconstructions posited empirical characteristics of the singing-speaking voice of the ancients, the emperor’s Shijing melodies simply negated the commensurability of this voice with modern theater.

This negation fractured the space of qupai. Because of their form, the three songs in ritual plays that I have examined would have been heard by their audience as representing exactly how “ancient” songs sounded at the real-life court. Like the apparent audiovisual synchronies in sound films, the coincidence between a song’s dramatization as ancient within the play’s plot and its distinctly archaic non-qupai setting “swallows [the song] up,” to borrow Chion’s description of sound films, into the play’s diegesis.¹⁷⁶ Such a phenomenal song is now also diegetic, because it lets the spectator hear it as if her ears were inside the play’s world.

But diegetic singing both negated the pervasive use of qupai and negated this negation. The space of qupai split into a diegetic “source music” and a qupai “underscore,” only so that the auricular identification between spectator and characters at a moment of diegetic singing could bridge across it. Previously, singing in Kunqu had not been predicated on any diegetic/nondiegetic divide, thereby precluding any bridging over. Now, the existence of the bridge both presupposed and negated the divide: hearing a phenomenal song diegetically rather than as a qupai enacted this divide by allowing spectators to cross it. This dialectics of division and connection meant that the bridge functioned not only by standing on both sides of the proscenium but by rising out of each side. Diegetic presentations of ancient songs negated Kunqu’s normative economy of interchangeable qupai, but by giving them the same form as the nonhistoricist Qing court ritual songs, they at the same time negated their being reabsorbed into any specific time and place. Hearing diegetic songs, the imperial ear could varyingly immerse itself in the worlds of ancient kings, peasant hunters, and fantastic fledglings only because, unlike evidential language reconstructions, it was not tethered to any empirical, sensuous commitment. The imperial ear became a gateway that connected the two sides of the proscenium only by transcending them altogether.

This transcendent ear underpinned a voice-extracting regime even more blatant than that outlined by Hegel’s alphabetism and post-Thermidorian language politics. To wit, Amiot was hardly the only Qing resident whose approach to “la musique des Chinois” shifted along Tomlinson’s “music” vs. “songs” binary in the eighteenth century. As mentioned above, the Qing’s official music treatise Lülü zhengyi (1714) arose from Yūn-c’y’s vision of using theater to nullify social, historical, and ethnic distinctions and thus legitimize the Qing rule of China. Upon completion, Lülü zhengyi was also incorporated into the Lüli yuanyuan (Origins of pitch pipes and the calendar, 1723) project, in which music theory joins mathematics and astronomy in probing the laws of the cosmos.¹⁷⁷ Such universalist aspirations vanished two decades later, when the Qianlong Emperor commissioned Yūn-lu for a 120-juan expansion of Lülü zhengyi (1746), which focused instead on taxonomizing different musics at court.¹⁷⁸ Besides Confucian ritual songs and Manchu maksi dances,¹⁷⁹ the expansion cataloged the songs and instruments of the empire’s conquered peoples in the exact order in which they met the emperor’s ear in both historical and ritual time.¹⁸⁰ This ethnography expanded as the Qing expanded.¹⁸¹ It culminated in a set of seventeen song manuscripts presented by the Imperial Household Department in 1796 to commemorate the Qianlong Emperor’s promised abdication in favor of his son after sixty years on the throne (see table 1).¹⁸² Drawn from the Manchus, Mongols, Tibetans, Burmese, Vietnamese, and Nepalese, the songs were written in their original languages with translations into Manchu and Chinese, and sometimes into Mongolian, a working language of the Board of Colonial Affairs.¹⁸³

Representing many different voices, these songs were unified through a remarkable feat of materiality: all were written in powdered gold on an expensive type of dark paper (ciqing zhi; see figures 7–9). This is exceptional, because virtually every other manuscript produced in this configuration at the court was a hand-copied Buddhist scripture, or sutra.¹⁸⁴

Why were these songs written as if they were sacred texts? The answer is phonocentrism, specifically the extraction of voices through alphabetic writing. The sutras include numerous Sanskrit or Pali incantations known as dhāraṇı̄ (tuoluoni in Chinese, tarni in Manchu) or mantras. Chanted and written for their cosmic powers, dhāraṇı̄ were typically left untranslated in sutra translations, only transliterated: they crossed linguistic bounds not as banally meaningful words, but as profoundly meaningless sounds.¹⁸⁵ Translating sutras was political under the Qing, as Tibetan Buddhism was instrumental in controlling the Mongol confederations in the Eurasian Steppe, the empire’s gravest threat since its founding.¹⁸⁶ Thus, at a time when evidential learning was making philology the dominant paradigm of eighteenth-century Chinese scholarship, the Qianlong reign saw the production of many imperially commissioned philological studies—almost all related to dhāraṇı̄ and transliterations.

