Social Dynamics in the Highlands of Southeast Asia (Handbook of [PDF]

racy' were western concepts applied to the Nagas and, as such, did ...... Lazcano Nebreda, S., 1998, “La cuestion de l

52 downloads 44 Views 4MB Size

Recommend Stories


The Future of Southeast Asia
Before you speak, let your words pass through three gates: Is it true? Is it necessary? Is it kind?

Kingdoms of Southeast Asia
Never let your sense of morals prevent you from doing what is right. Isaac Asimov

Trends in Southeast Asia
Where there is ruin, there is hope for a treasure. Rumi

terrorism in southeast asia
When you do things from your soul, you feel a river moving in you, a joy. Rumi

Nationalism in Southeast Asia
Everything in the universe is within you. Ask all from yourself. Rumi

Piracy in Southeast Asia
The wound is the place where the Light enters you. Rumi

Islam in Southeast Asia
What we think, what we become. Buddha

Trends in Southeast Asia
Your big opportunity may be right where you are now. Napoleon Hill

Trends in Southeast Asia
If you want to go quickly, go alone. If you want to go far, go together. African proverb

Trends in Southeast Asia
Make yourself a priority once in a while. It's not selfish. It's necessary. Anonymous

Idea Transcript


Social Dynamics in the Highlands of Southeast Asia

Handbook of Oriental Studies Handbuch der Orientalistik Section Three Southeast Asia

Edited by

V. Lieberman M. C. Ricklefs

VOLUME 18

Social Dynamics in the Highlands of Southeast Asia Reconsidering Political Systems of Highland Burma by E. R. Leach

Edited by

François Robinne and Mandy Sadan

LEIDEN • BOSTON 2007

On the cover : Rukyen Palaung Woman (Ta-An), Northern Shan State. Photograph by François Robinne This book is printed on acid-free paper.

ISSN 0169-9571 ISBN 978 90 04 16034 7 © Copyright 2007 by Koninklijke Brill NV, Leiden, The Netherlands. Koninklijke Brill NV incorporates the imprints Brill, Hotei Publishing, IDC Publishers, Martinus Nijhoff Publishers and VSP. All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher. Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill NV provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA. Fees are subject to change. printed in the netherlands

For my father, with love, respect and gratitude M.S.

Measuring a gong offered by the dama (wife takers) to the mayu (wife givers) during a Jinghpaw wedding ceremony at Myitkyina, Kachin State. Photograph by FranDois Robinne

CONTENTS Preface ......................................................................................... François Robinne and Mandy Sadan

ix

List of Contributors ....................................................................

xvii

Introduction: Notes on Edmund Leach’s analysis of Kachin society and its further applications ......................................... F. K. L. Chit Hlaing (F. K. Lehman)

xxi

Part One Historical Contexts of E. R. Leach’s Fieldwork The biographical origins of Political Systems of Highland Burma ... Robert Anderson On the continuing relevance of E. R. Leach’s Political Systems of Highland Burma to Kachin Studies ..................................... Maran La Raw Translating Gumlau: History, the ‘Kachin’ and Edmund Leach ....................................................................................... Mandy Sadan

3

31

67

Part Two Comparative Approaches in Assam and Laos Politico-ritual variations on the Assamese fringes: Do social systems exist? ........................................................................... Philippe Ramirez Naga ethnography and Leach’s oscillatory model of gumsa and gumlao ............................................................................... Pascal Bouchery Interethnic systems and localized identities: The Khmu subgroups (tmoy) in North-West Laos ..................................... Olivier Evrard

91

109

127

viii

contents

From Kettledrums to Coins: Social transformation and the ow of valuables in Northern Laos ....................................... Guido Sprenger Political hierarchical processes among some highlanders of Laos .................................................................................... Vanina Bouté

161

187

Part Three The Kachin Subgroups Rethinking Kachin wealth ownership ........................................ Ho Ts’ui-p’ing The Missing Share: The ritual language of sharing as a ‘total social fact’ in the eastern Himalayas (northwest Yunnan, China) ....................................................................... Stéphane Gros Transethnic social space of clans and lineages: A discussion of Leach’s concept of common ritual language ......................... François Robinne Postscript: Reconsidering the dynamics of ethnicity through Foucault’s concept of ‘spaces of dispersion’ .......................... François Robinne and Mandy Sadan

211

257

283

299

Bibliography ................................................................................

309

Index ...........................................................................................

325

PREFACE François Robinne and Mandy Sadan This volume originated from a panel held to mark the ftieth anniversary of the publication of Edmund Leach’s classic study Political Systems of Highland Burma: A Study of Kachin Social Structure (1954), which was organised for the EUROSEAS Conference held at the Sorbonne University, Paris, in September 2004. The ftieth anniversary, in combination with the location of the conference in France, presented a timely opportunity both to reconsider the signicance of Leach’s seminal text as well as to continue his original scheme of making the work an engagement between Anglophone and Francophone anthropology (which, in Leach’s case, meant an engagement with Lévi-Strauss in particular). However, we are delighted that we have been able to include contributions in this volume that succeed in breaking down those limiting parameters of discourse and which incorporate other spatial, relational, discursive and cognitive models of ‘Kachin’ than Leach was able to present. All of the papers presented at the conference have been included in this volume, with two additional contributions from Vanina Bouté and Guido Sprenger. The Introduction by U Chit Hlaing (Professor F. K. Lehman) has been developed from comments that were made on the day in the key role that he played as Discussant. For ease of reference, Leach’s work is referred to as Political Systems throughout. The eldwork underpinning Political Systems was begun as World War II approached, and was continued in its midst. The post-war environment in the Kachin region of northern Burma, however, was not amenable to the pursuit of academic research by Western anthropologists and, after publication, Edmund Leach was largely forced to divert his interests to other places and anthropological subjects (Anderson, this volume). Nonetheless, Leach’s name was established by the publication of Political Systems and its impact and implications extended far beyond the Kachin case-study at its core. Indeed, the theoretical perspectives of the work have proved to be signicant in the development of Social Anthropology in the longer term, not least through Leach’s analysis of social change considered through the articulation of structural variation and relational dynamics. The contributions that Leach’s work has made

x

françois robinne and mandy sadan

to Southeast Asian Studies, to Social Anthropology, and to theories of the relationship between Anthropology, History and Ethnicity have been extensive. Indeed, it is hard to think of other, similar texts that might warrant such an intensive reconsideration in the form of conference panels and published volumes after fty years. The fact that such an exercise has proved to be so useful after such a time attests to the ongoing signicance of Leach’s work. However, the value of this reconsideration arises from the fact that, whilst the text remains a standard of university reading lists and academic bibliographies to this day, some of the more contentious aspects of Political Systems (which are detailed fully in the chapters to follow) are as well known as its groundbreaking ideas on relational identities. Indeed, two generations of students can attest to the fact that Political Systems is a perennially difcult work to critique. Up to now, most criticism of Political Systems has been conducted at an abstracted theoretical level in the absence of empirical eldwork data from the Kachin region. Fifty years after publication, however, new research has been conducted in a number of settings related directly to the Kachin social environment, as well as with other ‘highland’ communities in Southeast Asia, the ndings of which are of pertinence to some of Leach’s central ideas. These developments make a reconsideration of Political Systems at this time a relevant and useful exercise, of benet to our understanding of Historical Anthropology in general and Southeast Asian Studies in particular. It is intended that this volume will assist the process of opening up Leach’s text to important new readings on issues such as social dynamics and identity, postcolonial studies, the politics and ethics of eldwork and of community representation, amongst others. In writing Political Systems Leach hoped to challenge a number of theoretical approaches that were then current in British Social Anthropology. He hoped to contrast his work with the essentialist approach still prevailing in much academic writing originating from the colonial sphere; he also established a critical distance from Durkheim and Malinowski and ideas of the duality of the sacred and the profane. Most signicantly, however, he hoped to challenge the claims of structuralism as a universal and a-temporal model by positioning Political Systems theoretically around the concept of “unstable equilibrium”. Yet, paradoxically, the strength of this conceptual approach appears also to be its greatest weakness and has resulted in a highly ambiguous book in which the data and the theory seem especially ill at ease with each other. Indeed, one might with justication accuse Leach of reducing

preface

xi

the Kachin sphere to a kind of intellectual laboratory without any expression in reality because of the way in which he moulded his case study to a theory, rather than the other way round. Furthermore, the large degree of difference evidenced between Leach and local analysis of ‘Kachin’ social practices goes beyond the simple manifestation of a ‘disagreement’, suggesting that something more essential and substantive in the politics of representation is at play in local objections to the work and the basis of its underlying construction (Maran La Raw, this volume). Leach himself vindicated such a distance by a self-professed indifference towards ‘the facts’, commencing Chapter VIII of Political Systems, ‘The Evidence from Kachin History’, with the statements that anthropologists always faced problems as to “what to do with the facts” and that facts “frequently bored” him; a problematic position to say the least given the intrinsic (and thus uneasy) relationship between History and Anthropology maintained throughout the book. Some fty years later, this volume explicitly attempts to test ‘the facts’ presented in Political Systems. However, our purpose is not to prove or disprove ‘facts’ alone in support or otherwise of Leach’s arguments. Neither do we reject the signicance of the critiques of Leach made by those such as LéviStrauss, Friedman and Nugent, amongst others, whose theoretical insights highlighted important issues relating to Political Systems over the course of three decades following its publication. Rather, the aim here is to reassert the theoretical and empirical importance of any ‘facts’ by considering in detail throughout the volume how Leach’s notion of structural oscillation ‘ts’ across both space and time, in Kachin society and beyond, and to raise both theoretical and empirical questions in places where that t is not seen. The perspective adopted is not that facts and theory should be placed in a hierarchical opposition with each other, but that they should be intertwined in any representational discourse. The chapters presented here do this in a way that is unique in relation to Leach’s work, demonstrating a considered blending of theory and empirical data. Such a critique, however, does not mean rupture; this is not simply a work of deconstruction, an attempt to list the mistakes that sprinkle Leach’s demonstration. It consists, on the contrary, of an attempt to reect on the perspectives that Political Systems continues to offer, and to locate those in an appropriate historical timeframe, cognizant of the gap of fty years and the local and regional transformations that have taken place during that time. Thus, we inscribe ourselves in terms of metamorphoses, rather than in terms

xii

françois robinne and mandy sadan

of rupture stricto sensu. Our ability to lay claim to this is in no small measure due to the support and insights of U Chit Hlaing (Professor F. K. Lehman), to whom we express here our sincere gratitude, and who, by agreeing to comment on our papers during the conference and to reformulating those comments for our Introduction, has helped to create a more direct connection between ourselves and Leach than might otherwise have been possible, and which was surely necessary. Many thanks are also due to Professor Maran La Raw for his personal insights into Leach’s relationships with Kachin people, and his encouragement of the preparation of this collective volume. Because of its location at the crossroads of China, India and Southeast Asia, the Indo-Chinese massif, of which the Kachin region is a part, is the site of a complex intersection of populations. These varied societies even today display exceptional linguistic diversity and organise a wide range of socio-political systems and religious practices—a diversity that was possibly even more nely drawn in Leach’s day. Communication between these different social spheres, combined with the dynamics of economic, matrimonial and commercial networks, produced those very social and cultural differences that seemed to conrm to Leach that the region and its people existed in a state of “unstable equilibrium”. The historical strategic importance of this area and its resultant entanglement in decades of local, regional and global political and military conict resulted in the transformation of this “unstable” region into a ‘restricted area’ for most of the twentieth century. Only in the 1990’s was a slight loosening of this restrictiveness seen, and it is as a result of this that some researchers were able to breach previously narrow boundaries of contact and to extend considerably the range of investigation possible in relation to historical and anthropological studies of ‘the Kachin’ and related communities. This renewed direct encounter informs many of the chapters in this volume and enables the theoretical and empirical engagement with Political Systems to go some way beyond that which was possible in previous critiques. In order to understand the ambiguity that characterizes Political Systems we have tried in this volume to explore Leach’s text in the context of the intellectual and political environments in which Leach found himself. For example, Leach’s idiosyncratic representation of historical context is given much attention. In Political Systems historical contingency is reduced to the presentation of structural variables within Kachin society; these then underpinned the transformation of Leach’s Kachin case-study from being a local demonstration to being a work of general

preface

xiii

theoretical importance, as was Leach’s aim. As will be discussed in the chapters to follow, Leach’s ‘distance’ from the historical facts, however, also led to some distance being created between Leach’s analysis of other important aspects of Kachin society, particularly the diversied nature of economic exchanges and the importance of networks underlying any expression of ‘social oscillation’. Furthermore, whilst Leach attested that the Kachin borrowed their hierarchical system, he did not clearly demonstrate the process involved; in particular, he did not identify the nature of the relationship between highlands and lowlands, despite his description of the ecological context of the Kachin region at the beginning of the book. All of these issues require greater historical penetration than Leach was prepared to accept was necessary when Political Systems was published. The historical context in this volume is considered initially by reecting on the extraneous and personal intellectual limitations placed on Leach’s vision before and at the time of writing Political Systems (Anderson), Leach’s understanding of Kachin notions of historical discourse and his desire to give an historical frame to his anthropological theories (Sadan), and the contemporary acknowledgement given by the Kachin themselves to Leach’s published work (Maran La Raw). Many of the later chapters also give much attention to historical context, blending this with anthropological analysis in a way that demonstrates the development of the inter-relationship between these two disciplines since Leach’s rst (largely unsuccessful) attempt to bridge the divide. The concepts of “structural oscillation” and “unstable equilibrium”, centred upon a presentation of the gumsa-gumlao model in Kachin society, represent both the best and the worst aspects of Political Systems. Maran La Raw’s chapter goes far beyond demonstrating the inadequacies of Leach’s understanding of the vernacular Jinghpaw categories of gumsa and gumlao to give a fuller description, extending analysis to that of khau wang magam. This was an exchange system amongst the chiey lineages according to which the wife-givers become the wife-takers, leading to the idiom that “each participant marries its granddaughter”, and discussion of which signicantly transforms our understanding of the generalized system of exchange in Kachin society. Other comparative perspectives presented in this volume also help us to qualify Leach’s assertions about the gumsa-gumlao model. The analysis of social and political organisation amongst three societies in Assam (Ramirez), as well as the discussion of Naga political systems (Bouchery), point out the limitations of the model proposed by Leach beyond the Jinghpaw and Kachin sub-groups.

xiv

françois robinne and mandy sadan

In the same way, the analysis of the acephalous political systems of the Khmu (Evrard) and of the Phunoy (Bouté) of Laos demonstrate that social change in these cases is not due to internal social structures but arises mainly for a range of exogenous reasons. In these chapters, it is in particular the increasing pressure from economic factors, such as trade and control of economic networks, migration and the movement of populations, together with the development of social mobility, land ownership and land pressure resulting in the “inequitable allocation of the lands”, that produces radically different social and political structures in neighbouring social landscapes. A distinction between internal and external dynamics of social change is also at the centre of analysis focused on the circulation of wealth amongst the Jinghpaw of Yunnan (Ho Ts’ui-p’ing) and the Rmeet/Lamet of Laos (Sprenger). In both cases, the development of a market economy generates similar consequences, notably a process of individualization, which is then linked with the metamorphosis of goods that are circulated. In both chapters, one can observe change from a vertical system of ritual exchange and the sharing of gifts into a horizontal system based on social mobility through access to ownership. The demonstration focused on the Jinghpaw of Yunnan concentrates on the orescence of opium production amongst the Jinghpaw during British colonial rule and in times of political instability; the demonstration focused on the Rmeet/Lamet of Laos analyses the ow of bronze drums and colonial silver coins within Laos and in Rmeet society. In both cases, the internal and external interactions of society, respectively with political instabilities and commercial exchanges, are considered in detail. Continuing from the analysis of these two chapters, which focus on the concomitant evolution of the circulation of wealth, wealth ownership and debt, the example from Drung society of northwest Yunnan (Gros) focuses on “what is exchanged” rather than the exchange strictly speaking. The notion of ritual language developed by Leach is correlated with Mauss’ concept of “total social fact”. The circulation of wealth proves to be a signicant vector in the attribution of social positions, but its inclusive dynamic operates at different levels where individuals and communities, nature and supernature, appear bound by the same logic. In a complementary way, the transethnic dimension of generalised exchange is then considered, allied to the notion of ritual language (Robinne): beyond the hierarchy in which wife-givers and wife-takers are linked, and beyond the exogamous clanic as well

preface

xv

as ethnic belongings which may be claimed, the generalised exchange system is seen to transcend the inner limits to which it is generally reduced by outside observers. Following the intentions of Leach himself, the present volume extends the analysis to different communities and to different multi-ethnic contexts all along the Indo-Chinese massif. One of the main difculties the editors had to face was to decide upon a conceptual or spatial conguration for the collective volume. This was particularly difcult given the signicance of both internal and external dimensions of social change in the various demonstrations, which made both conceptual and spatial criteria in themselves unsatisfying. If spatial distribution nally prevailed in the volume, it is by no means a categorical choice: on the contrary, the three sections into which this volume are divided do not primarily gather together neighbouring groups but, rather, juxtapose some conceptual proximities developed by the authors. In the same manner, the postscript (Robinne & Sadan) suggests that we might consider the social in terms of village partnerships rather than in terms of interethnic relationships. By substituting the notion of sets of communities for spaces of dispersion, the notion of social change ts more closely to the dynamics that characterise the crossroads of Southeast Asia in general, and the Kachin region in particular, which was the focus of Leach’s observations. François Robinne Mandy Sadan

LIST OF CONTRIBUTORS Robert Anderson is Professor of Communication in Simon Fraser University, Vancouver, Canada. Trained as an anthropologist at the University of Chicago, he has worked in South and Southeast Asia on subjects as diverse as tropical forestry (“The Hour of the Fox”) and rice cultivation systems (“Rice Science and Development Politics”). He is currently undertaking a study of guerrilla warfare in frontier regions in the 1940s. Pascal Bouchery completed his PhD thesis in Anthropology in 1995 at the Université Paris X, France: ‘Les Hani: Introduction à l’étude d’une population tibéto-birmane du Yunnan en relation avec la Chine’. He is the author of “Les systèmes politiques naga”, ( Journal Asiatique, 1988). His current academic interests are the Tibeto-Burman speaking peoples of the Indo-Myanmar border area, political anthropology and kinship systems. Vanina Bouté completed her PhD in Anthropology at the Ecole Pratique des Hautes Etudes, Paris, France: ‘En miroir du pouvoir. Les Phounoy du Nord Laos: ethnogenèses et dynamiques d’intégration’. She is currently Lecturer in the Ethnology Department, University of Provence. She is particularly interested in social change and representation. Her most recent publication is “Empowerment through acculturation: Forgetting and contesting the past among the Phunoy in Northern Laos” (Southeast Asia Research, 2006). Olivier Evrard is an anthropologist and researcher at the Institut pour le Développement (IRD), Paris, France. He obtained a PhD in Social Anthropology at Paris 1—Sorbonne in 2001. He lived for extensive periods of time with Khmu populations in Northern Laos between 1994 and 1997, and then went back to this area several times for shorter periods between 2001 and 2006. He is the author of Chroniques des cendres, anthropologie des populations Khmou et des relations interethniques du Nord Laos (Editions IRD, Paris, 2006). His main areas of interest include the social organisation of Mon-Khmer speaking populations in Southeast Asia, mobilities, and interethnic relationships. He is currently living

xviii

list of contributors

and working in Northern Thailand as an invited foreign researcher at the Social Research Institute, Chiang Mai University. Stéphane Gros received his PhD in 2005 from the University of Paris X at Nanterre, France, and is afliated to the research team “Milieux, cultures and societies of the Himalayas” (UPR 299) of the Centre National de la Recherche Scientique. His dissertation, ‘The Missing Share. Exchange and Power in Yunnan’s Borderlands’ on the Drung people of Northwest Yunnan in China, aims at situating the Drung within the context of global society and in a framework of present and past relations with neighbouring societies. It tackles the issue of political, social, economic, and religious transformations, in particular since the founding of the People’s Republic of China (1949). His current research considers with a comparative perspective along the SinoTibetan border issues of identity politics and social transformations in the face of modernisation and commoditisation. Ho Ts’ui-p’ing is Associate Research Fellow at the Institute of Ethnology, Academia Sinica and an adjunct faculty with the Institute of Anthropology, National Tsing Hua University, both in Taiwan. Her research interests include personhood, materiality, social change, and historical anthropology. Her areas of interest, including archival research, are Southwest China and the adjacent border region with Burma. She has published on person, time, and material culture among the Jingpo nationality in Yunnan Province, People’s Republic of China. Recently, she co-edited the volume State, Market and Ethnic Groups Contextualized (2003) and edited a double-issue entitled “Life-Cycle Rites, Objects and Everyday Life” of the Journal of Chinese Ritual, Theatre and Folklore (2005). She is currently writing a monograph tentatively entitled Rethinking the Kachin and serves as co-organizer of the Chiang Ching-kuo Foundation funded project “The Conversion of Chieftains: Territorial Gods, Chieftain Lineages and the Retention of Indigenous Identity in Chinese Border Areas.” Maran La Raw is a research anthropologist, concerned since 1962 with social-structural changes in Kachin society in the postcolonial era. An emeritus faculty member from the University of Illinois, Urbana, USA, where he received his PhD, he had the privilege of knowing both E. R. Leach and his Kachin language assistant. He is currently nishing

list of contributors

xix

a political science textbook written in Jinghpaw-Kachin, as requested by Kachin leaders. F. K. L. Chit Hlaing (F. K. Lehman) is Professor of Anthropology, of Linguistics, and of Cognitive Science in the University of Illinois at Urbana, Champaign, USA. His current research projects are the interethnic trade networks of the China-Burma border and the algebraic theory of cultural systems. He has published two monographs: The Structure of Chin Society (1963 and later editions), and Kayah Society as a Function of the Shan-Burma-Karen Context (1967), as well as many articles. Philippe Ramirez is a researcher at CNRS (Centre National de la Recherche Scientique). He has worked for several years on the political structures and ritual systems of Central Nepal. He has also undertaken an anthropology of Nepalese Maoist representations. He is at present carrying out research on hill communities of Northeast India. Amongst other publications he is the author of De la disparition des chefs: une anthropologie politique népalaise (CNRS, 2000) and editor of Resunga: The mountain of the horned sag: Two districts in central Nepal (Himal Books, 2004). François Robinne completed his PhD in Social Anthropology in 1985 at the Ecole des Hautes Etudes en Sciences Sociales (EHESS Paris, France). He is Senior Researcher at the Institute of Research on Southeast Asia (IRSEA-CNRS). He has published two monographs: Fils et maîtres du Lac. Relations interethniques dans l’Etat Shan de Birmanie (CNRS, 2000) and Prêtres et chamanes. Métamorphoses des Kachin de Birmanie (L’Harmattan, 2007). Mandy Sadan is British Academy Postdoctoral Fellow in the School of Anthropology and Museum Ethnography (Pitt Rivers Museum), University of Oxford and Junior Research Fellow, Wolfson College, University of Oxford. She completed her PhD in History at the School of Oriental & African Studies, University of London, in 2005: ‘History & Ethnicity in Burma: Cultural contexts of the ethnic category ‘Kachin’ in the colonial and post-colonial state, 1824–2004’. She has spent extended periods of time since 1996 living in Burma working with Kachin communities on a variety of projects relating to cultural representation in the Burmese state. A number of articles arising from this research have been published and she is presently working on a monograph. She has also worked on research projects at both SOAS and

xx

list of contributors

the Pitt Rivers Museum considering the history of colonial collecting and ethnographic representation of Arunachal Pradesh and Tibet. Guido Sprenger is a Postdoctoral Fellow at the Institute of Ethnology, Academia Sinica, Taipei, and earned his PhD at the University of Münster, Germany. He has been doing eldwork among Rmeet (Lamet) people in northern Laos since 2000. His works include “Erotik und Kultur in Melanesien” (Erotics and Culture in Melanesia, 1997) and “Die Männer, die den Geldbaum fällten: Konzepte von Austausch und Gesellschaft bei den Rmeet von Takheung, Laos” (The Men who cut the Money Tree: Concepts of Exchange and Society among the Rmeet of Takheung, Laos, 2006). Besides ritual, social organization, and concepts of the person, he is particularly interested in the way systems of value and exchange communicate and change over time.

INTRODUCTION: NOTES ON EDMUND LEACH’S ANALYSIS OF KACHIN SOCIETY AND ITS FURTHER APPLICATIONS F. K. L. Chit Hlaing (F. K. Lehman) Most of the remarks in this Introduction were originally made when I acted as Discussant for a panel at the 2004 meeting of EUROSEAS in Paris convened to mark the ftieth anniversary of the publication of Political Systems of Highland Burma: A Study of Kachin Social Structure by Edmund Leach. Most of the papers in this volume were originally presented on this panel, and my comments shall proceed more or less in the order in which the papers were originally presented. Two papers were added after the conference (by Bouté and Sprenger), and about these I shall comment less extensively. However, as with my more limited comments on the paper by Pascal Bouchery, this is also because I wish to avoid repetition of comments that are made elsewhere in the Introduction. Much of my purpose in making these comments is to relate the discussion of Leach’s ideas to general themes in the anthropology and history of Southeast Asia, and to recent and current work in formal anthropological, linguistic and cognitive theory. The panel commenced with a paper by Robert Anderson, which is presented in this volume as “The biographical origins of Political Systems of Highland Burma”. Anderson’s chapter presents an overview of the background to Leach’s eldwork in Burma and some of the issues arising from it and, as such, raises all sorts of questions. For example, given Leach’s work experience in China and his knowledge, however limited, of the Chinese language, it has always seemed to me odd that in Political System he deals hardly at all with the Kachin on the China side of the border. We are now beginning to amass ample materials on the ‘Kachin’ in China in Western languages (some of them presented at this panel by the Taiwanese anthropologist Ho Ts’ui-p’ing, other materials by my late former pupil Wang Zhusheng, and my current student Zhang Wenyi). It is clear in retrospect that a good deal is to be learnt from the China side, not least of Kachin-Shan relations and the complexity of Kachin ethnicity. Furthermore, in Burma Leach worked in close proximity to the China border and I cannot imagine

xxii

f. k. l. chit hlaing (f. k. lehman)

that he did not cross into China during his Burma period of study and military service; it is a puzzle that he mentions almost nothing of it. In a similar fashion, it is also somewhat odd to me, as an ethnographer of the Chin, that Leach does not refer to what was already known of Chin social structure and marriage (in many ways so similar to that of the Kachin), considering that his entry into research in Burma seems to have been facilitated by H. N. C. Stevenson, himself a considerable gure in the anthropology of the Chin. It is also surprising that what Anderson calls Leach’s “considerable knowledge of trading networks in China” plays no obvious role in Political Systems or his papers on Kachin, in as much as the border trading networks have had a lot to do with Kachin-Shan connections. Above all, however, for me, being also a former mathematician and a worker in mathematical anthropology, it is puzzling how little of Leach’s mathematical training is brought to bear by him in his analysis of the complex structure of the Kachin marriage system. After all, in Leach’s 1980 lecture at the University of Washington, he gave every indication that his famous model of oscillation between Kachin political models was inuenced by his knowledge of Pareto, and a realisation that ideas of homeostasis were, in a technical sense, a relevant consideration. In addition, he also said that his engineering background inuenced his anthropology. Yet, mathematical notions were rst explicitly invoked in Leach’s published writings only much later than Political Systems, namely, in Rethinking Anthropology (1961), where they are applied mainly to the notion of ‘function’ with little, if any, direct reference to Kachin matters. As indicated in the 1980 lecture, it is clear that Leach was informed about technical matters in the relationship between ‘system’ and ‘structure’, and that this certainly bore upon his remarks in Political Systems and elsewhere on the problem of how one talks about the structure of a whole social system. I myself have dealt with this part of Leach’s work in my 2004 paper on the ‘Globality’ hypothesis of social and cultural structure, from precisely this mathematical angle (Chit Hlaing 2004, 2005). So, it seems to me that Anderson’s end-question ‘g’ (‘Why did [Leach] decide to write a ‘technical book’ particularly for anthropologists, minimizing the inuence of the geopolitical and historical context, in spite of the fact that those were the things he knew so well, and that the audience for it would have been even larger than for Political Systems? ’) is, in many ways, central for relating Leach’s Kachin work, and, in fact, his anthropology more generally, to his pre-anthropological background, and to the development of formal anthropological theory.

introduction

xxiii

Sometimes, at least, Leach goes astray, as if that technical background were less than it might have been. Now and again, Leach, like LéviStrauss, employs the term ‘transformation’ when discussing ‘structure’, as if he had no idea of its correct computational meaning, namely, as a ‘map’ from a given family of structures into the same family only. That, assuredly, has a lot to do with his attempts (as seen in his 1970 Paper on ‘Kachin Concepts of Sin’) to deal with formal similarities between the marriage-cum-political system and the cosmology as if they were ‘the same structure’ of ‘asymmetric exchange’ under transformation (see Lehman 1977). One further point may be added here relating to the conduct of Leach’s eldwork, which throws some additional light upon his Kachin work by what it reveals about his attitude to Kachin people themselves. Once, during a three-way correspondence between myself, Leach and Maran La Raw, (who was then my PhD pupil), Leach became aware of La Raw’s analytical work on aspects of Kachin society and, in particular, on matters of Kachin cosmology and religious ideology, about which Leach had written. Strange to say, Leach indicated in a private communication to me, that he was rather put out to think that a Kachin might presume to reect analytically upon such things, it being, in his view, the proper object for scholars with a background in formal comparative (possibly implying Christian-based, Classics-based) religious thought. Although I have the distinct impression that Leach liked the Kachin people, it also seemed clear that, one might say, he liked them ‘in their place’. I have known any number of British excolonial-military ofcers from the old Indian Empire whose attitude to the ‘natives’ was of this sort. One must bear in mind that such attitudes were not at all incompatible with the social ideology of otherwise anti-colonial and liberal, even leftist British writers of that era, such as George Orwell regarding the Burmese, and which, as Anderson shows, Leach was also. That Leach was capable of such prejudices is independently clear from the tone and content of the review he made of British Social Anthropologists of his day in the Annual Review of Anthropology (1984). It seems to me, reecting upon all this, that Leach’s willingness, nay eagerness, to dismiss Kachin views about their society and its history, as in his dismissal of Kawlu Ma Nawng’s book on the gumlao rebellion, The History of the Kachins of the Hukawng Valley (1942), in favour of an analysis entirely motivated by his own ‘take’ on British and Continental social theory (see my remarks, below, upon Mandy Sadan’s paper in the present volume) may be of a piece with

xxiv

f. k. l. chit hlaing (f. k. lehman)

such attitudes. Such an attitude cannot but have played a part in the way Leach conducted his eldwork and in his rôle as an ethnographer (as a ‘participant observer’ amongst the Kachin, say, in the spirit of Malinowski’s or Firth’s work, which inuenced Leach’s practice). For, whatever else ‘participant observation’ may mean, as current cognitively-oriented theorists and practitioners of ethnography understand it, it must entail the involvement of one’s informants/consultants with one’s work intellectually. That means recognition of the possibility that at least some may be or may become one’s intellectual colleagues (I have written on this matter elsewhere, e.g. Chit Hlaing 2000). My comments on Mandy Sadan’s chapter in this volume will begin with and focus mainly on the question of why Leach was motivated to downplay the idea that the gumlao system might have been a consequence of historical developments, in favour of his well-known idea of a built-in oscillation between gumsa and gumlao. My remarks on the gumlao system as such are based here upon materials in Wang Zhusheng’s book on the Jingpo of China (1997), on still unpublished work by Dr. Maran La Raw, and also my brief paper on the Kachin in the Encyclopedia of World Cultures (1993). On evidence, Leach’s analysis of this matter does not hold up. It is now clear that the idea of a gumlao rebellion has documentable historical support, as I shall argue, and that his view of the classical (‘thigheating’) Kachin chieftainship as being modelled upon the role of a Shan prince or Sawbwa (tsau-fa) was wrong. This can be seen directly from Kachin ethnographic and linguistic evidence (for example, in Ola Hanson’s Kachin-English Dictionary [1954]). Why, then, did Leach build his model and ignore some of this evidence, which was already available to him when he wrote Political Systems? Anderson (see my previous remarks) makes a good case for the proposition that Leach was intent upon radically reconstructing anthropological social theory. Nevertheless, Leach must have been inuenced by the widely accepted Radcliffe-Brownian theory that rejected any historically oriented analysis of the organisation of society in favour of a certain idealisation of the idea of ‘structure’. Leach was concerned in the rst instance with the fundamental idea of (social) structure. The standard approach at the time was that structure was idealized to a situation in which all elements of the system are deemed mutually to reinforce one another so that the system is, as nearly as one pleases, static and unchanging. Thus, structure was made to commute with stability, while change was equated with what the

introduction

xxv

late Victor Turner (1960) called ‘anti-structure’ (c.f. Raymond Firth’s 1961 distinction between ‘social structure’ and ‘social organisation’, and Claude Lévi-Strauss’ treatment of classical, pre-modern societies as ‘cold’ or ‘outside of history’). I deal with all this in my paper on the technical notion of ‘structure’ by British social anthropologists, who were by no means all Radcliffe-Brownians, such as Firth or, for that matter, Leach. Thus, for Leach, a proper account of the Kachin social system, or rather, its political order, had to show somehow that its ideal form was a stable, ‘unchanging’ system in a classical or ‘premodern’ context. Yet, the problem remained as to what he was to do about the gumsa-gumlao alternation. On rst view, this seems to be a matter of ‘dynamics’ taken as the antithesis of structure. Leach’s solution, therefore, was to treat this alternation as a built-in oscillation; a process that had no directionality. To manage this, he called upon the idea of ‘dynamic equilibrium’, where the term ‘equilibrium’ is taken to solve the problem. Ever since Maran La Raw’s earliest work on Kachin ethnography (1967), we have known that this model (motivated as I have shown above) is factually incorrect. In particular, the very terms it employs are wrong. Leach confuses two different, if similar sounding words: gumtsa and [ gumchying] gumsa. It is the latter term that is proper for the classical Jinghpaw system of ‘thigh-eating’ chiefs; the former term is used for the kind of chieftainship sometimes adopted by Kachin chiefs who had taken over a Shan principality or were trying to set themselves up on the model of a Shan prince. Moreover, since, as Leach himself noted (1960), a Kachin chief would lose his Kachin standing by modelling himself as a prince and making his Kachin followers ‘subjects’, there arose yet another category, gumrawng gumtsa, which was a sort of compromise between the competing models. Rawng is the Jinghpaw verb ‘to be proud’, and, as such, indicates a sort of hubris, as well as the fact that a chief ’s authority is not hereditary in the Kachin system, and that he does not acknowledge allegiance to any other chiefs. In addition, as Maran points out, the Leachian oscillation model characterises gumlao as a ‘democratic’ version of social organization, in which the idea of chiefs and aristocracy is rejected. Leach borrows this term from his sometime mentor-colleague Stevenson (1943), who used it to characterize the Chin political order of Tashon. But, in the latter instance, it is clear that (I once spoke with Stevenson about it and he assented), although there was no one chief of Tashon, there were several self-proclaimed chiefs in Tashon, as also in Hakha. In the

PART ONE

HISTORICAL CONTEXTS OF E. R. LEACH’S FIELDWORK

THE BIOGRAPHICAL ORIGINS OF POLITICAL SYSTEMS OF HIGHLAND BURMA Robert Anderson Edmund Leach spent his longest period of anthropological eldwork anywhere, in Burma. From that experience, undertaken between 1939 and 1945, he wrote a PhD dissertation in 1947 and published his celebrated book on the political systems of the Kachin in highland Burma in 1954. Of course, this was not typical eldwork at all. His time in Burma was punctuated by preparations for war, followed by three years ghting in northern Burma, where more than ve nations had sent troops. Not only did Leach never return to Burma, he also discouraged readers and researchers from learning about the intellectual and experiential origins of his book, at least for twenty years after its publication. Even his preface to the book was cryptic about its origins. These origins, however, are important for the history of the social sciences because the book had such a wide impact. Working in war conditions had a particularly intense impact on Leach’s thinking. Inevitably, his understandings of Kachin-Shan political life in northern Burma would have been shaped by the tensions he witnessed in those communities and the conicts generated by the war. This chapter is part of a larger project that I am undertaking to explore

Acknowledgements: I am very grateful for stimulating and candid conversations with Lady Celia Leach, Louisa Brown, Stephen Hugh-Jones, Alan Macfarlane, Nur Yalman, Stanley Tambiah, F. Kris Lehman, Dame Marilyn Strathern, Caroline Humphrey, Sir Jack Goody, Maran La Raw, John Barnes, Frederic Barth, François Robinne. I am particularly grateful to Mandy Sadan for her close editing of my draft text. These discussions were possible while I was a Visiting Fellow at Clare Hall, Cambridge 2004–2005, an honour for which I am most grateful to the Fellows of that lively institution. Prepared for the September 2004 EUROSEAS conference, this essay is a preliminary inventory of both published and unpublished sources (including the Leach Papers in the Modern Archives of Kings College, Cambridge, and the Burma les in the British Library and the UK National Archives). My debt to archivists at these places will be fully explained later. However, I have not made explicit use here of the Leach Papers at Kings College Archives in Cambridge, nor the military, administrative, or diplomatic les, all of which I have studied. A full treatment of that material is planned in my forthcoming book whose working title is Highland Burma, Edmund Leach, and Anthropology.

4

robert anderson

the origins of Political Systems of Highland Burma (hereafter referred to as Political Systems). Here, however, my intention is to review the accessible sources in order to establish what is publicly known about Leach in the period in which he wrote the book, and to describe Leach’s later reections about his earlier work. Of course, what is publicly known now was and is surrounded by some oral history, but little of that lesspublic information is introduced here. ‘What is publicly known’ refers mostly to what Leach chose to release, slowly, and what others have learned and consolidated.1 We should note from the outset that Political Systems achieved its reputation long before the details about Leach’s early life were made public (most of them by Leach himself ). The book was, thus, judged without reference to its origins, except among those with whom Leach talked privately about his Burma experience. Nevertheless, readers knew ‘just enough’ to speculate. This chapter is intended to ‘stabilize’ a scaffold upon which to build a fuller portrait of both man and book, using less-public sources, where a powerful book and biography of a person come together. Political Systems may have been the most widely read anthropological work in English in the second half of the twentieth century, along with translations of Lévi-Strauss. Moreover, it has been widely read among social scientists outside anthropology. In 1978 Leach asserted that Political Systems was “certainly my most inuential work”.2 But, as he was unable to return to Burma, despite requests for permission to do so, the frontier receded for him more and more into the highland mists. From approximately 1962, for the next forty years, no other western anthropologists obtained permission to study there.3 1 Most of the information here comes from the valuable source book on Leach’s life, Stanley Tambiah, Edmund Leach: an anthropological life, Cambridge University Press, 2002. Tambiah has assembled comments from a large number of students and colleagues, but, unlike mine, his focus was not on the origins of Political Systems. I also acknowledge the scholarly benet of the major work of Stephen Hugh-Jones & James Laidlaw [eds], The Essential Edmund Leach, New Haven, Yale University Press, 2000 (two volumes). The introductions written by them for each chapter in these two volumes are an excellent map for the large intellectual landscape across which Leach worked. 2 Edmund Leach, “Glimpses of the Unmentionable in the History of British Social Anthropology”, Annual Review of Anthropology, Vol. 13, 1984, p. 18. Also author’s conversation with Stephen Hugh-Jones, Cambridge, September 2004, in which he explained that this remark was not just Leach’s conclusion in 1984, but had been repeated by Leach in conversation with many people about Political Systems in the 1970s and 1980s. 3 In my forthcoming project I shall be looking for anthropologists (from any country, including Burma) who were able to make studies in highland Burma. If any, I predict the number will be very few.

the biographical origins of POLITICAL SYSTEMS OF HIGHLAND BURMA

5

Leach’s Experience Before Burma Born in November 1910 to a large and prosperous Lancashire family with nineteenth century roots in Argentina, Edmund Leach learned about the operation and inuence of kinship long before it became one of his academic subjects. Surrounded by uncles and aunts, and one of a lineage of twenty one men who attended Marlborough School, he was ready to seek some distance from his large family (particularly his mother) when it was possible. At age nineteen he won a scholarship to Cambridge to study mathematics, and within a year explained to his father that he preferred engineering. Determined to do well in the engineering exams, he nevertheless read and discussed Sigmund Freud, Karl Marx, Bertrand Russell, and the anthropologist Bronislaw Malinowski. But he felt restless, nding it difcult at this age to concentrate. Visiting Germany in 1931, twenty-one year old Leach was troubled by what he saw there, and wrote home, mentioning his British and German friends who were attracted to anti-fascist and communist movements. He wrote to his father from Cambridge in 1933 that he might be good enough to qualify for the colonial service in East Africa or Sudan, but was not likely to qualify for the more demanding and prestigious Indian Civil Service. He knew he was expected to graduate with rst class marks, and in the end did so in 1933. He remained conscious of his rigorous undergraduate training in mathematics and engineering all the way through the writing of Political Systems, twenty years after his rst Cambridge period ended, frequently referring to the intellectual value of this kind of training for anthropologists. Engineer-in-Business in China Wanting to avoid working as a gas works engineer in Britain (that was the concrete offer of employment he had received), he applied to and was accepted by Buttereld & Swire, who were among the dominant British general trading rms in Hong Kong and China. Thus, he left the London ofce three times a week during his rst nine months with Swire and went to study Chinese language and culture at the School of Oriental & African Studies (SOAS) in the University of London. Once in China in late 1933, he was stationed in Shanghai, Tsingtao, Peking, and Chungking. How well he spoke or read Chinese is not known, but it is relevant for his experience later on the border between Yunnan and the Kachin Hills (he later said “I did not speak good Chinese but

6

robert anderson

enough to get around”).4 Leach described his early life thus: “I had lived in various parts of China between 1933 and 1936 as a member of the British commercial community.”5 He travelled widely, often on Swire’s river steamers on the Yangtse River, and as a serious hobby collected Sung pottery and jade, particularly in Beijing. Not particularly athletic by his own report, he nevertheless climbed four of China’s ve sacred mountains, a capacity he put to use regularly in northern Burma, where he would climb up and down 4000 feet in the same day. At an embassy party in Beijing in 1936 he met Kilton Stewart, an American psychiatrist and keen amateur anthropologist. He also met Owen Lattimore in China, though it is to be determined what kind of inuence Lattimore may have had on Leach’s outlook.6 Stewart’s inuence, however, is better recorded: he had already done an anthropological study of the Noone people in Malaya, and exuded such enthusiasm that Leach agreed to go with Stewart to Botel Tobago, a Japanese-held island off Taiwan. Over the course of ve weeks between December 1936 and February 1937, Leach made detailed and elegant drawings of house and boat designs, and took numerous photographs; the experience on that island left ‘an indelible impression’ on him, according to Leach biographer Stephen Hugh-Jones.7 When he returned, aged twenty-seven, to London in mid-1937, he met his childhood sweetheart Rosemary Upcott, who had just married anthropologist Raymond Firth. Meeting Firth through Upcott, Leach was drawn into the anthropology seminar at the London School of Economics (LSE), led by Bronislaw Malinowski. Kilton Stewart had also returned to London from China and was on the fringe of that circle; he gave illustrated lectures on life among the Yami on Botel Tobago, both to anthropologists and at the prestigious Royal Geographical Society.8

4 Adam Kuper, “Interview with Edmund Leach”, Current Anthropology, August–October 1986, p. 375. 5 “In Formative Travail With Leviathan”, talk at the LSE, 1973, published in R. M. Berndt [ed], special issue of Anthropological Forum, vol. 4, no. 2, 1977, p. 192. 6 I am grateful to Professor Sir Jack Goody at Cambridge University for suggesting that this connection be studied. 7 Stephen Hugh-Jones, “Edmund Leach 1910 –1989”, a memoir prepared by direction of the Council of Kings College, Cambridge, 1989, p. 5. 8 His rst popular publication was probably “The Yami of Koto-sho: a Japanese colonial experiment”, The Geographic Magazine, October 1937.

the biographical origins of POLITICAL SYSTEMS OF HIGHLAND BURMA

7

In Malinowski’s Charmed Circle, London The LSE seminar attracted an extraordinarily cosmopolitan circle of bright students, the majority of whom had a strong interest in Africa, and who, like their teacher, were committed to empirical eldwork methods and learning languages in the eld. Leach was determined not to continue with Buttereld and Swire, and so became a student of anthropology, “one of the members of the last cohort before Malinowski left for the U.S.”9 In Leach’s words later, “Nearly all the members of the seminar, staff and students alike, had some sort of eld experience in the environment of a relatively ‘primitive’ culture.”10 Of course, Leach did not then know Malinowski would shortly leave for Yale University, and he developed a proposal in late 1937 to write a dissertation under his supervision about Malinowski’s favourite subject: technology and culture in Oceania. A few months later, he realized Malinowski was leaving and knew he would have to be supervised in another topic by someone else at LSE. By mid-1938, an alternate plan was in place for Leach to travel to Iraq for eld work. Leach had much to admire in Malinowski, a man twenty-six years his senior and who had also studied mathematics and physics before moving to anthropology.11 Malinowski resolutely remained a central European intellectual in the British atmosphere of the emerging eld of social anthropology, and established in London a certain social and intellectual distance from Oxford and Cambridge. He also deliberately cultivated a wide network of internationalists. This distance and network appealed to and encouraged Leach. Leach wrote student papers referring to the engineer’s structural outlook on cultural questions, and showed his keen interest in the philosophy of science in seminar discussions. His teachers included Raymond Firth of New Zealand, trained in economics, and Meyer Fortes of South Africa, trained as a psychologist. Among the seminar regulars in 1937 were Noel Stevenson, an administrator stationed in northern Burma and writing a ‘diploma thesis’ at the LSE, and L. R. Ogden, also an anthropology student in

9

Tambiah, 2002, p. 33. Ibid. Quoted p. 36. 11 Michael Young, Malinowksi: odyssey of an anthropologist, New Haven, Yale University Press, 2004. 10

8

robert anderson

the LSE, on leave from the Burma administrative service in the Frontier Areas.12 These people were to inuence him greatly until the writing of Political Systems. Interlude in Iraq Having re-framed his project under supervision by Raymond Firth, the heir to Malinowski at the LSE and nine years older than himself, Leach nally left London in 1938 for his dissertation eldwork. His research was running in parallel with “a love affair with an archaeologist”, as he later said, which resulted in a “broken relationship” in Iraq mingled with “a touch of dysentery”.13 After the Munich crisis, war loomed and Leach thought Iraq looked as though it would become a battleground soon [ he was not wrong]. Though he did not stay long, surprisingly his brief eldwork in 1938 was published as an LSE monograph, featuring his own photographs, nely drawn maps and diagrams of water-wheels.14 He returned to work in London for about ten months in 1938–39 as an assistant to Raymond Firth. During this period, Kilton Stewart introduced him to Celia Buckmaster, a young painter, and daughter of a barrister; this relationship was to have great consequence, as we shall see. At this time, Noel Stevenson persuaded Leach that a study in northern Burma should be the source of his dissertation, and made some administrative arrangements for him to stay in the southern part of Kachin territory near Bhamo.

12 Leach got to know Ogden, who reappeared in Rangoon after the war, having courted death on the frontier a number of times. He is described in Rangoon from a young girl’s point of view in Maureen Baird-Murray, A World Overturned: a Burmese childhood, 1933–47, London, Constable, 1997. Baird-Murray was abandoned by her parents (or so she thought) in 1942 and was orphaned. Ogden was considered responsible for her by her father, and, indeed, showed up at the convent orphanage in Kalaw in 1945, and took her under his wing in Rangoon, enabling her to travel to live with relatives in Ireland. She provides a fascinating account of this semi-retired Frontier Service ofcer’s life in Rangoon in 1946–47. 13 Adam Kuper, “An Interview with Edmund Leach”, Current Anthropology, 1986, p. 376. 14 Edmund Leach, Social and Economic Organization of Rowanduz Kurds, London, LSE Monographs, no. 3, 1940.

the biographical origins of POLITICAL SYSTEMS OF HIGHLAND BURMA

9

A Doctoral Student Goes to Burma In August 1939, with Europe on the brink of war, Leach arrived in Rangoon. He was twenty-nine, an age when many doctoral students are completing their eldwork. When war was declared in September, he was on a ferry heading for Bhamo, a multi-cultural frontier trading and administrative town at the centre of Noel Stevenson’s Kachin Regeneration Scheme. The Kachin Regeneration Scheme had been Stevenson’s inspiration in 1937–38, drawing Governor Archibald Cochrane into a web of focused funding for dozens of small incomegenerating projects. Leach had intended to study its impact, but the war and Stevenson’s absence rendered the scheme inactive. Stevenson was still on ‘compassionate home leave’ to deal with the accidental death of his youngest son, but Leach met the District Commissioner R. S. Wilkie, ICS, “dressed in grey annels and a sports coat sitting in front of a roaring re 6000 up in the air”, and their relationship clicked.15 (Though Kachin leaders would assert that the centre of the Scheme was the small administrative town of Sinlum, the government ministries from which Stevenson obtained his resources were based in Bhamo.) Though it was in ‘Burma proper’ or ‘Ministerial Burma’, Bhamo was the entrepot on the river for the whole of the Kachin region, as well as the conduit for trade with China.16 When Leach arrived, it was a rich and booming frontier town through which passed many tons of war equipment and material for the Chinese Army (since the Japanese had captured the east coast ports). At a 1973 LSE seminar, Leach described the situation he found in the Kachin Hills in 1939 –40, where “law and order were enforced by an extremely inefcient and corrupt military organization known as the Burma Military Police.”17 Moreover, the specialized Burma Frontier

15 Edmund Leach, “Letter from Bhamo” in Hugh-Jones & Laidlaw, The Essential Edmund Leach, pp. 217–119. 16 Few administrative arrangements could have been more confusing to Leach: “Ministerial Burma” was to the south and west of Bhamo, across the river, and administered by a combination of Burmese elected ministers from 1937, and British civil servants. “Frontier Areas” in which Bhamo lay were administered separately, and by the Frontier Administration who took pains to distinguish themselves from Rangoon, though their salaries came from the same Accountant-General’s ofce. But the towns (municipalities really) of Bhamo and Myitkyina on the river to the north operated as if they were in Ministerial Burma. In this strange context the same ofcial person functioned in two jurisdictions on the same day. Thus Bhamo was in “Burma proper”. 17 “In Formative Travail. . . .”, p. 193.

10

robert anderson

Service administered the Kachin-Shan border areas separately. Rangoon exercised little authority there, as most administrators in ‘Scheduled Areas’ were seconded from the Indian Civil Service (e.g., Wilkie) and reported directly to the Governor.18 These administrators, however, were ICS men whom “India” had chosen to forget and Rangoon preferred to ignore, except when necessary: the frontier was far away from their administrative power and, so long as the frontier ofcers kept the Chinese at the border, moderated disputes among the Kachin and Shan, and levied the appropriate taxes, there was little interference.19 Leach immediately did nine months eldwork from October 1939 to June 1940 in villages in the Sinlum Hills at elevations between 5000 and 6000 feet. He focused on the 130 households of Hpalang, most of them practising taungya shifting cultivation. This area, though two days walk from Bhamo, was under the administration of the Frontier Service. At rst he had a translator, but eventually he released that person, according to Maran La Raw (this volume).20 In the spring of 1940 he began to write a draft monograph from his notes, saying in a 1986 interview that his original objective had been a “socio-economic study of domestic production and networks of trading”, and that “segmentary lineage and cross-cousin marriages didn’t come into the story at all.”21 His considerable knowledge of trading networks in China would doubtless have been applied here, not least because of the dominant position of Chinese traders in this frontier area. In his rst letter home, Leach noted that he was close to a mountain in Yunnan that he had climbed in 1936, and had a Yunnanese cook with whom he could speak the Szechwan dialect he had learned while living in Chunking. In fact, it is likely that the focus and the result of his eldwork would have been quite similar to his monograph on the Kurds of Iraq, then being published by the LSE in London; some features of Political Systems, such as the mapping of agro-ecological zones and articulating their complex socio-economic boundaries, have a striking similarity to the 1940 monograph about Iraq.

18

Ibid. A large part of the material to be included in my forthcoming book will be the extremely uid cross-border situation into which Leach stepped in 1939 –41. 20 Author’s conversation with Dr. Maran La Raw, Paris, September 2004. 21 Adam Kuper, “An Interview”, 1986, p. 377. 19

the biographical origins of POLITICAL SYSTEMS OF HIGHLAND BURMA

11

Leach in Pre-combat Mode Leach described the frontier as a crusted hierarchy of ofcials and military, noting the strong ethnic-language community boundaries and the frequency with which those boundaries were crossed. We do not know if he had read George Orwell’s Burmese Days, which had appeared rst in the US in 1934 and then in the UK in 1935; in 1936 Orwell had published another critical piece about Burma, Shooting an Elephant.22 As Leach was a voracious reader in China, he may already have seen Orwell’s writings about the ‘crusted hierarchy’ then, or at least later, on return to London in 1937. “But things were on the move,” said Leach in 1973. “Even if there had been no war with Japan, some type of genuine independence would almost certainly have been achieved by the main Burmese population long before 1945. But what would have happened in the frontier areas is a moot point.”23 Leach said in the same seminar that it was perfectly obvious to most people “that the British would get pushed out of Burma. It became a military crime to express these views.”24 One is tempted to think that Edmund Leach would have expressed those views in 1940 and 1941 and that this would be known in the small, gossipy British circles of northern Burma. Leach said that “at this point (in 1940) Noel Stevenson once again became a signicant factor in my life . . . Stevenson was largely responsible for persuading the Governor that it would be feasible to leave a network of spies and guerrillas [in place] until such time as the British armed forces could return.” Seven years older, very knowledgeable about frontier Burma, Stevenson had a wide and effective network. Leach “soon found myself embroiled with various shady characters known by letters of the alphabet and numbers.”25

22 Orwell was a policeman for seven months ( December 1926 to June 1927) not far from Bhamo, on the northern railway line at Katha, a town Leach would visit many times fteen years later. It is believed that Katha provided the model of ‘Kyauktada’ in Burmese Days. A most interesting essay on Orwell’s experience in Katha and its remnants in 1997 is Steven Martin, “Orwell’s Burma” found at http://orwell.ru/library/ novels, etc. One wonders also if Leach read Maurice Collis, Trials in Burma. London, Faber & Faber, 1945. Collis was a District Magistrate in Rangoon in the 1930s, and his critical and reective study of colonial legal culture was written in 1938, but not published until after the war. 23 “In Formative Travail…”, 1977, p. 58. 24 Ibid., p. 59. 25 Ibid., p. 195.

12

robert anderson

Nearing the end of this eldwork in early 1940, Leach wrote to Celia Buckmaster, inviting her to marry him when he came home for a visit in 1940: the proposal was typical in that, at that time, the war was not expected to last long. When his home leave was cancelled due to the war, he asked her to come to Rangoon for the marriage. She agreed, and he travelled the long river journey from Bhamo to Rangoon to meet her in July 1940. Celia returned with Edmund to the Sinlum Hills to enable him to complete his dissertation eldwork. Meanwhile, he was instructed to come to Maymyo during October and December 1940 to begin his military training; the new couple stayed in Maymyo, with its pleasant hill station climate, for the winter. For much of 1941, before the Japanese invasion, we have little idea what Leach was doing or thinking, although the couple’s daughter, Louisa, was born in October 1941. Maymyo was a completely militarized and civil service town. It was the location of the Bush Warfare School and Leach must have met those training ofcers as he was preparing for bush warfare. He attended a conference in Maymyo on boundary settlement with the Chinese ofcials. As he held a military ofcer’s commission, Leach was subject to its discipline. At rst, “because I spoke the Kachin language and the Kachins were, in effect, the Gurkhas of the Burma army,” Leach said he was used as a recruiting ofcer, “which was weird as my political sympathies were not in that direction at all; but I was under orders.”26 One does not yet know whether such a brief period of nine months of eldwork, completed before he reached linguistic competence, would have warranted a degree, or how this would have compared with other doctoral eldwork projects at that time. Both he and Evans-Pritchard published LSE monographs in 1940 based on their short sojourns in Sudan and Iraq. But Leach worked fast and wrote constantly, so no doubt he covered considerable ground. How accurate his understanding was in the 1939 –45 period has been addressed in the accompanying chapters by Mandy Sadan and Maran La Raw. A Dramatic Change in Plans When the Japanese captured Manila, Hong Kong, Bangkok, Java, and Singapore, and then Moulmein in lower Burma in January 1942, panic 26

Kuper, “An Interview…”, 1986, p. 376.

the biographical origins of POLITICAL SYSTEMS OF HIGHLAND BURMA

13

swept through the British population and its dependents. Rangoon was evacuated and taken by the Japanese in early February. The oil elds south of Mandalay at Yenangaung were destroyed in order to deny them to the Japanese forces. Chiang Kai Shek and Joseph Stilwell both visited the senior British eld commanders in Maymyo. People began to leave Maymyo and Mandalay, so Leach and other ofcers left quickly and drove their families in a bus to Shwebo, an airstrip in open country north of Mandalay, where they then felt abandoned (except for comfort and food given to them by a nearby Anglo-Burmese family). Finally, “in the midst of derailments caused by the Japanese advance” said Leach later, he lost the photos, notes, and draft monograph about Hpalang and the Sinlum Hills.27 That monograph was to have been his dissertation. Three days later, a plane came, and because she was cradling their infant daughter, Louisa, Celia Leach was allowed to y with her to Calcutta, leaving Edmund Leach and the men behind. The others (women and children able to walk included) decided to walk through to India and China, and many of them died on the way.28 The men were instructed to stay behind or walk out by the northern route. In April 1942, Japanese forces captured Lashio in the Shan States, thus cutting off the Burma Road to Yunnan and Kunming (parts of which had been destroyed by British troops, including Leach, who had destroyed a major bridge himself ). Japanese airplanes had control of the air and its forces held Akyab, the only big port on the Bay of Bengal, as well as the aireld at Myitkyina. Leach said he was ordered back to the Sinlum Hills, east of Bhamo; from there, military les show that he engaged in sabotage to destroy and deny roads and bridges to the Japanese (who had complete air superiority). He was now cut off from an escape to Assam. Instead, he walked (with others, accompanied by mules) along a well-known, old smuggling track to Kunming, and from there was own back to Calcutta to be treated for malnutrition and dysentery. The fact that he was not among the thousands who walked to Assam, like most of his British ‘comrades’ (Stevenson, Ogden, Wilkie, Leyden, et al.) did, meant that a particular, dangerous and gruelling bonding experience on the trek did not take place between Leach and senior Frontier Service ofcers with whom he eventually worked. Going out through China marked Leach off from most of the others.

27 28

Tambiah, 2002, p. 42. Ibid., pp. 43–44.

14

robert anderson

While resting in Calcutta he wrote a new version of his dissertation monograph from memory. The Army then suddenly joined him up with Noel Stevenson again, this time at a little airstrip in Assam: the fact that they both knew something about northern Kachin territory threw them together. From then on, Leach was on the move for the next three years, organizing guerrilla warfare among Jinghpaw speakers along the frontier, acting as quartermaster for the Kachin Levies at Putao (the most northerly and high altitude aireld in Burma, carelessly not captured by the Japanese). Then he was given a commission, rst in the “Army of Burma”, and then in the military’s Civil Affairs Organization. Now he was in a world very familiar to Leach from his work in business in China: salt, rice, currency, kerosene, medicines, opium, etc., all wrapped in the language of logistics, payments, and distribution. He moved around from Putao, up and down the Salween River, to Sumprabum in the south, then to Myitkyina, west to the Hukawng Valley, further west to the Chin Hills, and back to the Chinese border, criss-crossing these hills and valleys as the situation demanded, working in the zone commanded by US General Joe Stilwell. Leach said later that he was moved to ‘the mysterious X List’ by 1942, but was never actually a member of SO(E), or Force 136, nor was he later involved in “either of the Wingate operations, or with V Force, but peripherally my war was muddled up with all these things as well as others.”29 He later stated that he held a commission in the so-called “Army of Burma” and was known as “Major A.B.R.O” and that he was senior in the Civil Affairs Organization as “a DCCAO”, or a Deputy Chief Civil Affairs Ofcer. That means he held a military commission in the Army but was responsible for civil administration in ‘free Burma’ and ‘liberated Burma’ and ‘re-captured Burma’ until July 1945, starting with the most northerly part of Myitkyina District in 1942. Leach’s Writing During the War When we examine the Leach papers and diaries in Kings College, Cambridge, they can be calibrated with Leach’s military and civilian war record. He worked in both military and civil administration com-

29

In “Formative Travail….”, p. 192.

the biographical origins of POLITICAL SYSTEMS OF HIGHLAND BURMA

15

munities, and was mentioned in ofcial dispatches from and to the ofce of the Governor of Burma (in exile in India). His administrative life intertwined with R. C. Wilkie, John Leyden, and Noel Stevenson, and they kept in touch after the war through letters and visits. He met Governors and Generals, and managed coolies and killers (the title of Ian Fellowes-Gordon’s book on Kachin guerrillas organized by Leach in this period is ‘Amiable Assassins’). The project I am undertaking in the Leach Papers is to develop a chronology of his Burma activities and to trace the geography of his movements and stays, then to estimate the inuences on him in London from 1945 while writing his radical 1947 dissertation, and then to weigh the inuences that changed that dissertation into the nal manuscript of Political Systems in 1952–1953. This provides the basis for weighing the kind of experience he had against the kind of things he learned and wrote about the frontier. Thus, Leach’s experience in China, Britain, and Iraq is relevant. In addition to the Leach Papers, there are references to Leach and the specic context in the war surrounding him (or to Stevenson, Wilkie, Ogden or others) in London in the British Library and the National Archives, where I have read about 9,000 pages of these les, looking for references to Leach. Without intending to, I have learned much about guerrilla warfare in the 1940s in highland Burma. The main Leach sources during the war that are available in the Kings College Archive are the Olive Notebook, the Red Notebook, the Blue Indian Notebook, and the Blue British Notebook, and the Huge Black Folder (ERL 4/29).30 I list only some of the documents here, noting that there are relevant pages tucked in notebooks and loose sheets showing key information, such as his destruction of secret telegrams and orders for ve tons of rice, alongside sketches of textile designs. The following are some of the larger titled texts: 1) There is a 5 page, typed, untitled essay on the political history of Sinlum Hills in the Kings College Archives, which addresses the same questions as his eldwork. In 1942 Leach rewrote a draft monograph about Hpalang while on sick-leave with dysentery in

30 I am grateful to Stephen Hugh-Jones for introducing me to the Leach Papers, and to Kings College archivists for facilitating my use of them. See also Stephen HughJones, “Edmund Leach 1910 –1989” Cambridge Anthropology: special issue on Sir Edmund Leach, vol. 13, no. 3, 1989 –90.

16

robert anderson

Calcutta. “But during his subsequent extensive travels in the Kachin Levies operation, he was fated to lose that document too,” according to Tambiah.31 So, this untitled essay was probably written soon after that loss, perhaps in early 1943. Under military regulations, a personally written report was less likely to raise curiosity if it were untitled; the Sinlum Hills were rmly in Japanese hands throughout the war, though Leach’s Kachin scouts from Putao moved in and out of them. 2) 1943 September, “Report by Major E. R. Leach on Economic Conditions in Namtammai Valley.” Leach was sent, with others, over the 5000ft pass to this frontier area east of Putao to rescue a British ofcer who had been taken hostage by Chinese troops, and Leach wrote these observations in that context. 3) 1943 September, E. R. Leach, “Miscellaneous Word List Recorded at Kawglangbhu on the Namtammai River due east of Fort Hertz.” “A high proportion of the Nung with whom I was dealing at the time spoke Kachin, and all my transactions with them were through Kachin interpreters,” wrote Leach in this document. This list does not contain words of great military signicance, though Leach was there to interview Kachin and Nung (and Chinese) observers about the kidnapping from which the British ofcer had to be rescued; the interviews were to establish grounds, if any, for discipline and compensation. A linguist’s study of the quality of this list would be rewarding. 4) 1943 November [?], E. R. Leach, “Jinghpaw Kinship Terminology: an experiment in ethnographic algebra,” accepted for publication in a 1945 issue of the Journal of Royal Anthropological Institute, Vol. 75, pp. 59 –72: this issue was actually printed and distributed in 1948. Leach spent a month in late 1943 in a Calcutta hospital due to malaria and dysentery, and this was most probably the occasion of writing this very technical and experimental essay. It would have been written sui generis because it is very unlikely that he had access to any recent writing about kinship (or even anthropology in general). He made only minor corrections on the proofs in 1947 before publication. A linguist’s study of the terminology in this paper would establish Leach’s degree of competence with Jinghpaw at this time.

31

Tambiah, 2002, p. 42.

the biographical origins of POLITICAL SYSTEMS OF HIGHLAND BURMA

17

5) 1944, E. R. Leach, “Sexual Behaviour Under Wartime Conditions in Eastern India”: 3 pages in pencil. Leach spent ten days in Delhi in hospital in 1944, and this essay may have originated during that sojourn: US and British commanders had initiated a joint campaign against sexually-transmitted diseases in eastern India and Burma, and this may have been part of that campaign. 6) 1944 September, Major E. R. Leach, “Tour Notes—Nogmaung”: Leach accompanied a senior British ofcer to tour unoccupied northern Burma and report on economic and agricultural conditions, and this report is part of that tour (and an implicit corrective to some of that ofcer’s impressions). The key centre of Japanese occupation was Myitkyina, and, after a prolonged battle, they withdrew from it in September 1944, permitting much freer movement. 7) 1945, Major E. R. Leach, Army Military Record (largely the medical record of his military service). Doctoral Student, Home for a While When Mandalay was re-captured by the Allied Forces in March (heavily fought) and Rangoon (unopposed) in May 1945, Leach, almost thirtyve years old, was brought home on compassionate family leave— although, he thought, temporarily. He stayed in Britain from July, and was then demobilized six months later. He lost little time returning to his family and studies. He registered as a doctoral student under the supervision of Raymond Firth, and produced the following fascinating early paper on his war experience three months after arriving in London. He claims the gures given in it are hypothetical, not statistical: 8) 1945 November, E. R. Leach, “Currency Ination in Kachin Tracts”, for Rayomd Firth seminar, November 1945, (typed with marginal notes). Leach was deeply involved in economic-warfare involving trade, jobs, and currency, and was very knowledgeable concerning the opium economy. This paper will be useful for interpreting his eld notes, which are full of references to these variables in his work. I note that the numbers in the paper correlate well with the military les, so there is no good reason to suppose that the data and argument were as ‘hypothetical’ as Leach suggested. 9) 1945 November, “Memorandum on Anthropological Research in Reconstruction and Development of Burma”, presented to the Secretary of State for Burma by the Royal Anthropological Institute.

18

robert anderson [ No month/date, but context suggests late 1945]. Leach was a member of the Board/Council of the RAI by 1946, before he received his doctorate. In London from 1945, Leach was probably consulted regarding this memo, and may even have written part of it. Noel Stevenson had proposed a new Department of Ethnography for the University of Rangoon in 1942, and would have discussed his plan with Leach; such a department was established in Rangoon in 1948.

Leach’s private/professional correspondence in 1945–1947 reveals the beginning of a more professional community of anthropologists, faced with exciting new research opportunities. During the period in which Leach wrote his dissertation, he considered an opportunity to return to northern Burma. Encouraged by Stevenson, but interested on his own account, he considered doing ‘applied anthropology’ in the context of a planned railway corridor through Kachin State. Although Stevenson dropped abruptly out of the ofcial picture in early 1947, Leach’s old friend John Leyden replaced Stevenson as senior ofcial for Frontier Areas in early 1947.32 Leach considered his project possible only if he had other funds for his work to make him independent of the 32 The Leach-Stevenson relationship warrants more attention if one is to understand the origins of Political Systems. The two men had many conversations about anthropology. Stevenson wrote “The Case for Applied Anthropology” in 1944, and published “The Economy of the Central Chin Tribes” at the Times of India Press, Bombay, 1943. This latter work was recommended to others by Leach, although, in spite of his own practice in Burma during the war and Sarawak in 1947–1950, he did not approve of applied anthropology. Stevenson became, by June 1945, the chief secretary to the Governor of Burma. He was exceptionally well informed about policy and politics because his wife was cipher clerk in the Governor’s Ofce in Rangoon when that ofce was reestablished in 1945, [according to les in the British Library]. In 1945 he was awarded the OBE, and in 1946 Stevenson was appointed Director (in Rangoon) of the Frontier Areas Administration, a post he had coveted for a long time. After that, his career in Burma ended abruptly and unceremoniously in January 1947. Stevenson and Leach communicated and met in London while Leach was considering returning to Burma in 1946–47, and they appear to have maintained an irregular communication during the next few years. But up to the writing of Political Systems in 1952–3 (after Leach had achieved a doctorate) Stevenson was still working toward a university degree at Oxford, an opportunity he had not had during his entire career in Burma nor while at the LSE in 1937. Stevenson’s lack of formal qualications and credentials was one of the barriers to their professional relationship, and Stevenson’s higher status in Burma, both in military and administrative terms, would not have been unnoticed by Leach. Stevenson could not have felt exempt from Leach’s criticism of British administration in Burma. Leach would not have done some of the things he did in northern Burma were it not for Stevenson. This will be a subject of my forthcoming book Highland Burma, Edmund Leach, and Anthropology.

the biographical origins of POLITICAL SYSTEMS OF HIGHLAND BURMA

19

Government of Burma. He applied to the Horniman Trust (for funds administered by the Royal Anthropological Institute, of which he was now a board member). He was fully briefed by both Stevenson and Leyden on the complicated situation on the frontier. His logistical support would come from the new frontier administration (under Leyden), but was still dependent on Rangoon’s cooperation. And the British were on their way out in 1947; even troops were being withdrawn. His referees for this project, which was given ofcial government support, were Raymond Firth and Sir Rupert Howorth of the Lord Chancellor’s Ofce (Leach said his ‘grandfather gure’ had been Sir Henry Hoyle Howorth, scholar, art collector, traveller, and uncle of his mother). Leach nally deposited his dissertation for typing and examination in February 1947, and the exam itself appears to have gone ahead in March. Over 600 pages in length, it must have been written (and typed) like Beethoven composed music, in a white heat. It has a huge bibliography because Leach was permitted to do a ‘library-based’ dissertation due to the special circumstances of the loss of his notes. It is called “Cultural Change, with specic reference to the hill tribes of Burma and Assam” (unpublished PhD dissertation, University of London, 1947). Although supposed to be library-based only, it contains much direct evidence and a trenchant, running-critique of recent colonial, educational, and missionary practice, along with a nearly total dismissal of the popular theories of anthropology on which most of that practice was based. Alfred Radcliffe-Brown of Oxford was the external examiner. Readers of the dissertation at the University of London have been relatively few in number, though at least two microlm versions were approved by Leach and deposited in foreign university libraries like Columbia and Bergen. Leach later recalled that Malinowski himself was preoccupied (when Leach was closest to him in 1938–39) with “the application of his functionalist method to what were described as situations of culture contact.”33 This was precisely an approach Leach took in his 1947 dissertation: how cultural contact leads to cultural change. His experience with the dissertation more clearly revealed to Leach the limitations of the method and approach of the culture-contact model (he was interested in a German geographer’s work at the time). One wonders how much of the dissertation was shaped by his supervisor Firth’s instruction;

33

Tambiah, 2002, p. 414.

20

robert anderson

given Leach’s age and independence of mind, I would say ‘not much’, except for Leach’s own notes following a meeting with Firth saying that it was to be cut by a third. Dr. Leach Changes Plans at LSE as Burma Changes Just after completion of his dissertation, the Horniman Trust decided not to fund Leach’s work; behind the scenes, members of the Royal Anthropological Institute and the powerful Burma Ofce in Whitehall suggested both that the project was no longer a good idea and that Leach was unsuitable for it (considering local memory of his military activities), whether or not the railway line was built. Leach was rst keen and then not keen on the project, and declined to continue to look for other funds when the Horniman Trust turned him down. He then seems to have let go of Burma. His family’s reluctance, his traumatic wartime experience, and the Trust’s rejection all inuenced him. But Leach’s change of plan in 1947 has to be calibrated against regional geopolitics as Burma’s independence approached. Again, he was kept informed by Noel Stevenson (now out of government altogether) and John Leyden (now in full control) throughout this period, both very knowledgeable about current policy. There had been a nal break of the Independence crisis in 1946 in Rangoon, when Viceroy Mountbatten, on London’s instruction, appointed General Aung San to the Executive Council. The British administration, including the new Governor Sir Hubert Rance (Mountbatten’s former deputy commander) made it clear that the Anti-Fascist Peoples Freedom League (AFPFL), led by Aung San, would be the most important intermediary to negotiate with British power moving towards 1948.34 Aung San had a rapidly expanding party-based militia ‘Peoples Volunteer Organization’, armed and led by former Burma National Army veterans. The British accepted him because he was very powerful and capable of negotiating with the Communist Party of Burma, but close enough to them to be well-informed and, thus, to negotiate without them, too. In this tense and contradictory scenario, Kachin units that had fought against the Japanese were moved into the Burma Army, along with units of the Burma Independence Army who 34 The following paragraph relies on information in Martin Smith’s excellent Burma: insurgency and the politics of ethnicity. London, Zed Books, 2001.

the biographical origins of POLITICAL SYSTEMS OF HIGHLAND BURMA

21

had fought with the Japanese. AFPFL activists were working for their cause in Bhamo and Myitkyina, and were trying to recruit support in the hills for Burma’s independence. The Shan king-maker inside the AFPFL, Sao Shwe Thaike, elected President of the ‘Supreme Council of United Hill Peoples’, was asked to be President of the new Union of Burma on independence in 1948. It does not appear that the Kachin had an intermediary of the calibre of the Sawbwa Thaike. One AFPFL plan allowed Kachin autonomy over the Kachin State, but minus the boundaries of the two most important administrative and market towns of Bhamo and Myitkyina. Kachin leaders were increasingly aware of the ambitious AFPFL negotiations conducted by the Shan and Karen people with Aung San, especially after the 1st Panglong Conference in January 1946.35 There was evidence that insurrections against the AFPFL were going to be attempted, with the help of sympathetic Britishers loyal to Karen soldiers and ofcers with whom they had fought in the war: but the approach appears to have been disorganized and was revealed to authorities. Far away in London, Prime Minister Attlee asked in early 1947, why would he consider independence for the Shan or Karen just when he was trying to stave off independence demanded by the Indian princely states like Hyderabad? What logical conclusion about autonomy would the Kachin leaders draw from all this? Suspicious Burmans wondered if the British might be trying to line up independent territories on the frontier with China, having realized that they would have to give up central Burma. Were the British going to use independence, or delaying independence, to bargain for those independent frontier territories and a role in them? In this atmosphere, Britishers who advocated Kachin autonomy like Stevenson, or who were close to Kachin leaders, like Leach, were enfolded in that suspicion, particularly in Rangoon. Karen leaders continued to make pathetic petitions to the Burma Ofce in London, but these were disregarded. Sensing this futility, others plotted armed struggle. As Leach was nishing his dissertation and considering the offer to return to Burma to do research, Governor Rance cabled London to conrm his commitment to one united Burma, and General Aung San visited Attlee in London in late January 1947 without a representative of any minority present. Kachin leaders heard of this, and

35 Kachin versions of this historic process in 1945–1948, based on Jinghpaw sources, will doubtless modify this simple narrative.

22

robert anderson

cabled London to say they “would not regard any agreement as binding.”36 After attending the 2nd Panglong Conference in February 1947, (where Leach’s friend John Leyden was in charge of proceedings), the Kachin found that the new constitution would allow for a 19 member Kachin Council, with half the members to be elected Kachin from the hills, but the other half had to include non-Kachin representatives from the towns (deferring to AFPFL activists in Bhamo and Myitkyina, and to signicant minorities, including Shan, Indian, Chinese, etc.). Leach appears to have realised that the Jinghpaw speaking peoples would not get an independent state and would have only as much autonomy as could be negotiated within a federated Burma (which he knew would be dominated by Burmans). This was not unlike Noel Stevenson’s position on Kachin autonomy, but Stevenson made more public his suspicion about Burman and British intentions. In fact, Stevenson’s advocacy for greater Kachin autonomy (in part as recognition of their part in the war) ran into the limits of the new British realpolitik. Leach’s knowledge of the Kachin (and theirs of him) was thought to be a risk in this atmosphere, even if Leach was not prepared to go as far as Stevenson. With the door to Burma closing, Leach changed direction quickly in spring 1947, and accepted an applied anthropology contract with the Colonial Ofce for a study of the newly-acquired territory of Sarawak (previously under the control of a British Rajah and a commercial company). Firth had met the new Governor of Sarawak in 1946 and learned of the British Government’s intent to manage this resourcerich, multi-ethnic country more wisely than the Rajah and Company had done, to make it a colonial ‘showcase’. Under the supervision of Raymond Firth and funded by the Colonial Ofce, a series of detailed studies were conducted by other anthropologists. Leach based his proposals on ve months of eldwork in Sarawak, where he travelled extensively in late 1947: 10) 1947 Sarawak Project for Colonial Ofce, published as Edmund Leach, Social Science in Sarawak, Colonial Research Studies, no. 1; London, HMSO, 1950. See also Leach’s statement of qualications written for the purpose of this contract, in which he details

36

Smith, 2001, p. 78.

the biographical origins of POLITICAL SYSTEMS OF HIGHLAND BURMA

23

his anthropological and war experience. Leach travelled extensively through Sarawak during July–November 1947, and the direction in which he pushed the project suggests how he would have inuenced the development of anthropology in Burma, were he to have been there. Students at the LSE worked in the eld in Sarawak under his supervision. His colleagues accepted his importance and inuence, enhanced by Firth’s evident condence in him; by 1950 Leach had been promoted from Assistant Lecturer to Reader at the LSE, and was elected to be Chairman of the Executive Committee of the Royal Anthropological Institute from 1948 to 1952. 11) 1949 BBC talk on “The Kachins of ‘Burma’ ”, 18 October 1949. This radio talk refers to his war experience on the frontier, but reveals little about it. He ‘anthropologizes’ the Kachin situation, making it almost timeless and saying little about the present. But he was very much in the semi-ofcial communication network, tracking the careers and political fortunes of Kachin leaders he knew during the war through the transcription of letters in Kachin supplied to him by old frontier colleagues. Reticence about the war was common at this time, particularly among those who were in Burma. 12) 1950 August correspondence with the Yale University Human Relations Area File Project (HRAF ) pointing out errors in their les on border-area languages, geography and ethnicity, while Leach criticized the drawing of ethno-linguistic boundaries along the frontier: not that there should not be any [there were] but that the HRAF experts had got it wrong. His old army friend R. S. Wilkie showed Leach a number of informative letters he had received from Burma, written in the Jinghpaw language in 1950 –51, and typed by a translator. 13) 1951 Leach was still reading in preparation for his coming book. “There are one or two documents which I have never managed to trace, but in general, excluding ephemeral publications issued by the missions, I think I have at one time or another probably read nearly everything that has been published in English, French, or German about the Kachin Hills Area in the past 130 years” [Political Systems, p. 312]. There is little evidence of regular communication with Jinghpaw-speaking people in northern Burma, but certainly the occasional visit and letter.

24

robert anderson

Leach resigned from LSE in 1952 to write Political Systems, apparently able to withstand the economics of ‘the writer’s life’.37 While seeing the book through a typist in London, he offset his reduced salary by completing a major contract with UNESCO in Paris, focused on a technical mission to Ceylon. He also travelled to UN ofces in Rome and Geneva for this study. This may be the origin of his interest in Ceylon, where Leach went two years later to do eldwork. Though a regular at LSE during this period of leave, he had little to do ‘ofcially’ until he accepted an appointment at Cambridge in 1953. 14) December 1952, Edmund Leach, “Report to Unesco on Technical Assistance”, 16 December 1952, (in fullment of obligations under contract No. ss/350.772), Kings College Archives, ERl 2/6. See also draft of this report, notes to Ralph Phillips [of UNESCO], both of which express Leach’s critical views of current development policies and practices in bolder form than in the nal version. Tambiah comments on this critical position on development, “Leach’s neglect or amnesia regarding Malinowski’s programmatic advocacy of his version of ‘practical anthropology’ joined with the study of the dynamics of the colonial encounter may be, I surmise, linked to his negative attitude toward ‘applied anthropology’ in the service of alleged economic and social development of subject native populations.”38 What he planned for Burma in 1947, and actually did in Sarawak in 1948–50, was indeed applied anthropology, but he nevertheless contested current development thinking and practice. He said it had to be less Euro-centric, less supercial, and less top-down. The actual date of completion of the manuscript of Political Systems is 19 January 1953, shortly after his UNESCO work concluded. Because

37 There is a curious contrast between his resigning and taking a year off from LSE in 1952 to write Political Systems and his later statement that he was interested in Cambridge because it offered him a higher salary than LSE. Because his father’s will was not yet executed (and his mother was still alive), said Leach, “I was rather hard up . . . I was in debt” (source: notes of Susan Drucker-Brown’s interview with Edmund Leach, Cambridge, April 1984. Quoted in Tambiah, 2002, p. 50). But in his 1984 review essay “Glimpses of the Unmentionable in the History of British Social Anthropology” Annual Review of Anthropology, Vol. 13, 1984, pp. 9 –10, Leach commented acidly about the willingness of well-paid academic anthropologists elsewhere to give up such jobs and accept lower pay if they could move to Cambridge (or Oxford). His case seems to have been the very reverse, and perhaps he overlooked that. 38 Tambiah, 2002, p. 447.

the biographical origins of POLITICAL SYSTEMS OF HIGHLAND BURMA

25

of the nature of printing at the time, there was little room to change the manuscript after it was set in type. He was re-appointed to the LSE to teach at this time, during which Meyer Fortes (known since his Malinowski seminar days in 1937–38) continued to encourage him to move to Cambridge. Later that year, he accepted an appointment as Lecturer at Cambridge. Within a year, however, in mid-1954, Leach was off to do eldwork in the irrigated hillside villages of Ceylon, during which time the Political Systems book appeared. After the 1954 Publication of POLITICAL SYSTEMS There are dozens of references by Leach to Political Systems in post-1954 papers, letters, and publications. In reply to these communications, Leach sometimes wrote or said something that enlarges our understanding of his pre-1953 experiences and how they shaped the book, but it was twenty years until he began to talk personally in public about the book’s origins. The most important of his publications about Burma after Political Systems are the 1960 essay which sets Political Systems in a larger context of the regional powers and their cultural inuence, and the 1962 essay on the political future of Burma, written after the March 1962 coup in Rangoon.39 It is the disconnection of the specic Kachin-Shan analysis from its regional context and the geopolitical inuences that is striking in Political Systems, and Leach puts this context and inuence back in the picture in these two later works. He conceded in 1987 this intellectual disconnection in Political Systems as an error (though he may have spoken informally about this much earlier, as the disconnection is quite obvious). Mandy Sadan has written eloquently on this subject in an accompanying chapter (this volume). 15) In 1973, Leach presented “Sovereignty and the Northern Frontiers of Burma” as the Kingsley Martin Lecture, Cambridge; this was a

39 In 1960 Leach wrote his inuential essay “The Frontiers of ‘Burma’ ”, Comparative Studies in History and Society. Vol. 3, no. 1, 1960, pp. 49 –68. In September 1962 Leach completed “The Political Future of Burma”, published in Bertrand de Jouvenel (ed.) Futuribles: studies in conjecture. Geneva, Droz, 1963, but originally written for and presented at the Congress for Cultural Freedom meetings in late 1962 at the Institut Universitaire de Hautes Etudes Internationales, in company of Edward Shils, Oskar Morgenstern, Karl Popper, Michael Postan, Bertrand de Jouvenel. Leach refers to ‘falsiability’: is this his rst brush with Popper? Source: Kings College Archive, ERL 1/62.

26

robert anderson

succinct summary of Leach’s views of Burmese history from 1885 up to the Ne Win coup in 1962. 16) In 1973, Leach wrote an essay for an LSE seminar on anthropological practice in colonial situations; Leach, Firth and others were invited to address the charge that British functionalist anthropology was ‘peculiarly well suited’ to the tasks of colonial administration. Leach used this occasion to discuss the circumstances of Political Systems with more candour than he had yet done publicly, and I have already quoted from this essay.40 17) Leach gave a lecture called “Political Systems of Highland Burma: a retrospective assessment” at the University of Washington in Seattle in March 1980, wherein, according to notes taken at the lecture, he conceded that his book had been somewhat ‘reductionist’ in that it used categories like gumlao, gumsa, Shan, Kachin, and Burmese “while leaving out the play of historical forces on their usage.”41 He conceded that the categories were used variably, during his time in the eld in Burma and subsequently. He said he had been intent in 1954 to apply Pareto’s concept of an optimum built into homeostatic model societies, to develop a thesis of oscillation between the gumsa and gumlao tendencies in Kachin life. This conceptual narrowing, part of a doctoral student’s attempt to x data to a theory, helps partially to explain the loss of the rich context that he could have included in Political Systems. Of his thinking before writing Political Systems, Leach said in 1985 that “I never subsequently practiced as an engineer, but my engineering background has inuenced all my anthropology. I tend to think of social systems as machines for the ordering of social relations or as buildings that are likely to collapse if the stresses and strains of the roof structure are not properly in balance. When I was engaged in eldwork I saw my problem as trying to understand ‘just how the system works’, or ‘why it held together’. In my own mind these were not just metaphors but problems of mechanical insight.”42 This is consistent with his fascination

40 “In Formative Travail with Leviathan”, Anthropological Forum, 1977, Vol. 4, no. 2, pp. 190 –197. 41 Tambiah, 2002, pp. 429 –30; the notes were taken by C. F. Keyes at the University of Washington. 42 “Glimpses of the Unmentionable in the History of British Social Anthropology” Annual Review of Anthropology, Vol. 13, 1984, pp. 9 –10.

the biographical origins of POLITICAL SYSTEMS OF HIGHLAND BURMA

27

with the calculus of textile and boat design, of which he made elaborate drawings, just as he did with diagrams of the algebra of kinship. He appears to have worked out his conceptual problems graphically, and appears to have had a photographic memory. The idea of stresses and strains in proper balance is a key concept for Leach, and gives us an idea of what he believed was driving social change. In November 1985 Leach taped an interview with Adam Kuper, published as “Interview with Edmund Leach”, Current Anthropology, (vol. 27, no. 4, August–October, 1986 pp. 375–381). Kuper is an historian of anthropology, and there is passing reference to Political Systems. The discussion evidently led to a public treatment of his eldwork experience, about eighteen months later: in April 1987 he lectured on ‘Tribal Ethnography’ at a professional anthropological meeting at the University of East Anglia in Norwich, wherein Leach admitted that at the time of writing Political Systems in 1951–53, “I still accepted the conventional view that my task was to discuss the indigenous social system of which I myself was not a part. Thus the missionaries and colonial administrators and the British military recruiting ofcers were not really part of my story. I see now that this was a mistake.”43 One concludes that his extraordinary knowledge of the situation in Kachin State in 1939 –45 was channelled through a spiral of formalization from 1945 onward, in which the 1947 dissertation sets out on paper a good deal, but not all, of what he knew about the interplay of gumsa-gumlao dynamics within the missionary, juridical, socio-economic, and military context of the 1930s and 1940s. There is then a greater ‘anthropological’ formalization between the completion of the dissertation in 1947 and the submission of the Political Systems manuscript in early 1953. In this process, even more of the relevant geopolitical, military and colonial administrative context is dropped away, leaving the axioms or formulae that Leach thought were underlying Kachin political life and kin relations on their own. This became known to others as ‘the indigenous social system’. Ironically it was the whole context, and its interplay with the indigenous formulae, that he knew best. But something made him set most of this hard-won dissertation

43 “Tribal Ethnography: past, present, future” lecture to Association of Social Anthropologists, [ published in Cambridge Anthropology, vol. 11, no. 2, 1987, pp. 1–14]; in quoted in Tambiah, pp. 347–48.

