182 17 10MB
English Pages 262 Year 2012
MODALITIES OF CHANGE
MODALITIES OF CHANGE THE INTERFACE OF TRADITION AND MODERNITY IN EAST ASIA
Edited by JAMES WILKERSON and ROBERT PARKIN
Berghahn Books New York • Oxford
Published in 2013 by Berghahn Books www.berghahnbooks.com © 2013 James Wilkerson and Robert Parkin All rights reserved. Except for the quotation of short passages for the purposes of criticism and review, no part of this book may be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system now known or to be invented, without written permission of the publisher.
Library of Congress Cataloging-in-Publication Data Modalities of change : the interface of tradition and modernity in East Asia / James Wilkerson and Robert Parkin, editors. p. cm. Includes bibliographical references and index. ISBN 978-0-85745-568-0 (hardback : alk. paper) 1. Ethnology—East Asia. 2. Social change—East Asia. 3. National characteristics, Asian. 4. Civilization, Modern. 5. East Asia—Civilization. 6. East Asia—Social life and customs. I. Wilkerson, James, 1951– II. Parkin, Robert, 1950– GN635.E18M34 2012 306.095—dc23 2012012478 British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Printed in the United States on acid-free paper ISBN 978-0-85745-568-0 (hardback) ISBN 978-0-85745-571-0 (ebook)
Contents List of Figures, Tables and Diagrams
ix
Preface xi James Wilkerson and Robert Parkin
Introduction Modalities of Change: The Interface of Tradition and Modernity in East Asia
James Wilkerson and Robert Parkin
1
People’s Republic of China Chapter 1
The House, the State and Change: The Modernity of Sichuan rGyalrong Tibetans
Wang Ting-yu
21
Chapter 2
From Kinship to State and Back Again: Lineage and History in a Qiang Village
37
Liu Biyun
Chapter 3
Embroidery Speaks: What Does Miao Embroidery Tell Us?
Ho Zhaohua
Chapter 4
Tensions between Romantic Love and Marriage: Performing ‘Miao Cultural Individuality’ in an Upland Miao Love-Song
Chien Mei-ling
Chapter 5
Modalities of the One-Child Policy among Urban Migrants in China
Chang Kuei-min
Chapter 6
The ‘Culture’ of World Cultural Heritage
Eveline Bingaman
62
93
117 134
Taiwan and Vietnam Chapter 7
‘Amis Hip-Hop’: The Bodily Expressions of Contemporary Young Amis in Taiwan
Futuru C.L. Tsai
151
viii
Contents
Chapter 8
Contesting Memory: The Shifting Power of Narration in Contemporary Paiwan Contexts
Li-Ju Hong
180
Chapter 9
Ethnicity as Strategy: Taiwan State Policies and the Thao
197
Yayoi Mitsuda
Chapter 10
On the ‘Third Morning’: The Continuity of Life from Past to Present among the Nung of Northern Vietnam 209
Afterword
N. Jenny Hsu
Performance as a Mechanism for Social Change
James Wilkerson
Notes on Contributors
225 241
Index 245
List of Figures, Tables and Diagrams Figures 2.1 Part of the zongbei of Izi coufang, the oldest stele in the village erected in 1788
44
3.1 Left-sleeve embroidery: Butterfly-Mother
76
3.2 Right-sleeve embroidery: Fortune-and-Peace
77
3.3 Ob ait duf sleeve embroidery by Grandmother Niu, 1930s to 1950s
82
3.4 Hob mongl sleeve embroidery, 1930s to 1950s
84
3.5 Hob mongl sleeve embroidery, 1930s to 1950s
85
6.1 Monument dedicated to the Inscription of Lijiang Old Town on UNESCO’s World Cultural Heritage List
141
7.1 The malikoda dance of the A’tolan Amis
157
7.2 The Fly Lady Fly play by Laciensi (2003)
159
7.3 Lakayakay’s modern dance in 2003
163
7.4 The malikoda allows for variation with some kinds of limitations
168
7.5 The modern dance of Laciensi in 2005
170
7.6 Female traditional dance movements versus Laciensi’s traditional dance movements
171
10.1 Baskets are brought by households related to the maternal side or bung lang 210 10.2 The maternal grandparents of the newborn arrive with the va, a bamboo-woven platform
210
10.3 The maternal grandparents prepare the va and an extra basket containing zok va, paper-cut flowers
214
Tables 3.1 Names of decorated sections of Shidong Miao jackets
69
3.2 Main categories of festival jackets in Shidong
69
3.3 Various styles of ordinary daily wear jackets in Shidong
71
3.4 Styles of hob mongl festival jackets
72
3.5 The stitches used in Fortune-and-Peace
78
3.6 The symbolic meanings of the imagery in Fortune-and-Peace
80
4.1 Marriages into and out of the Vangf Dof and Ghad Dlongb hamlets of marriage group 4
98
x
List of Figures, Tables and Diagrams
5.1 Scale of migrant population (ten thousand)
118
5.2 Migrant birth-control mechanisms
122
7.1 The structure of the age organization of A’tolan Amis (as of 2003)
155
Diagram 3.1 Shidong Miao jacket
68
Preface With one exception, the chapters in this volume originated from a series of presentations given at Tsing Hua University, Hsinchu, Taiwan, in the summer of 2006 (Eveline Bingaman’s chapter was written subsequently). While no theme was suggested at the outset, it quickly emerged that most of the papers dealt with the interface between tradition and modernity, mostly in the Chinese cultural sphere, as well as in Vietnam. In fact, the papers also mostly dealt with minorities in all the main states represented, namely the People’s Republic of China, Taiwan and Vietnam. While this is not a book about minorities specifically, it is the case that these chapters tell how groups that are minorities in relation to the Chinese politico-cultural sphere in which they exist are confronted by and are confronting a modernity that, if it does not come ultimately from this sphere, at least is channelled through it. The historical colonization of minorities by Chinese can be compared to European colonization, especially of originally under-populated areas like North America, Australasia and Siberia; in both cases, the colonization has become established and has changed into permanent settled life, which has changed local indigenous minorities too. This can be contrasted with European colonization of Asia and Africa, where indigenous populations were sufficiently large to bring an end to that colonization eventually and attain independence. Groups like the Qiang, Miao, Paiwan, Amis, Thao and Nung, dealt with in this volume, are essentially ‘fourth-world’ peoples every bit as much as tribes in the Amazon or in India. The cultivation of minority ‘traditions’ by Beijing and Taipei has been balanced by modernity introduced to such minorities by or through central political control. However, this is not to say that these minorities lack agency: some of the impetus for change is internal, and they are able to select some ‘modernizing’ influences from outside while rejecting others. Perhaps it is especially in this rejection that their own ‘traditions’ are emphasized, often in revamped form. In any event, such considerations provide the background to the present collection of papers, which have all been revised for publication here. The editors would like to thank all the contributors for their work and their patience in responding to their various requests for revisions, Sue Jollow for language editing across the entire revision process, and Eveline Bingaman for language editing in the last round of revisions. The editors also thank the Center for Humanities and Social Sciences and the Institute of Anthropology, National Tsing Hua University, and the College of Hakka Studies, National Chiao Tung University, for the funding that made the seminar possible, and Berghahn Books for agreeing to publish the work. Parkin would also like to
xii
Preface
thank Wilkerson for the latter’s invitation to visit Tsing Hua as a Writer in Residence in the summer of 2006, as well as Tsing Hua University for funding a most enjoyable and fruitful trip. He would also like to thank Futuru Tsai for taking him on a visit to the Amis to witness their dancing. Wilkerson would like to thank Parkin for kindly agreeing to participate and to work with all involved with patience and persistence. Both Robert Parkin and James Wilkerson thank Eveline Bingaman for collaborating in the compilation of the index. Finally, the editors and contributors would like to dedicate this volume to the memory of one of their number, Li-Ju Hong, who sadly and tragically died while it was in preparation. She was especially kind and helpful to Parkin during his 2006 visit. She is dearly missed by all who took part. This volume is dedicated to her memory. James Wilkerson and Robert Parkin
Hsinchu and Oxford, July 2012
Introduction Modalities of Change: The Interface of Tradition and Modernity in East Asia James Wilkerson and Robert Parkin
Anthropology and Social Change Change is a factor in the experience of any society. This has always been the case, at least in recorded history, despite the tendency to think that there is something special about the modern age – for example, that change is now more rapid, more far reaching, more violent. However, it is not something social or cultural anthropologists have always been comfortable in dealing with. This was most marked in the British school of functionalism of Malinowski and Radcliffe-Brown because of their dismissal of the speculative history they saw in evolutionism (though Malinowski in particular was well aware of the impact of colonialism on the colonized). There is a contrast here with Durkheimian functionalism, which had remained partly evolutionist, and most of this group’s works incorporated at least some, often rather bare reflections on how the West was different from the rest (good examples are Robert Hertz’s works on death and right-handedness, 1960). Indeed, some famous texts from this milieu are clearly structured in evolutionary terms, such as Mauss’s works on the gift (1954) and the person (1985), and his joint work with Durkheim on classification (1963), though differently from the evolutionism of the British school (cf. Parkin 2001, Ch. 13). However, identifying and charting change is one thing, accounting for it another, and neither functionalist nor structuralist anthropology were particularly good at the latter theoretically, despite occasionally being willing to recognize the fact of change itself, to describe it and even to predict it, as both Malinowski and Lévi-Strauss foretold the disappearance of the primitive, and therefore of anthropology’s subject, as they saw it. Yet this is not just a matter of the blindness produced by theoretical fixations or an inherent inability of such anthropologists to account for change: rather, it is because of the often-concealed structuralism that informs even work that is
2
James Wilkerson and Robert Parkin
not explicitly structuralist. Nineteenth-century evolutionism itself, although ostensibly history, was in reality often closer to structure because of the fairly rigid stages, usually but not always of progress, through which the whole of humankind was supposed to move eventually; this touched Marxism as much as the non-revolutionary evolutionism of the British nineteenth-century school. For the theoretical structuralist as much as the British functionalist, change does not matter much because it cannot explain much, since at root we all share the same characteristics and propensities: in the case of structuralism, for example, we all allegedly share a tendency to think in terms of binary oppositions, or to stage rituals according to the same basic pattern, or to distinguish those we may or should marry from those we may not. Although Lévi-Strauss distinguished between hot and cold societies, that is, societies that respectively do and do not see their histories as different from themselves, this too could be seen as a structural difference as much as a historical one. In any case, his preferred form of narrative was always myth, where his structuralist method could be put to full use. Fundamentally, though, structuralism in whatever form raises the question of what precisely we mean by change. Under its influence (and in general terms, that predates Lévi-Strauss), anthropologists have often focused more on stressing the continuities over time in social forms, cultural practices, identities, and so on, than on, identifying mechanisms of change, reflecting the fact that they are not, in the last resort, historians. This often remains true, however much individual anthropologists may cooperate with historians, examine archives themselves, or collect oral histories and life stories. Even in such studies, the fact of change is often of less interest than how the past and the present are articulated within a single model or process. Seeking to go further – as often is the case in Europe, where the past is well documented – always threatens to reduce anthropology to a form of social history. But then again, this is more or less what Evans-Pritchard advocated in claiming – in opposition to Radcliffe-Brown, but under the probable influence of Marett, his earlier teacher – that anthropology was a form of historiography, not of science. However, other trends in anthropology have dealt extensively with change, initially especially among colonized or once-colonized peoples as a result of the impact of colonialism. This started with the anti-evolutionist Malinowski himself, though his disquisitions on this matter were always coloured by his generally successful attempts to obtain research funding from the Rockefeller Foundation. Another key early figure was his student, Lucy Mair (e.g. 1969), who wrote extensively on change in Africa, especially in the period of independence. Many of her essays interrogate the likelihood of tradition surviving post-colonial modernization trends and programmes. The Manchester School led by Max Gluckman, and the policy-oriented RhodesLivingstone Institute in what is now Zambia with which many of its members were associated, both distinguished themselves from orthodox functionalism by their greater concern with social process and with the vast changes that were taking place in African societies right under their noses as a result of the colonial experience. There were similar trends in other parts of the world like
Modalities of Change
3
the Americas, leading to Bohannon declaring already in 1967 that ‘One of the largest literatures in all anthropology is that devoted to acculturation and culture change’ (1967: xi). However, Bastide, in a slightly later work ([1971] 1973), saw this as an American intellectual and political obsession, as well as being a process that merely produced acculturated elites who would inevitably turn against the colonialism that had spawned them sooner or later, becoming ‘encysted’ within their domestic societies in the pursuit of their own interests rather than those of the formerly colonized population at large. This is the socalled ‘Marginal Man’ phenomenon, a good example being Abner Cohen’s work (1971, 1974) on acculturated creoles, the descendents of repatriated slaves, in Sierra Leone in the years before independence, whose Anglicized upbringing and origins, as well as their use of freemasonry as a form of organization, allowed them to distance themselves from the African ‘tribal’ population as handmaidens of the British colonialists. Bastide advocated instead a Marxist, anti-Cartesian approach to change focusing on the socioeconomic development of whole populations rather than the merely cultural assimilation of elites. However, Bastide also called ‘development sociology [sic] … merely the latest expression of evolutionary theory’ (1973: 158), since there is the same sense of certain peoples in the world having to ‘catch up’ with progress, sustained by the frequent blindness of the development worker to the adaptiveness of the supposedly ‘maladaptive’ practices of the developed (a particularly striking example is Fairhead and Leach 1997, on forest conservation in West Africa; also, for example, Nanda 1994 on the Bonda ‘tribe’ in Orissa, India). For many social anthropologists, accordingly, development has become just another social situation in which the dynamics between the developers and the developed can be studied, and not a prescription for ‘progress’ – itself a deeply problematic term. Social change is perhaps easier to deal with intellectually in Europe and other parts of the world with a literate tradition, ample historical records and a strong sense of history. Certainly there is no shortage of such studies, especially in Eastern Europe, which, having been classified as politically oppressed (and therefore as backward in its way), now risks being defined as endlessly struggling through a difficult economic transition that has privatized and deindustrialized state production, de-collectivized agriculture, slimmed down the welfare state, and produced unemployment or casualization on a large scale. This weighs especially heavily on the former industrial proletariat (Dunn 2001; Kideckel 2002), as well as on women (Pine 2001, 2002), who seem more affected by these changes than men in the region. However, as Kristmundsdottir (1999) has shown, this can also affect a West European society like Iceland, where the recent feminization of professions like teaching, ostensibly permitting women to enter the world of work in unprecedented numbers (including in politics), has actually caused them to fall in both prestige and salary levels. A focus on tradition and modernity and their interrelations brings its own conceptual and definitional problems. We should not forget that tradition is
4
James Wilkerson and Robert Parkin
not just ‘there’, but apt to be invented (Hobsbawm and Ranger 1988). In a sense this is always true, given the lability, even fragility of culture, which has come to be seen as an ongoing process rather than an over-reified ‘thing’, as well as the fact that it is produced and reproduced by living people in living societies. However, tradition is often felt to be invented backwards, as it were, in order to serve present-day agendas (discourses of nationalism are a good example of this). As for modernity, we need only recall Bruno Latour (1993) questioning whether anyone has ever really been modern. In any case, even if we accept that the modern does have meaning, it can be difficult if not impossible to draw the line between tradition and modernity: where does the latter really start? Yet there has also been a reaction against attempts to deconstruct change and history completely as concepts in anthropology and other social sciences, or to see culture, identity and social forms as created in a vacuum with regard to the past. One example here is John Peel (1989), who opposes some aspects of Abner Cohen’s work on West African ethnic and religious identities as representing what Peel calls ‘presentism’. Another is the Anthony Smith (1986) notion of perennialism, or the recognition that, although no identity is either primordial or essential, a number of identities have been around for an awfully long time, like the Greeks, Jews, Persians, Chinese and Japanese. People’s own perspectives on history are now more important in anthropology than history itself, and there has therefore been a tendency to evaluate how much continuity with the past is claimed in, for example, nationalist discourses, since these are not merely ideological fabrications, and some continuities always exist. Even in extreme cases, basic categories of thought, norms and values tend to survive the most far-reaching disruptions in people’s lives (e.g. colonialism, the Holocaust), but in more settled times too, change may well modify rather than undermine key beliefs. One excellent example of this is Edwin Ardener’s article, ‘Witchcraft, Economics and the Continuity of Belief’ (1970), in which he shows how, among the Bakweri of Cameroon, modernity was at first resisted because witchcraft beliefs ruled out demonstrations of wealth from agriculture or trade, though these were later embraced under the twin pressures of the colonial government and the more ‘acculturated’ Bakweri who had lived away from the area and acquired different attitudes to wealth. But although these new activities generated significant profits, these were used not for modern consumerism, but to buy more effective witchcraft medicine. Any work on modernity in anthropology is compelled to take ‘tradition’ into account, however that may be interpreted, and whoever does the interpreting. All work on witchcraft now seems to have to confront the impact of modernity on it while also showing the extent to which modernity has to compromise with traditional beliefs and modes of contestation marking the structural fissures in a community, with ostensibly modern courts often having to make allowance for such beliefs, and even sometimes acting as if they reflect reality (Comaroff and Comaroff 1999; Geschiere 1994; Green 2005; Strathern 1982). While, for example, Tanzania has retained colonial-era laws against witchcraft accusations
Modalities of Change
5
(Green 2005), in the Central African Republic it is witchcraft itself that is outlawed (Unreported World, Channel Four, 2011).
Change in East Asia East Asia is clearly a region that has known massive economic and social change in recent decades. Choosing a particular key date may be arbitrary when it comes to tracking social processes, but national unity and socialist revolution in China and Vietnam, as well as the alternatives of capitalism and quasidemocratic reform in Japan and Taiwan (though top-down rather than bottom-up processes, unlike largely in the West), have clearly done much and are still doing much to shape the way the region is today. These changes have altered social forms to some extent, such as the disruption to family organization through rural–urban migration, but less so cultural norms and values, whether in relation to kinship or to ritual and religion (where the latter is permitted by the state). As the chapters in this collection all show to a greater or lesser extent, even here, in this period, tradition has sharp elbows that continue to poke through the fabric of modernity. Indeed, in articulating the past and the present, tradition and modernity, however defined, these chapters all adopt a sharp ethnographic focus by building on earlier, but still quite recent work on economic and social change in the region (e.g. Garnant et al. 1996; Goodman 1997; Ikels 1996; Milwertz 1997; Oi 1999 – all works by non-anthropologists. Feuchtwang 2002 and Latham 2002 are two recent anthropological studies). Much of this change has, of course, been rapid, although the whole period since the collapse of the Chinese Empire in 1911 has seen far-reaching changes in China, from the rise of the nationalist republic, through the Japanese invasion and communist revolution, to the Cultural Revolution and the reaction to it, which eventually ushered in the contemporary and rather paradoxical period of capitalist economic growth promoted by a communist party that still dominates politically. The recent economic crisis of 2008 and after is expected to enhance China’s standing in the world rather than diminish it, for even though the global recession has affected China too, the world is looking to China to lead it out of recession, America is hoping it will continue to fund its national debt, and the IMF is grateful that it has agreed to bolster its credit lines. The last fifteen or so years in China especially, with its phenomenally rapid economic growth being followed by the current steep recession, reminds one of Hann’s edited volume, When History Accelerates (1994), which grew out of the collapse of communism in Eastern Europe in which China did not take part (at least as far as party dominance is concerned). Nor should we forget the other states represented here, namely Vietnam and Taiwan. The former can be described as a mainland China in miniature for its doi moi reformism, which has also led to quite rapid, though similarly fluctuating, economic change. The latter, by contrast, is long used to being a capitalist outpost in the Chinese cultural and
6
James Wilkerson and Robert Parkin
geographical sphere. For Taiwan, indeed, change has been more gradual, though no less fluctuating in terms of economic cycles, with democracy and civil society emerging belatedly out of Kuomintang dominance. One aspect that appears to characterize economic development in this region, as in neighbouring South-East Asia, is state sponsorship. Whether key economic ministries in Japan or South Korea, the communist party in mainland China and Vietnam, or the highly disciplined, even authoritarian statist regimes in Singapore and Malaysia, the Adam Smith principle of the state setting the conditions for growth but not intervening directly in economic activity and letting a mercantilist civil society do the work itself – basically a bottom-up approach – tends to be replaced here by top-down direction combined with the market-driven flexibility characteristic of capitalist systems. This in itself allows the contemporary People’s Republic of China to appear as just another state in the region, rather than emphasizing its distinctiveness because of its continued rule by a communist party. Added to this relative uniformity is a continuing if often patchy claim in many of these countries that the state will look after the basic needs of the whole population so long as the latter pursues economic activity, rather than expecting people to stand on their own feet as much as possible as in the West (America consistently, Europe post-Thatcher). Ong has described this implicit bargain between state and people very effectively for Singapore (1999), where is it perhaps most marked, but it can also be found elsewhere in the region, if not everywhere. Ong rightly derides attempts by some writers, exemplified by Huntington’s ‘clash of civilizations’ hypothesis (1996), that modern Asian states are perpetuating precolonial forms of oppression (or at best an excessive but still traditional communitarianism) because of allegedly irreconcilable cultural contrasts with genuine Western liberalism. Instead Ong sees them as new polities concerned as much with comprehensive welfare as with political oppression: not even the more authoritarian states such as Malaysia and Singapore are police states in the usual sense of the term, and the focus is increasingly placed on building a modern but Asian middle class to maintain achievements to date and build on them. This in itself suggests permitting enough freedom of thought to encourage innovation and lateral thinking in the interests of the overall society and economy, even where democracy per se may be a low priority. The main exception here, Japan, clearly owes its democracy to the American postwar occupation, and even this democracy has been characterized by the dominance of a single party since 1945, combined with continued veneration of the emperor. In brief, Ong sees in modern Asian states a series of alternative modernities – alternative especially to the West, but also to each other – rather than the continuity of Asian politico-cultural forms from colonialism to postcolonialism in a different guise, as writers like Huntington would have it. A comparison with Eastern Europe – similarly a region of rapid economic, political and social change – is instructive. Katherine Verdery (1996) has cautioned against the possibility that post-socialist societies in the region will simply become clones of the western half of the continent. Again we may be
Modalities of Change
7
faced with alternative modernities here, not least as this was socialism’s own view of itself. For Thelen (2011), however, the real issue in Eastern Europe is the failure of anthropology to capitalize on the perfect opportunity the region presented to study social change as it happened. The emphasis on the ‘social’ is critical here: change was seen as primarily economic, the stress being on the privatization or dismantling of socialist economic systems, practices and institutions, and changing property rights, especially in land through decollectivization; much less attention was paid to political changes (the emergence of multi-party democracy) or to social changes per se, which, in the work of figures such as Pine and Kideckel (discussed above), are seen almost as epiphenomenal on sudden economic disadvantage for women and the industrial proletariat respectively. Thelen links this tendency to the influence of Kornai’s neo-institutionalist theories of socialism (1992), in which she sees close parallels with rational choice theory (for Kornai, institutions are interpreted as being formed by the choices and interrelations of the actors within them, not as reified abstractions locatable nowhere). While there is some sense to Thelen’s criticisms, she does not mention work on post-socialist ethnic violence, which asks whether such violence is new, was foreshadowed under socialism itself or represents a revival of pre-socialist ‘primordialist’ hatreds. Nor does she cite Bornemann’s work on identity and its loss (1993) – an odd omission, given that both he and she worked in East Germany, where that loss is felt exceptionally keenly – nor work on religious revival in Eastern Europe (e.g. Hann 2010). Clearly there are also continuities between the socialist past and the postsocialist present, for example, in the use of networks to obtain jobs and other resources, or the blurring of public and private, of the official and the personal, in respect of precisely such networks, as well as land rights (Thelen ibid.; Dunn 2001). Once again one feels that structuralism and even functionalism are being introduced through the back door: people behave the same, whatever their exact political and economic circumstances or sociocultural forms and practices, and the danger of over-dichotomizing Eastern and Western Europe, socialism and capitalism, socialism and post-socialism, is ever present. The experience of East Asia may be different from that of Eastern Europe, not least because states in the former region have generally not undergone the same sort of far-reaching changes in the political system and institutions. The articles by Feuchtwang (2002) and Latham (2002) cited earlier, both from a book on post-socialism, the former on the fossilization of the Chinese Communist Party, the latter on the significance of the new consumerism, give some indication of the differences here. Both authors take as their baseline the obvious fact that China represents the irony of a politically authoritarian and monopolistic socialist party not merely permitting but encouraging a capitalist economy, an entrepreneurial spirit and, in effect – if not so deliberately as in Malaysia and Singapore – the emergence of a middle class, a bourgeoisie, in fact, despite this social group being the canonical socialist class enemy. In Feuchtwang’s words (2002: 196), ‘reform and economic growth required repudiation of socialism, but the continuity of the Communist Party required
8
James Wilkerson and Robert Parkin
its retention’. Feuchtwang then asks whether this makes the Communist Party in China a ‘remnant’ or ‘fossil’ – perhaps a ‘survival’ – since it exists as a factor of power in the country but has lost its moral compass and much of its legitimacy, as the cracks in its monopoly have steadily opened up. Like Steinmüller more recently (2011), Feuchtwang points to the irony, extending to cynicism, that marks popular attitudes to the party and its functionaries – which is not to say that it or they have lost their legitimacy completely. Both Feuchtwang and Latham see the party as having abandoned class struggle in favour of nationalist patriotism in which the country is encouraged to make progress, overtake the West and achieve prosperity, although ideologically this may also be represented as the Chinese and capitalist road to socialism. For Latham, who rejects crudely materialistic arguments that the new consumerism is a ‘social palliative’ that makes party rule tolerable, it is precisely the party’s focus on transition that gives it legitimacy today because it claims ultimate control over it. If that transition were to stop, it would signal the party having lost control and undermine its legitimacy as a necessary institution. As Feuchtwang points out, this also means that the party retains an evolutionist view of transition and progress drawn ultimately from its Marxist roots. Another irony identified by Feuchtwang is the position Mao has attained as a spiritual icon, with amulets and the like, offering protection much like a traditional ancestor, despite the Communist Party’s rejection of ‘superstition’ as part of its evolutionary trajectory. Yet the party also retains its dominance in other ways, not all of them blatantly repressive. While, as Latham also shows, there are cases where party men have been involved in the rehabilitation of traditional ancestor halls, temples and genealogies on their own initiative (which still may give the party legitimacy locally), Feuchtwang draws attention to the party’s often successful attempts to incorporate, at one level or another, whatever civil society organizations manage to emerge, as it struggles to retain its monopoly of power. Latham himself remarks on the Chinese consumers’ association being state-directed, as well as the party’s rejection of consumer choice where the media are concerned, seen most recently in attempts to control the internet. For Feuchtwang, on the other hand, many critiques of the party, even in Tiananmen Square in 1989, begin within a party tradition of revolutionary renewal in which bottom-up initiatives always have a chance of success. What, then, of differences between China and Eastern Europe – can we even speak of post-socialism in the former case? One obvious difference is that the break in Eastern Europe was more complete and far reaching: although many regional communist parties were reinvented as parliamentary socialists, they had to reconstitute themselves and compete as such in democratic elections; this did not happen in China, where the party and its monopolistic organization continued despite the change from Mao to Deng (whose marketoriented reforms, incidentally, anticipated those in Eastern Europe by more than a decade). For Feuchtwang, while the experience of socialism in Eastern Europe was of its gradual attenuation and loss of legitimacy until a crisis was
Modalities of Change
9
reached, again this did not happen in China, where Feuchtwang writes instead of its ‘saturation’ through, primarily, the Cultural Revolution. For Latham, the transition to the market is now over in Eastern Europe; in China it must continue, at least ideologically, to give the party legitimacy, as we have seen. For both Feuchtwang and Latham, however, the transition is still seen as fundamentally economic rather than social or (except marginally) political, thus further justifying Thelen’s critique of theories of post-socialism generally. For Feuchtwang the key trope is fairness, that is, people’s sense of what is acceptable as economic activity (socialist regulation, proper legal contracts, morally obligated reciprocity) and what is not (bribery, privileged and purely instrumental personal connections); for Latham it is consumerism, despite his refusal to connect it with party legitimacy. Latham’s argument also depends partly on a notion of deferred gratification: this, an aspect of Mao’s utopianism, is no longer necessary, thanks to reform and capitalism. Yet, as the papers in the present volume indicate, not all change is far reaching, nor, when it is, is it solely economic in kind. Thus alongside economic changes, we also find in this volume changes caused by politics, artistic creativity, cultural imports, social attitudes and popular resistance. And as we shall also see, changes in one area make possible, and may actually lead to, changes in others. Yet throughout, elements of tradition persist, being changed by, but also in their turn modifying, other influences. Thus although there may be a temptation to see tradition as indigenous, change as external, this would only lead us back to the functionalist weakness of viewing change as only external to a society. In fact, none of these states or societies are intrinsically static: they each have their own dynamics, being receptive to outside influences, yet selecting them and adopting them according to internal, self-generated imperatives. In that sense, tradition impacts on change as much as vice versa.
The Chapters Wang Ting-yu’s ‘The House, the State and Change’ explores the meaning of modernity for a village in north-eastern Sichuan Province that has shifted from a primary dependence on agriculture over to a dependence on village tourism and employment in a nearby town. Before the early 1950s and the changes imposed by the Chinese Communist Party, Zhoukeji village had served for centuries as an administrative centre for a local chiefdom (tusi). The rGyalrong Tibetans are Tibetan Buddhists, but speak a language that is classified as a Qiang language in the Tibeto-Burmese language subfamily. In the Qing dynasty era, because it was an administrative post, a trading centre and an agricultural village, Zhoukeji village included three groups of residents: rGyalrong Tibetans, who owned land and farmed; Chinese traders, who did not own land but lived off commerce; and Chinese who served in the local chiefdom administration. Then as now, named houses have been classified by language: the names for Tibetan houses use Tibetan language surnames; the
10
James Wilkerson and Robert Parkin
names for Chinese houses use Chinese language surnames. No two houses share the same surname. There are additionally unnamed houses, which are offshoots of and subordinate to named houses and are either eventually reabsorbed into a named house or recognized as a named house when a surname is available and the house has passed through three generations. Although the precise details are sketchy, named houses were a fundamental unit in the local sub-bureaucracy of the local chiefdom in the Qing dynasty era. Succession to the head of a named house required the approval of local officials who headed the sub-bureaucracy, and the house undoubtedly served as an administrative unit in taxation, policing, and perhaps the military. As Wang Ting-yu describes it, the house was never displaced as a basic social unit despite the upheavals and disruptions that followed the establishment of the People’s Republic of China. Despite land reform and collectivization, the reliance on agriculture continued across the first three decades of the People’s Republic of China until it was replaced during the post-reform eras by tourism and wage labour. Nowadays most who live within the village either work in the tourist industry or in wage labour in the village or a nearby market town. Also, outmigration of young people is sizable, and now the population of those still living within the village is aging. Parallel to these changes, the reform and postreform eras have included a revival in Buddhist ritual practices, especially those associated with the house. Nevertheless, considering the revival of Buddhist ritual practices within the context of the economic changes, Wang Ting-yu concludes that, even accepting that there is an important literature documenting Buddhist modernism elsewhere in the world, modernism in this particular Tibetan village is driven by secular economic change. Liu Biyun’s ‘From Kinship to State and Back Again’ focuses on Nzidabu, a Qiang village in north-east Sichuan as studied not long before the disastrous 2008 Sichuan earthquake levelled the village and whose subsequent rebuilding has totally altered its physical layout. The local Qiang language is one of several in the Qiang subdivision of the Tibeto-Burmese language subfamily. The village is far removed from the nearest market town, and agriculture remains the primary local means of livelihood. The coufang (literally, ‘people under the feet of one’s ancestors’) is the Sichuan Chinese term used to refer to a lineagestyle corporate grouping which still occupies an important place in Qiang social life, and whose history can be traced back through local genealogies, stone stele, and grave inscriptions across at least half a millennium. The coufang originally existed historically within the context of first a local chiefdom (tusi) and then a military colony (tun), where in both cases the coufang apparently fitted either directly or indirectly into the sub-bureaucracy. The subbureaucracy was of course swept away with the elimination of the military colony shortly after the establishment of the People’s Republic of China, and with it any role of the coufang in village administration. Although the ritual responsibilities of the coufang were also suppressed in the first decades of the People’s Republic of China, that ritual side quickly revived during the reform and post-reform eras.
Modalities of Change
11
Nowadays, the ritual activities of the coufang revolve around birth, marriage and death, and the practical demands of ritual work and the still firm ideological assumptions of human relatedness. First, in terms of birth and marriage, household succession is constrained by the presence within any single choufang of a limited number of houses, and so a house lacking a successor couple is filled – by uxorilocal marriage or other similar means – by couples that come from houses in that coufang that already have a successor couple. Second, in terms of death, the coufang is an important unit in the allocation of ritual obligations in the form of prescriptive gifts and labour sharing on ritual occasions; it also serves as the unit in the ownership of coufang cemeteries. Third, the coufang is highly flexible and adaptive as a whole, in either splitting into separate single coufang or merging several coufang to meet changing conditions in terms of exogamy and ritual obligations. Fourth, the coufang, as an extra-domestic group, makes use of a distinctive ideology (Heaven for males, Earth for females, and Seed for children) that prevents the sorts of controlled affinity that result in systematic intermarriage down the generations between designated coufang. Ho Zhaohua’s ‘Embroidery Speaks’ is concerned with the lowland Miao of Shidong village, Guizhou Province. During the Qing dynasty, Shidong village was a significant commercial enclave located along a major river trade route, where it served as a trading centre between the Miao, who lived beyond the reach of direct state rule, and the Han Chinese traders, who came to Shidong mainly for trade. The commercial enclave, which for much of the Qing dynasty was a region of ‘raw Miao’ that was beyond Qing dynasty administration, was and is still subdivided into separate territorial sub-groupings (Ho Zhaohua 2011). These territorial sub-groupings were and still remain endogamous. The constituent villages of the sub-groupings were and are exogamous. Marriage was and remains in classification, norm and practice between the children of classificatory cross cousins. Each territorial grouping is marked off from all other territorial groupings by differences in the design of women’s costumes. Differences in the social status of individual women is marked out by differences in the quality and quantity of their silver jewellery and needlework, which attest to the actual extent of and potential for family wealth manifest in its public display. Despite a history of low female literacy, use of Chinese characters, whether to represent spoken Chinese or Miao, is widespread. These words and other iconic images stitched into the female costumes link female costumes to Miao historical memory and cosmology, and provide accounts of the trials and tribulations of Miao life. Women’s costumes are integral in and even a measure of the importance of every Shidong social boundary, but still leave room for the exhibition of individual achievement. Today, the theme of exotic national minority culture as a marketable commodity is now a firmly fixed feature of Shidong Miao economic and social life. The slightly more than twelve thousand Miao residents of Shidong are visited annually by about fifty thousand tourists. Shidong’s landscape and architecture attract tourist attention and serve as backdrops to documentaries
12
James Wilkerson and Robert Parkin
and movies. Pieces of Miao female costumes or even entire costumes are widely available at shops nearby international hotels throughout major cities in southwest China. Taiwanese traders have created a regional, and even global, trade network in Miao female costumes. World-class museums certify their public value in their own purchases and displays. Shidong women’s costumes have become icons both nationally and internationally of the People’s Republic of China’s national minorities. Chien Mei-ling’s ‘Tensions between Romantic Love and Marriage’ describes institutionalized, competitive love-song performances among the upland Miao of Fangf Bil village in eastern Guizhou Province. The language of the Miao of Fangf Bil village belongs to the northern subgroup of the central Miao. As Chien Mei-ling describes these songfests, although institutionalized as a kind of ‘joking relationship’ associated with courting, these love-songs also express individual feelings and identities as a counterpoint to positive marital restrictions and ethnic identities. She describes these marital restrictions as a ‘Dravidian-style’ of classificatory bilateral cross-cousin marriage. Although villages are supposed to be endogamous, about half of all marriages were and still are elopements. Elopements generally occur between men and women from different villages but without otherwise violating the marital restrictions. In Chien Mei-ling’s depiction, the Fangf Bil Miao do with their love-songs what the Shidong Miao do with female costumes. That is, love-songs express a sense of the individual without undercutting social obligation. Some married people still take part in the courting songs even after marriage, and not necessarily with their own spouses. Still, expressions of individual choice in love-songs only surface within the frame of individuals pairing off in relationships that are in accordance with approved forms of affinity. Moreover, the themes given in the sung performances that Chien Mei-ling extensively quotes and translates cover explicit disquisitions on the subjects of individuality and social responsibility. Included are discussions of how both flirting and marriage are to be conventionally performed and culturally understood. Of particular interest are the Miao sung disquisitions on the ultimate subordination of a carefully framed individuality and emotional autonomy within an equally carefully framed life of social obligation. There is a sense of traditional individuality distinct from the modernity of the tourists and female clothing traders. Chang Kuei-min’s ‘Modalities of the One-Child Policy among Urban Migrants in China’ addresses the question of responses to the one-child policy among urban migrants in the People’s Republic of China. Although the subject is not that of national minorities, a large slice of the younger population of the national minorities have also joined this floating population; also, although the population control regulations they live under are broader than for others, they remain consistent with the spirit of coercive controls experienced by others in the People’s Republic of China. Regardless of the specifics of the backgrounds of the urban migrants, Chang Kuei-min sees a clash of tradition and modernity in so far as the new policy allows no leniency towards either traditional childbearing practices (for example, the focus on sons, and on several children)
Modalities of Change
13
or reproductive rights as generally understood by international human rights. Chang Kuei-min places special emphasis upon resistance to the control system in the hometown, where it is normally possible for strict enforcement of the one-child family, since it is urban migration that allows its evasion. Chang Kuei-min’s chapter thus balances a bi-local view that alternately reflects the perspectives of the hometown and the urban residence. However, she also emphasizes the broader notion of migrant workers as composing in aggregate a nationwide ‘floating’ population, with the obvious potential for providing a space for novel forms of modernity and individualism. These migrants can, if not always then at least frequently, evade state control over a woman’s reproductive future and the consequences of her past reproductive behaviour. For example, a different child is registered in each of the parent’s respective registration books. Alternatively, the incomes from urban wages are turned to good use in easing the way to legal status for illegal births. In these acts, it is possible to see the application of subtle performative manipulations in instrumental moments in confronting and overcoming the intended state punitive measures aimed at coercive enforcement of the one-child policy. This sense of modernity is anchored nowhere and yet everywhere; its individuals act as agents in pursuit of a now lost reproductive autonomy. The irony is that the one-child policy as applied to urban migrants places them in situations where it is possible to evade state restrictions, but, when evaded, place the parents in a situation where the processes of negotiating solutions produce the very autonomy of action outside state supervision and control that the one-child policy was explicitly intended to prevent. Eveline Bingaman’s ‘The “Culture” of World Cultural Heritage’ describes the experience of the residents of the Yunnanese town of Lijiang, and the various levels of government of the People’s Republic of China with UNESCO’s World Heritage Committee. Lijiang and its environs serve as home to a large population of Naxi speakers, whose language belongs to the Tibeto-Burmese language subfamily. The attractions of the architecture of the town, and the costumes and religious customs of the Naxi, have made what was once a quiet and economically ‘backward’ town into a major national and international tourist destination. Eveline Bingaman places special emphasis upon a conflict between the World Heritage Committee and the management of Lijiang as a world heritage site as being a problem of language. The World Heritage Committee’s desire to preserve traditional culture – incorporating relatively fixed and universalistic ideas of the value, stability, and yet also the fragility of culture – conflicted with the desire in Lijiang and among officials at various levels within the People’s Republic of China to capitalize on the town’s status as a world heritage site. A large proportion of the local population relocated themselves on their own initiative out of the ‘old town’ and into the ‘new town’, and now rent out their original homes in the old town to professional tourism providers. This explosion of tourism and commercialism eventually provoked in its turn the World Heritage Committee to issue what amounted to a ‘yellow card’, threatening to revoke the town’s status as a world heritage site. That is,
14
James Wilkerson and Robert Parkin
the ‘authenticity’ and ‘integrity’ – the key words for the World Heritage Committee in discussing ‘traditional culture’ – of the ‘traditional culture’ of the ‘old town’ is being undermined by the very activity – tourism – that is most likely to provide the funds for conservation. However, since the World Heritage Committee does not generally grant world heritage status to sites that cannot finance their own conservation, and, since this requires a strong local income from tourism, the inhabitants of Lijiang and the various government levels of the People’s Republic of China were caught in a vicious circle. Eveline Bingaman’s chapter thus shows an entirely different modality of modernity. Instead of the government of the People’s Republic of China attempting to wield physical and financial coercive control in human reproduction, it is the World Heritage Committee that is attempting to wield its own physical and financial coercive control over Lijiang. Still, the frame for the conflict turns on mutually exclusive understandings of ‘culture’. For the former, ‘culture’ is a resource to be mobilized for transforming the economic circumstances of the residents of Lijiang as well as, more generally, transforming the People’s Republic of China into a world economic – and political and military – superpower. For the latter, however, the purpose of the world heritage sites is to preserve traditional culture and their traditional communities. However, the terms ‘authenticity’ and ‘integrity’ – the key elements in the World Heritage Committee’s sense of traditional culture – translate poorly into Chinese and there has been little or no effort invested in effectively anchoring these terms in development policies and practices for Lijiang. That is, rather than words like ‘culture’ successfully performing some form of work with good reason to expect predictable results, the words fail in their appointed tasks. Instead, economic development in Lijiang through the tourist industry has led to the physical and cultural saturation of the ‘old town’ by a ‘floating’ population of tourists and outside entrepreneurs who profit from those tourists. Ultimately, Lijiang represents a modality of modernity as miscommunication and the individualism that realizes itself through the consumption of tradition. Futuru C.L. Tsai’s ‘Amis Hip-Hop’ illustrates an emphasis on an ‘ethnic minority’ community in Taiwan where villagers still perform for one another within the context of an Amis male age-grade system, and where village elder age-grades still act as judges for the younger age-grades. The Amis are an Austronesian-speaking people whose homeland is concentrated in the coastal region of south-east Taiwan. Beginning with the point that the age-grade system has historically always placed great emphasis upon performance, Futuru C.L. Tsai goes on to observe that this emphasis has shifted in recent decades from formal instruction in codes for proper behaviour to an emphasis upon humour and creativity as a style for meeting the challenges of modern lives. He also importantly documents that the seemingly peripheral role of women as an audience for men is much more proactive than casual observation might suggest, and that women are now moving directly into the actual performances. This is all done through the juxtaposition of a still vigorous Amis traditional dance and song culture alongside a new tradition of selective borrowing of
Modalities of Change
15
styles of dance and music originating from elsewhere in the world. These borrowings, in addition to Afro-American hip-hop, principally come from native North Americans, Pacific Islanders, and others in the Fourth World. For Futuru C.L. Tsai, such borrowings describe a situation where modernity is neither resisted nor opposed to tradition, but rather where ethnic and cultural ‘otherness’ is not only modern, it is intrinsic to this sort of modernity in both its traditions as well as its social changes. What is being communicated to the performers and their audiences is thus a style that challenges the dominant style of modernity by the dominant Han Taiwanese, and that even asserts that a Taiwanese modernity that excludes or ignores the Amis and their style of modernity would be neither Taiwanese nor modern. Li-Ju Hong’s ‘Contesting Memory’ records a simmering confrontation between historical memory and written history in the Paiwan village of Puleti. The Paiwan are also an Austronesian-speaking people whose homeland is in the mountainous region of south-west Taiwan. The Paiwan emphasize genealogical memory that is closely implicated with a hierarchical social structure. The traditional negotiated verbal genres sustained an internal hierarchy between aristocrats and others through the absolute sense of status given in myth (milimilingan) and status endogamy against the negotiable sense of status given in legend (taucikel), and the potential incremental advances or retreats in relative status played out along the edges of status endogamy. Like Eveline Bingaman’s chapter, Li-Ju Hong’s chapter treats the topic of modernity and social change as a problem of communication. Written history began its invasion with ethnohistorical descriptions authored by Japanese researchers during the Japanese colonial era, but which apparently only began to intrude back into Puleti in recent decades since the end of KMT singleparty rule. Since then, one Paiwan and several other Taiwanese began to come up with their own written versions of Puleti village history. In Li-Ju Hong’s account, it appears that these histories assume national literate political ideologies more than strictly oral indigenous accounts when it comes to what is and is not an ethnic group. As these written sources have begun to be pulled into oral performances at marriage negotiations, written communication poses a new challenge to the process of oral negotiation and renegotiation of Puleti hierarchy. Although it is impossible to predict how written communication will eventually become a part of this village constitution, in the instance of Li-Ju Hong’s interlocutor, Muni, it appears that members of the community are now attempting to achieve for their communities in writing the sorts of negotiated histories heretofore achieved only verbally. Somewhat different from Lijiang, the Paiwan of Puleti are walling out what the villagers perceive as the corrosive potential of modernity and its individualisms. Yayoi Mitsuda’s ‘Ethnicity as Strategy’ describes Taiwan’s numerically smallest officially recognized Austronesian-speaking people, the Thao. The Thao are located in a single village next to Sun Moon Lake in a valley in central Taiwan. They now have a population of about five hundred, although the exact numbers have in recent years dramatically increased with changing
16
James Wilkerson and Robert Parkin
fashions in official identification as ethnic Thao. From their earliest investigations, Japanese ethnographers had difficulty in agreeing on the classification of the Thao. This lack of consensus continued over the decades of Nationalist Party single-party rule. As Yayoi Mitsuda describes it, rather than just recounting this history of classification as a now dated debate between academics, disaster and politics were importantly at work in the recent reclassification. As for the disaster, there was a 7.6 Richter-scale earthquake in central Taiwan on 21 September 1999, whose epicentre was located a mere 12.5 km west of the Thao community in Sun Moon Lake, central Taiwan. The media gave special coverage to the plight of the Thao, linking earthquake recovery efforts to their ethnic survival. On the one hand, there was considerable media coverage of the Thao ulalaluan (a bamboo basket whose contents represent patrilineal ancestors for each member family), used to mark the boundary between who was and was not Thao in earthquake recovery efforts, and to shinshi (female spirit mediums), who came to represent the ethnic distinctiveness of Thao religious ritual. Both the ulalaluan and shinshi are historically prominent in village and domestic ritual. On the other hand, however, the Thao have widely adopted a broad range of external religious traditions, and the Thao language is no longer spoken by their young people. In fact, the Thao are considered by the academic community to be highly assimilated with the surrounding Taiwanese who had long ago become the majority population in Ito Thao. As for the politics, the opposition Democratic Progressive Party seized on the survival of the Thao for their own purposes. Supporting the needs and interests of the Thao became a barometer for political commitment to the value of all ethnic groups, and especially so for those for whom the survival of their distinctive way of life was at special risk of disappearing. For their part, and notwithstanding internal discord, the Thao became adept at presenting and even marketing their own distinctive way of life as a value in and of itself. Mandated with the responsibility to preserve and promote traditional Thao culture and indigenous language, a government-sponsored research and empowerment team ‘was reorganized into the Thao Tribal Cultural Development Association’. The association took control of all formal interactions with the media, government, and other outside organizations, and actively campaigned for official recognition of the Thao as the tenth indigenous ethnic group. Not long after the March 2001 presidential election, when single-party rule by the Nationalist Party was broken by the election of Chen Shui-bien, the Thao were the first ‘new’ indigenous group to be officially recognized since the Japanese colonial era. N. Jenny Hsu’s ‘On the “Third Morning”’ focuses on change and continuity among the Nung of northern Vietnam as manifest in the life-cycle rite ‘Third Morning’, normally held on the third day after the birth of a child in the household of its father. The Nung language is classed with the Central Thai languages, which in turns belongs to the Tai-Kadai language family. The Nung of Dai An commune migrated from Longan Chieftaincy Prefecture (Longying
Modalities of Change
17
tuzhou) in Guangxi Province, China, some two centuries ago during the Qing dynasty era. The descendants of the shared ancestors who remained in China belong to the Zhuang national minority, the largest national minority in the People’s Republic of China. Although there are noteworthy differences, considerable continuities are visible between the Nung and Zhuang, and especially so in religion and marriage. The house itself is a ritual space containing significant spirits, some of whom are present on ‘Third Morning’, and ritual life generally is centred on kin and ancestors. On the one hand, the ‘Third Morning’ ritual and other related religious activities are still maintained as a charter for a distinctively Nung identity and its continuity with the past. In the ‘Third Morning’ rite, the Nung have maintained a strong continuity with past practices in the face of colonialism, war and, more recently, the Doi Moi reforms. Especially since the Doi Moi reforms, the Nung of Dai An commune have entered an intensive phase of economic transformation. Nevertheless, except for minor technical changes in the objects used and the room where spirits are kept, the effects of social change are limited in the ‘Third Morning’ ritual. These technical changes include the scheduling of the ritual performances and the location of the domestic shrines. The schedules of performances have altered to suit office work schedules, while government imposed changes in domestic architecture has resulted in the relocation of domestic shrines. The above social changes are of little moment for the ‘Third Morning’ ritual. It is still performed for the first-born of every family. In the textual, eventual, and social emergent qualities of performance, the keying of the footing of the child as a participant is to his or her fate and the implications of that for other family members. Among the Nung of Dai An commune, children are under the care of me mu or va mu, who controls the births of children and then their individual fates up to the age of fifteen. Mei mu is the goddess of flowers (huawang, in Chinese). At the beginning of the performance as event, the wife-givers have the special responsibility in the Third Morning ritual for ‘bringing the basket’ (au lam). The basket contains the sacrifices to be offered to the goddess of flowers as the central part of the ‘Third Morning’ ritual. What is expected to emerge from the sacrifice to the flower goddess in particular is that she will care for the fate of the newborn child up to fifteen years of age. Since the wife-givers provide – or potentially withhold – the sacrifice, the well-being of a child is in the hands of the natal family of the child’s mother. That is, the well-being of a first-born is keyed through ‘bringing the basket’ to the natal family of the child’s mother. However, another purpose of the ‘Third Morning’ ritual involves the emergent quality of text that takes place at the end the sacrifice, when the family and extended kin sit down to a banquet. The given names for all the members of the family are all changed following a principle of teknonymy so that each name includes as one of its two elements the given name of the first-born. Additionally, since the second element of the given name is a kin term that refers to the relationship between the child and the speaker, each member of the household moves terminologically
18
James Wilkerson and Robert Parkin
up a generation (for example, ‘mother of so-and-so’). Domestic reproduction is thus formally acknowledged to depend upon women and affinity. Finally, in his Afterword, James Wilkerson considers how the ethnographic chapters in this volume may be informed by the idiom of performance (in ritual, but also, for example, in labour migration, marriage, embroidery or ethnic group organization) and the notion of the ‘footing’, drawn from Erving Goffman, which positions social actors in relation to such performances and validates their right to take part. Wilkerson suggests that there is a continuum in performance between the conservative and the innovative, and he uses this to compare both the performances described in these chapters and their wider ethnographic contexts. One key distinction here is certainly that between authoritarian regimes in the People’s Republic of China and Vietnam, concerned to validate their own power rather than offer their populations the level of autonomy that would challenge that power, and Taiwan, now a more democratic state that can accommodate innovation in performance without detriment to its own existence. Basically, though, the question for Wilkerson is the impact of individual performances: do they leave society the same at the end of the performance as at the beginning, or do they produce change within it? Clearly this dichotomy can be related to the more conventional ones – between tradition and change, or tradition and modernity. Together, therefore, these chapters build on recent studies on both southeast China (Harrell 2001) and on Taiwan (Cauquelin 2004; Faure 2001). There are clearly some common themes here. One is cultural performance and modes of expression, whether in embroidery (Ho Zhaohua), love-songs (Chien Meiling), dancing (Futuru Tsai) or ritual (N. Jenny Hsu). Many of these traditional forms are being ‘modernized’, occasionally for narrowly practical reasons, but mostly because their practitioners are consciously reaching out to the world through their art, and reacting to – both absorbing and (re-)creating – global flows of culture. Another theme is kinship, whether in relation to descent and the house (Wang Ting-yu, Liu Biyun, Li-Ju Hong), affinal alliance (Chien Mei-ling, N. Jenny Hsu), or the one-child policy and family policy more generally (Chang Kuei-min). These studies can be seen as building, in their various ways, on the study of kinship and marriage in China from Marcel Granet via Maurice Freedman to Charles Stafford. A third theme is minorities and identity, whether in south-east China (Wang Ting-yu on Tibetans, Ho Zhaohua and Chien Mei-ling on the Miao, Liu Biyun on the Qiang), Taiwan (Futuru Tsai on the Amis, Li-Ju Hong on the Paiwan, Yayoi Mitsuda on the Thao), or Vietnam (N. Jenny Hsu on the Nung). Throughout too there are concerns with the authenticity and preservation of tradition in the modern era, highlighted especially in Eveline Bingaman’s chapter. All in all, the volume draws attention to the contingency of views of modernity, which is often imagined rather than realized, and is often imagined as a challenge, even a threat, rather than an opportunity.
Modalities of Change
19
References Ardener, Edwin. 1970. ‘Witchcraft, Economics and the Continuity of Belief ’, in Mary Douglas (ed.), Witchcraft Confessions and Accusations. London: Tavistock Press. Bastide, Roger. [1971] 1973. Applied Anthropology. London: Croom Helm. Bohannon, Paul. 1967. ‘Introduction’, in Paul Bohannon and Fred Plog (eds), Beyond the Frontier: Process and Cultural Change. Garden City: The Natural History Press. Bornemann, John. 1993. ‘Uniting the German Nation: Law, Narrative, and Historicity’, American Ethnologist 20(2): 288–311. Cauquelin, Josiane. 2004. The Aborigines of Taiwan: The Puyuma from Headhunting to the Modern World. London and New York: Routledge Curzon. Cohen, Abner. 1971. ‘The Politics of Ritual Secrecy’, Man (n.s.) 6: 427–48. ———. 1974. Two-Dimensional Man: An Essay on the Anthropology of Power and Symbolism in a Complex Society. London: Routledge & Kegan Paul. Comaroff, Jean, and John L. Comaroff. 1999. ‘Occult Economies and the Violence of Abstraction: Notes from the South African Postcolony’, American Ethnologist 26: 279–303. Dunn, Elizabeth. 2001. ‘Carrots, Class and Capitalism’, in Michał Buchowski et al. (eds), Poland beyond Communism: Transition in Critical Perspective. Freibourg: University Press. Durkheim, Emile, and Marcel Mauss. 1963. Primitive Classification. London: Cohen & West. Fairhead, James, and Melissa Leach. 1997. ‘Webs of Power and the Construction of Environmental Policy Problems: Forest Loss in Guinea’, in R.D. Grillo and R.L. Stirrat (eds), Discourses of Development: Anthropological Perspectives. Oxford and New York: Berg. Faure, David (ed.). 2001. In Search of the Hunters and their Tribe: Studies in the History and Culture of the Taiwan Indigenous People. Taipei: Shung Ye Museum of Formosan Aborigines. Feuchtwang, Stephan. 2002. ‘Remnants of Revolution in China’, in C.M. Hann (ed.), Postsocialism: Ideals, Ideologies and Practices in Eurasia. Cambridge: Cambridge University Press. Garnant, Ross, Guo Shutian and Ma Guonan (eds). 1996. The Third Revolution in the Chinese Countryside. Cambridge: Cambridge University Press. Geschiere, Peter (with Cyprian Fisiy). 1994. ‘Domesticating Personal Violence: Witchcraft, Courts and Confessions in Cameroon’, Africa 64: 323–41. Goodman, David S.G. 1997. China’s Provinces in Reform: Class, Community and Political Culture. London and New York: Routledge. Green, Maia. 2005. ‘Discourses on Inequality: Poverty, Public Bads and Entrenching Witchcraft in Post-Adjustment Tanzania’, Anthropological Theory 5: 247–66. Hann, C.M. (ed.). 1994. When History Accelerates: Essays on Rapid Social Change, Complexity and Creativity. London and Atlantic Highlands (NJ): The Athlone Press. ——— (ed.). 2010. Religion, Identity, Postsocialism: The Halle Focus Group 2003–2010. Halle: Max Planck Institute for Social Anthropology. Harrell, Stevan. 2001. Ways of Being Ethnic in Southwest China. Seattle and London: University of Washington Press. Hertz, Robert. 1960. Death and the Right Hand. London: Cohen & West. Hobsbawm, Eric, and Terence Ranger. 1988. The Invention of Tradition. Cambridge: Cambridge University Press. Ho Zhaohua. 2011. ‘Gifts to Dye For: Cloth and Person among Shidong Miao in Guizhou Province’. Hsinchu: National Tsing Hua University (Ph.D. Dissertation).
20
James Wilkerson and Robert Parkin
Huntington, Samuel P. 1996. The Clash of Civilizations and the Remaking of World Order. New York: Simon & Schuster. Ikels, Charlotte. 1996. The Return of the God of Wealth: The Transition to a Market Economy in Urban China. Stanford: Stanford University Press. Kideckel, David A. 2002. ‘The Unmaking of an East-Central European Working Class’, in C.M. Hann (ed.), Postsocialism: Ideals, Ideologies and Practices in Eurasia. Cambridge: Cambridge University Press. Kornai, Janos. 1992. The Socialist System: The Political Economy of Communism. Oxford: Clarendon Press. Kristmundsdottir, Sigridur Duna. 1999. ‘“Father did not answer that question”: Power, Gender and Globalisation in Europe’, in Angela Cheater (ed.), The Anthropology of Power. London and New York: Routledge. Latham, Kevin. 2002. ‘Rethinking Chinese Consumption: Social Palliatives and the Rhetorics of Transition in Postsocialist China’, in C.M. Hann (ed.), Postsocialism: Ideals, Ideologies and Practices in Eurasia. Cambridge: Cambridge University Press. Latour, Bruno. 1993. We Have Never Been Modern. London: Harvester Wheatsheaf. Mair, Lucy. 1969. Anthropology and Social Change. London: Athlone Press. Mauss, Marcel. 1954. The Gift: Forms and Functions of Exchange in Primitive Societies. London: Routledge & Kegan Paul. ———. 1985. ‘A Category of the Human Mind: The Notion of Person, the Notion of Self ’, in Michael Carrithers, Steven Collins and Steven Lukes (eds), The Category of the Person: Anthropology, Philosophy, History. Cambridge: Cambridge University Press. Milwertz, Cecilia Nathansen. 1997. Accepting Population Control: Urban Chinese Women and the One-Child Family Policy. Richmond (Surrey, U.K.): Curzon. Nanda, Bikram Narayan. 1994. Contours of Continuity and Change: The Story of the Bonda Highlanders. New Delhi: Sage Publications. Oi, Jean C. 1999. Rural China Takes Off: Institutional Foundations of Economic Reform. Berkeley: University of California Press. Ong, Aihwa. 1999. ‘Clash of Civilizations or Asian Liberalism? An Anthropology of the State and Citizenship’, in Henrietta Moore (ed.), Anthropological Theory Today. Cambridge: Polity Press. Parkin, Robert. 2001. ‘Durkheimian Evolutionism in the Work of Marcel Mauss’, in his Perilous Transactions: Papers in General and Indian Anthropology. Bhubaneswar: Sikshasandhan. Peel, John. 1989. ‘The Cultural Work of Yoruba Ethnogenesis’, in Elizabeth Tonkin, Maryon McDonald and Malcolm Chapman (eds), History and Ethnicity. London and New York: Routledge (ASA Monographs, 27). Pine, Frances. 2001. ‘Who Better than your Mother? Some Problems with Gender in Rural Poland’, in Haldis Haukanes (ed.), Women after Communism. Ideal Images and Real Lives. Bergen: Institute of Gender Studies, University of Bergen. Pine, Frances. 2002. ‘Retreat to the Household? Gendered Domains in Post-Socialist Poland’, in Chris Hann (ed.), Postsocialism: Ideals, Ideologies and Practices in Eurasia. Cambridge: Cambridge University Press. Smith, Anthony. 1986. The Ethnic Origin of Nations. Oxford: Blackwell. Steinmüller, Hans. 2011. ‘The State of Irony in China’, Critique of Anthropology 31(1): 21–42. Strathern, Andrew. 1982. ‘Witchcraft, Greed, Cannibalism and Death: Some Related Themes from the New Guinea Highlands’, in Maurice Bloch and Jonathan Parry (eds), Death and the Regeneration of Life. Cambridge: Cambridge University Press. Thelen, Tatjana. 2011. ‘Shortage, Fuzzy Property and Other Dead Ends in the Anthropological Analysis of (Post)Socialism’, Critique of Anthropology 31(1): 43–62. Verdery, Katherine. 1996. What Was Socialism, and What Comes Next? Princeton: Princeton University Press.
Chapter 1 The House, the State and Change: The Modernity of Sichuan rGyalrong Tibetans Wang Ting-yu
Introduction Modernity is both global and plural, now as much a phenomenon of the countryside and indigenous peoples as it is of urban residents. In Africa, South Asia and South America, for example, religion is an important conduit through which indigenous people adopt modernity (Comaroff and Comaroff 1993; Kapferer 2002; Taussig 1980). For other indigenous peoples, secular, everyday life experiences provide such a conduit. This chapter explores modernity as experienced by indigenous Tibetans in a rural community in Sichuan Province in the People’s Republic of China. Here secular experience is the means through which modernity is realized, and the state plays a major role in organizing such experiences. The first image when we think about Tibetans may be of Tibetan Buddhism, which has become the main feature of Tibetan culture. When the Tibetan people face changes in their culture, Tibetan Buddhism also changes. As Melvyn C. Goldstein and Matthew T. Kapstein showed in their book Buddhism in Contemporary Tibet (1998), sometimes Tibetan Buddhism is able to survive or perhaps revive after change, but sometimes it does not. In fact, they were unable to uncover any positive relationship between Tibetan Buddhism and social change. However, we can still try to understand how the Tibetan people face changes by examining their secular lives. My discussion focuses on ‘the house’, the basic unit in both domestic and community life. Historically the house has played a central role in both domestic and community social reproduction, but it is now an aspect of life that is seriously being challenged by contact with the modern world. The house as a social unit has long been a subject of anthropological interest (see Lévi-Strauss [1949] 1969, 1987, [1983] 1994; Carsten and Hugh-Jones 1995; Keane 1995).
22
Wang Ting-yu
This chapter looks both at the material aspects of houses – Tibetan houses in Sichuan – and at how spaces are used and their symbolic meanings. However, greater weight is given to the historical aspects, particularly the effects of social changes accompanying the economic liberalization of China’s former socialist command economy. The chapter is divided into four sections. The first describes my fieldwork site, Zokjibu, a Tibetan-speaking community in Sichuan. The second outlines the place of house names in social life in the village and region, both historically and more recently. This outline includes their role in the bureaucratic and subbureaucratic organization of the local chiefdoms or tusi, which dated from the Ming and Qing dynasties and only came to an end in the 1950s. The third section describes a typical Tibetan house, and both the continuities and transformations in its material and symbolic aspects which have occurred since the time of local chiefly rule through to the various major periods under Chinese Communist Party rule. The final section outlines the major social changes that have taken place in the village and to its houses and house names as consequences of state-introduced changes to the local agriculturally based economy leading to Zokjibu’s dependence upon the wider market economy.
Zokjibu Village Maerkang County is located in the north-western corner of Sichuan Province, along the south-eastern rim of the Qinghai-Tibetan Plateau. Maerkang is home to speakers of Han Chinese, Qiang (a Tibeto-Burman language) and rGyalrong Tibetan. This part of Sichuan is known in Chinese as Situ, a reference to ‘The Four Tusi’. The term tusi encompassed both local chiefs and their local chiefdoms. The tusi system was the primary means through which the Chinese state ruled local minority populations across south-west China from the Ming dynasty, in some places continuing right up to the beginning of communist rule. The four tusi originally composing Situ were Zhuokeji, Soumo, Songgang and Dangba. The main road connecting Sichuan with Tibet runs through this region and was an important supply route during the Jinchuan War between the local Tibetans and the Qing state from 1747 to 1776 (Dai 2001). It was also important for trade and immigration. The local Tibetan population was constrained in working as traders by their hereditary service obligations to the tusi, the local chiefs. At the same time, the tusi system resulted in the local Tibetan population becoming more closely integrated into the Chinese civil administration. Over the last several hundred years of Chinese imperial rule all these factors worked together, with the significant effect of encouraging Han Chinese immigration into the region for purposes of trade. The fact that opium was a legal crop in the region further encouraged this, and thus also provided the local Tibetan chiefs with literate Chinese government functionaries.
The House, the State and Change
23
The Zhuokeji tusi covered about half of today’s Maerkang County, a considerable part of which now comprises Zhuokeji township. The township consists of three administrative villages that each administer a number of smaller natural villages: (1) Charmi, named after a nearby mountain, and (2) Larjok, named after the adjacent river, which each have three villages within their jurisdiction; while (3) Siso, named after a nearby ravine, has two: Siso village and Zhuokeji (Zokjibu) village. I use the rGyalrong Tibetan name, Zokjibu, to distinguish Zhuokeji village from any other local area sharing the ‘Zhuokeji’ element in its name. Zokjibu is located at the confluence of the Somo and Larjok rivers (the Larjok River flows into the Somo River), the former flowing westward, the latter to the north. The name ‘Zokjibu’ consists of zokji (highest ranking), derived from the nearby former official residence of the Zhuokeji local chief, located on the east bank of the Larjok River, and bu (village). On the eve of the establishment of the People’s Republic of China in 1949, the residential area of Zokjibu consisted of only about twenty households concentrated on the west bank of the Larjok River, across from the official residence of the Zhuokeji tusi. This location near the official residence resulted in a number of Han Chinese coming to reside in the village. About one-third of its population consisted of permanent Tibetan residents who had house buildings, house names and land in Zokjibu. Another third were permanent resident Chinese-speakers, whose ancestors settled down in Zokjibu during the Qing dynasty and who also had house buildings, house names and land in the village. The final third were temporary residents serving the local chief, who also had house names there, but who had house buildings and land in other villages. Zokjibu village now has about seventy households and has expanded beyond its original nucleated residential area. Fifty-one households are located in the original residential area, but there are also two households in an apartment building for a nearby power plant, four located next to the official residence of the tusi, five along the north bank of the Somo River, and finally eight at the foot of Charmi Mountain. Although the number of households has more than tripled since communist rule began, migration out of the village has been a major factor affecting village demography. There are a number of aspects to the mobility of the local population. Traditional marriages across a wide area and between different language communities have occasioned a definite degree of population mobility from well before the time of communist rule. More recently, many children and teenagers have been going to live outside the village (for example, to Chengdu or Jiuzhaigou Valley) so as to be able to attend secondary school, and about half of all residents between twenty and fifty years of age have now relocated for work-related reasons. The result of this has been an aging of the remaining resident population. There are three major religious sites in Zokjibu: the dandarling gompa, associated with the Nyingma school of Tibetan Buddhism; the Chank’skiyu
24
Wang Ting-yu
community altar, where people pray to the local mountain deity; and the graveyard, where local resident Tibetans and Chinese both bury and commemorate their dead.
Gompa The Tibetan Buddhist gompa, called dandarling, is located south-west of the village on the west bank of the Larjok River. There is currently no resident lama. Prior to the abolition of the local chiefdoms in 1951, eight houses southwest of the gompa provided accommodation for the lamas serving as ritual specialists.
Community Altar A single village altar to the local mountain deity, Chank’skiyu, is located on the east bank of the Larjok River, halfway up Mount rGyazu, south-east of the tusi’s official residence. The altar was originally located at the summit of the mountain but was relocated twenty years ago, closer to the official residence. Both the original and the more recent sites overlook the whole village, a requirement for an altar to a mountain deity. Villagers visit the altar to pray and make offerings each New Year and on other occasions to seek good fortune or avoid misfortune.
Graveyard In the past, burials were normally conducted at a single main graveyard on a mountainside north-west of the village beside the Somo River. The site was known locally as darndasda (graveyard). Zokjibu residents have always buried their dead on their own land after funerals. Funerals are conducted in a manner partly Tibetan, partly Han Chinese.1 Graves in Zokjibu usually face north, towards Mount Charmi, or very occasionally east. Until ten years ago, there was a prohibition on burials on the east bank of the Larjok River, the side of the official residence. Graves are seldom visited except on Chinese Tomb-sweeping Day and Chinese New Year, when the graves of both agnatic and affinal ancestors are visited, food is offered to the ancestors by putting it into the fire, and wine is offered by pouring it around the fire. Paper money is burnt at the same time. The rites conclude with a communal meal shared by the living descendants and the spirits of the dead.
The House, the State and Change
25
House Names The system of house names shows how Zokjibu was organized, with the house as its central social unit, which was itself historically organized by the Zhuokeji tusi. There are two sorts of houses: those that are named and those that are not. This section only discusses named houses; unnamed houses are subordinate to named ones. There are two categories of house name in Zokjibu: Tibetan, and Han Chinese. A house name is a marker which differentiates the members of different houses, although house-name conventions are changing in ways discussed further below. An increasing number of houses lack their own distinctive house names, and the relationship between houses and house names is therefore changing: a house name in a community does not need to be exclusive to a single house. There are fourteen traditional Tibetan house names. Villagers usually address each other only by their given name, but the various Tibetan given names are usually heard frequently because they will have been taken from Buddhist texts by a lama. So if the villagers want to refer to a specific person who is member of ‘X’ house, they will use the house name. The house name for Zokjibu Tibetans encompasses the house’s various buildings, land and, in the past, taxes and labour service owed to the tusi. To inherit a house name is at the same time to inherit everything just mentioned. Succession to a named house is accompanied by the division of the family of that house. One part of the family will inherit the named house, and the rest of the family will inherit or find another named house. The effort involved in inheriting a named house and finding new houses for other family members often delays family division. Inheritance of a house occurred in one of three ways: a child of a deceased house head could (1) inherit the original house; (2) marry into another house; or (3) inherit a house that otherwise lacked an heir. For example, two generations ago the rGyazu family had two brothers: one inherited the original rGyazu house, while the other married into the Jodig’si house. The brother who had inherited the rGyazu house had two daughters and one son. One daughter inherited the rGyazu house, the other daughter married into the Barjing house, and the son inherited the Chinie house, which would have had no heir. Any inheritance arrangement of houses formerly had to be approved by the tusi. Villagers report that in the past the tusi controlled both the land and house names. Without a house name, there was no access to land to farm for a living. Tusi had the power both to give the house name to any family in Zokjibu, and also to take its house name away. The tusi needed to control the numbers of houses, as well as those who worked and fought for him, and tax payments. The tusi could not let the numbers of houses increase so much that they could not be controlled, nor let a house die out through having no heir.
26
Wang Ting-yu
In Zokjibu, Chinese surnames were also attached to individual houses, just like the Tibetan house names. Each house had its own name and house building, and also gave some duty to the tusi. The inheritance arrangement or division of these houses also needed the approval of the tusi. The main difference between the houses with a Tibetan house name and those with Han surnames as a house name was that the latter usually did not have their own land; the people from them worked as silversmiths, blacksmiths, dressmakers, merchants, and so on. After 1951 the tusi declined, and their powers over house building, house names, land and tax completely disappeared. Today the people in Zokjibu usually hold the houses which are more traditional, or older, in higher regard, because they represent the ideal image of a Zokjibu house, one whose house name, land and house building were all granted by the tusi. More and more families have begun to separate from their natal houses and now do not even have house names for their new houses. This does not fit the ideal of a Zokjibu house. Only the fourteen houses that have a Tibetan house name and the six houses that have a Han house name enjoy a higher reputation and are considered to be examples of the ideal traditional house. Houses that split away from their original house need at least three generations to prove that they retain a good relationship with the latter and have a stable financial situation: if so, their reputation will increase, and they may finally come to be seen as ideal houses. The K’jidobu family is an exception to the usual rules for the inheritance of named houses. The senior brother and sister of this family decided to leave their original house and build two new houses. The original house was not divided, nor was a new house created. Community residents now refer to the house built by the sister as d’mi k’jidobu, meaning ‘the female K’jidobu house’, and to that built by the younger brother as d’za k’jidobu, meaning ‘the male K’jidobu house’. These two houses share the same house name and the same reputation: they are both considered as older houses. A person’s given name usually corresponds in type with the type of their house name. Someone belonging to a house with a Tibetan house name will be given a Tibetan name. This naming rule also applies to people belonging to the houses that have a Han house name. Traditionally, the people of Zokjibu could have both a Tibetan given name and a Han given name, but never use them at the same time. People in Zokjibu always choose to use the given name which corresponds to their house name.2 Those belonging to the houses with Tibetan or Han house names are not entirely separate from each other. Intermarriage and marriage outside Zokjibu are both complicating factors, which can result in a co-mingling of the Han Chinese and Tibetan naming practices. Additionally nowadays, both Tibetan and Han given names are sometimes selected for educational purposes and good fortune. Modernity too is encouraging the mixing of the two naming practices. The use of Han given names has increased dramatically in recent times, with
The House, the State and Change
27
improved opportunities for education and work outside the area for a large proportion of the community. Nowadays, there is even a movement in the opposite direction: people who historically have had a Han house name and given name are assuming Tibetan names to express their ‘Tibetan identity’ more accurately.3 It can be seen that some aspects of the house name and given naming practices are open to change, while others are not. A house name represents the relationship between the people and their house, and this still exists in the everyday lives of people in Zokjibu. But marriage, education and working outside have created the variety of given names for them. This is therefore an instance of hybridity that Zokjibu people can change as necessary. Nowadays the number of new houses is steadily increasing, but there has been little or no increase in the number of house names, especially Tibetan house names. Nonetheless, Zokjibu people use their social values, the reputation of a house, a new house’s relationship with other houses and their financial circumstances to recognize the new houses. This is one way in which they are facing the changes brought about by modernity.
