UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The birth of the Chinese population : a study in the history of governmental logics Thompson, Malcolm 2013

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


24-ubc_2013_fall_thompson_malcolm.pdf [ 7.61MB ]
JSON: 24-1.0073963.json
JSON-LD: 24-1.0073963-ld.json
RDF/XML (Pretty): 24-1.0073963-rdf.xml
RDF/JSON: 24-1.0073963-rdf.json
Turtle: 24-1.0073963-turtle.txt
N-Triples: 24-1.0073963-rdf-ntriples.txt
Original Record: 24-1.0073963-source.json
Full Text

Full Text

THE BIRTH OF THE CHINESE POPULATION: A STUDY IN THE HISTORY OF GOVERNMENTAL LOGICS by Malcolm Thompson  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in The Faculty of Graduate Studies (History) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) June, 2013  © Malcolm Thompson, 2013  ABSTRACT  It was only in the early twentieth century that China discovered that it had a population, at least if a population is understood not as a number of people but instead in terms of such features as relative levels of health, birth and death rates, sex ratios, and so on—that is, as an object with a specific rationality that can be managed and improved. In 1900, such a conception of the population did not exist in China; by the 1930s, it was utterly pervasive. How did this transformation take place? This dissertation argues that it occurred at the level of techniques of governing and systems of knowledge production, and explains it from the perspective of changes in the institutional and epistemological forms by which interventions into other people's activities are organized. The installation of populationist practices into China is tracked in four sites: 1.  The problem of “race efficiency”—formalized in this period as the cost in “race energy” of producing a given increment to a population—and analyses of the effects of different kinds of social organization on the production of life.  2.  The institutional division of population registration into censuses (“statics”) and vital statistics (“dynamics”)—in a word, the formation of a statistical system based on mechanics.  3.  Public health, whose object of care is not patients but the collective life of the population and its conditions of existence.  4.  The problem of the China's “rural surplus labour-power” in relation to the formation of a national economy.  This dissertation shows how the privileged position of the population in political and economic reflection in Republican China carved out a field of governability by which it was possible to enchain a variety of previously disconnected fields of activity into a single logic, the axiom of which was the capitalist accumulation of life.  ii  TABLE OF CONTENTS  ABSTRACT...................................................................................................................................ii TABLE OF CONTENTS..............................................................................................................iii LIST OF TABLES..........................................................................................................................v LIST OF FIGURES.......................................................................................................................vi ACKNOWLEDGEMENTS.........................................................................................................vii CHAPTER ONE—INTRODUCTION...........................................................................................1 Which “Population Problem”?..............................................................................................1 The Population–Family–Economy Nexus.............................................................................6 Toward a History of Governmental Logics.........................................................................16 Fields of Governability........................................................................................................20 Capital and Capitalist Governmentality..............................................................................28 Chinese Nationalism and Its Apparatuses...........................................................................36 “Agency”.............................................................................................................................40 Outline of the Dissertation...................................................................................................46 CHAPTER TWO—“RACE EFFICIENCY”................................................................................51 Introduction.........................................................................................................................51 The General Problem of Race and Efficiency.....................................................................53 The Racial Efficiency of the Chinese..................................................................................59 Ubiquitous Life....................................................................................................................68 Governing Indirectly...........................................................................................................76 The Nation as Entrepreneur of Itself...................................................................................83 The Logical Structure of Biopolitical Governing................................................................91 CHAPTER THREE—“THE BOOKKEEPING OF CHINESE...................................................96 HUMANITY”: LIVING CAPITAL AND VITAL STATISTICS Late Imperial Population Registration.................................................................................96 The Creation of Vital Statistical Capacity.........................................................................104 The Form of Vital Statistical Knowledge..........................................................................118 Living Capital....................................................................................................................128 V = $, or: Populationism and Capitalism..........................................................................137 CHAPTER FOUR: THE VIRTUAL OBJECT OF PUBLIC......................................................143 HEALTH, OR: THE PROBLEM OF LIFE Public Health as an Apparatus...........................................................................................143 iii  The Manchurian Plague.....................................................................................................153 Research on Life and Death..............................................................................................159 The Apparatus of Environmental Management.................................................................163 The Nature–Population–Society Nexus............................................................................183 CHAPTER FIVE—“THE ECONOMIC VALUE OF CHINA'S................................................187 RURAL POPULATION” The Immanence of Health and Wealth..............................................................................187 Quantitative Economics, Equilibria, and Indifference......................................................192 Rural Economics and the Man-Work Unit........................................................................205 The Buck Survey and China's Rural..................................................................................210 Surplus Labour-Power Rural Statistics...................................................................................................................224 The Chinese Agrarfrage....................................................................................................232 “Harmonizing Labour and Capital”...................................................................................245 Conclusion.........................................................................................................................247 CHAPTER SIX—CONCLUSION.............................................................................................250 The Population as the Subspace of Governing..................................................................250 The Rebirth of the Chinese Population.............................................................................254 BIBLIOGRAPHY.......................................................................................................................271  iv  LIST OF TABLES  5.1  Basic farm and population data, Yangtze Rice-Wheat Area.............................................214  5.2  Total Man-Labour Requirements, Yangtze Rice-Wheat Area...........................................214  5.3  Ratio of population to ME by farm size, Yangtze.............................................................216 Rice-Wheat Area  5.4  “Adult Male Unit” conversion ratios................................................................................219  5.5  “Standard” vs. Chinese age distributions..........................................................................230  v  LIST OF FIGURES  3.1  Location of vital surveys conducted as part of J.L. Buck's...............................................111 land utilization survey (1929–1933)  3.2  Organizational chart of the Nanjing Municipal Vital Statistics.........................................113 Coordinated Office  3.3  Group photo of the staff of the Nanjing Municipal Vital..................................................114 Statistics Coordinated Office  3.4  Nanjing Municipality, deaths by week, July 1934–June 1935..........................................125  3.5  Dot map, “Distribution of Plague, 1910”..........................................................................128  4.1  The microscopic scale of plague, the human scale, and the..............................................156 scale of the population.  4.2  Schema of an organic system............................................................................................166  4.3  Methods of air circulation; triangular vs. radial street organization;................................180 instrument for measuring ground temperature  5.1  Agricultural areas of China...............................................................................................213  5.2  Areas for which population returns were submitted by 1931 under.................................226 the 1928 census  5.3  Comparative chart of land-area and population by political unit......................................226  5.4  Age- and sex-distribution pyramid....................................................................................229  vi  ACKNOWLEDGEMENTS  Principal thanks are clearly due to my Supervisor and committee members: Timothy Brook, who took me on as a student even though I was coming to the field from a different discipline, and who has gently but insistently steered me toward being something like an historian; Robert Brain, who has always seen immediately the logic behind the odd historical connections I sometimes make; and Glen Peterson, whose course on the social history of modern China inspired me to make the move to this discipline in the first place. I was very lucky to enter the Ph.D. program at the University of British Columbia at a particularly fortuitous moment, in 2005, when the Chinese history program came to be populated by an especially delightful and energetic group of people. My fellow students at UBC have always provided me with useful feedback. I'm thinking especially of Desmond Cheung, Noa Grass, David Luesink, and Tim Sedo; I would be remiss not to acknowledge as well Frederick Vermote, Heidi Kong, Nick Simon, and Craig Smith. David Luesink and Amber Wang deserve special thanks for making my family's adjustment to living in China much easier than it would have been otherwise. A number of faculty members at UBC have helped me out at important moments, both with feedback on my work and with the all-important factor of funding. In addition to my committee members, whom I've already mentioned, I'd like to thank Alison Bailey, Alejandra Bronfman, Timothy Cheek, Bill French, Janet Giltrow, Thomas Kemple, Carla Nappi, John Roosa, and Danny Vickers. I have been lucky to have had many opportunities to present this work in a number of contexts. Thanks are due especially to TJ Hinrichs, Rebecca Karl, Michelle Murphy, and Leon Rocha for inviting me to speak at their universities. My interlocutors at these presentations and others have been extremely useful in helping me adjust and hone the arguments made below,  vii  especially Toby Lincoln, Akiko Ishii, Daniel Asen, Mary Brazelton, Ken C. Kawashima, Ritu Birla, Chen Li, and Wu Yiching. Of course, responsibility for all errors herein resides solely with me. As far as my research time in China is concerned, Xu Jilin was instrumental in securing an affiliation with East China Normal University in Shanghai, all-important for Visa purposes. For facilitating the collection of research materials, I'd like to thank the staff at the Shanghai Municipal Library and the Shanghai Municipal Archives, both entirely pleasant places to conduct research. Much closer to home, the staff of the UBC Asian Library (especially the Chinese Librarian, Jing Liu) has been extremely important to this project, whether in terms of finding obscure materials buried deep in the caverns of UBC's holdings or, indeed, acquiring new materials on my behalf. Thanks are due, as well, to the staff of the Interlibrary Loans Office of UBC Library, especially, but not only, for getting me a complete run of Dongfang zazhi 東方 雜志 (Eastern Miscellany) from 1904 to 1948 on microfiche. This project has been financially supported by the Social Sciences and Humanities Research Council (SSHRC)'s Canada Graduate Scholarship–Doctoral, the UBC History Department's Summer Fellowships, and the UBC Faculty of Arts Graduate Award. This dissertation was completed while employed in the Department of History at the University of Toronto. Many people there were most welcoming, especially Ken Mills, Carol Chin, and Nhung Tran. Finally, and most important by far: Megan, Adrian, and Lauren—sine quo non.  viii  CHAPTER ONE INTRODUCTION  WHICH “POPULATION PROBLEM”?  