28

robert anderson

knowledge aside in Political Systems, except in a few suggestive footnotes, leaving the published version an elegant but less complex work than the dissertation. The intricacies of the kinship systems and their interplay with the gumsa-gumlao and Kachin-Shan political dynamics now seemed to stand on their own, set in Leach’s carefully-described ecological situation. It is quite difcult to assign weight concerning which things are more important and less important. My work now necessarily includes the private Leach, interested in psychoanalysis and committed to his family and friends, the cloak and dagger Leach in the eld, making notes on the order of battle of the Chinese Army, the restless anthropologist Leach writing kinship articles in his hospital sickbed, and the anti-colonial and non-military Leach in the eld, working out word lists or diagramming loom designs, preparing for a new kind of anthropology in the middle of a wild and sometimes boring war, and the writer-broadcaster-public intellectual Leach. The guiding questions for the next stages of my research are as follows: a) Did the intensity and unpredictability of the war favour a Kachin social and kin relations structure that had to be particularly adaptive to rapid change in the economy, sudden swings in political fortune for households, and sudden (and lethal) shifts in alliances? b) How did the war around him inuence his perception of Kachin life, and how did his British ofcer status in the military’s civil administration (1941–1945) inuence Kachin perceptions of Leach—thus shaping what they told him and showed him? c) Did Leach’s reaction and resistance to the common thinking about ‘tribes’ as closed systems embolden his claim that there was ‘no such thing as a Kachin’? Did the skilful and advantageous use by Jinghpaw-speaking people of that same administrative ‘closed system’ (the ‘tribe’) contribute to Leach’s own resistance to it? What did Kachin intellectuals think of this ‘tribal’ closed system approach, which colonial administrators and missionaries had adopted? d) When Leach wrote his dissertation in London (1945–47), did he run into old Burma hands (or other anthropologists) who preferred thinking of ethnicities and linguistic groups as bounded or coherent tribes? Did these encounters further propel him to describe Kachin territories and peoples as ‘unbounded’, and further push him toward an ‘open system’ concept, reinforcing his conviction that he should

the biographical origins of POLITICAL SYSTEMS OF HIGHLAND BURMA

29

proceed with his anti-colonial, independent-minded, even radical, approach? e) The ambient intellectual climate of the late 1940s was increasingly charged with the ‘systems’ idea, forged during the war and articulated more and more by mathematicians like von Neumann and Turing, et al.: did Leach encounter such ideas in journals, seminars, pubs, newspapers? Were the ideas of Norbert Weiner’s cybernetics available to him? f ) When Leach wrote about ‘political systems’ in 1951–3, what did he understand them to be, and did he think he had re-framed his dissertation in line with a new kind of social science? Was he involved in debates and dialogues at the LSE and across London in this period that helped him to articulate, defend, and rene his ideas? g) Did communication with Burma and about Burma in this pre-Political Systems period (1945–53) inuence Leach in the formulation of his argument? Why did he decide to write a ‘technical book’ particularly for anthropologists, minimizing the inuence of the geopolitical and historical context, in spite of the fact that those were the things he knew so well, and that the audience for it would have been even larger than for Political Systems? Did responses to Political Systems show Leach that the wider context in fact interested people and was essential for a full understanding of the situation in highland Burma (even the receding situation, as it indeed was?). Do his essays about Burma in the 1960s show that he quite quickly conceded that point, though never losing sight of his search for underlying structures? Like the chapters in this book, answers to these questions contribute to a deeper understanding of the origins of Political Systems of Highland Burma, and may assist us to understand its very widespread reception. Whether Leach’s understanding actually corresponds to the situation of Jinghpaw-speaking people in highland Burma in the period 1939 –1953 is another kind of question, to be answered by a number of people, including the Kachin people themselves using their historians, memories and records. We have to ask what is meant by ‘corresponds to the situation’, how Kachin descriptions might make more sense or different sense than Leach’s; Sadan’s chapter here speaks eloquently to that difcult question. What is established in this book is that these questions about Leach’s life circumstances and intellectual development are well worth studying, because the ideas in it have such wide and fascinating applications.

ON THE CONTINUING RELEVANCE OF E. R. LEACH’S POLITICAL SYSTEMS OF HIGHLAND BURMA TO KACHIN STUDIES Maran La Raw Introduction A brief overview Sir Edmund Leach’s study of Kachin social structure appeared as Political Systems of Highland Burma in 1954, and proceeded to have substantial inuence over the study of human society and culture. There are indications that it may well have been one of the most widely read works among published anthropological literature during the second half of the twentieth century (see Anderson, this volume). The reception the book drew from Kachin readers was no less profound, and it is the purpose of this chapter to give an overview of that reaction, spanning the last fty years.1 We begin this review of Kachin reaction with the observation that by 1960, Kachin concerns had gelled into a coherent set of issues to address, and that out of that consensus grew a research agenda that has been maintained, to the extent possible, since then. At a later point in this chapter I will explain the basis of this statement. The second observation that must be made early in this chapter is that it will be written from the perspective of a person who has been part of the Leach-Kachin story since 1954. I want to discuss briey below the issues that underlie these two observations. 1 Acknowledgements: I owe a debt of gratitude to Pungga Ja Li and Maran Ja Gun, my long-time, trusted research colleagues on Kachin culture and history. Through long talks in cramped quarters in obscure places we have kept up with the agenda of 1960. To follow us, there are youthful Kachin students of anthropology now. I also want to thank my wife Kit, aka NHkum Ja Mun, for her unfailing support and editing skills, and my daughter Mina for showing strong interest. Finally, I want to thank Mandy Sadan for getting me involved in this volume, and for the editorial help. This paper and the ones to follow are dedicated to the memory of my mother, Hpauyam Mahkun and my Uncle C. Htang Wa; in 1960, they shaped the course I have followed. And special thanks to the memory the Rev. Duwa Zau Tu of Pangmu-Sinlum; what great interviews he gave!

32

maran la raw The beginning of concerted Kachin-Leach interaction in 1954

There are four important reasons why Kachin reception of Political Systems in 1954 was enthusiastic but, at the same time, neither haphazard nor lacking in focus and direction. First, in 1948 Burma gained independence from Britain and the Government of Kachin State came into being. At the highest level of Kachin State Government in Myitkyina, and the Ministry of Kachin State in Rangoon, sat Kachin administrators who had been in the British colonial administration at signicant levels. From the standpoint of high-level administrators in the colonial and immediate post-colonial regimes, there was substantial dovetailing. To a man, they knew Edmund Leach as a British anthropologist who spent the years 1939 –45 among the Kachin people. One of them was the language assistant that Leach felt obliged to use for only three months. The point here is that, among some highly inuential Kachins were those in positions of responsibility who felt the need for an assessment of Political Systems as early as possible. Second, the post-colonial era started with a form of parliamentary democracy. By 1954, Kachin political leaders realized that they probably would not get what they stipulated in a meeting with General Aung San at the Panglung Conference in 1947.2 At Panglung, Kachin delegates stipulated that they wanted a federal system with a politically autonomous Kachin State in exchange for agreeing to co-found the Union of Burma. By 1954 they were convinced that constitutional reform was necessary. The unexpected pressure to assess the relevance of Political Systems quickly, especially in matters relating to the frontiers of Kachin State, came from this angle. Leach was, after all, a colonial era British anthropologist and it was not insignicant that Kachin delegates at Panglung were, for the most part, gumsa chiefs. This characteristic of Kachin political leadership continued through 1962, when Burma’s armed forces overthrew the elected government. Third, there has been a tradition of private individuals collecting information about Kachin culture and history, and producing self-published monographs and books in Jinghpaw Kachin. These reports gave detailed descriptions of social systems and their historical backgrounds at the local level. Kawlu Ma Nawng’s (1942) work on the gumlau Kachins 2 This is the preferred spelling in Jinghpaw today of the town that in Robert Anderson’s chapter, and elsewhere in historical colonial records, is spelt Panglong.

on the continuing relevance to kachin studies

33

of Hukawng Valley was not an isolated instance. Noteworthy in this genre were books by Sumlut Gam (1955) and Htingbai Naw Awn (1956). So, there was a strong perception from the rst moment of publication that Political Systems was an incomplete work that contained errors. This in itself had an impact on the research agenda already mapped out before 1954 to study Kachin culture and history. The effort seemed vigorous and focused after 1954. Fourth, by 1954 there was a generation of college-educated youth with deep intellectual commitment to matters concerning the political future of the Kachins, a situation that approached outright agitation with the ceding of three Kachin villages when the central government, dominated by ethnic Burmans, settled the boundary with China in 1955. Political Systems was heatedly discussed in this circle in 1954. It is no accident that, after General Ne Win’s coup d’etat in 1962, Kachin political leadership passed into the hands of college-educated young leaders. These factors pushed the Kachin to respond to Political Systems by developing a research agenda of their own. This was implemented in 1960 with the appointment of the author as the rst Kachin State research anthropologist. But a brutal civil war began with Ne Win’s coup d’etat, a state of affairs that by the end of the 1960s had worsened into an outright civil war against the non-Burman co-founders of the Union of Burma. As the 1960s ended, the author, who was in formal studies in social anthropology in the U.S. at the time, was removed from his appointment. The seriousness of this action reected the seriousness of the Kachin rebellion. However, this did not end Kachin research on culture and history; it became more centred on the efforts of one person. Due to war conditions in Kachinland, the former government anthropologist became, by default, the keeper of the Leach-Kachin agenda. It became my responsibility to present the topics and issues to be discussed with Kachin colleagues in safe locations in Kachinland. As an example, between 1991 and 2004 twelve such meetings were held. After the meetings of 1997, activities began to be accelerated. There is again today a focus and sense of immediacy that has begun to approach the situation in 1960. One characteristic of the work following the appearance of Political Systems in 1954 is the sense that our anthropological research has been a group effort all along. For this reason, I frequently use the plural “we” instead of the singular. Topics mentioned in this paper have been touched upon frequently in group

34

maran la raw

discussions, and additional data have been requested from participants with access to the regions concerned. While I assume responsibility for the anthropological reading and analysis presented here, I acknowledge explicitly that the work could not go on without the contribution of co-workers from within Kachinland. Our report on the “gumsa-gumlau” systems that are central to the theoretical thrust of Political Systems is approaching conclusion now as a fully collaborative effort. Such is the background to this chapter. I will now return to the beginning and tell about the purpose and scope of what will be covered in the following pages. The onset of the Leach-Kachin story This chapter will iterate and elucidate how Political Systems has continued to have inuence on the study of Kachin culture and its institutionalized functions, such as kinship, marriage and alliance, political organization, and change and adaptation, etc. However, this is not a review of Political Systems; it is a statement of our understanding of Political Systems’ limitations. This chapter is about what Kachins felt they should do after 1954 when they initially examined the book. I want to accentuate the fact that we have added much to our understanding of specic phenomena, such as gumlau and gumsa societies. There are actually varieties beyond that which Leach showed in both of these systems, and I will attempt to summarize how they differ and how they are similar. It may even appear at times that this chapter is working to annotate his work. This is not surprising considering that Political Systems serves as a synthesis of published literature that spans from Neufville (1828) to Kawlu Ma Nawng (1942). The text of Political Systems is a substantial reservoir, whether one is concerned with early accounts as history, or with using them as an outline to amend and enhance the value of the material by means of research. These are matters of great priority in our research today, and these issues are taken up in Part I of this chapter. This chapter is also the report of a Kachin who, after having read Political Systems in 1954, became a social anthropologist and communicated with Leach. It is an account of a Leach-Kachin story spawned by circumstances just mentioned. It does not pretend to be full or comprehensive, and it is one person’s recollection of what actually should be written by many. Each of us who has shared interest in Political Systems has found the last fty years to be difcult and frustrating at times, and yet we are also lled with gratitude, since our research would have been far more difcult without this impetus to our work.

on the continuing relevance to kachin studies

35

The Leach-Kachin encounter—an example Let us briey consider an example related to the issue of common identity that Kachins share, and the foundation from which such things may develop. In his work, Leach characterized the over-arching term ‘Kachin’ as referring to communities that are discrete linguistic categories: Maru, Lashi, Zi, Lashi, Nung, Rawang etc., (see Political Systems p. 44), but that occur within a larger pattern of inter-related political and cultural systems. It was not obvious that he felt he could read the possibility of potential for any one specic development beyond what he saw. And Leach did not care to speculate. Even as Leach laboured toward the publication of Political Systems, the ethnic heterogeneity in the very area of his concern (pp. 41–45) erupted into a “one people, one state” (amyu langai, mungdan langai ) movement. The colonial administrative districts of Bhamo and Myitkyina became Kachin State after independence from Britain in 1948, and it is clear that happenstance was not the reason. While Leach concerned himself with the structurechanging, transformative phenomena of the gumlau-gumsa relationship, it is clear that he did not consider the potential for political development in the region marked by linguistic differences, but cultural and political integration. This has always been a point of curiosity for Kachin students who share his interest. In 1947, on the eve of independence from Britain, ‘Kachin’ had become a shared identity that allowed them to exclaim together “We are one people, one culture, one nation, and we demand one state”. If Leach felt that during his eldwork of 1939 –45 he saw no evidence that stood out to suggest such a development, then what happened in 1947 must be a transformation that was earlier only an embedded potential. But it did come about, and the question when the process began cannot now be ignored. Can we presume that the literature he traversed did not contain the kernel for this transformation? That appears most unlikely. He considered the “intervariability of culture and (social) structure” (p. 287) within the context of phenomena that had happened. We suspected early on that intervariability might also be an integrating link capable of maturing into an awareness that can support shared political objectives. Was Leach unaware of this potential, or was he just not interested? We shall revisit this question briey in Part III under the section on “Commonalities”. This situation illustrates why Political Systems continues to be a fundamentally important work. The questions we are interested in asking are not excluded a priori by its contents and thrust.

36

maran la raw A note on the post-colonial anthropological agenda

Burma gained independence from Britain in 1948, and in the constitution that created the Union of Burma the colonial administrative districts of Bhamo and Myitkyina became Kachin State. The Government of Kachin State was set up with Myitkyina as the capital. The administrative structure in 1960 was made up of the Secretary of the Kachin State Government, a Deputy Secretary, three Assistant Secretaries and department heads. The Deputy Secretary, Duwa Kumje (Maran) Tawng, was a Kachin who had been in the Frontier Areas Administration under the British, and had the same type of assignment as Leach during the war. Chyauchyi Htang Wa became an Assistant Secretary with anthropological research among his responsibilities. He had been Leach’s only Jinghpaw language assistant. (More about this later.) The Department of Public Information was under Assistant Secretary Duwa Htang Wa. As soon as the Government of Kachin State came into being, inquiry committees to collect information on Kachin customary law, culture and history were set up through this department. The purpose of the customary law inquiry was to collect material from all regions of Kachin State for later codication. The British had already built the foundations for this by adopting the southern gumsa system for uniform application in Kachin areas. The inquiry to gather information about culture and history was begun due to popular mandate. During months of preparation (1945–46) for a national manau (a highly ritualized, ceremonial communal dance) to celebrate victory over the Japanese, Kachin leaders—gumsa chiefs, gumlau leaders, school teachers, government workers and youth groups, representatives from religious and other community groups, etc.—met in Myitkyina and produced a framework of objectives that included political and cultural-educational development. This became known generally as “amyu lam ga shaka”, testament or mandate of the people.3 With the appointment of a government anthropologist, the two branches of inquiry—customary law and cultural history—were consolidated into a program of anthropological research. Salang Waje La Doi, who had been responsible for organizing the inquiries and keeping records, was also appointed to the new department. Finally, Sara Hkyeng Nang, a schoolteacher with expertise about the gumlau area was also hired as assistant research ofcer. 3

Personal communication. My father was one of the recorders.

on the continuing relevance to kachin studies

37

Conjunction of perspectives Political Systems is an important junction for topics that might justiably be investigated at one time or another as separate topics, but might also need to be examined with a common focus on Kachin culture. For instance, for Kachins, the post-colonial era has involved giving the highest priority to the matter of preserving their culture and to searching continuously for ways to maintain connection between the meaning of tradition and contemporary life. Despite the destructive nature of the game of political dominance so characteristic in the post-colonial era, Kachins have remained determined not to cease to exist as a culturally distinct people. In the Kachin view, although political struggles address current reality, the remedy they apply to the situation is premised upon staying connected with traditional culture. What very tradition-oriented societies such as the Kachin are engaged in doing about the post-colonial condition may be quite different from the discourse on the same condition in the West. Similarly, in Political Systems Leach sought to make theoretical statements intended for the community of British social anthropologists; should these be germane to Kachin interests in any direct way? Furthermore, beyond a purely Kachin context, the conditions of colonization and post-colonialism are discussed today largely in postmodern terms. From the standpoint of what Kachins are trying to accomplish politically at present, it is not clear whether the conceptual milieu of post-modernism will detract them from their task or not. Yet, Leach’s position within colonial-era British social anthropology makes it difcult not to want to look at current debates on these conditions by Western scholars. By the same token, some might look with interest upon on the state of Kachin scholarship that has revolved around Political Systems since 1954. Political Systems is contested by Kachins Another aspect in the complexity of the Kachin-Political Systems relationship is the fact that Kachins in 1954 contested some of Leach’s interpretations. The ‘relatively passive subjects’, as Giddens has it (1996, p. 8), talked back, quite uncharacteristically for that time, and they have sustained their contention over the last fty years. Part II deals with three such issues that have been contested. Can this, in a sense, be construed as challenging the hegemony of British power and inuence on them? Did they perforce put themselves in post-traditional society

38

maran la raw

in 1954 (in the sense of Giddens 1996, pp. 8–9)? This angle may yet prove vital when Kachins begin to cast what they have learned into a framework akin to social anthropology. We do not know when this will occur, but I think it is inevitable. The outline and objectives of this chapter Before proceeding further, I would rst like to survey the areas that Kachins have been concerned with in their work since 1954, and where Political Systems has served as a guideline. They have added much indepth information, and it is an intriguing prospect to speculate, given the fuller picture we now have, how much of his original analysis Leach would uphold. This shall be dealt with in Part I of this chapter. The second objective deals with the specic issues about which Kachins have contested Leach’s reading of their culture. This shall be dealt with in Part II. Part III will summarize the implications of work done after 1954, including the perceived relevance of ongoing Kachin work. My goal is to discuss, however briey, the relevance of current work for those outside the Kachin context, and the strategic position of our current work in relation to current and future expectations of Kachin people. Narratives of recollection The relationship between Political Systems and what Kachins have been doing subsequently is both an interaction as well as a case of crossfertilization. It is an interaction to the extent that Political Systems is a source of ideas for things to do, for issues to examine and for areas of information to be expanded. It is cross-fertilization because Kachins learned to look at ethnographic information from the standpoint of analysis and interpretation. Quite importantly, the association between Political Systems and Kachin reaction towards it is also a story that involves real persons, and their recollections can provide much needed clarication of Leach’s points. Thus, wherever appropriate, I will insert part of my recollections in narrative form. (Recollection of the beginning of research following publication of Political Systems) Sometime during the summer of 1960, I was shown to a desk, a chair and empty bookshelves in a room in the Kachin State Government building in Myitkyina. I was informed that I would be the rst government

on the continuing relevance to kachin studies

39

anthropologist. My host was Chyauchyi Htang Wa, Assistant Secretary of Kachin State Government, and from that moment on he was my supervisor. My department would be under his administration. He placed on my desk a well-thumbed and marked copy of Edmund Leach’s book, Political Systems of Highland Burma: A Study of Kachin Social Structure. We both sat down and he explained that before the war he had been the chief assistant to Mr. H. N. C. Stevenson, the British colonial ofcer stationed at Sinlum. Stevenson had assigned him to be the language assistant to British anthropologist E. R. Leach. This was in 1939 and the location was to be the village of Hpalang, just a short distance away from Sinlum. Leach however, used his services for only three months and sent him back to Stevenson. Chyauchyi Htang Wa said some things about the book were exciting, other things were quite bothersome. He said it would be my job to begin sorting things out. He had tried to do so since 1954, and his notes would also be delivered to me. I told him that I had also read the book in 1954, and that I had been mostly bothered by it. This pleased him clearly. He said he had looked for someone to take this position since 1956, but had not found anyone. He spoke with apparent satisfaction and anticipation, and ended with a trace of affection. He might be my new boss trying to full his dream, but his new anthropologist was also his own nephew. He married one of my mother’s parallel cousins, and was aware that my mother started teaching me about Kachin culture and history at the age of seven. Leach’s only Kachin assistant had just hired me to begin the Kachin-Leach journey. Myitkyina 1960 4

What follows from this point on is a summary of the work that began, more or less in a formal sense, at Myitkyina in 1960. I will cover only some of the major themes, my purpose being to show in context the extent to which Political Systems is relevant to Kachin research work by showing the extent of our accomplishments after 1960. The report

4 I had the occasion to ask both Duwa Htang Wa (1960) and Edmund Leach in person concerning this brief tenure of the former as Jinghpaw language assistant. Duwa Htang Wa felt there might have been two reasons for Leach using him for only three months: rst, he was assigned by Stevenson, apparently without prior consultation with Leach; second, Mr. Stevenson and Duwa Htang Wa were both colonial government employees, a situation Leach might have been sensitive about. Edmund Leach told me in 1981 that he wanted to become competent in the use of Jinghpaw in eldwork as soon as possible. I asked him if he had bristled at Stevenson’s decision to assign a language assistant without consulting him. Leach chuckled a bit, paused, then answered, “There were these other problems also.” We left it at that. Anyway, war came and for Leach, both eldwork and language studies were cut short within about four months.

40

maran la raw

makes frequent references to the structure and organization of Political Systems. Part I: Additional Information from Research after Publication of POLITICAL SYSTEMS, 1954 Gumlau and Gumsa issues in 1960 At the outset of my direct involvement in research, two issues were uppermost. The rst issue concerned the list of features comparing the two political systems, gumlau and gumsa (pp. 204–7). Was it a composite that Leach had extracted from published literature? What were the sources? To what extent was this a construct? The second issue concerned the details (pp. 199 –201), which clearly implied an actually violent breakaway, and if Leach were giving abstracted information this would not be included. So, where did this occur and in relation to which parent systems? The gumlau issue has proved to be far more complex than anyone expected. Today, fty years after Political Systems appeared, there is still no overall synthesis of the variations within gumlau society. The sort of structural description Leach employs as a model, the model gumlau system, has one major difference from the other gumlau societies. Whereas the model gumlau society broke away from its parent system due to problems inherent in the parent system, the gumlau societies among the Tsasen and Duleng communities of Kachin State maintain that their differences with gumsa society were ideological, such as who was entitled to invoke which nat-spirit. Kawlu Ma Nawng’s (ibid.) account of this was “not fully comprehensible” to Leach (p. 199). My colleagues and I feel today that the points of difference between gumsa-gumlau must be heard from the Tsasen and Duleng gumlau group. The gumlau accounts given in Political Systems represent largely the gumsa adherents’ viewpoint, including that of Kawlu Ma Nawng (ibid.). The work is still in progress. 1. Gumlau system report Jinghpaw, the lingua franca of the Kachins, has had a standardized orthography since 1895 based on the romanized system of writing developed by Dr. Ola Hanson, the American Baptist missionary. It is still used today and the spelling should have been gumlau, not gumlao. Leach reports (pp. 57–59) that the gumlao (sic) system, “the republican

on the continuing relevance to kachin studies

41

system, so-called”, occurs among the Jinghpaw dialect groups of Duleng (all) and Tsasen (some). He included the Singpho ( Jinghpaw of Assam, called ‘Mungnun’ in Jinghpaw) in Tsasen.5 Problem 1: On page 162 of Political Systems, Leach uses the denition “aristocrats” for magam in the context of the gumsa system, leading one to believe that the term refers to hereditary aristocrats. But magam actually means “leader” and is used both in gumlau, where it is woiawn magam—“leaders who exhort to lead”, and in gumsa systems as du-magam—“chief-leader” or equiv. “leaders who reign as chiefs”. In terms of usage it will be prudent to include the proviso that in gumlau areas the term magam will not likely be used without the qualifying expression woi-awn. This may be in part due to the tendency in gumsa societies to abbreviate du-magam to magam, which is what, most likely, misled Leach. Gumlau system distribution is geographically cohesive, whereas the gumsa system is spread out over a wide area. For many gumsa societies, due to remoteness from any gumlau community, there is the tendency to say magam for du-magam. Problem 2: On pages 204–207 of Political Systems, Leach listed the primary distinguishing characteristics of the gumsa and gumlau systems. He notes there that in the gumlau system, lineage seems to fragment easily, and that the marriage alliance mayu-dama system appeared to be in the process of trying to develop an endogamous circle involving three or four lineages to form an exclusive alliance. Two additional problems occur in relation to this: rst, in the gumlau areas that he mentioned, I was unable to nd any evidence of three or four lineage endogamous marriage alliance circles. Second, I did nd a three lineage mayu-dama circle in the gumchying gumsa system, the hkau wang magam system. In this particular gumsa society, the mayu-dama alliance is spatially located and the alliance relationship rigidly xed, rendering the lineage-descent dimensions less relevant and fragmentary. This is a structural similarity between a gumlau society and a gumsa society. These issues need to be further considered in relation to the particular gumlau society Leach appeared to have been addressing. First, through Kawlu Ma Nawng (ibid.) he described a gumlau society that had rebelled

5 For reference please see pages 204–7 for characteristics, 231–8 for distribution and 198–204 for information on the origin of the system, in Political Systems. I reiterate that this is not a point-by-point comparison with Leach on the treatment of this system. Rather, this section shows to where we have taken the discussion.

42

maran la raw

and broken away from a gumsa society. In the large gumlau societies of Tsasen-Singpho and Duleng, this legend is missing. Second, the above societies, as stated, also have a well-formed concept of leadership, woi-awn magam which the society Leach described did not have. Third, there is the problematic issue of difference between Leach’s model gumlau community and the other gumlau societies. The break away would have to be from a gumsa society, which would be Jinghpaw-speaking. Since Duleng and Tsasen-Singpho dialects are Jinghpaw-speaking but are already gumlau, the location of this rebellious gumlau society will have to be sought outside the ‘non-rebellious’ gumlau societies and alongside a Jinghpaw-speaking gumsa society. This is an issue of enormous importance in resolving the history of Kachin political systems in the upper Hukawng Valley (Kawlu Ma Nawng, ibid.). Where is the ethnographic centre of Leach’s model gumlau society? The remaining issue concerns structural differences between Leach’s model gumlau society and the other gumlau societies on the one hand, and the structural similarities between Leach’s model gumlau system and the gumchying gumsa system on the other. This is with specic reference to the non-robustness of lineage systems, the seeming preference for a circular marriage alliance system, and the high probability that the dialect of Jinghpaw spoken by Leach’s model group is likely the same as that spoken in gumchying gumsa society. If this assumption is borne out, we are talking about Leach’s model gumlau society being a small and marginal group that had probably been assimilated into the surrounding societies by 1954. Regardless of the status of the model gumlau society, Leach was concerned with the instability of systems, and outcomes in the form of structural change. What we have been able to do since 1954 is to add to the information about Kachin gumlau social systems as such. On the other hand, we have not been directly concerned with Leach’s theory itself about gumsa-gumlau systems being in oscillation. 2. Gumsa System Report This report will focus on the structural and functional features among the varieties of the gumsa system. Political Systems gives no information about variation within the gumsa system.6 6 For reference please see pages 204–7 for characteristics, 231–8 for distribution, and 198–204 for information on origin.

on the continuing relevance to kachin studies

43

The gumsa system was the most widespread of Kachin political systems; a closer examination also reveals substantial variation within the gumsa group. It is also clear that the differences are related to adaptation to different situations. In 1965, I chided Leach (in Kunstadter, 1967, pp. 138–40) for not looking into the phenomenon of gumsa system variability. I contend that it is the gumsa political ideology that appears to be pervasive during both the colonial and postcolonial eras; I suggest further that the Kachin political leadership that grew out of the postindependence civil war (after 1962) has functioned within the gumsa concept of salang hpawng (council of elders), and du ni (uniformed ofcerleaders, the default chiefs of today). The noteworthy feature is the replacement of hereditary du-magam (leadership of hereditary chiefs) structure with the hierarchical structure of the uniformed militarypolitical leadership of today. Toward the end of World War II it was clear that a new vision was needed to deal with the onset of the modern world of post-colonial existence. Gumsa society was the system that had the capability to produce it. Three fundamental reasons support this statement: 1. Contact with agents of the West, missionary and colonial, had occurred essentially only in the southern gumsa regions of Bhamo, Myitkyina and northern Hsenwi in the Shan States. The onset of literacy, education and broader awareness began here. 2. In the regions concerned, customary law was applied uniformly and, with it, the level of comprehensiveness that would permit colonial authorities to apply it to regulate the entire Kachin Hills area. The secondary effects of this were, through institutions that became uniform through British adoption, e.g., the Kachin Hills Regulation, colonial administrator’s Handbooks for the Kachin Hills, etc., the strengthening of the traditional unity of the Kachin groups, and putting Jinghpaw language at the centre as the lingua franca, including for education. Earlier, Jinghpaw was a common language by its usefulness for communication across linguistic boundaries, but now it was a requirement for success. 3. Missionary success and colonial policy had the effect of enhancing the notion that the Kachin tribes, so-called, were one politically cohesive group. This would be a critical factor as the empire ended in 1948. The gumsa system had hatched and nurtured the foundation of pan-tribal unity.

44

maran la raw Gumsa system variations

Kachins felt in 1960 that the dearth of information regarding variation within the gumsa system was the most serious shortcoming in Political Systems, especially in relation to the events of 1939 –1947 when Kachins emerged as one politically united group. They lined up with the Allies in the war against the Japanese, who were backed by the majority Burmans, and in 1947 they informed General Aung San at Panglung that they were an ethnic nation-state. Our feeling concerning these developments was that the transitions from tradition to modernity and from colonial to post-colonial society were connected via the hinges provided by gumsa ideology. This accounts for the apparently larger effort given to the study of gumsa society, as the following will show. Gumchying gumsa society was practised at the very headwaters of the Hkrang and Mali rivers in the southern ank of the Himalayas. The unique thing about this society was that all social institutions were stringently structured, as though a static system of rules governing social relations. This happens to a degree unparalleled in other gumsa societies. Three lineage groups, a xed number, form a mayu-dama marriage alliance circle, the hkau wang magam system, which excludes everyone else. Responsibilities to each other in the circle are obligatory. Although the three lineage groups claim to be hkau wang magam or hkau wang gumsa magam chiefs, everyone is locked in and constrained by the rules of the alliance circle. There is nothing for a chief to do over and above the rules of the alliance, with the result that, in practice, authority resides in the hkau wang alliance itself. Notable summary points are as follows. Lineage: descent relations and lineage grouping are redundant by rigid alliance rules. Authority: resides in the rules of the alliance, totally pre-empted from persons. Situated: Alliance partners occupy specic village groups; does not grow big. Closure: alliance is a closed system. Domain: not large, very hard to grow without stressing the alliance. Special problem: small gene pool and in-breeding, related health problems.

on the continuing relevance to kachin studies

45

The characteristics of hkau wang magam This is an alliance system of gumsa society. It involves three participating members where each member is a du-magam or reigning chief family. Each member is also part of a lineage segment, such as Maran or NHkum, etc., and the alliance is used within a xed geographic location. The primary principle of operation in this alliance system is the cycle of exchanges, with each cycle specifying a sequence of mayu-dama pairs in the cycle. The system stipulates that each third cycle repeats the rst cycle of mayu-dama relations. This is expressed by the rule “Each participant marries its granddaughter ( gashu).” The following illustrates the system. 1. A, B, C are the participant lineage segments. 2. Each mayu-dama pair is an ordered relation of two participants, thus (A,B), (B,C) and (C,A); the rst member of the pair is dama and the second the mayu. 3. The mayu-dama relation runs on a system of cycles, as follows: Cycle 1: ((A,B), (B,C), (C,A)) Cycle 2: ((B,A), (C,B), (A,C)) Cycle 3: ((A,B), (B,C), (C,A)) The striking features are that Cycles 1 and 2 are mirror-images of each other, and that Cycles 1 to 3 together form the unit of recursion. Two profoundly interesting features emerge: the symmetry of the mayu-dama relationship, and why the hkau wang system can stipulate “marrying one’s granddaughter ( gashu)”. Symmetry of mayu-dama alliance Consider the relationship between participants A, B: In Cycle 1, A is dama and B is mayu, Cycle 2, B is dama and A is mayu Cycle 3, A is dama and B is mayu, and so on. Therefore, over the three cycles of exchange which constitute the hkau wang system, the relationship between A and B is symmetric. The relationship is asymmetric only if a particular instance of the mayudama relationship is isolated from the system of exchange itself. Political Systems refers to this marriage alliance system only in a footnote on p. 136. This is the kind of evidence that convinced us of shallow analysis. (See Part II Sec. 3, below.)

46

maran la raw

Marrying one’s ‘granddaughter’ as a feature of system regulation In Cycle 1, A is male and B female; descent is through A Cycle 2, B is male and A female, descent is through B Cycle 3, A is male and B female B female in Cycle 3 is the offspring of B male of Cycle 2, who has the same lineage identity as the female B of Cycle 1. Therefore, the Cycle 1 female B is in fact the grandmother of Cycle 3 B female. This result would not be possible if the exchange relationship between A and B were not symmetric through pairs of cycles. Gumrawng gumsa In Political Systems Part II Chapter V, Leach tried to cope with local level differences in Kachin culture, which he called “a welter of variations” (p. 101). Our position is that a lot of these differences are rooted in the fact that in gumsa society there are system-level differences. This report summarizes the three most striking types, of which the previous system is one. Note that gumrawng gumsa—the proud gumsa, is different from gumrawng gumtsa—the proud following; I prefer zaw gumsa for the latter. (Recollection of what my grandfather said about the importance of gumrawng gumsa chief rank) “If you are a duwa you are a duwa even if you reigned for only one day. You are above other people.” The Duwa of Pangmu, 1962.

Gumrawng gumsa system—the “proud, paramount gumsa chief ” system, found especially on the eastern ank of Bhamo district, is quite different from the preceding type. A chief acknowledges no higher authority, and all authority resides in the person of the chief. This system shows the greatest degree of internal stress, as will be discussed later. Characteristics of the gumrawng gumsa system Kinship and grouping: open, martial groupings, conducts raids. Marriage alliance: Mayu-dama relationship is not xed (see below). Authority: resides in the person of the chief, no overlord acknowledged. Subjects: all followers are zaw—“subjects”, the reigning chief is at the top. Domain: up to 10 –12 villages Problems: leans toward hereditary succession; this is punishable by death (see gumshem)

on the continuing relevance to kachin studies

47

Gumrawng gumsa mayu-dama case report Source: Duwa of Pangmu Date: marriages occurring between 1926–54 Symbols: F—female offspring, M—male offspring According to sibling order and Lahpai offspring to spouse 1. F—married Maran clan M 2. M—married Maran clan F 3. F—married Maran clan M 4. F—married Maran clam M 5. M—married Lahtaw clan F 6. M—married Maran clan F 7. M—married Maran clan F 8. M—married Lahtaw clan F 9. F—married Lazum (Marip) clan M Points to note: A. 3 Lahpai M took Maran F, and 3 Lahpai F went to Maran M. B. All Maran spouses are from different families and localities. C. The variables of the gumrawng gumsa mayu-dama marriage alliance are family units, not large lineage segments. Here is the evidence. This example shows that the mayu-dama system does not divide kingroups into opposing kin segments. Marriage rules penetrate far deeper into the fabric of society than Leach apparently thought, and if separate mayu and dama segments or clusters do form as a consequence, they would likely be small. A point of theoretical signicance is that descent structure may be segmentary, but the effect of the marriage relation does not produce another segmentary kin system. What this means for Leach’s theory of structural inconsistency in the mayu-dama system is a question that needs to be addressed at some point. Zaw gumsa This information on zaw gumsa is original with this chapter; Political Systems has no mention of it anywhere. Zaw gumsa, also called gumrawng gumtsa (where gumtsa indicates the multitudes or citizenry), literally means “chieftainship by the people”, and it is the system that has departed the most from the characteristics of the gumchying gumsa system. Two striking characteristics dene this system: rst, authority is a concept and the chief and the entourage must both accept what is the

48

maran la raw

authority that is to be exercised through the chief; second, the chief represents authority in the sense that he is applying a power that resides outside himself and that his people accept that he uses authority with their consent. The Duwa-Zaw relationship has reached the point where authority and its application have become two separate things: authority is now abstract since it has to be thought about, but the consent to use authority has become a practical matter. This situation ties the chief and the people into a power-sharing relationship. This gumsa system developed in North Hsenwi (Shan States) and I believe that it enabled Kachins to envision a role as a co-founding nation for the post-colonial Union of Burma. Characteristics of zaw gumsa or gumrawng gumtsa Kinship & grouping: very strong, very vigorous. Marriage alliance: as in gumrawng gumsa. Authority: abstract, requires mutual consent from chief and subjects. Subjects: equal with the chief. Domain: very large, as many as 60 –65 villages in a domain. Problems: settlement in the valleys meant living next to non-Kachins, i.e., Shans. Zaw-gumsa system and Shan princedoms North Hsenwi State was a large Shan princedom with the Saopha of Hsenwi on the throne. All zaw-gumsa chiefs acknowledged the suzerainty of the Saopha, and there were elaborate symbolic gestures to express mutual accommodation. Kachin chiefs in general provided police service, a function the Saopha did not maintain. Over and above this, Kachin duwa were expected to be available for consultation. The princedom did not levy tax on the Kachin as it did on Shan subjects. The Saopha did not interfere in the exercise of political authority by the duwa in Kachin communities. The Saopha did not have a class of Shan aristocrats under him that would correspond to the Kachin gumsa chiefs. The Kachin political agenda under zaw-gumsa chiefs contained two main objectives: one, symbolic acceptance and accommodation with Shan suzerainty; two, to ourish as a cultural nation (kulturnation in German) as against a nation-state (staatsnation) society. The latter system is clearly impossible without contesting the Saopha’s authority. They did not. Instead, they developed a vigorous political culture to support a cultural nation.

on the continuing relevance to kachin studies

49

Approximately one-half of Kachin delegates who met with General Aung San at Panglung in 1947 to plan for Burma’s independence came from the North Hsenwi State. This is quite phenomenal, and most of them were zaw-gumsa chiefs. A more complete analysis is being prepared at present. In any case, the topic is outside the scope of this chapter. The demographic data specially collected by the author’s relatives in 1963 indicates that, at that time, there were thirty-eight “zaw” gumsa type chiefdoms in North Hsenwi State, with a total of 350 Kachin villages in this political system. This was the largest of the gumsa societies summarized here. A follow-up survey would be very worthwhile, especially because Ne Win abolished Shan Saopha princedoms immediately following the coup d’etat in February 1962. All Shan Saopha were imprisoned, where they died. While Kachin gumsa chiefdoms were not abolished by edict, as happened with the Shan Saopha, large-scale, forced relocations of rural populations rendered chiefdoms as territorial entities meaningless. A study of Kachin political institutions in Shan States is in the planning stage. Gumshem Magam There is no mention in Political Systems of this characteristic, which happens only in the gumrawng gumsa system. (Recollection of my grandfather talking about gumrawng gumsa chiefs) Gumshem Magam Shi wundap e gazut dai Wunhting p mung gumbrut dai Mayan hkinyut wu dai Gumshem baw hkrat di dai. Duwa of Pangmu 1962 Free translation: “The Gumshem Chief. He sits xed in the replace (a symbolic posture of authority). He tears up the family structure. He messes up succession. This gumshem head will have to fall.” I rst learned about this aspect of social regulation among gumsa Kachins through personal experience in 1944. A gumsa chief, the father of my friend in Nawng-ang Village above Namhkam, North Hsenwi State, was assassinated by his younger brother, also a gumsa chief. Apparently, the younger chief became violently ambitious and tried to remove his brother, a more successful chief, out of the way. He was killed immediately by the elders of his village. This became a point of great curiosity. It happened rarely, five reported cases, and only in the gumrawng gumsa chiefdoms whose locations formed a half-rounded rim around the Namwan Tract formed by the Shweli River as it enters Burma to the

50

maran la raw west. In 1946 I began gathering information on this phenomenon. By 1962, I had discussed the problem at length with four elders who knew about the gumshem problem well. The following characteristics have been compiled from the records of these conversations. The Problem: An ambitious gumsa chief, who does everything to secure his own position, and to arrange for his oldest son as the only possible successor, is a gumshem chief. His goal is to create a rigid hereditary succession. The elders usually take care of removing the gumshem chief quickly. It is clear this is the accepted, custom-sanctioned solution. The Circumstances: All attested reports of assassination of gumshem chiefs occurred in domains where both Kachin and Shan composed the chief ’s entourage. The settlements were relatively large, composed of 10 –15 villages. Food production was largely by irrigated rice cultivation, usually assuring surplus.

A number of factors apparently combine to produce severe stress in the strictly traditional framework. Within animistic tradition and practice, only a reigning chief can make offerings to madai nat, the spirit of the sun, who is accorded a shrine in front of the chief ’s house. In fact, the location of the madai shrine must occupy a higher ground, certainly never lower, than the chief ’s house. It seems innocuous enough, but it became a major point of stress in the Kachins’ concept of the legitimacy of gumsa chiefs during the nineteenth century (see report to appear). During the nineteenth century, Kachin settlements reached valleys and lowlands already occupied by Shans practising irrigated rice cultivation. In many instances, due to political instability further south in the Shan-Burman political locus, Kachin settlements close to the Shans were largely invited to settle as military allies in the beginning and, thus, these occupations became re gau ga —“land given to invited allies” (‘A History of Kachin Expansion into Shan Country’. Manuscript, 1997). Valley settlement created certain problems for Kachin religious cosmology, which is adapted to the highlands. First, a re gau chiefdom is not considered ancestral land, hence madai worship became tenuous as well as contentious. Add to this the problem of many chiefs becoming Christians and no longer wishing to maintain madai shrines. Needless to say, chasms and ssures developed within this particular gumsa society. Finally, these gumsa chiefs had to accommodate the arrival of the British colonial power in 1886, earlier than the hinterland cousins. From the standpoint of traditional understandings of Kachin culture, these new situations produce severe and difcult stresses. The gumshem magam problem combines that stress with a political ideology that does not acknowledge a higher authority.

on the continuing relevance to kachin studies

51

Concluding the Gumsa Report There is no mention about the gumshem magam in Political Systems, and yet, it is probably the most powerful and striking development attesting to structural change, and where the reasons can be clearly delineated, amongst Kachin political systems. In addition, the circumstances should be looked at from a comparative viewpoint involving other societies. This gumsa report began with a capsule on gumchying gumsa, then gumrawng gumsa, zaw gumsa, and concluded with a summary on gumshem magam as a related issue. We suggest that this view represents the pattern of development of Kachin political systems. Given the changes in the political ideology that marks the zaw gumsa, we can suggest that the modern era has arrived. What we nd here that is striking is that a chief need not be resident in person, nor present in ritual regalia within a village for his authority to apply, and that his authority applies only by consent of his subjects. This is the conceptual microcosm of an open society, and a strictly modern phenomenon. It is conceivable that the ideology of the zaw gumsa variety of gumrawng gumsa society has reached a stage of development where authority is separate from the dumagam (chief ). When this is coupled with the concept that the ruler rules with consent of the zaw (public, following), the idea of a state is most likely already incipient in the ideology. At the time Kachinland was declared a British colony in 1886, the contact points were gumrawng gumsa chiefs; some gumrawng gumsa chiefs resisted the British through to 1928. During the colonial period, the administration avoided structural changes to the political system, except to make customary laws uniform. During World War II, they led a seemingly united people in support of the Allied war efforts, and in the post-colonial parliamentary democracy, to a man the elected members of parliament were gumrawng gumsa duwa. When the realization came that constitutional reform was necessary, they led the people to support it. They were the dominant political leaders until the takeover by Ne Win in 1962, at which point younger, well-educated political leaders stepped up. But they were still called “duwa”. The civilian gumsa leadership form may have ended, but a uniformed gumsa leadership emerged to replace it. It was only in the year 2002 that the leadership of the Kachin Independence Organization, the political arm of the movement opposed to the junta, began calling itself the “gumrawng gumtsa asuya”. The intended meaning today is gumrawng (free) gumtsa (citizenry) asuya (government, a Burmese loanword).

52

maran la raw

It appears that when in 1960 we felt that Leach’s lack of coverage of the variations within the gumsa system was regrettable, we did not quite understand just how important that feeling was. Hindsight now clearly shows that the most recent form of political leadership, the armed and uniformed variety, can only be meaningfully studied within the context of gumsa ideology itself. In 1976, when the resistance leaders changed from an essentially separatist political agenda to a political reform agenda, the course of political ideology returned to where it was in 1962, when the leaders all had roots in the gumsa ideology of colonial Kachin society. Part II: Contested Aspects of POLITICAL SYSTEMS There are three problems that Kachin readers of Political Systems perceive as issues they are compelled to contest. The rst problem ( pp. 159 –72) is concerned with the use of verbal expressions as evidence in support of the existence of a rigid rank/class structure. We contend that Leach has no case here. The second problem is about incomplete and shallow analysis, especially in relation to kinship. We feel that this is regrettable since information for deeper analysis would have been available. The third issue concerns external models, such as slavery, feudal rank/class systems, etc., which Leach employs without any effort to show why the models t the Kachin social systems, such as by demonstrating specic networks of institutions. We were compelled to ask, for instance, where the location for slavery is in our culture. Our impression frequently is that native traditions have no bearing on the selection and use of external models. We do not propose to belabour any of these issues, as the examples offered are sufcient to make the points. However, I want to iterate the reasons why they deserve to be contested. Suggestions as to what probable causes account for these perceived problems will be covered in the nal part of this chapter. 1. The Problem of Part II Chapter VI, Political Systems In Chapter VI, ‘Concepts of Rank and Class’, Leach began (p. 159) with this statement: “What makes the Kachins particularly interesting from an anthropological standpoint of view is that they have a society which is simultaneously segmentary and class stratied.”

on the continuing relevance to kachin studies

53

The point of this statement is the assertion in kinship theory that these two types of organizational structure do not co-occur. The key to his theoretical construction in Political Systems is his belief that Kachin society has class stratication, and that this is a primary attribute of the gumsa social system. Providing evidence to support this position is the purpose of Part II of Political Systems. The feeling that the case he presented did not hold water was widespread soon after 1954, and a careful analysis of this topic was in our agenda of 1960. Before Chapter VI, Part II, Leach built his case slowly from chapters on concepts of territory, of aggregates of persons, afnal relationship and incest, property and ownership, leading to the chapter on rank and class. This latter chapter is at the heart of his demonstration that Kachin society is rank/class stratied. There are many serious problems in Part II, but the most serious aw occurs, we believe, in Chapter VI, where he attempts to pair unrelated words to create semantic constructs to show that a class structure is implied in the meaning of these expressions. A sufcient command of Jinghpaw might have prevented these mistakes. Examples of this can be given below: On page 162, he cites ma gam as a lower class of aristocrats since there is magam for aristocrats (actually, it means ‘leader’); ma, he is aware, means “child” and so, ma gam must mean a lesser magam or aristocrat. The problem is ma gam in Jinghpaw is the equivalent of “Hey kid” in English. There is no such thing as a “ma gam amyu”, linguistically or in terms of implied social categories in kinship. Furthermore, the collective noun expressions darat daroi (common soldiers, distinguished from commanders) and yu maya (spectators, distinguished from performers) are used to derive the class of “residual commoners” (op. cit.). Combined with du (chief ), he points out there are now three classes: expressions for chief, commoners and, elsewhere, he presents evidence for slaves, mayam. Similarly, he paired two different words ga, soil, the surface that is worked in growing, with hpaga, trade, a completely unrelated word, to derive the association that land has economic value when it is exchanged for alliance commitments. This kind of improbable word association occurs many times. Breaking down the problem In the rst case, the Jinghpaw writing system is incomplete. It does not contain any notation to show tone differences among words, and

54

maran la raw

does not show glottalized stops in the word-nal position. The script is non-phonemic in that it leaves out necessary phonetic features to distinguish between words. It creates false similarities that do not exist in the spoken language. The key would have been to consult a native Jinghpaw speaker, but we recall that Leach only used the language assistant with whom he was provided for three months, condent that he was enough of a linguist to manage by himself. He relied on the only Kachin dictionary in existence at the time ( personal communication). He made similar mistakes in his Kachin kinship terminology analysis (Freedman 1967, pp. 125–152). The second problem is the situation where, in the preceding four chapters, respectively on territorial division, aggregates of persons, afnal relationship and incest, and property ownership, Leach was not successful in nding clear-cut evidence that a rank/class structure, in fact, exists in these institutions of Kachin society. He needed to establish that there was evidence in the chapter on rank and class. Why was he trying to nd the evidence in word associations? In functionalist thinking in British social anthropology one needs to observe particular social actions or relations as the ‘natives’ do them, and learn the meaning of the relations and acts from them. One way in which these relations and acts can be observed is in language; if there is a principled way of talking about a semantic domain, such as a rank/class system, there will be ready expressions for that domain in the language. This was what Leach was after. The last aspect we consider is related to the fact that Leach already had his model gumlau society as the outcome for which he wanted to account. It was the causal factors used to account for the process that produced the structural change that he was trying to nd. In the terminology of information technology, his task was output-driven. Relying in part on Lévi-Strauss (Political Systems, pp. 287–88) he suspected that the asymmetric marriage alliance system of the Kachin might provide the answer. A good hypothesis to start, but perhaps there might be other causes, such as the rigidity and restrictiveness of the hkau wang magam gumsa society (see section on gumchying gumsa). This would be a far simpler solution. This is his case, in a nutshell: kinship theory says that when a society has both a segmental lineage system and an asymmetric marriage alliance system, rank/class structure is concomitant. However, what we have demonstrated thus far, and will continue to do so (see below), is that, considered at a more technical level, Leach has no case.

on the continuing relevance to kachin studies

55

Leach considered himself as having established that there is rank/ class structure based on Part II Chapter VI. He remarked later (p. 287): “[ T ]he most important theme in this book is, it seems to me, my documentation of the relationship between Kachin mayu-dama marriage system and the class structure of Kachin society.” He went to the extent of characterizing Kachin class structure as having “caste-like rigidity” (p. 159). He did not nd any evidence observing living Kachins and the institutions of their culture, but he thought he found the evidence, in their language, working alone with a Kachin dictionary in London ( personal communication). Kachin readers of Political Systems contest this. Leach wrote about one aspect of their culture, and it seems his reading was rather alien to their experience. 2. Supercial Investigation of Kinship In as much as Political Systems depends on kinship to construct its theory, the treatment of the topic is supercial. Leach ( pp. 287–88) seems to have accepted Lévi-Strauss’s assertions as regards the characteristics of different kinship systems, including their tendencies and fate. What we want to argue here is not that Leach might have been wrong, but rather that he did not dig deeply enough into the Kachin kinship system. Our work since 1954 has been motivated by the awareness that there is serious inadequacy of kinship information in Political Systems. Furthermore, given the need to help young Kachin people currently to understand kinship, we must ll the gap. So, again, it is the continuing interaction between Political Systems and our present concerns that has produced the information summarized below. In the summary of the gumchying gumsa society (above) we presented an outline of a society whose organization is anchored in an inexible and fully prescriptive social relations system. The society is bound to actual locations and is an exclusive alliance. In the perception of Kachins, this is the oldest and most traditional social system, a model out of which the other systems have developed and from which they have gradually moved away. In the comparative study of Kachin social systems, the gumchying gumsa occupies the point from which the others emerge. This is our tradition. The gumrawng gumsa system shows marriage alliance, the mayu-dama relationship, to be not rigidly structured, but apparently functionoriented as against being structure-oriented, as it is in the hkau wang

56

maran la raw

system. While in the hkau wang system the mayu-dama relationship is a structural formula, in the gumrawng gumsa system it is a functional one. We will elaborate on these points in this section. Structural principles of Kachin kinship The structure-organizing principles shared by all Kachin social systems are: 1. patrilineal descent grouping; 2. exogamy; 3. mayu-dama marriage alliance system. That is, clan identity is traced through the male line, one does not marry into one’s own clan, and that exchange of women for marriage is one-way directed for each instance of exchange. Functional principles of kinship Functional principles provide the means of establishing kin-group based relational status with any other Kachin. This is the system of kin classication. Since 1960, one of the more striking advances we have made has occurred in investigations of kinship. The ramications of this work are reected especially well in its relevance to contemporary social relations. Kinship is deeply rooted in tradition but it also provides the kernel of group identity and underlies the expression of nationalistic character today. The evidence from Kachin kinship is that the part of its system that adapts and adjusts is the specic capability to redene classiability. See Part III (below), on identity and ethnicity. By kinship rules, no Kachin is a stranger to another. The fundamental bedrock of functional principles is classication. The most specic rules of classication are those based on genealogy and work in combination with the mayu-dama relationship. Three kin classes are thus generated: descent group (neither mayu nor dama), mayu (not common clan, not dama) and dama (not common clan, not mayu). This works as follows. Descent group—common descent identity; exogamy refers to this group. Mayu group—mother’s clan/family group Dama group—father’s sister’s husband’s clan/family group. Since the mayu-dama relationship is not reciprocal or two-way directed between any two actual family units, we have the basis for dening non-cognate kin (not of the same descent group) into a mayu sector and a dama sector. The rst produces wives and the second takes one’s sisters as wives.

on the continuing relevance to kachin studies

57

The next fundamental principle, the most essential principle today, is reclassication of kin categories and the conditions under which this is allowable. This has the effect of redening the mayu-dama relations into two sub-domains: one, those kin prohibited for marriage, and two, those kin preferred for marriage. Next, within each sub-domain some may be reclassied to change status, others may not. We may dene these as follows. Genealogically specic kin Mayu domain Marriage prohibition: mother, mother’s sisters, mother’s genealogical half-sisters Marriage preferred: mother’s brother’s daughter Reclassication: the prohibited kin categories are genealogically specic (hard-wired); any female currently in the prohibited category but who is not actually genealogically related (not in the matrix of hardwired groups) may be reclassied. This is the basis of the distinction between classicatory kin against genealogical kin. Dama domain Marriage prohibition: father’s sisters, father’s half-sisters, father’s sister’s daughter Marriage preferred: none with father’s sister’s female offspring. Reclassication: non-genealogical or purely classicatory kin. Reclassied mayu Any kin not prohibited by the exogamy rule and the above rules may be reclassied for the purpose of establishing a new mayu-dama relationship. Reclassication and the boundary problem We have discussed how kinship functions within the society as the foundation structure of social relations; the remaining question is, what happens at the edges of the ethnic community where Kachins deal with other people? Is there softness and indeterminacy? I have been interested in situations where Kachin and non-Kachin couples marry, in order to ask what happens to their kinship at the boundary. In the cases of mixed marriages that led to formal adoption of the alien spouse, Kachin elders have apparently decided that the rules and rituals of new baby christening apply. And this has started to become

58

maran la raw

standard practice. So, an existing cultural mechanism has acquired the boundary function as well. Foundation of gumsa success I have presented data to show how the gumrawng gumsa social system is a re-analyzed and modied version of the hkau wang magam system of mayu-dama relationship. The essence of each system is in the structure for the former, and in the domain for re-classiability (function) for the latter. Now we can review the illustrative case given in the gumrawng gumsa summary of marriages in the family of the Duwa of Pangmu. (See Part I Section 2 Gumsa Report). Leach missed these aspects of kinship entirely. Hence, the feeling was that his account of the system was extremely shallow, and to a Kachin with a formal background in anthropology, unacceptable. Continuing relevance of Kachin culture Custom and tradition set down the above rules prohibiting marriage to certain kin. Frequently, young people today ask, “If I can marry my mother’s brother’s daughter, why can’t I marry my father’s sister’s daughter?” The answer has two parts: 1. The kinship term for one’s wife says specically that she is, by classication, mother’s brother’s daughter. 2. Kinship term for father’s sister’s daughter says exactly that she is father’s sister’s daughter. Now would call your father’s sister’s daughter your mother’s brother’s daughter? ( Recollection of conversation with a cultural studies teacher in Kachin State) Our kinship terms are grounded in our culture, and the meaning is not random and manipulatable. If you go and tell your father’s sister’s daughter that you will be addressing her as your mother’s brother’s daughter, you are an impossible liar. Young people get it every time. Cultural studies teacher, 2004

3. Slavery in Kachin Class Structure The third issue Kachins feel they must contest is Leach’s handling of slavery. (Reference: Appendix III, pp. 299 –303.) Leach characterized “slave” as being at the bottom-rung of the ladder in the distribution of rights. Our research did not produce such a category of person.

on the continuing relevance to kachin studies

59

The situation, from his standpoint, can be summed up in the form of a two-part proposition followed by a deduction. This is a simulation, we think, of his reasoning: Proposition one: the gumlau system is the result of changes due to instability in Kachin social structure; the instability appears to stem from property relationships attendant upon the mayu-dama system. This is the major premise. Proposition two: kinship theory says that segmentary lineage systems coupled with asymmetric marriage alliance systems tend to be associated with rigid, caste-like rank/class structures. This is the minor premise. Deduction: The Kachin kinship system combines segmentary lineage (patrilineage) with asymmetric marriage alliance (mayu-dama), and they have a word (mayam) for slaves; therefore, it follows that they have a rigid caste-like rank/class structure. This is the conclusion. Observations It is true generally that wherever there is slavery as a social institution there is a caste-like rank/class structure. The word mayam opened the gates for Leach to reach the conclusion that the institution of slavery not only existed in gumsa society, but that it was part of the societal mechanism of class structure that caused instability in that society. Kachin readers of Political Systems contest this interpretation and give three reasons. First, during research for the Master’s degree thesis in history at Delhi University, Maran Brang Di (1994, unpublished thesis) examined British records concerning skirmishes with Kachins (Singpho) in Assam. Out of this research has come the alternative view on slavery in Kachin society. The essentials of this view are as follows. During the mid-thirteenth century, the Ahom Shan kingdom of Hukawng Valley extended its empire to eastern Assam and existed until the end of the eighteenth century. This empire relied on irrigated rice production using Assamese (Khasi) slaves for labour. When the Shan empire ended, Kachin settlement in Assam increased and, at the same time, raiding for Assamese slaves began. Maran Brang Di (op. cit.), based on exhaustive research on British documents still in India, established that Kachins in the Hukawng Valley did not know wetland irrigation farming and hence realized the value of Assamese slaves. So, slaves were brought over to cultivate rice in production areas in Hukawng Valley that the Kachins now dominate.

60

maran la raw

The second point is that Kachin tradition did not have in its culture a specic institutional slot for slaves. The consequence is that Assamese slaves were put into the “being adopted—mayam” slot. They worked with the “lords”, lived in the same long houses and ate from the same pots. Following language acquisition, they were given the clan name of the “lord”. There was no restriction on intermarriage. The fate of Assamese slaves was a transitional one in Kachin societies of the Hukawng Valley. Incidentally, the word “yam” as a part of given names for adopted children is still widespread over the entire Kachin area. The third point is that we have a situation where slaves and slavery entered the lexicon of Kachin culture but without a specic institution of slavery in the social system. If Leach needed slavery to demonstrate the existence of caste-like class structure in the Kachin social system, it seems that the meaning of slavery alien to Kachin cultural experience had provided what he needed. In the years ahead this issue will continue its role as a link between Political Systems and students of Kachin culture and history. It still commands enormous interest. The topic of slavery is raised again in the next section regarding the hegemonic role of Political Systems. Part III: Implications and Larger Issues In this concluding part of the Leach-Kachin story, I want to highlight four areas one, a retrospection of Political Systems and the Kachin reaction; two, the commonalities that link the Kachin situation to other societies and cultures that were also colonized; three, a note on identity and ethnicity; and four, the hegemonic role of Political Systems. In retrospect In this chapter I have tried to show how Political Systems has provided us with a blueprint as well as a gauge. It is a blueprint in that we have followed the organization of the book in a manner of speaking, and it is a gauge in the sense that our priorities are responses to what we perceive to be the inadequacies and problems of the work. For instance, I have felt all along that Leach’s handling of both variation within gumsa society and kinship were seriously inadequate, and I believe my colleagues and I have added much to enhance the understanding now possible in these areas.

on the continuing relevance to kachin studies

61

I have summarized the reasons why we have contested his analysis in the area of establishing the presence of a rigid and caste-like rank/class structure. As regards slavery and Kachin social institutions, we challenged the too casual manner of his analysis. My conclusion is that our work is far from being complete, but I also maintain that we plan to go on, and that our agenda follows from the priorities of 1960. I am quite aware also that the over-riding purpose for Leach in writing Political Systems was to advance his version of social anthropology. It is clear that through Political Systems he has provided indispensable help as Kachins labour to keep their culture and to continue their history today. We are candid about the fact that these two purposes do not frequently overlap. On the other hand, the relationship between the purposes is symbiotic and catalytic. What we gain from our association with Political Systems depends on how well we understand that relationship. My feeling after fty years is that the benets we have received have been substantial. Today, Kachinland is subject to forces of political dominance coupled with forces of globalization. The combined assault of these forces has caused unprecedented political and economic penetration into Kachin society. Does anyone know how to cope with this situation? There is little that a small ethnic nation can do against the destructive part of global economic forces, especially because there is no unied political control centre. The pressure on Kachins is enormous, and their response has come from where they know how to respond, and this is to strengthen the connection to our traditional culture. We have already shown that Political Systems is a major stakeholder in this particular agenda. Its role as blueprint and gauge shall continue. The commonalities Giddens (1996 p. 116) observes: “All the major British anthropologists in the post-World War II generation built their careers on the basis of studying colonized cultures.” Sir Edmund Leach was among them. In the 1950s and 1960s serious moves occurred in British social anthropology to break away from the orthodoxy of Radcliffe-Brown and Malinowski, and Kuper (1996 pp. 135–158) names Leach and Gluckman as the leading mavericks. In this way, Political Systems is placed crucially in the central pathway of these changes. It has a complex relationship

62

maran la raw

to the Empire and with it Kachin history; these are inseparable consequences of being colonized. This historical involvement reveals commonalities and invites comparison with other colonized cultures. Asad (1973 p. 18) has an interesting question: if anthropologists of the colonial era claim political neutrality, does that not render them unable to envisage and argue for a different political order for the subordinated people? We suggest that this question reminds us that, whatever our feelings toward Political Systems as a work about Kachins, Leach was also a colonial era anthropologist. Forster (in Asad 1973 p. 31) argues that colonial era anthropology suffers theoretically because it studies safe subjects but excludes the colonial administration. Asad (1973) is devoted to the specic consideration of the relationship between “anthropology and the colonial encounter”. So this particular viewpoint looks at Political Systems as the work of a colonial era anthropologist. But from this point the problem of colonial era anthropology itself bifurcates. On the one hand, there is current interest in writing cultures (Clifford & Marcus 1986), post-colonial theory and questions (Williams & Chrisman 1994; Chambers & Curti 1996), and, on the other, there is the problem of identity and indigenous rights in post-colonial states ( Dean & Levi 2003). Historicization, re-invention and re-connecting There are also situations where small national communities confront, whether they were colonized at one time or not, the uncertainties of current reality. In order to remain distinctive, special efforts are made to adhere to traditional culture by ethnic communities such as the Kachin, who are often surrounded by larger, different communities. Fisher (2001) presents the situation among the Thakali of Nepal, where itinerants and expatriates seem constantly to re-invent the sense in which they are staying attached to traditional culture. Confronted with a situation where they become separated in both time and space from tradition, they redene their connection. The boundary between a conceptualized centre of tradition and its distant members is uid. In Oceania, Thomas (1997) reports that past events involving contacts with outside agents of change are historicized in a manner that contains relevance and advantage, and these representations re-enter contemporary life as history or precedence. These situations resemble the Kachin situation today. And we are quite interested as well as hopeful that we can begin to interact with

on the continuing relevance to kachin studies

63

those studying the phenomenon of re-connecting with tradition. This could be mutually valuable. This particular section has dwelt upon the theme of how Kachins are studying their culture and history with help from Political Systems, a colonial era work. And their goal is to obtain acute understanding of their culture and history for the benet of future generations. There is no sense, at this point, that they are trying to produce a version of social anthropology or critical theory. And yet, some of what they have come up with will be striking even for those who study colonized cultures in other settings and with other objectives in mind, such as social anthropology and critical theory. The next example will illustrate this. Identity and ethnicity among the Kachin One of the enduring enigmas of the post-colonial world is the problem of characterizing the relationship between identity and ethnicity. An extremely good summary is given by Keyes (in Bareld 1997 pp. 152–54). The positions cited there are made by anthropologists based on studies of specic communities. Among the most obvious dimensions of identity and ethnicity is their connection to international ethnic conicts; Eller (1999) attempts to provide an anthropological perspective on this connection. The important contribution of Benedict Anderson (1991) is the concept of communities expressing nationalistic sentiment as imagined communities. Although both tribes and ethnic groups are imagined communities, the difference between them may be that while an ethnic group imagines itself, a tribe has been imagined by others. Wyatt MacGaffey (in Mudimbe 1997, pp. 56–7)

We suggest that three questions are involved here. On what basis do outsiders recognize a group as a community? On what basis do members of a community recognize each other as members? If a dening algorithm D for a specic community K were written, and then shown to the members of K, will they be able to look at D and then imagine themselves as being represented by D and say “Hey, that’s us”? Our current work on Kachin identity is concerned essentially with question three. At the core of our thinking is that identity involves real persons and a representation of self-concept. When Kachins imagine themselves as an ethnic community, what are the assertions or truisms that are at the core?

64

maran la raw

My analysis begins with these assertions or truisms as the backdrop: one, the more socially interactive the members of a community are, the more integrated they will be; two, the more integrated they are, the more effective the community will be; three, the more effective a community is, the more likely it will be strong and will endure longer. The key is to have a basis for the rst assertion, as the others are built upon it. Assume that Kachins are highly interactive: what is at the core of this? Kinship is clearly the answer and it does two things Any two Kachins can establish kinship status readily, and the kinship status provides a preset schema for interaction. Thus, two people will turn out to be A, of the same descent group, or B, mayu, or C, dama. There is no other possibility. Once this is determined, the two “strangers” are part of a familiar kin network. A model of Kachin identity contains a variable and a simple algorithm: I. Lineage clan identity, call it L; there are seven main clans—the variable. II. A simple algorithm [A] that performs these functions: The model of information processing may be characterized as follows: Is the other person: 1. in the same clan group? If yes, then kahpu-kanau ‘brother’, if not —next rule 2. in the mother’s clan group? If yes, mayu (bride-giver), and if not—next rule 3. in the father’s sister’s husband’s clan group? If yes, dama (bride-taker). 4. If the answer is “No” to rule 3, the person is not a Kachin. There is no other possibility within the Kachin community. In a formal sense then, each Kachin has an identity code that can be represented as: 5. (L, [A]) where neither variable is empty (i.e.: neither symbol is equal to 0) Furthermore, Kachin culture provides a suitable schema of behaviour towards each type of kin. To give a trivial example, with “brother” you divide loads, with mayu you carry his load, and with dama you give him your load to carry. Given (L, [A]) recruitment of a random Kachin into a kin group is automatic, and once the analysis is completed, then attached social relations are inherited. That is, once kin classication is determined, the recruit is already tted into an existing pattern of social relation.

on the continuing relevance to kachin studies

65

The tension that comes with uncertainty is removed. It is difcult to imagine that this does not contribute to cohesion. Please note that in Part II, Section 3, a specic mention of the boundary problem is made. This has bearing here, since the rituals of adoption are based on kinship. On the hegemony of Political Systems Giddens gives an extremely insightful analysis of the relationship between modernity and tradition. He characterizes modernity as standing in opposition to tradition, and that, in this sense, modernity is post-traditional. Modernity has generally rebuilt tradition as it dissolved it. But there is a critically important concomitant: the continuing inuence of tradition within modernity that remains obscure so long as “modern” meant “Western”. This continuing inuence of tradition is “the unexamined basis of Western hegemony over other cultures.” (Giddens, 1996, pp. 8–9). I suggest that this continuing inuence of tradition within the modernity of Leach’s social anthropology is the culture of peer expectations. In writing Political Systems, Leach sought to dissolve and replace the orthodoxy of Malinowski and Radcliffe-Brown (tradition) with his theory (modernity). However, there is an element of tradition that was not detached, and that affected his work, and this was the tradition of writing ethnography in conformity with the expectations of his peers, the community of British social anthropologists. Why did Leach not consult Kachins on his language data? He did not do so because it was not part of the traditional practice of British social anthropology. Why was his analysis of kinship and marriage alliance systems shallow? Traditionally you must identify what kinship types and alliance types these “other people” have. Beyond that, one could choose to go deep, or not at all. Apparently, something of the “buttery collecting” tradition remained with Leach. He did not go deep. Why employ external models (feudal structure, slavery, rank/ class stratication, caste-like rigidity, etc.) without demonstrating the existence of specic institutions in Kachin culture that accommodated and supported them? Surely one trained in functionalist theory would insist on this?

66

maran la raw Are Leach and Political Systems hegemonic?

Yes he is, and it is, to the extent that “the subjects” are compelled to talk back and to challenge the unexamined basis of the tradition of the peer culture with which Leach was associated when he wrote Political Systems. This is the basis of Western hegemony on us. I do not believe that Leach did things that I have contested because of questionable motives. I believe that there was peer-based tradition, which inuenced how he did his work. Finally, if I am correct in this assessment, then I have shown that problems caused by this type of hegemony may be remedied with effort. But we have only started to show proof of this. Works in progress Research and writing are in progress in the following areas: the overview of gumsa political ideology from the pre-colonial era to the present; a deeper analysis of the kinship system; the gumlau history of TsasenSingpho and Duleng; a participants’ view of the history of Shan-Kachin mutual accommodation in Hukawng valley; and current educational and research plans involving youth of Kachin State. We will be glad to share with others information about our progress in these areas as our research develops.

TRANSLATING GUMLAU: HISTORY, THE ‘KACHIN’ AND EDMUND LEACH Mandy Sadan1 Many colonial ofcials, military ofcers and missionaries produced texts during the nineteenth and early twentieth centuries in which the ethnography of the Kachin peoples of Burma was discussed and dened. Some of these authors were adventurers of greater or lesser standing and abilities, but others wrote from the vantage point of extensive, long-term involvement in this region. Most well known of the military as distinct from missionary writers was Major C. M. Enriquez (see also Robinne, this volume). Major Enriquez in many ways exemplied that class of concerned, well-informed, amateur academic who largely dominated the textual sphere of discourse on the category Kachin during the period of British colonial rule in Burma. Enriquez’ most well known book was Races of Burma (1920), in which he wrote about the physicality and ethos of Kachin men recruited into the Indian Army, but he also wrote memoirs, such as Burmese Arcady (1923), and continued to write in the local press in Rangoon into the 1950s under the penname of ‘Theophilus’.2 His books typify a genre of colonial writing in which ideas of ethnic type were elucidated for a wider audience, and which accompanied the political formulation of ethnic categories and their representation in the public sphere of empire. In the 1920s, many such works had pretensions towards academic objectivity whilst still privileging the insights born of long-term local residence and empirical experience acquired in government service. Tensions between these and

1 I am indebted to Pungga Ja Li for sharing with me so much of his valuable research into Kachin ritual performance during 1996–99 in Burma, which underpins the ndings of this chapter. All errors in interpretation are entirely my own. This work was also only possible with the help and support of Daing Ze, Hkanhpa Tu Sadan, Aung Seng, and members of the Yup Uplift Committee. My thanks also to Elizabeth Dell at Brighton Museum, UK, for her support of the research that was undertaken during this time. I am also extremely grateful to both U Chit Hlaing (Professor F. K. Lehman) and Professor Robert Anderson for their careful reading of early versions of this chapter and their detailed suggestions as to how it could be improved. 2 See also A Burmese wonderland: a tale of travel in Lower and Upper Burma, Calcutta: Thacker, Spink & Co., 1922.

68

mandy sadan

other more formally academic modes of knowledge and representation became more clearly drawn by the 1930s, when Leach rst went to Burma, as distinctions were made between popular works of ethnographic description and the increasingly professionalized demands of pure anthropological discourse. When Edmund Leach published Political Systems of Highland Burma in 1954, the rupture in the case of ‘Kachin’ ethnography became irrefutable. This work signicantly transformed the textual domain of the category ‘Kachin’ and set it on a new trajectory as a globalised ethnographic term. One may quite appropriately set the term ‘Kachin’ in this context inside the inverted commas that are colloquially termed ‘scare quotes’. Writers of the Enriquez era were used to employing the term Kachin as a shorthand for a collective of peoples whose Kachin identity had been formalised progressively during the period of direct and indirect colonial rule from 1886 onwards, through the activities of the military (e.g. in the creation of the Kachin Ries, a composite of Jinghpaw, Nung, Lisu, Lachik and Zaiwa recruits), the colonial civil administration (e.g. the Kachin Hills Regulations, which created administrative uniformity out of a region widely known to ofcers for its population complexity) and the Christian missions (especially through attempts to standardise literate forms in the face of a multiplicity of local dialects). Kachin in these spheres was known to be a complex and diverse category of peoples, languages and social forms, even if downplaying the signicance of such diversity increasingly became a pre-occupation of colonial ofcials and burgeoning Kachin nationalist political leaders alike. When Leach catapulted the term into an academic Other-world, however, ‘Kachin’ became an anthropological trope in the sense that it became the centre of a discourse about its meanings whilst being dissociated from this kind of epistemological complexity. This chapter attempts to evoke this distinction by variously enclosing the term in scare quotes or not. When used, the implication is that it is the ‘dissociated’ form of the term that is relevant to the point being made; when not used, the implication is that the underlying historical and societal formulation of this ethnic category is somehow relevant and being referenced. In this latter case, the most common usage of the term Kachin evokes a contemporary grouping comprised of six main sub-groups: Jinghpaw, Zaiwa (Atsi), Lawngwaw (Maru), Lachik (Lashi), Lisu and Nung-Rawang (increasingly identied separately). This representation is also a gross over-simplication of the term Kachin, but will sufce for present purposes, a fuller discussion being outside the scope

translating GUMLAU

69

of this chapter (see Robinne, this volume, for explication of some key aspects of this multi-group complexity; Sadan 2005 & 2007). Yet, the transformation of the term ‘Kachin’ from populist to formal academic discourse created a new kind of problematic for educated Kachin elites in their engagement with discourses on ‘Kachin’ by Other, as has been so clearly elucidated by Maran La Raw (this volume). Dr. La Raw has discussed the anthropological conundrum that the work created, but at another level this problematic centred upon contested uses and abuses of historical ‘facts’: disputed representations of how identities can be transformed over historical time and the role, if any, that history might play in that process. However, again as Dr. La Raw has demonstrated with regard to kinship terminology, it also related to a problematic of translation and the transposition of conceptual categories from ‘the eld’ to academe. There has been, of course, a relatively well-known academic debate on the anthropological meanings and implications of Leach’s text in relation to ‘Kachin’. This chapter, however, will primarily consider the impact of the historical dimensions of Leach’s work, and the difculties that his theoretical construct of socio-political oscillation has engendered for analysis of historical Kachin ethnicity. Edmund Leach, Burma and British Social Anthropology Edmund Leach went to Burma in 1939 to do eldwork for a doctorate in anthropology at the London School of Economics. He found himself based in the Sinlum region, close to the border with China, an area long of strategic importance in relation to the historically signicant trade routes that traversed the Kachin region from China to India. Shortly after his arrival, World War Two broke out and he found himself embroiled in the war in Burma. For the rest of the war, he travelled extensively in the Kachin region on a variety of ofcial and military duties, many of which it would seem he undertook against his own best instincts, such as recruitment of Kachin soldiers into the Indian Army (see Anderson, this volume). During this chaotic time, Leach famously lost his eldwork notes and much of the data that he had collected during the initial nine months of his most concentrated eldwork. This loss has bedevilled analysis of Leach’s most famous text. It is tempting to hypothesise that the loss inuenced the transformation of his PhD thesis of 1947 into the publication referred to here as Political Systems in

70

mandy sadan

1954. Supercially, it might seem to be the reason why Leach’s initial research ndings were by 1954 transformed beyond recognition to being primarily a theoretical engagement with Lévi-Strauss’ work on structuralism and an onslaught against the then-state of British social anthropology. Increasing abstraction may have been brought about by a loss that, surely for lesser mortals, would have created a signicant black-hole in eldwork data. Much has been made of this in critiques of Leach, although U Chit Hlaing comments persuasively that the signicance is perhaps overblown given that Leach had prolic powers of memory, bordering on the photographic.3 Certainly, the detailed ethnographic comments drawn from memory were moulded quite ruthlessly into an anthropological theory of socio-political oscillation. But, by 1954 Leach seems to have been a passionate advocate for substantive change in British social anthropology, and this requires no qualication relating to the loss of eld notes: it was this concern, it seems, that dened the agenda of this particular work. Nonetheless, it has long been contended that one of the most signicant aspects of Leach’s Political Systems was that he made such efforts to integrate anthropological theory into an historical framework. This was an important methodological shift at the time, and even led eminent historians such as Professor Keith Thomas, who later included analysis of the dichotomy of highland-lowland in his major work Religion and the Decline of Magic (1985) to praise Leach’s sensitivity towards the historical (1963). Yet the shift in the historical framework Leach presented in 1954 is perhaps more difcult for historians to accept in cases where one looks to the text for real historical insights into the nature of Kachin ethnicity. Such a desire is not necessarily motivated by a preference for the empiricism of contingency as against an analysis of social structure and process (see Sprenger, this volume), it rather denotes a concern that the integration of the historical in the text, in fact, does not ‘work’. Leach (1997 [1954]: 227) described Chapter VIII of Political Systems, ‘The Evidence from Kachin History’, as being in some ways ‘at best a complete waste of time’ and even as a chapter that could be omitted by the reader. According to Leach’s own bibliography in Political Systems, the historical framework evolved entirely from the secondary sources of ofcial government publications, monographs and published journals covering the period from 1828

3

Personal communication.

translating GUMLAU

71

onwards. This is at odds historically with the doctoral dissertation that Leach completed in 1947, which contains a great deal of detailed local information and signicant counter-narrative to the published archive upon which Leach later appeared to rely.4 That this reects a shift over time into a theoretical approach in which Leach himself declared his general disinterest with ‘the facts’ raises serious questions as to what role, if any, history truly has in Leach’s work, despite the claim that it was ground-breakingly cross-disciplinary.5 History in this new historicoanthropological model became a-history, merely an extended temporal framework against which the tenets of anthropological structuralism could be tested. Historical ‘facts’ determined no part of anthropology’s outcome, and no such causal affects, the stuff of historical contingency, external inuences and the like, should be sought from them. There are, thus, great difculties in Political Systems in distinguishing those aspects of the historical framework that are deemed to be based on empirical observation and those that have been moulded into a theoretical discourse to address particular interests of British and francophone anthropologists in the early-1950s. Discussing the ‘what is said’ in Leach is, therefore, evidently no easy or straightforward matter on either the empirical or theoritical level. Even after fty years there remain a large number of paradoxes, ruptures, discontinuities and disunities in Leach’s text. There is clearly an epistemological rupture in Leach’s analysis between the empirical data and his idealist interpretation, which Leach acknowledged. It will no doubt always prove impossible to unify completely the discontinuities in Leach’s work, both historically and anthropologically. Of course, it is not the case that these approaches necessarily and innately preclude each other, but clearly Leach’s text is neither a harmonious nor particularly successful aggregation of two domains of knowledge. Despite this, and maybe even, paradoxically, because of it,6 Political Systems remains a seminal work that continues to inuence studies of ethnicity in areas far beyond the Kachin region; Leach’s inuence on looking at the relational in the formulation of ethnic identities remains an outstanding contribution to this subject. For historians and anthropologists alike, therefore, it is worthwhile penetrating further 4 My thanks to Professor Robert Anderson and Professor U Chit Hlaing for their comments relating to Leach’s doctoral thesis and the Leach archive. 5 My thanks to Professor U Chit Hlaing for his comments on this issue. 6 Professor Robert Anderson, personal communication.

72

mandy sadan

what might underpin the historical conceptual as well as difculties that permeate Leach’s work and the discontinuities that exist between his representations of historical and anthropological data, including from Kachin points of view. Leach, Socio-Political Oscillation and History We learn from the recently published works of Stanley Tambiah (2002), Stephen Hugh-Jones and James Laidlaw (2000) that Leach revisited his Kachin study repeatedly over his lifetime, and in many different contexts, ranging from academic debates to radio presentations and lectures. It would be a misrepresentation of Leach to fail to mention that he himself provided a forthright self-assessment of the limitations of this text in his 1980 lecture at the University of Washington, ‘Political Systems of Highland Burma: A Retrospective Reassessment’. Tambiah (2002: 429) states: Leach openly admitted to his audience that his book . . . had postulated categories such as Burmese, shan, Kachin, gumlao and gumsa, while leaving out of consideration the play of historical forces on their usage

Stanley Tambiah has stated (ibid.: 122) that it is a disservice to Leach to isolate his model of Kachin in Political Systems from his article ‘The Frontiers of Burma’ (1960). As Tambiah states: [ M ]ost specialists of Southeast Asia have not acknowledged or have ignored this piece, although it does in an original, prescient and provocative manner discuss the spatial ordering and political dynamics of pre-colonial Shan polities and their relations with ‘hill peoples’. . . .These themes have been revived and even replicated in some modern texts dealing with the history of map- and boundary-making, census classication, and the reication of tribes and minorities as separate entities

‘The Frontiers of Burma’ provides a broader framework for Leach’s statements on ‘Kachin’ at a regional level. If one were to make a study of Leach’s corpus of work, it would, indeed, be a mistake to fail to consider this article, and Leach’s later critical statements of his own historical position. However, the focus here will remain the enunciative formulation of ‘Kachin’ made primarily in Political Systems as a seminal text. It is this text that, in more detail than any other work, seeks to dene the nature of Kachin society through its own verbal codes. As stated, it was partly Leach’s attempt to integrate the historical into an anthropological theory that was deemed signicant by many. The

translating GUMLAU

73

historical framework Leach created underpinned his representation of a relational model of ethnicity. He believed that ethnic categories such as ‘Kachin’ were not discrete entities and that only when distanced from the presentist analysis of anthropological eldwork could their transformative aspects become clear. A précis of the historical position outlined in Political Systems might run as follows. Leach makes a number of points relating to his use and interpretation of ‘History’ throughout the text. He sees social anthropological data as being inherently and primarily sets of historical incidents upon which order is imposed through the ctional (idealist) construction of systems (1997 [1954]: xi). He feels that the ambition of his work in Political Systems differs in temporal scale, and that when the timeframe of social anthropology is extended, the ctive nature of these constructed equilibrium systems becomes apparent. To Leach, history seems to become a vast space in which relationships and systems are conceived by social anthropologists as if they unied these historical incidents into unied, a-temporal entities, rather than systems that could change and ‘oscillate’ between different extremes. In his notion of historical social oscillation, Leach seeks to avoid the notion of historical developmental cycles, as these would also be implicitly coherent and integrated over time (ibid.: xii). Similarly, historical determinism should be rejected ‘in any shape or form’ (ibid.: xiii). The order imposed upon historical incidents in as if structures are no more than their verbal ordering, the verbal construction of category being the social domain through which order appears to exist. History could have no explanatory force upon these systems, and the occurrence of, for example, gumlau revolts had no historical agency other than the system itself.7 The longue durée in this context was merely a temporal framework in which directionless oscillation could be modelled more extensively. In the Kachin case, Leach found these verbal categories to be open to re-interpretation in different contexts according to a host of factors, such as ecology, economy and agency, such that the categories in themselves could appear chaotic and discontinuous over time and space. However, when located spatially, the juxtaposition of these categories with neighbouring groups (dened by the use of different verbal categories) were seen, according to Leach, not to operate in a historically

7 My thanks to Professor U Chit Hlaing for his comments. Use of the spelling gumlau rather than gumlao reects a preference for using the Jinghpaw convention for this term.