Material Aspects of the House: ka, sjangku and rahbo The house is the basic domestic and community social unit for Tibetans in Zokjibu. For ease of discussion, a distinction will be made between the material and religious aspects of the house, although it will quickly become clear that these are two aspects of a single indivisible whole. The prototype of a house building in Zokjibu village is a three-storey building made of wood and stone. The ground floor consists of a pigpen, stalls for cows and a storage area for farm tools. The first floor consists of a kitchen, living room and bedroom. This is where members of this house sleep, and it generally serves as the main living space for day-to-day life: communal space takes up 90 per cent of the first floor, and in some cases the remainder is given over to one or two small bedrooms. The top floor usually has a toilet, a balcony and a Buddhist altar. Some houses also have a spare bedroom there for short stays. The roof area above the top floor is usually used to dry crops and store possessions. The first floor includes three features whose presence is definitive of a ‘normal’ house in Zokjibu: the ka, the sjangku and the rahbo. I visited several houses when I came to Zokjibu for the first time. Residents always invited me to sit in the biggest room of the house, which had at least one stove. When I asked what the room was called, some replied, in Sichuan Chinese, that it was the ‘living room’, others that it was the ‘kitchen’. When I asked for the Tibetan term, I was always told ‘We usually call the room “ka”. It means the biggest room, where there is a stove in the house building’. Eventually it emerged that the living room and the kitchen were once the same room but began to be separated into two separate rooms about twenty years ago.
28
Wang Ting-yu
Up to three benches will be placed along the walls around the sjangku or fire (see below), forming a U-shape. The term for the benches is k’rbian, meaning a bed that can also be used for sitting on. Some households use chairs instead, since they are easier to move around and are cheaper. Even then, the chairs are always arranged to form a U-shape, just like the k’rbian. These days the wall that is left without seating is used for a television set or a cupboard. The ka is used for such communal activities as sitting, eating, drinking, chatting, singing and dancing. The ka also contains an area for women to cook food, boil water, and make tea and wine. Only a very few households cook food for cows and pigs at the sjangku. A separate stove is usually located on the ground floor near the pigpen for this purpose. Zokjibu residents told me that the ka is also a way of classifying people, saying ‘ka refers not only to a place, but also to the seating order’. There are actually two seating orders, ordinarily clearly seen at meal times. People are separated into one of three categories: (1) seniors and guests (both men and women from outside Zokjibu village); (2) junior men and male relatives from outside the house; and (3) junior women, female relatives from outside the house, and children. The seniors and guests have separate seating for men and women, men to the right of the ka and women to the left. I was also told about another order of seating, said to have been current while the tusi still ruled. In this version, people were further categorized according to their social status. Lamas and nobles held the highest status, the seniors an intermediate status (they also have more relatives in the village), and young men the lower status. Seating arrangements for occasions other than domestic feasts were simply extensions of the seating order in the ka. Whether at domestic meals or public meals, women must kneel while serving men, and in the past could not eat until the men had finished. Men cannot serve themselves rice or tea. A woman who leaves a man to serve himself with rice or tea will be criticized and laughed at for her bad manners. One day I attended a funeral and joined in the lunch; a woman – a good friend of mine, but not a member of the host’s family – jokingly told me to help myself to the rice, which I did. I met the same woman the morning after and she told me that her mother-in-law had spent the night criticizing her for her lack of manners. Her mother-in-law was upset because of the embarrassment her daughter-inlaw had caused by allowing a man to help himself to rice. Apart from funerals and other formal public occasions, the seating order is not rigorously followed. The women of a household can sit and eat with the men of that household, but they cannot sit with guests and are still expected to serve rice and tea to all adult men and any guests. On formal occasions, close female relatives and women of the household all eat last: ordinary female guests eat at the same time as men, but they do so in separate rooms. Women guests need not be waited on. In the past, the location of a fire made in a trivet in the centre of the ka was known as the sjangku: it is commonly said that ‘the fire of the sjangku will never be extinguished’. Nowadays iron stoves are used, which use less wood and
The House, the State and Change
29
produce less smoke, and some houses have now replaced their fire stove with an electric stove kept in the ka. The relationship between ka and sjangku is an interdependent one: if a stove is not located at the centre of the ka, it cannot be called a sjangku. Similarly, if the largest or main space of a house does not have a sjangku, the space cannot be called a ka. Villagers call the rack that hangs above the sjangku (fire) a rahbo. One of its purposes is to prevent the flames from the sjangku reaching high enough to spread elsewhere. It is also used to store utensils and ingredients. Finally, the rahbo serves as a metaphor for authority within a house. Residents say: ‘Anyone with the power to order others under the rahbo wields the authority in the house.’ One aspect of the symbolic meaning of a ka is how inheritance is associated with the iron trivet (used as a stove) and hearth fire of the sjangku. Whether male or female, only one person in a generation can inherit a house. This person takes control of the house building, land, cattle, tax payment (for tusi) and, most importantly, the iron trivet of the sjangku. If a successor’s parent is poor and propertyless, he or she must at least be left the house building and the iron trivet of the sjangku: inheritance of the iron trivet of the sjangku indicates that the heir is the proper heir to that house. Ancestor worship is another aspect of the relationship between the sjangku and the continuity of the house. On the fifteenth day of the seventh lunar month, villagers must worship their ancestors in front of the house building.4 Pork, rice, paper money, tea and wine must be prepared and burned in a circle, marked out by using ash from the sjangku. Villagers call this ash sjangkalabgi. They also burn the same offerings to ghosts, who do not belong to any household in Zhoukeji and are located outside the circle marked by the sjangkalabgi. Villagers believe that only their ancestors can enter the circle of sjangkalabgi and take the offerings: other ghosts cannot. Villagers told me that this custom was borrowed from the local Han Chinese, because half of all villagers have a Han Chinese ancestor. The symbolic meanings outlined above regarding seating customs, authority and inheritance indicate the centrality of the ka as a social unit, but both domestically and for the community, changes in recent decades have made that centrality more virtual than actual. Only two or three households still use the rahbo. This is because many villagers are redecorating their houses for tourism and more modern sensibilities (both require cleanliness and tidiness), and keeping the rahbo would interfere with this. Similarly, only one household still uses the iron trivet in the sjangku; as already indicated, all the others have replaced the trivet with an iron stove, which is safer, more fuel-efficient and produces less smoke. Nowadays almost every house has a living room and a kitchen: can either be called a ka? Some villagers call their living room a ka, but for others the ka is the kitchen. Generally speaking, the stove that people consider the sjangku has also been located at the centre of the living room. Chinese-style stoves have for
30
Wang Ting-yu
the most part replaced the Tibetan-style stoves (iron trivets), but the area surrounding the Chinese-style stoves tends to be quite grimy and thus unsuitable for receiving guests: therefore the living room is where family members and guests alike spend their time. Some people were already redecorating their living rooms in a more traditional Tibetan style, not only to display their social status to other villagers, but also to attract tourists. The form of the ka is thus changing, but residents still arrange the seating following the traditional ka pattern and regard this arrangement as symbolizing the legitimacy of house inheritance. Nowadays, although faced with the effects of modernity and its changing form, the meaning of the ka is still relevant to villager’s lives and is kept active. Thus we can say that the notion of the house can be inherited and is remembered by Zokjibu people.
The Religious Aspects of the House: choko and rewakang In Zokjibu, a house has two major religious places: the choko, an altar; and the rewakang, the space where the mountain god is worshipped. The choko is always located on the second floor, the highest place in a house, with an open balcony. People place Tibetan Buddhist statues and tantric scrolls in the choko. These statues and tantric scrolls are usually bought in the market, but they must be ritually blessed by a lama to become effective. Villagers believe that, if they have a choko full of statues and tantric scrolls, they will acquire more Buddhist merit than those who do not. Another important function of the choko is its use as a shrine for funerals. If a member of a household dies, villagers summon several lamas to come to the house and carry out the necessary rituals as soon as possible. Once the lamas arrive, the funeral can begin. The lamas chant and carry out rituals for the deceased in front of the choko. A funeral can last at least six to seven days, and they must eat and sleep at the choko until it is over. The choko area is sometimes used as a bedroom when there are a lot of guests or villagers staying in the house, but only men can sleep there. A woman cannot sleep in the choko, even as a guest. Indeed, women are not particularly welcome near the choko, and if they do go there, they must leave as soon as possible. Not every house has a choko. Some houses have more than two children living at home or want to take in more tourists, and so rather than a choko they build another bedroom. When such houses need to conduct a funeral, they use an empty room on the second floor as a temporary choko. Some villagers told me that, if a family’s ancestors had been Han Chinese immigrants, their houses would not have a choko on the second floor anyway, only a shrine in the living room on which to place a statue and thangka of the Buddha. In other words, the choko also represents the identity of their ancestors. But even this association is in many cases redundant: almost half the houses in the village do not have either kind of altar.
The House, the State and Change
31
The rewakang is a shrine located at a wall near the balcony where branches of pine or cypress are burned for the mountain god. The rewakang must be out in the open because it is believed that the smoke from the rewakang can convey people’s sincere wishes to the mountain god. Almost every house has its own rewakang, and there is a communal rewakang located at the edge of the village, called cank’skiyu, already mentioned as one of the religious sites of Zokjibu village. Villagers burn branches of pine or cypress in the rewakang every fifteen days, or on the first and fifteenth days of the lunar calendar, praying to the mountain god in the local Tibetan or Sichuan Chinese dialect. The prayer always mentions the name of the mountain god and presents a petition for the safety and well-being of the entire family, including any children or siblings living outside the village. Villagers burn flour called rewaba with the branch of a tree. The cank’skiyu is used in the same way as the rewakang in the house buildings, except that it has another, different altar called the skiyu. The rewakang plays an important role during funerals. The lama must feed all the deities (including the Buddha, the Bodhisattvas and the mountain god), the ghosts and the deceased three times a day, just as human beings eat three times daily. The ghosts and the deceased must be fed both outside the house and on the roof, while food must be offered to the deities at the rewakang because only the smoke of the rewakang can convey wishes and pleas to them. Today, some houses that do not have a rewakang use a suitably sized container as a temporary rewakang at funerals. Villagers said that in the past every house had a rewakang, but all rewakang and choko were destroyed during the Cultural Revolution (1966–1976). Villagers have been allowed to rebuild their rewakang and choko since 1982, and have done so in two ways: the traditional way is to make a chimney in a wall to release the smoke through the roof. The other way is simply to make a hearth against a wall of the balcony. The houses that rebuilt their rewakang in this way had been built before 1982, and villagers did not want to remove a whole wall to build the traditional type. Some houses have still not rebuilt their rewakang. This illustrates how villagers are facing modernity in their everyday lives by changing aspects of the house-building structure, rather than their religion: that is, they change the outer form, but the traditional meaning is still connected with most people’s lives. The ka is a good example: its form has changed in many different ways, but its traditional significance remains in everyday life. The part it plays in how to classify people, both within a house and between houses, and the ideas about house inheritance are both at the core of people’s lives. The Zokjibu house system as it exists today shows us how these Tibetans are facing changes. When the force of the Qing Empire reached this area, the state ruled this house system through the tusi, and Zokjibu villagers also used this system to procure land and other properties. After 1949, the house system changed along with decline of the tusi’s rule. There are no longer any tusi or heirs of the tusi living in Zokjibu. According to the structure of the house
32
Wang Ting-yu
system, it should collapse without the system as its centre of power – that is, the tusi and its statecraft – behind it. However, the house system remains the main structure of people’s social and economic relationships, and people still design their houses based on the ‘ideal’ model of a house. In the next section, the economic changes in Zokjibu will be described.
The Economy As we can see, the house system was not only a political means to organize people’s social relations in Zokjibu, but also an economic one. For example, who could own land, cattle and houses, who would inherit them, who needed to pay the tax, and which career people could choose – all of these were determined by the house that people belonged to. But after 1949 the control of this system began to loosen, and the most significant change came from the mode of production. The traditional mode of production of Tibetans in Zokjibu was a mixture of agriculture and herding. The major crops were wheat and highland barley, grown in the valleys; yaks were herded in mountain pastures. During the first half of the twentieth century, some residents traded locally grown opium for gold, silver, coral or cash, and at other times for wheat or highland barley. People in Zokjibu report that the earliest trade in the area was in medicinal goods, which were bought and sold in Dujiangyan for tea, rice and salt.5 During the time of local chiefly rule, wheat and highland barley served as the major staples, although a few wealthy households ate rice imported from Dujiangyan. The wheat was ground up and processed into noodles or bread at one of the two flour mills, sometimes resulting in long delays. The noodles were a special sort called gmji bei bei (beggar’s noodles) found only among rGyalrong Tibetans. Wheat flour was also used as offerings in rituals to the mountain god or to Buddha, called rewaba. But although of great importance, the wheat crop was but a fraction of that of the highland barley. Highland barley is processed in one of two ways. It is ground into flour and eaten with tea, butter, salt or sugar. Secondly, it is used in two different types of alcoholic beverage, one being distilled. Both kinds of beverage are usually drunk when guests visit or at Chinese New Year. Both wines are necessary to the ritual process. Rather than being used for daily consumption, most alcoholic beverages are used as offerings to the Buddha, the dead and other spirits. In the local Tibetan Buddhist tradition, highland barley and other grains (such as wheat or rice) can be used for any ritual occasion. The lama throws the grains as offerings to the Buddha. A house will generally use twenty-five kilograms of highland barley in a single year, and if there is a funeral, it will need fifty kilograms. Meat eaten traditionally consisted of pork and yak, and less commonly duck or chicken. Fish are almost never eaten, since they are released to acquire Buddhist merit. Each house keeps at least one pig and one milch cow. Yaks are
The House, the State and Change
33
also important, though not a requisite for each house. Yak wool is made into ropes and blankets, the meat is usually offered to guests, and the milk is used to make high-quality butter. Merchants also use yaks as pack animals when travelling over the mountains. Pigs are slaughtered in late autumn, and relatives are invited to help and to join in the subsequent feast. The male guests help with the slaughter proper, while the female guests help with the cooking. Still, the status of pigs is very low. They must be kept in pigpens. Cows are important for providing dairy products used in both rituals and everyday life. Their milk is used in two kinds of tea, both of which are served to guests and ritual specialists. The first is a mixture of salt, butter, walnuts and milky tea; the second omits the butter. Butter is a necessary ingredient in roasted barley flour in general, and is put in the flour used for the ‘dolls’ that feature in rituals in particular. The cow enjoys higher status than the pig. Cattle are allowed to roam freely in the grasslands and are fed ‘dolls’ of roasted barley flour. Today, wheat and corn are the main agricultural crops. Highland barley is almost never grown, but purchased in the market. The amount of wheat grown is insufficient to provide the noodles and bread needed for the whole year. Corn is considered unfit for human consumption; it is sun-dried on the cob, then stored to feed to cows and pigs when the winter snows make grass and other forage inaccessible. Vegetables are sometimes intercropped with the grain crops, but most people buy their vegetables directly from the market. Although the villagers do not have enough land to plant highland barley for their own use any more, wine and flour from highland barley are still necessary for everyday life, as well as for religious purposes. If villagers want to eat pork at times other than the autumn, they buy it from the market. The number of guests who will be invited to join a feast where pigs are slaughtered is decreasing, and nowadays only includes a few relatives and friends. Although villagers still have several yaks on the mountain pastures, they usually sell these in the market. Because they have less opportunity to eat yak and drink yak’s milk, they use cow’s milk instead. Villagers go to market when they want to eat yak, where there is fresh yak meat every day. Nowadays, people also usually carry things by car or motorcycle. Today the daily staple is rice, which the government has long subsidized. The earlier subsidized provision of rice began to be replaced with a cash subsidy in 2004, but there is still a lot of rice stored in the village. Rice is also used in rituals and at funeral feasts. The power of the tusi quickly declined after 1951. After 1954 taxes were paid to the new government instead of to the tusi. In 1958, private land and harvests in Zokjibu were all appropriated and placed under a people’s commune. There followed the familiar story of that era for rural China in general. One was given food, no matter how hard one did or did not work. Meat was readily available, and every day was ‘like a feast’. This lasted for two months until the food ran out, and starvation conditions soon followed. Food shortages
34
Wang Ting-yu
continued in this part of China until 1982, which saw the beginning of decollectivization. By the 1990s, the villagers’ main income was no longer derived from agriculture or herding, but from woodcutting and the sale of wood, and its transport by truck. Villagers used this income to buy food and grain. In 2000, however, the government began to carry out a policy to prohibit woodcutting in order to protect the forest and the environment. As part of its environmental protection policy, the government began to withdraw the land of Zokjibu residents, and instead subsidize rice and wheat according to the acreage that each family was going to lose. Many villagers with farms in the mountains or on the mountainside have therefore been prevented from farming anymore because it is too hard and too far away to do so. The maximum farm size is now ten acres; the smallest landholdings do not even amount to one acre in size – in other words, the main source of food is no longer the fields, but the open market and the government subsidy. Tourism began in 2000 with initially just a few people being involved in this work. Now almost all the households in the village compete to accommodate the tourists. Tourists are usually taken to visit the official residence of the local chief and see the beautiful nearby scenery. A few years ago, a novel about a Tibetan tusi was produced as a television series (Alai 2004, [1998] 2005). The story is set in the time of the last two Zhoukeji tusi, and the official residence became an important set for the filming. It was also once briefly the residence of Mao Zedong and the location of a meeting he attended. The village residents hoped that the novel, the television series and the story of Mao’s visit would attract lots of tourists. The first ten days of both May and October are national holidays in the People’s Republic of China. At both times, up to fifty tourists stay in the village, bringing in a substantial income for many residents. What the tourists are willing to pay for is the experience of ‘traditional life’ in Zokjibu, in particular its ‘traditional Tibetan food’.
Conclusions The above account suggests that the dynamic of modernity in Zokjibu comes not through Tibetan Buddhism but through everyday life, especially as a consequence of the way in which the state makes its presence felt. The process of becoming modern in Zokjibu involved two stages. The first was facing the Qing’s imperial state; the second was facing the nation-state, which is the PRC. In the first stage, the empire ruled Zokjibu through the tusi, and in turn the tusi used the house system to control the people of Zokjibu. The sub-bureaucracy of the imperial state was created and integrated into Zokjibu’s social structure. In the second stage, the nation-state was ruled by different levels and different departments of government, for example, the township and county governments, the police, the religious bureau, and the Communist Party. All of these bureaucracies and sub-bureaucracies come from outside Zokjibu; most of
The House, the State and Change
35
them are not even Tibetan. They have not been integrated into Zokjibu’s social structure, yet their impact on the village is as strong as was the tusi’s. So, we can say that the power of the state and the market has greatly influenced Zokjibu residents to become modern. It is only in relation to religion and tourism that Zokjibu represents anything ‘traditional’ nowadays. Ironically, the result is that village residents realize that in Zokjibu they can live in a way that is actually a distinctively rural and indigenous form of modernity. Life is increasingly taking on suburban aspects. Residents in general, and men in particular, make frequent visits to the county seat to buy food and other consumer items, and drink tea or wine. Virtually every house has at least a telephone and television set, and mobile phones are spreading rapidly. Evenings are spent in front of television sets; feasts are becoming increasingly rare. Nonetheless the house exists in both memory and reality. Looking at its material and social aspects, it becomes evident that it remains the centre of villagers’ social lives. Every day, the people of Zokjibu reproduce the idea of a ‘normal’ house – with a sjangku, ka and a house name – both internally and on public occasions. Although the house itself has undergone some changes with modernity, its symbolic significance has not changed. In particular, it can be seen that the changes are ones like the mixture of given names – things that have less connection with the inheritance of houses, thus allowing them be changed or replaced. The house name always represents the descent of a house, but a given name cannot do so. Having a mixture of given names, Tibetan and Chinese, is thus a reasonable strategy for villagers. This indicates that, although villagers will choose to make some changes to adapt to modernity, these do not extend to the reproduction of the ideal house. It is in such ways that Zokjibu people are facing modernity.
Notes 1. The Tibetan part consists in finding a lama to do the chanting and villagers to sing the mani, a Buddhist mantra, for the deceased. The Han Chinese part consists in burning paper money for the deceased and using a Han Chinese coffin for the burial. 2. In the past, some villagers who were businessmen would have a Tibetan name and also a Han Chinese name, so that they could easily deal with both Tibetans and Han Chinese. The Darmojobu family is an exception: they also have a Han Chinese surname, Yang. This is because this family provided the chief’s translator. Other villagers would not mix Tibetan and Han Chinese names at the same time. 3. The ethnic identity policy which began in the 1950s identified all the people in Zhuokeji as Zangzu (Tibetan). Therefore, they all now think of themselves as Tibetan, including some Zhuokeji people who have a Han Chinese surname and personal name. Some villagers still think that those who have Han Chinese surnames are not pure Tibetans, but a hybrid of rGyalrong Tibetan and Han Chinese. 4. Many people in Zokjibu now use the Chinese lunar calendar to arrange their religious activities. In the past, they would ask a lama and arrange religious activities in accordance with the Tibetan lunar calendar. This is written in Tibetan script, which almost none of them can read. If a lama is present the villagers still do this, but as there are no lamas
36
Wang Ting-yu
living in Zokjibu on a day-to-day basis, villagers use a Chinese lunar calendar bought from the market. They have learned to read Chinese at school or elsewhere. 5. Dujiangyan is a Han Chinese area north-east of Chengdu, the nearest rice-growing area.
References Alai. 2004. Chen ai luo ding (TV series). Shanghai: Oriental Satellite TV. ———. (1998) 2005. Chen ai luo ding. Beijing: People’s Literature Press. Carsten, Janet, and Stephen Hugh-Jones (eds). 1995. About the House: Lévi-Strauss and Beyond. Cambridge: Cambridge University Press. Comaroff, Jean, and John Comaroff (eds). 1993. Modernity and its Malcontents: Ritual and Power in Postcolonial Africa. Chicago: The University of Chicago Press. Dai Ying-Cong. 2001. ‘The Qing State, Merchants, and Military Labor Force in the Jichuan Campaigns’, Late Imperial China 22(2): 35–90. Goldstein, Melvyn C., and Matthew T. Kapstein (eds). 1998. Buddhism in Contemporary Tibet: Religious Revival and Cultural Identity. Berkeley: University of California Press. Kapferer, Bruce. 2002. ‘Sorcery, Modernity and the Constitutive Imaginary’, in Bruce Kapferer (ed.), Beyond Rationalism. New York: Berghahn Books, pp. 105–28. Keane, Webb. 1995. ‘The Spoken House: Text, Act and Object in Eastern Indonesia’, American Ethnologist 22(1): 102–24. Lévi-Strauss, Claude. (1949) 1969. The Elementary Structures of Kinship. Boston: Beacon Press. ———. 1987. Anthropology and Myth: Lectures 1951–1982. Oxford: Blackwell. ———. (1983) 1994. The Way of the Masks. Seattle: University of Washington Press. Taussig, Michael. 1980. The Devil and Commodity Fetishism in South America. Chapel Hill: University of North Carolina Press.
Chapter 2 From Kinship to State and Back Again: Lineage and History in a Qiang Village Liu Biyun
Introduction The Qiang, numbering about three hundred thousand, are one of the fifty-five officially designated ‘national minorities’ in the People’s Republic of China; they speak a Tibeto-Burman language.1 This chapter presents the coufang (族房), lineage, or ‘people under the feet of one ancestor’, as the central social category and pivotal social grouping in Qiang social life.2 Examination of the coufang graveyards uncovers the historical process whereby the village was incorporated into the imperial military network, as well as aspects of the transition to Han literacy. This chapter argues that the coufang played, and continues to play, an important role helping villagers to survive political turbulence and difficulties in their social lives. This was the context in which Han cultural features, such as ancestral halls, use of Chinese script, Han surnames and generation names, began to be seen. But because the Qiang coufang maintains its specific cultural logic of lineage composition and social relatedness, it displays hybridity in many dimensions. The implanting of imperial institutions in the early Qing dynasty opened up a road to sinicization and imperialization for the Qiang, and the local social groups, the coufang, played an important role in mediating between the villagers and the changes imposed from outside.
Historical Background The majority of the Qiang live in the mountainous regions of north-western Sichuan, along the upper reaches of the Minjiang River, an upper tributary of the Yangtze River. The nearest metropolis is Chengdu, the provincial capital, on the eastern side. The Qiang often reside in close proximity to, or interspersed with, Tibetan communities. This chapter looks at Nzidabu village in Lixian
38
Liu Biyun
County, in Aba Tibetan-Qiang Autonomous Prefecture (Aba Zangzu Qiangzu Zizhizhou) in Sichuan.3 Lixian County is located on the western edge of the predominately Qiang region of the prefecture, where it borders the predominantly rGyalrong Tibetan region in the same prefecture. Given their proximity to and intermixing with other nationalities, study of the Qiang necessitates a consideration of their relationships with the Tibetans and Han Chinese. More specifically, their culture is described as being transitional, since, at the most general level of dispersal, they are physically located between the Han Chinese and Tibetans and have been profoundly influenced by their interactions with both of these communities (Wang Mingke 2003).4 The Qiang propensity to form lineages has clear similarities with Han Chinese practice. Nevertheless, the Qiang lineages, the coufang, found in rural agrarian villages, have their own distinctive social organization and cultural meanings. Many areas in the mountainous region of north-west Sichuan were, historically, predominantly Qiang, Tibetan or Han, and remain so. Before the Qing dynasty (1636–1911), Chinese imperial control in north-west Sichuan was limited to the area along the south-east edge of the Tibetan Plateau – the present counties of Wenchuan, Maoxian and Songpan – mostly along the upper reaches of the Minjiang River. The Chinese imperial courts of different dynasties established centres of administration in this region, which were mostly governed by military organizations and were responsible for levying taxes from the local Han subjects, while the local non-Han populations continued to be ruled by the local chiefs.5 As the Qing presence intensified in this region, more and more local towns and villages, even those under the jurisdiction of the tusi, came under the direct administration of the empire and were incorporated (編戶, bianhu) into a system of regular households that paid taxation (里甲, lijia).6 The situation was quite different in another part of the same region, along the Zagunao River, a western tributary of the upper reaches of the Minjiang River. The Zagunao River is located in Lixian County; half the population in the western part of the county is rGyalrong Tibetan.7 From the early Ming (1407) to the early Qing (1752), what is now Lixian County was an area under the jurisdiction of the Zagu tusi. Zagu tusi families were involved in a complex network of kin and affines with the four other tusi in the Situ area and with others in the south-west of this region. The entangled relationships between these aristocratic families resulted in numerous instances of regional unrest.8 The Zagu tusi regime, Tibetan rather than Qiang, was forcibly replaced by decree of the Qing due to the final tusi’s notoriety for his propensity to attack neighbouring territories.9 The former Lixian area, in which the Zagu tusi regime was situated, was reorganized in 1752 and became part of Lifan Sub-Prefecture (理番廳, Lifanting). Lifan Sub-Prefecture consisted of a large swathe of territory containing five military units (屯, tun) and four tusi in the hands of stateappointed military officials.10 Nzidabu village was located in Jiuzi, the only tun
From Kinship to State and Back Again39
of the five in which the population was identified as Qiang in the early twentieth century. The other four tun and the four tusi were identified as being populated by rGyalrong Tibetans. One of the aims of the Qing Empire in establishing the tun military unit was to train and deploy local troops: when wars were being waged elsewhere, the trained local people could be assigned to fight for the empire. A separate politically motivated goal was to redeploy the headmen formerly under the tusi and retain them as leaders in another form.11 Although Confucian education and government temples had been introduced into the market towns during the reigns of earlier dynasties, not until the early Qing could the children of tusi and people under tusi jurisdiction attend the Confucian schools under the aegis of an imperial policy.12 Under the Zagu tusi regime, Nzidabu village was subordinated to the Tibetan chiefs’ rule and influenced by Tibetan culture and Buddhism. During the Lifan Sub-Prefecture period, Nzidabu village, along with the other neighbouring Tibetan villages, provided soldiers and even commanders for the empire to pacify any unrest in the borderlands.13 Local history on the one hand resides in the narratives and memories of the villagers, and in stories related to ancestors – a condensation of space and time. On the other hand it is mutely inscribed on the tombstones in the village. 14 Both reveal the village’s involvement with the Qing empire’s frontier affairs. In contrast, the lineagecentred coufang, based on common ancestry and still maintained by present-day villagers, exhibits more of the influence of the past empire and of the associated Han culture. This chapter discusses evidence of how the script and power of the empire entered this village: the Qing empire provided the Qiang villagers with the experience of being governed and incorporated into the administrative system, and in so doing introduced literacy into the village. For their part the Qiang absorbed the power of the empire and the consequent effects through its interface with its own social institution, the coufang. Comparison of the Qiang coufang with Han lineage practice reveals some seemingly common but essentially different features.15
Hybrid Landscape: The Coufang Graveyards Nzidabu village is located at mid altitude in the mountains along a tributary of the Zagunao River. It is physically and socially subdivided into the Upper, Lower and Little Hamlets. A road zigzagging from the foot of the mountains to the three hamlets, still without asphalt, has only been opened up in the last two decades. In 2006 the village had a population of approximately four hundred comprising some ninety-five households. There are presently eleven coufang in the village. Each coufang has both a Qiang name and a Han Chinese surname. There are only six Han surnames among the eleven coufang, meaning that they cannot be distinguished on the basis of their Han Chinese surnames, though each coufang has a distinct Qiang name which differentiates it from all the others.16
40
Liu Biyun
The different coufang are roughly nucleated, with the majority of the households of each coufang concentrated within one of the village’s three hamlets. Larger coufang have more than ten households, smaller coufang only a few. The larger coufang have large enough numbers of people to stage the lifecycle rituals of their members without problems; those coufang with only a few households seek out opportunities to cooperate and even merge with other coufang to obtain the numbers needed to stage life-cycle rituals. In general, a coufang has its own graveyard, located in the hamlet where member households of that coufang live, in some cases beside a neighbouring coufang’s graveyard. The traditional Qiang way to build a tomb was, and still is, to pile up stones and clods of earth to cover the coffin, which lies underground or in a hole dug into rock at the foot of a hill. Traditionally there was nothing obvious to identify the deceased: people knew their ancestors’ tombs by memory.17 In contrast, about forty-five tombs with tombstones clearly visible to an observer were erected in the village between 1788 and 1942. Villagers reported that tombstones with inscriptions were limited to being on the tombs of government officers during the Qing dynasty and early twentieth century, a witness to the ‘glory days’ of villagers joining the wars and the history of the empire on the peripheries.18 Some coufang have their ancestral hall, citang (祠堂), located within their graveyard; some have no ancestral hall but have a zongbei (總碑) or ‘communal stele’, almost all in the graveyards. An ancestral hall is a square room without a roof but with walls each about 5m long, about 2m high and made of clods of earth, and with a doorway in one of the walls. On a small flat stone inside, opposite the doorway, is a stele inscribed with Chinese characters, for example: 余氏門中歷代昭穆高曾遠祖之神香位 Yushi menzhong lidai zhaomu gaozeng yuanzu zhi shen xiangwei Tablet of the ancestral spirits of each generation of Yu
The Chinese surname Yu has been used rather than Nziwa, the Qiang name of the coufang. An ancestral hall has the same function as the zongbei, both being places where people within the same coufang can come together and make offerings before gathering in a nearby field to enjoy the food and wine offered to all the ancestors after the offering ritual has been concluded. All members of the same coufang or Han lineage gather for ancestral rituals, but the Qiang ancestral hall functions differently from that of the Han, and especially from those found in southern China since the middle of the Ming dynasty. These are more likely to exemplify the cosmos of the Confucian ethic of filial piety: the descendants change the ancestor tablets generation by generation, and the incumbent ancestors-on-tablets are always within four generations of their own generation, in accordance with the neo-Confucian ideology approved by the imperial courts. The practice among ordinary Han people of building ancestral halls in southern China experienced changes from the Song to Ming dynasties, including moving the site from graveyard areas into dwellings, and descendants of the same lineage accumulating and managing
From Kinship to State and Back Again41
landholdings to provide for sacrifices to the ancestors (see Faure 2007). Qiang ancestral halls have never been moved and remain in graveyards, where the ancestors belong and need to return, nor have coufang members developed communal permanent property for the ancestral rituals. 19 Qiang villagers gather in their ancestral hall only three times a year: at the end of the lunar year, layue sanshi (臘月三十); at qingming (清明) and at qiyue ban (七月半), the so-called ‘ghost festival’. To feed and satisfy the ancestors with food and wine, to make the living all happy together on those days, and to receive blessings from ancestors through eating and drinking the offerings in turn are the primary goals of the ancestral rituals for the living. The tombstones, ancestral halls and communal steles in the graveyards play important roles in Qiang villagers’ ritual lives with their ancestors, though they are becoming weathered and gradually being eroded as time passes.20 Four pairs of stone hitching posts, xiama zhuang (下馬樁), remain standing in the graveyards: two of the four are outside the doorway of two different ancestral halls, one within the broken walls of an ancestral hall, and the fourth in front of a tombstone.21 People on horseback were required to dismount and tie the horses to these hitching posts. The elders told me that in the past people could not enter the graveyard on horseback: commoners had to respect the authority of the lineages by dismounting and staying low when passing by. The xiama zhuang thus exemplify the official status the dead formerly enjoyed, but they have lost significance in the present context: villagers certainly do not make offerings there during ancestral rituals. There are two other places in a coufang’s graveyard, both now almost invisible, where villagers still make offerings to respect the deceased in ancestral rituals. One is the crematorium, which is used after a bad death: villagers reported that it was very rarely used now, although historical records mention cremation in the other Qiang areas since the Later Han dynasty.22 Making offerings in front of a crematorium or its former site is the first step in the ancestral ritual; those who are cremated are ‘bigger’ than ancestors who died normally and were not cremated. Some coufang have never owned a crematorium: they used that of another coufang when in need, which historically helped to establish a specific relationship between them. The other place is a corner set aside for the invisible ‘servant’ of the dead, xifuzi (媳婦仔, ‘daughter-in-law’) in Sichuan Mandarin, zi11jue55n 11se55gu55n 11 in Southern Qiang. The Qiang term means ‘she who carries water on her back and cuts firewood’, tasks that a son’s wife is supposed to do. Making offerings to xifuzi shows that the living do not dislike ‘her’, and indeed show her respect during the ancestral rituals. The role the xifuzi plays exemplifies the hierarchy of different generations of ancestors: the most recently deceased serves as xifuzi and needs to take care of all the more senior generations of ancestors. Departure from the graveyards after ancestral rituals and commensal activities have finished occurs in a sequence fixed by tradition: the descendants of the first coufang to arrive in the village have the privilege of being the first to e
e
42
Liu Biyun
leave. I observed the ancestral rituals of Nziwa coufang on layue sanshi, the thirtieth day of the twelfth lunar month in the Upper Hamlet. A Nziwa coufang member asked a child to go to the zongbei of Rudi coufang to see whether that coufang had left its graveyard yet. Rudi is considered the first coufang to come to the village, at least in Upper Hamlet, and so they must leave first. Only then can the others follow. Even Izi coufang has to wait, although their graveyard is next to that of Rudi and shares the same entrance, a ‘door’ made of thorns along the fence between their fields. Rudi members opened the thorny door, filed through and left the graveyard, then put the door back and set off fireworks there. This signified their leaving. Then Izi members did the same, finishing by setting off fireworks to signify their departure. Setting off fireworks ends the ritual: it stops the spirits or ghosts following the living back to their homes. The graveyard in Nzidabu village is thus a space of living history where the social hierarchy is constantly being rebuilt and played out within and between the coufang. People gather in a particular graveyard because of their coufang membership: their departure from the graveyard is ordered in accordance with the arrival time of their coufang ancestors. Coufang practice of social relationships in present-day village life, especially its symbolic and ritual dimension, is strongly linked with their past ancestors. The hybridized sources of the social hierarchy are reflected in the sequencing. The earliest always enjoys the greatest respect, whether a dead individual or a coufang, and conversely when making offerings to the dead: people say ‘the crematorium is the “biggest” ’ and make offerings there first because they regard the threats posed by those who died a bad death to be much greater than those who died a normal death. Letting those who died a bad death enjoy the offerings first will please them and not result in harm to the living. The descendants of the coufang that arrived in the village first have the highest status, as demonstrated by the order of departure from the graveyard. The most recent ancestor to enter the world of the graveyard has the lowest status and must serve as a servant to all the senior ancestors until he or she is ‘upgraded’ or replaced by a newcomer from a later generation. These practices and their various associated meanings probably originate partly from Tibetan Buddhism and some Qiang traditions, but the available evidence does not allow clear differentiation. It remains unclear whether any ancestral halls were built or whether the custom of all coufang members assembling for ancestral rituals began before the time of imperial military posts. The remains of ancestral halls, tombstones and xiama zhuang are evidence of the influence of the imperial presence and Han culture in the frontier region. The fact that the steles inside the ancestral halls are written in Chinese characters is connected with the time the villagers started to use Han surnames, which can be traced to late in the Qing dynasty’s Qianlong period. Careful consideration of the implications of the Chinese characters and cultural practices associated with the ancestral halls, or similar, uncovers features displaying an interesting hybridity.
From Kinship to State and Back Again43
Transition to Literacy: Use of Chinese Script The places in the graveyard where ritual practices occurred and occur, some no longer having any visible distinguishing feature, display a hybridity of Qiang, Han, and Tibetan cultures. The Chinese character inscriptions in particular, on the steles or tombstones, are a result of the village’s involvement in imperial military history. The oldest stele in the village is the zongbei of Izi coufang, erected in 1788, thirty-six years after Zagu tusi was replaced by the Qing court. It is quite large – at least a metre high. The inscription on the front of the stele looks like a genealogy (see Figure 2.1). The inscription is written in Chinese characters without punctuation, but cannot be read as Chinese. None of the villagers could say exactly what is written. The first sentence is: 先理來 依止物 xian li lai yizhiwu
An old man of Izi coufang who recognized the characters thought they said that Izi was the first coufang that had come to the village. This is based on the local pronunciation of the Chinese characters and their rough correspondence with the pronunciation of possible Qiang terms. He regarded yizhiwu as a Chinese transliteration of his coufang’s name Izi, which is also called ‘Iziwuo’ in Qiang. He wondered too whether Izi or Rudi was the earliest coufang in the village.23 Other combinations of characters on the stele look like names, each of three or four Chinese characters, written vertically and listed row by row from top to bottom. The reading order appears to be from right to left, in that the characters 男, nan (son), 孫, sun (grandson), 玄孫, xuansun (great-greatgrandson), and so on, are listed on the far right of each row. A few combinations appear to correspond to the names of some present-day coufangs, for example: 茶物, chawu, to Tsawuo; and 別肉吉, bierouji, to Biemeijei. If this is a genealogical record, as the serial kinship terms indicate, and if it also contains many coufang names, this suggests a genealogical relationship among these coufang. Most of the coufang named on the stele are, however, not recognizable to the villagers, and the relationships indicated do not neatly correspond with present-day inter-coufang relations. This may be evidence of former groups that have long since left the area, an avenue for further research in other Qiang areas. Other questions to pursue are why villagers in early Qing times chose to inscribe steles with information about their internal relationships in Chinese characters, and how a stele like this ‘became’ the zongbei of Izi coufang. The forty-five old tombstones reveal the emergence of a more fluent usage of Chinese, and even contain a citation from the Confucian Classics. The tombstones were mostly those of exemplary military officers under the Qing dynasty; one was that of a scholar-official. The official rank titles inscribed on the tombstones of the military officers were all ranks of the military tun established by the Qing court, in descending order: 守備, shoubei, assistant
44
Liu Biyun
brigade commander; 千總, qianzong, company commander; 把總, bazong, squad leader; and 外委, waiwei, detached squad leader. These ranked positions fall into two categories: those associated with the basic contingent (額設, e-she), and those in a supplementary contingent (增設, zengshe).24 From the early eighteenth century to the 1920s, Nzidabu village, a sub-unit of Jiuzi tun, had villagers promoted: four shoubei, two qianzong, three bazong and two waiwei in the basic contingent;25 two shoubei, one qianzong and one waiwei in a supplementary contingent; and two qianzong, one bazong and two waiwei by hereditary descent. The title awarded to the scholar-official was 監生, jiansheng, Student by Purchase at the imperial academy.26 These military officers and literati were principally from five coufang in the village. The tombstones all record the titles and names of the dead in Chinese characters, and detail the kinship relations and names of those who erected the tomb. Some simply transliterate their Qiang names on the front; some tombstones have epitaphs inscribed on the back.
inscribed on the back. Figure 2.1. Part of the zongbei of Izi coufang, the oldest stele in the village erected in 1788
From Kinship to State and Back Again45
Such epitaphs elaborate brief biographies of the deceased military officers in a Chinese-style genre. Simpler ones describe how their direct patrilineal (sons) or collateral agnatic descendants (nephews) miss the deceased. Others honour the dead for their meritorious contributions as military personnel for the Chinese Empire, including special contributions made in wartime. These cases include names other than those of immediate kin. My translation of an example of an epitaph is given below. The title awarded to the deceased is no longer legible due to erosion. The first few words are 軍功行營, jungong xingying, ‘militarily meritorious brigade’. The rank of the son who erected the tombstone was zengshe shoubei, indicating a shoubei in the supplementary contingent. It has been known since ancient times that to continue and narrate [the achievements of one’s patrilineal ancestors] well is a fulfilment of the virtue of filial piety.27 Although we dare not say we have done well in these two, we will try to represent the whole life of our grandfather. There is no one who would not feel pity for our grandfather for the difficulties he experienced building his family and career. And who knows that he contributed his labour to the Emperor for battles in Nanjing (南京) the first time; Guizhou (貴州) the second; Xiushan (秀山) the third; [not legible] the fourth; Jingyan (井硯) the fifth; Diexi (疊溪) the sixth; and finally, Shangluhua (上路花) the seventh. He was in charge of the provision of salt and foodstuffs in the General Banner (旗, qi) of the Five tun at each of these times. Who else knows that, after he came back home, our grandparents, unlike those who know nothing and lack motivation, worked very hard and were very frugal. They did agricultural work to establish the family and bought farms to found their careers. The hardships of our grandparents would cost many words to relate. Unexpectedly, the heavenly decrees came and [they] went back to the divine road. We owed [them] much during [their] lives and can only lament them after [their] passing. It is just like the saying: the parents will not be present when the children want to take care of them. Now we are inscribing these words on this stone in order to let our offspring know the whole life of our ancestors.28 (A tombstone of the coufang Yang, erected in 1924)
This inscription begins with a citation from the Confucian classic Zhongyong (中庸) about expressing one’s filial piety by following the achievements of one’s ascendants and making them widely known. This provides a reasonable explanation for why the descendants erected the tombstones. The story then proceeds directly to the experience of the deceased joining the battles for the Emperor; the war zones are listed in order, and his responsibility as an officer of very high rank is described. After retiring from military service, the deceased came back home and built his family – another period of working very hard. One can see and appreciate the virtues of the deceased. The Chinese description of his death is very refined: 天書下召, ‘the heavenly decrees came’. This has strong religious and imperial overtones. No matter whether short or long, written earlier or later, the Chinese on all the epitaphs is very fluent in expression. The authorship of these epitaphs is doubtful due to the villagers’ limited literacy in earlier times, and it is possible to speculate that there were professional ghost-writers living in the nearby
46
Liu Biyun
town providing this service. The ornately fashioned tombstones show a wide similarity or commonality of styles beyond the mountainous village, which supports this idea. This is testimony to the transition which the content and subjects of inscriptions underwent. The Izi zongbei inscription is the village’s oldest, dating from 1788. The earliest tombstone with an epitaph was erected in 1815. Compared with the Izi zongbei inscription, although both are written in Chinese characters, the Yang epitaph, erected in 1924, is neat and in a Chinese genre, while the other is unintelligible. The Izi zongbei recorded names, whether of persons or of groups, and social relationships, using Chinese characters merely to transliterate Qiang names. In contrast, the names on the Yang tombstone are Chinese-style, and the epitaph is fluent and well written, as are all the other epitaphs. The Yang epitaph commemorates, praises and even flaunts their ancestors. Other examples demonstrate changes in naming customs, another aspect of villagers’ use of the Chinese script. Villagers nowadays only use Chinese names: there are no exceptions, even in the present generation of seventy- to eightyyear-olds, but the names appearing on tombstones present a different picture. About half are Chinese-style, consisting of the surname and a given name of two Chinese characters. The others are Qiang names transliterated with Chinese characters. These vary in length and corresponding number of characters, for example, Shibaoyesi (石保業四), Akaji (阿卡吉), Sidanbie (思旦 別), Chawuchu (察勿初), Worubao (我茹保) and Alai (阿來). Most are prefixed by a character which is a Chinese surname. This clearly indicates the Qiang naming practice before and during the period of military garrisons. Chinese characters served to record the Qiang pronunciation – as shown in these examples and those on the Izi zongbei discussed above – at a transitional stage to Chinese characters recording Chinese names. A pai zibei (排字輩) fashion also emerged in the mid to late Qing. This is a Han custom in which the first character of the two-character given name of each male member of a particular generation of a family (or lineage) is the same – they all share a particular character, usually auspicious or elegant. The character is chosen to create a sequence forming a poem when many generations have had male offspring and their names are listed. Such zibei poems usually consist of lines of five characters. The Qiang in some coufang used Chinese characters to imitate this aspect of Han culture as practised in Han lineages. Zibei poems are carved on some tombstones, as well as being practised in the naming custom of present-day villagers. Five coufang in the village have zibei poems, four consist of four five-character lines, and the other one consists of two seven-character lines.29 The Chinese used in these poems refers to the empire or state, and encourages people to be useful to the imperial court, or wishes for the continuity and prosperity of the family. It was reported that only male members in a coufang have names with zibei elements, but I observed at least one little girl born a few years ago in Yang coufang whose name shares the zibei character. I observed two coufang that still practise this naming custom.
From Kinship to State and Back Again47
The written records show the development of the use of the Chinese script in Nzidabu village during the one hundred and fifty years from Zagu tusi’s replacement to the termination of Lifan Sub-Prefecture (1752–1913), a period which also saw a transition from phonetic to semantic borrowing, indicating the adoption of the Han writing system. This transition is reflected in the change in the villagers’ naming customs. It is directly related to the military role villagers played in the Qing empire, which saw the influence of Han culture and its writing system gradually permeate the village. The military provided channels for the villagers to achieve higher social positions and brought them glory, and the erection of tombstones after a death gradually became fashionable. Villagers began to erect permanent tombstones and display their ability to write Chinese to manifest their social status.
Coufang and its Members As mentioned above, the coufang in Nzidabu village is a social unit sharing ritual practices: coufang members have the same ancestors, and they all visit their coufang graveyard together. Male members have zibei names in common. The coufang is a coherent grouping in these senses, and there is a relationship of mutual respect between different coufang. In the coufang graveyards, a gradual though not full-scale process of sinicization can be observed in the fashion to erect ancestral halls, xiama zhuang and tombstones with inscriptions, as well as the use of Chinese characters and writing genres. The focus can now be turned, first to within the coufang to see how it is constituted and the cultural logic on which it is based; secondly, to the relationships between the eleven coufang to see how they interact with others within the village as localized social groups living on a high mountain; and thirdly, to the reaction of the villagers involved in these social relationships when facing the Qing military context. In other words, the historical construction of social relationships will be traced through the lens of the present-day society.
Membership There is a saying in Nzidabu village: sandai de qinqi, wandai de coufang, 三代的 親戚, 萬代的族房, ‘kinship is of three generations, coufang is of ten thousand’, meaning that kinship only extends over three generations, but coufang relations continue generation after generation and last forever. Another saying, yidai qin erdai biao, 一代親, 二代表, ‘the first generation are close kin, the second generation are maternal cousins’, means that kinship lasts until the third generation. After the third generation kin ties are lightened and loosened, and thus no longer constitute a bar to marriage. These sayings have two implications: one is that those within two generations are regarded as close kin and cannot intermarry; the other is that, no matter how many generations pass, the
48
Liu Biyun
relationship between people within the same coufang will never change and coufang members can never marry each other. This is also reflected in the villagers calling a coufang ‘people under the feet of one ancestor’: the descent of people from one ancestor is an unchangeable fact, and there is no possibility of people within such a group marrying each other. Although coufang members are descended from one ancestor, segments have developed over generations, a natural result of families dividing. People in the same coufang can identify each segment but have no clear memory of when division occurred, nor of who was involved. Remembering one’s own segment is not difficult because members of the same segment visit the common tombs of their ancestors and later assemble with other segments within the same coufang at a nearby ancestral hall or communal stele. The unit under the coufang is a household: hu (戶) or jia (家) in Sichuan Mandarin, and ji55 in southern Qiang. Generally speaking, a hu or jia refers to a social unit comprising people living and eating in the same house who share livestock, land and property, as well as subsistence and ritual obligations; the house itself is also included. A hu or jia might typically consist of a married couple, their children, the parents of the coufang in which the couple settle, and unmarried siblings; they all belong to the same coufang. Marriage results in a change of social status and a substantial residential rearrangement for the married couple, as well as their coufang membership; it depends on negotiation and consent between the two parties to the marriage. A marriage involves either the man or woman marrying into one of the two associated jia parties, and the marital couple and their subsequent children then belong to that jia and thus its coufang, and inherit that surname. A fenjia (分家), a division of a jia, usually follows the marriage. This refers to the situation of a member of a jia finding another place to build up his or her own jia and the taking away of his or her share of the household properties mentioned above. The member who stays at the original jia and builds up his or her marriage within that jia is said to chengding menhu (承頂門戶), namely ‘succeed and head the (original) hu’. As a member household of a coufang, a jia or hu should shoulder its obligation to its coufang, since only then will its membership within the coufang be socially recognized – and the same is the case with a newly built jia. If fenjia only results in two residential and living units which do not shoulder their individual responsibilities to their coufang, then the two substantive jia are still regarded as just one in terms of coufang membership. For example, a woman of Tsawuo coufang married a man of Ruwabo coufang uxorilocally, that is, the marital couple stayed in and inherited the woman’s jia. They are each the only son and only daughter of their own respective jia, and they had just one son after marrying. Their only son therefore ‘heads’ (頂, ding) both his mother’s and father’s jia. He has two surnames and two given names: ‘Wang Yong’ and ‘Yang Mingyong’, each used as appropriate when dealing with the affairs of the two coufang he belongs to. He also carries out the ritual obligations in both coufang. Hence, the two coufang – Tsawuo and Ruwabo – both recognize his and his families’ membership. By comparison, another man of Izi
From Kinship to State and Back Again49
coufang has married and left his father’s house and lives with his wife and children in another house, but on ritual occasions he is not regarded as separate from his father’s jia. His father no longer farms, and his only unmarried brother lives with their father and has not established a jia; they do not yet have fenjia. Therefore, at weddings or funerals within his coufang, this man ‘heads’ the one jia and gives only one set of gifts on behalf of his father (including his unmarried brother), himself and his jia. Thus fenjia encompasses not only a division of the members and properties of a household, but also ongoing individual obligations to their common coufang and consequent confirmation of the social status of a jia. In summary, marriage, through arrangements for post-marital residence, is the second decisive factor in the attribution of membership to a jia and a coufang, the first being one’s birth. Post-marital residence may be either uxorilocal or virilocal: married couples have the choice of joining the coufang of either the husband or the wife, according to the most efficient distribution of descendants and labour. This is a specific feature of Qiang coufang membership: members are descended from common ancestors, not exclusively along patrilineal lines, but ambilaterally. This is in stark contrast to the Han concept of what constitutes a lineage.
Marriage Prescriptions There are more elaborations interwoven in coufang, kinship ties and marriage prescriptions. In addition to the coufang relationship, the notions of zhong (種, ‘seed’), tian (天, ‘heaven’) and di (地, ‘earth’) give a certain patrilineal cast to exogamy. The gender of the parents is marked: ‘heaven’ refers to the father, ‘earth’ to the mother. And the ‘seed’ passes from the father to his children and his children’s children. Some important implications follow from this. One concerns children who have only one parent in common. Half-siblings who share the same father but have different mothers have ‘the same heaven, but different earths’ (同天不同 地); similarly, half-siblings who share the same mother but have different fathers have ‘the same earth, but different heavens’ (同地不同天). The terms ‘heaven’ and ‘earth’, referring to father and mother respectively, are usually expressed and realized through Chinese terms. Translated into Qiang, the terms are: ‘wa11yie55a11r 51, wa11mie11de55n 11d 51’ (same father, different mothers) and ‘wa11mie55a11r 51, wa11yie55n 11d 51’ (same mother, different fathers). The Chinese is more metaphorical than the straightforward description in Qiang. ‘Heaven’ and ‘earth’ involve different degrees of transmission of the ‘seed’ of the father and further prescribe marriage rules. For example, male ‘ego’ transmits his ‘seed’ to both his sons and daughters. His sons are same-sex male siblings to each other, his daughters are same-sex female siblings, and his sons and daughters together are opposite-sex siblings. From the position of the third generation, different marriage prescriptions emerge. The children of same-sex male siblings are considered to share the same ‘seed’ and thus may e
e
e
e
e
e
50
Liu Biyun
not intermarry; the shared ‘seed’ is from their father’s father – their common grandfather. In the case of the children of opposite-sex siblings, the children of male siblings are called ‘jiujiu di wawa’ (舅舅的娃娃, ‘children of mother’s brothers’), and the children of female siblings ‘niangniang di wawa’ (孃孃的娃 娃, ‘children of father’s sisters’). The tie of having the same ‘seed’ shared by the male and female siblings is still strong enough that none of the children from these parties may intermarry either. The case of the children of same-sex female siblings is diametrically opposed to this: their relatedness comes through their mothers rather than their fathers as the immediate source of their ‘seed’. Thus, because they have different ‘seeds’, intermarriage is allowed. Therefore, the existence of a male sibling is decisive regarding whether or not the children of siblings can intermarry. Where male siblings who have the character of ‘heaven’ and transmit the ‘seed’ are involved, their children are also regarded as sharing the same ‘seed’, whether directly (through fathers and fathers’ brothers) or semi-directly (through fathers and fathers’ sisters or mothers and mothers’ brothers), and they cannot marry each other. If only female siblings who both have the character of ‘earth’ are involved, they become the carriers of different ‘seeds’ transmitted by their husbands respectively, and the children of sisters are then regarded as not sharing the same ‘seed’ and can thus marry each other. This is reflected in the saying ‘the first generation are close kin, the second generation are maternal cousins’, and further differentiates the biao cousin category into two subcategories: xiebiao (血表), cross cousins; and yibiao (姨表), matrilateral parallel cousins – xiebiao cannot marry each other but yibiao can, although this does not happen very often. In summary, there are two prescriptions regarding marriage. The first is that marriage between people of the same coufang is absolutely forbidden. Secondly, even if people are from different coufang, an additional qualification similarly prohibits marriages between kin related through ‘heavens’, but not between those related through ‘earths’. A distance of three generations ends the marriage prescriptions between people in cases where they are related through one’s ‘heaven’ and the other’s ‘earth’.
Birth Order, the ‘Old House’ and Zhongzhibu Birth order functions in a jia or hu, and decides the hierarchical relations between the jia or hu within the same coufang from a long-term perspective. There is a Sichuan Chinese saying that Nzidabu villagers use to refer to the youngest son’s or youngest brother’s status: yaofang chu laobei, 么房出老輩. The yaofang is the youngest son or brother and his family, while the laobei are those in senior generations and thus of higher status. This saying can be roughly translated as ‘those of the senior generation always appear in the jia of the youngest son or brother (of the previous generation)’. This is used to explain why a man of younger age than another is in a higher kinship position. This
From Kinship to State and Back Again51
occurs because the youngest son or brother is often the last to marry, while at the same time the son of the oldest son or brother may be old enough to marry. In such cases two young men may be close in age, but one belongs to the previous generation. Over generations the gap between the ages of the descendants of the oldest and youngest sons of one particular generation can become wider and wider, until there may be a gap of several generations between kin who are close in age. This saying can be applied to the descendants of siblings in each generation, and can also be used to describe relations between the segments within the coufang. A question arises at this point. How is the Chinese term fang (房) understood in this context compared with its use in the Chinese context? Does the Chinese usage coincide with the concept of a segment of a coufang? The term fang in the context of the Han lineage specifically refers to the relationship between a son and his father: each son of the father is a single fang among other fang, the other sons. Only a son is a fang to the father, a daughter is not; and only the contiguous generation can be a fang – a grandson is not a fang to his grandfather (Chen Chi-nan 1990: 131–32). The meaning of the term fang, as used in the above saying, is similar to the Han genealogical concept in these senses, but the similarity seems to end there without the further implications that the Han may exploit, such as being the basis on which to share out property or ancestor worship obligations equally. On the other hand, the segment of a Qiang coufang is not a strict basis for tracing genealogical relations but a loose one, indirectly indicating the once-existing siblingship, as the fang in the saying above is. The way the birth order traditionally functions in a jia is that the youngest son or brother has the privilege of inheriting his parents’ house. Any other older sons must each move to another house, one into which they have married or built themselves.30 This traditional privilege of the youngest son or brother gives the term yaofang a special emphasis, and over generations the house each yaofang inherited, or laofangzi (老房子, ‘old house’), in turn becomes the oldest and most respected place which the descendants of other fang think they were divided from. The ‘old house’ is held in the same regard as the laobeizi (老輩/伯 子, ‘senior people’). An important document was reported as having been kept within the ‘old houses’, but nobody has ever seen one, not even those aged in their eighties. Nonetheless, there remains some evidence pertaining to this in an ‘old house’ in Lower Hamlet: a wooden tablet, consisting of two wooden sheets inserted into a wooden base which is located on the ancestral shrine of the ‘old house’. There should be a sheet of paper inserted between the wooden sheets, but people say the document disappeared during a turbulent age, a time they cannot pinpoint clearly. The paper is said to have been a written genealogy, a zongzhibu (宗支簿). This recorded the marriages into and out of the whole coufang, and it was put on the ancestral shrine to report those members who were becoming part of, or leaving, the coufang to the coufang ancestors in the ‘old house’. Due to the loss of such documents, and with it the practice, many villagers have thrown their wooden tablets away or burned them as firewood.
52
Liu Biyun
Citang Yinzi, or Silver for the Ancestral Hall Similar to the disappeared zongzhibu, which exemplified the coufang ancestors’ authority over the flow of coufang members, is the practice of compensation for taking women out of the village, including out of the aegis of coufang ancestors. This compensation, called citang yinzi (祠堂銀子, ‘silver for the ancestral hall’), is paid by wife-takers from other villages at all marriages that cross village boundaries in the name of their coufang ancestors.31 The wife-giver keeps the citing yinzi temporarily in their house, then, during the ancestral rituals on layue sanshi (the last day of the lunar year), the funds are used to buy food such as peanuts, melons or homemade sausages. These are distributed to all coufang members when they share the offerings together in a place near the ancestral hall. It is worth noting that the village boundaries were formerly marked by Tibetan Buddhist stupas (pian33da51 in Qiang, tazi 塔子 in Sichuan Mandarin) and another landmark, a kuen55xin55 Buddha (櫆星菩薩).32 The five stupas, which were destroyed in the 1960s, were situated around the periphery of Nzidabu village and served as markers of the village boundaries; although no longer visible, their former locations are known and they symbolically remain in the collective memory. One of the five stupas, with a kuen55xin55 Buddha nearby, was located en route to the market town. A marriage in which the woman married outside the village meant crossing the border represented by the stupa and kuen55xin55 Buddha.33 The citang yinzi thus symbolizes social relatedness across space, in which the power of the ancestors is interwoven with that of Buddhist deities and no doubt other religious powers. In summary, the coufang in Nzidabu village has its own cultural logic. The ‘people under the feet of one ancestor’ are specifically comprised of household units rather than individuals. The household unit is essential for maintaining the coufang – it does not matter whether the son or daughter inherits the household as long as someone does so. People therefore have relatively free choices to marry uxorilocally or virilocally, and thus there is no strict genealogical line based on either patrilineage or matrilineage. There are specific rules concerning the privileges and obligations of household members. And, in the name of the ancestors, the coufang ritually takes care of the flow of its members. As observed above, there are also elements of Tibetan Buddhist culture and Han culture present.
Relations between Coufang As exogamous social units, the eleven coufang in Nzidabu village intermarry and historically have developed specific relations with each other. Some relations are based on a putative siblingship between two coufang, some on the need for labour or the use of a crematorium, and some on a mixture of reasons.
From Kinship to State and Back Again53
Those cooperating with each other on a needs basis are called hecou (合族, ‘merging of coufang’) in Sichuan Mandarin. Two coufang that have merged call each other jiamen rather than coufang: they do not need to observe the exogamy rule of a coufang with each other, nor do they need to change their coufang names or surnames. As a result of the historical hecou, members of both coufang have the right to ask each other to provide free labour – one or two individuals per household – when they hold a wedding or a funeral, occasions which both require large numbers of people to assist. For those coufang with very few households, the hecou is both a convenient and inevitable option. Within the Yang coufang is a group that was originally called Relie. Relie had only one household and therefore merged with Yang and nowadays is always within Yang. Similarly, Nzewuo also had only a few households and had no sons to do the agricultural work on the farms, so it merged with Rudi. No intermarriage takes place between Relie and Yang, or between Nzewuo and Rudi. Another reason for hecou concerns crematoria. The merging of Izi and Yang, for example, was due to Yang having no crematorium of its own. Lu merged with Izi for the same reason, and also because it had too few households to hold rituals (Lu is no longer an independent group but is now always within Izi). Izi, Lu and Yang can mobilize one another whenever one of their member households has a wedding or funeral. They can still intermarry because they are not the descendants of shared ancestors. Another example in recent decades involved newcomers coming to the village and wanting to be part of the community. A Tibetan man surnamed Bai came from a neighbouring village after marrying a woman from Nziwa coufang. The Bai family joined the Nziwa with the approval of Nziwa seniors and other members, and participated in every wedding or funeral in Nziwa to fulfil his coufang obligations. This shows that cooperation with other coufang or joining a coufang are useful and friendly options that Nzidabu villagers have taken, and can continue to take, to overcome many difficult situations. Another form of coufang cooperation similar to hecou is said by villagers to have arisen in association with the military garrison system set up by the Qing empire: at that time Tsawuo had once helped Ruwabo to provide young men for the conscription, and Ruwabo reciprocated by giving Tsawuo land, so that these two coufang still have close relations concerning land. They did not merge with each other but only exchanged manpower and land, and they still intermarry. The relationship between Tsawuo and Ruwabo exemplifies cooperation based on mutual interests in the past.34 Another example associated with avoidance of military conscription occurred in Izi coufang. The Han surname of Izi is ‘Wang’, but there is also a segment with the Han surname ‘Yu’. A few generations ago, one of the current conscription rules was: ‘sanding chou yi, wuding chou er’ (三丁抽一, 五丁抽二). That is, if there were three sons in a household, it needed to send one to be a soldier; if there were five sons, then it had to send two. At that time one Izi household had three sons, and to escape conscription the youngest of the three
54
Liu Biyun
was sent to live with a doctor in the market town. He inherited the doctor’s surname of Yu, which his descendants still have to this day. A few years ago the descendants in this segment started discussions about reverting to their original Wang surname.
Conclusion The hybrid landscape and the use of Chinese characters in the graveyards of Nzidabu village demonstrate clearly that this was a village on the periphery of the Chinese Empire. Its ‘history’ is still visible, although now broken and weathered, and no longer active in the villagers’ present-day lives. It provides clues for an improved understanding of the social relationships between and within the present coufang in the village. Generally speaking, the gaitu guiliu (改土歸流, ‘replacing tusi by mobile administrators’) in the Qing dynasty put villages on the imperial periphery on the road to incorporation into the empire, with no way back. Before gaitu guiliu, direct contact with the imperial polity was limited to contact with the local tusi and their families. The common people were not subjects of the empire and had no direct obligation to the government; they only went to Chengdu seasonally for economic activities.35 The gaitu guiliu planted the imperial institution into the village, making the household a unit for conscription and taxation. On the one hand, the villagers started being constrained under the direct control of the imperial power, while on the other hand they had access to the trajectory of social mobilization and to change further their power relations and social positions at the village and/or local society levels. It is perhaps appropriate to understand the gaitu guiliu as an early example of a process of ‘modernization’: from the chiefdom of aristocratic headmen to the institution in which villagers became imperial subjects. The discussion in this chapter has mainly concerned the coufang rather than the households. Field observation of the present-day village shows that the coufang is actually a more pivotal localized social grouping than the household in both ritual and social contexts. Stories about ancestors and narratives about the graveyard landscape are mostly presented regarding the coufang units. Apart from kinship relatedness and the exchanges of labour in daily life, the crosshousehold relationships between different coufang are still traced back to the unit of the coufang, for example, when coufang merge with other coufang due to their having too few households. There remain gaps in our understanding of the relationships between households within a coufang concerning the affairs of conscription and taxation: more data is needed to see the picture more clearly. But the method Izi members used to escape the conscription requirement for soldiers, that is, sending their children above the threshold number to other households, is evidence strongly suggesting that such strategies must also have existed between the households within a coufang, which provided a greater kinship network resource for the households to exploit.
From Kinship to State and Back Again55
The aim of this chapter has been to discuss the influences of the imperial powers and Han culture on this village located in the high mountains on the periphery of the empire on the one hand, and to compare the differences between and common features shared by the Qiang coufang and the Han lineage on the other. It has not been possible here to clarify the historical origin of the Qiang coufang – whether the communal worship of ancestors is an imitation of Han culture or derives from the Qiang’s own traditions predating the gaitu guiliu. Nevertheless, from the observable material, the differences between the Qiang coufang and the Han lineage prove to be greater than their common features, some of which appear to be imitation and borrowing. Both claim common ancestral origins, and both have practices of mutual help and the common worship of ancestors. They differ in their genealogical concepts, ritual practices and associated religions or cosmologies, as well as internal property and obligation relations. Regarding genealogy, although present-day Qiang put more emphasis on patrikin, there has been no development of a strict patrilineal line. The composition of ‘people under the feet of one ancestor’ is multiple: descent is ambilateral, and thus people within this group are better regarded as cognates rather than agnates; the adoption of a child of close kin or a different coufang to succeed to the household is another common occurrence. This is at extreme variance with the genealogical concepts of the Han. Other Qiang coufang features discussed – such as the building of ancestral halls, pai zibei and even writing genealogies (zongzhibu) – should be regarded as the result of imitation and borrowing from Han culture, as was the use of Han surnames and Han names. This was accompanied by the villagers’ involvement in military history on the periphery of the empire: the military role villagers played in the Qing Empire saw the influence of Han culture and its writing system gradually permeate the village. The most immediate lingering effect is in the use of Han names. At present, the coufang, the social grouping based on Qiang kinship ideology, is still operating, and the relations between coufang established by their respective ancestors in the past are still being followed. But the halo of the ancestors inscribed on the tombstones has already faded, and the villagers probably only notice its existence when some observers from outside come and show a great interest in it. The glories of ancestors marked out in words seem to belong to the past and are not influential today. A major Sichuan earthquake occurred in May 2008, resulting in serious damage to Nzidabu village: most of its houses either collapsed or became too unsafe to live in. Many villagers moved to the foot of the mountains where Nzidabu village is located, with the resources and help of the government or institutions outside. This village is in some senses dissolving, although twothirds or perhaps half of the households have stayed up on the mountain. This might be a chance for future research using a historical perspective, to see whether and to what extent the coufang can maintain its ritual and social significance after experiencing such a huge disaster. And that will surely give us
56
Liu Biyun
a more comprehensive understanding of the coufang within its more recent context.
Notes 1. The Qiang described here speak the southern Qiang dialect of the Qiang branch of the Tibeto-Burman subfamily of Sino-Tibetan. The language spoken by the nearby Tibetans is the rGyalrong language, which is also a member of the Qiang branch of TibetoBurman language subfamily ( Matisoff 1991: 482; also Ting-yu Wang, this volume). I do not give the tone marks for the Sichuan Mandarin terms; for the Southern Qiang terms, I give the tones in superscript, following village pronunciation as closely as possible. 2. People in the village explained to me that the term coufang means ‘people under the feet of one ancestor’ (個祖宗腳下的), that is, all the descendants of the same ancestor. The Southern Qiang term is du55na55. Coufang is the Sichuan Mandarin pronunciation. I asked some old men how to transcribe coufang into Modern Standard Chinese (MSC), and they said cou was probably zu 族, and fang the character fang 房. Currently, the term in most common use locally is the Sichuan Mandarin coufang. This is the term I use in this chapter. 3. Nzidabu is a transliteration of the Southern Qiang pronunciation of the village name. The situation there changed a great deal after the 2008 Sichuan earthquake. Lixian is next to Wenchuan County and sustained serious damage during the earthquake. Although, thankfully, none of my villagers died as a result of the earthquake, the composition of villages, arrangement of buildings and other landscape details described in my earlier MA thesis have become sealed in the past. 4. Wang Ming-Ke, in his book The Qiang Between the Han and the Tibetan (2003), points out that ‘Qiang’ as a term for groups has referred to the western Others of the Han Chinese since the thirteenth century bce. He discusses the historical processes of the westward movements and shrinking of ‘the belt of Qiang population’, and concludes that the term ‘Qiang’ was representative of peoples living on the western peripheries of the Han. Peoples subject to the Han for whom the term ‘Qiang’ was used were prone to geographical mobility, and the meanings of the term itself differed from dynasty to dynasty. As a result, the use of this term to indicate people in general, as well as actual groups, should be considered in each particular historical context: ‘Qiang’ has been a significant term in the past incorporating ‘non-Han’ and ‘non-Tibetan’. See Wang 2003, Chapter 6. 5. The local chiefs in south-west China had been pacified by the Chinese dynasties since Yuan times (1271–1368). The policy of the first emperor of the Yuan dynasty was to allow the local chiefs to maintain their independent regimes and systems of inheritance as long as the chiefs showed their loyalty to the Chinese court. This was the primary form of the tusi (土司) system in which tusi meant both the chief and the system as a whole (Li Shiyu 1990: 467). The people under tusi rule were not ordinary subjects of the Chinese Empire: they were subject to regular payment of tribute to the emperor and were required to provide soldiers to the empire in wartime, but otherwise were permitted to live under their own laws within the areas under tusi jurisdiction (Ran Guangrong et al. 1984: 228–36). 6. Historical records indicate that the households and neighbourhoods were willing to be incorporated into the Han units (漢甲, hanjia) and become regular Chinese subjects (民, min). In these same records, these segments of the population are called Qiang min (羌 民). DGMZZ (1831) 1992: 347–49. 7. In the 1990s the population there was 30% Qiang and 45% Tibetan, the remainder being mostly Han. See Lixianzhi 1997: 141. 8. See TZZLLFTZ (1866) 1992: 771–77.