Something like the problem, or question, of population (人口問題 renkou wenti) made its first appearance in China in the very early years of the twentieth century as a problem of numbers: How many people does China have? A sustained discussion of the size of the Chinese population and how to count it—the discussion which most immediately preceded the period I will be concerned with in this dissertation—can be dated to the period 1903–1912. In 1903, Liang Qichao 粱啟超 (1873–1929), in exile in Japan and probably the most famous voice of political reform in China at the time, published an analysis of historical population statistics in China in Xinmin congbao 新民叢報, one of the several major periodicals he organized during his long career.1 In 1904, the American statistician W.W. Rockhill (1854–1914) published An Inquiry into the Population of China,2 which was translated into Chinese in 1911 with a 1 Liang, “Zhongguo shishang renkou zhi tongji 中國史上人口之統計 (Statistics on China's Historical Population),” 1903. Prior to this essay, the only references to population that I have found are tables of figures, devoid of discussion. In 1897, Liang Qichao's Shiwubao 時務報 published a translation of Tojō Teikichi 古城貞吉's estimate of China's population: “Zhongguo renkou 中國人口 (The Population of China),” 1897. It was presented without interpretation. From 1900 until about 1907, the emerging periodical press regularly published, without attribution, tables or series of population data, but with little or no explanation of what they meant or why they were important; see, for instance, Anon., “Zhina renkou biao 支那人口表 (A Table of the Population of China)” in Qingyi bao 清議報 (June 26, 1901), another Liang Qichao organ published from Tokyo; Anon., “Zhongguo renkou biao 中國人口表 (A Table of the Population of China),” in Xuanbao 選報 (Selected Reports, November 11, 1901); Anon., “Ge sheng renkou xianzai shu 各省人口現在數 (A Count of the Present Population of Each Province),” in Zhejiang chao 浙江潮 (Zhejiang Tide, February 17, 1903); and Anon., “Zhongguo renkou zuijin zhi tongji 中國人口最近之統計 (The Most Recent Statistics on the Population of China),” in Xinmin congbao, March 12, 1903. For an analysis of the conditions for the emergence of many of these periodicals, see Zhang Qing 章清's essay, “Wan Qing 'sixiangjie' de xingcheng yu zhishifenzi de 'gonggong kongjian' 晚清 '思想界' 的形成與知識分子的 '公共空間' (The Formation of the Late Qing 'Intellectual World' and Intellectuals' Public Space),” 2007. 2 Rockhill, 1905 (1904). 1  commentary by an unknown author who styled himself Ming Shui 明水.3 The latter also translated and commented upon the work of the Japanese scholar Negishi Tadashi 根岸佶 (1874–1971) on the same problem.4 In 1912, after the results of the 1910 census of households undertaken by Qing dynasty's Minzhengbu 民政部 (Ministry of Civil Affairs) became known, Rockhill published another study, “The 1910 Census of the Population of China.”5 Chen Yujing 陳裕菁's 1912 essay, “Zhongguo hukou wenti 中國戶口問題 (The Problem of China's Administrative Population,”6 begins with Rockhill's analysis and is, as far as I can tell, the first attempt to arrange a variety of estimates of China's historical and contemporary population, from both Western observers and Qing statistical sources, into some kind of coherent series.7 The primary purpose of the 1910 census of households—which would have been followed in 1912 by a census of individuals, but for the collapse of the Qing dynasty and its replacement by the Republic—was to establish a demographic basis for the size of the Provincial Assemblies (諮議局 Ziyiju) created under the New Policies (新政 Xinzheng) reforms initiated in 1905, along the lines of the American model of a census. A larger provincial population would result in a larger Provincial Assembly, which would occupy a proportional number of seats in the National Assembly (資政院 Zizhengyuan), first convened in 1910 (in advance of the results of the  3 Ming, “Zhongguo renkou wenti 中國人口問題 (The Question of the Chinese Population),” 1911a. 4 Ming, “Zhongguo renkou wenti, xu diwu hao 中國人口問題, 續第五號 (The Question of the Chinese Population, cont’d.),” 1911b. 5 Rockhill, 1912. 6 The term used for “population” here, 戶口 hukou (“households and mouths”), is a specifically administrative category of very long standing in imperial Chinese state practice. Its proper domain is the state. I will return to this category in Chapter Three. The other, new, term for population, 人口 renkou, is, as we will see, a more properly natural-scientific category, though it will not—indeed, cannot—exist apart from a practice of governing in which the state is centrally involved. 7 Chen, 1912. Chen's sources start with Hubu 戶部 (Ministry of Revenue/Households) returns for 1743, run through the estimates given by such Western observers as the French Jesuit Jean Joseph Marie Amiot (1718–1793) and the British consular explorer Alexander Hosie (1853–1925) and the returns from the 1910 Census. 2  census).8 In another register, a kind of proto-Malthusian notion of overpopulation in relation to resources as a cause of social disorder first appeared in print in China at basically the same time. Two anonymous articles appeared in the flagship journal of the constitutional monarchist movement, Dongfang zazhi 東方雜志 (Eastern Miscellany):9 “On China's Condition of Order or Disorder Arising from the Size of Its Population” in 1904,10 and “On the Connection between China's Condition of Internal Order or Disorder and Its Population” in 1907.11 Similar articles followed in other venues.12 From these rather humble beginnings, however, the problem of population was very rapidly distributed into a whole series of other problems and forms of knowledge, some old and some new, and maintained in this distribution only superficial and tenuous connections to its original, more or less enumerative, appearance. Very quickly, that is, we find the problem of population being inscribed into the field of social life and practices in a number of ways, at a variety of levels. In 1900, the idea that interventions at the level of a population could be used to induce changes or adjustments at the level of general social organization was completely absent from China. For all the provenance, density, and sophistication of the tradition of governmental thought in China, an immanent relationship between a given state of a population and a set of possibilities of development had never been specified. No major transformation in this regard is 8 The basic official documents pertaining to this census were prepared and published by the Minzhengbu in 1910 as the “Memorial on the Regulations Governing the Census of Households” (調 查戶口章程摺 diaocha hukou zhangcheng zhe). For an extended discussion of the planning for and process of this census, see Hou Yangfang 侯楊方, Zhongguo renkou shi 中國人口史, 第六卷: 1910– 1953 年 (The Population History of China, Vol. 6: 1910–1953), 2000: 20–31. 9 For a useful précis of the political parties and purposes behind Dongfang zazhi, see Judge, Print and Politics: 'Shibao' and the Culture of Reform in Late Qing China, 1996: 29. 10 Anon., “Lun Zhongguo zhiluan youyu renkou zhi zhonggua 論中國治亂由於人口之眾寡,” 1904. 11 Anon., “Lun Zhongguo zhiluan yu renkou zhi guanxi 論中國治亂與人口之關係,” 1907. 12 For instance, Anon., “Lun hukou guoshu zhi huan 論戶口過庶之患 (On the Harm of Overpopulation),” 1908. 3  entailed by the fact of a discussion of the size of China's population, nor even by significant plans to reform a census machinery. By the 1930s, though, the notion that the state of the population was inextricably connected to everything else was absolutely pervasive. Every significant governmental project launched in Nationalist China from that date onward already presupposed the existence of this web of connections.13 Within a couple of decades, it had come to constitute no less than the fundamental matrix of governmental activity. As I will show in the following chapters, it was effectively no longer possible not to govern with reference to it. The rapidity of this transformation did not occur as an expansion of the kinds of discussions summarized above; it was the effect of a series of displacements and a proliferation of the sites in which the population's “effects” were detected. What actually constitutes the problem of population in its real historical existence is the formation and composition of this distribution of knowledges and practices of governing. Indeed, the problem of population proper, if I may call it that, really emerges only when discussion moves beyond the question of numbers of people and begins to concern the determination and management of some new thing, some x, which is given the name “population.” It is the task of this dissertation to describe the formation in China of the governmental ensemble to which a problem of population correlates. We will not be able to give any one definition of this x, since it appears as different things in different contexts, composed of different elements and related to different variables. Nor, then, would we able to define a population and then proceed on the basis of this definition to determine the problem's “proper” dimensions, to differentiate a “real” from a merely putative or “so-called” population problem, real and effective interventions from misguided or “pseudoscientific” ones. Such periodic adjustments are certainly important as an impetus for change in 13 This claim relies on a distinction between governing and violence. Certainly the brutal suppression of the Communists was the other major Nationalist project in the late 1920s and 1930s, but those campaigns were not really “governmental,” stricto sensu. For a discussion of how a properly governmental apparatus involves “processes of subjectivation” rather than of simple destruction, see Agamben, What is an Apparatus? and Other Essays, 2009: 11–12. 4  the internal history of the various knowledges involved, but they are not the present problem. Instead of an analytic approach, which would begin with a correct determination of the population, I will take a descriptive approach to the problem itself: When people speak of dealing with problems of population, and take concrete measures to intervene in it, what is it they are speaking of? What are they actually doing? Only sometimes, it turns out, does the matter turn on the question of a number of individuals. Even then, this quantity is usually a form of appearance, a momentary distribution, of something else. The problem is not ultimately one of numbers: its various aspects, corollaries, and tributaries do not converge into, and are not reducible to, the problem of a number of units in relation to questions of subsistence or relative geopolitical strength. It is a question rather of questions like the following: What are the effects on a population of its conditions of existence? What historical, cultural, biological, environmental, necessary, or accidental factors have given the Chinese population its present form, extent, and characteristics, and placed it in the specific relationship it maintains with the economy and the possibility of development? Where can one find effective points of intervention, in order to steer the population toward development? How and where can governmental efforts be most efficiently and judiciously applied, not to the population as from outside of it, but within the field of processes which the population reveals and determines? How, finally, can the various machineries of governing be calibrated to the population, and to each other by reference to it, to maximum effect? None of these questions had any meaning whatever in the context of late imperial Chinese governing, and nor could they have, for reasons I will explore in Chapter Three. But they are the essential questions in the Republican period. It is less a matter of increasing or decreasing numbers than of the effective use of the life of the population, and the implantation of regulatory mechanisms at each point which bears upon the creation and distribution of that life.  5  I will not, then, accept the “demographic postulate” as a simple fact, but rather, suspend belief in populations, so that I may examine what different apparatuses of knowledge tell us about them, without reducing the investigation to the question of how well or how poorly they capture an object that we somehow—but how?