Theyaskifweeatfrogs GaroEthnicityinBangladesh

See discussions, stats, and author profiles for this publication at: https://www.researchgate.
net/publication/288756780
They ask if we eat frogs: Garo ethnicity in Bangladesh
Article · January 2007
CITATIONS READS
36 3,916
1 author:
Ellen Bal
Vrije Universiteit Amsterdam
46 PUBLICATIONS 390 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Migration, livelihood and SRH of young female migrants at Dhaka View project
PEOPLE: People-Centred Development Approaches in Practical and Learning Environments View project
All content following this page was uploaded by Ellen Bal on 09 February 2017.
The user has requested enhancement of the downloaded file.

The International Institute for Asian Studies (IIAS) is a postdoctoral research
centre based in Leiden and Amsterdam, the Netherlands. Its main objective is to
encourage the study of Asia and to promote national and international co-operation
in this field. The geographical scope of the Institute covers South Asia, Southeast
Asia, East Asia, and Central Asia. The institute focuses on the humanities and the
social sciences and, where relevant, on their interaction with other sciences.
The Institute of Southeast Asian Studies (ISEAS) was established as an autonomous

organization in 1968. It is a regional research centre dedicated to the study of socio-
political, security and economic trends and developments in Southeast Asia and its
wider geostrategic and economic environment.
The Institute’s research programmes are the Regional Economic Studies (RES,
including ASEAN and APEC), Regional Strategic and Political Studies (RSPS), and
Regional Social and Cultural Studies (RSCS).
ISEAS Publishing, an established academic press, has issued more than 2,000
books and journals. It is the largest scholarly publisher of research about Southeast
Asia from within the region. ISEAS Publishing works with many other academic
and trade publishers and distributors to disseminate important research and analyses
from and about Southeast Asia to the rest of the world.
00 TheyAsk_EatFrogs Prelims By: ROS 2 Size: 6" x 9" J/No: 06-15540 5/7/07, 4:46 PM Fonts: AGaramond
First published in Singapore in 2007 by ISEAS Publishing
Institute of Southeast Asian Studies
30 Heng Mui Keng Terrace
Pasir Panjang
Singapore 119614
E-mail: publish@iseas.edu.sg
Website: http://www.bookshop.iseas.edu.sg
First published in Europe in 2007 by

International Institute for Asian Studies
P.O. Box 9515
2300 RA Leiden
The Netherlands
E-mail: iias@let.leidenuniv.nl
Website: http://www.iias.ni/
All rights reserved. No part of this publication may be reproduced, stored in a

retrieval system, or transmitted in any form or by any means, electronic, mechanical,
photocopying, recording or otherwise, without the prior permission of the Institute
of Southeast Asian Studies.
© 2007 Institute of Southeast Asian Studies, Singapore.
The responsibility for facts and opinions in this publication rests exclusively with the
author and her interpretations do not necessarily reflect the views or the policy of the
publishers or its supporters.
ISEAS Library Cataloguing-in-Publication Data

Bal, Ellen.
They ask if we eat frogs: Garo ethnicity in Bangladesh.
1. Garo (Indic people)—Bangladesh—History.
2. Garo (Indic people)—Bangladesh—Ethnic identity.
3. Minorities—Bangladesh.
I. Title
II. Title: Garo ethnicity in Bangladesh
DS393.83 G37B17 2007
ISBN 978-981-230-446-9 (hard cover)

ISBN 978-981-230-591-6 (PDF)
PHOTO CREDIT: The photograph used on the front cover is reproduced with the
kind permission of the author, Ellen Bal.
Typeset in Singapore by Superskill Graphics Pte Ltd

Printed in Singapore by
CONTENTS
Acknowledgements vii
Glossary xi
1 Introduction 1
2 The History of a Persistent Image 27
3 ‘The Importance of Being Garo’: Garo Narratives of Self 50
4 Peoples without History? 67
5 ‘Dual were Dual, Kochu were Kochu’: Garos Divided 88
6 Negotiable Boundaries, Negotiable Identities 111
7 Garos and Christianity 132
8 Garos and the State 158
9 Summary and Conclusion: From Tribes to Ethnic Minorities 209
References 217
Index 233
About the Author 243
ACKNOWLEDGEMENTS
They Ask if We Eat Frogs is a revised and updated version of my Ph.D. thesis
which I defended in June 2000 at the Erasmus University in Rotterdam and
which appeared under the title “‘They Ask if We Eat Frogs’: Social Boundaries,
Ethnic Categorization, and the Garo People of Bangladesh”. Although I
opted for a new subtitle to the book, I chose to keep the main title the same.
To anyone familiar to South Asia, the question whether one eats frogs needs
no further explanation. It symbolizes a whole world of prejudices and
hierarchical relations between South Asia’s ‘majority populations’ and so-
called tribal minorities. ‘Frog eaters’ supposedly stand for the most primitive
of South Asians, those who have not learnt ‘civilized’ behaviour. To ask if one
eats frogs goes beyond ‘objective’ curiosity. No matter how naive the inquirer,
notions of inequality and simplicity underlie the question, and result, at the
least, in feelings of embarrassment and uneasiness of the person addressed. In
other words, ‘they ask if we eat frogs’ is the briefest possible summary of
unequal and uneasy relations that have for decades marked Garo images and
self-perceptions, and their relations to many others, non-Garos.
For this project I carried out research between 1993 and 2000 and visited
Bangladesh more or less on a yearly basis. Between March 1994 and March
1995, I spent a full year in the field. In that year I established most of my
contacts with Garos and others, formed a number of valuable friendships,
and experienced a heartwarming hospitality and willingness of Garos to
welcome me into their lives and share their thoughts with me.
During the past thirteen years I met numerous people who contributed
in all sorts of ways to this outcome. Without their help and support this book
would not exist. Hereafter I shall mention a number of them without even
the smallest suggestion that the list includes all. My gratitude first goes out to
the Garos in Bangladesh who opened their houses and hearts to me; who
willingly spoke up and shared their life stories with me; far too many to be
mentioned by name. Without their generosity, curiosity, enthusiasm and
hospitality this research would not have been possible. I would especially like
to thank the villagers of Bibalgree who time and again welcomed me and the
viii Acknowledgements
occasional friends and relatives I brought along. I am particularly indebted to

mamu Nehru, mani Chameli and their daughter Puti, who took me into their
family and treated me as one of theirs.
I am very grateful to the brothers of Taizé in Mymensingh. The gate to
their home was always open. They inspired me by their work and commitment
to the people in Bangladesh. Here I would particularly like to thank
Br Guillaume, always full of stories and always out of time to share them all,
possessing the priceless talent to see the beauty in all and everything. Br Erik
helped me in a number of ways, among others, as co-editor of the book of
photographs Manderangni Jagring: Images of the Garos of Bangladesh and as a
travel companion during one of my many enjoyable journeys through
‘Garoland’. It was always good to discuss my research with him.
In Dhaka, Professor Anwarrulah Choudhury helped me to navigate the
inevitable bureaucratic hurdles and to obtain a research visa. Professor Sirajul
Islam, Dr Ratan Lal Chakraborty and Dr Kibriaul Khaleque, whose sudden
passing away came decades too early, offered me invaluable advice on the
research project.
My friends in Dhaka made my stay even more enjoyable. I was very
fortunate to find a pleasant and stimulating environment in which to live
with the volunteers of the Swedish Swallows, Karin and Maria, who also
invited me to join their large circle of friends. Others who need to be
mentioned here are the many members of the Chisim family of T.B. Gate,
Bernard, Eva, Ineke, Juise and Zeba, Sanjeeb, Sister Elizabeth, Sumon, Tanvir
and the workers of Nokmandi.
I also want to thank my two colleagues in the ‘Garo field’, Robbins
Burling and Erik de Maaker, and my colleagues at the Faculty of History and
Arts of the Erasmus University of Rotterdam, in particularly Andrea Tyndall,
Arjan de Haan, Bart Hofstede, Gijsbert Oonk, Hanneke Hoekstra, Jeroen
Blaak, Jolien Harmsen, Karin Willemse, Manon van der Heijden, P.W.
Zuidhof, Rick Dolphijn, Stef Scagliola, Ulbe Bosma, and Wilfred Dolfsma.
The faculty funded much of my research, at home and abroad. Additional
funding was provided by the J.E. Jurriaanse Stichting, Netherlands Foundation
for the Advancement of Tropical Research (WOTRO), and the Trustfonds.
Since August 2001, I have found myself a new circle of colleagues. The
Department of Social and Cultural Anthropology offers a friendly and
stimulating environment to work in. I came to know my new colleagues too
late to share many of my ideas about this particular project with them,
although some of their influence has seeped into this revised edition, but I am
looking forward to working with them and to be part of each other’s projects
in future.
Acknowledgements ix
I feel particularly indebted and thankful to my research companion

Suborno Chisim, who jokingly referred to me as “boss”, but who really was
my research partner. Suborno turned out to be much more than a good
interpreter, research assistant and key informant. He became a very good
friend and as committed to the project as I was. I feel that this book is his as
well as mine. Willem van Schendel was much more than a distant supervisor
to this project. He also inspired my work through his own and the force of his
ideas. His ongoing support and belief in the relevance of the project and its
outcome were fundamental to a positive outcome.
I want to thank Wim Stokhof, Paul van der Velden and Marloes Rozing
of the International Institute for Asian Studies (IIAS) in Leiden, for encouraging
and enabling me to have my work published. At the ISEAS, Triena Ong and
particularly Fatanah Sarmani were responsible for bringing it to a good end.
I experienced that my life as a researcher is inextricably bound up with
my life as a whole. To feel well is to work well. I am particularly thankful to
my friends Annemarie Uhlenbeck, Esther van Beelen, Fennie Posthumus,
and Sandra Bos, who supported me with their friendship and enthusiasm,
and to Kathinka Sinha-Kerkhoff, who encouraged me both with her friendship
and with her intellectual support. I hope we will continue working together
for many years to come. Finally, thanks to my sisters Marlon and Pien and my
parents Nellie and Simon Bal, who spent some unforgettable weeks in
Bangladesh. Erik and Jip, let us keep travelling, together.
GLOSSARY
The following is a glossary of Bengali, Garo, and English terms as they are
used in the text. Most of the terms are Bengali. The Garo terms are indicated
with a (G)
Abeng Division of Garos and language spoken by Garos in

the lowlands of Bangladesh (G)
Achik Lit. hill, or hill man; also name of a division of Garos;
language spoken by Garos in the hills (G)
Achu Grandfather, also used for elderly men, or forefathers
(G)
Adivasi Lit. original inhabitant; commonly used as a synonym
for the communities known as tribes
Ajong Mother’s elder sister (in Abeng language) (G)
Ambi Grandmother, also used for elderly women or
foremothers (G)
Asong Country (G)
Atong Division of Garos living in Bangladesh; also name of
their language (G)
Bangal Derogatory naming of Bengali (used by Garos)
Bazaar Marketplace
Borkondazes Armed retainers
Bormon Caste title of many Hindu minorities in East Bengal
Chatchi Kin group; also relative or guest (G)
Choudrie Landlord; revenue officer
Chu Rice beer (G)
Dakari.ka Culture (G)
Desh Country
Dol Group (in Bengali language). Here it is also used to
refer to particular Garo groups or divisions
Fakir Muslim mendicant or missionary
Fouzdar Mughal tax collector
xii Glossary
Gari Car, cart

Ghor jamai Son-in-law of the house. Husband moves in with his
in-laws
Gotro Clan or tribe
Hartal General strike
Hat Market
Hookah Water pipe
Jat Most commonly translated as caste; other meanings
are endogamous community, religious, social or ethnic
group
Jati See: jat
Jhum System of cultivation by which jungle on the hills is
cut down, generally in the cold season (December–
February). The remains of the trees are left until March
when they are burned
Jumma Collective name of the inhabitants of the Chittagong
Hill Tracts
Kaka Paternal uncle
Kushuk Language or dialect (G)
Lungi Male circular, ankle-length, skirt-like garment worn
by men; it is also worn by Garo women who fasten it
differently from men
Ma’chong Matrilineal kin group (G)
Madong Lit. married to mother; refers to forbidden marriages
within exogamous kinship group (G)
Mahajan Moneylender
Mahari See: ma’chong
Mama Maternal uncle
Mandi Lit. man, human being; Garos of Bangladesh call
themselves Mandi (G)
Mani Maternal aunt
Marak Garos are primarily divided into three exogamous
groups: Sangma, Marak, and Momin (G)
Midi amua Sangsarek ritual in which supernatural beings are
worshipped (including Gods and lesser spirits) (G)
Mimang kam Ritual ceremony (G)
Mimang Spirit (G)
Moghul Islamic Empire in India from the sixteenth to the
nineteenth century
Momin See: Marak (G)
Glossary xiii
Mujibite Follower of Sheikh Mujibur Rahman

Naib Lit. a deputy, landlord’s agent and rent collector
NGO Non-governmental organization
Nokma Chief of a clan (G)
Nokna Garo heiress (G)
Pagol Mad, lunatic
Pahari Hill people; also indication of one particular minority
Pan Betel-nut
Pandit Hindu priest
Para Village neighbourhood, hamlet
Pargana (or pergunnah) Revenue division; an estate
Partition Creation in 1947 of two independent states – India,
with a Hindu majority, and Pakistan, with a Muslim
majority – out of the territory formerly ruled by the
British
Pi’sa Child (G)
Puja Worship, adoration
Purdah Lit. seclusion of women, custom of segregating the
sexes
Raja King; sometimes an estate-holder, or a landlord, or a
chief
Razakaar Collaborator with the Pakistan army
Ruri Garo way to refer to Bengalis, irrespective of their
religion
Ryot (raiyat) Tenant
Sangma See: Marak
Sangsarek Religion of the Garos; believer in the Sangsarek religion
Sannyasi Hindu mendicant or missionary
Sanskriti/kriti Culture
Sarkar Government
Scheduled Tribes Tribes listed by the Indian Government after
Independence
Shongho Organization, association
Shomaj Community
Sthaniyo Bengali for local (also tanio, taina)
Swaraj Self-rule
Swarajist Supporter of self-rule
Taka Bangladesh currency
Tanka System of cropsharing whereby a tenant had to pay a
fixed quantity of crop to the landlord
xiv Glossary
Tebhaga Lit. three shares; movement by sharecroppers for two-

thirds of crops from landholders
Thana Administrative unit in rural Bengal; a police station
Upojati Lower order of the nation; sub-nation
Wana krita ceremony in which sacrifice to spirits is made (G)
Wangala Garo harvest festival (G)
Zamindar In Bengal, holder of a large estate, usually one farmed
by sharecroppers or sub-tenants; created by the British
administration as part of a feudal system of revenue
collection
Zenana Seclusion of women; separate quarters for women which
men cannot enter (except for close relatives)
1
INTRODUCTION
FIXED CATEGORIES, FROZEN IMAGES

During a brief visit to Bangladesh, in December 2005, I spotted a recent
edition of the national encyclopaedia (the Banglapedia), in the showcase of a
professor of history at Rajshahi University.1 I eagerly searched for a reference
to the Garos, among who I had carried out research between 1993 and 2000,
and to my satisfaction, discovered four whole pages devoted to their history,
culture, religion, and habitat. To me this meant they have a place in the
national representation of Bangladesh. My pleasure, however, at once gave
away to disappointment, as I discovered one more image frozen in time; one
that might as well be found in a nineteenth-century colonial report on Garos
or another tribe. The Banglapedia described the Garos as follows:
Their faces are round, hair and eyes black, foreheads extended to eye area,
eyebrows deep, eyes small, noses flat and jaws high. Beards rarely grow on
their cheeks and they almost have no hair on their body. … The natural
habitats of the Garo people are the hills, hillocks, deep forests and places
near fountains, springs, and other water bodies. Animals, reptiles and birds
are their closest neighbours…. MIRZA NATHAN, a Mughal army
commander, remarked that Garos eat everything except iron. There is some
exaggeration in this statement but in fact, they eat all animals except cats,
which is their totem. They live in an isolated world and within their own
geographic, economic and cultural boundaries and follow their own
customary norms.2
1
01 TheyAsk_EatFrogs Ch 1 By: ROS 1 Size: 6" x 9" J/No: 06-15540 4/16/07, 10:40 AM Fonts: AGaramond
2 ‘They ask if we eat frogs’: Garo Ethnicity in Bangladesh
This is a common depiction of a so-called tribe in South Asia. Hundreds of

other groups that fall into the tribal category are often described in a similar
vein, both in popular and administrative publications, and frequently also in
academic accounts. Somehow, dominant discourse of tribe has undergone
noticeably little change since the British began their arduous (and impossible)
task to classify the Indian population into neat categories of castes, tribes,
religions, races, etc.
This book is about social categories, images, and identities. It is about
their construction and disappearance, their malleability and continued
existence. We cannot think about the world unless we imagine ourselves and
others in separate compartments. The recognition that any category does
violence to the complexity of social reality does not change the fact that some
classifications, better than others, serve to understand social processes and
identity formation. Serious problems arise when categories, and consequentially
the boundaries that separate them, are taken for granted and are viewed as
timeless and unchangeable. When their labels carry notions of inferiority and
no longer correspond with experienced realities. When, as various contexts in
Asia and elsewhere have shown, social categories take on a life of their own,
once authenticated by the state or science or both.3
Globally, race is a social category that still wields monumental power.
Even though it is as much a product of human perception and classification
as nations or ethnic communities are, “in many societies, the idea of
biologically distinct races remains a fixture in the popular mind, a basis for
social action, a foundation of government policy, and often a justification
for distinctive treatment of one group by another”.4 Race exists as a cultural
construct and informs people’s actions, whether it has a “biological” reality
or not.5 Tribe is another notion that often carries primordial and essentializing
connotations, yet has a great bearing on social reality. As a social category,
it greatly resembles popular perceptions of race. The present study is a
historical investigation into the category of so-called tribes of South Asia,
or, as they are being called in a more politically correct fashion, “Indigenous
Peoples” or adivasis.
Willem van Schendel pointed out that the ways in which South Asian
tribes have been depicted – as if they share a number of “essentially tribal
characteristics” that are fundamentally different from, even opposite to,
“civilized” society – show a striking similarity with Orientalist representations
of people from the Orient, as described by Edward Said.6 Similarly, as
suggested by Van Schendel, can we refer to the “complexes of signs and
practices which organize social existence and social reproduction” of so-called
tribes, as the tribalist discourse.7
Introduction 3
In his seminal work entitled Castes of Mind, Nicholas Dirks argues

that a historical analysis of caste shows that “caste (again, as we know it
today) is a modern phenomenon, that it is, specifically, the product of an
historical encounter between India and Western colonial rule.” He
continues that “it was under the British that ‘caste’ became a single term
capable of expressing, organizing, and above all ‘systematizing’ India’s
diverse forms of social identity, community and organization.”8 Dirks and
other scholars have made a significant contribution to the scholarship of
caste through careful scrutiny of India’s colonial and modern history.9 I
shall come back to this in greater detail in the next chapter. Here it is
important to note that a similar transformation of tribal studies has not
taken place. Tribalist discourse has continued to have a great impact on
much of the academic research, policy making, and popular imagination
of tribes in South Asia.
Nevertheless, an increasing number of scholars have indeed questioned
the usefulness of the concept as an analytical tool and expose its many
downsides.10 Among them, researchers of hill tribes or upland people in
Southeast Asia have made particularly insightful contributions. Since
Edmund Leach’s seminal publication on the highlands in Myanmar, in
which he convincingly demonstrated that processes of group formation and
identification in the hills were far more complex than the earlier essentialist
tribal studies suggested, many have continued to criticize the concept of
tribe and, since more recently, also the dualistic framework of mutually
exclusive categories in which tribes/uplanders and lowlanders usually are
studied.11 Unfortunately, however, social scientists have divided the world
into convenient academic areas. Even though “a rethinking of ‘regional’
systems of knowing is under way”,12 tribal studies in South Asia have only
just benefited from the empirical and theoretical insights produced by their
colleagues working on Southeast Asia.
TOWARDS A SITUATIONALIST APPROACH OF TRIBE

This book is based on the premise that tribalist representations should not
simply be replaced with more nuanced historical and ethnographic accounts.
They also need to be scrutinized in connection with the contexts in which
they are produced and the purposes that they serve. In other words, we need
to examine why and how tribalist discourse has come into being, and how
representations of tribes/tribals have served particular agendas (of themselves
and others) and have had real effects in the shaping of self-perceptions,
identities, and development.13
For this very purpose, this study has adopted a situationalist approach to
culture and identity. It proposes a conceptual shift from tribe to ethnic group
or community, and from culture to boundary.14 Notwithstanding the ongoing
debate, contemporary conceptions of ethnicity generally incorporate a dynamic
and relational perspective. They examine, but do not presuppose, prevalent
notions of superiority and inferiority. A change from a tribal to an ethnic
perspective thus means a shift from a static to a dynamic approach to social
groupings and boundaries, without the construction of yet another generic
term which has incorporated notions of homogeneity and inferiority.
This research is heavily indebted to Barth’s classic work, Ethnic Groups and
Boundaries. Almost four decades ago, Barth developed a model for the study of
ethnic relations that conformed to Leach’s suggestion to remove culture from
the front stage of ethnic studies.15 He proposed a change of focus from “the
cultural stuff that it encloses” to the “boundary that defines a group” and argued
that continuity of a group depends on the maintenance of that boundary rather
than on the cultural characteristics of that group.16 Barth still holds a special
position in the studies of ethnicity and is often held as the first who introduced
the shift of focus from static to interactional approaches to ethnicity.17
Barth’s understanding of ethnicity also implied that ethnic groups are
defined from within, from the perspective of their members. Instead of listing
traits of “objective culture”, which members often share with non-members
anyway, he defined ethnicity as categorical ascriptions which classify individuals
in terms of their “basic, most general identity”.18 This also suggests that
cultures may change without removing ethnic boundaries. Or, the other way
around, that old boundaries may disappear and new ones come into existence,
without significant cultural change preceding such changes. In other words,
boundaries produce culture. Since Barth, with this seminal work on ethnicity,
made a path-breaking contribution to the thinking on ethnicity, many scholars
have insisted that the relation between culture and ethnicity is far more
complex than Barth suggested; that we also need to take the limitations of
choice and freedom into account. Culture matters too, and, in the words of
Thomas Eriksen, “ethnic identities are neither ascribed nor achieved: they are
both. They are wedged between situational selection and imperatives imposed
from without.”19 And thus, from the many available definitions of ethnicity,
the following is picked, which defines an ethnic group as a “reference group
invoked by people who share a common historical sense (which may only be
assumed), based on overt features and values, and who, through the process
of interaction with others, identify themselves as sharing that style”.20
A major drawback of anthropological studies of ethnicity is that very few
of them “really undertake the task of showing how ethnic distinctions emerge
Introduction 5
in an area; how initially homogeneous groups are historically split into two or
several distinctive ethnic groups”.21 (Or, how lose collections of distinctive
groups develop into one ethnic community). However, in order to understand
ethnicity, a historical perspective is fundamental. Only then can we uncover
the complex, socially, economically and politically embedded, processes of
identity formation and cultural change.22
Anthropological research on tribes in South Asia commonly focuses
on the “cultural stuff ” without taking historical processes of boundary
construction or the wider socio-economic and political contexts into account.
Historians of South Asia, on the other hand, not often study tribes. Therefore
this book breaks away from previous studies of tribes. It neither studies
Garo culture as such, nor their social organization. Instead it deals with the
evolution of Garo identity/ethnicity and with the progressive making of
cultural characteristics that support a sense of “Garo-ness”, in the context
of the complex historical developments in South Asia and the world.23
By means of such a historical examination, it hopes to contribute to
contemporary research on South Asia’s tribes. It addresses the following
central research questions: What does the particular history of the Garos of
Bangladesh tell us about processes of identity formation in the region?
What does it tell us about prevailing notions of tribe in South Asia, and the
usefulness of tribe as an analytical category? And how can close scrutiny of
their history help us understand contemporary minority-majority conflicts
and even violence in the region?
WHY A HISTORY OF THE GAROS OF BANGLADESH?