From Kinship to State and Back Again57
9. At many points the local gazetteer, TZZLLFTZ, describes the Zagu tusi and their subjects as Tibetans. One passage describes the ancestors of Zagu tusi as being the sons of a Tibetan woman who married the Han commander stationed in the fort in the Tang dynasty, while in the ‘Indigenous Culture’ (夷俗, yisu) section another directly mentions that those aborigines in Zagu and Suomo were tubo (吐蕃, Tibetan). See TZZLLFTZ (1866) 1992: 754–55, 767. On the replacement of the Zagu tusi regime, see SCTZ (1816) 1967: 3111; TZZLLFTZ (1866) 1992: 764. 10. The five tun were: (1) Shang Mengdong tun 上孟董屯, (2) Xia Mengdong tun 下孟董屯, (3) Jiuzi tun 九子屯, (4) Zagunao tun 雜谷腦屯 and (5) Ganbao tun 乾堡屯. The other four tusi were: (1) Suomo tusi 梭磨土司, (2) Zhoukeji tusi 卓克基土司, (3) Songgang tusi 松岡土司 and (4) Dangba tusi 黨壩土司. In addition, there were also other districts that were within the jurisdiction of the Lifanting: six li 里, nine ku 枯, ten zhai 砦 and three fan 番; only people living in the six li were recorded as Han subjects. See TZZLLFTZ (1866) 1992: 656–57. 11. MGSPXZ (1924) 1992: 143. 12. See Baoxianzhi (1748) 2001: 358–59, 393; TZZLLFTZ (1866) 1992: 736, 749. 13. When interviewing the elders in the village, I commonly heard it said that their ancestors had once ‘fought upward to Taiwan, downward to Tibet, and inland to Gansu, Lanzhou and Yunnan’ (上打過臺灣, 下打過西藏, 深打過甘肅蘭州雲南二省). Checking the ‘Loyalty and Righteousness’ (忠義, zhongyi) section in the TZZLLFTZ, one can also find many people from the five tun recorded as dying in wars for the empire in other war zones. See TZZLLFTZ (1866) 1992: 730–32. 14. Villagers’ memories of their ancestors includes the stories of their going to join armies, defeating enemies with their intelligence and bravery, contending for the political power of military governance, and their clumsiness regarding courtesy when invited to the parties held by Han officers. Details and analysis of the stories of ancestors and epitaphs on village tombstones can be found in Liu Biyun, 2007. 15. The neighbouring rGyalrong Tibetans practise a different kinship system, without an emphasis on lineage. See Ting-yu Wang’s chapter in this volume. 16. The one exception is the coufang Yang: it has only a Han surname and no Qiang name. The fact that Han surnames do not differentiate between lineage groups (coufang) is a significant detail when comparing the Qiang and the Han Chinese, and examining the relationships between the Chinese empire and its peripheries. Ebrey 1996. 17. There are now some gravesites which record details of the deceased. 18. My investigation of the tombstones showed that they did not always match what the villagers remember, which was that only people who had been government officers had had tombstones erected; some nine had been erected by officials for patrilineal ascendants who had not had official status; in sixteen cases neither the deceased nor the person or persons who erected the tombstone had had any official title. There were another seven which were too illegible to determine. Present-day villagers pay little attention to the inscriptions on the old tombstones. Few, in fact, have ever read them or know what is written there: they know the stories of their glorious ancestors through their oral traditions. For a comparison of oral stories and local history documents, see Liu Biyun 2007. 19. During my fieldwork, I never heard villagers mention any corporate relationship based on communal property, only on common descent. The concept of the common descent of the Qiang villagers is discussed below, and ‘monetary tribute’ to coufang ancestors is mentioned in the context of citang yinzi, 祠堂銀子, ‘silver for the ancestral hall’. This was the only mention of money involving the whole coufang I heard. Comparison of the significance of and practices surrounding Han and Qiang ancestral halls is an issue needing further exploration. 20. People seem to be unconcerned by this: it is the ritual practices themselves that are significant rather than the physical landmarks or the legibility or otherwise of particular inscriptions. Two examples illustrate this. I saw the communal stele of Izi coufang on two
58
Liu Biyun
different occasions. The first time it was standing upright by the village road. When I next saw it one or two years later, it had been hit by a truck and left lying on the field there. I observed that some members of Izi coufang felt unconcerned about leaving it in its fallen state, considering that it did not matter. One old man told me he wanted to put it upright again and when he had the time he would find some people to help him do so. The other example is that of an old and beautiful tombstone of Nziwa coufang, erected in 1815. It was hit by a large rock rolling down from the top of the mountain and was broken in two. The pieces were just placed together at the original place without any plan to repair them. All the ancestral halls I personally observed were broken and overgrown inside, with weeds higher than a normal person, and the inscriptions on many tombstones have become eroded and can no longer be clearly read, but the villagers do not see any need to renovate them. 21. Not all graveyards originally had these stone hitching posts. 22. See Houhanshu (n.d.) 1972: 1021; Baoxianzhi (1748) 2001: 434; DGMZZ (1831) 1992: 335. Also, according to villagers, the crematorium is used when there is a bad death, that is death from sickness, a difficult birth, before the age of sixty, and so forth. Some villagers additionally said that cremation has declined since the 1950s, although a bad death during my stay in the village was still followed by a cremation. Evidence from a local gazetteer indicates that cremation was more widely practised during the tusi era (TZZLLFTZ [1866] 1992: 766). The practice of cremation suggests influence from Tibetan Buddhist culture. 23. This is relevant here because the first three characters, 先理來 xian li lai, are literally ‘first put-in-order come’. 24. According to historical records, the regular arrangement in Jiuzi tun consisted of 1 shoubei, 2 qianzong, 4 bazong and 8 waiwei from the basic contingent, and 1 shoubei, 1 qianzong, 1 bazong and 2 waiwei from a supplementary contingent. See TZZLLFTZ [1866] 1992: 758. English translations of these official titles are taken from Hucker (1985: 433, 561). The arrangements regarding the basic contingent caused a dispute among military officers and some ‘want-to-be’s, finally triggering a local insurrection called xiamengtun muni zhi luan (下孟屯穆逆之亂), an upheaval of Muzusuolang (穆租索朗) in Xia Mengdong tun. This event led to a Qing court decision to pacify this area; an ancestor of Nziwa coufang was deeply involved in this event and became a target of the empire. He escaped with the help of relatives and local people, and spent most of his lifetime outside the village. His son, born outside, finally returned to the village and was the greatgrandfather of a man of Nziwa coufang, one of my primary informants. An official version of this event is recorded in TZZLLFTZ (1866) 1992: 651–52, 826; the local version and details of that ancestor’s tombstone are discussed in detail in my MA thesis (Liu Biyun 2007: 32–37). 25. The term ‘basic contingent’ refers to the contingent of conscripts which met the basic quota set for the village. 26. A jiansheng – also called 民生, minsheng – was a Student by Purchase Fourth Class, a subcategory of Student by Purchase (例監, lijian) in the Imperial Academy. Lijian were men admitted to the Imperial Academy without having passed any level of the civil service recruitment examination, in recognition of their contributions of grain or money to the state (Hucker 1985: 150). Other titles or designations of scholar-officials I observed on the tombstones were 文生, wensheng, student, and 撰書, zhuanshu, scribe (an unranked sub-official). 27. This phrase is taken from Zhongyong (The Doctrine of the Mean), one of the Confucian classics. YYGJSSDBXY [n.d.] 1952: 18–20. 28. There seem to be some inconsistencies here. The kinship terms inscribed on the front of this tombstone (A) indicate that it was the son who erected the tombstone for his parents, but the writer of the epitaph on the back addresses the dead as ‘grandparents’. A reasonable supposition is that it was actually the ‘grandsons’ who erected the tombstone for their grandparents on behalf of their father, i.e. the ‘son’ of the ‘grandfather’; this is
From Kinship to State and Back Again59
supported by the fact that there is another tombstone (B) next to this one (A), on which the names of the dead are the ‘son’ and the ‘son’s wife’ on (A). Tombstone (B) was erected by the ‘grandsons’ the year after the first tombstone (A) was erected. There are many other examples of inconsistencies of this kind in the village. 29. The zibei poem of Nziwa coufang is ‘現開華國瑞, 文德重興邦, 俊秀朝廷選, 禎祥發錦江’ (xiankai huaguo rui, wende zhong xingbang, junxiu chaoting xuan, zhenxiang fa jinjiang); that of Izi is ‘福德龍光明, 永遠鳳林芳, 輔弼繼文武, 秀玉至朝堂’ (fude long guangming, yongyuan fenglin fang, fubi ji wenwu, xiuyu zhi chaotang); that of Rudi is ‘開木萬騰源, 榮 華總後先, 衍慶芝蘭茂, 家聲自此綿’ (kaimu wanteng yuan, ronghua zong houxian, yanqing zhilan mao, jiasheng zici mian); that of Ruwabo is ‘開文萬代德, 家傳裕後昌, 吉 星增福壽, 永世慶長春’ (kaiwen wandai de, jiachuan yu houchang, jixing zeng fushou, yongshi qing changchun); that of Yang is ‘萬福廷邦國家用, 茂蘭香芝春時芳’ (wanfu tingbang guojia yong, maolan xiangzhi chunshi fang). 30. The following example illustrates the youngest son’s privileged right to the ownership of a house. A fifty-year-old man told me how he bought his present house from his youngest brother. This man is the second of four brothers. The oldest brother left the house after his marriage, the third one married his wife uxorilocally, and the youngest one left the village to study and subsequently decided not to come back and settle there with his family. The second brother wanted to keep their natal house, but he did not have the right to it. He finally gave his brother 2,400 renminbi and still owns the house. This happened twenty years ago, although other people said that the man exaggerated the cost. I heard of other examples in the village demonstrating the convention of the youngest brother or youngest son having the privilege of inheriting the house. 31. The wife-taker should pay many fengfeng (封封), a gift of cash presented in a red envelope, to the parties of the wife-giver to thank those who give special assistance at the wedding. If the wife-taker is from another village, one of the fengfeng, called citang yinzi, is given to the coufang ancestors. If the wife-takers are from within the village, there is no need for them to pay the citang yinzi. The money (silver) is now more symbolic than practical, only amounting to 1.2 renminbi. 32. The villagers described to me what the wooden structure housing kuenxin Buddha had looked like and where it had been located, and said that there had even been a literate man who inscribed four Chinese characters, 櫆星菩薩, on it. A teacher at a senior high school downtown, approximately a dozen kilometres away, told me that the 櫆星 figure had been designed by Kongming (孔明), a famous late Eastern Han dynasty politician. 33. Like the stupas, the kuen55xin55 Buddha was destroyed in the 1960s. 34. Further investigation of this topic is an important and potentially fruitful direction for future research. 35. Relevant historical records mentioning these economic activities can be found in Houhanshu (n.d.) 1972: 1021; Huayang guozhi (354) 1972: 144–45; Baoxianzhi (1748) 2001: 436; DGMZZ (1831) 1992: 335.
References Primary Historical Literature Chang Qu 常璩. (354) 1972. Huayang guozhi 華陽國志 [Records of the Lands South of Mt. Hua]. Taipei: Hongye Shuju. Chen Kesheng 陳克繩. (1748) 2001. Baoxianzhi 保縣志 [Gazetteer of Bao County]. In Sichuan fuzhou xianzhi 四川府州縣志 vol. 17, Gugong bowuyuan, ed. Haiko: Hainan Chubanshe, pp. 303–441. Fan Ye 范曄. (398–445). (n.d.) 1972. Houhanshu 後漢書 [The History of the Later Han Dynasty]. Taipei: Yiwen Chubanshe.
60
Liu Biyun
Wu Gengmei 吳羹梅, and Zhou Zuoyi 周祚嶧. (1866) 1992. Tongzhi zhili lifanting zhi (TZZLLFTZ) 同治直隸理番廳志 [Gazetteer of Lifan Sub-Prefecture under direct jurisdiction during the Tongzhi reign period]. Repr. in Minguo Songpan xianzhi 民國松潘縣志 [Gazetteer of Songpan County in R.O.C. times], Zhongguo difangzhi jicheng bianji weiyuanhui, ed. Chengdu: Bashu Shushe, pp. 641–832. Yang Jiayi 楊迦懌, and Liu Futing 劉輔廷. (1831) 1992. Daoguang maozhou zhi (DGMZZ) 道光茂州志 [Gazetteer of Maozhou during the Daoguang reign period]. Repr. in Minguo Songpan xianzhi 民國松潘縣志 [Gazetteer of Songpan County in R.O.C. times], Zhongguo difangzhi jicheng bianji weiyuanhui, ed. Chengdu: Bashu Shushe, pp. 317–405. Yang Fangcan [et al.] 楊芳燦等. (1816) 1967. Sichuan tongzhi (SCTZ) 四川通志 [General Gazeteer of Sichuan]. Taipei: Jinghua Chubanshe. Zhang Dian 張典, and Xu Xiang 徐湘. (1924) 1992. Minguo Songpan xianzhi (MGSPXZ) 民國松潘縣志 [Gazetteer of Songpan County in R.O.C times]. Zhongguo difangzhi jicheng bianji weiyuanhui, ed. Chengdu: Bashu Shushe, pp. 1–316.
Secondary Literature Chen Chi-Nan 陳其南. 1990. ‘Fang’ yu chuantong zhongguo jiazu zhidu: jianlun xifang renleixue de zhongguo jiazu yanjiu 「房」與傳統中國家族制度:兼論 西方人類學的中國家族研究 [The ‘fang’ and the Traditional Chinese Lineage Institution: And Studies of Chinese Lineage in Western Anthropology]. In Jiazu yu shehui: Taiwan han Zhongguo shehui yanjiu de jichu linian 家族與社會:臺灣和中 國社會研究的基礎理念 [Lineage and Society: Fundamental Research Concepts in Taiwan and China]. Taipei: Lianjing Chubanshe, pp. 129–213. Ebrey, Patricia. 1996. Surnames and Han Chinese Identity. In Negotiating Ethnicities in China and Taiwan, Melissa J. Brown, ed. Seattle: Washington University Press, pp. 19–36. Faure, David. 2007. Empire and Ancestors: State and Lineage in South China. Stanford: Stanford University. Hucker, Charles O. 1985. A Dictionary of Official Titles in Imperial China. Taipei: Southern Materials Center. Li Shiyu 李世愉. 1990. Luelun tusi zhidu yu gaitu guiliu 略論土司制度與改土歸 流 [Brief Arguments on the Tusi Institution and the Policy of Gaitu Guiliu]. In Zhongguo gudai bianjiang zhengce yanjiu 中國古代邊疆政策研究 [Research on the Policies of Old China to its Frontiers], Ma Dazheng, ed. Beijing: Zhongguo Shehui Kexue Chubanshe, pp. 465–94. Liu Biyun 劉碧雲. 2007. ‘Sandai de qinqi, wandai de zufang’: Sichuan lixian shuitang qiangren de qinshu yu shehui guanxi. 三代的親戚,萬代的族房:四川理縣水塘 羌人的親屬與社會關係 [‘Kinship of Three Generations, Coufang Ten Thousand’: Kinship and Social Relations in Shuitang Qiang Village in Lixian County, Sichuan]. Taiwan: National Tsing Hua University, Master’s thesis. Lixianzhi bianzuan weiyuanhui 理縣志編纂委員會. 1997. Lixianzhi 理縣志 [Gazetteer of Lixian County]. Chengdu: Sichuan Minzu Chubanshe. Matisoff, James A. 1991. Sino-Tibetan Linguistics: Present State and Future Prospects. Annual Review of Anthropology 20: 469–504. Ran Guangrong 冉光榮, Li Shaoming 李绍明, and Zhou Xiyin 周錫銀. 1984. Qiangzu shi 羌族史 [History of the Qiang]. Chengdu: Sichuan Minzu Chubanshe.
From Kinship to State and Back Again61
Shen Zhifang 沈知方 ed. [n.d.] 1952. Yuyi guangjie sishu duben xueyong (YYGJSSDBXY) 語譯廣解四書讀本, 學庸 [Reading of the Four Books: The Great Learning and the Doctrine of the Mean]. Taipei: Qiming Shuju. Wang MingKe 王明珂. 2003. Qiang zai han zang zhijian: yige huaxia bianyuan de lishi renleixue yanjiu 羌在漢藏之間:一個華夏邊緣的歷史人類學研究 [The Qiang – Between the Han and the Tibetans: A Historical and Anthropological Study of the Huaxia Borderlands]. Taipei: Lianjing Chubanshe. Wang Ting-yu 王廷宇. 2012. ‘The House, the State and Change: The Modernity of Sichuan rGyalrong Tibetans’ (this volume).
Chapter 3 Embroidery Speaks: What Does Miao Embroidery Tell Us? Ho Zhaohua
Introduction Clothing is a cultural medium through which people speak: it is a complex phenomenon deeply involved in identity and cross-cultural boundary setting. Examining the concept of ‘style’ contributes to our understanding of material culture. So how is the making of clothes significant for cultural and social reproduction? McCracken suggests we rethink the relationship between culture and consumption. He argues that the social sciences have been slow to see the relationship between these, and slower still to take seriously the idea that consumption can contribute to cultural and social reproduction (1990: xi). The clothing examined in this chapter is not merely a consumption good, but a system of cultural ideology: clothing is shaped, driven and constrained by cultural considerations. The designing and making of clothes is an entirely cultural enterprise, and what stitching techniques are used and how the style is accomplished are matters charged with cultural meaning. As my informants there told me: ‘In Shidong, we don’t pay attention to food, and we don’t care whether we live in a dilapidated house, but for you and your family to have any dignity, you have to dress well. That’s why women must be able to do embroidery. It is as important as your life.’
The Miao in Shidong Shidong is a small river-port township in the middle reaches of the Qingshuijiang River in Taijiang County in south-east Guizhou Miao and Dong Autonomous Prefecture in Guizhou Province. According to the Fifth Population Census (2000), the population of Shidong was 12,332, of whom more than 95 per cent were ethnic Miao.1
Embroidery Speaks: What Does Miao Embroidery Tell Us?
63
The Miao constitute one of China’s official major ethnic minorities; most live in Guizhou, Guangxi, Yunnan, Hunan and Sichuan provinces. The name ‘Miao’ may mislead one to regard these approximately eight million people as a single group, but it should be remembered that they are considered to be one of China’s major ethnic groups as a result of the ethnic minority policy (minzhu shibie, 民族識別) which the government of the People’s Republic has pursued since the 1950s. People designated ‘Miao’ are also found in a number of South-East Asian countries, and there is a small diaspora further afield. There are some eight million speakers of the three Miao languages and the thirty to forty highly mutually intelligible dialects which form the Miao branch of the Hmong-Mien (Miao-Yao) language group in the Sino-Tibetan language family (QMDZDBW 2000: 143–49). Shidong Miao speak the East Guizhou dialect.
History Shidong has had long experience of contact with Han culture. Historically, this region was deeply impacted by the Chinese Empire, politically and militarily. The Miao proved themselves difficult to assimilate and control. Essentially they retained their sense of being an independent people in their own vigorous cultural landscape, in complete contrast to their being engulfed by a dominant Han Chinese culture, a scenario that might be readily imagined. Chinese historical records from the Ming and Qing dynasties (1368–1911) use the terms ‘Miao’ and ‘Man’ (a Han term for ‘barbarian’) to designate the indigenous peoples in regions populated by the Miao and Yao, and the Qing government used the Great Miao Wall to isolate the Miao people. Such documents suggest that Shidong was a wild place where barbarians lived, and lacking cultural accomplishments. Indeed the Qing military action of 1728, when Ertai sent Zhang Guangsi’s troops into the region, was the first time that imperial power reached deep into Miao areas. Although Zhang Guangsi was renowned for his violent military extermination of native peoples in Guizhou, this Qing military force was unsuccessful in colonizing the region. The Qing government employed a policy of ‘using barbarians to rule barbarians’ ( yi yi zhi yi, 以夷制 夷) to assimilate and sinicize the Miao, and a Confucian school ( yixue, 義學) was established near Shidong in 1730 to assimilate the Miao ideologically into the Chinese administrative system. By 1726, river ports such as Shidong were serving as trade centres along the major transport route between Guizhou and Hunan provinces, dealing in items such as timber, tung oil, opium, cotton fabric and coal. Merchant guild halls2 were built along both banks of the Qingshuijiang River in the mid-nineteenth century to promote business and trade. Chinese merchants and soldier-settler colonists began to settle the territory along the river in the late Qing period (Taijiang Xianzhi 1994). Informants told me that by Republican times the river was crowded with more than one hundred vessels every market day, that is,
64
Ho Zhaohua
every six days. When Chinese commerce was forced to move inland during the Anti-Japanese War, Shidong again became a major trade centre. An elementary school was established in Shidong under the late Qing dynasty in 1906, and a girl’s school was opened providing access to Chinese literacy for Miao women in 1914.3 Nevertheless most Miao women had no opportunity to be formally educated until the 1980s. Women in traditional China were generally deprived of a classical education because of their gender (Chakrabarti 1998). Chakrabarti quotes a survey of rural areas carried out by John L. Buck in the 1930s, which found that only 2 per cent of the female population aged 7 years and above had ever attended school, and that only 1 per cent could read Chinese (Buck 1937: 373). Even after 1949, according to the 1982 census, 90 per cent of mainland Chinese women aged 55 to 59 were still illiterate (Chakrabarti 1998). Shidong women can hardly be considered an exception. Local people explained that, although school facilities had been established, Miao people considered that from a moral point of view it was not suitable for girls to stay at school all day long with boys: that would be considered lacking a proper upbringing. Furthermore, local people believed that girls and boys needed different forms of education to become ‘cultivated’: as a local saying goes, ‘Men learn to read to be an officer; a woman who reads will become hungry and thirsty; only needlework can feed a hungry belly and make one rich.’4 This attitude led to high rates of illiteracy among Shidong Miao women. More than fifty thousand tourists a year are now attracted to this relatively tiny town. There has been no difference between the clothing of Shidong Miao men and Han Chinese men for decades. Miao women, however, still ordinarily wear their distinctive traditional Miao clothing. Their gorgeously embroidered jackets worn for festive occasions, such as the Sisters Rice Festival in the third month of the lunar calendar and the Dragon Boat Festival and races in the fifth lunar month, have made Shidong a focus of attention all over the world. Many textile lovers come all year round to enjoy and study the exquisite workmanship and intriguing stories the imagery depicts. Commercially, Shidong embroidery now enjoys a high profile and is a valuable commodity in the art and tourist industries (Oakes 2003).
Miao Culture The Miao attach great importance to maintaining their own cultural selfidentity: expressions of Shidong Miao cultural identity lie at the heart of their integrating rice production and (through marriage and childbirth) social reproduction into their social construction of time, particularly during various festivals. This integration is the basis of the social and cultural fabric (Ho 2009: 7–59). Shidong Miao use the Chinese lunar calendar,5 but the calendar does not explain the particular dates or the various rituals, festivals and ceremonies which construct Shidong Miao social life, that is, time as a social construct.
Embroidery Speaks: What Does Miao Embroidery Tell Us?
65
Rather than being enveloped by Chinese culture, its calendar and the attendant sequencing of the year’s activities, Shidong Miao people made adjustments to maintain their own cultural ideals, accommodating Chinese culture, including elements of its construction of time, without surrendering their cultural identity. Traditional expectations regarding Miao social roles are broadly expressed in a Miao idiom: ‘Men are responsible for food, women for taking care of clothes; both are responsible for bringing in money.’6 To be a cultivated person with the ability to make money is considered a sign of success. One informant, Wu Yongsheng, listed the traditional attributes that characterize (now and historically) an accomplished Miao man: being able to read and write Chinese, and sing Miao songs; and serving in government or doing business. In contrast, the traditional criteria for an accomplished woman are: having mastered fine skills such as embroidery, weaving, dyeing and making festival clothes; and being able to sing, dance, speak well, tell Miao stories, and bring in money. A cultivated Miao woman possesses the highly sophisticated ‘household’ skills and modes of communication which represent the local knowledge system. Some aspects of this are presented in the discussion below, addressing the issue that McCracken raises.
The Social Significance of Embroidery Oakes observes of current-day Shidong that embroidery ‘represents highly skilled work that can generate cash income for households’ (2003). He was principally concerned with investigating the commercial embroidery industry in Shidong but also observes that ‘embroidery is a common but unevenly mastered skill’; nevertheless, ‘regardless of the varying skill levels, all women were expected to learn embroidery’ (ibid.). He notes that good needle-workers could still command considerable respect among the women of Shidong (ibid.), although he does not comment on its importance for a woman to enjoy understanding and respect. Weaving and sewing traditional clothes clearly remain valued skills: girls start learning the art of embroidery at six or seven years of age. After a woman marries, it is her duty to make clothes for not only her husband and children but her husband’s parents as well, otherwise she will not be considered a good wife or mother, and far more seriously, simply not a Shidong woman. In fact a girl who does not master a repertoire of embroidery skills will be harshly criticized by her mother and ridiculed by her relatives and neighbours. One young woman, Zhenmei, had gone to work as a teenager, which left her no time to spend practising and improving embroidery.7 As a result, her embroidered clothing did not meet what is considered an appropriate standard, and she suffered public disdain. She told me it was a painful experience; she felt that many people ridiculed her secretly, and her mother constantly criticized her. She disliked the gossip intensely, but could not escape the situation without
66
Ho Zhaohua
making a great effort to improve her embroidery skills or becoming rich so as to be able to hire skilful women to make clothes for her. But the repercussions did not end there; they went on to affect her marriageability. Her poor embroidery skills and the insufficient number of items she had completed led her to being considered ‘a lazy girl’, and she could not find a marriage partner in Shidong – only a poor family relatively distant would consider her. In the end she married a Han Chinese man from Kaili City – not the ideal choice for a Shidong Miao woman. Oakes’s observations and Zhenmei’s case reflect the central role that doing embroidery plays for local women. This is clearly indicated by commonly heard Shidong Miao sayings which Zhenmei told me: dot sux ait nangl ax jangx fangb ait nangl dot vut ax jangx naix ait hmub dios gid jus ait nongx ait nangl ait nongx ait nangl jus dlas dlas xiangl xiangl
If a person does not learn how to make clothes, she will not become a person like we are; If a person does not learn how to make clothes well, she will not be considered an accomplished person. If a person learns how to master embroidery, she has grasped the skills of survival and living. If a person has mastered the skills of survival and living, she has grasped the chance for success and fortune.
A woman must be able to make clothes displaying her embroidery skills within the design guidelines and style constraints of her community. Jackets in all of eleven styles are the most valuable part of her dowry when a Shidong Miao woman marries.8 A ‘requesting clothes’ ritual is one of the highlights, an occasion to see these jackets displaying the financial and craft standards of the bride’s family, but also a highly charged occasion when village women are overcritical. No matter whether the new couple is accorded admiration or disdain, it places huge pressure on the bride’s family. They resort to borrowing beautiful clothes from sisters or relatives to meet quality and quantity standards, social expectations, and to satisfy the groom’s family. People are well aware of this – every family does it – but the ‘deception’ is not exposed. Instead, they say that such ‘beautiful clothes are hard to make’. One informant, San Meibao, pointed out: ‘To maintain one’s dignity is more important than keeping clothes; clothes will eventually fall to pieces – relations have to be maintained for the long-term, and not get lost.’ Mastery of various techniques, such as stitches and ornamentation, are basic criteria for determining the best embroidery work. Beyond this, being able to create ‘fresh’ images is a most important element in making clothes express one’s insights with heart and mind.9 A new fresh look is greatly admired and inspires imitation: that is, it is the Shidong Miao form of being ‘fashionable’, while the freshness is considered a metaphor for the vitality and life energy of youth. Why are ‘freshness’ and ‘fashion-setting’ important assessment criteria? Another informant, Xiaomei, told me: ‘Shidong people think that humans are the same as trees; if they keep shedding their old bark it will keep life continuing
Embroidery Speaks: What Does Miao Embroidery Tell Us?
67
and growing.’ She said Shidong people think it impressive if someone can create a fresh new look on clothes. But she also commented on the months, sometimes years of work involved in impressing people – it is a hard task. ‘In order to be admired, we have to endure all the hard work of embroidery, weaving and dyeing; all our efforts are to avoid suffering the fate of being ridiculed. … In Shidong … women must be able to do embroidery. It is as important as your life.’
Embroidered Miao Clothes Chinese anthropologists have identified more than a hundred kinds of Miao clothing in their field investigations, each with their own characteristic features: ‘There are many kinds of Miao clothing, just as there are many kinds of beautiful flower’ (Clothing and Ornaments of China’s Miao People 1985: 4). For the Shidong Miao, clothes are a cultural representation expressing their collective identity, not just that of the individual. A Shidong Miao proverb says: ‘Different kinds of animal have different patterns on their skin’; the function of embroidering clothes is to create images appropriate to their own cultural identity. Mais hliob Hua’s case outlined below by Mais hliob Sanb illustrates the substantial importance of this and of the assimilation process by which a bride is transformed into a member of the Shidong community by changing her clothes – particularly her jacket and apron styles, and hairstyle. Mais hliob Sanb said: ‘Mais hliob Hua married into Shidong from Ngul Kib. The styles and colors of her clothes were quite different from Dangk Vongx. So every time people gathered for dancing and singing, she had to endure being stared at and criticized. She had to make great effort to change her clothes’. I once met Mais hliob Hua and asked her about the changing-clothes event. She was happy to tell me: Before it all happened, I didn’t know that I would marry into your place, Dangk Vongx. When I was a girl, I worked very hard doing embroidery and making clothes. I liked being appreciated as a capable girl. Unexpectedly, the embroidery and clothes in my place were different from yours, so the clothes I brought were unsuitable. I felt ashamed to wear them. It was very uncomfortable to be stared at all the time. I asked Mais hliob Sanb to teach me to do clothes like yours. It took me several years to change my clothes. The process was very tiresome; I was marrying into a wrong place.’ (Narrated by Mais hliob Hua, August 2007)
Assimilation, or making outsiders part of the in-group, and also quarantine, or excluding former insiders, both play a significant role in controlling marriage by enforcing the boundary between other and self (see Ho 2011: 413).
68
Ho Zhaohua
Basic Classification of ‘Style’ in Shidong Miao Women’s Clothing Traditional women’s clothing falls into two categories: special clothing for festivals and special occasions; and ordinary daily wear. The difference between the two is mainly determined by the patterns and images embroidered and stitches used on the jackets worn over a long or medium-length pleated skirt. Festival jackets are gorgeously embroidered. By contrast, clothing worn daily has less and simpler embroidery. Traditionally, older women wear clothes that are as plain and conservative as possible. Marriageable girls and newly married women should wear bright, gorgeous, sumptuous jackets for festivals. No matter whether for festivals or for ordinary daily wear, the form of a jacket and the colours used remain the same. A jacket has straight sleeves 40–45 cm long, front panels about 75 cm long, and the back is about 60 cm long. It is made by combining a number of rectangular pieces of fabric with no arc-shape cut in the armpit, as shown in Diagram 3.1 and Table 3.1.10 Women wear the left lapel of the collar over the right one.11 Embroidery decorates specific set areas on the lapels, collar, shoulder line and sleeves; only the images and details depicted can vary.
765
567
10
1
2
2
1
4
3
8
Diagram 3.1. Shidong Miao Jacket (by author)
9
Embroidery Speaks: What Does Miao Embroidery Tell Us?
69
Table 3.1 Names of decorated sections of Shidong Miao jackets No. Miao
Chinese
English
①
ghab mongl gangb
衣袖花
main sleeve body
②
hfud mongl gangb
袖頭花
upper sleeve
③
ghab dab hmub; ghab dab dux 小花邊
④
ghab dlok
前襟及領背堆花 lapel and collar central area
⑤
guf jil seet
脊背松針花
shoulder section border
⑥
guf jil gangb
肩上花
shoulder section
⑦
bangx
領、襟部位邊花 woven lapel and collar outer borders
⑧
hlat
帶
belt
⑨
linx pat
腳踩的布
twill: handwoven cloth
⑩
linx dix
手打的布
qi: handwoven cloth
narrow border
The body fabric is black; the main colours used for decoration are various shades of red and blue. Two major categories based on these decorative colours are ub xok, red clothing, and ub zek, blue clothing. The term dlub indicates bright colours and is often used instead of xok to mean ‘red’; zek indicates dark colours, usually blues. Jackets are divided into the various categories set out in Table 3.2 and Table 3.3. Table 3.2 Main categories of festival jackets in Shidong Miao term
Main stitches
Comment
hob mongl 破線繡花衣 Subcategories:
satin stitch
hob mongl dlub
red satin stitch
– very expensive – made with stencil patterns – unsuitable for silver ornaments because the looser thread would break under the weight
hob mongl zek
blue satin stitch
70
Ho Zhaohua
Table 3.2 continued Miao term
Main stitches
Comment
ob ait See below 兩層花衣 There are 3 kinds of ob ait – each can be further subdivided into blue – zek (dark), and red – dlub (bright). 3 Subcategories:
All 3 kinds of ob ait have 3 strips on the sleeves. The upper and lower strips always have horizontal patterning, i.e. like bands going around the arm. The middle section has vertical patterns, resembling stripes going up and down around the arm at that part.
satin stitch 1, ob ait pul duf 兩層夾破線繡直條花衣 for all strips
– very expensive. – upper and lower strips are made with stencil patterns. –m iddle section always uses satin stitch.
* satin stitch – u pper and lower strips are 2. ob ait duf made with stencil patterns. 兩層夾破線繡與挑花直 for upper & lower strips; 條花衣 * mixture of satin stitch & countedthread for middle strip 3. ob ait pul 兩層夾挑花直條花衣
*satin stitch for upper & lower strips; *countedthread for middle strip
- less expensive. - upper & lower strips are made with stencil patterns
khait mongl 紮花花衣 – dlub (bright) and zek (dark) are added to distinguish two different styles of khait mongl, red and blue
Basically countedthread
geometric patterns using countedthread
Embroidery Speaks: What Does Miao Embroidery Tell Us?
71
Table 3.2 continued Miao term
Main stitches
Comment
hob jub 不破線花衣 - has both red and blue colours
Stem stitch
- special style for marriageable girls - style closely resembles hob mongl, but uses stem stitch. Satin stitch is looser: stem stitch is firmer, and so can take the weight of silver ornaments
There are also five styles of ordinary daily wear jackets, listed in Table 3.3, which fall into two main styles: ib ait, and ub jub bangx. The first, ib ait, can be further subdivided into four. These jackets are for wearing outside the home, appropriate for visiting friends or relatives.12 Older women often wear these jackets for festivals. Table 3.3 Various styles of ordinary daily wear jackets in Shidong Five styles of ordinary daily jackets Local name
Local name of subcategory
Sleeve embroidery details of four ib ait variations
1. ib ait 一層花衣 Sleeve embroidery with one strip of satin stitch
ib ait dlas dlas 一層子孫花衣
lower strip – satin stitch middle part – counted-thread stitch with vertical children images represent their crossing a bridge over a river/on a road patterns upper strip – counted-thread stitch depicting seven houses with women or flowers between them
ib ait deik dent 一層破線繡花衣
lower strip – satin stitch middle and upper parts – countedthread middle part – usually depicts flowers, birds and butterflies upper strip – depicts seven houses with birds and flowers between them
72
Ho Zhaohua
Table 3.3 continued Local name
Local name of subcategory
Sleeve embroidery details of four ib ait variations
ub mongl dlub 一層破線繡亮色 花衣
Similar to above, but simpler – usually worn by older women. The houses remain but only have one flower between each of them. The embroidery is less dense. Strips are bright blue satin stitch
2. ub jub bangx
ub mongl zek 一層破線繡暗色 花衣
Strips are dark blue satin stitch
ub jub bangx 織帶衣
Sleeves have belt-weaving decoration at the shoulder end. The simplest style of all.
Festival jackets are usually embroidered in satin stitch using patterns bought in the market. Ordinary daily wear uses household designs, and the stitching is mostly counted-thread: precision is the prized quality. Some styles combine both kinds of stitch and design. For example, the upper and lower satin-stitch strip-design elements of the ob ait pul style come from the market, the middle vertical sections of counted-thread from the household. Interestingly, the Shidong Miao consider that different approaches to design can be combined within one style of clothing. The bright hob mongl festival jackets, for example, incorporate different images and patterns visuals onto sleeves using any of six design types (Table 3.4). The sleeves usually have symmetry of design: if the left sleeve has a thematic pattern, the right sleeve is usually the same. Table 3.4 Styles of hob mongl festival jackets Overall Design-Structure Styles Ref. No.
English name & comment
Local name and Chinese term
1
thematic ib daib bod – requires high quality skill and aesthetic 一個寶 expression
2
2 strips ob bax – requires less time and labour 雙排 – people consider this style less attractive and of less value
Embroidery Speaks: What Does Miao Embroidery Tell Us?
73
Table 3.4 continued Ref. No.
English name & comment
Local name and Chinese term
3
3 strips – of less value
bib bax 三排
4
2 strips, with 1 narrow dividing strip ob bax jat ib daib ghad daib – requires high quality skill and aesthetic humb expression 兩排中夾一條小花邊
5
2 strips, with 2 narrow dividing strips ob bax jat ib daib ghad daib humb 兩排中夾兩條小花邊
6
3 strips, with 2 narrow dividing strips bib bax jat ob daib ghad daib – requires a lot of time and labour humb – very highly valued 三排中夾兩條小花邊
Patterns for Embroidery In the Shidong area, jacket sleeves are decorated with highly elaborate embroidery: Miao women have a great interest in designing and creating effects to be admired and imitated. Embroidery design falls into two major categories: the first is hnent hmub, comprising the fundamental patterns and techniques passed from mother to daughter in the pattern portfolios called bend hmub. Embroidery in this category requires precise needlework that is considered to represent character and morality. It falls into the general class of hvot, ‘stroke’, meaning a single stroke of a pen. The hvot are counted-thread geometric designs originating within households. Precision stitching is highly valued. The second category of design, gangb hmub, comprises the splendid, beautiful jackets worn by the upper class, representing their wealth and abundance. The word gangb means ‘beautiful insect’. This category encompasses figurative images such as butterflies, birds, fish, dragons, flowers, and so on. Gorgeous images and dedicated, expressive skills are highly prized. Designers draw, cut and sell ‘paper patterns’ or stencils in the market. Only five or six women in Shidong are able to do this, whereas almost every woman can produce ‘counted-thread’ embroidery, so these pattern designers have a great impact on Shidong Miao fashion. The women buying them incorporate the images they depict into their own design compositions. One pattern designer, Grandmother Sanmei, told me that paper patterns only became a commodity after the 1980s Reform and Opening Up era.13 Before that people had always come to her home to ‘request patterns’. As a little girl she learned how to draw, cut and make new patterns from the older
74
Ho Zhaohua
women, and now women in that village often asked her to help. She married during the Cultural Revolution. When she and her husband could not make enough money from growing rice alone, poverty pushed her to sell ‘paper patterns’ in the market. She says only a few women in Shidong township can design ‘paper patterns’: Grandmother Tao is skilled at thematic and detailed designs, Grandmother Ying is good at neat, meticulous ones, and her own are more dramatic. The buyers choose paper patterns based on how mature their embroidery skills are: ‘Only a skilful woman would choose thematic or dramatic imagery. A skilled embroiderer who does so can then lead the fashion. Everybody will admire her when she dresses up and attends the festival parades.’ A good design must incorporate both variety and consistency: the designer must be able to put complicated elements into a composition harmoniously, and ideally create an innovative look. Grandmother Sanmei had to create completely new patterns once every two to three years and frequently make new arrangements of the elements of her current offerings: ‘Seeking fresh innovative designs is what drives making new clothes in Shidong.’ San Meibao observed: ‘Why do we Shidong Miao have so many festivals … for women to contest and see whose clothing is most beautiful? If someone has a new design, we follow up as soon as possible: this is “fashion”. And we have to think hard to create new designs which other people don’t have! Are we fools? Or we are just very smart and never tire of this competition?’ According to Grandmother Sanmei, ‘We can’t change the style [of our jackets], but we have to have a variety of different patterns and designs on different jackets. By doing that, we can have face.’ This is how Shidong Miao women enjoy status and respect.14
Imagery and Stitches on Miao Clothes As mentioned above, a Shidong Miao woman’s social status is inextricably connected with her embroidery, the stitching and images being key factors providing a Shidong Miao woman with access to standing and prestige. Delicate needlework reflects personal qualities; dense stitches and control of tension are among the most highly sophisticated aspects of Shidong Miao embroidery. The direction of the stitching is dictated by the details of each part of the imagery or pattern. Sometimes chain stitch is added to highlight outlines and create a strongly linear effect; nice even chain stitching is considered impressive. Shidong satin-stitch embroidery is also called poxianxiu, literally ‘broken thread embroidery’. Its secret is the use of a single strand of untwisted silk separated from two-ply silk thread, or ‘breaking’ (that is, splicing) a thread into six to eight very fine threads. Shidong Miao women use starch to stiffen and prevent the fine thread from fraying and to keep the silk bright and shiny (Tomoko Torimaru 2008: 54–55).15 Most Shidong women do satin stitch well; you must have something new or rare if you want to impress someone. The
Embroidery Speaks: What Does Miao Embroidery Tell Us?
75
stitching of hob mongl-style festival jackets is briefly discussed below, and then the imagery of a particular historical example is examined in some detail and socially contextualized.
A Historical Example Illustrating Cultural Boundary Writing I have long been exposed to Miao embroidery, having been a museum curator at the Chinese Textiles and Clothing Cultural Center (CTCCC) at Fu Jen Catholic University since 1990, and in 1999–2000 I edited a book, ‘Miaozu Fushi Tuzhi – Qiandongnan (Miao Costumes of Southeast Guizhou – Illustrated Research Directory)’, showing many examples of Shidong Miao embroideries impressive in their beauty and technique. The embroidery attracted me because of the ingenious techniques, and also because of some strange Han characters included in the composition, discussed in detail below. This phenomenon seems to be unique to Shidong; I have never observed or heard of this in other Miao embroidery, and this unusual feature led to my doing fieldwork there. I have seen more than a thousand pieces of Shidong Miao clothing and embroidered items, among which a particular example stands out for a number of reasons, not least because this was the most beautiful among many contenders: a pair of antique panels which originally would have decorated the sleeves of a woman’s festival jacket. They had become pieces in a private collection in Taiwan.16 To facilitate discussion below, the left sleeve design is called ButterflyMother, the right sleeve Fortune-and-Peace. The two panels are asymmetric in design: the Butterfly-Mother sleeve comprises two wide strips divided by a narrow strip, while the right sleeve, Fortune-and-Peace, is thematic. This leftright asymmetry is quite unusual. This intrigued me, so I took photographs of it and took questions with me on fieldwork trips to Shidong.
The Butterfly-Mother Sleeve The Butterfly-Mother embroidery, approximately 25 x 30 cm, was originally a sleeve decoration on a hob mongl zek jacket (see Figure 3.1). It depicts the Butterfly-Mother and Niul Jil Bongx’s creation story, a story familiar to all Shidong people. Put simply, when there were still no creatures in Heaven and Earth, a butterfly grew from the heart of a maple tree. Her name was MeiBang, and she grew to be beautiful. At that time, there were no people. She had a romance with a water bubble and laid twelve eggs. But although the butterfly could lay eggs, she could not incubate them. Birds living on the hillside, Niul Jil Bongx, came to help Butterfly-Mother incubate the twelve eggs. From that time, Heaven and Earth started to have all things, including humans. The sleeve embroidery elaborates this story through its structure and images. The structure of two wide strips with a narrow dividing strip divides
76
Ho Zhaohua
the space into three. The middle dividing strip is simply decorative. Five frontview Butterfly images are arranged vertically in the centre. Small figures appear immediately next to these, left and right. The imagery in each strip is built symmetrically. Both have frontal images of Butterfly-Mother in the centreline, that is ‘tummy-up’ (see A, D). The upper and lower strips each have two lateral views of Niul Jil Bongx (B, C, E, F), wild and beautiful birds, on the left and right. These also have highly elaborate symmetry. The centre ButterflyMother clearly divides the overall pattern, creating a mirror effect,17 with some asymmetric details filling sub-areas.
Figure 3.1. Left-sleeve embroidery: Butterfly-Mother (photo by Fang Siu Nang; permission by Fang Siu Nang) Butterfly-Mother has many highly sophisticated and unique decorative details, such as the flower-bud-shaped decoration along the upper and lower edges not commonly seen on Shidong clothing. The design is mainly accomplished using flat stitch (qizhen), a Han stitch, and the most common satin stitch in Shidong.18 Extensive ‘long-and-short’ satin stitch (chanzhen), a Han stitching effect, is used to decorate the birds’ necks and wings in the lower strip (E, F): stitches of two different lengths are staggered, thereby creating the bleeding colour effect. Couching stitch (zhazhen), a Han stitch, usually fixes stitches to form the pattern. Here, however, it is used expressively over satin stitch, a flat stitch, to create the decorative effect on the birds’ wings and tails (B, C, E, F). Two
Embroidery Speaks: What Does Miao Embroidery Tell Us?
77
contrasting colour threads are twisted together to produce a rich colour mixture and texture on the birds’ necks and legs (E, F) on the lower strip. The extensive variety of stitches and their highly sophisticated use give the Butterfly-Mother design a high quality.
Imagery and Stitches of the Fortune-and-Peace Sleeve The right sleeve panel, Fortune-and-Peace, appears to convey an exotic Han Chinese theme in its imagery; the stitches are highly sophisticated and rare in Shidong Miao embroidery. This sleeve is in the thematic style of a hob mongl zek (i.e. dark) Shidong Miao jacket, but the layout used to deploy the images is essentially the same as that of the left sleeve, namely two wide strips with a narrow dividing strip: the middle ‘dividing strip’, at K in Figure 3.2, is less immediately obvious. Very few jackets break with convention and have just one sleeve with a thematic design.
Figure 3.2. Right-sleeve embroidery: Fortune-and-Peace (photo by Fang Siu Nang; permission by Fang Siu Nang) The Fortune-and-Peace imagery has both Han Chinese and Shidong Miao elements, all intensely stylized. These are realized with a variety of stitches, listed in Table 3.5. The images A to E and G are in Han style; F and I to K are in Miao style.19 The Fortune-and-Peace embroidery clearly demonstrates that Shidong Miao have appropriated Han Chinese motifs into their own cultural artefacts.
78
Ho Zhaohua
Table 3.5 The stitches used in Fortune-and-Peace Area
Image depicted
Style Stitch
Feature
A
lotus (growing in blue and white container)
Han
weaving stitch – a Han stitch, nashaxiu: metallic warp strips are picked up with a threaded needle creating a woven effect.
Used to enrich the lotus leaf texture. Weaving stitch is common in Miao outer edge designs, but rare in sleeve embroidery.
blue and white container
Han
satin stitch for the body and couching stitch (zhachen) for decorative effect
Using negative space to reveal the cloud shape of the contour line along the ‘shoulder’ level of the basin is quite rare in Shidong. The bleeding effect ( yunran) in the blue and white is a technique found on Chinese imperial clothes and sumptuous costumes.
a. lion (playing Han with one coinlike ball) b. two coin-like balls; c. round-shaped figures above and below
a. satin b. chain stitch (suoxiu) c. ‘long-andshort stitch’ (chanzhen)
The same negative space technique used on the blue and white container is used on the lion’s body. The roundshaped figure above closely resembles the Chinese figure Liu Hai.20
C
a crane turning Han its head
‘long-andshort stitch’ (chanzhen)
Shidong Miao use birds for decoration a lot, but rarely include cranes.
D
downwardfacing peony
Han
‘long-andshort stitch’ (chanzhen)
This is rare in Shidong design.
B
Embroidery Speaks: What Does Miao Embroidery Tell Us?
79
Table 3.5 continued Area
Image depicted
Style Stitch
Feature
E
deer looking upwards
Han
This technique not only fixes the loose silk, but also intensifies the patterning on the deer, creating a rich texture.
F
flower/s and bird/s
Miao satin stitch
G
a growing lotus Han
H
coin with sash decoration
Han
I
butterfly – vertical ‘tummy-up’ view
Miao satin stitch The ‘tummy-up’ with decorative butterfly is the image couching stitch most typical of Shidong regional design.
J
frog – ‘tummy-up’ view
Miao ‘long-andshort stitch’ (chanzhen)
Frogs are common in Shidong Miao design but the decorative details on the body are not a Miao style.
K
a bird, a cat, and an animal with one head and two bodies
Miao satin stitch – with couching stitch on bird and cat.
The bird wing, the face of the cat, and the animal all use twisted colour threads for decorative effect
satin stitch – with couching stitch (zhazhen) added over it for decorative effect
satin stitch
The flower is the image most typical of Shidong regional design. The lotus is rare in Miao embroidery.
satin stitch This image is rare in with decorative Miao embroidery. couching stitch (zhazhen)
The Han-style images at A, C, E and G, especially the lotus designs, rarely appear on other examples of a Shidong Miao woman’s jacket from the same period in the collections at Fu Jen Catholic University Museum. A deer would be used on a girl’s jacket, but not appear on an adult woman’s jacket. The crane is another uncommon image. The form of the peony at D is similar to a pud bangx flower21 in full bloom; its stitching, however, makes it different. Shidong embroidery did use tiger and lion images on sleeves, being wild animals that
80
Ho Zhaohua
are considered to bring fortune. But the traditional design does not depict a ‘lion dancing’, wu shi, as at B. One informant told me that people in Shidong began having wu long wu shi, ‘dancing dragons and dancing lions’, to ‘make the Chinese New Year lively’ from the late Qing dynasty. Thus the lion image at B depicts a relatively new scene – an instance of shifting boundaries. Similarly, the large figure above the lion in the same area B resembles the ButterflyMother images at D, but also closely resembles the Chinese figure Liu Hai.22 It is common to see a cat and bird – as in the middle strip – and to see two animals combined to fit into one form, but the split representation of an animal, using one head to join two bodies, is rare in Shidong embroidery. The image most typical of Shidong regional design is the flower in F. Most images are shown in frontal view, some in lateral view. The peony faces downward, the frog upward, and the butterfly is shown vertically, in the ‘tummy-up’ front view. The Han Chinese motifs depicted in the Fortune-and-Peace panel have been de-contextualized and have quite different symbolic meanings and stories associated with them. The imagery is summarized in Table 3.6. Table 3.6 The symbolic meanings of the imagery in Fortune-and-Peace Area Image depicted
Style
Han meaning
A
lotus growing in blue and white container
Han
continuous greater emphasis on peace continuous growth
B
lion playing with a coin-like ball
Han
an official career
symbolizes Chinese New Year festival
fortune
Shidong Miao use copper sequins to symbolize money. Round-shaped figures usually symbolize Butterfly-Mother.
longevity
In Shidong the crane is thought similar in shape to the egret, which steals fish from farm pools, so the crane symbolizes a person having ability to find food.
round-shaped figures (above and below)
Han
C
crane
D
downward-facing Han peony
Miao meaning
associated Considered a ‘flower in with wealth full bloom’, meaning a and fortune mature female able to give birth
Embroidery Speaks: What Does Miao Embroidery Tell Us?
81
Table 3.6 continued Area Image depicted
Style
Han meaning
Miao meaning
E
deer looking upwards
Han
an official career
simply a beautiful animal for decoration: see footnote 23
F
flower/s and bird/s
Miao
decoration
fertility
G
a growing lotus
Han
continuity and purity
decoration
H
coin with sash decoration
Han
fortune
decoration
I
butterfly
Miao
fortune
fertility
J
frog
Miao
fortune
transformation and fertility
K
a bird, a cat, and a Miao split animal representation
decoration
decoration
The Fortune-and-Peace sleeve embroidery images are clearly not constrained by the Chinese readings of images that were originally Han Chinese. For example, Shidong Miao consider becoming an official a form of success, but they do not use images of the lion and deer to represent it – the deer is simply a beautiful image used decoratively.23 The peony is associated with wealth and fortune in Chinese culture; in Shidong, however, it is considered a ‘flower in full bloom’, meaning a mature female able to give birth. The way Han Chinese motifs have been incorporated into the Miao design structure – putting a highly elaborate asymmetric design into a symmetric arrangement – achieves something completely new for sleeve embroidery.
Han Chinese Characters on Shidong Miao Clothes Two outstanding features of ‘top of the range’ embroidery from Shidong in the early twentieth century are the stitches and the incorporation of Chinese characters into the visual display. Even now, older Shidong Miao women wear embroidered jackets decorated with a mixture of images and Chinese characters. Most of these women, aged over forty, had not received any formal schooling and are unable to read Chinese. Why had they embroidered these particular characters onto their traditional clothes? What meaning did they have for Shidong Miao people?
82
Ho Zhaohua
Shidong Miao women at that time accessed Chinese characters by ‘asking for words’ from the Taoist priest, teacher, father, brother/s, or other literate males.24 Appropriating words that belonged to an ‘elite’ was an act of ‘being literate’, at that time considered something that women in Shidong had no right to be. In Chinese script, Chinese characters represent morphemes, the minimal unit of form and meaning. Shidong Miao women used them in at least two distinct ways, both of which constitute a form of ‘literacy’. The first was to borrow particular characters to represent and convey the meanings of longer Han Chinese phrases with which local people were variously familiar. The second was to use similarly pronounced Han Chinese characters to ‘write’ the Miao dialect, which did not then have a written language. Moreover, embroidering characters was an expression of Miao agency. The Miao believed that the characters had intrinsic magical power: by embroidering particular Han ‘words’, they were incorporating them into the Miao religious system. They could thus have the Heavenly Thunder God fulfil people’s desires and achieve their cultural ideals.
Figure 3.3. Ob ait duf sleeve embroidery by Grandmother Niu, 1930s to 1950s (CTCCC 1956) (photo by CTCCC; permission by CTCCC)
Embroidery Speaks: What Does Miao Embroidery Tell Us?
83
Many examples of the morpheme 大, ‘large; great’ (da) appear in the blank areas, in the lower strip of the ob ait duf sleeve embroidery shown in Figure 3.3, made by Grandmother Niu (now over eighty years old) in the 1930s to 1950s. The characters are in proportion and follow the correct contour lines, and the stroke order is correctly suggested, but some of these characters are depicted as mirror images of the normal form. Was this a mistake, or was it done deliberately – and if so, why? Most female informants25 thought depicting characters was simply a popular fashion at that time and said this was a mistake made by women who had not learnt to write characters and who were unaware of the correct stroke order, or indeed of any need to depict this and other details. One male informant, Long, offered a different explanation, saying that the visual incorporation of characters was not a women’s decision but had originated with the ‘landowners’ (地主, dizhu). This elite group’s understanding was that blessings and merit all come from the supernatural. Depicting the mirror images of characters allowed the Thunder God to read them from his heavenly perspective. Long said such ‘text’ was not for decoration; it possessed the power to communicate with the Thunder God, a phenomenon and motivation which went far beyond ‘fashion’. Why did the Shidong Miao need a Han Chinese character text to communicate with their Thunder God? Long explained that the Thunder God had already given his allegiance to the Chinese emperor and had subsequently become Hongjiang County Grand Master (縣太爺). The Miao should make requests to him using Han Chinese characters, and the Thunder God would then present the written request to ask the Chinese emperor’s approval. A case in which it was difficult to make the distinction between words and phrases led to the identification of examples of the borrowing of particular characters to represent longer phrases and convey their meanings. The embroidery in Figure 3.4 shows a number of characters: 早, zao, early; 吉, ji, auspiciousness; 大, da, great, large; 換, huan, change; 合, he, ‘join/combine’; 春, chun, spring; 相, xiang, mutual/ity; and 成, cheng, accomplishment. At first glance, in various combinations the phrases seem unintelligible and meaningless. This puzzled me for a long time. San Meibao told me that when women saw new characters on another woman’s clothes design, they were all eager to have the ‘new words’ to do their own embroidery, otherwise they would consider themselves ‘out of fashion’. This may well have been the case, but it did not explain this particular selection of characters. I then showed the photographs to two former schoolteachers, people considered members of the Miao elite. The first, Wang, a retired junior-high-school teacher, pointed out that these characters are abbreviations of set Han Chinese phrases: 成語: 早, zao, early, is a reference to 早生貴子, zaosheng guizi, ‘to get pregnant as soon as possible’, said of a newly married woman. 吉, ji, ‘auspicious/ness’, is an abbreviation of the phrase, 大吉大利, literally ‘great auspiciousness great prospering’. 大, da, ‘great/large’, is an abbreviation of 大戶人家, dahu renjia, ‘a large [i.e. successful]
84
Ho Zhaohua
household of people living together’. 合, he, ‘join/combine’, indicates the phrase 陰陽相合, yin yang xianghe, Yin and Yang conjoined, usually referring to marriage. Also 春, chun, always ‘spring’, refers to the season with its associations of new life, fertility and growth; 相, xiang, ‘mutual/ity’, refers to mutual interactions expressing unity, and indicates marriage; and 成, cheng, ‘accomplishment’, says that all wishes are successfully manifested, done. He could not, however, say what 換, huan, ‘change’, meant or referred to.
Figure 3.4. Hob mongl sleeve embroidery, 1930s to 1950s (photo by Fang Siu Nang; permission by Fang Siu Nang) The second informant, Long, a retired elementary school principal, then told me: ‘Such mysteries are not hard to solve as long as one looks at them from the Miao point of view’, adding that ‘One must pay attention to precisely where the various characters are embroidered.’ There are eight large characters visible on the Hob mongl sleeve embroidery in Figure 3.4. Each character occurs on top of or next to the following: (1) the fire-dragon; (2) a bird; and (3) the woman’s head. There are four characters for the fire-dragon: 早, zao, ‘early’; 吉, ji, ‘auspiciousness’; 大, da, ‘great, large’; 換, huan, ‘change’. There are also three characters for the bird: 春, chun, ‘spring’; 相, xiang, ‘interaction’; and 成, cheng, ‘accomplishment’. Lastly, there is one character for the woman’s head: 合, he, ‘join/combine’. The Shidong Miao consider that the fire-dragon relates to the supernatural. A bird is considered male and as having agency in sex and reproduction. The female figure represents those who become pregnant: she is the addressee for the accomplishment of social reproduction, represented by the 成, cheng.
Embroidery Speaks: What Does Miao Embroidery Tell Us?
85
These characters on the fire-dragon and the associated morphemes are references to blessings from the heavenly Thunder God. The character 換, huan, ‘change’, also means ‘change one thing for another’ and can be understood here as referring to a Shidong belief that they can bargain with the heavenly Thunder God.26
Figure 3.5. Hob mongl sleeve embroidery, 1930s to 1950s (photo by CTCCC; permission by CTCCC) Shidong Miao people were and possibly still are able to interpret and invest these characters with significant meaning, without grammatical formative structures, such as phrases or sentences, to effect communication between Heaven and humans. This encompasses consideration of precisely where the relevant characters are embroidered. A quite different, clearly intentional appropriation of Chinese characters occurs to represent Miao phonemes, to express and thereby manifest Miao cultural ideals. An example is the character 疆, ‘boundary’, which appears on the dragon’s face in Figure 3.5. Shidong Miao use its Mandarin pronunciation, jiang, ‘to transcribe’ (i.e. to ‘write’), for the similarly pronounced Miao word jangx, ‘done’, ‘accomplished’, ‘success’. The situation this ‘accomplishment’ pertains to is indicated by the other characters nearby: a 合, he, ‘join/combine’, on the water-dragon’s leg, and a mirror image of the character 明, ming, ‘bright’, on the dragon’s jaw. One explanation for such mirror images has been discussed above. Long Lin offered an alternative, suggesting that this character should be separated into its two components: 日, ‘sun’, considered male, Yang; and 月, ‘moon’, considered female, Yin. The meaning conveyed by depicting these three characters together can then be understood as ‘the [requested] interaction of
86
Ho Zhaohua
yin and yang is done’. The 疆, jiang, character is actually incomplete – it is missing a stroke. Long Lin commented, ‘Our women don’t have culture; they can embroider well, but they cannot turn embroidery into writing.’ This echoes a remark often heard from the local women themselves: ‘Sorry, we don’t have culture.’27 Shidong Miao have obviously incorporated Han Chinese characters – aware of the associated morphemes – into the Miao cultural system. The morpheme is a unit that migrates easily between two languages when they come into contact. This was readily perceived as symbolic capital – capital which Shidong Miao landowners could make use of to consolidate their social status within Shidong Miao society.
Magic Embodied in Stitching and Morphemes Michèle Lamont and Virág Molnár distinguish symbolic and social boundaries in their article, ‘The Study of Boundaries in the Social Sciences’. They call boundaries ‘tools by which individuals and groups struggle over and come to agree upon definitions of reality’. They suggest that examining symbolic and social boundaries allows us to capture the dynamic dimension of social relations (2002: 168). It is significant that Shidong Miao landowners use symbolic capital from the ‘other’ culture to establish self-identity within the in-group, in fact within the very medium that publicly identifies Shidong Miao – their embroidered jackets. In the Shidong Miao case, stitches and morphemes became media with which to both cross and veil the cultural boundary between Miao and Han. As Epstein states, ‘symbolic boundaries also separate people into groups and generate feelings of similarity and group membership’ (1992: 232). In summary, ‘new’ stitching and the depiction of Chinese characters became symbols of elite status and modernity, imposing a symbolic boundary within Shidong Miao society and constituting valuable capital which landowners used to separate themselves from the lower classes.
Stitching Alterity Versus Self-Identity Grandmother Ying gave me a surprising answer when first shown photographs of Butterfly-Mother and Fortune-and-Peace, saying this was not a special case, and recalling it as a ‘fashion’ when she was a little girl (about sixty to seventy years ago) in her grandmother’s time (1910s to 1930s). But even so, at that time it was still relatively unusual; she had seen wealthy landowner’s daughters dress in jackets with asymmetric decorative designs, an instance of ‘high fashion’ in the local area. How did exotic stitches become valuable? Grandmother Ying pointed out that only a landowner’s daughter could have so much time to embroider such elaborate and beautiful stitching, adding that local people call embroidery like
Embroidery Speaks: What Does Miao Embroidery Tell Us?
87
Butterfly-Mother dit zhux huab, ‘landowners’ designs’. There was a gap between rich and poor households at the time it was made. It would not have been possible for a woman whose family was poor to achieve embroidery of such quality; it needed not only time but also money. This piece would have involved buying highly prized colourful silk threads. Similarly, San Meibao too recalled that, when her mother was young, her grandfather, a successful merchant during the same era, had hired skilful women to embroider new clothes for her mother. She said that this Butterfly-Mother depiction was a fashionable detail in her grandmother’s jackets, and in Shidong such ‘landowners’ designs’ used to be seen on high-quality jackets. As indicated above, it is difficult to distinguish precisely which particular embroidery stitches were originally Han Chinese: what is clearly revealed is the process of cultural interaction and the porous nature of some cultural boundaries. Some stitches are those most commonly used by Han people, and these only appeared in Miao embroidery relatively recently. Shidong women consider satin and chain stitches as Miao stitches because of how they have been modified and transformed, thereby creating effects slightly different from those in Han-style embroidery and imbuing them with their own distinctive style and spirit. Shidong people believe that, through needlework, a woman gives part of herself to the embroidery. Regarding when and how various aspects of Han Chinese style and stitching first arrived in Shidong, Grandmother Ying suggested that ‘They might have come with luxury trading items … such as curtains. Or more probably they were seen in costumes and stage decorations used in Xiang Opera Theatre. This was very popular in my grandmother’s time.’ Evidence supporting this idea is the bleeding effect (yunran) seen in the blue and white container in the Fortune-and-Peace example, which as already noted was a technique found on Chinese imperial clothes and sumptuous costumes. Grandmother Ying commented that Butterfly-Mother is a smart design: ‘It gives everyone a sense of freshness at first glance, and even those experts at embroidery would be very impressed’. Moreover, ‘The two sides of the design are in quite contrasting styles … one side uses Han details to elaborate a Miao design; the other is using Miao details to assimilate Han images into local design’. In addition, ‘She has been able to present two different techniques at the same time; only a very capable needle-worker dare to do so’. Her explanations removed my doubts, coming as I was with an outsider’s point of view. I asked some women whether a Shidong Miao woman would embroider a pair using different design approaches on each. Their answer was in the affirmative. This kind of embroidery would be done to show off her highly sophisticated techniques, and equally importantly, if wanting to impress everybody. Women used what they called ‘Han’ stitches and images to create something new, fashionable and unique to stand out from the work of others. The purpose was to consolidate social position and class. Grandmother Ying also remarked that when it comes to the logic and thinking behind this use of
88
Ho Zhaohua
clothes to impress people, there is little difference between those times and now.
Speaking Alterity Versus Shidong Miao Cultural Identity The use of the character image was a widespread phenomenon in Shidong embroidery from the late Qing dynasty to the 1980s. The selection and incorporation of Chinese characters into their own cultural matrix by Miao who originally lacked their own written language cannot be understood as simply an indication of sinification, or of subjection to the Chinese state. Elite women accessed a symbolic resource to signify their social status. The embroidery of the traditional jackets women still wear is an important, even critical means whereby Shidong Miao women access respect and status, self-affirmation and identity in their social context.
Conclusion Shidong Miao have obviously incorporated Han Chinese characters – aware of the associated morphemes – into the Miao cultural system. The morpheme is a unit that migrates easily between two languages when they come into contact. This was readily perceived as symbolic capital that Shidong Miao landowners could make use of to consolidate their social status within Shidong Miao society. Incorporating characters from another tradition onto clothing as decorative elements is nothing new. Shidong Miao, however, have incorporated Chinese characters into their embroidery in a quite particular way, reflective of their own strong culture. Based on my fieldwork, I understand this Shidong Miao cultural phenomenon of the incorporation of Han Chinese images and techniques as representing the symbolic capital of a local Miao landowning class, exploited in order to consolidate their position in Miao society. Nevertheless, Shidong Miao culture is neither a substance nor an exclusive resource for the dominant class; it is constituted through the more subtle exercise of symbolic power to create social order. The Shidong Miao jackets enable them to expand their symbolic power and cultural capital to transform social status. In this context a question that arises is whether the images on Shidong Miao embroidery represent an independent ‘literacy’? Using Chinese characters does not signify individual embroiderers’ literacy; rather, it becomes symbolic capital for distinguishing the social class boundary. Nevertheless these characters constitute an abbreviated written language. Shidong people use single characters to evoke a phrase or idiom, thereby conveying a meaning. They also use characters with suitable Han pronunciations to transliterate and thus ‘write’ phonetically similar Miao-dialect words and their meanings,
Embroidery Speaks: What Does Miao Embroidery Tell Us?
89
representing a significant blurring of the boundary between image and word. That the Miao of Shidong have lived under the regime of the Chinese Empire and nevertheless vigorously maintained their traditions is a testament to social continuity. They have used clothing to write their own culture and history. Ethnic minorities can have agency. Will these traditions continue to have meaning for Shidong Miao people in the face of ‘modernity’? This remains to be seen, but it would seem to stand a good chance of doing so.
Notes 1. From the Qiandongnanzhou zhengfu website (Qiandongnan Autonomous Prefecture Government) www.qdn.gov.cn (accessed 23 November 2010). This population figure is lower than that given in the 1991 Fourth National Population Survey, which records Taijiang County’s population as 141,621, of whom 13,982 were from Shidong. This may reflect migration to major cities. This phenomenon and the direct impact on traditional clothing-related practices are discussed in Ho 2011: 269. 2. The Jiangxi Merchant Guild Hall (江西會館) was established in Mahau Township (馬號 鄉) in the 1850s; the Hunan and Hubei Merchant Guild Hall (兩湖會館) was established in Shidong in 1877. 3. Miaozu Jianshi Bianxiezu 1985; Qiandongnan Miaozu Dongzu Zizhizhou Difangzhi Bianzuan Weiyuanhui 2000: 22–26. 4. Dail dial bul lil sux ait ghab lail; dail ad bul lil sux nongx nangs; ait hmub ait gheb gid jus dlas dlas xiangl xiangl. 男孩讀書學當官;女孩讀書像吃草(飢餓) ;繡花做家務才得富 享貴。 5. Shidong Miao cosmology also has a binary opposition between yeng and yangx. These two concepts are similar to the Chinese concepts of Yin and Yang, but are also distinct in many ways. See Ho 2011: Chapter 2. 6. Dail bad ait nongx, dail mais ait nangl, seix seix ait zangt. 男管吃、女管穿、一起找錢。 7. I met Zhenmei while doing fieldwork in Shidong in the summer of 2002. I later attended her wedding, one of eight weddings I attended during my fieldwork in Shidong in 2006– 2007. 8. These styles are: 1. hob mongl dlub 亮破線繡花衣; 2. hob mongl zek 暗破線繡花衣; 3. pul duf dlub 亮兩層破線繡花衣; 4. pul duf zek 暗兩層破線繡花衣; 5. ob ait duf dlub 亮兩層夾 破線繡與挑花直條花衣; 6. ob ait duf zek 暗兩層夾破線繡與挑花直條花衣; 7. ob ait pul dlub 亮兩層夾挑花直條花衣; 8. ob ait pul zek 暗兩層夾挑花直條花衣; 9. khait mongl dlub 亮紮花花衣; 10. khait mongl zek 暗紮花花衣; and 11. hob jub 不破線花衣. See Table 3.2 for details. The jackets also include silver decorations. Along with cloth and embroidery, silver is the other fundamental element in cultural valuing, marriage trajectories, a woman’s personhood etc. in Shidong Miao. See Ho 2011. 9. When I tried to learn some embroidery during my fieldwork, old women in the village always told me: ‘Working slowly and deliberately will accomplish beautiful work.’ The mayor of Shidong made a somewhat similar observation when discussing the commercial embroidery situation with Oakes (2003): ‘When they’re (women) making it for money they go too fast, rather than working from their hearts’, adding that this had resulted in poor quality work. 10. This method is called ‘square clothing system’ ( fang yi xitong, 方衣系統). 11. This has implications regarding what imagery can be worn close to the body, but space restrictions preclude further discussion here. See Ho 2011: Chapter 6. 12. Old, worn-out jackets, dyed blue, are worn at home. 13. Tim Oakes explores this period and more recent impacts of external government policy and economic forces such as the exoticization and commodification of ‘authentic
90
Ho Zhaohua
nationality culture’: ‘The idea of modernization and progress based on preserving nationality tradition, an idea now enthusiastically embraced by local leaders throughout Guizhou, may in fact be resulting in a new form of internal colonialism’ (1999: 53). I am looking at something quite different: the internal sense of identity of local Miao people and the highly articulated cultural matrix that lies behind and informs their motivation and choices. 14. A separate core design element is the expression of Shidong Miao religious references: cultural ideals pertaining to social reproduction are highly coded into all designs. These aspects relate, for example, to the sequencing of various images (such as outlined in Table 3.2) on different areas of jackets and associated restrictions. Space constraints preclude further discussion here. For details, see Ho 2009, 2011. The discussion below of the creation story depicted on the Butterfly-Mother sleeve touches on its religious content. 15. The hob mongl style involves using untwisted silk to do satin stitch over a paper pattern stencil pasted onto satin fabric ground cloth with adhesive glue. After any necessary ‘pruning’ of the pattern, the embroidery begins. This local use of poxianxiu is a feature that differentiates Miao satin stitching from its use in Han embroidery. This is discussed further below. 16. I tried to trace and find the woman who had originally done this embroidery, but after following many false trails was ultimately unsuccessful. 17. Shidong Miao women use paper pattern stencils to produce this symmetrical mirror effect. 18. Miao and Han women share many common embroidery techniques, and it might seem strange to call such a seemingly basic stitch as flat stitch a ‘Han’ stitch and call satin stitch a ‘Miao’ stitch, but there is a simple principle behind this. Shidong Miao women make this distinction: if the stitch was borrowed from Han embroidery historically sighted, and they have not made any subsequent changes or modification, then they consider that stitch a ‘Han’ stitch. If, however, they have made some modification to a stitch or technique, they consider those modified stitches have become theirs. This is why Shidong Miao consider both satin and chain stitch – stitches actually found in various forms around the world – as belonging to ‘Shidong’. I discovered while doing fieldwork that children’s headwear, and the border areas of such women’s accessories as aprons, mostly featured the fashionable Han stitches, but that these were much more rarely used on women’s jackets. It seems that sleeve designs were generally more conservative and serious. 19. Apart from the visual composition aspects and embodied meanings, Shidong people consider the paper stencils and household bend hmub collections as comprising their range of motifs: any motif from beyond the repertoire of these basic visual components will be considered a Han image. For details, see Ho 2011: 274–83. 20. Liu Hai is a highly popular Chinese folk figure, representing prosperity and wealth. 21. A pud bangx flower is a flower presented face-up with the pistil in the centre and petals spread around it. 22. See note 19. 23. This was commonly reported by my informants. Nevertheless this detail deserves further investigation. 24. I was told this by many informants while doing fieldwork in 2006–2007. 25. Among them were San Meibao, the lady whose house I stayed in while doing my fieldwork, and Grandmother Ying, a lady in her late sixties who I first met in 2002. She was also one of Tim Oakes’s informants. 26. Shidong Miao have an unusual response if a request or desire is unfulfilled: they can ‘steal blessings’ from the heavenly Thunder God. 27. These statements are very much at odds with what I heard in their songs, saw in their delicate craft and read in the history of this area. Perhaps they are being somewhat disingenuous: a case of ‘not having culture we tell outsiders about’ or ‘not having a
Embroidery Speaks: What Does Miao Embroidery Tell Us?
91
culture we think outsiders – especially present-day “hyper-charged” electronic outsiders – would consider culture or of value’.