74

mandy sadan

predictable or determined way, but within a framework that dened the parameters of oscillation, which in extreme cases could lead to the integration of a Kachin person/community into a non-Kachin group and a complete shift into a domain where other verbal categories of social order predominated. This, he felt, was a repetitive, albeit nonpredictable process, that extended historically beyond the temporal constraints of the reifying anthropological present. An historian might suggest that Leach developed this model from his own chaotic experience of eldwork during wartime in 1939 –43: his experience suggested to him that ethnicity was a chaotic construct. Certainly, at this time, large numbers of Kachin groups expressed discontinuities in their conceptual understandings of political and social systems; communities appeared and/or were conictual with each other, and the models of neighbouring political centres of power were themselves frequently the agents of change from ‘within’. Yet, such an analysis also has to take into account the very determined revisionist onslaught against British structuralist-functionalist anthropology with which Leach was engaged by 1954. The biggest criticism, therefore, must be the apparent ease with which Leach seemed willing to incorporate or discard historical evidence purely to substantiate his theoretical model, which itself was primarily addressed not to ‘the facts’ but was intended to challenge the interests and conventions of British social anthropologists. Conceptual and Verbal Categories GUMSA and GUMLAU Political Systems is an extremely problematic text among literate, urban, educated Kachin people and has retained this status apparently since the moment of its rst publication (Maran, this volume). Edmund Leach seems to have had a difcult personal relationship with some of his Kachin co-workers, preferring it seems, to go it alone in his research and to dispense early on with translators. There were then further distancing leaps from local discourses ‘on the ground’ to thesis to publication. The Kachin academic Dr. Maran La Raw, who had responsibility for anthropological research in the newly independent Kachin State, has already discussed some of the (to Kachin people) glaring errors of fact and interpretation in the text, and has stated that it was felt immediately that these errors should be addressed, but that it was not

translating GUMLAU

75

clear by what means this could be done. 1950s and 1960s Burma was very distant in time and place from the postmodern ‘linguistic turn’ that could have helped in the formulation of local counter-narratives to a work such as this if, indeed, such an approach was appropriate. Furthermore, the very globalisation of the term ‘Kachin’, the fact that it was now an identity that was familiar to large numbers of people who had otherwise no concern with Burma, made responses to the text even more difcult to formulate: it at once both privileged the Kachin peoples whilst being felt in some signicant ways to misrepresent them. It is clear that this difculty was not caused simply by a distancing in time or geography; local misgivings were/are not a simple consequence of the text being ‘difcult’ or of the details being separated from the time at which it was written. One also gets a sense of discomture about the text locally from the apparent lack of contemporary local social memory attached to the gure of Edmund Leach himself, as an individual in contact with Kachin society who entered peoples’ lives and had a lasting impact upon them. The lack of social memory around Leach is in marked contrast to the personalities of key missionary gures, such as Ola Hanson, who is one of a number of people constantly gured in narratives of culture-contact with Europeans and Americans before, during and after the Second World War; H. N. C. Stevenson likewise in secular Chin society. This subdued local account of Leach today is, thus, not inevitable. It would be possible, should they choose to do so, for local Kachin educational and political elites to promote Leach’s work as a mark of status, as proof that the Kachin people have distinction in the wider world. That this is not done seems to relate rst of all to some very heartfelt opinions concerning the need to ‘get one’s facts right’ and that Leach has failed to do this on a number of signicant points. Although Political Systems has been translated into Burmese, but not into Jinghpaw, the text seems to be a restrained vector of public Kachin discourse. There are some very clear contemporary nationalist concerns that go some way to explaining this attitude to the text, which in no way relates to its accuracy or otherwise as an interpretive model of Kachin society in the 1930s–50s. Dr. La Raw’s chapter, however, shows effectively how, even when nationalist concerns are accorded signicance and have clear inuence on the interpretation of Leach’s work and the development of counter-narratives, there remain large areas of contention, irrespective of these kinds of concerns and which clearly

76

mandy sadan

transcend them. Nationalist ideology renders the relational interpretation of ethnicity problematic for many Kachin people: if ‘Kachin’ people may easily ‘become’ Shan, this is obviously at odds with the reifying discourses of the nationalist community, which was already in armed conict with the central government by 1961, just seven years after the publication of Political Systems.8 My own contact with Leach’s text whilst in Burma was limited, if not curtailed, by this context. Whilst this is a ‘difcult’ text because of its potential implications for Kachin people in their present political difculties, it nonetheless does not take long to uncover the relational aspects of contemporary Kachin ethnicity, with its host of sub-groups and assimilative practices (see Robinne, this volume). To assert that nationalist ideology alone presently denes the main local framework by which Leach’s analysis is challenged, and that local criticisms can, therefore, be treated with less consideration than is their due, is a serious misjudgement. Local reluctance to privilege Leach’s work partly relates to his representation of historical detail and his failure to amplify local constructs of history, despite his use of Kachin (in this case, Jinghpaw) ‘verbal categories’. Where Leach has attempted to make these amplications, his own linguistic limitations have resulted in serious errors of interpretation. In this way, he has failed to reect appropriately the conceptual constructs that he claimed to be elucidating, despite his emphasis on ‘verbal categories’ in local languages. The core terms privileged by Leach as ‘verbal categories’ relevant to the oscillation model, were gumsa and gumlau (or gumlao, as he preferred, following the colonial convention in spelling). These terms are highly problematic historically, as Leach admitted in 1980 but did not realise and/or acknowledge in 1954. Gumsa is presented by Leach as being a Kachin social system under the aegis of hereditary chiefs. Gumlau is, in contrast, a system with an egalitarian, non-hereditary system of lead8 I was aware of one Kachin student who attempted to reverse this model in his MA dissertation in History and thus to demonstrate that Shan people could equally, therefore, once have been Kachin. He was advised by his supervisor to abandon this approach as the thesis would not pass the academic censor, it being more likely to be perceived as a work of Kachin nationalist history. It is not currently possible to publish a history of The Kachin in Burma; local histories have all to be written from the perspective of lineage and family histories. This undoubtedly inuences the availability of such accounts in local languages (see Robinne, this volume), yet it is but one example of how the central censorship mechanism fails to appreciate the nature of cultural output in minority languages, as this emphasis on lineage clearly accords with the maintenance of awareness of kinship afliation as an important vector of nationalist identity.

translating GUMLAU

77

ership. Leach’s contention was that both were essentially ‘ideal types’ or ‘verbal categories’ and that, in fact, there was an innate tendency for Kachin communities to oscillate between these two over time. Historical incidents in and of themselves had no causality in relation to these terms, other than to provide a vague a-historical perception of the ‘nature’ of Shan leadership as a stimuli for oscillation; but as a systemic feature this perception could be abstracted from any real incidents in which outside agency might initiate change. Despite Leach’s representation, there are clearly incidents of historical causality in the appearance of these terms throughout the colonial archive that he privileges bibliographically. The archaeological-genealogical analysis of discursive formations that Foucault (1972 [1969]) advanced could be applied to the transmission of local models of gumlau into the colonial archive. A number of references to this term can be found extending over the same timeframe with which Leach was concerned in his thesis and book, which covered the colonial encounter with Kachin from the 1820s to the end of the Second World War. The local ‘auto-ethnographies’ of meaning that accompanied the transference of these terms from Jinghpaw ‘informants’ into the colonial archive, the various enunciations of them that were made by government or military ofcers to superiors, and their subsequent entanglements in multiple discourses of power relations, can also be explored. These historical eruptions of the term gumlau cover a range of occurrences from the purely shamanistic to those overtly referencing the political experience of colonial expansion in the region. Such historical transformations are evident with the terms today, where Kachin nationalism has in recent years transformed the concept of gumlau into an ideological narrative of revolt against despotism. In recent years, the term gumrawng gumtsa (which Dr La Raw in this volume has termed zaw gumsa) has been used locally to describe the ideal Kachin polity: a proto-democratic, enlightened leadership that does not exclude the possibility of heredity where appropriate. This model is set against what has recently tended locally to be referred to as a despotically-inclined form of leadership, termed gumchying gumsa. The stand-alone usage of gumlau, or the description of it as part of a system that excludes the concept gumrawng gumtsa, is considered inappropriate: gumlau is, in contemporary Kachin discourse, considered merely a historical process of revolution, of overthrow and upheaval, not as a system. However, contemporary Kachin discourse attributes to gumrawng gumtsa many of the characteristics that Leach attributed to gumlau.

78

mandy sadan

The contemporary political contexts that can go some way to explaining these apparent discontinuities are manifold: the repositioning of modern notions of ‘democracy’ in relation to ‘traditional’ concepts of political order; the experience of prolonged communist interventions in the Kachin region; the interpretation of ‘rebellion’ and ‘overthrow’ within the main Kachin armed resistance movement under the leadership of the Kachin Independence Army. From this contemporary perspective it would be an endlessly frustrating task to try to ‘prove’ or ‘disprove’ Leach’s model of gumsa-gumlau. Because of Leach’s privileging of ‘verbal categories’ as a primary tool of analysis, one gets tied into an endless cycle of attempting to unify historical notions to smooth out the discontinuities that are so often encountered between Leach’s construct and the situation ‘on the ground’ over a historical period. The criticism that can be made of Leach’s historical interpretation of gumlau and gumsa, therefore, is that he treats these verbal categories as having stable meanings over time, even though the societies that utilise the terms are in constant modes of transition and change: Leach does not come close to dealing with their historical modality. Furthermore, the historical modality of the terms is not revealed by their representation as singular lexical items; they become disconnected from the conceptual nexus within which the terms are related to each other by the account and interpretation of them by Leach as singular items. The rst point of confusion for many Kachin people in discussing Leach’s work and the conceptual rupture they perceive, is that use of the stand-alone terms gumsa and gumlau is inappropriate in a socio-linguistic way. These terms seem to carry local meaning only when paired as a couplet: for example, gumchying gumsa, gumle gumlau or gumrawng gumtsa. The conceptual modalities of these ‘verbal categories’ may, therefore, be approached best by considering them as part of couplet phrases. It is this aspect of the gumsa-gumlau issue that will now be considered. Couplet Phrases and the Modality of Historical Discourse Kachin ( Jinghpaw) makes much use of couplet phrases, as will be discussed, and this has many implications for the development of verbal labels for conceptual categories. Some might feel that the attention given to the use of couplet terms as part of a local cognitive model of gumsa and gumlau is attempting to emphasise unduly a merely semantic or poetic point. However, if one fails to take account of this couplet form, one fails to appreciate an important level of Kachin ( Jinghpaw)

translating GUMLAU

79

historical discourse by which these terms can begin to be approached according to a more truly ‘indigenous’ system of meaning and countermeaning: the embeddedness of couplet forms, parallelism and dyadic sets in Jinghpaw oral ritual language as recited by spirit priests (dumsa, jaiwa etc.). A fuller discussion of the nature and context of non-Christian religious ritual and ritual language in the Kachin region is beyond the scope of this chapter (see Sadan 2005; Robinne 2007). The chaotic nature of Leach’s wartime experience meant that he would rarely if at all have been able to witness large-scale, community-based ritual performances in practice, such as the manau (Sadan 2000), which he comments in Political Systems was an aspect of research that would have beneted his analysis (1997 [1954]: 119, fn 23). Leach’s early decision to dispense with translators for his anthropological research in those early years meant also, and incontrovertibly, that he could not access this deeper linguistic domain other than through colloquialised renderings of myth as folklore. It would have been impossible for him to penetrate these layers of meaning in ritual language without the signicant intervention of local intermediaries and local intuitions on this subject.9 The key point is that Jinghpaw ritual language (which term is not used here in the sense that Leach, or Gros, this volume, use it; likewise, I do not engage here with anthropological theories of ritual for the sake of the present argument) is structured upon a culturally intuitive semantic and euphonic model of couplets. This form of ritual language structure is found elsewhere in Southeast Asia and beyond, where dyadic sets of sentences, or a system of parallelism in which two juxtaposed utterances should be understood in conjunction with each other for the meaning to be fully understood, are seen to be relatively frequent (Fox 1988). The occurrence of these and other forms of lexical-semantic coupling and doubling, and their cognitive implications, have been the subject of a wide anthropological and linguistic literature. Their prevalence in monosyllabic languages, however, has attracted particular

9

Professor Robert Anderson points out (personal communication) that Leach would have had military translators during the period 1942–45. However, my own experience of working with translators on matters related to ritual performance who have no particular insight into or interest in this form of language (which has never functioned as a popular communicative form but has always, rather, been an esoteric linguistic domain for the specialist) means that (especially, but not exclusively in the present socio-political environment in which the form is further marginalised) it is extremely difcult to nd suitable translators for this task.

80

mandy sadan

attention, for example in relation to the ‘heaviness constraint’, where monosyllabic languages may demonstrate a stylistic preference for longer forms through the doubling of monosyllabic lexical items, which in turn provide both depth and precision to the language.10 Figure 1 gives the rst lines of a studio recording of the myth of origin of gumlau and gumrawng gumtsa, which was recorded in 1992 as part of the cultural documentation and archive programme of the Yup Uplift Committee, Myitkyina, under the aegis of Pungga Ja Li. This was a studio recording undertaken by a volunteer, Hkan Naw, who had formerly been a spirit priest or dumsa, and who had made this recitation previously in the conduct of local spirit practices, but who had, in recent years, converted to Christianity. He was, therefore, somewhat nervous about making the recitation, but agreed out of empathy for the cultural objectives of the archive project. This extract demonstrates a very typical usage of couplets in Jinghpaw ritual language. Figure 1: Introductory section of a dumsa recitation on the origins of gumlau and gumrawng gumtsa Opening section of the recitation Chyup Chya Nga, recorded in a studio in Njang Dung, Myitkyina, Kachin State, by dumsa Hkan Naw on 14th April 1992 001 Nnga e . . . hkawp e ya, 002 hkringwa ga bai jasat yu gaw, 003 dumsa mahku bai shaprat yu gawq, 004 hkawp hpan wa Ningsang e Karai wa e ya e, 005 laijum ga bai jasat nna me, 006 ka-ang matut bai yawn na rai sai, 007 hkringwa ga bai shaprat nna, 008 shing-ra madat bai gawn na, 009 ya ningdung wap na nang ginnip, 010 ninghpang dap na nang ginchyip, 011 nang hpan Ningsang wa, 012 ya hkringwa e yawn ai marawng mung e nang kawn yan, 013 dumsa chyai ai mara mung nang ningtsa Karai hpan Ningsang kawn ginchyip nna hpyan bara, 014 mara mung rawq marawng mung hkrawq ya bara, 015 n nga n mai rai nna nga,

10 My thanks to Professor U Chit Hlaing for his comments and for making me aware of this hypothesis. He also suggests that gumle may be an example of a doublet in this case.

translating GUMLAU

81

Figure 1 (cont.) 016 n dumsa n mai rai nna dumsa gumrawng gumtsa nga ai ginshi rai nna yawn, 017 gumlau paw ai ninghpang gawn na rai sai nga ma ai English free translation: [ Honoric exhortation] Let’s chant the words of the dumsa again, let’s compose the tune of the dumsa again. Creator, the dumsa is going to chant the story of human beings and compose again the instructions from the ancestors. If the top is not completely covered, if the bottom is not complete, I hope that you will forgive my errors. If everything is not complete, free me from my misfortune and remove the maraw. I must chant, I must dumsa. I have been told to chant the brief history of revolution, the origins of gumrawng gumtsa

Figure 2 presents a tabulated summary of the main functional relationships between the couplets in this part of the recitation. Figure 2: Examples of couplet relationships in Jinghpaw ritual language11 Dyadic set

Occurrence

Hkringwa ga Dumsa mahku Jasat Shaprat

002

Laijum ga Ka-ang Matut Ningdung Wap

Semantic relationship

003

Parallel meaning [noun phrases] 002/005 003/007 Euphonic—complementary meanings [verb] Hkringwa 005 007 Parallel meaning [noun ga phrases] Shing-ra 006 008 Couplet [compound noun] Madat 006 008 Euphonic—complementary meanings [verbs] Ninghpang 009 010 Euphonic—antonymic relationship [nouns] Dap 009 010 Euphonic—complementary meanings [verbs]

11 Notes: Hkringwa ga and yawn are italicised because they occur twice but in different sets; Line 1 is an opening exhortation and thus does not form part of a dyadic set; Line 4 does not form part of a dyadic set because it is at this point that the dumsa conates the terms Hpan wa Ningsang, the name in animist ritual for the original creator spirit, and Karai Kasang, the term that has been used to denote the Christian God. It is in this line that the dumsa implicitly establishes his Christian beliefs, although in a manner that does not interfere with the form of the recitation; Line 11 appears not to be part of a dyadic set but this seems rather to be a consequence of the punctuation choices that were made in the transcription process; it should rather be seen as part of the same phrase that continues into line 12 and forms a dyadic set with line 13.

82

mandy sadan

Figure 2 (cont.) Dyadic set

Occurrence

Ginnip

Ginchyip

Marawng

Mara

Yawn

Kawn

Yan

Hpyan

Rawq

Hkrawq

Gumrawng Gumlau

Yawn

Gawn

Semantic relationship

009

010/013 Euphonic—second word only carries meaning; ginnip is a contracted form [verb] 012/014 013/014 Couplet (order reversed in line 014) [verb phrase] 012 012 Euphonic—complementary meanings [verb] 012 013 Euphonic —synonymic relationship [verb] 014 014 Euphonic—complementary meanings [verb] 016 017 Euphonic—complementary meanings [ part of noun phrase + verb here used as a noun] 016 017 Euphonic—complementary meanings [verb]

Much linguistic research needs to be conducted on these ritual recitations, but the occurrence of couplets and other forms of dyadic relationship suggests that, beyond the purely linguistic, there is a broad historical and anthropological signicance to these forms, occurring as they do particularly in complex, multi-ethnic spaces across Southeast Asia and beyond (Fox 1988). When viewed as a discourse rather than as a set-piece genre, oral ritual language becomes a key tool for the historical exploration of identity formation, maintenance and transformation in complex ethnographic settings such as the Kachin Hills. It is the contention here that the omission of the study of ritual language in context, or of couplet performances of ‘verbal categories’, helps to explain the rupture in the conceptual understanding of these terms in Leach’s work and their historical misrepresentation. When studying these ritual forms in ritual idiom, it is the multi-temporal and multi-vocal nature of the historical discourse that becomes apparent. Terms resonate with multiple historical associations and diverse usages, usually deliberately constructed by the spirit priests or dumsa themselves. The conceptual categories underlying the verbal labels, to which Leach preferred to attach signicance, become more obvious. Figure 3, which relates the same lines of the gumrawng gumtsa

translating GUMLAU

83

recitation cited previously, demonstrates the difculty of translation where a conceptual category is deemed encapsulated by a single verbal label. In this performance context, the lexical items are conceived as being on the one hand deliberately ambiguous, but on the other hand as relating conceptually to a very precise cognitive and semantic domain. As stated, the spirit priest who was making the recitation on this occasion had previously converted to Christianity, and was concerned about making the performance. The recitation is, thus, loaded with ambiguous layers of meaning, as he appeals to the Christian God for understanding of his errors, but done in language that evokes traditional appeals to ‘animist’ spirits at the commencement of a performance. It was a characteristic of the opening statements of any such spirit recitation that the priest should make explicit any circumstances affecting the performance that might surprise or offend the spirit concerned, in order that their potential malevolence could be assuaged (Sadan 2002). The extract shown could be paraphrased as follows: the spirit priest makes an appeal to the Christian god and/or ‘animist’ spirits for forgiveness of his ‘errors’ (sins? verbal mistakes? misguided agreement to participate in the performance?); he justies his performance to God/spirits by stating that his intention is to explain the origin of ‘revolution/upheaval’ and ‘democracy/removal of despotic chiefs’; this was done at a time of heightened nationalist conict in the early 1990s, before the ceasere between the government and the KIA in 1994. Despite the clear contemporary layer of meaning, the recitation was made in a wholly ‘authentic’ narrative mode with ritual lexis, the one exception being Karai Kasang, as explained in the footnoted notes to the text. Signicantly also, the prex gum- for all the key referents, and the ability of this prex to mutate them into other couplings throughout the rest of the recitation not detailed here, demonstrates the coherence of the major referents as part of a historical discourse upon the chiey system of rule.12 Some colonial ofcers appear to have recognised this tendency towards couplet usage, but failed to conceptualise adequately the cognitive aspects of these terms, being too bound up in their own minds with the colonial definition of ‘revolution’ for gumlau. For example, the Deputy Commissioner at 12 I am extremely grateful to U Chit Hlaing (Prof Kris Lehman) for this point, which was made in response to a presentation of parts of this chapter at the EUROSEAS Conference, Paris 2004).

84

mandy sadan

Myitkyina, Mr W. A. Hertz, wrote in his Report on a visit to the Hukawng Valley in March –April 1907:13 ¶62—From Gatawk to Magwitawng: From Magwitawng to the Malikha (Irrawaddy) is said to be a journey of ten days over three high ranges of hills. There is very little lowland paddy cultivation here and the people live mostly by taungya cutting. There are about a hundred and fty slaves in the village and 22 guns. The akyiwas will not admit that their villagers are kumlaus, and say that they are kumlaukumsas, which is contradiction in terms, as a kumsa is the opposite of a kumlau, the two words being antithetical. However, the double word expresses what they mean, namely, that the people are republicans in sentiment, but have not openly declared themselves kumlaus.

Precise unitary translation of these terms is inappropriate because the historical modality of the couplets used, and their multi-temporal trajectories, make this so. Figure 3 demonstrates an attempt to ll-in ‘gaps’ of meaning in the gumlau origin recitation by making the transcription an inter-textual object, footnoting terms with both a historical dictionary and a local intuitive commentary. Not only did the translation prove too difcult without an adequate level of interpretive intervention from the spirit priest himself, but my attempt to use an historical dictionary published rst in 1906 (Hanson 1954 [1906]), the very same as used by Leach, was objected to quite vigorously, not for the obsolescence of the work, which remains to this day a key marker of Kachin transitions to modernity in local development discourses for the way it represented a transition to ‘literacy’, but, paradoxically, for the impact its use would have in pre-dening terms in ways that reduced their current and historical multi-vocality. As seen in the example cited, couplets in ritual historical discourse possess a lexical multi-vocality that, in this case, can encompass themes such as rebellion, disorder, religion, conversion, history and present politics, yet without having to make signicant structural or lexical adaptations to contemporary language or discursive modes. These are not separate strands of meaning, but are implicitly cross-referencing and mutually signicant discourses. It is also the nature of couplets that they help to hone in on cognitive or conceptual categories in monosyllabic languages, such as Jinghpaw, giving greater precision to the process of word formation than might otherwise be possible.14 When Kachin 13 14

British Library, Oriental & India Ofce Collections [OIOC]: Mss Eur/F/116/82. I am grateful to U Chit Hlaing (Professor F. Kris Lehman) for this point, which

translating GUMLAU

85

people object to the stand-alone usage of the term gumsa, therefore, in the model that Edmund Leach presented of their culture, I argue that it is this deeper cognitive level of meaning that is being invoked in the objection, not simply a contemporary re-politicisation of meaning. It is the inability to transfer this intuited and experiential knowledge from local discursive models to the realm of ‘Other’ that constitutes one of the difculties that Kachin elites face in their attempts to contest Leach’s model. Conclusion Edmund Leach’s work on the ‘Kachin’ has great signicance for the study of ethnicity in Southeast Asia and beyond. Its contribution to our understanding of the relational aspects of identities ensures that its value as an academic text is unlikely to diminish in the foreseeable future, although the authors of this volume hope that the nature of the discourse about it may see some transformation. However, in critiquing the work we are right to question its use and abuse of history, amongst other things, for the text is not neutral in its impact upon local communities and their desire to represent their sense of historic-ness along indigenously meaningful lines. Whilst Leach was later to recognise the limitations of his historical reductionism and the error of presenting ‘verbal categories’ as unchanging, a-historical forms, it is also necessary to understand more deeply the nature of the ruptures that were created by these errors. In so doing, we may help to reposition the text in new lines of analysis, as well as to uncover forms of interpretation that may yet reveal more nuanced understandings of the role that history may play in anthropological constructs of ethnicity.

he made in response to a presentation of this paper at the EUROSEAS conference, Paris 2004.

86

mandy sadan

Figure 3: An attempt to make an inter-textual object of a Jinghpaw ritual recitation Nnga e . . . hkawp1 e ya, hkringwa ga bai jasat2 yu gaw3, dumsa mahku4 bai shaprat5 yu gawq, hkawp hpan wa Ningsang e Karai wa6 e ya e, laijum7 ga bai jasat nna me, ka-ang matut8 bai yawn9 na rai sai, hkringwa ga bai shaprat nna, shing-ra10 madat11 bai gawn12 na, ya ningdung13 wap14 na nang ginnip, ninghpang15 dap16 na nang ginchyip17, nang hpan Ningsang wa, ya hkringwa e yawn ai marawng18 mung e nang kawn19 yan20, dumsa chyai ai mara21 mung nang ningtsa Karai hpan Ningsang kawn ginchyip nna hpyan22 bara23, mara mung rawq24 marawng mung hkrawq25 ya bara, n nga n mai rai nna nga, n dumsa n mai rai nna dumsa gumrawng gumtsa26 nga ai ginshi rai nna yawn, gumlau27 paw ai ninghpang gawn na rai sai nga ma ai. 1

An honorary epithet or appellation for nats or great chiefs (Hanson 1954: 312). To walk lit.—P. Ja Li interpreted this metaphorically as meaning ‘to chant’. 3 Hkahku pronunciation of yu ga—exhortative. 4 A sound, a note, a noise (Hanson 1954: 420); a tune ( Ja Li). 5 To beget, procreate, generate—sometimes a couplet with shalat—to beget, to give birth (Hanson 1954: 647); here meaning ‘compose’, as in a tune (P. Ja Li). 6 The dumsa was a convert to Christian minister who had converted from animism some years previously. He was, therefore, concerned about making this recitation, and so, at various points, attempts to conrm his belief in God (translated by Hanson in a Christian sense as Karai Kasang) and to justify his reasons for making this recitation. 7 Possibly a combination of the preformative lai (Hanson 1954: 352) and variant spelling of chyum, which has a similar meaning to that of a divine revelation (used by Hanson in Chyum Laika as a translation for ‘Bible’). 8 To connect, join, link (Hanson 1954: 440). 9 To tell, relate, as a story (Hanson 1954: 732). 10 When used as a couplet with ka-ang this phrase has the meaning of the earth that is the appropriate/natural/original habitat for humans. P. Ja Li states that here it has the meaning of the original humans/ancestors. 11 To listen, to harken as an instruction (Hanson 1954: 409); P. Ja Li states that it has the meaning of ‘instruction’ here. 12 To narrate, or relate as news, a story or fable; to detail (Hanson 1954: 174). 13 The end (Hanson 1954: 130). 14 To be bulging, baggy or wrinkled; g. to be loose or vague, applied to the mind (Hanson 1954: 708). 15 The beginning (Hanson 1954: 536). 16 To protrude, jut out; g. to show pride, to be overbearing (Hanson 1954: 107–8); conveys the idea of concern lest anything should be incomplete and not dealt with fully in the recitation. 17 To be arranged, be put in proper order (Hanson 1954: 89). 18 The Maraw nat or protrusions, juttings (Hanson 1954: 433–4); P. Ja Li interprets this as the Maraw spirit. 19 Possibly means to look after, tend with special care (Hanson 1954: 241). 20 To be unrolled and spread out, to ow in an uninterrupted manner; g. to be freed, loosened, as the tongue of a mute (Hanson 1954: 720). 21 Fault or guilt (hanson 1954: 430). P. Ja Li interprets this as ‘sin’; cf: Hanson (p. 434) states that marawng is a couplet with marang, suggesting that marawng mara could be a derivation from this or a reinterpretation. 22 To unroll and atten out; g. to grow, open up, expand, as the mind (hanson 1954: 555). 2

translating GUMLAU

87

23 Could this be a variant of bra, to be scattered, dispersed, or to be apart, loose (Hanson 1954: 73)? 24 To untie, unbind, loose; mara raw, to explain, offer an apology, a defence; see also maraw raw ai (Hanson 1954: 576). 25 To agree, concur, give assent to; to be dry, gruff, grating as the voice (Hanson 1954: 328). 26 Hanson (1954: 166) gives gumrawng gumtsa a more pejorative tone than is usually used (cf.: gumchying gumsa is usually used to describe an oppressive ruler, whilst gumrawng gumtsa is today interpreted as a form of proto-democratic inclination). Hanson states the historical usage was that gumrawng gumtsa was proud, and thus acknowledging no authority or law; gumrawng gumtsa ga, a place without a ruler. These meanings have today become associated with gumlao in distinction to the usage here which is to overthrow (see below). 27 (From lau, to tempt) to cause trouble, mischief, disturbance, to rebel, revolt; gumle gumlau n. trouble, rebellion (Hanson 1954: 163).

PART TWO

COMPARATIVE APPROACHES IN ASSAM AND LAOS

POLITICO-RITUAL VARIATIONS ON THE ASSAMESE FRINGES: DO SOCIAL SYSTEMS EXIST?1 Philippe Ramirez For its vast cultural diversity and the multiple interactions among its human groups, Northeast India, i.e. Assam and its adjoining areas, may be described as a paradise for the anthropologist. For the same reasons, however, when interpretations and hypotheses have to be considered, when some “order” has to be envisioned, paradise may turn into hell. The discrepancies between ethnic labels, social patterns and cultural features, the internal differentiations within groups, the overlapping of political and ritual models, all make the social scientist doubt whether any assertion can be put forth, whether any social system exists at all. The recent owering of ethnic politics in the region adds an additional dimension, which does not make the task easier. As a matter of fact, the anthropological complexity that characterises Northeast India pertains to a wider Southeast Asian area, which also covers northern Burma. So it is natural that Political Systems of Highland Burma is called upon by anyone attempting to nd coherence in the great anthropological maelstrom of present-day Assam. The scepticism of Leach, which inspired him to the general idea of dissociation between political systems and “tribes”, has the great merit of conjuring essentialism, which always threatens the anthropologist. Essentialism is a pervasive trend in the times of ethnic revivalism in which we are living, and particularly so in present-day Northeast India. The area experiences multiple identity claims, the discourses as well as consequences of which may be usefully analysed with the tools provided in Political Systems. The discursive essentialism of ethnic movements sometimes runs up against such obvious cultural disparities ‘on the ground’ that even hard-liners have to admit their signicance. Ethnicism assumes a perfect correlation between a territory, language and sociopolitical system. As is obvious to all, correlations with this degree of

1 Among the people who made this eldwork possible, I would like to thank more particularly Samiran Boruah in Guwahati, Sarvajit Thaosen, Snigdha Hasnu and A. K. Langthasa in NC Hills, and Kangbura Senar in Diphu.

92

philippe ramirez

adequacy are rarely realised in Assam. Nevertheless, one cannot reject ethnic categories too easily and only on the pretext that they do not correspond to actual cultural or social entities. Every contemporary anthropologist would agree that is not cultural homogeneity that makes an “ethnic group” but a perception. It is through such representations that people feel they belong to a coherent and perennial entity. In this sense, the Assamese ethnic groups, or “tribes” as they are called and call themselves, are real. Thus remains the task of analysing the competing articulations of identity discourses and the “reality” of other domains, a task which was central in Political Systems. I will briey present three cases taken from Tibeto-Burmese speaking people living at the fringes between hills and plains in Central Assam. These cases are drawn from research that focuses mainly on the historical and anthropological foundations of ritual geography in an area that, before the British Raj, was alternatively or conjointly under the political authority of three main states: the Ahom, the Jaintiya and the Dimasa (or “Kachari”). Some historical background is necessary before proceeding to the anthropological data. From the thirteenth century onwards, a number of warriors of Tai origin, the Ahom, gradually conquered the lowlands of the Brahmaputra valley. There they became acculturated by the former elite, who were Hindu by religion and Assamese-speaking. In Central Assam they faced competition from two hill-based states: the Jaintiya kingdom, which was culturally close to the Khasi, centred on the Eastern Meghalaya plateau and whose authority extended far below the foothills; the Dimasa kingdom, which till the sixteenth century controlled the greater part of south-central Assam before slowly receding towards the southern hills of present-day North Cachar district. The precious Ahom chronicles, the burañjis, provide valuable testimonies on the relations between the rising Ahom power and the existing states.2 They also give several indications on a number of people and chiefs who lived on the southern foothills of the Brahmaputra valley and whose allegiance gradually shifted from the Jaintiya and Dimasa to the Ahom. Thus, on the fringes of its dominion, the Ahom sovereign (swargadeo) recognised a number of petty kings, the rÊjÊ povÊli, who begged for his protection and were required to assist him in war. These chiefs, signicantly called dÊntiyÊliya rÊjÊ (“kings of the margins”), generally controlled the lower portions

2

See Baruah, 1985, particularly p. 369 ff.

politico-ritual variations on the assamese fringes

93

of the routes linking the hills to the plains via a number of marts at the bottom. These small dominions played a “bridge and buffer role”, which has to be kept in mind when considering the shaping of local cultures and identities.3 The rst example that I will explore pertains to the people who call themselves Dimasa (“Sons of the large water”). Numbering around 100,000, they speak a Bodo-Garo language and live in the hills bridging Meghalaya and Nagaland. In British times they fell under the label “Kachari”, which covered a fair portion of the Tibeto-Burmese speaking groups of the plains, including the people who today refer to themselves as Bodo. These Dimasa display a kind of anthropological strangeness in their descent system, which is dual, combining male clans and female clans respectively passed on by men and women. Dimasa conceive of themselves as the descendants of rulers and subjects of the Dimasa kingdom. Ahom chronicles attest the existence of “Timisa kings” (khun timisÊ),4 who ruled over a large area of Middle Assam, initially from Dimapur on the Western foothills of present-day Nagaland. They were driven by the Ahom towards the south to their present habitat, where in 1540 they founded the Hedamba kingdom. Despite the scarcity of data on this kingdom, the archaeological vestiges in successive capitals (present-day Dimapur and Maibong) indicate a fair level of economic strength and centralisation. From at least the eighteenth century, the kings and the elite went through a process of Hinduization, apparently under the inuence of the Bengali brahmans attached to the court. Following the nal invasion of the area by the Ahom in 1745, the kings and part of the commoners settled further south in the plains of Cachar. Culturally speaking, present-day Cachar Dimasa can hardly be distinguished from the Bengali majority and they seem to make little attempt politically to assert their identity. In the hills, however, Dimasa remained demographically dominant whilst cohabiting with Hmar-Kukis and Zemi Nagas. As a scheduled tribe, they obtained the creation of an “autonomous council” in 1952 and of an autonomous district (NC Hills) in 1972, where they benet from special reservations. In some ways, Dimasa people of North Cachar Hills face few problems when dealing with their identity. They share a fairly high

3 4

On this notion, see B. K. Roy Burman, 1994, pp. 81–91. Barua, 1985, p. 86.

94

philippe ramirez

linguistic and cultural homogeneity, all claiming their link with the former Hedamba kingdom. In comparison to their neighbours, their society can be described as a “sealed society”. The status of Dimasa is strictly acquired by descent: a Dimasa is dened by his descent from both a male clan and a female clan, adoption within a Dimasa clan is impossible, and union with a non-Dimasa forbidden. In the latter case, such an act exposes the whole community to impurity ( gushu), and it generally results in ex-communication. Nowadays, Dimasa girls may on rare occasions be allowed by their patriclan to marry a foreigner after being puried in their clan sanctuary, but even in this case any offspring will be denied Dimasa status. Finally, under the special provisions conceded by the Indian State, matrimonial affairs, like other “customary” affairs and land issues, legally fall within the authority of the autonomous district council. This judicial autonomy reinforces the “exclusive identity” of Dimasa. This does not mean that the internal order of the society is the object of a consensus among Dimasa. Put simply, despite the harsh protective measures enforced towards external inuences, Dimasa society has recently experienced dramatic changes in its cultural, economic and demographic features. Among others, the decline of shifting agriculture ( jhum) as the main source of revenue, and the shift from the villages to the towns have deeply transformed the mode of living and the human geography of the area. These changes have necessarily resulted in inconsistencies that have provoked debates between conicting views on “tradition”. A few years ago, the autonomous council was asked to rule on the exact meaning of the word which refers to the main Dimasa ritual institution, the daikho, which for the moment we will only dene as a “worship circle”. The debate pertained to the concrete case of a priest attached to a daikho who had planned to move his home to another location, and intended to shift the place of worship as well. Some of the follower lineages strongly objected to this. They resorted to etymology, postulating that “dai-kho” was in fact “mdai-kho”, “god-abode”, or the “house of god”, and that, consequently, it could not be displaced; the priest on his side argued that a daikho was simply a “worship spot”, independent of any locality, and thus the offerings could be made anywhere.5

5

The court could have included Political Systems in the exhibits: because, as a matter

politico-ritual variations on the assamese fringes

95

So what is a daikho? The question is central to any attempt to describe the ritual institutions of the Dimasa and has tormented at least one anthropologist before me. Relying on the data she collected in the seventies in northern North Cachar, Danda stated that the whole of Dimasa country is under the authority of “area-gods” who have “non-structured sanctuaries” in different localities, called daikho.6 The area-god controls the life of the villagers who live in its area. However, Danda noted several elements which pointed towards clanic features of the daikho: in some localities, certain clans claimed to be afliated to another daikho than the local one, worshipping the rst one, the second or both. She concluded that the “daikho gods” were formerly clan-gods but that, due to migrations, they became area-gods.7 The elements I collected in southern North Cachar in 2002 differ signicantly from those of Danda, but, as we will see shortly, they pose the same problems of interpretation. I need to say rst that I came across two particular types of hurdles. First, the indigenous descriptions of the daikho vary greatly from one informant to the other; second, the main functionary of the institution, the jonthai priest, strictly avoids all kind of impurities, including contact with foreigners, i.e. all “non-Dimasa”. So the information one can obtain originates only from laymen. All informants described the daikho as a spot where collective worship of local deities was held either on a clan-basis or on a pan-Dimasa basis. People are supposed to gather regularly in a daikho with their agnates to sacrice to their tutelary deity, and should any disaster affect the Dimasa area, such sacrices would be held simultaneously in all daikho. It is not clear whether the daikho corresponds to an ancient administrative unit, but everybody attributes the institution to the ancient kings. Furthermore, although the most commonly given etymology of the term is “house of god”, i.e. sanctuary/temple, some translate it as “container of nes”, the place where in ancient times the judiciary nes were collected. According to these informants, the offenders were ned and puried only in the daikho to which they were attached.

of fact, daikho is often spelled mdai-kho, where mdai means “deity”, a term strikingly similar to the Kachin Madai, the chief of the sky spirits which has an altar in the headman’s house (Leach, 1954, p. 136). Kho can mean a “place” or a “container”, like the granary, maikho. 6 Danda, 1978, 126–127. 7 Ibid.

96

philippe ramirez

The Dimasa land is said to be divided into twelve daikho, each presided over by a god and under the responsibility of a priest, the jonthai, selected from a particular clan and who cannot in any case leave his post. The activities of the jonthai are monitored at the centre by two religious ofcers, the Great jonthai, jonthaima, and the puricator, gisia, who are themselves appointed by the council of the forty Dimasa male clans, the calis. At rst glance, therefore, the daikho system as described by the Dimasa may look well codied, coherent and in tune with the ancient political context. However, beyond these general principles, all kinds of detailed descriptions may be heard, and, as one tries to go more deeply into the subject, the rst impression of coherence gives way to a feeling of true confusion. For instance, a seemingly simple question such as “Who appoints the jonthai ?” gets several different answers: some say they are appointed by the great priests ( jonthaima-gisia), others that they are chosen by the daikho followers, sometimes by divination, sometimes without, or that they inherit the function from their father. Similarly, (but this is more common in India), an inquiry about the list of the twelve daikho does not generally go beyond the same two or three names.8 One wonders, then, if it is possible at all to draft a general sketch of the system, to clear up a minimum number of common practices, or even to draw a simple map of the daikho. More troublesome still is the absence of certainty concerning the relationships between daikho, gods, clans and territories. Although daikho worship displays many clanic features, the Dimasa seem not to be sure whether the rst criterion for afliation is the clan or the residence. The touchstone of the problem is obviously migration: what happens if a family migrates far from the daikho it used to attend? Some insist that it would have to perform in the original daikho, while others assert that it would attend the nearest daikho. This ambiguity possibly reects the equal importance of the principles of descent and residence in present-day Dimasa society. I am not convinced that, as Danda suggested, clan afliation gives way to locality afliation and that clan-gods became locality-gods. Dimasa social structure undoubtedly went through major transformations, especially after the disappearance of the local State and its centralising dynamics; this is a subject too broad to be

8 Authors writing on the Tay dodecarchies have noted that the number 12 is purely conventional; cf. Condominas, 1980, p. 270 and Izikowitz, 1962.

politico-ritual variations on the assamese fringes

97

elaborated upon here. Nevertheless, the relationship between descent and residence is always a complex one, and all the more so when one considers its ritual features. In any clanic society, local gods may be adopted by immigrating clans, thus becoming clan-gods, or clans may bring in their own “portable” tutelary god and “settle” it in their new locality. A right-holder clan always develops an osmotic relationship with its locality, so that the locality gets identied with a clan, and a clan with a locality.9 Thus, we may remark that, although Dimasa make both the clans and the daikho the two main pillars of their society, they feel uneasy when asked to account verbally for the relationships between the two because, unavoidably, contradictions arise. I suggest that most of these contradictions are due to the fact that the anthropologist, trying to crosscheck discourses with “real facts”, generally confounds at least two, if not three domains of facts, or logic: the domain of identity discourse; the domain of general rules; the domain of practice. Each domain has its own coherence, which cannot be simply translated to the others. When Dimasa dene their society as having forty patriclans and twelve daikho, it should not be taken as an assertion of the actual number of existing clans and daikho. Similarly, if it is found that at the larger scale, the clan is the organising principle of daikho worship, it does not rule out the possibility that lineages within a clan will attend different daikho on a territorial basis. As a matter of fact, it seems that the actual practices reect all the possible arrangements of clanic and territorial worship. As an illustration, one very peculiar case may be described, that of the ideal village of Semkhor, which stands somehow outside the daikho complex, but which, paradoxically, helps us to understand its nature. Semkhor, a remote locality on the fringe of the Naga hills, is a strictly sealed village, multi-clanic and fully endogamous, very timid towards foreigners; these features make it a kind of microcosm of Dimasa society, a picture

9

Further investigations may reveal the exact relationships that the Hedamba State entertained with the Dimasa clans. The Dimasa term for “male clan” is sengphong, which literally means “holder of the sword”, i.e. sacricial sword which Dimasa distinguish from the ghting sword, tang. One of the rst Dimasa kings held his power from the discovery of the sword Ranachandi—an epithet of the Goddess (Gohain, 1977, pp. 11–12). This let imagine a “clanic state”, and a territory which was partaken among several clans, each with their ritual centre, the daikho. Possible links have also to be explored with the sword hengdÊn, which the Ahom sovereigns handed down to their dignitaries and vassals (e.g. Bhuyan, 1990, p. 62).

98

philippe ramirez

which is conrmed by the fact that, according to some Dimasa, within its limits seven salt wells are to be found, which corresponds to the seven original Dimasa clans.10 This suggests the existence of former clanic monopolies over salt, and maybe a central daikho. Semkhor has adopted a ritual formula that enables it to combine the principles of clan and territory: each clan worships its own tutelary deity in a specic spot outside the village.11 I think this peculiarity is far more than idiosyncratic. In a certain way, it is the strict ritual separation between clans that constitutes the village unit: a paradigm that suggests that it is not in the strict adherence to a unique set of norms that the coherence of the daikho complex may be found. Thus, the apparent uncertainties of Dimasa discourse on the ritual institutions neither point to deculturation (to a “detribalisation”) nor sustain the idea of a chaotic, non-structured reality. I have no doubt that the Dimasa, when the time arises, know exactly what they have to do, and that rules in the true sense preside over the practices. What needs to be underlined in the light of this data is that the general description—and possibly perception as well—of the institution by the actors themselves gives little clue to its operation. This is not to suggest, as Leach suggested in the introduction to Political Systems, that world-view and practices are fully dissociated. They are certainly articulated by a “missing link”, which substance may be searched for in the historical transformations that followed the vanishing of the Hedamba state: Dimasa would still view their society as a centralised clanic state. In comparison with the Dimasa, the second society I would like to consider displays almost antinomic features. The 400,000 Karbi, scattered in the hills and plains between the Meghalaya plateau and the Naga hills, do not possess very deep and corporate clans. They have apparently developed no centralised political institution, except the Karbi-Anglong autonomous district council (1952), which administers the hilly section of their territory. They are divided between at least three cultural areas. Although the quality of Karbi is theoretically based on the belonging to a Karbi clan, outsiders may be adopted through a “purication” rite. If they did not speak relatively similar languages, all part of the Kuki-Naga family, the unity of the Karbi would seem to be almost purely “ethnic”, i.e. widely self-ascriptive. In other words,

10 C.f. Pascal Bouchery, 1988, on the importance of salt in the relationships between the neighbouring Nagas and the Ahom state. 11 Danda, op. cit., pp. 126–127.

politico-ritual variations on the assamese fringes

99

compared to the “sealed society” of the Dimasa, Karbi society could be qualied as an “open society”. The Karbi were made famous by a monograph written by Edward Stack in the early twentieth century on “the Mikirs”, as they were called by the British and the Ahom.12 The “Mikirs”, as the Karbi today, were divided into ve clans, which are represented in all Karbi inhabited areas and which have little other role than xing the exogamic rules. Politically, three types of local government are found among the Karbi. The Karbi of the plains, called Dumrali by the Hill Karbi, live outside Karbi Anglong district, within Assamese-dominated or multi-ethnic settlements. As many other plains tribes, they do not fall under the Sixth Schedule of the Indian constitution, which provides the scheduled tribes with the capacity to run an “autonomous council” of their own. So, among the plains Karbi, there exists no traditional authority above the hereditary village chief (bangthai ), who is stricto sensu a ceremonial chief but who also often assumes the charge of gÊU burÊ, the usual ofcial village chief in Northeast India. In Karbi-Anglong district, the autonomous council appoints village chiefs (sarthe) in all areas and deals with land as well as customary issues.13 In the eastern half of the district, the system is relatively simple, with no intermediate levels. The old council of village elders (me), which is said to have been formerly the real seat of authority, seems now to be in decay. In western Karbi-Anglong, however, formerly under the sphere of inuence of the Jaintiya kings, there remains above the usual village chiefs a sophisticated institution which the Karbi call the lindok system, by reference to its head functionaries, the lindok. As in the case of the daikho among the Dimasa, the descriptions of the lindok system by the Karbi vary greatly and often look incoherent at rst sight. As a matter of fact, the rules differ from one place to another, but even within a single area the perceptions differ, particularly in the relationships between functions and territories, as well as in the territorial setting itself. Let us focus now on what happens in this particular area.

12 The term “Karbi” does not seem to appear in the Ahom chronicles. It seems that “Mìkir” was used instead; e.g. Barua, 1930, p. 285, for a mention in the eighteenth century. 13 Most villages are referred to by the clan name of their present or former chief, a situation which seems to be peculiar to the Karbi area.

100

philippe ramirez

Every Karbi agrees on the existence of three main territorial divisions, which are always listed in the same order of precedence: Rongkhang, Chinthong and Amri. Each division is governed by a pinpomar, a council of pinpo, who are representatives appointed by their peers from a number of constituencies. Within the pinpomar, representatives of a constituency are headed by a lindokpo, who is endowed with several royal aspects and referred to as a recho, i.e. raja. A very peculiar feature of this system is that during their tenure, the pinpo come to live in a “capital town” (rongbong), a specic, isolated locality which is internally structured into wards corresponding to each constituency and centred round a sacred enclosure. The pinpomar appoints local chiefs, the habe, who are each in charge of several villages: their prime responsibility is to supervise rituals and settle cases pertaining to customary laws, i.e. mainly matrimonial affairs. Habe regularly refer to the pinpomar, which thus acts as a paramount court for customary affairs. So, at rst sight, the lindok system appears to be a well structured “tribal state”, with clear-cut representation and decision processes as well as territorial attributions. However, as in the Dimasa case, the geographical structure of the institution is much more difcult to discern. It is not a surprise that ordinary villagers are unable to describe it in all its details: what proportion of British or French voters can clearly picture the structure of the local governments they are supposed to elect? Yet, one faces the same difculties in making the various accounts given by the functionaries themselves coincide. The number of habe varies, the names of their territories vary, spatial groupings vary, and so on. So, if at the most general level the system is described as structurally homogeneous and spatially continuous, when it comes down to the details it appears spatially discontinuous and incorporating a number of local peculiarities. In the Karbi case, centralisation does not involve levelling. Historical facts may account for some of the political discrepancies. Western KarbiAnglong remained under Jaintiya suzerainty long after the imposition of Ahom rule in the eastern part (seventeenth century). Later on, the east was more directly administered by the British than the west, which like other “primitive inhabited areas” was qualied as “excluded” or “partly-excluded” and not subject to tax settlement. On a more local scale, differences were numerous. Nevertheless, the heritage from different administrative regimes does not sufce to explain the gaps between representations and practical settings.

politico-ritual variations on the assamese fringes

101

An even greater heterogeneity characterises the content of collective rituals, even in the lindok divisions. This may look striking, as one of the main functions of the lindok institution is to monitor these rituals. Rongker, the prosperity festival, which is cited by all as the most important ritual, as “the” festival of the Karbi, is not associated with the same territorial levels and with the same type of deities in all localities. Literally, “Rong+ker” means “village+public worship”. Stack described it as an annual village worship dedicated to two types of deities: rstly the “Great Hill” (Inglongpi ), the god of the hill where the village stood; second, the “Hundred Gods” (Arnam pharo), which Stack interpreted as a collective term. Both were asked to let villagers use their territory during the year in exchange for sacrice. In exceptional cases, like attacks by tigers, Great Rongker (Rongkerpi ) may be organised on a mauja basis (several villages).14 A Karbi writer I interviewed in the district headquarters recently gave me a quite similar account: during the annual Rongker, the village propitiates rst the three “great gods” (Hemphu the creator, Mukrang, his rst follower and brother-in-law, Rasinya, Hemphu’s sister) and then the “local” ones. On a less regular basis, groups of villages, belonging or not to the same territory, may perform Vophong Rongker for their own benet. Another intellectual from the same area stated that annual Rongker was held conjointly by several villages. Other Karbi of all extractions held varying views, including the inexistence of a territorial basis for this ritual. In Chinthong division, functionaries in charge of the management of all local ritual asserted that no village Rongker took place in their area, but only a general Rongker organised at the division level. Finally, the plains Karbi of Kamrup (Guwahati area) do not perform Rongker at all, their main annual collective ritual being Dehal pÖjÊ, addressed to Tamlong alias Buda GohÊi. In Hindu Assamese terms this would mean a “temple worship” (deval pÖjÊ) in honour of Siva. It is even difcult to understand if the present Karbi conceptualise something looking like local deities, to which this annual worship would be addressed, and who would give us a clue about the Karbi symbolic landscape. Educated Karbi denitely seem to under-estimate these deities in favour of non-localised encompassing “Great god(s)”, particularly Hemphu, evoked with a mixture of Hindu and Christian tones.

14

Stack, pp. 31, 42.

102

philippe ramirez

More generally, Karbi seemed puzzled by my questions about “local deities”. Only twice did I obtain a denite statement that conrmed the existence of “area deities” (longri arnam) who were addressed during Rongker. The general impression is that Rongker is denitely a collective ritual, that it is not clanic at all but territorial, but that according to the locality, it is held on different territorial levels and according to different procedures. Even in the western area, under the lindok regime, the sophistication of a central authority at the higher level does not imply that practices at lower levels are similar and tightly regulated, and it does not imply as well that inhabitants of a particular locality are aware of their neighbours’ practices. The rationale behind the lindok institution is not itself obvious. One could argue that it is a relic of a vanished political system, with no remaining coherence. It could also be seen as “articially” maintained by the Karbi, as a virtual symbol of their unity, a sort of pastiche of a State. In this case, the lindok system would form a pure variant of Geertz’s “theatre-state”, which would have relinquished all concrete governmental functions and a great part of its territorial rooting.15 I won’t go so far, because several clues show that the lindok and habe still hold aspects of concrete power. In matrimonial affairs at least, it seems that they are still able to impose their decisions. As for the rituals, they still “send orders” to the local headmen when the time comes, asking them to start the performances. However, it is true that everything looks as if the main preoccupation of the apparatus is to ensure that, in the ritual domain, things are done, whatever the way they are done. Parallels can be drawn with what we have called above the Dimasa “identity discourse”. Through the lindok system it is an ideal image of the “traditional” Karbi society that is enacted, however faithfully or not it reects the actual practices. While entertaining comparable representations on the ideal order of society, Dimasa and Karbi conceive the relationships between this internal order and the outer world in diametrically opposite ways. Where the Dimasa stem by birth from a matrix (the forty clans) that must be protected from the introduction of all external elements, Karbi internal order is a mould where external elements can be melted. Among Dimasa and Karbi the disorder implied by contact with the alien is dealt with through the same operation: purication ( Dimasa: thar-ba, Karbi:

15

Geertz, 1980.

politico-ritual variations on the assamese fringes

103

thekal ).16 However, purication, which in both cases restores order, does so in two opposite ways and thus brings forth opposite outcomes: the Dimasa purication aims solely towards the exclusion of the foreign substance, the pollution, be it in the case of polluted mourners or of polluted wrongdoers, who were formerly “puried” after being ned; the Karbi purication washes the foreign element of its alien-ness, thus enabling its integration. It may without too much risk be compared with a baptism. Karbi commonly assert that a Karbi is someone who believes in the supreme god Hemphu and who follows the rules instructed by him: so, in principle, anybody can become a Karbi. This idea is much more than a modern progressive slogan. The process, already mentioned by Stack one century ago, is quite common and I have documented a recent case of collective conversion from Garo to Karbi, not far from Guwahati.17 A Garo village isolated among Karbi had difculties nding matrimonial matches; so, by a simple act of collective purication and clan name attribution, all of its inhabitants became Karbi. Such a case is a clue to one of the several processes through which Karbi diversity—and, in fact, Assamese diversity—could historically have been established. The set of rules that Hemphu instructed and that one has to follow as a Karbi is hardly detailed beyond the respect of clan exogamy. But it is the declaration that matters much more than the content: for contemporary Karbi, the ultimate criterion is the creed in the rules. I cannot for the moment assess how ancient is this attitude, and how far its Christian tones have something to do with the conversion of a part of the Karbi to Christianism during the twentieth century. Nevertheless, such a model of identity building is not completely at odds with what seems to be an older model in Assam, and elsewhere in India, in which a group identity—ascribed identity at least—is determined not by certain cultural features but by the allegiance to a political authority or to a political order. In this case, cultural heterogeneity does not prevent coherence of the group in terms of identity or social relation16 Purication seems to play a central role in these two societies, both at bottom and central levels, with specic functions and meanings compared to the classical Hindu purication. Some clues indicate a possible common background: for instance, the Dimasa thar-, “to purify”, might bear some link with the Karbi term kathar, which designates a village priest in Kamrup and the central priest in the lindok area, who is primarily considered as the “rains priest”. 17 Stack, op. cit., p. 23.

104

philippe ramirez

ships. However, the modern Karbi introduced a new dimension in the old model: not only cultural features but also political afliations are secondary to the assertion of identity. Karbi are neither those who live in the autonomous district nor those who are governed by a peculiar set of institutions (e.g. the lindok system): they are all those who claim to be Karbi. The Karbi identity model as proposed here may be particularly contrasted with the third case I will describe, which represents what I call the “old identity model”. In such a model, externally or internally ascribed group identity was determined by political afliation. For instance, at the advent of the British Raj, the Karbi of Eastern Meghalaya were identied as Bhoi. From a Khasi point of view, “Bhoi” referred to several linguistically distinct groups living at the northern periphery of the Khasi-Jaintiya polities and whom the Khasi and Jaintiya chiefs claimed as their subjects.18 Among these groups fell those who today call themselves Lalung or Tiwa (c. 35, 000): they inhabit the northeastern corner of Meghalaya, speak a Bodo-Garo language, and display a matrilineal descent system, like their Jaintiya and Khasi neighbours.19 The present Tiwa do not identify with Bhoi but with a confederation of principalities at the margin of the hills and plains. The Ahom chronicles mention several petty kings on the southern fringe of the Brahmaputra valley who, in the eighteenth century, alternatively fought or paid allegiance either to the Jaintiya kings or to the Ahom, before nally being forced to submit by the Ahom.20 Ahom and Jaintiya kings entrusted them with the task of ensuring free access to the routes and markets linking the Jaintiya kingdom to the plains. The chronicles did not culturally qualify these chiefs, but the deorÊjÊ of Khala, Gobha, Neli and Sahari still exist today as symbolic rulers of the Tiwa. Deputies of each division within each kingdom annually visit their king and are re-appointed by receiving a turban. According to Tiwa traditions, the four remaining kings were part of twelve deorÊjÊ who each ruled a division of the present Tiwa territory.21 Whether the different kings had a similar status in the past is not known, but the present ritual arrangements ascribe a senior 18 Gurdon, 1907, pp. 62–63, is the rst to have noticed the pluricultural nature of Bhoi. What was generally described as a section of the Khasi included, in fact, not only Khasi speaking people but also Mikir and Lalung. 19 On the Lalung, c.f. Gohain 1993. 20 For Ahom testimonies, see Bhuyan, 1933, pp. 160 –165. For Jaintiya versions: Shadap Sen, 1981, pp. 130 –149. 21 Sharmathakur, 1985, p. 75.

politico-ritual variations on the assamese fringes

105

position to the raja of Gobha. He is now considered as “the King of the Tiwas”, and receives allegiance during a great annual fair ( Jonbil melÊ) which gathers people together from the whole region.22 Most of the plains Tiwa speak Assamese and are patrilineal. They claim to have come from the hills to ee the imposition of matrilineality by the Jaintiya kings. Their religion shows as many features common with Hill Tiwa as with the Assamese Hindus.23 In fact, under the British and for some time after Independence, they were referred to as “plains Lalungs”. However, they now strongly reject the term Lalung, which they consider derogatory. Their economic condition and levels of literacy are higher than those of the hill people and their elite are very active in ethnic politics, so that they have now become the reference point of Tiwa-Lalung ideal culture. Among the immigrants who recently came down from the hills, as among educated sections of the hill people, the Tiwa label and cultural models are gaining momentum. This is the result of a particular kind of reversal. Until recently, plains Tiwa were somehow looked down on by the hill people (Hajowali ) and designated as Makdoliving, a term meaning “those who have lost their creed”. So, the formerly peripheral section of the Tiwa, the plains people, escaping their reputation as a de-culturized community, have now become the new reference for Tiwa culture, while the ‘original’ hilly stock is taking on the appearance of a ‘backward’ leftover. However, in terms of identity politics, nothing has changed in the sense that plains and hills people have no doubt that they form a single entity, that they are of the same kind, the main criterion being their recognition of the same kings to which ritual homage is paid every year. The three Assamese cases that I have too briey and too roughly sketched, conrm the validity of one of Leach’s main assumptions in Political Systems: the cohesion of societies cannot be established without, paradoxically, assuming the dissociation of identity, cultural patterns, variations and practices. As we have seen, each of these levels is organized according to a specic logic, which does not imply that it is isolated from the other levels. Indeed, the Assamese anthropological landscape shows how the relationships between identity, cultural patterns and practices may be built in very different ways according 22 Jonbil mela is also attended by Khasi and Karbi speakers of the area. Further investigations will conrm if they are considered and consider themselves as “subjects” of the Gobha king. 23 On the Tiwa, see Sharmathakur, op. cit.

106

philippe ramirez

to places and times. Dimasa combine a very strong identity and strict enrolment rules, both founded on clanic principles, with a exible system of ritual interactions. Among the Karbi, the exibility of ritual apparatus is, in contrast, associated with an acceptation of cultural and political variations and a widely declarative identity. Dimasa and Karbi situations would represent two diverging evolutions of an older pattern, exemplied by the Lalung-Tiwa, where cultural heterogeneity is second to political allegiance, which forms the real basis of identity. In certain ways, this old logic is being reversed in the present ethnic politics. Where a group was formerly dened in relation to a paramount entity, a king or a State, now the group is posed a priori and subsequently negotiates its relations with the State—i.e. India. Members of the group may have different views about the legitimacy of that State and the attitude to be adopted towards it. Some Karbi or Dimasa may acknowledge the present political arrangements, others may ask for a wider autonomy, while others again may reject the sovereignty of the Indian State over their people and ask for independence. In all cases, however, the selfascribed identity precedes the debate over political status. As a matter of fact, it is through the permanent invention of original relationships between identity, cultural variations and practices that societies evolve and often regenerate themselves. Obviously, the discourses that people produce about their own society always tend to amalgamate culture, structure and practices into a single perennial essence; such a discourse is not to be thrown away as a mere ction, it is one of several aspects of social reality. It is on this particular point that Leach’s approach might have to be reconsidered to t contemporary political situations. In Political Systems, Leach did not really speak of identity per se, and he denitely did not speak about identity politics. It is, however, impossible today to avoid what has become a major factor in the shaping of the societies of this region. Thus, for Leach, the “cultural situation” resulted from “historical incidents” and was only a “dress” of the “social situation”.24 Whatever objection can be made against this idea, the fact is that today, people rely largely on cultural appearances in their identity building. Cultural differences between groups may have very concrete implications for their political relationships, as well as for the relationships inside each group. In present-day Northeast India, identity perceptions are largely

24

Leach, 1954, p. 16.

politico-ritual variations on the assamese fringes

107

shaped by the nineteenth century nation model of “one territory, one history, one language, one culture, one people”; cultural similarities and dissimilarities dene spaces that certain groups claim appropriate, and where they try to impose certain types of social relations, certain institutions, certain laws. Through this realization of a culture-based identity model, the correspondence between a social structure and a cultural area may well become a reality. Political unity, whatever its articiality, may induce a cultural homogeneity when, for example, a scheduled tribe is given autonomy over a particular district where, among other things, a uniform, indigenous educational curriculum will be imposed. The minimization of both cultural differences and ethnic labels was fully legitimate in the 1940’s to 1950’s: the deconstruction of colonial categories represented a true progress of knowledge by rendering more correctly the large degree of independence between political structures and cultural forms. Fifty years later, the tools designed by Leach remain largely efcient, on condition that they are reshaped in order to take into account new political contexts, in which the political construction of identities determines a closer dependency between culture and social structures.

NAGA ETHNOGRAPHY AND LEACH’S OSCILLATORY MODEL OF GUMSA AND GUMLAO Pascal Bouchery In Political Systems of Highland Burma Edmund Leach interpreted the concepts of gumlao and gumsa as political models in Kachin society. In this chapter, I will question Leach’s attempt to establish his oscillatory model as a general theory of social change applicable to segmentary societies in this part of Asia by comparing it with neighbouring societies of the Indo-Burmese border, especially the Nagas. I will argue here that, while some of the Naga systems can be viewed as gumsa-like organizations, a model such as Leach’s gumlao as dened in Political Systems cannot be found anywhere in the Naga Hills, according to the ethnographic material collected to date. This assertion, which corroborates F. K. Lehman’s ndings about the Chins of Burma (1963), casts some doubt on the validity of Leach’s oscillatory model. Leach started the concise comparison he made of Kachin and Naga modes of governance in Political Systems by remarking that the Naga systems, like the Kachin’s, were characterized by the existence of two contrasting forms of village government. The Sema Nagas, for example, with their powerful hereditary chiefs, could easily be contrasted with the Angami Nagas, whose villages were described in early ethnographic reports as being run on a more ‘democratic’ basis. However, Leach went further than this by equating the political organization of the Semas with the gumsa model, while considering Angami political organization to be a gumlao-like model.1 However, in so doing, the author of Political Systems seems to have been misguided by the colonial sources he was using. This can be demonstrated in a number of ways. First, and most obviously, at the time of British political expansion in the region, ‘autocracy’ and ‘democracy’ were western concepts applied to the Nagas and, as such, did not have equivalents in local languages. Second, ethnographic works carried out under the British regime in Assam were undertaken rst

1

1972 [1954]: 232.

110

pascal bouchery

by ‘soldier ethnographers’, such as Colonel R. G. Woodthorpe and Captain John Butler, and then by ‘administrator ethnographers’, like J. H. Hutton and J. P. Mills.2 As government ofcers observing tribal modes of government, their attention naturally focused on the operation of secular power, particularly on local leaders who exerted political inuence and were able to help the British Government as political intermediaries in the enforcement of colonial law. Groups in which such prominent gures could be found, such as the Sema and the Chang, were inevitably labelled ‘autocratic’, while the vast majority of tribes, such as the Angami, Lhota, Rengma, Ao, Sangtam, Tangkhul, Zemi, Kabui (Rongmei), Maram, Mao and Konyak of the Tenkoh group, were considered ‘democratic’.3 A good example of this peculiar blend of ethnography and colonial administration is provided by Mills who, as Ofciating Deputy Commissioner of Assam, wrote in 1922: Turning to the polity of the village, different tribes have very different customs. Among the Semas a system of hereditary chiefs exists; (. . .). The Changs have a system of chiefs very like that of the Semas (. . .). The Konyaks too have hereditary chiefs in the Thendu section of the tribe, though not in the Tenkoh division (. . .). On the other hand, the Ao and the Tangkhul villages are governed by bodies of elders representing the principal kindreds in the village, while the Angami, Rengma and Lhota and apparently Sangtam villages are run on lines of democracy, a democracy so extreme in the case of the Angami that, in view of his peculiar independence of character, it is difcult to comprehend how his villages held together at all before they were subject to the British Government.4

Naga Political Systems—An Overview Though mainly drawn for pragmatic reasons, this dichotomous approach towards supposedly ‘democratic’ or ‘autocratic’ systems has never been questioned, even by anthropologists such as Fürer-Haimendorf (1936, 2 The two authors wrote monographs on the majority of groups living in presentday Nagaland. 3 For the Angamis, Lhotas, Rengmas, Aos, Sangtams: Hutton, 1921a: 353, Mills, 1922: XXIII; for the Tangkhuls: Brown, 1874, Watt, 1887, in Elwin (ed.) 1959: 456, 469; Mills, 1922: XXXIII; for the Zemis, Lyangmeis and Rongmeis (formerly Kacha): Mills, 1926a: 28; for the Maos and Marams: Hutton, 1921a: 353; for the Konyaks (Tenkoh group): Mills, 1926a: 28. 4 Mills, 1922: XXXIII.

naga ethnography

111

1939, 1969), and Leach himself entirely relied on it in Political Systems. Not surprisingly, however, it does not correspond to the Naga conception of power, which is deeply embedded in religious beliefs and does not make a distinction between political and religious spheres. Different means of gaining social respectability and inuence based on individual qualities are socially valued among the Nagas: charisma, success in hunting, headhunting or war, and social ascent through the accomplishment of a codied series of Feasts of Merit, for example. However, in many cases individual success is ultimately related to luck, and luck itself is generally derived from connection to some magical source of power, be it through the possession of luck-stones (micrometeorites) or by the invocation of deities through ritual action. Therefore, in traditional society, the most respected form of authority is conceded to those people who can manifest by the efciency of their ritual action that they are favoured by the gods (especially those controlling cosmological elements) and who, as such, stand as mediators for the benet of their own communities. If we take into consideration the religious basis underlying such basic political concepts as ‘inuence’, ‘merit’ and ‘renown’, I suggest that we may also consider them valuable tools in the interpretation of Naga political systems. By so doing, we observe that the institution of village-based, hereditary chieftainship, which can tentatively be called ‘ritual hereditary chieftainship’, has existed everywhere in the Naga Hills, the chiefs holding their title by virtue of direct descent from founding ancestors, and their political power being backed by supernatural sanctions.5 The following can also be seen to be true of all chiefs of this kind. 1) As representative of the village founder’s kin, the chief (Angami Tevo or Kemovo, Lhota Ekyung, Tangkhul Avunga, Zemi Kadepeo, Kabui or Rongmei Matai ) is considered the sole ‘owner’ of the village territory. Ownership here is not to be taken in the Western sense. Rather, it means that he and his lineage or clan hold exclusive territorial rights because their ancestor, at the time of the foundation of the village,

5 For the Angamis: Fürer-Haimendorf & Mills, 1936; For the Maos (Memis): Watt, 1887, in Elwin (ed.), 1959: 461; for the Zemis, Lyangmeis and Rongmeis (formerly Kacha): Soppitt, 1885: 9; Bower, 1951: 79; for the Tangkhuls: Horam, 1977; for the Lhotas: Mills, 1922: 96; for the Konyaks: Mills, 1926a: 28; Fürer-Haimendorf, 1969: 62–63; Hutton, 1965: 23.

112

pascal bouchery

set up an alliance with the true ‘owner’ or ‘master’ of the area, some genius loci, who accepted human settlement in exchange for an annual cult. This does not give the chief the right to dispose of the land at will, although he may sometimes intervene in its distribution, for instance by reallocating cultivable land to landless villagers. The chief ’s function and title are transmitted along the patrilineage, most often from father to eldest son. The chief, together with the village council, which consists of representatives from each clan of the village (often the heads of the eldest families), handles the overall affairs of his community. He and the village council constitute the village court, and all disputes pertaining to land, theft, divorce, inheritance, fornication, adultery, murder, etc., are heard and tried in it. The chief may or may not be the most inuential person in that assembly, but he usually presides over it. The chief ’s main functions are ritual. The ritual action of the chief, expressed through the performance of collective ceremonies, is mainly devoted to village welfare, but also to village safety by ensuring that it is protected from headhunting raids. The ritual action of the chief is seen as the perpetuation of the actions of his ancestors, who made human settlement possible through the alliance rst established with protective deities. This is the condition sine qua non for the community’s survival and perpetuation. Through the supposed efciency of his ritual action, the village founder’s kin is responsible for village welfare and he is perceived, therefore, as also holding the village’s destiny in his hands. For this reason, the chief himself is also frequently assimilated to a principle of fertility, such as Tevo amongst the Eastern Angami, and, as such, subject to all kinds of prohibitions in order to preserve his integrity and ritual efcacy.6 As a token of gratitude and respect for his benevolence and status, the villagers help to build and repair the chief ’s house and cultivate his elds. He also receives as an exclusive privilege the thigh or foreleg of any four-footed animal that is killed.

2) 3)

4)

5)

The set of principles upon which the institution is based is remarkably uniform throughout the Naga Hills, but the perception of it by Western

6

Fürer-Haimendorf & Mills, 1936.

naga ethnography

113

researchers has been biased by the disproportionate importance attached to the performance of political power by colonial administrators. While the institution has been very stable over time, the secular power or inuence actually attained by Naga chiefs has uctuated much more. As a result, contradictory reports are common in the colonial sources. For example, military ofcers of the nineteenth century reported that Lhota, Rengma and Kacha hereditary chiefs exerted considerable political power; little of this remained at the beginning of the twentieth century when the ethnographies of the three groups were written, and at which time their political organization was classied as ‘democratic’. Furthermore, prior to the Second World War the Semas and the Changs were considered to be ‘autocratic’ groups because some of their villages where run by powerful chiefs. Yet, at the same time, their biggest communities were organized, as in other groups, in a way that was considered by ethnographers to be ‘democratic’. Naga ethnography is very rich and extends over approximately one and half centuries, which allows for some historical perspective. Interestingly, colonial sources point to a general correlation between the secular power of chiefs and the process of territorial expansion. This is well documented for the Semas and the Changs. They were the last of the Naga groups to colonize vast areas of uninhabited land in the central part of the Patkai Range bordering Burma, thus enabling the process of colonization to be observed by British administrators.7 In pioneering areas where newly founded villages were most numerous, the representative of the village founder’s kin, or the village founder himself, acted as an undisputed leader, while the principle of ultimogeniture ruled the transmission of chieftainship. As in the Kachin context, elder brothers had to choose between remaining subordinate to the youngest brother in their village or becoming themselves chiefs by founding a new colony, which, by that time, was rather common. In such cases, the leader or Akekao, who was the actual founder of the colony, was regarded as owner of the land and, thus, received the noble part of big game and beneted from free labour on his lands from the 7 Hutton, 1921b: 3, 4, 8. The process of territorial expansion among the Nagas has been documented by Hartwig (1970) from a Marxist perspective. The author estimates that ‘land territorial and economical appropriation’ (Territorialöknomischen Erschlissung des Landes) was mostly completed in what constitutes present-day Nagaland around the beginning of the nineteenth century, but occurred later in the highest and less exploitable zones of the Naga Hills, where the Semas and Changs now live (1970: 44–46, 222).

114

pascal bouchery

other villagers.8 He had the nal word in the village council and can be considered as taking decisions in the fullest sense of the word, assuming military responsibilities in case of conicts with enemies. His relation to other villagers was seen as one between father and adopted sons (they called him apu, ‘father’; he called them mughemi, ‘orphans’). However, when all available land had been divided between communities in a given neighbourhood, the political situation appeared quite different. First, elder sons started to contest the undivided rights of the youngest, claiming their share of land and trying to be recognized as legitimate successors. Each of them had his own faction of followers or ‘adopted sons’ who called him father. This process was also accompanied, according to Hutton, by a gradual change in the rules of transmission of chieftainship from ultimogeniture to primogeniture.9 Later on, other inuential people who were not of the chief ’s kin could in turn enter into an apu/mughemi relation with some villagers. As Mills remarked: “The result is a splitting up of rights to the point of extinction, and a general levelling up all around.”10 By the turn of the twentieth century the ‘true’ autocratic pattern existed only east of the Tizu river, where Semas were still migrating in numbers at that time. In areas of older occupancy, such as the Dayang and Kileki valleys, the political inuence of chiefs was largely tempered by other people of the community who merited such respect (warriors, donors of Feasts of Merit, charismatic personalities), and Mills provided the following outline in 1926: “There are no real chiefs at all, and the tendency is for the villages to be run by such people as have wealth and inuence, and can shout loudly.”11 The situation was very much the same among the Changs. Around 1936, only in two newly founded villages, Yamrup and Yonyimti, did chiefs exert real political power, while factionalism prevailed in older ones.12 Elsewhere in the Naga Hills, this process seems to have occurred earlier, and could still be remembered among some ‘democratic’ groups such as Zemis, Rengmas and Lhotas. When the British rst encountered them, the Lhotas had powerful village chiefs, or Ekyung, whose function

8

Hutton, 1969 [1921b]: 147–148. Ibid.: 148–149. 10 Hutton, 1926a: 28. 11 Ibid. 12 For the Zemis or Kacha Nagas: Soppitt, 1885: 9 –10; Bower, 1951: 71. For the Changs: Hutton, 1965: 23; Mills, 1926a: 28. 9

naga ethnography

115

was always transmitted along the patriclan. They organized warfare, managed overall village affairs with the village council, and usually received, among other privileges, the biggest share of loot acquired from raids. However, little of this remained some fty years later when J. P. Mills wrote a monograph on this group, published in 1922. By this time, the chieftainship institution had vanished to such a point that the title name itself, Ekyung, was hardly heard. In some villages, people still remembered that in the past it was customary for the chief to receive free labour service from his co-villagers on his lands, but afrmed that this right had ceased to be asserted several generations ago.13 Cases of the complete disappearance of such chiey privileges had been reported among the Semas over one or two generations.14 GUMSA or GUMLAO? So dened, Naga village-based political organization appears to have been neither gumlao nor gumsa. The so-called ‘democratic’ system of ritual hereditary chieftainship found in most Naga groups cannot be equated with the gumlao model since, according to Leach, the absence of hereditary chieftainship is an essential characteristic of this model. Similarly, the Sema organization that Leach assimilates to the gumsa model is not based on a sharp distinction between aristocratic and non-aristocratic lineages, but on the relationship between one chiey individual and his followers and which does not give rise to any durable stratication of lineages. It also remains a village-based structure, for, while several communities are linked for defensive purposes by bonds of mutual assistance, the chief of the protecting community is in no way considered the ‘owner’ of a territory that would encompass the totality of allied village lands, as would the Kachin paramount chief. It is true that both models are ambivalent to some extent. A gumsa community need not be hierarchical, and the gumsa chief need not be autocratic. What set the chief and aristocrats apart among the Kachins, as well as among the northern Chins, were symbols of social position, not material advantages. Nonetheless, the essential distinction between two social strata based on lineage afliation remained, and these have classically been named in the literature ‘aristocrats’ and ‘commoners’. 13 14

Mills, 1922: 96. Mills, 1926a: 28; Hutton, 1965: 23–25.

116

pascal bouchery

Furthermore, gumsa organization comprised not only a village-based, but also a territorial-based unit, including a cluster of neighbouring villages organized into a pyramidal structure: every village had its own chief or headman, but the chief of one of them would also be considered the paramount chief of a politico-territorial domain (mung) comprising the lands and people of all allied villages. In other words, gumsa organization included two distinctive attributes of what anthropologists call a chiefdom. Chiefdom organizations do exist among some Naga tribes, but, very surprisingly, they were ignored by Leach in his attempt to extend the gumsa-gumlao dichotomy to neighbouring societies. The Konyak, Wancho, Nocte and Tangsa groups, who occupy the northernmost part of the Naga Hills, have both stratied lineages and paramount chiefs (Ang, Lowang, Lungwang respectively) whose authority extends over a territorial domain consisting of a group of allied villages, and often behave like true autocrats. In striking contrast with other groups, their political power is in no way related to a process of territorial expansion, as it was in the Sema or Chang cases. Neither is the chief in these groups seen as the representative of the village founder’s kin: wherever the latter is found, he most often acts as an advisor of the chief, remaining subordinated to his authority.15 Parallels with the Kachin gumsa organization described by Leach and the chieftainship organization of northern Chins as reported by Lehman (1963) are evident. For example, a major distinction separates chiey clans (called Wangham among the Konyak Thendu) from commoner clans (Wangpen). Marriages of men of commoner status with women belonging to Wangham lineages or clans are prohibited, while men of chiey clans can take wives of the commoner stratum as secondary spouses. Such unions are responsible for the existence of several minor aristocratic clans (Wangsu, Wangsa among the Konyak Thendu). Village headmen are invariably members of aristocratic lineages and clans (major or minor), while paramount chiefs are recruited from chiey families of the Wangham stratum only. Such a political structure is, as in the Kachin gumsa case, a chiefdom in line with the model conceptualized by Sahlins as a ‘conical lineage chiefdom’, that is, one in which rank is determined by genealogical distance from the chiefdom founder or

15

Dutta, 1978: 92, 195, 206–207.

naga ethnography

117

ruler. As among the Kachins and Chins, in daily life the basic distinction between aristocrats and commoners is a matter of symbolic privilege more than economic advantage, and headmen are not necessarily the most inuential men in their communities. Even the authority of paramount chiefs can be tempered by the village council, or effected only in their own ward or clan, as in the village of Wakching studied by Fürer-Haimendorf.16 Economic goods are accumulated but largely redistributed through collective ceremonies. Oral traditions collected so far among northern Naga groups unanimously attribute an external origin to the chiefs’ families, and, thus, by extension, to the social stratication as it exists today. Most local legends recount a process by which a chief was acquired from another village, either by submission or what is often more advantageously presented as an ‘invitation’. As to its ultimate origin, the Noctes are quite explicit in connecting their chiefs’ genealogies with Taï invaders from Burma: The rst chief came from the Hukong valley of Burma. His name was Khunbao. With a few attendants he crossed the Patkai hills and followed a very difcult route to a place called Sajuok, near the present Laptang village. The inhabitants of Sajuok, whom he subdued, accepted him as their Chief. He had two sons Khunlung and Khunlai. They were succeded by Tangthok and Tankam. The present chiefs of Namsang, Borduria and Laptang claim descent from Khunbao.17

This story clearly refers to the origin of the Ahom State whose founder Sukapha, a Shan prince coming from Mong Mao in Upper Burma, succeeded in establishing a small kingdom in the Brahmaputra valley by defeating the Chutias in 1228. Khunbao is a Shan title for ‘Prince’ that primarily applies to the ruler of a realm or princedom (möng). Historically, the lands of the Konyaks, Wanchos and Noctes were situated in the immediate vicinity of the three successive capitals of the kingdom, Abhaipur, Choraideo and Garghaon, and the Noctes of the lower area traditionally exploited salt mines, which were of strategic interest to the Ahoms. Not surprisingly, the Ahoms claimed control over the Naga salt. When they set up their capital at Garghaon in 1253, there were several clashes with the Nagas over the collection of salt from the salt licks; these happened to lie in Naga territory. However, the Nagas were nally subdued and the Ahoms established their political

16 17

1969: 63. Dutta, 1978: 190.

118

pascal bouchery

domination over the foothills; this rst took the shape of a tax imposed on the villages exploiting the salt mines.18 Later, political relations became much more intricate, although this was apparently to mutual advantage. The salt mines were under the indirect control of the Ahoms, and the salt manufactured in the low hills continued to be shared between the State and the exploiting villages until at least 1819. The Nagas owing allegiance to the Ahom Kings were claimed as subjects of the State, and, as such, beneted from military protection. In this way, the Ahoms managed to secure a buffer zone and prevent raids on their capitals by uncontrolled Naga groups. Economically speaking, submission led to some substantial advantages for the Naga groups: several of the chiefs received grants of irrigated lands (khat, reported in the Royal Court Chronicles, the Buranji, as Nagakhat ) and of shing waters (bheel ) on the plains. Not only did the Ahom kings renounce their right to collect taxes from them, but they also allowed the Naga chiefs to levy taxes for their own benet. This constituted an important source of income, and special administrative agents (katakis) were appointed for the purpose.19 Politically speaking, these Naga chiefs were treated not only as tributary allies, but were also conceded privileges and titles of the Ahom aristocracy. Furthermore, those having received grants of lands also enjoyed assignment corvee labour ( paik) like the ordinary Assamese nobility.20 The Buranji mention at least one occurrence of an Ahom princess being married to a Naga chief, reecting a common way of contracting political alliances with other princedom courts or States in a Tai-Shan context.21 Also reported in the Royal Chronicles is the King’s habit of going hunting with ‘allied Khunbaos’, which is a clear indication that honoric titles normally restricted to the ruling aristocracy were conceded by the Ahoms to some Naga chiefs and were used by them in ofcial documents.22 18 According to Elwin (1959) taxes claimed by the Ahom kings from their Naga ‘subjects’ consisted of slaves and elephant tusks. 19 The right conceded to some tribal communities to receive payments from specied villages in the foothills, provided the payment of annual tribute to the Ahom king, is known as posa, and was applied elsewhere in the foothills surrounding the Brahmaputra valley, for instance in areas inhabited by the Nishis. 20 Mackenzie, 2001: 91. 21 Bhuyan, 1933: XXX, Condominas, 1980: 269; Leach, 1972 [1954]: 253. For instance, political relations between the Ahom and the Manipuri were tied up with matrimonial relations and princess exchanges. 22 Bhuyan, 1933: 195.

naga ethnography

119

The result of this is very similar to that which Leach reported for the Kachins: an emulation of the Shan princes by Naga chiefs. When rst encountered by British ofcers, all Nocte chiefs designated themselves as Khunbao or Sofa (= tsao pha, a political title of Taï origin normally applying to the ruler of a möng).23 Their chiey genealogies were all ultimately connected to Ahom sovereigns, and the legends associated with the foundations of their lineages incorporated cultural elements borrowed from the Ahom, such as the king’s title (Swargadeo) and the gure of the ruler descending on earth by a silver or golden ladder.24 Among other cultural elements pointing to Taï origin are the use of thrones as specic privileges of chiefs, common in all northern groups but unknown elsewhere in the Naga Hills (most elaborate ones, decorated with umbrellas, are found among the Noctes, who claim their use is a legacy from the Ahom kings), as well as cock ghts as part of the chief ’s installation rituals.25 The military support of the Ahoms had considerable impact on traditional political life in that part of the Naga Hills, especially after the Ahoms were taught the use of gunpowder following the unsuccessful intrusions by the Mughals during the reign of Suhunmunga (1497– 1539). Allied Naga villages were provided with rearms for defensive purposes, sometimes with heavy artillery as in the case of Borduria who received two canons from the Ahoms.26 Following this, the Nagas established in the foothills rapidly took control of the gun trafcking in the entire hill region. As reported in the Buranji for 1665, when allied chiefs called on military help from the royal army, it could result in the destruction by cannon of entire villages.27 Most oral traditions of the northern Nagas in the lower region recount the arbitration of the Ahom king in conictual successions to the role of chief, though no mention of such a direct intervention can be found in the Buranji.28 This backing by the Ahom Court of northern Naga hereditary chieftainship in the foothills rapidly turned to the advantage of a few small villages, such as Borduria, Namsang and Laptang, whose ruling families managed to take over the political control of the entire northern

23 24 25 26 27 28

Butler, 1847: 155; Dalton, 1872: 41. Dutta, 1978: 193–194, 283; Gait 1926 [1906]: 286. Dutta, 1978: 80, 197; Fürer-Haimendorf, 1969: 59; Srivastava, 1973: 12. Dutta, 1978: 9. Gait, 1926 [1906]: 156. For instance Dutta, 1978: 205–206.

120

pascal bouchery

hilly area. Traditionally among the Nagas, the widespread institution of headhunting resulted in the constitution of small, political realms uniting several villages for defensive purposes. Most often, a powerful and heavily populated village provided military protection, while at the same time exercising a right of overlordship over tributary villages in the wider territory. However, and insofar as colonial sources may be considered reliable, the size of such political realms rarely exceeded fteen or so communities. Nonetheless, Borduria and Namsang each comprised some sixty houses by the turn of the twentieth century, and extended their political control over thirty ve and fty or more villages respectively, some of them in Burma, all believed to have been subjugated by force.29 It is doubtful that such a favourable situation could have taken place without any external help, as some of their subordinated villages, such as Laju, already comprised more than three hundred households by that time. More likely, the control of the rearms trade in the hands of Borduria and Namsang since the sixteenth century played a decisive role in what can be viewed as a process of political expansion. This process was consolidated in the political realms of northern Naga groups by the placement of an aristocrat. This man would act as chief in every newly subjugated village and would be close kin of the paramount chief, related in most cases through a collateral branch (such as a paternal nephew or cousin), or of an inferior status (such as the son of one of his commoner spouses). This customary rule points to the probable origin of social stratication amongst the northern Nagas as being those families of chiefs appointed or backed by the Ahoms. In the case of the Noctes, at least, there is little doubt that the Ang aristocracy as we know it today spread from the villages of Laptang, Borduria and Namsang, whose ruling families controlled the salt extraction for the benet of the Ahom kings. Conclusion The northern Naga system has many similarities with Kachin gumsa organization, as well as with the northern Chin chiefdom systems (Haka, Zahau, Sukte Kamhau, Sailo, Tashon, Zothung, and so on). It can be considered without hesitation a gumsa-like political model in the

29

Dutta, 1978: 191–198.

naga ethnography

121

sense that it emerged historically from the attribution of special titles, functions and privileges by Tai (or, in the case of the Chins, Burmese) lowland centralized powers to some tribal chiefs, and attempts by the latter to emulate their rulers. Such an historical development seems to have been fairly common in continental Southeast Asia, especially in areas politically dominated by the Tai, as similar processes have been identied so far among the Achang, the northern Chin, the Karen, the Lawa, the Muong, the Palaung, and possibly others.30 On the other hand, a gumlao model, such as dened by Leach, does not exist among the Nagas, whose political institutions are always centered upon the prominent gure of a hereditary leader or chief, even in groups where lineage stratication is absent. Indeed, the gumlao model even appears atypical in the region for at least two reasons. First, despite Leach’s assessment, it remains historically isolated, being restricted to the Triangle Region in what is today northern Burma, and the Jingphaw and Atsi groups of the ‘Kachin’, and is said today to be virtually extinct.31 Second, the gumlao model does not t into the general framework on which chieftainship is based in both Naga and Kachin societies. Leach says that the thigh-eating chief must be a member of a lineage considered ‘older’ than any other, and is said to ‘own’ (madu aï ) the village territory (lamu ga, literally sky and earth). As such, he receives the thigh of all four-footed animals killed on hunts or during collective ceremonies, as well as free labour service from the villagers on his lands, including help to build or repair his house without reciprocal obligation. Most important among his privileges and attributes is the monopoly of the cult of the Madai nat, the guardian of village welfare, which is seen as a distant relative of the chief and whose shrine is kept in the chief ’s house. All of the above makes the Kachin duwa an equivalent of the Naga ritual chief, since the basis of chieftainship in both societies lies in the essential link between the representative of the village founder’s lineage and the village guardian spirit, which provides the thigh-eating chief ’s legitimacy. This model of the chief as being not only owner and distributor of land, head of his community and commander in war, but also high priest, responsible for offering sacrices to the territorially rooted guardian deities whose 30 Leach, 1972 [1954]: 85; Lemoine, 1978: 867–68; Lehman, 1963; Marshall, 1922: 127–129; Steinman & Sanidh Rangsit, 1939; Robequain, 1948: xvi; Condominas, 1980: 264; Lowis, 1906: 21–22; Cameron, 1912: xxxiii–xxxiv 31 La Raw, passim.

