Drama Kings: Players and Publics in the Re-creation of Peking Opera, 1870-1937

The publisher gratefully acknowledges the generous contribution to this book provided by the Asian Studies Endowment Fund of the University of California Press Foundation.

Drama Kings

Players and Publics in the Re-creation of Peking Opera, 1870–1937

Joshua Goldstein

UNIVERSITY OF CALIFORNIA PRESS

Berkeley    ·    Los Angeles    ·    London

University of California Press, one of the most distinguished university presses in the United States, enriches lives around the world by advancing scholarship in the humanities, social sciences, and natural sciences. Its activities are supported by the UC Press Foundation and by philanthropic contributions from individuals and institutions. For more information, visit www.ucpress.edu.

University of California Press

Berkeley and Los Angeles, California

University of California Press, Ltd.

London, England

© 2007 by The Regents of the University of California

Library of Congress Cataloging-in-Publication Data

Goldstein, Joshua, 1965–

    Drama kings : players and publics in the re-creation of Peking opera, 1870–1937 / Joshua Goldstein.

        p. cm.

    Includes bibliographical references and index.

ISBN 978-0-520-24752-9 (cloth : alk. paper)

    1. Theater—China—Beijing—History. 2. Operas, Chinese—China—Beijing—History. 3. Mei, Lanfang, 1894–1961. I. Title.

PN2876.B37G65 2007

792.0951'156—dc22                                          2006036179

Manufactured in the United States of America

15   14   13   12   11   10   09   08   07   06

10   9    8    7    6   5   4   3   2   1

This book is printed on Natures Book, which contains 50% post-consumer waste and meets the minimum requirements of ANSI/NISO Z39.48–1992 (R 1997) (Permanence of Paper).

Contents

List of Illustrations

Acknowledgments

Introduction

PART ONE. (RE)FRAMING THE GENRE, 1870–1919

1. Late Qing Institutions of Peking Opera

2. From Teahouse to Playhouse

3. The Experimental Stage, 1895–1920

4. May Fourth Realism and Qi Rushan's Theory of National Drama

PART TWO. PEKING OPERA TO NATIONAL DRAMA, 1920–1937

5. Landscape and Figure, Nation and Character

6. The Limits of Reform

7. The Gendering of National Culture, Or, The Only Good Woman is a Man

8. Nationalization through Iconification

Epilogue

Notes

Bibliography

Index

Illustrations

Additional illustrations, keyed to the chapters of this book, and a glossary of Chinese characters, are posted at http://www-ref.usc.edu/~jlgoldst

1. Tan Xinpei

2. The Four Famous Dan, from Northern Pictorial

3. Lithograph of teahouse brawl, by Wu Youru

4. Teahouse floor plan

5. Guangxu-era teahouse

6. Seating at the Reform Customs Theater

7. Mei Lanfang in contemporary dress as title character in Deng Xiagu

8. Mei Lanfang in contemporary dress as Li Renfen in A Thread of Hemp

9. Mei Lanfang in (new) ancient costume in The Goddess Scatters Flowers

10. Mei Lanfang in (new) ancient costume as in A Thousand Pieces of Gold for a Smile

11. Shang Xiaoyun as Bojidi in Modengjia Girl

12. Advertisement for Mei Lanfang cigarettes

13. Chart of dan roles from late Qing to the Republic

14. Xun Huisheng as Li Feng in Meilong Town

15. Cheng Yanqiu as Zhang Huizhu in Tears on a Barren Mountain

16. Mei Lanfang riding with Mayor James Rolph in San Franscisco

17. Mei Lanfang at reception in Hawaii

18. Mei Lanfang receiving honorary doctorate from Pomona College

19. Set of stamps commemorating Mei Lanfang, 1962

Acknowledgments

As with many other aspects of my life, my karmic debts as an academic are beyond my capacity to summarize in these acknowledgments, much less actually repay. I would have to begin, however, with my mentors and classmates at the University of California, San Diego. Joseph Esherick and Paul Pickowicz demonstrated patience bordering on the masochistic when they decided to accept a fidgety welfare advocate into their program with the hope of making him into a Chinese historian. I'm still hoping to become one some day, and continue to rely on their advice and scholarly acumen in pursuing that goal. Dorothy Ko opened intellectual doors for me that I would never have thought to look behind and has provided insights from a conceptual plane that I am occasionally able to glimpse but will never be able to attain. Takashi Fujitani was incredibly generous with his time, energy, and guidance, particularly during some unexpectedly trying moments. Masao Miyoshi, James Holston, George Lipsitz, Lisa Lowe, and Suzanne Cahill were great teachers who helped me form the basic ideas behind my research.

As important as these mentors were in training me, my classmates were even more so in sustaining me. Conversations with Madeleine Yue Dong and Michael Chang have been invaluable, and both of them have tried their best to save me from intellectual embarrassment. Sue Fernsebner, Xiao Zhiwei, Andrew Morris, Eric Cazdyn, Elena Songster, Cecily McCaffrey, Liu Lu, Mark Eykholt, Wang Liping, and Julie Broadwin all read chapters, brought new sources to my attention, and helpfully critiqued my work; and Jim Cook was a guardian angel.