References Buck, John L. 1937. Land Utilization in China. Nanjing: University of Nanjing Press. Chakrabarti, Sreemati. 1998. ‘Women and Adult Literacy in China’. Electronic document, in Across the Himalayan Gap. http://ignca.nic.in/ks_41036.htm, accessed 5 July 2006. Cultural Palace of Nationalities 民族文化宮. 1985. Zhongguo Miaozu fushi 中國苗族 服飾 [Clothing and Ornaments of China’s Miao People]. Beijing: The Nationality Press. Epstein, C. 1992. ‘Tinker-bells and Pinups: The Construction and Reconstruction of Gender Boundaries at Work’. In M. Lamont and M. Fournier (eds), Cultivating Differences: Symbolic Boundaries and the Making of Inequality. Chicago: The University of Chicago Press, pp. 232–56. Guizhousheng Taijiang Xianzhi Bianzuan Weiyuanhui 貴州省台江縣志編撰委員 會. 1994. Taijiang xianzhi 台江縣志 [Historical Documents of Taijiang County]. Guiyang: Guizhou Renmin Chuban She. Ho Zhaohua 何兆華. 2009. Zhaochi yu zhaochuan: Guizhou Shidong Miaoren de Shijian yu Shehui zaishengchan 找吃與找穿:貴州施洞苗人的時間與社會再 生產 [The Social Construction of Time: Shidong Miao Calendar and Cultural Construction], in Minsu Quyi, Journal of Chinese Ritual, Theatre and Folklore. Vol. 166 (2009/12): 7–59. Ho Zhaohua 何兆華. 2011. ‘Gifts to Dye For: Cloth and Person among Shidong Miao in Guizhou Province’, Ph.D. dissertation. Xinchu: National Tsing Hua University. Lamont, Michèle and Virág Molnár. 2002. ‘The Study of Boundaries in the Social Sciences’. In Annual Review of Sociology 28: 167–95. McCracken, Grant. 1990. ‘Clothing as Language: An Object Lesson in the Study of the Expressive Properties of Material Culture’. In Culture and Consumption. Bloomington and Indianapolis: Indiana University Press, pp. 57–70. Miaozu Jianshi Bianxiezu 苗族簡史編寫組. 1985. Miaozu Jianshi 苗族簡史 [A Short History of the Miao]. Guiyang: Guizhou Minzu Chubanshe. Oakes, Tim. 1999. ‘Selling Guizhou: Cultural Development in an Era of Marketisation’. In Hans Hendrischke and Feng Chongyi, eds, The Political Economy of China’s Provinces: Comparative and Competitive Advantage. New York: Routledge, pp. 31–72. ———. 2003. Dragonheads and Needlework: Textile Work and Cultural Heritage in a Guizhou County. Provincial China 7(2) (2003): 151–77. Qiandongnan Miaozu Dongzu Zizhizhou Difangzhi Bianzuan Weiyuanhui 黔東南 苗族侗族自治州地方志編纂委員會 (QMDZDBW). 2000. Qiandongnan Miaozu dongzu zizhizhou zhi—minzhuzhi 黔東南苗族侗族自治州志—民族志 [Historical Documents of Qiandongnan Miao and Dong Autonomous Prefecture]. Guiyang: Guizhou Renmin Chubanshe. Torimaur, Tomoko. 2008. One Needle, One Thread: Miao (Hmong) Embroidery and Fabric Piecework from Guizhou China, University of Hawaii Art Gallery, Department of Art and Art History. Zhongguo Kexueyuan Yanjiu Suo Guizhou Shaoshu Minzu Shehui Lishi Diaochazu 中國科學院民族研究所貴州少數民族社會歷史調查. 1964. Guizhousheng
92
Ho Zhaohua
Taijiangxian Miaozu de Fushi 貴州省台江縣苗族的服飾 [Miao Clothing in Taijiang County Guizhou Province]. Zhongguo Kexueyuan Minzu Yanjiusuo Guizhou Shaoshu Minzu Shehui Lishi Diaochazu & Zhongguo Kexueyuan Guizhou Fen Yuan Minzu Yanjiusuo 中國科學院民族研究所貴州少數民族社會歷史調查組 與中國科學院貴州分院民族研究所出版 (Published by the Minority Social and Historical Investigation Group in Guizhou Province, Institute of Ethnology of Academy of Social Science and Guizhou Division of Institute of Ethnology of Academy of Social Science).
Chapter 4 Tensions between Romantic Love and Marriage: Performing ‘Miao Cultural Individuality’ in an Upland Miao Love-Song Chien Mei-ling
Introduction Pierre Clastres notes that primitive societies are not ‘incomplete’, even without a state, literacy or written historical records (Clastres 1987). In line with this way of viewing and understanding non-modern or non-Western society, this chapter focuses on a peripheral society that has retained its autonomy by performing individuality within its own cultural contexts. This, however, does not imply an approach towards the morally autonomous individualism found in modern Euro-American societies. This chapter explores the local expression of individuality in the love-song performance of an upland Miao society in eastern Guizhou in south-west China, illustrated using extracts from a song of over four hundred lines performed as a duet between two female and two male singers.1 Sentiment has long been a topic of both attention and contention in the anthropology of kinship, but it is often treated as following from kinship relations (Radcliffe-Brown 1924; Homans and Schneider 1955). This chapter adopts an alternative performative perspective, exploring the sentiments expressed in the love-song performances of upland Miao people. Miao lovesongs place thematic emphasis upon humour, sentiment, and social and kin relationships by employing the rhetorical force of various poetic devices. I give special attention to a thematic analysis of the tensions between ‘romantic love’ (personal sentiments) and ‘marriage’ (affinal alliances), and outline how this Miao love-song poetically reveals a very common theme: the encounter between individual identity and social ideals. The dialogic relationship between the individual and society has long been discussed by Durkheim, Mauss, Dumont and others. Examining further the claim that ‘individualism pertains to a particular historic-cultural
94
Chien Mei-ling
conceptualization of the person or self’ (Rapport and Overing 2000), this chapter concludes that Miao individuality is expressed within performances of Miao love-songs and the social dynamics they embody.
The Individual and Society in Anthropology There is a long history of debate on the topic of the ‘individual’ versus ‘society’ in the anthropological literature. I will not review the entire debate here but concentrate on outlining the general approach of French anthropology, in particular Durkheim, Dumont and Mauss, as well as the work of the British anthropologist Marilyn Strathern. Much of this work gives a significant role to the individual, and to the encounter between the individual and society. From their work, one can see how the concepts of ‘individual’, ‘individuality’ and ‘individualism’ developed by considering them within their specific ethnographic, cultural and historical contexts. An article by Rapport and Overing in their Social and Cultural Anthropology: The Key Concepts (2000) reviewing Durkheim’s notion of the individual is my starting point. Émile Durkheim was the ‘key exponent of collectivist narratives which subsumed the individual actor within grand-societal workings’ (Rapport and Overing 2000: 179). He saw human beings in terms of the concept of homo duplex. As Rapport and Overing note, for Durkheim the individual ‘led a double existence; one rooted in the physical organism and one … in a social organism. … Between the two there was constant antagonism and tension, but through inculcation into a public language and culture humankind was capable of rising above mean (animal) individuality and becoming part of a collective conscience in which the (sacred) traditions of a society were enshrined’ (ibid.: 180). The narratives of Durkheim, Marcel Mauss and Louis Dumont all combine individualism and individuality. Again citing Rapport and Overing (2000: 180), in ‘A Category of the Human Mind’ (1979), Mauss shows ‘how society exerted its force on the physiological individual: how, through collective representations and collectively determined habitual behaviours, it submerged the individual within “a collective rhythm” ’. Combining universal individuality with cultural individualism, Mauss proceeds to outline an evolutionary account of how people in different ages and societies have been aware of themselves differently as individual beings, and how these differences can be traced back to different forms of social structuration. Mauss dismisses self-awareness early on in his paper as of no interest to him, considering it to be basically psychological and a human universal: what he is interested in is social constructions of the person. What constitutes ‘an individual’ is not a universal – it has its own expression in different societies. In his Essays on Individualism (1986), Dumont agrees with Mauss that ‘the Western notion of the individual—an autonomous actor, bearing supreme moral value—is an exceptional stage in the evolution of civilisations’ (Rapport and Overing 2000: 181). For Dumont, Hinduism offers a social insight: ‘the
Tensions between Romantic Love and Marriage
95
crucial difference between the Hindu world renouncer and the modern individualist is that the former can [ideologically] continue to exist only outside the everyday social world’, while the notion of the modern individual is central to Western society (ibid.). On the basis of Melanesian and English models, Marilyn Strathern discusses how indigenous conceptualizations of the individual and society differ in different cultures and in different historical periods. Her theoretical position regarding the individual and society is clearly articulated in After Nature, a work on modern English kinship: [The English model] depicted mid-twentieth century English kinship as a model for the reproduction of individuals and suggested a contrast with the Melanesian interest in the reproduction of relations. As individuals, persons in the English model do not symbolize whole social entities and cannot be isomorphic with a collectivity or a span of relationships. Rather, individuals are held to exist as parts of numerous different systems—a part of the kinship system, part of a naming system, part of society—and do not replicate in total any one systemic configuration. I referred to the conceptualization as merographic. … We are now in a position, I think, to give this merographically conceptualized English person its aesthetic or iconic dimension. … The individual person who is the microcosm of convention becomes elided with the individual person who makes his or her own choice. In the process, this figure will present a different kind of image, a composite, a montage, of itself. (Strathern 1992: 125–27, my emphasis)
Four points follow from the above quote that deserve to be highlighted. The first is that there is a tension, or antithesis, between the individual and society. The second is the valuing of the individual. Strathern found that different cultures understand what an individual is quite differently. It seems that Durkheim saw individualism ‘as a value’, although not explicitly stating this (Lukes 1973: 338ff.; Dumont 1986: 16). Moreover, Mauss and Dumont address autonomy and morality as the defining values of the Western individual. Comparing the Melanesian model with the English one, Strathern found that in Melanesia individuals are seen in terms of the relationships between them, that is, as producing the relationships which constitute their collectivity, their sociality; in mid-twentieth-century English society individuals were simply understood as being various discrete parts of the whole system. Given that her English subjects were a ‘mid-twentieth-century’ group, a third point Strathern’s quote raises is the question of modernity. Individualism has to different degrees underlain anthropological thought since Durkheim, being explicit or implicit in revolutionary or progressive ideas. It has long generally been considered a feature of modern Western society. Subsequent debate on the nature of the individual brought about a consideration of modernity. As Dumont notes, ‘for Mauss it is sometimes as though everything else were leading up to modernity’ (1986: 4). In fact, Durkheim and Dumont himself both have an evolutionist basis to their ideas. Most subsequent discussion contrasts ‘traditional’ forms with modern ones, or is evolutionist in structure.2
96
Chien Mei-ling
The fourth point of interest in Strathern’s study of contemporary English kinship is the shift towards the aesthetic and iconic dimension. Her considering the relations between individuality and sociality from such a perspective might also be traced back to her earlier study of Melanesian societies: Strathern’s notion of the ‘dividual’ being was drawn originally from McKim Marriott’s work on India (1976), and she applied this notion to Melanesia (Strathern 1988: 348–49, f.7).3 Far from being regarded as unique entities, Melanesian persons are as dividually as they are individually conceived. They contain a generalized sociality within. Indeed, persons are frequently constructed as the plural and composite site of the relationships that produced them. The singular person can be imagined as a social microcosm. (Strathern 1988: 13, my emphasis)
This novel perspective on the individual addressed by Strathern, aesthetic or iconic, offers a new method for uncovering the characteristics and value of the individual. In fact, this perspective is closely related to what I am doing in this chapter, namely examining language in use and the perfomative dimension of social life.
The Turn to ‘Language in Use’ The turn towards ‘language in use’ has its roots in the fields of philosophy and linguistics. Wittgenstein examined ordinary language usage: he ‘turned away from meanings and toward speaking habits’ (Urban 1996: xii). Picking up this line of thinking, John Austin observed that ‘people do much more with words (“performativity”) than merely talk about the world (“constativity”)’ (ibid.). The Russian linguist Roman Jakobson’s teleological view of language also contributed to this trend. For him, change in sound ‘therefore, is teleological. It must be measured … against the referential goal of communication, or if the example is from verse [sic], against the poetic function.’ Such a multifunctional, polysystemic view of language is a radical departure from Saussure, for whom langue was homogeneous in both function and system (Caton 1987: 231). Moreover, M.M. Bakhtin’s ‘homogeneous versus heteroglossic poles of language’ could also be built into the Jakobsonian model (ibid.). Bakhtin’s emphasis on the system of aesthetics in speech genres deserves more attention. Michael Holquist introduces Bakhtin as follows: Since the time of Kant, we have with ever increasing insistence perceived system as a closed order rather than as an open-ended series of connections. System for Kant meant not only the rigorous application of a fully worked out and absolutely coherent set of categories. System also implied that no major question should be treated in isolation: thus, any consideration of reason had to answer demands not only of logic or epistemology, but of ethics and aesthetics as well. It is in this latter sense only that Bakhtin’s thought might be labeled systematic: the sense he seeks to invoke when he calls … for an ‘open unity’. (Holquist 1986: x, my emphasis)
Tensions between Romantic Love and Marriage
97
Following the development of this pragmatic view, a critical perspective has emerged which explores the poetics and performance of language and social life, the aesthetic aspect. Bauman and Briggs (1990: 79) comment on these new perspectives as follows: ‘the turn to performance marked an effort to establish a broader space within linguistics and anthropology for poetics—verbal artistry… . A focus on the artistic use of language in the conduct of social life— in kinship, politics, economics, religion—opens the way to an understanding of performance as socially constitutive and efficacious, not secondary and derivative.’
Background There are roughly nine million Miao currently spread across a large part of the massif that covers the Yun-Gui Plateau in south-west China and surrounding uplands in both south-west China and northern parts of South-East Asia (in Vietnam, Laos and Thailand). The Miao within China are usually divided into five groups on the basis of their language and geographical distribution.4 The earliest known relations of the central Miao group with the Han Chinese occurred during the Song dynasty (Yang 1998). From the Yuan dynasty onwards, Chinese written records refer to Miao in this area as ‘Black Miao’. The region where the central Miao are located, ‘an unbroken strip of Black Miao’, did not become widely known until the Chinese forced their way into Miao territory in the Qing dynasty period of the Yongzheng (雍正) reign (1724–1736). Fangf Bil is a Miao village perched high on a hillside in the upper reaches of the Qingshuijiang River.5 It forms part of the northern subgroup of the central Miao (Yang 1998: 99), and administratively is part of Fanzao township, Taijiang County, in South-Eastern Guizhou Self-Governing Autonomous District in Guizhou Province.6 Local people speak an eastern Miao dialect of the MiaoYao subfamily of Sino-Tibetan. The people of Fangf Bil call themselves Hmub (cognate with Hmong, the usual self-designation). The village is composed of over 330 households with a population of almost 1,500 (1998–2000). It is divided into eleven hamlets (vangf ), whose names refer to various nearby geographical features. The residents of any single hamlet are generally the agnatic descendants of a lineage subsegment, and share a common Han Chinese surname. The naming system is patronymic. For both males and females, their name is composed of their individual name preceded by their father’s name, and their father’s name preceded by his father’s name. Han Chinese surnames came into use only in the eighteenth century and are seldom heard in everyday Miao discourse. They are nevertheless in keeping with the patrilineal spirit of Fangf Bil naming practices. The eleven hamlets are organized into five patrilineal marriage groups within which marriage is forbidden. The five marriage groups are designated by the five Han Chinese surnames of Zhang (張, Zix Zangb, literally ‘Family Zhang’), Tang (唐, Zix Tangf ), Wang (王, Zix Uangd), Yang (楊, Zix Iangf ) and
98
Chien Mei-ling
Tai (邰, Zix Taif ). There is no exclusive correspondence between the name of a marriage group and the surnames of the households and people belonging to it. For example, not everyone in the Zhang marriage group has the surname ‘Zhang’, and there may be people in other marriage groups with this surname. Hamlets, surnames and patrilineal marriage groups are all, in one way or another, organized around ancestral groups. Those in a marriage group generally live in close proximity, share a common male ancestor, and act as patrilineal descent groups in relation to ancestral rites and the possession of agricultural land. The surnames and hamlets coincide roughly with the marriage groups, but only the marriage groups correspond directly with the ancestral groups. The combined population of marriage groups 2 and 4 composes over 90 per cent of Fangf Bil’s total population. The proportion of intermarriages between them therefore far exceeds the proportion of intermarriages between any other marriage groups and marriages outside the village. The statistics in Table 4.1 are based on the family genealogies of the Vangf Dof and Ghad Dlongb hamlets. It is clear that women marry in both directions. The affinal classifications of this village seem to constitute something approaching a binary structure. Table 4.1 Marriages into and out of the Vangf Dof and Ghad Dlongb hamlets of marriage group 4 Other marriage group
Marriages into Vangf Dof and Ghad Dlongb
Marriages out of Vangf Dof and Ghad Dlongb
1
4 (4.2%)
2 (2.2%)
2
84 (89.3%)
71 (80.06%)
3
0 (0.0%)
0 (0.0%)
Zix Hot hamlet of marriage group 4
0 (0.0%)
1 (1.1%)7
5
2 (2.1%)
3 (3.3%)
Marriages out of Fangf Bil village
4 (4.2%)
11 (12.4%)
Totals
94 (100%)
88 (100%)
Most marriages take place within the village, and the society is organized around classificatory bilateral cross-cousin marriage. The kinship terminology of this village has a similar structure to the Dravidian-type kinship terminology in its ideal and practice of prescriptive cross-cousin marriage. Most women in Fangf Bil still marry either their classificatory matrilateral or patrilateral crosscousins: that is, both FZD/MBS or MBD/FZS marriages occur.8
Tensions between Romantic Love and Marriage
99
Nevertheless, the local practice of cross-cousin marriage is still related to how kin are classified in a community which clearly distinguishes between near and distant kin. Thus for female ego, bilateral cross-cousin marriage is not with actual FZS or MBS but occurs between members of patrilineal descent groups: • w ho are related to one another as classificatory patrilateral cross-cousins (FZHBS, FZHFBSS, or FZHFFBSSS); or • who are equated terminologically with the FZS; or • who are related to one another as classificatory matrilateral cross-cousins (MFBSS, or MFFBSSS); or • who are equated terminologically with the MBS. That is, there is again both classificatory FZD/MBS and classificatory MBD/ FZS marriage. The binary organization of Fangf Bil is given classificatory reality in the distinction between gad ghat (agnates) and khait (affines).9 These are relative categories of the sort found in virtually all societies, rather than sociocentric groups; there is no true dual organization here, despite the quasi‘binary organization’ in this village. Marriage is prohibited between gad ghat (or simply ghat) but permitted with khait. The centrality of the relationships between ghat and khait in Fangf Bil is indicative of the importance of kinship in village social processes. Traditional formalized flirting and what I call ‘delayed transfer marriage’, that is, duolocal post-marital residence, are the last specific Miao cultural phenomena I introduce here, though they are far from being the least important. They significantly demonstrate some fluid aspects of Miao marriage which exist simultaneously with the system outlined above, and seem to have done so for a long time. Courting is commonly understood in the Western tradition as an activity performed with marriage as its goal. Miao flirting is a far more complex phenomenon and is certainly not solely linked to the goal of marriage. Flirting in a Miao village is an institutionalized cultural practice, an expression of local social and cultural arrangements of place, time and person. It partly fits in with the stipulations of the affinal alliance pattern and partly gives expression to the fluid nature of the interplay between personal sentiments, as well as the relatively rigid cross-cousin marriage which is the social ideal. In contrast with some other Miao villages in eastern Guizhou that permit daytime flirting, in Fangf Bil this traditional activity is only permitted in the evening, but it is also allowed during the daytime following certain ritual activities such as New Year or harvest festivals. Evening flirting takes place in the living room of the house of the young woman being flirted with, or outside but close to the house. Daytime flirting activities take place at different ‘flirting places’ belonging to specific marriage groups, where opposite-sex affines flirt with each other in conjuction with affinal alliances. Formal flirting is the most important social activity in Fangf Bil village and is practised exclusively by ‘the young’. The local indigenous conceptualization of ‘the young’ who may and do participate encompasses those from as young as
100 Chien Mei-ling
ten years old, as well as adult men and women who have married within the previous few years. The latter do not usually flirt with their spouses, though this does occasionally occur. This inclusion of recently married people is a critical difference from the Western idea of courting, or courtship, and will be discussed further below. Flirting involves young men and women spending a great deal of time together, sometimes in groups, sometimes in single pairs, talking, singing and exchanging gifts.10 Sex is expressly prohibited. Miao flirting is clearly not simply a prelude and pathway to marriage: married people also participate in this quite distinct cultural activity (Chien 2009a). There are, however, specific rules for married men and married women. A married woman can flirt with her opposite-sex affines until she has delivered her first child. In contrast, a married man can participate much longer than most women: he is not prohibited from flirting with his opposite-sex affines until his children reach their teens. Unmarried men or women (single or divorced) are considered to be ‘young persons’ who can attend flirting activities freely, as long as they flirt with someone appropriate, that is, in conformity with prescriptions of affinal alliance. There is some fluidity in the kinship terminology – for example, sometimes a particular person can be described by two different kinship terms, each having different practical implications. Like ‘flirting’, duolocal residence also demonstrates the complex nature of local marriage. The bride does not usually go to live with the groom immediately after their wedding but returns to stay with her natal kin soon after it instead. She only visits the groom’s house on festival days or to assist her husband’s family do farm work. She does not live with her husband in his house until she has her first child. During the duolocal period, the bride and the groom can both freely attend flirting activities – interacting with other opposite-sex cross-cousins – on their own as they did before they married. The solid, clear binary structures of cross-cousin marriage are juxtaposed with the fluid aspects of marriage, flirting and residence that are clearly seen in the customs of traditional formal ‘flirting’ activities and duolocal residence. This juxtaposition and the complexity of the social domains associated with flirting and marriage find clear expression in Miao love-songs.
Local Love-Songs Many Miao songs were published during the 1950s in the series Minjian wenxue ziliao (民間文學資料, ‘Miao folklore data’), reprinted in the 1980s. The lyrics of these and the Miao songs published in Miaozu hunyin ge (苗族婚姻歌, ‘Marriage Songs of the Miao’, Tang [1959] 1986) are public records expressing the Miao collective experience of marriage, including the personal ‘individual’ experiences of bride and groom before and after marriage. Songs in Fangf Bil fall into two categories: ‘old’ songs, hxad lok, and ‘young’ songs, hxad vangx. Most ‘young’ songs present similar themes; the villagers further classify them into four subcategories based on performance details.11
Tensions between Romantic Love and Marriage 101
The ‘old’ songs are generally about Miao mythology, the origins of villages, lineages, clans and ancestors. The song discussed below is in the ‘young’ song category. These generally have a joyful, joking tone, but at times the tone can also be sentimental, lonely or forlorn. Among other things, they present the inherent tension between traditional marriage and individual personal sentiments.
Methodology Most of the Miao songs I collected in Fangf Bil village came from natural settings in the sense that almost all were performed as part of the ‘flirting’ activities in conjunction with the celebration of certain festivals or ritual activities. I recorded such songs on many occasions in both ritual and nonritual settings. This love-song performance was recorded with a tape recorder, and then transcribed using the Miao phonetic system.12 I chose to rely on a small number of local villagers as key cultural consultants to assist me with this – a twenty-year-old girl, her parents and some other relatives – because transcription is extremely tedious, time-consuming work, and the interpretation, translation and analysis of the more than four hundred lines required mutual understanding, trust, familiarity and patience on behalf of both the cultural consultants and myself. We worked together for several days and nights replaying the tapes many times, and transcribed the entire performance.
Love-Song Performance The extracts cited below come from a love-song performance that took place in the early spring of 1999 in connection with the seasonal festival celebrating the planting of rice. The song, of more than four hundred lines, was performed as a duet between two female singers and two male singers. The social context was that many people, male and female, young and old, from Fangf Bil and four neighbouring villages gathered on the hillside fields to watch bullfights. The event was hosted by Fangf Bil. After the bullfights most of the older men and women and the children left, but the young men and women stayed for the traditional ‘flirting’. Men and women from different villages sat or stood close together, some in groups, most in single couples, happily talking, and some, though not all, singing to each other. Love-song performances by some of those who were skilled singers attracted a large audience. More than one of these performances could be happening at any one time. The two female singers were from Fangf Bil, the two male singers from another village, not very far from Fangf Bil – about 40 or 50 li. Similarly, apart from the singers, there were in effect two audiences: one of women from Fangf Bil, the other of men from the other village. The people of both villages speak
102 Chien Mei-ling
the same Miao dialect and wear the same style of Miao clothes.13 Like the male and female singers, the men and women in the audience were also classificatory affines to each other. The two male singers (both forty years old) were already married and had become fathers. One female singer (twenty years old) was unmarried; the other (nearly twenty-five years old) had married but not yet become a mother and was still living with her natal kin. We – myself and some young ladies, Fangf Bil villagers, who were accompanying and assisting me – knew that the female singers were descendants of the same patrilineal group, but did not know their actual genealogical relationships to the male singers. During the break, however, and at the end of the song performance, we heard the singers address each other using prescriptive cross-kin terms. This particular love-song was sung entirely by men and women as a duet.14 In total, 450 lines were sung and 28 lines were spoken. They were grouped into verses of various lengths from 4 to 8 lines, mostly of 6 or 7 lines. The sequence of these lines was not prescriptive. This love-song performance took the form of a spontaneous competition among the singers. They had a ‘data bank’, a store of lyrics and verses, but they did not simply memorize and reproduce a long set sequence. They had to listen carefully to what their competitors, the other singers in the duet, were singing. If they did not do so, or were careless, they could get lost among the possible responses and so fail to select an appropriate response to what had just been sung. They could repeat some lines if appropriate. Some lines or verses were gender-specific, some were not. Nouns or pronouns could be changed to express gender-specific contents (Chien 2007). The duet performance lasted for more than an hour. It was not at all serious or formal. The singers talked to each other, or joked with the audience when they had finished their own lines, and the audiences talked to each other while enjoying the show. They mostly liked commenting on the performance, comparing the skills of the singers, and discussing the contents of the lines with the other members of the audience. The content was much more important than the quality of the singing delivery. But any singer who did not pay attention to when it was their turn to sing, or did not come up with a good line with which to respond, left themselves open to being criticized and commented on negatively by the audience, something not good for their reputation.15
Thematic Foci Miao love-songs place thematic emphasis upon humour, sentiment, and social and kin relationships. A particular feature is their performative expression of the local culturally determined tension between the individual experience of ‘love’ – equated with the flirting setting – and local cross-cousin marriage, which is the ideal. To summarize briefly, this love-song describes an encounter between two women and two men meeting each other at just such a traditional flirting event.
Tensions between Romantic Love and Marriage 103
On the one hand, the men and the women both express the same shy but joyful emotions they experience when attending such occasions. On the other hand, they tease themselves and their partners about the dialogical relationship between marriage and flirting. They finish the performance by expressing their feelings: forlornness about parting and being alone. Interwoven with the ‘story line’ outlined above are four recurring themes: (a) marriage; (b) flirting; (c) the two different local relationships between marriage and flirting; and (d) individuals’ experience, through which the aesthetic dimensions of marriage and flirting for the Miao are brought to the fore. (a) Marriage is an obvious thematic thread permeating the love-song. Two particular aspects of local marriage are directly referred to: the individuals’ desire for marriage, and the collective ideal of cross-cousin marriage. The following lines are an example of the song and singers expressing individuals’ longing for marriage. Viewed somewhat paradoxically from the Western tradition, this longing occurs in the context of assured marriage in keeping with tradition, as referred to in line 349, and line 157 further below: (male singers) 347 Xud dongf mongf dot jut
(male singers) Do not say that you will marry others.
348 Ninx dongf lel bib mek
Just say you will be with us forever.
349 Ghet benf liangf hvib naik
We have to become husbands and wives, anyway.
(female singers) 352 Ceit daik las baf pent, muldiangf.
(female singers) We come here intentionally to sing with you, cross-cousins.
353 Yel bib mongf lab diuf mek zix
Bring us to your homes. They (your parents) do not like us, but we still want to marry you.
354 Neff hfif sax vob diangt
The flown bird will return to the cage.
Some lines more explicitly refer to or articulate details of cross-cousin marriage, for example: (female singers) 154 Bib mul-diangf naik jut
(female singers) We are cross-cousins.
155 Bib def diangf dak niangt
We separate temporarily then return to sit together again and talk a while.
156 Bib nal diangf dak ghent
Our mothers come to say something to the boys.
104 Chien Mei-ling
157 Bib naf neif jek dint
We will become husbands and wives anyway.
158 Bib del diangf liak ciet
Do not mind if we have to part right now.
There are implicit allusions to the prescriptive alliance involving village endogamy: (male singers) 359 Hvid cet liuk lel lait
(male singers) We come here just now as done for thousands of years.
360 Liuk gheb mangk lel ghet
We still cannot take you as our wives even in our next lives.
366 Ghat but vangf hveb bif
You girls will marry someone else within your own village.
367 Vob diut dak jut liuk
Become other men’s wives.
The young men cannot marry the young women they are singing to; village endogamy requires that both parties marry within their own village. The emphasis on the relatively prescriptive cross-cousin alliance aspect of marriage is realized both by the lyrics and also by the repetitive use of the prescriptive affinal kin terms throughout the entire song, indicated in bold in the extracts. (b) Flirting. There are various illustrations of flirting and institutionalized formal flirting occasions, this being the second thematic thread. Some lines vividly describe the flirting scene. Descriptions of the singing and talking mirror what occurs in real life. Line 385 can be understood in two ways: as a reference to how flirting also occurs in the living room of the flirted girls, or to talking more ‘after we marry’: (female singers) 343 Dak lait iauf dak ceit, muldiangf
(female singers) Once you have come here, come and sing together, cross-cousins.
344 Ceit daik las baf pent
We come here on purpose to sing with you.
(male singers) 383 Ghaib qub mangk juk ghof, mul-diangf
(male singers) Like chickens, we cannot stop pecking the rice, cross-cousins.
384 Bib pot mangk juk jib
We cannot stop talking to each other.
385 Dleit diut ghab diuk nail
Leave something to be said inside the house.
Tensions between Romantic Love and Marriage 105
Some lines describe a scene of everyone gathered for an event such as the one where this performance was underway, and the playful atmosphere that is characteristic of such flirting events: (male singers) 428 Bib sint sint ghenf jek wek
(male singers) We just sit together in groups.
429 Sint sint ninf jek cangk
Just sit together creating this spectacular scene.
430 Dit lit deif liuk deik
The girls do not want the boys immediately.
431 Deif liud daif rangk dok
They throw the boys away like straw or firewood.
432 Qit wab iauf dak ninf
The boys are very put out, so they come to tease the girls.
(c) Two Relationships between Marriage and Flirting. The two themes outlined above address marriage and flirting. The most pervasive thematic focus in the song, however, is the conflict inherent in the relationship between marriage and flirting experienced by the individuals concerned. Two different relationships are presented. In the first, marriage and flirting are sequentially connected; the logical progression of flirting is towards cross-cousin marriage. This is the sentiment expressed in line 45: (female singers) 42 Naik liat dangk git qet
(female singers) People thinking of each other come to rest midway.
43 mul-diagngf eb
Cross-cousins.
44 Niongk liat dangk git qet
The girls want to sit and talk with the boys, so they rest midway.
45 Mak dint mongf vux diut
If you have wives, boys, then go home and take care of them.
Another example is at lines 90 and 91: (female singers) 88 Daik sat het deif dot
(female singers) No matter what the truth is, we will say the boys have wives.
106 Chien Mei-ling
89 Hlaib sak het deif dot
No matter whether it is false, we will say the boys have wives.
90 Diangt zix lius mek dint bongf heix
Go home and part with your wives.
91 Diuf niongk hveb lel ghob
Only then will we two be willing to accompany both of you.
In contrast with the themes outlined above, many lines present a more complex and perhaps problematic model of the relationship between marriage and flirting, a dialogical one. Generally in these scenarios, the flirting does not lead to marriage; in that regard both are simply independent of each other. The next two extracts both vividly express the second complex relationship between marriage and flirting. The first two lines, 29 and 30, are like a Greek chorus in their effect; line 30 offers another vivid metaphor portraying the flirting scene, and the pleasure of those there: (female singers) 29 Liob huf yiok
(female singers) The boys love to talk to us.
30 Ghas hfaf seik
The boys and girls are talking and singing together like ducks playing joyfully in the water.
31 Bongt neif meb niangt gait nongt
We do not know why we sit next to you all the time.
32 Bongt lif dak jut liuk
We do not know we are accompanying others’ husbands.
Another example is: (male singers) 263 Benf vux ob vak liuf
(male singers) The girls are good looking and speak well, but have two hearts.
264 Lal vux ob vak ghek
Like the good fields, which can grow millet twice each year.
265 Ib pit hvib qongk bongf
One heart accompanies their husbands.
266 Ib pit hvib dak lif
Another heart accompanies us.
(d) Personal Sentiments. Another obvious thematic thread is the various individual personal sentiments, feelings and emotions – including shyness, joy, solitude, longing and passion – in connection with marriage or flirting. Some
Tensions between Romantic Love and Marriage 107
lines describe the shyness with which both the boys and girls approach flirting, for example: (male singers) 22 Naik liuf ment let lab.
(male singers) Strangers will become familiar with each other by singing and talking.
23 Meit ob lab nit nil.
Just come to say a few words.
24 Lab def lab det jaf.
But we do not know how to talk to you.
25 Xid xeit meb gid weif.
We feel very shy to you.
(female singers) 72 Ob lik git ment let xef
(female singers) The girls do not love to talk,
73 Naik liuf ment let lal
until meeting the strangers like you.
74 Diut ux qib ux naif
Look, we do not wear our clothes appropriately.
75 Heb ghaf heb det ghaf
Put on one sleeve of our coat only.
76 Heb neif heb det rinf 16
Put on one sleeve of our coat only.
77 Xid xeit meb gait weis
We feel very shy to you.
Some lines describe the joy and delight associated with flirting:17 (female singers) 60 Liuf del let
(female singers) When we hear the drums, we feel delighted.
61 Deif del ghait
When we hear the words of the boys, we feel happy.
Other examples are: (male singers) 238 Nef huf yiok
(male singers) The girls love to talk to us
239 Ghas hfaf saik
like ducks joyfully playing in the water.
(male singers) 196 Dad nef mak qet ninf
(male singers) The girls will fool people.
197 Xongx juf daik qet liuf
The boys are happy about that.
108 Chien Mei-ling
In addition to joy and similar emotions, the lyrics also express strong longing, often in combination with a sense of anticipated solitude or forlornness. These sentiments mostly arise in contexts of the second, complex relationship between marriage and flirting: (female singers) 277 Mek hveb at nit hfuf
(female singers) Your words are just talk.
278 Mek hvib at yat yiok
Your hearts are far away.
279 Juk eb fat hongx vongf
Your hearts are like a river flowing to the falls.
280 Juk hob fat hongx bif
Like a fog in the midst of the hillsides which has lifted.
281 Mek zod but mek bongf
You will go back to be with your wives.
(male singers) 284 Nef cent niangt diek vik, muldiangf
(male singers) The bird is singing in the nest. cross-cousins
285 Dliut diut dot jut mongf
Let the girls marry others.
286 Nongf heit ob dak nif
We two can only sit together a while, and talk a while.
Finally, desire and passion are expressed and alluded to in the song. This, however, is not done as explicitly as are the emotions and feelings outlined above. Only a few lines have such descriptions and allusions, for example, in line 226: (male singers) 224 Meb dlab let lil rak renf
(male singers) You two are good at talking.
225 Lab qet juf wak wek
Your hearts are distant.
226 Ment ghad ghuf dek deif
Girls touch the boys with their bodies.
227 Dlek xongx dangf dlek dok
You desire to be close to the boys, as you desire to be near the fire in the winter cold.
228 Diuf xongx nongf wek yef
We two boys are all alone.
Another example is line 106: (female singers) 106 Mangk hongx xangt bif deif.
(female singers) We just do not want to let go the boys’ hands.
Tensions between Romantic Love and Marriage 109
Discussion The various themes and scenarios presented in the lyrics of this love-song, examples of which are given above, provide complex expressions of the nature of Miao marriage and flirting, and more. They include both the entangled relationships between marriage and flirting, and associated individual personal sentiments. This section discusses the role of love-songs such as this one – with its specific aesthetic and iconic performance – in constructing Miao social life, as well as social reproduction. The Miao love-song emphasizes a central, highly visible social feature of Fangf Bil village: affinal alliance or cross-cousin marriage. This is clearly noticeable in certain linguistic and social phenomena, such as the Dravidian kinship terminology, in conjunction with the ideal and practice of prescriptive cross-cousin marriage. Citing my own earlier research, the principle of prescriptive marriage is particularly obvious in the changes in address terms for cross relatives. Unmarried male speakers are addressed as but, male crosscousin, and mul, cross-cousin; unmarried female speakers are also addressed as mul. This is an abbreviated form of mul diangf (FZS/MBS), that is, an unmarried woman whose relationship with her cross-cousins – including potential husbands – is thereby indicated. She may, and indeed is expected to, marry a male having the same relationship with her. (Chien 1999: 46)
In effect this means that classificatory cross-cousins within the village are under very strong pressure to marry among themselves. This situation is highly visible performatively within the love-song duet and the iconic content of certain lines. Singers and audience are all classificatory opposite-sex affines. This is social evidence that instantiates the core structure of the Miao terminology system and its bilateral marriage system within the village. The ideal of affinal alliance is a thematic focus within the text, highlighted both implicitly and explicitly. For example, in lines 154 to 157, we read: ‘We are cross-cousins to each other. … We will become husbands and wives anyway. Do not mind if we have to part from each other for now’, as in village endogamy, for example, lines 366 and 367: ‘You girls marry someone else within your own village’ (emphasis added). The lyrics reinforce this with the repetitive use of affinal kin terms by both male and female singers – mainly but, mul, mul diangf, maib yut, indicating cross-cousins – throughout the more than four hundred lines. The consistent presentation and reaffirmation of local cross-cousin marriage contrast with lines which make highly visible the coexistent fluid nature of Miao marriage and the associated conflict in flirting contexts. The tension inherent in marriage and flirting in this Miao village is obvious. Elopement is a social dimension that throws some light on this. In this Miao village, much like before 1949, marriage occurs after a brief period of flirting; it may or may not involve romantic love and is divided more or less evenly between public
110 Chien Mei-ling
marriages and elopements. This is not just a relatively recent or post-1949 development (Chien 2005a). Public marriages are conducted according to – and must comply with – the positive rule of prescriptive cross-cousin marriage with village endogamy. Most elopements, however, violate village endogamy. They may or may not be between cross-cousins; they are very occasionally with non-Miao. They occur without the prior consent or knowledge of the bride’s family and are finalized when she crosses the threshold with her groom in the middle of the night. A bride considering elopement must weigh the uncertainties and far greater ambivalence of a cognatic marriage against the alternative: the certainties and normal ambivalence of a restricted marriage. Nevertheless elopements do occur – in fact, in approximately half of all marriages – highlighting Miao ambivalence about the collective marital ideal. This also raises the issue of individuality: individuals strongly resist the severe social constraints represented not just by cross-cousin marriage, but by cross-cousin marriage with village endogamy (Chien 2005a). The recorded love song performatively highlights the uncertainties about marriage and flirtation, in particular, the two different relationships between marriage and flirtation as experienced by individuals. In the first the natural development of the flirting is towards the local ideal of cross-cousin marriage: they are connected. For example, lines 42 to 45 and 88 to 91: ‘If you want to court the girls, go home to separate from your wives first. If you don’t, then let’s stop our flirtation and you guys go home and take care of your own wives.’ In these lines, flirting and being married cannot exist at the same time, even though on the surface it seems that extra-marital flirting is in fact underway at that moment. In complete contrast, there are lines which express the much more fluid and negotiable relationship which also exists between marriage and flirting in this Miao village. Marriage does not necessarily follow flirting, nor preclude it. Similarly, extra-marital flirting occurs (institutionalized extra-marital flirtation; see also Chien 2009a); flirting simply does not naturally or necessarily stop because of marriage. The two are simply independent of each other for the individuals involved. Lines 29 to 32, sung by the women, illustrate this using a rhetorical strategy: ‘The girls do not know why they desire to sit next to the boys always; they do not know their flirting companions are somebody else’s husbands.’ And again, lines 263 to 266 use a clearly poetic metaphor: ‘The girl has two hearts: one is for her husband, and the other is for the boy.’ The lyrics poetically reveal the fluid, conflicting and dialogical features of marriage and flirtation. Finally, individual personal sentiments, which also highlight the ambivalence between marriage and flirtation, are another major feature and thematic focus. The emotional element is significant. These elements are also seen in nonfestival flirting activities. These not only index how kinship and affinal alliance are valued, and their social connotations: a polite tone and playful, cheerful atmosphere is also created as Miao boys and girls take conversational turns while flirting (Chien 2005b).
Tensions between Romantic Love and Marriage 111
The personal emotions presented through the lyrics, however, are more complicated than simply playful interaction. In addition, personal feelings such as shyness, joy, solitude, longing and passion are also performatively presented, and thereby legitimated. In contrast to the joyful emotions that are mostly linked to marriage-connected flirtation, sentiments such as personal longing and feeling solitary are usually expressed in the context of the more complex dialogical relationship between marriage and flirtation. Lines 277 and 281 create a strong impression of such lonely feelings using metaphors: (female singers) 277 Mek hveb at nit hfuf
(female singers) Your words are just talk.
278 Mek hvib at yat yiok
Your hearts are far away.
279 Juk eb fat hongx vongf
Your hearts are like a river flowing to the falls.
280 Juk hob fat hongx bif
Like a fog in the hills which has lifted.
281 Mek zod but mek bongf
You will go back to be with your wives.
In these lines, the personal feelings – forlornness, anticipation of parting and loneliness – are expressed metaphorically by images from nature such as the ‘river flowing to the falls’ and the ‘fog in the hills which has lifted’, that is, gone away. They also express ambivalence over how those who participate in the extra-marital, ‘flirtation’-style of Miao courtship feel, this time focusing on the negative aspect: ‘Your words are just talk. Your hearts are far away… . You will go back to be with your wives.’ The emotions lyrically expressed in this lovesong, which belong to the personal, the individual sphere, are not simply a panhuman or universal psychological ‘thing’ – they are closely linked to the tension and ambivalence inherent in the local flirting and marriage context. And it is this sociological context which informs and celebrates the nature of ‘Miao individuality’.
Conclusion: Performing Miao Individuality Following Strathern’s aesthetic and iconic approach, studying individuals in their specific cultural context, this chapter has focused on an example of the performative use of language in the conduct of social life, thereby foregrounding Miao individuality. Tensions such as the antithesis between the individual and society, the value of the individual and the modern sense related to the indigenous conceptualization of individuality have been discussed in anthropology since Durkheim. Miao love-songs offer an opportunity to rethink this debate: they imaginatively expose the tensions between local individuals and their society.
112 Chien Mei-ling
Tensions between individuals and society are common, it would seem, in most human societies. A Durkheimian view suggests that the individual is always struggling with an awareness of his or her individuality as opposed to collective forces and representations on the level of ideology.18 The Miao in Fangf Bil face this dilemma too, as is clearly demonstrated by certain social institutions, such as duolocal residence and institutionalized extra-marital flirtation, and brought into high relief performatively in the contrast and tensions between marriage and flirting, expressed in the lyrics of love-songs. The content and thematic emphasis of Miao love-songs – particularly the dialogical relationships between flirtation and marriage – make the tensions between individual experience and sentiment and affinal alliance clearly visible. Different cultures have different concepts of the ‘individual’: the Western notion of the modern individual is of someone morally autonomous. Hindu world renouncers ideologically exist outside their social world but sociologically remain part of it. Fangf Bil Miao have their individuality, which has its own aesthetic linguistic dimensions: the love-song discussed above highlights individual personal experience and sentiments, while at the same time showing the ambivalent position of the individual with regard to a central aspect of local society. Rapport and Overing (2000: 185) describe individuality as ‘tied inextricably to individual consciousness, to that unique awareness and awareness of awareness, which is the mark of human embodiment. … human beings come to know themselves within the world by way of cognition and perception, thoughts, feelings and imaging, which are unique to them.’ Faced by the collective social ideal of marriage, Fangf Bil Miao villagers address their consciousness to performatively highlighting the political nature – in the sense of social dynamics – of feelings and imagination, that is, of individuals’ sentiments, emotions and desires. In this way they performatively legitimate the experience of the individual, thereby countermanding the strong forces of the collective ideal of marriage. Significantly, like Strathern’s notion of the ‘dividual being’ in Melanesian society, the Miao indigenous conceptualization of individuality is the ‘paired individual’. There are recurrent pervasive linguistic references to people in pairs throughout the more than four hundred lines of this Miao love-song. Consider lines 224 to 227: (male singers) 224 Meb dlab let lil rak renf
(male singers) You two are good at talking.
225 Lab qet juf wak wek
Your hearts are distant.
226 Ment ghad ghuf dek deif
Girls touch the boys with their bodies.
227 Dlek xongx dangf dlek dok
You desire to be close to the boys, as you desire to be near the fire in the winter cold.
Tensions between Romantic Love and Marriage 113
228 Diuf xongx nongf wek yef
We two boys are all alone.
The thematic emphases here are individual personal desire, strong longing and loneliness. In every line, however, the singing voice and the addressee are in fact not ‘one person’, but ‘two persons’. I argue that this is a performative device which demonstrates that the individual is essentially a social being, and further, that individuals identify themselves that way, happily. In brief, Fangf Bil Miao villagers performatively express their unique indigenous individuality using this text. Individualism has long been considered a definitive aspect of modern, Western ideology. Consequently debate on individuality has also involved the concept of modernity (Dumont 1986: 4). This chapter has argued that the Miao view of individuality also demonstrates a sense of the modern. It has presented observations of the collective social ideal of affinal alliance, the Miao awareness of individual personal identity and the tensions between individual sentiments and affinal alliance in various institutionalized contexts, all performatively instantiated in the love-songs. The sense of the modern present in Miao individuality as indicated in the love-song lyrics is, however, not a development in the direction of an extreme, strictly Western individualism. Miao individuality is characterized by an ongoing negotiated competition between a collective ideal – cross-cousin marriage within the village – and individual sentiment, including the playfulness and thinly veiled erotic nature of extra-marital flirtation in the local Miao flirtation and feelings of personal solitude expressed by the local love-song lyrics. The complex but essentially resolved relationship between Miao society and the Miao individual is performatively legitimated. And it is important to note that the audience consists of precisely those ‘young people’ who will give birth, literally and figuratively, to the next Miao generation. This can be linked to Dumont’s rethinking of Western individualism. Dumont saw totalitarianism as an attempt to ‘turn the clock back’ and restore the collectivism that Western individualism had undermined: Indeed, totalitarianism expresses in a dramatic way something we keep running into in the contemporary world: individualism is all powerful on the one hand and, on the other is perpetually and irremediably haunted by its opposite. … This coexistence in the contemporary ideology of individualism and its opposite comes forth more forcefully than ever at the present stage of research. In this sense, the individualistic configuration of ideas and values characterizes modernity, but it is by no means coextensive with it. (Dumont 1986: 17)
Examining how individualism relates to a local, historical and cultural view of the person or self, I conclude that ‘Miao individuality’ has a substantial aesthetic dimension in the performance and social dynamics of this Miao love-song in the form in which it has been conventionally transmitted, both before and after 1949; moreover, ‘Miao individuality’ is essentially modern.19
114 Chien Mei-ling
In response to the general approach of the classical debates and theories on Western individualism and modern development in the field of anthropology, and in line with Dumont’s reflective aspect and with the linguistic, aesthetic dimension articulated by Marilyn Strathern, this chapter further suggests that the performance is an imaginative outlet whereby these contemporary Miao villagers in eastern Guizhou can release emotional and ideological tensions between the ideal of marriage and their individually lived experiences embodying personal feelings, eroticism and various freedoms. Secondly, the lyrics and performance of the love-song clearly demonstrate that Miao villagers construct ‘individuality’ and ‘modernity’ in their own terms, both in their lives and performatively. Miao individuality challenges the whole evolutionist positioning of ‘individuality’ and ‘individualism’ in the historical Western anthropology tradition, outlined with its thinly veiled implied heirarchy predicated on Western individualism occupying a superior and more advanced position. There is an irony to this: contemporary Miao villagers’ vibrant maintainence of their own particular individuality is itself an expression of cultural individuality in international society. In this sense, it constitutes a modern-day ‘morally autonomous individualism’.
Notes 1. I collected this material at an institutionalized flirting event that took place in conjunction
with a bullfight during my village-based fieldwork related to marriage and flirting conducted for more than fifteen months between November 1998 and February 2000. 2. This is clear from Dumont’s own words (1986: 9, 16, original emphasis): ‘A system of ideas and values current in a given social milieu I call an ideology. I am calling the system of ideas and values that characterizes modern societies modern ideology … . Certainly, Durkheim saw individualism quite clearly as a value, but he did not work it indelibly into his vocabulary: he did not adequately emphasize the distance created by this value between modern man and all others; only by failing to do so could he come, in the passage from The Elementary Forms that Descombes pinpoints, to imagine that modern societies might go through a communal “effervescence” similar to that of Australian tribes.’ 3. According to Marriott, the South Asian theory of a person was as ‘dividual’ or ‘divisible’. Each person absorbs different material influences (1976: 11). 4. The five groups are the southern, central, western, eastern and northern Miao (Yang 1998). 5. I am using the dialect pronunciation of Miao in eastern Guizhou for Hmub terms, in which the final consonants are not pronounced but indicate tones. For example, I refer to the village of Fang31 Bi11 as ‘Fangf Bil’ in the discussion and the song extracts, following village pronunciation as closely as possible. 6. The central group comprises the northern subgroup in the Wuyang and Qingshuijiang river basins, a southern subgroup living in the mountain region of the Duliujiang River and an eastern subgroup in the transitional region between the Yun-Gui Plateau proper and adjacent uplands. 7. The case was identified as a ‘real’ sibling marriage, which should have been avoided. 8. FZD stands for father’s sister’s daughter; MBS, mother’s brother’s son; MBD, mother’s brother’s daughter; FZS, father’s sister’s son. Hereafter, I follow the kin abbreviations:
Tensions between Romantic Love and Marriage 115
F=father, M=mother, B=brother, Z=sister, G=sibling, E=spouse, S=son, D=daughter, P=parent, C=child, e=elder, y=younger, ms=man speaker, ws=woman speaker, etc. (Barnard and Good 1984: 4). 9. Gad ghat, agnates, ‘hosts’, refers to the in-group, and khait, affines, ‘guests’, is used to refer globally to non-in-group members and other outsiders. 10. Flower belts or clothes are exchanged as gifts. 11. The four subcategories of the ‘young’ songs are diut hxad vangx, qint hxad, et hed and iof het. The love-song discussed below belongs to diut hxad vangx. 12. The Miao phonetic system is composed of 31 consonants, 26 vowels and 8 tones. 13. Wearing the same style of Miao clothes is in fact a strict requirement for marriageability (Chien 2005a, 2009b). 14. Unlike Western ‘duet’ performances, the four never all sang together. 15. Reputation is extremely important in Miao society (Chien 2005a, 2009b). 16. The meanings of lines 75 and 76 are the same. 17. Joyful emotions are linked in some lines with joking and teasing, but it is difficult to convey the jokes across languages. 18. Though this is different from Mauss (see above). 19. That is, both before and after 1949.
References Barnard, Alan and Anthony Good. 1984. Research Practices in the Study of Kinship. London: Academic Press. Bauman, Richard, and Charles L. Briggs. 1990. ‘Poetics and Performance as Critical Perspectives on Language and Social Life’, Annual Review of Anthropology 19: 59–88. Caton, Steven C. 1987. ‘Contributions of Roman Jakobson’, Annual Review of Anthropology 16: 223–60. Chien Mei-ling (簡美玲). 1999. ‘Relationship Terms, Cross-Cousin Marriage, and Gender Identity: The Fanpaizai Miao of Eastern Guizhou’. Paper presented at the Workshop on Kinship and Economy on the Yun-Gui Plateau, Institute of Ethnology, Academia Sinica, Taipei, 19 May 1999. ———. 2005a. ‘Miao Elopement in Eastern Guizhou: Ambivalent Collective and Individual’ (貴州苗人的私奔婚:集體與個人的曖昧), Taiwan Journal of Anthropology 3(1): 49–86. ———. 2005b. ‘Conversation, Joke and Alliance: An Analysis of the Miao Daily Talk of Courting’ (語言、戲謔與聯姻:貴州苗人平日游方說話的分析), NTU Studies in Language and Literature 62: 247–380. ———. 2007. ‘ “You two and we two are life-long mates”: Gender Images and Pairing Ideals in Miao Songs from Southeastern Guizhou’ (「你倆是我倆一輩子的丈夫」 :黔東南苗族情歌語言的兩性意象與結伴理想), Journal of History and Anthropology 5(2): 115–49. ———. 2009a. ‘Extramarital Court and Flirt of Guizhou Miao’, European Journal of East Asian Studies 8(1): 135–50. ———. 2009b. Sentiment and Marriage among the Miao in Eastern Guizhou (貴州東部高 地苗族的情感與婚姻), Guiyang: Guizhou University Press. Clastres, Pierre. 1987. Society against the State: Essays in Political Anthropology. New York: Zone Books. Dumont, Louis. 1986. Essays on Individualism: Modern Ideology in Anthropological Perspective. Chicago: University of Chicago Press.
116 Chien Mei-ling
Holquist, Michael. ‘Introduction’, in Speech Genres and Other Late Essays: M.M. Bakhtin, Cary Emerson and Michael Holquist, eds. Austin: University of Texas Press, pp. ix–xxiii. Homans, George C., and David Schneider. 1955. Marriage, Authority, and Final Causes: A Study of Unilateral Cross-Cousin Marriage. Glencoe, Ill.: The Free Press. Lukes, Steven. 1973. Émile Durkheim. New York: Penguin Books. Marriott, McKim. 1976. ‘Hindu Transactions: Diversity without Dualism’, in Transactions and Meanings, B. Kapferer, ed. Philadelphia: ISHI Publications. Mauss, Marcel. 1979. A Category of the Human Mind: The Notion of Person, the Notion of ‘Self’, in Sociology and Psychology: Essays. Boston: Routledge & Kegan Paul. Radcliffe-Brown, A.R. (1924) 1952. ‘The Mother’s Brother in South Africa’, South Africa Journal of Science 21: 542–55. Rapport, Nigel, and Joanna Overing. 2000. Social and Cultural Anthropology: The Key Concepts. New York: Routledge. Strathern, Marilyn. 1988. The Gender of the Gift: Problems with Women and Problems with Society in Melanesia. Berkeley: University of California Press. ———. 1992. After Nature: English Kinship in the Late Twentieth Century. Cambridge: Cambridge University Press. Tang Chunfang (唐春芳). (1959) 1986. Marriage Songs of the Miao (Miaozu hunyin ge 苗族婚姻歌), Miao Folklore Data (Minjian wenxue ziliao 民間文學資料), Vol. 17. Guiyang: Association of Chinese Folklore and Art in Guizhou. Urban, Greg. 1996. Metaphysical Community: The Interplay of the Senses and the Intellect. Austin: University of Texas Press. Yang Tingshuo (楊庭碩). 1998. The Historical Process of Signs in Human Groups: The Case of Miao Nationality Names (人群代碼的歷時過程:以苗族族名為例). Guiyang: Guizhou Renmin Chubanshe.
Chapter 5 Modalities of the One-Child Policy among Urban Migrants in China Chang Kuei-min
Introduction During 1979 and 1980, the one-child-per-couple policy of the People’s Republic of China became a fundamental national policy and was enforced nationwide through the mobilization of the Chinese Communist Party (CCP) and the party-state (White 1990, 1994; Scharping 2003).1 Afterwards, birth planning was written into the PRC Constitution as the obligation of each married couple.2 Almost simultaneously, the leaders of the party-state launched an economic reform programme, vowing to lead the country in the direction of the policy goal of becoming a relatively well-off society (xiaokang shehui) based on ‘socialist construction’ in the next millennium. In the context of economic reform and China’s eagerness for modernity, and through a series of scientific discourses made by a small number of natural scientists and influenced by the dominant international population discourse in the 1970s, the country’s huge population was blamed for China’s backwardness (Grimes 1998; Greenhalgh 2003a, 2003b, 2005). Given that peasants constitute the majority of the population in the country, the one-child-per-couple policy was put into force without seriously taking into account the social resistance that was bound to happen,3 given that this policy was bound to cause tremendous demographic change and create social costs on an unprecedented scale. Despite the fact that the party-state has gradually changed its totalitarian appearance, withdrawing from almost every aspect of social life during the reform era, there is little sign that it is loosening its policy on population control. In this context, the regime still intrudes into each household, wielding its power over every woman’s body through the matrix of birth control. But although the state may never relinquish its control, equally society never ceases its resistance. Indeed, the contest between ‘modernity’ and a
118 Chang Kuei-min
‘traditional peasant childbearing culture’ has followed in the footsteps of migrants into the cities, where formerly the one-child policy did not counter much resistance from citizens due to the specific socioeconomic structure of urban regimes (Nie and Wyman 2005), with their attitudes of exclusiveness to all non-members.4 Mass rural-urban migration, stimulated by the economic reform policy and mostly undertaken by people of child-bearing age, thus turned the urban space too into a battlefield over birth control and resistance, therefore raising a fundamental challenge to China’s efforts to control its population growth. Table 5.1 shows the surge in rural-urban migration, especially in the late 1990s. Table 5.1 Scale of migrant population (ten thousand) Year
Migrant
Total population
Migrant population ratio (%)
1982
1,139
100,818
1.1
1990
2,952
113,368
2.6
1995
4,758
120,778
3.9
2000
14,439
126,583
11.4
2005
14,735
130,628
11.3
SOURCES: Guanyu 1982 nian renkou pucha zhuyao shuzi de gongbao; Guanyu 1987 nian quanguo 1% renkou chouyang diaocha zhuyao shuzi de gongbao; Guanyu 1990 nian renkou pucha zhuyao diaocha shuju de gongbao; Guanyu 1995 nian quanguo 1% renkou chouyang diaocha zhuyao shuju de gongbao; Guanyu 2000 nian renkou pucha zhuyao shuju de gongbao; Guanyu 2005 nian quanguo 1% chouyang diaocha zhuyao shuju de gongbao.
Since 1979 the Chinese state’s population discourses have tackled the issues of both quantity and quality (Greenhalgh and Winckler 2005). The central theme of the state’s population discourse in the last decade has been to stabilize low fertility and increase the quality of the newborn population. Having fewer children (only one child) is portrayed as ‘scientific, civilized and advanced’, and population policy serves the state’s developmental goal of transforming the ‘great populous country’ into ‘a strong country of human capital’.5 Parents are encouraged to fulfil their reproductive aspirations through one or two welleducated children to parallel the state’s assertion that the country’s prosperity lies in the quality (suzhi) of the next generation. This idea of good parenting has been promoted through school education and has gradually taken root among the population (Murphy 2004; Zhang 2007). In the past decade, a systematic propaganda campaign has been undertaken to transform the population’s reproductive behaviour. The Chinese state’s near obsession with civility and modernity is publicly visible, being seen in signs and posters everywhere. Its implementation is through the state’s ‘power of naming’
Modalities of the One-Child Policy among Urban Migrants in China119
in defining what good citizenship is and thus dividing the population into various subgroups by official naming practices (Anagnost 1997). How do we interpret the effects of these discourses in people’s daily lives? Researchers have pointed out the convergence of official and individual discourses and interpreted them as an ideological shift in the population (Murphy 2004; Zhang 2007). Normally, there is a time lag between the appropriation of an idea and its materialization, the process of which is complex and full of contradictions. In the Chinese case, the decline in the fertility rate has been occurring prior to the change in reproductive ideas, which further complicates our observations of the interworking of competing ideologies and thus their underlying power structures. If we were to take the interviewees’ words at face value, we would neglect the cultural continuity displayed in people’s unconscious everyday practices, despite the reality that a stable relationship between a cultural form and its content is no longer desirable for the agents (Liu 2000). In this sense, the idea of resistance is useful in examining the violation of a certain state policy, regardless of the protagonist’s rhetorical acceptance of the official ideology. Given the oppressive characteristics of the regime and the political sensitivity of the one-child policy, research examining resistance to the policy is still taboo in the People’s Republic of China,6 resulting in a lack of literature based on field research, both domestically and internationally. Although some Western scholars have endeavoured to sketch a picture of peasant resistance to birth control in rural areas (Wasserstrom 1984; Greenhalgh 1993, 1994, 2003; Anagnost 1995; Greenhalgh and Li 1995; White 1994, 2003), the resistance strategies of urban migrants have not yet attracted much academic attention, leaving a research vacuum whose exploration is the main task of this chapter. The first section defines reproductive rights in China and analyses how the principles of distributing these rights undermines the initial purposes for which they were granted. The second section depicts the birth-control mechanism and the intrinsic problem in its implementation. To show how migration has become the major parameter of peasant tactics in resisting the birth-control system, the third section analyses four case studies based on my field enquiries in Shenzhen, Dongguan, Shanghai and Kunshan, conducted in April 2005 and March 2006 respectively. The final section argues that the tactics rooted in migration are eating away at the efficiency of China’s system of population control, which is grounded in the dual registration system (hukou system). This has some implications for the future membership of peasant migrants in the cities and for the transformation of the CCP’s governance of China’s population.
Reproductive Rights in Question In this chapter, reproductive rights contain two concepts of citizenship rights, originally proposed by Marshall (1994) as two of the three developmental
120 Chang Kuei-min
phases of citizenship: civil rights and social rights. However, unlike the experience of Britain, where citizenship rights progressed sequentially with the development of the capitalist market economy, in China social rights were introduced by the state early in the 1950s. At that time, the Chinese Communist Party adopted the Soviet model of economic development, with planning flows for all goods and materials, while emphasizing the industrial sector and downplaying agriculture. Starting from the first five-year plan in the 1950s, the Chinese government progressively implemented population registration in all cities and rural areas by setting up a registration system linked to its control of all grain sales. Following the Soviet economic development strategy, the state separated the population into roughly two categories, urban and rural hukou, and severely controlled spatial mobility between urban and rural areas. In the urban areas, the state instituted a subsystem, the danwei system, which guaranteed the livelihood of all citizens, including food provision, education, jobs, medical care, housing and pensions. But in the rural areas the commune system, which broke down in 1984, offered only very limited social welfare by comparison. This dual hukou system remains the basis of population governance in China. Unlike social rights, certain aspects of civil rights were granted only in the mid-1980s, when the state loosened the prohibition on population mobility and imposed a series of reforms to property rights. Political rights, however, have remained strictly controlled until today. The term ‘reproductive civil rights’ (hereafter ‘reproductive rights’) refers to the freedom a couple has to decide the number and spacing of their children, which is an extension of the personal freedom intrinsic in civil rights – the sovereignty of one’s own body. Therefore, given that birth planning was written into the constitution as an obligation for all citizens, the continuous struggle for more than one child or for male children can be viewed as representing resistance to the state’s claim on women’s bodies. In talking about ‘reproductive social rights’ (hereafter ‘reproductive security’), I am referring to access to free or moderately priced reproductive services, which include the relevant perquisites, subsidies, medical care and reproductive leave, which are guaranteed only by one’s enrolment in the reproductive insurance set up by the state. Reproductive security has historically been viewed as part of female workers’ welfare in the People’s Republic and has gradually been established through the labour system, although mainly in urban areas. In the era of economic reform, the general inefficiency of state-owned enterprises could no longer afford employees’ social security. The financial burden was especially heavy for those mostly employing female workers. Thus, the Ministry of Labour and Social Security issued the so-called ‘Measures (for the Trial Implementation) of Child-bearing Insurance for Staff and Workers in Enterprises’ (qiye zhigong shenyu baoxian shixing banfa; hereafter ‘the Measures’) in 1994. Hence, enterprises’ responsibilities for ensuring workers’ reproductive security (and ultimately the state’s responsibility before the reform making
Modalities of the One-Child Policy among Urban Migrants in China121
state-owned enterprises financially independent) have been gradually transferred to society. The Measures were intended to relieve the burden on the state and establish the principle that reproductive risks should be placed on society as a whole, while still exempting individuals from shouldering the insurance premium. However, implementation regarding migrant workers seemed doomed to fail in the initial aims proclaimed in Article 1, which stipulates that female workers should be granted economic compensation, medical treatment and health care during their child-bearing years. This is due to the principles adopted in the articles. First, the Measures explicitly regulated the principle of localism in managing reproductive insurance (Article 3). In other words, the insurance is designed in accordance with local conditions, including the planned number of newborn children, economic development and the financial situation. Also, the maximum collection rate is set at 1 per cent of the total payroll for the enterprise (Article 4), indicating that only the local government will cover deficits in the reproductive fund. This has dissuaded some municipalities who have attracted mass migration, such as Beijing and Shanghai, from covering those whose permanent places of registration are beyond their boundaries. Moreover, compliance with the birth planning regulations is a prerequisite for receiving insurance benefits (Article 7); that is, those with pregnancies outside the state’s birth-planning policy are excluded from receiving any substantial support from the state. The two-sided classification of state planning has created social categories (most obviously the planned and unplanned) which have become the guiding principles of resource distribution (Greenhalgh 2003a). In other words, as long as there is non-compliance with birth control, any policy based on birth planning will inevitably leave a considerable portion of the population outside the scope of the state’s services, even if those policies are aimed to improve the women’s reproductive predicament. The thinking on birth control and the principle of distributing welfare based on one’s place of registration still act together as forces undermining the state’s attempt to provide reproductive security and its capacity to oversee a substantial amount of the female population in an era of vigorous population mobility.
Migrant Birth-Control Mechanisms In order to control the reproductive behaviour of the floating population, in addition to family-planning apparatuses at each administrative level, the state has combined the relevant departments to establish a comprehensive government mechanism, including joint jurisdiction by different local government authorities (both those of permanent residence and current residence) and cooperation between all departments concerning migrant activities in the cities. Table 5.2 lists the institutional planning relevant to birth-control regulations.
122 Chang Kuei-min
Table 5.2 Migrant birth-control mechanisms Agents of the state and fields of control
Methods Control and punishment
Birth Permanent Issue birth permit; planning residence issue marital and child-bearing status certificate; conduct propaganda; demand pregnancy check-up paper; collect excessbirth fine. Current residence
Public security
Examine marital and child-bearing status certificate; conduct propaganda; patrol.
Service and rewards
Loopholes
Provide contraceptives and health education.
Depend on migrants’ autonomous cooperation; difficulties in tracking down migrants.
Provide contraceptives and health education.
Depend on migrants’ autonomous cooperation; difficulties in tracking down migrants.
Permanent Birth permit residence required to register the newborn child.
Bribe, negotiated excess-birth fine, or one parent’s sterilization or permanent contraceptive in exchange for the newborn child’s registration.
Current residence
Few incentives for registration.
Marital and childbearing status certificate required in applying for residence permit; contraceptive measures required in obtaining local hukou.
Modalities of the One-Child Policy among Urban Migrants in China123
Table 5.2 continued Agents of the state and fields of control
Methods
Industrial and commercial administration
Marital and childbearing status certificate required in applying for work permit and business licence.
Vendors need no licence to conduct business.
Social security
Birth permit required in applying for child-bearing benefits.
Small amount Few migrants of pension have childsubsidies. bearing insurance; pension subsidies not enough to acquire compliance.
Control and punishment
Education
Service and rewards
Loopholes
Single-child tuition cut; score benefits in high school entrance exam.
Employer
Require marital and child-bearing status certificate before hiring; responsible for employees’ birth planning; accept inspection from local government.
Should pay for employees’ contraceptive expense.
Most businesses do not care about employees’ reproductive behaviour.
Hospital
Birth permit required in undertaking delivery; practice ‘remedial measure’ (abortion); practice contraceptive surgeries.
Issue birth certificate; provide pregnancy check-up service (with charge); practice contraceptive surgeries.
Many hospitals require no birth permit to undertake delivery.
124 Chang Kuei-min
Table 5.2 continued Agents of the state and fields of control
Methods
Mass organizations
Conduct propaganda and mobilization.
Landlord
Require marital and child-bearing status certificate before signing a lease; informant assistance.
Court
Freeze violator’s bank account; collect excess-birth fine.
Local law enforcement team
Hunt down and detain violators and accomplices; collect excess-birth fine.
Control and punishment
Service and rewards
Loopholes
Most landlords care about nothing but rents.
Difficulties in tracking down migrants.
SOURCES: Population and Family Planning Law (Zhonghua renmin gongheguo renkou yu jihua shengyu fa); Birth planning decrees and documents of various provinces and cities (Guangdong Province, Foshan City, Shanghai, Linyi County, Yulin County, Dongguan City, etc.); Field research.
The birth-control mechanism seems comprehensive at first glance, but in practice it depends greatly on migrant women’s voluntary obedience. According to the public statement of the former Minister of the National Population and Family Planning Commission, China’s floating population accounts for 10 per cent of the total population but is responsible for 60 per cent of ‘unauthorized births’ and even 80 per cent in the Pearl River Delta.7 This reveals the limited efficacy of the birth-control system on the floating population. The main reason is the failure of effective supervision,8 whose initial cause is, I argue, the fundamental principle rooted in China’s population governance – the territorial jurisdiction of the hukou system. In the next section, I shall describe further how this principle has created loopholes for resistance by peasant migrants in the cities to which they migrate.
Urban Resistance Brought to Light For a number of reasons, there has not been much academic research adopting ethnographic research methods or a sociological point of view in the study of
Modalities of the One-Child Policy among Urban Migrants in China125
the reproductive behaviour of migrant workers in the cities. First, the study of reproduction among migrant women is strictly confined to the area of reproductive health, and thus avoids dragging in two fundamental ‘national policies’ of China – the hukou policy and the one-child policy.9 Secondly, it is not easy even for local Chinese scholars to work with migrant communities because of the latter’s lack of trust in urban citizens, caused by the hostility that confronts them in the cities, especially when it comes to possible studies of their illegal violations of the one-child policy. Thirdly, the common image of the floating population’s reproductive behaviour carries the stigma of so-called ‘excess-birth guerrillas’ (chaosheng youji dui), the study of which usually focuses on the demographic goal of solving the alleged problem of the high fertility rates of migrant women, which has been proved to be a fallacy by Goldstein et al. (1997). This chapter deals with the reproductive practices of ordinary urban migrants and their tendency as peasants to resist the one-child policy, both structurally and culturally, thus trying to fill the gap in the empirical studies mentioned above. Based on in-depth interviews and participant observation, forty-four migrants (in all twenty-nine couples) were interviewed in 2005 (21– 30 April and 2–16 August) and 2006 (16–29 March) in Shenzhen, Dongguan, Shanghai, Kunshan and Beijing. Taken together, these migrants’ permanent places of registration covered a total of eleven provinces, and their professions in these five cities included production-line worker, office personnel, mid-level management positions in both foreign and domestic factories, taxi-driver and individual entrepreneur (getihu). In addition to basic information such as permanent place of registration, age, educational level, monthly salary, housing and work experience, each female informant’s reproductive history was recorded in detail, including the year of marriage registration, the cost of the birth permit, time(s) and location(s) of delivery, reproduction fee, extra-birth fine, contraceptive method and pregnancy check. Husband and wife were interviewed either together or individually. Each interview lasted from one to one and a half hours, and ten of the twenty-nine couples were interviewed twice. Moreover, during the 2005 national holidays I accompanied one of my informants back to her home village in northern Jiangsu Province, which provided me with an excellent chance to interact with the villagers as a visiting friend and thus acquire a better understanding of their views about birth planning. In this chapter, I argue that migrant workers have more latitude in resisting the state’s one-child policy and in implementing their ‘unplanned’ reproductive decisions thanks to the territorial jurisdiction inherent in the hukou system. The latter has created institutional loopholes in the birth-control system in an era of population mobility, thus turning the urban space into an effective sanctuary for continued reproduction. To expand on this argument further, I shall draw on the four case studies just mentioned, each describing how peasant migrants have insinuated themselves into these institutional loopholes in the matrix of birth control.
126 Chang Kuei-min
Rural Sister and Urban Brother Zhou went to Shenzhen in 1992 – the very year of Deng Xiaoping’s tour of the south, which launched a new wave of economic growth after the deadlock caused by the 1989 agitation. Zhou has a rural hukou, his wife Wu has an urban one. They got to know each other through a friend and married in 1994. Zhou and Wu, who are both from Sichuan Province, have been working in Shenzhen for more than ten years and normally go home once a year, like most migrant workers in the city. Wu did not change her registration status at marriage, and since she and her husband seldom live at his home she has not been asked to do so. Wu could envisage making this particular decision because social security for the urban hukou is much better than for the rural hukou.10 Their first child, a daughter, was born in 1996 and was registered in the father’s household. Four years later Wu gave birth to their second child, a son,11 who is apparently unauthorized. Therefore, Zhou invited the birth-planning staff in his village to a meal, hoping to find some way to legalize their son’s birth. ‘I was wondering, so I asked. They were talking without any ulterior motives, while I was listening there with a specific purpose.’ Thus, Wu registered their son in her household without paying any fine for violating the plan – which would have been 17,000 yuan in the year of their son’s birth. ‘I got out of it! I have taken advantage of this loophole in the state’s control’, said Zhou proudly. Due to the lack of exchanges of information between registration authorities, Zhou and Wu – one couple with two registration books – have a rural daughter and an urban son. What is worth mentioning here is that, although the children belong to different hukou, they both live together with their parents in Shenzhen, where having two separate hukou makes no difference to their education because of their outsider status – they can attend the neighbouring private school set up specifically for migrant children. The city regime’s exclusiveness to non-members has paradoxically made it possible for Zhou and Wu to raise both of their children together, without distinction of hukou.
Factories as Refuges Around the time of our first interview in 2005, Huang and Zhang were about to register their second daughter – ten years after the first one had been born – and were expecting to be fined. In the following interview, I asked Huang about the fine. He answered by saying, ‘I don’t know why … it seems to be the spacing … anyway, no one came to ask for the money’. According to the birth-planning regulations, all women should use contraceptive methods after their first child,12 and those working outside their permanent places of registration should send back pregnancy check-up papers every three months.13 However, Zhang did not obey the requirement to have
Modalities of the One-Child Policy among Urban Migrants in China127
an IUD (intrauterine device) inserted, as she was able to avoid this because of her long-term absence from her home village. ‘Their hands are not long enough to reach here!’, she said. During her pregnancy, she handed in a copy of her old record to the local birth-planning staff, who had been sent to the foreign factory and were supposed to check hundreds of female workers in, say, a single afternoon. ‘It is easier to fake in the factory because they won’t examine everyone carefully. There are too many people waiting!’ The birth-planning officers patrol their communities, checking pregnant women to see whether their pregnancies are authorized. ‘It was heard that some were found out … they would adopt measures of compulsion … we were in the factory when they came knocking at our door. When we went home, they were already off duty.’ Although the state has stipulated that all businesses should take responsibility for the birth planning of their workers and that migrants’ landlords should cooperate with the local birth-control authorities to implement the policy, these two parties put little effort, if any, into complying with these requirements. This is because foreign enterprises care much more about ‘material production’, while most landlords will not do anything to undermine the economic benefits they obtain from the migrants.14 The foreign factories, exempt from many substantial duties owing to their essential contribution to the local coffers, thus became a refuge for those attempting to escape the state’s surveillance of reproduction.