122

pascal bouchery

good will was believed to be necessary for prosperity, certainly applies for the Chins too, as seen, for instance, in Khuahrum cults. In a gumlao community, according to Leach, the lineage heads collectively ofciate in the cult of the nat associated with the foundation of the village, but never organize a sacrice to the Madai nat. This system, which denies any individual, exclusive access to the tutelary deities, is quite coherent with their absolute rejection of any status differentiation between lineages. However, it should be noted, it is also somewhat aberrant in the regional context where at least a kind of ritual prominence is conceded to the descendant of the village founder to the detriment of other lineages, especially those of later immigrants. So, in a way, the gumlao model is indeed revolutionary, and that is precisely how the rst British observers saw it: a localized, rebellious movement that had emerged as a reaction against the tyranny of some Kachin chiefs. Leach rejected this view, arguing that social change resulted from the structural instability of both egalitarian and hierarchical models, and suggested it must take place indenitely in a cyclical, non-linear way. This interpretation of Kachin political dynamics is commonly known as the ‘oscillatory model’. The Nagas and the oscillatory model If we admit that there is no equivalent of Kachin gumlao organization among the Nagas in general, we must now turn our attention to the Konyak system, for, in the context of Kachin society, and following Leach’s analysis, the gumlao system can be seen as an evolutionary byproduct of the gumsa model, which is structurally unstable. According to Leach, the very dynamics of traditional Kachin society lie in its tendency to oscillate between an ‘aristocratic model’, a form of organization under powerful chiefs which comes close to the Shan ideal of ruling princes, and a ‘democratic model’, a form of organization in which the claims of such dominance were rejected. Simply speaking, when people no longer saw advantage in the chiey system, they revolted. Historically, the change from gumsa to gumlao is supposed to have occurred when a sufcient number of people withdrew support from the former system, especially when the chief began to behave like a prince. Do we nd something equivalent among the Nagas? At rst sight, an opposition similar to that of the gumsa-gumlao system seems to exist among the Konyak Nagas studied by Fürer-Haimendorf (1939, 1969) through the distinction made by them between Thendu and Tenkoh

naga ethnography

123

groups. The two are easily distinguishable by their appearance and ornaments. While the Thendu group conform to the general model of chiefdom described so far among northern Nagas, in the Tenkoh group villages are more simply run by the village council. The Ang is only one of its members, and political power is often fragmented between sections of the village, wards (khel ) or bachelor’s dormitories (morung), each of them being free to contract separate political alliances with allied khel, morung or villages. As in the Kachin gumsa system, there are several reasons why the power of the Great Ang is unstable. 1) The Ang’s domination is legitimated by the general belief that his ritual action in relation to tutelary spirits will ensure good crops, peace or success in war. However, being subordinated to the preservation of village welfare, the chief ’s action always remains at risk of being contested in case of a series of failures. 2) The Ang’s domination is based on an unequal distribution of cultivable land and the monopoly he can assert over certain Feasts (as in the case of the manao monopolized by the Kachin duwa). Such ceremonies strengthen the relationship between the chief and tutelary spirits and convert the surplus collected from tributes and collective labour into social prestige. However, this monopoly can always be questioned, as, for instance, in cases where a shortage of land or grain affects the redistribution process. Furthermore, as Leach has shown in the Kachin context, the Ang can also be challenged by the emergence of leaders of smaller sections (hamlets, wards, bachelor dormitories), who might not openly contest the chief ’s supremacy but may start themselves behaving as true chiefs, albeit on a smaller scale, receiving from their followers the noble part of game and presiding over religious rituals for the welfare of their section. The recent history of many villages in the Tenkoh group provides several instances of either the decline of inuence of chiey lineages or the destitution of the chief. Some of them seem to have followed the fall of the Ahom kingdom and the subsequent Burmese invasion at the end of the eighteenth century. In Wakching, for example, a long lasting quarrel opposing aristocratic and commoner wards was nally settled by calling a chief from the powerful village of Chi. Nevertheless, the new chief never succeeded in imposing his authority beyond his own lineage over other morung, which remained virtually independent. In

124

pascal bouchery

Wanching, the Ang was simply deprived of his rights and privileges for having had an incestuous relationship. He was replaced by a member of the same lineage, who was designated not by himself, but collectively, by the villagers or their representatives.32 Extreme circumstances may even have led to true rebellions, as in Chen, where the villagers massacred the Ang and part of his family. Despite this, Naga ethnographic sources provide no indication of something resembling the shift from a gumsa-like system to a gumlao-like one in Leach’s sense. For an essential difference with the Kachin context lies in the fact that both groups, Thendu and Tenkoh, distinguish between aristocrats and commoners on the basis of descent, even if the distinction in Tenkoh villages is of minor importance from a practical point of view. As Fürer-Haimendorf notes in the case of the Tenkoh village of Wakching: The status differences between aristocrats and commoners was never disputed by the commoners who form the majority of the people of Wakching (. . .). They conceded without hesitation the right of the members of chiey clans to marry several wives, while accepting that monogamy was mandatory for commoners.33

In other words, rebellion, dened as the attempt within a society to disrupt the status quo and redistribute power and resources, does not imply structural change as the shift from gumsa to gumlao does, according to Leach. The northern Naga political systems rather parallel the situation prevailing among the northern Chins, among whom, due to the general recognition of status differences between lineages, Lehman could not identify a ‘true’ egalitarian model of the Kachin-gumlao type: Nowhere does anyone dispute the fact that bawi rank carries special privileges. People only question the basis for assigning rank, and even democratic headmen are generally of bawi rank.34

In contrast, among the Kachin gumlao, many villages were not only run exclusively on a village basis by headmen and councils of elders, but were composed entirely of commoners. Leach was absolutely right when stressing the intrinsic instability of the gumsa-like systems, and the fact that political units within Kachin

32 33 34

Fürer-Haimendorf, 1969: 57, 63. 1969: 63. Lehman, 1963: 141.

naga ethnography

125

society were in a continual state of ux. However, the gumlao model appears isolated, too, in the history of this region, as structural shifts of the gumsa-gumlao type seem to be rare or absent in neighbouring societies, such as the Nagas and the Chins, who also possess gumsa-like political organizations. Thus, the exact role of gumlao organization in Kachin political dynamics remains to be evaluated.

INTERETHNIC SYSTEMS AND LOCALIZED IDENTITIES: THE KHMU SUBGROUPS (TMOY ) IN NORTH-WEST LAOS Olivier Evrard Note For Lao place names, there is no ofcial transcription and spelling can vary greatly from one document or place to another. For names of provinces and districts, I have used the transcription of the 1995 national census, and for other names, a transcription based on English pronunciation. For the English spelling of Thai place names, I have used the ofcial transcription taken from the Romanization Guide for Thai Script (Royal Institute 1982). Unless specied, all translations from French documents are my own. Transcriptions of Khmu and Lao words follow phonetic English pronunciation, and do not take into account tones, vocalic length or consonant height. I would like to thank François Robinne, Guido Sprenger, Tim Wong and Mandy Sadan for their reading and comments on early versions of this paper.

Introduction In a short article published in 1978, Robert G. Cooper radically denied the relevance and usefulness of Leachian ideas and concepts for a comparative anthropology of Southeast Asia. According to Cooper, all attempts to apply a Leachian style of analysis outside of the Kachin Hills had “failed to uncover either social categories or mechanisms of change similar to those described by Leach” and had led to “false conclusions on the nature of ethnicity and interethnic relations in the area” (1978: 56). If Cooper was right, there seems little point in questioning the topicality of Leach’s framework fty years after Leach rst published Political Systems of Highland Burma. However, despite Cooper’s critique, I see at least two good reasons to re-engage critically with the Leachian model. First, Cooper reviews—sometimes mistakenly—the work of various scholars who have worked in Thailand and Laos, but he does not provide us, at least in this article, with personal data gathered in the eld. Following decades of social and political instability, most countries in Southeast Asia are now accessible again to researchers. This provides an excellent

128

olivier evrard

opportunity to carry out eldwork and gather rst hand observations, to revisit and revise a comparative anthropology of Southeast Asia. While Leach’s ‘oscillatory’ model is now widely criticized, both within and outside the Kachin Hills,1 a second and more important reason to revisit Leach’s framework is that his analysis of Tai cultural and political inuence among highlanders is still relevant. Leach was one of the rst authors to identify the ‘duality’ of Taization processes:2 the cultural inuence of the lowlands resonates within highland groups, but the latter use their own cultural resources and systems of values to transform this lowland culture into practices, behaviours and/or ideologies. In other words, Taization does not always lead to assimilation, but may also lead to the perpetuation, transformation, and even the increase (through fragmentation into subgroups, for instance) of ethnic boundaries. An example to illustrate this point is that, despite the long period of contact with Tai culture, Mon-Khmer groups of northern Thailand and northern Laos not only continue to hold specic identities, but also are distinctively known through various localized ethnonyms, which directly express their old relationships with the lowlands. This chapter focuses on the history of interethnic relationships and social change amongst Khmu villages (Mon-Khmer linguistic afliation) of the Nam Tha valley, in northwest Laos. It uses three kinds of data: accounts of the area by colonial explorers or administrators, ethnographic studies, such as those of Karl Gustav Izikowitz (and recently of Guido Sprenger) on the Lamet, or of Damrong Tayanin on the Khmou, and my own personal eld data collected since 1994. My purpose is not to demonstrate at all costs the validity of a Leachian model for another Southeast Asian ethnic group. Rather, I have used the model to facilitate my own understanding of the social reality and political history of the villages where I worked while researching my doctorate (Evrard 2001). In essence, I have shown that such a model holds heuristic value, through helping to nd relevant facts and, through comparison, avoiding false interpretations. What I demonstrate in this paper is that the differences between Khmu and Kachin relate mainly to their different interethnic contexts,

1 See, for instance, the radical critique developed by Zusheng Wang (1997). For neighbouring groups, see the work of Pascal Bouchery on the Naga (in this volume). 2 I borrow this term from Grant Evans (Evans 1991).

interethnic systems and localized identities

129

particularly their contrasting history of relations with lowland populations. Despite these differences, I argue against Cooper’s critique, and in support of the continuing topical value of the Leachian framework. I assert that Leach’s analysis allows a better understanding of how the history of relations with lowland populations has created cultural and political discontinuities in highland villages, which otherwise claim a common identity. I support these ndings through investigating the origin and characteristics of the numerous Khmu tmoy (subgroups), and show how these intra-ethnic solidarities still inuence local social and political dynamics. External Mechanisms of Social Change in the Nam Tha Valley: A Brief Overview Comprising eleven percent of the total population of Laos, the Khmu are the largest minority in the country and in some areas constitute overwhelming local majorities (see map 1). This is especially true around the Nam Tha valley in northwest Laos, where they are in contact with mainly Tai populations (Lao, Lü and Yuan) and other small Mon-Khmer groups, particularly the Lamet and the Samtao. In Nalae district (the focus of this study), Khmu villagers constitute nearly eighty percent of the total population, even without considering the mixed villages where some of them live alongside Laos people (see map 2). Khmu villages in this area share many economic and social characteristics with the Lamet dwellings of the west bank. Villages have an average size of approximately 250 people and 40 households, mainly located in upland mountain areas between 800 and 1200 meters. Most upland villages rely exclusively on slash-and-burn agriculture (with 8 to 15 year fallow periods) for their subsistence, and have periodic rice surpluses, which they barter with the neighbouring Tai populations for manufactured products. The kaang (house), klork (local patrilineages), ta’ (clans), and kung (village) constitute the basic social units of Khmu villages, with clan membership playing a mostly symbolic role (Lindell et al., 1979). From the villager’s point of view, there are no real, absolute differences in status between lineages of a locality. The only exception concerns the ritual status of the lineage of the founder of the village, whose male elder has the title of lkun (priest), and conducts the rituals for the village spirit. While differences of status can exist between individuals,

130

olivier evrard

Key

NORTH

international border provincial border district border main roads and tracks Nhot Ou

Mon-Khmer populations > 43 % Mon-Khmer populations > 73 % (from Taillard & Sisouphanthong 2000 : 35) OUDOMXAY provincial centre Houn district centre

PHONGSALY Samphanh

Mekong tributaries 50

0

CHINA

Booneua

100 km

Boontai Sing Khua

BURMA

Namor Long

Meung

LUANG NAMTHA

La

OUDOMXAY

Viengphoukha Nalaè

m Na

HUOIXAI Paktha

Na mB

eng

Tonpheung

Pha Oudom m Na

a Th

O

u

Beng

Nga Houn

THAILAND Pakbeng

LUANGPRABANG

c Olivier Evrard, 2005

Map 1: Mon-Khmer populations in northwest Laos

interethnic systems and localized identities

131

this does not usually apply to the whole lineage to which they belong. Important decisions are traditionally taken by the village assembly, which is made up of elders from each lineage. Inheritance is based upon the principle of equal repartition of goods, though prestigious items such as bronze drums remain the property of the lineage as a whole. Parcels of land are divided equally, with a supplementary share for the child who takes care of the parents (which is usually, but not always, the youngest). Nevertheless, historical records and ethnographic data tells us that differences in status amongst the Khmu villagers of the Nam Tha occurred through political and economic relationships with the lowlands. Three main aspects can be underlined: control over some goods obtained through trade and local specializations; temporary migrations of young males; titles of nobility given to some Khmu leaders by the Tai lords. The Khmu interviewed during this research say that they have always been dependent upon other populations to obtain the iron bars from which they make their agricultural tools. In several myths, it is said that before the arrival of the Tai in the northern peninsula, the Khmu used to cut the forest with stone axes. The iron bars were obtained through trade with Lao boatmen or through contacts with another Mon-Khmer population known as Samtao.3 Village priests in some villages had a ritual monopoly over working these iron bars. This feature, combined with the priests’ ritual privileges regarding the choice of parcels of land, must have placed them and their relatives in a favourable position within Khmu society, for they were able to exert some form of control over several aspects of the agricultural cycle. Another signicant process of social change has been the temporary migration of young males from this area to the lowlands, which is considered the most common way to gain prestige. Historically, most of the young Khmu and Lamet migrants went to work in eastern Burma, and after 1880, also in the teak plantations of northern Thailand. The French colonial administration became quite worried about these migrations, which they tried to monitor and control through an ‘agency’ built

3 This is a local name for a Palaung-related group whose members were famous blacksmiths (more on this below).

132

olivier evrard

in Chiang Khong at the end of the nineteenth century.4 In 1930, three to four hundred young Khmu men were continuing to arrive each year in Chiang Mai, where they would stay for an average of two to three years (Lebar 1965: 8). Lampang and Nan were also common destinations for these migrant labourers, but after the 1930s, they went mainly through the border areas of Chiang Saen and Chiang Khong. After several years in the lowlands, these young men usually brought prestige goods back to the Nam Tha Valley, such as bronze drums, gongs and buffaloes, which they used to pay bride price and to set up a family. In all the villages in which I have had the opportunity to work, the ‘richest’ men (the ones considered by the others as having accumulated more prestige) had spent several years in either Thailand or Burma. They usually possessed many buffaloes and gongs, and had brought at least one bronze drum back from their trip. Their wealth was conrmed and validated through rituals, as well as through the possession of larger amounts of land.5 Historically, it seems that in the villages where numerous young men had undertaken this kind of migration, an ‘upper class’ developed: the richest men received the title pia6 among the Khmu of the Nam Tha, and lem in Lamet villages (Izikowitz 1951: 347–352). These men often distinguished themselves by wearing symbols of Tai aristocracy, such as turbans or clothes decorated with silver. Special ceremonies were organised when a man received such a title, and in the case of the Khmu, a wooden drum was offered to his lineage by the lineage of his wife-takers (Lündström & Tayanin 1981: 175).7 Interestingly, such

4 Lefèvre Pontalis (1902: 113) wrote some few pages about the “exil des Khas [here Khmu and Lamet] de Louang Prabang et de M. Pou Kha [ Viengphoukha] que l’exploitation des forêts de teck attire depuis de longues années du côté de Xieng-Mai [Chiang Mai], sans qu’ils renoncent jamais d’ailleurs à l’idée de retourner dans leurs foyers” (Lefèvre-Pontalis 1902: 113). 5 Increasing one’s status turns economic capital into social capital. The higher the status, the better their ability to organise rituals (which implies the sacrice of animals), and own larger parcels of land. On buffaloes, prestige goods and relations with the outside world among the Lamet, close neighbours of the Khmu in the Nam Tha area, see the excellent analysis of Guido Sprenger (2005: 291–312). 6 They were then called ta pia, ta being the name used for every old man (more precisely, for every man who can be considered as belonging to the generation of father’s father). 7 Marriage with matrilateral cross-cousins (real or classicatory) is a social rule, and the symbolic opposition and hierarchy between em (wife-givers) and perha (wife takers) permeates nearly all aspects of Khmu social and ritual life. For the Lamet case, see also Sprenger, 2006.

interethnic systems and localized identities

133

ceremonies have also been organised more recently, when young men left their village to become soldiers throughout the Indo-China war. These migrations not only allowed some young men to accumulate prestige goods, but also constituted powerful vectors through which many aspects of Tai civilisation were adopted, including house construction, tattoo designs, clothing norms, and also many of the values and beliefs linked to Theravada Buddhism and with Tai conceptions of social order. The changing geopolitical environment of northern Laos reinforced this process. Tai lords were often at war either with each other or against powerful invaders, such as the Burmese, and were constantly attempting to secure the mountainous margins of their principalities (müang). Mon-Khmer populations often played the role of guardians at the borders of the muang, and sometimes served as vassals in the armies of the Tai princes. As reward, Tai rulers would bestow titles of nobility ( panya) to some of the most well-known or powerful Khmu chiefs.8 Some Khmu leaders even began to build pagodas in their own village, or fund the construction of pagodas in the lowlands to increase their own prestige (Lefèvre-Pontalis 1902: 147). In this way, a specic political context combined with frequent individual contacts with more sophisticated lowland civilisations, led to the creation of hierarchy networks amongst highlanders whose networks were, either wholly or partially, modelled ideologically on those of the Tai populations. At the village level, Tai inuences that were mediated through control over trade, temporary migrations or political agreements, favoured either the development of a local ‘aristocracy’ holding both economic and ritual power, or political instability due to tensions between these two aspects of legitimacy. During his stay amongst the Lamet in the 1930s, Karl Gustav Izikowtiz (1951: 347–348) noted that in the villages where many men had the title of lem, the authority of the village remia ( priest, or lkun in Khmu villages) was diminished, except if he himself had obtained the title of lem. Similar processes were also operating in Khmu villages. If the village priest was also a member of the richest

8 Their role as ‘guardians of the edges’ is commonly cited in the literature. For instance, on the Viengphoukha plateau, the Tai prince of Nan had concluded an agreement with three Khmu panya who were in charge of watching over the movements of the Burmese armies on the plains (Extraits du Journal de voyage de LefèvrePontalis et de Macey de Louang Prabang à Xieng Khong sur le Mékong, Archives du ministère français des Affaires Étrangères, Affaires diverses et Politiques 1815–1896, tome IV, volume 11).

134

olivier evrard

group of men, he succeeded in keeping political status, resulting in a relatively stable hierarchy (at least for one generation) between the local aristocracy—allied to the ritual power—and the commoners. Another situation that may occur is when ritual power conicts arise with younger men who have sufcient prestige to contest the legitimacy of the elder, and which subsequently leads to a split in the village. I encountered two clear examples of this scenario during my eldwork in the Upper Nam Tha valley. The rst such example occurred in a Khmu village named Konkud that split some sixty years ago. Two of the oldest and most respected men of this village in the present day, Ta Mom and Ta Tchoy, were at that time coming back from several years of work in the lowlands, the former in Thailand, the latter in Burma. Both were members of the lineage of the founder of the village and were potentially in a position to succeed the lkun after his death. Their return coincided with the split of their lineage into two branches. The ‘youngest’ branch, led by Ta Mom and Ta Tchoy, led the followers of several houses of other lineages to settle a new village, and did not return to Konkud until several years later. It would seem that these young men had accumulated sufcient economic power to contest the ritual power of the lkun. By founding a new locality, these two men tried to acquire a ritual position that they could not otherwise acquire in their own village. The name of the village indicates that this was not the rst time that such a scenario had occurred, as this name (Kon, child; koud, return, come back, enter) is said to symbolise the impossibility that villagers will leave forever, and that any split in the village population is only temporary. Data gathered on the other side of the Nam Tha, in a Lamet village, shows that such conictual relationships between young migrants and old ritual authorities still occur in the present day. In Ban Chomsy, for instance, a thirty-ve year old man who had worked in Thailand for ve years had returned and encouraged the other villagers to change their way of life by modifying their agricultural practices by growing cash crops. In February 1995, this man, one of the richest in the village, was the only one to possess a shpond and a large irrigated orchard. He was helping to convince the villagers to move downhill (as the local administration was requesting), but a majority of villagers, led by the previous chief of the village (who also had the title of priest), were still refusing to move. Finally, after a few months the young ‘progressive’ took his family with him and settled alone near the river. Two years

135

interethnic systems and localized identities

From Namtha

NORTH Kaengkhan Hatnalaeng

Key Khmu

Moktchong

Lamet

Puluang

Tai Sanghat Hato

mixed

Hatchala

Houeyfai

administrative centre Hatloi

Saloi

Takô deserted village

Phoutaen

affluent

g Haen Pouchai (2) Nam

Lao

Tonglahang Pouvieng Tako

Mokkam pang

Mouayou Na

Samnack Müang

Lampang

Mongmèng Koneti

Phulan pang

Phoutin m

Phousaengle Pouhong (2)

Peuane

Mokumtuan

Phoukhang

Saphut

Phulan Phahu Phusung

tmoy lü

Peung Konlang

Hatae Phuhon

Moktu Mokkha

Phouren Tongkatang Tongmok

tmoy yuan

Pusung pang Phavi 2 Phavi

Vienglao

Phulot (2)

tmoy lü

Om

Pouhou

Kongsalai

Vèn Namhaeng

Lampae

Konehine

mK

Hatlom

Thongthon

Based on the map of the district education services in 1995. Transcription follows English pronunciation

Na

Nam Tha

ho ne

NamTha

Tavan (2) several settlements

Pangbok

Poukho

ng

m

Na

Ya

Vat

Kanungphaet (2) Hun

Sakoub theung

Nalè

Hueykaï

Mongsouan Sakoub loum Kanuang

Sakoub kang

Salouang 2 Kanje m

Na

ip

Th

Salouang

Mongkhô (2) Salouang Pang Donethip

Phulom

Phulom pang Moktchod

Chomsi

Phuton

Phuheun Nguan

Hatdao

Konkud

Nong Kanha Püen (2) Konkham tmoy rok Mokloysato Mokkoud Mokprat Lapoum Mokloy (2)

Phayte (Lava 3)

Sangaen

tmoy rok

Chomsi 2

Lava

Chai(Lava 5)

Lavè (Lava 2)

Sakaen

Moktchak Mokchong

Moknjom (2) Chompon (Lava 4)

Koneveune

Tavan (2) Takeung

Lamet

Talang

Phukho

Hueykaen Lahang Namling to Bokeo (Pha Oudom district)

1

9 Km

c Olivier Evrard, 2005

Map 2: Villages of Nalae district: names and ethnic afliation

136

olivier evrard

later, many other households had followed him to settle a new village, for which he remains the naiban (administrative chief ) today. These two biographies offer examples of how Taization processes occur in Mon-Khmer villages of the Nam Tha valley. If some of the external mechanisms (namely, migrations of young males) described here are not mentioned by Leach for the Kachin, their impact at the village level has led to similar trends, namely to the appearance of traits of a class structure and to political instability, sometimes with the splitting of villages. However, what is relevant at the village level can be misleading on a wider scale, and, as such, we must examine to what extent Leach’s framework is relevant for an ethnohistory of the Mon-Khmer populations of the whole Nam Tha valley. Leach’s Framework and its Adjustment to the Local Interethnic Context Two key points should be made here. Firstly, it should be stressed that social inequalities do not ‘crystallize’ into the social structure of the Khmu to the same extent as Leach described for the Kachin. Indeed, nowhere amongst the Mon-Khmer populations of this area can linguistic categories such as “gumlao” or “gumsa” (Izikowitz 1969: 148) be found. Nor do we nd Khmu words to designate a domain or a domain chief. When rich men received specic titles, such as lem in Lamet villages or pia in Khmu villages, they obtained specic prerogatives, such as the possibility of choosing the best parcels of land (better orientation, better soil quality, closer to the village, etc.) or receiving a proportion of the nes that were paid during the resolution of conicts. Yet, this research has found that their titles were not hereditary from one generation to another, and that upward and downward mobility took place. Secondly, the idea of “vicious competitions” (Leach 1986: 194)— political competition for power and wealth—has less relevance to the Khmu, either from an economical or a political viewpoint. In the economic and ritual spheres, the concept of fertility and abundance takes precedence over the idea of surplus. In the case of the Khmu, (and this could be applied to other Mon-Khmer societies as well), it would then be excessive to speak about competitions between the houses of a village or about calculations aiming to maximise either the yields or the merit. This was one of the criticisms formulated against the theo-

interethnic systems and localized identities

137

retical models elaborated by Friedman (1979) and Kirsch (1973). They, for example, did not take into account the possibility of rapid reversal of fortune through demographic recession or other events.9 From the political viewpoint, the differences appear even more striking: violent contacts between Mon Khmer populations were rare, and neither the stories told by the Khmu elders nor any written document mentions competition between famous Khmu leaders supported by armed partisans carrying out vendettas. A number of additional differentiating factors must be sought to describe the interethnic context, and more precisely, its political and economic characteristics. In northern Burma, the Kachin have played the roles of turbulent allies or ‘political parasites’ (Leach 1986: 254) of the Shan. In some areas, they have even driven Shan communities from their homes and enslaved these groups to work lowland paddy elds. Conversely, the Khmu populations of northern Laos were driven forcibly from the lowlands (where they had probably already created some small kingdoms) in the fourteenth century following Tai usurpation of political power, despite the fact they had most probably been in contact with the Tai for a long period prior to this. This spoliation, then, gave birth to a ritual and hierarchical formalisation of interethnic relationships,10 which is not the case in areas where the Shan and the Kachin are in contact. Moreover, unlike the Kachin, the stability of this hierarchical relationship has not been threatened since the time of the foundation of the Lao kingdom. From an economic viewpoint, in several of the areas studied by Edmund Leach the Kachin were dependent upon the Shan for their supply of rice (especially in what Leach called ‘Zone B’, 1986: 235). However, the opposite situation often prevails in northern Laos. Because of the scarcity of at land, the political domination of the Lao could not be everywhere conrmed and secured by the development of large areas of paddy eld. On the other hand, some of the Khmu populations kept control over large and fertile upland territories, where slash-andburn agriculture produced quite good yields. This is especially true in

9 On the other hand, status change of an individual house by impoverishment is no argument against the notion of a status structure. In capitalist societies, high class families may drop out of their class the same way and just as quickly, but this does not change the fact of the class structure itself (Guido Sprenger, personal communication). 10 About the interethnic structure as it appears in New Year rituals in Luang Prabang, see Ajmer 1979 and Archaimbault 1973.

138

olivier evrard

the Nam Tha valley, which became, as written by Lefèvre Pontalis ‘the granary of Luang Prabang’ (1902: 140). Lao boatmen paddled upriver at the end of the monsoon to exchange iron bars, salt, clothes or jars, against paddy, cotton or forest products. This trade of manufactured goods against natural products was well in favour of the boatmen. In fact, it was so much so that some upland villagers, especially those settled near the mouth of the river and near the main Lao centres, were short of paddy for several months before the harvest. They had to borrow paddy from the Lao who, conversely, speculated on the price of the paddy that they had bought a few months beforehand. However, this was not the case everywhere, and until quite recently in the upper valley of the Nam Tha, most Khmu villages did not regularly suffer from shortages of paddy. In spite of these differences, it is still possible to adjust the model elaborated by Leach. From a geographical viewpoint, the dynamics described above gave birth to three different trends. Firstly, differentiation occurred between remote areas, usually higher in altitude, and regions that were closer to the economic networks of the lowlands. Such dynamics are already documented for the Chin (Lehman 1963), the Lamet (Izikowitz 1951), and the Wa (Scott & Hardiman 1900, cited in Lehman 1963: 27). Leach (1961) considered that his model could be applied to all upland regions of Southeast Asia. Lehman, for instance, showed that the Chin, who lived close to the lowlands and the Burmese populations, had peaceful relationships with neighbouring groups. In contrast to the “remote” (or northern) Chin, they did not need to develop complex social organization or hierarchical political systems to secure their access to the goods and technologies of the lowlands (Lehman 1963: 44–46). This does not mean that highland villages have a more ‘elaborated culture’ (Leach, 1961, cited in Lehman 1963: 45), as the proximity of the lowlands can also have a positive inuence, which does not necessarily imply an impoverishment of the culture of the highlanders. In northwest Laos, before the massive resettlements of the last thirty years (more on this below), the social organization of Lamet villages was more hierarchical in the remotest communities, and more egalitarian in those lying near the caravan path of the Viengphoukha Kha plateau. In the former villages, temporary migrations of young males were more numerous, and the numbers of individuals who had acquired the title of lem was accordingly higher (Izikowitz 1951: 99,

interethnic systems and localized identities

139

114, 347–348). I personally observed a similar trend in the Khmu villages on the east bank of the Nam Tha. Going further from the main river, and to higher elevations, villages are generally bigger, agricultural land is more intensively cultivated, and differentiation in status amongst villagers is more obvious. These villages, at least in the Nam Tha area, are also more prosperous, for the yields are better on the higher ridges of the mountains.11 Paradoxically, the cultural inuence of the lowlands is often more obvious in these remote villages than in those settled at mid-distance from the river or the main roads, which have usually kept more ‘traditional’ ways of life. A second trend concerns the merging of segments of lowland and highland villages, and the creation of multi-ethnic settlements along the main transport routes. This feature, also acknowledged by Leach, for instance in the area of Myitkyina (Leach 1986: 243–244), is a very old trend in the Nam Tha valley, and most probably occurred in the main uvial valleys of northern Laos and Thailand. Historically, seasonal trade in the lower Nam Tha valley encouraged some Lao boatmen to settle more or less permanently near their Khmu or Lamet suppliers and customers. Small trading posts developed, which were then progressively established upriver and further into the valley because of concurrence between the Lao boatmen (Izikowitz 1951: 27). At rst, such localities were usually multi-ethnic: Lao boatmen married Khmu or Lamet wives, and the latter brought with them some relatives to settle near the river. When the locality became properly established, additional Lao migrants would come to settle. A pagoda was then constructed and the multi-ethnic origin of the locality was then completely forgotten. Such processes are still occurring today in most upper sections of valleys, where one can observe quite recent villages that were founded by Lao boatmen who are married with Khmu women, and followed by several houses from each ethnic group (the Khmu usually being more numerous). The Lao houses commonly specialize in trade, while the Khmu practise slash-and-burn agriculture in the surrounding hills. The political context in which such villages are created today differs markedly

11 In Nalae, usually the higher the village is, the better the yield. One can also observe that the ratio of early/late species of paddy is much more in favor of the latter in the upland villages, meaning that villagers have less problems with food security between harvests.

140

olivier evrard

from earlier periods,12 however, the basic interethnic organization remains the same as it was in past centuries. A third phenomenon manifests itself more specically with the Khmu, and is less well documented. Through political alliances with different Tai chiefdoms, and exposure to different regional economic inuences, Khmu populations of the Upper Nam Tha valley acquired cultural peculiarities that led to the creation of subgroups named tmoy. This word may have little geographic meaning outside of the Nam Tha valley, but, in this area at least, it can be demonstrated that the tmoy clearly constitute a territorial system or, more precisely, a kind of cultural and political continuum (Évrard 2003). It seems that the tmoy of the western banks of the Nam Tha, which were more inuenced by Tai culture, acquired more stratied social systems than on the eastern banks and this feature played a signicant role during and after the Indochina war. The next section of this chapter goes into further detail, showing the history of the tmoy, and describes how their contemporary fate is a pertinent example to approach the topicality of Leach’s analysis for highland Southeast Asian peoples. The Khmu Subgroups, TMOY, in the Nam Tha Valley: Debates and Facts In Khmu language, the word tmoy has several inclusive meanings. It can designate guests, or strangers, or, more generally, people who share a common culture but who have different customs, language or peculiarities. This word is used to create specic ethnonyms by adding the name of a river (tmoy khong or khmu khong for those living near the Mekong; tmoy ou or khmu ou for those living near the Nam Ou), the name of an area (tmoy luangphrabang), a Tai population (tmoy yuan

12 Traditionally, the banks of the river in the uppermost parts of the valley were empty. Numerous rapids prevented traders from reaching the town of Luangnamtha by boat. After the war, and following the introduction of motorized boats, local ofcials encouraged the creation of new villages, both to develop trade with the Mekong valley, and to avoid the inltration of some guerrilla groups, which were still active in this area until the end of the 1980s. The current policy of the Lao State encourages and sometimes forces upland villagers to move downhill, and to mix with already existing lowland villages. This now constitutes the main factor leading to the creation of multi-ethnic localities in this part of the valley. These policies have directly contributed to the depopulation of the highlands, which has in turn reinforced the instability of the remaining upland villages.

interethnic systems and localized identities

141

or khmu yuan; tmoy lü or khmu lü), a linguistic characteristic (tmoy me and tmoy khat following the kind of negative particle that is used), or even a detail of dress (tmoy treal for those wearing short vests; tmoy vang for those wearing long shirts). Other Mon-Khmer groups such as the Lamet (tmoy lamet) can be included in this classication by opposition to the Tai, whom the Khmu pejoratively name tchae. William Smalley, an American missionary who conducted some linguistic eldwork in Luangphrabang Province between 1951 and 1953, rst mentioned this word in 1961, and identied several tmoy, or ‘subgroups’, amongst the Khmu populations (Smalley 1961 and 1965). A debate has arisen amongst specialists of Khmu culture concerning the nature and origin of such subgroups. For Kristina Lindell, who conducted research on Khmu culture at the University of Lünd for thirty years, the tmoy constitute localised subgroups that can be viewed as vestiges of old territorial organisation prior to the establishment of the Lao kingdom of Lan Xang in Luangphrabang (Lindell, Samuelsson and Tayanin 1979). Lindell identied seven tmoy and drew a map of their location in the northwest part of Laos (Lindell 1982). However, she does not provide any similar map for other parts of Laos, nor does she mention that Khmu people can name many other tmoy beyond those few she considered in her analysis. Finally, she does not acknowledge that one can identify cultural variations, especially from a linguistic viewpoint, that do not exactly match her classication. For instance, while all belonging to the so-called tmoy rok, villagers of Houn district (Oudomxay Province) use khat as a negative particle, while villagers from Nalae, Pha Oudom and Paktha districts mostly use phe. In Luangphrabang, some Khmu people use the term al, and others use am. Conversely, an American linguist, Franck Proschan (undated: 55), convincingly demonstrates that it is illusory to attribute a precise territory to each tmoy. Drawing on a list of tmoy names that he collected during his research among Khmu populations of Thailand, Laos, Vietnam and even America, he showed that it was impossible to account precisely for the total number of such terms. Indeed, instead of a simple relation between a subgroup and a name, one can observe that the same subgroup can be designated by different names according to the chosen criteria (type of housing, clothes, dialect) or that the same name can refer to mutually exclusive and geographically distant groups. Moreover, people may consider themselves as belonging to a different tmoy than the one attributed to them. Most of the time, people can easily tell which tmoy their neighbours belong to, but have difculties in labelling

142

olivier evrard

themselves with a specic tmoy identity. Proschan (1997: 98–99) thus asserts that the specic tmoy term should relate to the local and conversational context in which it is used. Paradoxically, however, Proschan’s analysis does not completely invalidate the data of the Swedish scholars of Lünd mentioned previously. On a local scale (i.e. the Nam Tha valley), the tmoy clearly constitute a system of territorialized identities. Villagers of the upper part of the Nam Tha valley are able to designate six tmoy, four of which are Khmu: tmoy yuan (Khmu yuan), tmoy rok (Khmu rok), tmoy lü (Khmu lü), tmoy kwaen (Khmu kwaen), and two others belonging to Palaung-Wa related languages: Lamet (tmoy rlamet) and Samtao (tmoy samtao). The territories of these tmoy —the Samtao excepted—are quite precisely delimited (see maps 2 and 3): on the right (west) bank of the river, the tmoy kwaen, youan and lamet are arranged from north to south; on the left (east) bank, are arranged the tmoy lü and rok. The territories of each tmoy are expanding into neighbouring districts: the Khmu kwaen and youan as well as the Lamet are also to be found in Viengphoukha and Huoixai districts, while the tmoy lü and rok occupy the entire mountain range separating the Nam Beng and Nam Tha valley (which includes the districts of Nalae, Beng, Houn and Pa Oudom). Geographical borders between each tmoy are quite well known by villagers themselves, and usually coincide with tributaries of the Nam Tha river. In many cases, these border areas are empty zones that take a full day to cross on foot. Indeed, the footpath that links Khmu villages to the valley are often in better condition than those linking villages belonging to different tmoy. Members of these tmoy used to distinguish themselves13 especially by their clothing: short white double-breasted jackets for the Khmu yuan (men and women); short blue jackets for the Khmu lü (men and women); short blue jacket with colourful stripes amongst the Khmu kwaen (men and women); long indigo shirts reaching the ankles (as a levite) for the Khmu rok (only for men). The hairstyle also varied from one tmoy to another: in Nalae District, the Khmu villagers say that the Khmu yuan men were the only ones who wore their hair twisted into a bun, while the Khmu rok used to shave their heads entirely except for a small tuft on the top of the head. Other distinctive features included the form of housing, basketry, and even the construction and function of some iron tools. The features described above allow people to designate the 13 This is still true today in some remote villages especially for clothing, but to a lesser extent than before.

interethnic systems and localized identities

143

tmoy using different criteria. For instance, instead of speaking about the Khmu lü and the Khmu yuan, a Khmu villager can sometimes talk about the tmoy klok (those who wear a white shirt) and the tmoy lün (those who wear an indigo shirt). Similarly, someone could speak about the Khmu rok by using the term tmoy vang (those who wear long shirts) comparing them to tmoy treal (those who wear short shirts). Despite the multiplicity of ‘identity labels’, and the various forms of classication that it allows, there is a denite consensus amongst Khmu villagers of the Nam Tha valley about the territoriality of such subgroups. It seems very doubtful, however, (and not grounded in any ethnographic or historical evidence), that following Kristina Lindell’s hypothesis, such intra-ethnic boundaries constitute the remnants of a political organization of the pre-Tai era. Rather, the tmoy should be viewed as both the imprint of Tai political systems in these mountainous borderlands, and as indicators of the intensity of the relationships between Khmu and Tai populations. The TMOY as an Imprint of the Tai Political System The Khmu population of the Nam Tha valley are called the Kha Kao (the old Kha) in the Lao chronicles (ofcial historical texts of the Lao Kingdom). They were chased away from Muang Swa (the old name of Luangphrabang) when the Lao seized power in this city, and took refuge in the area of Paktha. Later, the Lao king Fa Ngum resettled some of the Kha Kao, and asked those who stayed in the area to protect his kingdom from invaders (Pavie 1898). As mentioned previously, such alliances between Tai Kingdoms and their mountainous borderlands were quite common in northern areas, where they echoed, at least in Laos, the geographic fragmentation of ethnicities and the economic interdependency between the highlanders (who produced most of the paddy) and the Lao boatmen. Consequently, and despite the fact that the mountainous margins of the müang kept intact their own internal political dynamics, the Khmu became ‘connected’ to the Tai political system and eventually became subsumed within it. In this historical and political context, two main geopolitical divisions appeared amongst the Khmu populations of the Nam Tha valley and constituted the basis of the tmoy system. The rst division is between the east and west side of the valley. The Nam Tha river used to be the western border of the Lan Xang kingdom. At the conuence of the Nam Tha and the Mekong, the

144

olivier evrard

village of Paktha was used as a customs post. Downriver, the Mekong was subdivided into segments called mün and was controlled by the Lao administration, while upriver it was under the control by the Siamese (Lefèvre-Pontalis 1902: 100).14 Consequently, the Nam Tha basin was, for a long time, a kind of buffer zone, where Khmu populations gave allegiance to different Tai princes. Khmu of the right bank were linked politically to the Tai Youan of Nan kingdom (itself a vassal of Lan Na and then of Siam), hence the name Khmu Youan or tmoy yuan used to designate them. The Khmu of the left bank were under the control of Lan Xang, hence the term kha lao that was used to name these people in the past. This opposition predominantly concerned the lower part of the Nam Tha valley, while, in its upper part, Lao inuence was much weaker, due mostly to immigration of Lü populations from Müang La [Mengla] at the end of the eighteenth century. These populations, who settled near the banks of the uppermost part of the river, sometimes refused to pay taxes to Louang Prabang (Pontalis 1902: 142), as they were trying to create their own independent chiefdom. Consequently, the second geopolitical division to be taken into account is between the lower and the upper areas of the Nam Tha basin. The geographical limit of these two zones coincides more or less with the rapids that separate Louang Namtha plain from Nalae district (created only in 1983 and called tasseng nüa in the nineteenth century). Downriver from this point, the Lao (or Lü) paddlers were still able to navigate to and from Paktha and the mountains, which were controlled by Khmu populations owing allegiance either to Nan or to Louangphrabang. Upriver, there were no possibilities for waterbased transportation, and trade was carried out using mule tracks. In these areas, the Khmu owed allegiance to Sip Song Panna chiefdoms. However, a sense of opposition remained in place between the populations of the east and west sides of the Nam Tha. In the mountains lying west of the river (between the Nam Tha and Viengphoukha), Khmu villagers belong to the kwaen subgroup. The kwaen (as the tasaeng in Lao polities) were administrative subdivisions of the mountainous margins of the müang in the Sip Song Panna chiefdoms ( Lemoine 1997: 187). 14 One mün corresponds roughly to the distance that boatmen were able to row upriver, while a mün of wax (12 kilogrammes) burned. Today, Paktha is still a major military and customs checkpoint. There are still fteen mün between Louangphrabang and Paktha (called Lok, Seuang, Han, Tan, Hang, Nang, Phat, Krong, Teuan, Têt, Lè, Sit, Teun, Tôm, Daï ) and two others ( Dan and Than) have been added between Paktha and Huoixai.

interethnic systems and localized identities

145

They were given a great deal of autonomy, and were controlled by tribal leaders named ho kwaen. Conversely, in the mountains east of the river, Khmu villagers were considered as belonging not to the tmoy kwaen but to the tmoy lü. If, in these two cases, the Sip Song Panna inuence was predominant, the Khmu kwaen were placed under the control of Müang Sing, while the Khmu lü owed allegiance to the principality of Müang La [Mengla]. The inuence of Müang La extended much further south than Müang Sing, since Khmu villagers in Nalae district still consider today that the tmoy lü includes all the villages on the left bank until the river Yang (huaey yang), located in the centre of the district (see map 2). Indeed, Müang La used to control a great part of the Nam Beng valley (currently Oudomxay province) and this factor, added to the migration of Lü villagers towards the Nam Tha valley at the end of the eighteenth century, probably explains the geographical extent of the tmoy lü today. The TMOY as an Identity Continuum While the political history of Tai polities gives clues to understanding how the tmoy came to constitute a kind of territorial system in this area, it proves to be of little use in grasping the meaning of two other tmoy names, rok (one nds also the transcription of hoc or hok) and samtao. While the rst name designates a Khmu population, the second one is applied to a group linguistically related to the Palaung-Wa family, which is found in Laos and Thailand, but especially in the Shan States of Burma (Howard & Wattana Wattanapun 2001: 46). The word rok is still used today to designate the Khmu populations that are settled in the mountainous area at the intersection of the Houn, Beng and Pha Oudom districts. Few Khmu of this area would identify themselves as rok, for this name has a pejorative meaning of ‘savage’, ‘backward’ or ‘inferior’. Consequently, outsiders can delineate a ‘rok area’, but the relevance of this is lost when asking inhabitants of the rok community.15 Most of the criteria commonly used to identify subgroups in a given culture have not proven to be satisfactory in the

15 However, most of them acknowledge the use of this name by their Tai neighbours and some even tell proverbs in Khmu language about the so-called ‘rok’ subgroup. In one Khmu village of Houn district, an elderly man said once “Cüang mork, rok tang/ Cüang mork rok hor ”, which more or less means that Khmu rok are the people who are the servants of Cüang, the mythical Khmu hero.

146

olivier evrard

case of the Khmu, except maybe the forms of basketry that seem to be quite distinct from the other Khmu subgroups.16 If one tries to examine the criteria that are usually linked with the so-called rok identity, one sees that they often refer to an emptiness, absence or poverty. In other words, the rok identity is often dened relative to other subgroups by stressing its inferiority: the architecture is said to be more primitive; the women do not weave or embroider; silver jewels are scarcer than amongst other tmoy, etc. If some of these assumptions are wrong (architecture, for instance, can prove to be no less sophisticated than in other tmoy), others were certainly true, as I observed. For instance Khmu rok women do not weave, contrary to Khmu lü and Khmu yuan women. This is of great concern for local ofcials who are trying, with little success, to get a weaving school established in the Khmu rok area as part of a rural development project. In addition, iron or silver carved objects—pipes, for instance—are very scarce amongst the Khmu rok. Such pipes are mostly to be seen in the Khmu yuan area, where they are now considered as female objects and inherited from mother to daughter. Among the Khmu rok villages, traditional pipes were made of roots and only their extremities were sometimes decorated with iron or silver.17 It is most likely that pipes made from roots were quite common in the past and that silver pipes appeared progressively as a symbol of afuence and social sophistication amongst populations where outside activities had favoured the acquisition of silver coins. Conversely, their absence among the Khmu rok is probably the consequence of their isolation and their lack of economic integration within regional networks. A map of the tmoy and one of the main commercial routes of the nineteenth century (see map 3) clearly shows that the areas of the tmoy kwaen, lü and yuan were criss-crossed by several mule tracks. The main one linked Viengphoukha to Müang La through the small hamlet of Sop Ngim, which was lying near the conuence of two major tributaries of the Nam Tha, one coming from Viengphoukha, the other one from Namor. Sop Ngim was not really a major crossroads of regional

16 As much as weaving or embroidery, basketry is a meaningful identity marker, especially among the Mon-Khmer populations, and as such would deserve an indepth study. 17 A picture taken by the French traveller and journalist Alfred Raquez (1902: 227) shows Khmu rok people smoking such pipes.

interethnic systems and localized identities

147

trade, but was still important enough for Chinese traders to set up a permanent post there, where men and mules could rest (LefèvrePontalis 1902: 44). Besides this main axis, several secondary tracks also connected Viengphoukha to Ban Nalae or Ban Mo (now called Ban Donethip) on the right bank of the Nam Tha. During the rainy season, these secondary tracks allowed Chinese caravans coming from Müang Sing to fork east after Viengphoukha and to use uvial transportation to reach either Siam or Louangphrabang. In this context, the tmoy yuan and kwaen and even the Lamet (tmoy rlmet) were connected to regional trade, while Khmu populations from the east side of the Nam Tha were more isolated. This isolation and its consequences more likely explain why the name rok was attributed to them. The tmoy samtao constitutes a kind of opposite case compared to the tmoy rok. First, the identity does not correspond to a precise area in northwest Laos and has only a few members, most of whom are scattered in different villages. Secondly, it designates a population that used to be famous for its political and economic integration. Samtao themselves claim the name toumok (highlanders) and consider that samtao used to be mostly a political category. Indeed, during the nineteenth century, this term referred to a semi-autonomous political area linked to the chiefdom of Xieng Tong in Burma and inhabited by Palaung populations, known on the Lao side of the Mekong as Doi [ Doï, Doy], who were found mostly in Long district, Louang Namtha province. As noted by the French explorer Francis Garnier, the main area of the Samtao in Burma was inhabited by around 10,000 people and was quite famous for its production of ries made with the iron brought by Chinese traders. “The Doi Samtao were producing 3,000 ries a year and were giving more or less 200 of them to the prince of Xieng Tong for their allegiance”, wrote Garnier (1885: 416). The fact that some of the Samtao settled in Nalae district shows that their reputation as blacksmiths and ironworkers extended a long way from their place of origin. Nevertheless, this never constituted a massive migration in the Nam Tha valley; rather, Samtao seem to have followed the main commercial axis, and become mixed with other Mon-Khmer populations. Today, for instance, some Samtao families in Nalae district are still living in Ban Hatnalaeng, but only the oldest men still claim such a name, as their children and grandchildren now consider themselves to be Khmu. These two examples clearly show that the tmoy acquired a territorial meaning mostly through their relations with the locally dominant Tai identity, with which they become more or less imbued. In other

148

olivier evrard

Key

NORTH CHINA

Mekong tributaries current international borders main tracks at the end of the 19th century (drawing after archives of Pavie Mission, 1888-1896)

Mengla Nam La

provincial centre other locality Sop Ngim deserted locality or old name

Müang Sing

tmoi lü Boten tmoi kwaen

Luangnamtha

a

m

M

a Müang Long N

tmoi yuan

Namor

tmo rlmet

Xièng Kok Nam Fa

tmoi rok

Müang La Sop Ngim

Viengphoukha ha

Müang Mom Na

Huoixai

m

Ng

Nam T

BURMA

Oudomxay

Tang Ho ao

Müang Beng

Ban Hat Nam

Chiang Saen Chiang Khong

THAILAND

Nalae

Ban Mo

Müang Houn Paktha

Na

m

n Bè

g

Pakbeng

c

Olivier Evrard, 2005

0

Map 3: Khmu sub-group areas and old commercial networks

50 km

interethnic systems and localized identities

149

words, areas of variable size emerged by formalizing their relations in specic ways with the political environment, and by developing (or not) economic specializations or distinctive features. The case of the tmoy rok may seem different, but, indeed, it belongs to the same relational structure, named only in a negative way: this tmoy is dened by what could be called a ‘decit of relations’, or through a lesser inuence of the political and economic environment. Conversely, the label tmoy samtao refers to the propensity of this group to interact with their external environment, and their reliance on economic specialization. Beyond their linguistic differences, the so-called rok and samtao populations can be seen as two symmetric cases inside the same territorial and relational system, or, in other words, as two ends of a continuum that, if represented as a line, would include the other tmoy (rlamet, kwaen, lü and yuan) in intermediary positions. TMOY and Hierarchies: Geographical and Political Variations Due to the lack of ethnographic studies conducted in this area, it is difcult to know with certainty how various types of relationships with the lowlands inuenced local forms of social life. Conversely, we can only guess about how the latter reinforced and perpetuated the differences between the tmoy. Despite the fact that all the Mon-Khmer populations of this area possess the same social18 and economic (swidden agriculture) organisation, some clues indicate that local hierarchies were more obvious on the right bank, amongst the Khmu yuan and kwaen, than on the left bank, especially amongst the Khmu rok. Firstly, the villagers of the rok area clearly afrm that none of their ancestors had ever obtained the title panya from the Tai princes, while this title had been granted to several leaders of the Khmu yuan area. This is conrmed by the data gathered by the rst European explorers in this area, especially Pierre Lefèvre-Pontalis who visited the Nam Tha valley in 1893 and met three panya in the Khmu kwaen region (1902: 145–166). It was here that the Tai inuence had a great impact upon both territorial organization and the spread of Buddhism amongst local leaders, some of whom either constructed pagodas in their own

18 Local patrilineages, preferential alliance with matrilateral cross cousin, lack of political authority beyond the village level, etc.

150

olivier evrard

villages, or made donations to the pagodas of their Tai neighbours (Lefèvre-Pontalis 1902: 147; Raquez 1902: 241).19 A second clue lies in the toponymy of the highlands in the upper Nam Tha valley. I have already mentioned that the emergence of rich individuals in a village could lead either to a conict with the ritual power of the priest, or to a more or less stable hierarchy, where the priest and the rich jointly held political power over a population of commoners. In this second case, the hierarchy could inuence two or three neighbouring villages, with one of these being the historical and ritual centre. In such situations, one usually nds that the kung (villages) have the same name, being differentiated only by opposing adjectives:20 tal (low), trti (middle), and phe (high); prim (old) and me (new), nae (small) and nam (big) for instance. One also sometimes nds these toponyms embellished with images, such as in the example given by Damrong Tayanin (1994: 45–48): the two satellites or peripheral villages are distinguished by their size (Kon Salai koung nam: the big Kon Salai; Kon Salai koung nè: the small Kon Salai) and the ritual centre compared to a source, or more precisely to an ‘overow’ (Kon salai koung piak). All the lists available21 regarding the villages of the upper part of the Nam Tha valley suggest that the tendency towards the creation of small sets of villages with a common name is traditionally more frequent in the tmoy yuan and khuen than amongst the tmoy rok. For the Khmu, if a village keeps the same name as the village it has been created from, it means either that it does not yet have an altar for the roy kung (village spirit) and an lkun (priest) to propitiate it (meaning that it is not yet considered to be a permanent hamlet),22 or that it stays under the ritual power of the lkun of the mother village. In this latter

19 Unfortunately, the Khmu kwaen area is now completely empty since the massive resettlements that were undertaken by communist troops after 1975. 20 The names are given here in Khmu language but the toponyms are most often in Lao language. 21 Data of the 1995 census and of the survey done by the Ministry of Transport (IRAP, Vientiane, 1997) can be compared with those given by Damrong Tayanin (1994, 45–48). Damrong quotes from memory the villages he crossed during a survey done on behalf of the Lao government in 1968. His data have the great advantage of respecting the Khmu toponymy and give a transcription in international phonetic script. Karl Gustav Izikowitz (1951, 39 –40) gave the list of the Lamet villages registered in 1938 in the province of ‘Haut Mekong’ and in the principality of Louangphrabang. 22 In this case, the name of the village is usually followed by a specic term ( pang in Lao, re in Khmu) that indicates that the settlement is growing as an offshoot of the older village on one of its swidden elds.

interethnic systems and localized identities

151

case, the ceremony for the spirit of the village will take place in the ritual centre and will be conducted by the priest for the whole set of villages. Ritual hierarchies are then likely to be more developed in tmoy yuan than among rok villages. This feature coincides, as shown above, with the deeper political inuence of the Tai polities, and more regular connections to regional economic networks. Conversely, in the area less concerned by the Tai inuence, such as the tmoy rok, sets of villages could also be constituted, but following a different process and without an obvious hierarchy. This is the case, for instance, in the so-called tasaeng Sakaen, a subdivision of the district of Nalae that includes seven villages (Konkud, Mokud, Sakaen, Lapoum, Lava, Mokchong, Mokchak). During pre-colonial times, the tasaeng were semi-autonomous networks of villages in the mountainous edges of the müang. These groups of villages had always belonged to the same ethnic group, meaning that due to the numerous migrations and scattering of different peoples in these areas, the word tasaeng did not always have a territorial meaning. The colonial administration territorialized the tasaeng, which then became a subdivision of the district müang. Ofcially, the tasaeng as subdivisions of the district no longer exist in Laos and have been replaced since the end of the 1990s by entities called khet, which are usually larger than the previous tasaeng. What is important to our understanding here, however, is that the villagers of the tasaeng Sakaen consider that they have formed a single territorial entity for a very long time, certainly from well before the colonial period. A myth collected in Konkud village explains that these seven villages came to consider themselves as part of the same territorial unit, not to copy a Tai political model but precisely because they rejected it: Before building their dwellings, all the inhabitants of the tasaeng joined together at the top of the Head of Wild Buffalo Mountain to sacrice a buffalo. The blood of the buffalo spread on the slope of the mountain and its meat was shared among all the participants. The villagers took one buffalo’s rib and an old woman, Ya Phan Pheng, wrote an oath on it. By this oath, the villagers swore always to oppose themselves to the power of the lam. The buffalo’s rib was then buried on the top of the mountain. The villagers went to build their houses and to share the land by using the bamboo crosses, talae. Later on, however, the rib was discovered and stolen. That day we lost the knowledge of writing and we have since then suffered from the power of the lam.

The word lam refers to a specic politico-administrative position (which was often hereditary) in Tai polities. The lam, who were sometimes

152

olivier evrard

called pho lam ( pho, father, protector, patron and lam, translator, gobetween) were appointed by the local Tai ruler to keep control over the population of the müang. The lam had real power (tax collection, settlement of disputes) and constituted a kind of parallel power that facilitated relations between the ruler and his vassals (Lemoine 1997: 180). They were also in charge of relations with the highland populations in the margins of the müang (Reinhorn 1970: 1780). The myth above refers to the lam of Müang Beng and Müang Houn, two little müang of the Nam Beng valley (East of the Nam Tha river), and from where the main attempts to control the rok area seemed to have occurred. The refusal of the Khmu rok to accept the power of the lam nds interesting echoes in the Nithan Khun Borom (the Lao Chronicles), in which it is written that Fa Ngum, the rst Lao king, found support among the kha kao while ghting against the lam of Müang Hun (Hoshino 1986: 110 –111 and Pavie 1898). The end of the myth may refer to the colonial period, during which the power of the local Tai notables was bolstered by the French administration.23 One can also interpret this as an inuence of the Marxist historical analysis introduced and spread by the Pathet Lao during the war (on this issue, see Halpern 1964: 93–95 and 156–157, and also Guido Sprenger in this volume). This myth is also mixed with older mythical frameworks explaining how the Khmu were dispossessed of their power by the Lao: in another version, a dog eats the buffalo skin on which the Law was written, resulting in the loss of the written Khmu language. This example shows that Tai inuence did not follow the same pattern everywhere. In many cases, it allowed the creation and perpetuation of status inequalities amongst highlanders, while in other areas, such as the Khmu rok, it gave birth to local mythologies or territorial organizations based precisely on its negation. In other words, there was both acceptance and rejection of Tai inuence by the Khmu in different areas and such variations from region to region have played a crucial role in the recent history of northern Laos.

23 Most of the tasaeng in Northern Laos were then regrouping several highland villages under the control of one Tai village, a situation that contributed to exacerbating tensions between the highlanders and the lowlanders.

interethnic systems and localized identities

153

The Topicality of the TMOY: the War, the State and Rural Development Policy in Laos In past decades, numerous migrations and resettlements of Khmu villages in the lowlands, most of them implemented by the communist Lao State,24 have weakened, and sometimes nearly erased, the territorial basis of the tmoy. Highlanders have abandoned their previous territories and settled along the banks of the Nam Tha, or on the dusty edges of the main road linking Luangnamtha and Viengphukha, where one can now nd villages with mixed populations of different tmoy, sometimes containing both Khmu and Tai populations. This is especially true for the tmoy kwaen, whose previous territory is now entirely empty. In the meantime, however, tmoy identities have been treated as separate ethnicities and these identities promoted accordingly for tourists. For instance, it is quite common today to see hotels of Huoisai displaying signs indicating that the tourist who will boat up the Nam Tha river to reach Luangnamtha town (2 days trip) will “encounter numerous ethnic tribes, Lü, Yuan, Lamet, Rok and Kwaen, recognizable by their costume and the shape of their basketry”. It would go too far, however, to conclude that the tmoy remain only as essentialized folklore entirely emptied from any territorial and political content. Membership of a tmoy greatly inuenced the course of the 1945–1975 war at the regional scale, but to varying degrees depending on the geographic area and period. Khmu men got involved either with the communist troops or with the Royalist Troops backed up rst by the French (1945–1954) then by the US (1960 –1975). Local geopolitics have then undoubtedly contributed to dene relations between the villagers and the new State, their involvement in new local political structures, and their autonomy in the face of rural development policies implemented by the provincial administration. In other words, members of some tmoy have directly suffered from the new social and political order, while others have been able to prot from the dynamics of this transformative process. Between 1945 and 1975, almost the entire area of Laos became involved in wars at some time or other. Heavy bombing, artillery duels

24 On the issue of resettlement as the main tool for rural development policy in Laos, see Goudineau 1997 and 2000, or Évrard and Goudineau 2004. For a good overview of all the reports and articles written on that subject, see Baird and Shoemaker 2005.

154

olivier evrard

and infantry skirmishes were more numerous in the east along the Vietnamese border, but also occurred in the northwest throughout this period. Between 1946 and 1954, during the so-called rst Indochina War, battalions of Lao Chasseurs commanded by French ofcers were patrolling in the then-named ‘IVth Military Territory’ (covering more or less the current Luangnamtha and Bokeo province) to ght and disarm small ‘Lao-Issara-Vietminh’ groups (or LIV, in the language of the French ofcers). The latter were very mobile and were regularly attacking mule caravans (for opium), or Lao garrisons (for weapons). They found support amongst some of the highland populations, especially Akha and Khmu groups, among which they were diffusing nationalistic ideas and discourses. One interesting point underlined by some of the mission reports given by French ofcers to their superiors, concerns the methods used by colonial troops in the eld. They leaned on local hierarchies, and especially on the positive attitude of most of the panya towards the French, to progress in mountainous areas and to surround guerrilla groups. This strategy did not work everywhere, however, and some areas were entirely beyond the control of the French. This was notably the case for the Khmu rok of Müang Hun, known locally as kha khat,25 who were renowned for refusing any collaboration with colonial authorities. Interestingly, however, the Khmu of this area had welcomed and helped some French ofcers in 1945 while they were escaping the Japanese forces and helped organize guerrilla actions to regain control over the Lao territory (Caply 1966). It seems, then, that this Khmu subgroup, thought of by many as being more ‘backward’ than other Khmu, have always taken sides with guerrilla movements against regular forces. During the 1950s and the 1960s, a frontline progressively appeared on each side of the Nam Tha with the Nam Ha valley (tributary of the Nam Tha, on its right bank) as a northern limit. On August 13th 1950, a communist resistance front called Neo Lao Issara was created, subsequently renamed Neo Lao Hak Sat (NLHS) in 1955, along with a 12-point political program adopted by its leaders. From this time, military actions of communist troops were backed up with attempts to inuence directly the everyday life of the villages that they

25 This surname refers to the linguistic particularity of the Khmu of Müang Hun area who, or some of them, use khat as a negative word while Khmu living in Nalae or Vieng Phou Kha usually used phe.

interethnic systems and localized identities

155

had succeeded in controlling (e.g., initial attempts at agrarian reform, building of schools), and to spread Marxist-Leninist ideology amongst inhabitants. Fox (1986: 20) states that in 1953: “twenty-seven zones of operations had been established throughout Laos, mainly in frontier areas inhabited by tribal minority peoples, among whom Pathet Lao recruitment was particularly effective.” In northwest Laos, three zones covered some parts of the Viengphukha plateau, the core of the Nam Beng valley, and the area of Namo ( Deuve 1984: 35). In the following years, royalist forces took back control over the Viengphukha plateau, but the Nam Beng and the Namo areas came to constitute a single “rebel” zone, covering the Khmu lü and rok territories and entirely controlled by the NLHS troops ( Deuve 1984: 96, 166 and 210). The success of communist forces in these areas, and their ability to recruit many of the young men living there, can be explained by the conictual relationships between the Khmu lü or rok (and some of the Hmong populations of Namo) with the lam, whom I have mentioned previously. These lam were the Tai administrators—most of whom were Lü or Phouan—who were exercising their power over these areas in the name of the King of Lan Xang. The tax and territorial reforms introduced by the French in previous years had given more power to these Tai nobles, even though Tai villages were largely in a minority in these areas. The profound resentment against these reforms manifested itself rst in a kind of passive resistance (for instance, through a systematic underestimation by the highlanders of the number of inhabitants in each village in order that they should pay less tax), and was then transformed into open conict following the increasing presence of NLHS troops and Vietnamese military advisers. In 1953, one of the local Tai administrators of Müang Beng was killed in an ambush while he was trying to go to some Khmu villages, despite the warning sent to him by Khmu leaders who had let him know that they would not accept additional attempts to collect taxes or to search for escaped prisoners in their territory. In July 1954, a telegram sent by the governor of Luangnamtha to the French Commander of Louangphrabang, informing the latter that the mountainous area between the Nam Beng and the Nam Tha (the area of the tmoy lü and rok) was “100 % Viet ” and that “access to the highland villages is forbidden for lowland people” (Military Archives of Vincennes, le 10H5650). The political and military opposition on the two sides of the Nam Tha valley became increasingly obvious during the 1960s. Royalist

156

olivier evrard

forces settled several military camps near the right bank of the river and, following the takeover of Luangnamtha by communist troops in 1962, received much stronger military backing from the United States. Several Khmu lü and rok villages were bombed by the US Air Force during this period, and US Green Beret instructors actively recruited Lahu, Hmong, Khmu yuan and Lamet men of the Viengphukha plateau to train them in guerrilla ghting and commando techniques. Those paramilitary and multi-ethnic groups conducted several ambushes against NLHS troops, and even won back control of Luangnamtha town for a short period at the end of 1967, before being driven back by the NLHS and Vietnamese forces. It would be an overstatement to say that all Khmu and Lamet men living west of the Nam Tha river supported the actions of the royalist troops and anti-communist commandos. Commonly, only a few men in each village became soldiers, and their enlistment was motivated more by a pragmatic analysis of the local geopolitics, and sometimes by nancial motivation, than adherence to a real political project, such as the case of Khmu rok and lü in the eastern part of the valley. However, old cultural and political contrasts between the eastern and the western tmoy were brought ‘up to date’ and transformed by the course of the conict. A broad distinction was made by the new communist state between the vilason (heroes) and the satu (enemies) of the revolution, which still directly affects the relations of these populations with local administrations. One of the consequences of the conict in the Nam Tha area concerns the reversal of the ‘values’ or ‘images’ attached to the tmoy. Those who were considered as the most ‘enlightened’, because they were most inuenced by the Tai way of life, are treated as enemies by the new regime, whilst those previously considered the most ‘backward’ are given a new respectability through their association with the victory of the communist troops. Such a hierarchical reversal has obvious implications at the local level. For instance, in Nalae district, all the cao müang appointed since 1975 came from the tmoy lü,26 while at the 26 Things are changing, however, and for the rst time since the end of the war, a cao müang with a Khmu yuan origin was appointed in Nalae district in 2001. Some locals commented upon this nomination by saying that the period of ‘insecurity’ is now over on the east side of the valley and that the Pouthin area, where this man is from, will be a development focal zone in the next few years. Indeed, a track has been reopened there and it is now possible to link Vieng Phou Kha and Nalae in less than one day during the dry season.

157

interethnic systems and localized identities

From Namtha

NORTH

Key Khmou village Lamet village

tmoy kwaen

Lü or Lao village mixed village

Nam Tha tributary

Nam

Kh

Nam Tha

deserted site

on

recent resettlement zone, in-migrations still occuring district centre (4 villages: Vang Mi Sai, Done Sai Na Laeng, Konechan)

ng m Hae

Na

tmoy lü

tmoy yuan

Na

m

Peuane

tmoy lü Nam

tmoyyuan

g

Yan

tmoy rok tmoy rok

N

tmoy rok 0

10 km

tmoy rlmét Note : data gathered from education services of Nalae (1991 and 2002), 1995 national census and personal observations in the field (1995-2003) To Bokeo (Pha Oudom district)

c Olivier Evrard, 2005

Map 4: Resettlements and tmoy afliation in Nalae district

158

olivier evrard

provincial level, the Khmu rok have kept the most inuential positions. Between 1975 and 1998, two provincial governors were appointed in Luangnamtha and both of them came from Ban Mokkoud, a village from the left bank that was bombed three times by the US Air Force during the 1960s. In the neighbouring provinces, especially Bokeo and Oudomxay, Khmu rok men also received high positions in the provincial administration. Conversely, it is still unusual to nd highranking local ofcials recruited from Khmu yuan or kwaen, or even Lamet populations. Such political reorganization led to particular directions for policies on rural development, and, especially, ways in which the resettlement of highland villages was conducted. From the end of the 1960s, communist leaders offered material and logistical support to ‘allied’ highland populations who would agree to go down to the valleys and cultivate rice elds abandoned by those of the Tai-Yuan and the Tai-Lü populations who had ed to Thailand. Several hundred Khmu rok and Khmu lü families were then able to settle in the lowlands of Hun, Beng, Xay and Namtha districts. Some years later, the return of some of the Tai owners gave rise to land conicts in these areas. Most of the Khmu migrants had to give back their land, but they usually received some compensation from the provincial administration in the form of cattle, buffaloes or even rural development projects (Évrard 2002). During the same period, the continuation of an anti-communist guerrilla movement on the Viengphoukha plateau led to the authoritarian resettlement of numerous Khmu yuan and kwaen populations, along with Lamet, Hmong or Yao villages. Unlike the resettlement mentioned previously, these involved the forced relocation of entire villages, which were obliged to move under pressure from the Lao army. Most of these migrations took place between 1975 and 1985, and were especially numerous in 1977, when Lao troops, backed by Vietnamese forces, launched a vast security operation following the assassination of the governor of Viengphukha district. Most of the Khmu villages that are found on the banks of the Nam Tha river, or along the road near Viengphukha, were resettled under these circumstances. Unlike the Khmu rok or lü migrants, those who were forcibly removed did not receive any aid during the initial years, and most of them experience very difcult living conditions (Évrard 1997). The division of Laos into two opposing camps during the war led to a blurring of ethnic identities between different ethnic groups. In the case of the Khmu populations of the Nam Tha valley, these groups

interethnic systems and localized identities

159

drew upon previously existing intra-ethnic solidarities, which were ‘crystallised’ through their involvement with the war and transformed through political afliations with prevailing power structures in war and post-war contexts. The tmoy yuan and kwaen experienced brutal and signicant reductions in their territories, while the tmoy rok and lü have until recent times kept quite autonomous highland territories connected to the lowlands via familial and institutional networks. Conclusion This chapter has shown how Leach’s framework can still be used accurately for a contemporary anthropology of highland populations in Southeast Asia. If the concept of an ‘oscillatory model’ is misleading both inside and outside the Kachin Hills, the interactionist kind of analysis on which it relies appears well suited to the analysis of other groups, provided the specicities of the local interethnic contexts are taken into account. Political and cultural inuences of the lowlands bring deep changes into the lives and cultures of highland groups, but are nonetheless subject to a form of appropriation that is mediated and encoded through specic cultural and cognitive grids in order to be understood and adopted by the group concerned. Added to geographic, historic or economic features specic to each area, this leads to the creation of cultural discontinuities, specic ‘labels’ and intra-ethnic solidarities among populations that still claim a common identity. Brought about through historical processes, such phenomena can be well seen amongst the Khmu populations of the Nam Tha valley. Similar analysis can be conducted nearly everywhere in northern Laos, where hybrid ethnonyms are often derived from the Lao language to name highland groups or subgroups, either from Mon-Khmer or Tibeto-Burmese linguistic backgrounds. Instead of considering each of these populations separately, and trying to dene their cultural characteristics, Leach invites us to consider them as pivotal categories in a single interethnic context. What is especially interesting in the case of the Khmu is that, as in the case of the Kachin, such sub-groupings acquired opposing political meanings or values during the war. Thus, they allow a better understanding of how the Lao communist State came into being, and how it used and transformed, rather than erased, such intra-ethnic solidarities.

FROM KETTLEDRUMS TO COINS: SOCIAL TRANSFORMATION AND THE FLOW OF VALUABLES IN NORTHERN LAOS Guido Sprenger Like many debates in anthropology that have proved to be fertile over the decades, the one on Leach’s seminal tome (1977 [1954]) offers a plethora of perspectives. In order to open a window enabling comparison and critique, it is advisable to focus on specic issues and reduce the complexities of Kachin ethnography to basic questions. In this article, I refer to the discussion concerning systemic imbalance vs. historical contingency, using data from the Rmeet (Lamet) in Northern Laos.1 However, the subject does not stop short of discussions on ethnographic detail and interpretation, but expands to a more theoretical question: are changes in societies determined by the contradictions structuring local value systems, or by material forces that can be described in universally applicable terms of economics and power? This point was raised by a number of critical reevaluations of Leach’s book. John Donohue (1984) and David Nugent (1982) argued that the transformation of Kachin social organization could be explained by economic and political factors external to Kachin society. Victor T. King (1983) has pointed out that such an approach has two major disadvantages: rst, a systemic analysis enables comparison, while explanation by historical contingency isolates the Kachin case from a possible comparison of social structure and process; second, the empiricism of 1 Research was conducted in 2000–2001 and in spring 2002 in Luang Nam Tha Province, Laos, in the framework of the Research Group of the German Research Council ‘Cultural diversity and the construction of polity in Southeast Asia’ at the University of Münster. Additional research in 2005 was funded by the Institute of Ethnology, Academia Sinica, Taipei, and the Frobenius Institute, Frankfurt am Main. I want to express my gratitude to the members of the Institute of Cultural Research, Vientiane, its former director, Houmpanh Rattanavong, and Suksavang Simana, for making research possible. This paper was written during Postdoctoral Research at the Institute of Ethnology, Academia Sinica, Taipei. Earlier versions of this paper were presented for the research group “Rethinking boundaries: Historical and cultural processes at the Chinese Margins”, Institute of Ethnology, Academia Sinica, the 4th EUROSEAS conference, Paris 2004, and at National Tsinghua University, Xinzhu, Taiwan. I thank all discussants and Dr. Ho Ts’ui-P’ing for their suggestions.

162

guido sprenger

the “contingency” approach tends to veil some rather naïve theoretical assumptions of a mostly materialist type. Leach himself did not at all diminish the importance of external inuence, but debated its explanatory power: “The ultimate ‘causes’ of social change are, in my view, nearly always to be found in changes in the external political and economic environment; but the form which any change takes is largely determined by the existing internal structure of a given system” (1977 [1954]: 212). Seen from this point of view, it is not simply the structural imbalance of the hierarchized gumsa and the egalitarian gumlao systems that triggers the transformation of one into the other. Rather, shifts in economic and political conditions enable the acquisition of one model by the other. Yet, the models remain rmly local; thus, Leach describes the shifts in terms of the internal workings of Kachin social organization. He highlights the cultural and ideological conditions of the changes, an approach that a number of later authors confused with the denial of external inuence and the reduction of the social process to internal imbalances that produce determinable results. But this is not at all what Leach wrote. I largely follow Leach’s formulation quoted above in my interpretation of Rmeet (Lamet) data,2 although with some qualications: Leach can be read here as making a difference between the “form of changes” and their essence, supposedly represented by “causes”; this corresponds with a distinction of politico-economic vs. ideological features of society, with ideology always following or veiling power relations. This interpretation is consistent with Leach’s general approach, characterized by Kuper as seeing “the people’s model [as] a sort of screen behind which the actual competitive relations of community life are worked out” (Kuper 1983: 160; see also Leach 1977 [1954]: 16–17). But I argue that, the types of sociality of inside and outside used to determine the value of an act or a relationship, differ more sharply than Leach’s formulation suggests. He implies that external changes are cloaked with internal values and ideas in order to effect change, but I suggest that external elements acquire new meanings in the internal meaning system, while retaining some features that originally distin-

2 I cannot, though, go into a detailed comparison of my approach to that of Kirsch (1973). Kirsch had included Izikowitz’ Lamet data (Izikowitz 1979 [1951]) into his scheme for interpreting degrees of hierarchization and ritual feasting across the Southeast Asian mainland, but stressed somewhat different factors to determine the position of the Lamet than I do.

from kettledrums to coins

163

guished them (see also Sprenger 2005). Culture and ideology do not hide or distort social realities, but are the very media through which social change is achieved. Social change means to act upon a cultural interpretation of changed circumstances. So, while I think that Leach captures the dialectic of inside and outside correctly, I propose a clearer separation of external and internal factors, a distinction that seems warranted by the data. This is not to argue for a higher degree of isolation of the inside; in fact, the two spheres of sociality are complementary and, thus, indistinguishable. The terms “inside” and “outside” demand a denition of the type of boundary that separates them. It is important to remark that the terms are not merely analytical categories that hinge on conventional notions of cultural boundaries, but have emic validity. The boundary between them distinguishes categories of social and cosmological relationships, or, as I prefer, socialities, in which perception and agency are framed in the terms of a particular value system. This means that objects and acts occurring in an “inside” context acquire a meaning different from an “outside” context. Several inside/outside distinctions are found on what can conveniently be called levels of social organization in a specic geographic region: northern Laos and Thailand. I specically refer to the Rmeet village Takheung in Luang Nam Tha Province, Laos. The Rmeet are an ethnicity speaking a Mon-Khmer language and cultivating swidden elds mostly in upland villages. Three levels of social organization, all characterized by inside/outside distinctions from the Rmeet point of view, are important here: kinship groupings within Rmeet society; ethnic categories, in particular Rmeet and lowlanders; and states, in particular Laos and Thailand. At each level, the inside/outside relation is articulated in exchange. What representatives of groups give and take, what they receive and are deprived of, and how they do it, inuences the way the groups perceive themselves and their mode of reproduction (see Barth 1969, Lehman 1979). The denition of social entities by exchange has two aspects: the rst concerns exchanges between them; the second consists of exchanges that are exclusive to members of one category or group. These internal exchanges dene subcategories, parts of a whole, which form the larger, encompassing entity. What I want to illustrate here is the interplay of these types of exchange. The case in point is the ow of bronze drums and colonial silver coins within Laos and within Rmeet society: they are transferred from lowlanders to uplanders, thereby dening ethnic categories within Laos, whilst at the same time, at the state level, the

164

guido sprenger

transfer of coins is distinctive to Laos and former Indochina, in contrast to Thailand. Within Rmeet society, the coins move between houses as bride-price or are temporarily given to the dead at funerals; the rst is an example of inter-group exchange, the second denes a group of “brothers”. All these separate types of exchange reinforce each other and the systemic processes of which they are a part. The interplay of different categories of exchange provides clues to the specic understanding the Rmeet have of their own society and its history. This can be demonstrated by the comparison of Rmeet society in the 1930s with that of today. Politics and economy in the colonial era (1893–1954), the following Second Indochina War and the subsequent replacement of Laotian monarchy with a communist government in 1975, exerted a strong inuence on Rmeet society and its outside relations. Among other factors, these changes triggered the shift from kettledrums as major wealth and ritual items to French colonial silver coins. The shift signies the transition from a society with a class-like hierarchy to a supposedly egalitarian one. But, the (partial) substitution of kettledrums by coins is not determined simply by forces of politics and economy, but also by the meaning the Rmeet attribute to these objects in their exchange practices. The position I maintain here, is that both power structures and kinship practices are informed by the value system of a given society. External economic and political processes, indeed, do have a crucial inuence on these systems, but they do not produce universally unied effects. Rather, a particular value system produces responses in the form of social relations and acts that are intelligible in the terms of these values. Wealth from the Lowlands I focus in this chapter on two items of wealth, bronze drums and French colonial silver coins, both of which were acquired by the Rmeet through the lowlands. “Lowlands” are designated with the word yam, which covers Tai-speakers like the Lao and Thai, and which includes features of culture and society—state organization, wet-rice cultivation, Buddhism—and a specic type of superiority over uplanders. Drums and coins are not merely signiers of economic value, but derive their meaning largely from their use in rituals. Whilst bronze drums were considered high-status wealth objects in the past, their importance has declined today. Some of their signicance has been absorbed by French

from kettledrums to coins

165

colonial silver coins (Piastres de Commerce). The ritual use of both items is different, yet they share a number of features. The questions are: how do the uses of drums and coins relate to Rmeet society in different historical periods, and how did the change come about? Some of the answers have to be tentative. Ethnographic and historical data are insufcient to reconstruct what has happened in detail. The major source for the past is “Lamet”, Karl Gustav Izikowitz’ classic ethnography of Rmeet economics, rituals and society, rst published in 1951 and based on data collected in 1937 and 1938. Data on the present comes from my own eldwork, which includes the memories of older Rmeet people, collected in the years 2000 to 2005. The political and social events that occurred in the period in between not only helped to transform the ritual and social system of the Rmeet, they also shaped people’s present-day representations of their past. What the Rmeet say about how political and economic conditions affected their society is informed by the way they perceive their current relationship with the state and its neighbours. It is, therefore, not surprising that differences occur between present-day statements and the historical record. One of these differences involves the use of wealth items. Bronze drums helped reproduce a class society, while colonial silver coins are used in a society that claims to be egalitarian. Let us look rst at the use of bronze drums in the 1930s. Bronze Drums in the 1930s According to Izikowitz, Rmeet society of the 1930s showed traits of class distinction, with a particular group having access to limited resources, restricting access to their ranks and claiming positions of inuence. A number of male household heads were distinguished as lem, rich persons. As such, they held central positions in village decision making, the ning of misdeeds, the distribution of meat, rituals and the allotment of eld plots (Izikowitz 1979: 305). In particular, the latter supported the reproduction of their wealth: more productive plots that were easier to work increased the surplus of rice, the most important trade good of the Rmeet at the time. Status was conveyed by the other lem of the village in a ritual that resembled common blessing rituals. Strings were tied around the candidate’s wrists, and two chickens were killed. A new turban was wound around the new lem’s head; this latter act probably involved a mixture of familiarity and subjugation between the old lem

166

guido sprenger

and the new one, as the Rmeet usually avoid touching an adult’s head lest this might cause illness. The binding of wrists serves to x klpu, the personal “soul”, to the body. The lem status was determined by the actual wealth of a household: it could only be inherited insofar as wealth was inherited, and it could be lost due to impoverishment (ibid.: 116–7). The acquisition of wealth from outside Rmeet society, in particular by wage labour in Siam, reinforced the distinction of the lem from the rest of the population. Among the so-called Upper Lamet in the south west of the area, labour migration was common in the 1930s, and this went along with a stronger distinction between lem and non-lem: the lem as a group were practically endogamous. As bride-price was a major means of transferring accumulated wealth from one house to another, the lem preserved wealth by marrying exclusively among themselves (ibid.: 351). Thus, despite the instability of each individual lem’s status, the lem as a group attempted to cement their position by having privileged access to and control of resources, inuence over political decisions, and by practising endogamy. High-status persons also supplemented their positions by buying titles of nobility from Tai-speaking Lue. The values of this society stressed distinction in status and wealth. Characterizing the Rmeet in general, Izikowitz speaks of a “constant hunger for prestige”, satised by obtaining “bronze drums and other articles of luxury” (ibid.: 354). In this way, access to markets and labour opportunities in Siam and the lowlands interacted with Rmeet values and social organization. However, this does not mean that economic drives and wealth occupied the same position in Rmeet ideology as they do in Western notions of modernity. The articulation of status and its role in the reproduction of society was informed by the particular value system operative in Rmeet society. The central marker of lem status was the possession of at least two bronze drums. These drums originated from the Karen in Thailand, who traded them widely (Cooler 1995: 4). The drums belong to Heger type 3 (Heger 1902),3 a type with manifold ritual usages and widespread in Laos and beyond (for example, Goudineau 2000). The Rmeet also use the Karen word for the drums, klo. Siam was seen as the origin of the type of wealth that is focused in bronze drums, and

3 Izikowitz (1979: 330) identies them as Heger type 4, which is obviously a mistake. Both my own observations and those of Cooler (1995: 74) conrm that the drums are type 3.

from kettledrums to coins

167

the drums were mostly bought during labour trips there, especially by the Upper Lamet (Izikowitz 1979: 101–2, 351). One motive for many young Rmeet to work there was the nancing of their bride-price. Work in teak woods enabled contact with the Karen, who had played an important role in this industry since the late eighteenth century (Keyes 1979: 36). It is not clear how frequently the drums moved within Rmeet society. They were inherited down a patriline or sold. They were also a potential part of bride-price, though in no way obligatory. Other instruments, like gongs, seem to have been more common, although lem probably gave drums more frequently in marriage (Izikowitz 1979: 100–1, 108). Therefore, we can assume that there was some movement between patrilines after a drum had entered a Rmeet community. Still, the tie to a patriline was strong enough that it was preferable to destroy a bronze drum than give it to an outsider if its owner died heirless (ibid.: 108). Another type of wealth was the possession of buffaloes, although Izikowitz does not mention them as a measure for the ascription of lem status. Buffaloes were necessary to put the bronze drums into proper use: both were needed for the performance of rituals addressed to the ancestor spirits of the household, the drums in order to call the spirits and announce the ritual, the buffaloes as sacricial food. Buffaloes were, in the 1930s as well as today, obligatory parts of the bride-price (ibid.: 100). Both buffaloes and drums held central positions in the system of ritual reproduction of society. Households, especially those of rich men, were expected to perform a large sacrice for their ancestors annually. Illness in the household also provided an opportunity, although if it did not occur, sacrices were made after the harvest (ibid.: 243). The rituals involved the entire village community, as the village was closed to the outside during their performance. The meat was distributed to all villagers (ibid.: 305). Other rituals that restructured social relations were embedded in the larger ones: adoptions, lem initiations and the rituals for a newly acquired drum could all be performed during the days of ritual isolation (ibid.: 327, 331). Two of these additional ceremonies directly refer to the status hierarchy created through wealth: lem initiations and the rituals for a new drum. But, adoption relates to a different aspect of the ritual, the reinforcement of patrilineal relations. The ancestors addressed in the rituals are patrilineal ones, and adoptions are often motivated by the need to continue houses patrilineally. Izikowitz does not give details on the issue, but today adoptions are

168

guido sprenger

crucial when a couple has no son. In these cases, a male child from the same patriline, a son-in-law or a grandson, may become the successor. Thus, the buffalo rituals demonstrate the wealth of a house, especially the lem houses, thereby pronouncing the class structure. At the same time, they help to reproduce patrilineal relations that are indispensable for any house. The rituals thus conated the status difference with the reproduction of patrilines.4 The most forceful evidence of status, the ownership of the drums, also had a cosmological dimension: the drums have potentially dangerous spirits, which have to be appeased after acquisition and during rituals (Izikowitz 1979: 117, 243; for the closely related Khmu, see Lundström/Tayanin 1981). The importance of drums for the representation of ancestral relations at this time is still visible today: old gravestones of wealthy men sometimes display engravings of drums. These rare designs are often the only distinguishing feature of a grave that remains decades after the burial, when the person’s name has been forgotten. Today, only writing is used to single out an important man’s grave from the unmarked stones found on the others. As mentioned above, the buffalo sacrices of individual houses reinforced the village as an entity. The importance of the lem for maintaining Rmeet sociality is highlighted by the term used for commoners: to also means “body” or “animal”. This implies that, by contrast, only the lem are full human beings (ibid.: 116, 200). It is most probably their ability to perform rituals involving their village, thus maintaining proper relations with the dead, that assigned this quality to the lem. Thereby, individual houses came to represent the community as such by ritual, an effect found in other hierarchized societies in this area (Leach 1964, Kirsch 1973, Woodward 1989). This has also changed today: the taboos and restrictions, as well as the food distribution, only cover the house that sacrices and its extended family. The only rituals involving and closing off the village community are those for the village spirit. These do not stress prestige or the status of particular houses, and only a locally made, wooden drum is beaten. More than sixty years after Izikowitz, the society representing itself to itself by rituals stresses

4 On a comparative note, these feasts do not seem to have been used to expand a household’s sphere of inuence through exchange and the mobilization of a large network, as in Melanesian Big Man systems or some of the Feasts of Merit in other parts of mainland Southeast Asia (Leach 1977 [1954], Lehman 1989, Condominas 1969, Woodward 1989). Outsiders are excluded, and the meat distributed to all villagers.

from kettledrums to coins

169

equality over class distinctions. Among the means to achieve this is the ritual use of money. Multiple Currencies in Rituals Today Izikowitz does not report colonial silver coins as being specically ritual objects. Colonial money was used to pay bride-price, or nes for the breaking of ritual rules. Izikowitz also mentions that bamboo replicas of older types of money could be given as gifts to the spirits, for example after a successful hunt or before sowing (Izikowitz 1979: 196, 218). These replicas had the shape of decorated plates, 10 cm in length, and were called kmul, the generic word for money and silver (in my transliteration, gemuul). They probably represented the silver ingots of about the same size, called lim by my interlocutors. Izikowitz identies lim as copper coins, although he does not seem to have seen any originals (1979: 252). Yet, it is quite likely that the integration of colonial money into the ritual system started in his day, although it probably lacked the sophistication that it has in the present. This can be deduced from the fact that the colonial coins used today all date from the period between 1895 and about 1930. These coins, with their high silver content, ceased to be ofcial currency in 1932—several years before Izikowitz’ research was undertaken. At this time, the French Indochinese currency was bound to gold backing, and a new piaster, smaller and with a lower silver content, was issued (Touzet 1939: 133–4). Izikowitz does not mention any coexistence of the two different currencies. Yet, the piasters issued after 1932 seem to have completely disappeared from use, as has the kip of the Kingdom of Laos that replaced it after independence in 1954.5 At the time of my research, piasters were part of a system of currencies with diverse connotations, ritual uses, and validity. 1) Gemuul kesoong: cowry shells are the oldest known type of currency. They are not associated with any government, ethnic group or centralized organization, but were “money of the people” (gemuul pasason). They have been imported from faraway places (the closest

5 An older type of piaster, issued in 1885, had an even higher silver content, but as Laos became a part of French Indochina only in 1893, it probably never became common in the highlands (Thiollier 1930: 51–3).