I have also had the tremendous good fortune, both during and after graduate school, of being financially supported in writing this manuscript, both directly and indirectly, by a number of institutions. While I was at UCSD, the Gleich fellowship helped me through the most intensive year of research and writing. The Yale Council on East Asian Studies gave me the better part of a year to work on this and another book, and Valerie Hansen, Deborah Davis, and Helen Siu were always generous with their time and encouragement. And Franklin and Marshall College provided a semester's leave and summer stipends that allowed me to complete my manuscript.

During my time in Beijing, several scholars and Peking opera experts were extremely generous and helped me get my bearings. Professor Guo Weidong took me under his tutelage for a year of coursework at Beijing University and has been unfailingly helpful in guiding me through the intricacies of doing research in China ever since. Professor Liu Yigao was a patient teacher. Mei Shaowu and Tu Zhen made this stranger feel quite welcome in their home and at the dinner table one chilly day over ten years ago; and the opera aficionado Wu Xiaoru also gave me some of his valuable time.

Two years of working with the University of California Press have been crucial to making this book a reality, not simply as a material fact of publishing, but as (I hope) a readable work of scholarship. My greatest thanks go to Colin Mackerras and David Strand for their perceptive comments on the manuscript, as well as for providing the book with a title. Reed Malcolm, Kalicia Pivirotto, and Mary Severance held my hand each step of the way, and two anonymous readers at UC Press also provided crucial feedback. Finally, Erika Büky did an amazing job of editing the manuscript—any errors of writing and interpretation remain entirely my own, but that any sentence or footnote in this book is readable at all, let alone properly and consistently formatted, is due to her.

Finally, a constellation of friends and academics has helped me in sundry ways through the many years with this project. Doug Anthony, Tani Barlow, Jennifer Callahan, Eileen Chow, Kirk Denton, Sharon Marcus, Scott Schaffer, Dina Suggs and Mark Swislocki all were subjected, in either oral or written form, to parts of this book and were generous with their feedback. Charlotte Furth, foremost and most youthful among many other young and friendly colleagues at the University of Southern California, has been a fantastic guide through the fogs of academe. Michael Kowalski and his family always let me eat more than my share of potatoes gratin at Christmas supper, even if I spent most of Christmas afternoon being antisocial, hunkered down in the attic over a duffel bag full of photocopies. I dare not count how many times my sister helped me get an elusive library book or my mother and father emailed a crucial file, overnight mailed a paper, or otherwise bailed me out—in addition to just being a constant, supportive presence, quick to offer an ear or a weekend ski trip, even if I tended to turn both down. And the most thanks of all—too much to fit into a sentence, or a book, or a life—to Cynthia Freeman, for keeping me laughing and sane and crazy and going.

This book is dedicated to those who struggle with love to stay awake.

Introduction

Jingju [capital drama]? Pingju [Beiping drama]? Jiuju [old drama]? Guoju [national drama]? The people of the nation call it by different names. Before Beijing became Beiping it was called Jingju; afterward it was called pingju. The capital is old—a great place—so Jingju is also called old drama. Promoters of new drama [spoken drama] see Jingju and call it old drama. Promoters of Jingju think it should be called national drama. It is all the same drama, but the names are unclear. The name needs to be rectified.

Sun Danhan, Jingju? Pingju? Jiuju? Guoju? Xiju xunkan (Drama Biweekly), 1936

In July 2001 China Central Television launched its eleventh national channel, CCTV 11, dedicated to broadcasting traditional Chinese opera and music to all of China's thirty-four provinces and autonomous regions. Though its programmers boast more than two hundred kinds of Chinese musical drama in their broadcasting repertoire, the genre that dominates CCTV 11's fourteen hours of daily air time is Peking opera: daily programming usually includes more than an hour of Peking opera singing and music lessons and a broadcast of a full-length performance. Also, in the prime-time lunch and dinner hours, the voices of Peking opera's greatest stars of the 1930s can usually be heard singing their signature roles to a hungry nation. These programs, of which there are hundreds, involve today's star actors performing actions and gestures in sync with sound recordings from past decades, in a sort of reverse karaoke. For a historian researching Peking opera, these shows offer not only a wonderful insight into the vocal and staging techniques of actors past but also a daily reminder of the iconic stature of Republican-era actors in shaping Peking opera into a genre of national importance, then and today.

Peking opera is by far the most famous of China's approximately 360 theatrical genres. It is also the only one that, despite being named after a locatable city, is not merely categorized as a regional drama (difang xi) but put in a category of its own, either explicitly or implicitly placed under the rubric of national drama (guoju). Yet, despite its valorized position, the origin of Peking opera remains controversial, because Peking opera is a composite and ever-shifting fusion of several melodic and dramatic forms.¹ For many drama historians, this is precisely why discerning its aesthetic essence and combing its tangled past into a manageable history pose such an enticing challenge.