Crossing the Provincial Line Zhu and Liu have been working in a bag-making factory (where they got to know each other and married in 1997) since 1989 and 1993 respectively. They are one of the few inter-provincial couples in the factory because peasant migrants tend to form their own social circles according to their places of origin, despite the fact that the factory is a closed living space.15 Liu, originally from Henan Province, switched her registration to Zhu’s household in Guangdong Province after they got married, which unexpectedly brought her key advantages later in her childbearing history – Liu underwent permanent contraceptive surgery in 2004, after the birth of her second son. Liu gave birth to her first son in Henan by virtue of her mother’s suggestion that she might receive better care from her natal family. She did not have an IUD inserted because she was no longer registered there and was beyond the jurisdiction of the local birth-control apparatus. Four years later, Zhu’s family urged the couple to have a second child because they were already in their thirties – nearly too old to have children, in their view. ‘[The conception was] not unexpected, but should be considered as planned’, Zhu said. This time Liu decided to deliver her baby in Dongguan so that she could look after her elder son, who had to go to school then. Zhu explained:
128 Chang Kuei-min
[If we had been] at home, we wouldn’t have been able to deliver the child. [But we were] outside, there was no forcible control … generally speaking, the birth-control staff don’t look for those leaving the village. If you are at home, they will go to your house … unless you hide … normally you won’t be able to deliver [the unauthorized baby]. … But it still depends on the situation; those having only one child are allowed to deliver the second one as long as they don’t go out during the ‘movement’ [clearing up the unauthorized births] in April and August … . In the rural area, only one [child] is absolutely unacceptable. The general ideal is two boys, one girl.
They spent less than 100 yuan as a fee to register their second son without being fined, for Liu has undergone sterilization surgery.16 The story of Zhu and Liu, which is by no means unique, shows how territorial jurisdiction, grounded in the hukou system, can be used to undermine the state’s endeavours to control reproductive behaviour at a time of vigorous population mobility. Nowadays inter-provincial marriages have gradually become normal, and the state is doomed to face the predicament that more and more reproductive processes are skipping the boundaries of the fixed territorial jurisdiction of population control, thus moving beyond the state’s surveillance.
The Fees of Immunity Bao and Chen are inveterate rule-breakers of the birth-control system. Chen followed Bao to Dongguan with both parents’ consent in 2001 when she was seventeen, three years younger than the legal age of marriage, and gave birth to her first child, a daughter, one year later. ‘According to the regulations, the hospital shouldn’t have delivered our child. But since she was in labour, they couldn’t just ask us to leave,’ Bao said, describing their illegal childbirth. Not until they ‘got married’ in 2004 could they register their daughter, after paying a fine of 8,000 yuan. ‘The hospital certified my daughter’s birth because she was born there, after all … that’s still fine if they did not [issue her birth certificate]. We can ask someone at our home village to help [my daughter get registered] as long as we have money; that’s how it goes, the economic society’, Bao said.17 Asked if Chen had had an IUD inserted, the couple showed a sense of unease: ‘No …, we’ve been staying outside. … Last year we asked for leave and went home for several days. They [the birth-control staff] didn’t come to ask about that because we hadn’t been back for a long period of time.’ Moreover, Chen does not have to send back the check-up paper as long as they pay the local birth-control office a small sum of money (less than a hundred yuan) every six months. ‘My family helps to handle the payment. Don’t tell others about this, since it is illegal’, Bao said. Since Chen does not follow the regulations concerning contraceptives, she has not applied for the marital and child-bearing status certificate for the floating population, which is required by everyone leaving her or his permanent place of registration according to the regulations of the State Council.18 Her avoidance of this regulation has not caused her any trouble so far in finding a
Modalities of the One-Child Policy among Urban Migrants in China129
job or staying in Dongguan, which indicates that, so long as migrant workers do not have any dealings with the authorities where they reside, the surveillance mechanism symbolized by the certificate does not work. ‘Here is my idea: girls and boys are the same’, said Bao, trying to explain that his attempt to have a second child had nothing to do with the first child’s gender; ‘I saw with my own eyes that my auntie’s son died when she was in her fifties. Now she doesn’t know what to do [with her later years]. … The thing is, bearing the second child costs 20,000 to 30,000 yuan (including the fine), so you have to consider your own financial situation. We plan to have our second birth either this year or next’, Bao said, taking this for granted. They gave birth to a son the next year. The lifting of the ban on population mobility and the economic boom in coastal areas have created unprecedented economic opportunities for rural migrants. Once the peasants have increased their economic power, they are both more willing and more able to pay the fine for illegal births (Goldstein, White and Goldstein 1997). Economic migration like Bao’s and that of most other peasant migrants empowers their families through economic advantage, so that they can redeem their reproductive rights in the cities.
Conclusion The resistance to the imposition of the one-child policy among migrant workers and the demographic consequences for the Chinese population and Chinese social structure are vivid demonstrations of how the resistance of the weak impedes the intended policy goals of the dominant (Scott 1985). Nevertheless, the struggle is never an all-or-nothing victory by either side, as demonstrated by previous research (Greenhalgh 2003a; Scharping 2003; White 2003), for low fertility and an aging population are already producing the prediction of a heavy pension burden on the state. Similarly, the distorted sex ratio is causing difficulties in finding brides due to the increasing ratio of males to females. Finally, the citizenship of the unplanned population is aggravating the problem of social justice. In urban areas, the regular violations of the one-child policy by migrants have seriously damaged the efficiency of the state’s birth-control mechanism. Thus, inspired by the international discourse promoted in the Cairo Programme of Action and political reform in the direction of the rule of law, the state has gradually changed its methods of governance to improve the implementation of the population policy, thus making it less coercive and more rights-oriented (Greenhalgh 2001; Winckler 2002). One example of this is the delivery clinics for migrant women set up by the Shanghai Municipality, which charge only one-third of the average delivery fee, with vaccines for the babies thrown in.19 The implication of this shift from birth control to health service is that the state is trying to regain its control over women’s bodies by providing limited reproductive welfare, that is, by exchanging reproductive security for
130 Chang Kuei-min
reproductive rights. Interestingly, in the battlefield over birth control and resistance, social rights and civil rights are employed separately as tactics by both parties. We have seen on the one hand that peasants make use of their spatial mobility and the resulting economic empowerment to claim autonomy over their bodies. On the other hand, the state offers a package of social rights (reproductive welfare, pension, medical care, single-child education benefits, and so on) to bring them back into the population-control system. This change of strategy by the state is its response not only to social demands for equal citizenship rights, but also to the appeals staged at the UN’s International Conference on Population and Development, in which China is an important actor. Yet, the Chinese state simply addresses such issues as reproductive security instead of reproductive rights, which embrace a couple’s right of freedom to make reproductive decisions. Nonetheless, this welfare is only second class compared to that provided to proper urban residents. Thus, migrant women acquire this second-class citizenship in exchange for sovereignty over their bodies and also at the cost of the right to be treated equally as Chinese citizens, as is written in the constitution. Resistance to birth control has forced the state to refine its methods of governance, thus strengthening the state’s control while simultaneously reinforcing migrants’ second-class status in the cities. Ironically, the Chinese state’s project to transform society identifies and confronts tradition, but has been hindered by its own socialist past. The discourse surrounding citizenship and the improved treatment of the migrant population have come only as a reaction to the targeted population’s refusal to accept the state’s modernist dreams.
Acknowledgements The field research on which this chapter is based was conducted under a research project led by Professor Jieh-min Wu, ‘The Formation and Changing Forms of Differential Citizenship in China’s Emerging Market Society: An Analysis of the Migrant Labor’, financed by National Science Council, Taiwan, 2004–2006.
Notes 1. From the beginning of the 1970s, the Chinese government carried out a birth-control policy called wanxishao (connoting later childbirth, longer spacing and fewer children), which decreased the number of births allowed to each peasant woman from six to three (White 2003: 186). The one-child-per-couple policy was promulgated in 1979 and enforced nationwide one year later. Since then, the Chinese government has made slight adjustments to its birth-control policy, but the basic one-child principle has not been changed. 2. Article 49: ‘Both husband and wife have the duty to practice family planning.’
Modalities of the One-Child Policy among Urban Migrants in China131
3. The reasons for peasant resistance to the one-child policy are culturally and structurally intertwined, but included the desire for male children to continue the family line, carry out agricultural work and look after the elderly. In the reform era, the disentangling of peasant life from local cadre control has provided room for peasant resistance to the state’s birth-control policy to develop. For detailed accounts, see Wasserstrom (1984), Greenhalgh (1993) and White (2003). 4. The socioeconomic structure of the unit (danwei) system in the cities provided incentives for obedience to the one-child policy. Except for controlling the life chances of its members, the unit had a complete and well-regulated social security system, including medical insurance, education, pensions, food rationing, etc. Wasserstrom (1984), Greenhalgh and Li (1995), White (1990, 2003) and Nie and Wyman (2005) also mention the influence of economic and political structures on the effectiveness of birth control. 5. ‘Main missions of caring for the girls’, http://www.chinapop.gov.cn/zyzt/zhzl/ ganh/200403/t20040302_138443.html, 2 March 2004, accessed 6 May 2009. 6. Greenhalgh (1994) described in detail the difficulties she faced in conducting field research in 1988. What is surprising is that (on the basis of my interviews and contact with several Chinese scholars and graduate students from Guangzhou, Shanghai and Beijing during 2005 and 2006), even at present, almost twenty years since Greenhalgh’s research, this issue still goes beyond the political boundaries laid down by CCP rule. 7. ‘Minister of the State Population and Family Planning Commission, Zhang Weiqing’s speech about the population problem’, http://www.ymsrkw.gov.cn/Article_show1. asp?ArticleID=245, 24 January 2007, accessed 19 April 2010. ‘Guangdong to become the most populated province’, http://www.southcn.com/news/gdnews/nanyuedadi/ 200705120012.htm, 12 May 2007, accessed 19 April 2010. 8. Solinger (1999) observed the difficulties of inter-department cooperation in the city government, which have provided latitude for the collusion of migrants and officials. See also Zhang (2001). 9. To understand how the dual hukou system in welfare provision affects migrants’ health, see the insightful observations of Xiang Biao (2003). 10. For example, not until 2005 did Zhou’s home town start a medical insurance scheme for peasant residents to join. However, Wu has been enjoying a complete social security system as a citizen – as long as she keeps paying 20 yuan per month – even though she has lived out of the area for such a long time. 11. Strictly speaking, the son is the third child because Wu aborted her second pregnancy (caused by her IUD falling out) because of the inconvenience of taking care of two children at the same time and also ‘the principle of eugenics’. 12. Although the National Population and Family Planning Law stipulates the right of citizens to choose contraceptive methods, IUD insertion is the most commonly used method for its highly effective and provider-oriented characteristics. 13. The spacing is locally different – usually one month in the cities and towns, but three months in rural areas. 14. This statement is based on the field observation of eight foreign factory sites (with highly fluctuating worker numbers from 30 to 2,300 in March 2006) and oral confirmation by three factory managers. 15. The factory, which had 2,300 workers in March 2006, functions not only as a workplace but also as a closed living space, with its own generators, water-treatment system, dormitories, restaurants, grocery stores, playgrounds, kindergartens, security guards, etc. All single workers are required to live inside the factory, while married couples and their children are accommodated in the buildings (rented from local residents) right next to it. 16. In her field research in three Shaanxi villages, Greenhalgh (1994) observed how the state’s birth-control regulations, combined with peasants’ child-bearing ideas, informs the concept of ideal family size – two children per couple, one of whom is a boy – in the local practice of the birth-control policy. This community-based ideal family size became the consensus of the local cadres in enforcing birth control, which means that they will
132 Chang Kuei-min
not force those who have not achieved the ideal family size to undergo permanent contraceptive measures. A similar practice is mentioned in Zhu’s statement about his home village. This is why they could avoid the fine at the cost of Liu’s permanent contraception. 17. To register a newborn child, one needs the parents’ registration book and the birth certificate issued by the hospital where she or he was born. 18. Measures on Administration of Family Planning for the Floating Population (liudong renkou jihua shengyu gongzuo guanli banfa), Articles 7, 8. 19. ‘Shanghai established ten special arranged delivery spots for floating population’, http:// big5.xinhuanet.com/gate/big5/news.xinhuanet.com/autopub/2004-07/27/content_ 1658319.htm, 27 July 2004, accessed 11 September 2006.
References Anagnost, Ann. 1995. ‘A Surfeit of Bodies: Population and the Rationality of the State in Post-Mao China’. In Conceiving the New World Order: The Global Politics of Reproduction. Faye D. Ginsburg and Rayna Rapp, eds. Berkeley: University of California Press, pp. 22–41. ———. 1997. National Past-Times: Narrative, Representation, and Power in Modern China. Durham: Duke University Press. Goldstein, Alice, Michael White and Sidney Goldstein. 1997. ‘Migration, Fertility, and State Policy in Hubei Province, China’, Demography 34(4): 481–91. Greenhalgh, Susan. 1993. ‘The Peasantization of the One-Child Policy in Shaanxi’. In Chinese Families in the Post-Mao Era. Deborah Davis and Stevan Harrell, eds. Berkeley: University of California Press, pp. 219–50. ———. 1994. ‘Controlling Births and Bodies in Village China’, American Ethnologist 21(1): 3–30. ———. 2001. ‘Fresh Winds in Beijing: Chinese Feminists Speak Out on the OneChild Policy and Women’s Lives’, Signs: Journal of Women in Culture and Society 26(3): 847–88. ———. 2003a. ‘Planned Births, Unplanned Persons: “Population” in the Making of Chinese Modernity’, American Ethnologist 30(2): 196–215. ———. 2003b. ‘Science, Modernity, and the Making of China’s One-Child Policy’, Population and Development Review 29(2): 163–96. ———. 2005. ‘Missiles Science, Population Science: The Origins of China’s OneChild Policy’, The China Quarterly 182: 253–76. Greenhalgh, Susan, and Jiali Li. 1995. ‘Engendering Reproductive Policy and Practice in Peasant China: For a Feminist Demography of Reproduction’, Signs: Journal of Women in Culture and Society 20(3): 601–41. Greenhalgh, Susan, and Edwin A. Winckler. 2005. Governing China’s Population: From Leninist to Neoliberal Biopolitics. Stanford: Stanford University Press. Grimes, Seamus. 1998. ‘From Population Control to “Reproductive Rights”: Ideological Influences in Population Policy’, Third World Quarterly 19(3): 375–93. Liu, Xin. 2000. In One’s Own Shadow: An Ethnographic Account of the Condition of PostReform Rural China. Berkeley: University of California Press. Marshall, T.H. 1994. ‘Citizenship and Social Class’. In Citizenship: Critical Concepts. Bryan S. Turner and Peter Hamilton, eds. London: Routledge, pp. 5–44. Measures (for the Trial Implementation) of Child-bearing Insurance for Staff and Workers in Enterprises (qiye zhigong shenyu baoxian shixing banfa), No. 504, issued by the Ministry of Labour and Social Security on 14 December 1994.
Modalities of the One-Child Policy among Urban Migrants in China133
Measures on Administration of Family Planning for the Floating Population (liudong renkou jihua shengyu gongzuo guanli banfa), approved by the State Council on 6 August 1998 and promulgated by Decree No. 1 of the State Family Planning Commission on 22 September 1998. Murphy, Rachel. 2004. ‘Turning Peasants into Modern Chinese Citizens: “Population Quality” Discourse, Demographic Transition and Primary Education’, The China Quarterly 177: 1–20. Nie, Yilin, and Robert J. Wyman. 2005. ‘The One-Child Policy in Shanghai: Acceptance and Internalization’, Population and Development Review 31(2): 313–36. Population and Birth Planning Law of the People’s Republic of China adopted at the 25th meeting of the Standing Committee of the Ninth National People’s Congress on 29 December 2001. Scharping, Thomas. 2003. Birth Control in China 1949–2000. New York: Routledge Curzon. Scott, James C. 1985. Weapons of the Weak: Everyday Forms of Peasant Resistance. New Haven: Yale University Press. Solinger, Dorothy J. 1999. Contesting Citizenship in Urban China: Peasant Migrants, the State and the Logic of the Market. Berkeley: University of California Press. Wasserstrom, Jeffrey. 1984. ‘Resistance to the One-Child Family’, Modern China 10(3): 345–74. Winckler, Edwin A. 2002. ‘Chinese Reproductive Policy at the Turn of the Millennium: Dynamic Stability’, Population and Development Review 28(3): 379–418. White, Tyrene. 1990. ‘Post-Revolutionary Mobilization in China: The One-Child Policy Revisited’, World Politics 43(1): 53–76. ———. 1994. ‘The Origins of China’s Birth Planning Policy’. In Engendering China: Women, Culture, and the State. Christina K. Gilmartin, Gail Hershatter, Lisa Rofel and Tyrene White, eds. Cambridge, Mass.: Harvard University Press, pp. 250–78. ———. 2003. ‘Domination, Resistance and Accommodation in China’s One-Child Campaign’. In Chinese Society: Change, Conflict and Resistance. Elizabeth J. Perry and Mark Selden, eds. London: Routledge Curzon, pp. 183–203. Xiang, Biao. 2003. Migration and Health in China: Problems, Obstacles, and Solutions. Asian MetaCentre Research Paper, Series 17: 1–39. Zhang, Hong. 2007. ‘From Resisting to “Embracing?” the One-Child Rule: Understanding New Fertility Trends in a Central China Village’, The China Quarterly 192: 855–75. Zhang, Li. 2001. Strangers in the City: Reconfigurations of Space, Power, and Social Networks within China’s Floating Population. Stanford: Stanford University Press.
Chapter 6 The ‘Culture’ of World Cultural Heritage Eveline Bingaman
Introduction While the notion of culture may have been invented in Western academic circles, those circles no longer have monopoly over its usage. Even in academic circles, such as in the field of cultural anthropology itself, there has never been a single, standardized definition that holds the discipline together. The result of the notion’s entry into popular discourse has been a diversification of ‘culture’ as a concept so that its usage has been stretched to encompass the material and the immaterial, abstract notions as well as concrete practice, so much so that the term ‘culture’ seems to have become a catch-all term. What we have here is what Mayfair Yang (1996) refers to as a ‘travelling theory’ and what Arjun Appadurai calls an ‘ideoscape’. As Appadurai (1996: 36) argues: As a result of the differential diaspora of keywords, the political narratives that govern communication between elites and followers in different parts of the world involve problems of both a semantic and pragmatic nature: semantic to the extent that words (and their lexical equivalents) require careful translation from context to context in their global movements, and pragmatic to the extent that the use of these words by political actors and their audiences may be subject to very different sets of contextual conventions that mediate their translation into public politics.
The key terms in this diaspora that are being translated are ‘culture’, ‘cultural heritage’ and ‘cultural preservation’. The travelling theory is that culture is something that can and should be preserved. The notion of culture as it exists in popular discourse, while still wide and sometimes seemingly all-encompassing, does tend to have certain characteristics. The first is the idea of culture as something ‘unique and bonded’: ‘Western concepts assume “that people can be classified into mutually exclusive bounded groups according to physical and behavioral differences,”
The ‘Culture’ of World Cultural Heritage 135
based on putative shared ancestry’ (Linnekin and Poyer 1990: 2, cited in Picard and Wood 1997: 7). This idea leads to what I call the ‘bubble concept’ of culture. Imagining cultures as mutually exclusive gives the impression that they have solid, encompassing boundaries. This leads to the idea that, once a culture has been penetrated by some outside force, be it modernity, globalization or tourism, it will deflate until it disappears entirely under the weight of assimilation. The second characteristic of the notion of culture in Western discourse is that cultural diversity is valuable in the same way that biodiversity is valuable. This implies that its loss is a great tragedy, like the disappearance of polar bears or giant pandas. This is captured well in the opening lines of UNESCO’s Convention Concerning the Protection of World Cultural and Natural Heritage, which states that the ‘deterioration or disappearance of any item of the cultural or natural heritage constitutes a harmful impoverishment to the heritage of all the nations of the world’ (UNESCO 1972: 1). Related to this idea is the fear that allowing the disappearance of diverse culture may have unforeseen consequences that affect everything else along with it, much like the extinction of part of an ecosystem. Finally, the Western discourse contains the idea that to commodify and/or commercialize culture is to defile it, thus changing its meaning and destroying it. While to sell ‘cultural’ artwork is acceptable (e.g. masks from Bali, tie-dye tapestries from Yunnan, beaded bracelets from Thai hill tribes), to turn other aspects of culture into performance for the sake of profit or simply to ask for compensation for allowing the outsider to experience it is unacceptable. These themes shape the Western discourse that is present around the world and conflict with other discourses of culture, despite the stated common goal of cultural preservation. The Western discourse is felt in Lijiang – a World Heritage Site in Yunnan Province, People’s Republic of China – from two sources. The first is UNESCO’s World Heritage Committee through its designation of Lijiang as a World Cultural Heritage Site, and thus through its regulations and reviews determining whether it is properly maintaining that status. The second is through Western tourists, who now form a small minority of the ten thousand or so tourists who arrive daily, but still have a significant impact. In this chapter I shall leave aside Western tourists in order to consider the more obvious loudspeaker of this discourse: UNESCO and its World Cultural Heritage Committee.
The ‘Culture’ of World Cultural Heritage The Convention Concerning the Protection of the World Cultural and National Heritage (hereafter ‘the Convention’) was created at the United Nations Educational, Scientific, and Cultural Organization (UNESCO)’s 17th
136 Eveline Bingaman
general conference held in Paris on 16 November 1972. As the introduction to the convention reads: Noting that the cultural heritage and the natural heritage are increasingly threatened with destruction … Considering that the protection of this heritage at the national level often remains incomplete … Considering that parts of the culture or natural heritage are of outstanding interest and therefore need to be preserved as part of the world heritage of mankind as a whole … Having decided, at its sixteenth session, that this question should be made the subject of an international convention, Adopts this sixteenth day of November 1972 this Convention. (UNESCO 1972: 1, italics in the original)
With the adoption of the Convention, the intergovernmental World Heritage Committee (WHC) was created to oversee the protection of what were thereafter termed ‘Cultural and Natural Heritage of Outstanding Universal Value’. At the same time, the World Heritage Fund (WHF) was created. States were encouraged to become members, which entailed signing and upholding the Convention in their own countries, as well as making biannual ‘compulsory and voluntary’ donations to the WHF.1 At the same time, they were asked to submit to the WHC ‘an inventory of property forming part of the cultural and natural heritage’ within the boundaries of their nation. Sites submitted would then be assessed by the advisory boards to the WHC, the International Council of Monuments and Sites (ICOMOS) and/or the International Union for Conservation of Nature and Natural Resources (IUCN), to see whether they qualified as ‘cultural’ or ‘natural’ heritage as defined in Articles 1 and 2 of the Convention.2 If assessment revealed that the property submitted did in fact constitute a heritage of ‘Outstanding Universal Value’, it would be inscribed on what thereafter became known as the World Heritage List (hereafter, ‘the List’). First, ‘Outstanding Universal Value’ is defined as: cultural and/or natural significance which is so exceptional as to transcend national boundaries and to be of common importance for present and future generations of all humanity. (Operational Guidelines 2005: Paragraph 49)
‘Cultural heritage’ is defined in the Convention as: monuments: architectural works, works of monumental sculpture and painting, elements or structures of an archaeological nature, inscriptions, cave dwellings and combinations of features, which are of outstanding universal value from the point of view of history, art or science; groups of buildings: groups of separate or connected buildings which, because of their architecture, their homogeneity or their place in a landscape, are of outstanding universal value from the point of view of history, art, science; sites: works of man or the combined works of nature and man, and areas including archaeological sites which are of outstanding universal value from the historical, ethnological or anthropological point of view. (UNESCO 1972: 2)
The ‘Culture’ of World Cultural Heritage 137
For a site to be considered for inscription on the World Heritage List under the terms of ‘Cultural Heritage’, UNESCO has defined six ‘cultural criteria’, which reveal a great deal about its concept and definition of culture. Since Lijiang is the case study presented here, I will focus on the three of the six cultural criteria that were used in determining Lijiang Old Town’s acceptance onto the World Heritage List, specifically criteria ii, iv and v. ii. to exhibit an important interchange of human values, over a span of time or within a cultural area of the world, on developments in architecture or technology, monumental arts, town-planning or landscape design; iv. to be an outstanding example of a type of building, architectural or technological ensemble or landscape which illustrates (a) significant stage(s) in human history; v. to be an outstanding example of a traditional human settlement, land-use, or seause which is representative of a culture (or cultures), or human interaction with the environment especially when it has become vulnerable under the impact of irreversible change. (UNESCO 2005: 17)
Even a cursory examination reveals how problematic these criteria can be for translation, let alone policy making. This discrepancy between ideals and the realities of implication has been a recognized point of criticism for researchers and scholars in the fields of tourism and heritage management. As Ashworth and van der Aa have expressed succinctly (2006: 148): [A] dangerous gap is increasingly evident between the goal and the evolving reality stemming from [the List’s] implication. Such a discrepancy is structurally embedded in the convention and inherent in the processes of its application. It can thus only increase to the detriment of original goals, leading to an unsatisfactory list, inadequate management and lost opportunities unless quite radical revisions of the approaches, restructuring of the organization and strengthening measures takes place.
First, looking at the definition of Outstanding Universal Value (OUV), as the term implies it assumes a universalist approach – the idea that one theory can be applied to all humanity. Universalism as a theory has largely been abandoned in academic circles in favour of relativism because of the problems that arise in attempting to claim that one theory applies to all people. It also contains several ambiguous terms without explanation for their qualification. For example, what does it mean for a site to be ‘significant’ or ‘exceptional’ or even ‘important’? What are the terms and by whose definition? We are left to assume that UNESCO has the power to identify these things through some gaze of universalism that sees a single truth that can be applied to all human beings. As for UNESCO’s criteria of what constitutes cultural heritage, (ii) notes that sites should ‘exhibit an important interchange of human values’, which leads to the question, what exactly are these ‘human values’? The implication is apparently that there is a universal definition in existence, similar to the way that universal ‘human rights’ are defined by the Geneva Convention. How the term ‘value’ is to be defined is also problematic, especially as UNESCO does
138 Eveline Bingaman
not spell it out. Criteria (ii) also indicates UNESCO’s idea that culture is contained within static ‘culture areas’, which may be visualized as a map of the world, colour-coded, as are most maps indicating national boundaries (like a bubble), or what Ulf Hannerz calls a ‘culture map of the world’ (2004: 2). Many anthropologists have come to reject the notion of a ‘culture area’ as an isolated entity (Steedly 1999), especially those working in South-East Asia, where both ethnicity and culture have proven to be exceptionally ‘fluid, constructed, always changing, always contested’ (Wood 1997). Secondly, the reference in criteria (iv) to an illustration of a ‘significant stage in human history’ also utilizes an evolutionary explanation of the difference between cultures. Lastly, criteria (v) sums up the WHC’s goal, which rests on an assumption that cultural change can be stopped, or even reversed. Finally, with regard to UNESCO’s definitions of cultural heritage, it is important that none of these specifically refer to living culture, that is, to cultural practices. The definition is restricted to ‘monuments, groups of buildings and sites’. It makes no reference to the lives or culture of the people living within those sites. However, when UNESCO’s advisory boards review sites to evaluate how well preservation is being carried out, these additional aspects of culture are taken into account when determining the site’s OUV. As will further be shown later in this chapter, the fact that clear distinctions have not been made, and definitions of ‘intangible heritage’ have not been added to these, is problematic. There are two key words that UNESCO uses when deciding the overall value of a site, which is where the inclusion of non-material culture comes into play. According to UNESCO’s Nara Document on Authenticity, the term ‘authenticity’ itself refers only to determining the credibility of the information it receives about the purported cultural value of the site. ‘Integrity’, however, becomes a more controversial concept when placed in contexts outside the Western discourse of culture as ‘bonded’. Calculating the ‘integrity’ of a cultural site includes everything deemed as ‘culture’ within its geographical boundaries. This includes the lives and daily practices of those living there. While UNESCO does make concessions that some change is ‘normal’ when evaluating a site, they still make decisions on whether that change is ‘within an acceptable range’.
Tourism and ‘the List’ There are several misconceptions in popular understanding of how the UNESCO World Heritage programme works in practice. The general impression is that sites are found by UNESCO representatives who are seeking out sites considered to be unique and valuable, which are then inscribed on to UNESCO’s List and singled out for efforts to preserve them. As a general idea this is not entirely incorrect, but it contains a few assumptions that are not quite accurate.
The ‘Culture’ of World Cultural Heritage 139
Contrary to popular conception, inscription on the list does not ensure international funding for the protection of the inscribed site. It is true that as a State Party to the convention, sites have the ability to apply for funding from the World Heritage Fund, but that ‘funding’ is limited to the WHC providing consultation by preservation experts and perhaps low-interest loans. It rarely includes actual handouts for preservation work (only in ‘exceptional cases for special reasons’; UNESCO 1972: 12). This is coupled with another misconception, namely that the WHC travels the world actively seeking ‘outstanding’ places to preserve. In reality, it is local sites that apply to the WHC asking to be added to the list. The application process is a long one and comes with a long list of requirements determining eligibility. One of these requirements is that local institutions and governments are already fulfilling the responsibilities of preservation. The WHC does not inscribe sites that cannot take care of preservation on their own. On the organization’s website, UNESCO explains what it believes to be the benefits for State Parties to ratify the Convention and have their sites inscribed on the List: ‘The overarching benefit of ratifying the World Heritage Convention is that of belonging to an international community of appreciation and concern for universally significant properties that embody a world of outstanding examples of cultural diversity and natural wealth.’ Along with such platitudes, the website also cites prestige, increased awareness and international cooperation as other benefits. However, the final section of this document acknowledges what applicants for World Heritage status already know: Finally, the inscription of a site on the World Heritage List brings an increase in public awareness of the site and of its outstanding values, thus also increasing tourist activities at the site. When these are well planned for and organized respecting sustainable tourism principles, they can bring important funds to the site and to the local economy. (UNESCO)3
The real benefit of inscription comes in the form of the UNESCO World Heritage Emblem, which notifies visitors that this site has received UNESCO’s stamp of approval and is one not to be missed. Despite whatever honourable intentions UNESCO may have had in 1972 when it first conceived the List, in practice it has become a form of branding for tourism development which now acts as a motivation for sites to apply to the WHC for inscription status. UNESCO itself is not blind to this irony and has its own response. As Christopher Alesevich (2008) clearly stated in an article entitled ‘Another Day in Paradise’: As with corporate logos created by graphic identity specialists, the World Heritage Emblem is to be displayed in particular ways. And as a result, certain royalties are expected, though not described as such. As the [Convention] puts it, ‘When commercial benefits are anticipated, the [World Heritage] Center should ensure that the World Heritage Fund receives a fair share of the revenues’.
140 Eveline Bingaman
For local sites hungry for development, kickbacks to UNESCO are a fair trade for the publicity they offer. In tourism development models, achieving inscription on the World Heritage list is often conceived as the first step. For example, in Izamal, Mexico, the Assistant Director of Tourism, Edgar Díaz, is quoted as saying: ‘By becoming World Heritage, we’ll have more investors. Upon having more investors, we’ll have more tourist infrastructure. That way, there would be greater tourism promotion, and you could have an economic influx; that is what the people need to support their families’ (quoted in Kugel 2006). Despite UNESCO’s recognition and gradual acceptance of the inscription– tourism nexus (Peters 2001), to non-locals and particularly Western observers the increased presence of tourism at heritage sites is often regarding as having an overwhelmingly negative impact on the site’s value. These concerns are in many ways valid and reflect a lack of management. They include things like stress on local resources, uneven access to the wealth produced by a tourism industry and disturbance to locals, who become ‘tourees’. However, the biggest concern is the abstract idea that a bubble has been penetrated which will lead to the deflation of a once thriving culture. UNESCO has had to deal with this built-in contradiction. The WHC has been forced to recognize the irony in the fact that the preservation they want is both made possible and obstructed by tourism development. By allowing for tourism at all, UNESCO is violating its bubble concept of culture.
Lijiang and UNESCO World Cultural Heritage China became a State Party to the World Heritage Committee by ratifying the Convention in 1985. For China, ‘the significance of UNESCO’s World Heritage stamp lies not so much in the approving of individual spots as in its restoration of China’s “five-thousand-year-old superior culture” to its rightful place’ (Nyíri 2006: 52). However, its blatant use as a marketing tool is hard to miss, especially when compared to other sites in South-East Asia where evidence of UNESCO’s inscription is limited to a plaque near the entrance. Lijiang’s local government began the application process for UNESCO inscription in 1995, the same year that the whole of Yunnan Province decided to pursue tourism as a route for development (Spensley 2003). In 1996 a major earthquake hit Lijiang, destroying 186,000 homes and damaging 300,000 more. While a clear statement has never been made regarding a connection between the earthquake and UNESCO’s acceptance of Lijiang’s application, it is generally assumed that the earthquake gave the situation a sense of urgency that lead to Lijiang’s inscription on the World Cultural Heritage List in December of 1997. Even now I have found that, when local Naxi speak of the recent change in Lijiang brought on by tourism development, they speak in terms of before and after the earthquake. The area covered in the inscription includes three sites: Dayan (大研), known colloquially as Lijiang Old Town; Shuhe, a small settlement located
The ‘Culture’ of World Cultural Heritage 141
four kilometres north-west of the Old Town; and Baisha, located eight kilometres further north at the site of Ming Dynasty frescos. As anticipated, upon inscription Lijiang’s tourist arrivals began to shoot up, nearly doubling from 1.1 million in 1996 to 2 million in 1998 (Li 2008).4 In 1999, advertisements were sent out by the local government asking for people to open up restaurants and guesthouses to accommodate the mass of tourism arrivals. However, tourism volumes quickly outpaced what the locals could provide, and Lijiang began to attract tourism entrepreneurs from other saturated tourism markets such as Banna and Guilin. While Lijiang had been running businesses catering to outsiders since the time of the caravan trails, the opportunity-minded Naxi quickly realized that, instead of running businesses themselves, they could rent out their Old Town houses to these entrepreneurs at exorbitant rates, several times more than their previous yearly incomes, and move into comfortable modern housing in the New Town. The result was that in a quick ten years, Lijiang had been transformed from a relatively sleepy minority town into a bustling tourism Mecca. It was then that UNESCO began receiving complaints.
Figure 6.1. Monument dedicated to the Inscription of Lijiang Old Town on UNESCO’s World Cultural Heritage List (photo by the author)
142 Eveline Bingaman
The ‘Yellow Card’ Rumour According to the WHC’s Operational Guidelines section on ‘Reactive Monitoring’ (UNESCO 2005: 169), all inscribed sites must be periodically reviewed by the WHC’s advisory boards to ensure that State Parties are effectively upholding the Convention at their inscribed sites. In practice this review is done every five years, meaning that Lijiang’s ten-year review took place in 2007. Unfortunately Lijiang’s review did not go well, and when the review board’s findings were reported to the 31st Annual Meeting of the World Heritage Council in Christchurch, New Zealand, rumours immediately began to spread that Lijiang had received a ‘yellow card’ (liangle huangpai, 亮了黄牌), that is, was being warned that it may be taken off the World Cultural Heritage List. According to UNESCO’s Decision 1449 regarding Lijiang Old Town: The World Heritage Committee, having examined Document WHC-7/31. COM/7B. Add 5,5 notes with concern the uncontrolled tourism and other development projects being carried out on the property, which might have a negative impact on its heritage values. (UNESCO 2007: Decision 1449)
What the supposed ‘yellow card’ actually amounted to was a verbal warning that raised three concerns:6 the amount of commercialism in the Old Town, the number of Naxi residents moving out, and the lack of management at Shuhe and Baisha, the two other sites also included in the Lijiang Old Town designation. In response to this verbal warning, the City of Lijiang hired a slew of researchers from a variety of backgrounds to help to compile a report. The first step was to review documents in all government departments. The next was to initiate a survey which involved dozens of researchers divided into groups responsible for such separate things as architecture, tourists, and shopkeepers. From their research, a report was compiled and presented to the local government for review. Next, it was presented to the provincial government for review before finally being presented to the WHC at the 32nd Annual Meeting in Ontario, Canada. The WHC’s response to the Christchurch report was immediately to send a ‘reactive monitoring team’ to gather more information. The team was sent to Lijiang in January 2008 and stayed for ten days. Their report was also submitted to the WHC in Ontario. The Lijiang City government report is not available to the public, or even to those who were involved in creating it. The report of the reactive monitoring team, however, became public in late 2008. In the report ICOMOS voiced its concerns clearly (UNESCO WHC-ICOMOS 2008: 20): Despite the legal framework being put into place, the lack of a vision for the overall identification of heritage values, tangible and intangible, may affect the implementation of these rules and regulations concerning procedures for the
The ‘Culture’ of World Cultural Heritage 143
approval of development projects … the conservation and management of the property continue to face severe threats to the values for which it was inscribed and these must be addressed by the State Party as a matter of urgency. In this regard, the mission suggested the site management authority to take a value-based management approach in the elaboration of a Conservation Master Plan and other conservation strategies … However, based on the on-site visits and observations, we believe that the site still retains outstanding universal value in spite of the large influx of the tourists … [The] conditions of integrity seem to be within the level of acceptable change … While authenticity of the site seems to be at risk, this is partly because of the change of lifestyle of the local community and partly because of the commodification and/or commercialization of the Naxi and Dongba culture. It should be stressed that, while the nomination dossier of 1996 referred mainly to the artistic and historic significance of the old town and its group of buildings, it is apparent that the Outstanding Universal Value of this property depends also on the extraordinary relationship between the tangible heritage and intangible heritage values, including its cultural identity. However the latter is at serious risk of losing its integrity if urgent steps are not taken.
Two key issues stand out in this long quote: the importance of Outstanding Universal Value in Lijiang’s inscription, and the assumed conflict between tangible and intangible heritage. I discuss these two issues below. One major problem the mission discovered was that Lijiang had been inscribed in 1997 without a statement of Outstanding Universal Value having been approved by the WHC. What Lijiang did have was a description of Lijiang as ‘an exceptional ancient town set in a dramatic landscape which represents a harmonious fusion of different cultural traditions to produce an urban landscape of outstanding quality’ (UNESCO 1997). What it should have contained are statements of Outstanding Universal Value for each of the three criteria under which Lijiang was inscribed. Without these statements, UNESCO is in an awkward position in which Lijiang has not actually violated anything they are able to cite, despite its ‘uncontrolled development’ and the exodus of local Naxi to the New Town. So now, ten years later, UNESCO is attempting to renegotiate in order to rectify this oversight, and to make it clear that its understanding of the Old Town’s value hinges upon ‘intangible heritage’ and ‘cultural identity’. What this means is that UNESCO is essentially changing the rules of the game. While it has not officially sanctioned Lijiang by giving it a ‘yellow card’, it has expressed its dissatisfaction with what Lijiang has done since being inscribed on the List in 1997 by sending a reactive monitoring team to explore the details. The result was that Lijiang’s ‘value’ needed to be restated in a manner going beyond the original value cited in Lijiang’s inscription, which designated Lijiang a World Cultural Heritage as ‘a cluster of buildings’ and never referred to that value as being contingent on the ‘intangible heritage’ now being cited.
144 Eveline Bingaman
Concepts of Culture in Conflict Now we can begin to see how different concepts of culture may create tension. While both the Chinese state and UNESCO are in agreement that culture needs to be protected and preserved, their individual discourses regarding what constitutes ‘culture’ places them in opposition to one another.
Development versus Preservation ‘You have to understand that for local officials, all they care about is Lijiang’s GDP (Gross Domestic Product), because that is all their supervisors care about. If Lijiang’s GDP continues to go up, their superiors are happy. If it slows or goes down, their supervisors want to know why and maybe they will lose their jobs.’7 Lijiang’s growth statistics are published on a monthly basis as evidence of the site’s successful management. For example, the 2007 report stated that Lijiang’s total GDP was 8.482 billion, up 13.9 per cent over 2006.8 It is a reminder that, within China, the state’s priorities are still first in line, and the importance of development is above and beyond anything UNESCO has to say. This highlights a conflict between UNESCO and the Chinese state in that, while the verbal warning ‘scared the living daylights’ (xia daole, 吓倒了) out of local officials, the local government takes a similar approach to the Convention as it does to minority culture: it will follow the Convention to the extent that it does not interfere with development. Right from the beginning, the supervisors of Lijiang wanted to make sure that preservation was not going to interfere with tourism. One major point of contention is the consistent problem of property borders. According to the Operational Guidelines, each site is to have a defined core zone and a buffer zone surrounding it (thus demarcating the boundary of the bubble). However, ‘In the process of elaborating a Conservation Master Plan for Lijiang Old Town, the relevant authorities attempted to reduce the core area and buffer zone of Dayan Old Town in order to allow the development of tourism-related projects at other sites of the property’ (UNESCO-WHC ICOMOS 2008: 12). This manipulation puts the Old Town Management Bureau in an extremely awkward position. As a bureau of the city government, its duty is to support government policies in the aim of increasing the city’s GDP. But the bureau is also charged with the duty of dealing with UNESCO and responding to its demands. It has, however, come up with an excellent strategy for easing the tension by appealing to UNESCO’s idea of culture as something bounded in history and tradition, and emphasizing Lijiang’s commercial importance in the past as a central stopover on the caravan trails. In an interview, He Shiyong (和士勇), head of the Old Town World Cultural Heritage Management Bureau, explained:
The ‘Culture’ of World Cultural Heritage 145
Lijiang is a business city. Most accurately it is a city of commercial culture. This goes back to the Han Dynasty [206 bc – ad 220], what we call the Ancient Horse and Tea Trail today. In the Naxi language, the Old Town has many names; their translations mean ‘warehouse’ or ‘a place for doing business’. If there was no business, there would be no Lijiang, so our commercial culture is also something that we need to preserve.
For the moment, Lijiang has managed to fight ambiguity with ambiguity by arguing that Lijiang’s ‘commercial culture’ extends to include modern-day tourism.
Integrity versus Good and Bad Culture The ‘bubble’ approach to culture of Western discourse promoted in Lijiang by UNESCO directly conflicts with the Chinese state’s idea that not all culture is ‘heritage’ and that some aspects of ‘culture’ per se need to be ‘overcome’. While the state considers the cultural changes that accompany development – increased education, access to technology, and a stronger connection between rural and developed areas – to be inherently good, for UNESCO each results in an additional puncture in Lijiang’s cultural bubble, allowing ‘value’ to escape like so many breaths of air. This is where the ‘intangible heritage values’ that ICOMOS cited in its reactive monitoring report clash with the goals of the Chinese state. The state’s ultimate goal of development is the eventual assimilation of ethnic minorities. They are allowed to keep their ethnic markers as long as they are ‘healthy’ (meaning harmless) and do not conflict with this goal of ethnic unity. Since UNESCO is not able to speak of its vision for Lijiang in terms less ambiguous than ‘intangible’ and ‘valuable’, we are left to guess what that vision actually looks like. When I try, I imagine a huge snow globe, the outer edges being the boundary lines of ‘the property’, as the Guidelines put it. Inside are the Naxi people, frozen in time, fully dressed in traditional clothes and engaged in petty business that does not extend outside that boundary. This is an exaggeration, of course, but does not seem all that different from the ambiguous messages UNESCO sends. The state, however, is much clearer in its vision for Lijiang. With its steady increase in GDP, Lijiang is a model for other economically backward but culturally unique areas.
Authenticity versus Commercialization The present situation also challenges the Western concepts of commercialization and commodification as being the enemies of authenticity, and reveals them to be subjective values impossible to measure. To begin with, it needs to be made clear that I was unable to find any Chinese term that adequately expressed the term ‘authenticity’ as it is meant in this context. The closest terms I came
146 Eveline Bingaman
across were zhenshixin (真實性) meaning ‘truthful’ or ‘reality’, and didao (地道) or yuanzhiyuanwei (原汁原味), both of which usually refer to food. When I interviewed Western tourists who used the word ‘authentic’, I tried to pin down what they meant by it. Not surprisingly none had a ready answer, and they mainly came up with a couple of key words like ‘old’, ‘traditional’ or ‘in its original form’. The word ‘authentic’ is tied to the fact that an item was not intended for commercial exchange, at least not outside a ‘local society’. Therefore, from the very beginning, tying market value or exchange value to a product immediately detracts from its ‘authenticity’ (Cohen 1988). However, if preserving culture means assuring its continued existence, Lijiang is ahead of the West in realizing that commercialization and commodification can be positive tools. As Fuat Firat (1995) argues: Cultures of all types – ethnic, national, regional and the like – that are able to translate their qualities into marketable commodities and spectacles find themselves maintained, experienced and globalized. Cultures that cannot or do not (re)present themselves in terms of marketable qualities, simulated instances, experience and products are finding themselves divested of members … [and] seem to vanish only to become museum items.
A good example of this in Lijiang is ‘dongba paper’. So-called ‘dongba culture’ refers to three things: dongba pictographic script, dongba dance and dongba music. The dongba script was traditionally written on a special kind of paper that was made specifically for collections of dongba texts. It was made from the bark of a particular kind of tree that was soaked, beaten into pulp form and finally dried on screens. Since indigenous religion became taboo during the Mao era, the production of this paper had stopped. In the 1980s, Resident Scholar at the Dongba Culture Research Institute, He Liming (和力民),9 had the idea of reviving this paper-making technology, mainly for the institute itself, since they were in the process of collecting texts and studying them, but had no original paper to copy them on. In recent years, starting somewhere between 2005 and 2007, a larger plan for dongba paper was conceived, which was to make a series of products from it that could be sold to tourists. This was originally a project of the Naxi Culture Development Company, but it has since been turned over to private ownership. Now in Lijiang there are a series of dongba paper shops, many with a resident dongba in traditional dress, who can write a prayer in dongba script on the paper for tourists who make a purchase. There is also one studio intended to educate visitors more deeply about dongba paper. Tourists can watch paper being made by a resident dongba and even make a sheet for themselves as a souvenir. For those just passing through there are a variety of products made from dongba paper: journals, binders, Nepalese-style paper lanterns and fill-in-the-blank business cards. I discussed the dongba paper shops with one of the originators of the idea, Xiao Jun (肖軍),10 one of the original members of the Naxi Culture Development Company. He talked about how it was their goal to preserve the technology of Naxi paper making that led to the diversification of products made with it. ‘Now everybody can use
The ‘Culture’ of World Cultural Heritage 147
dongba paper, not just dongbas writing religious texts’, Xiao Jun explained; ‘[t]his has been a successful revitalization of dongba paper’.
Lijiang Naxi and the Western Discourse In my interviews, when I asked Naxi people how they felt about the UNESCO designation, the most common answer I received was that upon hearing Lijiang had been inscribed as a World Cultural Heritage site, Naxi people were all very happy (hen kaixing) and proud (zihao). However, when I went a step further to ask about their impression of what inscription meant, I was met with confusion. ‘Why does UNESCO have this “World Heritage” programme? What is their purpose in saying places like Lijiang are a “World Cultural Heritage?” ’ is how I usually phrased this question. Sometimes I received the simple answer, ‘I don’t know!’, but most often the link between designation and tourism was cited: ‘UNESCO has this programme so that they can help places like Lijiang develop’ was the general understanding I often met with. ‘They want to help people like us!’ one informant, Li (李老師), a retired mathematics teacher from Shuhe,11 explained. ‘They tell the world that we are here so they can come and see for themselves, and that way we can develop!’ Regarding the concepts of authenticity and integrity, I found these hard to explain, let alone ask questions about. One of Lijiang’s responsibilities under the Operational Guidelines is to educate the public about cultural heritage and its importance to all of humanity. However, thus far the Old Town World Cultural Heritage Management Bureau has not done much of a job of indoctrinating local Naxi into these concepts. Perhaps the reason they accept that any attempt to do so would ultimately fail is because it would sound as if what UNESCO wants is for the Lijiang Naxi to remain poor, to continue to live in their Old Town houses without being able to fix them up or move to modern housing in the New Town – yet another irony of UNESCO’s attempt. As I was told in private by the head of the Management Bureau, ‘No one would be willing to live the way UNESCO wants them to, not even people from the countryside’. I found that to local Naxi, none of UNESCO’s or the state’s efforts have anything to do with keeping Lijiang the same, except perhaps the limitations on building styles. For the most part, both UNESCO and the state have been perceived by the local Naxi as actors accelerating change. This means that, in attempting to enforce the ‘bubble concept’ of culture, Lijiang is looking less and less like a ‘bonded and unique’ community.
Acknowledgements The author gratefully acknowledges the financial support of the Institute of Anthropology, National Tsing Hua University, and Institute of Ethnology, Academia Sinica which made this research possible.
148 Eveline Bingaman
Notes 1. Articles 15 and 16 of the Convention. 2. Since the focus of this particular research concerns how UNESCO regards culture, I will focus specifically on those aspects of the convention that deal with cultural heritage from this point on. In the convention itself, Cultural Heritage is defined in Article 1; Article 2 is specific to Natural Heritage. 3. From the UNESCO World Heritage website: ‘Benefits of ratification’, http://whc. unesco.org/en/164/ (accessed 14 March 2009). 4. Tourist arrivals have continued to grow. According to statistics I was given by Li Xianglin of Lijiang’s Tourism Bureau, in 2007 Lijiang had a total of 5.3 million tourist arrivals. 5. Document WHC-7/31.COM/7B. Add 5 is not a public document, so the actual specifics of the Advisory Board’s presentation to the WHC after their review are unknown except for information given to me by the Lijiang Old Town Management Bureau (Lijiang gucheng guanli ju, 丽江古城管理局) regarding their own reports to UNESCO in response to the review. 6. To date, no World Heritage Site has ever received a ‘yellow card’; instead, if preservation taking place at the site is not satisfactory, the site may be moved to the ‘List of World Heritage in Danger’ (UNESCO 2005: 177). The term ‘to be given a yellow card’ (liangle huangpai, 亮了黄牌), drawing on soccer terminology, seems to have been invented by the Chinese media. 7. Yang Jie Hong, 楊傑宏, Professor of Anthropology at the Lijiang Education College and one of the researchers involved in the Lijiang City research project; interviewed by the author, 22 October 2008. 8. From the Lijiang municipal government website: http://www.lijiang.gov.cn/pubnews/ (accessed 23 March 2009). 9. Interviewed by the author, 4 August 2007. 10. Interviewed by the author, 28 October 2008. 11. Interviewed by the author, 3 October 2008.
References Alesevich, Christopher. 2008. ‘Another Day in Paradise’, in US-China Today. University of Southern California US-China Institute. 7 March. Appadurai, Arjun. 1996. Modernity at Large: Cultural Dimensions of Globalization. Minneapolis and London: University of Minnesota Press. Ashworth, G.J., and Bart J.M. van der Aa. 2006. ‘Strategy and Policy for the World Heritage Convention: Goals, Practices and Future Solutions’. In Managing World Heritage Sites. Anna Leask and Alan Fayall, eds. Oxford: Elsevier, pp. 145–58. Cohen, E. 1988. ‘Authenticity and Commoditization in Tourism’. Annals of Tourism Research 15: 371–86. Firat, A. Fuat. 1995. ‘Consumer Culture or Culture Consumed?’ In Marketing in a Multicultural World. Janeen Arnold Costa and Gary J. Bamossy, eds. London: Sage Publications, pp. 105–25. Hannerz, Ulf. 2004. Transnational Connections. Chicago: University of Chicago Press. Kugel, Seth. 2006. ‘Preservation: sure it’s a good thing but …’. New York Times, Travel Section, 15 January: http://travel.nytimes.com/2006/01/15/travel/15journeys.html (accessed 11 February 2009). Linnekin, Jocelyn, and Lin Poyer. 1990. Cultural Identity and Ethnicity in the Pacific. Honolulu: University of Hawaii Press.
The ‘Culture’ of World Cultural Heritage 149
Li Xianglin. 2008. 着力优化客源机构--深度营销丽江旅游. Internal Report for the Lijiang City Tourism Bureau. 20 July. Nyíri, Pál. 2006. Scenic Spots: Chinese Tourism, the State and Cultural Authority. Seattle: University of Washington Press. Peters, Heather. 2001. ‘Making Tourism Work for Heritage Preservation: Lijiang – a Case Study’. In Tourism, Anthropology and China. Tan Chee-Beng, Sidney C.H. Cheung and Yang Hui, eds. Bangkok: White Lotus Press, pp. 313–32. Picard, Michel, and Robert E. Wood, eds. 1997. Tourism, Ethnicity and the State in Asian Pacific Societies. Honolulu: University of Hawaii Press. Spensley, Alys. 2003. ‘Under the Snow Mountain: Development in Lijiang and its Effects on Naxi Culture’. In Tourism and Development in Yunnan: Yunnan through Foreign Students’ Eyes, Vol. 1. Sam Mitchell, ed. Kunming: Yunnan Fine Arts Publishing House, pp. 11–31. Steedly, Mary Margaret. 1999. ‘The State of Culture Theory in the Anthropology of Southeast Asia’. Annual Review of Anthropology 28: 431–54. UNESCO. 1972. Convention Concerning the Protection of the World Cultural and Natural Heritage. Adopted by the General Conference at its seventeenth session. Paris, 16 November, English Text. ———. 1997. Advisory Body Evaluation No. 811. World Heritage List, Lijiang (China). ———. 2005. Operational Guidelines for the Implementation of the World Heritage Convention. United Nations Educational, Scientific and Cultural Organization Document WHC.05/2. Paris: UNESCO. ———. 2007. Decisions Adopted at the 31st Session of the World Heritage Committee; Christchurch, New Zealand. United Nations Educational, Scientific and Cultural Organization Document WHC-7/31/Com/24. Paris: UNESCO. ———. 2008a. Convention Concerning the Protection of the World Cultural and Natural Heritage. Presented at the 32nd Session of the World Heritage Committee; Quebec City, Canada. United Nations Educational, Scientific and Cultural Organization Document WHC-08/32.Com/7B. Paris: UNESCO. ———. 2008b. Decisions Adopted at the 32nd Session of the World Heritage Committee; Quebec City, Canada. United Nations Educational, Scientific and Cultural Organization Document WHC-08/32.Com/24. Paris: UNESCO. UNESCO WHC-ICOMOS. 2008. Mission Report: Old Town of Lijiang (China) (811) January 10–19, 2008. Presented at the 32nd Session of the World Heritage Committee; Quebec City, Canada. United Nations Educational, Scientific and Cultural Organization Document 32Com. Paris: UNESCO. Wood, Robert E. 1997. ‘Tourism and the State: Ethnic Options and Constructions of Otherness’. In Tourism, Ethnicity, and the State in Asian Pacific Societies. Michael Picard and Robert E. Wood, eds. Honolulu: University of Hawai’i Press, pp. 1–34. Yang, Mayfair. 1996. ‘Tradition, Travelling Anthropology, and the Discourse of Modernity in China’. In The Future of Anthropological Knowledge. Henrietta L. Moore, ed. London: Routledge, pp. 93–114.
Chapter 7 ‘Amis Hip-Hop’: The Bodily Expressions of Contemporary Young Amis in Taiwan Futuru C.L. Tsai
Introduction In this chapter, I employ hip-hop as a metaphor and model in order to interpret the bodily expressions of Amis young men in A’tolan village, eastern Taiwan.1 I argue first that the way mainstream Taiwanese people and some scholars misinterpret contemporary Amis dances as representing a loss of ‘cultural traditions’ is based on an imaginary of ‘nature’, of the ‘pure’, ‘simple’ and ‘marginal’ bodies of indigenous people. Secondly, however, the bodily expressions of Amis young men represent a complicated social and cultural context, within which they create an alternative style and form of ‘hip-hop’. Public stereotypes regarding indigenous peoples in Taiwan have mostly been propagated by the mass media. In this view, indigenous people live on the ‘periphery’, far away from Taiwanese Han society, and therefore, in contrast to the civilization of the Han, indigenes are ‘natural’ (Wang 1998: 232). Indigenous dances in particular represent the imposition of an inflexible view of the nature of the indigenes. However, the ‘natural’ images of indigenes can be further specified as two distinct views, which both contradict and correlate with one another. The public view, which is connected with tourism, sees indigenous dances as joyful (see Lin 1997; Huang Gueichao 1994; Li 2001a, 2001b). The second view, in keeping with the emerging worries of the ‘intelligentsia’, is that commercializing the ‘natural’ dances of indigenes translates into their losing their ‘cultural traditions’. Therefore, studies of indigenous dances have focused on dances in rituals and ceremonies (e.g. Li 1994, 2000, 2001a, 2001b; Ming 1997; Zhao 1997; Liu 2000; Huang Shiun-wey 1987). Nonetheless, as Maurice Bloch (1989, 1992) has suggested, the social cognition of rituals and non-rituals exist simultaneously in a society. If we only study a society through the unchangeable cognition of rituals, we will oversimplify social cognition in such
152 Futuru C.L. Tsai
a way that we can only understand the rituals, not the cognition of non-ritual events. Simply distinguishing ritual from non-ritual is unrealistic. Thus, a discussion of the bodily expressions of contemporary Amis young people in terms of their everyday lives is necessary. This chapter analyses the sociocultural meanings of three kinds of interrelated bodily expression performed by young Amis people in A’tolan village in Taiwan, namely ‘play’, ‘modern dance’ and ‘traditional dance’, as these bodily expressions form an ‘alternative’ Amis hip-hop. Before proceeding further, it is necessary to point out that Amis hip-hop is not directly related to the genre of ‘hip-hop’ as normally understood (involving, for example, rap-style recitation, hip-hop style dance movement, scratch-andmix DJ performances or graffiti art), but is both a metaphor and a model for the analysis. That is to say, the bodily expressions of contemporary Amis youth in Taiwan are subtly related to hip-hop, which originated in African-American youth culture: the key is ‘difference’. Young Amis men represent their subjectivity differently from what the public imagination dictates as losing traditions. The ‘difference’ consists in a pan-indigenous identity that they create themselves as they design their dance steps and incidental music. In other words, dance is a socialization process for young Amis by means of collective bodily expressions. Moreover, understanding the gender relations of Amis is key to understanding the young bodily expressions of A’tolan, in that women play a critical role in the dances of young male dancers, plays in which the young A’tolan Amis merge hip-hop and ‘traditional’ Amis dance steps within an Amis sociocultural context. In doing this, they have created a unique style and form of ‘Amis hip-hop’ that is inconsistent with both the public and the scholarly imagination – and even with older Amis’ styles of dancing.
Youth, Body and Hip-Hop Youth cultures have not received much attention in anthropology since the 1950s. Only recently have they been taken up again, due in part to the dynamic, unbounded and postmodern concept of culture in contemporary theory. The most important concept in this respect is ‘agency’, which has led scholars to comprehend the agentive creativity of youth cultures, among other topics. Anthropological studies of youth cultures during the first half of the twentieth century focused mostly on the socialization process, but tended to ignore the actual agency of young people themselves. Compared with anthropology, sociology, although paying more attention to youth cultures, saw them as subcultures within a framework of class conflict based on the concept of rebellious youth. However, approaches to the study of youth cultures have recently changed to focus on the cultures that are created and practised by young people themselves (Bucholtz 2002). Most anthropologists study culture from the perspective of adults or older informants over a long period. First, this implies that only adults or elders can represent ‘their’ cultures; secondly, it also
The Bodily Expressions of Contemporary Young Amis in Taiwan 153
suggests that the young people represent a generation that is foregoing ‘traditional cultures’ by living in the ‘modern’ world. Following these assumptions, it is easy to see how the bodily expressions of contemporary Amis youth might easily be regarded by the public and some of the intelligentsia, who see culture as a pre-set concept based on ‘traditions’ or ‘rituals’, as symptoms of cultural loss. However, the point of bodily expressions is not the body per se but the meanings of the body (‘as signifier is to signified’) (Hayles 1993: 181). If we ignore the meanings of the body, we will be handing over power and control of self to others (Stone 1991). Therefore, if the public and some of the intelligentsia regard the bodily expressions of contemporary Amis youth as a loss of culture, based on imagination but not on meaning, this would represent the power of control by one group over another. Hip-hop originated as a way for young people to express their subjectivity and difference in response or opposition to the power exercised by others over their space and movement. Hip-hop culture, which stemmed from the musical and dance styles of African-Americans, has spread around the world with global flows as a unique cultural product and become one of the popular cultures of contemporary youth around the world. Hip-hop culture is interwoven with the sense of being ‘cool’ and of ‘difference’ (Osumare 2001), and with a will to fight (George [2000] 2002: 9), thus providing young people with a common identity capable of transcending different races and ethnicities. Representing ‘difference’ signifies an expression of subjectivity, which has been crossing the boundaries of states around the world. Young people in many countries have also been developing their own ways of expressing their subjectivity by interweaving global hip-hop cultures. For example, Jamaicans construct the nationalism of anti-colonialism and a global identity through dancing as a process of seeking a national cultural identity along with the representation of nationalism in art forms (Thomas 2002). Similarly, Mexican youth living in the neighbourhoods of New York and New Jersey develop dances (sonideo bailes) that are ‘something new, exciting, and completely modern’, all in places where young people socialize, to express a shared identity as Mexican Americans (Ragland 2003: 338). However, technological and economic processes simultaneously bring about consequences in the form of capitalism – that is, American hip-hop culture exports its dancing and musical styles by parasitizing on the American hegemony of capitalism. On the one hand, American hip-hop leads the fashions and identities of youth, without adopting the rules of the general public or of adults; on the other hand, hip-hop is being led by the nose by the hegemony of capitalism, which has exported it to the rest of the world and thus constructed hip-hop as another kind of hegemony (Osumare 2001). Thus, if young people in different places want to create real subjectivities through ‘hip-hop’, not only to create their own styles but also to avoid the hegemony of American hip-hop, they need ‘real’ autonomies of the body. If bodies can be regarded as media, suggests Eric Michaels (1994), they can provide an example of ‘autonomy’. The key to the autonomy of Aboriginal media in Australia is that the Aborigines can
154 Futuru C.L. Tsai
manage their media in relation to their own cultural contexts. Michaels describes how the Warlpiri manage their own media through their kinship system. It is only from the viewpoint of this kinship system that we can understand how indigenous peoples produce and receive media. Similarly, realizing that the body of youth is indeed a medium and that the autonomy of any media comes from the cultural context of the society concerned enables us to comprehend how young people express and experience their bodies. In short, the subjectivity of young people may be based on the autonomy of their bodies within their specific cultural contexts. This is the origin of AfricanAmerican hip-hop. To understand this process among young Amis, it will be necessary to enter their sociocultural situation to understand and interpret their subjectivity and show how they represent it through bodily expressions.
The Sociocultural Context of Bodily Expressions of Amis Youth Definitions of youth differ across different societies and cultures (Bucholtz 2002). The category of ‘youth’ in A’tolan Amis society is indeed different from that of Euro-Americans. Historians have argued that the category of ‘youth’ is a relatively recent construction: up to the nineteenth century in Europe children were generally treated like small adults, and ‘youth’ and ‘teenager’ were not categories that were marked off as special in-between stages of life until after the Second World War (e.g. Mead 1939). For the Amis, age-grades have been an important part of the social structure for a long time, so the category of ‘youth’ (kapah, which also means ‘beautiful’ and ‘shining’) has a different history and different associations. Therefore, it is necessary to identify who the youth are in the sociocultural context of A’tolan Amis. The Amis are the largest population of Austronesian-speakers in Taiwan, with a population of over 140,000 at the end of 2003.2 Most of them reside in eastern Taiwan. Typically, the focus of the society and culture of the Amis is on the age organization, together with its functional aspects of military organization, training, politics and the kinship structure (Chen 1985). Although the societies and cultures of the Amis have a great deal of diversity in different areas due to different historical and geographical circumstances, the age organization and the kinship system – the latter being matrilocal and matrilineal (Wei, Yu, and Lin 1972) – are still the critical distinguishing features of each Amis village. Whether this is appropriate from the point of view of descent and lineages has been debated (Chen 1986: 69). Since women are regarded as the core of the kinship system and men as the core of the age organization, the ‘family (females)’ and the ‘age organization (males)’ are seen in terms of the distinction of sex (Lo 2005), or as both complementary and oppositional to one another (Huang 1989). A’tolan is a large village located on the south-east coast of Taiwan facing the Pacific Ocean.3 According to local stories and the names of the historical age
The Bodily Expressions of Contemporary Young Amis in Taiwan 155
sets, the Amis have been living in A’tolan for over two hundred and fifty years (Huang 1991; Huang and Lo 2001). Basically, A’tolan still maintains the age organization. When a boy reaches the age of about twelve, he can participate in the age organization as one of the Pakalongay (teenagers who serve others).4 Following the coming-of-age ritual five years later, he will go up one grade together with his kapot (the members of the same age set).5 Philip Gulliver (1968) identified two kinds of age system, one an age-grade system signifying individuals’ social roles according to their ages, the other an age-set system implying that those who are of the same age form a group like a corporation with specific duties, in which they remain for the rest of their lives. The Amis age system is generally regarded as one of age sets rather than of age grades (Huang 2005: 52–53). However, in A’tolan, the age organization works as both. The naming system of the age sets is the creation system, whereby every five years a new age set will be established with a collective name given to it by the village elders. For example, Laciensi is the collective name for the age set of those who passed the rite of passage and became Kapah (age grade of young men) in 2000 – in fact, the name Laciensi was given for ‘the millennium’. Besides the unique name for each age set, several age sets are categorized in the same age grade: from the youngest to the oldest, the grades are Pakalongay, Kapah, Matatapalay and Tu’as. In the age organization, each kapot should obey the orders of the age set above it: the upper age sets are responsible for training and looking after the lower age sets. Table 7.1 gives the overall structure of the A’tolan age organization.6 Table 7.1 The structure of the age organization of A’tolan Amis (as of 2003) Categories General Name by duties grade names Malitengay Tu’as (People (Old men) who are near the ancestors)
The unique Estimated name of age in each age set years
Ladihaf Lahonti (Men who take rest Lasinpay under the shadows of stone) Lakining
Above 78
Lahitay Lamintay Lasfi (Men who teach the men still inside the men’s house)
Lasingping Lakinmong
67~77
156 Futuru C.L. Tsai
Table 7.1 continued Malitengay Matatapalay Tukal (People (Middle(The pillars) who are aged men) Tapal near the (Accompany the ancestors) Tukal)
Malikoday (Dancers)
Lakinma
62~66
Latingko
57~61
Culal (Seeds germinating)
Lakocong
52~56
Lomlom (Making fire by rubbing wood)
Lacingko
47~51
Mikumoday (Gearing the village)
Lakensec
42~46
Mihiningay (Watchers)
Mihiningay Laencw (Watchers [those who watch and learn])
Kapah (Young men)
Sakakaay no Kapah Lakangcing 32~36 (Older brothers in the grade of youth)
Pakalongay (Teenagers)
37~41
Saka tosa no Kapah (Third level of the grade of youth)
Lakayakay
27~31
Saka toro no Kapah (Second level of the grade of youth)
Lakenca
22~26
Safafaay no Kapah (Younger brothers in the grade of youth)
Laciensi
17~21
Pakalongay (Teenagers who must serve others)
Pakalongay
12~16
It can be inferred that in A’tolan the age grade of Kapah represents youth, with four age sets, Lakangcing, Lakayakay, Lakenca and Laciensi. The interaction between these age sets is based on the principles of age and generation (Chen 1990). Political affairs in A’tolan are basically controlled by the age grades of Matatapalay and above. The kakita’an (chief ) is selected from Lasfi or Tukal. The kakita’an chooses some other older men from Lasfi and
The Bodily Expressions of Contemporary Young Amis in Taiwan 157
Matatapalay to act as his advisors or counsellors, who are responsible for decision making. The Mikumoday is in charge of executing the orders of the chief and his team, while the Kapah is managed by the Mikumoday to carry out the orders. The political system of the Amis in A’tolan is thus generally governed by the elders. Within it, the Kapah must comply within a hierarchy that provides one of the similar forms of power relationships that develop bodily expressions of hip-hop culture in the United States. The bodily expressions of young A’tolan Amis reflect age and generational relationships, especially in the malikoda (a style regarded as the ‘traditional’ Amis dance, in which the dancers take each other by the hand).
Figure 7.1. The malikoda dance of the A’tolan Amis (photo by the author) The malikoda is esteemed as relating to the strict ritual hierarchy (Li 1994, 2001a, 2001b; Ming 1997), physical training (Zhao 1997; Lin 1997) and the function of uniting people in the village (Ming 1997). The malikoda requires that people line up in accordance with the hierarchy. As for the Kapah, they are asked to perform the malikoda in unanimity and with majesty and great power. Many scholars have regarded this kind of traditional dancing as one of the best representations of Amis culture. However, it is not the only dance in Amis culture, especially for these Amis young men in A’tolan. If we value the malikoda as the ‘purest’ dance in Amis culture but ignore the others as merely symbols of a loss of culture and of secularization, we might easily miss the whole system of social cognition (Bloch 1989). Amis young men have developed many forms of bodily expression, which blend local character and other features, but also include so-called modern dances and plays.
158 Futuru C.L. Tsai
The modern dances and plays are the forms of bodily expression that others consider to be indications of a loss of culture. However, they are the forms that reveal the difference between Amis views and those expressed by the public and by some scholars, as well as the forms of bodily expression of the elders in A’tolan. In fact, the modern dances, the traditional dances and the plays are interrelated bodily expressions, but they take on different forms on different occasions and with different spectators. The stage performance is the most open of fields and is separated into two parts by Amis youth at A’tolan: one takes the form of traditional dances, the other of modern dances. As for the plays about Amis youth, they are usually not ‘performed’ on a stage. This raises the question of to whom the bodily expressions of Amis youth are being exhibited. In another words, who are the spectators? Below, I focus on the plays and the modern and traditional dances of Amis youth in A’tolan and consider the complex relationship between these bodily expressions and the spectators.
Plays The various plays of Amis youth may happen anywhere and at any time that Amis youth congregate. The forms of the plays vary, from the simplest, recreating songs, to various complex body performances. For example, Amis youth have re-created the lyrics of the Christmas song Jingle Bells as follows, and Amis youth sing it to seemingly ‘nonsensical’ body movements: My age is young; my body shape is slender; Mama asked me to kill the chicken, so I’ll pull out the chicken’s hairs [i.e. pubic hairs]; pull out the hairs; pull out the hairs; I’ll pull out the hairs; pull out the hairs; pull out the hairs; I’ll pull out the hairs.7
One night, two members of the kapot Laciensi visited the upper kapot Lakayakay and asked for their advice during the annual harvest festival, a custom of A’tolan Amis.8 After Lakayakay had given a lesson to Laciensi, and Laciensi had served rice wine to Lakayakay, the members of the latter asked their lower kapot, Laciensi, to perform something to please their kaka (older brothers and sisters) before they left. These two young men of Laciensi then immediately performed the re-created Jingle Bells in front of their kaka, singing the song with the body movements following the re-created lyrics as above. When they sang the words ‘my body shape is slender’, they drew a calabash shape with their hands and twisted their hips to symbolize the shape of the female body. When they sang ‘pull out the [pubic] hairs’, they approached the vice-leader of Lakayakay and unexpectedly tried to pull out his pubes, with him trying hard to prevent this without showing any anger. All the members of Lakayakay sitting alongside were extremely pleased by this scene. Everyone knew that this was merely a play. The play obscured the hierarchical relationships between the upper and lower kapot, although it was initiated within those relationships and because of them.
The Bodily Expressions of Contemporary Young Amis in Taiwan 159
Another kind of youth play that implies competition among kapots is basically a team play. For example, another night during the annual harvest festival, all the Kapah observed the custom of gathering in front of the sfi (men’s house) to practise the malikoda as usual. Breaks from practice were devoted to plays. On one occasion, the oldest of the kapots, Lakangcing, asked the lower kapots to act out some entertaining performances for everybody. Each kapot performed at the centre of the square, one by one. The first kapot was Laciensi, who immediately assembled in the centre of the square, one of them standing in the front of Laciensi acting as the ‘conductor’. Then they sang the very famous children’s ballad, Fly Lady Fly, to interpretive body movements: Butterfly, butterfly, beautiful butterfly; wearing golden ornaments on your head; wearing garish clothes; you love flowers; and flowers love you too; you can dance; flowers have the sweet honey.