—already know. The goal here is not to evaluate late imperial population registration and, anticipating the argument somewhat, let us call it the capitalist populationist regime, according to their greater or lesser approximation to an object the nature and structure of which we know well enough to use as a yardstick of evaluation. Rather, it is to examine the historical development of the technologies of knowledge through which this object is given in the first place. Thus, I am not addressing an historical-demographic problem, but the problem of the epistemological transformation which produced the very notion of population in terms of which alone the techniques of historical demography make sense. How, indeed, could we gain this object, with reference to which we evaluate this or that technology of knowledge, if not through the operation of some definite technology of knowledge? To be sure, where the notion of a population appears, it appears to apply universally, but this is different than its actual historical existence. Rather than be concerned with how a given knowledge—say, demography—attains the possibility of accuracy vis-à-vis a phenomena, or the obstacles that it faces or that it overcomes, or the reasons for its failure to achieve accuracy, the object here is the fact that, at a very specific historical moment, a great variety of people in China suddenly realized that they were faced with a problem they never knew existed, and devoted themselves to solving it. This, rather than whether any particular notion of population is correct or not, is the present problem. THE POPULATION–FAMILY–ECONOMY NEXUS But where is the break? Where do we find the passageway between one form of governing and  6  another? It can be found, I think, in a particular phenomenon, one that, moreover, was hardly unique to China, which is that, for a whole class of people in the second decade of the twentieth century, China suddenly appeared to have been stricken with a virtually inexhaustible series of “social problems” (社會問題 shehui wenti): the problems of women (婦女問題 funü wenti), of poverty and unemployment (貧窮 pinqiong and 失業問題 shiye wenti), of crime (犯罪問題 fanzui wenti), of peasants and rural society (農民 nongmin or 農村問題 nongcun wenti), of sovereignty (權力問題 quanli wenti), of minorities (小數民族問題 xiaoshu minzu wenti)—the list goes on and on.14 Obviously this was not the first time that China had poor people, or peasants, or women and families. Neither, clearly, did all of these disparate things suddenly really become problems just at this moment, where before everything had been fine. But it does 14 A rough sense of this can be gleaned by noting the dates of publication of the first book-length studies  of some of these problems: He Xiya 何西亞, A Study of the Problem of Banditry in China (中國盜匪 問題之硏究 Zhongguo daofei wenti zhi yanjiu), 1915; Du Junhui 杜君慧, Lectures on the Woman Question (婦女問題講話 Funü wenti jianghua), 1919; Piaoping jiren 飄萍吉人 (pseud.), The Problem of the Unemployed (失業者問題 Shiyezhe wenti), 1920; Feng Fei 馮飛, Introduction to the Labour Problem (勞動問題概論 Laodong wenti gailun), 1920; and Yi Junzuo 易君左, The Problem of Women's Employment (婦女職業問題 Funü zhiye wenti), 1922. Such a field emerged somewhat earlier in Japan—roughly, between the late 1890s and 1910, when the first discussions of this series of problems, under the general rubric of “the social problem,” took place—through the efforts of social scientists who were powerfully influenced by the sociological innovations of the German Historical School. See, for instance, Kanai Noburu 金井延, The Social Problem (社会問題 Shakai mondai), 1894; Tomeoka Kosuke 留岡幸助, The Charity Problem (慈善問題 Jizen mondai), 1898; the prolific Christian-Democrat Abe Isoo 安部磯雄, An Approach to the Social Problem (社会問題解釋法 Shakai mondai kaishakuho), 1901; the Christian educator Hani Mokoto 羽仁もと子, The Family Problem (家庭問題 Katei mondai), 1907; and Kawada Shiro 河田嗣郎, The Woman Question (婦人 問題 Fujin mondai), 1910. The role of Japan as an intermediary in the importation of “Western” knowledge—this concept, I hope, will make less and less sense over the course of the argument—has been well studied; see, for instance, Douglas Reynolds, China, 1898–1912: The Xinzheng Revolution and Japan, 1993; Lydia M. Liu, Translingual Practice: Literature, Culture, and Translated Modernity —China, 1900–1937, 1995, and idem., ed., Tokens of Exchange: The Problem of Translation in Global Circulations, 1999; Michael Lackner, Iwo Amelung, and Joachim Kurtz, New Terms for New Ideas: Western Knowledge and Lexical Change in Late Imperial China, 2001; and Joshua Fogel, The Role of Japan in Liang Qichao's Introduction of Modern Western Civilization to China, 2004. But it appears that many of the major early figures in the Chinese discourse of social problems drew their inspiration more immediately from American sociology and political economy of the 1900s and 1910s. The latter, in their turn, had been equally powerfully influenced by the German Historical School. I will return in more detail to these and other chains of transmissions often in the following chapters. 7  represent the almost simultaneous enchaining of all of these things into a common form, the emergence of a space in which they could all be addressed in the same terms, and taken as manifestations of a single underlying problem: that of the social transition to capitalism. The break, then, is to be found in a veritable discursive explosion of social problems. Each element in this series was immediately related to the others, and one could approach any of them through any of the others. Aspects of the woman problem could be linked to the problem of unemployment, or aspects of the crime problem could be solved by referring them to the peasant problem, and so on. In this respect, they were all formally equal. But there were two which were “universally present elements,” elements which were both necessarily included in discussions of any of the others and which together formed a kind of metaset. These were the population and the family. The mutual inscription of the one into the other—their having to do with each other—was not simply one new problem among others. It was the conjunction that, in many respects, by stitching together a new field of governability, made possible the rapid and profound transformation of the basic logic of governing in China. It was in exactly this brief period that the first book-length treatments of the population and the family problems were published in China: Zhongguo renkou lun 中國人口論 (On the Population of China) by Chen Changheng 陳長蘅 (1891–1987)—about whom we will hear a great deal more in the chapters that follow—in 1918, and Jiating wenti 家庭問題 (The Family Problem), a series of translations made and compiled by Yi Jiayue 易家鉞 (1899–1972) in 1920. By reading them together—I intend this reading to be more indicative than substantive—we are able to extract a general picture, an image, of the field of governability in terms of which alone does it make sense to speak of a Chinese population. When On the Population of China was first published, Chen Changheng had recently returned from the United States, where in 1917 he attained an MA in Economics from Harvard 8  University, and was employed as a lecturer in the Economics Department of Beijing University. Chen was the first person to bring to China some of the conceptual tools by which discussion in China on this topic could move toward considering how a country's “vital resources” (生命資料 shengming ziliao) could be fitted into a general national-developmental project.15 The book contains a lengthy discussion of the connections between the population problem and other social problems, but most important for our purposes, between the population problem and types of marriage and family organization. Here, Chen draws on the work of the Columbia University sociologist and economist, Franklin H. Giddings (1855–1931), who determined the principle of the historical evolution of the family on the basis of its more or less efficient production and use of its collective vitality.16 The earliest stage is the “religious-proprietary family” (同教共產的室 家 tongjiao gongchan de shijia),17 in which stifling and “arbitrary” (獨斷 duduan) parental control, patriarchy, and an overriding concern for social position decrease the stock of vitality in a family and channel it into useless, non-productive activities. With the emergence of liberalism  15 Other people brought more of these tools later, of course, and more systematically, but Chen does  have the distinction of being the first. In 1914, Zhu Jin 朱進, at the time an economics student at Columbia University, published a pair of essays in the journal of overseas students studying in the US: “Fuguo ce 富國策 (Political Economy)” and “Fuqiang siyi 富強私議 (Private Reflections on Wealth and Power),” 1914; his dissertation from Columbia was published in 1916 as The Tariff Problem in China; see Chu, 1916) which contain some of the same ideas, so technically he would be first, but Chen's was the first book-length treatment, and the first that anybody paid attention to. The book's preface is by Cai Yuanpei 蔡元培, founding Chancellor of Beijing University. At Harvard, Chen studied with Thomas Nixon Carver (1865–1961), about whom we will have a great deal more to say. He went on to a long career as a politically-involved intellectual and member of the Nationalist government. He was a founding member of both the Chinese Economics Association (中國經濟學社 Zhongguo jingjixue she) in 1923 and the Chinese Statistics Association (中國統計學會 Zhongguo tongjixue hui) in 1930, a professor at a wide variety of universities, and a member of the National Government's Legislative Yuan (立法院 Lifayuan) from 1928, in all of which capacities he played an important role in training and distributing China's corps of technocratic functionaries. Throughout the 1920s and 30s, he was regularly involved in scholarly debates on the population and related problems (which included almost everything). 16 Chen's source text is clearly Giddings' The Principles of Sociology, 1896; see especially the chapters on “The Patronymic Tribe and Folk,” 285–298, and “The Liberal–Legal Civilization,” 324–360. 17 The 室家 shijia here specifies the cohabitating family, the household-as-family, something like the nuclear family as distinguished from the lineage or the extended family. 9  (自由主義 ziyou zhuyi), society moves to the stage of the “romantic family” (愛情室家 aiqing shijia). This smaller family form is very good at producing vitality, but due to its detachment from extended networks and uprootedness from social structures, its members tend to be rash and faithless (澆薄 jiaobo), easily distracted and prone to anxiety (瞿然 juran), and generally dissolute (放蕩 fangdang). In short, the Romantic Family is an effective fount of vitality, but lacks any system of control. Such a system emerges with the final stage, the “ethical family” (倫 理室家 lunli shijia), which combines the production of vitality with its rational management, and which forms the “embryo” (胚胎 peitai) of truly civilized life.18 Predictably enough, this diachronic trajectory is overlaid with a synchronic class analysis: the large-family system, prevalent among both the traditional, “feudal” elites and the poor, proletarian, and peasant classes, produces too many children, and their children are all lethargic and stupid due to the oppressions they suffer, while the small-family system, that of the modernizing middle classes or bourgeoisie, produces energetic and intelligent children, but too few of them, and they are in danger of wasting their vitality in an excess of freedom. (Family reform, then, was not exactly a project of moralizing the poor, though it was that as well; it began as an effort at self-reform and autocritique by the newly self-identified middle classes.)19 Thus, the principle of the historical evolution of the family from lower to higher forms is to be found in its increasingly rational or economic self-organization, to the end of a maximally efficacious production of vitality. Yi Jiayue's The Family Problem, turning now to this, is a set of translations from American 18 Chen, Zhongguo renkou lun, 1918: 144–146. The same connection between forms of family organization and efficiencies of vital production is made at almost the same time by Zhou Zhangxian 周長憲 in the journal Jiating yanjiu (家庭研究 Family Research), founded by Yi Jiayue, in the article “Jiating zhi zuzhi ji shengming 家庭之組織及生命 (The Organization of the Family and Life),” 1922. 19 This is the aspect of this discourse that Susan Glosser focuses on in Chinese Visions of Family and State, 1910–1953, 2003. 10  and French writers on the family.20 About two thirds of the book is taken from Sociology and Modern Social Problems, first published in 1910 by another US sociologist, Charles A. Ellwood (1873–1946). It is well known that there was a good deal of debate in the late nineteenth and early twentieth centuries on the question of the family's origins, and that this formed a part of a much broader general discussion of the family in this period—on its functions, its history, its general forms and varieties, and so on—which doubtless had to do with exactly its repositioning within the general system of social government.21 According to one school, the temporal origin of the family was situated within the timespan of human history: human beings did not always live in families, and came to do so once the “obvious advantages” of that institution (and the equally obvious disadvantages of the state of “primitive promiscuity” in which they had theretofore lived) became clear. Properly speaking, then, according to this view, the family is a human invention. For Ellwood, by contrast, the origin of the family—since it is to be found in 20 Yi was a student at Beida while he was preparing these translations, during which time he also cofounded the Family Research Association (家庭研究社 Jiating yanjiu she) and its shortlived but influential journal, Family Research (家庭研究 Jiating yanjiu; see n. 18, above).Yi had a less highlevel but no less interesting career. He and his colleague Luo Dunwei 羅敦偉 wrote their own treatise on the family in 1921. (See Glosser, “'The Truths I Have Learned': Nationalism, Family Reform, and Male Identity in China’s New Culture Movement, 1915–1923,” 2002, for an extended discussion of the early career and work of Yi Jiayue, approached from a somewhat different perspective than that taken here.) Yi joined the Guomindang in 1924 and after the Northern Expedition was employed in 1928 in the Propaganda Office of the Hunan Provincial Bandit Suppression Office (which really means that he was professionally engaged in justifying the massacres of Communists that were occurring just in this period). Several scholars have tried to recuperate him as a male feminist—there is a case to be made—but his critique of patriarchy is made under the sign of such strong right-wing eugenic principles that I am skeptical. (On the other hand, Margaret Sanger was the same way, at least from the mid-1920s.) In the following, I will cite from Ellwood, except where there are significant divergences in Yi's translation. 