Although almost all parties agree that Peking opera began to cohere as a form sometime between 1790 and 1860, the term Peking opera (jingju, jingxi, or pingju) does not enter common usage until the early twentieth century, and it originates not in Beijing but in Shanghai:

Pingju: these two characters start being used after the Republic [1912]. It first appears during the early years of the Republic. When foreign students returned from abroad, they did not know what this thing pihuang was, but when they came back to China they first had to go to Shanghai, and all of Shanghai's best actors were brought down from Beiping. When they also saw that this thing, pihuang, was being sung in all the theaters, they simply assumed that this must be the Beijing melody-style (qiangdiao), so they called it jingxi [capital plays]. After the central government moved south [in 1928], Beijing [lit. northern capital] was renamed Beiping [lit. northern peace], and it was no longer good to call it capital plays, so they renamed it pingju, and this is the origin of the name pingju. Actually, the denizens of Beiping never called it jingxi or pingxi.²

The simple answer, then, according to Qi Rushan—probably the most influential, encyclopedic, and prolific expert on Peking opera in the twentieth century—is that Peking opera (whether written as jingju, jingxi, or pingju) is merely a label pasted on to dramas performed, for the most part, in a style comprising several dozen melodic themes otherwise known collectively as pihuang.³ This simple answer masks a technical complexity that is revealed the instant this label is peeled back. The above passage continues:

During the Qing dynasty, before the Tongzhi reign [1862–75], the most popular styles in Beiping were yiqiang, kunqiang, and bangziqiang. At this time bangzi and pihuang were not yet fully developed. Around 1860, bangzi and pihuang were competing neck and neck with kun and yi styles, and with all four styles roughly equal, no one style deserved to be named jingxi, so at that time there was no word jingxi. In fact, at that time [pihuang] was usually called erhuang, and even the word pihuang did not exist. Only at the end of the Qing did newspapers begin to use the word [pihuang] or huiban [the Anhui drama troupes’ style]. Anyone who wanted to hear pihuang performed might alternatively say they were going to hear huiban, but absolutely no one said huiqiang [Hui tunes or style]. The Anhui troupes before the Tongzhi period all mainly performed the kun style and did not specialize in pihuang until Cheng Changgeng, who did so at the end of his career. He first sang kun opera, and later added pihuang. At that time actors sometimes said the two characters, jingqiang [capital melodies], but this meant a specific style of Hui melody [i.e., it did not mean jingju or Peking opera].⁴

To avoid excessive terminological complexity, in this book I generally use the Anglicized term Peking opera. This term in fact brings its own significance to the discussion: it emerged during the era under study, and, being in English, bears the traces of colonial modernity, a context of great importance in shaping the genre's identity.⁵

What we call Peking opera is a modern construction: its parameters, performance, and disseminations were greatly affected by the conditions of colonial modernity. And though few scholars dispute that Peking opera is of relatively recent vintage by Chinese standards, its modernity is much more controversial. Even though the current scholarly consensus is that Peking opera did not coalesce until about 1845, it is still generally depicted as fundamentally isolated from issues of modernity and foreign imperialism—historical factors that, by any standards, were already shaping Chinese history. There is a powerful impetus to construe Peking opera as wholly traditional and purely Chinese: this is a picture of exaggerated contrasts, the outlines of which were etched during the Republican era (in particular the May Fourth era, 1915–25), amid the multiple crises of foreign imperialism, regional fragmentation, economic upheaval, and political and cultural revolts. But just a brief glance through the newspapers and tabloids of the late Qing and Republican eras reveals that Peking opera was the height of fashion not only in Beijing, but also in treaty ports such as Tianjin, Shanghai, and Hankou, urban spaces in which imperialist intervention, foreign cultural influence, and modern urban technologies were at their most visible and explicit. In the 1920s and 1930s, Peking opera stars set fashion trends, negotiated contracts with recording companies and movie studios, and socialized with foreign and Chinese dignitaries, bankers, and journalists. Nevertheless, Peking opera was seen as somehow in modernity but not of it. This apparent contradiction makes interpreting its development in these decades all the more fascinating.

This study is neither a search for Peking opera's historical origins nor the obverse, an attempt to strip away the obfuscations of false consciousness to reveal Peking opera as a purely modern manufactured myth. Rather, the question I am asking is this: Given that Peking opera was a composite of forms that was (and still is) in flux, what were the historical forces that made it seem urgent to all concerned—its aggressive detractors as well as its reformers and preservationists—to assign it a fixed essence, even though they were all quite aware that this very action represented yet another layer of influences reshaping the form? How did the changing discursive and social context of the late Qing and Republican eras—the explosion of mass-reproduced commercial print and visual media, the ascendance of discourses of the nation and the individual, changes in how gender was conceived and performed, and the economic and social experiences of colonial modernity—influence how Peking opera was perceived, produced, and performed? And how did efforts to define it influence what it could represent and how it could do so? These questions are all the more relevant because Peking opera throughout this period was arguably China's most commercially dynamic and widely advertised, watched, and written-about form of cultural entertainment. Thus this study, while concentrating on the transformations and conceptualizations of Peking opera as a genre, is ultimately concerned with illuminating issues of late Qing and Republican era history, particularly urban popular culture and gender construction, and their intersection with national formations.