While largely absent in the national history of Bangladesh as agents, rather
than frozen images, the Garos are hardly a new subject of interest. Since
John Eliot’s encounter with the Garrows in 1788–89, “as the first European
who has travelled among them”, numerous books and articles about Garos
have been published.24 The publications, however, mainly deal with the
uplanders living in the Indian state Meghalaya, in the district named Garo
Hills.25 They have formed the majority of all people known by the name
Garos. The lowland Garos of East Bengal have rarely been studied. Articles
and books are few and references in administrative reports are scarce. These
lowland Garos, however, have their own history(s). An international border
has separated them from hill Garos since 1947. The partition resulted in a
much stricter division than ever before. Although transboundary mobility
has never stopped, Indian and Bangladeshi Garos seem to be increasingly
developing into different directions. Nevertheless, differences between these
Garos have existed much longer.26 Lowland Garos have long been in contact
with Bengali culture; the natural environment (and climate) of the plains
require different agricultural methods and a different style of living; the
political status of the two regions also differed long before partition.27 At
present, the segmentation into (Indian) hill Garos and (Bangladeshi) lowland
Garos is also reflected in the names they give themselves. Bangladeshi Garos
call themselves Mandi, which means “human being”. They refer to the
Garos from the Garo Hills as Achik [hill person]. Such differences between
(Indian) hill Garos and (Bangladeshi) lowlanders legitimize the subject of
this investigation (Garos of Bangladesh) and clearly underline that Garos
have never constituted a single ethnic community.28
This research is not the kind of borderland study as advocated by Van
Schendel in his excellent work on the Bengal Borderland.29 My project was,
by and large, limited to the Bengal/Bangladesh side of the border that
separates the Indian (hill) state Meghalaya from the plains of Mymensingh in
Bangladesh. At the time of my field research, the Indian part of the Garo
borderland was closed to foreigners.30 Equally important was my special
interest in the situation of tribal communities in Bangladesh, who have
received far less attention than “the tribes of India”. I did however not begin
my study from the definition of a state as a natural self-enclosed unit.31 I tried
to show how not only the ongoing process of partition, but also the presence
of an international border, has had a major impact on the lives of the Garos
of Mymensingh. Chapter 8 provides a detailed analysis of that process. Here
I wish to limit myself to the claim that the 1947 Partition turned the Garo
tribe into an ethnic minority. That is, if we define minority in the words of
Eriksen, as “a group which is numerically inferior to the rest of the population
in a society, which is politically non-dominant and which is being reproduced
as an ethnic category”. Eriksen continues that “a minority exists only in
relation to a majority and vice versa, and their relationship is contingent on
the relevant system boundaries. In our present-day world, such boundaries
are almost always state boundaries. Majority-minority relations change when
state boundaries change.”32
Unlike some 90 per cent of the Bangladeshis, people like the Garos and
other so-called tribal communities are neither Bengali by ethnicity and culture,
nor Muslim by religion. The history of these “other peoples of Bangladesh”33
is one marked by “othering”, and by its extreme consequence: exclusion.34
Present-day Bangladesh has some 150 million inhabitants, Bangladeshis. It is
important to distinguish between Bangladeshis and Bengalis. While
Bangladeshis includes all citizens of the country, Bengalis refers to its dominant
ethnic population (some 98 per cent), who mostly live in Bengal, speak the
Introduction 7
Bengali language, and “feel” Bengali. They are not only found in India’s West
Bengal and Bangladesh, but also in Assam, Tripura, and other parts of India.
In contrast to their counterparts in West Bengal, who are mostly Hindus,
Bangladeshi Bengalis are predominantly Muslim.35
The partitioning of India in 1947, which resulted in the creation of India
and Pakistan (and since 1971, also Bangladesh) has had a fundamental
impact on the complex processes of identity formation in the region. In the
region of my investigations, much research has concentrated on Bengali
culture, self-identification and social arrangements. The so-called “other peoples
of Bengal” (and of Bangladesh) have no place in the contemporary written
history of the region. Their marginalized position is also reflected by the fact
that the number of groups and total population are not exactly known.
According to the population census of 1991, they make up more or less one
per cent of all Bangladeshis (1.2 million people). There are, however, reasons
to believe that this number is an under-estimation. Similarly, no one can say
exactly how many different ethnic groups there are. Estimates range from
twenty to fifty-six.36
I do not consider the peripheral position of these “other peoples of
Bangladesh” as a proper reflection of their numerical marginality or merely as
a short-coming of contemporary researchers and politicians, but as a serious
situation, which has many important socio-economic, cultural, and political
consequences. Their marginal position is not proportioned to their place in
national and regional history, in which non-Bengali peoples are practically
invisible.37 This study gives non-Bengali minorities a place in the (national)
history of the region.
TRIBALIST DISCOURSE AND ITS TERMINOLOGY

In Bangladesh, the most common alternatives for tribe are Scheduled Tribes,
Indigenous Peoples, aboriginals, adivasis, upojatis, paharis, and Jummas. Close
scrutiny of each term reveals a great number of drawbacks and misconceptions.
Below is a brief examination of each of these concepts.38
TRIBE
The English term tribe or tribal was first introduced in Bengal by British
colonial administrators and foreign anthropologists. Since then the word has
become more or less incorporated in the Bengali language. Many scholars
have pointed out that the term tribe in itself is highly problematic.39 There is
not one single definition which can be applied in the South Asian situation.
People generally understand who are the tribes and who are not. Nonetheless,
when they are asked to describe what a tribe actually means, the picture
which is presented rarely corresponds with the real situation. So-called “tribal
characteristics” such as being primitive, isolated, simple, undeveloped, believers
in local religions, or having specific tribal political and economic arrangements
rarely make any sense. Moreover, as we will see in Chapter 2, the term carries
clear evolutionist connotations. It refers to people in the “archaic stage of
development” and implies an image of “primitive” man as opposed to “civilized”
man. Tribals are thus seen as “inferior races”.40
While some authors argue that tribe was essentially a colonial construct,
a colonial category, others feel that this is not quite consistent with the ideas
about tribe in earlier anthropological writings.41 Whether a colonial construct
or not, its relevance did not, however, dissolve with the end of colonialism.
Much contemporary research shows remarkable similarities with colonial
ethnography. It has been noted that these studies are not only written in a
style which is reminiscent of colonial ethnography, but also from a functionalist
or even evolutionist perspective.42
THE SCHEDULED TRIBES OF INDIA
The partition of 1947 and the subsequent division into India and Pakistan
marked the beginning of distinct political developments in two (and
three, since the independence of Bangladesh in 1971) different countries.
In post-colonial India special national and state policies were formalized
in the constitution to “uplift” the “backward tribes”. The post-colonial
state showed the same systematizing urge that its colonial precursor had
displayed. In order to identify the people who qualified for preferential
treatment, an extensive list of all tribes was prepared. These people
have since been referred to as Scheduled Tribes (ST).43 In 1952, the
Commissioner for Scheduled Castes and Tribes admitted that no uniform
test to classify the Scheduled Tribes had been developed. Difficulties were
experienced in determining which tribe was to be included or excluded
from the schedule of tribes. Nevertheless, the commissioner did feel that
three features were common to all tribal people: a) they had a tribal
origin, b) they had a primitive way of life and habitation in remote and
less easily accessible areas, and c) they showed general backwardness in all
respects.44 The tautological nature of catchwords such as “tribal origin”,
“primitive” and “backwardness” to define tribes will be clear from the
foregoing. In India the designation “Scheduled Tribe” has gained enormous
socio-economic and political importance because of the special facilities
Introduction 9
which are provided for people who are included. In East Pakistan and
present-day Bangladesh, the state never bothered to collect systematic
information on its tribal population and never developed formalized
policies regarding “backward” groups.
INDIGENOUS PEOPLES AND ‘ABORIGINES’

Quite recently, another English term has gained popularity in both English
and Bengali: Indigenous Peoples. This expression has become particularly
popular since 1993, the United Nations Year for Indigenous Peoples. The
term has much relevance for it links South Asian groups to a large number
of peoples all over the world and provides them with a sense of a shared
identity which exceeds local, regional, or national boundaries. “Indigenous
Peoples” was a clear concept in the Americas, where it was developed and
first gained currency: the Amerindian population was indeed indigenous
compared to the later immigrants from Europe, Africa, and Asia. In South
Asia, the concept is less clear, as recent immigration from other continents
is not the issue and the idea of indigeneity is contested. Nevertheless,
theories of historical settlement continue to surface. Since a decade or so,
the status of indigenous people has also been claimed for Dalits on the
strength of the argument that Dalits were the original inhabitants of South
Asia and are a distinct people with their own culture.45
In a way, the term is very similar to the old concept of “aborigines” or
“aboriginal tribes”, introduced by the British during the colonial period.
Its connotations and social and political implications, however, differ
greatly. The British used the term aborigines to distinguish so-called
primitive peoples from the “modern Indian”, with the underlying
assumption that aborigines belonged to older, less advanced strata of
population who had somehow failed to keep up with progress. The
category did not just include people known as tribes today but also
different castes which were seen as having to some extent assimilated into
the surrounding Hindu culture. The term “Indigenous Peoples” shares
with “aborigines” the idea of early settlement but points to the old rights
to land and other resources flowing from that early settlement. In this
way, “Indigenous People” has become a marker of emancipation and
empowerment, and it provides a basis for worldwide networks of
organization and action. At present, popular issues such as human rights
and environment are attached to the category of Indigenous Peoples and
happily employed by the people included in this category to be on
political, socio-economic, and cultural agendas.
In Bangladesh, it is often argued, without much historical evidence, that

so-called tribal people are no more indigenous to the country than Bengalis.
In fact, the dominant discourse often reverses the positions by claiming that
Bengalis are the truly indigenous people of Bangladesh and others, such as
Garos, Santals, Tripuras, or Koches, are immigrants and therefore cannot
claim to be indigenous to Bangladesh.46 Interestingly, it is not the issue of
indigeneity but that of being socially and culturally marginalized which has
come to dominate the discussions about the Indigenous Peoples of Bangladesh.
ADIVASI
Adivasi is a common term in both Bengali and English.47 In Bangladesh, the
term was not popular until quite recently. Adivasi, coined from the Sanskrit
“adi” (meaning “beginning” or “of earliest times”) and “vasi” (meaning
“residents of ”), is a close translation of “aborigine”. It also incorporates the
idea that the adivasi are the original inhabitants of South Asia.48 Other
authors who prefer adivasi over tribe legitimize its application by the fact that
the term relates to a particular historical development which generated a
shared spirit of resistance that incorporated a consciousness of the adivasi
against the outsider.49 At present, adivasi has important political potency. We
could say that tribes have rejected their passive and exotic role and adopted
the role of a self-conscious actor on the social stage.
UPOJATI, PAHARI, AND JUMMA
Bengali terms for non-Bengali groups are upojati, pahari, or Jumma. Upojati
has connotations similar to the English word “tribe”. It refers to uncivilized,
less developed, and innocent peoples who live more or less isolated from the
“mainstream” of “civilized” Bengali society. The term is increasingly being
rejected by the peoples concerned. They argue that upojati is a derogatory
concept which suggests that they are of a lower order than the Bengalis, who
form a jati or nation, whereas an upojati is a mere sub-nation.
Pahari (or paharia) refers to both tribes in general and people who live
in hills or mountains. It has been argued that the name was originally given
by lowlanders to their hill-dwelling neighbours in the Rajmahal Hills in
Bihar.50 In Bangladesh, Pahari is also used to refer to one specific community
in northwestern Bangladesh. Pahari is thus a term which has once been
imposed by others on the basis of environmental aspects. The peoples
themselves use different designations. Yet hill-dwelling people have also
utilized the term pahari to forge a common identity and to distinguish
themselves from plains people.51
Introduction 11
A similar example is Jumma. Jhum refers to a particular type of cultivation,

and Jhumia (in Chittagonian dialect and Chakma language: Jumma) to the
cultivators. In the 1970s, Jumma was appropriated by the regional political
party of the Chittagong Hill Tracts, the JSS (Jana Sanghati Samiti), to refer
to all inhabitants of the Chittagong Hill Tracts. Jumma has become a way to
distinguish the inhabitants of the hilly regions of southeastern Bangladesh
from Bengali settlers in that area. Therefore the term refers to a shared
experience of marginalization, exploitation, and militarization, and to a social
or solidarity movement. It embraces groups of various linguistic, cultural, and
religious backgrounds.52
A crucial aspect of each one of the afore-mentioned terms is the assumption
that all people designated as “tribal” or “indigenous” somehow share basic
political, social, economic, and cultural characteristics with each other, enabling
scholars and policy-makers to group them all together into a single category.
But the essence of what tribal or indigenous might mean remains highly
elusive, and therefore undefined. This makes the category a catch-all for all
groups considered to be distant from the “observing self ”: it is a category
which explains more about the categorizer than about the categorized. This is
not to say that a shared designation cannot generate a sense of belonging to
a large (even worldwide) category of “like-minded” people. As we shall see in
the following chapters, the category that was once imposed on people
considered as tribal came to structure parallel experiences of marginalization
and discrimination among them, and was often appropriated by them as a
badge of identity.
THE GAROS OF BANGLADESH

At present, the Garos constitute less than ten per cent of the “other peoples”
of Bangladesh, an extremely marginal segment of the total Bangladeshi
population. Of these 80,000 to 100,000 Garos, a little over 14,000 people
live in Modhupur.53 This forested area is at least some fifty kilometres from
the Indian border. Stories about Bangladeshi Garos are generally about them.
This study largely concentrates on the people who are living in the
northernmost portion of the Greater Mymensingh district54 and on migrants
in Dhaka.55
Today, Garos can be found all over Bangladesh. For instance, since the
1950s, they have migrated to the betel leaf and tea plantations in Sylhet.56 At
present, Sylhet probably has around 7,000 Garos, but no one knows for sure.
From the 1960s onwards, and especially since the 1980s, many Garos have
also started to settle in Dhaka and in Chittagong, Bangladesh’s principal
cities. Dhaka is particularly popular. Many people – both men and women –
leave their villages out of necessity for jobs, others for education, or simply for
the adventure. Today, a couple of thousand Garos live in Dhaka. Many of
them are employed as household servants and in garments factories. Others
have found their way into the offices of NGOs and church organizations in
the capital. The latter often have their families living with them in the city.
Many of these people are doing well. Students form a third important section
of the Dhaka Garo community. More and more young men and women are
coming to Dhaka to study at one of the many colleges or at Dhaka University.
Since the 1960s, Garos have migrated to cities such as Mymensingh,
Chittagong and especially Dhaka in ever increasing numbers. They leave their
villages to look for work or to follow higher education (at colleges and
universities). Exact figures are not known.57
Present-day Bangladeshi Garos set themselves apart from the many other
ethnic communities of Bangladesh. We can safely argue that they form a
distinct ethnic community. They share feelings of being different from other
Bangladeshis and of belonging together, and ascribe their distinct identity to
a shared culture and traditions, language, history and experiences. In reality,
of course, like all of us, they are “complexly constructed through different
categories, of different antagonisms, and these may have the effect of locating
[them] in multiple positions of marginality and subordination, but which do
not operate on [them] in exactly the same way”.58 Nevertheless, hereafter, I
shall briefly give some background information on the Garos of Bangladesh.
I shall also introduce a number of characteristics that they themselves stress in
order to distinguish themselves from neighbouring Bengali- and non-Bengali
communities. I realize that such a description does great injustice to the
complexity of the Garo community, but feel that the unknowing reader needs
some information in order to have a handle on the subsequent writing. I hope
to make up for the simplicity of the ensuing description in the next chapters,
in which the process of ethnogenesis (the emergence of their ethnic identity) is
closely examined. The focus on one ethnic community was not based on any
preconceived notions of a distinct Garo identity. However, almost immediately
upon my arrival in the field, I found that Garo identity proved of great
importance to all people I talked to. I studied “an island”, and concluded that
the island was in so many ways an island, but not without also taking into
account the contextual conditions in which the community had come about.59
Most Garos live in a small strip of land in northern Mymensingh,
bordering India. In past times, these lowland Garos probably practised jhum
or slash-and-burn cultivation. Some elderly Garo villagers remember stories
about the Garos clearing the jungle and moving from one place to another.
Introduction 13
However, neither these villagers nor their parents themselves ever jhummed.
At the turn of the century, lowland Garos practiced plough cultivation and
grew wet rice on their fields, just like their Bengali neighbours.60 Another area
where Garos live is the Modhupur forest. Located in Mymensingh and
Tangail districts, about 150 kilometres north of Dhaka, this is one of the
largest forests of the plains. Detached from the rest of the Garo area, Modhupur
almost seems like a Garo island on the map of Bangladesh. No one knows
how this division of Garos over two separate regions came about. The
situation in Modhupur differs from the border area. The area is mostly
forested highland and requires different cultivation methods. Here, both wet
and dry rice are grown, and more recently pineapple has become a very
popular cash crop. In Modhupur, there are several problems related to land
rights. With the support of human rights activists, Garos have for many years
been trying to acquire the formal land rights that they feel they deserve. It is
important to realize that a great deal of interaction takes place between
Modhupur Garos and the people from the border area, and that Modhupur
people are far from isolated from the other Bangladeshi Garos.
Economically and politically, Garos and other ethnic minorities of
Northeast India and its bordering areas have fared differently than the
adivasis of Western and Central India. Subordination of the latter through
parallel colonial policy developments in forestry and agriculture resulted in
the emergence of a “sharply defined economic stratification in which a
majority of the adivasis became tenants, agricultural and landless labourers,
while the non-adivasis, the landlords, money lenders and timber merchants”.61
In Northeast India or the north and northeast of East Bengal, so-called
tribals were not transformed into one large subordinate colonial labour
force. This does not mean that they have always lived comfortably, but
neither have many of their Bengali neighbours. Today, the situation of the
Garos of northern Mymensingh does not differ significantly from that of
the Bengali peasantry.
Ever increasing dependency on the market economy and mechanisms
such as indebtedness, which operated as a means of downward mobility,
leading from landholding to landlessness, also applied to the poor (Muslim
and Hindu) Bengali peasantry and sharecroppers.62 By the end of the nineteenth
century, the Muslim Bengali peasantry formed 70 per cent of the total
population of Mymensingh but owned only 16 per cent of the land. Suranjan
Das, for example, describes how the Muslim peasantry was exploited by
Hindu landlords and moneylenders, and how economic grievances finally
became one of the causes that led to communal riots in Bengal during the
first half of the twentieth century.63
Garos thus never developed into one single economic subclass. Chapter
8 describes how in the early nineteenth century the local peasantry,
irrespective of their ethnic identity, collectively revolted against repressive
landlords or zamindars. Here we can also read that some of the landlords
who were holding large estates in and around the Garo Hills were Garos, or
at least partly “of Garo blood”. The stories which I collected during my
fieldwork reveal that among early twentieth century Garos landless labourers,
small peasants, landlords and big business men could be distinguished.
These days, not everyone owns a plot of land, but like the large majority of
the Bangladeshis, most Garos are still dependent on agriculture. There are
big landowners, peasants who farm small or middle-sized plots, sharecroppers,
and a great number of landless labourers among the Garos of Bangladesh.
This class differentiation is not a new phenomenon. In the early twentieth
century, there were also big landowners as well as landless labourers among
the Garo peasants.
Unlike their neighbours, Garos are matrilineal. This does not mean – as
is often mistakenly thought – that women rule all domains of Garo life. In
fact, Garo men play dominant roles in most public spheres. Matrilineal
means that each person belongs to the kinship group of the mother, not to
that of the father. People also take the mother’s name as their own. Closely
linked with their kinship system are Garo inheritance practices. Until recently,
property was passed from mother to daughters. Usually one daughter was
appointed as the main heiress, the nokna. These days, more and more parents
divide the property among all their children, even though this goes against
the traditional Garo law of inheritance. The position of Garo women is rather
different from that of their Bengali counterparts. Among the Garos, purdah
is totally absent. Garo women are much freer to travel than Bengali women,
and Garo girls and boys are never segregated in the way Bengali boys and girls
are. This does not mean that men and women perform the same tasks, or that
they are equal in all respects. Different tasks and duties are assigned to men
and women. Although Garo women can have professional careers, women’s
work is more restricted to reproduction and care, whereas men clearly dominate
in public arenas. Formal roles of leadership and authority are normally
assumed by men, especially in public.64
In regard to religion, the Garos also differ greatly from their neighbours.
They are the only people of northern Mymensingh who became Christians.
Missionary attempts to convert other minorities such as the Hajongs, Hodies,
Koches, and Banais have been unsuccessful. Today, Christianity is of great
importance to the Garo community. More than ninety per cent of the Garos
proudly consider themselves Christians. Both the traditional Garo religion as
Introduction 15
well as its followers are called Sangsarek. Some of the old Garos are still
Sangsarek. But unlike a couple of decades ago, outward signs of this religion
are difficult to find, even more so in the border area than in Modhupur
forest.65 Bangladeshi Garos are generally bilingual, speaking both the Garo
language (named Abeng) and Bengali. The various languages or dialects have
become overshadowed by Abeng, named after the linguistic division called
Abeng. This dialect or language has become the lingua franca of all the Garos
of Bangladesh. The importance of these different linguistic groups or divisions
has diminished greatly, and so have their languages. An important factor
which caused these changes was the introduction of education by missionaries.
When the Garos started sending their children to school, interaction between
the different groups increased. It is important to realize that these are all
important characteristics which influence the self-perception and feelings of
Garo-ness today and provide the foundation on which contemporary ideas of
Self are built. A historical perspective however reveals the fluidity and
changeability of these notions of identity.
BREAKING THE SILENCE

This study is based on a myriad of (at times conflicting) written and oral
sources: colonial accounts, post-colonial government publications, magazine
and newspaper articles, an extensive body of literature, missionary diaries and
other publications, fieldwork interviews, and participatory observation; each
source with its advantages and disadvantages. The subsequent chapters discuss
the contributions and drawbacks of the different sources. At this point I will
limit myself to a short exploration of what is perhaps the most significant
source material of this research: oral history.
For two reasons in particular, oral history is imperative to this study. It
opens up new areas of inquiry and it allows a shift of focus. For both the
nineteenth and the twentieth century, written documentation is scarce.
Although hill Garos received quite a lot of attention from both British
administrators and ethnologists, the lowland Garos of East Bengal were
hardly ever studied. An exception should be made for Christian missionaries,
who left us a great many reports and other historical data.66 Only in recent
years have the lowland Garos been increasingly studied by others than
missionaries.67 More importantly, oral history gives voice to people who have
remained outside colonial and national histories. Personal narratives provide
stories from within. In spite of this, an inside or emic perspective has
remained strikingly under-exposed in studies of “the other peoples” of
Bangladesh.68 Thus, although I used interview material to reconstruct parts
of the twentieth-century history of the Garos, their stories derived their

significance mostly from the fact that they mirror present-day perceptions on
Self and Other, and reflect contemporary ideas about their past. Such an
approach leads to the question of how to distil historical “facts” from
perceptions of the present, a question all the more complicated since there is
no material for comparison at hand. My use of oral history was based on the
premise that the value of oral history is directly related to the research
questions. As I was largely interested in Garo ideas and their perceptions of
their past and present identities(s), their personal accounts offered the best
possible answers.69
THE RESEARCHER(S) AND THE RESEARCHED