This mantric paradigm was established in 1751, when, on the verge of exterminating its final Mongol adversary and partitioning the Steppe with the Romanovs, the Qing issued Tongwen yuntong (Unifying writing and uniting rhymes), a treatise compiled by Yūn-lu on how to “unify” (tong) Manchu, Chinese, Mongol, and Tibetan transliterations of dhāraṇı̄. In his preface, the Qianlong Emperor complains that, when contemporary Chinese monks chant dhāraṇı̄, they sound drastically different from their Tibetan counterparts;¹⁸⁷ this was because many dhāraṇı̄ had been transliterated into Chinese characters during the Tang era (618–907), and the pronunciation of the same Chinese characters had changed since then—what modern linguists would describe as sound change. Whereas Chinese characters are nonalphabetic and hence phonetically imprecise and unstable, the emperor argues, the Manchu alphabet invented by the Qing’s founder Nurgaci in 1599 transliterates dhāraṇı̄ much more accurately and enduringly, because it “spells words by matching the voice with letters.”¹⁸⁸

Tongwen yuntong was not the first Qing-imperial publication to celebrate Manchu alphabetism. In Han i araha manju gisun i buleku bithe (Imperial Manchu dictionary, 1708), published by the Kangxi Emperor, for example, a group of officials argue in a postface that “the principle of writing pertains to sounds of the voice.”¹⁸⁹ Whereas Chinese characters, being nonphonetic, are difficult to learn and prone to mistakes and obsolescence, Manchu writing “matches sounds very well” and makes the empire’s laws forever comprehensible to all.¹⁹⁰ Yet because this postface was written in Manchu, it never preached these advantages of the Qing’s native writing to its conquered Chinese literati. Although the Kangxi Emperor’s preface to the dictionary was translated into Chinese, it completely defaces Manchu alphabetism by assimilating Manchu into the purported pictographic origin of Chinese characters.¹⁹¹

In contrast, Tongwen yuntong, written in Chinese, openly flaunts alphabetism—arguably even more aggressively than would Forkel or Hegel decades later. To resolve the mantric cacophony, the treatise ordains the Manchu alphabet as the “standard weight and balance” (quanheng) through which to commensurate all languages and scripts.¹⁹² As visualized in transliteration tables that place Manchu in the middle, the treatise uses Manchu letters to transcribe every possible sound and transliterate every possible written syllable in those other languages, so that dhāraṇı̄ can be transliterated between them through the Manchu bridge (see figure 10).¹⁹³ This Manchu alphabetism weighed heavily on Chinese, the only nonalphabetic and nonsyllabic script in Tongwen yuntong. To commensurate Chinese with the empire’s now alphabetized mantric polyphony, the treatise invented a “Chinese alphabet” by using Manchu letters to transliterate a select group of Chinese characters and convert them into “letters” (zimu).¹⁹⁴ Like literal alphabetic letters, these characters are stripped of any pictographic origin or logographic function and serve merely to capture speech sounds as intrinsically meaningless signs.

Tongwen yuntong heralded a series of imperial publications in which Manchu was used to rectify Chinese transliterations.¹⁹⁵ Each extolled Manchu alphabetism while singling out the Chinese lack of the same. From one preface to another, the Qianlong Emperor complains that Chinese characters not only constitute a phonetically inaccurate and unstable form of writing, but also perpetuate a defective mode of thinking. Because they use a nonalphabetic script, the emperor laments, the Chinese always fail to understand the raison d’être of writing. This line of attack was most clearly stated in his Chinese-Manchu bilingual preface to the 1772 expansion of his grandfather’s dictionary:

Thus, transliterating dhāraṇı̄ as well as Khitan, Jurchen, Mongol, Oirat, Tibetan, and Chagatai words into Chinese demanded not only that Chinese characters be better matched with non-Chinese sounds. To be part of the multilingual empire, the Chinese must learn that the true essence of writing is neither pictography nor logography, where each glyph denotes some “intrinsic meaning” (daci jurgan), but rather phonography (“voice-writing,” mudan acaburengge), where “there is no intrinsic meaning.” And heeding this lesson demanded that they extract intrinsically meaningless phonetic signs from their pictographs and logographs. All Qianlong-imperial philological publications posited Manchu as the shuniu (gateway) for enforcing this demand.¹⁹⁷ As the shu (door hinge) and the niu (handle), Manchu letters pried open the nonphonetic and therefore seemingly mute Chinese characters and extracted from them a sonorous voice. Only by proffering such a voice and thereby negating their silent pictography and logography could the Chinese join Manchu writing in negating the sonorous body of this voice, writing its incantations in powdered gold.

The seventeen song manuscripts were albums of similarly extracted voices. As shown in table 1, most of these songs were composed and/or collected from the empire’s various constituencies at different points in the emperor’s sixty-year reign, and they were never performed together in real life. Instead, it was the Qing’s ruling alliance between Manchu conquest elites and Chinese literati that conjured this polyphony of voices by gilding them onto sheets of dark paper—not unlike a stylus inscribing tracks on a vinyl disc. As such, this chorus existed only through a phonographic or voice-writing imperial ear that transcended “the celebration, the social context, the form, and the movement” of individual performances, as Herder would describe:¹⁹⁸ the empire created its constituent voices by extracting them toward its own transcendent ear. To be a constituent of the empire was to be heard by it, which was contingent upon providing raw materials for the imperial phonograph—a voice that negates its own sonorous body through writing while presupposing such a body to negate. The album featured no Chinese songs, but Chinese scholar-officials could join in producing its sutra-esque song manuscripts because a voice could be extracted from their apparently voiceless writing after all; they even proved perfectly self-extracting in this regard.

Just when the emperor was accusing them of not understanding the essence of writing, Chinese scholars outside Qing-imperial patronage were already undergoing what I have dubbed elsewhere a “phonographic revolution,” recognizing Chinese characters to be, in fact, phonographic.¹⁹⁹ Criticizing previous Chinese scholars for heeding only written characters while ignoring spoken languages, seventeenth- and eighteenth-century philologists approached ancient Chinese texts not as pictographs or logographs conveying meanings directly, but as intrinsically meaningless signs recording ancient speech sounds. Even the millennia-old narrative of the origin of writing was rewritten so as to become more phonocentric. The foundational text in Chinese grammatology, Xu Shen’s Shuowen jiezi (Explicating glyphs and analyzing characters, ca. second century), declared that the ancient sages had invented Chinese characters to mimic the shapes of natural phenomena, such as the footprints of birds and beasts.²⁰⁰ But Duan Yucai, who annotated Xu’s dictionary in 1815, flipped the script: “When the sages created written characters … first there was sound and then there was [written] shape.”²⁰¹ It was through a comparable phonographic epiphany that, nine years later, Jean-François Champollion would decipher the hieroglyphs on the Rosetta Stone, likely even inspired by Chinese philology.²⁰²

As it turned out, sanctifying the alphabetist principle of phonography did not even require an alphabet—let alone one that descended from the Greek alphabet (which Manchu did not), which Friedrich Kittler has deemed the earliest precursor to the gramophone on account of its supposedly groundbreaking notation of vowels in addition to consonants.²⁰³ With only a “largely nonphonetic script,”²⁰⁴ eighteenth-century Chinese philology arrived at the same phonocentric conclusion as contemporary European discourse, namely that writing serves to record sounds. Thus, pace Jacques Derrida and the Qianlong Emperor, phonocentrism was neither uniquely Western nor uniquely Manchu.²⁰⁵ Just as critiques of phonocentrism have inspired many to decenter Europe in the history and theory of writing, so it is time to decenter Europe in discourse about phonocentrism, not only regarding the philosophy of voice or history of linguistics, but also with respect to phonocentrism’s implications in imperialism and colonialism.²⁰⁶