170

guido sprenger

seashore is about 500 km away) and are now only used as ritual objects and gifts. On these occasions, in a manner subverting the quantitative character of money, the number of cowries is xed: in healing rituals, a single shell is given, while bundles of four replicas made of bamboo are used as gifts to the spirits. This means that the value addressed in these exchanges is not quantiable; they belong to a sphere of transactions separated from those involving quantication. 2) Gemuul man: ofcially inscribed “Piastre de Commerce”, these silver coins were issued by the French Indochinese government in the late nineteenth and early twentieth centuries. Some of the smaller denominations, such as gemuul thääb (50 cent) and gemuul selüng (10 cent), are only used as ritual objects. However, the large one piastercoins, weighing 27 grams, with a silver standard of 0.900—the gemuul man proper—can be employed as a regular means of payment. At the time of eldwork, the exchange rate was 50,000 Lao kip, which equals the selling price of 50 kg of rice, making gemuul man an expensive item. Their most important use is in bride-price. In case the bridegroom’s family is not able to raise the amount demanded, coins may be replaced by modern currencies. In this case, their market price is taken as the conversion factor. When wife-takers negotiate bride-price, they do not ask to lower the number of piasters but rather for a better conversion rate. The need for piasters in the highlands is such that forgeries have been coined, allegedly by Vietnamese manufacturers. These are made from lesser material and can easily be identied by buyers. As they are much cheaper, poor people occasionally buy them for healing or blessing rituals. The use of piasters is also reported from the Khmu (Evrard, pers. comm., Tayanin/Lindell 1999) and the Akha (Kaufmann, pers. comm. b), and it is likely that this applies to many groups in former French Indochina, although their function may differ widely. The coins are a common item sold on lowland markets patronized by uplanders. 3) Gemuul kip: the modern Laotian currency, issued in its present form by the socialist government since 1979, after the revolution. Due to ination, the denominations in circulation range from 100 to 5,000 kip. Bills of 10,000 and 20,000 kip were introduced in 2002. Kip are used in any trade or non-ritual (or rather, supralocal) payment, including taxes, although these can be paid in rice as well. Kip also gure in small amounts in rituals, but on these occasions they often demarcate a sphere of exchange different from the one associated

from kettledrums to coins

171

with piasters. One important example may sufce: during mortuary rituals, the agnates of the dead person display all their piasters on the dead body. Kip, on the other hand, are only seen as appendages of small, tree-like constructions placed near the head of the dead person. This signies that only piasters are valid in the realm of the spirits, while kip are not. Kip also offer an opportunity for the Rmeet to identify with the state. The 1,000 kip note depicts three women dressed as Lao, Khmu and Hmong respectively and representing the ethnic diversity of Laos. Although the Rmeet are not pictured, they clearly recognize kip as “the money of the Rmeet” (gemuul Rmeet). 4) Gemuul baht: the modern currency of Thailand is widespread in the Rmeet area, due to labour migration and trade across the border. Thailand, offering many more opportunities than Laos, attracts many male and some female Rmeet, sometimes resulting in permanent settlement there. Often, these border crossings are illegal, but, as mentioned above, this type of trafc has a long history. The money thus obtained is often changed into kip, as the higher baht denominations are unsuitable for buying cheap household items (the exchange rate of baht to kip at the time of research was 1:190–200). Baht are used for similar transactions as kip. Their ritual use, though, is restricted, as paper money features in rituals only in small amounts, and small denominations of baht are not in circulation. Still, this currency’s high status is occasionally mirrored in rituals: it may replace piasters in the mortuary rituals mentioned above, as it is valid in the afterlife and considered to be the “younger sibling” of the piaster. Piasters, Personhood and Social Organization Three types of rituals serve to elicit the special position of the piaster: the xing of klpu “soul” to the body in blessing and healing, the payment of bride-price, and the display of coins during mortuary rituals. These rituals have to be interpreted as part of a cycle, reproducing both persons and the part of social structure that is based on notions of kinship. To this end, a brief outline of social organization is necessary. Its basic unit is the house, which forms an architectural, social, economic and ritual unit. Most houses contain a couple with children and, in about half of the cases, a generation of grandparents. Outsiders can be incorporated by adoption and ritual. Houses are patrilineal and patri- or neolocal. They form mostly unnamed patrilines of two or

172

guido sprenger

three generations depth from the oldest living generation. Patrilines are organized around shared graveyards or grave clusters. They act as more or less sharply bounded groups in large rituals addressing the dead, in particular buffalo sacrices and funerals. The ancestors appear in two distinct forms, the benevolent, protective and de-individualized house spirit, and the potentially dangerous individual spirits of the dead who reside in the graveyard. Both entities are addressed in rituals that involve members of the lineage. These groups are exogamous, although they do not function as entities in marriage, which is decided upon rather at the level of houses (Sprenger, in press). The marriage system is based on asymmetric alliance, i.e. afnals are distinguished as wife-givers and wife-takers. Wife-takers are hierarchically subordinated to wife-givers: the givers are not only the origin of a wife, but the origin of the house founded or maintained by her and her husband; people refer to the origin of a man’s wife by saying: “His house came from that (wife-giver’s) house”. In rituals, wife-givers play a central role as reproducers of life. They provide their wife-takers with the ability to x life and reproduction to a particular place or body (Sprenger 2006b). However, independently of their performance in these rituals, they have the right to claim gifts and animal sacrices. The opposed values of afnity and descent—sometimes conicting, sometimes complementary—provide the dynamics for the reproduction of the Rmeet socio-cosmic order. Let us rst look at blessing and healing rituals, where piasters appear most frequently. Their name, dondeii (“to tie hands”), refers to the tying of white or coloured strings around the wrists of the addressees. They are performed when a person’s integrity is in danger or his relation to the groups to which he belongs changes: at the end of a healing ritual, after ancestor sacrices, before or after long journeys, as a welcoming for guests, when moving into a new house, and so on. The effect is to tie klpu to the body (see also Izikowitz 1941). Klpu is a personal invisible aspect of the person, as distinguished from a more impersonal life-force, pääm (“living movement”). Illness indicates that klpu has left the body. Upon death, klpu is transformed into what is called either klpu ii yoom (“klpu of the dead”) or phi ii yoom (“spirit of the dead”). Only humans, rice and buffaloes have klpu.6

6 The term “klpu of the rice” is not used in Takheung, where the rice spirit is exclusively called phi ngo. It is yet known in other villages. See also Izikowitz 1979: 243–247.

from kettledrums to coins

173

The addressees of a wrist-tying ritual are the person who is most affected by the occasion and often members of his or her household. On the other hand, there is hardly any restriction on who ties the wrists. Usually the equation applies that the more people attain, the better. One limiting factor is the size of the animal killed for the occasion; a chicken is killed on small occasions when only house members and their immediate relatives take part, while the killing of a large pig draws crowds, sometimes from beyond the village. At the height of a wrist-tying, the givers of the strings hand over of a bowl of gifts to the addressees. This bowl, called khan, contains the cooked animal or representative parts of it, besides other items: four cones made of banana leaves lled with owers, uncooked rice, two cups of liquor and piasters. Piasters gure prominently on these occasions, as they are placed on the animal’s body. Four is the number most often seen, but six or eight may be given as well; uneven numbers are ruled out. The addressees do not accept these valuables, but usually take only the banana leaf cones with them. Only if small amounts of kip instead of piasters are given will the addressees take the money. Then the addressees drink up the cups of liquor. The boiled meat is used to feed rst the klpu and then the body: those who tied strings earlier place small pieces of meat and boiled rice in the open hands of the addressees and pronounce blessing verses. This food only feeds klpu and is not consumed afterwards. A common meal concludes the ritual. In another part of the ritual, sometimes performed before the handing over of the gift bowl, the string givers dip the coins into the liquor and strike them on the palms of the addressees. These then smear the alcohol into their hair. This is done to enlarge their authority, to make others “afraid” (laad ) of them. At this moment, the coins are explicitly used to convey power and prestige. As in the handing over of the bowl, ownership of coins is not the issue, but the presence of them as objects that endow a gift with specic qualities is. In order to see how coins both maintain the klpu of the person and social organization, klpu needs to be understood as a relational entity. When a person gets ill, klpu has left the body, usually joining the spirits or a different house. The most prominent directions in which klpu goes represent the two major axes of kinship: descent and the wife-giving—wife-taking relationship. When the dead of the patriline demand a sacrice, they will at times take the klpu of a person and bring it to the graveyard. On the other hand, if somebody neglects or insults his wife-givers, his klpu will go to their house. In both cases, after a ritual healer has diagnosed the condition, klpu has to be retrieved through a

174

guido sprenger

healing ritual and wrist-binding—in the latter case, it is the wife-givers who receive the sacrice. The return of klpu is thus achieved by a ritual exchange that reinstitutes the proper relationship between a person and his hierarchically superior relatives: the patrilineal ancestors and the wife-givers. Piasters also play a major role in the constitution of the two axes of kinship. Their importance in alliance should be obvious: without piasters it is difcult to get married. In fact, any valid currency might be used as bride-price, but piasters provide the measure to evaluate the amount of money given and are, in fact, often used. The most common amount, given in a quarter of recorded cases, is twelve piasters. This number consists of two parts: four piasters form the khan, presented in a bowl to the living—in this case, the bride’s parents; the remaining eight piasters are called gab ña, which are given to the house spirit via the bride’s father, sometimes only on the day of the “big” wedding, sometimes already on the day of betrothal (“small wedding”), when the future husband moves into his bride’s house for several years of bride service. The act is sometimes explained as “attaching” the son-in-law to the house as a ritual unit. This implies that the klpu of the son-inlaw is now linked to the wife-giving house. It is the beginning of a relationship that may direct the klpu to the wife-giver’s house in cases when the son-in-law behaves disrespectfully. In principle, the relation works the other way around as well. The klpu of wife-givers could go to the wife-takers’ house if they had insulted them, but, in fact, this never occurs, pointing out the hierarchical nature of the relationship. The number of piasters in bride-prices often has a 1:2 ratio, with the larger part reserved for the house spirit. Sometimes it is even the case that all the piasters are employed as gab ña. The house spirit encompasses and protects the klpu of the residents of the house, and every time a person takes up residence there or leaves, the event has to be announced to the spirit by a pig sacrice. It is dened as “spirit of mother and father”, though it is not composed of particular named persons, but rather embodies a line of male ancestors with their wives. In this way, the descent principle enters the picture of matrimonial money exchange. Let us turn now to the third use of piasters, the mortuary ritual. As mentioned above, the dead appear in two different aspects. These spirits should be considered as relationships and not as bounded entities. Even in its common denition, the house spirit appears as a relationship rather than an individual person. At the level of ritual, this spirit

from kettledrums to coins

175

denes the unity of the house and maintains at the same time specic relations to other, similar units. These other houses are either agnatic relatives linked through common ancestry, or afnes whose klpu is linked to the house. Besides the house spirit, there are the phi ii yoom, the spirits residing in the graveyards. The graveyards are away from the village and situated in primary forest not used for cultivation; they are feared and avoided except for funerals. As long as the dead do not transgress their boundaries and the living do not enter the graveyard, the spirits are harmless. Only before a buffalo sacrice are the spirits called to participate. This is done at a place called kcuom on the fringe of the village, bordering the forest. Yet, sometimes the dead make an appearance in dreams or send a bad omen and, thus, express their desire to be fed with a sacrice. If angry, they may even appear as lightning that strikes a house or eld, provoking a string of elaborate rituals and taboo periods. In all these contexts, the phi ii yoom are individualized. They are located in single graves, are recognizable in dreams by their clothing and are called by name at buffalo sacrices. The fact that each dead person has his own “house” (ña ii yoom)—a wooden shack erected on the grave, complete with entrance, rewood, vegetable garden and a rice eld nearby—stresses the asocial character of the dead: while the house in the village is both an accumulation of people and a single unit related to other houses by descent and alliance, the “houses of the dead” are reduced to patrilineal relations in the graveyard and only have a single resident—something which would be considered peculiar and irritating if it occurred within the village. Piasters are used as gifts for both manifestations of the dead. In particular, the descent principle embodied by the dead in the patriline’s graveyard is constructed through money at the mortuary rituals. Between death and burial the deceased is laid out in state for one night in the house. During this phase of the ritual the body is surrounded by objects and persons representing his social person: relatives of all categories visit his house, as well as other people who feel linked to him. A rack is erected behind the head of the dead person, on which the silver ornaments of the family are displayed—a visible sign of social standing. The visitors belong to two categories: rst, there are afnes and friends bringing uncooked rice and an egg as gifts; the second category consists of the “brothers” of the dead. These are members of his patriline, or rather ‘grave cluster’, as well as more remote relatives and parallel cousins tracing the relationship via pairs of sisters. But, it is only the

176

guido sprenger

members of the grave cluster, including the house of the dead proper, who display their entire stock of piasters on this occasion. The coins, like the drums, are owned by houses, but are brought to represent patrilineal relations. All coins are placed on the body of the dead for a few hours and removed afterwards. They are given to the dead as a gift, to buy commodities in the spirit markets. The designation of the actors as brothers points to the descent principle behind this ritual act. Furthermore, the meaning of the ritual is dened by the Rmeet as based on symmetry: if one would not bring one’s coins to a dead “brother”, his house in turn would not come to do the same in one’s own house. A similar symmetry applies to the effect of the display: the dead person is expected to provide the “brothers” with more piasters in return, by making them successful and lucky. Symmetry characterizes relationships between patrilineally related houses in other contexts, as they are equated in respect to ritual roles. The hierarchic relations between afnes, on the other hand, are rather marked by asymmetric exchange in ritual. The display of coins, thus, reproduces the patrilineal unit and the values connected with it. In a manner similar to bronze drums, which are used to call patrilineal ancestors and are inherited, the objects become evidence of the descent principle. While the body itself is taken away the next day and buried, the display creates a kind of identication between the coins and the dead: although the Rmeet deny that the dead person is the permanent owner of the coins, they are brought into physical contact with him, much closer than any other object displayed at the funeral, and become his possessions at least for the time being. At the same time, the coins are not individualized, as the dead person is. They retain the quality that marks them off as money in exchange: they are widely convertible and may be used to reproduce many different relations. Any individual coin may be transformed into any other exchangeable good and may be replaced by any other single coin, without losing its value or changing its meaning in the process. The coins, thus, are external objects with specic features that are integrated into the ritual realm, under the condition that they retain the features that characterized their otherness in the rst place. In this regard, the coins as de-individualized representations of ancestral relations bear some resemblance to the house spirits, another aspect of the dead that remains within the society of the living. This conation of the identication of coins with the dead and the disappearance of individuality is illustrated by the statement

from kettledrums to coins

177

of one of my best interlocutors. According to him, large amounts of piasters have a spirit that consumes the dead during the display, thus making him an indiscriminate part of them. The display of coins at funerals also has an implicit function with regard to afnal relations. One of the effects of the superiority of wifegivers over wife-takers is their right to interminable claims for brideprice, even when the original recipients, the parents of the bride, have long passed away. But the funerals of the spouses—in particular of the last one to die—are the nal opportunities to ask for bride-price. On these occasions, the immediate wife-givers of the dead person’s house receive a gift consisting of a diminished bride-price: a hind leg of the buffalo killed at the funeral and three or six piasters (see also Sprenger 2006c). This gift publicly ends all demands, but the display of coins opens possibilities for future marriages: the coin stock becomes visible to visiting wife-givers, who, according to the marriage system, should provide brides to coming generations of the house; thereby, the ability of the house to pay future bride-prices is evident. Seen from this perspective, bride-price acquires additional meaning: with the coins, a representation of the dead is exchanged for a wife. Life gone by is turned into future life. Making the coins a gift to the house spirit of the wife-givers links the two principles of social reproduction. The klpu of the son-in-law, being part of the house spirit in his house, is also represented in the wife-giving house. Here, the descent principle (represented by the coins) is tied to the alliance principle. By this, the circulation of money feeds and renews the fundamental relations constituting Rmeet society; and it is exactly those features of money as a generalized medium of exchange detached from individual relations that renders this circulation possible. Yet, the concept of money differs from most modern-Western usages. The Rmeet use colonial coins to create and continue ongoing reproductive relations. But the coins also have the potential to isolate and bind, to cut off relations. One example is the nal gift wife-givers receive at mortuary rituals. Furthermore, relations that are potentially dangerous may be cut off by giving coins. At the burial, a 10-cent coin (gemuul selüng) is placed in the dead person’s mouth in order to prevent his klpu from talking to the living, leading their klpu away and causing illness. A similar idea is expressed by giving a coin to highly respected visitors or relatives from outside upon their leave: as their klpu is close to the village, it may stay there instead of moving away with the body; in order to avoid this, colonial coins of any denomination are given as a

178

guido sprenger

goodbye present in the form of pendants. This keeps klpu in the body, and the gift is called “to give klpu”. In both cases, the coins are used to x klpu in its place, but the condition for this is the removal of the coin from circulation: the coin for the dead remains in the grave, the pendant coin is pierced and, thus, devalued for exchange. From Kettledrums to Coins Today, possession of money is the central measure of wealth, among the Rmeet as well as among other inhabitants of Luang Nam Tha province (Kaufmann, pers. comm.),7 while bronze drums do not contribute much to a household’s status as being wealthy. What both items share in their respective historical context is not only the role of outstanding wealth item and ritual object. Both are viewed as central media to reproduce patrilineal, as well as afnal relations and hierarchies, with piasters occupying a privileged position among currencies. One feature that links drums and coins, while setting them apart from other forms of wealth like buffaloes or stored rice, is their limited visibility. Drums used to be hidden away, even buried in the forest, only to be taken out at funerals, buffalo sacrices for ancestors or the harvest (Izikowitz 1979: 243). In particular, the buffalo rituals for the ancestors were major occasions at which all wealth items were revealed to the community, including gongs and expensive, imported clothing. (ibid.: 330–1). Today, much less effort is made to hide the drums: they are usually placed in the sleeping room of the family, which may be open to guests or at least can be looked into, or in rice granaries. Although drums are still used and regarded as extremely valuable, they have lost some of their importance. Few present-day owners bought them in their lifetime; most drums are heirlooms. Although debates between brothers about the allocation of the drums belonging to deceased fathers occur, the instruments do not change hands the way they did in the 1930s. Furthermore, drums seem to have lost their obligatory role in rituals. When slaughtering a buffalo for the house spirit, the household head will hang up and beat a bronze drum in his house if he or one of his brothers owns one. If not, he uses gongs and cymbals, which are less 7 According to surveys conducted by Sylvia Kaufmann in the framework of the GTZ Food Security Project, Müang Sing, Nalae branch, 1997–2001.

from kettledrums to coins

179

valuable. Contrary to what happened in the past, this does not seem to make a difference. Since the mid-1990s, the district administration has been registering the possession of drums. Their sale is only possible with the administration’s permission. This indicates another shift in the meaning of drums: what have been objects of personal power and wealth are now items of national heritage. Yet, silver coins are hidden from public view, usually by placing them in bags or containers in the sleeping room or in rice granaries. They are not registered, and knowledge of their numbers is restricted. Like the drums in the past, they are only displayed in rituals involving fertility and the ancestors. The act of revealing objects of wealth has a comparable function. Both constitute a group of agnates by relating it to their dead. The dead themselves are hidden in the forest, in a place much feared and entirely secret in the past (ibid.: 105). The drums were used not only to call the dead to the buffalo sacrice, but also the living. The coins serve to activate and embody the patriline during funerals. The visibility of the drums and the coins gives evidence to social structure, including hierarchies of two types: the superiority of the wealthy over the poor, and the values at the core of the kinship system, with the ancestors presiding over the living and the wife-givers placed above the wife-takers. But, how did the change come about, and how does it mirror more general changes in Rmeet society? One reason is the changing sociopolitical circumstances of the Rmeet and their neighbours. The manufacture of drums in Burma ceased in 1924, but the effects of this were probably not immediately felt in Laos (Cooler 1995: 2). The war in Laos precluded the transport of valuables across the border for several decades. From the 1950’s to the 1980’s, trade was severely hampered by anti-communist governments in Thailand, boundary posts being often closed (Oldeld 1999). Another effect of the war was the displacement and migration of many highland communities. This not only lead to poverty and the selling of drums for money, like in the village Sepriim where all drums were sold before the entire village ed the war-torn area in the late 1960s (see also Cooler 1995: 4), it also stimulated the uplanders to exchange the bulky instruments for a more handy type of wealth, easier to transport and to hide, but which retained its value. This was the case with piasters. The preference for old piasters as wealth items over new ones or kip predates the war, but the historical development certainly supported the emergence of their highly complex role in ritual. What

180

guido sprenger

the drums lost in meaning and complexity, the coins acquired, though not in the sense of immediate translation: the coins became vital parts in the transformation of the older system. Another factor was the more equal spread of wealth within Rmeet communities. The Rmeet of Takheung do not know the term lem or any equivalent word for an upper class. Nonetheless, when talking about their past, they relate the distribution of wealth as the most important difference between past and present society. According to them, wealth in the past was concentrated in a few houses. These had lifted themselves from general poverty by working for the colonial power. In the view of the Rmeet of Takheung, the French colonial occupation appears as the rst and major source of wage labour and, thus, wealth imbalance. The importance of labour trips to Thailand is generally ignored in these accounts. The concentration of wealth was reinforced by a system of credit with high interest rates, granted by the rich houses to the poorer ones. High rates of interest are still extant in such loans—often 100% per annum—but today they are seen as less disruptive. Today’s society is described by many Rmeet as one of equality of wealth. This does not mean that wealth is, indeed, equally distributed, but it certainly implies that the value ascribed to wealth difference has changed. In the society Izikowitz describes, wealth difference was highly valued, and the wealthy institutionalized its maintenance by the rituals surrounding lem status. Today, there are many ways to display wealth, both ritual and non-ritual, but these displays are not used to maintain an explicit distinction between the wealthy and the rest of the community. What is more, the Rmeet hesitate to discuss individual wealth differences. While my interlocutors would willingly identify the rich houses of the past, they were hard pressed to name them for the present; typically I was told to look out for them myself. This way of muting status differences already existed in the past, as Izikowitz’ account suggests (1979: 117, 295), but it has gained prominence since his time. The present view of society is informed by the rhetoric of the revolutionary communist government, which claims to have abolished inequalities both between the rich and the poor and between upland minorities and the lowland majority. The government also discouraged displays of wealth in its formative years (Evans 2002: 177). Government rhetoric can be glanced in the statements that blame colonialism for the introduction of wage labour and wealth distinctions, obscuring the historical labour trips to Thailand. Yet, many Rmeet I talked to did not refer to the revolution as the major agent of change. Instead, they

from kettledrums to coins

181

stressed travel opportunities, growing condence and communication within Laos and between rich and poor. One interlocutor put it this way: Before, the poor were very poor, the rich were very rich. Today they are about equal. They see each other, they talk to each other. Before, they didn’t talk. The rich kept to themselves knowledge about how to do things (become rich), those who didn’t know were poor. (. . .) Now people observe each other, everybody does it, all are rich. (. . .) The people of the village decided to do like this (not the government).

What is important here are the two different models of society employed by the Rmeet, both of which they attribute to their own agency. The difference between them corresponds with the wealth items that represent them. One features strong wealth differences, manifest in drums that are themselves extremely expensive and, thereby, create a conspicuous and stark status divide. Bronze drums are objects made to be played and displayed. By their unique sound and size they effectively embody their concentration of economic and ritual value. Yet, in regard to exchange cycles, they move slowly: hardly an opportunity occurs that is important enough to transfer them. The other model is viewed as supporting equality and works with silver coins, items that allow for more nely graded distinctions between rich and poor. Although wealth differences are still very real in Rmeet society, the means to measure them allows more continuity. Furthermore, piasters are in intense circulation. Their use is much more diverse: they function not only as parts of bride-price and mortuary gifts, but also as a currency in the market economy. Each of these usages is xed to a particular context that conveys a specic meaning to the coins. Yet, the change from one model of society to the other does not mean that the Rmeet have shed their socio-cosmic values in order to model themselves after homo oeconomicus. The acquisition of wealth has not become more or less important as a motive, but the use of wealth, the way wealth items acquire meaning, inevitably expresses a specic value system, realized among others in rituals and kinship relations. Conclusion: the Regional Context A number of points have to be addressed in these concluding remarks: the distinction of Rmeet society and its mode of reproduction, as described above, from outside sources of wealth; the integration into the

182

guido sprenger

state via wealth, and the complementing distinction from other states, specically Thailand; and nally, the question as to how far Leach’s Kachin model helps us to analyse the Rmeet case. I have dealt with the complex relation between the Rmeet and the lowlands elsewhere (Sprenger 2004, 2006a). Sufce it to say that the internal reproduction of Rmeet socio-cosmic order depends on the integration of external wealth: bronze drums, money, buffaloes. Yet, the exchange spheres internal and external to Rmeet society are kept distinct. This is exemplied in the widespread origin myth of the money tree, which tells how drums, coins and other external products came into being among the Rmeet, through a distorted application of ancestral relations. However, the tree endangered the entire society and had to be cut, its top falling into the lowlands. The myth stresses the necessity to separate spheres of wealth production and ritual exchange (Sprenger 2006b). Thus, while the ultimate origin of wealth items is mythically located within Rmeet society, one of their distinctive characteristics is their integration from the outside—a feature that seems to be crucial for their effectiveness in the ritual exchanges.8 Today, bronze drums have lost their central role in reproducing a stratied society, and their foreign origin is no longer referenced in Rmeet accounts of them. While Izikowitz makes it clear that the drums had been imported from Thailand, some of my interlocutors claimed that the Rmeet actually knew how to make them and produced them themselves in the past. At the same time, the French origin of piasters is clearly recognized. However, even when the Rmeet distinguish inside and outside uses of wealth, money still works as a mode of integration into a larger entity, the state. The most prominent use of piasters in the past, as reported by the Rmeet today, was the payment of taxes to the King of Laos. It is not clear to which currency the Rmeet are referring in these memories, as the coins they use today were abolished in 1932; yet, they refer to the tax payments with the unequivocal term for piasters, gemuul man. The coins were checked for their authenticity in the village, and then brought to an administrative centre by village representatives. Although the heavy taxes levied during the time of the kingdom are spoken of

8 It is interesting to note that even for the Karen, the drums had a foreign origin. Originally bought or introduced from some unidentied people called Yu, in the nineteenth century the drums were entirely manufactured in the village of Nwe Daung, a Shan isolate among Red Karen settlements. The Shan smiths, who did not use the drums themselves, made them exclusively for the Karen (Cooler 1995: 51–56).

from kettledrums to coins

183

today as burdensome, they integrated Rmeet society into the larger unit of Laos. As mentioned above, the piasters were and are used by a number of upland societies in rituals, thereby creating a region of shared ritual devices. But, with regard to the Rmeet, integration was not only achieved by a common currency in a unied territory, as modern notions of state and money would assume. The Rmeet conceive the relation between king, society and currency differently. They claim that the King of Laos was of Rmeet origin. Thus, the king was at once a brother of the Rmeet and belonged to a higher level of organization. Drawing money from all places in the kingdom was one way to establish him—in the eyes of the Rmeet—as the center of the polity. This also indirectly linked the king to the exchanges constituting Rmeet village society: mortuary ritual, the xing of klpu and marriage. The—historicized—relation with the King of Laos and the use of coins today serve as a means to distinguish the Rmeet as inhabitants of Laos from Thailand, the most important national difference perceived by the Rmeet. The ritual use of piaster coins is one feature of this difference. The use of kettledrums rather pronounced the closeness of the areas, while the closing of borders during the war years and the changing wealth concepts emphasized the distinctions—a development that parallels the elaboration of Lao national identity. Thailand and Laos are also compared with regard to the value of their money, thereby clearly pointing out the differential meaning of state and leadership for the monetary exchange. The Rmeet explain the higher value of the Thai baht compared to the kip: the baht has the image of the living king on it; kip notes show either no specic person or the former president Kaysone Phomvihane, whose image is seen on bills issued after his death (see Evans 1998). These ideas all emphasize the Thailand-Laos distinction in terms of the Rmeet exchange system. In what respect does Political Systems of Highland Burma help to elucidate the Rmeet case? To employ the notion of “egalitarian vs. hierarchized” models of society is problematic: to equate the past version of Rmeet society with gumsa and the present-day practice with gumlao reduces the specics of the two cases to a contrast stemming from modern-Western political theory. Leach has presented the Kachin in relation to Shan principalities, while relations with bounded state-type polities—both during the kingdom and the communist government—have been crucial in the Rmeet case. There are also differences that touch upon the very systematics Leach analyzed for the Kachin. I have found no

184

guido sprenger

indication that wife-giver superiority in asymmetric alliance has been a factor in the emergence of the Rmeet class distinction. While long bride service and high bride-prices certainly contributed to the wealth of specic houses, its major source was wage labour and interest on credit. As for the ethnography, it is not altogether clear how marriage asymmetry leads from a relative hierarchy of ritual and values to the encompassing, absolute hierarchy of class and nobility; indeed, this link has even been debated for the Kachin (Maran, this volume). Also, the description of gumlao as egalitarian has been questioned. Lehman suggests that it is rather the sources and uses of prestige among the nobility that differentiate the Kachin systems (Lehman 1989). In the case of the Rmeet, the formal power of the lem remained restricted to the village, as was the social network of the buffalo sacrices. These restrictions on the reproduction of lem probably kept the lem system from developing into one with ranked descent groups, noblemen or chiefs. It is rather the difference in the use and valorization of wealth items that indexes the distinction between the two Rmeet systems, in a manner reminiscent of Ho Ts’ui-p’ings analysis of Kachin wealth ownership (this volume). Yet, both cases suggest comparison on a more general level. Both Kachin and Rmeet societies are equipped with opposing values regarding status distinctions. Their value systems should, therefore, not be considered as rigid structures but as driven by contextual and historical tensions between values (see Dumont 1980). These values shape dominant perceptions of one’s own society according to specic circumstances, which involves the integration and perception of neighbouring societies, their products and features. The Kachin value system allows for centralized and dispersed foci of ritual status and prestige, while the Rmeet system either stresses wealth distinction until it encompasses thevillage community (as when houses perform sacrices that affect the village) or suppresses the institutionalisation of such differences in ritual. In both cases, the relation between wealth and ritual is a major indicator of the respective forms of the value system. In any case, the dialectics of external inuence and internal interpretation and integration make Leach’s work an exemplary starting point for understanding the transformation of Rmeet society. In the course of history, the changing accessibility of wealth items went along with a shift in values concerning wealth and status differences. These values become manifest through specic Rmeet understandings of external and internal developments, attributing them (the values) pre-eminence

from kettledrums to coins

185

or subordination and a specic form according to the situation. In the days of the French-ruled kingdom, the availability of expensive ritual items, uneven access to wage labour and a state ideology that stressed the superiority of noblemen over commoners and of lowlanders over highlanders supported the public distinction of status. A government rhetoric stressing equality, the spread of silver coins and the disappearance of many drums intersected with an inclination to muting status distinctions. Yet, the reaction of the Rmeet to changes in the economy and in politics is specic to their system of representations, as shown in the ritual use of coins and drums respectively. There is imbalance in the system, which is at the same time the source of social dynamics; yet, the imbalance itself does not determine the when and how of social transformation. Specic value systems allow a range of different forms of society that are realized through interaction with economics, politics, and the way these are communicated through neighbouring societies. Changes in these external factors only contribute to social transformation insofar as they make sense in terms of the values in current use in a given society. It is Leach’s achievement to have pointed out this process.

POLITICAL HIERARCHICAL PROCESSES AMONG SOME HIGHLANDERS OF LAOS1 Vanina Bouté The principal themes used by different authors to comprehend Leach’s theory of socio-political oscillation between gumlao and gumsa societies are often given as land tenure or generalized exchange (cross-cousin marriage). However, neither of these criteria are valid to explain the existence of two types of social organization among the Phunoy: the rst one is a society organized according to the Tay Lu’s meuang system;2 the second one is nothing but autonomous villages with egalitarian lineages. I propose here a critical examination of the theory of sociopolitical oscillation in the light of the two kinds of social organisation displayed by the Phunoy. I will focus principally on the scattering of villages and the pressure on property to ask in what ways the proposals formulated among the Kachin and the Naga are not valid amongst the Phunoy, and why. Phunoy History: The Integration of Highlanders into the Lao Kingdom For the most part established in Phongsaly province, at the extreme north of Laos, the Phunoy comprise a group of approximately 35,000 individuals and speak a Tibeto-Burmese language. The mountainous zone of Phongsaly district is crossed from north to south by one of the most important rivers in northern Laos—the Ou river- and it is in this exact area where the Phunoy can be found. They are stock breeders and still practise shifting cultivation, despite the fact that their

1 A note on transcription: since no ofcial system of transcription of Lao or Phunoy into English script exists, I have used one that tries to reproduce as closely as possible the Lao pronunciation with sounds that exist in English. To simplify, I did not differentiate the two types of vowels, short or long. The origin of the word is specied, when a Phunoy word is used, with a ‘P’; words in Tai are indicated with a ‘T’. 2 The meuang is a fundamental territorial and political unit of the Tai populations, it can be translated as ‘‘principality’, and can refer to both the territory and its centre (the town, the village). For a denition of the meuang, please refer to J.-F. Papet, 1997.

188

vanina bouté

livelihood has been called into question by Lao national policies in the last ve years. The “Phunoy” (“small people” in Tai) entity is actually the result of the political and territorial remodelling of the area by the neighbouring Thai populations from the nineteenth century onwards. In the nineteenth century, due to repeated disturbances, the realm of Luang Prabang began to take interest in the territories and populations situated in their connes. Luang Prabang was, indeed, affected by various conicts: insurrections in the upper regions of the Ou river; internal wars in the principality of Sipsong Panna; invasions by armed gangs from Yunnan (known by the name of coloured ags, “black”, “yellow”, etc.) coming along the Ou river. The Siamese authorities encouraged the king of Luang Prabang to launch a major control policy in his territory through the political reorganization of the meuang satellites and the conrmation of the powers of governors and other competent authorities in the area (Smuckarn and Breazeale, 1988: 59). The inhabitants of the bordering mountainous areas were also the object of integration measures, not only because their control depended on the prosperity of the local Lao elite, but also because these border populations constituted a potential barrier against aggressions from neighbouring principalities. The border zones were reorganized through the bestowal of a special status on the mountain groups and, in addition, from the middle of the nineteenth century, the Phunoy language groups were also nominated as border guards. This was accompanied by the handing over of written documents called “The Books of the Land” ( peum kongdin, T), which established the limits of the territories for which the groups were responsible. The people in charge of guarding these documents also received the title of “Lords of the Land” (chao thi din, T). However, the territories drawn up in the “Books” (which shall henceforth in this chapter be referred to as domains), until 1950, were of varying dimensions depending on whether the village at the centre of the domain was on the left or right bank of the Ou river. The left and right bank villages had different external political orientations. Right bank villages had closer connections with the Tai Lu, a Tai federal organization known as Sipsong Panna with whom they had privileged relationships, whereas left bank villages fell under the Lao authority. On the left bank, each village received a manuscript and, therefore, became a domain in its own right. On the right bank, a domain was composed of an “elder” village ( pi, T) and the villages that were said to have evolved from it.

political hierarchical processes

189

The differences between right and left bank were noticeable on political, ritual and territorial levels. On the right bank, each village was composed of a certain number of localized clans. Among these, most of the ritual and political functions belonged to members of the founding clan, as they could associate their prior history with the territory. Members of the founding clan were given the right to better matrimonial compensations for their daughters’ marriages, as well as ownership of the best lands for shifting cultivation—and these lands were hereditary. On the domain level, two doyens, called “Lords of the Land” were entitled to some further specic prerogatives, including the right to lead rituals on behalf of all the villages of the domain (i.e. the domain of the clan of the doyen and, therefore, the clan considered the founder of the oldest village). On the left bank, each village was also composed of several clans, but all of these clans were believed to be founders of the village. Despite the fact that the Lao King of Luang Prabang had handed over the “Books” to only one of the clans, the doyen of each clan was called “Lord of the Land”, and these doyens celebrated all together the collective rights devoted to obtaining fertility for the elds and for livestock. Despite the fact that (as on the right bank) the village’s lands were divided between each clan, the lands were not owned by anybody in particular and all the families were free to cultivate lands located on another clan’s part. Several decades after their nomination as border guards and the handing over of “The Books of the Land”, the differences between the left and right bank populations were further consolidated by the French colonial power, established in this area since the beginning of the twentieth century. On the right bank, a large number of territorial and political restructurings, together with the ennoblement of many dignitaries, perfected the similarities with the Tai meuang: the inhabitants of this territory, called “Meuang Phunoy”, were notably exempted from taxes and chores applied to the neighbouring Tai population and were in charge of collecting their own taxes via their ennobled chiefs, called by the Tai title of “Panya”. On the left bank, however, the villages were not integrated by such changes. Even if the right and left bank populations endured similar policies, such as simultaneously being appointed border guards, being given the same “Books” and having some of their members entitled “Lords of the Land”, the consequences were quite different on either side of the Ou River. My aim is to understand the reasons for those differences, and I will attempt to show later on in this chapter how those differences

190

vanina bouté

arose. I argue that they were not a consequence of the left bank villages being isolated from the restructuring of the Meuang Phunoy undertaken by both the royal power and, subsequently, by the colonial government, but were the result of differences in the political territorial organization between the villages located on each side of the river Ou. These differences stemmed from a period before these populations fell under the inuence of the Tai population, and they seem to be the result of right bank villages being already organized according to a model quite similar to the Tai one. As such, the dominant powers seem to have preferred to rely on them for the constitution of the Meuang Phunoy. In order to address this issue, we should rst consider the following question: on the right bank, why do domains composed of several villages, dominated by members of the same clan, exist, while they do not on the left bank? The phenomenon I am about to address happened in a rather ancient period, for which there are no available documentary sources. Therefore, I can only proceed by adopting a hypothetical, deductive approach, mostly relying upon the analysis of authors who have worked on neighbouring populations and whose research has been concerned with the comprehension of similar or identical phenomena. By comparing the situations described by these authors, I make the supposition that a number of additional factors seem to be at the root of the creation of right bank possessive domains, as well as the hierarchization process seen there: demographic and territorial pressure, and the practice of scattering on the right bank. The Economic Data Examination of the differences in economic data is the simplest way to explain why a phenomenon of differentiation between individuals, lineages and clans happened in the right bank villages. Tannenbaum (1989: 69), for instance, explains differences in the political organization of lowland Tai societies (the hierarchical) compared with those of the highland populations (acephalous) by the following: in the mountains, the great uctuation of yields (dependent up amounts of precipitation), the number of workers available to clear a piece of the forest, etc., would represent an unstable base from which sufcient surplus required to secure a stable government could be produced; in the lowlands, the production of such a surplus would be ensured by the existence of

political hierarchical processes

191

regular agricultural production, which was in itself the result of those populations being settled upon territory and owning their own elds. Phunoy society on the right bank appears to be an intermediate case between these two situations. This is due to the fact that the Phunoy population became xed in the territory where they settled, as well as the fact that lands would be denitively attributed. Is this the nascent origin, therefore, of that process by which unequal relations emerged? Yet, it seems unlikely that the harvest would have allowed the collection of any surplus, or even that those harvests would have ever been steady; in the 1950’s, Harvard-Duclos reported that the lands were still unfertile, despite lying fallow for 50 years, and that only scrub thrived (1959: 8). Nevertheless, in spite of the soil’s aridity and the variability of the yields, the Phunoy chiefs could always count on good harvests, as they could depend upon free manpower and could, therefore, produce more than the household consumed and could even release some surplus. Signicantly, also, this enabled them to monopolize the ritual functions for sustained fertility. Furthermore, Tannenbaum’s analysis does not match the case of the Phunoy. For Tannenbaum, the surplus was the result of favourable ecological conditions and it only consequently allowed some individuals to grow rich and to claim any power. For the Phunoy, on the contrary, it was because the chiefs already had the authority that allowed them to have the population working for them that they had the possibility of obtaining a surplus, despite unfavourable ecological conditions. Nonetheless, it is interesting to remember the fact that the lands were privately owned. This would prevent any redistribution and allow families who owned the most fertile land to gain some surplus regularly; this was a factor that was likely to inuence the emergence of group stratication processes. As we will see later, in the left bank villages, where the lands are not appropriated, the organization of the village is relatively egalitarian and the clans do not maintain hierarchical relationships. On the other hand, I will try to show that the xation of the population on a territory (which, to me, implies to stop looking for new places to move and clear for shifting cultivation) seems to be a factor leading to a greater equality of the clans, contrary to Tannenbaum’s assertion. The wealth gained by an individual, lineage or clan, which would have allowed the latter to establish a differentiation between themselves and others over the long term could yet stem from factors other than agriculture. According to Donohue (1984: 70), the power of the Kachin

192

vanina bouté

chiefs living in societies organized under the gumsa3 political mode relied upon the exploitation of opium or precious metals; exploitation that would allow them to increase their wealth and to legitimize their authority. Nugent (1982: 516) shares this opinion by asserting that the reason for the weakening of the Kachin chiefs’ power at the end of the nineteenth century was a consequence of political and economic changes at the country level. It was, indeed, by preventing the Kachin chiefs from collecting taxes from caravans, practising slavery and rendering them unable to lead raids against the population of the valleys that the British colonial government greatly undermined the chiefs’ authority and then, consequently, gumsa social organisation in Kachin society. I can wonder, therefore, if the Phunoy chiefs controlled the trade of some valuables, as the Kachin chiefs used to do, thus gaining a superior economic status upon which they would have based their authority. However, research on this issue does not indicate the control by the Phunoy of any valuables allowing such an evolution. The Phunoy have never been important producers of opium; cultivation of this crop was mainly for self-consumption. Furthermore, in this region, the French military’s attention was never drawn to opium cultivation, even though the colonial power was eager to control the opium trade: this bears witness to the low volume of exchanges linked to the opium trade amongst the Phunoy. Neither was there any ore, the exploitation of which would have provided the Phunoy and their chiefs with the kind of prots that some Kachin could get by extracting jade. Although much ore can currently be found in Laos, in the Phongsaly region neither precious metals nor precious stones4 could be found at that time. The exploitation of salt, a unique precious mineral resource endemic to the region, belonged to the Tai Lu. As for trade, it was mostly controlled by the Yunnanese caravans that travelled throughout the region. Therefore, there were no locally found resources available, the trading of which the Phunoy could monopolize.

3 Organization dened by E. Leach as an ‘aristocratic’ type of organization with a politic entity represented by a territory called mung (c.f. the Shan’s mong), governed by a prince of aristocratic origin called the duwa and with the title of Zau (cf. the Sao of the Shan) ([1954] 1972 : 83–84). 4 Refer to A. Bernard (1990) on that subject. The author, who had established a record of the ore for each province of Laos, writes on the subject: “a Dutch, G. Van Wuysthoff, proved that Laos could easily rely on its own metallic products, by analyzing the trade possibilities with Laos, in 1641, selling Indian carpet in exchange for lacquer, gold, and honey. Yet, at the same period, its Company of the Dutch Indies exchanged copper and lead with Cambodia and Annam” (Bernard, 1990: 18).

political hierarchical processes

193

The Phunoy chiefs were unable to control either the trade in goods or the routes through the mountains. While noticing that it was, indeed, control over the trade routes that rapidly brought power to the Jinghpaw (Kachin), allowing them to become satellite feudal subjects of the Shan princes instead of their serfs, Evans reaches the following conclusion: “In Indochina the comparable groups were enabled to gain control of trade routes and therefore tended to become serfs of the upland Tai” (Evans, 2000: 268). But, no evidence can be found that the Phunoy would loot and take taxes from the Yunnanese caravans, as the Kachin used to do; these practices seem more suited to the Tai population of the valley. The majority of tracks followed by the Yunnanese caravans certainly crossed through the “Salt Route”, that is to say, by the valleys that are mostly inhabited by the Tai Lu (called the “Lu furrow”) spreading across the northwest to the southwest part of the province. Lefevre-Pontalis (1898: 222) remarked thus in 1894: “[ F]or several years Lu and Laotians were ghting for the monopoly of the customs taxes from the border. For the last 6 years it is true that the Laotians seemed to have no control because at the time of the caravans the Lu always arrived before them at the customs points and reaped all the associated monetary benets, simultaneously, the Lu would treat unfavourably any traders they encountered by chance.” A single track, rarely frequented, crossed the Phunoy territory, and its caravans brought iron bars and sulphur to the Phunoy to make hunting powder (Aymé, 1930: 94). Neis also noted that the salt and some of the Chinese textiles owned by the Phunoy were supplied by those Yunnanese caravans (1885: 61).5 Rather than sellers, the Phunoy were mostly buyers, and even then, the amount of goods that they purchased has to be put into perspective: “a part of the goods imported by the Chinese, up to Lung Prabang, is only passing in transit through the territory”, Aymé added (1930: 98). Lastly, for Leach, the people of the mountains—and a fortiori the chiefs—could still grow richer by becoming mercenaries for the Tai chiefs, or by blackmailing people of the valleys to prevent them from being looted (Leach, 1972: 45–46). Indeed, the Luang Prabang Royal Chronicles acknowledged that the populations from Phongsaly area were regularly enlisted as mercenaries. Moreover, observers from this region (Neis, 1885), as well as the Phunoy or Lu accounts, mention the conicting relationships between Tai Lu and the Phunoy. But, what

5

However, old people of the right bank and left bank villages recall the long trips taken by their parents to stock up on salt in the Tai Lu villages.

194

vanina bouté

was at stake in the war that led the two groups against each other? Was it regarding the control of the roads, about “blackmail”, about loot, and, in any case, was it a way to gain enough surpluses to ensure the power of the chiefs? It is not possible to determine the answers to these questions at this time. The economic factors, then, are either not sufcient or not well documented enough to allow us to draw a conclusion about the formation of the right bank domains and the acquisition of prerogatives by their chiefs. On the other hand, two issues attract our attention: the terminal settlement point of the populations and the inequitable allocation of lands, which were both described by Tannenbaum (1989) as factors contributing to long-term hierarchy amidst a society. Yet, analysis of the accounts concerning the foundation of the right bank villages implies that these two issues were the result of the scattering process and of the constitution of domains, two phenomena not present in the left bank villages. Examination of accounts concerning the foundations of both right and left bank villages may provide insights into the relationship between scattering, domain constitution, settlement of populations and attribution of lands. The Foundation Accounts: The Absence or Recurrence of the Phenomena of Village Secession On the right bank, whenever one asks how the villages and domains have been founded, the villagers generally relate a stereotypical discourse that seems to be connected to a pre-established model: the domains are composed of different villages, and the clan acknowledged as the founder of the oldest village of this domain is said to rule it. Generally, whenever tensions led to a confrontation between the chief of the village and his younger brother, the latter would question the authority of the former and leave the village, to be followed by some relatives and allies, after which they would found their own community.6 This process of secession could be repeated, either from the root village or from a village that had already split from the root village. In all cases,

6 Leach notes a similar fragmentation process in the Kachin’s society with the following difference: because of the ultimogeniture principle, the younger brother retains the authority and, therefore, it is the elder brother who tries to escape it by leaving and founding a new village.

political hierarchical processes

195

the same principal ruled the foundation of the new village: a member of the clan who founded the village of origin established his own community following a dissension. As a result, the clans that founded all the seceding villages were supposed to be the same, in as much as the conicts always involved the brothers belonging to the clan that founded the root village. The model of village foundation illustrated by the accounts seems to be veried in a number of cases. For example, the clan that founded the village of Houeylu is the same as the clan that founded the village it stemmed from, Ban Namhung; in the same way, the villages of Kunum Noy and Kunum Luang were founded by several clans that left Tapat village, and those clans happen to bear the same name, Siman, which is, indeed, the name of the clan that founded Tapat village. However, in some other villages, while the villagers do describe the origin of their village as the result of a split between two brothers, the elder and the younger, it seems that the founding clan of the village is different from the clan of the root village from which it had split. For example, the people of Thongpi village afrm that the Lawa clan is the founding clan of their village, whereas people from Xai village, while also founded by some families from Thongpi, think that the founding clan of their village is the Putin clan. This discourse does, therefore, contradict the facts in the aforementioned example (whenever this contradiction is brought up, the villagers remain evasive, usually stating that they are only repeating what they have been told by their ancestors). Let us consider as well the instructive example of Kiupork village. This village was founded rather recently, and the accurate memory of that event, as it was relayed to me, suggests to me that the account has not been excessively distorted. One old man from this village related the following: “Before, there was not enough land in the former village. All the poorest families nally took the decision to leave and settle on the other side of the hill. It has been really difcult. When the move happened, people were panicking, they did not know what to do and my grandfather organized everything. At the beginning, no one would listen but then, people did, and as everything went quite well, after they settled in the village, people asked him to be the chief ”. According to this account, the chief was not elected because of his lineage but because of his demonstrated leadership qualities and personal charisma. I may add also that in most right bank villages, alternative versions of the foundation stories are recounted during ceremonies dedicated to the spirits of Sky and the Earth (motha dat, mithon dat, P), and these versions

196

vanina bouté

obviously contradict the normative accounts such as were presented earlier. While accomplishing these rites, old people tell that their villages were founded by two or three clans (angtchu-angtcheu P—two clans being the minimum required) with those clans sharing the ritualistic functions and ensuring that fertility would be obtained. These founding clans are, therefore, considered equal: none of them obtain any supremacy and the accounts make no mention of any special leaders from the group. The particular manner by which villages are founded after some dissension between the elder and younger brother seems to be more a normative model recorded by the accounts than a constant sociological reality. Yet, whether a village was founded by a member of the clan of the root village or by another clan, it will systematically be depicted as the “younger” (nong, T) village in reference to the root village, thus called “elder” (ay, T). Whenever people are asked the reason why the name of the “younger” village founding clan is different from that of the root village, the answer is often that, indeed, it used to be the same, but that the clan split when the younger brother left the root village to found a new one. Thus, the Pongam clan used to be a part of the Lawa clan; in the same way, the Tongmumuya and Tongmumuba clans were originally the exact same clan. It is, therefore, possible to keep the ideal, according to which the founding clan of the root village is to be necessarily the founder of the other villages—the villages thus representing its domain. The principle by which a hierarchical distinction operates between the founding clan and the others allows the Phunoy to justify why clans of the same village cannot have access to the same lands, nor have the same rights. The founding clan, which would have the best lands and would request better matrimonial allowances for its daughters, was also in charge of the ritual functions of the group (let us remember that it was also the founding clan of the domain that provided the “Lords of the Land” who were in charge of the rituals ensuring the obtaining of fertility for all of the villages belonging to their domains). The appropriation of the best lands would be justied by the fact that the founding clan would distribute the plots between each of the clans of the village, or by the fact that the rst clan to arrive had chosen the best lands, leaving no choice to the last clans that had settled who were thus allocated the less fertile lands. This phenomenon of village division led to the distinction between founding clans at the domain and village level and is not dissimilar to

political hierarchical processes

197

the organization found in neighbouring Tai Lu villages. Moreover, it should be noted that the terms employed to describe the area occupied by each village within the Phunoy domain are of Tai origin. In the aforementioned cases, all the villages of one meuang are said to stem from the central village of this meuang. Let us consider the case of the Tai Lu villages of Bun Neua and Yo, respectively centres of their meuang. These two villages, considered to be the rst villages of those meuang, were respectively called “father” and “mother” villages (ban pho, ban ma, T), as they are supposed to have been founded by common ancestors. This is given as the explanation as to why the inhabitants of the two villages are not allowed to marry. The other Tai Lu villages, which are a part of the two meuang, were called “younger villages” (ban nong, T). Each of them is always composed of two clans (sing, T) organized hierarchically, as one is commonly designated “the chief clan” (sing nay, T) whereas the other one is called the “deputy clan” (sing hong, T).7 The functions of chiefs and ofcials for the rituals addressing the spirit of the meuang ( phi meuang, T) were exclusively reserved for members of the “father” and “mother” villages’ founding clans. Two types of account are, therefore, to be found in the right bank villages: stories of domain foundation depicting inequities between clans (one of which has to be systematically the founder); alternatively, stories of village foundation where the clans are said to share equally the ritual functions. How to explain the opposition between those two types of account? Accounts of Foundation on the Left Bank: The Pre-Eminence of the Chief and the Equality of the Clans This contradictory set of explanations for the manner by which villages and domains were founded (sometimes the equality of the clans at the time of the foundation being insisted upon and, at others, the pre-eminence of the founding clan and its chief), can equally be found in the left bank villages. The only difference is that, unlike the right bank accounts where there is typically one dominant version and the 7 The name of those clans can change. The opposition between the “chief clan” and the “deputy clan” is a specicity of the Yo village, but people from Bun Neua oppose the Phya clan to the Pee clan: there are both titles of nobility and the second one would be less important than the rst one. I refer here to the qualifying terms as those clans also have names that can be tree (sing lo) or animal (sing vang, the tiger) names.

198

vanina bouté

other version is given only as an alternative, on the left bank, these two kinds of explanation are given in the same story. The structure of these accounts is always the same, involving three characters who are the founders of the village as well as being the ancestors of the three clans who habitually comprise the village population. Take for example the story of the foundation of Kiu village: “One day, three hunters, Tao Khamsuk, Nyeum Phiban and Tao Pusum, departed for the forest. They were then attracted by a light that seemed to be requesting them to follow it. While chasing this light, they reached a clearing abounding with water. They decided to found a village at that location and are said to be the ancestors of the three clans: Keutchup, Tcheumang and Mantcha. Tao Khamsuk was the chief, Nyeum Phiban was his second, and Tao Pussum the layman in charge of the pagoda”.8 In these stories, the ancestors of all the clans of a village, always three persons, simultaneously found the locality: none of the clans can claim the status of founding clan. Yet, most of the time, those stories will tend to set a clan apart from the others—in this case, the clan of the chief. At the time of the handover of the “Books”, however, when the village was transformed into a domain, the slight differentiation between clans just noted does not become accentuated. First, I should emphasize that the domains located on the left bank differ from those on the right bank on account of their size. The inhabitants of the left bank villages explain it by the fact that their villages do not stem from the secession of other villages and that they have never been split.9 This situation contrasts markedly with the right bank villages, where even today the people are able to trace precisely the origin of each village, its successive moves, the root village from which it originated and whether it has been divided subsequently into other village units. Yet, we have seen that it is the reference to those villages’ secession stories that justies a

8 This account with these three clans is sometimes told in a different way: the rst clan would be the clan of the “good deaths”, the second the one of the “bad deaths” and the third one would have been in charge of separating them from each other. It is interesting to note that those separations of the clans in three functions can also be found in some of the founding accounts of the right bank. It is also said that one clan is the equivalent of the water, the other one of earth, and that the third one’s role is to maintain the balance between them. This may be a way to underline the necessity of a three-fold exchange, two terms being opposed, the third one contributes as a regulator. 9 This fact would explain the huge number of houses in the village located on that bank: around a hundred houses (versus an average of 40 houses only on the right bank).

political hierarchical processes

199

hierarchy between the clans and the villages within Phunoy society on the right bank. On the left bank, one cannot nd domains composed of several villages and, as a consequence, the ritual predominance of one clan upon a group of villages does not exist, as do neither any Lords of the Land at the supra-village level: each left bank village is a small domain in its own right. Therefore, when the King of Luang Prabang sent the “Books of the Land”, the books were handed over to the clan that was only slightly distinguished amongst the three and the “Lords of the Land” function was shared by all of them: among each of the three clans, there was a “Lord of the Land”. In the same way, those “Books” handed over by the king were decreed to be the property of the all the clans of the village, and all the clans had the same rights upon the lands cultivated by their members: for all land tenure, matrimonial compensations, and religious and political functions, the clans were nearly equal after the “Books” were handed over, with only a slight ritual pre-eminence being accorded to a descendent of the founding “chief ”. Lastly, it should be said that any person from outside the village who intended to settle in it, had necessarily to become afliated to one of the three existing clans, losing as a consequence his own regional clan identity.10 Whatever was the motivation for this practice within the left bank villages, a non-differentiation between the founding clans and the new settlers was the main result (differentiation could certainly have happened if the newcomers were to retain their former clan of origin). The absence of political and economic distinctions between the different clans in the same village, as well as between different villages on the left bank, whether occurring on the discursive or on the practical level and, furthermore, despite the intervention of external powers in the nomination of border guards and handing over of territorial documentation such as “The Books of the Land”, gives the impression that any process to institute a hierarchy or differentiation in the left bank villages encountered a sort of resistance. We can see the similarities between the different accounts regarding village foundation on the right bank and those of the left; it is mostly the accounts referring to domain foundation that tend to change. The right bank Phunoy accounts emphasize equality between clans at the

10 I do not know how those names are being adopted and what is the status of the families newly included. Anyway, this practice, together with the fact that the villages did not scatter, explains why the clans are always different from one village to another.

200

vanina bouté

time when the village was founded, the clans establishing the village sharing the ritual functions according to a complementary principle, but also, progressively, inequality begins to emerge with the constitution of the domain. At that time, the prerogatives of the founding clan of the eldest village increase and emphasize a hierarchy within each village and within the domain. These developments, presented in two distinct accounts on the right bank, are united in the same story in the left bank villages, where the equality of the clans (all founders) and the pre-eminence of the clan of the chief are presented all together. This becomes evident from the fact that, on one hand, all the clans have equal access to the land and each doyen of a clan is equally titled “Lord of the Land”, but on the other hand, there is a ritual pre-eminence for the members of the founding clan of the ‘chief ’. To conclude this point, it seems that the scattering process indeed favoured the constitution of the domains. The “Books of the Land” were afterwards handed over to members of the lineages derived from the elder brother of the clans considered founders of the root villages (i.e., from which other villages located in the domain would have split). Thereafter, a periodic ritual was performed by these individuals, called “Lords of the Land”, for the obtaining of fertility in villages in their domain. In situations where there was no scattering from a village, as is the case in the majority of the left bank villages, the “Lords of the Land” would only be in control of their own village space. The main difference existing between the right and the left bank villages is, therefore, the fact that, for the latter, the title “Lord of the Land” was given to all the clan chiefs. Therefore, the scattering process (or its absence) provoked a lot of distinctions between the two groups of villages situated on the right and left bank. On the right bank, the clans split to become, in each village, localized clans, whereas on the left bank, a clan was only represented in one village; on the right bank, wherever the villages and the clans were split up and the predominate ideology was connected to younger/ elder and rst/last settlers, great differences between clans and lineages appeared. In the opposite case, on the left bank, an acephalous village society predominated based upon clan equality.

political hierarchical processes

201

Scattering, Institution of Clan Hierarchies and Pressure Upon the Land Even though the scattering process indeed seems to have been central to the formation of right bank domains, it cannot be established as the only explanation for the creation of those territories, nor is it a sufcient datum to understand how, later on, those domains persisted. Yet, in the Phongsaly region, other villages experienced a similar scattering phenomenon without the appearance of any root village pre-eminence over other villages, nor the creation of hierarchy between the units in the domain relationship. For instance, the inhabitants of the Akha11 villages of Chakhamsaw and Chakhamdeng (Bun Tai district), where I conducted research, afrmed that they were descendents of Sanor village, but they never found themselves in a situation of subordination, ritual or political, towards Sanor. This is also true for some villages with small groups of Phunoy language speakers: according to its people, the “Pumon” village split up a century ago into two smaller agglomerations that became independent from their village of origin; the same thing happened to the ve Laopan villages located in the districts of Bun Tai and Samphanh, all supposedly originating from the same village. Similar phenomena can be found as well in some other provinces of Laos: Izikowitz (2001: 41) noted that Lamet villages were of a tiny size and that this was a consequence of regular divisions. Lastly, the Phunoy left bank villages most likely originated from the right bank villages—even if they have no common memory of it. In order to explain the fact that every scattering process does not necessarily involve the constitution of a domain, I am willing to present the analysis proposed by Leach for the Kachin example. For Leach, the fact that in Kachin society some villages live in an autonomous fashion, whilst others are gathered into agglomerations subordinated to a root village, is mainly the consequence of a scattering process combined with population increase and land shortage. Leach writes: “Wherever the density of the population is remarkably low, the very small autonomous villages are the big majority; the chief of each of these villages claims his own independence. . . . In the regions where the density of the population is higher, then there is no way for the sons of the chief to establish some independent domains without stepping on the rights,

11

Another important ethnic group of Tibeto-Burmese speakers in northern Laos.

202

vanina bouté

already existing and acknowledged, of some other chief. In that case, whoever decides to move anyway would have no other choice but to settle on a relative’s domain, with a status of subordinate” (1972: 202). According to Leach, the traditional way of resolving conicts and the will to be apart from a chief ’s supervision are linked to population increase and land shortage, inasmuch as the absence of free space suitable for cultivation will not allow for the foundation of new communities. Kirsch shows complete agreement with this theory by asserting that, in a context where demographic pressure is low, departures from the village facilitate the levelling of differences within the root villages themselves: “Older sons of a ‘chief ’ set up ‘colonies’ . . . draining off some of the more highly motivated from the parent village—leaving the less motivated behind” (1973: 29). These alternative propositions seem, at rst glance, to t the situations of both right and left bank villages perfectly well. The territory occupied by the right bank Phunoy, more or less equivalent to the south of Phongsaly district, was densely populated (this was before the great migration movement of the 1960’s that affected the region). According to Roux, in 1924 the Meuang Phunoy was composed of thirty-six villages, whilst there were twenty-eight in the same zone in 2003.12 Furthermore, not only were the villages in 1924 more numerous than today, they were also larger: “Whereas the most populated of the villages belonging to other races do not exceed ten shacks, most of the P’u-noi housing schemes are composed of more than 50 shacks” (Roux 1924: 456). Some other French observers at the beginning of the century made similar remarks: the Nam Pung village was composed of sixty to eighty houses (Guillemet and O’Kelly, 1911), Phongsaly village, more than one hundred and fty (Cheyrou-Lagreze, 1921). On the contrary, demographic and territorial pressure on the left bank of the Ou River seems to have been extremely low. The left bank villages are more recent than those on the right bank (Alexandre and Eberhardt, 1998: 51); some inhabitants even afrmed that their ancestors came from the right bank, even though none of them could remember precisely which village they had come from. Maps of the

12 I have taken into consideration for 2003 the area that corresponds to the ancient limits of the Meuang Phunoy in 1924; I have subtracted from the 2003 numbers the number of habitants of the Phongsek village (that was not included in H. Roux in the Meuang Phunoy) as well as the number of habitants of the villages that originated from it and settled by the Ou River bank (Hatkao in 1970, Hatao in 2001).

political hierarchical processes

203

region designed by the French at the start of the twentieth century only indicate the big villages and do not mention those on the left bank. We can, therefore, make the supposition that, at that time, demographic and territorial pressure was not an issue at all. One exception among the left bank villages to the general absence of hierarchical development as a result of scattering should be mentioned: the people of the “Sen Sukhwa the Great” (Sen Sukhwa Luang) told me in 2002 that eighty years ago (meaning the year 1920), conict arose between the chief of the village and his younger brother. They shared open “Books of the Land” (composed, as previously mentioned, of two parts: the ofcial letter and the description of the territorial limits). The youngest left and founded a new village, called “Sen Sukhwa the Little” (Sen Sukhwa Noy). The inhabitants of this village belonged to three different clans, but it was only from within the clan of the younger brother that a person could be acknowledged as “Lord of the Land”. This person (and his descendants) was, therefore, the only one entitled to proceed annually to the division of the plots between each household before slash and burn of the forest. The two villages, which formed the same domain according to their inhabitants, celebrated a common cult every ve years for the obtaining of fertility for the elds in the forest located at the border of the two villages. The way the two villages of Sen Sukhwa are organized strongly reminds us of the organization of villages on the right bank of the Ou River. The founding clan is entitled to the ritual prerogatives, and the title “Lord of the Land” can only be bestowed upon one of its representatives. Here, indeed, is a scattering process leading to the creation of a clan’s domain and the institution of a clan hierarchy. But, how do we explain the subordination of the Sen Sukhwa Noy village to the root village in spite of the fact that, with territorial pressure on the left bank being low, this village could have settled independently on another place, at least according to the process described by Leach? Does it mean that my previous conclusion concerning the signicance of very low demographic and territorial pressure upon the left bank could be incorrect? If so, what could explain why such scattering processes did not occur in other villages of the left bank? Moreover, nowadays, demographic and territorial pressure has become extremely important on the left bank: each village is composed of almost one hundred families and are often located in close proximity to one another; yet even so, they still do not scatter. I shall now consider this problem before re-examining the singular case of Sen Sukhwa village.

204

vanina bouté

Fixation in Territory Delineated by the “Books of the Land” The most likely hypothesis to explain why the left bank villages did not scatter is provided by the handover of the “Books of the Land” by the royal power: the delineation and allocation of the land was divided up between the pre-existing villages, preventing the foundation of a new village on non-monopolized lands. When the “Books of the Land” were handed-over, on the right bank they delineated a territory already marked by the scattering process (a model of elder and younger villages). The pressure on the land being ancient, the villages that were newly founded had to settle on the territory of the villages from which they originated.13 These new villages would have kept the original cult location, as the favoured spirit of the founding clan brought its protection to the entire group of villages in the territory. In this way, we can surmise that the cult strengthened the prestige (power, prerogatives) of the founding clan of the original village from which the cult originated, not only over the inhabitants of the root village but subsequently, following division, for both villages. Thus, scattering, within the same territory, reinforced the ritual pre-eminence of a clan that, from thenceforth, accomplished the cultic rituals for a group of villages and clans. The handover of the “Books” would have been a consequence of the pre-existing belief in the pre-eminence of the supra-village founding clan. On the contrary, because of the low demographic density due to the recent settlement of the inhabitants, on the left bank each village formed an autonomous entity and, thus, constituted a domain of its own.14 In fact, the scattering process has been rarely in operation since the handover of the “Books”. By comparing the maps drawn up by the French during the twentieth century with those more recent maps made in the last decade, we notice that the villages did not scatter

13

It is most likely that these new villages were, in fact, the small eld house of a root village and later, would have formed an autonomous group. H. Roux noticed, at the beginning of the 20th century, that: “each village is divided in 2 parts: the original village, which never moves, and the village of cultivations, where the adults, the women, a part of the livestock and other pet animals settle during the entire length of the work in the elds.” (1924: 457). 14 While discussing with the left bank villagers, they found it difcult to imagine that, in the past, a village would have to split up as, in their opinion “the space to cultivate was important enough, there was no reason to leave the village”.

political hierarchical processes

205

any more, neither did they spontaneously move, or, if they did, it was always within their domain.15 We can, therefore, suppose that it is this “xation” of territory that prevented the phenomenon of splitting in the villages of the left bank. The fact that the territory of each village was xed by the “Books of the Land”—leading to the allocation of all the lands in spite of the absence of any pressure on the land—did not enable the village to split. The case of Sen Sukhwa Noy shows that it was possible to form a new village, even if it had no other option but to settle on the lands of the village from which it stemmed. Why did such a similar phenomenon not occur more often on the left bank or, supposing that it did, why did the inhabitants not recall the ancient divisions? It seems that the handover of the “Books” not only delineated a zone (composed either of several villages or of one), but also attributed some functions (those of the “Lords of the Land”) and contributed to the mode of social organization as well. This type of organization being, on the left bank, equitable, scattering did not become necessary, as there was no conict with authority. Indeed, it was because the scattering process did not begin in the left bank villages that no supra-village organization existed and each village was, therefore, autonomous at a political and ritual level. When the “Books” were given, the clan of the founding chief only had prerogatives upon his own village and the difference with the other clans was not so pronounced. The “Books”, therefore, would have reinforced an egalitarian system within left bank society. The role of the chief being small (he only had ritual prerogatives), and decisions being taken by the council of elders and all the clans, via their doyens, being charged with the control of the eld’s fertility rites, there was not any singular authority to be questioned, and then, no reason for leaving the village.16

15 The Khunsuk village, located at the North of Phongsaly town, did split up 80 or 90 years ago, according to its inhabitants. The village resulting from the scission did settle on the territory of the original village. 16 This idea of an equality between the clans considered as necessary is still widespread nowadays, even if the left bank villages endure a pressure on the land more important than it has ever been on the right bank ones. However, the inhabitants keep explaining the scattering of the plots and their free allocation by the fact that “it has always been like that, because there has always been much space for the shifting cultivation”. The ideas, here, prevail over the facts. It is, in my opinion, an indication of the refusal of a monopolized control on the production by a minority via the plots’ repartition, as this latter leads to the instauration of a hierarchy between the families.

206

vanina bouté

The equality between the clans might have been reinforced because of the impossibility of founding a new community on the lands of the root village. Indeed, this model seems appropriate to similar situations within other highland populations: the lands being xed (and with no split in the village), the members most likely to question a chief ’s authority and who, if they had not been confronted by land pressure would have left and founded new communities, instead have no choice but to question the authority of the village chief (Kirsch, 1973: 29–30). This results in a decrease in the status differences within a village, the chief ’s authority fades to the benet of the village assembly, even though the founding clan can still be symbolically acknowledged (Bouchery, 1988: 301). Whether such villages are egalitarian from the start or become so following a democratization process (due to the progressive loss of distinction between the founding clan and the others), the lands, together with the ritual and political functions, become more fairly divided between all the clans. Because of this ‘democratic’ mindset, potential agents of conict would be less likely to question authority or leave the village. The Impact of the Colonial Power Finally, we still have to comprehend the sole village division that occurred on the left bank: the Sen Sukhwa division. Towards that goal, let us examine some contingent events. At the beginning of the twentieth century, the left bank villages, isolated from the military posts and barely indicated on the maps, were not of any interest to the French settled in the region, and the administration of the zone was left to the Lao dignitaries of the meuang Hun. But, in the 1920s the colonial administration took the decision to build a track to link together Phongsaly city (which then became the county town of the 5th French Military Territory in 1929) with Dien Bien Phu. This track passed through the territory of the left bank Phunoy but close by only one village, Sen Sukhwa, which, therefore, became a possible stopover on this long route. The French then decided to ensure that this village would be a safe stopover whilst in transit through this region. In order to ensure the loyalty of the village and the proper functioning of the authorities, they entitled the chief of Sen Sukhwa village to become chief of the two villages of Sen Sukhwa, and gave him some advantages; the chief remained the meuang Hun chief ’s subordinate and his authority did not

political hierarchical processes

207

exceed the limits of his own village (Mordant, 1934: 874). Nowadays, the inhabitants of Sen Sukhwa Noy relate that dissensions in the village resulted from the nomination of the chief of the village to this prestigious role: “In Sen Sukhwa Luang the family of the chief of the village did not pay taxes, so the others became irate, the younger challenged his brother to a ght and, followed by the malcontents, he came here to avoid the taxes”. The reasons why some families left and settled in a new location would have been caused by unequal relations, newly instituted. However, it was not to seek a more egalitarian environment that the younger brother of the chief would have taken the decision to migrate: unlike the accounts of other left bank villages, the foundation of Sen Sukhwa Noy reported a unique founder. The sole clan to be considered that of the chief was the clan directly related to the man who founded the village, and his representative was the unique bearer of the title “Lord of the Land” and holder of the “Book of the Land”. The two villages of Sen Sukhwa Luang and Sen Sukhwa Noy, the clan of the “Lord of the Land” being the same (the clan of both the elder and younger brothers), were a part of that clan’s domain and this dominant clan beneted from certain prerogatives. Therefore, it seems probable that the scattering reinforced the status of the dominant clan in the original village. This was achieved rstly by increasing the ritual authority of the chief of that village, and secondly by the fact that the new village was founded by a member of the same clan as this chief, thus illustrating further the capacity of the chief ’s family members to assume the responsibilities of leadership. The case of the two Sen Sukhwa villages is, therefore, a pertinent illustration of how external intervention, in this case by the French colonial administration, could introduce a hierarchy among the clans. However, the new communist government of the Pathet Lao abolished the privileges of these chiefs in the 1960s, and the socio-political organization established previously progressively collapsed.17 In this village today, the hierarchy between the clans, which used to be visible through the fact that the “Books” and the title “Lord of the Land” belonged to the representative of only one clan, is no longer seen. Unlike the other left bank villages where the doyens of each clan still maintain the title of “Lord of the Land”, there is no longer a representative bearing 17 After the Geneva agreement in 1954, the communist forces of the Pathet Lao held the government of the province of Phongsaly.

208

vanina bouté

this title, nor is there a ceremony to celebrate this function in either of the two villages of Sen Sukhwa. In Sen Sukhwa Noy it is said that the “Book” became something negative, dangerous for its owner and his family, and it was consequently taken to the pagoda to neutralize its evil powers. As for the “Book” of the root village (Sen Sukhwa Luang), it underwent a similar process: “The other villagers used to tell me that it was my ‘Book’, and that it was not for them to deal with nor should they participate in the rites. Me, I did not want to celebrate the rites on my own: if, for instance, I could not produce enough offerings to present, it becomes too dangerous for me and my family. So, I buried it in the wood to get rid of it” its former owner relates.18 *

*

*

When the “Books of the Land” were handed over to populations speaking the Phunoy language, who were also given the role of border guards, the Phunoy encountered different situations depending on the locality. In the villages on the right bank of the river Ou, populated by more ancient inhabitants, there already existed a hierarchy, probably arising from the process by which villages split, combined with pressure upon the land. Local representations reinforced the distinction between elder/younger and rst/last settlers, and established the division process as a constituent of the pre-eminence of one clan over the others. On the contrary, in the villages on the left bank, which was a zone more recently populated, no hierarchy existed between the clans in spite of the ritual pre-eminence of the representative of the founding clan: the “Books” were divided between the clans of the village, and a member of each clan bore the title “Lord of the Land”. The “Books”, therefore, did not have the same effect. By xing the populations on the territory that was allocated to them, they conrmed a reality: the hierarchy of the right bank clans and, conversely, the absence of hierarchy on the left bank. It must be noted in this context, the royal power and, at the start of the twentieth century, the colonial administration, relied more upon populations that represented, according to their assumptions, a hierarchically organized system, closer to their own concepts of authority, and it was these populations who, thus, assisted them in their control of the region—the ones of the right bank.

18

This interview was given during my eld enquiries in 2002.

PART THREE

THE KACHIN SUBGROUPS

RETHINKING KACHIN WEALTH OWNERSHIP * Ho Ts’ui-p’ing This chapter examines Kachin wealth ownership in order to offer an alternative to Leach’s “oscillation” model of social change for Kachin communities across a wide expanse of upland Burma, as set out in his Political Systems of Highland Burma (1954). Although my account benets from both Leach and his critics, it differs in three important respects. First, it introduces primary sources that provide wider coverage of Jingpo in Yunnan for the time immediately before and just after Leach’s own eldwork.1 Second, my own ongoing eldwork, which began late

* The writing of this chapter beneted from discussions of the original draft at two occasions: the workshop of “Rethinking Boundaries” held at the Institute of Ethnology, Academia Sinica, Taiwan, 18 August 2004; and the panel of “Reconsidering Political Systems of Highland Burma: A Study of Kachin Social Structure—Comparative Approaches 50 years on from E. R. Leach” at EUROSEAS (4th Conference of the European Association of Southeast Asian Studies), Paris, 1–4 September. At the workshop I beneted from exchanges with and comments by Yang Shu-yuan, Chiang Bien, Guido Sprenger. For the panel, I especially thank the panel organizers, Mandy Sadan and François Robinne, including the former for her efforts during the editing process and the latter for his continuous kindness and support in inviting me to join this group of scholars. I additionally thank Robert Anderson, Stéphane Gros, and the panel reviewer F. K. Lehman for their interest in and comments on my original draft. Finally, I am very grateful to my husband James Wilkerson for numerous discussions about the initial writing and subsequent revision of the various versions of this chapter. Any remaining shortcomings are solely my responsibility. 1 My rst hand material from the Jingpo in China includes data from Jinghpaw speakers and Zaiwa speakers in Dehong Prefecture in the People’s Republic of China. The research on the Jingpo includes eighteen-months of dissertation research between 1988 and 1991 through funding from the United States Committee on Scholarly Communication with the People’s Republic of China and an exchange program between the University of Virginia and the Yunnan Nationalities Institute. Between 1995 and 2005, I carried out eight brief research trips funded by “Exchange, Life-cycle Rites and Personhood: Regional Research on the Chinese Southwest Nationalities” project under the Thematic Research Program of “Highland and Lowland Societies and Cultures of Monsoon Asia”, Academia Sinica, and personal annual research projects of the Institute of Ethnology, Academia Sinica, Republic of China. In this paper, where appropriate for reasons of context, I continue the earlier practice of using the term Kachin, I use Jingpo to refer to all the Jingpo nationality under the Chinese classication of nationality, including Jinghpaw, Zaiwa, and Langvo speakers. Hanson’s (1954 [1906]) romanization is used for Jinghpaw language terms (indicated by [ j]; the Dehong Language Committee’s romanization is used for Zaiwa terms (indicated by [z]); and, Pinyin romanization is used for Chinese terms (indicated by [c]).

212

ho ts’ui-p’ing

in the 1980s on Jinghpaw and Zaiwa speakers of Jingpo nationality across the border from Burma in Yunnan, makes it possible to focus afresh on the cultural classications of property in general and wealth objects in particular that are so much at the heart of Leach’s oscillation model. Third, in aggregate, these additional descriptive materials provide a fresh documentary standard against which to measure Leach’s oscillation model and the alternate formulations of his critics from the perspective of recent relevant debates within anthropology. Briey, the perspective I have arrived at is that Kachin social change between the 1850s and 1950s revolved around monetized markets whose histories predate the arrival of British colonialists, and a major shift in those monetized markets in association with the imposition of British colonial rule. Nevertheless, the Kachin were not simply passive witnesses to their own history. Rather, they brought, and are still bringing, cultural understandings of wealth objects to engagements with their new circumstances. Thus, in contrast to Leach’s oscillation model and similar to his critics, the conclusion I reach is that external events triggered fundamental changes in Kachin society that undermined the gumsa hierarchical social order. Nonetheless, like Leach’s oscillation model, and different from his critics, my alternative interpretation still includes a central role for Kachin culture in Kachin social change, including especially cultural classications of wealth objects. Leach’s Oscillation Model and its Critics Edmund Leach’s Political Systems of Highland Burma has long served as a touchstone for research in upland Southeast Asia (Woodward 1989: 121). Furthermore, Leach’s monograph has long been inuential even more widely for the study of social change elsewhere around the world (see, for example, Kuper 1983 [1973]: Chapter Six). Leach’s oscillation model asserts that people in upland Burma had a range of possible social orders from which to choose, including: the hierarchical gumsa, the egalitarian gumlao, and the monarchical Shan. No single community completely adhered to any one of these social orders, instead tracking relatively closer to one or the other, and adhering to any one of the social orders for only a limited stretch of time before turning in the direction of another social order. Leach’s oscillation model further asserts that, although external forces encouraged Kachin gumsa to move toward monarchical Shan hierarchy, internal

rethinking kachin wealth ownership

213

cultural factors encouraged gumlao equality. More specically, the motive behind a swing in the direction of a monarchical Shan hierarchy was “imitating Shan.” In contrast, the motive for a swing in the direction of an egalitarian gumlao social order arose from ambitious seekers of power who exploited the paradoxes and contradictions lodged in afnal relations and ultimogeniture (1977 [1954]: 262), and where justication for their actions was located in myths that recounted and legitimized egalitarian gumlao revolts (ibid.: 263). Criticisms of Leach’s oscillation model break down into two kinds. One sort emphasizes factors Leach saw as essential to social change, the other emphasizes factors Leach did not see as essential to social change. Criticisms of the former sort turn on Leach’s “imitating Shan” account of the motivations toward a hierarchical social order. One attraction of Leach’s argument about “imitating Shan” is that it foregrounds “human desire” as a motive for human action. Unfortunately, in seeing “human desire” as more-or-less innate and divorced from culture and political economy, Leach frees himself from the obligation to give an account of the specically Kachin culturally-situated desire to imitate the hierarchical Shan, a criticism that has already been widely made (see Kirsch 1973, Lehman 1977, 1989, Woodward 1989, and Woodward and Russell 1989).2 Likewise, one attraction of Leach’s argument about myths as social contracts is, indeed, that it introduces a culturally situated desire for a particular social order. Be this as it may, human desire might account for why Kachin social change moved in one direction; it could not account for social change in both directions. Nevertheless, the bulk of the criticism levelled against Leach’s oscillation model concentrates on possible external factors that alternately encouraged gumsa hierarchy and gumlao equality. Three events from the period beginning in the middle of the nineteenth century and ending

2 For instance, Kirsch agrees with Leach’s thesis on the manipulation of individuals, but moves beyond that position. He argued that the “motivational basis of the dynamism found in these upland Southeast Asian Societies” is that “the individuals within these groups are seeking ‘ritual efcacy,’ ‘potency,’ ‘enhanced ritual status,’ or some such religiously dened goal, not seeking simply to possess ‘political power’ ” (Kirsch 1973: 3). Kirsch shifts Leach’s emphasis on the individual’s “political” motivation to that of “religion,” because he wants to “highlight more general internal systemic consistencies than Leach would allow” (Kirsch 1973: 3). I provide elsewhere (Ho 1997: 53–66) a more detailed review and critique of my own and other scholars’ understandings of Leach’s views about culture.

214

ho ts’ui-p’ing

in the middle of the twentieth century have been emphasized. Two of these external events were decidedly political (reviewed in Giersch 2001 and Atwill 2003 on Muslim rebellion especially). First, beginning in the middle of the 1850s, Yunnan entered into a long period of political instability in association with wider crises besetting China’s Qing dynasty. Second, in 1885, the British imposed colonial rule upon northern Burma. One important shared feature of these two events was that the classic kingdoms of Burma and China had by then lost their previous ability to inuence politics in upper Burma. An additional, third economic event took place in close conjunction with the above two political events. That event was the orescence in the production, trade, and consumption of opium. This event provided the Kachin with a major new source of cash income (Renard 1996, Bello 2003, Trocci 1999, Zhang 1990, Qing 1998, 2005), which probably resulted in a period of increased Kachin wealth. Leach referred to all three of these events in Political Systems of Highland Burma. Still, since Leach did not see any of these events as essential to his oscillation model, his descriptions of these external events were correspondingly brief and fragmented. Furthermore, even years after the publication of the rst edition of the monograph, Leached continued to reject vigorously suggestions by his critics that external events, and especially the introduction of opium, might be relevant for an evaluation of his oscillation model (1977, 1983: 195–197, 788). Tellingly, a central rationale Leach explicitly cited for spurning the importance of the above three external events in modelling Kachin social change was the importance he gave to providing a ‘cultural” account for social change. David Nugent (1982) and Jonathan Friedman (1987), two of Leach’s most prominent critics, concentrated on the importance of these wider events in the history of upland Burma and argued for a lineal history of Kachin social change as being most importantly organized around the above three external events. In so doing, they identied the introduction of opium as the single most important event for the century that began in the middle of the nineteenth century and lasted until the middle of the twentieth century. Nugent’s argument includes three steps. First, the presence of the gumsa social order up to the middle of the 1850s was a consequence of the introduction of opium. Second, after the 1850s, the political instability across the border in Yunnan and the imposition of British colonial rule in upland Burma precipitated gumlao revolutions at the

rethinking kachin wealth ownership

215

end of the nineteenth century. Third, Nugent goes on ultimately to relate the introduction of opium to wider global forces. Friedman gives a much different account of the consequences that arose from the introduction of opium. First, instead of linking the introduction of opium with the presence of the gumsa social order, Friedman attributes the gumsa social order to the longstanding “predatory structure” of the Kachin. Specically, the organization of the gumsa social order was around a strategy of “exploiting the ow of wealth in the larger system by extortion” (1979: 13). Second, the role of the introduction of opium only became pivotal early in the twentieth century. Consequently, the introduction of opium was “associated in fact with the spread of gumlao organization” (1987: 17). There is special merit to Friedman’s account. First, Friedman’s criticisms of Nugent is that the latter’s attribution of the gumsa social order to the introduction of opium leaves no role for the obvious historical importance of trade in upper Burma long before both the imposition of British colonial rule and the introduction of opium. Additionally, though certainly present, the importance of the opium trade prior to the 1850s relative to trade in other products is to misconstrue the historical sources. These comparative merits to Friedman’s account notwithstanding, his attribution of the presence of the gumsa regime to the predatory relationship between the Kachin and pre-British regional trade is arguably itself overly narrow in its concentration upon predation. First, Kachin interactions with regional markets were not limited to predation alone, but also importantly included production for trade, trade proper, and wage labour in association with that production and trade. Second, trade in this part of the world had a long history of using money and money substitutes, including Burmese and Chinese currencies and specie. Finally, although of great importance, the introduction of opium was not the only impact that the imposition of British colonial rule had upon Kachin market activities. In sum, it is not easy to dismiss a priori the possible broader relevance of monetized markets and their relationship to wider state formations in the historical continuities and discontinuities behind Kachin social change. The Kachin economy is historically, and still remains, agrarian. Nonetheless, there is still solid evidence that markets and monetization played a wider role than heretofore realized in Kachin social change. Chinese and English sources both describe vigorous and widespread Jingpo in Yunnan and Kachin efforts to “look for money” (YNSBJZ

216

ho ts’ui-p’ing

1985a, 1985b; Chen 1941, Enriquez 1920, 1923, 1933; Renard 1996, Qing 1998, Kuang and Yang 1986, Leach 1977, Webster 2003). There are reports that Kachin in eastern Bhamo collected tolls from caravan traders (Huang 1976: 85, Leach 1977 [1954], YNSBJZ 1985a: 129). As early as 1871, in addition to producing rice and pigs for trade, Kachin were already producing opium for sale, whose income was then used to buy salt, dry sh, needles, buttons, cloth and clothing (Anderson 1871: 221, 261). Some decades later, shortly after the institution of the Nationalist Chinese Government in 1911, the Chinese state began to impose a tax in the Zhidan mountain, to be paid in opium, which each annually taxed two liang of opium for a total of about 800 liang of opium (YNSBJZ 1986a: 10).3 Opium had been an important cash income, but not the sole cash income for the Jingpo since the late nineteenth century. They also mortgaged and sold their land to the Han.4 Since the latter half of the nineteenth century, this area of northern Burma and the China border has always been in ux politically. The massive Han immigration to Yunnan since the mid-nineteenth century, voluntarily or under government encouragement, caused the series of rebellions, revolts and unsettlement (Atwill 2003, Lee 1982, Giersch 2001), as well as other chances to make money. My Jingpo “uncle” in Dehong in his early 70s talked about how the powerful Gudong headman made his wealth from the geopolitical space the Jingpo occupied at the state’s peripheral. One does not know how many head of cattle the Gudong chief had. We are not directly under his ruling. But, in our village, one family was tending some of his cattle. In our village only, he had at least 40 to 50 heads of cattle. He earned his wealth from being the leader mediating between the Han and the Jingpo, the tribute money, rice, opium or cattle by the villages (including many Han villages) under his ruling. After every successful mediation of conict, the Han gave the chief several heads of cattle, guns etc in return. He also grew opium, but not much.