Although I have not made my presence in the field and its influence on the
outcome of this research a subject of investigation, I have chosen to give
account of that presence in the text. Here, I briefly discuss the background of
my investigations and my relations with the Garos.
Of the seventeen months that I spent in Bangladesh, between November
1993 and November 2000, I lived in Bangladesh for one full year (from
March 1994 until March 1995). An important part of the field research was
carried out in Dhaka. Here I also met Suborno Chisim, who became my
research assistant. We first met in November 1993, at a seminar about
“tribals” and their “Christian identity”. Suborno, who is himself a Bangladeshi
Garo, proved a good interpreter at the time, and, as turned out later, a very
competent research assistant. We developed a close bond of mutual trust and
friendship and soon he became my key informant. His interest in the subject,
his knowledge of the Garos of Bangladesh, his proficiency in English, Bengali,
and Garo, his remarkable memory, his communication skills, and his ability
to grasp the intentions of my investigations made our cooperation very
successful. To acknowledge his contribution to this research, I use “us” and
“we” when I refer to the work that we did together. When I use “I”, it is to
refer to my personal reflections, interpretations, and conclusions. Of course,
I remain responsible for the errors in this study.
Throughout my stay in Bangladesh, Suborno and I visited hundreds of
Garo families and individuals in villages all over Bangladesh. Some we met on
a frequent basis, others we saw only once or twice. We conducted interviews
with people from different social and economic strata and collected a total of
eighty formal interviews. Another important part of the research took place
in the countryside. During the hottest months of the year, from June until
September, we stayed with a Garo family in the village of Bibalgree.70 Suborno
Introduction 17
and I picked this village from the approximately 330 Garo villages in
Bangladesh.71 It was, however, not my intention to conduct a village study.
Rather, we needed a place to stay which was more or less centrally located in
the Garo border belt, and a family who would be willing and able to
accommodate us. Bibalgree suited that purpose very well.
The quality of our relations with the people was imperative to the success
of the fieldwork. During the first few weeks in Bibalgree, I made my first
acquaintance with many Garos in and around the village. In the village itself,
we conducted a small survey about the number and composition of households,
but mostly to find a good reason to go from house to house and become
acquainted with the villagers. Our host family, who provided us with a home
throughout our field research and during every visit since, was related to
Suborno. Suborno referred to them as mani (maternal aunt) and mamu
(maternal uncle). Other members of the household were their little daughter,
a paternal uncle (kaka), the local schoolmaster, one or two nephews from
India, and one or more domestic servants. Kaka was the only additional
household member who lived there on a permanent basis. Others came and
went. Since I had adopted a fictitious kin relation as Suborno’s younger sister,
I could always address people in the way he did. We all needed a few weeks
to get used to each other, and for me to learn some of the basic skills to live
in the village (such as bathing at the tube well, eating and dressing properly,
and “digesting the rice”). Slowly I transformed from a “baby” who copied
Suborno in much of his behaviour, into someone who, in the eyes of the
villagers, seemed just like a Garo.
Soon after we had settled in, Suborno and I started to conduct interviews,
mostly with elderly villagers, both men and women. Their stories are crucial
for this study. Being a woman and a Westerner did not seem to complicate
the fieldwork notably. Being a Westerner (and therefore perceived as a
Christian) made it easy to establish rapport.72 For reasons that will become
apparent in the following chapters, Garos feel close to (Christian) Westerners,
whom they consider reliable allies.73
Gender relations among Garos are not quite as restricted as among the
neighbouring Bengalis. I had easy access to both men and women and
Suborno and I could move around freely, on foot, and even by bicycle. I
noticed that being a young female oftentimes was advantageous for my
relations with the Garos. People seemed to take my presence easy. This
became strikingly clear when “my professor” paid a short visit to Bibalgree,
and when one of our village friends went as far as to cut a hole in one of the
(mud) walls of his house, in order to bring out a table on which he wanted
to serve our meal. He could not bear to serve dinner on the floor to a
professor. With me they obviously cared much less about “respectful

behaviour”, which made it easier to talk freely and openly.74
Since my knowledge of Bengali was very basic and I could hardly make
myself understood in Abeng, I always needed Suborno as an interpreter and
could not communicate independently from him beyond a superficial level.
This meant that a man was present during most of the informal conversations
or any of the formal interviews. I do not believe, however, that this lead to a
strong male bias in the research outcome. In general, Garo women are as
outspoken and uninhibited as men,75 and never were we confronted with
women shying away from conversations. I also never heard of people hinting
at us being a couple. It seemed that our informants clearly took our relation
as a professional one.
ORGANIZATION OF THE BOOK

This study is divided into three parts. The first part shows how the tribalist
discourse is shared in the post-colonial era by outsiders and insiders. It
introduces etic and emic perceptions of what it means to be Garo, and also
ascribes a cultural content to so-called tribes (who, for example, are imagined
as “frog eaters”, primitive and isolated), as well as a basis for social and
political claims. Whereas Chapter 2 focuses on the history of the tribalist
discourse, Chapter 3 examines a number of Garo narratives of Self. An
underlying question is how a discourse which was developed in the days of
British hegemony could so easily endure into the post-colonial period.76
Part two is specifically devoted to the history and constitution of Garo
boundaries – both external and internal. It recounts the process of
categorization of ethnic groups in the colonial context and its aftermath in
East Pakistan and Bangladesh, and recalls that boundaries are social
constructions that can always be crossed. Chapter 4 is about how nineteenth-
century colonial researchers and administrators observed boundaries between
Garos and others. It challenges the suggestion that Garos have always formed
one distinct category of people. Chapter 5 underlines the assertion that the
Garos have only more recently begun to consider themselves as one distinct
ethnic community, belonging together on the basis of a shared identity and
culture. The chapter is largely based on interviews with elderly Garo villagers
who explain how, until a few decades ago, Bengal’s Garos were a diverse
collection of different linguistic and cultural groupings rather than one
distinct ethnic group with one collective identity. Chapter 6 looks at the
relative fluidity of relations between Garos and others, the variation in how
Garos see others and how (im)penetrable boundaries between Garos and
Introduction 19
others really are. This chapter is based on the presumption that ideas about
inter-marriage mirror ideas about Self and Other.
The third part analyses the nineteenth- and twentieth-century history of
the Garos, with particular focus on the process of ethnogenesis. It closely
examines the historical context and developments that have played a significant
role in the Garos becoming the distinct ethnic group they now are in
Bangladesh. Chapter 7 explores how the arrival of Christian missionaries and
the introduction of Christianity have influenced Garo self-perceptions and
group formation. Chapter 8 investigates the role of the state in the ethnogenesis
of the Garos of Bangladesh.
The study encompasses two centuries. It begins with the first European
encounter with Garos and ends today. Such a long-term perspective serves the
two most important purposes of this book: to examine the tribalist image of
the Garos and to unravel the intricate processes by which Bangladeshi Garos
have come to constitute a distinct ethnic group or people; in other words: to
shed light on the ethnogenesis of the Garos of Bangladesh.
NOTES
1. Sirajul Islam, ed., Banglapedia: National Encyclopedia of Bangladesh, Volume 4
(Dhaka: Asiatic Society of Bangladesh). The encyclopedia can also be found on
the Internet: <http://www.banglapedia.org>.
2. Ibid., p. 331.
3. See also Laura Dudley Jenkins, “Another ‘People of India’ Project: Colonial and
National Anthropology”, The Journal of Asian Studies 62, no. 4 (November
2003): 1145–46. For an elaborate analysis of similar state classification projects
in Southeast and East Asia, see Charles Keyes, “Presidential Address: ‘The
Peoples of Asia’ – Science and Politics in the Classification of Ethnic Groups in
Thailand, China and Vietnam”, The Journal of Asian Studies 61, no. 4 (November
2002): 1163–203.
4. Stephen Cornell and Douglas Hartmann, Ethnicity and Race. Making Identities
in a Changing World (Thousand Oaks, London and New Delhi: Pine Forge
Press, 1998), p. 23.
5. Thomas Hylland Eriksen, Ethnicity and Nationalism. Anthropological Perspectives,
2nd edition (London/Sterling, Virginia: Pluto Press, 2002), p. 5.
6. See Willem van Schendel, “The Invention of the ‘Jummas’: State Formation and
Ethnicity in Southeastern Bangladesh”, Modern Asian Studies 26, no. 1 (1992):
103, n.34. Edward Said explained Orientalism as a whole system of knowledge
about a place called the Orient which defines its people, geography, moral
character, destiny, history, future, and character. And although it is partly
empirical, it is largely imaginative. Said, Orientalism (London, etc.: Penguin
Books, 1995).
7. Here, discourses are understood as “complexes of signs and practices which

organize social existence and social reproduction. In their structured, material
persistence, discourses are what give differential substance to membership in a
social group or class or formation, which mediate an internal sense of belonging,
an outward sense of otherness”. This definition is from Richard Terdiman as
cited by Laurence J. Silberstein, “Religion, Ideology, Modernity: Theoretical
Issues in the Study of Jewish Fundamentalism”, in Jewish Fundamentalism in
Comparative Perspective. Religion, Ideology and the Crisis of Modernity, edited by
Lawrence J. Silberstein (New York: New York University Press, 1993), p. 9.
Despite its frequent use, discourse is often left undefined. For a good overview,
see Sara Mills, Discourse (London and New York: Routledge, 1997).
8. Nicholas B. Dirks, Castes of Mind. Colonialism and the Making of Modern India
(Princeton and Oxford: Princeton University Press, 2001), p. 5.
9. Among others, see Susan Bayly, Caste, Society and Politics in India from the
Eighteenth Century to the Modern Age. The New Cambridge History of India IV, 3
(Cambridge: Cambridge University Press, 1999); Ronald Inden, Imagining India
(Oxford: Basil Blackwell, 1990); Gyan Prakash, “Science ‘Gone Native’ in
Colonial India”, Representations 40 (1992): 153–78.
10. Several anthropologists, sociologists, political scientists, and historians have
criticized essentialized and reified notions of tribe, and ensuing practices and
processes of exclusion. A number of these scholars are mentioned here and in
following chapters. Cf. Crispin Bates, “ ‘Lost Innocents and the loss of Innocence’:
Interpreting Adivasi Movements in South Asia”, in Indigenous Peoples of Asia,
edited by R.H. Barnes, Andrew Gray, and Benedict Kingsbury (Ann Arbor,
Michigan: The Association for Asian Studies, 1995), pp. 102–19. See also André
Béteille, “The Definition of Tribe”, in Tribe, Caste and Religion in India, edited
by Romesh Thapar (Delhi: Macmillan, 1977), pp. 7–14; Susana B.C. Devalle,
Discourses of Ethnicity. Culture and Protest in Jharkhand (New Delhi, etc.: Sage
Publications, 1992).
11. Edmund Leach, Political Systems of Highland Burma: A Study of Kachin Social
Structure (Cambridge: Harvard University Press, 1954). See also for example,
Tania Murray Li, “Introduction”, in Transforming the Indonesian Uplands:
Marginality, Power and Production, edited by Tania Murray Li (Australia, etc.:
Harwood Academic Publishers, 1999), pp. xiii–xxiv; “Marginality, Power and
Production: Analysing Upland Transformations”, ibid., pp. 1–44; and “Relational
Histories and the Production of Difference on Sulawesi’s Upland Frontier”,
Journal of Asian Studies 60, no. 1 (February 2001): 41–66.
12. Willem van Schendel, “Geographies of Knowing, Geographies of Ignorance:
Jumping Scale in Southeast Asia”, Environment and Planning D: Society and
Space 20 (2002): 664. See also Henk Schulte Nordholt, “Locating Southeast
Asia: Postcolonial Paradigms and Predicaments”, in Asia in Europe, Europe in
Asia, edited by Srilata Ravi, Mario Rutten and Beng-Lan Goh (IIAS: Leiden/
ISEAS: Singapore), pp. 36–56.
Introduction 21
13. Murray Li, “Marginality, Power and Production”, p. 1.

14. Cf. Susana Devalle, who also points out that the term ethnicity seldom features
in anthropological literature on tribes. See Devalle, Discourses of Ethnicity, p. 34.
15. Fredrik Barth, ed., Ethnic Groups and Boundaries. The Social Organization of
Cultural Difference (Bergen and Oslo: Unversitets Forlaget/London: George
Allen and Unwin, 1969). See also Eriksen, Ethnicity and Nationalism, pp. 36–
40.
16. Fredrik Barth,“Introduction”, in Barth, Ethnic Groups and Boundaries, pp. 9–39
(with special reference to pp. 14–15).
17. Hans Vermeulen and Cora Govers, “Introduction”, in Hans Vermeulen and
Cora Govers, The Anthropology of Ethnicity. Beyond “Ethnic Groups and Boundaries”
(Amsterdam: Het Spinhuis, 1996), p. 2.
18. Eriksen, Ethnicity and Nationalism, p. 37.
19. Ibid., p. 56; see also Richard Jenkins, Rethinking Ethnicity. Arguments and
Explorations (London, etc.: Sage Publications, 1997), pp. 121–22.
20. Peterson Royce, Ethnic Identity, p. 27.
22. In her valuable contribution to tribal studies in India, Susana Devalle also
underlines the fact that ethnicity is primarily a historical phenomenon. See
Devalle, Discourses of Ethnicity.
23. Other useful and interesting studies of ethnicity are numerous. Here I mention
a select number which proved helpful to my comprehension of the subject:
Michiel Baud et al., eds. Etniciteit als strategie in Latijns-Amerika en de Caraïben
(Amsterdam: Amsterdam University Press, 1994); Stephen Cornell and Douglas
Hartmann, Ethnicity and Race. Making Identities in a Changing World (Thousand
Oaks, London and New Delhi: Pine Forge Press, 1998); Lola Romanucci-Ross
and George A. De Vos, eds., Ethnic Identity. Creation, Conflict and Accommodation,
3rd edition (Walnut Creek/London/New Delhi: Altamira Press, 1995; Anya
Peterson Royce, Ethnic Identity. Strategies of Diversity (Bloomington: Indiana
University Press, 1982).
24. John Eliot, “Observations on the Inhabitants of the Garrow Hills. Made during
a Public Deputation in the Years 1788 and 1789”, Asiatic Researches 3 (New
Delhi: Cosmo Publications, 1979. First published in 1794), pp. 17–37. Sipra
Sen mentions some 582 references on the Garos in her extensive, but by far not
exhaustive, bibliography: Sipra Sen, The Tribes of Meghalaya (Delhi: Mittal
Publications, 1985), pp. 14–62.
25. In 1872, the Garo hills became a separate district under the name Garo Hills, as
a part of Assam. In 1972, the Garo Hills, together with the United Khasi and
Jaintia Hills districts, became the full-fledged state of Meghalaya.
26. Major A. Playfair separated “those who inhabit the Garo Hills district, and those
who reside in the plains and are scattered over a very wide area of country”.
A. Playfair, The Garos (Gauhati and Calcutta: United Publishers, 1975, first
published in 1909), p. 59.
27. According Chie Nakane, for example, this division into hill and plains Garos
represents the prime division of Garos: “The Garo of the plains have become
more sophisticated by closer contacts with the lowland peoples and many of
them have adopted Christianity. They seldom come into contact with the hill-
dwellers and live in an entirely different ecological and cultural environment
from the latter. Thus the Garo may be divided roughly into two main categories
as hill-dwellers and plain-dwellers.” Chie Nakane, Garo and Khasi. A Comparative
Study in Matrilineal Systems (Paris: Mouton and Co., 1967), pp. 21–22.
28. To the world they are known as Garos, and to the world they present themselves
as Garos. For that reason, I also opted for Garo instead of Mandi, but I chose not
to translate Mandi in the interview fragments.
29. Willem van Schendel, The Bengal Borderland. Beyond State and Nation in South
Asia (London: Anthem Press, 2005).
30. The permit requirement for Assam, Meghalaya and Tripura was lifted in 1995.
I completed my one-year field research early that year.
31. Van Schendel, The Bengal Borderland, p. 201.
32. Eriksen, Ethnicity and Nationalism, pp. 121–22.
33. For the notion of “the other peoples”, see also Van Schendel and Bal, “Beyond
the ‘Tribal’ Mind-Set”.
34. I derived the concept of “othering” from Lamia Karim, “Pushed in the Margins:
Adivasi Peoples in Bangladesh and the Case of Kalpana Chakma”, Contemporary
South Asia 7, no. 3 (1998): 310.
35. According to the 1991 census, Muslims comprise 88.3 per cent of the population;
Hindus, 10.51; Buddhists, 0.59; Christians, 0.33; and others, 0.27 per cent.
36. For example, Mahmud Shah Qureshi, who has edited the most comprehensive
collection of studies on non-Bengali Bangladeshis, estimates their total number
at thirty-one. See Mahmud Shah Qureshi, “Foreword”, in Tribal Cultures in
Bangladesh, edited by Qureshi, Mahmud Shah (Rajshahi: Institute of Bangladesh
Studies Rajshahi University, 1984), p. xv. I also came across an estimation of 41,
which was mentioned at a seminar about tribal identity and Christianity held in
Dhaka in November 1993. A foreign missionary who has been living in Bangladesh
for over twenty-five years and has widely travelled the country, estimated the
number of different ethnic groups at fifty-six. According to him all “other
peoples” number over 31 million of people.
37. The inhabitants of the Chittagong Hill Tracts, who have been engaged in a
struggle with the governments of East Pakistan and Bangladesh since the 1960s,
are a bitter exception.
38. For a recent evaluation of the terms that are used in West Bengal and Bangladesh,
see also Willem van Schendel and Ellen Bal, “Beyond the Tribal Mind-Set:
Studying Non-Bengali Peoples in Bangladesh and West Bengal”, in Contemporary
Societies: Tribal Studies. Volume V: Concept of Tribal Society, edited by Georg
Pfeffer and Deepak Kumar Behera (New Delhi: Concept Publishing Company,
2002), pp. 121–39; “Bhumika: Name Ki Eshe Jay?”, in Banglar Bohujati:
Introduction 23
Bangali Chhara Banglar Onnyanyojatir Proshongo, edited by Willem van Schendel

en Ellen Bal (Calcutta: International Centre for Bengal Studies, 1998), pp. 7–
25.
39. Researchers who have contributed to the discussion about the validity of the
term tribe are, among others, Bates, “Lost Innocents”; Béteille, “The Definition
of Tribe”; Devalle, Discourses of Ethnicity; Binay Bhushan Chaudhuri, “The
Myth of the Tribe? The Question Reconsidered”, The Calcutta Historical Journal
16, no. 1 (1994): 125–56; Binay Bhushan Chaudhuri, “Tribal Society in
Transition: Eastern India 1757–1920”, in India’s Colonial Encounter. Essays in
Memory of Eric Stokes, edited by Mushirul Hasan and Narayani Gupta (New
Delhi: Manohar, 1993), pp. 65–120; M.H. Fried, The Notion of Tribe (Menlo
Park, Calif.: Cummings Publishing Company, 1975); M. Sahlins, Tribesmen
(Englewood Cliffs, N.J.: Prentice Hall, 1968).
40. David Hardiman, The Coming of the Devi: Adivasi Assertion in Western India
(Delhi, etc.: Oxford University Press, 1995), p. 14.
41. See, for example, Devalle, Discourses of Ethnicity; Chaudhuri, “ ‘The Myth of the
Tribe’?”, pp. 125–56 (with special reference to pp. 132–34); and B.G. Karlsson,
Contested Belonging. An Indigenous People’s Struggle for Forest and Identity in Sub-
Himalayan Bengal (Lund: Department of Sociology, Lund University, 1997).
42. Karlsson, Contested Belonging, pp. 44–45.
43. These same policies were applied to the Untouchables, Harijans or Dalits, who
have henceforth been addressed as Scheduled Castes (SC). Numerous publications
about the Indian reservation policies have been published since then. For an
extensive account, see Marc Galanter, Competing Equalities. Law and the Backward
Classes in India (Delhi, etc.: Oxford University Press, 1984).
44. Report of the Commissioner for Scheduled Castes and Scheduled Tribes. Volume 1
(New Delhi: Manager of Publications, 1952), p. 11.
45. Ruth Manorama, “The Situation of Dalit Women”, Indigenous Affairs 2 (1995):
4–7.
46. Tanvir Mokammel’s documentary on the Chittagong Hill Tracts, entitled Teardrops
of Karnaphuli (2005) which has been banned from Bangladesh, includes an
interview with a Bengali settler who refers to the local hill people as Thai and he
makes it very clear to the interviewer that the Bengalis are the true indigenous
people to the area.
47. The term adivasi gained popularity in pre-Independence India and continued to
be used in India after 1947. It is a recent term which probably originates from
the Chhota Nagpur region in Bihar (and present-day Jharkhand) in the 1930s
and was popularized at a wider level in the 1940s. Hardiman, The Coming of the
Devi, p. 13.
48. G.S. Ghurye argues that because many of the so-called aboriginal tribes came to
their present habitat from somewhere else in South Asia, they cannot be considered
autochthonous to their present home. Nevertheless, he adds that although they
may not belong to exactly the same area which they are now occupying, they still
are the autochthones of South Asia. Thus, to this extent they can be called
adivasis or aborigines. G.S. Ghurye, The Scheduled Tribes, 3rd edition (Bombay:
Popular Prakashan, 1963), pp. 11–12.
49. See, for example, Hardiman, The Coming of the Devi, pp. 15–16.
50. Stephen G. Gomes, “The Paharias in Bangladesh: A Case Study of Assimilation
and Identification with Policy Implications”, in Tribal Cultures in Bangladesh,
edited by Mahmud Shah Qureshi (Rajshahi: Institute of Bangladesh Studies,
1984), p. 140. Stephen G. Gomes, The Paharias: A Glimpse of Tribal Life in
North Western Bangladesh (Dhaka: Caritas, 1988).
51. Prashanta Tripura, “The Colonial Foundation of Pahari Identity”, Journal of
Social Studies 58 (1992).
52. Cf. Van Schendel, “Invention of the ‘Jummas’ ”, pp. 95–128.
53. The Bangladesh Population Census of 1991 mentions 68,210 Garos, which is
probably too low. Garo members of the two largest Christian denominations,
the Baptists and Roman Catholics, totalled 65,076. Not included in these
figures are Garos living in the tea gardens of Sylhet, members of other
denominations, and Sangsarek, Muslim, and Hindu Garos. Catholic Directory of
Bangladesh (Dhaka: Bishop’s Conference of Bangladesh, 1992); Yearly Report of
the GBC to the ABMS, 1991–1992 (n.d.). For the Modhupur figures, see
Robbins Burling, Strong Women of Modhupur (Dhaka: University Press Limited,
1997), p. 13.
54. When this study mentions “northern Mymensingh”, it really refers to the six
northernmost thanas of the former Greater Mymensingh district. This also
includes the districts of Sherpur and Netrokona.
55. The majority of the villages are located in the thanas Haluaghat, Dhubaura,
Durgapur, Netrokona, and Kolmakanda. Nowadays Garos are also found in
Sunamganj and Moulvi Bazar, two former districts of Greater Sylhet. Sunamganj
Garos have been living there for a long time. The Garos in Moulvi Bazar are all
migrants from Greater Mymensingh who came to look for work in the plantations.
56. In 1956, for example, missionary records of the Catholic Church repeatedly
referred to the migration of Garos from Haluaghat to tea estates in Sylhet. See
for example, the Chronicles of Biroidakuni Mission for 10, 23, 27 February and
27 March. These migrations are attributed to oppression and thievery by Bengali
immigrants, and to dire poverty.
57. Only one study focuses on these migrants. Its authors estimate the total number
of Garo migrants in Dhaka at 3,000. Nokmandi Prakashana, A Census of Garo
Housemaids and Others in Dhaka (Dhaka: Nokmandi Prakashana, 1994).
58. Stuart Hall, “Old and New Identities, Old and New Ethnicities”, in Culture,
Globalization and the World-System, edited by A.D. King (Basingstoke, England:
Macmillan, 1991), p. 57.
59. Gerd Baumann, fair enough, writes about most [is his word] contemporary
studies of ethnicity, that “we have, in effect, created a little island; we study this
island, and we usually conclude that the island is, in so many ways, an island.
Introduction 25
What a bore.” Gerd Baumann, The Multicultural Riddle. Rethinking National,