In this respect, Amiot has brought us a long way. Heeding his auricular comparisons has allowed me to sketch the shared phonocentrism of Europe and the Qing. This phonocentrism privileged neither “music” nor “songs”—neither mind nor body in the Cartesian sense, and neither literacy nor orality in the Ongian sense, dichotomies that Jonathan Sterne has critiqued as the “audiovisual litany.”²⁰⁷ Underpinning it was rather the dialectics of extraction, manifest as phonography. As in Hegel’s double negation, the voice negates the “deaf- and dumbness” of mimetic gestures and pictographs with its ephemeral, intrinsically senseless sounds, only to further negate its own sounds with the “deaf- and dumbness” of alphabetic writing. Thus, neither embodiment nor abstraction—neither sound nor silence—sufficed the voice of phonocentrism. Instead, the voice must continuously extract itself from its body while continuously presupposing such a sonorous body from which to self-extract, in order to be heard.

And heard not just by anyone. Like Herder’s absolute music, rooted in yet growing out of Volkslieder, the voice’s self-extraction stipulated a teleological growth toward a transcendent audition. Herder’s shockingly similar denigrations of the sharp-eared Chinese and the Deaf elucidated his dialectical theory of hearing as the “middle sense,” one always detaching itself from the bodily senses so as to reflect on them from a higher plane. This transcendent hearing defined the Hegelian-Thermidorian fantasy of a liberal society: through the convention of senseless alphabetic signs, all voices must negate their experiential or epistemic idiosyncrasies—only so that all experiential and epistemic differences can be equally tolerated, in contradistinction to the “hieroglyphic” conformity and stagnation of “Chinese philosophy.” In China, meanwhile, the rise of diegetic songs in ritual plays reflected the way the Qing-imperial ear touted itself as a bridge connecting the empire’s different sonic and scriptorial worlds by rising above them. Inclusion in the multiethnic imperial polyphony depended on proffering a voice commensurable to this transcendent ear, which required the phonographic extraction of voices through Manchu as the commensurating “standard weight and balance.”

What this comparison of late eighteenth-century Europe and the Qing reveals is the way in which phonocentrism served hegemonic ends—not by silencing anyone, but by “giving them a voice” and listening to them. Granted, Sicard’s dismissing of an uneducated deaf person as incommensurable with the moral order was blatantly discriminatory. Yet less conspicuous, and thus potentially more devastating, than the violence of being deemed incommensurable is that of being heard—commensurated—by an extractive ear. An exoticism certainly permeated the Enlightenment idealization of the perceived ability of Deaf people to communicate with natural gestural signs without interference from the hearing. But now that being phonographically audible to an extractive ear was indispensable for Hegelian-Thermidorian pluralism and progress, Deaf people were silenced precisely by being heard. Just as being granted a voix (vote) by the state depended on having one’s voix (voice) heard through a phonographic literacy test, hearing Deaf people similarly entailed the double negation of voice-extraction, turning them into a lack—a “silent statue”—by negating the sign language they had invented for themselves as that of a nonsentient automaton, only so that the “beneficent hand” of state-endorsed education could negate this lack by prying them open. Even by refusing such prying, a deaf person would still be “an utter nonentity in society,” still subject to its phonocentrism as a lack.²⁰⁸ Regardless of their self-recognition, Deaf people were already being assimilated according to the hearing world, already subject to a negation that foreclosed any “outside” to phonocentrism. They were afforded no other destiny than negating this negation—or, in Sicard’s words, “compensating for” their deafness and extracting a voice out of themselves toward the benevolent prying ear.²⁰⁹