My approach to Peking opera is indebted to the works of numerous scholars, most obviously Eric Hobsbawm and Terrence Ranger's work on invented traditions, an idea that sparked its own veritable subgenre of cultural history based on the now almost axiomatic assertion that the very appearance of movements for the defense or revival of traditions, ‘traditionalist’ or otherwise, indicates a break in historical continuity, a modern rupture with the past.⁶ Yet, though I argue that Peking opera's formation as a genre is intimately interwoven with technologies of colonial modernity, I also hope to reframe the premises of the invented-tradition approach. It is often assumed that behind every invented tradition lurks a discernible ideological agenda, or that unveiling the manipulated and constructed nature of such practices somehow demystifies them and weakens their ability to mobilize participant-spectator affect; in most instances neither assumption has proved correct, and certainly neither applies well here. Indeed, many of the artists and scholars of the period seem to be perfectly aware that they are constructing Peking opera, (re)interpreting it, and shaping it into a tradition; and they seem remarkably comfortable with the paradox of inventing tradition—was this not what it meant to be an active participant in one's own culture, history, and nation? The beautifully oxymoronic neologism invented tradition ascribes a kind of manipulative agency to a tradition's inventors and a false consciousness to its consumers that requires reconsideration.

Rather than build on the dubious opposition of modernity and tradition, I instead approach Peking opera as an object of a certain kind of knowledge production, enmeshed in the context of colonial modernity. Evidence for the assertion that Peking opera's development and spread was shaped by technologies and practices of colonial modernity is plainly laid out in the chapters that follow, but a brief description of what colonial modernity implies is in order here. A framework most prominently championed by Tani Barlow and a recent generation of postcolonial scholars, it proposes that Western-modeled modernization and colonialism must be grasped as integrally interwoven processes involving discursive powers that increasingly connect at key points to the globalizing impulses of capitalism.⁷ This framework proposes neither a wholesale importation of modernity from the West (an assertion untenable in most cases, and certainly in that of China, where elements of arguably modern commercial, urban, and social relations date back to at least the late 1500s) nor a simple substitution of an alternative national modernity—a distinctly Chinese modernity, say, which shifts issues of inequality, class, and gender into an essentialized nationalist framework. Rather, the term highlights how the context of colonial domination compelled the reorganization of institutions, technologies, and practices so as to address and negotiate its threat, resulting in a translation of colonized societies’ production and reproduction processes into frameworks interpellated by the dominant powers (in China's case both Western and Japanese).

A colonial-modernity framework is thus highly compatible with the analysis of colonialism as a process of knowledge construction pioneered by Bernard Cohn and expanded on by scholars applying and modifying the theories of Michel Foucault, such as Timothy Mitchell and Partha Chatterjee. In his work on colonial India, Cohn recognized that colonialism involved far more than merely the brute force of invasion, that in coming to India, [the British] unknowingly and unwittingly invaded and conquered not only a territory but an epistemological space as well. The ‘facts’ of this space did not exactly correspond to those of the invaders. Nevertheless, the British believed they could explore and conquer this space through translation: establishing correspondence could make the unknown and the strange knowable.⁸ This process of reclassifying facts not only imposed British notions of property rights and laws on Indian socioeconomic relations but also reached into the cultural practices and history of the colonized: Cultural forms in societies newly classified as traditional were reconstructed and transformed by and through this knowledge, which created new categories and oppositions between colonizer and colonized, European and Asian, modern and traditional, West and East.⁹ Vernacular languages were classified into systematic grammars, artifacts and texts into archeological taxonomies, the laws of the Mughal Empire and Hindu and Muslim communities into corrupt forms of despotism and theocracy that needed subjugating to Britain's more judicious rule. Mitchell, Chatterjee, Barlow, and other postcolonial scholars have added to this insight by showing that these processes were not simply the acts of colonizers on the colonized, but that anticolonial nationalists, nativist modernizers, and the semicolo-nized elites of China often participated in similar projects of knowledge production, with the goal of helping to forge more powerful and manageable state forms that might ultimately contribute to the struggle to resist colonial domination.