21 The corpus of the general discussion is so broad that any attempt to cite its key texts is bound to be arbitrary. Jacques Donzelot's The Policing of Families, 1979, remains, to my mind, the best theoretically informed treatment of “the family problem” in this period. Also excellent, though rather less well known in the English speaking world, is Donzelot's L'invention du social, 1994. It is easier to cite the basic texts concerning the family's origins. See especially J.J. Bachofen, Das Mutterrecht, 1861; J.F. MacLennan's Primitive Marriage, 1865; Lewis H. Morgan, Systems of Consanguinity and Affinity of the Human Family, 1868, and idem., Ancient Society, 1877; Friedrich Engels, The Origin of the Family, Private Property, and the State, 1884; E. Westermarck, The History of Human Marriage, 1891; and J. Lubbock, The Origin of Civilization, 1902. Freud's Totem and Taboo (1913) can, in many respects, be seen as a contribution to this debate. All of these texts were discussed and sometimes translated in China in the 1920s. 11  mammals and especially primates22—lies outside the span of human history proper.23 The human family owes its existence, in biological terms, to the (for simplicity, Spencerian) principle of greater and greater economization of energy over time, and to the need which humans have of caring for their young in order not to incur “the expenditure of such enormous amounts of energy in mere physiological reproduction.”24 What then is the family? It is the site of a kind of primodial economization that is constitutive of humanity as such: It is the effect of an economization that is broader and deeper than humanity itself, and the site of an ongoing economization whose refinement defines progress (and thus humanity as an agent and bearer of progress). As such, the family “is almost as much a biological structure as it is a social structure,” and no other social institution occupies exactly this position.25 In his translation, Yi makes the relationship rather more equitable: “The family is, therefore, both a biological structure and a social structure.”26 Whence derives the family's “primary function,” the basic, underlying imperative to which all of its myriad social functions are subordinated: “continuing the life of the species,” or the reproduction of the population.27 If, as I suggested above, the general position of the family within strategies of government was changing in just this period, a schematization such as this provides a way to understand or manage the family in the context of the reorganization of the productive activities of society that attend, and in many ways comprise, the “rise of modern industry”: 22 Some writers say birds live in families, others are skeptical. 23 Ellwood, Sociology and Modern Social Problems, 1910: 83–84. 24 Ibid., 87. 25 Ibid., 76. 26 Yi Jiayue, Jiating wenti, 1920: 3. It bears noting that this formulation is absolutely generic in discussions of the population; see, for instance, Chen Da 陳達, Renkou wenti 人口問題 (The Problem of Population), 1935; Hu Jianmin 胡鑑民, “Renkou bianqian yu shehui bianqian 人口變遷 與社會變遷 (Population Change and Social Change)”, 1932; Pan Guangdan 潘光旦, Renwen shengwuxue luncong 人文生物學論叢 (Collected Discussions of Sociobiology), 1928a, and idem., Zhongguo zhi jiating wenti 中國之家庭問題 (The Problem of the Family in China), 1928b. 27 Ellwood, Sociology and Modern Social Problems, 1910: 76. 12  [I]t is evident that the family has a very important relation to the industrial activities of society, and industry a very important bearing on the family. Primitively all industry centered on the family, [and] modern industry … is but an enormous expansion of primitive housekeeping … The very word economics means the science or the art of the household.28 All that need be said about the delegation of the industrial activities of the family to other industrial institutions is that the movement is not one which need cause any anxiety so long as it does not interfere with the essential function of the family, namely, the birth and rearing of children … And the removal of industries from the home, even such essential industries as the preparation of food, is to be regarded as a boon if it gives more time to the parents, especially to the mother, for the proper care and bringing up of their children.29 Certainly such a perspective, in which the small, nuclear, consuming family is positioned as the pinnacle of social evolution, is subjectively self-serving for the modernizing reformers and social scientists who utilized it. But more importantly (since the fact the some group of people fancies itself better than everybody else does not yet consitute a major problem of social power), it forms a grid of analysis by which to evaluate and find opportunities for reform in any family whatsoever, including the bourgeois family. Once (the) economy is installed behind the family as such, it is no longer a question of making poor, proletarian, peasant, and feudal families more like middle-class families, except incidentally, except insofar as these latter are further along a trajectory on which all families are placed. The family becomes a properly economic problem, both in terms of its internal economy and in terms of its connections to the general social processes of production, exchange, and consumption: Where could economies be found? How could “subsidiary” activities—socialization, education, and food production—be socially redistributed in order to support the family in its really essential role?30 28 Ibid., 79–80. The word “economics” is included in English in Yi's translation (1920: 8). The Chinese term for economy, 經濟 jingji, carries no such etymology. 29 Ellwood, Sociology and Modern Social Problems, 1910: 79–80. 30 Leon Rocha at Cambridge University is currently studying those utopian reformers in China, such as Zhang Jingsheng 張競生, who took such a logic to its conclusion, calling for the complete dissolution of the family, given that, if things were arranged correctly, every function of the family could be accomplished more efficiently if it were disassembled (the disassembly of the family into a series of analytically separable functions being the key fact here). See, in this connection, Zhang Weici's “Emancipating Women by Reorganizing the Family,” translated in Lan and Fong, eds., Women in Republican China: A Sourcebook, 1999. In Yi Jiayue's later works, he tends more and more to this position. 13  We can see that the privileged position of the population in the series of social problems does not derive from the fact that China's population was so enormous, so “obviously” a problem, but from its isolation as a zero level of social problems as such, what all forms of social organization were organizations of. The family's privilege derives from its position as the hinge between the biological and the social, the social “in the first degree.” Further, it was impossible to consider one without the other: the family was the immediately given form of the content given by the population. Since it was the primary site of the emergence, cultivation, and disappearance of life (at least in present conditions), it was clear to everybody that if they wanted to act on the population, they had to go through the family. We can see from these texts, then, that from the very start, in China, the population problem is immediately the family problem, and vice versa. To summarize, we have two central phenomena here, neither of which have any correlate in Chinese thought on the family prior to the twentieth century: First, the diachronic and synchronic schematization of family forms according to their differential capacities to produce, cultivate, and distribute vitality; and second, the linking of this “internal economy” of the household immediately to a social division of labour, a general economy. The family and the population are put into relation with economy in two ways simultaneously: first, economy in the “pure” sense of the rational allocation of the factors of production of an enterprise with a view to the maximization of abstract utility,31 and second, “the” economy in the classical or politicaleconomic sense of the general social processes of production—and, of course, the drive that cathects them all under a capitalist logic, accumulation. If, as I have suggested, the population is the zero level of social problems and the family is the social in the first degree, then economy is both their essential condition of existence and the principle of their articulation with each other.  31 This will be a central concept in the following chapters. 14  One need only note at this point the family's position here is completely different than it had been or could have been where the general system of governmental problems was differently organized. The family is thus repositioned by being imbued with the population. Obviously, the publication of two books—however interesting or important they may be— does not constitute a major historical transformation. But my claim is simply that these texts contain all the elements one needs to outline the form of the problem. They give in advance the outline of a formal transformation whose conversion into an actual transformation will be the subject of this dissertation. They establish the series of relationships between diverse phenomena and a way of approaching them that would eventually characterize Chinese bourgeois-nationalist governmentality in general—a series that was put in abeyance, certainly, after the defeat of the Guomindang and the formation of the People's Republic. But it did not disappear completely, or at least it reemerged in the late 1970s. Due to the peculiarly decentralized condtions obtaining in China in the first half of the twentieth century, and the simultaneous operation of multiple networks through which knowledge was circulating globally, populationism entered China at a thousand different points. It was not in any sense imposed from above; it was progressively insinuated by a multitude of actors into widely dispersed areas of social life. In the chapters that follow, exhaustiveness is neither possible nor the goal. Rather, by analyzing a few central domains of governing, I wish to show how one domain after another was cathected to a general national-developmental and biopolitical project by way of their reorganization by and in terms of a population–family–economy nexus. What I mean by “nexus” is the series of points at which acting on any one of the elements is already, automatically, acting on and inducing effects in the others.  15  TOWARD A HISTORY OF GOVERNMENTAL LOGICS How shall we explain this? It cannot be explained satisfactorily by changes at the level of society and the economy; such changes, while tremendous, and not unrelated to the change that will concern us here, cannot necessarily have produced just this understanding of the population. Nor can it be explained solely by the importation of new ideas (though, again, these undoubtedly played a role), because some of the changes took place at levels of which no clear idea existed.32 But the rapid and relatively thorough biopoliticization of a social field as broad and dense and that of China is no mean feat. It can't have taken the form of a total transformation of everything. Rather, it took place through the installation of a new “formaltranscendental matrix” within practices of governing.33 It is a change that takes place at the level of techniques of governing and the systems of knowledge production that accompany them. I will attempt to describe the birth of the Chinese population in terms of a history of governmental logics, that is, of changes in the epistemological and institutional forms by which interventions into other people's activities are organized. But what is a governmental logic? The first thing to note is that this term is intended to reveal a general level of possible historical change, of which the transformation out of which the Chinese population emerges would be a “case.” So, before getting to the case, I should describe this level in its generality. We can approach it by way of a series of questions, which have the considerable advantage, I claim, of being askable of either side of a given change over time. 1.  What are the criteria according to which a given intervention can be evaluated as successful or unsuccessful?  2.  What effects are meant to be induced in a given social field by the activity of governing?  32 We will see in Chapter Four how this is true even of the central term itself: life. 33 On the notion of a formal-transcendental matrix, see Žižek, “Against the Populist Temptation,” 2006: 567. 16  3.  What are the elements of the field of governability to which a given kind of governing refers? That is, what is on the agenda of government? What are the pertinent realities or variables, through or in terms of which one is compelled to act? Further, what specific relation is established between the activity of governing and the set of things to be governed?  4.  How are they related to one another? How is their reciprocal organization taken into account?  5.  What effects of social order or of economic reproduction are seen to be induced by good administration, and what ills are generated by corruptions or failures of practice?  6.  What is the general relationship between social order and good government?  