Chinese scholar-officials and literati under the Qing might appear to have been a success story: instead of pining for any outside to phonocentrism, they showed that they too could be phonocentric. But for those whom the Qing did not have to co-opt in order to feed, finance, or facilitate its imperial projects, the price of phono-extraction was dear. Taiwanese Indigenous peoples, for example, were always paraded in visual representations of the Qing’s ethnic diversity, sometimes even shown to be making music (see figure 11).²¹⁰ Yet, like the many Indigenous “Miao” peoples in southern China of whom visual depictions also proliferated, they were absent from musical and philological representations at the imperial level.²¹¹ Such an absence was not because they, like Chinese writing, were apparently voiceless: Chinese officials and settlers in Taiwan described and transcribed Indigenous songs in local gazetteers, romanticizing them as living reverberations of the idyllic ancient time.²¹² Nor was it because they had no writing with which to record their voices: the Latin alphabet brought by earlier Dutch colonists was used by several Indigenous communities until the nineteenth century. In fact, the extant corpus of such writing, the Sinkang Manuscripts, fulfilled phonography’s extractive promise all too well, consisting entirely of deeds documenting the leasing out or selling of land by Indigenous communities. According to Lee Wen-liang, it was Chinese settlers who, fearing lawsuits and Qing policy inconsistencies, demanded that Indigenous peoples disgorge their voice together with their land in writing.²¹³ Hence, the moment in which they were phonographically heard, Indigenous peoples were already made superfluous to their land, which was reconceptualized as a wilderness to be tamed, tilled, and taxed through colonization.²¹⁴ As Robert Nichols puts it, colonial dispossession must commemorate Indigenous peoples as “original owners” in order to retroactively legitimize the commodification and loss of their land to settlers: in a double negation, “Indigenous ownership,” a retronym that negates traditional nonproprietary relations to land, is presupposed only in order that this ownership can be then negated, in turn, by sales to settlers.²¹⁵ Accordingly, Taiwanese Indigenous peoples could be heard by the Qing only as already being dispossessed—through voice-writing contracts. They were heard only as having their Indigeneity—a relation to land that lay outside Chinese property laws and customs—already negated, hence no longer needing to be heard as Indigenous.

Thus, in both cases of phonocentrism examined here, a transcendent ear compelled its subjects to extract voices out of themselves in order to maintain its hegemony. Yet it portrayed such nonreciprocal extraction as inevitable and beneficial for its subjects. Extraction was benevolent: it served as a means of tolerating difference, of including Deaf people in the social contract, of promoting imperial diversity, and, according to Qing officials wanting to assimilate Taiwanese Indigenous children through Chinese schools, of protecting these innocent subjects from the swindling low-class Chinese settlers (jianmin).²¹⁶ Through such all-inclusive benevolence, the ear foreclosed any possible outside to its phono-extractive regime—any possible outside to audism and settler colonialism—and constructed a world where there was nothing it could not pry open from its transcendent position. Even those marginalized from the extractive centers were not excluded: instead, they were assumed to be already moving toward the ear, already being assimilated as a lack that simply needed a helping hand. From the ear’s perspective, protecting these subjects could take no other form than furthering their assimilation—that is, listening to them even more—except that such assimilation could never be complete. Compensating for these assimilated lacks required voices to be extracted from them, yet such extractive compensation must presuppose that lack, that opening from which voices can be mined. After all, to negate its sonorous body with a silent phonography, the voice must presuppose such a body to negate, which meant that it must first use sound to negate silent mimesis. Hence, the voice must constantly presuppose an outside to phono-extractivism only so it can negate this outside and reaffirm its fealty to the ear again and again. All is foreclosed to the subaltern under phonocentrism except for being perpetually turned into a subaltern, perpetually compelled to speak to a hegemonic ear that always transcends its self-extracting reach.

But must the subaltern speak?²¹⁷ Must subalterns continue to extract a voice for the ear’s benefit, only to further their own subjugation as an extractable lack? The compulsion to speak and to be heard under an extractive phonocentrism resembles what Glen Sean Coulthard (Yellowknives Dene) critiques as the “politics of recognition” in North American settler colonialism. “In situations where colonial rule does not depend solely on the exercise of state violence, its reproduction instead rests on the ability to entice Indigenous peoples to identify … with the profoundly asymmetrical and nonreciprocal forms of recognition either imposed on or granted to them by the settler state and society.”²¹⁸ Undergirding colonial recognition is a two-pronged assumption—“the legitimacy of the settler state’s claim to sovereignty over Indigenous peoples and their territories on the one hand and the normative status of state-form as an appropriate mode of governance, on the other”—that denies Indigenous peoples any alternative.²¹⁹ Whether through schemes of recognition, redistribution, or reconciliation, the settler state compels Indigenous peoples to speak to its own ear in order to protect their livelihood, even though such livelihood entails a relation to land that is incommensurable with colonial capitalism.²²⁰ This “incommensurability,” pivotal in Indigenous studies, is therefore at complete odds with the late eighteenth-century European narrative of incommensurability critiqued by Tomlinson. The latter narrative ultimately assimilated the non-European world as a domesticated—albeit rigidly racialized—difference into a universalized Eurocentric taxonomy of “music” vs. “songs” (or musicology vs. ethnomusicology). In contrast, Indigenous incommensurability, according to Jodi Byrd (Chickasaw) and Michael Rothberg, does not bespeak “a ‘natural’ product of cultural difference but derives from established power differentials.”²²¹ Rather than vying to feed more voices to the extractive ear, which is “more than willing to ‘listen in’ on the subaltern, whether through surveillance, bio-piracy, or reified forms of consumptions,” Indigenous incommensurability seeks to disengage from it.²²² As Coulthard argues through reading Frantz Fanon, “those struggling against colonialism must ‘turn away’ from the colonial state and society and instead find in their own decolonial praxis the source of their liberation.”²²³