While such an approach may sound abstract, I attempt to present a generally chronological narrative, grounded in an array of evidence that demonstrates these notions to be historically concrete. Chapter 1 describes the basic institutions of Peking opera in the nineteenth century. It introduces a variety of economic and social institutions—Qing court patronage, commercial acting troupes, schools for actors, and commercial theater networks—to demonstrate how Peking opera, from the moment it coalesced as a recognized dramatic form, was as much enmeshed in the social and political structures of the Qing imperial metropole as it was in a network of touring and commercial expansion. These developments were made possible through technologies, such as steamships, railroads, and newspapers, that point to the importance of the context of colonial modernity. In this highly commercialized and prestigious convergence between Qing imperial and colonial modern networks, the social dynamics of gender performance played a central role in Peking opera's ascendance: the dominating presence of the laosheng actor (who took the roles of older male characters, models of late Qing masculinity) was crucial to the form's social and cultural elevation.

Chapter 2 focuses on the teahouse theater as a microcosm of late Qing social relations and cultural tastes, then traces some of the basic shifts in the organization of Peking opera theaters as social spaces catalyzed in the early twentieth century by the introduction of the new-style playhouse centered on the proscenium stage. The influence of architectural space and audience socialization are discussed throughout the book and are also the primary focus of Chapter 6. These themes will be immediately familiar to historians of modern European and American theater and popular culture: the works of Norbert Elias and Michel Foucault on the social disciplining of the public and of Jean-Christophe Agnew, Jeffrey Ravel, James Johnson, and Lawrence Levine on Parisian, English, and American theaters of the eighteenth and nineteenth century all serve as bases of comparison.¹⁰ Another useful point of reference is provided by Timothy Mitchell's analysis in Colonizing Egypt of the integral connection between disciplinary practices and modes of representation. Following Foucault, Mitchell argues that the colonial reorganization of Egyptian military forces, schools, villages, and cities all involved processes of ordering space to make populations legible and manageable, a process he calls enframing: Enframing is a method of dividing up and containing…Within these containers, items can be isolated, enumerated, kept.¹¹ Mitchell points out that the disciplining effect of enframing is inseparable from its representational effect—the construction of reality as an exhibition, of social spaces as material realizations of conceptual models. The supposed superiority of enframing technologies is based on this effect: the organized colonial city is said to be superior to the squalid and disorderly native one, the modern school of self-managed learning superior to the religious one of masterly authority over pupils, because these spaces are material realizations of legible conceptual models. The new playhouse theater functioned similarly. With its detailed blueprints, numbered and assigned seats, and division and isolation of activities—purchasing tickets at ticket booths, snacking and chatting in the lobby, watching attentively in the auditorium—it was an explicit technology of enframing aimed at a new Chinese public of national citizens, its self-disciplining order clearly superior to the uncivilized, status-riven chaos of the teahouse. In later chapters, I elaborate on Mitchell's concept of enframing to describe a shift in the dominant practices of representation in Republican-era urban China more generally. Great efforts were made to create and enforce a demarcation between social reality and cultural representation, a division that would be crucial in reshaping Peking opera as a genre, and gender as it was performed on the stage.

Chapter 3 places the changes in the Peking opera theater in a broader historical context of late Qing and early Republican efforts to reform the Qing polity into a Republican nation-state. From the 1890s to the 1910s, a vibrant concatenation of reformist experiments converged in the Peking opera theater, both in the onstage productions of plays—which increasingly treated contemporary and global issues using all the technologies available in the new-style playhouse—and in the audience, particularly with the integration of women into what had previously been the all-male space of the public theater. Theaters were at once spaces of social liberation and gender experimentation and sites over which various social fractions—male actors and the newly emerging female actresses, male and female fans, foreign-educated students, and wealthy patrons—competed for control, while drama as a medium of cultural communication came to be seen as crucial to shaping citizens’ political attitudes and social behavior. This atmosphere of experimentation dissipates rather abruptly, however, in the May Fourth era, and chapter 4 is devoted to explaining the various factors that contributed to Peking opera's transformation from a relatively loose and elastic dramatic form to a much more rigidly delimited genre used almost exclusively to represent traditional Chinese subject matter. While chapter 4 starts with a discussion of May Fourth intellectual attacks on Peking opera as an inherently feudalistic and backward art form—arguments quite familiar to those acquainted with the dominant historiography of the May Fourth movement as the pivotal event of modern Chinese cultural radicalism—it also suggests the need to revise our dominant narrative of the era. Indeed, from the study of a popular cultural form like Peking opera, we find that the May Fourth moment was as much defined by the discursive engagement of intellectuals reductively characterized as conservatives as by self-proclaimed radicals. More important, the dichotomies so characteristic of this moment (for example, tradition versus modernity, Confucianism versus science and democracy, Chinese versus Western) were not the enlightened products of a Chinese cultural avant-garde—a notion that, despite its centrality to Chinese historical materialist orthodoxy, is profoundly idealist. Rather, they took hold so quickly and pervasively because they articulated experiences already embodied in urban daily life practices and public spaces, including—but not at all limited to—commercial theaters.