Ethnic, and Religious Identities (New York and London: Routledge, 1999) pp.
145–46.
60. Playfair, The Garos, p. 35.
61. Ashesh Ambasta, Capitalist Restructuring and Formation of Adivasi Proletarians.
Agrarian Transition in Thane District (Western India) c. 1817–1990 (Unpublished
Ph.D. thesis: Institute of Social Studies, The Hague, 1998), pp. 14–15. See also
Devalle. Discourses of Ethnicity.
62. See for example, Adrienne Cooper, Sharecropping and Sharecroppers’ Struggles in
Bengal 1930–1950 (Calcutta and New Delhi: KP Bagchi & Company, 1988);
Taj ul-Islam Hashmi, Peasant Utopia. The Communalization of Class Politics in
East Bengal, 1920–1947 (Dhaka: University Press Limited, 1994).
63. Suranjan Das, Communal Riots in Bengal 1905–1947 (Delhi, etc.: Oxford
64. See also Burling, Strong Women of Modhupur, pp. 242–46.
65. Sangsareks believe that the world is populated by mite. Mite is generally referred
to as “spirit”, but some mite were so powerful that they are better thought of as
gods. These mite can cause illnesses by biting people. The priests, or kamal,
knows how to perform sacrifices in order to cure the victims. Sacrifices were also
performed at various points of the annual cycle, and they were a central part of
village festivals. See, for example, Burling, Strong Women of Modhupur, pp. 53–
56.
66. Since the end of the nineteenth century, missionaries from a variety of Christian
denominations have worked and lived with Garos. See also Chapter 7 of this
study.
67. Important contributions have been made by Robbins Burling and Kibriaul
Khaleque, the two anthropologists largely concentrated on the Garos of
Modhupur.
68. In this study, I use the concepts “emic” to refer to views from within or insider
views, and “etic” to refer to outside perceptions or outsider views. Although
there has been much debate about the two notions, these are the most common
definitions. See Thomas N. Headland, Kenneth L. Pike and Marvin Harris,
Emics and Etics. The Insider/Outsider Debate (Newbury Park, etc.: Sage
Publications, 1990), p. 22.
69. For an elaborate discussion about the use and validity of oral history as a source
for historical research, see Paul Thompson, The Voice of the Past (Oxford: Oxford
70. To safeguard the privacy of my informants, I have used pseudonyms both for the
village Bibalgree as well as for my informants. Suborno Chisim, who I consider
a fellow researcher rather than an informant, is referred to by his genuine name.
71. By Garo village, I mean a village which numbers or used to number a significant
number of Garos. With the exception of Modhupur villages, very few villages are
entirely Garo.
72. Garos consider Westerners Christians. After all, everyone must have some religion
or the other, and Westerners can never be Muslim, Hindu or Buddhist, so must
be Christian.
73. Cf. Kibriaul Khaleque’s hard work to be accepted by the Garos during his field
research. Khaleque found it very difficult to establish friendly relations with the
Garos. People did not easily accept or trust him, being a highly-educated Bengali
Muslim. See Kibriaul Khaleque, “My fieldwork Experience in a Garo Village of
Bangladesh”, in Pains and Pleasures of Fieldwork, edited by Anwarullah Chowdhury
(Dhaka: National Institute of Local Government, 1985), pp. 207–23.
74. Cf. Joan Neff Gurney, “Female Researchers in Male-Dominated Settings.
Implications for Short-Term Versus Long-Term Research”, in Experiencing
Fieldwork. An Inside View of Qualitative Research, edited by William B. Shaffir
and Robert A. Stebbins (Newbury Park, etc.: Sage Publications), p. 56.
75. Cf. the title of Robbins Burling’s ethnography The Strong Women of Modhupur.
76. Cf. Lionel Caplan, who ask himself similar questions about the Gurkhas. Lionel
Caplan, Warrior Gentlemen. “Gurkhas” in the Western Imagination (Providence
and Oxford: Berghahn Books, 1995), p. 8.
2
THE HISTORY OF
A PERSISTENT IMAGE
INTRODUCTION
One afternoon in February 1995, the Garo student Rosie told us the following
anecdote:
“Today we had ethnology class about the different stages of human
civilization. The teacher described how food habits of people developed
stage by stage. At one moment she said that the ancient people ate frogs,
snakes, and dogs. This was derived from the chapter on barbarism, savages,
like this. All students were staring at me. I sat in the back of the classroom
and they all turned around and stared at me. I was feeling so uneasy; madam
noticed it too. I told her that the students are studying ethnology, but they
are not broadminded.”1
At the time of the interview, Rosie was a student of anthropology at Dhaka
University. In this interview fragment, she referred to that morning’s lecture,
in which the teacher explained how food culture has developed from primitive
to civilized in an evolutionary manner. When the teacher discussed the diet
of the so-called ancient people, the students started staring at Rosie, associating
her with those primitive people. The behaviour of Rosie’s fellow students
leads to the question of why an intelligent girl like her, who had made it all
the way to Dhaka University and who dressed or behaved no differently from
27
her fellow students, was so easily included in the category of primitives. The
answer is shockingly simple: they knew Rosie to be a Garo, and Garos are one
of Bangladesh’s many ‘tribes’, upojatis, adivasis, or Indigenous Peoples. To this
very day, many Bangladeshis imagine these communities as inherently
unsophisticated, simple, primitive people without a history.
The following fragment from the same interview also illustrates the
primitive image of contemporary Garos:
“They [Bengalis] ask if we eat frogs or snakes. That is alright. But they ask
more stupid questions about our dress. They can see that we wear the same
clothes as they do, but they still ask us if Garo women cover the upper part
of their body, and if they wear very short clothes. I don’t want to say
anything more about what else they say. Their questions are so strange.
They seem to know nothing. I don’t mind if they want to know about us,
but they ask it in such a way that we don’t seem to be human beings. They
want to insult us by asking stupid questions. That’s what I don’t like.”
This part illustrates, for example, how Rosie’s Bengali dressing style has
virtually no impact on the image of the scantily dressed Garo woman.
Despite the fact that Bangladeshi Garos have been fully dressed (often in
Bengali dressing style) for a long time, and no Dhaka dweller ever meets a
half-naked Garo woman anywhere, the image of the scantily dressed Garo
still exists. Similarly, Rosie’s classmates saw what she was wearing and still
wondered how Garo women dress.2 The case provides a striking example of
how perceptions of human societies as ranked according to their level of
civilization have persevered in the classrooms of Dhaka University.3 The
anecdotes are therefore significant for two reasons: they illustrate the
reproduction of scientific discourse in academic Bangladesh and its social
manifestations.
Although, as pointed out in Chapter 1, an ever increasing number of
academics have made strong cases against colonial and post-colonial
categorizations on the basis of caste and race, not quite as many researchers
have taken up an interest in the notion of tribe, its conceptualizations,
and its consequences. This chapter explores the tribalist discourse, its
origins and its remarkable persistence, and demonstrates that contemporary
tribalist discourse is firmly rooted in colonial perceptions of the Indian
sub-continent. I hope to show that we cannot understand contemporary
views of the Bangladeshi Garos and the hundreds of other tribes without
close scrutiny of colonialist perceptions of Indian society and their attempts
to explore and classify its population. The chapter concentrates on three
major themes: it first deals with the so-called observers of India, their
backgrounds, objectives, and confusion. Who were they and what did
The History of a Persistent Image 29
they see? Next, it deals with the processes of data collection and data
ordering, and describes how colonial observers gathered, organized, and
reproduced data in discrete bounded categories (such as races, castes, and
tribes). Third, it discusses a number of typical “tribal characteristics” in
general, and of the Bangladeshi Garos in particular, and identifies blind
spots and fallacies in the tribalist discourse.
For our reconstruction of colonial history, we are largely dependent on
the accounts of colonial eyewitnesses. However, their perceptions of India
and its people were not mere reflections of Indian social reality but selections
of what they found important or necessary, coloured by nineteenth century
European perspectives, scientific theories, cultural differences, and power
positions. This chapter includes different layers of historiography and
interpretations. It includes those first-hand observations as well as colonial
and post-colonial interpretations of those colonial eyewitness accounts and
fact-gathering projects. It is important to realize that for those evaluations of
British observations cum interpretations, and the alternative theories that
were built on those assessments, those critical scholars depended on these
same British data too.
OBSERVING OBSERVERS
Publications on the Garos are of four main kinds: administrative, missionary,
administrative-cum-ethnological, and professional anthropological accounts.
Attention is given to the last two types here, although distinction between
one type and another is not always as unambiguous as the categorization
suggests. We will see that in the course of time professional anthropologists
monopolized Garo research, and that Indian anthropologists rather than
Europeans came to dominate the field of Garo studies. This process took
more than one century and coincided with general developments in colonial
writings on India. The historiography and ethnography of the Garos can only
be fully understood if the general histories of colonial and post-colonial
categorization of the Indian sub-continent are closely scrutinized.
By the late eighteenth century, three major traditions in the colonialist
approach to Indian society could be distinguished: the Orientalist, the
missionary, and the administrative.4 Each tradition was characterized by its
own set of ideas and theories, tied up with the roles and believes of the
observers belonging to that tradition.5 The early Orientalists derived their
thoughts from the study of texts and cooperation with pandits (Hindu
priests) and scholars of the Hindu scripts. By the late eighteenth to early
nineteenth century, this had led to an image of India as a static, timeless, and
spaceless society. As the early Orientalist observers did not distinguish between
the prescriptive, normative statements derived from the Hindu texts, on the
one hand, and the actual behaviour of the people, on the other, they saw
Indian society merely as “a set of rules which every Hindu followed”.6 “The
outcome was the image of a static society – a society that did not move either
in space or in time scale – a society permanently divided, primarily into two
religious communities, the Hindus and the Muslims, who had two different
legal codes and thereby two different cultural and social traditions. The
Hindus in their turn were thought to be dominated by the Brahmans, whose
power was based on a monopoly of knowledge and who ruled over a hierarchy
of castes.”7
The missionaries, who started to develop their views of India in the late
eighteenth century, also saw India as a society in which religious ideas and
practices were at the core of all social structures. Contrary to the Orientalists,
however, they did not admire the civilization and religion represented in
the texts, but considered India as a corrupt and, in many ways, absurd
society. Chapter 7 discusses missionary perception, ambitions, and attitude
in greater detail.
The colonial administrators formed the third category. With the
establishment of British hegemony in the late eighteenth century, interest
in, and knowledge of, India developed rapidly. As time went by, British
administrators became increasingly aware that the early Orientalist perception
of the Indian sub-continent could not be sustained as these officials found
themselves confronted with a “bewildering variety of peoples, histories,
political forms, systems of land tenure, and religious practices”.8 These
feelings of bewilderment lingered on into the twentieth century. Note the
following remarks of an early twentieth-century observer who found how,
soon after their arrival, untravelled Europeans would have to acknowledge
India’s great variety:
It is a familiar experience that the ordinary untravelled European on first

arriving in India finds much difficulty in distinguishing one native of the
country from the other. To his untrained eye all Indians are black; all have
the same cast of countenance; and all, except the “decently naked” labouring
classes, wear loose garments which revive dim memories of the attire of the
Greeks and Romans. An observant man soon shakes off these illusions, and
realizes the extra ordinary diversity of the types which are to be met with
everywhere in India.9
Different observers, with different backgrounds, personalities, objectives, and

interests, presented different pictures of India. They found themselves in the
middle of a complex society they tried, for various reasons, to make sense of.
In the course of the nineteenth century, a heterogeneous data collection came
into being.
These administrative researchers observed with different objectives: a
great many of them wanted to make a contribution to scientific developments
back home. These “scholar-officials were trying to place themselves in the
vanguard of contemporary scientific thought”.10 Various officials feared that
much valuable information would forever be lost if not recorded soon. In the
light of this concern, it is not surprising that colonial ethnology gained much
of its importance in the 1860s and 1870s and became a scientific discipline.11
Other observers compiled, enumerated, and categorized India primarily to
facilitate administrative rule. In order to govern efficiently, systematic
information about Indian society became essential. The British needed to
know better where they were and who they were ruling. This administrative
need for systematic information came to be felt particularly strongly during
the second half of the nineteenth century when the nature of British colonialism
changed and India came under direct rule of the British crown.12 In practice,
these objectives to contribute to the scientific and the administrative body of
knowledge frequently coincided.
‘SCIENTIFIC’ OBSERVATIONS
In 1884, the Government of Bengal recommended to the Government of
India the appointment of H.H. Riley “to conduct an inquiry into castes and
occupations throughout Bengal”. The concern was vented that “if it is not
undertaken now, a mass of information of unsurpassed interest will be lost to
the world”.13 L.A. Waddell, a high official in the Indian Medical Service, is a
good example of one of these concerned observers. He took his personal
scientific task very seriously and conducted his own private research with the
aim to contribute “towards fixing the physical type and racial affinities upon
the only trustworthy basis, namely, precise measurement”. Waddell was
particularly worried that “this unique mass of material which is thus available
for solving such important problems lying at the very basis of civilization and
culture is being allowed to disappear unrecorded!”14
At the time, in nineteenth-century Europe, the evolutionary perspective
on human development was widely accepted. And even though Western
racism used biological differences to justify colonization and slavery,15 racial
differences did not merely serve the need for an ideological justification of
colonialism. In the words of Susan Bayly, “much of this scholarship was not
‘colonial’ at all in the usual sense, being conceived as a contribution to
broad debates in social theory and ‘scientific’ ethnology, rather than being
focused solely on questions of how to ‘know’ and subjugate Indians as the
ethnographic ‘other’.”16
At present, colonial perceptions and understandings of India are often
summarized with the term Orientalism or Orientalist discourse. In spite of
the increasing acknowledgement of India’s heterogeneity, colonial perspectives
on India continued to be based on a clear dichotomy between the colonizer
and the colonized, and views of the sub-continent did not lose their Orientalist
character. That is, if we understand Orientalism as the approach which
“focused most sharply on the cultural essences of subordinated societies,
ascribing them singular qualities, and individuating them with reference to
each other in order to create a universal typology”.17
Contemporary researchers like Partha Chatterjee argue that the European
presence in India was in fact based on these fundamental differences, in
other words, on “the rule of difference”.18 However, since Orientalism refers
to a dichotomization between a civilized Europe and a primitive Orient, it
does not help us to understand the segmentation or categorization of India’s
population that increasingly came to preoccupy the colonizer and resulted
in colonial representations of a strictly hierarchically ordered Indian
population. Colonial accounts of India became marked by discussions of
how to order Indian society. Whereas Europeans were evidently the most
developed members of mankind, among the Indian population a whole
range of less and lesser developed categories of people could be distinguished.
This process of segmentation is better explained by anachronism than
by Orientalism. Anachronism ranked different societies according to how
much in time they were behind Europe. Whereas Orientalism explains the
oppositional thinking of the “civilized” Europeans about the “primitive”
Indians, anachronism explains the internal hierarchical ordering of the
Indian population. “It was precisely the intersection of the two that defined
and gave force to colonial categories.”19 Anachronism thus explains the
compartmentalization of the Indian population into primitive and less
primitive segments, which finally resulted in the distinction between the
two fundamentally different categories of castes and tribes (the most primitive
of Indians).
Early anthropologists saw in this latter, most primitive society “their own
society (as they understood it) seen in a distorting mirror”. They believed that
their type of society was witnessing a revolutionary transition and “looked
back in order to understand the nature of the present, on the assumption that
modern society had evolved from its antithesis. To study the tribes would
reveal much about their own past”.20 This can be illustrated by what Waddell
wrote at the turn of the century: “Unfortunately for science, however, no

steps are being taken to record the rare vestiges of prehistoric society which
still survive here; but which are now being rapidly swept away by our
advancing civilization.”21
No matter how distorted these ideas may seem to us today, they had a
major impact on how colonial observers perceived the world around them. It
is thus in this broad scientific context that we have to understand the various
colonialists’ attempts to understand Indian society. And it is this dominant
mental framework that, at least partly, explained why “[a]ll European sources
on early colonial Asia suffer from similar prejudices, blind spots, and
distortions.” Fortunately, “[t]he distance in time makes it easier for us to spot
these shortcomings than those inherent in contemporary perspectives on
Asian societies.”22 Nineteenth-century scientific practice in Europe was thus
an important factor in the process of fact-gathering and interpretation.
Colonial observers were influenced by the spirit and scientific insights of their
epoch. The contemporary body of scientific knowledge influenced colonial
perceptions and interpretations of the world and its population and played a
weighty role in the objectification of India’s population.23
OBJECTIFYING INDIA
In 1891, Risley wrote that “[t]he relations of different castes to the land, their
privileges in respect of rent, their relations to trade, their social status, their
internal organization, their rules as to marriage and divorce, all these are
matters intimately concerned with practical administration.”24 This comment
illustrates nicely the need for modern states to collect data. The colonial state
needed information in order to rule India. Such fact-gathering projects
required new skills to measure and the narrowing of vision. Dipesh Chakrabarty
commented, for example, that “[o]ne symptom of its modernity was that its
techniques of government were very closely related to techniques of
measurement.”25 Significant consequences of the second prerequisite, the
narrowing of vision, were a loss of information and simplification of an
otherwise very complex reality, as James Scott convincingly argued in his
seminal work on state formation and simplification:
Certain forms of knowledge and control require a narrowing of vision. The
great advantage of such tunnel vision is that it brings into sharp focus
certain limited aspects of an otherwise far more complex and unwieldy
reality. This very simplification, in turn, makes the phenomenon at the
centre of the field of vision more legible and hence more susceptible to
careful measurement and calculation.26
Thus while the ever growing body of ethnographic data (in the form of
administrative reports, linguistic and other surveys, population censuses,
articles and monographs) led to the negation of India as a homogeneous
society, objectification again reduced the complexity. Fact-gathering projects
of the colonial state went hand-in-hand with the simplification of a complex
Indian society, such as the classification of its population according to religion
or caste background. Three categories stood out both in colonial and post-
colonial reflections on India, of outsiders as well as Indians themselves, and
functioned as significant objectifying tools. These are caste, race, and tribe.
Hereafter I shall discuss each of them separately, and argue that compared to
the ever increasing attempts to deconstruct notions of caste and race, tribe has
received remarkably little attention.
CASTE AS A SOCIAL CONSTRUCT