Figure 7.2. The Fly Lady Fly Play by Laciensi (2003); the person on the right is the conductor (photo by the author) While they were singing this song, all of the members of Laciensi followed the conductor’s body movements. The whole play was similar to one of the famous ‘team plays’ (Tuan kang) of the ‘China Youth Corps’ (CYC or Jiuguotuan),9 with its imitative movements (Daidongchang). However, unlike the imitative movements of the CYC style, Laciensi did not ask everybody present (the other kapots in attendance) to follow their movements, but only those belonging to Laciensi. Everyone was laughing heartily at that moment. Their play has subtle differences from the imitative movements of the CYC style, in which the conductor would ask everybody present to follow the conductor’s dance together. If a follower were to laugh out loud instead of participating in
160 Futuru C.L. Tsai
the imitative movements, he would be regarded as ‘unsocial’ or ‘antisocial’. This is different from the atmosphere of Amis youth in an awkward predicament, even though the forms of the Amis play and of the imitative movements are similar. After the performance by Laciensi, Lakayakay also came into the centre of the square and performed another play whose form was similar to that of Laciensi, but with a different song and different body movements. Their body movements seemed even more ‘illogical’, including merely rolling on the ground and barely singing the song, but using nonsense words instead. The Lakayakay received more laughter than Laciensi. After Lakayakay’s performance, Lakangcing performed, and then Pakalongnay. There are two contextual distinctions between the CYC style of followalong moves and the Amis youth’s play. First, the most important distinction is that the Amis youth’s play is based on hierarchical relations. Secondly, however, all the members of each kapot know each other very well; in other words, the Amis youth’s play is performed in private, with body movements and voices to please everyone. Moreover, the Amis youth’s play reflects the subtle competition among the kapots. The more laughter a kapot receives, the more it claims the status of ‘the good kapot’. The Amis youth’s play respects the hierarchical relations in the age organization system, but it also permits the creation of unbounded body expression. The plays among the kapots could also be transferred to the intra-kapot plays, which is a play within the kapot. The plays of the A’tolan Amis youth are varied and innovative. The ‘spectators’ of a play’s bodily expressions are the ‘people on their own side’; in other words, they are basically performing for themselves. Everyone is not only a spectator, but also a performer. Furthermore, ‘people on their side’ still can be sorted into several levels, from the most private occasion (the intra-kapot) to the semi-public arena (plays among several kapots). The plays of A’tolan Amis youth represent an outward similarity to the public youth play, but there are remarkable differences. The differences stem from Amis youth re-creating their bodily expressions based on the incorporation of external bodily expressions into the Amis cultural context.
Modern Dances The modern dance competition is usually held on the first day of the annual harvest festival at A’tolan. Generally speaking, about 70 per cent of the spectators are Amis from A’tolan, the rest being visitors passing through and a few amateur anthropologists of indigenous culture. On these occasions, the public and private spheres of the A’tolan Amis overlap. These public or private spheres are distinguished on several levels depending on different subjects, from age sets all the way up to age grades, the age organization, the village and outsiders (visitors). The A’tolan Amis’s definition of ‘modern’ and ‘traditional’ dances is based on whether or not a specific dance must be coordinated with a specific song. A
The Bodily Expressions of Contemporary Young Amis in Taiwan 161
modern dance is one where the dance movements and the song are independent of one another; the traditional dance, on the contrary, is one where they must be combined. This means that a modern dance can utilize audio equipment such as compact disc or cassette players, and the dancers do not sing themselves; the traditional dance, by contrast, requires the dancers to sing the song themselves while they dance, without any accompaniment. The A’tolan Amis greatly value the modern form of dance. Indeed, the modern dance competition during the annual harvest festival is one of the most highly regarded competitions for each kapot. Several weeks before the festival each year, somebody in each kapot takes charge of selecting the music and arranging the dance steps, which the members of the kapot practise every night. Since most of the young people work outside A’tolan village, they can usually only start practising their modern dance a few days, or even just one night, before the festival. On the last night before the festival, each kapot practises its modern and traditional dances until midnight, and some of them even practise all through the night. Since 2001, the modern dance competition at the A’tolan annual harvest festival has been separated into two groups: youth and middle age. Those organizing the competition made this decision because Laciensi won first place in modern dance for the two years following their incorporation as a kapot in 2000. Their modern dances were in different styles from the older kapots and attracted huge applause. Some older kapots therefore proposed that it would be fairer if the modern dance competition were separated into two groups according to age grades, otherwise middle-age kapots would always be the losers. The kakita’an and his advisory team accepted the proposal. Some older kapots of Kapah such as Lakangcing, Lakayakay and Lakenca, gradually stimulated by Laciensi, began to work hard selecting the music and designing an arrangement of modern dance steps that could compete against their safa (younger brothers and sisters) Laciensi. Especially since 2002, when Lakenca and Lakayakay were combined into a temporary kapot because of a shortage of members, they have been very aggressive in arranging and practising their modern dances. On the last night before each annual harvest festival, most of the members return to A’tolan from elsewhere in Taiwan and gather together to practise. It would be completely wrong to assume that these older ‘young men’, who are not professional dancers, feel tormented by having to practise and memorize the steps and tempos of the music, while simultaneously sweating in the hot and windless summer nights, with all their mothers and aunts sitting around them, laughing and poking fun at them. They do not regard it as toilsome or embarrassing but actually practise with ever-increasing vigour. They are entirely focused on the competition that is to take place the following day. The modern dances of Lakenca, Lakayakay and Lakangcing are becoming better and more varied under the pressure of fame; they do not want to be the losers forever, and would feel especially embarrassed to be beaten by their safa (younger brothers). For example, in the modern dance of Lakayakay in 2002,
162 Futuru C.L. Tsai
their music consisted of two parts: the first of African music with a strong beat, the second of an Amis folk-song. The steps of their modern dances were choreographed within these two parts. In the first part, they held two bamboo sticks in their hands and followed the African drumbeats, twisting their waists, holding the sticks, and so on. Suddenly the music switched to the Amis folksong with light beats and a brisk rhythm. Their movements also became spry and vivid. Furthermore, they danced towards the kakita’an and the advisory team, announcing loudly one by one that they were the Amis braves of A’tolan. In the next step they blended a unique dance, the kulaku or ‘bravery dance’, using umbrellas as symbols of spears and shields, and danced around the square. When they passed the tents of each kapot, everyone was going crazy for their dance, and even some of the dancers’ mothers began dancing with them. Lakayakay took second place that year.10 In 2003, Lakayakay came back again after their defeat. The modern dance of Lakayakay was still being arranged by a member nicknamed ‘Iron Coach’.11 That year, the modern dance of Lakayakay was also separated into two parts: the first part was collocated with the Amis folk-song Zacateca, which was recreated in the genre of New Age music by Enigma as Return to Innocence; the second part also followed African-style music with a strong beat. The longer bamboo sticks were used in the dance, tied with turtledoves’ feathers to signify spears. In the first part of the modern dance, Lakayakay came on stage with the steps transformed from kulaku, guarding a little boy who was the nephew of one of the kapot. This little boy was carrying a ‘wild pig’ on his shoulders – actually a pink toy pig, but representing a wild pig in the dance (see Figure 7.3). Once the pig had been set down at the centre of the square, the music suddenly turned into an African style with intense beats. Their body movements also became extremely exaggerated, Lakayakay dancing around the pig like hunters. The dance integrated many other elements this time, including the ‘traditional’ Amis dances, the steps of a popular group from Korea called the Cool Dragons, the stretching of tongues in the Maori haka,12 and twisting the waist and hips like an African butterfly. Most of these dance steps were gleaned by Iron Coach from watching the Discovery television channel. The Lakayakay dance certainly won enormous applause again that year, although it took only third place. According to a village elder who was one of the judges, the reason that Lakayakay did not win first place as the people of the village had expected was as follows: ‘Lakayakay performed very well; however, your members’ steps were untidy. If you had enough time to have more practice, you would win first place.’ It is obvious that Lakayakay reached a level of achievement in the modern dance competition that could be attributed to both the pressure of Laciensi and the arrangements of Iron Coach, as well as the involvement of the whole kapot. Before 2002, Lakayakay always gave up in the modern dance category, and then were usually given a dressing down by the upper kapots or the elders, totally without effect. However, things have changed since Laciensi underwent a modern dance revolution.
The Bodily Expressions of Contemporary Young Amis in Taiwan 163
x
Figure 7.3. Lakayakay’s modern dance in 2003 (photo by the author) Laciensi have won first place in the modern dance competition for several years since they officially became a Kapah in 2000. They ranked number one from 2000 to 2003, and again in 2005. Only in 2004 did Lakayakay finally take first place, but Laciensi took an oath to win back first place in 2005, and they did. Since Laciensi was named, most of its modern dances have been arranged by a woman whose nickname is ‘Strawberry’. She was the classmate of some members of Laciensi when they were Pakalongay from 1995 to 2000, and Strawberry has been with Laciensi ever since. Laciensi could always defeat their kaka (elder brothers) in the modern dance competition because of Strawberry’s critical role; not only did she major in dancing arts in college, but she is a sort of ‘under the table’ leader (Tsai 2005). In addition, the members of Laciensi respect the modern dance competition very much, and they always try to overcome any difficulty by practising their modern dance several days before the annual harvest festival. When the Laciensi passed their initiation ritual to become Kapah, they also performed two modern dances, the first one a transformation of the Maori haka,13 a popular dance among baseball players in Taiwan, the second a blockbusting modern dance for the competition. The latter was arranged by Strawberry, the incidental music being a brisk song transformed from a Puyuma folk-song by a female singer called Samingad ( Ji Xiaojun, 紀曉君), herself a Puyuma.14 Their modern dance, with this song, integrated several movements, including dances in Slavonic folk style, the female dance moves of Amis elders, some hip-hop styles, and some other original steps devised by Strawberry. The next year, Strawberry chose another song, by Biung, a Bunun singer.15 The Laciensi still incorporated many elements of Amis traditional dance steps, including the malikoda and original steps, into their modern dance.
164 Futuru C.L. Tsai
In 2002, Laciensi’s modern dance commenced with a song about a baseball game sung by a Han singer, Zhu Toupi (Pig Head Skin). For the dance, Strawberry designed body movements inspired by the Flying Fish Dance of the Tao (Yami) people of Orchid Island,16 and referring to a wrestling episode from Laciensi’s initiation ritual in 2000. For their modern dance in 2003, Laciensi also selected Zhu Toupi’s music but re-created it based on an Amis folk-song. This time, the dancing style was as cute as the movements of children. From 2000 to 2003, Laciensi always won first place in the modern dance competition at A’tolan. Although they were defeated by Lakayakay in 2004, they came back in 2005 with another modern dance that integrated some hip-hop styles and a lot of modifications from original Amis dance steps in A’tolan. Their dance was similar to that of Lakayakay, who also integrated many different elements into their modern dance. Even though both kapots transformed the traditional dances of the malikoda, or the elders and women, into their modern dances, they both performed them at an ever-increasing tempo. As for Lakangcing, however, they are heading towards middle age, and so they do not feel the same urgency to compete with their safa in the modern dance. Instead, they face the pressure of losing the traditional dance competition. But surprisingly, Lankangcing selected English-language dance music for the modern competition in 2003. Their modern dance was mostly a reprise of the cheer-leaders’ moves, but integrated with a tug-of-war session in the opening. The biggest success in their modern dance performance was this tug of war because the performers consisted of several ‘families’. Comparatively speaking, the cheer-leaders’ dance style did not obtain as many cheers as the tug of war. The aspect that was most different from the performances of Laciensi and Lakayakay was Lakangcing’s inclusion of their wives and children in the dance. Most of the members of Lakangcing are married, and their wives started to appear either in the modern or traditional dance competition. The dancers of Lakayakay and Laciensi were still mostly male. However, the dances of the other groups represented the complex image of the female in A’tolan Amis culture, discussed at length below. As mentioned above, the public or private spheres of the modern dance competition are distinguished on several levels, depending on different subjects: each level might be in a public or private sphere relative to the other levels. To the village, the age organization and the age grades, the age set is a unity: the modern dance competition of each kapot is within the private sphere of the village, the age organization and the age grades. In other words, the kapots do care about their performance in front of the Amis people of A’tolan and are especially concerned about the opinions of the on-stage judges, as well as the other kapots in their own age grade. However, the outsiders see only the whole village as a unity, the age sets being meaningless to them. It is at this moment that the age sets become an obstacle. Actually, the kapots do not care what outsiders think of their performances. One of the leaders of the A’tolan kapot Lakangcing remarked:
The Bodily Expressions of Contemporary Young Amis in Taiwan 165
After visiting A’tolan’s Harvest Festival, outsiders may wonder, ‘How come the A’tolan tribe is so modern?’ Or they think we A’tolan Amis have been losing our traditions. But the fact is that … this is exactly the unique tradition of the A’tolan tribe. So, the chief says that … the youth must do their best to demonstrate … to try to please the elders. That means how to present your body movements. And so, we are living in this moment, in this era. What you can show to the elders … is what is most familiar to you. We A’tolan Amis have a special character. … You have to show yourself off in front of the elders. Through the form of the modern dance, you show off your best to everybody. We also combine the traditional and modern … when we present our dances. I think this is the wisdom of our elders in the A’tolan tribe. It’s the wisdom. … I don’t agree with the viewpoint that … ‘the A’tolan tribe has been losing their traditions’. As a matter of fact, the idea is from our heritage … that during the annual harvest festival … the young people have to please the elders. And, that is what continues until today. There’s nothing wrong with that.
In one sense, Amis youth in A’tolan are maintaining their independence from the imaginations of outsiders; in another sense, they are creating some ‘new’ and ‘different’ bodily expressions by integrating various elements and including the heritage of traditional dances, to display something distinctive from what the middle-aged and the elders produce. Even by doing so, the elders and the middle-aged do not feel that this is inappropriate. On the contrary, they feel happy to see something different, as the kakita’an of A’tolan noted: On the first day of the festival, all performances are Amis. The second day is also the same. There are a lot of guests visiting on the first day, but there’s less variation of our traditional dances. If this year’s festival is the same as last year’s, the elders would feel a lack of variation. Even so, much of the so-called ‘modern dance’ does not change much. Although traditions cannot be forgotten, we also must follow the modern age into the future. Now we see that our children are willing to participate with so much originality, when elders are proud to say, ‘Look, that’s my child!’ Then, we all feel very happy. The youth have been changing the boring part of the ceremony.
The kakita’an does care about outsiders’ views very much since he is the symbol of the village: he certainly cares about visitors’ feelings, but the feelings of the elders are still most important. What they feel is the key to the youth’s dances. The outsiders’ view of the youth’s modern dances as a loss of tradition or as their commercialization for tourism makes no sense at all to the youth themselves, nor does it to the elders of A’tolan. In fact, from the youth’s perspective, since outsiders have never been spectators, how can the suggestion that they are merely trying to please the visitors be taken seriously? The criticisms and interpretations of the visitors and certain ‘scholars’ have never counted in the context of the village. To the A’tolan Amis, especially the young, the content of outsiders’ imaginations are not important. What is important is how many cheers they receive from the Amis in A’tolan, especially the elders.
166 Futuru C.L. Tsai
Traditional Dances As mentioned above, the distinction between modern dance and traditional dance in the A’tolan Amis context is whether the singing and body movements (dance) are separated or coordinated with one another. The traditional dances can be differentiated into two kinds. The first is the malikoda, which links the hands. There are fourteen songs in A’tolan of this sort with different dance steps. These kinds of dances are conventionally regarded as the ‘traditional’ Amis dances, as they coordinate singing with specific dance steps (Li 1994, 2001a; Liu 2000; Huang 1994). The other kind of traditional dance might be categorized as ‘non-malikoda’, and can be further subdivided into two categories. One is Militepulay’s dance, which is performed by the middle-aged wives of the Mihinigay to Tukal age grades. The other traditional dances are basically derived from Militepulay with added body movements from the daily life experiences of former times, such as digging the soil, sowing, hoeing weeds, reaping, hunting, firing the bow, fishing, and so on. At this level, the traditional dances may have many variations. Both the malikoda and non-malikoda dances can be counted as traditional dances since they coordinate singing with specific dance steps and body movements, but they also have many variations. A scholar who is an Amis from another village has categorized the Amis traditional dances into three types: ‘entertaining the divinities’, ‘entertaining the people’ and ‘entertaining oneself ’ (Zhuang 2003: 45). According to these categories, while the malikoda and the Militepulay’s dance can be classified as ‘entertaining the divinities’, the other traditional dances are for ‘entertaining the people’ and ‘entertaining oneself’. However, the traditional dances in fact represent much more complicated variations than these three simple categories allow, especially as performed by the young people. As for the genre of malikoda, there are both transformed and non-transformed versions. In the nontransformed version, there are basic steps, the linking of the hands and a hierarchical line-up; in the transformed version, the young people always create some ‘fancy steps’, though they may be inspired by the originals by permission of the upper kapots and the elders. The malikoda is danced in a big ‘C’-shaped formation following the sequence of the age grades, the youngest kapot being the last. In other words, the young people are always in the second half of the malikoda. The malikoda are focused on the movements of the legs, that is, the lower half of the body. Each of the malikoda songs requires its own incidental steps. However, the dancers can create some fancy steps for most of the malikoda, which do not interrupt their basic structures. These fancy steps generally include ‘bending the upper half of the body intentionally’, ‘lifting the legs higher’, ‘magnifying to twist the hips’, ‘fragmenting the steps’ and so on, all such moves being done to strengthen the power or to increase the complexity of the movements. If the dancers improvise the steps too freely and the original structure of the malikoda is destroyed – for example, by going in the wrong direction, producing the wrong cadence, or if
The Bodily Expressions of Contemporary Young Amis in Taiwan 167
only one person makes a fancy step – then that person will be reprimanded by the upper kapots and the elders. Therefore, in so far as the two simultaneous conditions are fulfilled – ‘no disruption of the structure’ and ‘the collective moves’ – the dancers can still express their innovation and creativity. Although Amis youth do not dance the Militepulay, in the other traditional dances they transform the malikoda and Militepulay, as well as adding many body movements relating to the daily life of former times, such as cultivating, fishing and hunting. This level of traditional dance is permitted with greater variation than the malikoda and the dance of Militepulay can have. Amis youth can arrange steps freely assembled from different Amis folk-songs. Furthermore, the body movements of this sort of traditional dance represent the close relationship between the Amis people and their lands (Li 2001a, 2001b). These dances are also the greatest source for Amis youth’s bodily expressions, especially those of Laciensi and Lakayakay, which appropriate many body movements from these kinds of traditional dances into their modern dances. Even in the traditional dance competition, the young people always re-create movements that are different from those of the elders. For example, in the traditional dance of Laciensi in 2003, one movement signified reaping, which several dancers exaggerated, thrusting out their chests to the extreme, raising their hips high, and clenching their hands tightly. The exaggerated movements made the spectators laugh boisterously. Nobody would have thought that they were dancing the wrong dance. In short, the bodily expressions of the traditional dance of young Amis have in one sense inherited many symbols and meanings from the original Amis dances; in another sense, they have also simultaneously re-created and transformed the originals through a number of variations. Although traditional dances could never have the same number of variations as the modern dances and the plays, young Amis still have their own innovatory space in which to perform their bodily expressions.
Amis Hip-Hop as Metaphor While from the perspective of outsiders the young people of A’tolan may seem to be showing that they are not really ‘traditional’ Amis, in reality they are mocking these external imaginations through their own bodily expressions. That is, in general the young people merge and transform enormous and varied resources based on the ‘traditions’ of the A’tolan Amis into their bodily expressions. Young Amis can then agreeably exhibit their subjectivity and creativity with their bodies, even within global flows, in their own social and cultural context. On the other hand, by referring to and respecting the heritage of the elders and ancestors of A’tolan, they continually manipulate them with their own imaginations, informed by the flows of global media, to create their own bodily expressions that are different from those of the middle-aged and the elders of A’tolan. Moreover, behind the bodily expressions of the young
168 Futuru C.L. Tsai
dancers are the female images, the real source of creativity. Finally, the recreation and transformation process of Amis young people’s bodily expressions is a sort of socialization process, not only moulding, but also representing the shifting identities among different levels within the age set, the age grade, the age organization, the village and ultimately a worldwide pan-indigenous identity.
Figure 7.4. The malikoda allows for variation with some kinds of limitations (photo by the author)
The Bodily Expressions of Contemporary Young Amis in Taiwan 169
Expressing Difference As mentioned above, the public imagination sees the bodily expressions of the Amis youth in two preconceived ways: on the one hand, they must be ‘natural’ and ‘primitive’; on the other hand, in common with the views of certain ‘anxious intellectuals’, they are allegedly showing a loss of ‘culture’ and ‘tradition’ to the forces of ‘commercialization’, ‘touristization’ and ‘secularization’. Faced with these two public imaginations, Amis youth show tremendous ingenuity in representing the imaginations of outsiders, as if they were indeed as the outsiders think of them. However, the imaginations that emerged from the ‘Others’ are not those of the ‘spectators’ in the context of the Amis youth’s bodily expressions. If one wants to be a ‘real spectator’, one must first become an ‘actor’, that is, one must enter into the social and cultural context of A’tolan Amis. Yet, in front of the middle-aged and elders, young Amis also present their ‘independence’ in terms of bodily expression. For example, although the malikoda and other traditional dances are based on the body movements of the lower half of the body, Amis youth still create and transform them with plenty of upper body movement, not to mention the modern dances, in which they borrow, transform, create or re-create vast and varied body movements. Through the flows of global mass media, as with music and television, they select the kinds of music and moves that are based on their interpretations of themselves and their own culture, and then either merge or transform them with the dance heritage of A’tolan Amis into their bodily expressions. The results of the transformation and creation processes represent the unique vitality and diversity of A’tolan Amis youth. Amis youth plays present similar differences in terms of the different kapots, as well as similar plays in the public mainstream in Taiwan. The emergence of Amis youth plays out of hierarchical relations in the age organization system is not entirely new, nor even a wholly contemporary phenomenon. According to the ethnography of a Japanese scholar in the 1930s, which described the relationships between the different age sets in terms of plays: When the A’tolan juveniles reach a certain age, they are called Pakalongay. They get together in the men’s house in the village, be [sic] forced to serve their predecessors and the elders, or have work imposed on them. Five year[s] later, they will be upgraded to become the Kapah. Before becoming the Kapah, the upper age set choose a certain morning to teach them how to dress in the new clothes with many wicked plays. In the meantime, a competition of holding the hairs, fighting and playing with each other, called maraorotai would be held. At the time of sunset, the Pakalongay were forced to eat the millet cakes: if they refused, they would be punished by the upper grades. This is the upgrading ritual to the grade of Kapah, which is called misakapot. Passing through these rites, the Pakalongay could then be categorized as Kapah and receive their kapot’s name from the elders. (Furuno [1945] 2000: 50)
170 Futuru C.L. Tsai
According to this description, hierarchical relations among age sets are very rigid. However, over seventy years later in A’tolan, although relations among age sets remain rigid, there has been a change in the direction of a kind of joking relationship. Amis youth plays are a kind of representation of these joking relations (Tsai 2004). Ethnic Chinese youth do not stage similar plays to those of Amis youth when they gather together. For Han youth, these kinds of plays would be staged only in order to become acquainted with strangers; but for Amis youth, they are staged by and for those who already know each other intimately. The difference is that the plays of the Amis youth both intensify the unity of being a common age set and obscure the hierarchical relations of the age grades more generally. Some Han youth might parody an ‘old-fashioned’ play, especially those who are the so-called ‘seventh-grade’ (Qinianji) generation of young men in Taiwan.17 However, what makes the Amis youths’ parody different has to do with the context of their performances, along with the fact that their audience are the older age groups, and that they effectively amuse the audience and thus relax hierarchical relations. Amis youth transform and integrate many other elements along with the heritage of A’tolan to create a certain style of body expression which seems similar to, but is in fact totally different from, the public’s and some scholars’ imaginations. Moreover, Amis youth, especially in the modern dance, create an alternative body expression to that of the middle-aged and elders by absorbing and transforming several musical and dancing cultures from different parts of the world, accessed through the global flows of media and images. Although the styles of Amis youths’ bodily expressions are different from those of the middle-aged and the elders, still the youths’ expressions are accepted and appreciated by the middle-aged and the elders of A’tolan.
Figure 7.5. The modern dance of Laciensi in 2005 (photo by the author)
The Bodily Expressions of Contemporary Young Amis in Taiwan 171
Young Men as Actors, Young Women as Directors As I have mentioned already, although the ‘youth’ here are mostly males, Amis dances are generally regarded as relating to women, who are also the driving force behind the Amis youths’ bodily expressions. It is essential to explain this phenomenon in terms of the gender relations of the Amis. The gender relations represented in Amis youths’ bodily expression seem inexplicable from the perspective of the complementary and oppositional gender relations of Amis. No matter whether in the modern dance, the traditional dance or the play, all of them, either directly or indirectly, are related to the image of the ‘female backstage director’. Amis traditional dance is basically directed more closely to the land: not only are the movements always focused on the lower half of the body, which is nearer to the ground (Li Hungfu 2001a: 18), but the postures and movements are mostly transferred from the body movements involved in cultivation. Furthermore, traditionally the image of land, cultivation and the major crop, millet, are connected with women in Amis society (Huang Shiunwey 1989). In other words, the Amis traditional dance is mostly a representation of the image of the females. Regarding modern dance, Amis youth also blend and transform many traditional dance movements into the modern dance, especially those movements relating to the cultivation of the land, that is, to an image of female labour.
Figure 7.6. Female traditional dance movements versus Laciensi’s traditional dance movements. The left is the traditional female dance transformed from the cultivation movements; the right is Laciensi’s traditional dance with the movement similar to the females on the left (photos by the author)
172 Futuru C.L. Tsai
The modern dances of Amis youth are mostly created by females, either directly or indirectly. For example, as mentioned above, since 2000, Laciensi’s modern dances have been arranged by the young woman called Strawberry. Lakangcing’s modern dances are also mostly directed by a woman called ‘Little Flower’. For these two kapots, the women direct the modern dances, though most of the dancers are young men. In the case of Lakayakay, although the modern dances are arranged by a male member, namely ‘Iron Coach’, in fact several women influence Lakayakay’s modern dances indirectly. First, Iron Coach’s mother is the master of the Blue Star Dancing Group of A’tolan, and Iron Coach has learned a lot from his mother, who in fact plays a backstage directing role for the Lakayakay. Secondly, Iron Coach has always been surrounded by many other middle-aged female dancers. Another event, which similarly represents the image of females backstage directing the Amis youths’ bodily expressions, personally involved me. I participated in Lakayakay’s modern dance as a dancer in 2002. Since I was putting on some weight at that time, I did not notice that my belt had not been put on properly and was slipping off. While we were dancing in front of the elders, my belt loosened into what looked like a tail. Although, as a ‘professional’ dancer, I continued dancing, it was a very embarrassing experience. At that time, there were two unfamiliar ‘older sisters’ who successively came onstage and tightened up my belt. However, none of my ‘good’ male friends from other kapots did anything to save me. This was not a unique event: I have seen many similar situations unfold in this manner. For instance, at the initiation ritual for the Pakalongay to be upgraded to the Kapah, it is the mothers who dress their children with the kayap (the Amis young man’s skirt), the symbol of becoming the Kapah. No matter what kind of dress or dance, the women play a critical role backstage in directing the bodily expressions of young Amis. That is, the women direct the youths’ bodies to perform in accordance with an essentially female image. It seems that the ‘complementary’ and ‘oppositional’ gender relations of the Amis cannot entirely explain the gender relations represented by the Amis youths’ bodily expressions. Especially from the ‘oppositional’ point of view, the age organization is generally regarded as a male domain, the kinship system as a female domain. How could the women become such a critical part of the activities of the male age organization as the dancing competition of the age sets? Moreover, why do male Amis youth not consider the women’s backstage directing oppressive in the same way that women in patriarchal societies consider men’s ‘backstage directing’ of women’s bodies oppressive? For example, when Euro-American male film directors use women actors’ bodies to represent their own stereotypical images of what women ‘should’ be like, feminists often see this as suppressing the women’s own subjectivity. The simple answer to this question is that gender relations in Amis society probably go beyond the concepts of simple ‘complementarity’ or ‘opposition’. Recently, an idea has emerged that the distinction of gender in Amis society is not oppositional but hierarchical, though still based on complementary relations (Lo 2005). Although this seems to explain gender relations through
The Bodily Expressions of Contemporary Young Amis in Taiwan 173
Amis youths’ bodily expressions, the concept of hierarchical gender relations is still obscure. This begs the question of how hierarchical relations could not be regarded as oppressive. In my view, this can be answered by looking at the Amis concept of the family, especially relations between siblings as the core of the family. Although the contemporary kinship system of A’tolan Amis is not formally divided into matrilineages, the idea of the family and sibling relations still forms the core of ‘relatedness’, this being quite different from Han society, which is patrilineal (Chen 1986; Lo 2000; Huang and Lo 2001). Siblings in the Amis ‘family’ have different roles based on gender, especially in ancestral rituals and kinship-related activities (Lo 2000). In short, the core of an Amis family consists of this form of differentiation. The representation of gender relations in Amis youths’ bodily expressions presents a similar sibling relationship to that of the family: that is, the age set or the age organization is in the image of a family. Moreover, from the linguistic representation in Amis, this also implies that the age organization is to some extent an extension of the family. The family in the Amis language at A’tolan is called luma, and the annual harvest festival is called kiluma’an, literally ‘getting together as a big family’. The common root of both words is luma, ‘family’. The annual harvest festival mainly consists of the activities of the age organization, again suggesting that the age organization is a certain kind of family. From this point of view, gender relations in the representation of the Amis youths’ bodily expressions also actually resemble the relationship between siblings. Thus, the different duties of the two genders of siblings in Amis youths’ bodily expressions translate into women taking on the role of directors, the young men that of actors, all of which is based on the notion of the family in the context of A’tolan Amis.18
Shifting Identities: From the Age Set to Pan-indigenous Identity The annual harvest festival and the age organization have become symbols of contemporary Amis in Taiwan (Guo 2002), including among young people in A’tolan, who participate in the age organization and the annual harvest festival partly for this purpose. However, the processes of the Amis youths’ identity politics are much more complicated. Their identities shift in the context of their bodily expressions and on several levels, namely the age set, the age grade, the age organization (the village) and a Taiwanese pan-indigenous identity. The age set is always regarded as a unity within the age organization, and its identity is represented through participation in the practice and performance of modern dance, traditional dance and plays. Amis youth have been socialized through this process, where individuals are gradually integrated into the kapot. Meanwhile, the identity of the age grade emerges from the sociocultural context of A’tolan Amis, that is, the Kapah. Young people need to present their
174 Futuru C.L. Tsai
bodies as Kapah, literally ‘beautiful and shining’. They select forms and styles for their dances that have heavy beats and brisk rhythms, on the one hand to distinguish themselves from the middle-aged and the elders, and on the other hand to present the ‘beautiful’ and ‘shining’ quality of youth. Whenever the Amis youth perform ‘wonderful’ dances, the elders on the stage always say something like kapah no niyaro, that is, ‘Look! Those are the young men of our village!’ Young people in the Amis sociocultural context are not above resistance, but they ultimately represent the whole village. From the forms of Amis youths’ bodily expressions, they present not only the Kapah identity, but also a ‘pan-indigenous’ identity in Taiwan and even throughout the whole world, especially in the genre of modern dance. Amis youth either borrow or transform different music and dance movements from multiple indigenous or minority cultures into their modern dances, including African, African American, Maˉori, American Indian and other indigenous peoples of Taiwan.19 Some dance steps are even borrowed from non-indigenous cultures such as the Cool Dragon of Korea. Nonetheless Amis youth still transform these into the imagination of the Amis modern dance: for example, the dance steps of the Cool Dragon are focused on the movements of the legs, then on stepping heavily on the ground, similar to Amis traditional dance, especially the malikoda. In other words, the Cool Dragon’s dance movement is transformed and recreated in the context of the Amis at A’tolan. In the context of shifting and multiple identities, therefore, Amis youth create a sense of their own subjectivity and identity in different contexts. The integration and borrowing of multiple cultures yields a kind of ‘new’ identity and subjectivity. Similarly Smitha Radhakrishnan (2003) describes a transcultural dancing group called Surialanga, based on the common living experiences of its Indian female members and Zulu male dancers, and creating a new dance blending Zulu genres with Indian Bharatnatyam dance, which they performed at South African President Nelson Mandela’s inauguration, thus transforming the distrustful experiences of Indians and Zulus into something more positive. Judy Flores (2001) also points out that the Chamorro in Guam adapted dances that were regarded as representative of Chamorro culture from their Polynesian and Micronesian neighbours, based on Chamorro social structure, to create the Chamorro dance. Through this cultural process of integrating and exchanging, a sense of belonging is created for the identity of the ‘Pacific Islanders’, as well as an understanding that Guam is not merely a colony of the United States, but is stepping forward and, indeed, away from it. Similarly, in the transformation and adaptation of music and dances from other cultures, Amis youth have been creating their own subjectivity in the complex processes of identity construction and distinguishing themselves from ‘outsiders’. The ‘outsiders’ cannot become the spectators of Amis youth’s performances unless they enter an Amis sociocultural context. Only then will they have any chance to understand the creativity and subjectivity of Amis youth of A’tolan, as well as to realize the centrality of female images in their own bodily expressions.
The Bodily Expressions of Contemporary Young Amis in Taiwan 175
Discussion and Conclusions Cultural reproduction is a dynamic and continuous process. For example, Mary Masayo Doi (2002) describes how contemporary traditional dance in Uzbekistan is now regarded as a symbol of ethnic identity, though it was originally transformed from the dance training used by the Soviet regime in official attempts to change the conditions for Muslim women at the beginning of the twentieth century. This dance training of the Soviet government gradually became the symbol of Uzbek identity in resisting the Soviet regime. After Uzbekistan gained independence from the Soviets in 1991, the traditional dance was finally considered to represent both the identity of the state and the nationality of Uzbekistan. In another example, Daniel Miller (1992) argues that, although the fashions and forms of soap opera in Trinidad appear to be influenced by American soap operas, its prevalence in Trinidad is actually based on a complicated local concept of ‘Bacchanal’ (including the truth, the scandal beyond the truth, the confusions and the disorder). The ‘Bacchanal’ concept in both the soap opera and the carnival gave vent to identity when Trinidadians were faced with the effects of globalization, including their colonial history, the discovery of oil and the subsequent depression in the oil industry. Thus, the local context and the global process are interactive. Similarly, contemporary Amis youths’ bodily expressions should be put in both the global and local sociocultural as well as historical contexts to be understood. They blend many ‘Other’ or ‘outside’ elements with their ancestors’ heritage into specific bodily expressions to create a certain position which diverges from public representations of them. This difference is exactly the source of ‘anxiety’ for those outsiders who fear the destruction of tradition. However, the Amis elders and the middle-aged of A’tolan do not see Amis youths’ bodily expressions as neglecting the ways of ancestors, but rather as representing the beauty of the village. In other words, outsiders cannot become spectators of Amis youths’ bodily expressions unless they can immerse themselves in the context of the A’tolan Amis. For Amis youth, the bodily expressions represent the performance of subjectivity and identities blended with the global flows of music and dance movements under the specific sociocultural environment of A’tolan Amis. This essay does not describe contemporary Amis youths’ bodily expressions in the genre of ‘hip-hop’, but employs it as a metaphor and a model in order to interpret their bodily expressions. That is, in the context of social and cultural flows, Amis youth are shouldering the creativity and peculiarity of the ancestors in order to exhibit their agency, subjectivity and identities. In so doing, they are creating an alternative style and a form of ‘hip-hop’ which presents their complex identities and unique gender relations, subverts the public image of them, and has become ‘Amis hip-hop’.
176 Futuru C.L. Tsai
Acknowledgement I would like to thank DJ W. Hatfield, Robert Parkin, Eli Alberts and Christian Anderson for editing the English language version of this essay. DJ W. Hatfield, Robert Parkin, James Wilkerson, Teri Silvio and Ku Kunhui also gave me many constructive comments on the draft. However, I take full responsibility for any unclear descriptions or arguments. Also, I feel an immense gratitude to the Chiang Ching-Kuo Foundation for International Scholarly Exchange, who granted me a scholarship to present this essay at the 2005 American Anthropology Association annual conference in Washington DC. My documentary film, entitled Amis Hip Hop, can be viewed and downloaded at http://www.oz.nthu.edu.tw/~d929802/amishiphop/index.html.
Notes 1. Amis is not the plural of Ami but is one of the main self-designations. However, Ami is used by Han Taiwanese and Japanese, which is why some articles describe Amis as Ami. 2. Figures from the web database of the ‘Council of Indigenous People’, http://others.apc. gov.tw/popu/ 9310/aprp5g02.htm, accessed 11 November 2004. 3. Actually, A’tolan is the Amis name of the village, the Chinese name of the administrative division being Dulan. In Dulan, the populations of Han and Amis are almost equal. The total population of Dulan was over 3,000 at the end of 2003. 4. Today we would say that he ‘can’ participate in the age organization, but in the past the appropriate verb would rather have been ‘should’. Because of the circumstances of social and historical change, there is no longer any ‘requirement’ to become involved in the age group. However, in the context of cultural revitalization, more and more Amis youth are now choosing to participate in the age organization. 5. Kapot can also be used to refer to an age set. 6. ‘Traditionally’, girls did not attend the age organization. Once married, they would participate in the age organization along with their husbands. In contemporary times, most unmarried girls participate in the age organization with their male friends, but they still do not officially attend the kapot until they marry and become Matatapalay along with their husbands; teenage girls are called Lakaying, the age set of girls. In addition, the wives of Mihiningay to Tukal are called Militepulay, which is not an age set but an age grade. 7. ‘Pull out the hairs’ is pronounced qu bamao in Mandarin, but Amis youth sang it as ji bamao, which means ‘pubes’. 8. Originally, according to the custom of Amis at A’tolan, Laciensi should visit the kapot Lakenca, but because the members of Lakenca were too few to form a kapot, they were combined with Lakayakay to be a kapot at that time. 9. The China Youth Corps ( Jiuguotuan) is a corporation set up in 1952 on Chiang KaiShek’s order, his son Chiang Ching-Kuo being the first chairman. The corps held many activities or summer camps for the youth of Taiwan to construct patriots according to its goals. There are lots of team plays designed by the corps that were performed in the activities or camps. CYC is almost a pronoun for ‘team plays’ in Taiwan. 10. I was told by a member of the advisory team that Lakayakay would have won first place if only I had not dropped my belt twice. 11. Iron Coach has been the choreographer of the modern dance for the Lakayakay since 2002. He works in Taipei, and is always visiting music stores to search for ‘suitable’ music
The Bodily Expressions of Contemporary Young Amis in Taiwan 177
for the modern dance. The nickname ‘Iron Coach’ was coined because every time he teaches the kapot the dance steps, he is very tough and vigorous, and seems not to get tired, acting as a coach just as hard as iron. But everybody trusts him and relies on him very much. 12. The haka is a traditional dance form of the Maori of New Zealand, which is called ‘Maori war dance’ in Taiwan. 13. There are several baseball players in Laciensi. 14. The Puyuma are an Austronesian-speaking ethnic group in Taiwan. The song was selected from Samingad’s first published album, The Voice of Sun, Wind, and Grasslands, for which she won an award as ‘Best New Singer’ at the 2000 Golden Music Awards in Taiwan. 15. The Bunun are an Austronesian-speaking ethnic group in Taiwan. They live in the mountainous area of central Taiwan. Biung was given the title of ‘Best Dialect Singer’ at the 2001 Taiwan Golden Melody Awards. 16. The Tao are an Austronesian-speaking people who inhabit the island of Botel Tobago (also called Orchid Island, or Lan Yu), located off the south-east coast of Taiwan. 17. The ‘seventh-grade generation’ (Qinianji) is a popular term referring to those who were born between 1980 and 1990 in Taiwan. The term also has the connotation of a generation that is fragile but innovative, as well as anti-traditional. 18. However, it is interesting that the dances of the middle-aged and the elders seem to present women as both actors and directors, unlike the presentations of the Amis youths’ bodily expressions. This difference requires further research. 19. Lakayakay borrowed American-Indian music to perform their modern dance in 2004.
References Bloch, Maurice. 1989. Ritual, History and Power. London and Atlantic Highlands, NJ: The Athlone Press. ———. 1992. ‘Internal and External Memory: Different Ways of Being in History’. Paper presented at the Edward Westermarck Memorial Lectures, Helsinki, 31 January. Bucholtz, Mary. 2002. Youth and Cultural Practice, Annual Review of Anthropology 31: 525–52. Chen Wende. 1985. ‘Kuanfuhou Riben xuezhe duei Ameizu de yianjiou’ [Studies of Amis People by Japanese Scholars after 1945], Thought and Words: Journal of the Humanities and Social Science 23: 2. ———. 1986. ‘Ameizu qingshu zhidu de zai tantao: yi danman bulou wei li’ [Reconsideration of the Amis Kinship System], Bulletin of the Institute of Ethnology Academia Sinica 61: 41–80. ———. 1990. ‘Danman Ameizu nailingzu zhidu de yanjiu yu yiyi’ [The Study of AgeSet Systems and its Theoretical Implications: A Taiwanese Case], Bulletin of the Institute of Ethnology Academia Sinica 68: 105–44. Doi, Mary Masayo. 2002. Gesture, Gender, Nation: Dance and Social Change in Uzbekistan. Westport, Conn.: Bergin & Garvey. Flores, Judy. 2001. ‘The Re-creation of Chamorro Dance as Observed through the Pacific Festival of Arts’. In Monographs of Taitung Austronesian Cultural Festival in 2001. Taitung: Taitung County, pp. 162–86. Furuno, Kiyoto. (1945) 2000. Taiwan yuanzhumin de jiyishenghuo [The Ritual Life of Indigenous People in Taiwan]. Translated by Yie Wanqi. Taipei: Indigenous Culture. George, Nelson. (2000) 2002. Hip Hop America. Translated by He Yingyi. Taipei: Business Weekly Publication.
178 Futuru C.L. Tsai
Gulliver, Philip H. 1968. ‘Age Differentiation’, Encyclopaedia of the Social Sciences 1: 157–62. Guo Qianting. 2002. ‘Zuqunxing yu wenhua rentong: chishang ameizu fongnian jieqing de chonggou’ [Ethnicity and Cultural Identity: The Reconstruction of the Amis Harvest Festival in Chishang]. MA thesis, National Taiwan University. Hayles, Katherine. 1993. ‘The Seductions of Cyberspace’. In Rethinking Technologies, Verena Conley, et al. eds. Minneapolis: University of Minnesota Press, pp. 173–90. Huang Gueichao. 1994. Fongnianji zhi lu [A Journey of Harvest Festivals]. Taitung: East Coast National Scenic Area Administration. Huang Shiun-wey. 1987. ‘Yiwan jiuluo zhi jiyi’ [The Rituals of Yiwan Village]. In The Study of Ritual and Folk Songs and Dances of Indigenous in Taiwan, Liu Bingxiung et al. Taipei: Academic Sinica, pp. 3–38. ———. 1989. ‘Cung suishi jiyi kan yiwan Ameizu chuantung shehui de hubuxing yu jiexiuxing’ [Complementary and Hierarchical Relationships between Two Forms of Traditional Ami Social Organization: A Symbolic Perspective Based on an Analysis of the Annual Rituals of Iwan Village], Bulletin of the Institute of Ethnology Academia Sinica 67: 75–108. ———. 1991. Dongbu haian Ameizu shehui wenhua ji diaocha yanjiu [Research on Amis Society and Culture on the East Coast of Taiwan]. Taitung: East Coast National Scenic Area Administration. ———. 2005. Yizuguan, diyuxing chabie yu lishi [Images of Others, Regional Variations and History among the Amis]. Taipei: Institute of Ethnology, Academia Sinica. Huang Shiun-wey and Lo Sumei. 2001. Taitungxian shi: Ameizu pian [The History of Taitung County: Amis People]. Taitung: Taitung County Government. Li Hungfu. 1994. ‘Xiandai Ameizu fennianji de hanyi yu wudau xianzheng gungneng’ [The Meanings of Modern Harvest Festival and the Functions of Symbolization of Amis People], Taiwan Indigenous Voice Bimonthly 6: 100–103. ———. 2000. ‘Ameizu Liloshe de Talatuas (Jizu yishi): Yixiang Wudao renleixue de guandian’ [The Ritual of Ancestor Worship (Talatuas) in the Amis Lilo Village: A Perspective on the Anthropology of Dance], Taiwan Indigenous Voice Bimonthly 25/26: 180–83. ———. 2001a. ‘Yuanzhumin wudao’ [Aboriginal Dances], Journal of Arts Education 124: 16–21. ———. 2001b. ‘Ameizu yishi shenti dungzuo yihan: Yi Fenbinxiang Ameizu Gangkocunfennianji weili’ [The Meanings of Body Movements in Rituals: The Amis Harvest Festival in Gangko Village]. In Monographs of Taitung Austronesian Cultural Festival in 2001. Taitung: Taitung County, pp. 208–13. Lin Mingmei. 1997. ‘Zungku dao haian, Malan dao Jian: Fengfu duocai de Amsizu Fengnianji wudao’ [From Valley to Sea Shore, Malan to Jian: Rich and Colorful Harvest Festival Dances of Amis], Performing Arts 55: 74–78. Liu Fengxue. 2000. Yu ziran gungwu: Taiwan yuanzhumin wudao. [Dancing with Nature: Indigenous Dances in Taiwan]. Taipei: Business Weekly Press. Lo Sumei. 2000. ‘Dulan Ameiren de saomujie: yige shehue wenhua de bianqian yu zhuanhuan zhi tantao’ [The Ancestral Festival of the Ami People in ‘Tolan: A Discussion of Sociocultural Change and Transformation], Eastern Taiwan Studies 5: 35–70. ———. 2005. ‘Xingbie qubain, jiexu yu shehue: Dulan Ameizu de xiaomi zhoqi yishi’ [Distinction of Sex, Hierarchy and Society: The Ritual Cycle of the Millet in the ‘Amis of ‘Tolan], Taiwan Journal of Anthropology 3(1): 143–84. Mead, Margaret. 1939. From the South Seas: Studies of Adolescence and Sex in Primitive Societies. New York: W. Morrow.
The Bodily Expressions of Contemporary Young Amis in Taiwan 179
Michaels, Eric. 1994. Bad Aboriginal Art. Minneapolis: University of Minnesota Press. Miller, Daniel. 1992. ‘The Young and the Restless in Trinidad: A Case of the Local and the Global in Mass Consumption’. In Consuming Technologies: Media and Information in Domestic Spaces. R. and E.H. Silverstone, eds. New York: Routledge, pp. 163–82. Ming Liguo. 1997. ‘Ameizu de yinyue wudao wenhua’ [Music and Dance Culture of Amis], Journal of Dance Education 1: 6–18. Osumare, Halifu. 2001. ‘Beat Streets in the Global Hood: Connective Marginalities of the Hip-hop Globe’, Journal of American & Comparative Cultures 24(1/2): 171–81. Radhakrishnan, Smitha. 2003. ‘African Dream: The Imaginary of Nation, Race, and Gender in South African Intercultural Dance’, Feminist Studies 29(3): 529–38. Ragland, Cathy. 2003. ‘Mexican Deejays and the Transnational Space of Youth Dances in New York and New Jersey’, Ethnomusicology 47(3): 338–54. Stone, Sandy. 1991. ‘Will the Real Body Please Stand Up? Boundary Stories about Virtual Cultures’. In Michael Benedict, ed. Cyberspace: First Steps. Cambridge, Mass.: MIT Press, pp. 81–118. Thomas, Deborah A. 2002. ‘Democratizing Dance: Institutional Transformation and Hegemonic Re-ordering in Postcolonial Jamaica’, Cultural Anthropology 17(4): 12–55. Tsai, C.L. Futuru. 2004. ‘Fame and Banter: An Alternative Transformation of an Amis Age Organization in Globalization and Capitalism Process’. Paper presented at NATSA Annual Conference at Hawaii University, 18–21 June. ———. 2005. ‘Virtual Memory: Cultural Changes of an Amis Age Set on the Internet’. Paper presented at the Asian Digital Cultures Conference, Taipei, 28–29 July. Wang Sungyin. 1998. Taiwan yuanzhumin yu xinwenmeijie: xingxing yu zaixian [Indigenous People of Taiwan and the Press Media: Images and Representations]. Taipei: Shiying. Wei Hueilin. 1986. ‘Taiwan tuzhu shehuei de buluo zuzhi yu quanwei zhidu’ [Tribal Organization and Authority System of Indigenous Societies in Taiwan]. In Essays on Indigenous Society and Culture in Taiwan, Huan Yingguei, ed. Taipei: Linking Press, pp. 111–40. Wei Hueilin, Yu Jingqiuan and Lin Hengli. 1972. Taiwansheng tungzhi Amei pian: jiuan 8 tungzhouzhi [Amis in the Cyclopedia of Taiwan: No. 8]. Taichung: Literature Committee of Taiwan. Zhao Qifang. 1997. ‘Ji Muosheng Yio Shouxi: Yi Jiouzu weili kan Taiwan de yuanzhumin shehei jiqi wudao’ [The Familiar and Unfamiliar: Indigenous Societies and Dances of Taiwan], Performing Arts 55: 70–73. Zhuang Kuoxin. 2003. ‘Taiwan yuanzhumin Ameizu Taibalang buluo daowei jieceng wudao zhijiaoyu yiyi’ [The Educational Meanings of Amis’s Miholol Dancing at Tribe Tafalong], Journal of National Taiwan College of Physical Education 12: 42–60.
Chapter 8 Contesting Memory: The Shifting Power of Narration in Contemporary Paiwan Contexts Li-Ju Hong
Introduction When I began my fieldwork in a Paiwan village in southern Taiwan several years ago, I was greatly intrigued by the lack of correspondence between various historical documents and local oral history. It is hardly surprising to anthropologists to find written text and social history in contention with one another. It becomes more interesting, however, when they are intertwined with ‘social reality’, that is, with what is presently happening around us in a village. Rightly or wrongly, while researchers often attempt to draw a line between a historian’s or an anthropologist’s own written texts and the lives of those in the village where he or she is conducting research, that boundary is now often blurred and subject to challenge. In her study of the juvenile culture of Samoans, Margaret Mead’s aim ([1928] 1961) was to draw comparisons with contemporary American juvenile culture at a period of supposed crisis in American society in the 1920s. Thus, Mead’s work on Samoans was initially written for an American audience. However, her work has since been widely read by Samoans themselves, and, despite the heavy criticisms of it (e.g. Freeman 1983), in more recent years it has become a reference work for Samoan cultural revitalization. This is just one example of the fate of many ethnographic writings that are becoming known to a much wider audience in this era of postcolonial spirit and mass transnational communication. The upshot is that these works have also become intertwined with the social history of the researched community; that is, contemporary local readers were not part of the original target readership, but now their children and grandchildren are increasingly aware of written resources on their own cultures, and that awareness has its own consequences for village life.
Contesting Memory 181
The Paiwan are an Austronesian-speaking indigenous people of Taiwan. As of 2010, they had a population of approximately eighty-six thousand, living in more than seventy villages in the southern area of the Central Mountain Range of the island. Traditionally subsisting on slash-and-burn agriculture of millet and taro with supplementary hunting, the Paiwan have a hierarchical social system divided into the classes of aristocrats and commoners.1 In 2003, while I was visiting the Paiwan village of Puleti, the eldest daughter of a mamazangilan or aristocratic family, who I shall call Muni, made a remark that triggered my interest in the transformation of oral histories into written histories: Look at this piece of paper. It is a genealogy from the Japanese colonial era, which was recorded in Japanese [switching to Hokkien, Muni then adds parenthetically]: ‘I can’t recognize this either.’ [ghwa31-ma24-khuã53-b 24] This is number 207 and 208. So they put us together. What is this? Paiwan? It is our Taljalub [Tariarp] family. And this is the Rovaniau family. And altogether it is heavily involved with the entire Qalangiyan genealogy. Once this genealogy has been reorganized and published, it is all connected with us people of Puleti. But we did not know this before. If we do not work on our village history, the history of Puleti is up to other people’s interpretation. (Puleti, 13 August 2003)2
e
Muni was a member of the village gazetteer-writing team, sponsored and trained by the government to work on the history of Puleti village. She was looking at an account from the Research Report (Rinji [1915] 2002), which I will discuss further below. Muni herself also collected historical materials and other information concerning present-day Puleti village. The details that puzzled her exemplify one of the social realities resulting from the transformation of oral ‘texts’ into written documents. That social reality is that oral histories play a very particular and important role in Paiwan society, a role in which written texts are incongruous. Conversely, these incongruities may be regarded as themselves reflecting the recognition, or alternatively the denial, of various texts as a new means of local contestations of status and hierarchies among these closely related Paiwan villages. In an effort to understand better the dynamics of oral versus written history, in this essay I look at two topics. The first topic is what happens when oral history is transformed into written history. Written histories, which in one way or another are all attempts to create a master narrative overwriting all others, is never a singular act but rather is embedded in a hierarchy of acts, all of which are subject to negotiation and contestation. I have found the same to be true of oral histories. This leads to the second topic; that is the implications and consequences for Paiwan oral histories in the process of the Paiwan people reclaiming and authoring their own written history. I begin with a brief overview of Paiwan oral historical narratives and their role in Paiwan society. I then offer a summary and discussion of four written texts, all of which turned collectively authored indigenous oral histories into written histories individually authored by three outsiders and one Paiwan. A
182 Li-Ju Hong
close look at these texts and their vastly different interpretations of a single event, the founding of Puleti village, demonstrates Muni’s predicament and reveals how deeply the agenda of an author colours their recounting of ‘the facts’. Lastly, I conclude with a description of the complications the Paiwan project faced in its efforts to reclaim and author the Paiwan’s own written history.
Paiwan Oral History In the past, sociolinguistic debates about the differences between oral and literate societies focused on aspects of social solidarity, interpersonal relationships, modes of social memory and concepts of reality. Ong (1977: 18–19) suggested that a purely oral society has greater social solidarity and closer interpersonal relations, since orality requires more face-to-face communication. Similarly, Lévi-Strauss (1982: 163–87) stressed the view that an oral society is closer to reality than a literate one. Goody (1977: 41–46) rejected all this on the grounds that the dichotomy these theories draw between orality and literacy is too sharp. More recent approaches, such as Halliday’s (1989: 98–101), treat both oral and written traditions as creative of meaning and social reality, although in different ways. Most importantly, while orality creates a world of happening, writing creates one of objects. He further points out, based on studies of language learners, that written texts tend to define a relatively static society or a cosmos as a fixed product or structure, whereas oral traditions tend to treat a society or cosmos as process. The Paiwan are very much concerned with the origin narratives and genealogies of family names, since these are related to the status of individuals, houses (umaq) and villages (kinayan). Regardless of rank, stories of origin and genealogies of Puleti village residents are narrated on several important occasions where status is an issue, and sometimes simply for entertainment. Village elders, especially those who are respected and possess good memories, are the main narrators. These narratives constitute knowledge passed down generation after generation to younger people when they are mature enough to participate in these occasions. These venues are important in the negotiation and renegotiation of status through efforts to legitimize power by imposing a certain genealogical interpretation. The statuses of individuals, houses and villages are often recounted during rituals for the naming of newborn children, marriage negotiations, wedding ceremonies and a wide range of daily concerns, such as drawing fishing and hunting boundaries, food distribution and justifying the privileges of using certain patterns in tattoos, sculptures, textiles, and so on. There is a principle of ‘precedence’ that is central to Paiwan narratives of the origins of villages and family names, best exemplified by the Paiwan words umaq (house) and vusam
Contesting Memory 183
(first-born). Umaq denotes the corporeality of a house, the house one is born into, and also one’s grave.3 Vusam basically means ‘seed millet’, but it is also used to refer to the ‘first-born’, the ‘headman’ and more generally the system of inheritance. When referring to the ‘first-born’, it refers either to a son or daughter as the candidate for succession to the house. According to tradition, the first child of a Paiwan family, male or female, especially among aristocrats (mamazangilan), would inherit the house name (ngdan na umaq), the house itself and most of its assets. Also inherited were claims to fish and animals caught by other lower-ranking villagers. The ownership of particular patterns – such as the motifs of the sun, snakes and feathers – were the privileges of aristocrats and their close relatives. In return, first-born children were expected to provide assistance to their younger siblings, for example, in building new houses when the latter had to move away from their natal houses (Wei 1960; Shi 1971, 1976; Jiang 1999). Significantly, the purpose of narratives regarding village origins and house names in Puleti village is invariably flexible and strategic. Heroic achievements or important alliances from earlier generations are often recited as evidence to justify a higher status that is then reinforced by genealogy: narrators could selectively trace lines back to the more prominent sides of a family without restricting themselves to maternal or paternal lineality. Expressions of status are especially evident in naming practices. For example, the head mamazangilan’s name is Taljalep Tanuvag. Taljalep is his inherited aristocratic house name and Tanuvag his personal name. The personal name of Tanuvag is considered prestigious and is reserved for the Taljalep family and its affines. This principle of the prestige of the past in naming in the present is wide ranging. On the one hand, the names of ancestors who died a bad death were often avoided by their descendants. On the other hand, the recitation of a family’s personal names during a marriage negotiation is a highly symbolic and effective way of claiming both the eminence of a family’s history of past alliances and its present status. Personal names from important alliances in certain previous generations are therefore remembered in order to ensure that the family’s descendants will stand out in future competitions over status (Ku 2002). As Eric Hobsbawm indicates (1997), oral traditions and family memories are seldom comparable with the sort of collective memory that is supported, directly or indirectly, by the nation state: We must be aware that this is so, particularly at a time when alternative ways of preserving the past – oral tradition, family memory, everything that depends on the effectiveness of intergenerational communications which are disintegrating in modern societies – are disappearing. In any case the history of large collectivities, national or other, has rested not on popular memory, but on what historians, chroniclers or antiquarians have written about the past, directly or through school textbooks; on what teachers have taught their pupils from those schoolbooks; on how writers of fiction, film producers or the makers of television and video programmes have transformed their material. (Hobsbawm 1997: 275–76)
184 Li-Ju Hong
What Hobsbawm points out is indeed happening in Puleti village, but in different ways and contexts. When discussing history, attention should be focused on who is doing the narrating or writing, and to whom, where, how and why these various texts of memory are being presented in terms of power struggles. Earlier research ( Jiang 1983; Ku 2002) showed that, along with attempting to maintain status by land tenure, tribute-collecting and privileges regarding certain ceremonial objects, Paiwan aristocrats specifically seek to consolidate their status by resorting to several types of oral tradition. Various topics covered in that oral tradition range from narratives of the origin myth and the relocation of the village to genealogies of important family names. The stories covering these topics are commonly narrated following one of two specific genres: milimilingan and taucikel. On the one hand, milimilingan is often understood as myth or legend, stories passed down from the ancient days that are not verifiable by the speakers themselves. On the other hand, taucikel is understood as accounts that relate events known to have happened to either a speaker or at least to someone known personally. As a privileged genre that is uniquely Paiwan, researchers often emphasize milimilingan’s fictive quality (e.g. Hu 2004). This is in contrast to taucikel, which is considered to be more faithful to actual events. I argue that this mythical quality of milimilingan, whether in oral or written form, exerts great symbolic significance through its structure and plots, and can be very consequential for social reality. That is, the Paiwan people use milimilingan to establish some sort of conjuncture between narrated and social realities. The characteristics of milimilingan seem quite unique and have long been noticed by outsiders. The Research Report (Rinji [1915: 111–12] 2002) commented on the contrasts between history and myth in its account of milimilingan. Four different meanings of milimilingan were described. The first sense refers to it as ancient history: Paiwan people called ancient historical deeds stories of milimilingan; they were similar to Japan’s jindaiki [Kamiyo Nanayo, the mythical period, known as ‘the seven generations of the divine age’] or Book of Genesis. For example, the ancestor of the headman was born from the eggs of a sun, or by singing. That ancestor created human beings, millet, taro and pigs. Or there were two suns in the sky, or two siblings left after the Big Flood, and the brother and sister married and became the origin of their group. These are considered the stories of milimilingan.
The second sense of milimilingan is as an adjective that describes something incredible, modern or creative: southern groups of Paiwan people refer to things as milimilingan now. Such things as phones, photos, music players, watches, films and so on. When they are surprised by an object, they call it milimilingan.
The third sense glosses milimilingan as a deity who created millet, taro and human beings:
Contesting Memory 185
Beside these, there was a bronze-handled knife, a milimilingan, a family heirloom of the headman, Ruvaniau, in the Kuvulj group. At the top of the bronze-handled knife, there was a human face figure. It was said that milimilingan was a supernatural deity. A long time ago their ancestor, Sara Ac, left his hometown of Paumaumap to supervise his land. A deity named milimilingan followed him, singing everywhere to create millet, taro and human beings. He is considered to have made many contributions to their tribe, and in order to commemorate him, after he died, Sara Ac imitated his figure and put it on the top of a spear.
The fourth and final sense of milimilingan identifies the name of the deity of iron: It was said that milimilingan died in Pinavavu Acan; reportedly, his skeletal remains changed into iron. For this reason, they used milimilingan as the name of the deity.
There are two points that should be noted about the above four meanings of milimilingan. The first point is that the author tried to understand this kind of genre in terms of both his own Japanese legend (jindaiki) and the Christian Bible. The second point is that his explanation of milimilingan recognizes it as a tool which can be linked to the origin story of the Paiwan aristocrats. This latter correlation with aristocratic status and its negotiability emphasized the sanctity of this genre among the Paiwan and underlined the importance of the aristocratic family in Paiwan origins.
Four Different Pasts The previous section suggests that the Paiwan people place specific importance on written and published versions of their origin narratives and that this emphasis is understandable only in the context of a hierarchical social system that consists of aristocratic and commoner houses. Underneath this visible dichotomy of two ‘classes’ is the more encompassing ideology of the house. Two elements are central to this house ideology: primogenitary succession, and memory of the past based on both personal names and house names. The origin narrative itself consists of both genres, milimilingan and taucikel. Although some sources define milimilingan as more fabulous and taucikel as more factual in terms of content, milimilingan is also the more privileged genre in narrating the origins of villages and chiefly houses. Origin stories usually follow a certain structure, although they come in many different versions. For example, the typical story of how the first founding family moved to Puleti village generally involves a misfortune. The details of the two specific misfortunes vary from one version to the next, but the structure of the plot remains the same. The earlier discussion of oral narratives as a means of social competition for status raises the general question of what happens when such narratives are written down and published. I tentatively address this question by presenting
186 Li-Ju Hong
four written texts on Puleti village from different historical periods in the twentieth century, and authored under different circumstances. Specifically, I look at how a single event, namely the origin of Puleti village, is described in four different written texts based on oral history.
Text 1 In the Japanese records, the genre of milimilingan was understood first through a comparison with the traditional Japanese genre of jindaiki. Secondly, the term itself was said to be derived from the name of an ancient deity and to be rooted in the story of an aristocratic family, thus further reinforcing its association with authority and high status. Japanese researchers expressed a keen ‘interest’ in the milimilingan genre. This interest was kindled by the activities of the Provisional Research Association for the Study of Old Customs in Taiwan, established by the Japanese colonial government in 1901. The mission of the Association was to collect information about the customs of the peoples in Taiwan in order to facilitate the governance of the island. Led by Okamatsu Santaro, a law professor at Kyoto University, and a colleague, the association focused on Taiwan’s local customary laws. Its purpose was to resolve problems regarding land, property and local customs, but members of the association also collected information on daily lifestyles and religious practices (Cheng 1974; Liu 1975: 9–11; Fujii 2001; Zheng 2002a, 2002b). Under the association, an Office of the Aborigines was established in 1909, its investigators including Kojima Hoshimichi, Ino Kanori and Mori Ushinoshuke, all experienced contract anthropologists. It mainly investigated the so-called ‘raw aborigines’, since both ‘the acculturated aborigines’ and the ‘cooked aborigines’ were regarded as falling within the realm of Han Chinese society.4 Mori Ushinoshuke emphasized that, in their investigations, everyone should make efforts to understand the viewpoints of the aborigines objectively and try not to interpret them from a Japanese standpoint (Zheng 2002b: 40). Origin narratives and the records of family names constitute a great part of this report, and the genres of both milimilingan and taucikel are included. We do not know how the documented stories were selected, or whether or not some versions were excluded. However, conflicting narratives did appear in the Research Report for which no explanations were offered. A description of Puleti village is available in the Research Report (Rinji [1915] 2002). It classified Puleti village as part of the Vuculj subgroup of the Paiwan. Subsequent anthropological research since then has continued to follow this classification. The entry of Puleti village (referred to as Puljti) begins with a brief description of its geographical location and population: This village is situated on the left bank of the upper Puljti River, at the foothills of Mount Tjagaraus. The elevation is about 2,000 [ Japanese] feet. To the north and
Contesting Memory 187
across the river are the villages of Kaviyangan and Kulaljuc. There are about 80 households in Puljti, and this village is about five [ Japanese] miles to the east of its administrative centre, Chaozhou. The original meaning of the name Puljti is now lost. (Research Report, Rinji [1915] 2002 V[1]: 11)
Later in the Research Report, in documenting the history of the founding and development of Puleti village, the anonymous author focused on the conflicts between the two houses of origin, Pataljinuk and Tjuruvu: The founding of this village seems traceable to very ancient times, but exactly when is hard to estimate. According to oral tradition, Satjagaraus came down from a mountain and Satjukutjuku came from the plains. The two married and established a village here. This was the founding of the Ilavure [Tjuruvu] house (the house was later merged into the Taljialep house, while the old building turned into a ritual house, known as vinᕈacan). … Afterwards two brothers, Djaljadjaljai and Saljimlji, arrived from the plains area. … Djaljadjaljai settled down in Puljti and founded the Pataljinuk house (the old building also turned into vinᕈacan now). The villagers honoured Djaljadjaljai as chief as well. Later on, the Pataljinuk house lost to the Tjuruvu house (later merged by the Taljialep house) over land rights. The house head Tjuku wrapped all the gata (beads) in a pack and left with her followers. They built a new village called Tjaᕈau and founded the Kazangiljan house. (ibid.: 65–66)
In addition to internal strife and fission within the village, the Research Report always paid special attention to inter-village relations, in accordance to the original colonial and administrative mission of the association.
Text 2 In his The History of the Paiwan People (2001), the Paiwan scholar Tong Chunfa (Masegseg Jingror) writes the origin history of Puleti village from an entirely different perspective. Tong, a Paiwan aristocrat and devout Christian, was recruited in 1994 by the Taiwan Historical Research Commission to write a volume for the series Taiwan Aboriginal History. The goal of this publication project was to support a movement of Aboriginals writing ethnic history from their own perspective. Tong Chunfa describes the history of Puleti village as follows: According to legend, Taljiyalep house used to be called Curuvu house. It was descended from a hatched egg of a hundred-pacer snake. There were seven members [siblings] of the house. The eldest was in charge of the house affairs, while the second one moved to the plains area for further development. In the Japanese era, plains people [descendants of the second sibling] still came to Puleti village to pay vadis (tribute gifts) and kazelu (tax). Afterwards some villagers killed the people who came from the plains to pay tribute, and the offended plains people thereafter stopped paying tribute. Among the remaining five siblings, one younger brother travelled in the direction of Mt. Tjagaraus looking for new agricultural land. The dog that accompanied him stopped at the Old Kuljaljau site and refused to move on. So the
188 Li-Ju Hong
brother erected a landmark and invited other siblings to move there. Some of them later moved further on to the eastern foothills of the Central Mountain Range. … One day the villagers of Lariyang went farming in the mountains. Nearby there was a lake said to be full of spiritual potency. Tjuku, the eldest daughter of the chief, came to watch the people farming and was drawn into the lake by the mountain deity; nobody was able to rescue her from the lake. Saoepe, the eldest daughter of another chief, was riding a swing in the late afternoon dusk. She suddenly soared high, flew in the direction of Chaozhou [in the plains] and disappeared. The chief of the Curuvu house was badly shaken by the two incidents; he believed these were supernatural punishments. He thenceforth changed the house name into Taljiyalep and moved out of the original residential building. (Tong 2001: 80–81)
It is noteworthy that Tong does not mention a dispute between the Pataljinuk and the Tjuruvu houses, but relates in detail the unfortunate incidents in the old settlement to account for the change of the chiefly house name from ‘Tjuruvu’ to ‘Talijalep’. In Tong Chunfa’s version (2001: 241), it was the ‘Rovaniaw’ family that led the villagers safely to a new settlement; ‘Rovaniaw’ was the ancestor of Katu Camak, the official head of this village. In this version, Tong Chunfa skilfully placed Katu Camak in a higher position, thus elevating the ‘Rovaniaw’ family to a higher position than ‘Talijalep’. In other words, it established that, as a descendant of the village founder, even though he is not a member of an aristocratic family, Katu Camak still has the right to sit as headman. I have compared the entries for other Paiwan village histories and noticed that the discrepancies between the Research Report and The History of the Paiwan People on the origin of Puleti village is typical of other discrepancies between these two works. The Research Report often gives highly detailed descriptions of conflicts and disputes between different houses of a village or between villages, by listing exhaustively all the names of rival and allied houses and villages. By contrast, in The History of the Paiwan People, the statuses of the head mamazangilan are not subject to much challenge, and disputes in the villages are seldom described. While Tong Chunfa’s book presented the ideal of ‘aboriginal people writing aboriginal history’, Tong himself felt that time constraints had left this account of Paiwan history incomplete. While the Taiwan Historical Research Commission was attempting to transform ethnohistory into an authoritative agency in the academic arena, Tong Chunfa felt that the most valuable cultural heritage in Paiwan history was still missing from the published book. He argued that the Taiwan Historical Research Commission project should continue to investigate related data in the field, such as maps, oral documents and archaeological investigation, thereby making itself an authoritative agency in Paiwan society.
Contesting Memory 189
Text 3 Pan Lifu’s volume The Civilization of the Paiwan Kavulungan: A Society Without a Written Language or Monetary System (1998) attempts to override the surface chaos of multi-vocal oral narratives with the authority and power of written texts. The book was sponsored and published by the Pingtong County government. Pingtung County is Taiwan’s southernmost county and has the largest population of Paiwan. The book has provoked great controversy in many Paiwan villages. Pan Lifu was a self-taught local historian who, although not Paiwan himself, claimed to have an unspecified partially indigenous background. He had his own political agenda for ethnic consolidation and revitalization, and was generally better known for his political commitments than his career as a historian. For example, he strongly opposed the installation of a bronze memorial statue of Chiang Kai-shek in the school where he taught, which brought his teaching career to an early end. Eventually, he even became one of the founders of the Democratic Progressive Party for South Taiwan. He also enthusiastically campaigned for office in many elections, but was never successful. Pan Lifu’s version of the origin history of Paiwan aristocrats focuses on the story of mamazangilan, which was the term for Paiwan aristocrats (1998). As he describes it: About 1100–1200 ad, mamazangilan was born from an urn, so he is a leledan, which mean[s] an infant from an earthen jar. When he came out, Atjitjan, a Paiwan, discovered him. So Atjitjan arranged for the care of this child; one person took care of his bathing, another responsible for his feeding; so mamazangilan was inherently taken care [of] by others.
Pan’s political agenda heavily influenced his writing. In his history, he consolidated a body of diverse oral traditions about an ancient ‘Big Flood’ and identified Mount Kavulungan as ‘the’ safe haven during the flood for all the Paiwan villages. He advocated the abandonment of the ‘erroneous name Paiwan’ and the adoption of a new group name, ‘Kavulungan’, taken from the holy mountain of Kavulungan.5 Under the sponsorship of the Pingtong County government, Pan’s tactic was to organize a series of meetings with the heads of the major chiefly houses in the area, at which the latter were asked to recite their respective house histories. During the meeting, Pan would also enthusiastically lobby these chiefs to consent to the name change from Paiwan to Kavulungan. According to the minutes published in the book, many of the chiefs who attended the meeting were supportive of Pan’s agenda. The project, however, came to an abrupt halt with Pan’s untimely death. In his attempt to create a pan-Paiwan identity, Pan attempted to overwrite local histories with a universal myth. While his account still strongly supports Paiwan hierarchy, it also glosses over any rivalry among houses or families in an effort to paint a picture of a united people. Thus, the specific founding of
190 Li-Ju Hong
Puleti and other villages, despite their importance in determining hierarchies and status, are not mentioned. Pan Lifu’s The Civilization of the Paiwan Kavulunjun invoked a rhetoric that was more sensational than logical; the discourse is neither scientific nor Paiwan. Due to the arrogant and politically popular rhetoric of the writing, with many colour photographs, all put into a coffee-table kind of book, Pan’s book has enjoyed a wide circulation among Paiwan villages. However, both the content of the book and Pan’s way of holding meetings to amass support from Paiwan chiefly houses remain controversial.