Part 2 of the book is dedicated to describing Peking opera after the May Fourth moment. In the 1920s and 1930s a concerted effort was made to construct the genre as a model of Chinese national culture. In these decades the Peking opera stage was dominated by male dan actors (male performers of female roles), epitomized by the Four Famous Dan (Mei Lanfang, Cheng Yanqiu, Xun Huisheng, and Shang Xiaoyun). Chapter 5 takes as its subject the Peking opera star, both as a national celebrity and as a character on the stage. As stars and their images circulated throughout the nation, they became subjected to endless gossip and scrutiny, publicity that manifested the close interconnection between a growing commercial investment and the public's increasing emotional investment in these national figures.¹² The attention paid to stars’ personal lives and political commitments in the tabloid press not only informs us of the networks of patronage and influence to which stars were indebted; it also intersects with an increasing emphasis on the actor's ability to convey the deep emotional interiority of characters onstage. This concentrated attention on the actor as a public citizen who specialized in crafting and portraying the psychologically complex individual was, of course, related to other social and institutional trends, and this discussion continues in chapter 6, which describes the continued attempts to reform theaters and acting schools in order to foster the proper cultural habits of self-regulating citizenship.

The changing terrain of gender performance and homo- and hetero-sexuality within the theater converge in the description in chapter 7 of the rise of a new form of female role known as the huashan (lit. flower gown/garment). If women were becoming increasingly influential in urban theaters, both as actresses and audience, why was the Peking opera stage overwhelmingly dominated by male dan actors? Borrowing in part from the works of Judith Butler, Eve Sedgwick, and Faye Dudden,¹³ this chapter explores the ramifications of the representational practices of enframing for constructions of sex and gender, arguing that this regime of representation clearly had uneven effects on the public perception and reception of male and female actors.

Chapter 8 goes on to describe the rivalry between the two most famous male dan actors, Mei Lanfang and Cheng Yanqiu. Both men undertook tours of Western countries as a way of raising Peking opera's status as a globally significant cultural form. Peking opera's journey to the United States and Europe reveals remarkable parallels between the discourses of cultural nationalism that recast Peking opera as national drama and the discursive categories of Orientalism.¹⁴ Mei Lanfang's eventual victory in this rivalry also tells us something of the weakness of Republican-era state institutions (with which Cheng Yanqiu was more cozily aligned) when compared with the diffuse power exerted by the commercial media (Mei Lanfang's most actively cultivated and supportive source of patronage).

The dominant thread of this narrative, which spans more than seven decades, is Peking opera's steady rise to national prominence, a rise that is full of tensions and transformations inflected by China's shift from a troubled empire to a troubled republic. Early Peking opera was influenced by Qing court patronage: during Tan Xinpei's heyday in the late Qing, the genre was closely associated with the patronage of the Empress Dowager Cixi, renowned as an opera addict. Yet when the Qing dynasty fell, to a chorus of denunciations, Peking opera continued to thrive. It was beloved by both Mao Zedong and Zhou Enlai, who had an expensive load of Peking opera costumes and other paraphernalia shipped to the remote Communist base area of Yan'an during the war against Japan in the late 1930s. During the same years, the Japanese puppet government of Manchukuo dubbed Peking opera its official national drama, hoping to use it as part of a project of cultural assimilation by playing on Peking opera's historical connection to the Manchu court and its great popularity among Han Chinese. And at the same time, in remote south-western China, where the Guomindang (GMD) built their base during the Anti-Japanese War, Peking opera was being promoted and warmly received as a key part of the Chiang Kai-shek government's cultural programs promoting national resistance against Japan. In other words, within a few decades Peking opera went from being closely associated with the Qing imperial court—whose existence and legacy Republican leaders attacked as obstacles to Chinese nation building—to being a cultural form so integrally tied to the idea of a Chinese nation that governments across the political spectrum all saw it as a powerful tool for building cultural nationalism. This book is devoted to tracing some of the sociopolitical and discursive processes responsible for shaping Peking opera between these two historical moments. We will be as interested in the historical forces that motivated Peking opera's construction into a monument of Chinese national culture as in the aesthetic timbers out of which this edifice was built. It will be useful, therefore, to present a brief sketch of the genre—an introduction to the basic melodic and stylistic components from which Peking opera was formed.¹⁵