Whereas the early Orientalists viewed religion as the prime mover of social
relations, in the mid-nineteenth century many British officials felt that not
only religion but also caste was an important key to understanding India.
Many professional researchers, administrators, and laymen perceived caste as
the very core of Indian society, and it was caste that divided an important
segment of the Indian population: the Hindus. These ideas resulted in the
widely held assumption that an all-India system of caste classification could
be developed, as was subsequently done in many of the population censuses
that they undertook.27 In the 1901 census, an attempt was made to set each
caste down according to its place in Hindu society. This caused so much
resistance that in the 1911 census it was again decided to present the castes
in alphabetical order.28
Modern scholars have not only explored how colonial perceptions were
coloured by Orientalist ideas, they have also researched how colonial
categorization resulted in the simplification of otherwise very fluid, dynamic,
or perhaps even non-existent categories. Particularly, caste has attracted much
attention.29 Two sets of questions about the importance of caste feature in
these modern reflections on caste. The first relate to the intrinsic existence of
caste. What are castes? When did they come into existence? What has their
relevance been for Indian society? And how has colonialism influenced the
caste system? An increasing number of contemporary researchers have argued
that India was never compartmentalized in such fixed and distinct categories
as colonial representations of the caste system suggest. Researchers like Ronald
Inden have gone so far as to argue that the caste system was in fact a “colonial
construct”. “Caste had become essentialized and turned into the substantialized
agent of India’s history.”30 Others, like Susan Bayly, favour the more moderate
suggestion that caste should not be understood as a mere “invention” but
rather as “a meeting ground between Indian reality and colonial knowledge
and strategy.”31
The second set of questions deals with the importance of caste in the
colonial understanding of India. These questions are particularly relevant
since large numbers of British Indians were not part of that caste system.
Muslims, evidently, were not, and neither were the so-called aborigines (later
known as tribes). Bayly, for example, argues that colonial observers were well
aware of this. She puts the importance of caste in a different perspective and
brings the concept of race into the discussion. Not caste but race was of major
concern to the colonial observers.32
DISCOURSE OF RACE
Throughout the nineteenth century, one distinction played a particularly
significant role in colonial perceptions of the Indian population. This idea,
with Risley as an important spokesman, amounted to the differentiation
between “immigrant Aryans” and “indigenous non-Aryans” or “Dravidians”.
The civilization of India began with the arrival of Indo-Aryan peoples from
the north some time around the second millennium B.C. Until the end of the
eighteenth century, differences between Aryan and non-Aryan “races” were
attributed to linguistic and environmental characteristics, but since the turn
of the eighteenth century they became increasingly explained in evolutionary
terms.33 Although no consensus existed about the exact relation between race
and caste, the Aryan race symbolized racial superiority, and the degree of
absorption into this civilized race matched the level of civilization.34 In such
an approach, castes and tribes were perceived as racial categories. People’s
place in the Indian hierarchy was directly linked to the level of absorption
into the Aryan caste system. And from this follows that tribes are, in racial
terms, essentially different from caste Hindus.
The consequences of such ideas have been manifold. Even today they
may lead to the suggestion that Northern Indians are essentially more
civilized than Southern Indians, and that those people who were never
absorbed into this race of civilized Aryans are the most primitive of all
South Asians. Moreover, this idea of a superior race invading the sub-
continent and either absorbing the original population into their caste
system or pushing them into the fringes of the region, inaccessible jungle or
hill areas, still lies at the basis of (South Asian) contemporary perceptions of
South Asia’s tribes.35
Colonialists’ preoccupation with race resounds clearly from the population
censuses. Even in the fourth decade of the twentieth century, the census of
Bengal and Sikkim (of 1931) devotes one substantial chapter to “Caste, Tribe
and Race”. It is, however, added that this section “is perhaps the most
unsatisfactory and troublesome of all the enquiries undertaken during the
census”. Regarding the concept of race, the author of the census report, A.E.
Porter, commented that it is vague and it might almost be said that as many
different racial classifications exist as there are writers on ethnographic
subjects.36 Concepts of race, nation, caste, and tribe continued to puzzle
colonial researchers until far into the twentieth century.
Thus when Skaria contends that by the 1860s colonial officials routinely
distinguished between the castes and tribes of India, and that they saw both
categories as fundamentally different, he disregards the discussions,
disagreements, and misunderstandings over the terms and categories. Skaria,
however, qualifies his own statement by following up on Bayly’s suggestion
that the colonial writers’ primary concern was not caste but race.37 “[T]he
distinction between high and low castes was really a distinction between
peoples of supposedly superior and inferior racial endowment.”38
Nevertheless, although race continued to occupy the minds of the
colonizers and to feed their discussions, and terminological confusion lingered
on into the twentieth century, separate categories did crystallize further in the
course of colonial rule. At the time of the 1931 census, one was sure of one
basic segmentation of the Indian population: its division into Hindus, Muslims,
Primitive Tribes, and Others (that is, Christians and Buddhists). Garos
indisputably belonged to the category of Primitive Tribes.39
CATEGORY OF TRIBES
Earlier I contended that to approach colonial representations of India as the
reflections of one single uncivilized Other by one civilized Self would be far
too simplistic. The colonial observers, who themselves were a diverse lot,
distinguished several more and less developed social groups among the Indian
population. There were Brahman Hindus at the top of the hierarchy, the so-
called primitive tribes at the very bottom, a myriad of castes, tribes, and semi-
tribes in between, and a great deal of confusion about how to order them.40
Notwithstanding the debate and confusion, however, there appears to have
existed a clear consensus about the most primitive of all Indians. They were
“those aboriginal tribes, out of which the whole series of caste was fashioned
by slow degrees, through the example and under the guidance of the
Brahmanical priesthood”.41
As time went by, and Indian ethnology was increasingly monopolized by
professional anthropologists – who focused their attention on these primitive
tribes – a dualistic framework of analysis was developed.42 Albeit in many
forms, this dichotomy between tribes and others has continued to influence
the thinking about so-called tribes (and the ensuing practice) to this very day.
Several scholars have pointed out that also before British colonial hegemony,
Indian civilization was characterized by a complementary co-existence of
settled agricultural communities and an “alien sphere of the jungle”.43 These
more or less inaccessible niches, however, were often used as the bases for
social, economic, and political initiative. People living in forests or hills, or
even those who were practising settled agriculture, were certainly not confined
to these spaces, and the transgression of boundaries and the moving in and
out of people and resources was by no means an accidental process.44
Objectification and social evolutionist ideas seem to have played a significant
role in the fixing of categories and the dichotomization of hill and forest
peoples and others. In this way, we could say, science became reality instead
of the other way around.
The classification of “aboriginals” or “primitive tribes” has lived on in
the category of tribes, Scheduled Tribes, adivasis, upojatis, and so on.
Colonial discourse placed them at the bottom end of society and today they
are still considered the most primitive of South Asia. Apart from the fact
that the terminology of this category varied (from Aboriginal Tribes, Primitive
Tribes, or Tribes to Scheduled Tribes, adivasis, upojatis, Indigenous Peoples),
this tribalist discourse has remained strikingly similar. Although a number
of researchers have indeed critically reflected on the issues of tribes and
“tribal categories”,45 no real post-colonial debate about the category tribe
has developed.
There are, however, various reasons for a serious rethinking of the category
of tribes. Particularly the contemporary debates on caste and race provide
some of the basic arguments for an evaluation of the tribalist discourse, and
help to reframe current perceptions of tribes. Firstly, because those researchers
of caste have demonstrated how imagined categories have become social
realities. Secondly, by introducing the concept of race in Indian historiography,
these scholars put the relevance of caste in a different perspective. They argue
that the colonial preoccupation with race rather than with caste provides an
explanation for the creation of the tribal category. This tribal category did not
merely function as a wastebasket for all those groups that colonial observers
could not place. Nineteenth-century racial theories lie at its very basis.
INDIGENOUS PEOPLES
It may be argued that with the introduction of new concepts such as adivasi
or Indigenous Peoples, an influential evaluation of the tribal category indeed
did occur. For example, it has been maintained that the new terminology
carries none of the derogatory connotations of tribe or its local variants
(upojati, pahari).46 My study of the Garos shows, however, that the introduction
of these new terms also did not foreshadow the end of the tribalist discourse:
1) the new labels do not have fundamentally different connotations; 2) the
shift in terminology also has not resulted in a deep transformation in the
theoretical approach; and 3) the very notion of one category of people sharing
a number of characteristics has remained.
Many advocates of the concept of Indigenous Peoples argue that these
people need special rights and protection if they are to preserve important
aspects of their cultures and to develop some form of political autonomy.47
They hold that all Indigenous Peoples worldwide share that their territorial
claims are not respected by governments, that they are at risk of enforced
assimilation or physical termination, and that they are in need of special
socio-economic or political protection in order to preserve their way of
life.48 Analytically however, the concept is not accurate.49 For the culturally,
linguistically, and ethnically diverse population of South Asia, the notions
of indigenousness are especially problematic. Of the hundreds of ethnic
and linguistic groupings that make up South Asia’s population, only
certain groups are considered indigenous. 50 In practice, they are the same
people who used to be known as tribes.51 My own research indicates that
in many ways the Indigenous Peoples discourse is a continuation of the
tribalist discourse. The locus of Indigenous Peoples or adivasis is still their
origin and habitat, and closely related to these characteristics, their isolation
from civilized society, their exclusion from modern life, and their link
with nature. The assumptions that these peoples cannot take care of
themselves and will disappear if they do not receive special care is very
similar to the paternalist attitude towards “naive childlike tribes”. And
although the term indigenous carries no notions of primitiveness as such,
in practice adivasi or indigenous is often just another label for the old
category of Primitive Tribes.
In some ways, the introduction of the new terms does signify an important
turning point in the tribalist discourse. The notion of indigeneity offers these
people (who are indeed generally marginalized in national politics) a vehicle
on the basis of which “marginalized tribal communities” can assert themselves
and claim rights on the basis of this identity.52 The category of Indigenous
Peoples stretches far beyond the borders of South Asia. It provides a political
platform, worldwide network, and a reason to ask for support and attention
from international aid organizations, human rights organizations and other
international associations. Another important difference is that adivasis or
Indigenous Peoples are terms often (but not always) much preferred above
other derogatory designations such as tribes by the people themselves. And
instead of being passive objects, categorized and labelled by others, the new
concepts also carry notions of self-worth and social and political emancipation.
Chapter 3 analyses how ideas of indigeneity have been incorporated into
Garo discourses of identity and help them to upgrade their self-esteem and to
feel socially, politically, and culturally safer.
It is by no means my intention to disclaim the social, economic, and
political importance of the concept Indigenous Peoples. A major setback is
however that the new concepts continue to incorporate the idea that Indigenous
Peoples are fundamentally different from modern societies, and that they can
all be grouped together on the basis of a number of similar characteristics. In
other words, the new terminology provides the modern foundation for the
continuation of dichotomization and “othering”.
THE GAROS OF BANGLADESH

We shall now take a closer look at the situation in Bangladesh. An important
shortcoming of most historical and ethnographic examinations of the tribes
or indigenous peoples in South Asia is that they only address the Indian
situation. The history of Bangladesh has followed a different trajectory and
we can only understand processes of marginalization and othering of
Bangladeshi “tribes” through a careful contextualization of these processes.
This chapter concludes with a short examination of the tribalist discourse in
Bangladesh and of how it has been reflected in colonial and post-colonial
images of Garos.
As mentioned before, this study is largely based on research carried out
between 1993 and 2000. In December 2005 I returned to Bangladesh for the
first time since then. I did not spend enough time among the Garos to assess
whether their lives had changed significantly, but some of my friends in
Dhaka indicated that things had improved a little, that Garos were now better
known than before, and that they feel they are being taken more seriously. All
the more astonishing was the description of the Garos in the Banglapedia.
The sections quoted in the introduction to Chapter 1 unambiguously expose
some of the chief dichotomies in tribalist discourse and show that Van
Schendel’s recapitulation of the tribalist image, which still dominates the
thinking of many Bangladeshis today, is also a blueprint for descriptions of

Bangladeshi Garos.Van Schendel writes that:
Such representations of the ‘tribe’ are based on the assumption that all tribes
share characteristics that are fundamentally different from, even opposite
to, those of civilized people. Principal among these are ‘childish’ qualities
that betray a lack of socialization: immoderately emotional behaviour (revelry,
sensuality, extravagance, cruelty, fear of the supernatural) and naivety
(credulity, incapacity to plan for the future).53
For example, Abdus Sattar, once called East Bengal’s “pioneer in social
anthropology”, voices this tribalist image in an almost caricatural manner.54
Sattar, who published a number of works about “the primitive tribes of
Bangladesh” during the 1970s, considered the uplift of the tribals by means
of education essential because, “isolated and left behind, the tribes will
become more inward looking and aggressive”.55 Although it would be quite
unfair to some other researchers to view Sattar as exemplary, his books can
still be easily obtained in Dhaka – where I even encountered some copies in
some Garo households – and influence people’s ideas of Bangladeshi
minorities.56 About the Garos, Sattar wrote, for example, that “[t]hey are
living in the grooves [sic] of forests and lonely lands under the shadows of
modern industrial complexes, yet zestfully retaining their own tribal identities.”
On the same page, he added that “many of them have received education and
come into the light of civilization.”57
Sattar’s writings on the Garos (or on any other Bangladeshi tribe for that
matter) not only reveal the tendency to present Garos as totally different, they
also illustrate another important feature of today’s tribalist discourse: tribes’
“formulaic quality”. His works repeat “certain formulas, stock phrases, and
ideas” that are echoed time and again in works on so-called tribal populations.58
Rosie’s reference to ideas of primitive eating and dressing habits is another
example of such tribalist stock phrases and ideas. In an environment where
food restrictions are intimately linked with ideas of purity and impurity, and
culinary habits are directly linked with status, the culture of eating and
drinking is taken as an indicator of the degree of civilization.
The same applies to the clothing of Garos. The presumed nakedness of
the Garos suggests an uncivilized character.59
Other tribalist “stock phrases or ideas”60 are the notion of “unchangeable
tribes frozen in time”, their peripheral or marginal location, their isolation
from the wider world, their rigid group boundaries, their inextricable link
with nature, and their childish, unbridled nature. In the 1960s, the
anthropologist Pierre Bessaignet wrote that “[a]lthough extensive settlements
and long standing civilization existed in the Indus plains and the Ganges
basin as early as the seventh century B.C., most other parts of this vast
subcontinent still lay covered with forests – the home of wild beasts and
scanty tribes.”61 This peripheral habitat and the unbreakable link with nature
also feature prominently in modern debates about the Indigenous Peoples
and their knowledge of nature. In the following sections, some “typical” Garo
characteristics that reveal a highly formulaic quality and degree of timelessness
are discussed.
PEOPLE WITHOUT HISTORY

Sophisticated people have long histories of civilization. Tribes have no such
history. Sumit Guha commented that “[t]heir history is seen as constituted
by two events – one which took place thousands of years ago, when they
were driven into the forests, and the second in contemporary times, when
they were being excluded from them; between these two they dwelt
untouched by history.”62 Consistent with this perspective is the absence of
comprehensive historical studies of the Garos. True, some notions of their
history (of migration) are incorporated in the ethnological accounts.
A. Playfair, for example, refers to a legend according to which Garos once
inhabited Tibet. From there they started the voyage which ended in the
present Garo Hills and surrounding plains.63 Playfair, who wrote perhaps
the most famous and influential work on the Garos, seemed well aware of
the highly speculative character of these historical assumptions. Nevertheless,
this uncertainty disappeared from many of the later works on Garos, and as
time passed, the Garos lost much of their history. Note, for example, the
following description that was derived from the Mymensingh District
Gazetteer of 1978: “The traditional history of the Garos is very obscure.
There is, however, an agreement on the point that they came from Tibet.
They exchanged some women for trees, built rafts and eventually crossed a
river to enter the Garo Hills.”64
LIVING IN ISOLATION
Another important tribalist notion is that tribes have always lived in isolation,
far away from civilized societies. In 1929, a colonel from the Britsh army
wrote that “The Garos are a wild people, having little to do with either the
Khasias or the plainsmen … and were probably driven into these regions
either by early invaders from India or by the pressure of other migrating
tribes.”65
In 1984, Mahmud Shah Qureshi introduced the most comprehensive

overview of Bangladeshi tribes with the comment that “these ethnic minorities
are scattered groups often living in hilly areas or peripheral zones.”66 A glance
at the map of Bangladesh supports these ideas. Most non-Bengali minorities
live in the border areas of the country. Peripheral habitat is also one of the
Garo characteristics. They live in the margins of the country: border areas,
forest, or hills. This does not mean that they have lived there forever. Moreover,
concepts such as periphery and marginality are relational and dependent on
the perspective one takes. Until the partitioning of India and Pakistan in
1947, the Bangladeshi Garos did not live in an international borderland. This
notion of peripheral habitat becomes really tribalist when it is directly linked
with ideas of civilization. Note what The New Nation published about the
Garos in 1988:
The Garo migrated from north-east China three or four thousand years ago
through Tibet, Coochbihar and Assam respectively and being driven out
from everywhere ultimately settled in the inaccessible dense forest of the
Garo hills and led a tight primitive life completely cut off from the outside
world till almost the advent of the British in this country.67
Stories like these suggest an innate primitiveness of tribes which only
changes when they get in touch with the “outside world”. They also explain
why F.A. Sachse found that plains Garos, who have lived in close proximity
to Bengalis, retained perhaps “some, though not all, primitive customs of
their cousins in the hill”.68
THE PRIMITIVE OTHER

The most important aspect of the tribalist discourse is that tribes stand for the
primitive Other, in contrast to the civilized Self. Principal among “tribal”
characteristics are “ ‘childish’ qualities that betray a lack of socialization:
immoderately emotional behaviour (revelry, sensuality, extravagance, cruelty,
fear of the supernatural) and naivety (credulity, incapacity to plan for the
future).”69 The civilized Self is rational, sophisticated, and contained; the
primitive Other shows opposite traits. The example of Rosie reveals how such
ideas are maintained even when experiences indicate otherwise. Such
stereotyping may take different forms. Garos were both described as fearsome,
dangerous headhunters, as well as simple-minded, friendly, and truthful.
Garo women were once described as “the most unlovely of the sex”, whereas
they now symbolize sensuality and sexuality.
Earlier in this chapter, it was suggested that primitive societies provided
pioneer anthropologists with an image of their own society seen in a distorting
mirror. Primitive man constituted the antithesis of these pioneer

anthropologists. “Primitive society therefore must have been nomadic, ordered
by blood ties, sexually promiscuous and communist.” And “primitive man
was illogical and given to magic”.70 Close scrutiny of colonial perceptions of
India reveals that the colonial observers did not merely project their counter
images on “primitive tribes” like Garos. Reality was far more complex. Note,
for example, Sachse’s characterizations of both Garos and Muhammadans.
He writes about the Garos that they are noted for their simplicity,
straightforwardness, and truthfulness, whereas “the character of the
Muhammadan masses is full of inconsistencies. They have for long been
untruthful and dishonest”.71 In this example, Muslims rather than Garos
seem to figure as the significant Other of the British.
TRIBALIST DISCOURSE AND THE

EXPERIENCED REALITY
The initial Orientalist image of India as a static society – moving neither in
time nor space – developed into a much more complex picture in the course
of the nineteenth century. Different colonial observers (who formed a far
from homogeneous category themselves) became increasingly aware of the
complexity of Indian society. Influenced by scientific developments back
home, their personal interests, and the need of the colonial state to objectify
India, they involved themselves in complicated projects to quantify and
segment the Indian population. At the basis of what seem clear-cut divisions
today, such as the Hindu-Muslim divide, Hindu castes, or the (primitive)
tribes versus (civilized) other dichotomy, lies a complex story of bewilderment,
debate, disagreement, and a great deal of confusion. Interestingly, the attempts
of the colonial state to enumerate and classify the Indian population led to
increasingly fixed categories.
Even though contemporary debates about caste and race criticize the
presumed fixed and essentialized make-up of both concepts, a similar debate
has never developed on tribes. The next chapters attempt to show how,
similarly to castes, the present-day category of tribes (upojati, adivasi or
Scheduled Tribes), which has outlived the British Raj, is the outcome of the
complex interactions between colonial projects to objectify India and local
action and responses. As a social category, tribe has a great impact on the lives
of its members, and the consequences of the tribalist discourse are manifold.
Although widely used, the images of tribe, upojati, adivasi, and Indigenous
Peoples render various problems for the people concerned, and none of the
generally accepted and/or applied terms suits social reality all too well.
The case of the Garos is a clear example of how ideas and images can
become social reality. The following chapters further examine the processes
of marginalization and othering that characterize the history of Garos and
other tribes in Bangladesh. We shall also see how tribalist discourse has
played an important role in those processes: how it has influenced seeing,
thinking, and acting, not only by outsiders, but also by the people involved;
how it has enabled rulers to exclude so-called tribals from central domains
of power; and how it has presented its members with many obstacles, in
addition to the social and economic problems that so many Bangladeshis
have to deal with today.
NOTES
1. When I quote from an interview, I use citation marks.
2. The perseverance of the “scantily dressed” Garo woman is all the more remarkable
when we read that, as early as the 1870s, H.H. Hunter noted that “among
the people living to the south of the Túra range there is a slight emigration to the
plains. The emigrants usually settle near the Gáro border, just within the
boundary of Maimansinh; many of them have now learned to till the ground
like ordinary Bengálí husbandmen, and also to dress in a somewhat similar way”.
In 1909, the English administrator Major A. Playfair, also the author of the first
monograph on the Garos, commented on the female Garo dress that Christian
as well as plains Garos wore “clothing similar to that of Bengalis and Assamese”.
See W.W. Hunter, A Statistical Account of Assam. Vol. II (London: Trübner & Co,
1879), p. 160; Playfair, The Garos, p. 25.
3. Cf. Adam Kuper, who found that notions of primitive societies “have persisted
until very recently (indeed, still survives, if no longer within mainstream
anthropology)”. Adam Kuper, The Invention of Primitive Society. Transformations
of an Illusion (London and New York: Routledge, 1991), p. 1.
4. The first recorded observations on India are from the third century B.C. The
scattered foreign records which were produced during the following centuries
were written by travellers, including Greeks, Romans, Jews, Chinese, and later
Arabs, Turks, Afghans, etc. See Bernard S. Cohn, “Notes on the History of the
Study of Indian Society and Culture”, in Structure and Change in Indian Society,
edited by Milton Singer and Bernard S. Cohn (Chicago: Aldine Publishing
Company, 1968), p. 4.
5. Although it is possible to broadly distinguish these three categories, it is also
necessary to recognize that these categories are far too simplistic, and that also
within each category, individuals or segments had different agendas and attitudes.
For such differences among missionaries, see Geoffrey A. Oddie, “ ‘Orientalism’
and British Protestant Missionary Constructions of India in the Nineteenth
Century”, South Asia 17, no. 2 (1994): 27–42.
6. Cohn, “Study of Indian Society and Culture”, pp. 6–8.

7. Sekhar Bandyopadhyay, “Caste in the Perception of the Raj. A Note on the
Evolution of Colonial Sociology of Bengal’’, Bengal Past and Present 104 (1985):
58.
8. Cohn, “Study of Indian Society and Culture”, p. 13. See also Bandyopadhyay,
“Caste in the Perception of the Raj”, pp. 56–80.
9. T.C. Hodson, India. Census Ethnography, 1901–1931 (New Delhi: Usha, reprint
1987, first published in 1937), p. 9.
10. Susan Bayly, “Caste and ‘Race’ in the Colonial Ethnography of India”, in The
Concept of Race in South Asia, edited by Peter Robb (Delhi, etc.: Oxford Press,
1995), p. 167.
11. Kuper, Invention of Primitive Society, p. 1. For an elaborate account on the
relations between British anthropology and the colonial government, see Adam
Kuper, Anthropology and Anthropologists. The Modern British School (London and
New York: Routledge, 1991), with special reference to Chapter 4: “Anthropology
and Colonialism”.
12. See for example, Bandyopadhyay, “Caste in the Perception of the Raj”, p. 57.
13. Quotations are from a letter of the Government of Bengal to the Government of
India, as recited by H.H. Risley, The Tribes and Castes of Bengal. Ethnographic
Glossary. Volume I (Calcutta: Firma Mukhopadhyay, 1981, first edition 1891,
p. iii.
14. L.A. Waddell, “The Tribes of the Brahmaputra Valley: A Contribution on their
Physical Types and Affinities”, Journal of the Asiatic Society, Part III. 1900.
Anthropology, no. 1 (1900): 2–4.
15. See, for instance Jan Breman, ed., Imperial Monkey Business. Racial Supremacy in
Social Darwinist Theory and Colonial Practice (Amsterdam: VU University Press,
1990); Albert Memmi, The Colonizer and the Colonized (London: Earthscan
Publications, 1990, translated from Portrait du Colonisé précédé du Portrait du
Colonisateur, 1957); Ashis Nandy, The Intimate Enemy. Loss and Recovery of Self
Under Colonialism (Delhi, etc.: Oxford University Press, 1990 (1983)); Said,
Orientalism; Mineke Schipper, De Boomstam en de Krokodil. Kwesties van ras,
cultuur en wetenschap (Amsterdam: Van Gennep, 1995).
16. Bayly, “Caste and ‘Race’ ”, p. 214. The lengthy article by Crispin Bates in the
same collection also substantiates this argument. In his conclusion, Bates contends
that “[a]lthough the colonial discourse of caste and tribe in India may have been
hegemonic, it was not always uncontested, and it would be a mistake to regard
it solely as the effect of a larger project aimed at ‘normalising’ the sociology of
India in order to render it more susceptible to administrative control.” Crispin
Bates, “Race, Caste, and Tribe in Central India”, in Robb, ed., Concept of Race
in South Asia, p. 256.
17. Ajay Skaria, “Shades of Wildness. Tribe, Caste, and Gender in Western India”,
in Modern Asian Studies 56, no. 3 (1997): 727.
18. Partha Chatterjee as reproduced by Skaria, “Shades of Wildness”, p. 727.
19. Skaria, “Shades of Wildness”, pp. 727–29.

20. Kuper, Invention of Primitive Society, pp. 4–5.
21. Waddell, “Tribes of the Brahmaputra Valley”, pp. 2–3.
22. Willem van Schendel, “Introduction”, in Francis Buchanan in Southeast Bengal
(1798). His Journey to Chittagong, the Chittagong Hill Tracts, Noakhali and
Comilla, edited by Willem van Schendel (Dhaka: University Press Limited,
1992), p. xx.
23. I have derived the notion of “objectifying India” from Cohn. See Bernard S.
Cohn, “The Census, Social Structure and Objectification in South Asia”, in
An Anthropologist among Historians and Other Essays, edited by Bernard S. Cohn
(Delhi, etc.: Oxford University Press, 1987), pp. 224–54.
24. Risley, Tribes and Castes of Bengal, p. vii.
25. Dipesh Chakrabarty, “Modernity and Ethnicity in India”, South Asia XVII,
Special Issue (1994): 147.
26. James C. Scott, Seeing like a State. How Certain Schemes to Improve the Human
Condition Have Failed (New Haven and London: Yale University Press, 1998),
p. 11. Scott’s book is a challenging exposé about the causes and consequences of
modern states’ standardization projects or “large-scale social engineering”.
27. Cohn, “Census, Social Structure and Objectification”, pp. 242–43.
28. See Census of India, 1921. Volume V, Bengal, Part I, Report (by W.H. Thompson)
(Calcutta: Bengal Secretariat Book Depot, 1923), p. 345.
29. For an elaborate discussion of recent scholarship on caste, see: Bayly, Caste,
Society and Politics.
30. Inden, Imagining India, p. 66.
31. Bayly, “Caste and ‘Race’ ”, pp. 165–66. I agree with Bayly that researchers from
the second category have convincingly demonstrated that the interaction between
colonial perceptions and indigenous responses influenced the self-perception of
indigenous social or ethnic groups and resulted in the articulation of community
identities and the creation of caste histories. See, for example, Bandyopadhyay,
“Caste in the Perception of the Raj”, pp. 56–80; Sekhar Bandyopadhyay, Caste,
Politics and the Raj. Bengal 1872–1937 (Calcutta and New Delhi: K.P. Bagchi &
Company, 1990); Sekhar Bandyopadhyay, Abhijit Dasgupta and Willem van
Schendel, “Introduction”, in Bengal. Communities, Development and States, edited
by Sekhar Bandhyopadhyay, Abhijit Dasgupta and Willem van Schendel (New
Delhi: Manohar, 1994), pp. 1–16; Swaraj Basu, “The Colonial State and the
Indigenous Society: The Creation of the Rajbansi Identity in Bengal”, in
Bandyopadhyay, Dasgupta and Van Schendel, eds., Bengal, pp. 43–64; Lucy
Carroll, “Colonial Perceptions of Indian Society and the Emergence of Caste(s)
Associations”, Journal of Asian Studies 37, no. 2 (February 1978): 233–50;
Cohn, Anthropologist among Historians; Cohn, “Study of Indian Society and
Culture”, in Singer and Cohn, eds., Structure and Change; Frank F. Conlon,
“The Census of India as a Source for the Historical Study of Religion and
Caste”, in The Census in British India. New Perspectives, edited by N. Gerald
Barrier (New Delhi: Manohar, 1981), pp. 103–17.
32. Bayly, “Caste and ‘Race’ ”, pp. 165–67.