I will suggest a series of initial answers to some of these questions, not at all in order to settle them once and for all, nor to indicate that they are easily answered, but in order to give the reader a sense of the method I employ in this dissertation. The shift from one set of answers to another defines the level at which the analysis will proceed; retracing the process by which one set of answers is replaced by another, then, constitutes the history itself. Schematically, then: For late imperial statecraft, what was sought was to maintain the conditions of normal production and to reproduce a given social order; to the degree that particular measures did this (which was always a matter of debate), they would be interpreted as successful.34 For the political-economic form of governing with whose history I will be concerned here, on the other hand, the goal is to produce the possibility of economic growth, and to direct and channel the resulting social disorder in a structural and constant manner (which is not at all the same thing as reproducing a social order). In terms of the field of governable 34 The activities of, in particular, the Qing state prior to the mid-nineteenth century in attempting to maintain the conditions for an ideal type of normal production have been well covered in, among other works, Pierre-Ètienne Will, Bureaucracy and Famine in Eighteenth-Century China, 1990; Will and R. Bin Wong, Nourish the People: The State Civilian Granary System in China, 1650–1850, 1991; Peter Perdue, Exhausting the Earth: State and Peasant in Hunan, 1500–1850, 1987; R. Bin Wong, China Transformed, 1997a: 105–126 et passim; and Lillian M. Li, Fighting Famine in North China: State, Market, and Environmental Decline, 2007. In The Making of a Hinterland (1993) and in a variety of recent works, Kenneth Pomeranz has made the difference between late imperial statecraft and political-economic governing at this level—the transition between which occurs in exactly the period covered by this dissertation—very clear, though his focus is on governing an economy, rather than a society. For a valuable summary of late imperial governmental activities designed to reproduce a social order, see R. Bin Wong, “Confucian Agendas for Material and Ideological Control in Modern China,” 1997b. 17  objects, we might take our example from the realm of law. In late imperial Chinese law, the “positive terms” (if we can call them that) were the concentric degrees of proximity or distance of relationship. There was no abstract “subject of law,” whose rights and duties derived from a universal and formal predication of “Man”; every obligation and reciprocity depended on one's position within a series of hierarchies and lateral relations, the terms of which are given by degrees of mourning obligation (kinship) and statutorily defined rank relations (such as master and servant). By contrast, the new legal codes that began to be introduced into China in the twentieth century—first under the Qing in 1908 and culminating in the Civil Code of the Republic of China (中華民國民法 Zhonghua minguo minfa) promulgated on May 23, 1929, both of which were modelled in the German Civil Code of 1900—begin with a predication of “Natural Man” (自然人 ziran ren) and proceed to derive from that predication the universe of legal facts. The foundation of one's inclusion in a structure of legal relationships is the fact of being a person, and this fundamentally restructures the field of objects in terms of which legal governing must operate.35 Within the horizon of late imperial statecraft, the family was one element in an ascending series of orderable domains, and the effects of poor governance in any one of them could ripple outward to infect the others in either direction.36 In the twentieth 35 For useful materials on the jurisprudential aspects of this change, see Jérôme Bourgon, “Shen Jiaben (1840–1913) et le driot chinois à la fin des Qing,” 1994; idem., “Uncivil Dialogue: Law and Custom did not Merge into Civil Law under the Qing,” 2002; and idem., “Rights, Freedoms, and Customs in the Making of Chinese Civil Law, 1900–1936,” 2004. I have benefitted greatly from conversations with Jérôme Bourgon on these matters. For an English translation of the Civil Code of 1929, see Hsia Ching-lin, et al., 1931. The original text can be found in Zhongguo di'er lishi dang'anguan, Zhonghua minguo shi dang’an ziliao huibian 中華民國史檔案資料匯編 (Compilation of Archival Materials on Republican History), 1999: 317–459. 36 Ultimately, I suggest, this derives from the schematization of the interconnectedness of a scalar series of domains to be found in The Great Learning (大學 Da Xue), which set the general parameters and form of governmental thought from its canonization in the Southern Song dynasty: “Only when things are investigated is knowledge extended; only when knowledge is extended can our thoughts be sincere; only when our thoughts are sincere our minds be rectified; only when our minds are rectified can we cultivate ourselves; only when we have cultivated ourselves can our families be regulated; only when our families are regulated can our states be well governed; and only when our states are well governed can All Under Heaven be pacified” (Chan, A Source Book in Chinese Philosopy, 1969: 86–87, translation modified). 18  century, as I've suggested, the family is repositioned vis-à-vis other social-organizational forms, becoming rather a central point of exchange between the natural-biological domain and the social-political domain. In the statecraft tradition, social order is the effect of a ceaseless activity of governing; if governing disappeared, the possibility of social order would disappear with it. For political economy, a preexistent social order is imagined to be recognized, registered, taken account of, and (if they are to be successful) respected by practices of governing; governing as a specialized activity could disappear and social order would remain.37 Thus, if one wants to define a form of governing, as opposed to a political theory or an ideology, one should not look to the explicit principles according to which it understands or limits itself, or how it differentiates itself from other species of its own genus. One should look instead to the relationships it addresses and which organize its activities. One should look to what it makes immanent to what else. These questions have little to do with the standard questions asked by liberal and conservative (or Marxist, for that matter) theories of social power: What justifies it? What are the bounds beyond which it cannot pass without thereby becoming “illegitimate” and tyrannical? We also cannot accept as our terms of reference the pregiven terms the permutations of whose relations political theory studies, such as the subject (whether legally or culturally defined), the individual, the family, the community, society, the nation, or the state. Sometimes governmental practices take these terms as given, and sometimes they don't. The history of a form of government will be tied not to a particular set of political or social universals but to a field of governability, which I must now define.38 37 This is linked historically to the emergence, right in the same period as political economy and as part of the same transformation, of the concept of “civil society” as a transhistorical constant within a liberal analytics of power. (On this, see Foucault, The Birth of Biopolitics, 2008: 291–316.) It is essential to note that this civil society is not at all the same thing as the “society” that is the bearer and site of “social problems.” This latter category really only came into existence at the end of the nineteenth century, through a process much like the one I will try to describe herein; see Gilles Deleuze's Foreword to Donzelot, The Policing of Families, 1979: ix–xviii. 38 I do not mean “governability” as an idiosyncratic translation of Foucault's term gouvernmentalité, for which the established “governmentality” is perfectly adequate. By this term I mean to indicate a field 19  FIELDS OF GOVERNABILITY 天下惟器而已矣 The world consists solely of apparatuses. — Wang Fuzhi (1619–1692)  A good deal of Michel Foucault's work (not all of it, to be sure, but certainly his work in the 1970s) can be characterized as the specification of these fields of governability, and the description of the historical conditions of their emergence and transformation, a field being something like a charged space in which a number of elements coexist in such a manner that an action performed on any one of them affects all of the others: by acting on any element, one adjusts the coordinates of the whole system that is thereby defined. The purpose of the kind of specifically governmental reflection that Foucault is trying to isolate and describe in these studies, then, would be to organize interventions in such a way that one can distribute effects through an entire field by the rational, reflected deployment of measures which would take the reciprocal organization of its elements into account. The most famous of Foucault's studies along these lines is the first volume of The History of Sexuality, and sexuality is as good an example as any of the kind of activity that pertains to a field of governability. It is not unrelated, either theoretically or historically, to the topic of the present dissertation. We misconstrue Foucault's analysis of sexuality if we take it, in the first instance, as an object; it exists first as a field in which objects are in relation with each other. The order of derivation is important: It is not the case that sexuality was what “enabled” women's bodies, children's sexual practices, marriage and reproduction, and the perversions—to take Foucault's examples—to be drawn together onto a common terrain. On the contrary, it was the drawing constituted by the governability of its elements, be they objects, sets of objects, or relationships. That is, such a field exists in virtue of the strategic intent on the part of one who governs—the nature of this one being indeterminate—to adjust the field toward the attainment of a given goal. As soon as this strategic intent disappears or changes, so too does the corresponding field of governability. See immediately below for some examples of this term. 20  together of these elements through techniques of governing and knowledges devoted to quite other ends and quite different objects that produced this new thing, sexuality. At a first pass, then, sexuality would be the field in reference to which it becomes possible to say that any modulation or fluctuation in any one of women's bodies, children's sexual practices, marriage and reproduction, or the perversions instantly and immediately redounds upon the others. It is formed as their common substance, what makes them all “have to do with each other.” Families, according to Foucault, played a very precise role in the constitution of this field in the eighteenth and nineteenth centuries in Europe and America. They were central sites of elaboration for the tactics and interventions that brought about the tendential shift from an axiomatics of alliance to an axiomatics of sexuality. As Foucault puts it: The family cell, in the form in which it came to be valued in the course of the eighteenth century, made it possible for the main elements in the deployment of sexuality [the aforementioned] to develop along its two primary dimensions: the husband–wife axis and the parents–children axis.39 A variety of knowledges and interventions came to be arrayed along each of these axes. This kind of family drew a whole series of disparate problems into its orbit, and this new orbit formed the condition, the “historical a priori,”40 of the emergence of something called sexuality: “'Sexuality' was taking shape, born of a technology of power that was originally focused on alliance.”41 It only became an object, strictly speaking, with a knowledge and a practice of governing specific to itself, quite late: roughly, the 1870s,42 which would be the point at which a possible object of a science “crosse[d] the threshold of epistemologization,” as Foucault puts it in The Archaeology of Knowledge.43 We could imagine this as a transition from ground to figure, a process of the formation of the objects of sciences by a curvature of social space induced by 39 Foucault, The History of Sexuality, Volume I, 1990: 108. 40 Foucault, The Archaeology of Knowledge and The Discourse on Language, 1972: 142. 41 Foucault, The History of Sexuality, 1990: 108. 42 The centrality of this decade to the transformations I am describing in this dissertation will become clear as I proceed. 43 Foucault, The Archaeology of Knowledge, 1972: 190. 21  the collective action of technologies of knowledge and governing, and this curvature occurs through and on the basis of the instrumentalization of the family in a variety of ways. In his lecture course at the Collège de France in 1978, published recently as Security, Territory, Population, Foucault outlines a similar history of the population. (Again, this will not be unrelated to China in the early twentieth century.) We know that historically, in the major European languages, “population” designated the process of populating a place before “the” population was the name of an object.44 For it to become an object, a very particular series of transformations have already to have occurred. The Physiocrats, for instance, in relation to the grain trade—here I am summarizing the first several lectures—have already to have resituated the fundamental activity of government at the level of “needs” (besoins), and the basic level of its articulation at that of production and the labour process rather than that of exchange.45 In another area, the town or the city—with its circulations and desires—has to have displaced the feudal territory with its fixity and obligations as a “primary problem” of governmental reflection.46 In addition, certain technical innovations have already to have been developed around contagion, illness, and death: mortality tables, vaccination, discoveries in probability and actuarial mathematics, and so on.47 None of these problems—of “the town, scarcity, and 44 The Chinese term 人口 renkou does not appear to have been used to translate “population” before 1900; the late imperial administrative category of hukou has a very different meaning (see n. 6, above). 45 For an excellent discussion of the Physiocrats in precisely this connection, and in relation to the emergence of “the economy” as an object of government, see Rosanvallon, Le capitalisme utopique, 1979: 34–62. (Especially interesting is Rosanvallon's discussion of the role of Quesnay's interpretation of the nature of the Chinese emperorship in the development of his theories.) It is worth noting that Marx, in Theories of Surplus Value, also describes the Physiocrats as occupying an absolutely central transitional position: “The analysis of capital, within the bourgeois horizon, is essentially the work of the Physiocrats. It is this service that makes them the true fathers of modern political economy. In the first place, the analysis of the various material components in which capital exists and into which it resolves itself in the course of the labour-process … The Physiocrats transferred the inquiry into the origin of surplus-value from the sphere of circulation into the sphere of direct production, and thereby laid the foundation for the analysis of capitalist production” (Marx, Theories of Surplus Value, 1978b: 222, emphasis in original). 46 Foucault, Security, Territory, Population, 2007: 64. 