Text 4 Jiang Hai’s Drifting for Two Thousand Years ( Jiang 2000: 253–54) is another controversial publication. Jiang Hai is a journalist who claims no indigenous background, but was befriended by the Ruvanuyeav chiefly family. His book focuses solely on the prestigious house of Ruvanuyeav of Chalaabus village. He tries to synchronize on the one hand the chronology of the Common Era, and on the other hand major events named in the oral traditions of both the milimilingan and taucikel genres, and the life histories of consecutive heads (vusam) of Ruvanuyeav. Jiang Hai’s account of the origin of the house of Ruvanuyeav of Chalaabus village locates the house at the centre of Paiwan history. In his description of the origin of this family, he reports that: The God of the Sun happily fell into a water pool. The God of the Sun found that this is a beautiful place. He then created two eggs in an urn; suddenly, one male and one female broke their shells and emerged. … The God of the Sun prepared a copper pot to speed up the growth of these two infants. Soon, they became a handsome man and a beautiful girl. He made them husband and wife and thereafter [their issue] reproduced generation after generation. Now in Chalaabos, the old village site, there is still a pile of stones called Fulaw, which was the original place of the bronze copper. That’s why the house of Ruvanuyeav consider this is a sacred place. ( Jiang 2000: 18)
Jiang gives an alternative account of the history of the Ruvanuyeav family when talking about their history of migration to Puleti village. He wrote that after the Great Flood it was God’s will to ask the Ruvanuyeav house to move first to Checherutan, and was also God’s will to ask them to move back to their place of origin at Chalaabos ( Jiang 2000: 18). However, for the sake of fertility and reproduction, God changed their dialects so that some members of the Ruvanuveav family were unable to communicate with other members. This resulted in a major conflict, and so certain Ruvanuyeav members moved yet again (ibid.: 15–16, 19). Jiang Hai’s version was greatly influenced by Christianity. The creation of two siblings is similar to the creation of Adam and Eve. Also, the story of the
Contesting Memory 191
great flood and description of migration and a single language becoming different dialects reflect biblical accounts of the Tower of Babel. These Christian themes reflect the influence of Jiang Hai’s close friend Gao Defu, a preacher, who is the second son of the Ruvanuyeav chiefly house. Focusing on the Ruvanuyeav house as the centre of Paiwan history, Jiang concedes that any history from the point of view of a single house is bound to provoke argument, but he contends that there is no better way to deal with it ( Jiang 2000: Preface).
It is worthwhile to consider some of the more important features of these four texts. With regard to the Research Report, this was compiled during the Japanese colonial era for the purpose of better management of the rebellious aborigines. It was certainly neither intended nor envisaged that it would be circulated among or read by the Paiwan villagers. Alternatively, the self-evident intention of Tong Chunfa’s much more recently published work, The History of the Paiwan People, was to present a more or less orthodox history from the Paiwan point of view. This work is the sole example by a Paiwan author. Pan Lifu’s central concern was to go beyond local variation and promote a unified sense of Paiwan ‘civilization’. This led him to focus on myths, allowing his history to be broad enough to incorporate historical diversity. Jiang Hai’s Drifting for Two Thousand Years is an example of how the active support and endorsement of publications of written histories by Paiwan aristocratic families is used to enhance their own status vis-à-vis other contesting families. It is noteworthy that, although written over the course of nine decades, all four of the above published histories result from government projects. Each written history was intended to answer the governance needs of their respective periods, yet as they make their way back to the community itself, all have their relevance for contemporary social realities. This is evidenced in the remarks made by Muni (cited at the beginning of this chapter), someone who is, on the one hand, well informed of both Japanese and Chinese written texts, and, on the other hand, very familiar with the oral narratives of the Paiwan elders. The intention of Muni’s project to reclaim Paiwan history is an attempt to write a ‘true’ history through the eyes of the Paiwan within the master narrative tradition. When confronted with discrepancies between these various texts, she is able to manoeuvre and utilize her knowledge of both texts: to proclaim the house status of her village (a present-day agenda), or to pursue historical truth by drawing supplementary evidence from the village elders’ narratives (a form of historicism). However, because this act of writing history runs counter to the spirit of history as a concept in Paiwan culture, the endeavour in itself has met with local resistance. This resistance reveals the tension that has emerged in the social reality between the oral and the written. What really worries many Paiwan elders about the increasing emergence of written Paiwan records is the perception that any published genealogy or narrative will diminish or remove their flexibility and credibility in contesting status.
192 Li-Ju Hong
One such example of local resistance to written texts is the many years of effort expended to compile a written genealogy of the Qalangiyan clan in Taiwu County, where many Paiwan reside. This project was initiated by a Taiwan ministry legislative representative of Paiwan ethnicity. This still unpublished genealogy adopts the format of the Han Chinese genre of zupu (records of ancestors) and includes an anthropological vocabulary. Nevertheless many families are not satisfied with the structure of this written genealogy and dispute its validity.
Conclusion: Whose Subjectivity? There are many different ways of interpreting social reality. Historical researchers try to establish some sort of equivalence between actual past events and written documents. Anthropologists are known to criticize historical researchers for their narrow focus on only written documents at the expense of oral accounts. Milimilingan is often classed as a literary, rather than a historical genre. However, I would argue that interactions between human agency and texts are equally complex in both oral and written traditions. There is significant evidence that people use narratives of the past – including history – to legitimize the present. Oral accounts are like written history in serving as a kind of memory system. There is a difference between the two in so far as the subjectivity of oral histories surpasses that of written histories. Oral tradition, passed down from generation to generation, relies heavily on the re-presentation, reinterpretation and performance of the narrators, who have control over the ‘texts’ and constantly exert their agencies. On the other hand, written texts, especially widely circulated public ones, are able to escape the control of the original author and drift into contexts for which they are not intended. This is one reason why in some societies, like the Paiwan, oral histories are considered to have more authority than written ones. An examination of mele koihonua (genealogical, chanted panegyrics) as opposed to mo’olelo (historical narratives) in Hawaii led Valeri (1990) to assert that texts used for legitimizing the present depend for their efficacy on context. The Hawaiian mele koihonua genre focuses on continuity, establishing hierarchies on the basis of the birth orders and marriage alliances of the firstborn of families. In contrast, the mo’olelo genre stresses change and discontinuity, especially when describing the ten generations from King La¯loa to Kamehameha I, whose power did not derive from their priority in the birth order but from younger siblings challenging their older siblings (Valeri 1990: 165–67). These two types of narrative are indexical of two aspects of the same historical processes. Similarly, milimiligan and written histories are now intertwined as different parts of the same Paiwan history. Long-term memory likewise plays a significant role in Paiwan oral narratives. Maurice Bloch (1992) noticed that it is important to research both
Contesting Memory 193
the internal and external memories of human society. He suggested that, by researching both internal and external memory patterns, one can reveal different ways of understanding how the individual is connected with his society in history. Bloch suggests that folk theories of memory, by internal and external memory, can reveal the individual point of view and his or her position within his or her specific group or community. By doing so, researchers might grasp the complexity of how culture and history are embodied in the minds of humans, and how the cognition of the reality of the here and now is shaped. My own fieldwork in Puleti village suggests that statuses in Paiwan social hierarchy are never stable and that oral narratives exemplify universal contestation rather than partial unification. Like culture, history is a living thing, a process in itself. Temporary compromise and consensus over particular versions of history are sometime reached, but more often than not they only serve to fuel future disputes. For example, one marriage negotiation consisted of more than seven meetings between the families of the potential groom and bride, but without reaching a conclusion. Elders from both families are generally the central players on these occasions. However, as in this instance, written genealogies are now gradually being introduced into marriage negotiations. This provides fuel for the younger and better-educated generation to challenge the authority of the elders, who are now able to resort to the various written records and publications, and demand that the authority of these new sources of knowledge about the past be placed alongside the authority of oral traditions. Thus, Muni’s work is a ‘mission impossible’. The written history she produces, even though different from previous written histories in its attempt to write history without an agenda, will still become a part of this game once it enters circulation. Thus, Paiwan efforts to empower themselves by writing their own history before it is written by others embeds an irony. This means that the interplay between individuals in Paiwan society and history is essentially being ‘rewritten’.
Acknowledgements This research was initially supported by a Summer Fieldwork Grant from the Institute of Anthropology, National Tsing Hua University, Hsinchu, Taiwan, in 2002, and later by a Fulbright Doctoral Research Fellowship, and a National Science Council Research Grant from 2005 to 2006. I am very grateful to Chiang Bien, Ku Kun-hui, Lin Ying-chin, Wang Ming-ke, James Wilkerson and Michael Herzfeld for their suggestions during the process of writing. I also wish to acknowledge the helpful discussions after my presenting an earlier version of this essay at a Harvard Yenching Lunch Talk and at the Third Meeting of the European Association of Taiwan Studies. Special thanks go to Wang Cheng-hua and Stephan Feuchtwang, my discussants on these occasions. This paper could not have been revised without the assistance of Robert
194 Li-Ju Hong
Parkin, writer-in-residence at Tsing Hua University’s Writers’ Workshop in 2006. Lastly, I appreciate the aid of the villagers of Puleti and my informants, especially Muni, for valuable details. To safeguard people’s privacy, I have changed some place names and personal names.
Notes 1. Council of Indigenous Peoples: Paiwan, Introduction. http://www.apc.gov.tw/portal/ docList.html?CID=B68D98A9742D94C7 2. From author’s field notes, 13 August 2003. The documents Muni was referring to were from photocopies of household registrations from the Japanese colonial era. In this chapter, there are three important family names: Taljalub, Camak and Rovaniau, all belonging to the Qalangiyan genealogy. However, the names of these houses have been recorded using a variety of spellings, which can easily be confused. For example, Taljalub is sometimes recorded as Taljiyalep or Taljialep. Ruvaniaw has been written as Luvaniyau, Ruvangiyan, Ruvanuyeav or Ruvaneyeav. Alternate spellings for Camak include Camake, Camek and Camage. In this chapter, I will not discuss many details of the relationship between them, but focus on four different texts that refer to them. 3. Before the Japanese colonial authorities prohibited the practice in the 1920s, the Paiwan buried the deceased members of their families beneath the slate floors of their houses ( Jiang 1999). 4. For administrative purposes, the Japanese colonial government followed an earlier tripartite classification of the indigenous populations of Taiwan that had been devised by Qing officials. According to the degree of being ‘civilized’, Taiwan indigenous peoples as a whole were divided into (1) least civilized ‘raw aborigines’, (2) most civilized ‘cooked aborigines’ and (3) an intermediate classification of ‘acculturated aborigines’. 5. Kavulungan is claimed to be the origin of many Paiwan villages. In the northern region of Paiwan territory there are two mountains summits that the Paiwan consider the origin places of their legendary ancestors. The two mountain summits are approximately five kilometres apart from one another as the crow flies. The northern summit is commonly known as Kavulungan, while the southern summit is called Tjagaraus. The exact designations of the two names, however, are in some cases interchangeable and may differ from village to village.
References Bloch, Maurice. 1992. ‘Internal and External Memory: Different Ways of Being in History’, The Edward Westermarck Memorial Lecture, Suomen Antropologi 1(92): 3–15. Cheng Chilu. 1974. ‘Linshi Guanxi Diaochahui yu Taiwan gaoshanzu yanjiu’ [Committee on Customary Laws and Research on Taiwan Aborigines], Taiwan Fengwu 24(4): 9–24. Freeman, Derek. 1983. Margaret Mead and Samoa: The Making and Unmaking of an Anthropological Myth. Cambridge Mass. and London: Harvard University Press. Fujii, Shizue. 2001. Taiwan yuanzhumin shi: zhengze pian [The History of Taiwan Formosan Aborigines: Policy Formulation]. Nantou: Taiwansheng Wenxian Weiyuanhui. Goody, Jack. 1977. The Domestication of the Savage Mind. Cambridge: Cambridge University Press.
Contesting Memory 195
Halliday, M.A.K. 1989. Spoken and Written Language. Oxford: Oxford University Press. Hobsbawm, Eric J. 1997. On History. New York: New Press. Hu Tai-Li. 2004. ‘Paiwanzu xugou chuanshuo de zhenshi’ [The Reality of Paiwan Imaginary Legend]. In Pingdong chuantong yishu: Pingdongxian chuantong yishu yantaohui lunwenji [Traditional Arts: Traditional Arts Workshop Proceedings, Pingtung County]. Chinese Folk Arts Foundation, ed. Yilan: Center for Traditional Arts, pp. 157–90. Jiang Bin (Chiang Bien). 1983. ‘Paiwanzu guizu zhidu de zai tantao, yi Dashe weili’ [A New Approach to the Aristocratic System of the Paiwan], Zhongyang Yanjiuyuan Minzuxue Yanjiusuo Jikan 55: 1–48. ———. 1999. ‘Muzang yu ximing: Paiwanzu de liang ge jiyi jizhi’ [Burial and Naming: Two Memory Mechanisms of the Paiwan]. In Shijian, lishi yu jiyi [Time, History and Memory]. Yinggui Huang, ed. Nankang: Institute of Ethnology, Academia Sinica, pp. 381–415. Jiang Hai. 2000. Piaoliu liangqian nian, ruvaneyeav jiashi [Drifting for Two Thousand Years: The Ruvaneyeav Genealogy]. Pingtong: Pingtongxian Wenhuaju. Ku Kunhui. 2002. ‘Ming, wu yu qiexu: xi Paiwan piuma de anli’ [Name, Object and Hierarchy: An Example of West Paiwan Piuma]. Paper presented to the conference on Objects and Material Culture, Taizhong, June. Nankang: Institute of Ethnology, Academia Sinica. Lévi-Strauss, Claude. (1979) 1982. The Way of the Masks. Translated by Sylvia Modelski. Seattle: University of Washington Press. Liu Bingxiong. 1975. ‘Riben xueren zhi gaoshanzu yanjiu’ [ Japanese Scholars’ Research on the Aborigines in Taiwan], Zhongyang Yanjiuyuan Minzuxue Yanjiusuo Jikan 40: 5–18. Mead, Margaret. (1928) 1961. Coming of Age in Samoa. New York: Morrow Quill. Ong, Walter J. 1977. Interfaces of the Word: Studies in the Evolution of Consciousness and Culture. Ithaca: Cornell University Press. Pan Lifu. 1998. Kavulungan Paiwanzu wenming: yi ge meiyo wenzi jinqian de shehui [The Civilization of Kavulungan Paiwan People: A Society without a Written Language or a Monetary System]. Pingtong: Pingtongxian Wenhuaju. Rinji Taiwan Kyukan Chosakai. [1915–22] 2002. Banzoku Kanshu Chosa Hokokush [Research Report on the Customs of the Aborigines]. Kojima Yoshimichi and Kono Kiroku, eds. Taihoku: Taiwan Sotokufu Rinji Taiwan Kyukan Chosakai. Chinese Translation: Taiwan Zongdufu Linshi Taiwan Jiuguan Diaochahui [1915–22] 2002. Fanzu guanxi diaocha baogaoshu Vol. V [Research Report on the Customs of the Aborigines, Vol. V]. Bin Jiang, ed. Nankang: Institute of Ethnology, Academia Sinica. Shi Lei. 1971. Fawan: yi ge Paiwanzu buluo de minzuxue tianye diaocha baogao [Su-Paiwan: An Anthropological Investigation of a Paiwan Village]. Monograph No. 21. Nankang, Taipei: Institute of Ethnology, Academia Sinica. ———. 1976. Taiwan tuzhu xuzuxing qinshu zhidu: Lukai Paiwan Beinan san zuqun de bijiao yanjiu [Cognatic Kinship Systems among Formosan Aborigines: A Comparative Study of Rukai, Paiwan and Puyuma]. Monograph No. 23. Nankang, Taipei: Institute of Ethnology, Academia Sinica. Tong Chunfa. 2001. Taiwan yuanzhumin shi: Paiwanzhu shi pian [The History of Taiwan Indigenous Peoples: The Paiwan]. Nantou: Taiwansheng Wenxian Weiyuanhui. Valeri, Valerio. 1990. ‘Constitutive History: Genealogy and Narratives in the Legitimation of Hawaiian Kingship’. In Culture through Time. Emiko OhnukiTierney, ed. Stanford: Stanford University Press, pp. 154–92.
196 Li-Ju Hong
Wei Huilin. 1960. ‘Paiwanzu de congzu zuzhi u jieji zhidu’ [Lineage and Hierarchy of the Paiwan], Zhongyang Yanjiuyuan Minzuxue Yanjiusuo Jikan 9: 71–108. Zheng Zhengcheng. 2002a. ‘Linshi Taiwan Jiuguan Diaochahui chuchuang zhizi de chouhua yu yunzuo’ [Planning and Operations of the Commission on Taiwan Customary Laws at its Initial Stage], Taiwan Wenxian 53(1): 135–66. ———. 2002b. ‘Rizhi shiqi Linshi Taiwan Jiuguan Diaochahui de yuanzhumin diaocha 1909–1922’ [Report on the Aborigines from the Commission on Taiwan Customary Laws 1909–1922], Taiwan Wenxian 53(4): 25–48.
Chapter 9 Ethnicity as Strategy: Taiwan State Policies and the Thao Yayoi Mitsuda
Introduction Studies of ethnicity in multi-ethnic societies have been one of the central issues in anthropology and sociology since the 1960s. This has stimulated studies of indigenous people in Taiwan, which have been popular both academically and politically. The Thao were officially recognized as a discrete indigenous group in September 2001. Before then they were regarded as being part of the Tsou, another indigenous group in central Taiwan. The Taiwanese government had classified indigenous populations into nine groups, so the ‘newly’ recognized Thao people became the tenth. This was a welcome event for the Thao, and it also represented an important breakthrough for other ethnic groups who had been petitioning the government for recognition as indigenes. Indeed, the number of Taiwan indigenous groups officially recognized by the government has increased since 2001.1 A state’s policies have the power to determine the various identities of its citizens. A colonial government in particular can affect ethnic identities and relationships among ethnic groups through its strategic policies. Prior to 2001 the Taiwanese government had used a classification of indigenous populations based on one used during the Japanese colonial period in the fifty years after 1895. We therefore need to take the Japanese colonial government’s policies concerning the Thao into account to understand better how they came to obtain recognition by the Taiwanese government. The Thao are a very small group of approximately seven hundred people and are much more heavily sinicized than other indigenous groups in Taiwan.2 This chapter therefore addresses the following questions: How did the Thao achieve this recognition as a separate indigenous group? What strategies did they follow in doing so? And, how does their ethnicity interact with state policies? It first outlines how the Thao have been treated historically in the
198 Yayoi Mitsuda
classification of Taiwan indigenous groups, and the process of their achieving official government recognition in 2001. The following section presents a sketch of how the main Thao village has developed into a major tourist spot, and how Thao culture has been transformed into a tourist resource by both Taiwanese business people and the Thao themselves. Moreover, I ask how this situation has affected Thao people’s sense of their own ethnic identity.
The Process of Official Recognition of the Thao Non-committal Recognition of the Thao before 2001 Indigenous people in Taiwan are commonly understood to be Austronesianspeaking ethnic groups, but the problems of classifying them and of differentiating between mountain and plains peoples have generated considerable controversy. The numbers of subgroups identified and the ways of describing how they relate to each other differ according to historical period: Chinese Qing dynasty rule; the period of Japanese occupation; and the postSecond World War period. Under the Qing dynasty, Taiwanese indigenous people were classified into three groups according to their relations with the Chinese authorities: shengfan (‘raw/uncooked’ barbarians), huafan (‘semi-cooked’ barbarians), and shoufan (‘cooked/ripened’ barbarians). This last group referred to plains indigenes. The terms ‘Thao’ and ‘Sun Moon Lake’, the latter being where the Thao live, were not officially recorded during the Qing dynasty; the Qing government called this region Shuishalian and classified the indigenes who lived there as huafan. There are official records of the Shuishalian huafan because they had been paying taxes to the Qing government since quite early times. But it seems that the Qing government was not particularly interested in differentiating between ethnic groups; historical materials from this period only record that there were twenty-four she (settlements where the indigenous people lived) and make no mention of which indigenous groups these huafan consisted of (Chan 2000: 88–89). Kai Yiu Chan (2000), a historian, points out that the ‘Shuishalian’ region actually comprised a very large, complex area and could not contain just a single ethnic group. Chan also suggests that the Thao might be directly related to the indigenous populations in these twenty-four settlements, but is sceptical whether this group was identical to the present-day Thao. During the Japanese occupation, scholars such as Ryuzo Tori’i and Kanori Ino began to work on a more detailed classification of Taiwan indigenes. While they were more or less in agreement regarding some groups, such as the Atayal, Bunun, Paiwan and Amis, there were some discrepancies in their respective accounts of how the Thao should be characterized. For example, in the early Japanese colonial period the Thao and the Saisiyat were both generally classified as kaban ( Japanese term for huafan). The Saisiyat started to be regarded as seiban ( Japanese term for shengfan) in 1911,3 but the Thao
Ethnicity as Strategy: Taiwan State Policies and the Thao 199
continued to be classified as kaban (Suzuki 1932; Fujisaki 1930). Other scholars at that time, however, such as Ino and Ushinosuke Mori, regarded the Thao as part of the Bunun or Tsou rather than as kaban (Ino 1904; Mori 1918). In 1913 the Japanese government adopted a classification containing just seven seiban groups, based on Mori’s classification.4 This became the official standard and lasted until the end of the Japanese colonial period. Mori regarded the Thao as a subgroup of the Tsou, but the Japanese government did not adopt this categorization, classifying the Thao as a jukuban ( Japanese term for shoufan) group rather than as a part of the Tsou. Another influential classification was made by Nenozo Utsushikawa et al. (1935), who classified the seiban into nine groups. Utsushikawa et al. were basically researching the seiban and seem to have considered the Thao as a jukuban group.5 After the Second World War, the Department of Archaeology and Anthropology at National Taiwan University (formerly the Taihoku Imperial University) classified indigenes into nine groups following Utsushikawa et al., and gave the first listing of Chinese and English names corresponding to the Japanese names of different Taiwan indigenes in 1953 (Ruey 1953: 37–38). The Taiwanese government also adopted this nine-group classification as its official classification the following year.6 Jukuban groups were considered to have been assimilated, but the Thao were not included in this category. Strange to say, at that time the Thao were registered as the Tsou, a classification that lasted until 2001. Some Taiwanese scholars suggested that the Thao should be recognized as a separate group (e.g. Chen et al. 1955), but the Kuomintang government did not pay them much attention. On the whole, therefore, the definition and recognition of the Thao has historically been a very grey area and a matter of some confusion.
Government Recognition in 2001 The late 1990s saw a shift towards democracy and an independence movement in Taiwan, and the government began to pay attention to issues related to Taiwan’s indigenous population. The government gradually began to accept academic opinion on the Thao that it had previously mostly ignored – opinion supporting their status as a separate indigenous group – and began to regard the Thao in this way.7 In this context there were two important factors that led the Thao to success in achieving government recognition. The first factor was the catastrophic earthquake measuring 7.6 on the Richter scale that hit Taiwan on 21 September 1999. The epicentre of the earthquake – known in Taiwan as ‘9/21’ because of the date on which it occurred – was Jiji Township, just some 12.5 km west of Sun Moon Lake, where the Thao live.8 The damage caused by the 9/21 earthquake was extremely severe, and areas near the epicentre became the focus of public attention. Sun Moon Lake had become one of Taiwan’s most famous tourist spots, so the
200 Yayoi Mitsuda
media were quick to report any damage there. Moreover, the government strongly supported post-earthquake reconstruction in this area to restore the tourist industry. The process of reconstruction of the houses and daily lives of the Thao received widespread attention, giving rise to the impression that their rituals and culture were peculiar to them. The effect was that after the earthquake the Thao began to be seen widely as a distinct indigenous group, not only by scholars, but also by the general public. The second crucial factor was the general election held in March 2000. The Kuomintang had been the ruling party in Taiwan in the fifty-five years since the Japanese colonial government evacuated the island in 1945. In 2000, however, Chen Shui-bian of the Democratic Progressive Party (DPP) won the presidential election, leading the first ever peaceful democratic transition of power to an opposition party in Taiwan’s history, thereby marking a decisive end to the Kuomintang’s monopoly of the central government. The DPP policies which differentiated them from the Kuomintang included a strong focus on social welfare involving the rights of subaltern groups in society, such as women, children, labour and indigenous people. The DPP has consistently been associated with supporting rural education and the protection of ‘local’ languages and cultures.9 It was in these political circumstances that the Thao became the tenth indigenous group: in 2001 the DPP government modified the former classification of ‘nine indigenous groups’ that had lasted for more than fifty years. Thus the period from 1999 to 2000 was a turning point for the Thao: the 9/21 earthquake acted as a catalyst in spreading academic recognition of the Thao as a separate indigenous group from academic circles to the general public; moreover the shift of government from the Kuomintang to the DPP made official recognition of the Thao much easier. How, then, did the Thao react to this change in their circumstances?
Survival Strategies of the Thao The Thao Village Ita Thao,10 essentially the only Thao village,11 is located on the shore of Sun Moon Lake in Nantou County in central Taiwan. Sun Moon Lake is the largest lake in Taiwan and has been a famous tourist spot since the Japanese colonial period. When the Japanese began to develop the tourist industry there, they isolated the Thao so that the latter could not move away from the lake, and other people (e.g. Han) could not move in (Deng 2000: 152). This policy at least guaranteed the Thao their own space and way of life, allowing them to maintain their traditional customs, such as music, dance and the traditional dugout canoe; these became the focus of tourist interest and ironically caused a rapid shift from a traditional barter economy to a monetary economy. During the period of Kuomintang government, a large number of Chinesespeakers moved to Ita Thao to profit from the tourist industry, and the Thao
Ethnicity as Strategy: Taiwan State Policies and the Thao 201
have since become a minority in their own village. Moreover, the economic position of the Thao has declined because such newly arrived residents had and continue to have large amounts of capital to invest and are generally good at business. As the Chinese-speaking population increased, the distinctive Thao way of life receded. Even the name ‘Thao’ has now come to be used by both Han Chinese and the Thao themselves as a way of selling souvenirs to tourists.
Thao Ethnicity As outlined above, tourists have been consumers of the culture and tradition of the Thao since the Japanese colonial period, to the extent that the Thao have now become a minority in their own village. The question arises as to whether this situation has affected their own sense of their ethnic identity. How do Thao people now identify themselves as ‘the Thao’? When I was staying at Ita Thao, I often heard the phrase, ‘Guan si huan-a’ (‘We are barbarians’).12 There are far more polite words meaning ‘indigenous people’ in Taiwanese, and in other regions where indigenes live huan-a might be considered a very rude word, but the Thao themselves tend to use it, meaning ‘barbarian’. Hsieh Shih-Chung (1987) points out that many Taiwanese indigenous people have a ‘stigmatized identity’, meaning that they have adapted outsiders’ stereotypes of themselves. Interestingly, however, although the Thao call themselves huan-a, they do not have a negative self-identity, nor do they try to hide their ethnic background. The word huan-a is an important code for the Thao to differentiate between themselves and other Taiwanese or lang.13 When they use the words huan-a and lang, Thao are conveying something quite different: they feel their ethnic identity as something quite positive. This may be because they have been demonstrating their indigenousness to tourists for a long time. The ‘Thao culture’ that has been presented to tourists for them to consume seems to form the basis of Thao ethnicity; its authenticity is, however, rather dubious. The ‘traditional Thao costumes’ sold at Sun Moon Lake and the dances performed there are not actually part of Thao traditional culture, nor are the dancers always Thao. A number of local people, Thao and other indigenous people, told me that the dances have been borrowed from other indigenous groups, such as the Amis. Certainly the actual life of the Thao has long ceased to be the same as their ‘traditional’ lifestyle. They have therefore been open to incentives to borrow from other cultures in order to be able to perform for tourists. A 1998 ‘Writing the History of Villages’ programme, supported by the Community Empowering Society (CES) and the Taiwan Provincial Government, served to strengthen the Thao’s sense of solidarity ( Jian Shi-lang 2000). The aim of this programme was to collect and reconstruct collective memories in the local area, and Ita Thao was one of ten villages selected (Chen
202 Yayoi Mitsuda
Ban 1998). Through this programme, a ‘Thao Team’ was organized which actively collected early photographs and other data on Thao traditional life. In June 1999, the Thao Team was reorganized into the Thao Tribal Culture Development Association, the first organization of the entire Thao tribe ( Jian 2000). The association became the main unit for the Thao to talk to the media, the government and all other organizations from the outside. One of the association’s main goals was to achieve government recognition of the Thao as a separate indigenous group. Another goal of the association was and remains the recovery of the traditional culture of the Thao.
Strategies of the Thao The association has been working on ‘the reconstruction of Thao traditional culture’ since 2001. The strategic approach has had two main objects: revitalizing the Thao language, and Thao ritual performance. I first stayed at Ita Thao in 2001, and have visited every year since, spending every summer vacation there. I have hardly heard anyone speak Thao in daily life. According to a survey conducted by Lin Xiu-che in 2001, only 8 per cent of Thao had a ‘fluent’ or ‘ordinary’ ability to speak the Thao language, and as many as 69 per cent ‘cannot speak’ the language (Lin 2001: 39). This situation is quite different from that of other indigenous groups, where surveys have found that more than 50 per cent of young people communicate with the older generation in their mother tongue.14 With regard to the proportion of the general Thao population who are not able to speak their mother tongue, we need to pay attention to the remarkable population increase of Thao people in the past decade: there were only 263 Thao people in 1999 when Lin Xiu-che drew up a summary report (ibid.: 38), but their population had already increased to 520 by June 2004.15 In other words, the population had doubled in just three years. The reason that the population increased in such a short period was that many people went to register as Thao for welfare benefits.16 For example, when an indigenous woman marries a Taiwanese man, the children generally take the father’s Taiwanese surname and do not register as indigenes; but in this period many of these children expressly changed their surnames to their mothers’ and registered themselves as Thao. Generally speaking, most of them are not living in the village, nor have they ever spoken Thao in their lives. Thus, as the population of the Thao increases, the percentage of the overall population who can speak the language decreases. Given that many elderly people have passed away during the years since, the percentage of fluent speakers of the Thao language must be much lower than 8 per cent now. The older members of the Thao community and the association members feel a sense of crisis at seeing their language disappearing. The association has accordingly held classes in the Thao language a few times, but without much success. When I interviewed some of the younger people, most of them said,
Ethnicity as Strategy: Taiwan State Policies and the Thao 203
‘How can the Thao language be useful? We are so few people’. As a result, all the Thao in Ita Thao are very fluent speakers of Taiwanese, and most of them do not use any Thao words in their daily lives. Most of the younger generation do not even know how to address their father and mother in Thao. Although most Thao can hardly find any motivation to learn their own language, they still understand the importance of their mother tongue when faced with people from outside, especially government officials. In 2001, Chen Shui-bian, then the President of Taiwan, came to Sun Moon Lake to announce that the Thao had been officially recognized as the tenth indigenous group. The Thao had prepared a song in the Thao language for the occasion and practised it under the guidance of an elder who spoke the Thao language very well. I witnessed the ceremony and stayed around for a while afterwards: I observed that soon after the ceremony, Thao people I spoke to hardly remembered a single phrase of the song. They simply seemed not to care about it anymore. When scholars try to classify ethnic groups or when people identify their own group, a group’s mother tongue is one of the most important markers: language embodies many factors of a culture such as classification and cultural ideas. The Thao language, however, no longer works as an inside marker for the Thao; they do not speak the language very well, and for most of them it has just become a performative instrument to show to outsiders. Secondly, the most obvious aspect of Thao culture is its traditional religion and rituals. Most Thao follow a combination of Buddhism, Confucianism and Taoism, like almost all Taiwanese, and a few households are Christian. This is quite different from other indigenous groups in Taiwan, who generally follow Christianity. Although the Thao have been deeply affected by Taiwanese religion,17 almost all of them still believe in their traditional religion at the same time. Thao traditional religious rituals are mostly put in Thao shamans’ hands, as described below. The most obvious artefacts of the traditional Thao religion are the ulalaluan and the shinshi. The ulalaluan is a kind of a bamboo basket, which in general every household has. The Thao put the clothes and ornaments of their ancestors in the ulalaluan and worship it as a symbol of the spirits of the Thao who have passed away. But the ulalaluan is not only a religious object, it is also something which signifies that a person is Thao. This was clearly demonstrated when the Thao built forty temporary dwellings for themselves after the 9/21 earthquake, and the association decided that only those households that had an ulalaluan would be entitled to move into the temporary housing. In other words, having an ulalaluan is the most important marker indicating a Thao household. A shinshi is a female ‘shaman’ who alone can serve and console the spirits of the Thao.18 Since the ulalaluan is used in almost every ritual, Thao bring their ulalaluan to any place where a ritual will be conducted. The shinshi collects all the ulalaluan from the Thao, sets them out properly and starts praying to them. Generally, the only people who may touch anything in the ulalaluan are the
204 Yayoi Mitsuda
shinshi and adult Thao; although the rituals are often conducted in public, children, guests and tourists cannot touch an ulalaluan. Almost all Thao rituals are performed in the morning, and those who bring the ulalaluan are mostly housewives. However, the owner of an ulalaluan is the male head of the family, since Thao society is patrilineal. While today the shinshi mainly act as spirit mediums and no longer claim to have the ability to treat illness, expel evil spirits or stop the rain – abilities that earlier shinshi were said to have had (Tang 1996) – the shinshi still play quite an important role in Thao people’s religious life. Moreover, media reports on the Thao have increased since the 9/21 earthquake, and their rituals have become one of the most important foci of outside attention. The rituals are the most – perhaps even the only – outstanding and distinguishing performances given by present-day Thao; the shinshi, who perform the rituals, have readily attracted considerable attention, and all five current shinshi have become important informants for both the media and academic scholars. The change in the circumstances surrounding the shinshi has caused other changes within Thao society. Most of the older men think the shinshi have been paid too much attention recently and that this has made them become quite arrogant and neglectful in paying the respect they owe such men. The association, however, generally promotes the shinshi and the rituals they perform to attract media and government attention. The association invites the media when important rituals are held. As a result, the rituals have become rather showy: the association often has to give the shinshi new costumes, and the cost of performing the rituals has been increasing. In recent years, the association has been cooperating with the Sun Moon Lake National Scenic Area to perform the Thao New Year ritual for visiting tourists. The eyes of outsiders have become very important for both the Thao and the association, and they have started to care about how their rituals can be used to attract more media and scholarly interest, and how financial aid can be obtained from the government. Local rituals have therefore now become a big theatrical event where the shinshi performs for the tourists, media and officials, as well as for the Thao people themselves.
Conclusion With the exception of religious rituals, other aspects of traditional culture, such as the language, costumes and arts of the Thao, have almost been lost. Yet the Thao successfully achieved official recognition by the government, using their strategies very effectively. Until the 1980s the only reason why the Thao could not be added to the list of Taiwanese indigenous groups was that they were ‘too small a group’ (Hsieh 1987: 110).19 The Thao, however, effectively transformed this negative categorization into that of ‘an endangered group facing extinction’ through the experience of the 9/21 earthquake, thereby attracting the attention of the
Ethnicity as Strategy: Taiwan State Policies and the Thao 205
general public. ‘Survival’ itself became one of their most important survival strategies. The population of the Thao has actually increased markedly since their official recognition, but this fact is seldom mentioned by the Thao, who prefer to continue being considered ‘an endangered group’. According to the direction the Taiwanese government has taken in indigenous affairs so far, it is undeniable that they tend to essentialize in how they recognize indigenes. The government clearly expresses the fact that ‘objective conditions’ may be taken seriously. Thus, the people who think they are indigenous are required to prove their uniqueness with evidence such as their language, religion, customs or other cultural features marking their difference from the Han Taiwanese.20 And in spite of these policies, ironically the first group the government recognized as an honourable tenth was the Thao, who were one of the most sinicized indigenes. The generally recognized negative features, such as ‘small population’, ‘dying mother tongue’ or ‘sinicized lifestyle’, are somehow working very well as a strategy for the Thao to argue their uniqueness. Although Thao people are not superior to other indigenous groups with these objective conditions, they have quite a strong sense of their own ethnicity as huan-a, and their long period of using their indigineity to appeal to tourists has both shaped and reinforced this. Therefore, their recent plan for the ‘reconstruction’ of their ‘traditional’ culture was intended not only to strengthen what has come to constitute the basis of their ethnic identity, but to perform it for the outside world. Barth (1969) considers that recognizing the existence of ethnic boundaries and determining how a boundary has been maintained is more important than investigating the internal constitution of the ethnic group in question. The way the Thao sustain, and exploit, the religious world symbolized by the ulalaluan and the shinshi effectively stresses the boundary between themselves and the Tsou or other Taiwanese. As Thomas Eriksen has said, identities are ambiguous, and this ambiguity is connected with a negotiable history and a negotiable cultural content (Eriksen 1993: 73). In the case of the Thao, their ethnicity has long had a performative dimension, and they are accustomed to displaying and performing their indigeneity to the outside world. This might be one reason why the Thao are so ready to cooperate with the National Scenic Area in demonstrating their rituals. The ethnicity of the Thao is not simply based on their ‘traditional’ culture: their performances for others constitute an essential element, and their ethnicity as ‘a small group’ is also emphasized very strategically.
Notes 1. The following year the Kavalan became the eleventh, and by the end of 2010 there were fourteen recognized indigenous groups in Taiwan. 2. Executive Yuan Council of Indigenous Peoples, March 2011 census figures. http://www. apc.gov.tw/portal/docDetail.html?CID=940F9579765AC6A0&DID=0C3331F0EBD31 8C2D84A7D3680077179, accessed 16 April 2011.
206 Yayoi Mitsuda
3. The Bureau of Aboriginal Affairs published a compact English pamphlet, Report on the Control of the Aborigines in Formosa, in 1911. In it, the Japanese colonial government divided ‘the savages in Formosa’ into nine groups which followed Ino’s classification (Bureau of Aboriginal Affairs 1911). The Saisiyat were classified as seiban for the first time in this pamphlet. 4. Mori actually had a ‘six-group classification’; he did not regard the Saisiyat as a seiban group. The Japanese government, however, added the Saisiyat to Mori’s classification, making a seven-group classification. The reason why Mori regarded the Thao as a subgroup of the Tsou was not clearly recorded, but Thao’s famous origin legend of ‘chasing a white deer’ tells that the Thao came from Mt. Ali where the Tsou live; the similarity between the names ‘Thao’ and ‘Tsou’ also might have been an important factor influencing the Mori classification. 5. Two volumes were published by Taihoku Imperial University in 1935. Utsushikawa et al.’s book, Taiwan takasagozoku keito shozoku no kenkyu (The Formosan Native Tribes: A Genealogical and Classificatory Study) was one of them. It describes all nine seiban groups in great detail. In another volume, Gengo ni yoru Taiwan takasagozoku densetsushu (The Myths and Traditions of the Formosan Native Tribes (texts and notes)), published by the Institute of Linguistics the same year, the Thao were classified as a distinct jukuban group for linguistic reasons. 6. Prior to 1954, the Kuomintang government had not had much time to determine its indigenous policies, and therefore the Taiwan indigenes remained divided into the seven groups of the Japanese classification (see Wang 1967). 7. For example, the textbooks for elementary education (國民小學原住民鄉土文化教材) (1995) edited by the Ministry of Education, and The History of Formosan Aborigines (台灣 原住民史) (2000) edited by the Historical Research Commission of Taiwan Province, treated the Thao as one of Taiwan’s indigenous peoples. 8. http://en.wikipedia.org/wiki/921_earthquake, accessed 26 August 2006. 9. For the programmes supported by the DPP, see http://newcongress.yam.org.tw/dpp/ action.html, accessed 26 August 2006. 10. The official place name of this area is Riyue-cun (日月村) or Dehua-she (德化社), but in recent years the Thao have started to call their village Ita Thao. Ita means ‘we’, and thao means ‘people’. 11. There is another village where some Thao live in Shuili, the neighbouring township area, but its inhabitants are very few in number and hardly have Thao ethnicity. Although officially recognized and also registered as Thao, they have lost their traditional religious practices. Because their traditional religion is the core element that defines people as Thao, most of the Thao in Shuili lack a strong sense of ethnic identity. I will therefore not discuss them further here. 12. Currently, Thao people speak Taiwanese in their daily life; this phrase is also Taiwanese. 13. This word means ‘human’ in Taiwanese and is used to refer to Chinese-speakers. 14. http://www.apc.gov.tw/upload/publish/monthly/第59期/p10.htm, accessed 26 August 2006. 15. Executive Yuan Council of Indigenous Peoples, June 2004 census figures. http://www. apc.gov.tw/portal/docDetail.html?CID=940F9579765AC6A0&DID=3E651750B40064 6722F14D1141CE39D0, accessed 17 April 2011. 16. In Taiwan, government social policies provide indigenous people with benefits in university entrance examinations, scholarships, examinations for public employment, and so on. 17. Most Taiwanese are adherents of a combination of Buddhism, Confucianism and Taoism. There are only four families who follow Christianity among the Thao: others keep ancestors’ tablets or statues of Taiwanese deities, like most Taiwanese. 18. The number of shinshi is generally maintained at around five to seven. When any of them dies, a new one may be recruited. There were five shinshis at the end of 2009.
Ethnicity as Strategy: Taiwan State Policies and the Thao 207
19. According to Chen Chi-lu, at that time the population of the Thao was under 200 (Chen et al. 1996). 20. For example, these conditions are seen in Article 3 of Yuanzhuminzu rending fa (原住民 族認定法, ‘Recognition Act for the Indigenous Peoples’, draft).
References Barth, Fredrik, ed. 1969. Ethnic Groups and Boundaries: The Social Organization of Culture Difference. Boston: Little, Brown and Company. Bureau of Aboriginal Affairs, The 臺灣総督府民政部蕃務本署. 1911. Report on the Control of the Aborigines in Formosa. Taihoku: Bureau of Aboriginal Affairs, Government of Formosa. Chan, Kai Yiu 陳計堯. 2000. ‘Shilun riyuetan diqu yuanzhumin de lishi qianyi’ 試論日 月潭地區原住民的歷史遷移 (A History of Aboriginal Migration in the Sun Moon Lake Region, 1815–1934). Taiwan Historical Research 臺灣史研究 7(1): 81–134. Chen Ban, ed. 陳板 主編. 1998. Dajia lai xie cunshi 大家來寫村史 (Writing the History of Villages). Nantou: Culture and Education Division, Taiwan Provincial Government 台灣省政府文化處. Chen Chi-lu et al. 陳奇祿等. 1955. ‘Riyuetan shaozu minzuxue diaocha chubu baogao’ 日月潭邵族民族學調查初步報告 (Ethnological Primary Researches among the Thao of Sun Moon Lake, Formosa). Bulletin of the Department of Anthropology 考古 人類學刊 6: 26–33. Deng Shian-yang 鄧相揚. 2000. Riyuetan shi hua 日月潭史話 (History of the Sun Moon Lake). Nantou: Sun Moon Lake National Science Area 交通部觀光局日月 潭國家風景區管理處. Deng Shian-yang and Mu-tsu Hsu 鄧相揚與許木柱. 2000. Taiwan yuanzhumin shi: shaozu shipian 台灣原住民史:邵族史篇 (The History of Formosan Aborigines: The Thao). Nantou: Historical Research Commission of Taiwan Province 台灣省 文獻委員會. Eriksen, Thomas Hylland. 1993. Ethnicity and Nationalism: Anthropological Perspectives. London: Pluto Press. Fujisaki Seinosuke 藤崎 濟之助. (1930) 1931. Taiwan no banzoku 台湾の蕃族 (Barbarians in Taiwan). Taipei: SMC publishing 南天書局. Hsieh Shih-Chung 謝世忠. 1987. Rentong de wuming: Taiwan yuanzhumin de zuqun bianqian 認同的污名:台灣原住民的族群變遷 (Stigmatized Identity: A Study on Ethnic Change of Taiwan Aborigines). Taipei: Independent Evening Newspapers Press 自立晚報社. Ino Kanori 伊能 嘉矩. 1904. Taiwan banseishi 台湾蕃政志 (History of Indigenous Policies in Taiwan). Taipei: Taiwan Sotokufu 台湾総督府. Jian Shi-lang 簡史朗. 2000. ‘Nantouxian riyuecun cunshi de caozuo moshi, guocheng yu kunjing’ 南投縣日月村村史的操作模式、過程與困境 (The Mode of Operation, Process and Difficulties of Recording History in Riyue Village, Nantou). In Cunshi yundong de mengya 村史運動的萌芽 (The Beginnings of the Village History Movement). Wu Mi-cha et al. eds. 吳密察等編 Taipei: Tangshan 唐山. Lin Xiu-che 林修澈. 2001. Yuanzhumin de minzu rending 原住民的民族認定 (Ethnic Recogniton of Indigenous People). Taipei: Council of Indigenous Peoples, Executive Yuan 行政院原住民族委員會. Ministry of Education 教育部. 1995. Guomin xiaosue yuanzhumin xiangtu wenhua jiaocai: shaozu 國民小學原住民鄉土文化教材:邵族 (Elementary School Indigenes Rural Culture Teaching Material: The Thao). Taipei: Ministry of Education 教育部.
208 Yayoi Mitsuda
Mori Ushinosuke 森 丑之助. (1912) 1918. Taiwan banzoku zufu 台湾蕃族図譜 (Pictures of Taiwan Indigenes). Rinji Taiwan kyukan chosakai 臨時台湾旧慣調査会. Ruey Yih-fu 芮逸夫. 1953. ‘Kaogu renlei xuejie xiaoxi: Benxihua Taiwan tuzhu gezu zhongxiwen mingcheng’ 考古人類學界消息 本系劃-臺灣土著各族中西文名稱 (News: Chinese and English Versions of Names of Taiwan Aborigines). Bulletin of the Department of Archaeology and Anthropology 1: 37–38. Suzuki Sakutaro 鈴木 作太郎. 1932. Taiwan no banzoku kenkyu 台湾の蕃族研究 (A Study of Taiwan Barbarians). Taipei: Taiwan Shiseki kankoukai 台湾史籍刊行会. Tang Mei-chun 唐美君. (1958) 1996. ‘Riyuetan shaozu de zongjiao’ 日月潭邵族的宗 教 (Religion). In Riyuetan shaozu diaocha baogao 日月潭邵族調查報告 (Ethnological Researches among the Thao of Sun Moon Lake, Formosa). Chen Chi-lu et al. 陳奇 祿等. Taipei: SMC Publishing 南天書局. Utsushikawa Nenozo et al. 移川 子之藏. 1935. Taiwan takasago-zoku keitou shozoku no kenkyu 台湾高砂族系統所属の研究 (The Formosan Native Tribes: A Genealogical and Classificatory Study). Taipei: Institute of Ethnology, Taihoku Imperial University 台北帝国大学土俗人種学研究室. Wang Jen-Ying 王人英. 1967. Taiwan gaoshanzu de renkou bianqian 臺灣高山族的人口 變遷 (Change of Formosan Aborigines). Zhongyang yanjiuyuan minzuxue yanjiusuo zhuankan 中央研究院民族學研究所專刊之十一. Monographs No. 11. Taipei: Institute of Ethnology, Academia Sinica 中央研究院民族學研究所.
Chapter 10 On the ‘Third Morning’: The Continuity of Life from Past to Present among the Nung of Northern Vietnam N. Jenny Hsu
Introduction This chapter describes the sham-ne (literally ‘third morning’) rite, a ritual observance of the Nung people celebrating the birth of a couple’s newly born first child. This observance is ideally held on the third day after the birth, from which the name ‘third morning’ comes, but the exact day may be modified to accommodate people’s work schedules and holidays, which follow the state calendar. Nevertheless, indispensable aspects of this ritual, such as who brings what and who hangs up the va where, have remained constant, generation after generation. I argue that this practice, carried on for at least a hundred years, demonstrates aspects of how the Nung continue to live their lives in ways that do not, and are unlikely to, fluctuate with the changing social environment. On the third day after a couple’s first baby is born, the household of the newborn baby’s father invites other households to the ‘third morning’ rite: the new baby is introduced to all the households, and each person receives a new kinship term in relation to that baby.1 This ritual observance only occurs once in each generation, when the first child is born. The va (see below) is also set up for the first child of a generation. For subsequent children, the household does not hold a feast, but its members pour wine, light incense and place a whole cooked chicken on the va. Some said that the household can still hold a ritual observance, but not as extravagant as the one for the first-born. If the first child has been difficult to raise, two more paper flowers are placed on the va for the second child and so on. A most important aspect of this rite is gift exchange or au lam (literally ‘bring the basket’), involving certain items being loaded into a woven bamboo basket and hung over a carrying pole. These baskets are brought by households related to the maternal side or bung lang, ‘the side at the back’ (see Figure 10.1).
210 N. Jenny Hsu
Moreover, the household of the maternal grandparents of the newborn arrives with the va, a bamboo-woven platform (see Figure 10.2), which is central to their belief in a female goddess me mu (or va mu).2 Each household carries out different duties determined by how they are related to the baby.
Figure 10.1. Baskets are brought by households related to the maternal side or bung lang (Note: Circle=female, square=male, diamond=both. All shaded in grey are the bung lang of the female in hash)
Figure 10.2. The maternal grandparents of the newborn arrive with the va, a bamboo-woven platform (28 May 2007, Dai An commune, Lang Son province, photo by the author)
The Continuity of Life from Past to Present among the Nung of North Vietnam 211
The Nung discussed here reside in several hamlets in Dai An commune, Lang Son province.3 The Nung were officially recognized as an ethnic group by the Socialist Republic of Vietnam in the late 1970s, but references to them can be traced back to at least Qing dynasty documents.4 They are believed to have migrated from the present-day Guangxi province in southern China about two hundred years ago. Their ancestors are said to have cultivated rice and other crops for self-sustenance, and to have brought anise trees to plant and products to sell for financial profit, activities they still practise today. Local Nung have experienced different phases of change, connecting their relatively remote world to a more global environment. In the hamlets where I did my research, an electricity network was set up in the late 1990s, enabling every household to have lights at night, a television and even a telephone. Anise is still a major economic resource, but locals complain how little profit they make from selling it to China, where it is made into medicinal oil. Rice is cultivated mostly for a household’s own consumption; it is rarely sold in the market as a commodity. Besides working in the fields, people also find other means of earning a living. Men and women, usually those in their forties, may sign up for a political position in the local echelon. High-school education is mandatory: younger people in their teens or twenties have started to migrate to other cities to live, study, perhaps join the army, or work wherever work is available to support their families, whether in the provincial cities or near the Chinese border, and usually in the telecom or construction industries. Within this social context, which differs from that experienced decades ago, the Nung continue to hold life-cycle rites and ritual observances for births, marriages and funerals, as well as other celebrations for certain events through the course of the year. The Nung celebrate their events regularly, every month from the first until the eleventh lunar month. The Lunar New Year is one of the important holidays. During the second lunar month, when not much daily work is performed as it is the beginning of a New Year, rites are performed by different ritual specialists. A ritual may also be performed when a misfortune befalls someone, regardless of what time of the year it is – for example, a funeral. The ‘third morning’, a birth-related rite, is an example of the first category of rite. This chapter will show that, while the Nung adjust their lifestyle to adapt to the present, they continue to maintain important aspects of their lives, such as the ritual observation of ‘the third morning’, from the past.
Who the Nung Are The Nung in present-day Vietnam speak one of the Central Tai languages, one which belongs to the Tai-Kadai language family (Lebar 1964; Li 2002). Other groups who speak similar languages to the Nung are the Tay in northern Vietnam and the Zhuang in Guangxi. They are renowned especially as skilled
212 N. Jenny Hsu
cultivators. Their traditional houses are propped up on stilts; nowadays, due to the scarcity of construction material, people building a Nung house retain the basic layout but omit the stilts. The 1999 census records the total Nung population in Vietnam as 856,412. This was about 8 per cent of the total national minority population and 1.1 per cent of the entire population. The Vietnamese government officially declared the Nung to be one of fiftyfour ethnic groups in 1979 and took measures to preserve their indigenous language (Hà and Lã 1984; Hoàng 1992; Pelley 1998). The Nung in Vietnam live along the present border with China, and their ancestry can be traced back to Zhuang ethnic groups in the south of China’s Guangxi province (Barlow 1987; Nong 2005; Teng 2002). According to the Summer Institute of Linguistics (SIL), different ethnic names exist for Nung in Vietnam, such as Nong, Bu-Nong, Highland Nung, Tai Nung, Tay and Tay Nung. Apart from having various names, the Nung in Vietnam speak different dialects, the names of which correspond to their autonyms. These names or references to their various dialects may reflect their different locations of origin in southern China (Fan 2005; LeBar et al. 1964). Lebar et al. (1964: 236) found that ‘some of the subgroups of the Nung in Vietnam are named for their places of origin in China, e.g. the Nung Inh, Nung Loi, Nung An, and Nung Chan’. More recently, Fan (2005) points out that: ‘names of the Nung subgroups correlate to where they come from in China. For instance, the Nung Inh migrated from Longyingzhou (龍英州). Longyingdong (龍英峒) was established during the Song dynasty (around the twelfth and thirteenth centuries), re-named Longying Tuzhou (龍英土州) during the Yuan, and retained the name Longying Tuzhou in the Ming and Qing. It is located in present-day Daxingxian (大新縣) in Guangxi’ (Fan 2005: 120–21). When the Nung gradually settled in Vietnam, it was mainly in Cao Bang and Lang Son provinces in north-east Vietnam. At different periods, especially during the war with America, some migrated to or were resettled in Bac Giang, another north-eastern province, and Bac Ninha, a Red River Delta province, or further south to Lam Dong, a Central Highlands province, and Ho Chi Minh city (Saigon), or even abroad.
The Nung, their Hamlets and their Lives I carried out my fieldwork in three hamlets in Lang Son province: ‘the hamlet on the mountain foot’, ‘the hamlet next to the bamboo’ and ‘the hamlet on the step’, roughly located at the base of the surrounding mountains.5 The main road, passable by motorcycles and small cars, leads to the village centre and is accessible by each hamlet. The quickest route to the outdoor market where produce and daily necessities are purchased is to climb the mountain on foot, which takes around thirty to forty minutes for an average adult.
The Continuity of Life from Past to Present among the Nung of North Vietnam 213
The locals distinguish between the hot and cold months. The temperature in the cold months – usually from November to February – ranges from 8°C at night to about 15°C during the day. In the hot months, spanning June to October, the temperature rises to 30°C or more under the sun, cooling down to around 18°C at night. Rain in this area mainly falls during April and May, when the first season for rice cultivation begins. Rice is the main cultivated product, especially on wet paddies; dry fields are planted with corn, beans, peanuts, and so on. This economic crop is harvested twice a year, from the sixth to the eighth lunar months and in the eleventh and twelfth lunar months. Anise grown on trees is more productive and produces more oil in winter. The trees may grow up to five metres, and the locals have to wrap their feet around the trunk to climb up to the branches, on which they perch to pick the star anise.6 Who the Nung can marry is based on asymmetric affinal alliance. The wifegivers and wife-takers have specific roles. The Nung nowadays often marry in their mid- or late-twenties; some think it is more important for a son to have a decent income before settling down with a wife. A Nung does not necessarily have to marry another Nung; the younger generation especially, who are more exposed to life outside the hamlets, might marry a non-Nung as long as the two people are ‘matched in their destiny’, a matter ascertained by consulting a Taoist priest.7 After their wedding, the young couple usually lives with the husband’s parents. If the couple does not yet have a child, it is possible for the wife not to live with her husband’s family every day. The wife stays at her husband’s house, helps out and tries to integrate, but she can visit her natal household for one day once every two weeks or so, or even stay over for a couple of days. Sometimes, if the husband is away for a long period, his wife may stay with her own parents most of the time. The young couple’s first child, however, must be born at the wife-taker’s household, or at the very least, the ritual observance of the ‘third morning’ for every first child must be held there.
The Ritual Observance of the ‘Third Morning’ The focus of this chapter is on the ‘third morning’, a ritual observance that is held around the third day after a couple’s first child is born. A ritual specialist does not have to be present, but the maternal grandparents and the households of the extended family on the maternal side, comprising the so-called bung lang, need to ‘bring the basket’. The maternal grandparents also prepare the va and an extra basket containing zok va, paper-cut flowers, representing the child being born (see Figure 10.3). Zok refers to the flowers to be placed on the va. Paper flowers are also made and used in rituals, especially those for treating the sick. They are referred to as zok mu in these rituals, since they are flowers to be presented to me mu.8
214 N. Jenny Hsu
Figure 10.3. The maternal grandparents prepare the va and an extra basket containing zok va, paper-cut flowers (28 May 2007, Dai An commune, Lang Son province, photo by the author)
Invitations and the Making of the Va In general, around two days before such a gathering, the baby’s paternal grandfather starts visiting each household related to him and invites them to attend a lunch, which the young people in these households are asked to help to prepare. The extended family of the baby’s father, including married women, does not need to bring anything to the ‘third morning’: they just attend, but they can give the newborn’s household a chicken later, when the baby is a few months or even years old. The household of the baby’s mother also go around and invite their relatives, and busy themselves making the bamboo va to be carried to the newborn’s house and preparing items for ‘bringing the basket’. The maternal grandfather of the newborn baby is responsible for carrying the va from his own house to the baby’s, but this is only the case when his spouse is alive. Otherwise, the brother (or another male relative) of the baby’s mother and his wife can fulfil this duty: it is vital that ‘those who prepare and tuck the va on their back must be a couple’. The basket itself is round in shape with a separable cover and made with bamboo strips. This basket, loaded with cooked glutinous rice coloured yellow
The Continuity of Life from Past to Present among the Nung of North Vietnam 215
which is spread out on banana leaves laid in the bottom, and a small, square bamboo basket, thu, is placed on one side of the pole.9 On the other is a live chicken in a woven bamboo cage. Inside this thu are two bowls of yellow sticky rice, a bottle of rice wine flavoured with ‘yellow ginger’, a cooked chicken and two zok va. The gender of the cooked chicken is the same as that of the newborn baby. There are two flowers, one red and the other yellow, each glued on to a bamboo stick. The colour of the sticky rice comes from a type of ginger called ‘yellow ginger’, which is also used to dye rice wine. The ginger is said to create ‘heat’, a substance the new mother must replenish in her body: she needs to be ‘warmed up’ for her own health and to be able to have sufficient milk to breast-feed her baby during the post-natal period. For the month after giving birth, the mother eats a diet of chicken, pork, or a selected variety of vegetables, all cooked with this ginger, but without any water. The flowers are not only an essential element of the ‘third morning’ rite, but are also seen in a series of rituals concerning the conception, birth and growth of a child.10 These items are offered to me mu, who is believed to look after anyone under fifteen years old. After a person reaches the age of fifteen, she or he will be looked after by me sheng. The Nung believe that me sheng in the heavens gives birth to every person. I asked if the va would be used for worshipping me sheng after each person reaches that age. The villagers often pondered for a fleeting moment, then told me that ‘the va is only for worshipping me mu’. When I interviewed the villagers, what the transition from me mu to me sheng involves was often not made clear to me, although they were always clear about the age boundary, fifteen, regarding the worship of these two gods. In some households a piece of red paper, usually quite washed out, had been pasted on the wall just above the platform suggesting a title ‘sheng mu’, 聖母, for the me sheng (originally 壽 妣南山;南堂太白六國婆花皇聖母妙扶人;福如東海). The households of the bung lang also ‘bring the basket’, but only yellow or red sticky rice and the live chicken are given to the baby’s household. A man (rarely a woman) from each of these households carries such a basket with a carrying pole.
The Day On the day of the ‘third morning’, the father (or one of the male relatives) of the baby’s mother sets out from his house, shouldering the carrying pole with the basket. This man tucks his arm through the shrine platform and walks swiftly in silence; he smiles but does not make any sound. The Nung say that the person carrying the va should remain silent when walking towards the baby’s house. No one should greet or congratulate him cheerfully before he attaches the shrine to the wall and performs the first offerings to the me mu. This silent procession is said to prevent children from being mischievous and crying too much.
216 N. Jenny Hsu
Everyone who sees him arrive at the baby’s house smiles and whispers, ‘Here comes the grandfather (or uncle)’. The man enters the room of altars and hammers nails into the walls to hold the new shrine, with the help of other men from the baby’s paternal side. Then the cooked chicken is taken out of the basket along with the red and yellow flowers and placed on the new va.11 He then pours rice wine into the two cups (prepared by the household of the baby) and leaves the wine bottle on top of the shrine. The two bowls of yellow rice are placed on the platform while incense is lit on the ancestors’ altar. After the ritual observance, the rest of the rice wine in the bottle is only consumed by the post-natal mother and her husband; the bowls of rice and cooked chicken, on the other hand, are eaten by anyone who attended the occasion. Chicken brought from the bung lang or raised by the paternal household can be cooked immediately and presented on the ancestors’ altar. All the men – the uncles and grandfathers from the paternal and maternal sides, and the father of the baby – finally sit down near the ancestors’ altar to have some tea and create the atmosphere of having a new generation. Friends and relatives now have no hesitation about talking or laughing out loud. The whispers – calling out each other’s new kinship terms – become cheerful greetings. Everyone has an updated name reflecting their relationship to the new baby. The new father’s parents are now called gong and meh, as they have truly become grandparents. The new mother’s parents become boh lau and me lau. People are figuring out how to now address whoever steps into the house. The Nung use kinship terms to refer to generations within three earlier (+3) and later (–3) generations. Beyond the +3 generation, the villagers may have records in their genealogy book, but they refer to such people using the collective term ‘distant ancestors’. Villagers follow a teknonymy naming system which extends to everyone: people are addressed by a term combining their relationship to the first child plus the child’s name. When the locals receive guests who are not part of their kinship network, Vietnamese terms such as anh, bà, chi., chú or ông are usually used, depending on the observable age-relation between the person addressed and the addressee.12 A midday meal starts, when things become boisterous. Yellow glutinous rice is provided for everyone to eat their fill, and toasts of regular rice wine seem never to stop. While this is going on, the women go into the mother’s room to take a look at the baby.13 This ritual observance usually ends in the early afternoon.
Bung lang The bung lang are defined at the very moment of the marriage.14 Bung lang refers to those households that are related to the wife-giver, while those on the wife-taker’s side become bung na. Lang and na are Nung words used to describe
The Continuity of Life from Past to Present among the Nung of North Vietnam 217
the ‘back’ and ‘face’ of the human body, of a house, and the day or moment ‘after’ and ‘before’, respectively. During certain events, such as the ‘third morning’, funerals, or rituals to cure a sick person, one member of the bung lang will ‘bring the basket’ to carry rice-based foods, rice wine and chicken to the household of their bung na. Each Nung house falls into the bung na or bung lang category, depending on whether it is their son or daughter who is getting married. When a son gets married, this household becomes the wife-taker in contrast to the wife-giver’s household; when a daughter marries out, they become the bung lang of that wife-taker’s household.
The va for me mu The Nung traditionally built their houses on stilts, creating three levels inside each. The front of the house always faces the paddies and the back faces the mountains. The average height of a stilt house from the ground to the highest point is about six metres, and the width is about four to five metres. The ground level of a stilt house is traditionally a space for buffalos and pigs or other domestic animals; the middle level is where people receive guests, rest, eat, cook, store things, and have various altars; the top is another storage space for food for livestock, such as hay or unthreshed rice. Nowadays, as construction materials have become scarce, people building new houses rarely use wooden stilts to hold up the house.15 The space at the bottom for raising domestic animals has thus been removed from the house itself and replaced by a self-standing one- or two-storey shed.16 The hearth, rooms for women, beds for men and altars are indispensable elements of a Nung house.Entering by the front entrance, one steps over the threshold and sees a wall straight ahead, at the rear of the house; behind this is the hearth where meals are prepared. The hearth, referred to as ‘top of the fire’ in Nung, is believed to be where the ‘hearth goddess’ is situated. This hearth goddess is called me hen fai, literally ‘the woman near the fire’. There are several taboos associated with this goddess. For example, one should not step directly over the three-legged stand for pots, and banana leaves or paper with words written on it should not be burnt under this stand, or else one will have lots of wrinkles or not be able to learn well at school, respectively. These taboos were observed most strictly in the house of the ten, a ritual specialist. Traditionally, each household sends this goddess back to the sky before the Lunar New Year and takes her back to the hearth in the New Year. The main facility consists of a rectangular area, from two to six metres wide, with a surrounding frame of cement or wood on the floor. Enclosed chambers are located on one side of the house in which to place the altars and rooms for women.17 Altars for people to venerate their ancestors always face the front of the house, in general near the front entrance. The beds, but usually not the rooms for men, are usually on the side opposite the altar room.18
218 N. Jenny Hsu
The altar, which is usually placed on a table, is erected when the house has just been built. Altars are used to worship the ancestors and the female deity or me nang. A smaller platform, the va, is hung on the wall of this room; as we have seen, it is set up a few days after the first child is born to worship every Nung person’s me mu. This va is a bamboo-woven platform attached to the wall with strings. Its average size is 20 by 30 centimetres, woven from a particular soft bamboo that is easy to bend. On top of this people place two wine cups, three sticks of incense, and red and yellow paper-cut flowers, one on each side of the platform. The va used to be hung on the wall near the hearth; now only a small number of Nung households still do this, and it is now usually mounted on the wall near the ancestors’ altar, sometimes in a room. As one woman said, the protruding platform would hinder the path of people walking from the hearth to the main room. A household may move it into the altar room. The Nung say that the goddess or me mu is in charge of how a person lives in the present-day world.19 And each goddess has her individual character. If a baby cries excessively during the night or seems to be sick, it is said that her or his me mu is craving for something. The household then immediately pours rice wine into small cups placed on the va and lights sticks of incense; then a whole cooked chicken is prepared to ‘quieten’ the goddess and hopefully, simultaneously, the baby. A similar arrangement of foodstuffs and other items are regularly offered on the va while one is still young and whenever a household makes offerings to its ancestors for different rites throughout the year, especially any involving a ritual specialist, or ten.20 Any kind of unhappiness or bad luck a person experiences, especially concerning love or career, can imply that her or his mother goddess is upset or desires something. For example, if men or women do not have good marital matches in their twenties, at the end of the lunar year the family will first consult a ten to make offerings to the me sheng of this person during a ritual. The ritual is always designed for the welfare of every member of the same household. Items are meticulously prepared for offering, including an offering of real flowers (usually blossoms attached to a branch) to give this particular person a chance of marriage. However, not all rituals are conducted with the sole purpose of indulging the me sheng. What I describe above is how in general a ‘third morning’ ritual occasion is conducted. People may, however, choose to alter some aspects to accommodate the demands of modern life, which is organized around the state-used sevenday-week calendar schedule. This contrasts with how local adults still carry out each aspect of their lives according to the lunar calendar. They abide constantly by certain elements that appear to be intrinsic to this ritual observance, such as where the va should be set up, who ‘brings the basket’, who eats what, and so on, on the ‘third morning’.
The Continuity of Life from Past to Present among the Nung of North Vietnam 219
Conclusion As stressed throughout, the ‘third morning’ is ideally held on the third day after the first child of a generation is born. When a household has difficulties holding this ritual observance exactly on that day, it consults a Taoist priest to decide on another good day; this must occur within the first month of the child’s life. The only ritual observance of the ‘third morning’ I was able to attend was held ten days after the baby was born because the mother gave birth in the provincial city and could not immediately return to her husband’s home in the hamlet. I was told they were also wishing to arrange such a day on a Sunday, despite most people involved living their lives according to the lunar calendar. Both the maternal and paternal grandmothers of the baby were teachers at an elementary school, and they worked six days a week. Sunday was therefore the best day for them and even their colleagues, for whom no kinship relation was involved, to attend this special occasion. In cases like this involving some delay, the villagers still refer to this celebration as ‘third morning’. At the ritual observance I attended, although the date was slightly postponed, the mother who gave birth in the city carried her ten-day-old son on the rough roads back to her husband’s house for the ritual observance: to mount the va is indispensable for a household. The exact timing of this ritual observance is of secondary importance, as long as it is held within the first month after the baby is born, but the va must be hung up inside the baby’s father’s house and yellowdyed sticky rice be brought over by the bung lang. These practices are still carried out regularly ‘because this is the way our ancestors had done’, as I was told constantly when asking people why they perform these traditions. To continue the line of a household is vital to the Nung. Those who attend the ‘third morning’ will acquire a new kinship term with reference to the firstborn baby of every generation. People would be figuring out and calling each other’s new title, as described in ‘The Ritual Observance of the “Third Morning” ’ in this article, which gives the rite a cheerful atmosphere. But more importantly, this is done to celebrate a new generation in the family. For the sprouting new babies in the same generation under the same household, a rite as such is not necessary. The kinship terms for each relative would not change at this moment: the household merely needs to offer a cooked whole chicken and light sticks of incense on the existing va. This new generation is also perpetuated through households of the wifegivers, whose ties to the baby are strengthened. The maternal grandfather is usually responsible for bringing the va to his new grandchild’s household, while others on the side of the bung lang take yellow sticky rice to the rite. The basket of sticky rice is not unique to the ‘third morning’, as in other rituals described in my thesis (Hsu 2007), a different colour of rice is offered. The bamboowoven va is the focus of this rite.
220 N. Jenny Hsu
The location of each person’s me mu is important in Nong religion. She is to be pacified under crisis circumstances, such as when the baby has problems for a long period of time. As a young person grows up, and throughout her or his life, different types of rites and rituals are performed: some rituals are to pray for a nice marriage, some are to cast away sickness, and others are for general blessings at the beginning of a lunar year. For some rituals, if a household is not rich enough to invite a ritual specialist and hold a feast for the guests, they may give them up this year or cut down on the items prepared for the ritual. These rituals all involve the mother goddess, me mu, who is worshipped on the va and can be pleased through the chanting power of the ritual specialists and the offerings of the household. In turn, she will not only quieten down the baby, but also bless the female or male who lives in this world. As stressed already, me mu refers to the goddess who is in charge when one is under fifteen, while me sheng is for those above fifteen. However, a specialist is not required for the ‘third morning’. In fact, at the ritual I was invited to, the maternal grandparents of the baby could not see or touch the baby because the father is a Nung ritual specialist, ‘a man with bright eyes’ as the locals put it. He and his wife must avoid contact with newborn children because babies carry ‘unclean matter’ that would interfere with their powers. They can see the baby after he is one month old. Moreover, a person consulting a shaman should be aware of this and should also avoid being near a newborn child for at least seven days before the ritual is performed. The celebration of the beginning of a life, the ‘third morning’, must be carried out as an act of settling in me mu of the newborn baby. Although those who are to attend may have different responsibilities today and cannot gather together on precisely the ‘third’ day after the birth, the rite must still be held without fail. The Nung in Lang Son have not been secluded inside their own world since their migration from China around two hundred years ago. In recent decades especially, they have been slowly catching up with the development of the rest of society, through the media, public facilities and political apparatus, although their lives are still considered reserved and simple. But while adopting the conveniences of modern life, the Nung have not cast aside all the ways of their ancestors: these are apparent in their rituals and ritual observances, such as the ‘third morning’ described in this chapter. Modernity has not affected how the Nung practise their traditions, but it has modified how they celebrate the birth of a new life and also a new generation. That is integral to how the line of a household is preserved: for the Nung, as a person one can live one’s whole life under the good hands of one’s goddess.
Acknowledgements Some of the material presented here comes from Chapters 2, 3 and 5 of my thesis ‘The Flow of Rice, the Flow of Life: Exchange of Rice-Based Foods of
The Continuity of Life from Past to Present among the Nung of North Vietnam 221
the Nung in Lang Son Province, Socialist Republic of Vietnam’ (2007). I thank my host family and all the other households in the hamlets of Dai An commune, Lang Son province, in Vietnam. I thank my thesis advisor, Professor James R. Wilkerson, and Professor Ho Ts’ui-p’ing and Eli Alberts for editing. The fieldwork was supported by a joint Graduate Student Writing Award from the Institute of Ethnology, Academia Sinica in Taipei and the Institute of Anthropology, Tsing Hua University in Xinzhu. I also acknowledge with gratitude the Subsidy for Students for International Visiting, provided by National Tsing Hua University and the Ministry of Education, which supported the initial writing of my thesis on which this chapter is based. However, I am solely responsible for the content.
Notes 1. Basically all the households are related and so all will receive new kinship terms. Guests like myself are rare in the village. I was known as the adopted daughter of my host family. 2. In the rest of this chapter, I will use the terms va, me mu and bung lang because the translations do not entirely reflect their true meaning. The translations ‘third morning’ and ‘bring the basket’ are used for the terms sham ne and au lam respectively. 3. All Vietnamese names and place names are written here without the diacritics. 4. The documents referred to here are the genealogy books kept by each family, which record the names and birthplaces of family members. The earlier records are in traditional Chinese characters, which the locals cannot understand. After realizing I could read Chinese, whenever the chance arose the locals were interested in showing me their books. More recent records are written in Vietnamese. 5. These fictitious names have been chosen for the hamlets to protect the privacy of the local residents. The three hamlets and nine others belong to the village of Dai An. Along with another eight villages, they form the commune of Dai An. The population of ‘the hamlet on the mountain foot’, the main hamlet I stayed in, was recorded as 93 in December 2005; this figure included some working or studying away but having their national residence registered in the hamlet. Some babies were born in this period, bringing the population to 96 by the time I left the field in May 2006. 6. A day of diligent climbing and picking can yield about 30 kilos. Younger members of a household usually do this work, but I have met sixty-year-old ladies who could still climb the trees. Those who pick the star anise need not be from the same household; others can be hired to help out, especially at harvest times. 7. To determine if two young people ‘match’, their names and details of when they were born are taken by their parents to a Taoist, who has a book explaining the ‘destiny’ such details determine. If, for example, the Taoist says that a woman will have bad luck in her life, bringing misfortune to whoever she marries, she may have great difficulty in finding a husband. 8. See my Master’s thesis, ‘The Flow of Rice, the Flow of Life: Exchange of Rice-Based Foods of the Nung in in Lang So’n Province, Socialist Republic of Vietnam’ (2007). 9. The thu must be made on the day of the ‘third morning’. Traditionally every family then kept the basket to use in future rituals, when it would be used as a container to ‘retain the spirits’ of family members for their health. Nowadays few households actually use a basket as the container because a basket gets broken over time. More often, when there is a ritual, a plastic bag is used to replace the thu. 10. The colour of these flowers is consistent for each ritual. For further details, refer to my thesis (2007).