33. Ibidem, p. 168.
34. Cf. John Nesfield versus H.H. Risley. Nesfield denied any general difference
between so-called Aryan and aboriginal blood, arguing that Aryan blood had
been mixed with the indigenous by the time the caste system evolved. Risley, on
the other hand, maintained that the primary distinction was one of race. He
argued that social ranking was a matter of purity of blood, with the communities
of pure Aryan and pure aboriginal stock at the top and bottom respectively.
Census of India, 1911. Volume I, India, Part I Report (by E.A. Gait) (Calcutta:
Government Printing, 1913), pp. 380–81.
35. Sumit Guha, Environment and Ethnicity in India, 1200–1991 (Cambridge:
Cambridge University Press, 1999), p. 19.
36. Census of India, 1931. Volume V, Bengal and Sikkim, Part I (by A.E. Porter)
(Calcutta: Central Publication Branch, 1933), p. 421.
37. Skaria, “Shades of Wildness”, p. 727 and pp. 729–30.
38. Susan Bayly as quoted by Skaria, “Shades of Wildness”, p. 728.
39. Census of India, 1931. Bengal and Sikkim, p. 441.
40. The following are only two out of many examples: The Aguri are described by
Buchanan as “tribe which makes pretences to be of Khyotryo dignity”. In Risley’s
“Tribes and Castes of Bengal”, the A’guri return as a “cultivating and trading
caste [ital. mine]”. This may also suggest that people who were once considered
a tribe gradually transformed into a caste through Hinduization.
Another example is the Kaibortos. Buchanan writes that “the chief part of
the pure Hindus of Bengal, that have settled in this district, are the dubious tribe
of Kaibortos”. Further on he describes the Jhalo as “fishermen of the Kaiborto
caste”. Risley, who provides an extensive account of the Kaibartta, refers to
them as a “sub-caste of Kewats in Bengal”. Francis Buchanan, as compiled by
Montgomery Martin, The History, Antiquities, Topography, and Statistics of Eastern
India; Comprising the Districts of Behar, Shahabad, Bhagalpoor, Goruckpoor,
Dinajepoor, Puraniya, Ronggopoor, and Assam. Volume V, Rangpur and Assam
(Delhi: Cosmo Publications, 1976, first published in 1838, pp. 529–31 [Hereafter
referred to as Martin, ed., Eastern India]; Risley, The Tribes and Castes of Bengal.
Vol. I, pp. 12–13 and 375–82.
41. “Mr Nesfield’s Theory of the Origin and Nature in Indian Caste”, Census of
India, 1901. Volume I, India, Ethnographic Appendices (by H.H. Risley) (Calcutta:
Government Printing, 1903), pp. 231–32.
42. This term is from Ambasta, Capitalist Restructuring, p. 2. A number of scholars
of Southeast Asia have produced interesting studies on upland-lowland
dichotomies, in which they demonstrate that the uplands “have been constituted
as a marginal domain through a long and continuing history of political,
economic, and social engagement with the lowlands”. In other words, “Marginality
must therefore be understood in terms of relationships, rather than simple facts
of geography or ecology”. Tania Murray Li, “Introduction”, in Transforming the
Indonesian Uplands: Marginality, Power and Production, edited by Tania Murray
Li (Australia, etc.: Harwood Academic Publishers, 1999), p. xvii. See also

“Relational Histories and the Production of Difference on Sulawesi’s Upland
Frontier”, Journal of Asian Studies 60, no. 1 (2001): 41–66.
43. J.C. Heesterman, The Inner Conflict of Tradition. Essays in Indian Ritual, Kingship,
and Society (Chicago and London: The University of Chicago Press, 1985),
p. 118.
44. Guha, Environment & Ethnicity, p. 4.
45. See, for example, Ambasta, Capitalist Restructuring; Crispin Bates, “Lost Innocents
and the Loss of Innocence: Interpreting Adivasi Movements in South Asia”, in
Indigenous Peoples of Asia, edited by R.H. Barnes, Andrew Gray, and Benedict
Kingsbury (Michigan: Association for Asian Studies, 1995), pp. 103–20; Bates,
“Race, Caste and Tribe”, pp. 219–59; Chaudhuri, “The Myth of the Tribe?”, pp.
125–56; Devalle, Discourses of Ethnicity; Karlsson, Contested Belonging; Guha,
Environment and Ethnicity; Skaria, “Shades of Wildness”; Van Schendel, “Invention
of the ‘Jummas’ ”, pp. 95–128.
46. Jan Breman, for example, points out that the local designation for the tribes of
Surat district (in South Gujarat), namely, kaliparaj [dark coloured], has been
replaced by the term adivasi: “a term which has no pejorative meaning and
which, in its sense of ‘original inhabitants’, has become the official name for all
such categories of tribals”. Jan Breman, Of Peasants, Migrants and Paupers. Rural
Labour Circulation and Capitalist Production in West India (Delhi, etc.: Oxford
University Press, 1985), p. 115.
47. Corresponding opinions lie at the basis of the special policies for Scheduled
Tribes, designed to socially, economically, and politically uplift those tribal
peoples “whose primitiveness, backwardness, or isolation made them deserving
of special treatment”. Galanter, Competing Equalities, pp. 151–52.
48. Thomas Hylland Eriksen, Us and Them in Modern Societies. Ethnicity and
Nationalism in Mauritius, Trinidad and Beyond (Oslo: Scandinavian University
Press, 1992), p. 6.
49. There has been much debate about the notion of “indigenous” and
“indigenousness”. For a particularly critical evaluation of use of the term for the
South Asian context, see André Béteille, “The Idea of Indigenous People”,
Current Anthropology 39, no. 3 (April 1998): 187–91; for a more general
recapitulation of arguments, see for example, Andrew Gray, “The Indigenous
Movement in Asia”, in Barnes, Gray and Kingsbury (eds.), Indigenous Peoples of
Asia, pp. 35–58.
50. Questions of authenticity and indigenous status are particularly complicated for
Northeast India and Bangladesh, where many of the so-called tribal groupings
are not considered (nor consider themselves) indigenous to their present habitat,
but are believed to originate from places like Tibet or China. A well-known
critic of the use of “indigenous peoples” in the Indian context is André Béteille.
See “The Idea of Indigenous People”, Current Anthropology 39, no. 2 (1998):
187–91.
51. For example, the Independent Commission on International Humanitarian

Issues simply translated indigenous in Asia as tribal. Gray, “The Indigenous
Movement in Asia”, p. 38.
52. See for example, Bengt K. Karlsson, “Anthropology and the ‘Indigenous Slot’:
Claims to Debates about Indigenous Peoples’ Status in India”, Critique of
Anthropology 23, no. 4 (2003): 403–23.
53. Van Schendel, “Invention of the ‘Jummas’ ”, p. 103.
54. A.K. Nazmul Karim (Head of the Dhaka Department of Sociology of Dhaka
University) described in his foreword how Sattar as a revenue officer of the
government of East Pakistan came in touch with the aboriginal tribes of that
part of the world and how he ably demonstrated, in his several publications,
“how to scientifically handle social anthropological materials”. In Abdus Sattar,
In the Sylvan Shadows (Dhaka: Bangla Academy, 1983, first published in 1971).
55. Abdus Sattar, Tribal Culture in Bangladesh (Dacca: Muktadhara, 1975), pp. 5–
6. From the same author, see also The Sowing of Seeds. The Sociology of Primitive
Sex (Dacca: Abdeylebros & Co, 1978); In the Sylvan Shadows (Dhaka: Bangla
Academy, 1983, first ed. 1971).
56. For an overview of studies about the so-called tribes of Bangladesh, see Quereshi,
ed., Tribal Cultures in Bangladesh. For a recent collection of articles for both East
and West Bengal, see Van Schendel and Bal, eds., Banglar Bahujati.
57. Sattar, Sylvan Shadows, p. 191.
58. This concept is derived from F. Padel and quoted by Lionel Caplan. Caplan,
Warrior Gentleman, p. 8.
59. Cf. Willem van Schendel, “A Politics of Nudity: Photographs of the ‘Naked
Mru’ of Bangladesh”, Modern Asian Studies 36, no. 2 (2002): 341–74.
60. Cf. Van Schendel, “Invention of the ‘Jummas’ ”, pp. 104–05.
61. Pierre Bessaignet, “Tribes of the Northern Borders of East Pakistan”, in Pierre
Bessaignet, Social Research in East Pakistan, edited by idem. Second revised
edition (Dacca: Asiatic Press, 1964), p. 172.
62. Guha, Environment and Ethnicity, p. 19.
63. Playfair, The Garos, pp. 7–14.
64. Bangladesh District Gazetteers, Mymensingh (Nurul Islam Khan, ed.) (Dacca:
Bangladesh Government Press, 1978), p. 58.
65. L.W. Shakespear, History of the Assam Rifles (Calcutta: KLM, 1977, first published
in 1929), p. 30.
66. Qureshi, “Foreword”, p. xv.
67. Ali Nawaz, “The Tribal and Other Links of Bangladesh”, The New Nation,
21 February 1988.
68. F.A. Sachse, Mymensingh, Bengal District Gazetteers (Calcutta: Bengal Secretariat
Book Depot, 1917), p. 40.
69. Van Schendel, “Invention of the ‘Jummas’ ”, p. 103, n.34.
70. Kuper, Invention of Primitive Society, p. 5.
71. Sachse, Mymensingh, pp. 40–41.
3
‘THE IMPORTANCE OF BEING GARO’:
GARO NARRATIVES OF SELF
‘DID SHE NOTICE THE BAD THINGS?’

To an outsider, Bangladeshi Garos come across as a close-knit, harmonious,
peaceful and hospitable community.1 In view of their efforts to uphold such
an image, this is no great surprise. On several occasions, I noticed anxiety
about me, an outsider, looking backstage. More than once, when Suborno
and I returned from the village to Dhaka for a couple of days, Suborno was
confronted with some of his friends worrying about what I, an outsider, had
seen and come to know. “Did she notice the bad things?”, they would ask.
And then they would instruct Suborno to only show me the positive sides of
their community. Such behaviour is by no means exceptional for people or
communities that are subject to discrimination, and who, like Suborno’s
anxious friends, tend to magnify differences between themselves and others,
and to play down discord and hide problems within the group in order to
maintain a positive image for the outside world.2
This chapter centres on Garo narratives of Self and examines how these
are influenced by, as well as inform, tribalist discourse. Soon after I began
my research, I found that being Garo, whatever that entails, is an important
component of the social identity of most Garos. Whether they are villagers
or city dwellers, male or female, high class or low class, educated or
illiterate, all share a strong ethnic awareness. Being Garo determines to a
50
Garo Narratives of Self 51
large extent their self-perception, how they organize their lives, and how
they relate to others, within and across ethnic borders. This shared identity
does not mean, however, that one and all experience and express their
identity in the same way, nor that it has the same relevance for everyone.
Close scrutiny of emic discourses at work amongst the Garos reveals a
heterogeneous understanding of Self. This chapter analyses five different
narratives of Self and ensuing self-expressions.
My study of the Bangladeshi Garos revealed that dominant notions of
tribe have a strong bearing on emic discourses of Garo-ness and serve as a
guideline or touchstone for Garo images and representations of Self. Unlike
the previous chapter, which presented a historical analysis of the genesis of a
static image, on the basis of a largely body of etic accounts of tribes, this
chapter concentrates on contemporary narratives of Self. The focus on the
present day is largely dictated by the lack of emic historical accounts that
reveal how in the past (people known to be) Garos thought of themselves and
others. Ethnic relations with the majority population have for long been
highly asymmetrical in terms of access to political power and economic
resources, as I will also discuss in much greater length in later chapters. Here
it is important to take account of the fact that this asymmetry is also reflected
in ideas and representations of Self. As a consequence, emic descriptions of
Garo-ness need to be assessed with as much caution as the historical
publications of outsiders.
The full range of Garo ideas and presentations of Self will not be
considered here. Instead I shall discuss five prevailing narratives of Garo-ness.
Two of these are, what I call, private narratives of Garo-ness3 and three of
them are public presentations meant for a non-Garo national and international
audience. In this way I attempt to differentiate self-reflections and presentations
that come close to how people really think and feel, and other presentations
with an agenda.4 I decided to concentrate on written and spoken narratives
rather than other self-representations.5 During my research, I paid much less
explicit attention to the many other ways – non-verbal communication and
silences, or cuisine and clothing – that display how people experience and
express their identity. I did, evidently, notice how people dressed, what they
ate, how they lived, how they organized the cultural shows and performed
their dances and traditional rituals. Here, such non-verbal representations are
only presented as illustrations. I was fortunate enough to be able to see
behind the curtains, although not always, not everywhere, and not with
everyone. As time passed by, relationships of mutual trust were developed and
my network of Garo friends, acquaintances, and “relatives” slowly expanded.
People like Suborno’s anxious friends began to accept me and to confide in me.
It is important to note that the five narratives are not mutually exclusive.
Communication lines amongst the Garos (also between city dwellers and
villagers) are generally short and well-maintained, and people are often well-
informed about Garos in other parts of Bangladesh. Notions of what Garo
identity and culture are supposed to mean also travel widely. At times some
of the ideal-types even become self-fulfilling prophesy, when people not only
begin to believe in them but also start acting upon them. The same applies to
dominant, pejorative views held by others.6 And the other way around,
representations of Self have an obvious impact on etic perceptions of people.
Although by no means easy to unravel, there is a clear connection between
imagined and experienced identities and images.
PRIVATE NARRATIVES
1. ‘BORN A GARO, ALWAYS A GARO’
Six weeks after I commenced my research in Bangladesh, Suborno and I took
off for the countryside and found ourselves a place to stay in the village of
Bibalgree, Haluaghat thana. I soon discovered two things about Garo identity.
First, that Garo-ness was important for all villagers, and second, that the
villagers took this identity as a fact of life. This applied not only to Garos but
to their neighbours as well. Villagers with different ethnic backgrounds
frequently met and interacted, but nevertheless remained conscious of each
other’s ethnic and religious backgrounds. Differences were obvious enough,
and unlike the younger city generation, who are busy (re)inventing Garo
traditions, elderly villagers did not worry about losing their distinct culture.
One explanation for the absence of such worries – which are, as we shall
see, frequently ventilated by urban middle class Garos – lies in the primordialist
and essentialist perspectives on Garo ethnicity.7 The villagers generally explained
their distinct Garo identity in terms of birth, bloodlines, and race. To the
question why people were Garo, what made them different, answers were
fairly uncomplicated: “A Garo is a Garo”, or “my parents were Garos, so I am
Garo”. The elderly Garo woman Sammati (between 60 and 70 years old) used
an interesting comparison to explain why a Garo is a Garo: “It’s by birth. Like
with the buffalo. If a buffalo gives birth to a baby, it will be a buffalo again.”
A Garo is a Garo because he or she is born as one. Suborno later added to
Sammati’s explanation that “a buffalo is a buffalo. Even if it is raised by a lion,
it will never be a lion.” In the previous chapter, I showed that race and racial
characteristics are important aspects of the tribalist discourse. “Typical Garo”
looks are both considered relevant signs as well as products of their
distinctiveness. Garos themselves also consider their physical appearance as

an important facet of their ethnic identity. It points at the timeless and
essentialist quality of their separate identity.
Racial continuity and purity are directly linked to another typical tribalist
feature, the presumed remoteness of tribes. A well-known Bengali journalist
of environmental and human rights issues underlined this isolated character
of Bangladeshi adivasis (as he called them) and he told us that because of this,
the Garos have been able to preserve their original culture. The journalist
further said that:
“He [referring to Suborno] is an adivasi because he has not mixed much
with Bengalis. He has retained his original culture, identity, his blood type.
So, he is more original than I am. … We [Bengalis] are not pure. We have
mixed so much. My blood, there is no Bengali blood, it’s mixed blood.”8
Although the journalist generally takes a pro-adivasi position in public
debates, the underlying tone is paternalistic, and the reasoning is essentially
tribalist. He continued the interview with the argument that adivasi pureness
accounts for their simplicity (almost childlike innocence) and inability to
deal with the cunning Bengalis who invaded their traditional territories. He
pointed out that “the people [Bengalis] who lived around these tracts
gradually encroached upon the trees and lands; and they could successfully
encroach upon lands, cut trees, grab lands, because of the simplicity of the
people who lived in these areas.”
Primordialist Garo narratives of Self are remarkably similar. They too
refer to the dichotomy of “pure” versus “mixed” and present Garos as a pure
or homogeneous jat or jati,9 in contrast with Bengalis, who constitute a
“mixed race”. For example, my host (“uncle”) in the village commented that
their very purity accounts for physical similarity among Garos, and absence
of purity would explain why offsprings of mixed marriages look so different.
In the perception of Garos, racial purity has many positive connotations.
Pureness equals virtue, unity, and cooperation: all highly valued qualities.
This explains why the mixed character of Bengalis, which is reflected in their
physical dissimilarity, accounts for the vast differences in their thinking,
morality, ideology, organization, and cooperation, and is looked down upon.10
In villages, boundaries between Garos and others are clearly marked and
have been for many decades.11 It is probably also for that particular reason
that village Garos do not question their ethnic identity. Differences between
them and their Bengali neighbours are obvious enough, and a distance has
been well maintained. Why should the villagers question Garo identity if
differences are so much part of their daily lives; when people see their
“feelings of being Garo” materialized in so many practical differences? After

all, they speak the Garo language, practise the Sangsarek or Christian religion,
and keep pigs in their gardens. Moreover, they have experienced the
consequences of their distinct status on several dramatic occasions during the
past decades (see also Chapter 8).
The aforementioned examples also point to two other interesting aspects
of the primordialist narrative of Self: firstly, that Garos resort to the tribalist
discourse to distinguish themselves from Bengalis but attach different meanings
to these concepts and ideas; and secondly, that comparisons are based on
dichotomization: pure versus mixed, loose versus restricted, naive versus clever.
Whereas “pure” implied “isolation” and “naivety” for the Bengali journalist
(and many others), Garos take it as a sign of unity, common origin, and
history, and a strict adherence to community norms and values (in contrast to
“looseness”). Where freedom of movement is seen as indecent behaviour,
Garos consider it a sign of liberty and progress. Instead, to their minds purdah
symbolizes backwardness and gender inequality.12 Garos avail themselves of
opposites but turn them around to their own advantage: Garos are clean,
Bengalis dirty; Garos are honest, Bengalis unreliable; Garos are peaceful,
Bengalis are aggressive; Garos are quiet, Bengalis loud; Garos are generous
and Bengalis stingy. It is, in this context, the Muslim Bengali who stands for
the Significant Other. In Chapter 6, I show that, where other outsiders are
concerned, Garo systems of classification operate on a much more ambiguous
inclusion/exclusion basis.13
The primordialist narrative of Garo-ness presents Garo identity as a mere
fact. It prevails among (elderly) villagers, who seem to face few problems in
understanding why they are different from their neighbours and who experience
no gap between what they feel and what they see. To them their distinctiveness
is obvious; it is a given, a fact of life. Next I show that many of the
experienced distinctions (read overt signs) seem to disappear when Garos
move to the cities, particularly Dhaka, and (need to) integrate more fully into
Bengali society. Often this process leads to the uncomfortable belief, especially
among some young educated and ethnically conscious Garos, that Garos are
losing their culture and, as the ultimate consequence, themselves.
2. IDEAS OF A LOST CULTURE

The second private narrative revolves around the argument that Garos are
in the process of losing their unique identity and culture.14 Johnson, who
was born in 1949 and who left the Garo area at the age of twenty, has been
living in Dhaka since 1975. He explained why Garos no longer practise

their culture:
“We are not practising Garo culture. Actually, we are just remembering
Garo culture. Sometimes we just practice it for enjoyment. … Actually, we
don’t have any opportunity to practice it. What was our culture? To drink
wine, to worship the sun and the moon? Do you think that is possible now?
We are Christians now, we believe in God.”
Johnson himself is not confused about his Garo identity. He makes fun of the
traditional Garo lifestyle and explains that Garo culture no longer agrees with
their modern lifestyle. What was their culture anyhow? “To drink wine and
to worship the sun and the moon?” Johnson remembers the traditional
culture well but chooses not to practise it. Examples like this, however, worry
many young and educated Garos, who have temporarily or permanently
settled in Dhaka and who are well adjusted to and acquainted with middle-
class Bengali lifestyle. If Garos lose their distinct culture, what makes them
different from other Bangladeshis?
MOVING TO DHAKA
Since the 1960s, and increasingly since the 1980s, Garos have started to
migrate to big cities such as Dhaka and Chittagong. Particularly Dhaka has
become very popular. The need for jobs has induced both men and women
to leave their villages. Others have come to the big cities for higher education.15
For many Garos, city life has an adventurous ring to it. Sometimes villagers
believe that the big city promises a more exciting life. So we can find men,
women, and even children leaving their homes simply because they find
village life boring. However, like everywhere else in Bangladesh, proper jobs
are difficult to find and migrants discover that the city does not offer the kind
of opportunities they anticipated. They often settle for less than expected,
and sometimes they return to their homes in the villages.16
Today, Dhaka is home to a couple of thousand Garos who are engaged in
a variety of ways. Many Garos are household servants. Often they work as
cooks, nannies, drivers, or guards. Other Garos work in garment factories,
and quite a lot of girls and women are employed in beauty parlours. We can
find hundreds of these employees living in the cramped workers’ lines that
surround the posh areas such as Banani, Gulshan, and Baridhara. If possible,
they live together in the same quarters, and if a Bengali family moves out, it
is likely that a Garo family moves in.
Other Garos have found their way to one of the many (Christian) non-
governmental organizations (NGOs) or church organizations in the capital.
They often have their families living with them in the city. These Garos
have developed into a new middle class. Although they appear best adjusted
to Bengali (middle class) lifestyle – sometimes their children cannot even
speak the Garo language – many of these Garos are particularly conscious
of their ethnic identity. Often they hold influential positions among the
Bangladeshi Garos.
A third important section of Dhaka Garos are the students. More and
more boys and girls are coming to Dhaka to study in one of the many colleges
or at Dhaka University. They live with relatives or stay in one of the student
hostels that are spread all over town. Like the new Garo middle class, these
young students are much better acquainted with and integrated into Bengali
society than their parents in the villages. At the same time, they are very
conscious of their Garo identity. How can we explain that the discourse of the
lost culture prevails particularly among these students or educated middle-
class Dhaka residents?
FEELING GARO, ACTING BENGALI

Like village Garos, many Dhaka residents, particularly the higher educated
ones, argue that Garos and Bengalis are different. They find it hard to relate
to Bengalis and prefer Garo company. Although they are familiar with
Bengali lifestyle and consider Bangladesh as their country and homeland,
time and again they find themselves confronted with feelings of being different.
In the following interview fragment, Semion, a Garo student from Dhaka
University, reports his sense of loneliness in this Bengali-dominated
environment:
“At first there was the charm of city life. I had this dream of going to the
city, so to be there was exciting. But because I am Garo and had always been
with Garos, it felt bad sometimes to see Bengalis everywhere. Bengalis on
the right, Bengalis on the left. In front and behind there were Bengalis. I
cannot be free with them in the way I can be with Garos.”
Semion was not the only Garo who reported feeling lost and lonely. Most
migrants told us how they went searching for other Garos soon after they had
arrived. Rohil was among the first Garos who came to Dhaka. He was offered
a scholarship to study there. He came in 1968, at the age of seventeen. In
1973, Rohil started a Garo organization with the aim of bringing together all
Dhaka Garo. For five years, the Garo Progothi Shongho (progressive Garo
organization) ran successfully, but then people started to lose interest. In