47 The history of statistics is, of course, the the subject of a substantial scholarly literature. Ian Hacking's work (especially The Emergence of Probability, 1975 and The Taming of Chance, 1990) has been very 22  epidemics” (and risk)—developed with explicit reference to the population, but collectively they produced the curvature in the social and epistemic field by which “the” population could make such a “remarkable entrance” into political and economic reflection in the second half of the eighteenth century, an entry that is widely noted at the time.48 In the first instance, then, population is the field in which taxation, migration, trade and the circulation of goods and currency, the relative importance of economic sectors, city design and urban planning, policy, illness, and a host of other phenomena are conjoined to each other: The population is therefore everything that extends from biological rootedness through the species up to the surface that gives one a hold provided by the public. From the species to the public; we have here a field of new realities in the sense that they are pertinent elements for mechanisms of power, the pertinent space within which and regarding which one must act.49 In a schematic way, then, this is what I mean by a field of governability. With this in the background, then, we can move to our real topic: the formation of just such a field of governability in China, the carving out, for the first time there, of a series of relationships between family forms, demographic phenomena, and a national economy—a series of relationships that defines a certain moment in the development of what I think we can call a capitalist world-episteme. With reference to the definition in China of a nexus of immanence involving the population, the family, and the economy, I suggest, I can not only cast the history useful in fleshing out some of Foucault's claims, historically speaking, and in tracking the development of important related aspects of modern governmental knowledge. The transition in the nineteenth century from amateur science toward systematic vital statistical work is the subject of Andrea Rusnock, Vital Accounts: Quantifying Health and Population in Eighteenth-Century England and France, 2002, while the history of the incorporation of these techniques into large-scale, statecentric systems, and the transformations that are wrought by that relocation, have been covered in works like Theodore M. Porter, The Rise of Statistical Thinking, 1820–1900, 1986; Alain Desrosières, The Politics of Large Numbers: A History of Statistical Reasoning, 1998; and Libby Schweber, Disciplining Statistics: Demography and Vital Statistics in France and England, 1830–1885, 2006. 48 Foucault, Security, Territory, Population, 2007: 67. 49 Ibid., 75, emphasis added. Further, biopolitics—which, for Foucault, is the specific form of power that arises from these developments (which distinguishes it from a number of other current usages of this term)—“will derive its knowledge from, and define its field of intervention in terms of: the birth rate, the mortality rate, various biological disabilities, and the effects of the environment” (2003: 245, emphasis added). 23  of China in this period in a new light, but I can also go some way toward a general characterization of capitalist governmentality in the early twentieth century. Thus, the goal is not to track the development of understandings or ideas of the population, nor their progressive approximation to some notion of population that we might recognize as scientific or “within the true.” Hence, the corpus that is defined by this problem cannot be limited to those texts and debates in which it was immediately or explicitly thematized. One would hardly do justice to the importance of the problem by studying the texts of those who might be considered population experts (though naturally one can hardly ignore them). The goal, rather, is to investigate the transformations introduced into a wide variety of governing practices by the installation of the population in them. In a sense, then, my purpose is to determine the corpus that would be the trace and effect, the body, of the apparatus50 of population management in Republican China. The problem is not “What did this or that person or group of people think or do about the population?” but “What apparatus must have come to exist, such that this corpus exists?” The approach taken, then, is quite close to what Foucault, in The Archaeology of Knowledge, calls “the analysis of the episteme.”51 There, in the process of differentiating his own approach from that of others concerned with the history of knowledge, he describes an episteme as the total set of relations that unite, at a given period, the discursive practices that give rise to epistemological figures, sciences, and possibly formalized systems … The episteme is not a form of knowledge (connaissance) or type of rationality which, crossing the boundaries of the most varied sciences, manifests the sovereign unity of a subject, a spirit, or a period; it is the totality of relations that can be discovered, for a given period, between the sciences when one analyses them at the level of discursive regularities … [I]t is what, in the positivity of discursive practices, makes possible the existence of epistemological figures and sciences … In the enigma of scientific discourse, what the analysis of the episteme questions is not its right to be a science, but the fact that it exists.52 The episteme, therefore, is a descriptive artefact of a given configuration of knowledges, 50 I discuss this term immediately below. 51 Foucault, The Archaeology of Knowledge, 1972: 193. 52 Ibid., 192. 24  institutions, practices, and strategies, a specific and historical system on the basis of which it is not just possible but necessary that certain objects appear (such as a population in China). How, then, are such configurations put together? In the early 1970s, Foucault shifted his point of analysis from the specification of epistemes to the description of apparatuses (dispositifs), these latter being the historical systems on the basis of which it is possible to describe the appearances and disappearances, shifts and displacements, of epistemes and the objects and fields they involve.53 In an interview conducted soon after the publication of the first volume of The History of Sexuality, Foucault describes an apparatus as a thoroughly heterogeneous set consisting of discourses, institutions, architectural forms, regulatory decisions, laws, administrative measures, scientific statements, philosophical, moral, and philanthropic propositions … Such are the elements of the apparatus. The apparatus itself is the network that can be established between these elements … [T]he nature of an apparatus is essentially strategic, which means that we are speaking about a certain manipulation of relations of forces, of a relational and concrete intervention in the relations of forces, either so as to develop them in a certain direction, or to block them, to stabilize them, and to utilize them. The apparatus is thus always inscribed into a play of power, but it is also always linked to certain limits of knowledge that arise from it and, to an equal degree, condition it. [It is] a set of strategies of the relations of force supporting, and supported by, certain types of knowledge.54 An apparatus, then, produces, and unfolds in relation to, a field of governability, which in turn is a set of things to be governed, and a set of relationships determined to be operative between them, organized by a given set of intended effects and strategies, and a way of being related or relating onself—on the part of the one who governs—to this set so that these effects can be induced. At a formal level, we can approach this from another direction. In a commentary on the external chapters of the Yijing 易經—the same commentary that provides the rather striking epigraph to this section—the seventeenth century philosopher and historian Wang Fuzhi 王夫之 53 This shift is rather better known as the shift from archaeology to genealogy; there is a significant specialist literature about its nature and consequences, which I will not delve into here. 54 Foucault, Power/Knowledge, 1980: 192–194. 25  discusses the relationship between the Way (道 dao), or a Way (which might correspond to a logic, as I am using that term here), the activity of ordering (治 zhi), and the instruments (器 qi) of rule in a manner very reminiscent of Foucault's work: Not yet having both bows and arrows, there can be no Way of archery; not yet having both vehicles and horses, there can be no Way of charioteering. If there is not already sacrificial wine (牢醴 laoli), ceremonial currency (璧幣 bibi), 鐘磬 zhongqing,55 and 管紘 guanhong,56 there can be no Way of Rites and Music. Without yet having a son, there is no Way of being a father, and without yet having a younger brother, there is no Way of being an elder brother. The ways of possessing and failing to possess the Way are manifold. Thus, if one lacks an apparatus (器 qi)57 one lacks a Way as well … Thus the ancient sages could order apparatuses (治器 zhi qi), but they could not order the Way (治道 zhi dao). What is called a Way is the ordering of apparatuses. What is called virtue or efficacy (德 de) is the achievement of a Way. What is called putting into effect (行 xing) is the completion of apparatuses. The breadth of the utility of apparatuses is what is called their transformability and flexibility (變通 biantong). What is called an undertaking (事業 shiye) is the manifesting of the effectivity of apparatuses … Thus, for the sage, seeing to the benefit (善 shan) of the people consists solely of ordering apparatuses.58 A Way, then, follows upon and is a function of the establishment of an apparatus. It is the effect of the putting into relation of concrete things, rather than what generates them. (As though the Way of archery generated bows and arrows; on the contrary, it is only once there are bows and arrows that there can be a Way of archery, which can then reflexively redound upon the formation of particular bows and arrows. If there were no bows and arrows, in what would the Way of archery consist?)59 55 An ancient ceremonial musical instrument. 56 Another instrument, made of bound pipes. 57 The more usual translation of this term is “instrument”, “machine,” or “organ.” Thus, a 武器 wuqi is a weapon (lit. an instrument of war) and 電器 dianqi refers to electrical equipment in general. But in the sense of a machinery that makes it possible to act in a particular way, it is like an apparatus in the Foucauldian sense; hence the present translation. 58 Wang, Zhou-Yi waichuan, 45a–b. 59 I will return to this notion of government in greater detail in Chapter Three. I am indebted here to the studies of Wang by François Jullien (Procès ou création: une introduction à la pensée des lettrés chinoises, 1989; Figures de l'immanence: Pour un lecture philosophique de Yi king, 1993; and The Propensity of Things: Toward a History of Efficacy in China, 1995) and Jacques Gernet, La raison des choses: Essai sur la philosophie de Wang Fuzhi (1619–1692), 2005, even though they disagree with each other, and even though I do not think (as Jullien apparently does) that this authorizes a general schematization of “the Chinese way of thinking.” 26  By the early twentieth century, of course, the population had already crossed the threshold of epistemologization, in the sense that it had already, by the time anyone in China took notice of the governmental technologies that gave birth to it, emerged from out of their collective operation and had been reflexively put to work within them, and it traveled along with them. In fact, the first iteration of the population problem in China occurred before the installation of its particular apparatus. It is usually assumed that if a new notion becomes intelligible to a group of people operating within a horizon that does not at first include the conditions for it, this change must take place at the level of a series of realizations or translations, by which the new is at first encoded in the terms of the old. To this assumption corresponds such questions as: How did Chinese people “get it”? Through what cognitive operations was the epistemological space for such a notion produced? One then looks to translations, debates, and so on, for evidence of a gradual approximation or convergence toward the modern. Once they get it, they proceed to reorganize their activities accordingly. Such will not be the approach taken here. The discussion of the population in the first decade of the twentieth century had something to do with biopolitics (but what exactly is not obvious), but however far these discussions may have taken their participants toward an apprehension of the biopolitical, what happened in the 1910s and 1920s was not due to them. I will argue that the conditions of the biopolitical in China were established by transformations in areas that did not have anything explicitly to do with these earlier discussions. The formation of the epistemological conditions of there being a Chinese population that appears as such to Chinese people does not precede its installation into the grain of governmental activity in China. Things become objects of knowledge not because internal or external obstacles to perceiving them are removed, but because of modulations in the positive order of knowledges. I am dealing here not with a process of bringing into the order of knowledge an object that had previously lain outside it, but with the replacement of one order of  27  knowledge by another. CAPITAL AND CAPITALIST GOVERNMENTALITY A population is, as Marx points out in the Grundrisse, “an abstraction.” Specifically, an abstraction from “the classes of which it is composed. These classes in turn are an empty phrase if I am not familiar with the elements on which they rest. E.g. wage labour, capital, etc.”60 Little is gained by simply pointing this out, though. Even if it is granted that, by being abstract, the population is somehow not real, we still have a tremendous edifice of entirely real social phenomena that are concerned with it. There are two very different ideas about the role of abstraction in thought, and by extension in historical explanation, and if this project trafficks in the abstract, it will be important to be clear about which one I mean. The first and more familiar one—clearly, then, the one I will not be using—holds that the abstract is formed by the progressive removal of the concrete. Abstraction signifies a process of progressive subsumption of more and more particulars under a general form or summarizing idea. In this figuration, the abstract and the concrete are two poles, and as one approaches one, one moves away from the other. The more abstract something is, the more removed it is from the concrete, and it is better to stick as close as possible to the concrete. But there is another, more dialectical notion of abstraction, in which the abstract is “written into” the very grain of the concrete, and always at work within it. Take money, for instance. Nothing can be more abstract than money (except value, which it represents), and yet it is impossible to get through the most concrete activities of daily life without it. And value, to the notion of which we now turn, is yet another step more abstract, but the whole system of capitalist governmentality that will form our object in the following chapters is organized around it. Once the abstract dimension of things governs their 60 Marx, Grundrisse, 1973: 100. Note the very precise relationship here: the classes of a capitalist society rest on the forms of economic relationships, rather than these forms constituting instruments in the hands of the different classes. 