PEKING OPERA IN THE NINETEENTH CENTURY: A MELODIC PATCHWORK

In the early Qing dynasty (1644–1911), the most popular forms of drama in Beijing were kunqu and yiqiang, both of which originated in the Southern Yangzi (Jiangnan) region. Handling wen (civil) themes and generally centered on a romance involving a beauty and a scholar, kunqu was accompanied by gently paced music supplied by a small orchestra called the wenchang (lit. civil stage), which was led by two flutes (dizi) and included minimal percussion. Kunqu very rarely directly portrayed martial (wu) themes, and, although full of graceful physical movement, included little in the way of exciting acrobatics.¹⁶ In contrast, other dramatic styles might stress wu themes (or alternate between wen and wu). The orchestras for the highly acrobatic martial plays, the wuchang (lit. martial stage), were dominated by drums and gongs. Kunqu was the only form of drama in the Qing to hold the privileged designation ya (elegant, refined; cf. yabu, lit. elegant register). All other dramatic forms, whether portraying civil or military themes, were designated as either huabu (lit. flower registers) or luantan (assorted melodies) and viewed as predominantly popular forms. The yabu designation is usually associated with imperial court favor, and, through most of the Qing, kunqu was the primary genre staged within the Forbidden City. Yet some huabu forms were popular with the court as well. Yiqiang was performed inside and outside the palace grounds throughout the Qing. Other drama forms, like bangzi, clapper operas (a broad category of highly percussive dramas with a variety of regional forms), were seen as cruder and were not permitted in the palace; this more polarized difference was expressed not through the ya and hua opposition but by the opposition of ya and su (plain, popular, crude).

In the 1780s, a newcomer from the west (Sichuan and Shaanxi), qinqiang—an exciting and more su style—arrived in Beijing. Qinqiang was a regional subtype of bangzi. The leading exponent of qinqiang was Wei Changsheng. Wei was a charismatic male performer of female roles, a dan (women were generally prohibited from performing publicly in the Qing, though troupes of actresses sometimes performed in households and at private affairs). Wei Changsheng quickly became notorious for his bawdy antics, and qinqiang was banned from Beijing in 1785. Wei then took the style to Jiangnan, where many of his techniques for portraying female roles, especially hairstyles and costuming—in particular the qiao, a sort of small stilt that gave the illusion of the actor's having bound feet, and was hence particularly useful for portraying flirtatious or lascivious female characters—were picked up by Jiangnan actors of various styles.

Wei's bangzi influence was not long absent from the imperial capital, however. In 1790, in celebration of the Qianlong emperor's eightieth birthday, a general call went out to Jiangnan opera companies to visit Beijing and join in the festivities, partly because of their mastery of kunqu, but also because Qianlong was entranced by the huabu dramas of the south, including yiqiang (also known as yiyangqiang, or jingqiang). These troupes, known as the Anhui companies, were a hit with both the court and Beijing urbanites generally, especially when spiced with Wei's qinqiang influence. Greeted by great success in the capital, many of the southern companies chose to remain in Beijing, though they continued to recruit most of their young apprentice actors from the Jiangnan region. The most successful and renowned of these companies were the size of a small battalion (including up to 150 actors, musicians, and stagehands) and came to be known as the Four Great Anhui troupes. One of the central melodic modes that these companies used (one that was dominant in yiqiang) was known as erhuang.

These troupes, with one or two rises and falls along the way, dominated the Beijing drama world for most of the nineteenth century and were the key institutions whose creative alchemy would, over a period of several decades, result in the development of Peking opera. By consistently absorbing fresh local styles and offering a blend of wen and wu, ya and su dramas, they managed to appeal to the tastes of an urban audience ranging from small local peddlers to high-ranking officials. Aside from kunqu, which was still a favorite among the literati, by far the most popular melodic style up to the 1830s was erhuang. The most esteemed troupe of the 1830s, the Sanqing (Three Celebrations) Company, was led by a dan actor, Gao Langting. Gao specialized in erhuang. Despite the fact that Gao was stocky and more than thirty years old, a venerable age for a dan actor, it was claimed that the moment he steps onto the stage he is the essence of womanliness, with not a whiff of brawny force. He always entrances those who see and hear him into forgetting he is a fake woman.¹⁷ The primary historical sources for this period, a dozen or so highly literary diaries of opera buffs, are almost exclusively dedicated to extolling and ranking the dan performers of the time (more than 160 are described), and there is no doubt that Beijing's opera world was overwhelmingly dominated by men who specialized in performing female characters. All the Anhui troupes, and most of the leading troupes in the other varieties of opera (including bangzi and jingqiang) had a dan at the helm until the 1830s.

Around this time several sheng (male role) performers of a local Hubei style called handiao became extremely popular in the capital, and the Anhui companies started recruiting them into their ranks. The core melodic mode for handiao, known as xipi, included many frenetic and bright melodies especially suited to acrobatic wu performances. With these xipi performers, the Anhui troupes presented programs that mixed kunqu, erhuang, and xipi. Far and away the most impressive figure of the day was the laosheng Cheng Changgeng (1811–79), an Anhui native who took over as head of the Sanqing company. Although great dan actors did not fall into obscurity, from around 1850 until 1910, laosheng actors bestrode the Beijing drama world, heading the majority of the capital's largest opera troupes. Similarly, kunqu did not disappear, but by the 1870s it was taking a backseat to xipi and erhuang. By the turn of the century the term pihuang—combining the pi of xipi with the huang of erhuang—was commonly used to describe the dramas performed by Beijing's major drama troupes.