As the term itself indicates, “turning away” does not hinge on critiquing or even dismantling the transcendent ear, which “would in fact recenter [the] power” of colonialism, argues Dylan Robinson.²²⁴ Indeed, such an exclusively immanent strategy perpetuates the same teleological assimilationism as phonocentrism: it implies that only by first assimilating to colonial capitalism, by negating any outside thereto, could Indigenous peoples then seek to dismantle it, negating the first negation. Yet this dialectical process, similar to the voice’s self-extraction, could never be complete: according to Dipesh Chakrabarty, if asked when the Indigenous or subaltern are sufficiently assimilated to trigger the dismantling of colonialism, the colonizers will always answer, “Not yet.”²²⁵ Thus, as Leanne Betasamosake Simpson (Michi Saagiig Nishnaabeg) puts it, “I am not so concerned with how we dismantle the master’s house, that is, which sets of theories we use to critique colonialism; but I am very concerned with how we (re)build our own house, or our own houses. I have spent enough time taking down the master’s house, and now I want most of my energy to go into visioning and building our new house.”²²⁶

Where does this leave settler musicology? As Robinson argues, the incommensurability between decolonization and colonialism means that, just as Indigenous peoples need exclusionary spaces “for the ‘working out’ of ideas, challenges, and futures by and for Indigenous folks,” the same “working out among themselves” must sometimes be demanded from settler allies, lest they only extract Indigenous knowledge.²²⁷ This distinction between Indigenous resurgence and settler allyship resonates with the one Kuan-Hsing Chen draws between the decolonization of the colonies and the “deimperialization” of the metropoles. Referencing only non-settler colonialisms, Chen contends that decolonization will not succeed unless “the colonizing or imperializing population [also] examine[s] the conduct, motives, desires, and consequences of the imperialist history that has formed its own subjectivity.”²²⁸ Deimperialization, from which lessons might be drawn for settler colonial allyship, involves “undercut[ting] the desire to identify with the empire and support imperialist projects.”²²⁹ Like Coulthard, Chen also articulates this desire through reading Fanon. According to Chen, it is not just the colonized who are bound with their colonizers through “a powerful desire for recognition,” manifest either as a nominally “disidentifying” resentment or as a fantasy of replacing the colonizers.²³⁰ Middle-class subjects in the metropoles and across the world are also compelled by imperializing projects such as Cold War containments and neoliberal globalization to identify with and seek protection under empires.²³¹ Such imperial identification even pushes peripheral subjects and postcolonial elites to pursue “subimperial projects” of their own, which Chen illustrates by analyzing how Taiwan’s “Southbound Policy” in Southeast Asia reanimated the apparatuses of Japanese imperialism, a former colonizer of Taiwan.²³²

Since empire compels imperial and colonial subjects to identify with itself, deimperialization entails a “turning away” similar to that of Indigenous resurgence—“a psychoanalytic movement toward a new object of identification.”²³³ For Chen, such a new object must not be commensurated with empire through either identification or disidentification: a relation to empire must not be its ultimate referent.²³⁴ In Indigenous decolonization, particularly in what Coulthard describes as “self-recognition,” this new object hinges on a revitalized Indigenous self not already being assimilated into settler multiculturalism as just another “difference.”²³⁵ Meanwhile, according to Chen, deimperialization hinges on a practice of “inter-referencing”—“multiplying the objects of identification and constructing alternative frames of reference.”²³⁶ If empire demands that all voices move toward its own transcendent ear, a way to decathect from it is to move in a direction that is neither parallel to, nor intersectional with, nor opposite to that of imperial extractions—but skew. And if it is necessary and sufficient that skew lines do not share a plane, then an “inter-referencing” space for “turning away,” incommensurable with any single planar frame of reference, will be found somewhere between two or more imperial ears, each constructing for itself a totalizing plane of voices to extract and transcend.