28  interactions (as it does in the case of monetized exchange), one can say that the abstract really, in fact, generates the concrete, and that the concrete is no longer explicable without it.61 Marx's theory of value is one of the most difficult elements of his analysis of capitalism. Whole schools and tendencies within Marxism have differentiated themselves from each other over their interpretations of it.62 With any luck, we will be able to avoid entering into the arcana of this debate—it would be of little interest to most historians in any case—but some understanding of value is necessary if we are to proceed with an investigation of capitalist governmentality, which is centrally, even entirely, concerned with it. “The wealth,” then, “of societies in which the capitalist mode of production prevails appears as an 'immense collection of commodities',”63 the essential property of each of which is its value. Value is what constitutes the general exchangeability in virtue of which alone it is a commodity in the first place, a constituent of “wealth in general,” as opposed to a mere useful article. Where before value had served as a measure of wealth, here value is introjected into the very definition of wealth: In order that something may be counted as wealth, it must be a commodity; insofar as it is a commodity, it is wealth, and vice versa. It is the property by which any material particular can be equated to any other, without thereby forming an dissociable property of any actual particular. Value cannot be detached from a thing like a “part.” Further, at no point is value (or capital as self-valorizing value) identifiable with any of the forms that it recursively takes—not even, or even especially not, money, quantities of which is how it appears in capitalist cost 61 Of course, in Republican China, it was possible to conduct a significant portion of one's daily activities without the use of money. But, as we will see in Chapter Five, this was a major problem for economic reformers at the time. The forms of knowledge by which they proceeded presupposed universal monetization, which presented all kinds of problems of transcoding: the calculation of wage-equivalents, book values of articles, and so on. 62 This debate has turned on questions like: Is the law of value universal, or specifically a feature of the capitalist mode of production? Is it something on the conceptual basis of which socialist demands might be made, or is it part of the problem? What is its relation to labour? I will return to this in greater detail in the Conclusion. 63 Marx, Capital, Volume I, 1990: 125. This is the first sentence of that work. 29  accounting.64 We can see this by noting a simple fact about all the symbolic formulae for the circulations of capital given by Marx in the second volume of Capital: capital itself never appears in them. M and M' (money), P (production), C and C' (commodities), L (labour-power), mp (means of production)—all of these forms of value are only “evanescent moments” of capital,65 which exists (but does not appear in any phenomenal form) only in virtue of their enchainment in a particular process (“value in process, money in process, and, as such, capital”).66 The first thing to do, then, is to dispense with the idea that Marx's analysis of capital is “materialist” in the sense in which this term is usually understood, and not just by non-Marxists: the sense in which everything is “really” concrete, and abstraction carries with it the taint of idealism, supposedly materialism's great enemy. Capital consists, in the first instance, of value, a feature that “no chemist has ever been able to isolate” among the material properties of anything.67 Marx's analysis of capital is not concretizing, in the sense of revealing that what is understood to be abstract is “really” concrete. It is the exact opposite: in the commodity, what appears as entirely concrete—a useful article—is, insofar as it is capital, shot through with value and its universe, the market. Capital, as Deleuze and Guattari have it, is “the abstract as such.”68 We have then also to dispense with the idea that the labour referred to in the much-abused “labour theory of value” is actual or concrete labour. Marx never actually says this, and if he did, his theory of value would deserve its reputation. As it stands, though, it does not. What Marx actually says is that what determines the value of a commodity, “on average and in the long term,” is the average socially-necessary labour time required to produce it within a given 64 This relies on the distinction between money of account and money as means of exchange. Money, then, appears in two ways on a balance sheet: as itself (“cash on hand”), and as the measure of everything else. 65 Marx, Grundrisse, 1978: 142. It is “real,” then, in the way that a Way is real for Wang Fuzhi. 66 Marx, Capital I, 1990: 173. 67 Ibid., 177. 68 Deleuze and Guattari, Anti-Oedipus: Capitalism and Schizophrenia, 1983: 261. 30  economic formation (which makes his analysis useless to anyone concerned with a capital over a short period—say, an annual accounting cycle). The form of the commodity, its combination of use-value (a set of actual properties) and value (an abstract quantity of an undifferentiated substance), links it, this “cell,” immediately to the global market by reference to which alone such an average can be discovered. Its value, supposedly interior to itself, is a complex function of its position within the world market, the global division of labour and the extent of the latter's systematization, at a particular point in time. Thus—this is a crucial point—the value of any particular commodity is determinable only by immediate reference to the whole universe of commodities, to the whole global field of capital that, in virtue of its form, it contains within itself, as a fetish. This is how the commodity both presupposes and constitutes not just this or that market, or even the global market, but the market as such, the universe of the commodity. We know, or at least most of us are prepared to entertain the possibility, that capitalist production is unique among all forms of production in the sense that it is undertaken for value, with exchange value already in view at the beginning: that is, it begins and ends in the abstract, with the concrete serving merely as its bearer. As Marx puts it, as far as capital is concerned, the capitalist, “or rather his pocket,” is just “the point from which the money starts and to which it returns.” The capitalist is “capital personified and endowed with consciousness and a will.”69 In this way, capitalist production is at the same time the production of things for the satisfaction of needs and the production of the abstract. One can thereby formulate an answer to a possible objection: Surely you are overstating the discontinuity, a discontinuity that, in any case, a whole scholarly literature has dedicated itself to stitching over. Surely capitalism is just the latest system—and the most efficient one—for the satisfaction of wants and needs, this last process being the definitive characteristic of forms of economic organization as such, of which genus  69 Marx, Capital I, 1990: 170. 31  capitalism would then be a species. In the light of the above, one might go so far as to suggest that this conception of economic activity (one which was only formalized within capitalism) is true of every mode of production except capitalism. In capitalism, the satisfaction of needs is instrumentalized in relation to the expansion of value. For capital, the satisfaction of particular needs is merely the means (and not even always the most felicitous means) for the accumulation of value—which is, according to the marginalist conception, satisfaction in general.70 I am aware, of course, that the universal category of choice in historical explanation these days is modernity, not capital. The above, though, puts us in a position to venture a claim about the relationship between these two terms that is, to my mind, more useful than either pole in the endless barrage of discussions about which is “deeper” or which one drives the other. Clearly I cannot hope to put a stop to this debate; it is too deeply engrained and too many people are invested in one answer or the other. Although I have my preference, all I wish to do here is to describe how I will put them into relation in this dissertation. That claim is: the modern is the form or apparatus of capital. This does not mean that “behind” the modern, the “real” motive force is capital, dissimulating itself into so many deceptions. Rather, capital exists as the modern. The modern is the form of life that results from the forms of social organization having been recoded as elements in a process of accumulation, from their being made tendentially to approximate to the optimal conditions for the accumulation of capital. Likewise, modernization would be the process of recoding. The modern provides the form in which capital can exist— that is, without which it would have no body, just as capital becomes the substance that the modern is comprised of. This is not to suggest that capital is—it is even less to suggest that capitalists are—the ultimate mover behind modernity.71 Rather, capital is the universal in the 70 This explains the oft-wondered at fact that commodities (surplus agricultural production in one country, for instance) that would satisfy perfectly real needs (starvation in another country) will be destroyed before they will be distributed in such a way as to block the further accumulation of value. 71 Although this claim should not be dismissed too hastily. One need only think of how quickly this or that cultural form will be jettisoned if it acts as hindrance to this accumulation, and of how cultural 32  particular, the “stuff” of the modern. To put it in Foucauldian terms, the modern plays a directly productive role vis-à-vis capital. The population (whether of China or as such) is, I suggest, exactly this kind of abstraction,72 one that exists in virtue of a given form of practice and knowledge. But it is not, for that, unreal: it really structures the concrete from within. One of the most important products of the process of capital, and one whose importance to our present concerns cannot be overstated, is the notion of labour-power, the abstract capacity to produce, analytically separable from any determinate act or kind of production. That is, capital produces labour-power—one abstraction to match another: just as capital is constituted by the abstraction of value as such from any particular material embodiment, so too is labour-power constituted by the abstraction of a capacity from any particular form of work.73 In doing so, it produces the population, which is, as I will argue in greater detail later, the general level of the human, human collectivity stripped of any particular social characteristic: prior to the family, to the ethnos, to the nation, the population concerns those features that may be predicated of any human being whatever, as a living being. The population appears within capitalist governmentality (and this is, after all, its first historical appearance) as the material bearer of this labour-power. We know—it has been said a million times—that the commodification of labour-power is one of the defining features of capitalism. Less often noticed is the fact that the practices must, it seems, find their justification in their rationality before they can be justified on solely cultural grounds. 72 As is, for that matter, “the” family. 73 The development of the science of abstract work has been admirably treated in M. Norton Wise, “Work and Waste: Political Economy and Natural Philosophy in Nineteenth-Century Britain,” 1989a (Part I) and 1989b (Part II)—I will return to this question in Chapter Five; for an excellent cultural history of the science of fatigue—the negative of the capacity to work—see Anson Rabinbach, The Human Motor: Energy, Fatigue, and the Origins of Modernity, 1992; Sean H.L. Lei has written a fascinating study of the transcoding of traditional Chinese disease categories in the 1920s, such that diseases which had been understood to be the result of overwork came to be understood in terms of a subjective inability to endure work: see Lei, “Weisheng weihe bushi baowei shenging? Minguo shiqi de bielei de weisheng, ziwo yu jibing 衛生為何不是保衛生命? 民國時期的另類的衛生, 自我與疾 病 (Why Weisheng is Not about Guarding Life: Alternative Conceptions of Hygiene, Self, and Disease in Republican China),” 2008, and “Why Weisheng is Not about Guarding Life: Alternative Conceptions of Hygiene, Self, and Disease in Republican China,” 2009. 