Though combining xipi and erhuang was typical by the 1870s, and the term pihuang was common by the 1890s, it was by no means a closed musical system. Throughout the late nineteenth century it was fashionable for pihuang troupes to perform alongside bangzi companies, resulting in mutual influences. Melodic variations were constantly being invented. The 1860s and 1870s saw a radical change in the main instrumental accompaniment, from the paired dizi (flute) to the huqin (a two-stringed fiddle).¹⁸ In the 1880s, the laosheng Tan Xinpei (1847–1917) began assimilating many erhuang melodies previously used only by dan characters into his performances.¹⁹ In the early twentieth century, Tan exchanged melodies with his friend Liu Baoquan, the era's most famous performer of dagu (lit. big drum song), a much simpler musical style that shared very little with Peking opera. Yet, according to the drama historian Wu Xiaoru, when Tan sang these tunes "from beginning to end everyone felt they were Peking opera (jingxi) and not dagu or anything else."²⁰

This brief sketch of Peking opera's melodic makeup substantiates that the composite dramatic form called Peking opera was indeed constantly changing throughout the nineteenth century. The project of identifying the genre's stylistic or musical essence is like trying to draw straight lines distinguishing blues from rock and roll, and both from R & B: it can spark meaningful discussions and generative critiques and histories, but in many ways the distinctions are irresolvable. The most productive and succinct approach perhaps follows the spirit of Wu Xiaoru's observation: like so many other cultural phenomena, Peking opera was hard to describe succinctly, but people felt they knew it when they saw it. Such a definition forces us to pay as much attention to how the term itself was deployed and defined as to the form and content of the dramas so defined and the processes by which they were produced and consumed.

PART ONE

(Re)Framing the Genre, 1870–1919

CHAPTER 1

Late Qing Institutions of Peking Opera

Cheng Changgeng's first performances on the Beijing stage in the early 1830s were disastrous. Alternately ignored and heckled by audiences, the proud young laosheng performer vanished from the stage for three years, during which he practiced assiduously.¹ When next Cheng opened his mouth onstage, at a prestigious private competition to perform one of the most challenging roles in the laosheng repertoire—that of Wu Yuan in the play Escape from Zhao Pass (Wen Zhaoguan)—Cheng unleashed the most powerful and influential voice of his generation, a voice which, it was said, could soar through the clouds and split rocks, its reverberations trailing deliciously around the rafters, and amid this superb, soaring clarity conveying still another deeper and more powerful note.² Cheng became identified with some of the most challenging male lead roles, including that of Guan Yu, the paragon of loyalty from the Three Kingdoms saga and a deity in the pantheon of the acting world. It was a role with strict protocols of makeup and costume that Cheng utterly transformed. Indeed, Cheng Changgeng was so central to the consolidation of Peking opera that several drama historians have called him its founder.³ A thrilling performer of heroic roles onstage and a commanding personality offstage, Cheng headed both the Three Celebrations Drama (Sanqing) Troupe and the Beijing Actors’ Guild for nearly twenty years. One brief biography from 1927 attempts to capture Cheng's daunting stature:

[Members of the acting community] called him Big Boss (da laoban), and he continues to be remembered by this name even today. He had no flaw to be ashamed of in either his artistic qualifications or his personal character…His stage presence was awesome, even inspiring fear. An anecdote conveys this. Once, at a private performance (tanghui), he performed Guan Yu in Attacking Changsha (Zhan Changsha). His expression was so awe-inspiring, his eyes so scorching, that the audience found it impossible to look at him directly. When he took the stage and struck his entrance pose, it was like a true god descending, and made the audience all unconsciously rise from their seats in reverence.⁴

On his deathbed in 1879, Cheng passed the leadership of his Three Celebrations Troupe to the laosheng actor Yang Yuelou (1844–90), bypassing one of his young students, the laosheng Tan Xinpei (1847–1917). It is said that although Cheng knew Tan was destined to become the most acclaimed laosheng of the next generation, he was convinced that under Tan's leadership the Three Celebrations would disintegrate, that Tan would be unwilling to make the sacrifices necessary to keep the troupe together through times of economic stress. Tan did indeed become the greatest star of his generation, leading the drama world in a full-speed charge down the path of individual stardom and commercialization.

That Peking opera was dominated in the decades of its consolidation (1850–1910) by performers of laosheng roles is remarkable; throughout the preceding century the stage had been ruled by male dan, cross-dressing male performers of female roles. Men performed the roles of female characters in all the styles that contributed to the mélange of Peking opera, from the refined (ya) style of kunqu, to the more folksy or crude (su) styles of xipi, erhuang, and bangzi.⁵ Male dan typically had brief careers, during which the most successful were admired for their elegance, youthful beauty, and charm; these graceful youths served as the muses for numerous compilations of laudatory criticism and misty poetry by aspiring literati. Indeed, most of the historical sources on eighteenth- and nineteenth-century Beijing drama are dedicated to young dan actors. Yet it was on the backs of the bearded laosheng that Peking opera rose from its