1979, its activities stopped.
It is particularly the lacuna between feeling different and a Bengali
lifestyle that gives rise to the complaint that Garos are losing their culture.
They feel different but find it hard to explain that difference. One distinctive
attribute, however, continues to play an important role in processes of
classification, both among Garos and non-Garos. Their physical appearance
often gives away their distinct ethnic background. Many of their Bengali
countrymen know little or nothing about Garos and may not recognize
them as such, but they often set them apart from Bengalis. I heard numerous
stories of Garos being mistaken as foreigners or recognized as upojati, or out
on the streets they may be called Chakma Chakma.17 Nevertheless, at the
same time these Garos realize that not all Garos can be told apart from
Bengalis so easily. Particularly in situations where Garos do not know each
other personally, it becomes apparent that appearance is composed of more
than physical characteristics. Behaviour, facial expression, comportment,
and dress all contribute to someone’s looks. The better assimilated in
Bengali society, the less easy it is to recognize a Garo. A young middle-class
Garo couple, Nihar and Lipi, told us how they often try to spot Garos in
the crowd, and how they, too, take a flat nose as an important indication.
They also acknowledged, however, that it is not always easy to recognize
each other and that physical appearance alone is not enough to establish
whether a person is “genuinely” Garo:
Nihar: “When I see any ‘flat nose’, I ask him where he is from. We have
special feelings for each other, not only among ourselves, but among all
Mongoloid people. … We also wonder whether he is a Garo or not. And all
tribal people are from a village, that’s why we wonder if he is in trouble and
needs some help.”
Lipi: “Once, we went to Mirpur. There we saw a rickshaw puller who
looked like a Garo. I thought he was Garo, so I asked him. But he wasn’t.
He was a Bormon from Lalmonirhat.”18
The following fragment is taken from a letter which I received from a Garo
friend. Here, he describes his journey with a companion to a Christian
mission in Savar, a town near Dhaka:
“We knew that we had to take a rickshaw to get there. People said there are
some Garo rickshaw pullers but we didn’t see any of them. So we took a
Bengali guy who knew where the mission was. As soon as we had proceeded
a few metres we saw a guy who looked like a Garo. He was smiling, standing
next to his rickshaw. We were sure that he was a Garo, but we do not know
why we felt like that. It felt very normal, and naturally we shouted: ‘Na
Mandi ma’ [Abeng language for ‘are you Mandi’]? He gave a big smile and
answered: ‘a·we’ [Garo for ‘yes’]. We felt bad that we hadn’t tried further,
otherwise we could have taken him. We told him that we had been looking
for a Garo rickshaw puller but that we had not seen him. He just smiled at
us. Later we told each other that we had made a mistake. We could have
taken that Garo. He would have been helpful in finding this place that we
didn’t know. Why we felt like that, I don’t know.”19
While travelling with Garos in Dhaka city or other places outside the Garo
area, I frequently witnessed similar situations. My companions were keen to
meet other Garos and would check them out when they thought they spotted
them. If we, on arrival in Haluaghat caught sight of a Garo riksha wallah, he
could be sure of clientele. The letter provides an explanation for such keen
search for other Garos: the two friends found themselves in an unknown
place and the presence of other Garos gave them some sense of security. They
were not alone.
While the city Garos worried about Garos losing their culture, they
showed no real interest in knowing more of Garo traditions. Instead they
preferred to distinguish themselves from Bengalis in new and modern ways,
which fit contemporary middle-class Bengali and Christian lifestyles. Today,
we can observe a fast development of all kinds of typical Garo art, songs, and
dances. Old Sangsarek festivals are remodelled to fit their relatively new
Christian lifestyle, new traditional dances are choreographed, and the radio
programme Salgetal [Achik for “new sun”] plays newly composed traditional
Garo songs. Worries about losing their culture have fed into the invention of
new (public) cultural events and performances. Hereafter I shall demonstrate
how these newly invented images, often put forward by the same middle-class
urban Garos, both interweave and play with dominant notions of tribe, also
to the advantage of the Garos themselves.
PUBLIC NARRATIVES OF GARO-NESS

The various different presentations of Self in leaflets, brochures, and
articles, for a broad public of non-Garo readers, reveal the image that
Garos like to uphold of themselves, but also how Garos comply with the
tribalist discourse by copying tribalist arguments and terminology, or by
bending the tribalist discourse to their own advantage. Below I discuss
three more narratives by Garos, about Garos, each grounded in a different
strategy. These strategies can be captured with the terms “compliance”,
“manipulation”, and “denial”.
1. A NARRATIVE OF MODERNITY
The following quotation is derived from the introduction to the first English
issue of the Mandi-rang-ni Chiti [Garo Newsletter]. It catchingly reflects the
need which has been felt more and more urgently by several Garos to improve
the knowledge of, and to influence the image of, the Garo people of Bangladesh.
It is somehow in order to introduce ourselves as a people, on behalf of all
who work in Dhaka, that we have decided to publish this letter in English
twice a year. Ignorance, at its best, creates nothing (at its worst, we know
how harmful it can be) – knowledge, on the other hand, opens up many
possibilities. We will be pleased to come into contact with you on a regular
basis and share with [sic] our culture and thoughts.20
This particular Garo newsletter was published in order to introduce the
Garos of Bangladesh as a group to expatriates in Dhaka, who employ a good
number of Garos in their homes as housemaids, cooks, gardeners, drivers, or
guards.21 The subject on page two of this four-page issue is captured under
the title “Who Are the Garos?” The answer to this question commences as
follows: “The Garos used to be known as head-hunters, and were the fear of
British and Bengalis alike. Now they have learned modern ways and have to
a large extent become literate.”22
Perhaps a primitive and fearful people once, Garos have now become
educated. They have adapted to modern society, and they have adopted
modern ways of living. In other words, they no longer are the primitive
people they once were. Garos may feel an increasing need to be placed on the
map of Bangladesh but feel uncomfortable with their traditional image.
The short article continues with an outline of some “typical Garo”
characteristics, which can also be found in practically all contemporary
anthropological accounts. The author mentions the ethnic and linguistic
origins of the Garos, their habitat, language, traditional clothes, the
disappearance of slash-and-burn cultivation (as a result of deforestation), and
the matrilineal kinship organization of the Garos. The fourth and last page
introduces Paraka, the organization that is behind the publication of the
newsletter. Here, it is written that:
The centre is run mainly by young people, these being the ones who most
keenly feel the threats and opportunities of modern society. The Garos have
moved from a close-knit, simple village society, complete with its own laws
and customs, to the modern international society within less than a hundred
years. Like so many other peoples in Asia, they had not participated in the
development which led up to our present-day society – they were swept into
it and have had to take it all in their stride to be able to survive. It is true
that the Garos have managed better than many other small indigenous
peoples to preserve their identity; but the Garo-identity is stronger among
the less educated than among those, every year more numerous, who make
it to the higher degrees.23
This quotation presents the image of simple, isolated villagers who have
developed into an internationally oriented community, well-adjusted to modern
life while preserving their distinct identity. Although Garos themselves were
no actors in these processes of change, they have managed to survive without
losing their identity. It also suggests that higher educated Garos (here educated
equals adjusted to modern society) are less ethnically conscious than Garos
who are less educated (and therefore less tuned to modern life). The small
newsletter, which by no means offers exceptional opinions, clearly shows how
Garos portray themselves to others in a fashion that so closely resembles
dominant notions of tribe. Garos have not developed their own independent
public discourse but fit images created and re-created by others. At the same
time, however, they distance themselves from that same tribalist image by
insisting that they have developed into a modern people. They deal with the
negative tribal image by asserting that they have outgrown that image.
2. GARO PRIDE AND VICTIMHOOD

Memories of victimhood constitute another important ingredient of Garo
public presentations. Often they stress their position as victims, by pointing
at the Bengali Muslims and the post-colonial state as their main enemies and
the cause of many problems, and by referring to the autonomous position
they once supposedly had. Throughout this book, I also argue that the history
of the Bangladeshi Garos is marked by various occasions on which Garos fell
prey to oppressive acts by local zamindars, the central state, or local neighbours.
Here, I want to examine the public representations of Garos in which they
conform to their tribalist image as helpless and naive, and call upon others to
protect them and to grant them special rights.
Two kinds of Garo victimhood stand out. Firstly, as a (Christian) tribal
or adivasi minority, Garos are victims of (Muslim) Bengali repressions and
post-colonial state policies and therefore in need of special protection. Secondly,
as tribals or adivasis, as children of the forest, Garos are inextricably linked to
nature, and since their very existence depends on these links, they will vanish
when the forests vanish. The following is a quotation from an article that was
published in a special issue of Chiring, a magazine distributed by the Dhaka
branch of the Bangladesh Garo Student Organization (Bagachas):
There is a saying that, if a child does not cry, even its mother does not want
to give it the breast, so the child cries in order to get mother’s breast. But the
sorrows of the forest dwellers are confined to the forest. The silent crying of
their hearts cannot be heard by the civilized people who live in the cities.
But the most regrettable thing is that in this modern age of civilization there
are some people who live in inaccessible hilly villages and there is nobody to
bother about them. … The cunning infiltrators have beguiled them of their
forests, have driven them out, and taken forcible possession of their lands,
those forest dwellers today, homeless and landless, move from one place to
another in search of forest and safety for life. They are so helpless, having to
live like foreigners in their own land.24
The author of this article alludes to many aspects of the tribalist discourse: he
compares forest dwellers, who live in inaccessible hills and forests, with
helpless infants, who have to cry in order to be nurtured. Their powerlessness
has rendered them homeless and landless, and they are dependent upon the
help of other, civilized people. In another magazine, a learned Catholic Garo
priest underlines the powerless position of the so-called aborigines. He writes
the following:
The act of ethnic cleaning of aborigines by the powerful political is going on
still. The aborigines being mostly illiterate, underdeveloped, poor, socially
disorganized, politically voiceless and physically helpless do not have the
least hope to survive the constant persecutions by the politically powerful
races except trust wholeheartedly in the divine providence.25
This quotation, too, emphasizes the helplessness and therefore hopelessness
of the so-called aborigines. This author suggests that these most “undeveloped
of all races” are left with only one hope: “divine providence” (that is, the
Christian God).26 Hence this discourse of Garo pride and victimhood capitalizes
both on the Indian reservation policies for Scheduled Tribes and on the
popular argument that Indigenous Peoples are in need of special protection.
3. PEOPLE OF NATURE
The victim attitude of Garos particularly concerns their “indisputable” relation
to nature. This argument goes perfectly well with the contemporary Indigenous
Peoples discourse, which centres on environmental issues and people’s rights.
Guha argues that “[i]n addition to the environmental concerns about
endangered species and ecosystems, there is also the parallel concern for
endangered cultures and ‘indigenous’ peoples – a concern perhaps strongest
in those parts of the world (such as the Americas) where such entities have
been most effectively triturated in the recent past.” Guha continues to argue
that with funds forthcoming, “entities could not be lacking, and lists of such
peoples have duly been generated, helped, of course by the poetic vagueness
with which they are often defined”.27
This argument is perfectly illustrated by a documentary about the Garos
of Bangladesh. This video was produced by a Bengali journalist in cooperation
with the Garos, to report on a number of environmental problems in the
Modhupur Forest, an area where some 14,000 Garos live. (The large majority
of Bangladeshi Garos, some 60,000 or more, live in the central northern
border area). In this documentary, entitled Mandi, a well-known local Garo
leader strongly argues that Garos are “children of the forest”. He contends
that the forest provides them with more than material resources; that, for all
aspects of life, Garos have always been inextricably bound up with the forest.
Without the forest they cannot survive.28
Although it is true that many inhabitants of the Modhupur forest have
suffered from repressive state policies and other developments that caused the
loss of land and the eviction of many peasants,29 it is equally true that the
border area, where most Garos live, is as scarcely forested as most other parts
of the country. There, Garos perform wet-rice cultivation, just like their
Bengali neighbours. One student was quite surprised by the documentary.
She commented that she had never known that Garos could not survive
without forests. Although her own village was not located in a forested area,
her family had been doing pretty well. The image of “children of the forest”
clearly shows how public representations do not necessarily coincide with the
ideas people have of themselves. Here, it even caught a Garo student by
surprise. It is by no means my intention to question the disastrous consequences
of the disappearance of forests and the forced evictions from land. My point
is to show how Garos strategically link up with dominant discourses and, as
a consequence, represent themselves in ways that may have equally less to do
with the experienced realities.30
RESISTANCE IN COMPLIANCE
Although contemporary Garos are very conscious of their ethnic identity, we
can discern a wide variety of Garo notions and representations of Self, both
in the private sphere as well as directed at non-Garo audiences. This chapter
discussed five different representations or narratives of Self, and tried to show
how each of them was clearly related to the tribalist discourse, as discussed in
Chapter 2. In addition, it argued that a closer look at each narrative revealed
how they were informed by the dominant tribalist discourse, but were not
identical to it. Instead, through compliance, resistance, and manipulation,

Garos twist and turn tribalist arguments in order to serve their own agendas.
Whereas the notion of primordial identity stresses an unbridled faith in a
fixed Garo essence, the discourse of lost culture reflects a great sense of
vulnerability but has also stimulated the invention of modern traditions,
through which, particularly educated urban Garos, can give expression to
their feelings of being different.
Also the three public narratives reveal how Garos conform to, but not
fully comply with, the tribalist discourse. The narrative of modernity presents
a picture of a group of people who have outgrown their primitive tribal
image. The narrative of Garo pride and victimhood uses the tribal image (of
being weak and helpless) to seek support in the Garo battle against hardship
and suppression. In a similar manner, representation of Garos as people of
nature involves the argument that Garos are intimately linked with their
natural environment and that they require a special status that allows for a
certain autonomy. Most dominant however, is the tribalist discourse. It
pervades all images and ideas that Garos hold of themselves and present to
others. The question remains whether the strategies and adaptations chosen
will serve the Garos well in the long run. As with any other fashionable theme
in the world of development and international politics, every popular issue or
perspective is prone to lose its popularity one day. In the meantime, Garos
have strengthened the social boundaries between themselves and others on
the basis of their tribal image and their distinct, essentially different, identity
and culture.
NOTES
1. This is more the case with Westerners than with Bengali outsiders. Unlike the
American anthropologist Robbins Burling or myself, for example, the Bengali
anthropologist Kibriaul Khaleque had to make a great effort to be accepted by
the Garos during his fieldwork in Modhupur. Kibriaul Khaleque, “My Fieldwork
Experiences in a Garo Village of Bangladesh”, in Anwarullah Choudhury, ed.,
Pains and Pleasures of Fieldwork (Dhaka: National Institute of Local Government,
1985), pp. 207–22.
2. Cf. Mineke Schipper, who writes how people are likely to stress their group
identity when they feel threatened by outsiders. They tend to keep problems in
the group, as they prefer to present a positive picture of themselves to the outside
world, especially when they feel that outsiders have said enough negative things
about them. Mineke Schipper, De boomstam en de krokodil. Kwesties van ras,
cultuur en wetenschap (Amsterdam: Van Gennep, 1995), pp. 223–24.
3. With private discourse I refer to daily performances and discussions that are not
visible and audible for outsiders. Public performance or discourse are consciously
produced for outsiders, non-Garos, to take notion of. The unit of analysis is the
social group, and not the individual.
4. My findings show some interesting parallels with Harald Eidheim’s classical
study of the Sami in Norway, who during the first months of his fieldwork, tried
their utmost best to show off their Norwegianness. Only after the Sami began to
trust Eidheim did he begin to see different cultural traits that set them apart
from Norwegians. The difference with the Garos however is that Garos felt no
shame of being Garo, and unlike the Sami, they were more than willing to
discuss their ethnic identity. At the same time, Garos also seemed well aware of
the pejorative image that others held of them and in certain situations, they took
great pains to present themselves as a perfect community. Harald Eidheim,
“When Ethnic Identity is a Social Stigma”, in Barth, ed., Ethnic Groups and
Boundaries, pp. 39–57; See also, Aspects of the Lappish Minority Situation (Oslo:
Scandinavian University Press, 1971).
5. Some researchers point out that these so-called non-state arenas for identity
constitution “provide the means by which indigenous populations articulate
their complex positioning vis-à-vis the nation-state, the market and social relations
of modernity”. See for example, Sarah A. Radcliffe, “The Geographies of
Indigenous Self-representation in Ecuador. Hybridity, Gender and Resistance”,
in European Review of Latin American and Caribbean Studies 63 (December
1997): 9–28.
6. Cf. Carole Nagengast and Michael Kearney, who found that the Mixtec population
of Mexico themselves to some degree believe the negative stereotypes that exist
about them. See Carole Nagengast and Michael Kearney, “Mixtec Ethnicity:
Social Identity, Political Consciousness, and Political Activism”, Latin American
Research Review 25, no. 2 (1990): 61–91. See also Eidheim, “When Ethnic
Identity is a Social Stigma”, and ibid., Aspects of the Lappish Minority Situation.
7. These observations are by no means typical for Garos only but clearly correspond
with the findings of many scholars of ethnicity. Take for example, George De
Vos’ who writes that “[s]ome sense of genetically inherited differences, real or
imagined, is part of the ethnic identity of many groups”. George A. De Vos,
“Ethnic Pluralism: Conflict and Accommodation. The Role of Ethnicity in
Social History”, in Romanucci-Ross and De Vos, eds., Ethnic Identity, p. 19.
8. Derived from a personal interview.
9. Abeng or Achik words for race do not exist. To refer to themselves, they generally
resort to expressions such as Garorang, Mandirang or Achikrang. Rang indicates
the plural. Often, Garos resort to the Bengali terms jat or jati. In the Bengali-
English dictionary, jat is described as: 1) one of the hereditary social classes
among the Hindus: a caste; 2) a religious community; 3) a class or kind of
anything; 4) a racial variety; a group possessing common qualities; breed. Jati is
explained as: 1) (biol) a group having common characteristics; a genius of
species; the human race; the species of beasts; 2) a religious community; 3) a
group of people having a common nationality, culture and political affiliation; a

nation; 4) a class of people following hereditary trade or profession: a caste; 5)
one of the main racial divisions of mankind; 6) a division among the Hindu
society according to caste distinction; 7) a kind, sort or variety; 8) birth: origin,
a Jew by birth.
10. “Purity of blood” also plays an important role in Bengalis’ perceptions of Self.
They prevent the “mixing of bad bloods” by a complex system of kinship
regulations and strict rules regarding female conduct and mobility (purdah). For
a discussion on Bengali perceptions of kinship and blood, see for instance, Lina
M. Fruzzetti, The Gift of a Virgin. Second Indian Impression with a New Introduction
on ‘Some Contemporary Issues in Context’ (Delhi, etc.: Oxford University Press,
1990); Lina Fruzzetti and Ákos Östör, “Bad Blood in Bengal: Category and
Affect in the Study of Kinship, Caste, and Marriage”, in Concepts of Person.
Kinship, Caste, and Marriage in India, edited by Ákos Östör, Lina Fruzzetti, and
Steve Barnett (Delhi, etc.: Oxford University Press, 1983); Santi Rozario, Purity
and Communal Boundaries. Women and Social Change in a Bangladeshi Village
(Sydney: Allen & Unwin, 1992).
11. R.W. Bastin described the relations between the different communities in
Mymensingh for the late 1930s and early 1940s as follows: “It is possible to find
adjacent hamlets inhabited by Muslims and Garos; in one the women will veil
themselves and skurry [sic] away at sight of a stranger while the pig is of course
anathema; in the other women will continue about their work and pigs will be
found rooting round the homestead, a striking example both of communal
conservatism and of general amity. In fact, if left to themselves cultivators live in
peace with neighbours of all communities.” R.W. Bastin, Final Report of the
Settlement Operations in Five Thanas of Partially Excluded Area of Mymensingh
1938–42 (Dacca: East Bengal Government Press, 1954), p. 40. I am very
grateful to Ratan Lal Chakraborty, who generously lent me his rare copy of the
report.
12. Literally, purdah means veil. Here, it refers to seclusion of women.
13. Thomas Eriksen refers to two different principles of inclusion and exclusion.
When they allow for differences of degree, he calls them analog. When they
operate on an unambiguous inclusion/exclusion basis, he speaks of them as
digital. In the present-day reality for the Garos, both principles are applied.
Eriksen, Ethnicity and Nationalism, p. 66.
14. Bengali words for culture are sanskriti or kriti, and in Garo language (Abeng) the
word dakari.ka is used to denote culture. Educated Garos generally resort to
these Bengali terms, and sometimes the English word “culture”. Garo villagers,
particularly the older people, talk about dakari.ka.
15. No figures about the number of school- (and college- and university-) going
Garo children are available. My personal experiences show that such figures
differ from village to village. Whether Garos send their children to school
ultimately depends on their income. Many Garos view education as the most
important way for social mobility and generally pride themselves on the fact that
their level of education is higher than that of the average Bengali. This applies
particularly to Garo women compared to Bengali women.
16. A good example was Suborno’s old classmate Riverson, certainly not from one of
the poorest farmers of Bibalgree. Riverson’s fascination for Dhaka made him
leave for the capital, just to discover that city life (and city jobs) were not quite
that great, as he told us after his return to Bibalgree.
17. The Chakmas, who mainly live in the Chittagong Hill Tracts, are the largest,
most well-known ethnic minority of Bangladesh.
18. Mirpur is one of the many fast-growing suburbs of Dhaka, and Bormon is a
caste title of one of the many Hindu minorities that are not considered part of
mainstream Bengali Hindus.
19. Personal correspondence. Letter dated 22 January 1997.
20. Mande-rang-ni Chiti, Easter issue (1996), p. 1.
21. The Nokmandi Survey of Garo household servants showed that almost half of
the 807 interviewees were employed by foreigners (47.5 per cent). Thirteen per
cent of these foreign employers were Asians, and other expatriates are mostly
from Europe, Australia, and the United States. Nokmandi Prakashana, A Census
of Garo Housemaids, p. 40.
22. Mande-rang-ni Chiti, op. cit., p. 2.
23. Ibid., p. 4.
24. Sergius Siram Marak, “The Indigenous People of Bangladesh Cry in the
Wilderness”, Chiring. International Year for the World’s Indigenous People Issue
(2 October 1993).
25. Peter Rema, “The Aborigines of Greater Mymensingh”, Angshidar (Partnership).
1993 International Year for World’s Indigenous People (1993), p. 50.
26. See also Eriksen, Us and Them in Modern Societies, p. 6 (and Chapter 2 of this
book).
27. Guha, Environment & Ethnicity, p. 4.
28. Mandi (produced by Philip Gain, n.d.).
29. See for example, R.W. Timm, The Adivasis of Bangladesh (London: Minority
Rights Group, 1991).
30. For further reading on human rights and environmental issues and the Garos of
Modhupur Forest, see for example: Philip Gain, The Last Forests of Bangladesh
(Dhaka: Society for Environment and Human Development (SEHD), 1998);
Timm, The Adivasis of Bangladesh.
4
PEOPLES WITHOUT HISTORY?
Historical questions and historical evidence are of genuine importance for

characterizing ethnicity – not, perhaps, by way of defining a common
history of some set of people but rather by way of defining the context of
ethnogenesis itself, the context of the virtual invention of an ethnic category.1
Etic and emic ideas of Garo-ness suggest clear-cut categories, well demarcated
by social boundaries, and labelled with unambiguous names. Close scrutiny
of past and present administrative and ethnographic accounts however, shows
that boundaries between the Garos and others have been much fuzzier and
more permeable than those investigations suggest. While the Garos and other
upland people were generally considered similar in their “primitiveness” as
pointed out in Chapter 2, these so-called hill tribes were not regarded as a
homogeneous lot. From the moment that the British began their explorations
of this region, they attempted to impose order on the bewildering diversity
they encountered. So many articles and books about the region’s cultural
variety have since been published that it has become a cliché.2 This chapter
concentrates on these attempts to classify the heterogeneous lot of “primitives”
into clear-cut races, nations, or tribes, with particular focus on the boundaries
between the Garos and others as they were applied by these colonial observers.
Through careful comparison I examine the dynamics and changeability of
social boundaries and hope to show that, in the words of Charles Keyes,
“[t]here is no logical reason why self-identification and assigned identity
67
should always coincide since the two identifications belong to different