33  very notion of labour-power only arises, and can only arise, as always-already commodified.74 And even less often noticed is the argument I will make: that the population is conjured into existence as a correlate of this development.75 If then, as Gayatri Spivak says, capital “decodes and deterritorializes the socius by releasing the abstract as such,”76 capital is not a “part” of any social formation. On the contrary, it is a kind of anti-social formation. It is not, therefore, an instrument in the hands of a European or Western bourgeoisie; Europe is just the place with the first historical experience of the abstract as such, i.e., capital. Capital is, rather, a formal-transcendental matrix that can be installed as easily (or with as much difficulty) in one social formation as in any other, all while leaving things, at first, and on a phenomenal level, much as they were before. In this sense (not in all senses), it doesn't matter what kind of social formation existed “before.” Capital is not the emanation of any particular culture. It did not arise organically from, and nor does it maintain any necessary connection to, “Western” cultural forms or trajectories. Thus, what I mean by capitalist governmentality (this would be my technical definition) is: that form of governing which approaches material particulars in terms of the adjustments that can be made to them so that they enter into the best possible relationship with the requirements of the accumulation of value in the abstract. That is, it works with the particular and the concrete (in this case, the field of Chinese social life as it has been produced historically) by way of their involvement in the processes that generate capital. Capitalist governmentality works at two levels simultaneously. First, the coding of the entire socio-natural system according to a universal form—this is the function of “science,” both natural and social77—and second, the 74 This argument is related to the work of Moishe Postone, who describes his work as a shift from “the critique of capital from the standpoint of labour” to a critique of labour as constituted within the form of capital. See Postone, Time, Labor, and Social Domination, 1994. 75 It is not sufficient that there be capital for there to be a population, as I will argue below, but it is necessary. 76 Spivak, Outside in the Teaching Machine, 1993: 62. 77 Thus I must necessarily be involved with the disciplinary subfield of the history of science, without 34  determination of the best and most efficacious avenues for improvement given by the particular configuration of this socio-natural field, in all of its historical contingency and messy particularity. Capitalist governmentality proceeds, then, by way of the abstract in the concrete and the given; it approaches the latter in terms of the manner in which it embodies the former. For this to occur, a tremendous positive labour of abstraction must take place, a coding of the concrete in terms of the abstract. This is what all the multitudinous surveys, censuses, investigations, and studies that I will analyse in the following chapters really do. We can call this the installation of the value-form into the system of governing.78 Capitalist governing, in order to get a hold on the universe of particular things, projects into them their abstract dimension, and then works at that level. With the development of capitalist governmentality in China, political knowledge takes on a new asymptote, a new formal structure in terms of which particular interventions are evaluated as more or less successful.79 This recoding of the concrete in China according to a new abstraction is the fundamental activity whose history I am writing. How are we to imagine that a reorganization of social life according to the requirements of accumulation would not call into existence a particular form of governing, one proper to itself? In fact, we do not imagine this, and we have a name for it: economic governance. The only mistake—and it is a very old mistake, but one that we somehow never tire of making—lies in taking it for the very model of governing itself, for that to which, like the economy itself, all actually wishing to write a history of science, strictly speaking. 78 This is one of the senses in which W.S. Jevons, one of the progenitors of neoclassical economics—to whom we will return several times in the following chapters—can define economics as “the science of Capitalisation” (The Theory of Political Economy, 1879: 241). 79 Thus, while the seemingly endless discussion of how much and at what levels China in this period “became capitalist” is useful, I need not take a position in it. By the capitalization of China, here, I do not mean the relative degree of presence of any particular set of institutions (legal structures, forms of family production, forms of business organization, state institutions, markets and contracts, and so on). Nor is it a question of what percentage of economic activity takes place under conditions that are generally recognized as capitalist (how many wage workers were there? how capitalized were Chinese businesses? how much of the surplus is appropriated by non-capitalist extraction versus capitalist exploitation?). Even if most of China was “non-capitalist” in these senses, by the 1930s all nationalist governing presupposed the schematization of capital. 35  governing would tendentially approximate, with only the removal of impediments. And, from within the horizon of intelligibility that capital constitutes—that of what Marx calls vulgar political economy—how could we not see things this way?80 CHINESE NATIONALISM AND ITS APPARATUSES Thus, I want to reemphasize a discontinuity at a point in history where a number of contemporary scholars have chosen to emphasize continuity. A wealth of scholarship has been produced around the strategies and social forms by which (some) Chinese elites came through the transition to modernity intact, that is, in the “same” position of social superordination that they were in before. Social networks and the institutions that structure them—native place associations, the scholarly system that had been organized around the civil examination system and the hierarchy of new educational systems that emerged out of its dissolution in 1905, local and translocal systems of patronage and influence, and so on—have been interpreted (correctly, I think, at a certain level) as the undergirding of social organization that carried elites into the modern, the ropes cast across the chasm that separates the nineteenth century from the twentieth.81 These, then, would be the loci of a continuity, the means by which the old survives 80 “Vulgar” here does not mean stupid or unsophisticated, but is used in a rather more technical sense. It refers to an approach that would seek to understand the historical emergence of an economic formation (in the case of political economy, capitalism) entirely from within the coordinates established by that formation. Thus, the transcendental schematization of economic life as such in terms of life, labour, capital, and the economically rationalizing subject that came to characterize what would be called neoclassical economics in the 1870s would represent something like the point of absolute vulgarization. I return to this in the next chapter, and in Chapter Five. 81 Some—by no means all—of the key works in this scholarship are R. Keith Schoppa, Chinese Elites and Political Change: Zhejiang Province in the Early Twentieth Century, 1982; Min Tu-ki, National Polity and Local Power: The Transformation of Late Imperial China, 1989; Susan Mann, Local Merchants and the Chinese Bureaucracy, 1750–1950, 1987; Mary Rankin, Elite Activism and Political Transformation in China: Zhejiang Province, 1865–1911, 1986; Byrna Goodman, Native Place, City, and Nation: Regional Networks and Identities in Shanghai, 1853–1937, 1995; Yeh Wenhsin, The Alienated Academy: Culture and Politics in Republican China, 1919–1937, 1990; and William Rowe, Hankow: Commerce and Society in a Chinese City, 1796–1889, 1984, and Hankow: Conflict and Community in a Chinese City, 1796–1895, 1989. Several studies of urban reform in various cities in China also stress the continuity of elite action; see Kristin Stapleton, Civilizing Chengdu: Chinese Urban Reform, 1895–1937, 2000; Michael Tsin, Nation, Governance, and 36  into the new, stamping Chinese modernity with its particularity. But we ought not to let a continuity at one level—the occupants of positions of social power—obscure a discontinuity at another—that of the general organization of relations of power. The other great mechanism of continuity is, of course, culture, a “Chinese culture” that is written into both these forms of social organization and the assumptions and intentions of the people who participate in them, and that “covers” the transition, introjecting it into a transition “within” Chinese culture. But here one should take careful note of an historical phenomenon that is, I suggest, closely correlated (both temporally and conceptually) to the emergence of the population in China: the global transmission of the concept of culture itself.82 As with the population, culture will not be treated here as a constant, something that exists as a kind of substrata of change (in the sense that one would say that a change takes place “in” culture). I will not accept the “cultural anthropological postulate”—that there is nothing controversial about stating that everywhere has a culture—just as I will not accept the demographic postulate. I will take seriously the fact that culture was marked out as a discrete field of social life, a field in which a specificity is to be found, at more or less the same time as the population, which functions at the level of the universal. I begin from the historical fact of the problem of culture, Modernity in China: Canton, 1900–1927, 1999; and Joseph Esherick, ed., Remaking the Chinese City: Modernity and National Identity, 1900–1950, 2000. 82 Andrew Sartori's Bengal in Global Concept History: Culturalism in the Age of Capital, 2008, provides a useful global history of the concept of culture (see esp. 25–67) before his more extended analysis of its discursive and political functions within elite nationalism in late-nineteenth and early twentieth-century Bengal. For the different ends to which the concept of culture (and its close correlate, civilization) could be turned in East Asia in the first half of the twentieth century, see Prasenjit Duara, Sovereignty and Authenticity: Manchukuo and the East Asian Modern, 2003: 89– 129. For a comparison of Chinese and Indian nationalisms in this period which attempts to adjust the analytical terms of subaltern studies, produced as they were out of work on Indian history, to the counterexample of China (it is certainly not my intention to reduce the differences, even if I choose to emphasize similarities at particular levels), see Duara again, Rescuing History from the Nation: Questioning Narratives of Modern China, 1995: 205–227. As a cultural historian (which does not mean that he accepts at face value the concept of culture; only that he focuses on the transformations that are refracted through it), Duara takes little account of the dimension of capital. On the global question of colonial and postcolonial national culture, it is hard to do better than Frantz Fanon's “On National Culture” in The Wretched of the Earth, 1963. 37  rather than from the actuality of the referent or its specific contents. One might, from this perspective, suggest that culture becomes the site for the articulation of specificity, the precious bastion, source, and repository of one's innermost and unique selfhood, precisely at the moment and to the extent that everything else is given over the universal logic of capital. Let us not be taken in: the fact that the specific contents of this or that culture—what M.N. Roy (1887–1954), a fierce critic of bourgeois nationalism, called a group of people's “superficial peculiarities”83— are unique dissimulates a much more important phenomenon, which is the inscription of every culture into a universal form, and the establishment of culture as a universal site from which it is possible to stake claims of particularity in the name of collectivities. This “politics of culture” (which can be traced back to the German Enlightenment's essentially hypocritical reaction against, first, Napoleon, and second, somewhat later, with the economic policies that Friedrich List has come to stand for, British commercial power),84 found its way to China in the same period, I suggest, as part of the same transformation that produced the Chinese population.85 A movement of specification at the level of content, then, occurs alongside and in the same process as a universalization at the level of form. However different the particular contents of cultures are, they are all composed of the same kinds of things, all having the same immanent 83 M.N. Roy, “Bourgeois Nationalism” (1923) in Selected Works of M.N. Roy, Volume Two, 1987: 125. 84 In the domain of economic policy, this reaction against Britain is doubled in the US with the trajectory of thought about trade policy that runs from Andrew Hamilton (1757–1804) through Mathew Carey's (1760–1839) Essays on Politcal Economy (1822) to his son Henry Carey's (1793– 1879) The Harmony of Interests (1872), which is as good a statement of economic nationalist principles as any (and not just with regard to international trade). This whole trajectory, and the debate between the British free-trade “school” and the American and German economists of the 1860s and 70s would be an essential structuring foil for working out economic policies in China in the 1920s and 30s. Some of the public debates around this question are discussed in Bryna Goodman, “Economics, Individual Freedom, and National Sovereignty,” 2011. Basically the same thing happened in India (and other places); see Manu Goswami, Producing India: From Colonial Economy to National Space, 2004. On the “hypocrisy” of the German bourgeois reaction to British free-trade cosmopolitanism, see Marx, “Draft of an Article on Friedrich List's Book: Das Nationale System der Politischen Oekonomie,” 1845. 85 Although there is not here the space to treat the question with the depth it deserves, there is a strong case to be made that such a politics of culture could not have existed in imperial China, being linked as it is to the category of the nation, which was absent until the twentieth century. 38  structure. This isolation of “culture” takes place within a very limited span of historical time, in a tremendous diversity of places, and by the activities of the agents of specifically bourgeois nationalist projects. In this period, one entered into the universal form of bourgeois civility not by abstracting oneself from one's particularity, but precisely through and by way of one's cleaving to it, just so long as t