In his recollection, people with skill making iron tools and utensils, woodwork or jewellery, and those who grew opium and raised cattle, 3

One liang (“Chinese ounce”) equals 37.7 grams. Land sale or mortgage was considered an important source of cash for the Chinese Jingpo by Chinese literature on Jingpo in the 1950s (YNSBJZ 1985a, 1985b, 1986a). Lead by social evolutionary theory, this research holds the argument that the development of private property ownership in different areas or villages is the most important criteria in developing into “landlord feudalism.” The typical explanation for this landlord feudalism holds that the penetration of Han Chinese into the Jingpo area causes mortgage and the sale of land. 4

rethinking kachin wealth ownership

217

could all nd money. There were other kinds of wage labour, too. Early reports additionally indicate that men sold their labour in various odd jobs, such as herding for lowlanders, transporting salt across the border, hiring out as “braves,” carriers, guides, and construction workers, as well as serving as military police and soldiers. Their employers were diverse. Early on, they worked for the lowland Shan and Chinese, the Dai local chiefdom (tusi [c]) government,5 for the British during the colonial era,6 for the American army in World War Two,7 and for the Nationalist Chinese government for surveying land and building roads. While women did not usually sell their labour to the British or Americans, women sold rewood and other mountain products to lowlanders. Outsiders also hired Kachin women as short distance carriers, including serving the lowland Dai inside the China border as short distance carriers between market towns (YNSBJZ 1985a: 127). In the 1950s, Kachin men and women also worked outside their communities in the agricultural slack season that occurred annually just before harvest to earn cash. Communities short of agricultural land were particularly dependent on cash income earned from lowland markets through trade, labour, and forest products. In short, whether the medium of exchange was money, or opium as a money substitute (see Leach 1977 [1954]: 151, 1983: 195), a trail of evidence documents the presence and importance of monetized markets in the Kachin Hills before, during, and after the arrival of the British.

5 Pictures taken in 1937 show Jingpo soldiers serving the Dai local chiefdom government ( Jiang and Jiang 2003: 107, 109). 6 Kachin began joining the Upper Burma Military Police Battalion at Bhamo in 1896–1897 (Dautremer 1916: 183–4, Dawson 1912: 66–68, Scott 1921 [1906]: 351, Enriquez 1923: 40). In 1917, they became part of the Eighty-fth Burma Ries and formally enrolled in the regular British Indian Army. Kachin soldiers fought for the British in Mesopotamia between 1917 and 1919 (Enriquez 1920: 46). 7 Beginning in 1942, Kachin involvement in Burma in World War Two, as levies and guerrilla soldiers in ghting against the Japanese, was signicant (Fellowes-Gordon 1957). That involvement included as British North Kachin Levies (NKL), who earned a reputation as expert at laying ambushes (Webster 2003: 49–56). Beginning in 1943, Kachin assisted the American Detachment 101 in guerrilla ghting as guides, escorts, and instructors in assassination techniques (pp. 158–162). As a unit within Detachment 101, under the command of Captain William C. Wilkinson, the Kachin won even greater acclaim for their “fearlessness, far-ranging surveillance skills, and uency in the jungle” (p. 162) Prior to the end of 1943, Detachment 101 had conscripted “thousands” of Kachin (pp. 162–163). Kachin recruits joined the “Kachin V force” to “act as a regular army version of Detachment 101,” in support of General Stilwell’s X Force re-invasion back into central Burma for creating diversions and gathering intelligence (p. 164).

218

ho ts’ui-p’ing Money and Culture

My emphasis upon the importance of the monetized market as a possible organizing force in social change is certainly not novel. To skip quickly forward to the more recent anthropological literature on the subject, the widely inuential Bloch and Parry edited volume Money and the Morality of Exchange (1989a), including especially their introduction (1989b), offers an obvious point of departure. Although there is no mention of Leach or the Kachin in this volume, it is important to note the continuation of Leach’s emphasis upon cultural accounts of economic phenomena. Their penultimate cross-cultural conclusion is negative insofar as the meaning of money is wholly dependent upon cultural context rather than attributable to features of cultural meaning that consistently arose cross-culturally. This even holds under colonial and other exploitative circumstances, though the cultural accounts turn in important ways to reect upon those colonial and otherwise exploitative realities. Bloch and Parry conclude that accounts that attribute cross-cultural meaning to money are reading such meanings from Western thought into ethnographic accounts. Robbins and Akin propose in their “An Introduction to Melanesian Currencies: Agency, Identity, and Social Reproduction” for their edited volume Money and Modernity: State and Local Currencies in Melanesia (1999) a two-pronged response to Bloch and Parry. Behind this response is the overall argument that, contrary to Bloch and Parry, money has unique cross-cultural features, despite the acknowledged presence of contrasting cultural responses to money, and despite the past problem of injecting Euro-American cultural notions into social science debates. One prong of Robbins and Akin’s response is that they nd unexpected similarities between local Melanesian currencies and globalized money. Sometimes such local currencies (the Kaliai studied by David and Dorthy Courts [1970 and 1977] are an important instance) can serve as “general purpose money” (Robbins and Akin 1999: 6, 11–12) insofar as such local currencies approach Simmel’s depiction of globalized money as “pure exchangeability.” This nding both calls into question Bloch and Parry’s conclusion that the meaning of money is culturally specic, as well as their related conclusion that accounts about the cross-cultural meanings for money are exclusive to Western thought. The underlying point is that accounts of social change cannot assume a priori that the introduction of money in and of itself will result in social change. The

rethinking kachin wealth ownership

219

corollary is, however, that some communities have the cultural means for controlling the ow of “pure exchangeability.” For this reason, and although currencies in general and money in particular commonly possess other features (“liquidity,” “divisibility,” “transportability,” and “concealability”), Robbins and Akin explore at greatest length the similarities and differences between local currencies and money and other objects of exchange. Robbins and Akin combine the works of Roy Wagner, Paul Bohannon, and Marshall Sahlins to come up with a new take on the notion of “exchange spheres.” Communities control “pure exchangeability” by keeping exchange spheres discrete. The cultural point is that exchange spheres contrast one another in terms of their organization out of discrete combinations of social relationships, classes of objects, and modalities of exchange. First, exchange in general is important in the society because it breaks up the ow of general human relatedness to create specic categories of social relatedness (Wagner 1977). Second, there is a relationship between the objects exchanged in terms of their cultural classication and the valuation of the exchange of one object for another in moral terms (Bohannon 1955, 1959). On the one hand, there can be an exchange of objects from within the same cultural class, in which case the moral weighting of the exchange will be either neutral or positive to all parties (conveyance). On the other hand, there can be an exchange of objects from across different cultural classes, in which case the moral weighting of the exchange is negative or otherwise adverse to one or more of the parties (conversion). Thirdly, in every exchange there must additionally be agreement between a particular mode of exchange (including, in Robbins and Akin’s usage, sharing, buying, delayed-return and exchange of exact equivalent (1999: 9) and a particular (pre-existing or expected) category of social relationships (Sahlins 1965). In the case of Melanesia, Robbins and Akin go on to observe that “pure exchangeability” can co-exist with exchange spheres by walling off such exchanges into enclaves, which are special sorts of exchange spheres that include the use of a local currency or money as a medium for value. This brings up the other prong in Robbins and Akin’s introduction: the shared features of globalized money and localized currencies that give them the ability to share features cross-culturally. Normally, there will be barriers erected between an enclave and other exchange spheres through one or more crucial distinctions in terms of social relationships, classes of objects, or modalities of exchange other than trade. Apparently, these barriers assume cultural arguments

220

ho ts’ui-p’ing

intended to give a moral weight to conventional community-wide attitudes toward “pure exchangeability.” The idea is that attitudes toward currencies and money will never be neutral, lurching between anxiety and desire concerning the importance associated with such barriers. Briey, currencies and money provoke anxiety because of their ability to run riot through a social order; alternately, they arouse desire because of this same power of “pure exchangeability”, regardless of situation. Finally, feelings towards money and other currencies involving “pure exchangeability” as a special class of objects can be ambivalent, because both share an intrinsic linkage between two dimensions: currencies “cannot be consumed” and instead “act most importantly as a means of exchange” (ibid.: 4). That is The movement of currencies tends, in fact, to be impeded by fewer barriers than that of other kinds of objects (Crump 1990: 92). This is so precisely because people are compelled to circulate them and because their use as a general medium of exchange makes them widely desirable. (ibid.: 5)

In sum, in terms of social change, money thus possesses in only comparatively extreme form the threat common to all currencies, since “the ow of currencies always threatens to exceed the controlling boundaries set up through these same social relationships” (ibid.). The formal denition that Robbins and Akin offer is: “What makes the role of money unique is that money can move against anything in any kind of exchange between people who stand in any kind of relationship to each other ” (ibid.: original emphasis). Money in particular and currencies in general are associated with the two additional features of substitution and display that Robbins and Akin note in greater or lesser detail. In each case, these two secondary features of money are important in enclaving. In both instances, attention shifts from their primary role as a means of exchange, whose unique value is in their exchangeability, and to secondary roles of money or currency. Robbins and Akin pay comparatively greater attention to display. Following Graeber (1996), Robbins and Akin argue that “Items of display index (or, in Graeber’s term “reect”) powerful actions completed in the past, and they make a claim on those who see them to treat their bearer as the kind of person who has wielded such power” (ibid.: 28). In other words, objects displayed are statements about persons and their relationships such that they express the potential power for future collective action in terms of past collective actions. In the special case

rethinking kachin wealth ownership

221

of the display of money, the power to display money or other objects that the power of money makes available have the further feature of being selectively conspicuous; that is, “. . . money is quintessentially a hidden thing and as such it comes to represent all of the unseen, internal capacities of persons” (ibid.: 28). There is, thus, the latent implication that, in its ability to be concealed, “money represents and enhances agency, an actor’s personal ability to get things done” (ibid.: 28). Other displays have their own respective powers and potentialities. Robbins and Akin place their emphasis upon adornment, although much of what they say is applicable to similar phenomena. With adornment, there need not be a backdrop, as with money, of a capacity for concealing individual abilities and powers. Leach makes this very point when he observes that, for the Kachin, ownership of movable property is not about “capital for investment”; rather it is “an adornment to the person” (Leach 1997 [1954]: 142). We will see, however, that Leach’s point is misplaced in one crucial respect, namely that “adornment to the person” can refer to past success in adventures, where money and other currencylike commodities are hidden or otherwise implicit elements in display. That is, if movable property is in this sense an adornment, potentially many other phenomena can likewise count as adornment. I would only add to the list negotiations—an often-important, spoken, performative portion of exchange—that are just as much a part of a transaction as the handing over of objects. Like money and, indeed, like other similar objects, negotiations are displayable as a form of adornment that, in certain performative situations including religious ones, evinces the power and potentiality of money and other phenomena afliated with its uses in exchange outside of that particular performative context. In thus paralleling money as display, negotiating in, say, a ritual context, displays the results of success in past market transactions. Additionally, adornments, implying the interjection of an object’s biography into performative contexts, mean that objects that move between exchange spheres in general, and between enclaves and other non-monetized exchange spheres, bring with them the powers associated with their “pure exchangeability.” I will now look into Kachin social changes brought on by colonialism in general and opium in particular. On the one hand, extensive research across upland Southeast Asia repeatedly documents that opium was once and sometimes still is pivotal to social grouping (see, for instance, Jösson 1998, 2001, Durrenberger 1989, Miles 1990). On the other hand, there is ample reason to question whether the opium

222

ho ts’ui-p’ing

economy or other related changes in monetized economies only acted on social change mechanically. Outsiders and outside wealth are indispensable to cosmology, and cultural understandings of exchange are central in constructions of personhood and sociality for the Jingpo in Yunnan along the border with Burma (Ho 1997, 1999, 2004). Early in the twentieth century, new concepts of individual ownership and new notions of personal display matched Kachin use of colonial money and opium in trade. The introduction of colonialism did not result in an inside-versus-outside contrast, such as found for “biznes” and “kustom” in Melanesia. Rather, it created a new sense of property, which legitimized the toppling of the previous rank order. With the above discussion in mind, I begin with photographs taken in the decades just before Leach’s eldwork, especially relevant for discussions of adornment-as-display. I next turn to Chinese language materials heretofore un-utilized in critiques of Leach’s Political Systems of Highland Burma.8 Review of these sources includes rst, a more-orless anecdotal description of the role that money and marketing played in the life of Muiho Mulat, a young woman who lived in the upland zone during the early 1950s. At the same time that she was aggressively involved in the pursuit of personal accumulation of private wealth, Muiho Mulat showed a culturally nuanced concern for the purchase of family goods and items of personal adornment. This coverage of the Chinese language social surveys from the early 1950s then moves on to an account of household conditions at the same time as Muiho Mulat’s trading adventures. This discussion establishes that Muiho

8 Beginning in the early 1950s, the People’s Republic of China regime initiated a series of social surveys for deciding policies of class struggle in frontier minority areas. Research in Yunnan on the Jingpo rst began in 1952, under the leadership of the Frontier Work Committee organized by the Yunnan Provincial Committee of the Chinese Communist Party and the Nationalities Affairs Committee of Yunnan Province. The purpose was to classify the nationalities according to their class relations of production. A rough, standardized social survey of production was used in each hamlet and village they went to, following their idea of the Marxist approach. For instance, the survey is done over a forty-day period in forty-one hamlets, 973 households, ranging in size from ve or six households to 80 or 90 households each hamlet in Luxi County (YNSBJZ 1985a: 1). The results of this survey are of three kinds. The rst kind is to describe certain households, hamlets, villages or areas, their means of production, modes of production, class relations in terms of production, and to count their household property, earning and expenditure. The second kind of writing is a survey of their customs and social structure. The third kind of writing is on history. This chapter use papers publishing the results of the surveys of the rst kind only.

rethinking kachin wealth ownership

223

Mulat’s trading was not unusual in being driven by cultural concerns about Kachin concepts of wealth, but rather that the same cultural concern was a pervasive motive across Kachin households even in the early years of rule in the People’s Republic of China. Photographs, Muiho Mulat’s adventures, and household surveys from the decades just before and after Leach’s own eldwork all provide a foil against which a critical re-reading of Leach’s account of Kachin wealth becomes possible. The Kachin obviously acted upon cultural assumptions about wealth objects in particular and property relations, the signicance of which Leach under appreciated. For this reason, and based upon my own ethnographic eldwork beginning late in the 1980s, I turn to a reconsideration of Jingpo cultural classications of wealth objects and property, and a comparison of these cultural classications with Zaiwa classications, hoping to show the cultural principles missing from Leach’s account. Finally, I return to my initial query about the history of Kachin social change with which this chapter began, giving my own account of the organization of exchange spheres and their growing engagement with enclaves. In my conclusion, I suggest that a different pattern of consumption based on a new concept of ownership emerged by at least the early twentieth century. In sum, then, the remainder of this chapter describes how Leach’s famous theory of Kachin social oscillation derives from his ethnographic understandings of wealth objects and property in general, as described in Political Systems of Highland Burma. I then give my own version of Kachin history, buttressed with a wider view of Kachin wealth objects and property more generally as situated within the British colonial era economy. Personal Adornment in James Henry Green’s Photographs James Henry Green’s photographs from the 1920s and 1930s document Kachin personal adornment. These photographs are useful for establishing the importance and extent of wealth for personal adornment.9

9 James Green began his service in the Eighty-Fifth Burmese Ries in 1918, appointed as commander of the Kachin Regiment in 1920, and served as Recruiting ofcer for “Hill Tribes of Burma” between 1922 and 1927 (Dell 2000b: 182). What he wrote about the Kachin and the pictures of the Kachin are from his witness of them from 1918 to the mid-30s (Dell 2000a). For interested readers, other photographs on the ethnic groups at the border of China and Burma taken in the mid-30s

224

ho ts’ui-p’ing

In Green’s picture numbered 0742, it shows the family of the chief of Namhkyek—the “most powerful chief in Kachin territory,” considered by Green—dressed and seated on a beautifully patterned mat (Sadan 2000b: 71).10 There are ve adults and two babies in the picture. Both men (chief and his brother) wore Shan style outts, with white turban and shoes. The three women all wore bracelets, rings, the velvet top decorated with many silver buttons and layers of necklaces, and leggings. The older women wore large-sized headbands, obviously using a large amount of cloth. The two younger women wore fullpatterned, hand-woven skirts. Two women were holding one baby each on their lap. The babies were all dressed in good clothing. One baby wore a fancily decorated hat. These are all costly wealth objects. These costly wealth objects are worn not only by the powerful chief ’s family posing for picture taking, but also by commoners, girls and ladies in the manau celebration and the market. The two Atsi (Zaiwa) ladies in photograph 0357 (Dell 2000b: 127), wear handsomely the hand-woven wool skirt with yarn woven patterns, long waistband and big turban made of market bought cloth, different kinds of necklaces and big earrings. The ladies in picture 0250 (Sadan 2000b: 72) were displaying their best, too, through their headdresses. Green gave a fantastic caption for this image “Two Old Kachin ladies dressed in the latest fashions at the Namhkyek manao.” Manau is the most important community occasion for personal display besides funerals. One woman’s headdress is made of beautifully patterned cloth styled in a turban that was about two and half times the size of her face. The accompanying description records: Everyone would wear their best clothes to dance at the rst day of a manau. These ladies are wearing Palaung headdresses, revealing that they were in vogue amongst Kachin women at this time. (Sadan 2000b: 72)

for the purpose of national boundary survey are also available. In 1935 and 1936, the Chinese anthropologist Rui Yifu, along with the photographer Yong Shiheng, left valuable pictures of the area. These pictures are now being digitalized by the National Digital Archival Program in Taiwan at the following website: http://ethno.ihp.sinica. edu.tw/. The Chinese historian and anthropologist Jiang Ying-liang’s photographs on the Chinese southwest nationalities in 1937 have now been compiled and published by his son ( Jiang and Jiang 2003). 10 As pointed out by Sadan, when she showed the pictures to the Kachin people recently, they “initially point out that he is wearing shoes.” Sadan very correctly interprets that wearing shoes was “a politicized act in both India and Burma, and, in this case, a sign of close contact with the British and evidence of some political aspiration.” (2000b: 72). However, I want to direct the attention to another viewpoint of the display of the wealth and social distinction here.

rethinking kachin wealth ownership

225

If the above photographs were all taken at the celebratory occasion, or posed for picture taking, this explains the strong characteristic of display; photographs 0293 (Dell 2000b: 107) of “Kachin girls selling spirit” shows another occasion for display: the market. Three girls are all in their best attire for their age, with necklace, earrings, bracelets and silver buttons decorating their clothes and bodies. Muiho Mulat goes to Market One early People’s Republic of China (1949–present) source (YNSBJZ 1985a: 90–91) gives a detailed account of the marketing activities of a woman named Muiho Mulat for the period beginning in February 1951 and ending early in 1952. Muiho Mulat rst sold two liang of her family’s opium to lowland Dai, from which she received in return a basket (luo) of peas.11 She then sold the basket of peas at the highland market town of Liangzi on the Chinese side of the border for 10 wen in bankai silver currency.12 Muiho Mulat nally used the 10 wen to buy 100 chicken eggs, which she then sold in the market town of Jiugu on the Burmese side of the border, and then used the resulting 20 wen to buy 10 zhuei of salt,13 20 bars of soap, and one cotton blanket, and to pay for tax and transportation expenses. Muiho Mulat made nine more cross-border trips between the Liangzi highland market and the Jiugu lowland market, continuing to trade in the same commodities. During these trips, she bought opium, chicken eggs, and mountain-grown luzi (betel nut) in upland markets and sold them for salt, soap, and blankets in lowland markets. When the Jiugu market was destroyed during ghting on the Burmese side of the border, she switched to trade exclusively on the Chinese side of the border, selling luzi in the lowland markets of Mangshi and Zhefang, and selling

11 Luo was the weight counter for things contained in baskets, such as peas or rice. Before PRC standardization, each luo of grain equaled to 30 jin (catty) in the mountain and 25 jin in the lowland. 12 1,000 wen equalled one liang of silver. Introduction of Bankai Yunnan’s ofcial currency took place in the Qing dynasty in 1887. The Bankai was recognized by the Republic of China (1911–1949) central government as Yunnan’s de facto ofcial currency by 1914 and was fully recognized as the ofcial currency by governor Long Yun in 1927. In April 1950, the People’s Republic of China ofcially banned the circulation of Bankai and replaced it with rmb. 13 Zhuei was the weight counter used in Yunnan. Before PRC standardization, each zhuei equalled three jin (catty) in the lowland and four jin in the highlands.

226

ho ts’ui-p’ing Table One: Items Mulat Bought at Market

Items Purchased

wen

12 liang of yarn Four sets of market bought clothes One female velvet top Yellow thread One silver bracelet Three sets of earrings Lacquer waistband Five beads for necklace Hair oil Two sets of red string Two sets of silver buttons Two pairs of straw sandals One piece of cloth for father One set of market bought clothes for a younger brother Total

5 24 12 1 8 2 1 6 0.5 3 1 0.5 8 3 75

salt and sugar in the highland market at Liangzi (YNSBJZ 1985a: 91). Muiho Mulat’s trading earned her a net prot of 125 wen and she used 50 wen to pay a ne levied against her elder brother (YNSBJZ 1985a: 90–91). She used the remaining 75 wen for 14 purchases. These 14 purchases included seven for personal adornment, three for weaving patterns for hand-woven skirts, and four for sets of clothes or cloth for herself and her family (see Table One). Muiho Mulat’s marketing was initially funded, importantly, with homegrown opium. In the end, except for the ne, the vast bulk of the other purchases were heavily tilted in favour of personal adornment for herself and her other family members. Early Era PRC Social Surveys Chinese language social survey accounts from the early 1950s for several communities in what is now the Dehong Jingpo and Dai Autonomous Region, just across the border from Burma, document household economic conditions for that time. In one “average” household in Nanjingli village cluster and one “wealthy” household in Huyu village cluster, both in Ruili County right on the Burmese border, it is possible to see the high degree of investment in “wealth objects,” and especially so for wealthy households (see Table Two).

rethinking kachin wealth ownership

227

Table Two: Chinese Language Social Surveys on Household Wealth in Zaiwa Communities in Ruili County from the Early 1950s Wealth Category

“Average” Household in Nanjingli Village Cluster

“Wealthy” Household in Huyu Village Cluster

Domesticated Animals

Three head of cattle One pig Four dogs Four chickens One cat

Twenty head of buffalo Fifty-ve head of cattle Thirteen pigs (raised jointly with other households) Fifty chickens Two dogs Two cats

Tools and Utensils (Agricultural Tools and others)

Five ploughs (in bad shape) Three hoes Two chopping knives Three at long knives Four sickles (three in bad shape)

Two ploughs Two hoes Two chopping knives Two long knives Two shovels One hatchet Fifteen carrying saddles

(Household Utensils) Two tripods (for woks) Two iron woks Two pots One steamer Three earthen jars Seven bowls One water container One copper spoon One teapot and tea cup Four glass jars Two lamps Bamboo baskets of various size for storage and for carriage Two sets of mortar and pestle

Nine iron woks of various sizes Two tripods One copper tea kettle One antimony tea kettle Two antimony bowls for food Five antimony plates for food Three enamel bowls Ten bowls of various size Ten big earthen jars Twenty three small earthen jars Two antimony spoons Bamboo baskets of various size for storage and for carriage

228

ho ts’ui-p’ing

Table Two (cont.) Wealth Category

“Average” Household in Nanjingli Village Cluster

“Wealthy” Household in Huyu Village Cluster

Cloth and Clothing Three cotton blankets Three shoulder bags Eight sets of purchased clothing One hand-woven skirt Two headbands One bamboo mat Four bamboo hats

Fifteen hemp bedrolls Eight cotton and wool coverlets Eight wool bedrolls One wool coverlet Three comforters Three umbrellas Five hand-woven skirts Three headbands One woollen hat Three sets each of purchased clothing for every household member

Ritual Wealth Objects

One gun Two diggers One long spear

Three large gongs One ceremonial knife with silver handle Four different kinds of gun

Furniture

One bamboo table Four bamboo stools

One large wooden cabinet Two wooden chests One wooden cofn

Cash and Jewellery

Four silver bracelets

400 dun in British Burmese coins Jewellery valued at 300 dun

Sources: For Nanjingli village cluster, see YNSBJZ 1985a: 140; for Huyu village cluster, see YNSBJZ 1985a: 153.

One major difference between the “average” and “wealthy” households that is relevant but can only be noted in this chapter, is that the latter owned sufcient agricultural land to rent some, while the former owned at most only enough land to work themselves. From their household wealth, it is also obvious that the “wealthy household” owned considerably more domestic animals, as well as every kind of object bought

rethinking kachin wealth ownership

229

from markets. This included more tools and utensils (iron woks and tripods of various sizes, tea kettles, pots, bowls, and baskets), furniture, cloth and clothing, ritual wealth objects (gongs and so forth), jewellery and money. Annual income and expenditure is presented for several “typical households” in Xishan Township in Luxi County (1985a: 83–101).14 First, cash income came from growing opium, selling wage labour, handwoven cloth shoulder bags and bamboo baskets. Second, additional cash income also came from trade between lowland and highland markets, as well as in cross-border markets in Burma, in such commodities as: domestic animals, opium, salt, cotton, soap, blankets, clothes, and yarn. Third, daily cash expenditures were primarily for salt, soap, opium, cotton, clothing, tools, and kitchenware. Fourth, there were “occasional expenditures” on “ritual wealth objects” for households with a wedding or funeral, and items of personal adornment—in the form of gongs, knives, guns, silk clothes, yarn, and jewellery. One obvious observation is that the line between “average” and “wealthy” was not only drawn in terms of ownership of the means of production, but included also the ownership of wealth objects that are or can be commodities. From the income and expenditure survey, the expenses for these wealth objects accounted for a signicant portion of consumption, especially for households with wedding and funeral expenses. In a case of one “exemplary household with wedding and funeral expenses,” the wedding expenses for a son were 94.71 percent of that year’s income, without counting the debt still owed the afnes, or 138.1 percent of that year’s income when counting the debt still owed the afnes.15 For a similar case surveyed in 1957, wedding expenditures

14 Not all reports do so. Among the four volumes of Socio-historical Survey of the Jingpo Nationality, volumes three and four provide no quantitative details on incomes or expenditures. 15 The report was done with the purpose of evaluating statistically whether people are wasteful or what kind of expenditure is wasteful. Considering the fact these expenses were counted with only their market value when weddings and funerals had taken place, without understanding the relationship or the mode of exchange behind the different objects used and the varied spheres of exchange involved, the gures are totally meaningless for our understanding of value. The reason I am still using these materials, contrary to the research in the 1950s, is to show the signicance of weddings and funerals to Jingpo life. I show the proportion of these expenses in their annual income rather than expenditure, as in their reports.

230

ho ts’ui-p’ing

still accounted for 105 percent of income (based on data of YNSBJZ 1985b: 15–16). Funeral expenses for the household of the deceased mother, who hosted the funeral, took up 62 percent of income (based on data of YNSBJZ 1985a: 83–84). Expenses used in weddings were mostly of three kinds for the man’s household. One kind was for the “wealth objects” used as bridewealth, an other was for food and drink to be consumed at feasts, and the third kind of expense (in money or the equivalent of money) was for ritual specialists. Bridewealth for a 1952 wedding of an ordinary family in Luxi County included: two heads of cattle, one large gong, two zhang (丈) of cloth,16 two sets of clothing (both purchased at local markets), and two kinds of bedrolls. For the wedding banquet, six pigs were slaughtered, two hundred bowls of liquor brewed, and seventy chicken eggs were consumed. In addition, uncertain quantities of the following objects were used: rice, glutinous rice, salt, luzi (betel nut). The above, moreover, included for the bride’s family one hind leg of a pig, thirty bowls of liquor, and 21 eggs. Finally, the ritual specialists were paid with ve liang of opium ( YNSBJZ 1985a: 87–88). Expenses for the “wealth objects” for the wife-givers, discounting food used in the feast, took up 53.8% of the total expenditure. The “wealth objects” used in a more wealthy household in the same county included ve head of cattle, three pi (匹) of cloth purchased from a market, three kang (亢) of money, one large gong (YNSBJZ 1985a: 84).17 The expenses for these wealth objects accounted for 75.9% of the total expenditure on the wedding. Rare but important information on the relationship and the mode of exchange involved in wealth objects and food used in weddings is unexpectedly available. In the just cited case of the wedding for an ordinary household (YNSBJZ 1985a: 87–88), the husband of the sister of the groom provided the cattle and the gong portions of the bridewealth. Additionally, the market-bought cloth came from an unspecied “cousin.”18 These afnal contributions count for 91% of the expenses

16 Zhang is a Chinese unit of measurement in length. One zhang equals 10 chi (尺) or 100 cun (寸). Each cun is equal to 1.23 inches or 3.1242 centimetres. 17 Pi is a Chinese counter for clothes, though what qualies as a pi in terms of width and length, varies from place to place. Wu Cheng-ming cites one pi of cloth as having a width between 0.9 and 1.15 chi and the length between 16 to 32 chi. Between 0.9 and 1.15 chi is equal to about 27 to 34.5 cm; 16 to .32 chi is equivalent to about 480–960 cm. 18 Since the source cited is in Chinese, the Chinese kinship term used does not specify

rethinking kachin wealth ownership

231

on “wealth objects” used in the wedding. Finally, agnatically related households and the households of village neighbours provided the rice for the wedding banquet. At the marriage for the wealthy household ( YNSBJZ 1985a: 84), the groom’s household turned over only ve of the stipulated nine head of cattle for the bridewealth, with the remaining four head of cattle still owed. For reasons not stipulated in the report, the groom’s household only “planned to give” three of the four head of cattle still owed. People did not own certain wealth objects as “possession” for sure, but rather owned the debt of the wealth, as Leach said (see section below on “Wealth in Upper Burma”). However, some other kinds of wealth objects, such as opium, market-bought clothes, and personal adornments, were different. For some households, opium smoking was the greatest expense. In 1954, opium consumption for one wealthy household in Huyu village cluster in Ruili County took up 46.7 percent of the yearly household expenses. ( There were no expenses for weddings or funerals that year) (based on data in YNSBJZ 1985a: 153). In one household in Luxi County, only the host took opium. His opium consumption took 6.9 percent of the whole household expenditure in 1953 ( YNSBJZ 1985a: 93). In a 1957 survey in Xishan Township, Luxi County, one hamlet as a whole used 12 percent of their expenditure that year in opium smoking ( YNSBJZ 1985b: 15). In the 1957 survey in a hamlet in Longchuan County, 9.3 percent of their yearly expenditure was used in opium smoking ( YNSBJZ 1985b: 35). Generally, men in Xishan Township in Luxi County wore clothing purchased from markets.19 Men usually wore the same set of clothing regardless of season, for at least a whole year. Men normally only purchased a new set of clothing after the old set was completely worn out, with only a wealthy man purchasing two sets of clothes in a single year. Women made their tops out of cloth purchased from a market, with some adding decorations of copper or silver buttons. In Xishan at

whether the “cousin” is a patrilateral cross cousin, or either a matrilateral cross or parallel cousin. It does make clear, however, that the “cousin” was not a patrilateral parallel cousin. In Jingpo kinship, these distinctions would be clearly drawn by the terms. My own reading of the passage is that the “cousin” referred to was most likely a patrilateral cross cousin, because market bought cloth is culturally a wife-takers’ prestation in the Jingpo classication of things (see below the section on “Wealth in Yunnan”). 19 The report also provides information on how the Han Chinese living next to the Jingpo with sewing skills will sell their labor to make clothing for the Jingpo ( YNSBJZ 1985a).

232

ho ts’ui-p’ing

that time, women wove the cloth for their own skirts ( YNSBJZ 1985a: 61), or purchased the cloth in Ruili near the Burma border ( YNSBJZ 1985a: 117). The women wove skirts using cotton with wool thread or yarn. People normally slept without covers, though some households had blankets woven from hemp and others could even afford blankets woven from “cotton from India” ( YNSBJZ 1985a: 117) or bedrolls, as mentioned earlier ( YNSBJZ 1985a: 87). The wool and yarn used in the hand-woven skirts, according to my research, were bought from markets (Ho 2000). When the weather turned cold, people wore woven shawls ( YNSBJZ 1985a: 61, 117). The better off also wore velvet tops with decorative buttons, leggings, hand-woven cotton and wool shoulder bags decorated with silver dollars, coins, or copper buttons (ibid.: 117). A large headband took up more cloth than a small one. The survey describes the cloth headband of a middle-aged woman, which was 1.5 zhang long or 468 centimetres in length. The large headband and the hand-woven wool skirts with patterns made of yarn were to some extent indicative of a household’s wealth (see below). The issues raised in the social surveys of the early era of the People’s Republic of China are multiple and suggestive of a variety of interesting tangents. Overall, the Chinese language sources show for the midtwentieth century that “wealth objects” differentiated between those who were wealthy and those who were poor. Aside from the regular living expenses, a signicant outlay of expenditures went, when possible, on the purchase of “wealth objects,” and especially those used for afnal exchanges. However, if we take into consideration the typical domestic economy, the wealthy owned a signicant number of “wealth objects”, beyond that needed for afnal exchanges. Those bridewealth objects that were for use as afnal exchanges, were circulated rather than accumulated, such as cattle and cloth given by the wife-takers but actually used and owned in bridewealth for the household of the wife-givers. Negotiation and delay-exchange were also possible for such bridewealth as cattle, gongs and clothing. That is, accumulation of a large portion of the “wealth objects” that the wealthy owned was for purposes other than afnal exchanges. The wealthy were instead accumulating “wealth objects,” household and personal wealth objects, for self-distinction, status competition and fame. These “wealth objects” include extra head of cattle, various kinds of wood furniture, additional items of personal adornment (such as jewellery, clothes and clothing, cotton and wool blankets) and consumption (such as opium), ceremonial knives, guns and money.

rethinking kachin wealth ownership

233

In sum, Muiho Mulat’s trading activities were something to which almost every Jingpo aspired. That desire was the pursuit of money, with its ultimate aim the distinction it brought through acquiring wealth objects. For Muiho Mulat, a girl, the self-distinction is further clearly made through personal adornment. However, it is also worth noting that she shares her money with the family generously. Aside from clothing bought for the family, forty percent of her prot is used to pay for her brother’s ne to the unspecied authority. Wealth in Upper Burma In Political Systems of Highland Burma, Leach devoted a lengthy passage in Chapter Four, “The Structural Categories of Kachin Gumsa Society”, to a section entitled “Concepts of Property and Ownership” (Leach 1997 [1954]: 141–154). In so doing, Leach attributed a role in social change to the cultural classication of property and ownership. This is because the cultural classication of property and ownership was an important grounding for action in rituals, associated exchanges, and their cumulative consequences for social change. Thus, “before we can understand the ritual implications of ‘owning a debt of ve wealth objects,’ we must comprehend something about the practical application, in the Kachin context, of ownership, debt and wealth” (ibid.: 104, original emphasis). Ownership. Ownership takes two major forms (ibid.: 141–142). First, madu ai (“to own”) is like “sovereignty” (“the rights of a ruler”), which is a kind of ownership “in the sense of ‘having rights over something or some person.’ ” Second, lu (“to eat”) or sha (“to drink”) is like “usufruct,” a kind of ownership “in the sense of ‘having, and therefore being able to enjoy for the time being’ ” (ibid.: 142). Two further points about Leach’s discussion of Kachin ownership deserve special emphasis. First, whatever the occasion, a recipient of an exchange has the rights and duties of lu or sha over the objects, while the giver retains the rights and duties of madu over those objects (ibid.). Second, these two major categories of ownership are as common to the Shans and Burmese as they are to the Kachin (ibid.). Debt. Leach grants the Marxist point that power relations revolve, in the nal analysis, around “the control of real goods and the primary sources of production.” However, Leach hastens to add: “The way in which particular goods and services are evaluated one against the other is a cultural phenomenon which cannot be deduced from rst principles”

234

ho ts’ui-p’ing

(ibid.: 141). Thus, at another level, the meaning of the Jingpho term hka (“debt”) comes quite close in Leach’s view to what anthropologists mean by the terms “sociological relation” and “social structure” (ibid.:). Interestingly, however, what Leach talks about when discussing hka is how it varies across what he calls “occasions.” One occasion for hka was trade. In Kachin trade, there was the use of currency or currency-like media (bullion, opium), where market transactions could convert any available sort of goods or services into other sorts of goods or services. Trade need not result in hka when the agreed upon price for and transfer of goods or services was completed at the same time. Hka was apparently present only when the nal transfer of goods or services remained incomplete. The handling of hka in trade could take on a local Kachin colouring. That is, in remote areas, there still persisted the practice for hka in trade to be calculated and certied using bamboo tally sticks as contracts. First, a length of bamboo was slashed crosswise, equivalent to the number of hpaga (see below) owed. Each slash potentially represented a different discrete category of wealth objects, where each slash represented an object of a specic category for which the debtor party owed repayment. Second, split lengthwise, each of the two identical halves of the length of bamboo preserved its record of half of each of the original crosscutting slashes. Each party to the trade then had their respective copy of what was a contract certifying the outstanding hka, although what the actual objects referred to for each slash had to be committed to memory. The important point is that there was no reduction of the repayment to a common index of monetary value. In addition to trade, Leach named ve other “occasions” involving hka. Hka of the sort relating to at least these ve occasions was generally set within relations between lineages (ibid.: 153). The ve “occasions” were: (a) marriages, (b) funerals, (c) in payment of ritual services by priests or agents, (d) on the occasion of a transfer of residence or the building of a new house, (e) as judicial compensation in settlement of any kind of dispute or crime. (ibid.: 147)

Hka involving judicial compensation seems to have been an especially wide-ranging occasion. Leach says, For the Kachin, legal claims and commercial claims are alike hka (debt). The only difference is that with commercial claims the items may be anything, depending on the circumstances of trade, while, with legal claims, the items are stereotyped according to the traditional pattern. (ibid.: 146)

rethinking kachin wealth ownership

235

And, more generally, Kachin tradition and ritual lays down what are the proper relations between individuals, that is to say, it species what obligations A has towards B and B towards A. Debts come into existence whenever anyone feels that these formal obligations have not been adequately fullled. (ibid.: 154)

Three further points about hka as it relates to judicial compensation with regard to debt merit brief mention here, though I will return to these points again later when discussing the cultural classication of wealth objects. First, feuds resulted from a failure to meet hka claims for judicial compensation. Indeed, feud and debt were also both referred to as hka (ibid.: 153). Second, negotiations for actual settlements of claims for hka judicial compensation took place within the framework of the class system: If the two parties in a debt relationship came within the jurisdiction of a single chief, then it is up to the chief to see that they come to some agreement about the terms of compensation. . . . Settlement when it is achieved is always the result of negotiations by third parties (kasa) who are usually persons superior in social standing to the principals. When two chiefs are in dispute, it may be difcult to nd anyone senior enough to act as kasa, and it is cases of this kind that are most likely to degenerate into feud. (ibid.)

Third, hka were only “scaled according to class” as formal principle, since: “In practice the payment depends on the economic standing of the defaulter not on his class status by birth” (ibid.: 148–149). Thus, a strategic choice between class and economic interests confronted a violator: whether to defend his economic interests by hardnosed bargaining, or to “validate” his class standing by paying what his class standing required. That is, “the validation of class status depends more than anything else on an ability to full correctly the gift-giving obligations that are proper to a member of that class” (ibid.: 149). Although haggling presented the holder of a status with an alternative means for meeting obligations, there was a loss of face and a risk of a lowered status. Leach concludes: “Paradoxically therefore it is often true, especially of the more enterprising individuals, that they pay as much as they can afford rather than as little as they can haggle for” (ibid.). Wealth. Leach classies wealth into the two categories, sut (“moveable wealth”) and “land.” (A discussion of land is beyond the scope of this chapter.) Leach further identies three sub-categories of sut: (a)

236

ho ts’ui-p’ing

shahpa (“ordinary perishable foodstuffs”) and forest products, which can be traded; (b) livestock, including especially nga (cattle); and, (c) hpaga (“goods and objects used in ritual exchange other than livestock”). However, in looking more closely into hpaga, Leach nds the above three sub-categories of sut were subject to other cultural considerations: . . . the notion of hpaga, in the sense of ‘ritual wealth object’, includes, not only all the items in category c [hpaga], but also the water buffalo in category b [nga] and certain items (such as opium, slaves and bullion) which in the past were extensively traded on the open market under category a [shahpa]. It is not true therefore to assert that ritual wealth objects have no ordinary commercial value. What is true is that some types of ritual wealth object have no ordinary commercial value, and that the value of a wealth object used in ritual exchanges is not in any case wholly determined by its ordinary commercial value in the open market. (ibid.: 144).

In fact, Leach sees that: “The relationship between hpaga the material fact, and hka the immaterial debt is rather similar” (ibid.). Hpaga thus included “the double sense of ‘trade’ and ‘ritual wealth object’” (ibid.). On the one hand, then, “In one sense hpaga simply means trade” (ibid.: 146). Yet, on the other hand, there were additionally ways in which the handling of hpaga on other occasions than trade pointed, as with hka, toward other heretofore unconsidered senses of hpaga. How Leach specically described the cultural means through which hpaga unites this “double sense” of trade and ritual revolves around three general points. First, he discusses hpaga as a verbal concept preserved in “poetic stanzas” with their own “poetic titles” that evoked the social rationale for the giving of the items named (ibid.: 148). Stated in general, these and other verbal concepts gave expression to and singled out for attention the stylized cultural classication of wealth objects and their relationship to the formal statuses and classes of the idealized social system (see also ibid.: 10–17). Second, though he also notes that there are some regional differences (ibid.: 147, note 62), Leach provided a more-or-less conventional Kachin list of hpaga, as well as several concrete instances of how they were suited to particular variants of the ve different occasions.20 The 20 Leach cites (1977: 147) J. L. Leyden’s note given in Kawlu Ma Nawng’s The History of the Kachins of the Hukawng Valley (1942: iv, 68) for a long list of traditional orthodox hpaga. That list is: “(1) a buffalo, (2) a gong (of several different types), (3) silver bullion, (4) a slave, (5) a cooking tripod, (6) n’ba (several types of shaped cloth which serve as male skirts, blankets and shawls), (7) an iron cooking pot, (8) sword (usually a dummy blunt edged one), (9) a spear (also usually a dummy), (10) a sheepskin coat, (11) a silver

rethinking kachin wealth ownership

237

value of the hpaga of this list had a radically different conguration from that for trade objects. For instance, the guns and pistols were often rusty or broken, and the spears and swords were not even genuine weapons but rather “blunt-edged models of no practical utility” (ibid.: 143). Although there is enough qualication (see ibid.: 144) in Leach’s account to allow for some trade in ritual objects, wealth objects divide into those for trade and which could serve as substitutes, and those for ritual (the other ve “occasions”) that were not traded. Alternately, the sense of “ritual objects” refers to certain hpaga as ritual objects, exchanged during the ve “occasions” other than trade. Third, all ritual hpaga could nevertheless have substitutes in one of apparently two ways. To begin with, there was a limited range of items where substitutions occurred, possibly in name, when hpaga of the proper category became unavailable. Specically, “ordinary cash frequently appears as a hpaga in a settlement but it is always a substitute for some items which circumstances make unobtainable, e.g. a slave, bullion, opium” (ibid.: 147). Alternately, it was possible to substitute in kind but, where it is explicit, the substitution was not in name. For instance, pigs substituted for water buffalo. The point is that: . . . Although the theoretical form of each hpaga is meticulously detailed, greater stress is laid on the number and title of the hpaga than on its outward form. The real payment is always a matter for agreement between the parties and here the principle of substitution (sang ai ) is all important. (ibid.: 148)

Leach asserts more generally that: What is of especial importance here is the exibility of the system. By manipulating the principle of substitution to its limits a poor man owning only a few pigs and chickens, and a rich man owning many buffalos can both appear to conform to the same formal code of gift giving. Although they do not in fact contribute goods of the same economic value, they do, by a ction, contribute the same hpaga. (ibid.)

Ritual Implications. Other than trade and judicial compensation, marriage is the only other “occasion” that Leach discussed at length in relation to

pipe, (12) opium, (13) a Chinese embroidered silk coat, (14) bead necklaces of a special type.” In judicial compensation, for instance, specic prescriptions about what objects to use and the order of their transfer were made to overcome thematically and processually the social rift (an example of the thematic and processual features of restitution for “cattle of a buffalo” is given on Leach 1977 [1954]: 147–8).

238

ho ts’ui-p’ing

the cultural meanings of hpaga. His discussion of marriage led him into further exploration of the relationship between ritual and its associated exchanges, with their consequences for social change. The difference between judicial compensation as a wide-ranging practice and marriage as a more specic practice turns, again, on the balancing of class and status interests, though with two important twists. First, whereas in principle the woman’s class status determined the debts incurred in a marriage, in practice the man’s status determined the agreed upon amount of such payments. Leach observed about the latent signicance of this for social change: The importance of this fact is considerable. We have seen that a crucial element in the structure of gumsa society is that when an individual marries out of his or her own social class it is normally the man who marries up and the woman who marries down. If bride price in such cases were xed according to the status of the bride, the system would break down, for the men of junior status would seldom be able to raise the necessary quality of cattle and hpaga. Nevertheless, despite what happens in practice, Kachin formal theory is that bride price is adjusted to the standing of the bride. It is a theory which permits a powerful chief to pick and choose among potential suitors for his daughters and to use their marriage as direct instruments of political alliance. (ibid.: 151, original emphasis)

And, most generally, . . . The concept of hpaga is of great signicance, for it permits structural rules which have all the appearance of rigidity to be interpreted very freely, thus opening the way for social mobility in a system which purports to be a caste-like hierarchy. (ibid.: 152)

Second, although hka claims could be met over a long drawn out process in either judicial compensation or marriage, there were markedly different attitudes toward these two occasions when the terms of the hka claims came to be negotiated and obligations met: It is especially debts between strangers that must be settled quickly otherwise the owner of the debt has a legitimate excuse for resorting to violence; in contrast, debts between relatives, especially afnal relatives, are not urgent matters. Indeed as between mayu and dama some debts are always left outstanding almost as a matter of principle; the debt is a kind of credit account which ensures the continuity of the relationship. There is thus a kind of paradox that the existence of a debt may signify not only a state of hostility but also a state of dependence and friendship. To the Kachin way of thinking co-operation and hostility are not very different. (ibid.: 153)

rethinking kachin wealth ownership

239

And, most generally, . . . . Hpaga are a kind of device for manipulating social status and they are used in a game which proceeds according to set rules. . . . In the theoretical system the value of any particular hpaga is ritualistic and symbolic; in real life the actual hpaga are only substitutes for the traditional objects. The real hpaga have both ritual and economic signicance at one and the same time. (ibid.: 154)

Finally, earlier in this same chapter is an important ethnographic insight about the relationship between movable property and personal distinction: Kachins do not look upon movable property as capital for investment, they regard it rather as an adornment to the person. The word sut which is used to denote riches can also be used adjectivally to mean ‘smart’ in the sense of ‘a smart coat’. Wealth objects other than ordinary perishable foodstuffs have value primarily as items of display. The best way to acquire notoriety as the owner (ruler) of an object is publicly to give possession of it to someone else. The recipient, it is true, then has the object, but you retain sovereignty over it since you make yourselves the owner (madu) of a debt. (ibid.: 142, my emphasis)

It is useful to relate the above overview of Leach’s discussion of the cultural classication of Kachin ownership and movable property back to the earlier summary of money, enclaves, and exchange spheres. Leach’s data appears to describe at least three exchange spheres, one of trade, another a marriage related exchange sphere, and the third one of judicial punishment. In the enclave of trade, the items of exchange are potentially unlimited, though ritual wealth objects are seldom included, and the social relationship is unlimited. In the enclave of marriage exchange for women, the items of exchange are of two kinds: one is the ideal exchange, both theoretically and in name, in which they exchange “ritual wealth objects” hpaga; the other kind is the trade objects hpaga, which are often used as substitutes for the ritual wealth objects. The mode of exchange is often delayed exchange. The exchange sphere of judicial punishment requires “ritual wealth objects” hpaga, both theoretically and verbally, but, in practice, they are always substituted for the trade objects hpaga. Their mode of exchange is more immediate than for marriage exchange, even though some room for time extension is still possible in comparison with trade sales.

240

ho ts’ui-p’ing Wealth in Yunnan and Its Display

My own recent eldwork in Yunnan provides supplementary understandings about the cultural classication of wealth objects not explored in Leach’s Political Systems of Highland Burma. Consideration of the cultural classications of wealth, and the ownership concept behind this wealth, for the Yunnan Jinghpaw and Zaiwa suggest four linked conclusions. First, from the cultural classication of their wealth, it shows that understanding the nature of hpaga is not sufcient for an understanding of their wealth ownership. Second, from a closer look at what and how things are exchanged, and the relationship created in the different sub-sphere of marriage exchange among the Jinghpaw and Zaiwa in Yunnan, I nd Leach overlooked the intricacy of Kachin concepts of wealth. Two reasons caused this ignorance: one was his “object-center model” of hpaga research; the other was his a priori assumption that trade objects hpaga equated with “pure exchangeability.”21 Third, lead by his oscillation model for social change and the assumption that trade objects hpaga was “all purpose money”, Leach reduced Kachin wealth objects to a dualism of ideal/ritual wealth objects vs. real/trade objects in substitution. Finally, by failing to explore more deeply how different kinds of substitution and varied strategies of display were involved in marriage exchange, Leach failed to see the difference between owning something relationally and owning possessively in Kachin society. In this section, I begin my discussion with a description of the Yunnan Jinghpaw, and then turn to a description of the Yunnan Zaiwa cultural classication of wealth (see Diagram One and Table Three). Afterwards, I describe in more detail the different sub-spheres in Yunnan Jingpo marriage exchange from objects exchanged, mode of exchange, and the relationship of exchange each sub-sphere builds. In order to engage my discussion with Leach, special attention is given to how substitution takes place and how people use different strategies of display in establishing their ownership. Finally, I will present my own idea about Yunnan Jingpo wealth ownership and its relation to understanding social change. The cover term in Yunnan Jinghpaw for wealth, which is common to the term described by Leach for Burma, is sut gan, which subdivides

21 “Object-center model” is used by Robbins and Akin in criticizing Bohannan’s model of exchange sphere (1999: 10).

pegva je [z] Household Wealth

hpaga rai [ j] Household Wealth Objects

pau je [z] Bridewealth

a rai [ j] Wealth Objects

Zaiwa

yvum wang je [z] House-Entering Wealth

igvun je [z] Valuable Wealth

shirung je [z] Dowry and Wife-Giver

chung je [z] Utensils

sharung shagau [ j] Dowry and Wife-Giver Wealth

a ja [ j] Valuable Wealth Objects

hpuja or hpaji hpaga [ j] Bridewealth

je [z] Wealth Objects

kun rai [ j] Utensils

Jingpo

Diagram One: Jingpo and Zaiwa Cultural Classication of Wealth

rethinking kachin wealth ownership 241

242

ho ts’ui-p’ing

Table Three: Wealth Terms for Yunnan Jinghpaw and Zaiwa Speakers Jinghpaw

Zaiwa

Translation

Items

a rai [ j]

je [z]

Wealth Objects

For Jinghpaw-speakers, “wealth objects” refers to one of the two major subcategories of wealth. ( The other category is “Valuable Wealth Objects” [sharung shagau ( j)]). “Wealth Objects” (a rai [ j]) subdivide into: (1) “Household Wealth Objects” (hpaga rai [ j]) and (2) “Tools and Utensils” (kun rai [ j]). For Zaiwa speakers, separate term ( je [ze]) that includes (1) “Household Wealth Objects” ( pegva je [z]), (2) “Valuable Wealth Objects” (igvun je [z]), and (3) “Tools and Utensils” (chung je [z]).

a ja [ j]

igvun je [z]

Valuable Wealth Objects

For Jinghpaw speakers, the term for one of the two major categories of wealth (the other category is “Wealth Objects” (a rai [ j]). “Valuable Wealth Objects” (a ja) subdivide into (1) “Bridewealth” (hpuja or hpaji hpaga [ j]) and (2) “Dowry and Wife-Giver Wealth” (sharung shagau [ j]). For Zaiwa speakers, the term for one of three subcategories of “Wealth Objects.” “Valuable Wealth Objects” (igvun je [z]) subdivide into (1) “House-Entering Wealth” ( yvum wang je [z]), (2) “Bridewealth” ( pau je [z]), and (3) “Dowry including WifeGiver Wealth” (shirung je [z]).

rethinking kachin wealth ownership

243

Table Three (cont.) Jinghpaw

Zaiwa

Translation

Items

hpaga rai [ j]

pegva je [z]

Household Wealth “Household wealth objects” Objects includes all market-purchased items used for all rituals, feasts, and festivals as well as opium, cattle, tea, jewellery, clothing, guns, ceremonial knives and any other commodities.

kun rai [ j]

chung je [z]

Tools and Utensils Tools and utensils include water tubes, woks, bowls, ploughs, hoes, knives, and stools.

-----

yvum wang je [z]

House-Entering Wealth Objects

Zaiwa term for “HouseEntering Wealth Objects” typically includes tripods and woks.

Bridewealth

“Bridewealth” includes such items as cattle, money, gongs, liquor, purchased cloth, clothing, coats, blankets, and jewellery.

hpu ja or hpaji pau je [z] hpaga [ j]

sharung shagau shirung je [z] Dowry including “Dowry including Wife-Giver [ j] Wife-giver Wealth Wealth” includes such items as two wife-givers’ baskets containing ritual grain seeds, ritual knife and spear, and hand-woven skirt, tripods, woks (Zaiwa name tripod and wok prestation separately as House-Entering Wealth), guns, ceremonial knives.

into: shapa (ordinary perishable foodstuffs; alternately called wunji sahka), yam nga (“livestock”), and hpaga (for Leach, “goods and objects used in ritual exchange other than livestock”). So far, the correspondences between Burma Jinghpaw and Yunnan Jinghpaw, and including the material in Ola Hanson’s dictionary (1906), are in accordance with one another. However, there are two anomalies. First, there are two other terms

244

ho ts’ui-p’ing

not covered by Leach, but whose presence in Burma is conrmed by Ola Hanson’s dictionary. One of the two terms is a rai (wealth objects), and the other of the two terms is a ja (valuable wealth objects). Second, the term hpaga occurs alternately as wealth objects in the term hpaga rai (household wealth objects) and in the term hpu ja (bridewealth; alternately hpaji hpaga). All household wealth objects are purchased through trade and from bridewealth prestations. That is, in daily usage they are hpaga rai, but as bridewealth they are hpu ja (alternately hpaji hpaga). The latter are included with further enclaves of the prescribed, the negotiable and the optional mode of exchange (see Table Four). These two terms are complemented by additional terms where the term hpaga is not used, including, for wealth objects, the sub-category of kun rai (utensils), and for valuable wealth objects the sub-category of sharung shagau (dowry). Except for the house-entering prestations (the tripod and wok), dowry includes neither household wealth objects nor wealth objects. The cover term in Yunnan Zaiwa for wealth is isut, which subdivides into howa (ordinary perishable foodstuffs), gau ngvui (livestock), and pegva je (Leach’s “goods and objects used in ritual exchange other than livestock”). For the Yunnan Zaiwa, the details are more complex. First, there is a single cover term for wealth objects ( je). Second, the category for wealth objects subdivides into pegva (household wealth objects), igvun je (valuable wealth objects), and chung je (utensils), and the sub-category for valuable wealth objects (igvun je) subdivides further into yvum wang je (house entering wealth), pau je (bridewealth), and shijung je (dowry). Again, like Yunnan Jinghpaw, except for including household wealth objects ( pegva je) as gifts in bridewealth of various modes of exchange, and including house-entering ritual objects (the tripod and wok of house-entering rituals) as dowry, the Zaiwa cognate of the Jinghpaw hpaga never refers to either bridewealth ( pau je) or dowry (shirung je). In sum, for Yunnan Jinghpaw and Zaiwa, trade objects in the form of hpaga for the Jinghpaw or pegva je for the Zaiwa make up an important component in bridewealth prestations of every mode of exchange. But, except for house-entering rituals, trade objects are not a component in prescriptive dowry prestations. Their classication is as “valuable wealth objects”. The point is that for the Jinghpaw and the Zaiwa in Yunnan, the sphere of marriage exchange involves “valuable wealth objects.”22 Through the exchange of the prescribed bridewealth and 22

I hasten to add that agreement on the classication of wealth for either the Yunnan Jinghpaw or Zaiwa is not, however, total. This is especially so for the general

• Cattle • Purchased cloth, clothing • Gongs • Money • Liquor • Full-patterned handwoven skirts • Furniture • Guns • Ceremonial knives • Jewellery • Opium • Tea(?) • All other commodities

Household Wealth Objects (including personal wealth objects)

• Cattle • Gongs • Pig’s hindquarters • Purchased cloth (silk) and clothing, including felt cushions • Beer • Liquor • Money

Prescriptive • • • •

Cattle Gongs (size) Money Liquor

Cattle Guns Jewellery Cloth and clothing • Opium • All other commodities, such as threshers, televisions, tractors

• • • •

Bridewealth Negotiable Optional

Dowry Prescriptive

• Wife-givers’ • Jewellery • Cloth basket and clothing containing ritual grain • Guns seed and ritual • Ceremonial knife and Knives spears • Other • Wife-givers’ Commodities, basket such as sewing containing machines, bed-quilts hand-woven skirts • Weaving Tools • Tripod and Wok (Zaiwa houseentering wealth)

Optional

Valuable Wealth Objects

Table Four: Classication of Household Wealth Objects versus Valuable Wealth Objects

rethinking kachin wealth ownership 245

246

ho ts’ui-p’ing

dowry, the negotiable bridewealth, and the optional bridewealth and dowry, marriage exchange makes the androgynous person in its social reproduction and the social distinction of the person and their house (Ho 1997: Chapter Five, 2004). Whether hpaga or pegva in their marriage exchange is a category of ritual wealth objects or trade objects or both, is insufcient or even irrelevant to our understanding of wealth ownership. It is actually more important to explore rst in detail the different enclaves of marriage exchange. Yunnan Jinghpaw and Zaiwa marriage exchange divides further into three enclaves. First, under the prescribed exchange of bridewealth and dowry, both household wealth objects and non-trade objects are enclaved. This rst enclave constructs the “body” part of the androgynous house person throughout the couple’s life-long exchanges (Ho 1997: 436–461, 482–498; Ho 2004). Theoretically, the last head of the prescribed bridewealth cattle is offered after the death of the married-in woman. In practice, this last head of bridewealth cattle can be delayed for generations.23 The second enclave of exchange of negotiable bridewealth uses household wealth objects only. This second enclave consolidates his and her house, constructs the metaphysical person—the soul component—and makes the social reproduction of the house person possible (Ho 1997: Chapter Five). After negotiation, the wife-takers can delay only certain parts of this debt, but return the rest of the debt in an agreed upon time; or they can choose to repay the debt whenever they have means before the agreed upon time to gain name. In the Luxi case in the early 1950s mentioned earlier (see above), the sister’s husband paid his bridewealth debt at his wife’s brother’s wedding.24 The third enclave of exchange uses household category of “valuable wealth objects.” First, some Jinghpaw assert that “valuable wealth objects” (a ja [ j]) only includes bridewealth; others consider that “valuable wealth objects” additionally includes “dowry and wife-giver wealth.” Second, other Zaiwa assert “valuable wealth objects” (igvun ze) is a more inclusive category that additionally includes all “household wealth objects.” In other words, inconsistencies in the usage of “valuable wealth objects” show Jinghpaw usage as more restrictive, while Zaiwa are more willing to cross the line between “valuable wealth objects” and “household wealth objects.” 23 This last head of bridewealth cattle is called “the funerary bridewealth cattle” in Jinghpaw (mayang ja [ j]), “cattle for the rewood” in Zaiwa (myithe no [z]) (Ho 1997: 457, 460–462). The complexities, the varied ways of sending this last head of cattle and a discussion of its signicance as a temporal strategy in constructing sociality, please see Ho 2004: 283, footnote 17, 308–314). 24 One of the head of cattle might be the prescribed bridewealth cattle of “cattle for constructing the house for the dead” (kario baw nga[ j], shimao no[z]or zangmo mau [z]),

rethinking kachin wealth ownership

247

wealth objects such as cattle, utensils such as threshers, and personal wealth objects such as jewellery and guns that are personally owned (see Table Four).25 This third enclave of exchange makes for the distinction, identication and fame of the households of the wife-givers and wife-takers, the bride and the groom. Only in this exchange sphere can the wealth objects given be household and personally owned. Jewellery coming with the woman can be sold or given to whomever she likes depending on one’s wealth. My Jingpo “aunt” has two strings of silver necklaces. She has two daughters and one daughter-in-law. Years ago, she remade the two necklaces into four. Each daughter and daughterin-law was given one. Last year, she gave me the last of her necklaces. A gun or ceremonial knife given to a man is his own property too. An old uncle had a gun many of his descendants desired. After his death, his son looked for his father’s gun, but could not nd it. It turned out that, before the old uncle died, he had already given it to one of his friend’s sons who had taken care of him a lot. This is just like Muiho Mulat in the early 1950s Chinese sources, whose prot out of the two liang of opium was all her own; she could do whatever she liked with the money. In these three sub-spheres of marriage exchange of Yunnan Jinghpaw and Zaiwa, substitution also occurs as Leach tells us. However, the substitution does not necessarily follow the direction of trade objects hpaga, substituting the ritual objects hpaga like Leach suggests, nor does it appear that all ritual wealth hpaga can be substituted by the trade objects hpaga, as Leach implied in explaining Kachin social change. which is required from the wife-takers. It can be from one of the married out daughter’s family or can be jointly given by more than one married out daughter depending on the original deal at marriage and the economic situation of the wife-takers (Ho 1977: 268–274, 2004: 289–293). 25 Culturally, a gun makes a very special kind of wealth object by its power in obviating and ending the wife-givers and wife-takers’ relationship. It is an “optional” bridewealth object from the wife-takers to reverse the direction of the ow of women, changing the original wife-givers’ position into the wife-takers among the Zaiwa. The occasion a gun is used in this way is called “to reverse the ladder going into the house” (Ho 1999: 201–203). On the other hand, it can also be a “prescribed” dowry object from the wife-givers to identify the wife-takers’ debt being nished. It is used as return prestation when the “funerary bridewealth cattle” or “cattle for the rewood” is sent by the wife-takers after the funeral (Ho 2004: 312–313). Even though it is the kind of prestation that is, by denition, a return “female” gift to a certain initiation of “male” gift, it is optional because the last “funerary bridewealth cattle” or “cattle for the rewood” can be delayed for generations or forever delayed. The power of guns in closing or nullifying the long term marital relationship in Zaiwa sociality obviously has a lot to do with its being the most desired wealth object historically in the Jingpo Hills.

248

ho ts’ui-p’ing

First, in the prescribed exchange sphere, when picking up the bride and the two dowry baskets of wife-givers’ wealth, at each step of the trip, the wife-givers create as many obstacles as possible before giving their consent to give away their woman and the wife-givers’ wealth baskets. Each obstacle requires liquor, “money” or a money substitute—purchased cloth—to buy the ‘let-go’ consent from the wife-givers (see also Levi-Strauss 1969: 265–268). No matter if it was money, liquor or cloth, the market value is always very small. Throughout the life-span of the woman, till her death, at every critical juncture of the couple’s life cycle—when she needs to be “puried” at the wedding in order to take the ladder going up to her husband’s house, when she nally has to move into her husband’s house to live, and when she or her husband dies—money or its substitute cloth is used in “purchasing” her detachment from the wife-givers. At each stage, either the woman or the representation of the composite partial person (in Strathern 1988 usage of “partible person”), in the style of partial “human gurines”—the “bride ladder” with breast carvings on top (Ho 1997: 96–97, 2004: 266–268, 281–287, 300–303), the “beer and liquor basket” with two legs (hkyingting lan [ j], tandving byap [z]) (Ho 1997: 483–487, 2004: 268–271, 281–289, 300–305) and “the honoric dead person gurine” (lup grawng [ j], guprong [z]) (Ho 1997: 489–498, 2004: 271–275, 289–293, 305–306)—is “bought.” What is most crucial in these “purchases” is that it is always done in a ridiculing, comic style with lots of laughter and the hustling and bustling atmosphere, like in a market. They make a scene of “buying and selling” in order to separate, step by step, the physical relatedness of the woman and the wife-givers’ wealth in baskets from the wife-givers, and nally to become part of the wife-takers’ wealth. In other words, not all the trade objects assumed to equate with pure exchangeability are used as substitutes—the poor can use the less expensive objects, such as pigs, to substitute for buffaloes, the rich can use ten heads of cattle to manipulate their status—as Leach suggested (see above). These household wealth objects of cloth, money and liquor are culturally specic wife-takers’ things in the prescribed marriage exchange. They cannot be used as wife-givers’ things at this stage of personhood construction. Furthermore, in focusing on the object only, Leach overlooks the signicance of this “buying and selling” drama as a strategy of display. Through this display, a new composite of relations in making the androgynous person is constructed. Second, in the negotiable exchange sphere, what is exchanged ritually, verbally or ideally is, indeed, different from what is actually exchanged

rethinking kachin wealth ownership

249

in practice, as Leach’s argument of “exible” society emphasizes. But, Yunnan Jinghpaw and Zaiwa negotiate exchange in a kind of entangled style that is reexive between what is ritually and verbally announced and what is actually given. In a stylized fashion, they bargain ferociously. The wife-givers inate their superiority by asking for an astronomical number of head of cattle, huge-sized gongs or enormous amounts of money; the wife-takers bargain back and forth in public, as hard as they can. The bargaining and the negotiation used to take days, or several trips, before the bridewealth was nalized. Bargaining was also a required stage before any real negotiation could begin. These phenomena are actually very structural, as pointed out by both Lévi-Strauss (1969: Chapter 14) and Valeri (1994). Societies with generalized exchange of women either as an ideal or ideological construct “negotiate” between wife-givers and wife-takers “as if ” wives were “purchased.” Valeri thinks this means that particular cultural meanings attributed to “purchase” include an important role for ideals and ideologies of exchange in the processes of social reproduction.26 Negotiations are always over the bridewealth, and these negotiations honour or distinguish the wife-givers and their mutual rank order, but can also challenge the wife-givers. The result of the negotiations is a sliding index of asymmetric exchange. The more debt negotiated, the more certain the marital relationship; the more debt cancelled after the negotiation, the more unstable the direction of the marriages. My eldwork establishes the presence of two qualications to negotiation that are each, in their own way, of special interest. On the one hand, when one is capable of changing the modality of exchange from delayed to immediate, when an outrageous request from the wifegivers is raised and met by the wife-takers, the “bargaining” no longer exists and the relationship changes. The exchange is no longer part of the sphere of negotiable exchange; it is in the sphere of optional exchange. Among the Zaiwa, the most famous stories about how wife-

26

Lévi-Strauss did not distinguish these two steps of negotiation of staged ritualized bargaining and negotiation. Valeri raises the important theoretical issue in the signicance of understanding the indigenous meaning of purchase from the case of Huaulu in Eastern Indonesia. Different from Huaulu’s purchase expressed in the bargaining between wife-givers and wife-takers over the wife-givers’ things and the wife-takers’ things, Yunnan Jingpo and Zaiwa purchase is expressed as putting out a staged purchase of verbal bargaining as if they were making a deal in the market between strangers. I suggest this staged purchase should be understood as a strategy of display that mirrors reexively the construction of person through the exchanged objects (Ho 2004: 322–323).

250

ho ts’ui-p’ing

giver and wife-taker relationships became reversed relate exactly to this mechanism (Ho 2004: 312–315). The stories have the following pattern: name-group A and name-group B were wife-givers and wife-takers to each other for generations. One day, a boy from group A wanted to marry a girl from group B. At the marriage negotiations, B requested the astronomical gure of thirty three head of cattle for bridewealth. Named group A not only agreed, but also fullled. There was neither bargain, nor delay. Starting from then, the path between the wife-givers and the wife-takers, between A and B, got reversed, the old relationship was nullied, and the new relationship was created by fullling the bridewealth in display. On the other hand, in its promotion of “civilization” (wenming 文明), the Chinese government and Communist Party label the Jingpo custom of “negotiating” marital prestations in general, and bridewealth in particular, as backward. Urban Jingpo additionally shy from customary “negotiations” and attempt to replace the practice with heavily toneddown versions of “negotiations.” Nonetheless, customary “negotiations,” with their emphasis on creating obstacles to the completion of the “negotiations,” continue throughout the countryside. In other words, for the Jingpo in Yunnan, ritual bargaining, or the staged purchase, is as real as nal negotiation in practice, and both are indispensable in constructing the metaphysical part of the androgynous person. Again, what is crucial in this ritual bargaining is not the use of the wealth objects for manipulating status, it is rather the strategy of display for the staged “purchase” through outrageous verbal bargaining, as if they were buying and selling commodities between strangers in markets. Without this display to de-familiarize the exchange parties, no real negotiation can begin. Unlike Leach’s thesis of a exible system that assumes the duality between ritual/ideal social orders vs. trade/real life, the real life negotiation of wealth objects in the marriage exchange of Jingpo in Yunnan depends on the staged purchase performance as its context. Third, for the optional exchange, the households of the wife-givers and the wife-takers give whatever household and personal wealth objects they like, depending on the bridewealth money they have nalized and how well-off and how competitive the two families are. Unlike in the prescribed bridewealth exchange, there is no “buying or selling” drama. Unlike in the negotiable bridewealth exchange, there is also no staged “bargaining” in public, only the display of the items of exchange. The bridewealth objects of big furniture, machines or tractors, dowry such as sewing machines, cabinets, bedding, clothing etc., are displayed in

rethinking kachin wealth ownership

251

a procession on the way to pick up the bride and subsequently sending her off. Only in this optional exchange sphere do people exchange objects as objects themselves, rather than “relational objects” in the prescribed and negotiable exchange sphere. In sum, we can see that household wealth objects used in marriage exchange appear to be in both the prescribed and the negotiable spheres indispensable among the Jingpo in Yunnan in their construction of personhood and sociality. They are “relational objects” that make the composition of the person and the social grouping. They require purchasing drama and bargaining ritual to make the exchange appear ‘as if ’ they were commodities with “pure exchangeability” (Wagner 1977, Strathern 1988). They are not “possession” owned by any one person, any household or named house group, and cannot expect to be accumulated (Damon 1993). Their importance in marriage exchange is not their assumed money-like value in overturning the ideal moral order of kinship, as Leach assumed about Kachin hpaga from the perspective of his oscillation theory. Only wealth objects exchanged as optional have the “property-like” quality (Damon 1993) of money, dened by Robbins and Akin as being able to “. . . move against anything in any kind of exchange between people who stand in any kind of relationship to each other ” (1999: 12, original emphasis). People gain these wealth objects either from money income or from bridewealth, which can be gained by high ranking wife-givers, or most interestingly, from the savings they make as a wife-taking family not owing much debt to their wife-givers. This is how my Dehong Jingpo “uncle” in his early 70s described such occurrences: I asked my Jingpo uncle how his family began to have any money, he said: Grandfather was a great ritual chanter. We had meat to eat.27 The only reason I can gure that we had some wealth was due to the fact that great-grandfather did not owe much bridewealth. Great-grandfather’s rst wife died before she had any children, hence he did not have to give any bridewealth cattle. Then he married a refugee Dai woman with a daughter already that cost no bridewealth. They gave birth to six daughters and only one son. The six daughters bring in bridewealth, but only one son needs to give bridewealth.

27 The payment to ritual specialists—the chanter, sacricer, and the receptionist—is in two forms. One form is money, which is usually in a small sum, the other form is the sacricial meat used in rituals.

252

ho ts’ui-p’ing

Their family did not start growing opium till the late 1940s. If there was any wealth at home, it was because they did not owe debt from bridewealth. From the angle of consumption, the wealth objects exchanged as optional bridewealth for Yunnan Jingpo are still of two kinds. One kind is the household wealth objects and utensils; the other is the personal wealth objects.28 The wealthy household shown in the early 1950s Chinese Jingpo survey data (see Table Two) owns not only many more livestock than the average household, but the gap between the two households in their ownership of wealth objects of utensils, furniture and money, and personal wealth objects of jewellery, cloth and clothing, is especially obvious. Many household wealth objects are to share among households, such as utensils, and are consumed in different feasts or nat spirit worship occasions. Families with tractors and threshers charge villagers going to markets, but at a very reasonable price. When I began doing eldwork among the Jingpo in late 1989, houses with tiled roofs were not at all common. Those houses with tiled roofs were considered well off. Among the Zaiwa, a family moving into a new tiled-roof house without feasting or carrying out certain rites to appease the spirits of gossip or curses will suffer from illness. The cure for it is usually to make proper sacrices to the particular nat spirits and, hence, an occasion for feasting. From the perspective of consumption, the nature of the ownership of the household wealth objects and utensils is still quite relational. On the other hand, personal wealth objects like adornments, cloth and clothing, such as Muiho Mulat bought and the ladies in the photographs taken by Green wore, are much more personally owned and consumed as well. Those who have personal wealth objects do lend to those who do not, without any hesitation, whenever occasions require. The fame built on the accumulation of personal wealth objects does attract gossip and attacks by jealous spirits. However, because of the fact that these kinds of personal wealth objects are traditionally often earned and owned by the owner, there is no obligation to share. Both the prevalence and the accumulation of this kind of wealth suggests a different and probably competitive kind of individuality and sociality, as Robbins and Akin suggest that globalized money brings.

28 It is interesting to see that all bought utensils, except woks and tripods, are used as optional marriage exchange items.

rethinking kachin wealth ownership

253

From a close look into the marriage exchange sphere of the Jinghpaw and Zaiwa in Yunnan, we can see the presence and importance of monetized markets, and the cultural means for controlling the ow of money. They enclave money into two kinds, one as relational objects in constructing personhood and rank order between houses, the other as possessive property in building the fame of individuals and houses in display and competition, or in reversing the rank order, bringing social change, such as gumlao revolt.29 While the desire for household wealth objects is built on the social agency of display, the display of personal wealth objects for personal consumption or fashion was probably new, having developed since the late nineteenth century, and might have brought anxiety to the collective in social reproduction. With the two liang of homegrown opium, Muiho Mulat was able to create a “new” role for herself under the politico-economic environment of the region then. I suggest here that the basis for building up this new role was the rise of a new concept of property that was not relational, but possessive. Three questions we still need to address include: what was the politico-economic context making the prevalence of possessive ownership possible? Was this politico-economic context the same before, during and after the British arrival in the Kachin Hills of Burma or close to the Jingpo Hills of Yunnan? How related is the opium economy to the emergence of the possessive ownership of wealth? Conclusion The power process historically involved in the value conversion of wealth objects before colonial contact was different from that which took place during and after colonial contact. For years, the Jingpo have enclaved money to the social agency of display in making distinctions between persons and households before money became more available

29 It is important to note that not all Jingpo in Yunnan with possessive ownership of wealth had experienced gumlao revolution. The gumlao revolution only swept through most of the Jinghpaw speakers villages in the bordering county of Yingjiang ( YNSBJZ 1986a, Gong 1988: 56, 147–9, 152–4). There is no recall of any kind of revolution occurring among the Zaiwa or Langvo speaker communities. I offered an explanation for this variation from the perspective of different personhood and sociality between the Jinghpaw and the Zaiwa in my dissertation (Ho 1997) and different strategies of time in grouping elsewhere (1999).

254

ho ts’ui-p’ing

to the masses. When opium became an important cash income, a different kind of “personal wealth object”, which built its value on personal consumption for individual distinction, began to emerge. Despite the strict prohibition against opium in China as a whole and southwest China in particular both before and after the opium war (1839–1842), research on opium indicates that the peripheral area of Yunnan ruled by local chiefs continued to grow, trade, and smuggle opium (Bello 2003, Kuang and Yang 1986, Lin 1980, Zhang 1990, Qing 2005). Research done in the late 1950s on the Jingpo area, shows that, on average, the income from opium growing amounted to 20% of the total household income (Qing 2005: 93–94). Partly due to the limited state power at the Chinese periphery, partly due to the Yunnan government’s attitude in seeing opium as a good revenue source for Yunnan nance, not only could the prohibition not be carried out in practice, attempts to do so were also often lax for reasons obviously related to these complicated local interests. The situation on the Burma side of the border under British colonial rule was little different. The British government had always had different policies towards opium for the lowland Buddhist Burman population and the highlanders before the annexation of Upper Burma in 1886. They strictly monopolized opium production in British India, and prohibited opium consumption in lowland Burma, but were lax on opium growing and consumption by the uplanders. Despite the issue of opium orders in the Shan State in 1923 and in the Kachin Hills in 1937, the area across the Salween was always conveniently outside jurisdiction, and continued to produce, smuggle and trade opium (Renard 1996). After independence in 1948, the Mienma [Myanmar/Burmese] government control over the northern Wa and Shan states was limited. Yunnan caravan traders and the Chinese Kuomintang armies had, since the 1950s, continuously made the opium grown and transportated in this area available to the whole world (Hill 1998, Litner 1997, Renard 1996). From the last half of the nineteenth century until the middle of the twentieth century, opium-related prosperity made for an unprecedented politico-economic context in areas where the Kachin and Jingpo resided. The Kachin and Jingpo were provided with a new source of cash income arising from this historical contingency. This new source of cash, arising precisely because the Kachin and Jingpo resided at the periphery of the periphery of the world system when opium had global value, made a difference to their everyday lives.

rethinking kachin wealth ownership

255

Earlier I raised the following question: could the motivation for gumlao revolution in the Kachin area have been elicited by a new vision of sociality and of individuals with a new concept of ownership? I suggest that the answer can only be revealed after more detailed research has been conducted into the dynamics of the various political regimes, the introduction of the new currency—opium—at this time, and the social agency of display behind the indigenous meaning of wealth ownership. This chapter begins to give some answers to the last topic. I argue in this chapter that the indigenous meaning of wealth ownership among the Kachin in making their social agency of display is key to understanding Kachin social change. From a re-analysis of Jingpo wealth objects, I argue that the competition between the desire for possessive wealth and relational wealth is an underlying reason for the “owering of gumlao” social order in the Kachin Hills, at least in the early half of the twentieth century (Friedman 1979, Maran 1967). In different regions of the Kachin/Jingpo area, what wealth means to the people, and how that meaning is brought into being to motivate social change, are important questions that need to be asked. This is true, whether or not it was because elites willingly gave up the chief ’s token privilege in “thigh-eating” (Maran 1967), or because regional inationary pressures for hosting chiey manau rites were too high to support the hierarchical social order (Lehman 1989). Leach denes Kachin ownership as being to “own the debt” and it is “to be noted that with few exceptions debts are deemed to exist between lineages rather than between individuals” (Leach 1977: 153). In other words, wealth can only be owned through display for the consumption of the related, and cannot be owned for personal consumption. The tension between household wealth and personal wealth intensied as more people were able to own and desired to accumulate and consume personal wealth items for display. This new concept of wealth ownership, based on individual possession, was in conict with the old agency of display for public consumption, especially when objects were purchased at the expense of cattle, the most important wealth objects. Mulat’s personal behaviour of consumption is not an exception in this area at this time, when a new vision of the individual began to emerge. A different reading of gumlao revolt as being caused by the tension between relational wealth ownership and possessive personal wealth ownership is, therefore, suggested.

THE MISSING SHARE: THE RITUAL LANGUAGE OF SHARING AS A ‘TOTAL SOCIAL FACT’ IN THE EASTERN HIMALAYAS (NORTHWEST YUNNAN, CHINA) Stéphane Gros Northwest Yunnan is where the eastern ank of the Himalayas, in a spectacular bend, turns abruptly south. This is the heart of what is called the Hengduan Mountains, where the main rivers of Asia ow southward in parallel gorges separated by high altitude mountain ranges. Here, the administrative boundaries of the Tibetan Autonomous Region, the provinces of Yunnan and Sichuan, and the country of Burma (Myanmar) meet. It is the home of Tibetans, but also of several Tibeto-Burman speaking peoples, such as the Naxi,1 the Lisu,2 the Nung and the Drung,3 amongst a few others. From the mid-nineteenth century until the mid-twentieth century, only a few explorers, geographers, botanists and scholars had crossed these mountains.4 Despite the increasing amount of research done during the last decade, we still know very little of northwest Yunnan’s history and ethnology, especially of its more remote parts, such as the upper reaches of the Salween river (Nujiang) and the easternmost source of the Irrawaddy, where the Drung people live.5 1 For recent studies on Naxi (and Moso) see for example Oppitz and Hsu (eds), 1998; Mathieu, 2003. 2 For the Lisu people of the upper Salween river valley, see the long introduction in Dessaint and Ngwâma, 1994. 3 Nu and Dulong are respectively the standardized form (and ofcial name in Chinese) of the names of these two groups. The Drung are part of a linguistic group together with the Nung and the Rawang of Burma. About the relationships between these groups, see Gros, 2004. 4 The French missionaries of the Foreign Missions Society of Paris were among the rst westerners to discover this part of the Sino-Tibetan border, and because of their long term relationships with the local populations, they were also able to provide detailed information about many aspects of this area, including its geographical, cultural and political particularities. See Gros, 1996, 2001. 5 The anthropological research I undertook about the Drung (Dulong) people, and the several eldwork assignments I conducted between 1998 and 2003 for a total of eighteen months, was largely funded by a Franco-Chinese Bilateral Grant (1997–1998), a Lavoisier Grant of the French Foreign Affairs (1999), the nancial support of the France Foundation (1999), and that of the Louis Dumont Fund for Social Anthropology (2003).

258

stéphane gros

Historically, the northwestern borders of Yunnan province lay at the juncture of both Chinese and Tibetan expansion. This remote area was coveted by Tibetans, Naxi and Chinese, and has been a theatre of unceasing conicts. For a long time, several political legitimacies coexisted, with the empowered parties exercising quite freely their rights on this territory. To this conguration was then added the progressive inuence of the Qing dynasty, and later on that of the Republicans with their respective colonial policies. Nevertheless, until the rst half of the twentieth century, this patch of land between Tibet, China and Burma was like a free zone, where only local chieftains retained any real authority. As with some other Tibeto-Burman speaking people still presently living in the area, the Drung, long known under the Chinese name of Qiuzi (Kiutzu), were caught between their more powerful neighbours, despite the isolation of their mountainous valley. While very little was known about these people, their valley soon became a pawn in imperial rivalries. As the British botanist, F. Kingdon Ward once wrote: It is certainly curious that no less than three empires should have laid claim to this wild valley. China has claimed it on the rather shabby plea that she was the rst to oppress the Kiutzu [Qiuzi]—taxation without compensation is a sheer oppression. Tibet claimed it on the ground that she had long been accustomed to extract slaves from that region. Britain’s claim rests, in part, on the perfectly absurd ground that the inhabitants want her to take it over. (1924: 190)

Since the end of the nineteenth century up the the 1930s, this area was considered an “un-delimited frontier.” China and Britain were not the only two rivals. Local rulers, Tibetans but also Naxi, added to the complex layering of spheres of power over this area. Here, interethnic relationships were framed by power relations between these neighbours, a conguration in which the Drung people could only occupy the lower end of the political hierarchy. Though not reported with much detail in historical documents, some Drung women and men were indeed at times taken away, or exchanged for oxen as we will see, to become slaves in an alien land. I will here argue that such facts can better be understood in this context by taking into consideration the whole system of goods exchange and political relationships, which reveals not only a social hierarchy, but also cultural values that are important for our understanding of the socio-political specicities of this area. In northwest Yunnan and its adjacent regions (upper Burma, eastern Tibet), I believe there is something like a “ritual language” that is

the missing share

259

common to these otherwise dissimilar groups. It seems that the way locals, whether Tibetan, Naxi or less politically centralised groups such as the Drung, understood the workings of political action was intimately linked with that of ritual action. This echoes Leach’s view of highland Burma and refers to his oft-cited formulation that “people may speak different languages, wear different kinds of clothes, live in different kinds of houses, but they understand each other’s ritual. Ritual acts are ways of ‘saying things’ about social status and the ‘language’ in which these things are said is common to the whole Kachin area” (1997 [1954]: 279). From this point of view, I will reect upon political relations between local populations and try to shed some light on their underlying principles. In this chapter, I will analyse the workings of political relationships and exchange networks in a regional system involving the circulation of both goods and people, but also of values regarding wealth and status differences. Using an ethno-historical approach, I will show how exchange, partaking and debt are, for the local societies of this very specic area where China, Tibet and Burma meet, key notions that articulate a more general politico-religious system. I will do so by combining ethnographic data collected mainly among the Drung, with historiographical sources in western and Chinese languages. My aim is twofold. On the one hand, I would like to try to characterize the notion of ritual language used by Leach in his book Political Systems and discuss its heuristic value. By so doing, I hope to contribute to a clarication of this notion so that we can better appreciate its place in Leach’s work. It is my conviction that it lies at the very heart of his anthropology. On the other hand, presenting my own analysis based on an ethno-historical approach to the relationships between the Drung and their neighbours on the Sino-Burmese border, I would like to make use of this notion of ritual language in an attempt to elaborate it further. The Notion of “Ritual Language” Ritual was an important focus of Leach’s later anthropological work, but even before this, a good deal of his approach in Political Systems depended upon his understanding of ritual. In fact, Leach did not attempt to dene ritual but rather to reconcile divergent views in an approach to ritual broadly conceived as a system of symbolic communication. I will here recapitulate the essential aspects of Leach’s “unorthodox”

260

stéphane gros

treatment of ritual—as he labelled it himself in the introductory note of the 1964 reprint of the book—and underline what makes the basic characteristics of his notion of ritual language.6 For the members of a group, the meaning of their actions is expressed in what Leach calls “ritual,” giving to this notion an extended meaning. Ritual, Leach stated, “serves to express the individual’s status as a social person in the structural system in which he nds himself for the time being” (1997 [1954]: 10 –11). It could be said that ritual is the homemade model, which refers directly to the more abstract order of reality. Leach views ritual in its cultural context as “a pattern of symbols” that “makes explicit the social structure” (ibid.: 15). The structure that is symbolised in ritual is “the system of socially approved ‘proper’ relations between individuals and groups. These relations are not formally recognised at all times [. . . and] this neglect of formal structure is essential if ordinary informal social activities are to be pursued at all” (ibid.: 15–16). “My thesis,” wrote Leach, “is that in ritual action and in myth the actor is ‘making statements’ concerning the same abstract order of reality as that with which the anthropologist is concerned when he uses technical jargon to describe some feature of social structure” (ibid.: 86). As the French anthropologist Jean Pouillon noted in his afterword to the French edition of Political Systems [1972], according to Leach, structure is a matter of concepts, not facts. And between the facts and the model elaborated by the ethnologist lies what we can call the “indigenous model.” This model reects the underlying order that is supposed to guide people’s social activities: “Ritual performances have this function [of reminding about this underlying order] for the participating group as a whole; they momentarily make explicit what is otherwise a ction” (Leach 1997 [1954]: 16). In other words, “ritual” as an aspect of culturally dened behaviour is attributed by Leach the qualities of a language, and such “verbal categories” as ritual actions and myth refer to an as if system of ideas, a “structure.”