Which finally brings me back to Amiot. Locating the French Jesuit anywhere near a position of incommensurability between eighteenth-century Qing and Europe can seem off the mark. It is much easier to count him among the cosmopolitan elites at both ends of Eurasia who extracted lands, labors, and voices from across the globe. The French mission subsisted on rents from its extensive landholdings in China, endowed by the Qing and the French Crown.²³⁷ For his writings, Amiot regularly received cash and wares from Bertin, customs and accounts having been cleared at Nantes and Guangzhou.²³⁸ The Jesuit himself also wanted to be remembered as a leisurely cosmopolitan. The makeshift frontispiece of Bibliothèque de l’Institut de France, Mss. 1515–1517, in which most of his letters to Bertin are now bound, features an engraved portrait of Amiot after a painting by the Italian Jesuit Giuseppe Panzi. A few months before sending the engraving to his sister in May 1793,²³⁹ Amiot composed a poem to his nephew, which ended up engraved beneath the portrait (see figure 12):

Dressed in a Manchu-style riding jacket or olbo, topped with the winter cap of Qing officials, and sporting a long beard (an emblem of elderly wisdom in Chinese iconography), the Jesuit is shown to commensurate worlds distant and different—“Frenchman, Manchu, [and] Chinese.” He had every right to check all three boxes. The “founding father” of Manchu studies in Europe, Amiot compiled the first Manchu dictionary in a European language, translated the Qing’s shaman manual, and even relayed a Manchu military song to Bertin in 1779.²⁴¹ Besides his enormous output on Chinese topics, Amiot’s “sinicization” is also evident in his later letters, where he distances himself from Europeans.²⁴² And what is more French than summarizing one’s life in alexandrines?

Yet an insecurity haunts this globalist self-image. The second line asserts that being in one of these groups does not mean “departing from” being in any other, the French word “déroger” commonly being used to describe actions unworthy of one’s social class.²⁴³ Apparently, Amiot feared that his multiple identities were incommensurable after all. This fear betrays him not as a cosmopolitan surveying and savoring the planet’s diverse riches, but as a survivor writing from the crevices between different empires, each orienting its subjects toward its own ear. Of course, Catholic missionaries in Beijing always had to navigate being both “apostles” of the church and “courtiers” of a heathen empire: the latter role legalized their presence in China.²⁴⁴ Yet Amiot’s marginality was exacerbated by both the Qing’s increasing hostility toward missionaries and rising anti-Jesuit sentiment in Europe. Though he successfully kept his mission in French hands after the Society’s suppression in 1764, Amiot, being the last French Jesuit in Beijing, saw his brethren pass away one after another. The French Revolution—that “modern philosophy” whose “poison” had infected so many, as Amiot bitterly reflected²⁴⁵—cut off his contact with the Republic of Letters, his patron Bertin escaping and dying in the Austrian Netherlands in 1792. When Amiot made his last will and testament in French and Chinese on November 9, 1793, he had “literally nothing” to leave behind, and mentioned no associate or inheritor other than three Chinese servants of whom we know only their names in French transcription.²⁴⁶