222 N. Jenny Hsu
11. An older va may exist if the father of the newborn baby is still living with its paternal grandparents. This older one was set up when this father (also the first child) was born into the house. 12. For further descriptions, see Chapter 3 of my thesis (2007). 13. The post-natal mother cannot leave her room until seven days after the baby’s birth. She defecates into the ground from the floor near the hearth as necessary. For one month, she eats particular dishes made with ginger as mentioned above, but cannot drink plain hot water or tea. She and the baby cannot bathe in water, but instead use towels to clean their bodies. 14. Those who are vital to ‘bring the basket’ are not only the households of the bung lang, but also certain households who play the role of an ‘adopted child’. A man or woman in these households could be adopted by his or her mother’s brother or a ritual specialist. I have described this further in my thesis (2007). 15. In the three hamlets I visited in 2006, newly built houses were generally made of cement. As wood had become more precious, in houses built within the previous ten years it had been replaced with cement, bricks, or just mud-plastered straw. 16. If the shed has two storeys, the upper one is used to store stacks of rice stalk hay. Buffaloes or pigs are housed at ground level. 17. In the old days, altars were situated in the upper storage space of a house. They are now mostly down on the middle level: the storage room has become the altar room. Some households still retain their altars up above. 18. When the villagers talk about who sleeps where, regardless of men or women, they only use the word ‘bed’, but beds for women are often placed inside enclosed rooms. Men and women have separate beds, and neither gender can sleep on a bed designated for the other, even when no one else is inside the house. This prohibition applies to siblings of different gender, and especially to those women who have married into a household. They are prohibited from sleeping on the beds that belong to, or formerly belonged to, any male kin in their husband’s household. In some large households, a room where only women can sleep extends into the area next to the hearth. 19. Also see Gao (2002, 2005) and Pan (2005) for their extensive discussions of this female goddess who is fundamental to Zhuang belief. According to Saul and Gregorson (1980: 206), the Nung who were resettled to the south in 1954 believed that ‘at the birth of a child, [a spirit] has given the child life and will guard him and rule his life. The parents are careful to perform acts of worship toward the spirit at birth. Later, however, when a person marries, this spirit no longer rules him.’ 20. Chicken is the usual oblation offered on the va, but exceptions do occur. I followed the ten to one ritual in early January 2006. Ten is one of the three ritual specialists often appearing in the Nung area. Instead of the usual whole cooked chicken, this house offered ginger on the va. The ten told me that the house has been doing this for a long time, even according to the eldest person in that family. While giggling at their offerings, the ten said that maybe the ginger would work to ‘calm down’ the goddess because it is so spicy that the goddess cannot ‘talk’.
References Barlow, Jeffrey G. 1987. ‘The Zhuang Minority Peoples of the Sino-Vietnamese Frontier in the Song Period’, Journal of Southeast Asian Studies XVIII(2): 15–46. Fan Hong-Gui. 2005. ‘The Dearest Siblings of the Zhuang Nationality in Southeast Asia’, Journal of Guangxi University for Nationalities 27(1): 119–24. Gao Yaning. 2002. ‘Guangxi Jingxixian Zhuangren nongcun shehui zhong me214 moùt31 (mopo) de yangcheng guocheng yu yishi biaoyan’ (廣西靖西縣壯人農村社會 中 me214 moùt31(魔婆)的養成過程與儀式表演, Initiations and Ritual Performances
The Continuity of Life from Past to Present among the Nung of North Vietnam 223
of the Me214 Moùt31 (Mopo) among a Zhuang Agricultural Society in Jingxi, Guangxi Province). MA thesis. Taiwan: Institute of Anthropology, National Tsing Hua University. ———. 2005. ‘Rice, Meat and Personhood in Jingxi’, Journal of Chinese Ritual, Theatre and Folklore 150(12): 27–68. ? Hà, Va˘ n Thu’ and Lã, Va˘ n Lô. 1984. ‘Va˘ n Hoá T ay Nùng’ (Tay and Nung Culture). Hà NÔ. i: NXB Va˘ n hoá. ? Hoàng, Minh Lo.’i. 1992. ‘MÔ. t Sô´ Phong Tu. c – Tâ. p Quán Liên Quan Ðê´ n Nhà C u’a ? C ua Ngu’Ò’i Tày, Nùng’ (Some House-related Customs and Habits of the Tay and ? Nung), Nguyê˜ n Thi. Th ao, trans. Ta. p Chí Dân Tô. c Ho. c 3: 47–52. Hsu, N. Jenny. 2007. ‘The Flow of Rice, the Flow of Life: Exchange of Rice-based Foods of the Nung in La. ng So’n Province, Socialist Republic of Vietnam’. MA thesis. Taiwan: Institute of Anthropology, National Tsing Hua University, Hsinchu. Lebar, Frank M., Gerald Hickey, and John K. Musgrave. 1964. Ethnic Groups of Mainland Southeast Asia. New Haven: Human Relations Area Files Press. Li Jing-fang. 2002. ‘Dongtai yuyan yu wenhua’ (侗泰與言語文化, Languages and Culture of the Dongtai). Beijing: Minzu Chubanshe. Nong Xian-sheng. 2005. ‘Relations between Being Refugees of Zhuang People after Nong Zhigao was Defeated and National Alienation in Southeast Asia’ (儂智高 失敗後壯族和儂氏逃難與東南亞民族異化的關係), Journal of Wenshan Teachers College (文山師範高等專科學校學報) 18(3): 206–10. Pan Chun-jian. 2005. ‘Rice and Kinship in the Zhuang Villages of North Bank of Yu Jiang River’, Journal of Chinese Ritual, Theatre and Folklore 150(12): 69–131. Pelley, Patricia. 1998. ‘ “Barbarians” and “Younger Brothers”: The Remaking of Race in Postcolonial Vietnam’, Journal of Southeast Asian Studies 29(2): 374. Saul, Janice E., and Kenneth Gregerson. 1980. ‘Nung Priests and Spirits’. In Notes from Indochina: On Ethnic Minority Cultures. M. Gregerson and D. Thomas, eds. Dallas: SIL Museum of Anthropology, pp. 201–14. Teng Cheng-Da. 2002. ‘Zhongguo zhuangzu yu yuenan dai, nong minjian zhushen xingyan bijiao’ (中國壯族與越南岱儂族民間信仰諸神比較, Comparison of Belief in Gods of the Zhuang Nationality in China and Dai and Non-Nationalities in Vietnam), Around Southeast Asia 2: 45–48.
Afterword Performance as a Mechanism for Social Change James Wilkerson
Comparisons of the ethnographic findings of the chapters in this volume that relate directly to the relationship between community-level social institutions, performance and social change suggest ways in which the volume speaks to more general issues of social change in East Asia. The notion of performance is key to this argument. In this respect, three formal features of performance stand out as especially important. In this concluding chapter, I shall first discuss these features in general terms, then show how they apply to each of the chapters in this volume, and finally draw some comparisons and contrasts between the various cases and their ethnographic contexts. First, closer consideration of the chapters suggests that one shared feature to which special consideration should be given is the ‘keys’ for the ‘footings’ of ‘ratified’ participants in a performance. The ‘key’ is ‘the set of conventions by which a given activity, one already meaningful in terms of some primary framework’ outside the performance, ‘is transformed into something patterned on this activity but seen by the participants to be something quite else’ (Goffman 1974: 44; see also Bauman 1984: 15, 38). The ‘footing’ is Erving Goffman’s term for ‘the position or alignment’ a participant takes within a performance (Duranti 2002: 296, citing Goffman 1981: 173; also Goffman 1979). Put more simply, and as discussed below in making further comparisons between the chapters of this volume, a ‘key’ for a ‘footing’ is simply a relationship in the classification of statuses in a corporate entity or social category that is outside a particular performance, but that valorizes the ‘footing’ of a participant within that performance. Lastly, the presence of a participant in a performance is ‘ratified’ by this keying in so far as he or she becomes ‘entitled and expected to be a part of the communicative event’ (Duranti ibid.: 298, citing Goffman ibid.: 226). As for the second feature, the extensive theoretical literature on performance notes that there is an ‘emergent quality’ to performance and that this emergent
226 James Wilkerson
quality is a potential mechanism for social change (Bauman 1984: 16, 38, 43– 45). This literature sees this ‘emergent quality’ as being visible across three qualities of performance: a performance’s text, event and social structure. The texts are internally coherent verbal or other meaningful acts, the events are fixed act sequences and/or other rules for performance, and social structures are set relations between the participants in a performance (ibid.: 40–42). The emergent quality for each is the difference between the fate of a text, event and social structure between the beginning and end of a performance. This emergent quality arises out of the processes of conflict, negotiation and consensus within and surrounding the performance through which a difference is achieved. As for the third feature, the literature on performance likewise suggests the presence of a ‘continuum’ from the ‘traditional’ (or ‘residual’) to the ‘innovative’ (or ‘emergent’) poles of performances (ibid.: 40, 47–48; citing, for ‘residual’ and ‘emergent’ culture, Williams 1973: 10–11). At the ‘traditional’ end of the continuum, the difference of a performance is a renewal of the extent of social structure, so that the everyday social structure at the beginning and end of the performance is the same. The emphasis is on social continuity and reproduction. At the ‘innovative’ end of the continuum, the difference is an everyday social structure at the end of the performance that is in some way substantially altered from that at the beginning of the performance. The emphasis is on social discontinuity and transformation. In comparing the ethnographic descriptions of each chapter and concentrating on these three specific formal features of performance, some surprising similarities and differences in community-level performances emerge across – in order of discussion – the six chapters on the People’s Republic of China (PRC), the three chapters on Taiwan, and the single chapter on the Soviet Socialist Republic of Vietnam. Chapter 1 by Wang Ting-yu on the Tibetans of Zhoukeji and chapter 2 by Liu Biyun on the Qiang of Nzidabu offer one set of comparisons limited to Sichuan Province. The authors describe the statuses of the house and the coufang respectively as keying the footings of life-cycle performances. The statuses in the house and coufang are rooted historically in former subbureaucracies, the house within a chieftaincy and the coufang within a military colony. It is important to note that these former sub-bureaucracies were integrated into an empire-wide organization of state authority and power over local communities, with its centre in the Qing dynasty being the imperial capital Beijing (surveyed in Brook 2010). Although perhaps counterintuitive, given the feudal associations of the term ‘imperial’, the legal person and all its encompassing categories represent one modality of modernity. That is, although there were differences in occupational statuses for chieftaincies and military colonies in Zhoukeji and Nzidabu, as well as elsewhere across the empire (Herman 2007; Took 2005), the social classification of statuses in the late imperial sub-bureaucracy of the Ming and Qing dynasties represented a sharp break from medieval China. That is, the ‘modern’ sub-bureaucracy of the Ming and Qing dynasties dispensed with many of the inherited occupational
Performance as a Mechanism for Social Change 227
statuses of previous dynasties, and the range of statuses keyed to and supplying ratification for sitting the imperial exams was much broader (Faure 2007). The person – always male – sitting for an imperial exam in these dynasties enjoyed suffrage as a legal person of a much broader range of statuses in comparison to earlier dynasties (cf. Elman 2000). Moreover, for both the Buddhist Tibetans of Zhoukeji and the Buddhist and Taoist Qiang of Nzidebu, the texts, events and social structures for ritual participants still place special emphasis on performances involving the cult of the dead. For both Zhoukeji and Nzidabu, the religious texts are written scriptures that circulate well beyond the community. The events are an elaborate number of ritual steps aimed at the care of the corpse and the afterlife of its soul. The social structures are hierarchical orderings according to gender, age and rank that mirror the traditional rankings present in community social institutions. Although the classification of houses in Zhoukeji differentiates between Tibetan and Han, the cult of the dead for Tibetan and Han alike is Tibetan Buddhist (Gouin 2010). For Nzidabu, however, the Buddhist side is apparently limited to community-wide, cross-coufang performances. Although the national minority identification of the village members is exclusively Qing, the cult of the dead that centres on the coufang nevertheless distinguishes between a Tibetan orientation towards the care of the body, and care of the ancestral soul, which is Confucian (cf. Brook 1989). Chapter 3 by Ho Zhaohua on women’s costumes among the Miao of Shidong, and chapter 4 by Chien Mei-ling on love-songs among the Miao of Fangf Bil offer a second set of comparisons limited to Guizhou Province. The female costumes among the Shidong Miao and love-songs among the Fangf Bil Miao each display controlled affinity that keys the footings of life-cycle performances, while the boundaries of the earlier lower-level imperial bureaucratic administrative divisions formed an internal frontier for the Chinese Empire that surrounded a ‘raw Miao’ enclave. While the subbureaucracy was absent from this enclave, commercial activity involving outside Chinese markets was and still is intensive (see Schein 2000 for the national perspective). It is likewise important to note that, in the case of the positive sense of controlled affinity (here meaning marriage between classificatory cross-cousins among the Fangf Bil Miao, and the marriage of the children of classificatory cross-cousins among the Shidong Miao), social mobility was in practice integrated – through the commodities produced, exchanged and consumed in Shidong and Fangf Bil in Guizhou – into an empire-wide organization of market mechanisms that fixed and still fix value well beyond the confines of Guizhou. Thus, despite the presence of this sense of affinity and the location of the former bureaucracy on the outside, there was and is an aspect of personal value and autonomy in women’s costumes and courtship. This commercialism, and the room it leaves for what Chien Meiling frankly calls a culturally specific sense of the individual, likewise suggests a sense of modernity within that enclave that was once unfettered by the direct application of the sub-bureaucratic mode of modernity.
228 James Wilkerson
As with Zhoukeji in chapter 1 and Nzidabu in chapter 2, the emergent quality of the texts, events and social structures of Shidong and Fangf Bil also centre on the performance of life-cycle rituals. The texts range from single characters, such as that for ‘accomplished or completed’ (cheng) that are sewn into the female costumes, to an elaborate oral mythology and life-cycle rituals that describe and enable the step-by-step cosmic cycling of souls between heaven and earth. Thus, although adopting significant elements from the Han Chinese written religious tradition, these elements are integrated into a clearly Miao oral textual tradition for life-cycle rituals. The event quality of life-cycle ritual performance includes women leaving their homes and parading and singing in public in their traditional costumes for community inspection and comment. Finally, as social structure the performances are homologous with everyday life. In sum, comparison of the ethnographic descriptions in chapters 1 to 4, all of which are located at the traditional end of the performance continuum in the PRC, reveals the presence of a contrast in modalities of modernity – subbureaucratic versus marital – that, in different ways, key and ratify the footings and participants in various life-cycle ritual performances. However, all these first four chapters likewise offer three points of underlying similarity. First, the sub-bureaucratic or maritally constrained senses of the modern individual within the respective communities are integrated into empire-wide political or commercial realities. Secondly, there is a shared emphasis on life-cycle ritual performances. The wider circulation of ritual texts is sometimes in the form of written Buddhist, Taoist or Confucian scriptures that cross ethnic, religious and even state borders, or, as in the case of the Miao, oral texts that are widespread among the Miao, extending beyond the confines of any single community. Thirdly, these sorts of footings and qualities of performance predate Western impact. Chapter 6 by Eveline Bingaman on tourism in Lijiang and chapter 5 by Chang Kuei-min on migrant workers in Shanghai offer yet a third set of comparisons involving respectively an ancient market town in north-western Yunnan and one of the world’s great metropolises located at the mouth of the Yangzi River. Both of these chapters cover the ‘innovative’ end of the continuum for the chapters on the PRC. Both chapters suggest, specifically, the presence of a ‘floating’ modality of modernity that is matched by a disengaged, even liminal sense of the individual. It is important to note in passing, however, that neither tourism (Brook 1999) nor migrant work are entirely recent phenomena in China (Skinner 1976). Rather, contemporary tourism (Lew et al. 2003) and migrant work (Fan 2008) began under significantly new circumstances associated with social changes in the PRC following the reform and postreform eras. That is, these performances occur within the special historical context of political adjustment to the consequences of the reintroduction of market economics during the reform and post-reform eras between 1979 and the present. First, in comparison with earlier eras before these, state control over child-bearing actually tightened (Greenhalgh 2008), although resistance
Performance as a Mechanism for Social Change 229
has been widespread, even though the demographic results are now provoking some second thoughts (Hesketh et al. 2005). Secondly, there has been a loosening in the political organization of state controls over physical mobility, which has been broadly embraced and matched with an enormous expansion of industry and an extensive upgrading of the transport infrastructure (He and Pooler 2002). Thirdly, a market economy has largely replaced the former Soviet-style command economy, which has not only turned China into the second largest economy in the world, but has also transformed China into a political and military world superpower (Wu 2010). Correspondingly, interstitial spaces have opened up outside the party and state apparatuses and bureaucracies, where innovative performances beyond any single community form a route through which incremental social changes are gathering momentum and even forcing further selective loosening along the edges of single-party Leninist-style state rule. These interstitial spaces have profound implications for the footings and qualities of performance. For tourists as for migrant workers, people native to one local community tour or work in another community. Specifically, the footings of tourists and migrant workers are less wholly grounded in hometown social institutions, yet not fully integrated into the social institutions of the communities where they tour or work. What holds for the interstitial footing of performances also holds for the qualities of such performances. The texts of the performances differ from traditional performances in two significant ways, both of which even emphasize silence or miscommunication. Still, they share a common rhetorical structure as negotiations. Thus, the language in the texts of performance may at first glance exhibit little continuity from performance to performance in so far as there is no fixed script. For one instance, Chang Kueimin describes the negotiating tactics in confronting violations of the one-child policy, ranging from simply silently evading enforcement by leaving the natal village, to resolving the matter with a friendly dinner with officials during a return visit to that village, or the selective paying of fines. For another instance, Eveline Bingaman describes the tactics of negotiation between international agencies and the government of the PRC. The definition of such terms as ‘traditional culture’ and ‘preservation’ differed between the representatives of the PRC and the World Heritage Committee, resulting in the negotiations involving more miscommunication than communication. In terms of events, even when taken as groups, migrant workers and tourists are atomized in so far as the tourist seeks personal satisfaction and the migrant worker personal profit. Finally, when it comes to the emergent quality of social structure, the saturation of the traditional social structures with the repetition of such performative encounters with tourists and migrant workers drives new responses in the form of the adoption of new routines in everyday social structure, meaning that social change saturates both the home and target communities, although in separate ways. Going one step further to compare all six chapters on the PRC together, the ‘traditional’ and ‘innovative’ ends of the continuum interact differently with
230 James Wilkerson
one another in, on the one hand, the sub-bureaucracy and the positive sense of controlled affinity of community social institutions, and on the other hand, in tourism and migrant work inter-community mobility. Even though sometimes co-present, the descriptions in the chapters of this volume suggest that the two ends of the continuum do not systematically conjoin in either their respective keying of footings or qualities of performance. This is not, however, to wholly discount variation between the different chapters. For instance, only in the case of Lijiang is the impact of tourism resulting in a readily visible wholesale reordering of a community-level social institution, while only in Nzidabu is tourism of little or no importance in such institutions. Still, in Zhoukeji, Shidong and Fangf Bil tourism is, apparently, just another source of income whose profits are to some extent reinvested in traditional social institutions. The situation is again different when migrant work is considered. In all six chapters on communities in the PRC, migrant workers are either working in these communities (Shanghai, Lijiang) or native to them (all the other chapters). In Zhoukeji and Nzidabu, the income of migrant workers only seems to provide more resources for funding the community-level performances at the traditional end of the continuum; however, in Shidong and Fangf Bil, which also benefit from tourism with seemingly little or no consequence for the traditional end of community-level performances, the outflows of migrant workers has begun to compromise controlled affinity and related performances. Women from Shidong and Fangf Bil who are migrant workers are not learning the requisite skills to make the costumes that are necessary to a good match, which increases the likelihood of marriages with outsiders they meet while away from their home community. In sum, in the chapters on the PRC, the interface between the traditional and innovative ends of the continuum sometimes threatens the traditional social institutions, but does not institutionalize a link between social institutions and the ‘floating’ sorts of modernity. A discussion of the footings and the emergent qualities of performance in chapters 7, 8 and 9 on Austronesian speakers in Taiwan suggests further instructive comparisons with the PRC, as well as in comparisons within Taiwan itself. In these three chapters, the keyings of the footings of the traditional and innovative ends of the performance continuum come together. To begin with, there are the keyings of the footings of participants in performances at the traditional end of the continuum. In chapter 7, Futuru C.L. Tsai describes the age-grade village system for the Amis in A’tolan village as one key. In chapter 8, Li-Ju Hong provides an account of a hereditary village hierarchy between aristocrats and commoners for the Paiwan village of Puleti as a second key. In chapter 9, Yayoi Mitsuda sketches out a village membership corporate organization for the Thao in Ita Thao as a third key. The local-level community social institutions for each of these three villages have a historical depth that goes back well before the imposition of Japanese rule over Taiwan in 1895. Next, there is also the keying of footings at the innovative end of the continuum for participants’ community-level performances that are informed
Performance as a Mechanism for Social Change 231
by the global flows of labour, art and identities at loose in the wider Taiwan society. Although a more recent phenomenon, this innovative keying follows a clearly discernable pattern from village to village. In A’tolan, the key is to wage work and residence outside the village. The knowledge gained from the outside world ratifies innovative performances in the annual harvest festival. Alternatively, in Puleti the keying is to the exclusion of literacy in schools. Ratification is similarly nullified when literacy is introduced to weddings and associated activities. Lastly, for the Thao of Ita Thao the keying of the footings is to the village as a tourist site and place of business. As an authentic endangered indigenous ethnic people, Thao entrepreneurs are ratified (although there are any number of impostors) to market parts of their identities, whether singularly in tourist souvenirs or as participants in community rituals for the media. Overall, in ranging between engagement, rejection, and accommodation, there is thus a sense in which the keying of the footing of participants in communitylevel performances in all three chapters on Austronesian communities in Taiwan selectively face the social phenomena of globalization, education and commercialism within the wider Taiwan society. In chapters 7 to 9, there is also a bringing together of the innovative and traditional ends of the continuum in terms of the emergent qualities of performance for the Austronesian communities in Taiwan. In A’tolan, the texts and events of the singing and the choreography of the dancing blend both Han Taiwan, Amis and fourth-world themes, and coexist with traditional Amis singing and dancing. As Futuru C.L. Tsai relates, in the emergent quality of performance for A’tolan and other Amis, ‘The “difference” consists in a panindigenous identity created by them as they design their dance steps and incidental music’. The work put into the preparations for the performances of the dances and singing, especially but not only at the annual harvest festival, as well as the performances themselves, speaks to a strong sense of culturally distinctive Amis communal work ethic such that, ‘For Amis youth, the bodily expressions represent the performance of subjectivities and identities blended with the global flows of music and dance movements under the specific sociocultural environments of A’tolan Amis’. In Puleti, the texts in the singing and oral storytelling for weddings and associated activities provide a negative barricade whose purpose is to wall out the written histories of the village that circulate in the wider Taiwan society. In Ita Thao, texts and events provide an image of what, in the wider Taiwan society, the Thao are imagined to have been and what an ethnic group should be like, even though they are in fact otherwise described as highly assimilated into the wider Taiwanese Hokkien society in religion, language and economic life. The social structure is realigned to the rights and duties involved in the survival of an endangered ethnic group, with the Thao Tribal Cultural Development Association playing a pivotal role in directing that realignment. Some performative qualities, like female shamanist performances, are consciously played up; other features, like male elder leadership, are pushed, at least temporarily, into the shadows. The social structure at the end of these performances thus potentially differs from that at
232 James Wilkerson
the beginning of the performances in the sense that, as Yayoi Mitsuda describes it, ‘The eyes of outsiders become very important for both the Thao and the association, and they have started to care about how their rituals could be used to attract more media and scholarly interest and how financial aid can be obtained from the government’. Thao ethnicity now exists, she adds later, ‘to perform it for the outside world’. Chapter 10 by N. Jenny Hsu on the Nung in Dai An Commune, not so far from their ancestral homeland in Guangxi Province in what is now the PRC, is the only chapter on the Soviet Socialist Republic of Vietnam. The keying of the footings of the participants in the ‘Third Morning’ ritual is located in the classification of kinship statuses. For the domestic sphere, the presence of all members of the family of the first-born child is normally obligatory. There are two other categories of extra-domestic kin who seem to attend optionally. The first are extra-domestic agnates. The second is one of the two categories of affines: wife-givers (bung lang) and wife-takers (bung na). Whereas attendance by the representatives of the wife-takers is not singled out as necessary, attendance by the wife-givers is crucial to the success of the ritual. It is important to note that these two categories of affines, together with the category of agnates, sets up a situation in which there is a latent ranked directionality of marriage running from wife-givers to agnates to wife-takers, although there is no evidence that there is a positive sense in which this controlled affinity should close back upon itself to constitute a closed, endogamous pattern of marriage. In terms of the performance as text, there is a similar concern with the astrological fate of souls as established in written religious texts. This is because the fate of an individual turns on what is called the ‘Eight Characters’ (ba zi, in Chinese). These eight characters are broken down into four pairs to mark out the year, month, day and hour of birth, which determines their astrological alignments. N. Jenny Hsu notes the continued importance of fate (without explicitly mentioning its connection to the the eight characters) in determining the suitability of potential marriages in the Dai An Commune. In addition, she notes that the fates are also important in life-cycle and curing rituals in general, where ritual intervention and sometimes the presence of a ritual specialist and his use of other written religious texts is necessary to ensure the protection of a soul. In terms of the performance as event, the fates of the souls of children under fifteen years of age are under the control of the flower goddess, who has the power to alter an astrologically fixed fate. The presence of a representative from the wife-givers and the sacrifices he brings and offers to the flower goddess are integral to all ritual interventions in the fates of a soul, as in the ‘Third Morning’ ritual. Lastly, in terms of social structure, the ‘Third Morning’ ritual raises the kinship statuses of the family members through the birth of the first child and the subsequent presence of the wife-givers. Stepping back to compare the six chapters on the PRC, the three chapters on Taiwan and the single chapter on the Soviet Socialist Republic of Vietnam as a whole, the similarities are highly suggestive. First, the chapters on both the
Performance as a Mechanism for Social Change 233
PRC and Taiwan all show that traditional community social institutions were explicitly intended to serve community control, defence and even, although this is not directly mentioned, warfare. Secondly, the same chapters also show that these traditional community social institutions are invariably of long standing and predate much if not all of what is conventionally known Eurocentrically as the modern era. Thirdly, the innovative end of the continuum for both the footings and qualities of performance are also present in one configuration or another in all the communities of all these chapters. However, there are also important cross-cutting differences between all of these chapters. At the traditional end of the continuum, the community-level social institutions differ. First, the former imperial sub-bureaucracy is still making its presence felt directly or indirectly in all traditional community performances in the chapters on the PRC, but there is no evidence of the direct presence of the former imperial sub-bureaucracy for any of the communities in Taiwan. This is probably because historically the Amis community of A’tolan and the Paiwan community of Puleti were probably still well beyond the reach of the imperial state bureaucracy right up to the time of the imposition of Japanese colonial rule, although Ita Thao was probably at least indirectly under the sway of the sub-bureaucracy in terms of indirect taxation. Secondly, the positive sense of controlled affinity present in the two chapters on the Miao in the PRC is absent from all the Taiwan chapters. The only chapter on Taiwan that stresses controlled affinity is that on the Paiwan in Puleti, where there is status endogamy. Still, there is no evidence for a positive sense of controlled affinity that closes back upon itself in an endogamous pattern of marriage. Nor is there in the case of the Nung in the Soviet Socialist Republic of Vietnam, where there is a certain positive sense of directionality, but where there is no indication that the directionality closes back upon itself to form an endogamous pattern of marriage. In terms of the footings for the innovative end of performance, there is an important difference between the ethnographic descriptions from the PRC, Taiwan and the Soviet Socialist Republic of Vietnam. In the chapters on the three Austronesian communities in Taiwan, the ‘floating’ and ‘fluidity’ of the footings at the innovative end of the continuum are subordinated to the traditional footings of community-level social institutions. That is, an institutionalized interface exists between the traditional and innovative footings. More than just this, the interface is realized in distinct yet related ways in each of the communities. This pattern is wholly lacking in the chapters on the PRC and the Soviet Socialist Republic of Vietnam.
Concluding Discussion It was noted in the introduction that, just as originally for Western Europe, accounts of social change for Eastern Europe and East Asia are often given in evolutionary, functionalist or structuralist terms. Since such terms of discussion
234 James Wilkerson
firmly anchor explanations of modernity in pan-human terms, other culturally specific mechanisms for social change are easily obscured. Still, as likewise discussed early in this introduction, the existing literature on social change in East Asia has uncovered at least three empirical conclusions that can be related to performance as a community-level mechanism for social change. The first empirical observation was that there have been relatively substantial changes in community-level social institutions over its recent history. The second was the continued presence of state sponsorship. The third was the limited changes in political systems and institutions, including an accommodation of existing regimes positively to market economics and negatively to an absence of outright regime change. Consideration of the above three empirical generalizations about social change in East Asia in light of the ethnographic accounts in the individual chapters of this volume suggest future lines of inquiry in terms of mechanisms of social change that are potentially specific to East Asia. Similarly, the same reflections highlight the limits of coverage of the chapters of this volume. For reasons of space, consideration will only be given to the footings of performances as keyed to relationships external to the performances proper. First, as for the empirical generalization about the extensive changes that have taken place in community-level social institutions, these can related to the discovery in the chapters of this volume that these very same institutions still provide the ‘keys’ to the ‘footings’ of community-level performances. Secondly, as for the empirical generalization about strong state sponsorship in East Asia, state sponsorship also shares an ideological unity with community-level social institutions in so far as market economies are subordinated to both. Thirdly, as for changes in political systems and institutions that accommodate themselves to the presence of market economies without provoking regime change, saturation of local community social institutions with the effects of market economics remains just as much a potential threat to such institutions as it does to political systems and institutions. In sum, then, the findings of the chapters of this volume suggest that the continuities between the three empirical generalizations for social change across East Asia as a region are likewise present in community-level social institutions. However, consideration of the history of these continuities also shows other ways in which these continuities were worked out in different historical contexts. The first different historical context in the chapters in this volume is the contrast between those that belonged to what can be called the old empire and those that belonged to the new. The old empire is that part that was long under firm control in terms of the presence of the sub-bureaucracy at the community level; the new empire – which need not be a part of the PRC – is that part that was newly brought under control or only brought under control in colonial circumstances in terms of the presence of later political systems and institutions. To begin with the old empire, the basis of eligibility for sitting the imperial exams on the statuses of the sub-bureaucracy provides both with a specific sense of modernity (see Faure 2007 for the perspective from the sub-
Performance as a Mechanism for Social Change 235
bureaucracy, and see Woodside 2006 for the perspective from the imperial exams). Both the imperial exams and the sub-bureaucracy filled in the gap that ran from the village to the county. A ‘constitution’ based upon orthodox texts related the ideologies of state sponsorship between a bureaucracy staffed by degree holders and a sub-bureaucracy peopled by those who were eligible but had not passed the imperial exams (Kuhn 2002). It is also widely known that the sub-bureaucracy depended upon an integration with and reliance upon community-level social institutions such as lineages, temple organizations and other local kinship and territorial organizations to fulfil its administrative functions. When possible, these community-level social institutions were used to push orthodox state ideology into the countryside, although these same institutions could also serve as a conduit for heterodoxy (Watson 1987; Sutton 2007). As for the new empire, the chapters on the Miao in Guizhou, the Nung in Vietnam and the Austronesians in Taiwan deal with situations that were all located on the edge of imperial rule. The Miao were defeated and included in the old empire in the late Qing, and the Nung and Austronesians were only brought under direct state rule in, respectively, Japanese (Meyers and Peattie, eds. 1984) and French colonial times. It has already been observed that the former sub-bureaucracy was never directly integrated into community-level social institutions. For the Miao and Nung, various forms of controlled affinity formed and still form the core of such institutions, whereas for the Austronesians they revolved and still revolve around a variety of extra-kinship arrangements. This includes an age-grade system (A’tolan), a village aristocracy (Puleti) and village membership corporation (Ito Thao), where controlled affinity was not and is not the central organizing principle. The authors of the chapters on the Austronesian communities in Taiwan discuss an annual harvest festival (A’tolan), weddings (Puleti) and female spirit mediums (Ito Thao), where there is no clear comparative pattern to the traditional end of the performative continuum. However, the Miao (Shidong and Fangf Bil), Tibetans (Zhoukeji), Qiang (Nzidabu) and Nung (Dai An Commune), where more or less controlled affinity is present, do share a comparative consistency at the traditional end of the continuum in the shared emphasis upon life-cycle rituals. Although the political system and institutions may have remained comparatively unchanged over time in the old empire, both the imperial exams and the sub-bureaucracy experienced profound changes. The imperial exams ended in 1905, even before the fall of the dynasty in 1911. From the formation of the nationalist state in 1911 until its defeat in the civil war in 1949, a new sub-bureaucracy was imposed over those parts of the former empire under its control. This nationalist state sub-bureaucracy was intended to replace the former imperial sub-bureaucracy (McCord 1999; Chen Yung-fa 1980). In many respects similar to the former imperial sub-bureaucracy in its police and taxation functions, which relied heavily upon household registries, the important difference during nationalist rule and subsequently during the PRC (1949 to the present) was that a single-party bureaucracy was put in place in
236 James Wilkerson
tandem with the sub-bureaucracy (or communes over the first three decades of rule in the PRC) to ensure separate national-level surveillance and control over that sub-bureaucracy. However, various additional adjustments have been made in communities to meet the consequences of a rapid expansion of market economies in more recent decades since the beginning of the reform era, resulting in a corresponding challenge to the effectiveness of single-party control over the countryside. Specifically, in terms of the Leninist-style Communist Party in the reform and post-reform eras in the PRC (Saich 2000), as well as in the era of doi moi reform in the Soviet Socialist Republic of Vietnam (Guay et al. 2004), the party apparatuses and state bureaucracies have attempted to (re)build surveillance and control through experiments with alternate arrangements. In the PRC, this adjustment took place through an addition to the political systems and institutions of organizations that Saich (2000) translates as ‘social organizations’ or shehui tuanti (in order to avoid the associations with the term for non-governmental organizations). The explicit purpose has been to combat the saturation of ‘floating’ and ‘fluid’ footings brought on by the effects of the expansion of market economies that the party and government lacked the means to supervise and control. These social organizations have an enormous range. On the one hand, they include national and regional social organizations that are intended to take over social responsibilities where Leninist states have discovered that party and state entities lack the flexibility or resources to confront. In these circumstances, the relevant party and state apparatuses and bureaucracies select social organizations that first register with the government and then apply for affiliation and sponsorship with a matching party apparatus or state bureaucracy. On the other hand, and although less well studied, these social organizations also include community-level social institutions that have applied for and gained party and state sponsorship (especially lineages in Han Chinese regions). In Taiwan between 1945 and the mid-1980s, single-party rule prevailed. Even from the early years of nationalist single-party rule over Taiwan, there were also organizations that operated very much like the ‘social organizations’ that currently exist in the PRC and the Soviet Socialist Republic of Vietnam. Since the end of single-party rule and free multiple-party elections, beginning in the mid-1980s (Wong 2003), Taiwan has become home to an enormous number and range of outright non-governmental organizations (NGOs). How these NGOs operate at levels of organization above that of individual communities has been fairly well studied (Huang 2009; Madsen 2007, 2011). However, how they operate in Taiwan at the community level has been less well studied, especially in regard to how they both integrate and interact with the traditional end of such institutions (but see Wu Yingqing 2010 for an indepth community study). The contrasts between state sponsorship (‘fairness’) and market economics (‘commercial value’), introduced in the discussion in the introduction of the differing conclusions of Feuchtwang (2002) and Latham (2002), centres on
Performance as a Mechanism for Social Change 237
whether the Communist Party of China has become fossilized in terms of the impact of political fairness, or whether its powers have eroded in the face of the impact of commercial gratification. When viewed from the perspective of community-level social institutions, this contrast can be placed within the historical context of the old and new empires and of Leninist-style states and democratic states. The answers that the essays of this volume provide to this contrast between state sponsorship and fairness versus market economics and commercial gratification also involve the ways in which and degrees to which the relationship between market economies and community-level social institutions is controlled from either above or below. All the chapters on the PRC and the Soviet Socialist Republic of Vietnam – although caution is called for in this latter case – agree that community-level performances keep the traditional and innovative ends of the performance continuum separate in concentrating on life-cycle rituals. This is a part of a state ideology associated with state sponsorship that still emphasizes a political system and institutions that are, even with the appearance of ‘social organizations’, top-down. Whether historically within the local community in the sub-bureaucracy in the old empire or on the fringes of what was once the old empire and soon became part of the new empire, ritual interventions facilitate the proper course of the cosmos as it affects the well-being of the living and the dead, in so doing still attempting to subordinate the effects of market economies to community social institutions. In the former instance, this is close in some respects to the realm of carceral regimes and the externally imposed rituals that are expected to fathom and remove the powers of a distant centre of organized state violence (Mueggler 2001). In the latter instance, this is close to the anti-state ideologies that attempt to deny and escape the state (Scott 2009). Alternately, however, all the chapters covering the institutionalization of community-level performances in Taiwan agree that the consequences of the globalized market economy and the ‘floating’ and ‘fluid’ dimensions now be directly confronted through community-level social institutions unmediated by the direct involvement of the state, and even that the way in which this confrontation with the market economy takes place can vary between the different Austronesian-speaking local communities. In sum, the pivotal difference between the chapters on the PRC and the Soviet Socialist Republic of Vietnam on the one hand, and those on Taiwan on the other, is in the location of the relationship between political systems/ institutions and community-level social institutions. In the modified Leninist states of the PRC and the Soviet Socialist Republic of Vietnam, an effort is made to keep the innovative end of the performance continuum away from local-level community social institutions and within the control of the political institutions of the party apparatuses and state bureaucracies. In democraticoriented government in Taiwan, an effort is made to confront the innovative end of the performance continuum within the community-level social institutions.
238 James Wilkerson
Yet, although the chapters of this volume are suggestive in terms of their continued commitment to ethnographic description as a still viable stimulus for theoretical renewal, no effort has been made here to provide comprehensive coverage of community-level performance as a mechanism for social change. For one thing, even given that all of the chapters included in this volume are on East Asian communities, no effort was made to ensure full coverage of all parts of East Asia. For another, the coverage does not include Han Chinese communities, which if included would have shown that traditional communitylevel performances in the PRC and Taiwan typically include footings that are keyed by community-level social institutions with the continued strong presence of the former sub-bureaucracy, and are not known to have controlled affinity that resembles either the two Miao cases of Shidong and Fangf Bil in the PRC, nor of the Nung in Dai An Community in the Soviet Socialist Republic of Vietnam. Similarly, this volume does not pretend to have provided comprehensive coverage of performance in East Asia, nor, for that matter, even of East Asia’s community performance. Performance embraces so much and includes so many genres that the best that can be said is that it includes an emphasis on traditional community-level performances balanced with limited comparative coverage of tourism and migrant work. This leaves out everything from new media and the Internet through to the Peking Opera and professional sports. Moreover, upon reflection, the chapters of this volume do not engage with the always-near presence of ‘social organizations’ or NGOs. Although missing, however, their shadows are visible enough, and future studies can be expected to engage with this issue more directly and thoroughly. Finally, there is the all-important subject of censorship. Although there is wide coverage of both the traditional and innovative ends of performance, there is no coverage of the issue of the suppression of such performances. This subject has real importance for this volume, since one of its more important findings is that the PRC and the Soviet Socialist Republic of Vietnam approach the control of community-level performances in sharply different ways from Taiwan. In the PRC and the Soviet Socialist Republic of Vietnam, all local community performance is potentially subject to censorship, whereas in Taiwan state control over the media is much more lax, although not without some controls. Significantly for the purposes of this volume, it is much harder in the PRC and the Soviet Socialist Republic of Vietnam to stage community-level performances at the innovative end of the continuum than in Taiwan. It is likewise much more likely for party and state apparatuses and bureaucracies to attempt to guide and exploit community-level performances to celebrate and valorize party and state legitimacies.
Performance as a Mechanism for Social Change 239
References Bauman, Richard. (1977) 1984. ‘Verbal Art as Performance’. In Verbal Art as Performance, with Supplementary Essays by Barbara A. Babcock, Gary H. Gossen, Rodger D. Abrahms, Joel F. Sherzer. Prospect Heights, Ill.: Waveland Press. Brook, Timothy. 1989. ‘Funerary Ritual and the Building of Lineages in Late Imperial China’, Harvard Journal of Asiatic Studies 49(2), 465–99. ——— . 1999. The Confusions of Pleasure: Commerce and Culture in Ming China. Berkeley: University of California Press. ——— . 2010. The Troubled Empire: China in the Yuan and Ming Dynasties. Cambridge: Harvard University Press. Chen Yung-fa. 1980. ‘Rural Elections in Wartime Central China: Democratization of Subbureaucracy’, Modern China 6(3): 267–310. Duranti, Alessandro. (1997) 2002. Linguistic Anthropology. Beijing: Peking University Press. Elman, Benjamin A. 2000. A Cultural History of Civil Examinations in Late Imperial China. Berkeley: University of California Press. Fan, C. Cindy. 2008. China on the Move: Migration, the State, and the Household. London: Routledge. Faure, David. 2007. Emperor and Ancestor: State and Lineage in South China. Stanford: Stanford University Press. Feuchtwang, Stephan. 2002. ‘Remnants of Revolution in China’, in C.M. Hann (ed.), Postsocialism: Ideals, Ideologies and Practices in Eurasia. Cambridge: Cambridge University Press. Goffman, Erving. 1974. Frame Analysis: An Essay on the Organization of Experience. New York: Harper and Row. ——— . 1979. ‘Footing’, Semiotica 25, 1–29. ——— . 1981. Forms of Talk. Philadelphia: University of Pennsylvania Press. Gouin, Margaret E. 2010. Tibetan Rituals of Death: Buddhist Funerary Practices. London: Routledge. Greenhalgh, Susan. 2008. Just One Child: Science and Policy in Deng’s China. Berkeley: University of California Press. Guay, Terrence, Jonathan P. Doh and Graham Sinclair. 2004. ‘Non-governmental Organizations, Shareholder Activism, and Socially Responsible Investments: Ethical, Strategic, and Governance Implications’, Journal of Business Ethics 52(1): 125–39. He, Zhaosheng, and Jim Pooler. 2002. ‘The Regional Concentration of China’s Internal Population Flows, 1892–1990’, Population and Environment 24(2): 149–82. Herman, John. 2007. Amid the Clouds: China’s Colonization of Guizhou, 1200–1700. Cambridge: Harvard University Press. Hesketh, Theresa, Li Lu and Zhu Wei Xing. 2005. ‘The Effect of China’s One Child Policy after 25 Years’, New England Journal of Medicine 353, 1171–76. Huang, Julia C. 2009. Charisma and Compassion: Cheng Yan and the Buddhist Tzu Chi Movement. Cambridge, Mass.: Harvard University Press. Kuhn, Philip A. 2002. Origins of the Chinese State. Stanford: Stanford University Press. Latham, Kevin. 2002. ‘Rethinking Chinese Consumption: Social Palliatives and the Rhetorics of Transition in Postsocialist China’, in C.M. Hann (ed.), Postsocialism: Ideals, Ideologies and Practices in Eurasia. Cambridge: Cambridge University Press. Lew, Alan A., Lawrence Yu, John Ap and Zhang Guangrui. 2003. Tourism in China. New York: Haworth Hospitality Press.
240 James Wilkerson
Madsen, Richard. 2007. Democracy’s Dharma: Religious Renaissance and Political Development in Taiwan. Berkeley: University of California Press. ———. 2011. Religious Renaissance in China Today, Journal of Current Chinese Affairs 40(2): 17–42. McCord, Edward A. 1999. ‘Local Militia and State Power in Nationalist China’, Modern China 25(2): 115–41. Meyers, Ramong H. and Mark R. Peattie (eds). 1984. The Japanese Colonialization of Taiwan, 1895–1945. Princeton: Princeton University Press. Mueggler, Erik. 2001. The Age of Wild Ghosts: Memory, Violence, and Place in Southwest China. Berkeley: University of California Press. Saich, Tony. 2000. ‘Negotiating the State: The Development of Social Organizations in China’, The China Quarterly 161: 124–41. Schein, Louisa. 2000. Minority Rules: The Miao and the Feminine in China’s Cultural Politics. Durham: Duke University Press. Scott, James C. 2009. The Art of Not Being Governed: An Anarchist History of Upland Southeast Asia. New Haven: Yale University Press. Skinner, G.W. 1976. ‘Mobility Strategies in Late Imperial China: A Regional Systems Analysis’. In Carol A. Smith (ed.), Regional Systems Analysis, Volume 1: Economic Systems. New York: Academic Press. Sutton, Donald S. 2007. ‘Ritual, Cultural Standardization, and Orthopraxy in China: Reconsidering James L. Watson’s Ideas’, Modern China 33(3): 3–21. Took, Jennifer. 2005. A Native Chieftaincy in Southwest China: Franchising a Tai Chieftaincy under the Tusi System of Late Imperial China. Leiden: Brill. Watson, James L. 1987. ‘Standardizing the Gods: The Promotion of T’ien-hou’. In David Johnson, Andrew J. Nathan and Evelyn S. Rawski (eds), Popular Culture in Late Imperial China. Berkeley: University of California Press. Williams, Raymond. 1973. ‘Base and Superstructure in Marxist Cultural Theory’, New Left Review 82: 3–16. Wong, Joseph. 2003. ‘Deepening Democracy in Taiwan’, Pacific Affairs 76(2): 235–56. Woodside, Alexander. 2006. Lost Modernities: China, Vietnam, Korea, and the Hazards of World History. Cambridge: Harvard University Press. Wu, Guoguang. 2010. ‘China in 2009: Muddling through Crises’, Asian Survey 50(1): 25–39. Wu Yingqing 吳映青. 2010. ‘Kuhai yusheng: Nanfangao yuye gongzuo minzuzhi’ 苦海漁聲:南方澳近海漁業工作民族誌 (Fishermen Talk about Fishing: An Ethnography of Work for Offshore Fishermen in the Village of Nanfang Harbor). Hsinchu, Taiwan: Institute of Anthropology, National Tsing Hua University (MA thesis).
Notes on Contributors James Wilkerson is Associate Professor of Anthropology at the Institute of Anthropology, National Tsing Hua University, Taiwan. He received his Ph.D. in anthropology from the University of Virginia. He has conducted research on religion, kinship and social history in the Penghu Islands, Taiwan, and in the Guangxi Zhuang Autonomous Region, People’s Republic of China. He is currently doing research on national minority language policy in south-west China. Robert Parkin is a Departmental Lecturer in Social Anthropology at the University of Oxford, and was writer-in-residence at the Institute of Anthropology, National Tsing Hua University, Taiwan, in the summer of 2006. He has also taught at the University of Kent in Canterbury, Oxford Brookes University and the Free University of Berlin. His main thematic interests are in kinship, religion and symbolism, ethnicity, nationalism and regional identity, and the history of French anthropology, on all of which he has published extensively. Wang Ting-yu is currently a Ph.D. student at the Institute of Anthropology, National Tsing Hua University, Taiwan. He received his BA from the Department of Chinese Music, National Taiwan University of Art, and his MA from the Institute of Anthropology, National Tsing Hua University in 2007. During 2005–2006, he conducted fieldwork in Ngawa (Ngaba or Aba in Chinese) Tibetan and Qiang Autonomous Prefecture of Sichuan Province, China. His Master’s thesis, ‘The “House” and Kinship of the Sichuan Zhuokeji rGyalrong Tibetans’, discussed the house system, kinship and social structure of the Zhuokeji rGyalrong Tibetan population, with a special focus on the relationship between the tusi (local chief ) and social structure in Zhuokeji village. His main research interests are in music and linguistic anthropology, historical anthropology and kinship studies. Liu Biyun is a Ph.D. student at the Institute of Anthropology, National Tsing Hua University, Taiwan. She received her MA from the same institute in 2007, and her BA from the Department of Chinese Literature, National Sun Yat-Sen University. Her research focuses on social groupings on the periphery of Han Chinese societies, especially the influence of Han culture on the Qiang minority in north-west Sichuan through the military colonization of the Qing Empire. Her research also includes a comparison of Qiang lineages with those in south China during the Ming and Qing dynasties. She has participated in
242 Notes on Contributors
research projects on the social life of the local literati in north Taiwan during the Qing dynasty, which included making analyses of household registration records during Japanese colonization. Her main interests are in historical, social and linguistic anthropology, and more recently, the agricultural predicament in modern Taiwan and its related cooperative organizations. Ho Zhaohua is an Associate Professor in the Textiles and Clothing Department, Fu Jen Catholic University, Taiwan, and also the Executive Director of the Textiles Design Division at the same institute. She holds a BA in Chinese Literature and a Master’s in Textiles and Clothing from Fu Jen Catholic University. She received her Ph.D. from the Institute of Anthropology at National Tsing Hua University in 2011. Her thesis is entitled ‘Gifts to Dye For: Cloth and Person among Shidong Miao in Guizhou Province’. Her scholarly interests include the anthropology of cloth, anthropology of art, Chinese fashion history, Miao ethnography, textiles in museum displays and reconstructions of Taiwan aboriginal textiles. Chien Mei-ling is Associate Professor of Cultural Anthropology at National Chiao Tung University. She received her Ph.D. from National Tsin Hua University in 2002. Her fields of specialization include kinship, gender and the anthropology of the emotions among the Miao (Hmub) in south-west China, and the Hakka in southern China and northern Taiwan. She is the author of Sentiment and Marriage among the Miao in Eastern Guizhou (Guiyang: Guizhou University Press, 2009), and co-editor of The Hakka: Formation and Transformation (Hsinchu: National Chiao Tung University Press, 2010). Chang Kuei-min is currently working on her Ph.D. in Politics and Anthropology at Columbia University in New York. She received her MA in Sociology and China Studies at National Tsing Hua University, Taiwan, in 2007. Her thesis, entitled ‘Birth Control Politics among Urban Migrants in China’, focused on how the one-child policy was received by the targeted groups and its implications for the ideological dominance of the Chinese Communist Party. She has also conducted extensive field research on the citizenship rights of peasant migrant workers in urban China. Her main interests are Han nationalisms and their cultural and political consequences, especially within the ‘Han population’. Eveline Bingaman is a Ph.D. student at the Institute of Anthropology, National Tsing Hua University, Taiwan. She first came to Asia in 2001 and lived in the Chinese cities of Hangzhou and Lijiang before moving to Taiwan to pursue her MA at National Tsing Hua University in 2006. Her MA thesis, ‘What is Culture in Lijiang?’, explores the different concepts of culture as defined by UNESCO’s World Cultural Heritage Program, the Chinese state and Lijiang’s cultural tourism industry, and how these different concepts are interpreted by the local Naxi population that is the target of cultural
Notes on Contributors 243
preservation efforts. Her current research interests include minority nationalities in China, ethnicity, kinship and development NGOs. Futuru C.L. Tsai was a training manager in a semiconductor corporation before he quit his job to pursue a Ph.D. in Anthropology, which he received from the Institute of Anthropology at National Tsing Hua University in 2010; he is currently an Assistant Professor at National Taitung University. His dissertation is entitled ‘Playing Modernity: Play as a Path Shuttling across the Space and Time of A’tolan Amis in Taiwan’. Futuru is also an ethnographic filmmaker and writer, who has produced three ethnographic films: Amis hip hop (45 min, 2005), From New Guinea to Taipei (80 min, 2009) and The New Flood (51 min, 2010), and two books: The Anthropologist Germinating from the Rock Piles (Shiduei zhong faya de renleixuei jia) (2009) and From A’tolan to New Guinea (Cong Dulan Dao Xingjineya) (2011). Li-Ju Hong was a Senior Lecturer at the Art Center of Commission of General Education, National Tsing Hua University, Taiwan. She held a BA in English Literature from the National Taiwan University, and an MA in Art History from the University of Maryland, United States. Her Ph.D. at the Institute of Anthropology, National Tsing Hua University was left unfinished due to her untimely death due to breast cancer on 5 January 2010. Her scholarly interests included art history, the anthropology of art, contemporary Taiwan art history, Paiwan ethnography and museum art. Yayoi Mitsuda received her Ph.D. in Anthropology from the Institute of Anthropology, National Tsing Hua University, Taiwan, and is now a Postdoctoral Fellow at the Institute of Ethnology, Academia Sinica, Taipei. Her dissertation explores the survival strategies of Taiwanese indigenous people as a subaltern group, and explains how they exploit the ‘smallness’ of their population and their unprivileged position to increase their visibility among Taiwanese indigenes. She is currently researching the fields of shamanism, studies of memories and the historical anthropology of Taiwan indigenes. Her most recent essay on the history and politics of Taiwanese indigenous movements will be published in the book, Taiwan since Martial Law, in 2011. N. Jenny Hsu holds a BSc in Zoology and received her MA in Anthropology from National Tsing Hua University, Taiwan, in 2007. She was a visiting student at the Institute of Social and Cultural Anthropology at the University of Oxford from 2006 to 2007. Her MA thesis, entitled ‘The Flow of Rice, the Flow of Life: Exchange of Rice-based Foods of the Nung in Lang Son Province, Socialist Republic of Vietnam’ (2007) focused on religious and ecological issues among the Nung ethnic group of northern Vietnam. With fieldwork experience and a passion for tea, food and the green lifestyle, she is a co-founder of ADELLA: Natural Food and BeverageTM, currently marketing organic Taiwanese tea and granola snacks.
Index A’tolan (Amis village), 151–158, 160–176, 231, 233, 235, 243 aesthetic dimensions of kinship, 95–97, 103, 109–114 of Miao embroidery, 72–73 affinal alliance, 18, 99–100, 109–113, 213 Miao, 99–100, 109–114 affinity, 11–12, 18, 227, 230–233, 235, 238 African-American youth culture, 152 age grades (Amis), 156, 160, 161, 164, 166, 170 age organization (Amis) 154–155, 160, 164, 168–169, 172–173, 176 agency, 84, 89, 152, 175, 188, 192 altar, 24, 30 Nung, 216–218, 222 rGyalrong Tibetan, Buddhist, 27 rGyalrong Tibetan, community (Chank’skiyu), 23–24 rGyalrong Tibetan, house (choko), 30–31 alterity, 86, 88 Amis (Taiwan), 14–15, 18, 151–179, 198, 201, 231, 233, 243 ancestors, 10, 16, 17, 23 Amis, 155, 156, 167, 175 Miao, 101 Nung, 211, 216, 217, 218, 219, 220 Paiwan, 183, 192, 194 Qiang, 39–42, 45–61 rGyalrong Tibetan, 24, 29–30 Thao, 203, 206 ancestral halls (citang, Qiang), 37, 40–42, 48, 52, 55, 58 Citang yinzi (Qiang, Silver for the Ancestor Hall), 52, 58, 59 ancestral rituals Amis, 173 Qiang, 40–42, 52 Anti-Japanese War, 64 Appadurai, Arjun, 134, 148 assimilation, 3, 63, 67, 87, 135, 145 au lam (bringing the basket, Nung), 17, 209, 221 Australia, 153 authenticity, 14, 18, 90, 138, 143–148, 201, 231 as a UNESCO World Heritage qualification, 142–147
Baisha, 141–142 barbarians, 63, 198, 201 bend hmub (embroidery catalogue, Miao), 73, 90 birth control, 119–131 birth order, 50–51 Black Miao, 97 Bloch, Maurice, 20, 151, 157, 177, 192–194 body, 117, 120, 153, 154, 158–171, 215, 217 boundaries, 11, 16, 52, 62, 67, 75, 80, 85–89, 121, 128, 131, 135–138, 144–145, 153, 180–182, 205, 215, 227 bubble concept of culture, 135, 140, 145, 147 bung lang, see wife-givers (Nung) calendar, 35–36, 64–65, 209, 218–219 Chinese Lunar usage among rGyalrong Tibetans, 35–36 Miao Traditional, 64–65 Nung, 209, 218–219 capitalism, 5, 7, 9, 153 Chank’skiyu, see altar, rGyalrong Tibetan, house chaosheng youji dui (excess-birth guerillas, Mandarin), 125 Chen Shui–bian, 200 chengding menhu (succeed and head the house, Qiang), 48 China Youth Corps, 159, 176 choko, see altar, rGyalrong Tibetan, community Christian influence among Paiwan, 184, 185, 187, 189, 191 among Thao, 203 citang (Qiang), see ancestral hall citizenship, 119–120, 129–130, 242 civil rights, 120, 130 Clastres, Pierre, 93 clothing (Miao), 62–68, 75, 91, 242 cognition, 112, 151, 152, 157, 193 colonialism, 1–6, 17, 90, 153 commercialization, 13, 135, 142–146, 151, 165, 169, 227, 231 commodification, 90, 135, 143, 145, 146 communal stele (zongbei, Qiang), 24, 27–28, 31, 40–48, 55, 58, 114, 231 commune, 16–17, 33, 120, also see Dai An Commune Community Empowering Society (CES), 201 competition, 74, 102, 113 Amis dance, 159–164, 167, 169, 172
246 Index
Confucianism, 39–40, 43, 45, 59, 63, 227, 228 education among Miao, 63 influence and education among Qiang, 39–45 conscription, 53, 54 consumption, 14, 32–33, 62, 211 contestations of status, 181, 185 coufang (lineage organizations, Qiang), 10–11, 37–59, 226–227 courtship (Miao), 100, 111, 227 crematorium (Qiang), 41–42, 52–53, 58 cross-cousin marriage, 103, 110 Cultural Heritage Management Bureau (Lijiang), 144, 147 cultural ideal, 26, 32, 35, 65–66, 82, 85, 90, 93, 98–99, 102–103, 109–114, 128, 131–132, 188 cultural preservation, 13–16, 134–135, 138–139, 144–149 cultural reproduction, see reproduction, cultural and social cultural revitalization, 176, 180 Cultural Revolution, 5, 9, 31, 74 Dai An Commune, 17, 210–211, 214, 221 dandarling (Tibetan Buddhist temple), see gompa danwei (work unit, Mandarin), 120, 131 darndasda (graveyard, rGyalrong Tibetan), see graveyard daughter-in-law (xifuzi, Qiang), 41 Dayan (Lijiang Old Town, PRC), 140, 144 delayed transfer marriage, see duolocal residence Democratic Progressive Party (DPP), Taiwan, 16, 189, 200, 206 descent, rules of (Qiang), 45, 49 development, 3, 16, 19, 120, 130–133, 144, 146, 149, 202, 231, 240 dividuality (M. Strathern concept), 96 dongba (shaman, Naxi), 146–147 Dongba Culture Research Institute, 146 Dragon Boat Festival (Miao), 64 Dumont, Louis, 93–95, 113–115 duolocal residence, 99, 100, 112 Durkheim, Emilé, 1, 19, 93–95, 111, 114, 116 earthquake Lijiang, 140 Sichuan, 55 Taiwan, 199 Eastern Europe, 3–9, 233 economy and economic systems, 5–11, 13– 14, 17, 22, 32, 54, 60, 90, 117–121, 126–131, 140, 153, 201, 211, 213, 231 education, 23, 27, 39, 64, 118, 120, 122, 126, 130–131, 145, 200, 206, 211, 231
embroidery design, 73, 83, 87 enclaves, 11, 227 endogamy, 15, 104, 109–110, 233 Eriksen, Thomas, 205 Ertai, 63 ethnic minority policy (China), 63 ethnicity, 15, 20, 138, 148–149, 153, 178, 189, 197, 199, 201, 203, 205–207, 232, 241, 243 ethnohistory, 188 exogamy, 11, 49, 53 family names (Paiwan), 182, 184, 186, 194 Fangf Bil (Miao village), 12, 97–102, 109, 112, 114, 227–230, 235, 238 fashion, 46–47, 66, 73–74, 83, 86 fenjia (house division, Qiang), 48–49 Feuchtwang, Stephan, 5, 7–9, 19, 193, 236, 239 filial piety, 40, 45 flirting, 12, 99–114 floating population, 12, 121, 124–125, 128, 132 footings (E. Goffman concept), 225–238 functionalism, 1–2, 7 funeral customs, 28, 30, 32, 33, 53, 211 gaitu guiliu (shift to direct rule, Mandarin), 54, 55, 61 gender relations (Amis), 152, 171–173 gender roles (rGyalrong Tibetan), 28, 30 genealogies, 8, 10, 43, 55, 98, 181–184, 191–194, 216, 221 Nung, 216, 221 Paiwan, oral versus written, 181–184, 191–194 Qiang, 43, 50–52, 55 globalization, 135, 175, 231 Goffman, Erving, 18, 225, 239 gompa (Tibetan Buddhist temple), 23, 24 governance, 57, 119, 120, 124, 129, 130, 186, 191 graveyards Darndasda (rGyalrong Tibetan), 24 Qiang, 37, 40–43, 47, 54, 58 Han Chinese, 11, 22–24, 29–30, 35–39, 56–57, 60, 63–66, 77, 80–83, 86–88, 97, 186, 192, 238 Han Chinese characters, use of, 43, 47, 75, 81, 83, 88, also see written script Han Chinese surnames, 26, 35, 37, 39, 42, 55, 57, 97 Han migration, 22, 30 Hawaii, 91, 148, 149, 179, 192 hecou (merging of lineages, Qiang), 53 hegemony, 153 heritage, 13–14, 91, 134–138, 140–148, 165, 167, 169, 170, 175, 188 heteroglossy, 96
hierarchy, 15, 41, 42, 50, 227, 230 Amis, 157, 158, 160, 166, 169, 170–173 Paiwan, 181, 185, 193 history oral, 180, 185–186, 189, 193 written, 15, 93, 181, 191, 192, 193 hitching posts (xiama zhuang, Qiang), 41, 58 Hmong, see Miao Hobsbawm, Eric, 4, 19, 183–184, 195 homo duplex, 94 house names, see named house huafan (‘semi-cooked’ barbarians, Mandarin), 198 hukou (household registration system, Mandarin), 119–122, 124–128, 131 hybridity, 27, 35–39, 42–43, 54 iconic (dimensions of kinship), 95, 96, 109, 111 identity, 4, 7, 17, 27, 30, 35, 62–67, 86, 88, 90, 93, 113, 139, 143, 152–153, 168, 173–175, 189, 198, 201, 205–206, 231, 241 imagery, 74, 77 imperial military network, 37 imperial sub-bureaucracy, 10, 22, 34, 226– 228, 230, 233–238 individual and individuality, 2, 11–12, 17–18, 26, 42, 48–49, 67, 88, 93–97, 100– 102, 106, 109–114, 119, 125, 140, 144, 193, 227–228, 232–236 inheritance, rules of, 25–26, 29, 30–32, 35, 48, 51, 54, 56, 167, 183, 226 innovation, 18, 74, 160, 177, 226–233, 237–238 Ino Kanori, 186, 198, 207 insurance, 120–123, 131 intangible heritage, 138, 143, 145 integrity (as a World Heritage qualification), 138, 145 iron trivet, 29 Ita Thao (Thao village), 203, 206, 230, 231 Jakobson, Roman, 96, 115 Jamaicans, 153 Japan, 5–6, 184 Japanese colonial era, 15, 16, 181, 186, 194, 197–201, 206, 233 Japanese colonial government, 186, 194, 200, 206 Jinchuan War, 22 ka (kitchen, etc., rGyalrong Tibetan), 27–31, 35 Kai Yiu Chan, 198 kaka (older brothers and sisters, Amis), 158, 163 kapot (age set, Amis), 155, 158–166, 169, 173, 176–177 keys (E. Goffman concept), 17, 225–234
Index 247
kinship, 5, 18, 43–44, 47, 49, 50, 54–55, 57, 59, 93, 95–100, 109–110, 154, 172–173, 209, 216, 219, 221, 232, 235, 241, 242, 243 kinship statuses, 232 kinship terminology, 43, 59, 98, 100, 216, 219, 221 Kojima Hoshimichi, 186, 195 Kuomintang (KMT), 6, 15, 199–200, 206 Lamont, Michèle, 86 landowner, see embroidery design Lang Song Province, Vietnam (Nung), 212–218 languages preservation of, Nung, 212 preservation of, Thao, 202–205 Lévi-Strauss, Claude, 1, 2, 21, 36, 182, 195 life-cycle rituals, 16, 40, 211, 226, 227, 228, 232, 235, 237 lijia (unit of late imperial sub-bureaucracy, Mandarin), 38 Lijiang (Naxi), 13–15, 135, 137, 140–149, 228, 230, 242 Lijiang Old Town, 137, 140–148 lineages, 38, 41, 46, 51, 101, 154, 235, 236, 241 literacy, 3, 15, 22, 37, 39, 43, 45, 59, 64, 82, 88, 91, 93, 182, 231 Liu Hai, 80 localism, 121 loopholes (in one-child policy), 122–126 malikoda (traditional dance, Amis), 157–159, 163–164, 166–169, 174 mamazangilan (aristocratic family, Paiwan), 181, 183, 188–189 Mao era, 8, 146 marriage, 2, 11–12, 15, 17–18, 25–27, 47–52, 59, 64, 66–67, 84, 89, 93, 97–114, 125–128, 176, 182–183, 192–193, 213, 216, 218, 220, 227, 232–233 material culture heritage, 138 Miao embroidery, 62–89 Qiang house, 27–30, 35 matrilineality, 154 Mauss, Marcel, 1, 19–20, 93–95, 115–116 me mu (mother goddess, Nung), 17, 210, 213, 215–218, 220, 221 me sheng (god of life and teenagers, Nung), 215, 218, 220 Mead, Margaret, 154, 178, 180, 194–195 memory, 11, 15, 35, 40, 48, 52, 182–185, 192–193 Mexican Americans, 153 Miao (PRC), 11–12, 18–19, 62–94, 97, 99– 116, 227–228, 233, 235, 238, 240, 242 migrant workers, 13, 121, 125, 129, 228–229, 230, 242
248 Index
migration, 5, 13, 18, 23, 89, 118–121, 129, 190, 191, 220 milimilingan (genre of oral history, Paiwan), 15, 184–186, 190 military, 10, 14, 37–39, 42–47, 53, 55, 57–58, 63, 154, 226, 229, 241 mobility, 23, 56, 120–121, 125, 128–130, 227, 229–230 modern dance (Amis), 160–165 Molnár, Virág, 86 morally autonomous individualism, 93, 114 Mori Ushinoshuke, 186, 199, 208 mother goddess, see me mu mountain god (rGyalrong Tibetan), 30–32 named house, 9–10, 22–27, 185 naming practices, 26, 46, 95, 97, 119, 155, 183, 216 naming system, see naming practices National Taiwan University, 178, 199, 241, 243 Naxi (PRC), 13, 140–149, 242 Naxi Culture Development Company, 146 Nenozo Utsushikawa, 199 Nung (Vietnam), 16–18, 209–224, 232–235, 238 Nyingma school of Tibetan Buddhism, 23 Nzidabu (Qiang village), 10, 37–39, 42, 44, 47, 50–56, 226–230, 235 Okamatsu Santaro, 186 one-child policy (PRC), 12–13, 18, 117–119, 125, 129, 130–131, 229, 242 origin narratives, 182, 185, also see milimilingan; taucikel origin stories, see origin narratives Outstanding Universal Value (OUV), 136–137, 143 Paiwan (Taiwan), 15, 18, 180–198, 230, 233, 243 Pakalongay (teenagers who serve others, Amis), 155–156, 163, 169, 172 pan-indigenous identity (Taiwan), 152, 168, 173–174 Pan Lifu, 189–191, 195 patrilineality, 16, 45, 49, 55, 57, 97–99, 102, 173, 204 performance, 13–14, 17–18, 93, 96, 97, 100– 103, 109, 111–114, 135, 158, 160, 164, 173, 175, 192, 202–205, 225–238 plays (Amis), 152, 157–160, 167, 169–173, 176 poetics, 96–97 policies, 14, 121, 144, 197, 200, 205, 206 political rights, 120 prestige, 3, 74, 139, 183 propaganda, 118, 122, 124 public and private spheres, 160, 164 Puleti (Paiwan village), 15, 181–190, 193
Qiang (PRC), 10, 18, 22, 37–61, 226–227, 235, 241 Qianlong period, 42 rahbo (fireplace rack, rGyalrong Tibetan), 27, 29 recognition Lijiang, 140 Thao, 16, 181, 197–205 reconstruction of Thao culture, 200, 202, 205 reform, 5, 7, 9, 10, 117–118, 120, 129, 131, 228, 236 Reform and Opening Up Era, 73 reproduction, social and cultural, 42, 47–48, 62, 64, 85, 175 reproductive rights, 13, 119–120, 129–130 reproductive security, 13, 118–130 Republican Era, 63 resistance, 9, 13, 117–120, 124, 129–131, 174, 191–192, 228 revitalization of culture, 147, 189 rewakang (household space for worship, rGyalrong Tibetan), 30– 31 rGyalrong Tibetan (PRC), 21–35, 38, 241 ritual practices, 10, 43, 47, 55, 58 ritual specialists, 33, 232 Nung, 211–213, 217–218, 220, 222 ritual versus non-ritual, 152 rituals and rites, 2, 24, 228, 231–232, 235, 237 Amis, 151–153, 169 Miao, 64, 98 Nung, 211, 213, 215, 217–221 Paiwan, 182 Qiang, 40–41, 53 rGyalrong Tibetan, 30–33 Thao, 200, 203–205 Ryuzo Tori’i, 198 Saisiyat (Taiwan), 198, 206 seating customs (rGyalrong Tibetan), 29 sentiment, 93, 115, 242 shamans (shinshi, Thao), 16, 203–206, 231 shehui tuanti (social group, Mandarin), 236 Shidong (Miao township), 11–12, 19, 62–91, 227–230, 235, 238, 242 shinshi (shaman, Thao), see shamans shoufan (‘cooked/ripened’ barbarians, Mandarin), 198–199 Shuhe (PRC), 140, 142, 147 Sisters Rice Festival (Miao), 64 Situ, 22, 38 sjangku (fire, rGyalrong Tibetan), 27–29, 35 social change, 226, 234 Tibetan secular, 21 social ideals, see cultural ideals social memory, see memory social reproduction, see reproduction, cultural and social
social security, 120 social status, 11, 28, 30, 47, 48, 49, 74, 86, 88 social structure, 35, 94, 227, 229 socialization process, 152, 168 soldier-settler, 63 spectators, 158, 160, 165, 167, 169, 174–175 spirit mediums, 16, 204, 235 state-owned enterprises, 120–121 stitches (of Miao embroidery), 74, 77 Strathern, Marilyn, 4, 20, 94–96, 111–116 structuralism, 1–2, 7, 233 subaltern, 200, 243 sub-bureaucracy, see imperial subbureaucracy subjectivity, 152–154, 167, 172–175, 192 succession, 10, 11, 25, 183, 185 surveillance, 127–129, 236 suzhi (quality, Mandarin), 118 Taiwan Historical Research Commission, 187, 188 Taoist priest, 82, 213, 219 taucikel (genre of oral history, Paiwan), 15, 184–186, 190 taxation, 10, 25–26, 29, 32–33, 38, 54, 187, 198, 235 tension, 74, 94, 95, 101–102, 109, 111, 144, 191 Thao (Taiwan), 15–18, 197–208, 230–235 economy, 200 Thao Tribal Culture Development Association, 202 thu (bamboo basket, Nung), 215, 221 Thunder God (Miao), 82–83, 85, 90 Tibetan Buddhism, 21, 23, 34, 42 time, Miao social construction of, 64 tombstones, 39–47, 55, 57–59 Tong Chunfa, 187–188, 191, 195 tourism, 9, 10–14, 29, 30, 34, 35, 64, 135, 137–148, 151, 165, 198–205, 228–231, 238, 242 traditional dances (Amis), see malikoda Trinidad, 175 Tsou (Taiwan), 197, 199, 205–206
Index 249
tusi (native chieftain), 9–10, 2–35, 38–39, 43, 47, 54, 56–58, 61, 241 ulalaluan (ancestral basket, Thao), 16, 203–205 UNESCO, 13, 135–149, 242 uxorilocal marriage, 48, 52, 59 Uzbekistan, 175 va (bamboo platform for worship, Nung), 17, 209–210, 213–222 Warlpiri, 154 welfare, 3, 6, 120–121, 130–131, 200, 202, 218 wife-givers, 17, 52, 59 Nung, bung lang, 209–210, 213, 215–219, 221–222, 232 wife-takers, 52, 59 Nung, 213, 216, 217, 232 World Heritage Committee (WHC), 12–15, 19, 132–149, 154, 198–199, 229, 240, 242 World Heritage Fund (WHF), 136, 139 World Heritage List, 137–139, 149 written history, see history written script, 15, 35, 42–43, 46–47, 57, 82–83, 88, 93, 97, 117, 120, 130, 146, 180, 186, 189, 191–194, 217, 221, 227–228, 231–232 xiama zhuang (Qiang), see hitching post xiaokang shehui (relatively well-off society, Mandarin), 117 xifuzi (Qiang), see daughter-in-law Yang, Mayfair, 134 yellow card, 142–143, 148 yi yi zhi yi (using barbarians to govern barbarians, Chinese), 63 Yongzheng Era, 97 youth cultures, 152 Zhang Guangsi, 63 Zokjibu (rGyalrong Tibetan village), 22–36 zongbei, see communal stele