6 It should be clear here that my aim is not to elaborate on the concept of ritual in anthropological thinking in general, but to frame the present discussion within Leach’s own denition. It is equally important to understand that Leach’s use of the notion of “ritual language” covers a broader range of social actions than what is generally implied by the ritual use of language, or religious language. This later and more exclusive acceptation of the notion (see for example Sadan, this volume) should not be confused with the meaning it takes in the present case.

the missing share

261

The example of the Hpalang community that Leach studied in detail can help in presenting Leach’s approach in terms that are more concrete. This community comprised several ethnicities, such as Jinghpaw, Atsi, Maru, Lisu and Chinese. Each of these groups—approached by Leach as “sectors” within the social structure—had its own language, religious practices and customary behaviour. But in Leach’s discussion of the principles of Kachin social structure common to all of them, such differences can only be understood as internal variations. Since the members of the Hpalang community were all part of one political system, they had to be able to communicate with one another about political status, and they did so in the language of ritual action. Precisely because Kachin society as a whole is made up of numerous sub-groups speaking diverse spoken languages we may expect that at a ritual level there is a rather simple stereotyped ritual ‘language’ which is understood by all sub-groups and in which issues of status are constantly represented in much the same way. Kachin ritual expression is relatively simple precisely because Kachin culture is complex. (ibid.: 102)

Being part of the same political system, people communicate about their social identity (status) through ritual means. There are, therefore, two associated aspects, which deserve further discussion: that of communication and that of identity. Ritual and Communication In Leach’s view, ritual was not limited to its religious (or transcendental) aspects. He clearly expressed his rebuttal of the classic Durkheimian dichotomy between “sacred” and “profane”: . . . [ T ]echnique and ritual, profane and sacred, do not denote types of action but aspects of almost any kind of action. Technique has economic material consequences which are measurable and predictable; ritual on the other hand is a symbolic statement which ‘says’ something about the individuals involved in the action. (ibid.: 13)

Ritual is, therefore, a category that encompasses different types of activity, and schematically all types of action can have a ritual aspect. It can be true of the way people dress, speak (and the language in which they choose to speak), etc. But, myth and ritual (in the restricted sense of ritual performances) are the two main types of action that are of primary importance:

262

stéphane gros “Myth, in my terminology, is the counterpart of ritual; myth implies ritual, ritual implies myth, they are one and the same. [. . .] As I see it, myth regarded as a statement in words ‘says’ the same thing as ritual regarded as a statement in action.” (ibid.: 13) “. . . [ R]itual action and belief are alike to be understood as forms of symbolic statement about the social order.” (ibid.: 14) “. . . [ M ]yth and ritual are one and the same. Both are modes of making statements about structural relationships” (ibid.: 264).

Leach’s approach is limited to myth as something that reects patterns of social relations. Ultimately, the only thing “rituals” have in common is that they are actions that communicate meanings, or, in some cases could create the very meanings they communicate. Therefore, it appears in all these formulations, that the roots for Leach’s later exploration of the concept of ritual (1968, 1971 [1966], 1976) were already there as the theoretical background of his Political Systems. Ritual, as he understood it, was a “symbolic communicative performance”; myths were a matter of verbal communication, whereas ritual performances were a matter of symbolic communication.7 He proposes that we look for similarities in patterns of communication rather than try to arrive at universally valid denitions of ritual. Ritual, Identity and Ambiguity Leach’s main goal was to offer an analysis of the Kachin system, and his formulation of the use of ritual to express status is embedded in the context of empirical political behaviour. However, ritual also appears as a means to deal with specic systems of relationships and to express an individual’s identity in the system. This is why, for example, everyday actions fall into the “ritual” sphere that Leach had in mind, and language change expresses the social positioning of actors: The two sides of this paradox [in some cases Kachins seem conservative about language but others seem almost willing to change their language as a man might change a suit of clothes] both exemplify the same social fact, namely that, in my terminology, for a man to speak one language rather then another is a ritual act, it is a statement about one’s personal status; to speak the same language as one’s neighbours expresses solidarity

7 The general discussion on the implications and limitations of such an approach goes far beyond the scope of this paper. Leach’s understanding of ritual action has been discussed in a stimulating way by Tambiah (2002: 350 –356), pointing out his two complementary perspectives, i.e. special behaviour in ritual occasions and behaviour in everyday life (presentation of self ).

the missing share

263

with those neighbours, to speak a different language from one’s neighbours expresses social distance or even hostility. (ibid.: 49)

In Leach’s work, ritual appears as a means of expression in a more general politics of identity. This politics of identity encompasses specic claims of status position in the socio-political hierarchy; ritual elements can be integrated into both religious and secular settings that affect the identity of participants. When Leach demonstrates the process by which Kachins have become Shans and Shans have become Kachins, he proceeds to show that this has been possible because Shans and Kachins share a common language of ritual expression, and can thus be thought of as members of “one society.” One could say that there is a ritual language of ethnic relations: in multi-ethnic societies, some “lingua franca” is needed to designate status positions between groups and individuals. This is a prerequisite for the contextual denition of ethnic identities, and Leach’s discussion of the uidity of identities is one of his important and well-known contributions to theories of ethnicity.8 Another important aspect underpinning part of Leach’s interpretation of the dynamic of the system is the ambiguity of the ritual language. Myth and ritual is a language of signs in terms of which claims to rights and status are expressed, but it is a language of argument, not a chorus of harmony. If ritual is sometimes a mechanism of integration, one could as well argue that it is often a mechanism of disintegration. (ibid.: 278) There are, as we have seen, a large number of ‘ritual acts’ which can be said to have the same meaning whether the actor is a Shan, a gumsa Kachin or a gumlao Kachin, but the inferences that are to be drawn from such acts will be entirely different in each case. That such ambiguity does not lead to intolerable misunderstandings is due to the essential vagueness of all ritual statements. Ritual and mythology ‘represent’ an ideal version of social structure. It is a model of how people suppose their society to be organised, but it is not necessarily the goal towards which they strive. (ibid.: 286)

As a “language of argument,” the “ritual language” allows manipulation by protagonists. Its potential ambiguity is essential to the social dynamic that Leach aimed at describing. In his approach to myth and

8 See Corlin (1994) who suggests that the “ritual language” of ethnic relations is patterned after a general conceptual model regarding the universe and man’s place in the world. The all-pervading and integrative marriage system also contributes to the ‘blurriness’ of ethnic categories; see Robinne, this volume.

264

stéphane gros

ritual, Leach rejected the functionalist view of a coherent whole composed of one mutually consistent set of myths that served as a charter for ritual action. Myth and ritual are both ways in which Kachin and Shan represent to themselves their social structure in its changing nature and with its contradictions. For Leach, there is no signicant difference between so-called magical rites and expressive political or technological action. Ritual is no more than a symbolic expression of status and is analysed in terms of the logic of linguistics. As he later on wrote in Culture and Communication (1976), we engage in rituals in order to transmit collective messages to ourselves, and these messages, Leach asserted, are always about the social order. For Leach, as he further developed in this book, ritual is a medium for the expression of cultural ideals and models that, in turn, serves to orient other forms of social behavior. But, on what grounds can one argue that there is one system of ritual expression? And, if this language is common and “ ‘represents’ an ideal version of social structure,” as Leach has it, what is this common structure? Or, to put it another way, what is the model, the underlying common order? Ritual theory has evolved since the publication of Leach’s Political Systems, and there has been criticism of the tendency to analyse ritual as a language as well as criticism of the attempt to analyse ritual as being communicative. After all, Leach himself pointed out that ritual is a category scholars have invented, and in large measure, theories of ritual depend upon the phenomena to which one chooses to assign the ritual label. But, I shall now address these questions through an analysis of the social and political situation of the borderlands of northwest Yunnan, with special reference to the rst half of the twentieth century. My intent here is to follow on from Leach’s intuition about the existence of a common ritual language and approach the interactions between the Drung people and their neighbours in much the same fashion. I propose to study the way in which the Drung were integrated into the multi-ethnic environment of the eastern Himalayas, and how, through their interactions with their neighbours, they produced as well as were ascribed their differences. In this context, debt seems to be an essential matrix for the construction and the expression of political relationships between groups, and it contributed to establishing or abolishing the frontiers between them. In this sense, the logic of debt and its ritual, political, and economic expressions constitutes a common language that enables all the groups to produce their differences, and which, in the process, also includes them in a common social system.

the missing share

265

Debt and Hierarchy In Leach’s manner, my perspective here is to consider the Drung in terms of their relations with their neighbours, not as an isolated social entity. Considering their geographical location and their socio-cultural characteristics, they could potentially be considered part of the Kachin world, in the sense that Leach used that category. The Drung (Dulong) people are a Tibeto-Burman speaking group ofcially recognized as one of the fty-ve “minority nationalities” (shaoshu minzu) of the People’s Republic of China. They are relatively isolated in the small valley through which the easternmost source of the Irrawaddy meanders, today an administrative division of the Dulong and Nu Nationalities Autonomous County of Gongshan (map 1).9 As they were ofcially recognized as the separate “Dulong minzu,” they were by this process distinguished from the Nung (Nu minzu) to the east, and from the Rawang that were on the other side of the national border. But, even up to the present, Drung have familial, cultural and linguistic links with these close neighbours that transcend natural and political barriers (see Gros, 2004). However, their inclusion into the Kachin social system is not a given. For most of their history, their relationships with political powers have been oriented to the East, with both Tibetan chieftains and the Chinese empire, rst indirectly through the ofce of tusi (indigenous chief ) granted to the Naxi rulers, and later on by means of direct administration. For the Drung people, power and authority have always been linked to the east since mythical times, and they nowadays often refer to this conguration as the time of the “Chiefs of the East.” It is these political relations with Tibetans, Naxi and Chinese that I am taking into consideration in the following analysis of the Drung people’s place in a regional system. The political situation of this area bordering Tibet and Burma was, by the mid-nineteenth century, under the authority of a Naxi indigenous chief (tusi ), who was at this time himself under the higher authority of a Chinese ofcial. To the north,

9 The total Drung (Dulong) population in China is estimated as 7,426 (National Bureau of Statistics of China 2002: 97), making up one of the smallest ofcial minority nationalities in China. More than 5000 Drung live in the remote Dulong valley bordering Burma.

266

stéphane gros

Beijing

Lancangjiang

Tibet

Kunming

Ridong

Chayu [Dzayul]

NORD

Yanjing

Sichua n

Deqin

Menkong N u j i a n g

N Benzilan

Kongdang

Gongshan

Yezhi

Jinshajiang

Dulongjiang

Bingzhongluo

Zhongdian

Kangpu

Putao [Hkampti] Weixi M a l i

Fugong

N m a i

H k a

Lijiang

Lanping

Bijiang M é k o n g

H k a

BURMA Lushui E r h a i

Liuku Dali S a l o u e n

Myitkyina

Legend

Yongping

Dulong

Xiaguan

20 km

Nu

Rawang © Stéphane Gros

Map 1: Geographical distribution of the Drung (Dulong), Nung (Nu) and Rawang in China and Burma

the missing share

267

Tibetans of Tsarong also extended their political inuence, and at some point the Naxi chief gave away to the Tibetans, nominally at least, his right to receive taxes from the Drung and the Nung. Therefore, for the population of the upper Salween valley and of the Drung valley, this resulted in overlapping spheres of authority. Nung as well as Drung were at the meeting point of the colonial expansion of their more powerful neighbours, whom they would sometimes resist by making ephemeral alliances with each other. More often, debt or even conscation of land would cause the families to move away. In this area, the Drung and the Nung occupied the lower end of the political hierarchy. An historically informed study of this area enables us to discover some aspects of the process of political integration. Yunnan’s borderlands had been the object of rivalry between Naxi, Tibetans and Chinese, so that the interests of several political authorities were simultaneously focused there. At the same time, the principal actors proted handsomely from the development of large-scale commercial activities (cf. Gros 1996). Debt appears to have been an institutional means for gaining political ascendancy and control over land, and it allowed the layering of various spheres of power. By the very means of debt relations there was a possibility of land appropriation, and the debt system, therefore, played a role in land distribution. In numerous villages in the upper reaches of the Salween, the Nung people had several creditors to whom they had to cede their land while remaining on the land as tenant farmers. As guarantee for rising debts, rights over land were ultimately transferred. Land could pass through different hands, and debt was one of the expressions of this mobility, implying changes in social relations. The creditor, having obtained the economic dependency of part of the population, was potentially a chief. Loans with interest were granted to locals by Naxi tusi, Buddhist temples, and some Tibetan chiefs as well. Commercial activities, like the selling of salt, several kinds of metal tools, cloth and wool, were also in their hands. The enslaved and the indebted represented two different degrees in a relationship of dependence. The relationship between the debtor and his creditor gave rise to a stable relationship of political dependency. This was the basis for a stratication of power and social organisation, thus revealing the strong link between political and economic relationships. In other words, the initial relationship based on exchange was transformed into an asymmetrical relationship, characteristic of dependency. As such, the creditor-debtor relationship becomes a central paradigm

268

stéphane gros

for understanding the grounding of relationships of dependence.10 Tibetan and indigenous chiefs (tusi ) exploited this very same process. Indeed, in this remote area up to the rst half of the twentieth century, impoverished peasants could not escape from the debt system. This was especially the case for the less powerful Drung and Nung who became potential slaves. However, it is important to specify to what extent, and in which cases, it might be possible to speak of slavery, for it seems that diverse modes of servitude coexisted. There were cases of external slavery: abduction, raids, all of which had as a target others, people external to the community of reference. Yet, there were also forms of internal slavery, in particular for debt. Debt slavery was described by missionaries and travellers in this part of the Sino-Tibetan fringes as an issue that concerned a great part of the population. It was present in eastern Tibet (Tsarong, Dzayul) as well as in the Tibetanized part of northwest Yunnan.11 Debt was closely associated with service labor, and in its extreme form, slavery. One of the missionaries who resided for many years at the border of southeastern Tibet specied that, amongst approximately six thousand families one could nd in Tsarong at that time: “270 only are tributary [i.e. “taxes payers”, khral-pa]; the others are farmers for the lamaseries [monasteries] or slaves of some rich owner” (Goré, 1923: 377). The farmers who did not have any means of paying their debts could become the slaves of their creditor, or settle their debt by giving one of their children. Those enslaved for debt could, in theory, free themselves; but rare were those who had the means to pay for their freedom. The creditor could transfer the debt to a purchaser, and the debtor thus changed creditor and his status was transmitted to his descendants. In these Tibetan areas, many slaves were Drung, or originated from the regions at the edges of the Himalayas (Upper Burma, Arunachal, Assam), but the debt phenomenon concerned the population as a whole, Tibetan or not.12

10

See Galey (1980: 145) on the creditor-debtor paradigm. For further details, see Gros 1996, 2001, 2005: chap. 3 and 4. 11 See Goré, 1923: 388–390; Ward, 1934: 52. 12 See Gros, 1996, 2001. One of the most precise descriptions of the situation in Tsarong is that provided by Rockhill (1881: 285–286), in which the ambiguity there can be between indebtedness and slavery, and the possibility of a transformation from one to the other is rather clear: “. . . While speaking of the Tsarong it is proper to note that the slavery exists there in a more aggravated form than in any other portion of Tibet. While now and then a poor Tibetan pilgrim, on his way through the Tsarong

the missing share

269

As mentioned previously, it also happened that Drung people were captured when victims of raids. However, some Drung also played an active part in a relatively important trade, geographically if not numerically, in which people were exchanged against goods, more often oxen. According to some oral accounts, the inhabitants of the south of the Drung valley took part in this trade more actively.13 These accounts accord with information provided by Father Fage who mentioned that: “Dijoux14 [. . .] are robbers, plunderers, and the trade of the slaves is almost entirely in their hands.” He continued by specifying that the Drung (and those who were at that time included in the same Qiuzi category) “like the ox esh passionately, and would not renounce any sacrice in order to get some. The Thibetans [sic] exploit this disproportionate craving and send to them each year hundreds of oxen; they receive in exchange the young slaves who receive our solicitude.”15 There is probably a degree of exaggeration in the remarks of the missionary because, at a rate of hundreds of oxen per annum, not many Drung would have remained. Nevertheless, it should be recalled that not only Drung people were exchanged. The slave trade also targeted adults or children who originated from other areas to the west. The French missionaries partly constituted their small community by purchasing slaves. These communities, therefore, included children (both boys and girls), generally non-Tibetans, who were orphans, victims of abduction or who had been sold by their indebted parents.16

to the Dokéla [ Dokerla], or to some other famous sanctuary, may become indebted to some one for the amount of his board or the like, and be obliged to work out by four or ve years of labor his little indebtedness, he at least eventually gains freedom; but such is not the case with most of the slaves in the Tsarong, who are taken from among the Lissus and other non-Tibetan tribes inhabiting the country. When one of these is in debt to a Tibetan and unable to meet the demands of his creditor, he becomes his “life servant” (ts’é yo [tshe-gyog]). The master has the right to sell, kill, or otherwise dispose of him; he is given a wife, or a share in a woman, and all the children born to him are slaves. Even if he should be able to get together enough to pay off his debt, the master may refuse it and count his labor as only a set-off for the interest of the sum due. The missionaries have bought and freed a number of these ts’é yo, but usually they have been able to buy only slave children; this class of person formed the nucleous of several of their little Christian communities.” 13 Due to space limitations I cannot here give all the details regarding this situation as it gures in Gros, 2005: chap. 4. 14 Father Fage is here using his own transcription of the name of a valley to the west of Dulongjiang in modern Burma. 15 Letter to M. Legrégeois, Bonga ( Thibet), July 23, 1857, Archives de la Société des Missions Étrangères de Paris, vol. 556 (2) [839]. 16 See Launay, 1902, I: 245. Archives de la Société des Missions Étrangères de Paris

270

stéphane gros

According to missionaries as well as the accounts of travelers at the turn of the twentieth century, the locality of Menkong in Tsarong was a center of the slave trade. In a short note, J. H. Edgar wrote that “Menkong before 1911 was an important slave market where men and women were bought and sold openly,” and according to the French Tibetologist J. Bacot, fathers, mothers and children could be sold separately (1912: 273). Most of the slaves were non-Tibetans and were most probably from the Drung and related groups.17 The living conditions of the slave and the rights that the master had over him could vary somewhat, but it is clear that slaves could be sold in Tsarong and its surroundings. The market extended beyond Tsarong, since we know that some Lisu took part in it, and that the Naxi indigenous chiefs (tusi ) of the Mekong valley owned a number of Qiuzi slaves who were freed on the order of the Chinese authorities during the rst half of the eighteenth century. All together, the origin and status of the “slaves” varied from one place to another, and included people living at the margins of society or who were in some ways excluded from it, such as orphans, robbers, prisoners of war, people who had been captured or were in debt. Many testimonies indicate that their living and working conditions were not particularly difcult. However, the relative discomfort of work and living conditions is not a sufcient criterion to identify or distinguish the status of the enslaved. To speak about slavery and slaves in the fullest sense of the term, it is necessary to admit the fundamental fact of exclusion. The status of the slave is marked by his total exclusion from a fundamental dimension of the society: that through which identity and status as a member of the community is articulated. As Testart (2001: 24) wrote: “the slave is a man without identity.” Within Drung society, for example, this exclusion refers to kinship. For this reason, it is often specied by the Drung themselves that the exchanges for oxen often concerned orphans, and more signicantly—in this patrilineal society—children who did not have a recognized father. By extension, one can consider that for the Tibetans in Tsarong, this exclusion referred to the community of language and culture, as well as religious identity, so that the slaves originated from groups considered

vol. 556 A(2). Letter from M. Renou to Mgr de Sinopolis, Tcha-mou-tong, January 28, 1859 [1379]. 17 See also Bailey, 1912: 338; 1945: 89.

the missing share

271

“ethnically” different and non-Buddhist.18 Thus, a difference remained according to the origin of the person. In this area, between Tibetans themselves slavery was initially a system by which persons could be pawned (pledged) for debts, and the enslavement was thus only the result of a transformation of the status of the pawned person. In the long term, the pawned person was likely to become a slave.19 What is of importance is that the trafcking of the enslaved proceeded between groups that were distinct by their identity and status, and that the slaves were generally, it seems, children or teenagers. However, the sale alone did not make it possible to ensure that they would always be slaves: in the case of debt, it is possible that we are closer to the workings of a pawn system. But if there was a rm sale and the individual who was sold was cut from his family ties, it would indeed be slavery. However, should they be pawned for debts or slaves, they would likely fall into the networks of the slave trade. In Tsarong, it seems that pawned persons could be sold or transferred, like slaves. Therefore, if we consider documented cases, slavery is found together with other types of service labor that do not imply a radical exclusion from society. In the case of a person pawned for debt, or of voluntary dependence or abduction, (which could, however, mean that the individual would actually be cut off from his familial ties), the bond with the community of origin is preserved, which implies the possibility of a return to that community. Similar facts are to be found on a regional scale, and a comparative study could be carried out from the borders of Yunnan to Assam. The great diversity found in status complicates such a task,20 but one can underline at least two signicant characteristics regionally: the important role of external slavery (by capture or abduction), and the phenomenon of internal slavery, which corresponds to the possibility of selling a member of one’s community or, more precisely, to having rights over a person and being able to transfer these rights. The sale of slaves has also been documented among the populations of Upper Burma. For example, among the Taron(g), adultery was punished by the sale of the two guilty parties (Mya-Tu, 1966: 33). Further to the south among the Maru, the sale of people to the 18

This point was highlighted by Lazcano Nebreda, 1998: 229 –231. About the pawning of persons, see Testart, 2001: chap. 3. 20 From a theoretical point of view, the discussion in Testart (2001: 115, 151–152) is to me the most rigorous. 19

272

stéphane gros

Kachin ( Jinghpaw) seems to have been relatively frequent (Pritchard, 1914: 528). It is probable that, in the rst case, the key issue was about selling into slavery, but the second case could relate more to sale for indebtedness, with, at least in theory, the possibility of releasing the pawned person from the enslaved state by repurchasing him/her later on.21 Nevertheless, there can also be no intention of repurchase, so that the (voluntary) sale becomes, in fact, nal. For the Kachin, the mayam, who according to Leach represented almost half of the population, were slaves; they were out of kinship bonds and could be sold [but see Maran, this volume].22 Leach had a different perspective and, according to him: “the majority were voluntary serfs—or even adoptive sons—of their Master, rather than movable goods” (Leach, 1968: 145, n. 2). The problem of voluntary dependence also exists in the case of the Drung. Some—without preliminary debts, without being sold—decided to put themselves at the service of a Tibetan chief. With regard to voluntary dependence in northwest Yunnan, many seem to have preferred the servile condition to freedom. “Slavery is all in all only a life time’s domesticity, without pledges, and many slaves, accustomed to this condition, prefer it to freedom” wrote Father Goré (1992 [1939]: 106). Outside of Tsarong, in the Tibetan communities of the Mekong valley, the slaves seemed better off. According to J. Bacot, serfdom would be a more correct term. He gives the example of a Tibetan who had seven wives and about fty slaves (he himself did not know exactly how many), with whom he shared his harvest. “This appearance of richness does not require at all the possession of a single taël. [. . .] No external difference distinguishes the Master from the slave. They wear the same clothes, eat together and are polite with each other” (1912: 273; emphasis added). To me, the ‘softness’ of the living

21 See, for example, this case mentioned in Pritchard (1914: 530): “He was sixteen years old, and had been a slave in Kachin country for two years. His people had sold him originally for a coat and cooking-pots. He had nothing but good to say of his Kachin owners. [. . .] Pritchard asked him why he did not run away from his Kachin masters, and he replied that, in the rst place, he had no wish to do so, as he was very well treated; and, in the second, his own people who sold him had told him that they would sooner or later buy him back” (also quoted in Leach, 1997 [1954]: 303). In this case, the sale is not denitive and the pawned person is still a member of his kinship group. 22 I follow Testart (2001: 151–152) on this point, based on the data provided in Appendice III in Leach (1979 [1954]: see also 160–162); See also Nugent, 1982: 519.

the missing share

273

conditions does not change anything with regard to status. Indeed, it is a matter of wealth, as the master is rich because of his dependents. The fact that it could be, as Bacot put it, just an “appearance of richness” is signicant. It follows that prestige is acquired thanks to the number of dependents or slaves.23 The slave, sharing the misfortunes as well as the good fortune of his master, also contributes to the master’s prestige. Ritual and Power By emphasizing the complexity of political and commercial interactions and the dynamic relationship between these contiguous societies, one can point out some fundamental aspects of their relationships. The importance of trade and debt invites us to see in exchange a means, as limited as it is, of access to power. However, it is a power that goes well beyond the economic domain. By recognizing the value that Drung people granted to some goods that they had secured by trading with their neighbours, one restores the active role of these goods within the exchange system. For example, gongs, which came from Burma, iron utensils (machetes, pots and tripods in particular) were all items of prestige that Drung traded with their neighbours. Indeed, they formerly constituted the principal set of goods that, locally, could enter into the composition of some forms of compensation, in particular those made for matrimonial exchanges. Among these goods, the most prestigious were certainly oxen. They were bought outside the Dulong valley because Drung people themselves did not raise them. Drung people and some of their neighbours needed oxen as a prestige good for matrimonial exchanges, as well as for communal sacrices. Locally, among the Drung, there were sufciently rich individuals who, for their own prestige, organized on their own behalf the sacrices of oxen, and would divide the meat that would be partaken. Because Drung people carried out their own sacricial rituals in their villages, they needed oxen, which they often got by exchanging those excluded by their society. The equivalence (or conversion) between ox and slave

23 About the Kachin, Leach (1968: 161) mentioned that chiefs gained their reputation by having many slaves, but there were many economic advantages to the servile condition. Contractual slavery was a means to obtaining economic credit and political protection from the chief.

274

stéphane gros

in the exchange, tends to show that the slave as well as the ox entered the category of ritual goods that ensure the prestige of the owner. From the Drung people’s point of view, oxen were the main goal of the trade that sometimes involved the circulation of people, often children and women. The exchange of persons occupied the highestvalue level of exchange. Furthermore, if “exchange” may be described by detailing what one person gives to another, there was a higher order of exchange, a larger set of relations that structured exchange itself. It is worth mentioning here how this structure of exchange and debt relationship was also used on another level by the Drung, by referring to their political allegiance to the Tibetan chieftains of Tsarong. Formerly, the Tibetans chieftains were considered by the Drung of the northern part of the Dulong valley as the “masters of the place” (mvli aqkang), and a tax was annually perceived as a compensation “to eat the land” (amra kai ). This relationship of subordination was initiated in a particular way. The Drung tell the story that only those in the north of the valley who had received their share of the meat of nine oxen that the Tibetans had once killed, became subordinate and had to pay the tax. The meat had been divided in the form of skewers and had been distributed in all the north of the valley. It is told that in the south of the valley, the distribution did not take place because the six porters (Drung people originating from the south of the valley), died on the way. Thus, “the tax was cut” (kri tot) where they died, this place becoming a geographical limit: not having received a share of meat, the inhabitants of the south of the valley would not pay the tax. The distribution of the meat among the inhabitants of the Drung valley legitimised the political ascendancy of the Tibetans who were recognized as “masters of the land” by the Drung. One could expect the distribution of meat to be carried out by those who had a claim to power. But, in that case, the status of the Tibetan chiefs seemed already acquired, and the distribution was but a legitimisation. According to Drung people’s accounts, the distribution of meat carried out by the Tibetans chiefs seems to have crystallised the hierarchy by conrming a relation of dependence. Interestingly, this very same process of meat distribution appears also in the relationship between the Naxi tusi (indigenous chiefs) and their dependents, the Lisu, Nung, and Drung people of the Mekong and Salween valleys. Some historical records mention that this chief, one of the main authorities in northwest Yunnan up to the beginning

the missing share

275

of the twentieth century, organized feasts for his dependents.24 Every year, all (the representatives of ) his subjects were required to come and bring him a present for the New Year, and he was to entertain all of them for three days. Every three years, he also organized feasts during which he distributed a piece of beef. These feasts seem to correspond to something other than a simple tributary relationship. What must be underlined in this case is that political centralisation was also a ritual process.25 Through these feasts, the status of the tusi and the political centralisation that he represented were conrmed and his prestige was enhanced. The extent of the festivals given by the Naxi indigenous chiefs seem to have been sufciently signicant for Roux to refer to them in the following terms: In a few days [at the beginning of August 1895], more than one million of these natives [Lisu, Nu and Qiuzi] will come, like every year at the time of Ho-Pa-Tsié,26 to bring to him, at a rate of approximately one taël per capita, their tribute in kind, in the form of wax, musk, powder of gold, medicinal plants, etc. On the other hand, he will give each of them a piece of meat, a bowl of rice and a cup of brandy; fteen oxen are hardly sufcient for these festivals, during which the savages perform in front of him dances and songs, as if adoring him. [. . .] Formerly, people of the Kioukiang [Qiujiang] valley had each year to provide him a slave. Currently this habit is abolished, but, to preserve at least the form and part of the value, they give him in place a statue made out of wax of the size of a twelve years old child (Roux, 1897: 227).

This reference to slavery once again stresses the importance of this institution and the form of dependence that could be established. If a slave is due, one leaves the sphere of exchange to enter an asymmetrical relationship characteristic of dependence. Therefore, power seems above all to be a matter of economic control rather than effective political authority. And control was essentially asserted over ritual goods (such as slaves and oxen), enabling one to assert one’s rank or claim for status. In northwest Yunnan, as in neighboring areas, the role and value of the ox was of primary importance: the ox can be seen as a good that enhanced status and prestige, as did slaves. The sacrice of the 24 There are historical documents that refer to the period of the seventeenth century, and later on to the period when the French missionaries were present (end of the nineteenth century). 25 On that point see also Sagant, 1990: 163. 26 Most probably Huobajie, the Chinese name for the Festival of the torches.

276

stéphane gros

ox and the sharing of meat was an expression of these local conceptions of prestige and wealth, common amongst several societies in the Himalayan border areas, as well as in mainland Southeast Asia and often refered to under the label of “Feast of Merit.”27 Moreover, the ritual slaughter of animals and the distribution of their esh appear to be the principle means through which one could acquire (or conrm) power and status. Stevenson (1943) rightly qualied the working of these feasts as a “prestige economy,” for one of their main concerns was with prestige and wealth.28 Such a sacricial practice remains a cultural characteristic of the Drung people today. Even though they abandoned it under the political constraints of the 1950’s, the meaning and cultural value of the practice is still very much alive. The Missing Share One of the many myths the Drung people have in their repertoire is especially important for understanding some essential values shared by the members of their society. This myth also tells us about the logic (and/or the morality) of interpersonal relationships: One day, after an abundant harvest, a man called Punggrin Pung decided to organise a great feast (dvruq-wa) that lasted for nine days and nights. All types of animals on earth joined the party. Then came the day to end the feast, meat and alcohol where nearly nished. The little that was left, according to custom, had to be shared equally. The number of participants was counted, and the meat divided.29 But, no matter how they would count, again and again, two shares were always missing. The two who received no meat decided to go hunting in the mountains. During the hunt, one of the two disappeared and became a spirit, Tsheu-pvlang, who demands offerings.

27 See Bouchery, this volume, about the Feasts of Merit among the Naga. These feasts could also accurately be refered to as “cults of fertility.” 28 See for example Russell (ed.), 1989. I nd myself much in accordance with Durrenberger (1989: 114) in the way he described the Lisu “ideology of honor”: “In highland Southeast Asia, there is an ideology of honor and wealth that can be translated into rank and prestige under certain circumstances. Where wealth and access to valued goods are scarce, hierarchic forms will develop; where they are widespread, egalitarian forms will develop. Both hierarchic and egalitarian forms are based on the same ideology, but the social forms are largely shaped by economic relationships.” 29 The animal could differ according to different versions.

the missing share

277

This man who became a spirit requires his due. He is one of the most feared amongst the Drung people. He appears as an always unsatised gure, the prototype of the kind of spirits with whom the relationship is that of debt: no sharing can be done without him potentially coming around, as if people’s debts towards him were permanent and could never be paid. This myth is often told in order to account for the origin of this particular spirit and to explain the origin of the sacricial feast that the Drung people used to perform yearly. The Drung name for the ritual sacrice of an ox is called dvruq-wa, literally “group-making.” Its very name underlines one of its essential aspects: according to the general principle, the sacricial victim is cut and its meat shared, the sharing network drawing the contours of the community.30 Among the Drung, this kind of sacrice was an opportunity for the one who was rich enough to own an ox to validate his prestige (koksang) and his wealth (karji ). Drung people did not raise their own cattle in the past, and oxen were obtained through trade relations outside their valley, especially with the Tibetans who exchanged them against goods or, more likely, individuals who would become slaves. Oxen were representative of one’s wealth, and such feasts were the principal means of political expression. On the one hand, they reinforced the cohesion of the local group as well as alliance relationships, but on the other hand, they could only enable one to afrm a provisional dominance, or prestige; this prestige could be challenged, as it was embedded in a permanent process of competition. This individualist competition is totally in opposition to hereditary status. Thus, for the Drung people, it was the language of sacrice and sharing that prevailed at the political level. It was at the heart of the local politico-religious dynamic, assuring social cohesion, reafrming

30 It is interesting to contrast the drung expression, “group making” (dvruq-wa) and the Kachin one “nat making” (nat galaw). Nevertheless, Leach (1979 [1954]) description of the Kachin nat galaw could be extended to the Drung context: “Thus from a certain point of view a Kachin religious sacrice may be regarded as a purely technical and economic act. It is a procedure for killing livestock and distributing the meat, and I think there can be little doubt that for most Kachins this seems the most important aspect of the matter. A nat galaw (‘nat making’, a sacrice) is almost a synonym for a good feast. But from the observer’s point of view there is a great deal that goes on at a sacrice that is quite irrelevant as far as butchery, cooking and meat distribution are concerned. It is these other aspects which have meaning as symbols of social status, and it is these other aspects which I describe as ritual . . .” (p. 13; see also pp. 172, 174).

278

stéphane gros

power relations and kinship ties. The ritual ensured the maintenance of the ties that linked the sacrice’s giver with his social horizon, his relatives, afnes and neighbours. At the same time, the function of sacrice was crucial for the internal process of competition and redistribution within Drung society. Similar to sacricial practices such as “Feasts of Merit” that emphasise sharing and the display of wealth, these feasts could create and/or validate distinctions regarding power and social status by transforming material wealth into social rank. And in some cases, these feasts were more about recognition or legitimisation than the granting of power.31 Nowadays, the ceremony of the ox sacrice has disappeared among the Drung people. But its underlying principle still prevails in the Drung people’s ideology and representation of power. What is still alive is the idea that any pretension to authority should be validated through generosity. This allows us to understand some aspects of the present day relationship of the Drung to the Chinese state (see Gros, 2005). Constant subsidies and assistance from the communist state, perceived as a welfare state, are an integral part of the validity of its power and authority. Without what can be seen as a constant dispensing of wealth, the state would lose its legitimacy and prestige—a prestige that it had gained, especially for the last generation, through constant assistance and help. This state, seen as a “provider of goods,” appears through the lens of the Drung’s vision as only an historical manifestation of a type of political power that depends on a principle of redistribution. Obviously, some fundamental principles of social relationships pertain even if the structure of which they were a part are being altered or destroyed. The “prestige economy” does not play a central political role anymore; it is relegated to the margins. In the Drung valley, attempts to effect conversion to Christianity have been relatively successful. “Religion”—Christianity—is being valorised in opposition to “Superstition”—popular religion. Christians gather regularly for the holydays of the Christian calendar, and at these gatherings, pigs or other animals are slaughtered and their meat is

31 One could refer for example (among many others) to Fürer-Haimendorf (1967); Lehman (1989); Woodward (1989); Jacobs (1991: Chap. 8); Kammerer and Tannenbaum (1996).

the missing share

279

shared among all participants. According to the Drung themselves, it is the same as “making the group” (dvruq-wa). These Christian feasts, and the community they contribute to creating are a contextual answer to the present economic, political and cultural situation. From that perspective, Christianity could constitute a new horizon. Conversion could be a way to go back to the time of sacrice, to a social cohesion in which a new “being together” becomes possible.32 Macdonald (1980) has pointed out the importance of the conceptual model of “creative dismemberment” among some Himalayan societies and argues that the model could directly be used and exploited by political and economic elites as an instrument of social control. In the context of political centralisation, the sacrice gives way to relations of domination, as we have seen. One can also argue that, through the sacrice, it is rather previous conditions that are conrmed and reestablished: the ties of reciprocal obligations, the stratied status in a social hierarchy, etc. According to Oppitz (1997), the creativity lies in myth, whereas in reality, sacricial practices can only re-create. But it could also be argued that through meat division, something else comes into existence, and is, therefore, created. For the Drung, the logic of sacrice was central politically; it was a means by which the contours of the society could be redrawn, a means of making and un-making the group. The mythical model is that of “the missing share.” That is to say, to paraphrase the myth, that the heart of the matter is a debt relationship. Humans have a debt toward the spirit that each sacrice can only temporarily appease. But among humans themselves, the feast giver’s generosity and the subsequent sharing of the meat, under the appearance of the gift, hides the debt. We have seen that debt relationships have played an important role in the constitution of local spheres of power in the larger region of northwest Yunnan. These relationships were formalised by ritual means and ritual came to assert the participants’ respective positions. In this case, it seems to me that ritual communication is not just an alternative way of expressing something. It is also the expression of things that cannot be expressed in any other way, and maybe, as Gellner (1999: 139) argued, “it is precisely because rituals do not simply say things that they acquire their power to persuade and legitimate.”

32 The problem of religious conversion among the Drung is the topic of a work in progress.

280

stéphane gros Conclusion

Because his broad conception of ritual as communication extended to many aspects of social life, the notion of ritual language used by Leach seems rather fuzzy. As such, the notion of ritual language can prove helpful in dealing with aspects of identity and political and religious relations in a multi-ethnic area, be it the Kachin Hills or elsewhere, such as in northwest Yunnan. Yet, despite its heuristic value, which enables one to point out similarities between otherwise dissimilar groups, it needs further elaboration. I have tried to show that for the Drung, meat division, and more generally sacricial feasting that involves the distribution of the esh of the sacriced animal, appears to be a model of internal and external power distribution.33 But it is also, more generally speaking, a model for inclusion and distinction. In other words, there is a double movement: at the same time as it creates or conrms social relationships, it also sets people apart by classifying and attributing social positions. In the case of the Drung, I see debt as a prototypical form of relationship that prevails in both relations between individuals and groups, and between people and supernatural entities. It is this ideology that expresses itself through ritual actions, such as meat division. Or to put it another way, the notion of debt is the principle of efciency that structures the world view and gives ritual performances their meaning at the same time. I am tempted to propose that the notion of debt could aptly characterize some essential aspect of the common structure on which this ritual language of sharing is articulated. I would argue that Leach’s notion of ritual language can be reformulated by placing the notion of debt as a central element of what this “language” tells us of the underlying order. In this specic case, the ideology that the ritual language expresses is in a broad sense an ideology of “power,” but authority and control only appear with control over the source of wealth, which includes ritual goods (oxen) and land. This ideology involves interrelated concepts of wealth and power that are central in the dynamic of social relations. This dynamic itself is supported by the logic of the debt that we nd in relationships between people, and between people and the other world.

33 This point was already made clear in a comparative perspective in Russell (ed.), 1989.

the missing share

281

To paraphrase Leach then, I would say that a common ritual language exists insofar as what it expresses through symbolical performances is a general ideology of power that reects not simply social relations, but relations of indebtedness. And this formulation, I believe, nds its roots in Leach’s work itself, partly in Political Systems and clearly in his concise Social Anthropology (1982). The practices surrounding debt relationships were analysed in detail by Leach when he dealt with the categories of hka (debt) and hpaga (trade, ritual wealth object). As Leach formulated it, almost any kind of legal obligation existing between two Kachins is likely to be described as a debt, and debts are expressed in terms of hpaga (for a discussion, see Ho this volume). He also underlined the important aspect that “the debt is a kind of credit account which ensures the continuity of the relationship. There is thus a kind of paradox that the existence of a debt may signify not only a state of hostility but also a state of dependence and friendship” (1982: 153). From Mauss to Lévi-Strauss, following a social approach to exchange, there is more in the exchange than just what is exchanged. What is “more” is the social relationship established in the process. In The Gift, Mauss insists on the imperativeness of the obligation to reciprocate, and even states that “the punishment for failure to reciprocate is slavery for debt” (1990: 42). It seems that Leach’s position was very much inspired by Mauss, with the nuance that the network of social relationships that one can observe through gift-exchange behaviour is a network of indebtedness. He makes this point clearly in Social Anthropology by writing: “persisting relationships only exist as feelings of indebtedness. From time to time every such persisting debt relationship needs to be made manifest in an actual gift transaction, but the relationship is in the feeling of indebtedness not in the gift” (1982: 154). In other words, the mechanism of gift-exchange is that of debt.34 Though not stated very explicitly, Leach’s concern in Political Systems was very much the notion of power, but in the limited sense of the “attribute of ‘ofce holders’, that is of social persons who occupy positions to which power attaches” (1997 [1954]: 10). But if power is a result of participation in interpersonal relationships, both the political

34 Gregory (1982: 19) also approaches the gift economy as a debt economy by underlying that what the giver desires in the gift-exchange are the social relationships themselves and not the things exchanged.

282

stéphane gros

and economic aspects are embedded in a global system. Leach’s analysis in terms of the logic of linguistics applied to ritual can be extended to this global system, and, in his own words: “the elds of kinship, economy, politics, law, religion etc., are all versions of the same thing. Each mode of expression of a particular relationship is metaphoric of all other” (1982: 158). In forging the notion of ritual language, it seems to me that Leach was formulating his version of Mauss’ “total social fact.” Both notions refer to the totality of society and its institutions, and to the mechanism by which individual interests combine to make a social system. In a way, Leach’s conclusion in Social Anthropology could, retrospectively, be used fruitfully as an entry point for understanding some important arguments in his Political Systems: “The thesis that runs through this book is that it is only when we come to understand that relationships between man and man and man and god are, at least in a metaphorical sense, the equivalent of economic indebtedness, that we can really appreciate how this transformation of economics into ideology through the mediation of kinship actually occurs. So it is really Chapter 5 [Debt, Relationship, Power], which elaborates a sociological theory of debt, which provides the keystone to my argument” (1982: 223). And it is very much the same argument that lies behind his treatment of his Kachin ethnography and his use of the notion of ritual language.

TRANSETHNIC SOCIAL SPACE OF CLANS AND LINEAGES A DISCUSSION OF LEACH’S CONCEPT OF COMMON RITUAL LANGUAGE François Robinne Primacy of Clans and Lineages over Ethnic Categories Leach starts the analysis in his classic study Political Systems of Highland Burma by considering the concept of ‘ritual language’. Stéphane Gros (this volume) previously synthesized the notion of ritual language with its social implications in the account he gave of debt among the Drung, whose representatives in Burma are integrated within various subgroups of the Kachin. This chapter acts as a complement to that of Gros in that it, too, is focused on the same notion of common ritual language. However, in this chapter I will extend the analysis to reect on the mayu-dama kinship system, which I consider as a transethnic space in which the Kachin subgroups of Kachin State in Burma interact. Refuting the dichotomous approach of sacred/profane developed by Durkheim, Leach denes the notion of ‘ritual language’ as ‘meaningful things’, whether or not they imply any supernatural or metaphysical conceptualisation. According to that denition, the exogamous clans and lineages appear to be a common ritual language shared by the Kachin populations of Kachin State, where social organisation is based on the general exchange kinship system, or where that system has been more or less imposed. Unlike the shamanic traditions of ritual performance, which have almost disappeared, and the traditional political structures that have disintegrated, the matrimonial system of the Kachin has, in contrast, survived and accompanied the process of social change. Although still based on the mayu-dama system, this matrimonial social organisation is not a system xed in time; it has metamorphosed both in space and time, and Leach denounces rightly the idea of a common ritual language when reduced to a ctive notion of stability applied to a global system. In the precursory study by Marcel Granet (1939: 240) and quoted by both Lévi-Strauss (1947) and Leach (1954), it was demonstrated that there were more than ve main Jinghpaw clans as described rst

284

françois robinne

by Gilhodes (1922: 162), but none of these authors reconsidered the general hierarchy inherent in the general matrimonial exchange between these clans: the Marip give to the Lahtaw, who give to the Lahpai, who give to the Nhkum, who give back to the Marip. Based on this myth—a universal myth considering its surrealistic stability—the system is more theoretical than effective. However, because of the segmentation process, and also because of the centralisation that broke down the traditional, localised monolineage group inhabiting a long house, marriage may occur nowadays between partners of different lineages originating from the same clan. Leach demonstrated the exibility of the mayu-dama system. He showed that the local understanding of clans and lineages is subject to manipulations like any other social sphere. Among the numerous examples on which his demonstration is based, Leach (1972 [1954]: 110–111) describes the conict between two lineage groups—the Maran Nmwe and the Nhkum Laga—where each one claims a supremacy over the other. Both refer to the hierarchy prevailing in the wife-giver wifetaker relationship, but the two groups interpret this kind of hierarchy according to their own interests: one side stands by the reality of the matrimonial exchanges, whilst the other has recourse to the mythical referent understood in the absolute. Generally speaking, the perennial nature of the clans and lineages from the colonial period up to now contrasts with the reication of ethnicity involved both in the process of Christianisation and in the contemporary pan-Kachin movement, at least indirectly, since it does not relate specically to the activities of their respective representatives. To contract a marriage, a Church membership sometimes may consider redhibitory the alliance between a bride and a groom who belong to two different Christian congregations, whatever may be the social adequacy of their clan criteria. Despite its metamorphoses and instrumentalizations, the mayu-dama system has been partly preserved by the ascent of Christianity amongst the Kachin, and the various notions of Christianism that have resulted. Considered from this perspective, therefore, on the one hand the mayu-dama system helps to give sense to the Kachin ethnonym when it is understood as a multi-ethnic community; on the other hand, it also contributes to the modern ideology of pan-Kachin identity, which is dened mainly with regard to the central Burmese power. The tendency for the Kachin lineages to become clans is a further factor helping to explain the survival and renewal of the mayu-dama principle in Kachin society. However, as stated, this is not without adaptation. For example, following the near disappearance of non-Christian

transethnic social space of clans and lineages

285

public ritual performance, the processes of centralisation and the wave of Burmanisation that have threatened to overwhelm other aspects of ‘Kachin’ social organization, a closed intra-clanic marriage is no more considered an incestuous marriage among the younger generations of Kachin. New conditions of education are partly responsible for this, as they have tended to obliterate the unimaginable transgression of the basic matrimonial rule, according to which, after having taken, we cannot give before a minimum of three generations have passed.1 It should be stated that, in all the cases observed above, the mayu-dama system appears to be instrumentalized according to local circumstances. Leach, too, denounced any structural stability (1972: 30, 318). In the same way, we can say that the notion of a common ritual language applied to the mayu-dama system is an artefact inherent in the process of social construction and reconstruction. Subsequently, the notion of ‘contradiction’ should not be opposed to the notion of ‘coherency’ and ‘coherency’ does not imply ‘stability’. It is from this perspective that we can understand how the mayu-dama system contributes signicantly to the social coherency of the so-called Kachin subgroups. The present-day ofcial meaning of ‘Kachin’ is asserted through the union of six ethnic groups, and this union is supposed to be symbolised by the monumental manau ceremony [see Sadan 2000a, 2005]. This ritual was traditionally performed around a set of sacricial posts that have in the contemporary situation metamorphosed into Christian symbols through which the traditional village unity has been extended to a pan-Kachin unity. Behind this representation of an ideal, and beyond the ofcial assertions, the problem remains to extract from the colourful folklore the social depth contained today in the word Kachin. Quoting Leach (1972 [1954]: 98): ‘the kinship networks extend the linguistic frontiers’. Fifty years after Leach’s analysis, my aim here is to show that the agnatic system of liation can be dened as a common ritual language. This is not only because each ethnic subgroup has its own exogamous clans, but also, and mostly, because the different clans and lineages intrinsically extend the social sphere from the articially considered delimitations of ethnic categories to the dynamics of their intrinsic relationships. In other words, the approach developed here reconsiders the traditional acceptance of the ‘ethnic’ (both as discrete

1 Concerning the possibility of a symmetric dynamic within the asymmetric cycle involved in the generalised exchange, see Maran La Raw’s development about the hkau wang system in this volume, and Robinne 2007.

286

françois robinne

entity as well as sets of communities) and the corollary idea of universal and a-temporal cultural markers. To support this hypothesis, I will focus the analysis on the confusion between the use of terms of ethnic category versus references to clans/lineages and its implications. Ethnic and Clanic Categories: Error and Confusion A confusion recurred among the colonial administrators and the missionaries who rst wrote on the Kachin. Quoting each other, most of them assimilated ethnic categories—or “tribes” according to the terminology—and clan/lineage categories. For example, we read in the rst volume of the monumental Gazetteer of Upper Burma published in 1900 the name of ‘ve Chinghpaw tribes’ associated with the ve main clans generally mentioned in the mythology: “The ve parent tribes are thus 1) Marips, 2) Lahtawngs, 3) Lepais, 4) ‘Nhkums, 5) Marans” (Scott and Hardiman, 1900: 373). In his Kachin-English Dictionary, Hanson (1954 [1906]: 172, 359), characterizes the Nhkum clan as a tribe and the Labang lineage as a clan; the Gauri are dened as a clan of the Lahpai tribe. In the same way, Grierson, author of the Linguistic Survey of India (1927, Vol III, Part II: 500) describes the Kachin as ve tribes whose names correspond to the ve main Jinghpaw clans: “The Kachins themselves are divided into ve tribes, which all claim to be descended from the grandson of the rst ancestor. These tribes are the Marips, Lahtawngs, Lepais, ‘Nhkums, and Marans”. A similar confusion was made by Hertz (1960: 81) among others. A noteworthy exception (albeit contradicted later on by the same author) was Enriquez (1933: 56) who wrote: “The divisions of the Kachin, loosely spoken as “tribes”, are, in fact, nothing more than clans. The ve main divisions (Marip, Lahtaw, Lahpai, N’Hkum and Maran) are really ve aristocratic clans descended from the ve eldest sons of Wahkyet Wa, the reputed Father of the Kachin race”. It is not time to discuss here the nal terms used in the quotation, but they reveal that Enriquez was caught up in the same stereotypes of the colonial period about ‘the Kachin’ as others. What we can say, however, is that this confusion between ethnic category and clan/lineage names is generally in marked contrast to the quality of the observations otherwise made by these same authors. The propensity to make clans and ethnic categories analogous is signicant, however, for the ethnic

transethnic social space of clans and lineages

287

reication of which authors of the colonial period, followed later on by researchers of the immediate postcolonial time as demonstrated by Aung Thwin (1998), were guilty. Caught in their own universe they were unable to consider fully the transethnic space that lay before them. In fact, this tendency seems to have been a confusion rather than an error stricto sensu: the investigator would ask his question according to the perceived environment and the people investigated would answer according to the terms used in the question and, often, in line with the wishes of the interlocutor. During my numerous periods of eldwork in Burma, which started in 1983, I have often been confronted with this crucial problematic of methodology. A few years ago, while working on interethnic relationships in the Southern Shan State, I adopted a comparative approach to the question of kinship terminologies. Because of the ethno-linguistic complexity in this area, combined with my linguistic incompetence, I was not able to speak any other language than Burmese and its closed dialects with my informants. Among the Pa-O, I did not understand at rst that they were reproducing for me the Burmese terminology, which I quoted as Pa-O terms. Unbeknownst to me, I was caught up in the dominant Burmese universe, as was Hackett some fty years before in the same area. More recently, while working in Kachin State, I understood that the confusion made by the British of the colonial period is, in fact, signicant of the secondary value that informants themselves attribute to ethnic categories. My investigations showed that, to the question “To which ethnic group do you belong?”, the spontaneous answer would invariably be the name of the clan and its lineage subdivisions. This means that, at least in an agnatic system such as among the Kachins, in its plural form, the lineage referent goes beyond the criterion of ethnic belonging. This postulation was developed in ways outlined below following extensive eldwork in Kachin State which started in December 1999. Homonymous Clans among the Kachin Subgroups After Marcel Granet, Leach showed that the Kachin (more accurately the Jinghpaw) clans numbered seven or eight, but it was generally recognized that there are ve main clans among the Jinghpaw Kachin sub-group: the Maran, the Marip, the Nhkum, the Lahpai and the

288

françois robinne

Lahtaw. Only these ve clans are generally enumerated by occidental observers. However, according to my recent eldwork investigations, as well as some previous occidental sources, it appears that these ve clans are not the social attributes of only the Jinghpaw: they are also common to other Kachin sub-groups. In some cases terminological homonyms are attested, which means that the same clan names are used by different ethnic groups. Some of my Atsi informants (otherwise known as Zaiwa) in Myitkyina, Mandalay and Yangon say, for instance, that their clans have the same names as those of the Jinghpaw. Leach (1972: 80 [1979: 53–54]) made the same remark concerning the Gauri ( Jinghpaw) and the Atsi (Zaiwa). Despite some orthographic variants, due mainly to irregularities in transcription, their lineages are homonymous. The fact is important enough to justify quoting the passage in its entirety: Atsi chiefs consider themselves to be all members of one lineage, namely Lahpai-Shadan-Aura. The Laphai-Shadan lineage of which Aura is a segment includes large numbers of very inuential Jinghpawspeaking chiefs. The Aura lineage itself includes the Gauri chiefs. Consequently, the Gauri chiefs, who speak a dialect of Jinghpaw, and the Atsi, who speak a variant of Maru, are always regarded as close lineage brothers. Leach adds about the commoners (ibid.: [1979: 54]): Many of the commoner lineages also are ‘the same’ in both language groups, thus: Gauri lineage name

Atsi lineage name

Dashi Jangma Mahka Sumnut

Dawshi Jangmaw Mahkaw Sumlut, etc. etc.

This identity is socially recognised. Dashi are lineage brothers to Dawshi. A Gauri Dashi might properly marry an Atsi Jangmaw but not an Atsi Dawshi.

Common Clans with Different Names The correspondences mentioned above are actually a list of terminological equivalences: different names for the same clans, or at least for exogamous clans.

transethnic social space of clans and lineages

289

Interviews carried out in Myitkyina with Maru (or Lawngwaw) speakers have contributed to showing that there are clanic correspondences with the Jinghpaw, a fact corroborated by Wawn Awng (s.d.: 65), a Jinghpaw writer. According to this research, the Lawngwaw share in common with the Jinghpaw the Marip and Lahpai clans. The correspondences may vary between the informants; this is due to the fact that the clanic equivalences between the ethnic groups are generally reduced to the effective matrimonial alliances within their own patrilineage, rather than to the normative possibilities inherent in the generalized exchange system. For this reason the following table is not complete: Jinghpaw Marip Maran Nhkum Lahpai Lahtaw Jasen Tangbau

Lawngwaw Lanin Je nin Yao nin Cao dung and Lao nin Banin Jein Phang Xao puy or Zawbuk

Labau Sau Nu, a Baptist professor at the Myitkyina College of Education, is a Lachid of the clan called La Nnan. He considers his own clan to be the equivalent of the Jinghpaw clan Marip. He adds that the clan Mading—fragmented into three exogamus lineages—corresponds to the clan Lahtaw, the clan Mang Shang to the clan Maran, and the clan Langawi to the clan Labang. According to him, a Lachid man cannot take a wife from his equivalent Jinghpaw clan. Furthermore, the account of Wawng Awng shows that the Lachid and the Lawngwaw have in common the Tangbau clan, and that the Jinghpaw and the Marip have in common the clan Lahtaw. A Rawang informer living in Mandalay says that his clan name, Agu Di, is equivalent to the Jinghpaw clan Marip. Garu Min, adept of the congregation of the Churches of Christ in Myitkyina, says that the Nung Lungmi—a Rawang sub-group or else ancestors of the Rawang according to the informants—also established some clanic correspondences with the Jinghpaw. The clanic correspondences are generally kept at the level of an assertion, and most of the informants have some difculty in establishing a list of equivalences beyond their personal cases. Nevertheless, the following correspondences have been recorded:

290

françois robinne Jinghpaw Marip Maran Nhkum Lahpai Lahtaw

Rawang and Nung Lungmi Rawang Yintang et Magawng Sangtung Garu ?

Enriquez (1993: 56) clearly pointed out the confusion between clans and ethnic categories: “Lazum, for instance, are sometimes counted as Gauris [a Jinghpaw southern sub-group], sometimes as Marips [a Jinghpaw clan]”. But it is Green (1934: 254, quoted by M. Sadan; see Sadan 2007), who rst detailed the correspondences between the clans of the different ethnic groups: There are many clans which are generally named after the valleys in which they live. Local groups are generally clans, although some local groups containing two or three clans are found [. . .] Some of the clans compare themselves to the main Hkahku clans as: The The The The The

Nanhku claim to be Maran, Sandong claim to be Lahtaw, Tingkan claim to be Lahpai, Yintangse claim to be Marip, Kumring claim to be Nhkum.

The Lisu Case The most striking correspondences are those made with the Lisu established in Kachin State. Contrary to the Jinghpaw and to the other subgroups considered to be Kachin, the basis of Lisu social organisation is not the mayu-dama general exchange system. It is possible for a male to take a female in any lineage except its own patrilineage: from Putao to Myitkyina, Lisu informants agreed in saying that they could take a wife from the mother’s side, but not from the father’s side. Some of them, interviewed downriver from Myitkyina at Waingmaw, explained that marriage with the bilateral cross-cousin is recommended: they call it fu shasha, which means, word by word, “to give in marriage + to cross two times”. Nonetheless, despite the Lisu kinship structure, and despite also the interferences caused by belief in Christianity and membership of such and such a religious congregation, which can affect the choice of a spouse, part of the mayu-dama structure is seen to be incorporated by the Lisu as soon as they have committed themselves in a matrimonial alliance with their Kachin partners. A Lisu man having taken a

transethnic social space of clans and lineages

291

wife from a Jinghpaw clan becomes, tacitly, a dama —which means a “taker”—of the clan of his wife, which in turn automatically becomes his clan’s mayu—which means a “giver”. According to this, the Lisu man and his siblings, real or classicatory—by extension all the kahpu kanau dened by Leach (1979 [1954]: 73–74) as “lineages which are treated as being of the same as clan as Ego”—are integrated in a matrimonial system in which one can continue to take wives from the other clan, but, on the contrary, one cannot give a wife to them in return. Once again, we owe to Green (1934: 254, quoted by M. Sadan see 2007) the oldest account describing the spread of this clanic correspondence among the Lisu. Because of variant forms of transcription, it has proved sometimes difcult to trace the equivalences in my own notes. Despite this difculty, it appears that the accounts obtained sixty years apart from each other establish the veracity of such correspondences, whether or not the terms are homonymous: Jinghpaw Marip Maran Nhkum Lahpai Lahtaw Lasang Tawn Kya

Lisu (Green) Ngaw Hpa Lair Mair (Myitkyina) Gu hpa and Waw hpa Zaw Kya Lair Mair (Bhamo) Mitung

Lisu (Robinne) Ngau hpa or Nga Zi hpa Wa hpa La hpa Li hpa and Nu

Chinese and Other Cases of Clanic Correspondence A detailed analysis of such matters in recent publications in vernacular languages indicates that the same pattern holds true for other cases of Kachin and non-Kachin clanic correspondence. Recalling the story of his own clan, Matsaw Nhkum Tu Yau (s.d.: 25) associates the Nhkum with the Rawang, the Maru, the Lisu, the Lachid, the Zaiwa, the Sam (Shan), the Miwa (Chinese), as well as with the Jinghpaw. Moreover, it seems that the sub-groups of Jingpo established in China also integrate themselves within this inter-ethnic clan dynamic. This dynamic is not only due to the classic phenomenon of pluriethnicity, it also involves the claiming of two or more ethnic identities by an individual, as demonstrated by Leach in a Kachin context and by Conrad in a Lisu context. Speaking about the Jingpo domains, Wang Zhuseng (1997: 93) (1997: 314, note 12) remarks that:

292

françois robinne Puguan, the formal name of Dazhai, is also the name of a principal clan that lives in Dazhai and its immediate vicinity. It is said that Puguan was one of the lineage segments of the Maran clan [. . .] Like all all other Jingpo names, Puguan has a corresponding Han Chinese name, Li.

And the same author also enumerates a corresponding list of Han Chinese names with some of the Jingpo clans ( Lahpai = Pai; Lahpang = Wang; N’kum = Kong, Li Yue; Marip = Sha, Shang, Ma), as well as with the different Kachin sub-groups: eight clanic correspondences between Zaiwa and Han Chinese names, ten between the Lachi and Han Chinese names, four between Langwo [Lawngwaw] and Han Chinese names (ibid.: 315–316, note 3).2 While not focused on that question, Wang Zhuseng’s analysis is sprinkled with these kind of correspondences, for example: “It was commonly recognized that the Zaiwa chief clans were from Jingpo’s Lahpai” (ibid.: 314, note 12). The different Kachin sub-groups of Burma, including the Lisu, give numerous examples of such clanic correspondences with the Han Chinese. The practice refers back to the networks that operated between different populations, and insofar as we may wish to understand the operation of some of these networks in both space and time, the study of matrimonial exchanges can contribute greatly in revealing to us some of their dynamics. Some differences appear between the Wang Zusheng informants on the China side and my informants on the Burma side. The people questioned at Myitkyina (where the Chinese community is important) state that equivalences between Kachin and Han Chinese clans are the following: Li Ja in Chinese is given as the equivalent of Marip in Jinghpaw, Tcho as Lachyawng, Tuang as Lahpai, Yang as Maran. The equivalent clans given are exogamous; according to my informants, a wedding between a male Jinghpaw Maran and a female Chinese Yang is quite impossible. Similar equivalences, underlying the same logic of social transethnic space, are also attested not only at a clanic level but also at a lineage level, with lineages deemed common to many clans. The examples are numerous: the Kadung lineage is common to the clans Magawng, Lahpai and Nhkum; the Kaja lineage is common to the clans Marip, Nhkum and Magawng; the Bau lineage is common to the clans Nding, Lahtaw and Hkasha. In addition to these examples of one lineage

2 Ofcially, the ethnonym Kachin gather six sub-groups known as Lisu, Jinghpaw, Rawang (including Nung Lungmi), Lawngwaw, Lachid and Zaiwa.

transethnic social space of clans and lineages

293

shared by three clans, I have also noted the name of a lineage among four clans: the Ndup lineage is common to the clans Lahtaw, Jasen, Hkasha and Marip.3 Past and present accounts of clanic correspondences between Kachin sub-groups are relatively rare. Nevertheless, they are signicant in that they relate to a more global social system. Some of the reasons why the phenomenon remained widely unnoticed are suggested here: rst, transethnic equivalence between clans was concealed by use of distinct vernacular terminologies; second, the pre-eminence of ethnic categories in determining our approach to the study of such phenomena has blinded us all, professional anthropologists and others alike, both within the said communities and without; third, and related to this, is that the dominant thought according to which the clans are understood as cultural markers of isolated ethnic groups has diverted our attentions. Clans and lineages are actually part of a larger generalized exchange system, part even of a cognitive process, without appreciation of which one may fail to understand the rise of the pan-Kachin movement. Caught myself in my own a priori assumptions—or in my own intellectual and political world as would say Aung Thwin (1998)—I experienced myself late on the interethnic and the transethnic dimension of the clans of the generalized kinship system to which they belong. The Transethnic Terminology of the System Despite scattered information, the sources cited corroborate my own eldwork observations that the mayu-dama system has an inclusive dynamic that imposes upon the groups that neighbour the Kachin in Kachin State. Recent developments in the system also show that current political concerns have led the logic to reach an extreme end, with the risk even that the kinship system could implode and, with it,

3 Of course, such clanic and lineage correspondences are not limited to the Kachin. According to Pascal Bouchery (1995: 27) in his study of the Hani, a Yunnanese TibetoBurmese population, some clanic correspondences are also attested with Chinese patronymics. The territorial conquests of the Han Chinese achieved by the end of the fourteenth century are cited by Bouchery as the origin of these correspondences: “The Chinese start to attribute Han patronymics to the local Tusi chiefs they administrated” [. . .] “At present time, [the Han patronyms] coexist with the traditional clanic names, but they are rarely supplanted, so that, most of the time, all patriclans have also Hani names”. And Bouchery also enumerates a list of twelve Hani clanic names aand their respective correpondences with the Chinese patronyms (ibid.: 47).

294

françois robinne

the principle of its exogamous clans. In the wake of the concomitant processes of Burmanisation, of nationalism, and of Christianisation, the tendency to substitute written tradition for oral tradition contributes highly to the obliteration of the collective memory that participates in the perennial construction of the generalised exchange system. After all, clanic correspondences still contribute to give coherency to the Kachin artefact and its predominant interethnic social dynamism. The terminology used to describe the mayu-dama system is signicant of the universality of the system, even among the ethnic groups like the Lisu whose social organisation is based upon an undifferentiated matrimonial system. The vernacular terms used to design the social partners—or kinship categories—bound in the mayu-dama system among the main “Kachin” sub-groups are as follows, without recourse to detailed technical terms that are not in use: Jinghpaw

mayu dama

kahpu kanau

kashu kahkri

Lisu Zaiwa

seqhpa mait mung zumaw (ou zaw mawn) yauhpu zamaw mung zumaw mayu dzemi

aiq nyiza agu aman

avi sza da au

? bik zaw phèdung phèshè

? semaw zeni pheli phela

Lawngwaw Lachid Nung Lungmi

Possible Conclusions and Examples to the Contrary Some possible conclusions drawn from what has been said can be proposed. 1) On the one hand, it appears that the ve main primordial clans of the current Jinghpaw mythology are common to every Kachin sub-group, sharing, with some variants albeit principally terminological ones, the same mythical stock,4 i.e. the Lawngwaw, the Nung Lungmi, the Rawang, the Zaiwa and the Lachid.

4 The variants recorded during eldwork and in recent publications in vernacular languages, mostly in Jinghpaw, show that they do not occur between ethnic categories. As for the clanic and lineage belongings, they interfere mostly at a clanic or individual level.

transethnic social space of clans and lineages

295

2) On the other hand, it appears that the kinship system of generalized exchange based on a hierarchy between wife-taking groups (mayu) and wife-giving groups (dama) prevails over and above any other consideration. According to this nding, the assertion of a common ancestor or not appears not to be decisive, rather than previous political structures or the relatively recent conversion to Christianism. Despite the structural and radical transformations effected by such developments, a new mayu-dama system is developed. 3) As for the Lisu, we know that their social, political and religious organisation, as well as other traditional cultural markers, contribute to distinguishing the Lisu community from the other Kachin subgroups. This demonstrates that the relationships between clans and lineages constitute, from a sociological point of view, a common ritual language between the different Kachin subgroups, justifying use of the label ‘Kachin’ and giving it coherency. 4) Finally, it can be added that the interethnic and transethnic dimensions of the clans and lineages contribute to reinforcing the concept of pluriethnicity, as developed by Leach and Conrad (1989 and 1992), i.e. the uidity and plurality of ethnic belonging that a person may claim throughout life. With regard to these tentative conclusions, two further reections should be made. First, if the interethnic correspondences of the clans and lineages are a reality, at the same time the transethnic assumption needs to be qualied. Numerous clans in every Kachin sub-group do not produce correspondences, and a wife keeps her clanic belonging when she gets married. Notwithstanding that the transethnic dimension of clans and lineages cannot be entirely generalized as a result, the imagined ethno-linguistic frontiers do not inhibit the dynamics of economic, religious and political exchanges, nor do they hinder the dynamics of social exchanges. The clanic correspondences create a community by which integration is facilitated regardless of linguistic variation. The second reection is a return to the notion of ‘common ritual language’. As developed by Leach, this implies on the one hand a multi-ethnic coherency, and yet, on the other hand, it appears to be in contradiction with an historical process. For this reason, therefore, I propose to substitute the notion of ‘common ritual language’ with

296

françois robinne

that of a common federative sphere versus a common operative sphere.5 By common federative sphere, I mean all political, religious, economic and social spheres of reference, more or less combined each other, more or less shared by different ethnic communities in contact. By common operative sphere I denote the process by which a community manipulates or instrumentalizes the federative sphere common to different communities in order to appropriate the social landscape, as is the case, for example, with the Jinghpaw over the Kachin subgroups, or, at a different level, of the Kachin in their relationships with the Burmese central power. In our post-colonial and post-Christian times, these spheres have been shared by numerous ethnic groups of the so-called Kachin stock. The social changes caused by both colonialism and Christianism, and determined as well by the relationship with the Burmese central power, have contributed to accelerating distance between the present and ancestral consensual forms. Supposedly being common to some 90% of the Kachin, Christianism became the new transethnic vector of the Kachin ideal, with the risk, noted by Gravers (1999: 80), of assimilating nationalism and religion. In this light, Christianism could be considered the main ritual language among the Kachin subgroups, but it needs to be considered in its plurality: Catholics versus Protestants, and their numerous components, Baptists, the Church of Christ, the Engel Church, the Assembly of God, the Anglicans and so on, each one dened more or less after ethnic criteria, each one having its own reformists or fundamentalists “sitting outside the mainstream”. Considered in its plurality, like shamanism, Christianism is certainly a common federator to the different subgroups; at the same time, members of the Churches generally claim to be of different religions, and Christianism still needs to overcome its intrinsic contradictions to become a common operator, artefact as such of a new Kachin coherency. We arrive at a paradox among the Kachin. On the one hand, Christianism claims to be the operator of unity, but it fails to have federative support on that religious basis. On the other hand, the mayudama system remains a common federative sphere between the different sub-groups, but it cannot be dened as the vector of the pan-Kachin process. The articulation between the substance of the social federative

5

From the concept of “opérateur commun” as developed by Amselle (1999: 28).

transethnic social space of clans and lineages

297

sphere, common to the Kachin subgroups, and between the operator of unity is still to be found, and only then can we fully understand the sociological depth of the pan-Kachin movement.

POSTSCRIPT

RECONSIDERING THE DYNAMICS OF ETHNICITY THROUGH FOUCAULT’S CONCEPT OF ‘SPACES OF DISPERSION’ François Robinne and Mandy Sadan Introduction: Ethnic Categories within Transethnic Spaces In most studies of ethnicity, basic complementarities are assumed to exist in the formulation of identities, whether one considers ethnic groups individually or in the framework of their interactions with each other. Indeed, the latter context in many ways delimits denitions of the former. In both cases, the notion of ethnic category is the focal point of analysis. The aim of this postscript is to challenge the proposition outlined above, not only by privileging the dynamics of networks rather than cultural differences in an attempt to understand the social coherency of multi-ethnic spaces, but also to oppose such ethnic presuppositions to the notion of “spaces of dispersion”, as suggested by Michel Foucault (1969: 53). Inevitably, one might argue, networks feed back into ethnic categories and their respective cultural markers. In this case, there may be a danger of simply reopening doors already forced ajar by Leach (1954) with his concept of “unstable equilibrium”, by F. K. Lehman (1967)1 with his theory of “ethnic categories as role systems”, and by Yves Conrad (1989) in his study of the Lisu of Northern Thailand and the claims of plural ethnicity. Nonetheless, the effort to address the primacy of ethnic category as a descriptive tool in the analysis of complex social environments is worthwhile. To illustrate how strong the inuence of ethnic presupposition or ethnic determinism may be, the following example may be given. Indeed, this example has been chosen because of the reputation of the author involved: Professor Victor King, justiably one of the most 1 We are very grateful to U Chit Hlaing—Professor F. K. Lehman—for his critical comments about the initial writing of this postscript. Any remaining shortcomings are solely our own responsibility.

300

françois robinne and mandy sadan

respected specialists working on the concept of ethnic identity and on the dynamics of interethnic relationships in Southeast Asia. King, in his work on West Kalimantan, pointed out that “bounded entities are fundamentally mistaken” (2001: 5), yet, at the same time, he also introduced an ethnic category into his analysis. The reader may be reminded briey of the case. In the region of Upper Kapuas, West Kalimantan, “where there has been considerable cultural exchange, intermarriage, religious conversion and assimilation,” Victor King applied the exonym Maloh—after the toponym Embolah—to people whose homogeneity in fact depends on the dynamics of networks. In other words, the idea these people had themselves about their social cohesion transcended the reductionist concept of ethnic category, and thus the necessity of being called by an ethnonym. Even for an anthropologist as eminent as Victor King, the lack of an ethnonym appears to be an intellectual dilemma in need of resolution (King, 2003: 209–212). On one level, a parallel may be drawn in the Burmese situation, since the “fabrication of ethnicity and colonial polity”, (to quote Rujaya Abhakorn (2000)), found its legitimacy in the two Constitutions of 1947 and 1974, and corollary developments during nearly half a century of civil war. Together with the tendency to institutionalize ethnic categories, the question of how to name oneself became particularly signicant for local ‘ethnic’ leaders who claimed and sought geopolitical recognition in the newly independent state. For example, discussions that focused on the choice of ethnonyms such as Kachin or Jinghpaw or Wunpawng, or on the choice between Chin and Zomi, or Palaung and Ta-an, demonstrated the ideological weight carried by the concept of ‘transethnicity’. The notion of ‘transethnicity’ used here may be dened as the dynamic of exchanges in which ethnic belongings are not stricto sensu determinant, but are rather, quoting King and Wilder (2003: 213), a sphere of interests that “cut across ethnic categories”; the transethnic dynamic challenges the notion of ‘multi-ethnicity’ by which social landscapes are deemed, of necessity, to nd coherency. Following Leach (1977: 60), transethnicity may refer to a somewhat arbitrarily dened area in which a social system exists, whatever may be the ethnic diversity of that area. If, according to the academic perspective proposed here, we put aside political ideology, it should be possible to separate ethnic determinism from other models and to substitute the analysis of supposedly discrete entities for the dynamics of networks. However, in so doing we are

postscript

301

forced to question again what kind of social continuity is attached to linguistic categories, in order to begin a process of substituting preexisting, dened categories with corollary ‘tables of differences’ in the analysis of transethnic ‘spaces of dispersion’. Juxtaposition and Superposition: The Limits of the Cartographic Exercise Since Leach (1977: 60) and his theory of a single social system embracing cultural and linguistic differences, researchers on Kachin society agree that not only does the ethnic category Kachin not correspond with an immediate linguistic family, neither is it a homogeneous grouping (Lehman, 1979: 232). Leach’s work is contemporary with that of theorists working on the concept of ethno-linguistics, as discussed in particular in the journal American Anthropologist (Silva-Fuenzalida, 1949; Buswell, 1950; Hymes, 1966). The aim of ethno-linguistic maps is to draw spatially the borders of isolated linguistic families and to establish on that base “tables of differences” (Foucault 1969: 53). These families are then represented as existing adjacent to each other, as if one were looking at stamps stuck on an envelope: a map for Austronesian subgroups, a map for the Austroasiatic subgroups, for the Tibeto-Burmans, for the Miao-Yao, and another one for the Thai-Kadai. Such spatially juxtaposed cartographic representations raise a number of questions: 1) To what extent do linguistic continuities accompany the process of social re-composition over time? For example, how are linguistic continuities expressed in the metamorphoses of Akha social organisation and its process of Christianisation (Kammerer, 1990),2 or in the remarkable persistence of the Jinghpaw mayu-dama system in the face of drastic political and religious changes? 2) To what extent do speakers of this or that language recognize an afnity with other, more distant speakers of languages in the same linguistic family? For example, Lisu people, when living in Kachin State, generally present themselves by modulating their identities according to the complex relationships they have with the Jinghpaw 2

Thanks to Vanina Bouté for this reference.

302

françois robinne and mandy sadan

and other close neighbours, rather than by referencing a relationship with the Lisu of Northern Thailand. 3) To what extent does linguistic diversity constrain the dynamic partnership of neighbours? For example, ethnic categories based on linguistic and cultural markers are exclusive, yet the generalised exchange system of the Jinghpaw, as demonstrated in this volume, is inclusive. In spite of the preference for juxtaposing linguistic maps, it might prove a valuable cartographic exercise to superimpose such maps upon each other; this would surely bring us closer to an appreciation of the linguistic complexity that is typically observed in the eld. The following map is a composite created from three maps that were originally presented separately by Jean Michaud (2000: maps 3, 4 & 5).3 The maps presented successively correspond with, according to Michaud, the “ethno-linguistic” families and their main “branches” attested in the Southeast Asian massif: (A) the ‘Austro-Thaï’ “ethno-linguistic” groups, in which four main branches are distinguished; (B) the Tibeto-Burman “ethno-linguistic” groups, totalling ten main branches; and (C), the MonKhmer “ethno-linguistic” groups, comprising ve branches. However, even if we set aside any questions about possible analogies between clans and ethnic groups and between exonyms and endonyms, which would complicate the cartographic exercise considerably, the complexity put forward even in this example still far from adequately represents the convoluted matrix of socio-linguistic relationships that, in fact, characterizes continental Southeast Asia. For example, recent eldwork in Shan and Kachin States has demonstrated that the occurrence of what we are used to calling ‘inter-marriage’ by reference to the ethnic origin of a couple is highly underestimated, since this criterion is not locally pertinent, or at least is secondary, with regard to the partnership spheres by which different villages are linked. Despite all the reservations about the cartographic exercise of ethnolinguistics expressed above, the main positive result of superimposing maps of linguistic families in this way is, paradoxically, the un-readability of the nal product. The illegible conation creates an apparent

3 In his introduction, Michaud himself presents the different linguistics families of highland Southeast Asia in a map.

Juxtaposition of linguistic families in highland Southeast Asia

postscript 303

304

françois robinne and mandy sadan

imbroglio, but it is precisely this that effects and gives meaning in terms of social coherency ‘on the ground’. From a sociological point of view, relationships are not between ethno-linguistic subgroups, but between villagers and partners, whose exchanges and networks contribute to the establishment of social cohesion, albeit an unstable cohesion, in a multi-ethnic landscape. What Kind of Social Continuity is Attached to a Linguistic Category? The strength and the ambiguity of the word ‘ethno-linguistic’ is that it presupposes a sociological continuity upon a linguistic category. It refers implicitly to the notion of a common sociological stock derived from a linguistic basis. However, it is clear that the linguistic criterion has to t also with the dynamics of economic networks, social relationships and religious interactions, each of which are daily involved in the process of social construction and re-composition. For example, research focused on the Austronesian world cannot be separated from the political contexts in which Taiwan and Australia are incorporated, for different reasons, in a dynamic approach along with Southeast Asia. We may have in mind the thesis developed by James Fox and Clifford Sather (1996), according to which ancestors appear to be a structural referent of social stratication in the Austronesian sphere. At the very least, such a theory, which proposes a common sociological stock on a linguistic basis, needs to be qualied by analysis that focuses on the peripheral points of that linguistic sphere. It would lead us to consider the situation of crossroads that characterize Southeast Asia, and the partnerships between neighbouring populations, whatever may be here and there the linguistic diversity. The French Anthropologist Jacques Dournes (1977) has demonstrated that the pötao sphere of inuence is inclusive, and its ramications go much farther than the Jörai category alone. Similar examples could be multiplied many times over using ethnographic sources to demonstrate, if necessary, the limits of linguistic continuities in terms of the identity process. Despite, or perhaps because of, the diversity of populations, it is necessary that social coherency be produced in such settings. Two examples may be mentioned here: (1) Kachin State, where the general exchange system as reinterpreted by the framework of Christianity appears to be a kind of ‘ritual language’, to use an expression taken

postscript

305

from Leach (see Gros and Robinne, this volume); (2) the Southern Shan State, where Buddhism and the undifferentiated kinship system appear to be common operators (see Robinne, this volume), through which the social landscape nds its coherency and on which is based the leadership process in which neigbouring communities are engaged. These two examples are detailed below: (1) In the case of the so-called ‘Kachin’, as described earlier in this volume, the process of Christianisation has contributed greatly to a re-imagining of the generalised exchange system. Even today, although the religious dimension appears to be pre-eminent in ‘Kachin’ identity over the mayu-dama relationship stricto sensu, the social dynamic generated by the cyclical exchange of women and of ritual goods remains in operation, and has been an important factor in the pan-Kachin process. This is because, at a basic level, the system of generalised exchange is inclusive and goes far beyond just the Jinghpaw, and even far beyond the Kachin. Indeed, in this way, the system contributes to introducing a transethnic dimension by which one can begin to understand how it is that social organisation appears to be resistant to both political and religious ruptures (Robinne 2007 and in this volume). (2) In comparison with the Kachin case, the populations settled in the Nyaung Shwe Township of the Southern Shan State have in common an undifferentiated kinship system, an acephalous political system and practise Buddhism, albeit with ritual variants; as with the Kachin subgroups, the partnership dynamic goes beyond the level of ethnic category. The coherence of the local social landscape has been strengthened by the evolution of a ritual in which ve Buddha images are circumnavigated around Inle Lake and, concomitantly, an economic cycle has developed based on a system of ve-day circulating markets. The interaction of these two religious and economic spheres not only contributes to the movement of people and of goods, it also constitutes political supremacy over the social landscape for those who control its articulation, at least symbolically. Having succeeded the Shan sawbwa administration and then the Intha symbolic and economic supremacy in the region, the Pa-O seem to have been engaged in the appropriation of the social landscape since the cease-re with the Burmese military in 1994. Although they actually belong to a common linguistic family with the Karen, the Pa-O do not imagine themselves as Karen and, indeed, clearly reject such an association, having no interest in the supposed linguistic afnities. Whilst they still have regular connections with

306

françois robinne and mandy sadan

the rice growing Taungthu in the south of the State, with whom they share a common origin, the Pa-O also have some cultural markers in common with the Shan, such as dress and titles, the latter of which are included in their kinship terminology.4 They imagine themselves to be partners of both their immediate neighbours (Shan and Taungthu), as well as being in a kind of ‘patron/client’ relationship with the Burmese central power. When relationships are considered, they are generally reduced to the politics of the central government towards the populations of the periphery, giving justication to the category ‘minority’ and reducing relationships to an opposition between ‘polis’ and ‘ethnos’ as dened by the Ancients (Brown, 1994). As decisive as it may be, that dichotomous, vertical and external interpretation constitutes just a facet of the Other. Besides the omnipresent level of analysis between lowlands and highlands, to which interethnic relationships are generally reduced, a second level of interpretation, consisting of analysis of local networks, is marginalised. Locally, however, the dynamics of political, social, religious and economic relationships are constitutive parts of the process of social coherency and identity. We may consider here again the Nyaung Shwe District, where the pagoda festivals, as well as novitiation ceremonies, implicate villagers irrespective of whether they are settled on the lake or on the mountains. The same point could be made concerning the fabrication of cheroots, for which the raw materials depend on the Pa-O to provide the outside leaves and on the Burmese for the tobacco; also, for the blacksmiths or carpenters who may have to climb the mountains from time to time to repair agricultural tools or to build large monasteries; again for the transhumance of water buffaloes whose owners are mountain dwellers and on which the lowland populations depend two times a year for ploughing activities, both for rice and for sugar cane. In every case considered here, the partnerships develop inclusive dynamics transcending the ethnic criterion. From this perspective, ethnicity becomes just a cultural marker in addition to, and just like, others. Locally, what predominates daily is less the ethnic referent than the partnership.

4 F. K. Lehman (personal communication) commented on a rst version of this article: “The Pa-O State was in every way a political organization of a Shan-type State, and recognised as such by other Shan States”.

postscript

307

Conclusion: Tables of Difference Versus Spaces of Dispersion It would be interesting and stimulating to focus both eldwork and analysis on transethnic spaces. It would entail forgetting for a while the framework of national connections and the spatial limits they produce. It would lead, however, to more fruitful observations on social intercourse, economic exchange, and religious networks, for example, by which villages and other communities situate themselves in dynamic partnerships. From that perspective, the component of ‘ethnicity’ would no longer be considered a kind of superior artefact covering other cultural markers; it would become a cultural marker amongst others. This does not entail denying the signicance of ethnic categories when they are claimed by the people themselves; it requires just that they be introduced when they appear relevant and that we should not assume them to be an a priori determinant of identity and social coherency. Such an intellectual endeavour may appear most improbable to those engaged in eldwork. However, such a response follows the logic denounced by Michael Aung Thwin (1998) in his critical reading of the reinvention of myths and legends by colonial historians “caught in their own intellectual and political words”. From an epistemological perspective, it is quite a challenge to postulate that ethnic entities have not only to be considered in terms of multiculturalism and transnationalism, as well as terms of ‘interity’, a neologism created by Jacques Demorgon (2005: 3) to contrast with ‘alterity’.5 Despite the new perspectives to which the theoretical debate on social dynamics leads, each one of these concepts harks back to the relationship between pre-existing, dened categories. Yet, all of these dynamics evidence the fact that ethnicity may be considered the result of a basically transethnic process emerging from a specic context, the great crossroads which is the Indo-China massif. As a result, we may more generally characterise and understand Southeast Asia as being a ‘space of dispersion’ rather than in terms of its continuity. Following Michel Foucault in his Archéologie du savoir (1969: 19, 49, 52–53, 88), we assume that, in spite of drawing up tables of differences

5 « L’interculturel factuel » met en pleine lumière que le culturel et le stratégique ne sont pas séparables et concernent tous les domaines d’interactivité humaine. L’occultation de cette intérité, pourtant fondatrice de l’histoire, tient à ce que les humains ont choisi de situer la centralité ailleurs, dans le choc des identités au cœur de l’altérité, comme s’il n’y avait pas toujours eu « intérité ».

308

françois robinne and mandy sadan

(tables de différences) around a centre that would be, in the circumstances, the ethnic categories, and, furthermore, in spite of considering interethnic relationships, such an approach would consist of describing a space of dispersion (“espace de dispersion”, p. 19), as well as a system of dispersion (“système de dispersion”, p. 53). This system would be the networks alluded to in this postscript. Original French version:

Robinne translation:

Une telle analyse n’essaierait pas d’isoler, pour en décrire la structure interne, des îlots de cohérence ; elle ne se donnerait pas pour tâche de soupçonner et de porter en pleine lumière les conits latents ; elle étudieraitdes formes de répartition. Ou encore : au lieu de reconstituer des chaînes d’inférence (comme on le fait souvent dans l’histoire des sciences ou de la philosophie), au lieu d’établir des tables de différences (comme le font les linguistes), elle décrirait des systèmes de dispersion.