Amiot’s scholarly bequests met with no better fate: the longer he stayed in China, the less impact his new writings seemed to have in Europe. Though his Mémoire continues to be read today, Amiot never succeeded in publishing its “Supplément” despite repeated appeals to Bertin.²⁴⁷ The same indifference befell his 1780s treatises on dance, which Amiot considered an essential part of Chinese music.²⁴⁸ “De la musique moderne” and the “Divertissements” transmitted many songs, yet the same clichéd “air chinois”—a qupai in origin—continued to be (mis)copied, from Jean-Baptiste du Halde’s Description de la Chine (1735) to Rousseau’s Dictionnaire de musique (1768), Diderot and d’Alembert’s Encyclopédie (1751–72), Benjamin de La Borde’s Essai sur la musique ancienne et moderne (1780), and Pierre-Louis Ginguené’s Encyclopédie méthodique (1791), as if no other Chinese song existed.²⁴⁹ Though Amiot remained a consultant on China to the century’s end, suspicion that the Jesuits were too sinophile had long been brewing in Europe, a sentiment echoed in Herder’s Ideen.²⁵⁰ Meanwhile, in China, whereas Lülü zhengyi (1714) acknowledged contributions by his missionary forefathers, Amiot’s name is nowhere to be found in any Qing-imperial publications, even though his admission to the empire was likely granted on account of his musical knowledge.²⁵¹ The only surviving documentation of his scholarly collaboration with any Qing personnel concerns experiments in electricity and ballooning with Hung-u, a low-ranking cousin of the Qianlong Emperor.²⁵² The legacy of this admittedly fascinating collaboration was limited: it failed to elevate the abysmal standing of European missionaries or reverse the empire’s deepening indifference toward European learning and trade.

Was Amiot “worthy of some prize as a writer,” as his alexandrines asked? If globalization is the standard, he must count as a failure. He failed to facilitate broader connections across Eurasia. He failed to herald deeper understandings between the lettered classes of Paris and Beijing. He failed to even produce intercultural musical knowledge that was consistently valued in either of the two metropoles. And yet, this failed and dejected missionary has inspired the “inter-referencing” work in this article, particularly through his unpublished writings, whose “impact factor,” as we would call it today, was practically zero by comparison with that of his Mémoire. Of course, as specified in my introduction above, it has not been by “writing about” or extracting the contents of Amiot’s comparisons that I have strung together such distant topics as late eighteenth-century European discourses on deafness, diegetic singing in Qing court theater, and Qing-imperial philology. It has rather been by “writing with” Amiot’s positionality—not a cosmopolitan consumer or a global historian but an outcast misunderstood, dispensed with, and even persecuted from multiple directions—that I have been able to compare the phonocentrism at both ends of Eurasia. In so doing, I have attempted to proceed “appositely,” to borrow Robinson’s term. Robinson defines “apposite” in opposition to extractive or “hungry” forms of listening to and writing about music, whereby an “active partner”—typically occupying a heteropatriarchal, bourgeois, and/or colonial positionality normalized as a view from nowhere—dissects and digests a “passive object.”²⁵³ In contrast, “apposite methodology” approaches listening as a grounded experience of intersubjectivity between “listener, music, and space” and uses writing to “share space alongside or move in relationship with another subjectivity.”²⁵⁴ This form of writing rejects the demand for any transparent medium or transcendent standpoint facilitating the mining of contents or affects. Rather, it seeks “a more proximal relationship between writing’s form (its materiality, its flow, rhythm, or space, and the way it structures time) and the form we sense in musical subjectivity”—a relational “writing with,” in lieu of the extractive “writing about.”²⁵⁵

By “writing with” Amiot’s comparisons of European and Chinese ears, I have admittedly stretched Robinson’s intersubjectivity (listener, music, and space) to also include writing, or at least historical writings about music and listening from nominally “cross-cultural” perspectives. Yet I believe that an “apposite” comparativism—one aimed not at mining the crevices between empires and extracting the boundary-crossing knowledge of those living within them, but at using writing to create and share an inter-referencing space with these liminal subjects—could be a fruitful practice. The comparativism I envision seeks not to commensurate distant worlds as if browsing at a World Market or an imperial archive, but rather to move with those not on the shelves: those who did not have their utterances recorded by a phonograph, who could not or would not satiate the transcendent ear, who moved to or fell through the cracks amid the global formations of markets, colonies, imperial identifications, and area studies. Rather than making the mute(d) submit themselves to the empire’s phono-extractions, offering to be their voice, or listening to them with an even more benevolent ear, apposite comparisons ask why one must be heard in order to matter at all. Writing with the experiences of incommensurability and inter-referencing between multiple imperial ears may be a way for settler and imperial allies to move in tandem with the decolonial work of “turning away” from those ears and toward building new houses.