Such an analysis would not try to isolate, in order to describe the inner structure, some islets of coherence; the aim would not be to suspect and to bring to light the latent conicts; it would consist of studying some forms of distribution. Or again: instead of reconstituting chains of inference (as we are used to do in the history of sciences or of philosophy), instead of establishing tables of differences (like the linguists do), it would describe systems of dispersion.

[ Foucault, 1969 : 52–53, italiques de l’auteur]

As structural components of the production of the social, these networks between villages, rather than between ethnic categories, are consubstantial to social unstable equilibrium, as demonstrated by Leach. Furthermore, these networks also participate in the hierarchical process by which people/communities are engaged in the appropriation of a social landscape. From this perspective, ethnic categories do, indeed, eventually have to be reintroduced, but only once the networks and their modes of articulation have been identied.

BIBLIOGRAPHY Abhakorn, Rujaya, 2000, “The fabrication of Ethnicity and Colonial Polity East of the Salween”, Proceedings of the Myanmar Two Millenia Conference. Yangon, Universities Historical Research Centre: 86–200. Akin, David and Joel Robbins (eds.), 1999, Money and Modernity: State and Local Currencies in Melanesia. Pittsburgh, University of Pittsburgh Press. Alexandre, Jean-Louis et Eberhardt, Nicolas, 1998, Des systèmes agraires de la rive gauche de la Nam Ou. Vannes, Comité de Coopération pour le Laos. Amselle, Jean-Loup, [1990] 1999, Logiques métisses. Anthropologie de l’identité en Afrique et ailleurs. Paris, Payot. Anderson, John, 1871, A Report on the Expedition to Western Yunnan via Bhamo. Calcutta, Ofce of the Superintendent of Government Printing. Anderson, B., 1991 (revised), Imagined Communities. London, New York, Verso. Ajmer, Goran, 1979, “Reconciling power with authority: an aspect of statecraft in traditional Laos”, Man, vol. 14 New Series: 734–749. Archaimbault, Charles, 1973, Structures religieuses lao. Vientiane, Vithagna. Aung-Thwin, Michael, 1998, Myth and History in the Historiographies of Early Burma. Paradigms, Primary Sources, and Prejudices. Athens, Ohio University Center for International Studies, Monograph in International Studies, Southeast Asia Series. Austin, John, 1962, How to Do Things With Words. Oxford: Clarendon Press. Asad, T. (ed.), 1973, “Introduction”, in Anthropology and the Colonial Encounter. New York, Humanity Books: 9–19. Atwill, David, 2003, “Blinkered Visions: Islamic Identity, Hui Ethnicity, and the Panthay Rebellion in Southwest China, 1856–1873”, The Journal of Asian Studies 62(4): 1079–1110. Ayme Georges, 1930, Monographie du V e territoire militaire. Hanoi, Imprimerie d’ExtrêmeOrient. Bacot, J., 1912, Le Tibet révolté, Vers Népémako, la terre promise des Tibétains suivi des Impressions d’un Tibétain en France. Paris: Hachette. Bailey, F. M., 1912, “Journey through a Portion of South-Eastern Tibet and the Mishmi Hills”, Geographical Journal, Vol. XXXIX: 334–471. ——, 1945, China-Tibet-Assam: A journey, 1911. London, Jonathan Cape. Baird, Ian & Shoemaker, Bruce, 2005, Aiding or Abetting: Internal Resettlement and International Aid Agencies in the Lao PDR, Probe International, téléchargeable sur www.probeinternational.org Baird-Murray, Maureen, 1997, A World Overturned: a Burmese childhood, 1933–47. London, Constable. Barth, Fredrik, 1969, “Introduction”, in F. Barth (ed.): Ethnic groups and boundaries. Bergen/Oslo/London: 9–38. Barua, G. C. Ahom Buranji, 1985 [1930], From the Earliest Time to the End of Ahom Rule. Guwahati, Spectrum. Baruah, SL. A., 1985, Comprehensive History of Assam. Delhi, Manoharlal. Bell, C., 1997, Ritual. Perspective and Dimensions. New York-Oxford, Oxford University Press. Bello, David, 2001, Opium and the Limits of Empire: The Opium Problem in the Chinese Interior, 1729–1850. PhD dissertation, University of Southern California. ——, 2003, “The Venomous Course of Southwestern Opium; Qing Prohibition in Yunnan, Sichuan, and Guizhou in the Early Nineteenth Century”, The Journal of Asian Studies 62(4): 1109–1142.

310

bibliography

——, 2005, Opium and the Limits of Empire: the Opium problem in the Chinese interior, 1729 – 1850. Cambridge, Harvard University Press, Harvard East Asian Monographs. Bernard, Auguste, 1990, “Les gisements miniers au Laos”, Péninsule 20: 3–98. Bernot, Lucien, 1967, Les Cak: Contribution à l’étude ethnographique d’une population de langue loi. Paris: CNRS. ——, 1982, “Les ‘nous’ et les ‘ils’. Contribution à l’étude des relations inter-ethniques”, Ethnies. [Paris/La Haye]: 277–292. (Reprint in Bernot Lucien, 2000, Voyage dans les sciences humaines. Qui sont les autres?, Paris, Presses de l’Université de Paris-Sorbonne, CREOPS: 311–324). Bhattacharjee, J., 1989, “Sequences in development in the Barak Valley”, in J. B. Bhattacharjee (ed.): Sequences in development in North East India (a study of tradition, continuity and change). New Delhi, Omsons. Bhattacharjee, T., Sociology of the Karbis, Delhi, B.R. Pub., 1986. Bhuyan, S. K. (ed.), 1990 [1933], Tungkhungia Buranji or the History of Assam (1683– 1826 AD). Guwahati, Dept. of History and Antiquarian Studies. ——. (ed.), 1962 [1933], Deodhai Assam Buranji. Gauhati, Department of Historical and Antiquarian Studies. Bloch, Maurice and Jonathan Parry, 1989, “Introduction: Money and the Morality of Exchange”, in Jonathan Parry and Maurice Bloch (eds.): Money and the Morality of Exchange. Cambridge, Cambridge University Press: 1–32. Bohannan, Paul, 1955, “Some Principles of exchange and Investment among the Tiv”, American Anthropologist 57 (1): 60–70. ——, 1959, “The Impact of Money on an African Subsistence Economy”, The Journal of Economic History 19 (4): 491–503. Borooah, Romy, 1975–77, The Wancho of North-East India: An Ethnographic Account. Unpublished M.A. paper, University of Illinois at Urbana-Champaign. ——, 1989, The Wancho of Northeast India in the 1970s and 1980s: Individuals and Structures in a Study of Change. Urbana, University of Illinois, Unpublished PhD Dissertation. Bouchery, Pascal, 1988, “Les systèmes politiques naga” Journal Asiatique 276, 3–4: 285–334. ——, 1995, Les Hani: Introduction à l’étude d’une population tibéto-birmane du Yunnan en relation avec la Chine. Paris, PhD thesis, Université de Paris X. Bower, U. G., 1951, Naga Path. London, John Murray. Brown, R., 1874, Statistical Account of the Native State of Manipur and the Hills Territory under its Rule. Calcutta (ed. Cf Bouchery). Brown, David, 1994, The state and ethnic politics in South-East Asia. London and New York, Routledge. Buswell, James O., 1950, “Some Comments on Hockett’s Language ‘and’ culture: A Protest”, in American Anthropologist, Vol. 52, 2, pp. 284–286. Butler, Capt. J., 1847, A Sketch of Assam with some Account of the Hills Tribes by an Ofcer. London (ed. Cf Bouchery). Cameron, A. A., 1912, “A Note of the Palaungs of the Kodaung Hill Tracts of Momeik State” Census of India 1911, vol. 9, pt. 1 app. A, pp. I–XIII. Caply, Michel, 1966, Guérilla au Laos. Paris, Presse de la Cité. Carrapiet, J. S., 1929, The Kachin Tribes of Burma. Rangoon, Superintendent Government Printing and State. Chambers, I. and L. Curti (eds.), 1996, The Post-Colonial Question. Common Skies, Divided Horizons. London, Routledge. Chen Bisheng 陳碧笙, 1941, Dianbian sanyi 滇邊散憶 (Remembrance of the Tribal Life along the Border in Yunnan). Taipei: 中國民俗學會(The Chinese Association for Folklore). Cheyrou-Lagrèze, Henri, 1921, Esquisses laotiennes. Rochefort, imprimerie Norbertine. Chit Hlaing, 2000, “Cultural Models (and Schemata) and Generative Knowledge Domains and How they are Related”. Paper presented at the 2000 Annual Meeting of the American Anthropological Association, panel on Cultural Models [to appear].

bibliography

311

——, 2001, “Aspects of a Formalist Theory of Kinship: The Functional Basis of its Genealogical Roots and Some Extensions in Generalised Alliance Theory”. Anthropological Theory 1, 2: 212–239 [special issue edited by D. B. Kronenfeld]. ——, 2004, “On the ‘Globality Hypothesis’ about Social/Cultural Structure: An Algebraic Solution”. Mathematical Anthropology panel, Proceedings of the European Meetings on Cybernetic Systems Research (EMCSR) Vienna. Revised and expanded in press, Cybernetics and Systems: An International Journal, special issue edited by Dwight Read. Clifford, J. and G. E. Marcus (eds.), 1986, Writing Culture. The Poetics and Politics of Ethnography. Berkeley, California. Collis, Maurice, 1945, Trials in Burma. London, Faber & Faber. Condominas, Georges, 1957, Nous avons mangé la forêt. Paris, Mercure de France [1969 Wir aßen den Wald des Geistersteins Gôo. Hi saa brii Mau-Yaang Gôo. Chronik des MnongGar-Dorfes Sar Luk. Frankfurt a.M.: Insel]. Condominas, G., 1980 [1976], L’espace social à propos de l’Asie du Sud-Est. Paris, Flammarion. ——, 1980 [1976], « Essai sur l’évolution des systèmes politiques thaïs », in L’espace social à propos de l’Asie du Sud-Est. Paris, Flammarion: 259–316. Conrad, Yves, 1989, “Lisu identity in Northern Thailand: a problematique for anthropology”, in J. McKinnon & B. Vienne (eds.), Hill Tribes Today. Bangkok-Paris, White Lotus-ORSTOM: 191–221. ——, 1992, “Identités et pratiques spatiales des Lisu du nord de la Thailande”, in J. Matras-guin & C. Taillard (éds.): Habitations et habitats d’Asie du Sud-Est continentale. Pratiques et représentations de l’espace. Paris, L’Harmattan: 19–53. Cooler, Richard M., 1995. The Karen bronze drums of Burma: Types, iconography, manufacture and use. Leiden (e.a.): Brill. Cooper, Robert G., 1978, “Poly-ethnic systems analysis: the relevance of Leach’s Political Systems to the ethnography of Northern Thailand and Laos” Journal of Sociology and Psychology 1: 56–64. Corlin, Claes, 1994, Common Roots and Present Inequality. Ethnics Myths among Highland Populations of Mainland Southeast Asia. Copenhagen, Denmark, Nordic Institute for Asian Studies, Report n° 17. Dalton, E.T., 1872, Descriptive Ethnology of Bengal. Calcutta, Ofce of the Superintendent of Government Print. Damon, Frederick H., 1993, “Representation and Experience in Kula and Western Exchange Spheres”, Research in Economic Anthropology 14: 235–254. Damrong, Tayanin, 1994, Being Kammu. New York, Cornell South East Asia Program. Danda, D. G., 1978, Among the Dimasa of Assam: an Ethnographic Study. New Delhi, Sterling. Dawson, G. W. compiled, 1912, The Bhamo District. Rangoon: Ofce of the Superintendent. Dean, B. and J. M. Levi (eds.), 2003, At The Risk of Being Heard. Identity, Indigenous Rights, and Postcolonial States. Ann Arbor, Michigan. De Heusch, Luc, 1971, Pourquoi l’épouser ? Et autres essais, Paris, Gallimard. Dell, Elizabeth, 2000 a, “Mapping Burma: the James Henry Green Collection of Photographs”, in E. Dell (ed.): Burma: Frontier Photographs 1918–1935. London, Merrell: 9–28. —— (ed.), 2000 b, Burma: Frontier Photographs 1918–1935. London, Merrell. Demorgon, Jacques, 2005, Critique de l’interculturel. L’horizon de la sociologie. Paris, Economica Anthropos. Deuve, Jean, 1984, Le royaume du Laos: histoire événementielle de l’indépendance à la guerre américaine. Paris, EFEO. Donohue, John J., 1984, “Ideology and Economy in Upland Southeast Asia: A Reappraisal of Political Systems of Highland Burma” Ethnos, vol. 49, n° 1–2: 62–79. Dournes, Jacques, 1977, Pötao: une théorie de pouvoir chez les Indochinois Jörai. Paris, Flammarion.

312

bibliography

Dumont, Louis, 1966, Homo hierarchicus. Le système des castes et ses implications. Paris, Gallimard [1980, Homo hierarchicus: the caste system and its implications. Chicago/London: Chicago University Press]. Durrenberger, E. P., 1989, “Lisu Ritual, Economics and Ideology”, in Susan Russell (ed.): Ritual, Power, and Economy: Upland-Lowland Contrasts in Mainland Southeast Asia. Dekalb: Northern Illinois University Center for Southeast Asian Studies, Monograph Series on Southeast Asia, Occasional Paper n° 14: 103–120. Dutta, P., 1969, The Tangsas of the Namchik Valley. Shillong, North-East Frontier Agency. ——, 1978, The Noctes, Shillong, Government of Arunachal Pradesh. Edgar, J. H., 1933–34, “Pygmies on the Salwin,” Journal of the West China Border Research Society, n° 6: 52–53. Eller, J. D., 1999, From Culture to Ethnicity to Conict. An Anthropological Perspective on International Ethnic Conict. Ann Arbor, Michigan. Elwin, V., 1959, Nagaland. Shillong, Advisors Secretariat. ——, 1969, The Nagas in the Nineteenth Century. London, Oxford University Press. Enriquez, C. M., 1920 [1933], Races of Burma: A Note on the Races and Recruiting Areas of Burma. Rangoon: Ofce of the Superintendent, Government Printing. ——, 1922, A Burmese wonderland: a tale of travel in Lower and Upper Burma, Calcutta: Thacker, Spink & Co. ——, 1923, A Burmese arcady: an account of a long & intimate sojourn amongst the mountain dwellers of the Burmese hinterland & of their engaging characteristics and customs, London: Seeley, Service & Co. ——, 2003, In Quest of Greatness, CC Naw Ja ed. Evans, Grant, 1998, The politics of ritual and remembrance: Laos since 1975. Chiang Mai: Silkworm Press. ——, (ed.), 1999, Laos. Culture and Society. Chiang Mai, Silkworms Books. ——, 2000, “Tai-Ization: Ethnic Change in Northern Indo-China” in A. Turton (ed.): Civility and Savagery. Richmond, Curzon Press. ——, 2002, A short history of Laos: the Land in between. Chiang Mai, Silkworm Books. Evrard, Olivier, 1997, “New villages of Luang Nam Tha province”, in Yves Goudineau (ed.): Resettlement and social characteristics of new villages: basic needs for resettled communities in the Lao PDR. Vientiane, Unesco-Pnud-Orstom: 5–46. ——, 2001, Émergence de la question foncière et relations interethniques au Nord Laos: mobilités, rapports à la terre et organisation sociale dans quelques villages taï et khmou de la vallée de la Nam Tha. PhD dissertation, University of Paris I-La Sorbonne. ——, 2002, “Politiques de sédentarisation et relations interethniques: les termes de la question foncière au Laos” Cahiers d’Anthropologie du Droit, 1: 189–236. ——, 2003, “Relations interethniques et solidarités intra ethniques: sens et actualité des sous-groupes khmou au Nord-Laos” Aséanie 11: 39–72. ——, 2006, Chroniques des Cendres: Anthropologie des populations khmou et des relations interethniques au Nord-Laos, éditions de l’IRD (in press). —— & Goudineau, Yves, 2004, “Planned resettlements, unexpected migrations and cultural trauma”, Development and Change, 35/5: 937–962. Fellowes-Gorden, Ian, 1957, Amiable Assassins: the Story of the Kachin Guerrillas of North Burma. London, Robert Hale Limited. Firth, Raymond, 1961, Elements of Social Organisation. Boston: Beacon Press. ——, 1973, Symbols Public and Private. Ithaca, NY: Cornell University Press. Fisher, W. F., 2001, Fluid Boundaries. Forming and Transforming Identity in Nepal. New York, Columbia. Fiskesjoe, Nils Magnus Geir, 2000, The Fate of Sacrice and the Making of Wa History. Unpublished PhD dissertation, The University of Chicago. Forster, P. A., “Review of the New Left Critique of Social Anthropology”, in T. Asad: Anthropology and the Colonial Encounter. New York, Humanity Books: 23–38.

bibliography

313

Foucault, Michel, 1969, L’archéologie du savoir. Paris, Editions Gallimard; [1972, The Archaeology of Knowledge, trans. S. Sheridan, London: Tavistock Publications]. Fox, J. (ed.), 1988, To speak in pairs: essays on the ritual languages of eastern Indonesia. Cambridge, Cambridge University Press. Fox, James J. and Clifford Sather (eds.), 1996, Origins, Ancestry and Alliance: Exploration in Austronesian Ethnography. Canberra, Australian National University, Department of Anthropology, Research School of Pacic and Asian Studies. Friedman, Jonathan, 1987, “Generalized Exchange: Theocracy and the Opium Trade”, Critique of Anthropology7 (1): 15–31. ——, 1998 [1979], System, Structure and Contradiction: The Evolution of “Asiatic” Social Formations. Copenhaguen, Nationalmuset [2nd ed.: Walnut Creek, CA, AltaMira Press]. Fürer-Haimendorf, C. von, 1939, The Naked Nagas. London, Methuen & Co Ltd. ——, 1967, Morals and Merit. A Study of Social Values an Social Controls in South Asian Societies. Chicago, The University of Chicago Press. ——, 1969, The Konyak Nagas. An Indian Frontier Tribe. Chicago, Holt, Reinehart & Winston. —— & Mills, J. P., 1936, “The Sacred Founder’s Kin Among the Eastern Angami Nagas” Anthropos XXXI: 922–933. Hymes, Dell, 1966, “On ‘Anthropological Linguistics’ and Congeners”, in American Anthropologist, Vol. 68, 1, pp. 143–153. Gait, E. A., 1926 [1906], A History of Assam. Calcutta, Thacker, Spink & Co. Galey, J.-C., 1980, “Le créancier, le roi, la mort. Essai sur les relations de dépendance au Tehri-Garhwal (Himalaya indien)”, Purusartha. La dette, n° 4: pp. 93–163. Garnier, Francis, 1885, Voyage d’exploration en Indochine. Paris, Hachette et Cie. Geertz, C., 1980, Negara: The Theatre State in Nineteenth-Century. Bali, Princeton. Gellner, D. N., 1999, “Religion, politics, and ritual. Remarks on Geertz and Bloch,” Social Anthropology, vol. 7, n° 2: 135–153. Giddens, A., 1996, In Defence of Sociology. Cambridge, Polity Press. Giersch, C. Pat, 2001, “A ‘Motley Throng’: Social Change on Southwest China’s Early Modern Frontier, 1700 –1880”, The Journal of Asian Studies 60 (1): 67–94. Gilhodes, Ch. [Father], [1922, The Kachins. Religion and Customs. Calcutta, Catholic Orphan Press]; 1996, Bangkok, White Lotus. Gohain, B. K., 1993, The Hill Lalungs. Guwahati, Anundoram Borooah Inst. Gong, Peihua, 1988, Jingpo zu shanguan zhi shehui yanjiu 景頗族山官制社會研究 (Social research of the mountain ofcial institution of the Jingpo nationality). Canton: Zhongshan University Press. Goré, F., 1923, “Notes sur les Marches Tibétaines du Sseu-T’chouan et du Yunnan,”Bulletin de l’École française d’Extrême-Orient, vol. 23: 319–399. ——, 1992 [1939], 30 ans aux portes du Tibet interdit. Paris, éd. Kimé. Goudineau, Yves (ed.), 1997, Resettlement and social characteristics of new villages: basic needs for resettled communities in the Lao PDR. Vientiane, Unesco-PnudOrstom, 2 vols. ——, 2000, “Ethnicité et déterritorialisation dans la péninsule indochinoise: considérations à partir du Laos” Autrepart 14: 17–31. ——, 2000, “Tambours de bronze et circumambulations cérémonielles: notes à partir d’un rituel kantou (Chaine annamitique)” Bulletin de l’Ecole Francaise d’Extreme Orient 87–2: 553–578. Graeber, David, 1996, “Beads and Money: Notes Toward a Theory of Wealth and Power”, American Ethnologist 23: 4–24. Granet, Marcel, 1939, Catégories matrimoniales et relations de proximité dans la Chine ancienne. Paris, Librairie Félix Alcan, 254p. Green, J. H., 1934, The tribes of Upper-Burma North of 24° latitude and their classication. Cambridge, Haddon Library.

314

bibliography

Gregory, C., 1982, Gifts and Commodities. London: Academic Press. Grierson, G. A., 1927, Linguistic Survey of India. Delhi, Varanasi, Patna, Motilal Banarsidass, vol. I, Part I, xvi + 517p.; Vol III (II), Specimens of the Bodo, Naga, and the Kachin Groups: 499–528. Gros, Stéphane, 1996, “Terres de conns, terres de colonisation. Essai sur les Marches sino-tibétaines du Yunnan à travers l’implantation de la Mission du Tibet,”Péninsule, tome 33, n° 2: 147–211. ——, 2001, “Ritual and Politics: Missionaries Encounters with Local Culture in Northwest Yunnan,” paper presented at the Association for Asian Studies Annual Meeting. ——, 2004, “The Politics of Names: The Identication of the Dulong (Drung) of Northwest Yunnan,” China Information, vol. 18, n° 2: 275–302. ——, 2005 a, “Le nouveau partage du monde. Pauvreté et dépendance au nord-ouest du Yunnan (Chine),” Moussons, vol. 8: 61–88. ——, 2005 b, La part manquante: échanges et pouvoirs dans les conns du Yunnan (Chine). Ethnologie des Drung dans leurs relations à leurs voisins, Unpublished PhD dissertation, Université de Paris X-Nanterre (France). Guillemet, E. et O’Kelly, K., 1916–1917, “En colonne dans le Haut-Laos” Revue Indochinoise, 1916: 9/10 (pp. 169–228), 11–12 (329–405); 1917: 1/2 (9–86), 3/4 (187–276). Halpern, Joel, 1964 a, Government, Politics and Social Structure in Laos. A study of tradition and innovation. Yale University, SEA Studies. ——, 1964 b, Economy and society of Laos: a brief survey, Monograph Series n° 5, Southeast Asia Studies, Yale University, New Haven. Hannay, Simon Fraser [Captain], 1847, Sketch of the Singphos or the Kakhyens of Burmah: the position of this tribe as regards Baumo, and in the inland trade of the Valley of the Irrawaddy with Yuman and their connection with the north-eastern frontier of Assam. Calcutta, W. Ridsdale, Military Orphan Press. Hanson, Rev. Ola, 1954 [1906], A Dictionary of the Kachin Language. Rangoon: Baptist Board of Publications. Hartwig, W., 1970, Wirtschaft und Gesellschaftsstruktur der Naga in der zweiten Hälfte des 19 und zu Beginn des 20 Jahrhunderts. Berlin, Akademie Verlag. Harvard-Duclos, Bernard, 1959, Etude socio-économique de la province de Phongsaly, unpublished document, 19p. Heger, Franz, 1902, Alte Metalltrommeln aus Südost-Asien. Leipzig, K. W. Hiersemann. Hertz, H. F., [1902] 1954, A practical handbook of the Kachin or Chingpaw language. Rangoon, Supdt. Govt. Parinting and Stationery. Hickey, Gerald C., (1982), Sons of the Mountains: Ethnohistory of the Vietnamese Central Highlands to 1954. New Haven, CT: Yale University Press. Hill, Ann Maxwell, 1998, Merchants and Migrants: Ethnicity and Trade among Yunnanese Chinese in Southeast Asia. New Haven, Yale University Southeast Asia Studies. Horam, M., 1977, Social and Cultural Life of the Nagas (The Tangkhul Nagas). Delhi, B. R. Publishing Corp. Hoshino Tatsuo, 1986, Pour une histoire médiévale du moyen Mékong. Bangkok, Duang Kamol. Ho Ts’ui-p’ing (He Cuiping), 何翠萍, 1997, Exchange, Person and Hierarchy: Rethinking the Kachin. Ph.D. dissertation. Charlottesville, Va: University of Virginia. ——, 1999, Shengming, jijie he buxiushehui de jianli: lun Jingpo, Zaiwa shijian de jiangou yu jiazhi 生命、季節和不朽社會的建立:論景頗、載瓦時間的建構與 價值 (Life, season, and immortal society: a discussion of time and value among the Jingpo and Zaiwa). In Shijian, lishi yu jiyi 時間、歷史與記憶 (Time, History and Memory). Huang Ying-kuei, ed. Pp. 157–228. Nankang: Institute of Ethnology, Academia Sinica. ——, (He Cuiping), 2000, Jingpo yangmaoqyun de gushi 景頗羊毛裙的故事 (The story of Jingpo hand-woven wool skirts). Paper presented at “Rethinking Communities

bibliography

315

Studies” conference. Institute of Ethnology, Academia Sinica, Taipei, Taiwan, Feburary 2–4, 2000. Ho Ts’ui-p’ing (He Cuiping), 2004, Cong Jingpo renti tuxiang tan ren yu wu di guangxi 從景頗人體圖像談人與物的關係 (The relationship between persons and objects in terms of Jingpo Human gurines). In Wu yu wuzhi wen hua 物與物質 文化 (Material and material culture). Huang Ying-kuei, ed. Pp. 261–333. Nankang: Institute of Ethnology, Academia Sinica. Howard, Michael and Wattana Wattanapun, 2001, The Palaung in Northern Thailand. Bangkok, Silkworm Books. Htingbai Naw Awn, 1956, Handbook for Pawng Yawng National (Privately printed, Taunggyi). Huang Chia-mo 黃嘉謨, 1976, Dianxi huimin zhengqyuan de lian ying waijiao滇西 回民政權的聯英外交 (一八六八-一八七四) (British Relations with the Panthy Regime of Western Yunnan [1868–1874]). Nankang: Institute of Modern Chinese History, Academia Sinica. Hugh-Jones, Stephen, 1989–90, “Edmund Leach 1910 –1989”, in Cambridge Anthropology: special issue on Sir Edmund Leach, vol. 13, no. 3. Hugh-Jones, S. & Laidlaw, J., 2000, The essential Edmund Leach. New Haven & London, Yale University Press, 2 vols. Hunter, W., 1879, A Statistical Account of Assam. Delhi, B.R. Pub., 2 vols. Hutton, J. H., 1921 a, The Angami Nagas, with some Notes on the Neighbouring Tribes. London, Macmillan. ——, 1921 b [2nd ed. 1969], The Sema Nagas. London, Macmillan. ——, 1965, “The Mixed Culture of the Nagas Tribes” Journal of the Royal Anthropological Institute, vol. 95, part I: 16–43. Izikowitz, Karl Gustav, 1951 [2001], Lamet: Hill Peasants in French Indochina. Göteborg, AMS Press. ——, 1962, “Notes about the Tai”, Bulletin of Far Eastern Antiquities 34: 73–91. ——, 1963, “Expansion” Folk 5: 173–185. Jackson, A., 1979, Na-khi Religion, An analytical appraisal of the Na-Khi ritual Texts. La Hague, Mouton [Religion and Society 8]. Jacobs, J., 1991, Les Naga, société et culture, translated from English by Helen Loveday. Genève, Éditions Olizane. Jiang Ying-liang 江應樑 and Jiang Xiao-lin 江曉林, 2003, Dianxi tusiqu zhuzu tushuo 滇西土司區諸族圖說 (Photographs of Ethnic Groups under Tusi Ruling in Western Yunnan). Dehong: Dehong minzu chubanshe. Jinghpaw Wunpawng Amyu Sha Ni A Ginru Ginsa Yu Hkrat Wa Ai Lam Ningpawt Ninghpang Labau Kadun. Myitkyina, Jinghpaw Wunpawng Htunghking hte Laili Laika Hpung Gingaw, 28p. dactylographiées (rst draft, in Jinghpaw, 2001). Jönsson, Hjorleifur, 1998, “Dead Headmen: History and Communities in the Southeast Asian Hinterland”, in Trankell I. & Summers L. (eds.): Facets of Power and its Limitations: Political Culture in Southeast Asia. Uppsala, Sweden, Uppsala University: 191–212. ——, 2001, “Does the House Hold? History and the Shape of Mien (Yao) Society”, Ethnohistory 48 (4): 613–654. ——, 2006, Mien Relations. Ithaca, NY: Cornell University Press. Kammerer, Cornella Ann, 1990, “Transformations in kinship among Akha (Hani) christians of Highland Northern Thailand” in Proceedings of the 4th International Conference on Thai Studies. Kunming, Institute of Southeast Asian Studies, vol. 1: 330–338. Kammerer, C. A. et Tannenbaum, N. (eds.), 1996, Merit and Blessing in Mainland Southeast Asia in Comparative Perspective. New Haven, CT: Yale University Southeast Asia Studies, Monograph Series 45. Kawlu Ma Nawng, 1942, The History of the Kachins of the Hukawng Valley. (translation and notes by J. L. Leyden). Bombay, privately printed. Keyes, Charles F., 1979, “The Karen in Thai History and the History of the Karen in

316

bibliography

Thailand”, in Charles F. Keyes (ed.): Ethnic Adaptation and Identity: The Karen on the Thai frontier with Burma. Philadelphia: Institute for the Study of Human Issues: 25–61. Keyes, C., 1997, “Ethnic groups, ethnicity” in T. Bareld (ed.): The Dictionary of Anthropology. Blackwell, Oxford: 152–154. King, Victor T., 1981, “Marxist analysis and highland Burma: a critical commentary”, Cultures et developpement 13: 675–688. ——, 1982, “Southeast Asia and the concept of ethnicity”, in King and Willder eds.: Special focus volume on Ethnicity in Southeast Asia in Southeast Asia Journal of Social Science, 10: 1–6. ——, 1983, “Imaginary Kachins (Reply to Nugent 1982)”, Man N.S. 18: 405–406. ——, 2001, “A question of identity: names, societies and ethnic groups in interior Kalimantan and Brunei Darussalam”, Sojourn 16: 1–37. —— and Wilder, William D., 2003, The Modern Anthropology of South-East Asia. An Introduction. New-York, Routledge and Curzon. Kingdon Ward, cf. Ward. Kirsch, Thomas A., 1973, Feasting and Social oscillation. A Working Paper on Religion and Society in Upland Southeast Asia. Ithaca, New York, Cornell University, Department of Asian Studies, Southeast Asia Program, Data Paper n°92. Kuang Haolin 況浩林 and Yang Liqiong 楊麗瓊, 1986, Jindai woguo shaoshu minzu diqu de yapian duhai wenti 近代我國少數民族的區的鴉片毒害問題 (The problem of the opium scourge in China’s ethnic minority areas during the modern period). Zhongguo jingjishi yanjiu 中國經濟史研究 4: 131–42. Kuper, Adam, [1973] 1983, Anthropology and anthropologists: the modern British school. London, Kegan Paul. ——, 1986, “An Interview with Edmund Leach”, Current Anthropology, Vol. 27, no. 4. Launay, A., 1902, Histoire de la mission du Thibet. Lille-Paris, Editions Desclée & Brouwer, 2 vols. Lazcano Nebreda, S., 1998, “La cuestion de la servidumbre en el sudeste de Tibet y regiones colindantes del Himalaya Oriental hasta la ocupacion China de 1950,”Boletin de la Asociacion Española de Orientalistas, vol. XXXIV: 221–237. Leach, Edmund R., 1940, Social and Economic Organization of Rowanduz Kurds. London, LSE Monographs, no. 3. ——, 1954, Political Systems of Highland Burma: a Study of Kachin Social Structure. Cambridge, Harvard University Press. [1967, 1979, 1986, 1997, London and N.J., Athlone Press; 1972 Les systèmes politiques des hautes terres de Birmanie, Paris, Maspéro]. ——, 1960, “The Frontiers of Burma”, Comparative Studies in Society and History III, 1: 49–68. ——, [1961, Rethinking anthropology]; 1968, Critique de l’anthropologie, translated from English by D. Sperber and S. Thion, Paris: P.U.F. [Collection SUP, Le sociologue]. ——, 1963, “The Political Future of Burma”, in Bertrand de Jouvenel (ed.): Futuribles: studies in conjecture. Geneva, Droz. ——, [1966] 1971,”La ritualisation chez l’homme selon son développement social et culturel,” in J. Huxley (dir.): Le comportement rituel chez l’homme et l’animal. Paris: Gallimard, Bibliothèque des Sciences Humaines: 241–248. ——, 1967, “The Language of Kachin Kinship: Reections on a Tikopia Model” in M. Freedman (ed.): Social Organization: Essays Presented to Raymond Firth. Chicago, Aldine: 125–152. ——, 1968, “Ritual,” The International Encyclopaedia of the Social Sciences, vol. 13: 520–526. ——, 1970, “The Concept of Sin Among the Kachin of North Burma”. Proceedings of the Eighth International Congress of Anthropological and Ethnological Sciences. Tokyo: Section S-9: 307–309.

bibliography

317

——, 1976, Culture and Communication. The logic by Which Symbols Are Connected. An Introduction to the Use of Structuralist Analysis in Social Anthropology. Cambridge, Cambridge University Press [ Themes in the Social Sciences]. ——, 1977, “In Formative Travail With Leviathan”, talk at the LSE, 1973, published in R. M. Berndt (ed.): special issue of Anthropological Forum, vol. 4, no. 2. ——, 1982, Social Anthropology. New York-Oxford, Oxford University Press. ——, 1983, “Imaginary Kachins” Man (N.S.) 18: 191–199, 787–788. ——, 1984, “Glimpses of the Unmentionable in the History of British Social Anthropology”, in Annual Review of Anthropology 13: 1–186. ——, 1987, “Tribal Ethnography: past, present, future”, Cambridge Anthropology XI, 2: 1–14. ——, 2000, “Letter from Bhamo” in Hugh-Jones & Laidlaw, The Essential Edmund Leach: 217–119. Le Bar, Frank, 1965, “The Khamu”, Final report to National Research Council, Bangkok, 27 avril 1965”, Human Area Files, Yale University. ——, 1967, “Observations on the Movement of Khmu’ into North Thailand”, Journal of the Siam Society 53: 61–79. Lee, James, 1982, “Food Supply and Population Growth in Southwest China, 1250– 1850”, Journal of Asian Studies 41(4):711–46. Lefèvre-Pontalis Pierre, 1898, Voyage dans le Haut-Laos et sur les frontières de Chine et de Birmanie (Mission Pavie—Géographie et voyages—tome V) Paris, Ernest Leroux. Lehman, Frederic K., 1963, The Structure of Chin Society. A Tribal People of Burma adapted to a non-Western Civilization. Urbana: University of Illinois Press. ——, 1967, “Ethnic categories in Burma and the theory of social systems” in Peter Kunstadter (ed.): Southeast Asian Tribes, Minorities and Nations. Princeton, Princeton University Press, vol. 1: 93–124. ——, 1970, “On Kachin and Chin Marriage Regulations”, in Man n.s. 5: 118–125. ——, 1977, “Kachin Social Categories and Methodological Sins”, in W. McCormack and S. Wurm (eds.): Language & Thought: Anthropological Issues. The Hague: Mouton: 229–250. ——, 1978, “Symbols and the Computation of Meaning”, in D. B. Shimkin, S. Tax, and J. W. Morrison (eds.): Anthropology for the Future. Urbana: University of Illinois, Department of Anthropology Research Series 4: 181–191. ——, 1979, “Who are the Karen, and if so, why? Karen ethnohistory and a formal theory of ethnicity”, in Charles F. Keyes (ed.): Ethnic Adaptation and Identity; The Karen on the Thai frontier with Burma. Philadelphia: Institute for the Study of Human Issues: 215–253. ——, 1984, “Remarks on Freedom and Bondage in Traditional Burma and Thailand”, Journal of Southeast Asian Studies 15, 2: 233–244. ——, 1989, “Internal Inationary Pressures in the Prestige Economy of the Feast-ofMerit Complex: The Chin and Kachin Cases from Upper Burma”, in Susan D. Russell (ed.): Ritual Power and Economy: Upland-Lowland contrasts in Mainland Southeast Asia. DeKalb, Northern Illinois University Center for Southeast Asian Studies, Monograph Series on Southeast Asia, Occasional Papers #14: 89–102. ——, 1990, “Outline of a Formal Syntax of Numerical Expressions, with Especial Reference to the Phenomenon of Numeral Classiers”, Linguistics in the Tibeto-Burman Area 13, 1: 89–128. ——, 1993, ‘Kachin’, Encyclopedia of World Cultures 5, East and Southeast Asia. New York, G. K. Hall (ed. Paul Hockings): 114–119. ——, 1997, “Forward”, in Wang Zhusheng: The Jingpo of the Yunnan Plateau, Tempe, Arizona State University, pp. xiii–xviii. ——, 2003, “The Relevance of the Founders’ Cult for Understanding the Political Systems of the Peoples of Northern South East Asia and its Chinese Borderlands”, in Nicola Tannenbaum and Cornelia A. Kammerer (eds.): Founders’ Cults in Southeast

318

bibliography

Asia: Ancestors, Polity, and Identity, New Haven: Monograph 52, Yale University Southeast Asia Program: 15–39. Lemoine, Jacques, 1978, “L’Asie orientale—Les Chinois Han; les ethnies non Han de la Chine, les Japonais, les Aïnou” in Encyclopédie de la Pléiade. Ethnologie Régionale II. Dijon, Gallimard. —— (ed.), 1997, Systèmes politiques des marches du monde chinois. Metz, Etudes Orientales/ Olizane, (Péninsule 35, Nouvelle Série). Lévi-Strauss, Claude, 1947 [1967], Les Structures Élémentaires de la Parenté. Paris, Mouton, Presses, Universitaires de France; [1969, The Elementary Structures of Kinship. Boston: Beacon Press]. ——, 1966, “The Future of Kinship Studies” (Huxley Memorial Lecture). Proceedings of the Royal Anthropological Institute for 1965: 13–22 [also his preface to the English translation (1969) of his 1947 book, as Elementary Structures of Kinship, R. Needham J. H. Bell, and J. R. von Sturmer trs. Boston: Beacon Press]. Lieberman, Victor B., Burmese administrative cycles. Anarchy and Conquest, c. 1580 –1760. Princeton, Princeton University Press. Lindell, Kristina & Samuelsson, R. & Damrong Tayanin, 1979, “Kinship and Marriage in Northern Kammu Villages: The Kinship Model”, Sociologus 29–1: 60–84. —— & Damrong Tayanin, & Svantesson Jan Olaf & Lundström Hakan, 1982, The Kammu Year: its lore, its music. Lund, Scandinavian Institute of Asian Studies & Curzon Press. Lin Manhong 林滿紅, 1980, Jindai zhongguo “jinkou tidai”gean yanjiu zhi yi, 1858– 1906 近代中國「進口替代」個案研究之一 (A case study of “import substitution” in modern China, 1858–1906). Zhongyang yanjiu yuan jindaishi yanjiu so jikan 中央研 究院近代史研究所集刊 (Bulletin of the Institute of Modern History, Academia Sinica) 9:385–432. Lintner, Bertil, 1997, The Kachin: Lords of Burma’s Northern Frontier. Thailand: Asean Film House. Löfer, Lorenz G., 1964, “Chakma und Sak”. Interenationales Archiv für Ethnographie L, 1: 72–115. Lowis, C. C., 1906, “A Note on the Palaungs of Hsipaw and Tawngpeng” Ethnographical Survey of India, Burma, n°1, Rangoon, Superintendent of Government Print and Stationery. Lundström, Hakan/Damrong Tayanin, 1981, “Kammu gongs and drums 1: the kettlegong, gongs and cymbals”, Asian folklore studies 40/1: 65–86. MacGaffey, W., 1997, “Kongo Identity”, in V. Y. Mudimbe (ed.): Nations, Identities, Cultures. Durham, Duke: 57. Mackenzie, A., 2001, The North East Frontier of India. New Delhi, Mittal Publications. Macdonald, A. W., 1980, “Creative dismemberment among the Tamang and Sherpas of Nepal,” in M. Aris et A. S. Suu Kyi (eds.): Tibetan Studies in honour of Hugh Richardson. Wardminster (England): Aris & Phillips: 199–208. Maran, Brang Di, 1994, The Early Relations of Kachins with English Government. Unpublished M.A. History Thesis, Delhi University. Maran, La Raw, 1967, “Toward a Basis for Understanding the Minorities in Burma: The Kachin Example”, in P. Kunstadter & J. M. Halpern (eds.): Southeast Asian Tribes. Minorities and Nations. Princeton, Princeton University Press, vol. 1: 138–40. ——, 1967, “Notes on the Algebra of Interacting Human Groups: An analysis of Kachin Clichés Regarding Certain Selected Topics”, Unpublished manuscript. Marshall, H. I., 1922, The Karen People of Burma: A Study in Anthropology and Ethnology, Colombus, Ohio State University. Mathieu, C., 2003, History and Anthropological Study of the Ancient Kingdoms of the Sino-Tibetan Borderland—Naxi and Mosuo. Lewiston: The Edwin Mellen Press [Mellen Studies in Anthropology, 11]. Matsaw Nhkum Tu Yau, 1995, Ginru ginsa labau the Nhkum amyu ni a labau laika ning nan, (editor unknown), 82p. (In jinghpaw).

bibliography

319

Mauss, Marcel, [1925, “Essai sur le don. Forme et raison de l’échange dans les sociétés archaïques” L’Année sociologique (NS 1)]; 1990, The Gift. The Form and Reason for Exchange in Archaic Societies, trans. by W. D. Halls, foreword by M. Douglas, New York/London: W.W. Norton. Michaud, Jean (ed.), 2000, Turbulent Times and Enduring Peoples. Mountain Minorities in the South-East Asian Massif. Richmond, Curzon Press. Miles, Douglas, 1990, “Capitalism and the Structure of Yao Descent Units: A Comparison of Youling (1938) and Phulangka (1968)”, in Gehan Wijeyewardene (ed.): Ethnic Groups Across National Boundaries in Mainland Southeast Asia. Singapore: Institute of Southeast Asian Studies: 134–148. Mills, J. P., 1922, The Lhota Nagas. London, Macmillan. ——, 1926 a, “Certain Aspects of Naga Culture” Journal of the Royal Anthropological Institute, vol. 56: 27–35. ——, 1926 b, The Ao Nagas. London, Macmillan. ——, 1937, The Rengma Nagas. London, Macmillan. Ministry of Transport and Communication, 1997, Integrated Rural Accessibility Planning for Luangnamtha province (IRAP), 5 vols. + one summary per district and one provincial summary. Mordant, Ct., 1934, “Monographie du Ve territoire militaire (Phongsaly)” Extrême Asie, n° 82, pp. 545–552; 83, pp. 589–596; 84, pp. 637–641; 86, pp. 629–636; 87, pp. 782–790; 89, pp. 873–878; 92, pp. 1036–1040. Mosko, Mark S., 1999, “Magical Money: Commoditization and the Linkage of Maketsi (‘Market’) and Kangakanga (‘Custom’) in Contemporary North Mekeo”, in David Akin and Joel Robbins (eds.): Money and Modernity: State and Local Currencies in Melanesia. Pittsburgh, University of Pittsburg Press: 41–61. Mya-Tu, M. et alii, 1966, The Tarons in Burma. Rangoon, Burma Medical Research Institute [Special Report Series n° 1]. Nakane, Chie, 1967, Garo and Khasi: A Comparative Study of Matrilineal Systems. Paris and The Hague: Mouton. National Bureau of Statistics of China, China Statistical Yearbook 2002. Beijing, China Statistic Press, 2002. National Statistic Centre, 1997 a, Results from the population census. Vientiane, R. D. P. Lao. Needham, Rodney, 1973, “Prescription”, Oceania 43: 166–181. Neis, Paul, 1885, “Voyage dans le Haut Laos” Le tour du monde. Paris, Hachette. Neufville, J. B., 1828, “On the Geography and Population of Assam”, Asia Research, XVI: 341. Nugent, David, 1982, “Closed systems and Contradictions: The Kachins in and out of History”, Man NS 17, 3: 508–527 (See also Leach’s reply: 1983, “Imaginary Kachins”, Man NS 18, 1: 191–199, and 18, 4: 787–788, and Nugent’s response to Leach, Man NS 18, 1: 199–206). Nung Lungmi Ginru Ginsa Htunghking ahtik labau [ Historical outline of the Nung Lungmi]. Putao, labau komiti, Nung Lungmi htunghking hpung, v + 128p. (In Jinghpaw, 2002). Oldeld, David J., 1999, Cold War relations between Thailand and Laos, 1950s–1991. Saensuk, Burapha University, Indochina Studies Research Report, no. 1. Oppitz, M. & Hsu, E. (eds.), 1998, Naxi and Moso Ethnography. Kin, Rites, Pictographs, Pictographs drawn by Mu Chen, Zurich, Völkerkundemuseum. ——, 1997, “The Bull, the Ox, the Cow and the Yak. Meat Division in the Himalyas”, in S. Karmay et P. Sagant (eds.): Les habitants du toit du monde. Nanterre, Société d’ethnologie [Recherches sur la Haute Asie, 12]: pp. 515–542. Papet, Jean-François, 1997, “La notion de mü:ang chez les Thaï et les Lao” Péninsule, 35 (2): 219–222. Pavie, Auguste, 1898, Recherches sur l’histoire du Cambodge, du Laos et du Siam. Paris, Ernest Leroux (Mission Pavie—Etudes diverses—tome II).

320

bibliography

Pouillon, J., 1972, “Postface. Leach, Lévi-Strauss et les Kachin,” in E. R. Leach: Les systèmes politiques des hautes terres de Birmanie, traduit par A. Guérin, Maspero [Bibliothèque d’Anthropologie]: 361–78. Pritchard, B. E. A., 1914, “A journey from Myitkyina to Sadiya via the N’mai Kha and Hkamti Long,” The Geographical Journal, vol. 43: 521–535. Proschan, Franck, [undated], The Names of Nations and Trives: Ethnonyms, Ethnic Identities and Ethnic Groups (Unpublished manuscript, 70p.). ——, 1997, “We Are All Kmhmu, Just the Same: Ethnonyms, Ethnic Identities and Ethnic Groups”, American Ethnologist 34–1: 91–113. ——, 2001, “People of the Gourd: Imagined Ethnicities in Highland Southeast Asia” The Journal of Asian Studies 60 –4: 999–1032. Qing Heping 秦和平, 1998, Yunnan Yapian Wenti yu Jinyan Yundong 雲南鴉片問題 與禁煙運動 (Yunnan’s Opium Problem and the Prohibition Movement). Chengdu: Sichuan Minzu Press. —— 秦和平, 2005, Xinan Minzu Diqu de Dupin Weihai ji qi Duice 西南民族地區 的毒品危害及其對策 (The Danger of Drug in the Chinese Southwest Nationality Area and our Policies in Dealing with it). Chengdu: Sichuan Minzu Press. Raquez, Alfred, 1902, Pages Laotiennes. Hanoï, F. H. Schneider éditeur. Renard, Ronald D., 1996, The Burmese Connection: Illegal Drugs and the Making of the Golden Triangle. Boulder, Colorado: Lynne Rienner Publish. Robinne, François, 1998, “Les immigrés Intha, la transhumance des bufes et la circumnavigation bouddhique du lac Inlé (État Shan de Birmanie), in P. Pichard et F. Robinne éds: Études birmanes en hommage à Denise Bernot, Paris, EFEO: 333–370. ——, 2000, Fils et maîtres du lac. Relations interethniques dans l’Etat Shan de Birmanie, préface de Charles Macdonald. Paris, Editions du CNRS, Collection ‘Chemins de l’ethnologie’. ——, 2007, Prêtres et chamanes. Métamorphoses des Kachin de Birmanie. Paris, L’Harmattan, Collection Recherches Asiatiques. ——, forthcoming, “Jeux d’échelle et enjeux: rituels processionnels en Birmanie bouddhique” in B. Brac de la Perrière, Y. Goudineau, M. Picard (éds.): Dynamiques identitaires (ethnicités, Etats, religions) en Asie du Sud-Est. Robbins, Joel and David Akin, 1999, “An Introduction to Melanesian Currencies: Agency, Identity, and Social Reproduction”, in David Akin & Joel Robbins (eds.): Money and Modernity: State and Local Currencies in Melanesia. Pittsburgh: University of Pittsburg Press: 1–38. Robequain, C., 1929, Le Tanh hoa, étude géographique d’une province annamite. Paris/Bruxelles, Les Editions G. van Oest (Publications de l’Ecole Française d’Extrême Orient). Rockhill, W. W., 1881, The land of the Lamas. Notes of a journey through China, Mongolia and Tibet. New-York, Century Company. Roux, E., 1897, “Aux sources de l’Irraouaddi. D’Hanoï à Calcutta par terre,” Le tour du monde 3 (N.S.), n° 17 to 23: 193–276. Roux, Henri et 1924, “Deux tribus de la région de Phongsali”, Bulletin de l’Ecole Française d’Extrême-Orient 24: 445–500. Roy Burman, B. K., 1994, Tribes in Perspective. New Delhi, Mittal Pub. Russel S. D. (ed.), 1989, Ritual, Power, and Economy: Upland-Lowland Contrast in Mainland Southeast Asia, Northern Illinois University, Center for Southeast Asian Studies [Monograph Series on Southeast Asia, Occasional Paper n° 14]. Sadan, Mandy, 2000 a, “The Kachin manau and manau shadung—development of an ethno-cultural symbol”, in T. R. Blurton & A. Green (eds): Burma Art & Archaeology. London: British Museum Press: pp. 119–131. ——, 2000 b, “The Kachin Photographs in the J. H. Green Collection: A Contemporary Context”, in E. Dell (ed.): Burma: Frontier Photographs, 1918–1935. London, Merrell: 51–66. ——, 2000 c, “The Kachin Photographs: A Documentary Record of Contact”, in E. Dell (ed.): Burma: Frontier Photographs, 1918–1935. London, Merrell: 67–94.

bibliography

321

——, 2005, “History & Ethnicity in Burma: Cultural Contexts of the Ethnic Category ‘Kachin’ in the Colonial and Postcolonial State, 1824–2004”, Unpublished PhD Thesis, School of Oriental & African Studies, University of London. ——, 2007, “Decolonizing Kachin: Ethnic Diversity and the Making of an Ethnic Category’ in M. Gravers (ed.), Exploring Ethnicity in Burma. Copenhagen: NIAS Press. Sagant, Philippe, 1990, “Les tambours de Nyi-shang (Népal). Rituel et centralisation politique,” in F. Meyer (éd.): Tibet, civilisation et société, Paris: Fondation Singer-Polignac, MSH: 151–70. Sahlins, Marshall D., 1965, “On the Sociology of Primitive Exchange” in Michael Bandon (ed.): The Revelance of Models for Social Anthropology. London, Tavistock: 139–235. Scott, J. George, 1921 [1906], Burma: A Handbook of Practical Information. London, Alexander Moring Limited. —— & Hardiman, J. P., 1900, Gazetteer of Upper Burma and the Shan States. Rangoon, Superintendent Government Printing, Part I, Vol. I. Shadap Sen, N. C., 1981, The Origin and Early History of the Khasi-Synteng People. Calcutta, Firma. Sharmathakur, G. C., 1985, The Lalung (Tiwas). Guwahati, Tribal Research Institute Assam. Silva-Fuenzalida, Ismael, 1949, “Ethnolinguistics and the study of culture”, in American Anthropologist, New Series, vol. 51, 3: 446–456. Smalley, William, 1961, Outline of Khmu Structure, American Oriental series, Essais; 2 American Oriental Society, New Haven (Conn.). ——, 1965, “Ethnographic notes on the Khamu”, Reproduced by U.S. AID Mission to Laos. ——, 1965, “Cian: Khmu’ Culture Hero” in Felicitation Volumes of Southeast Asian Studies. Bangkok, Siam Society, vol. 1: 41–54. Smith, Martin, 1991, Burma: Insurgency and the Politics of Ethnicity. London, Zed Books. Smuckarn, Snit & Breazeale, Kennon, 1988, A Culture in Search of Survival: the Phuan of Thailand and Laos. New Haven, Yale University Southeast Asia Studies. Sopitt, C. A., 1885, A Short Account of the Kachcha Naga (Empeo) Tribe in the Cachar Hills, with an outline Grammar, Vocabulary and Illustrative Sentences, Shillong, Assam Secretariat Press. Sperber, Dan, 1974, Rethinking Symbolism. Cambridge: Cambridge University Press. Sprenger, Guido, 2004, “Encompassment and its discontents: Rmeet and lowland Lao relationships”, in Gerd Baumann & André Gingrich (eds.): Grammars of identity/alterity: a structural approach. New York/Oxford: Berghahn. ——, 2005, “The way of the buffaloes: Trade and sacrice in Northern Laos”, Ethnology 44, 4: 291–312. ——, 2006 a, “Political periphery, cosmological center: the reproduction of Rmeet socio-cosmic order and the Laos-Thailand border”, in Alexander Horstmann & Reed Wadley (eds.): Centering the margin: Agency and Narrative in Southeast Asian borderlands. New York/Oxford, Berghahn: 67–84. ——, 2006 b, Die Männer, die den Geldbaum fällten: Konzepte von Austausch und Gesellschaft bei den Rmeet von Takheung , Laos. Berlin, LitVerlag. ——, 2006 c, “Bone transfers: Incomplete replacement in Rmeet ritual exchange”, Taiwan Journal of Anthropology 4,1: 79–112. ——, in press, “Do the Rmeet have clans?”, in Yves Goudineau & Michel Lorillard (eds.): Nouvelles recherches sur le Laos. Paris: EFEO. Stack, E., 1908, The Mikirs; from the papers of the late Edward Stack . . . (edited, arranged and supplemented by Sir Charles Lyall). London, Nutt. Steinmann, A. & Sanidh Rangsit, M. C., 1939, “Denkmalformen und Opferstätten der Lawa” Zeitschrift für Ethnologie, 71: 163–174. Stevenson, H. N. C., [1943] 1970, The Economics of the Central Chin Tribes. Bombay, Times of India Press.

322

bibliography

Strathern, Marilyn, 1988, The Gender of the Gift. Berkeley, University of California Press. Stuart-Fox, Martin (ed.), 1986, Laos: Politics, Economics and Society. London, Pinter Rienner. Sumlut Gam, 1955, History of Jinghpaw People (written in Jinghpaw, privately printed, Htingnan Village: 2 vols.). Taillard, Christian et Sisouphanthong Bouthavy, 2000, Atlas du Laos de la RDPL: les structures territoriales du développement économique et social. Paris, Reclus, coll. “Dynamiques du Territoire”, CNRS-Liber Géo-La Documentation Français. Tambiah, S. J., 2002, Edmund Leach: an anthropological life. Cambridge, Cambridge University Press. Tannenbaum, Nicola, 1989, “Power and its Shan Transformations” in S. Russel (ed.): Ritual, Power and Economy. Upland-Lowlands Contrasts in Mainland Southeast Asia. DeKalb, Northern Illinois University: Center for Southeast Asian Studies, Occasional Paper n° 14: 67–87. Tayanin, Damrong and Kristina Lindell, 1999, The traditional Kammu society, www.ling. lu.se/persons/Damrong/kammusoc.html. Tegenfeldt, H. G., 1974, A Century of Growth: the Kachin Baptist Church of Burma. Pasadena, William Carey Library. Testart, A., 2001, L’esclave, la dette et le pouvoir. Etudes de sociologie comparative. Paris, Editions Errance. Thiollier, Lucien Auguste, 1930, La grande aventure de la piastre indochinoise. Saint-Etienne, J. Bruyere. Thomas, K., 1963, “History and Anthropology”, Past and Present 24: 3–24. ——, 1985, Religion and the Decline of Magic: studies in popular beliefs in sixteenth and seventeenth century. Harmondsworth: Penguin Books Ltd. Thomas, N., 1997, In Oceania. Visions, Artifacts, Histories. Duke, Durham. Touzet, André, 1939, Le régime monétaire indochinois. Paris, Librairie du Recueil Sirey. Trocki, Carl A., 1999, Opium, Empire, and the Global Political Economy: A Study of the Asian Opium Trade, 1750 –1950. London, Routledge. Turner, Victor, 1960, The Ritual Process: Structure and Anti-Structure. Ithaca: Cornell University Press. Valeri, Valerio, 1994, “Buying Women but Not Selling Them: Gift and Commodity Exchange in Huaulu Alliance”, Man 29 (1): 1–26. Wagner, Roy, 1977, “Analogic Kinship: A Daribi Example”, American Ethnologist 4: 623–642. Wang, Zhusheng, 1997, The Jingpo: Kachin of the Yunnan Plateau. Foreward by F. K. Lehman. Tempe: Arizona State University Program for Southeast Asia Studies Monograph Series. Wawn Awng, 2003, Wunpawng Jinghpaw Htinggaw Mying hte Amyu Mying Ni [Thousands Names of the Kachin Family], Yangon, Laika Dip & Chywi, 79p. (In jinghpaw). Ward, (Captain) F. K., 1924, From China to Hkampti Long, London: Edward Arnold & Co. ——, 1934, A plant hunter in Tibet, London: Jonathan Cape. Watt, G., 1887, “The Aboriginal Tribes of Manipur” Journal of the Anthropological Institute, vol. XVI: 346–370. Webster, Donovan, 2003, The Burma Road: The Epic Story of the China-Burma-India Theatre in World War II. New York: Farrar, Straus and Giroux. Whitmore, John K., 2006, “The Rise of the Coast Trade, State and Culture in Early Dai Viet”, Journal of Southeast Asian Studies 37, 1: 103–122. Williams, P. & L. Chrisman (eds.), 1994, Colonial Discourse and Post-Colonial Theory. A Reader. New York, Columbia University Press. Woodward, Mark, 1989, “Economy, polity, and cosmology in the Ao Naga Mithan feast”, in Susan D. Russell (ed.): Ritual, power, and economy: upland-lowland contrasts in

bibliography

323

Mainland Southeast Asia. DeKalb, Northern Illinois University, Center for Southeast Asian Studies: 121–142. —— and Susan Russell, 1989, “Introduction: Transformations in Ritual and Economy in Upland and Lowland Southeast Asia”, in Susan D. Russell (ed.): Ritual, Power and Economy: Upland-Lowland Contrasts in Mainland Southeast Asia. Dekalb, Center for Southeast Asian Studies, Northern Illinois University: 1–26. Wunpawng Htung Hking Masa Tara Laika; Ginwang Wunpawng Htung Hking Rung [Kachin’s Customary Law], typescript, 19p. (In Jinghpaw, n.d.). YNSBJZ, Yunnan sheng bianji zu雲南省編輯組 (Yunnan Provincial editorial division): 1985 a, Jingpo zu shehui lishi diaocha 1 景頗族社會歷史調查(一) (A sociohistorical survey of the Jingpo nationality 1). Zhongguo shaoshu minzu shehui lishi diaocha ziliao congkan 中國少數民族社會歷史調查資料叢刊 (Series on the results of the sociohistorical survey of China’s national minorities). Kunming: Yunnan renmin chubanshe 雲南人民出版社 (Yunnan People’s Press). 1985 b, Jingpo zu shehui lishi diaocha 2 景頗族社會歷史調查( 二 ) (A sociohistorical survey of the Jingpo nationality 2). Zhongguo shaoshu minzu shehui lishi diaocha ziliao congkan 中國少數民族社會歷史調查資料叢刊 (Series on the results of the sociohistorical survey of China’s national minorities). Kunming: Yunnan renmin chubanshe 雲南人民出版社 (Yunnan People’s Press). 1986 a, Jingpo zu shehui lishi diaocha 3 景頗族社會歷史調查( 三 ) (A sociohistorical survey of the Jingpo nationality 3). Zhongguo shaoshu minzu shehui lishi diaocha ziliao congkan 中國少數民族社會歷史調查資料叢刊 (Series on the results of the sociohistorical survey of China’s national minorities). Kunming: Yunnan renmin chubanshe 雲南人民出版社(Yunnan People’s Press). 1986 b, Jingpo zu shehui lishi diaocha 4 景頗族社會歷史調查( 四 ) (A sociohistorical survey of the Jingpo nationality 4). Zhongguo shaoshu minzu shehui lishi diaocha ziliao congkan 中國少數民族社會歷史調查資料叢刊 (Series on the results of the sociohistorical survey of China’s national minorities). Kunming: Yunnan renmin chuban-she 雲南人民出版社 (Yunnan People’s Press). Young, Michael, 2004, Malinowksi: odyssey of an anthropologist, New Haven, Yale University Press. Yoya samain nhin yoya yinkhyemhu dhale thunsan mya [History and traditions of the Rawang]. Putao, Rawang Comittee for Culture et Edition, 260p. multigr. (In Burmese, n.d.). Zhang Pengyuan 張朋園, 1990, Luohou diqu de ziben xingcheng—Yun-Gui de xiexiang yu yapian 落後地區的資本形成:雲貴的協餉與鴉片 (Captial formation in an underdeveloped region: assistance loans and opium in Yunnan and Guizhou). Guizhou wenshi congkan 貴州文史叢刊1:50–74. Zhang Wenyi, 2004, The Circulatory Marriage: Kinship Systems of Jingpo in Jinzhuzhai, Yunnan, China. unpublished M.A. thesis, Beijing, Peking University. Zhang Wenyi, 2006, The Temporal Delay of Individual Memory of Trauma and the Cultural Time Recognition Device. MS, University of Illinois.

INDEX AFPFL (Anti Fascist People’s Freedom League) 20–22 Ahom xxxviii–xli, xlix, lii, 59, 92–93, 97–100, 104, 117–120, 123 Akha xxxiv, 154, 170, 301 Amri 100 Ang 116, 120, 123–124 Angami see also Naga xxxiv, xli, xlviii, 109–112 Ao see also Naga 110 Assam xxxviii–xxxix, xlix, 13–14, 19, 41, 59–60, 91–93, 95, 99, 101, 103, 105, 107, 109, 110, 118, 268, 271 Asymmetrical exchange see also exchange, marriage exchange, generalised exchange, symmetrical exchange xxiii, xxxi–xxxiv, xxxvii, xliv, li, 45, 54, 59, 172, 176, 184, 249, 267, 275, 285 Atsi see Zaiwa Attlee, Clement 21 Aung San 20–21, 32, 44, 49 Aung-Thwin, Michael 287, 293, 307 Austin, John xxvii, xlvii Ban Chomsy 134 Bawi xxxi, xliv, xlix–l, 124 Bhamo 8–13, 21–22, 35–36, 43, 46, 216–217, 291 Bhoi 104 Blacksmiths see also silversmiths 131, 147, 306 Boatmen 131, 138–139, 143–144 Bodo-Garo xxxviii, 93, 104 Books of the Land or Peum kongdin 188–189, 199–200, 203–205, 208 Botel Tobago 6 Brahmaputra xxxix, 92, 104, 117–118 Bride price or bridewealth 132, 164, 166–167, 169–171, 174, 177, 181, 184, 230–232, 238 Buckmaster, Celia 8, 12 Buddhism 133, 149, 164, 305 Buddhist xxvi, xxxix, xlvi, 254, 267, 271 Buffaloes xliii, 132, 158, 167, 172, 178, 182, 248, 306

Buranji (Ahom Chronicles) xl, 92, 118–119 Burma Frontier Service 8, 10, 13 Burma Military Police 9, 217 Burmanisation 285, 294 Butler, Capt. John 110, 119 Buttereld and Swire 5, 7 Caravans xli, xlv, 138, 147, 154, 192–193, 216, 254 Chang see also Naga 110, 116 Chiang Mai 132 Chin see also Hakha/Haka, Zahau, Socte/Sukte, Kamhao, Sailo, Tashon, Kuki, Zothung xxii, xxv–xxvi, xxix, xxxi–xxxii, xliii–xliv, xliii–xlv, lii, 14, 18, 75, 120–121, 138, 300 Chinthong 100–101 Christian missions or Christian missionaries see also Hanson, Ola and Smalley, William 19, 23, 27–28, 40, 43, 67–68, 75, 141, 154, 257, 268–270, 275 Christianism 103, 284, 295–296 Chyauchyi Htang Wa 31, 36, 39 Cloth 216, 224, 226, 228–232, 236, 243, 245, 248, 252, 267 Clothing see also earrings, headdresses, jewellery, necklaces 133, 142, 175, 178, 216, 224, 228–233, 243, 245, 250, 252 Cochrane, Governor Archibald 9 Coins see also currency, copper coins, silver coins, piasters, Piastre de Commerce, money, kip, gemuul 164–165, 170, 171, 173, 176–183, 185, 228, 232 Communist 5, 20, 78, 150, 153–156, 158–159, 164, 179–180, 183, 207, 222, 250, 278 Copper coins see also currency, silver coins, piasters/Piastre de Commerce, money, kip, gemuul 169 Cotton 138, 225, 228–229, 232 Couplets/Dyadic sets xxviii–xxix, 78–84, 86 Currency see also coins, copper coins, silver coins, piasters/Piastre de

326

index

Commerce, money, kip, gemuul 14, 17, 169–171, 174, 178, 181–183, 215, 218–221, 225, 234, 255 Dai see also Tai 144, 217, 225–226, 251 Daikho xli, 94–99 Dama, see also wife takers, marriage exchange xxxii, xxxvii, 41, 44–47, 55–59, 64, 238, 283–285, 290–291, 293–295, 301, 305 Dap see also lineage xxxvi–xxxvii, li Dayang valley 114 Debt xliv, xlviii, 24, 229, 231, 233–236, 238–239, 246–247, 249, 251–252, 255, 259, 264–265, 267–274, 277, 279–282, 283 Dehong 211, 216, 226, 251 Dimasa xxxviii–xxxix, 92–100, 102–103, 106 Donohue, John 161, 191 Drums (bronze) see also kettledrums, frog drums li, 131–132, 163–168, 176–183 Drung or Qiuzi (Kiutzu) xlv–xlviii, 257–259, 264–270, 272–280, 283 Duleng 40–42, 66 Dulong see Drung Dulong Valley 265, 273, 274 Dumsa see also jaiwa, spirit priest xxvii–xxviii, 79–82, 86 Durkheim, Émile vli, 283 Duwa xxx, 46, 48, 51 Dvruq-wa (‘Group-making’— Drung) 276–277, 279 Dzayul (Tibet) 266, 268 Earrings 224–226 Endogamy 41, 97, 166 Enriquez, Maj. C. M. (Theophilus’) 67–68, 216–217, 286, 290 Ethnographic algebra 16, 27 Exchange xxiii, xxxi–xxxiv, xliii, 32, 45–46, 56, 101, 138, 163–164, 168, 170–171, 174, 176–179, 181–183, 187, 217–223, 229–230, 232–233, 236, 239, 240, 243–244, 246–253, 258–259, 267, 269, 273–275, 281, 283–285, 289–290, 293–295, 300, 302, 304–305, 307

Fa Ngum 143, 152 Feasts of Merit see also manau/ manao xliii–xliv, 111, 114, 168, 276, 278 Firth, Raymond xxiv–xxv, xxviii, xlvii, 6–8, 17, 19–20, 22, 23, 26 Forest L, 131, 138, 175, 178–179, 190, 198, 203, 217, 236 Fortes, Meyer 7, 25 Foucault, Michel xxviii, 77, 299, 301, 307–308 Friedman, Jonathan 137, 214–215, 255 Frog drums see also drums (bronze), kettledrums li Funerals see also mortuary rituals 164, 172, 175–179, 224, 229–231, 234, 247 Fürer-Haimendorf, Christoph von xlix, 110–112, 117, 119, 122, 124 Garghaon 117 Garo xxxviii–xxxix, 93, 103–104 Gauri 286, 288, 290 Gemuul see also currency, copper coins, silver coins, piasters/Piastre de Commerce, money, kip 169–171, 177, 182 Generalised exchange see also exchange, marriage exchange, asymmetric exchange, symmetrical exchange 283, 285, 290, 294, 302, 304, 305 Giddens, Anthony 37–38, 61, 65 Gongs xliv, 132, 167, 178, 228–230, 232, 236, 243, 245, 249, 253, 273 Granet, Marcel 283, 287 Green, James Henry 223–224, 252, 290–291 Gumchying gumsa xxv–xxviii, xxix, xxxiii–lii, 41–42, 44, 47, 51, 54–55, 77–78 Gumlao see also gumsa-gumlao xxiii–xxvii, xxix–xxx, xli, xlv–xlvi, 32, 34, 36, 40–42, 54, 59, 66, 67, 72–74, 76–78, 80–84, 87, 136, 162, 183–184, 187, 212–215, 253, 255, 263 Gumlau see gumlao Gumrawng gumtsa xxv–xxvi, xxxv, 46–49, 51, 77–78, 80–82, 87 Gumsa see also gumchying gumsa, gumrawng gumtsa, zaw gumsa,

index gumsa-gumlao xxiv–xxvii, xxix–xxx, xxxiii, xxxv, xlvi, li–lii, 26–2, 32, 34–36, 40–56, 58–60, 66, 72–87, 109–125, 136, 162, 183, 187, 192, 212–255, 263 Gumsa-gumlao see also gumchying gumsa, gumrawng gumtsa, zaw gumsa, gumlao xxv, 27–28, 116, 122, 125 Gumshem also Gumshem Magam 46, 49–51 Gumyu-yu xxix Guwahati 91, 101, 103 Habe 100, 102 Hakha/Haka see also Chin xxv–xxvi, xxxi, l, 120 Han Chinese 216, 231, 292–293 Hanson, Rev. Ola (Kachin Dictionary) xxiv, 40, 75, 84, 86, 211, 243–244, 286 Headdresses 224 Headhunting xli, xlviii, 111–112, 120 Hedamba Kingdom 93–94, 97–98 Heirloom xliv–xlv, 178 Hertz, W. A. 84, 286 Hindu missionaries xl Hkan Naw (dumsa) 80–86 Hkau wang magam xxxi, xxxiii–xxxv, xxxvii, li, 41, 44–45, 54, 58 Hmong xlv, 155–156, 158, 171 Horniman Trust 19–20 House spirit 172, 174–178 Household Survey 223, 226–232 Hpaga xliii, xlv, 53, 234, 236–244, 246–247, 251, 281 Hpaji xxix–xxx, 241–244 Hpalang vlii, 10, 13, 15, 39, 261 Hsenwi 43, 48–49 Htingbai Naw Awn 33 Hukawng valley or Hukong valley xxiii, xxx, 14, 33, 42, 59–60, 66, 84, 117, 236 Hutton, J. H. 110–111, 113–115 Interethnic 127–129, 131, 133, 135–137, 139–141, 143, 145, 147, 149, 151, 153, 155, 157, 159, 258, 287, 293–295, 300, 306, 308 Iron bars see also iron tools, iron utensils 131, 138, 193 Iron tools see also iron bars, iron utensils 142

327

Iron utensils see also iron tools, iron bars 146, 147, 216, 227, 229, 236, 273 Irrawaddy river 84, 257, 265 Izikowitz, Karl Gustav 96, 128, 132, 136, 138–139, 150, 162, 165–169, 172, 178, 180, 182, 201 Jaintiya xxxviii–xxxix, 92, 99–100, 104–105 Jaiwa see also dumsa, spirit priest 79 Jewellery 216, 228–229, 232, 243, 245, 247, 252 Jinghpaw see also Jingpo, Jingpho and Singhpo xxv, xxviii–xxx, xxxii, xxxiv–xxxv, xli, xlv, xlvii, l–li, 32, 36, 39–43, 53–54, 68, 73, 75–81, 84, 86, 211–212, 240, 242–244, 246–247, 249, 253, 261, 272, 283, 286–292, 294, 296, 300–302, 305 Jingpho see also Jingphaw, Jingpo and Singhpo 234 Jingpo see also Jingphaw, Jingpho and Singhpo xxiv, 211–212, 215–217, 222–223, 226, 229, 231, 233, 240–241, 247, 249–255, 291–291 Jonthai 95–96 Kabui (Rongmei) see also Naga 110–111 Kabui see also Naga 110–111 Kacha see also Naga 110–111, 113–114 Kachari see Dimasa Kachin Hills 5, 9, 23, 43, 68, 82, 127–128, 159, 217, 253–255 Kachin Hills Regulation(s) 43, 68 Kachin Levies 14, 16, 217 Kachin Regeneration Scheme 9 Kachin State Government 32, 36, 38–39 Kamhao see also Chin 120 Kamrup 101, 103 Karbi 98–106 Karen li, 21, 121, 166–167, 182 Karenni or Red Karen 182 Kawlu Ma Nawng xxiii, 32, 34, 40–42, 236 Kayah xlviii, li Kettledrums see also drums (bronze), frog drums 163–168, 176–183 Khasi xxxviii–xxxix, 59, 92, 104–105 Khmu xlii, li, 127–129, 131–150, 152–156, 158–159, 168, 170–171

328

index

Khunlung and Khunlai 117 Kileki valley 114 Kingdon-Ward, F. 258 Kinship xxvi, xxxi, xxxix, 5, 16, 27–28, 34, 46, 48, 52–60,64–66, 69, 76, 230–231, 251, 270, 272, 278, 282–283, 285, 287, 290, 293–295, 305–306 Kip see also currency, copper coins, silver coins, piasters/Piastre de Commerce, money, gemuul 169171, 173, 179, 183 Kohima xli Konkud 134–135, 151 Konyak see also Naga xlix, 110, 116, 122 Kuki, see also Chin xlix, 93, 98 Kumje Maran Tawng 36 Labau xxxv–xxxvii, l Lachik or Lachid, Lashi 68, 289, 291–292, 294 Lahpai (clan) 47, 284, 286–292 Lahtaw (clan) 47, 284, 286, 288–293 Lalung or Tiwa 104–106 Lamet see Rmeet Lan Xang 141, 143–144, 155 Lattimore, Owen 6 Lawngwaw or Maru Lazum (clan) 68, 289, 292, 294 Lazum (clan) 47, 290 Lem 132–133, 136, 138, 165–168, 180, 184 Lévi-Strauss, Claude xxv, 4, 54, 70, 248 Leyden, John 13, 15, 18–20, 22 Lhota see also Naga 110–111, 113 Lindell, Kristina 129, 141, 170 Lindok 99–104 Lineage see also dap xxxi, xxxiii–xxxiv, xxxvi–xxxvii, l–li, 10, 41–42, 44–47, 54, 59, 64, 76, 94, 97, 111–112, 115–116, 119, 121–124, 129, 131–132, 134, 149, 172, 187, 190–191, 195, 200, 234, 255, 283–297 Lisu xliii, xlv, li, 68, 257, 261, 270, 274–276, 290–292, 294–295, 299, 301–302 Lords of the Land or Chao thi din 188–189, 196, 199–200, 205 LSE (London School of Economics) 6–10, 12, 18, 20, 23–26, 29 Luang Prabang 137–138, 144, 147, 150, 155, 188–189, 193, 199 Luxi County 222, 229–231, 246

Madai see also Mdai 50, 95, 121–122 Madu ai (‘to own’—Jinghpaw) 121, 233 Magwitawng 84 Maibong 93 Malinowski, Bronislaw xxiv, 5–8, 19, 24, 25, 61, 65 Manau or manao see also Feasts of Merit xliii, 36, 79, 123, 224, 255, 285 Mandalay 13, 17, 288–289 Manipur also Manipuri xxxix–xl, 118 Mao see also Naga 110 Maram see also Naga 110 Maran (clan) 45, 47, 284, 286–287, 289–292 Maraw (spirit) 86 Marip clan 47, 284, 286–287, 289–293 Market(s) xliii, xlv, 21, 170, 181, 215, 217–218, 221, 224–226, 229–231, 234, 236, 243, 248–249, 270 Marriage exchange see also exchange, generalised exchange, asymmetrical exchange, symmetrical exchange, mayu, dama, wife givers, wife takers xxxiv, 239–240, 244, 246–248, 250–253 Maru see Lawngwaw Mauss, Marcel 281–282 Mayam see also slave xxix–xxx, 53, 59–60, 272 Maymyo 12–13 Mayu see also wife givers, marriage exchange xxxvii, 41, 44–47, 55–59, 64, 238, 283, 285, 290–291, 293–295, 301, 305 Mdai see also Madai 94–95 Meghalaya xxxviii–xxxix, 2, 93, 98, 104 Meithei xxxix–xl Mekong river or valley 130, 140, 143–144, 147–148, 150, 270, 272, 274 Migration 95–96, 131–133, 136, 138, 144–145, 147, 151, 153, 157–158, 166, 171, 179, 202, 216 Mills, J. P. 110–112, 114–115 Ministry of Kachin State 32 Mizo xxxiv Modernity xxxv, 44, 65, 84, 166 Mon Khmer 128–131, 133, 136–137, 141, 146–147, 149, 159, 163, 302 Money see also currency, copper coins, silver coins, piasters/Piastre de

index Commerce, kip, gemuul 169–171, 173–179, 182–183, 215–222, 229–230, 232–233, 239–240, 243, 245, 247–253 Mong Mao xxxviii, 117 Mong or mung or meuang or muang xxxviii, 116–117, 119, 133, 143–144, 151–152, 156, 187–190, 192, 197, 202, 206 Mortuary rituals see also Funerals 171, 174–175, 177, 181, 183 Muiho Mulat 222, 225, 233, 247, 252–253 Mukrang 93 Myitkyina 9, 14, 17, 21, 22, 32, 35–36, 38–39, 43, 80–82, 139, 266, 288–292 Myth xlvii–xlviii, l–li, 79–80, 131, 145, 151–152, 182, 213, 260–264, 276–277, 279, 284, 286, 294, 307 N’Hkum also NHkum or Nhkum 31, 45, 284, 286–287, 289–292 Naga Hills or Nagaland xlix, 93, 97–98, 109–114, 116, 119 Naga see also Angami, Ao, Kabui, Kacha, Konyak, Lhota, Mao, Maram, Nocte, Rengma, Sangtam, Tangkhul, Tangsa, Tenkoh, Zemi, Sema, Chang, Kabui (Rongmei) xxiv, xl–xli, xlviii–xlix, 97–98, 109–125, 128, 187, 276, Nalae District 129, 135, 142, 144–145, 147, 156–157 Nam Tha valley 128–129, 132, 134, 136, 138–140, 142–145, 147, 149–150, 155, 158–159 Namhkam 49 Namhkyek 224 Namtammai Valley 16 Namwan Tract 49 Nat (spirit) 50, 50, 86, 121–122, 252, 277 Naxi 257–259, 265, 267, 270, 274–275 Ne Win 26, 49, 51 Necklaces 224, 237, 247 Neo Lao Hak Sat (NLHS) 154–156 Neufville, Lt. 34 Nocte see also Naga xlix, 116, 119 Nogmaung 17 North Cachar 92–93, 95 North Hsenwi 43, 48–49 Nugent, David xlv, 161, 192, 214–215, 272

329

Nung also Nung-Rawang, Nu, Nung Lungmi xlvi, 16, 35, 68, 257, 265–268, 274, 289–290, 292, 294 Omaha-type marriage system xxxiii–xxxiv Opium xxvi, xlv, 14, 17, 154, 192, 214–217, 221–222, 225–226, 229–232, 234, 236–237, 243, 245, 247, 252–255 Orwell, George xxiii, 11 Oscillation xxii, xxiv–xxvii, xliii, 26, 42, 69–70, 72–74, 76–77, 187, 211–214, 223, 240, 251 Ou river 187–190, 202–203, 208 Oxen 258, 269–270, 273–278, 280 Palaung or Ta-an 121, 131, 142, 145, 147, 224, 300 Panglong conference also Panglung 21–22, 32, 44, 49 Pangmu Duwa 31, 46–47, 49, 58 Pa-O 287 Pareto’s Principle xxii, 26 Pathet Lao 152, 155, 207 Patkoi/Patkai xli, 113, 117 Phongsaly province 130, 187, 192–193, 201–202, 205–207 Phunoy lii, 187–193, 196–197, 199, 201–202, 206, 208 Piasters, Piastre de Commerce see also currency, copper coins, silver coins, money, kip, gemuul 165, 169–179, 181–183 Pinpomar 100 Pragmatics xxviii–xxxiv Prestige economy xliii, xlv, li, 276, 278 Proschan, Franck 141–142 Pungga Ja Li 31, 67, 80 Putao or Fort Hertz 14, 16, 266, 290 Qiuzi (Kiutzu) see Drung Radcliffe-Brown, Alfred xxiv–xxv, 19, 61, 65 Rance, Governor Hubert 20, 21 Rangoon or Yangon 8–13, 17–18, 20–21, 25, 32, 67, 288 Rawang see also Nung 35, 257, 265–266, 289–292, 294 Rengma see also Naga 110, 113 Riang/Tippera/Tripura xli Ritual xxvi–xxx, xli–xlviii, l, 51, 67, 79–84, 86, 91–95, 97–99, 101–107, 111–112, 115, 121–123,

330

index

129, 131–134, 136–137, 150–151, 162, 164–176, 178–185, 189, 191, 196–197, 199–201, 203–208, 213, 221, 228–230, 233–240, 243–247, 250–251, 258–264, 273–283, 285, 295–296, 304–305 Ritual geography see also symbolic landscape 92 Ritual language xxviii, xlv–xlvii, l, 79–87, 256, 258–260, 263–264, 280–283, 285, 295–296, 304 Rmeet also Lamet 128–129, 131–136, 138–139, 141–142, 147, 150, 153, 156–158, 161–168, 171–172, 176–185, 201 Rongker 101–102 Royal Anthropological Institute 16–20, 23 Ruili County 226–227, 231–232 Sailo see also Chin 120 Salt xli, lii, 14, 98, 117–118, 120, 138, 192–193, 216–217, 225–226, 229–230, 267 Salween river (Nujiang) 14, 254, 257, 267, 274 Samtao 129, 131, 142, 145, 147, 149 Sangtam see also Naga 110 Sao Shwe Thaike 21 Sara Hkyeng Nang 36 Sawbwa or saopha xxiv, 21, 48–49, 305 Scheduled tribes 99 Segmentary lineage 10, 59 Sema see also Naga 109–110, 115–116 Semkhor 97–98 Shan xxi–xxii, xxiv–xxvi, xxix–xxx, xxxviii–xxxix, xlv–xlvi, li–lii, 3, 10, 13, 21–22, 25–26, 28, 43, 48–50, 59, 66, 117–119, 122, 137, 145, 182–183, 192–193, 212–213, 217, 224, 254, 263–264, 287, 291, 302, 305–306 Shan States xxxviii, xlv, 13, 43, 48, 49, 145, 254, 287, 305–306 Shweli river (Nam Mao) xxxviii, xl, 49 Silver coins see also currency, copper coins, piasters/Piastre de Commerce, money, kip, gemuul 146, 163–165, 169–170, 179, 181, 185 Silversmiths see also blacksmiths li, 182 Singhpo or Singpho xlv, 41–42, 59, 66 Sinlum 9–10, 12–13, 15–16, 31, 39, 69 Sipsong Panna 144–145, 188

Slave or slavery see also mayam xxx–xxxi, li, 52–53, 58–61, 65, 84, 118, 192, 236–237, 258, 267–275, 277, 281 Smalley, William 141 Sokte/Sukte see also Chin 120 Spirit priest 79–80, 82–84 Stevenson, H. N. C. (Noel) xxii, xxv, xli, xliii, 7–9, 11, 13–15, 18–22, 39, 75, 276 Stewart, Kilton 6, 8 Sumlut Gam xxxiii, 33 Sumprabum 14 Supreme Council of United Hill Peoples 21 Swargadeo 92, 119 Symbolic landscape see also Ritual geography 101 Symmetrical exchange see also exchange, marriage exchange, generalised exchange, asymmetrical exchange xxxi Tai see also Tai-Ahom and Dai xxxviii–xxxix, xlii, lii, 92, 117–119, 121, 128–129, 130–133, 135, 137, 140–141, 143–145, 147, 149–153, 155–156, 158, 164, 166, 187–190, 192–193, 197, 201 Tai-Ahom see also Ahom xxxix Tangkhul see also Naga 110–111 Tangsa see also Naga 116 Taron(g) 271 Tashon xxv–xxvi, 120 Taungthu 306 Tenkoh see also Naga 110, 122–124 Textiles, Chinese 193 Thakali 62 Tibeto-Burman and Tibeto-Burmese xxxviii–xxxix, xli, xlix, 92–93, 159, 187, 201, 257–258, 265, 293, 302 Tizu river 114 Tmoy 127, 129, 135, 140–147, 149–151, 153, 155–157, 159 Trade routes xli, 69, 193 Trade see also trade routes, caravans, hpaga xxvi–xxvii, xli, xliii, xlv, 9, 10, 17, 53, 120, 131, 133, 138–140, 144, 147, 165–166, 170–171, 179, 192–193, 214–217, 219, 222, 225, 229, 234, 236–237, 239–240, 244, 246, 248, 250, 254 Transethnic(ity) 283, 287, 292–293, 295–296, 300–301, 305, 307

331

index ‘Triangle’ (Mali and N’mai river region) 121 Tsarong (Tibet) 267–272, 274 Tsasen 40–42, 66 Turner, Victor xxv UNESCO 24 Union of Burma 21, 32–33, 36, 48 Upcott, Rosemary 6 Wa xli, l, 138, 142, 145, 254 Wahkyet Wa l, 286 Wakching 117, 123–124 Wancho see also Naga xlix, 116 Wealth objects 164, 212, 223–224, 226, 228–237, 239–248, 250–255, 281 Weddings 174, 229–231, 246, 248, 292 Wife givers see also mayu, marriage exchange xxxi–xxxiv, xxxvii, 132, 172–174, 177, 179, 230, 232, 243, 245, 247–251

Wife takers see also dama, marriage exchange xxxi–xxxiv, xxxvii, 132, 170, 172, 174, 177, 179, 231–232, 246–250 Wilkie, R. S. 9–10, 13, 15, 23 Woi-awn magam 41–42 Woodthorpe, Col. R. G. 110 Wool 224,228, 232, 267 Wunpawng l–li, 300 Xishan Township 229, 231 Yao

xlii–xliii, xlv, lii, 158, 301

Zahau see also Chin 120 Zaiwa xxviii, 68, 211–255, 288, 291–292, 294 Zaw gumsa see gumrawng gumtsa Zemi see also Naga 93, 110–111 Zothung see also Chin 120

Smile Life

When life gives you a hundred reasons to cry, show life that you have a thousand reasons to smile

Get in touch

© Copyright 2015 - 2024 PDFFOX.COM - All rights reserved.