cultural sets.”3
FROM FUZZY IDENTITIES TO FIXED COMMUNITIES

With the following fragment, Playfair begins his preface (in 1909) to the first
monograph ever published on the Garos. In his duty as Civil Officer of the
Garo Hills district, he was in daily contact with its inhabitants. His book,
which he simply entitled The Garos, was the first systematic account on the
Garos and probably became the most important source for his contemporaries
and many later writers. It has perhaps become the most famous of all works
on Garos and is still widely used as a reference work. Playfair starts his book
as follows:
In the following pages I have attempted to give a description of the general
characteristics, customs, and language of the Garo tribe. They are a people
who are little known to the outside world, and, though living in the midst
of a civilized province, have remained free from foreign influence in a
remarkable degree. This is due partly to the supposed unhealthiness and
inaccessibility of their hills, and partly to their natural conservatism. In this
connection I have the Hill Garos in mind, for those who inhabit the plains
belong to a different category, and have lost many of their tribal
characteristics.4
The first sentence of his book is the key to this chapter. Playfair writes that his
ethnography is an attempt to describe “the general characteristics, customs,
and language of the Garo tribe”. He refers to the Garos in the singular, as one
tribe, sharing a set of similar traits, customs, and one language, although he
does take some differences between hill and plains Garos into account. These
days, no one contests that contemporary Garos form a separate ethnic group.
Similarly, contemporary publications such as the People of India series often
mirror the strong ethnic or community feelings of the people described.5
Whatever debates arise about the content of ethnologies, such ahistorical
approaches generally consider ethnic or other social boundaries a (timeless)
fact. Nevertheless, if we compare Playfair’s observations with earlier nineteenth-
century colonial accounts, we find that the latter reveal a far more complex
picture than that of a single tribe with a shared set of characteristics, speaking
one language. However, none of those observers questioned their principal
distinction of Garos vis-à-vis others either.
A number of scholars of Northeast India and adjacent areas give account
of the fact that colonial classifications were simplified reflections of the
complex social realities of those days. For example Julian Jacobs and others
Peoples without History? 69
describe how in the early days of their contact with the Nagas (1830s), the
colonial observers struggled with the complexity of the situation they
encountered: “Hundreds, if not thousands, of small villages seemed to be
somewhat similar to each other but also very different, by no means always
sharing the same customs, political system, art or even language.” In their
subsequent attempts to create a classificatory system, the British thus “ ‘created’
the Naga tribes as relatively fixed groups each with a distinctive kind of social
organization and language”.6 The authors deal with issues of similarity and
differentiation, with problems of definition, and with the divergence between
emic and etic perspectives, and arrive at the conclusion that colonial
categorization and indigenous social organizations and self-perceptions did
not correspond.
Sudipta Kaviraj introduced the term “fuzzy communities” to describe
pre-modern conceptions of identity, community, and boundaries. According
to Kaviraj, precise ideas about how to relate to others were clearly defined for
certain aspects of their social lives (for example, on the village or kinship
level), but “such precision did not extend to other aspects of a person’s
identity or community: it was directed to only certain types of activities and
practices.” There was a lack of clarity about where one’s community began
and where it ended. This had to do with the fact that many collective
identities were not territorially based, and that communities were not
enumerated. Kaviraj argues that “[o]n being asked to name his community
(samaj), such a person could take, depending on the context, the name of his
village, neighbourhood, his caste, his religious denomination – but hardly
ever his linguistic group, not to speak of a nation.”7
Kaviraj critically evaluates contemporary perceptions of pre-modern
community identities and boundaries. However, a lack of available
documentation about emic perspectives of subaltern people seriously
complicates the reconstruction of how they perceived themselves and the
world around them. Nevertheless, with Kaviraj’s concept of fuzzy identities,
we can approach matters of identity and community boundaries in a different
way. It helps to explain the confusion of colonial observers and the differences
between their various accounts. It also provides a model for understanding
identities and encourages us to approach identities from another perspective:
people have different identities at their disposal and can, therefore, belong to
more than one social group. Whereas modern states employ all sorts of
methods to encourage overlap of these different social boundaries (linguistic,
cultural, religious) or attempt to subject them to national boundaries, the
concept of fuzzy identities shows that social boundaries and multiple identities
do not necessarily merge into another.
APPROACHING THE GAROS

The hilly border areas of what are now Bangladesh and northeastern India
have for well over a thousand years formed a frontier between South and
Southeast Asian civilizations.8 This frontier area (which by no means hindered
human interaction and cultural contact) is marked by a rather drastic change
in the natural environment. Here, lowlands abruptly give way to hills and
mountains. Of the numerous peoples inhabiting this region, the Garos are
believed to have been the first with whom the British came in touch.9 The
first mention of them dates from the British occupation of Rangpur and
Mymensingh.10 In 1765, the estates in the lowlands bordering the Garo Hills
were passed from the Mughal rulers to the British East India Company.11 The
hills known as Garo Hills, where most of the people called Garos were living,
did not come under direct colonial control for yet another century. The
British did not aim at bringing the area – which had so far never been under
any firm administrative control by a state – under their direct rule. Revenue
collection remained in the hands of the zamindars and their officials who
were henceforth left undisturbed in the management of their estates.12 However,
as time went by, the colonial state slowly extended its control over the whole
of northeastern India, and its need for information about its inhabitants
increased accordingly.
From the very first observations, all people known as Garos by outsiders
were approached as one group or people. Researchers studied their subjects
with the preconceived idea that the inhabitants of the Garo Hills belonged to
one and the same group, Garos, who distinguished themselves unambiguously
from other races, nations, or tribes in their neighbourhood. Regardless of the
confusion and many inconsistencies in their interpretations, these observers
never questioned their basic premises that all people called Garos belonged
together and constituted one distinct category or group of people. Questions
about how people defined themselves, on what basis they determined insiders
and outsiders, and how they maintained boundaries between themselves and
others, were not addressed. This particular conceptualization of the Garos,
seen as a permanent historical entity with clear cultural boundaries, suggested
that variation amongst them was made secondary to differences between
Garos and others.13
The absence of a written Garo language (or languages) turns the
reconstruction of their history into guesswork. These people themselves left
us no written records that provide any insight into how they perceived
themselves and the world around them. For twentieth-century history, we
can use oral history as a source of information, but for earlier history we are
dependent solely on records produced by outsiders. And, as already extensively
discussed in Chapter 2, those various colonial observations should be

understood against the cultural and intellectual background of the observers
and the tribalist discourse in which they participated.
Apart from these intellectual problems, there were other, more practical
complications related to the early colonial accounts about the Garos. Owing
to a number of factors, Europeans were in no position to gather much
information about the Garos until far into the nineteenth century. It was
extremely difficult to enter the interior of the hills.14 Researchers would have
had to face serious complications of rough terrain and the occurrence of
malaria, which was, until far into the nineteenth century, lethal to Europeans.15
Another problem was mentioned by the early-nineteenth century
Commissioner of Cooch Behar and Judge Magistrate of Rangpur, David
Scott. He wrote that “the jealous habits of the Garrows of the interior and
their ferocious treatment of strangers of every description renders it so
dangerous to attempt visiting their villages ….”16 Scott’s contemporaries
either had experienced violent attacks or were at least under the impression
that they would risk violent attacks as soon as they entered the hills.
As a consequence of the alleged inaccessibility of the interior Garo Hills,
information was gathered among people living at the edge of the hills or in
the adjacent lowlands. Contemporaries themselves also recognized that
knowledge about the Garos was limited. For instance, at the turn of the
nineteenth century, more than one century after Eliot’s publication of his
“Observations on the Inhabitants of the Garrow Hills” (1794), Waddell
complained that “[t]he subdivisions of this tribe have not yet been made out
at all satisfactorily.”17
This leaves us with the following question: if the available historical
sources are unreliable, incomplete, and prejudiced, what can they tell us
about emic perceptions of identity, social grouping, and boundary maintenance
of the Garos? Hereafter I shall try to show that, in spite of their shortcomings,
the accounts not only tell us about the Garos in the colonial imagination, but
to some extent also about Garos in their own experience. Men like Eliot,
Scott, Francis Buchanan, John Butler, C.S. Reynolds, Edward Tuite Dalton,
and several missionaries travelled parts of the Garo area as ethnologists avant
la lettre. Careful reading and comparison of their eyewitness reports uncover
a number of interesting details, lacunas, and differences that tell us more
about the people called Garos than one might expect. I will next discuss three
different issues: naming, units of analysis, and boundaries, and show how a
comparison on each of these three themes reveals a far more complex reality
than those individual observers acknowledged, and that etic and emic
perceptions of communities and boundaries did not correspond.
WHAT’S IN A NAME?
In an interview with Nihar and Lipi, a young Garo couple living in Dhaka,
Nihar told us the following: “When we go out they shout at us, ‘Garo, Garo’.
Yes, we are Garo, but the way they call us Garo is very insulting.”
This quotation presents a puzzling paradox: Garos are Garos, but to call
them so can be insulting. There is obviously more than one meaning attached
to the name and more than one way to using it. This comes as no surprise
when we take into account that Garo is a designation used by outsiders and
that it has long been considered a derogatory term. A detailed inquiry into
the meaning, origin, and usage of the name Garos reveals that, even though
both its meaning and origin remain vague and inconclusive, the name was
most probably introduced by outsiders, 18 and that people known as Garos
never used the name amongst themselves. At present, Bangladeshi Garos refer
to themselves as Mandi (which literally means “human being”),19 whereas
Indian Garos generally call themselves as Achik (hill dweller).20 Historical
data show that different groups used different names in the past, and that
there never was one name for all.
The fact that Garos are known by a name other than (one of ) their own
is not a unique phenomenon. It has even been contended that “few if any of
the tribal names appear to be indigenous”.21 Often, those names have primitive
or pejorative meanings, such as “naked” or “bushman”, and, more importantly
for my story, cover a large number of people who do not necessarily consider
themselves part of one group or people.22 Robbins Burling writes that Garos
of Modhupur forest no longer like to be called Garos. “This is the outsider’s
word, not their own, and since their experience with outsiders has sometimes
been unhappy, the outsider’s word has come to be resented.”23 During my
own research, I never noticed people feeling troubled by being called Garo,
unless, as we saw in the example, it was used in a certain, offensive, manner.
People recognize that they are known as Garos by others and often resort to
the name themselves, when talking to outsiders.24
As stated before, the origin and meaning of Garo are unknown. Garos
themselves frequently contend that they were given the name by outsiders but
no one knows its exact meaning, why and when it was given to them, and by
whom. Eliot, the first European who travelled among the Garos, was deputed
by the government of Bengal in 1788, “to investigate the duties collected on
theGarrow Hills, which bound the north-eastern parts of Bengal”.25 During
1788–89, he visited Garos of the southern side of the hills, the area which
seems to coincide pretty much with the region where I conducted my
research. In his article, Eliot refers to the people he studied as Garrows. He
wrote that “[t]he Garrows are called, by the villagers, and upper hill people,
Counch Garrows; though they themselves, if you ask them of what cast they
are, will answer, Garrows, and not give themselves any appellation of cast;
though there are many casts of Garrows, but of what differences I had no time
to ascertain.”26 Eliot found that Garrow was what the people called themselves,
or at least, presented themselves to outsiders. It is interesting to find that these
observations were much less complicated than those of many later researchers.
Eliot himself provides the explanation. He based his account on his experiences
with a small “Garo segment” only and had no time to explore “of what
differences” the many “casts of Garrows” were.
During the years 1807–14, Buchanan carried out an extensive survey in
the provinces of Bengal and Rangpur. He also visited the northern parts of
what he called “Garo country” and collected extensive information on the
inhabitants of the area bordering the northern part of the Garo Hills.
Buchanan’s accounts are particularly valuable. With his effort “to understand
ethnic identities from ‘within’, from the viewpoint of the people concerned”,
he distinguished himself from most of his contemporaries.27 Buchanan for
example mentioned that he made use of six Garo informants who explained
to him that the name Garo came from the Bengalis. He wrote that “[m]y
informants say, that Garo is a Bengalese word, nor do they seem to have any
general word to express their nation, each of the tribes, into which it is
divided, has a name peculiar to itself.” He also acknowledged that “what they
describe can be only considered as strictly applicable to that division of the
nation.” After all, all six of them were from the tribe that borders on
“Haworaghat” (called the Achhik or Achhikrong).28 His observations that the
Garo “nation” was divided into several divisions, or tribes, each with its own
name, were nevertheless no reason to forsake the name Garo altogether, let
alone to question the existence of a distinct Garo nation.29
Dalton, who partly based his account on Buchanan’s report, also toured
among the Garos himself (in 1848). Unfortunately, his account provides no
detailed information on where he was, on the length of his visit, or about his
methods of field research. Dalton, too, found that Garo was a name given by
outsiders. According to him, it was a term applied by the Hindus. About the
Garos he writes that “[t]hey consider themselves as forming three or four
nationalities with different names. Dalton mentioned the Nunyas, Lynteas,
and Abengyas. Nevertheless, he too continued to refer to them as Garos.”30
Half a century later, Playfair also found that the name Garo was never
used by Garos themselves except in conversations with outsiders. Playfair
wrote that the people would always call themselves Achik, Mandé, or Achik
mandé. He thought Achik to be the correct name and believed Garo to be a

mere corruption of the name of one of the sub-divisions of the tribe.31
The differences between the observations are striking, particularly between
the accounts of Buchanan and Playfair. Buchanan’s observations differ to a
much greater extent from the situation we encounter today than Playfair’s
observations, which are still applicable to contemporary Garos. Interestingly,
the inconclusive meaning and origin of the name never led to a re-evaluation
of its use. Waddell was the one exception to the rule. Unlike his contemporaries,
he referred to the Garos as Mandé. After all, he argued, this was what “this
large tribe” called itself.32
Contemporary Garos still lack a common name of their own to include
both Indian and Bangladeshi Garos. At the same time, however, they do
consider themselves a distinct ethnic or imagined community. It seems that,
in the course of two centuries, Garos have, to a certain extent, grown into the
label once introduced by outsiders.
GARO TRIBE OR TRIBES?

In the minds of nineteenth-century researchers, Garos constituted one distinct
category of people, whether they referred to them as a race, nation, or tribe.
No matter what divisions or variations they encountered, all so-called Garos
belonged together. At the same time, however, differences and divisions were
also observed and documented. In the previous section, early observers like
Buchanan and Dalton had recorded that Garos were divided into different
tribes, each with their own name. But there were inconsistencies in their
accounts. Buchanan, for example, wrote:
An individual of the tribe adjoining to Hawaraghat is called Achhik; but the
collective name or plural number is Achhikrong: the high hills of Mechpara
are occupied by the Abeng, with whom I could procure no interview, the
Zemindar having probably alarmed them. The Abeng may perhaps be
considered as subjects of the Company, as their hills are entirely surrounded
by the lands of the Mechpara Chaudhuri, and are not included in the
territory, which I have specified as belonging to the Garo nation; but I
believe they have always declined subjecting themselves to the decisions of
the courts in Bengal. The tribe bordering on Mechpara and Kalumalupara,
that occupies the high mountains, and retains an entire independence, is
the Kochunasindiya. This people also declined an interview, probably from
similar reasons. The tribe bordering on Susangga is called Kocha or Counch,
as Mr Eliot writes. From the account of that gentleman, these seem to
occupy only the lowlands, and to be tributary, and their territory is not
included in what I have considered as belonging to that nation as

independent. The tribe of the Garo nation, that borders on Asam is called
Nuniya. … The Nuniyas are also called Dugol.33
The quotation contains a number of interesting details. Firstly, Buchanan

distinguished five Garo tribes. These tribes not only inhabited different
geographical areas, their political status also differed (they were independent,
dependent, or tributary). Buchanan points out that he himself only got first-
hand information on one tribe: the Achhik. Similarly, Nagas, Rabhas, and
others many others, were also perceived as collections of several tribes.34
Nineteenth- and early twentieth-century researchers were obviously in doubt
and even confused about concepts such as nation, tribe, and caste (see
Chapter 2).
A closer examination of Buchanan’s use of “tribe” reveals a wide application
of the term. Sometimes Buchanan distinguishes between nation and tribe,35
but in other places he does not.36 Whereas he refers to the Kachharis, Mech,
and Rabhas as one tribe – the latter seems “to have been divided into different
branches”37 – Garos are described as a “rude nation” or people, divided into
different tribes.38 Buchanan realized that he was dealing with five distinct
groups among the Garos. He reported that each of those five tribes used a
different name for themselves, spoke a different language, differed in status,
and occupied their own territory. Although four of the five tribes spoke nearly
the same languages, the language of the Nuniyas was different. Buchanan also
established a hierarchy among the tribes. The Nuniyas were of highest rank.
“Their priests can officiate for all Garos; but the priest of any of the other
tribes cannot officiate for a Nuniya.”39
Interestingly, Buchanan was not the only author who described the Garos
as a collection of several tribes. Observers like John M’Cosh, William
Robbinson, F. Jenkins, Dalton, and Scott also used Garo tribes instead of the
singular word tribe.40 The latter wrote that the Garos (he writes Garrows)
are said to be divided into various Tribes of which the principal appear to be
the following. The Nooneas who live in the Assam frontier, and are governed
by Rajahs. The Hannahs who inhabit the country south of Habraghaut.
The Abings bordering on Mechparrah Kaloomalooparah and Kurreebarree
and the Koch who are said to inhabit the country east of the latter
pergunnah.41
He attributed this division to the natural environment of the country, which

logically separated the inhabitants of the different valleys. Moreover, each of
the tribes were
divided into different clans, each supposed to be descended from a common

ancestor and these again are subdivided into branches of the family, still
more closely connected together, who usually inhabit the same parah or
hamlet, some of which are also occupied by the descendants of strangers
who at various times have settled with the consent of the clan upon their
lands.42
Observations of a third contemporary, Dalton, largely correspond with
those of the mentioned writers. Dalton primarily distinguished between the
“uncivilized, unconverted, Garo” and “the Hinduized animal of the name”.
He mentioned that “[s]everal of the petty Rajas of Kamrup, whose estates
skirt the Kasia and Garo Hills, are Hinduized Garos, who have maintained
their footing in the valley during several changes of dynasty.” Dalton
himself wished to describe the first and undertook a tour among them in
1846. He acknowledged that “a great portion of the interior is quite
unexplored.” He related that the Garos consider themselves as forming
three or four nationalities with different names: “Of those subject to the
Gowalpara jurisdiction, or having communications with it, the most eastern
bordering on the Kasias, are called the Nunya; the central tribe are the
Lyntea, and the remainder are the Abengya.”43 About the differences between
these nationalities, he writes that the Nunya were the fairest of the tribes,
and that “the language of the Western Garos is unintelligible to the Nunyas.”
These Nunyas resemble the eastern neighbours of the Garos, the Khasis, in
feature, complexion, and language.44
What these three observers had in common was their reference to several
Garo tribes, and that they attributed these divisions to their geographical
dispersion. Nevertheless, the differences between these tribes were clearly not
enough to consider them distinct groupings, in the way they considered
Khasis and Garos distinct. In the perspective of the early colonial observers,
Garo served as an umbrella term for a collection of sub-divisions or tribes
who formed one distinct nation, inhabiting their own country.
FROM GARO TRIBES TO GARO TRIBE

As the end of the nineteenth century approached, the expression “Garo
tribes” gave way to “Garo tribe”. At the same time, however, at the turn of the
nineteenth century, instead of four or five Garo tribes, no less than twelve
geographical Garo sub-divisions were mentioned. For example, in his chapter
on “tribal organization”, Playfair divides the Garos into “those who inhabit
the Garo Hills district, and those who reside in the plains and are scattered
over a very wide area of country”.45 Like most others, he focused primarily on
the inhabitants of the hills, among whom he distinguished twelve divisions.46

He added that most of these divisions could be found in the plains as well.47
Some divisions were clearly distinct from others. About the Chisak, for
example, Playfair wrote that they “have much in common with the Awès, but
they have some distinctive features in dress and customs which proclaim
them a separate division of the tribe”. Other divisions show remarkable
similarities. The Matabengs, for instance, “claim that they are a distinct
division, but their language and geographical distribution make it more than
likely that in them is to be found a mingling of Abeng and Machi”.48 Playfair’s
contribution underlines how difficult it is to establish clear-cut sub-categories
of Garos and to define what distinguished one from the other. In the end,
however, all divisions formed one Garo tribe.
BOUNDARY ISSUES
If different Garo tribes or divisions constituted one social category, were
they in some way organized? Colonial accounts present no unequivocal
answer to the question of how these tribes or (sub)divisions related to each
other. Different observers held different, sometimes conflicting, opinions.
Scott, for example, wrote that they were “rich and individually free to such
an extent that they excitingly compare their country to a crab which they
affirm has no head”.49
This idea that Garos lacked any central leadership was also suggested by
M’Cosh and W.W. Hunter. M’Cosh pointed out that “like the Khassyas
they too are divided into numerous petty tribes, each chief of which has his
vote in the assembled council, though no one of them is independent of the
others.”50 And when Hunter related in 1879 that “[t]hey are subdivided
into many petty tribes or clans, each residing in its own village or villages
amid the hills,” he added that “[e]ach chief of the numerous petty clans has
his vote in the assembled council of the whole tribe, but no one is
independent of the others.”51
A rather different opinion of the Garo political system was held by
Reynolds. In his view, the relation between the different Garo divisions and
their chiefs was primarily voluntarily and clearly context related. Each chief
was entirely independent from the others. This certainly did not mean that
each chief lived in complete isolation. In cases of serious crime, for instance,
it was customary for a chief in whose jurisdiction the crime occurred to
invite neighbouring chiefs for advice. Moreover, “[i]n the event of an
inroad from the plains I believe all join under the most substantial
neighbouring chief to resist it.”52
Garo chiefs not only joined each other to resist incursions from lowlanders,
they attacked each other as well. For example, in 1847, Butler, a British
officer, wrote that
Passing through Lookee Dooar to the Jeypore stockade, at the foot of the
Garrow hills, I met with many Garrows who reside on the low hills
bordering Assam, and learned that they were frequently in great danger
from the highland Garrows; who, feeling secure in their mountain fortresses,
made occasional incursions into the territory of the former, and committed
acts of violence upon the British subjects located in the plains.53
Some three decades later, Hunter referred to an incident in which Garos in
British-held territory felt considerably threatened by “independent” Garos
from the hills:
Soon afterwards an attack was made on the protected village of Dámákchigirí,
in the southern hills, by the independent villages of Kákwágirí, Báwigirí,
and others, and several villagers were murdered. The reason of this attack
was supposed to be sympathy with the villagers of Pharámgirí, and
exasperation against some of the headmen of friendly villages near the
plains, who had rendered aid to Captain La Touche during his expedition
in the cold weather of 1871–1872. Shortly after the Dámákchigirí murders
were committed, an attack was also made on the Pharámgirí stockade by a
band of independent Gáros. Considerable alarm was felt by all the dependent
Gáros, and this feeling was communicated to the villagers in the plains.54
Both examples are about Garos attacking Garos. Hunter’s account presents a
story in which independent Garos were assaulting “friendly”, dependent,
Garo villages. The example suggests a relation between these attacks and the
political status of the villages concerned. Relations in the area were evidently
affected by the political situation and the role of the colonial state.
According to Playfair, the first few years of British administration provided
ample proof that Garos were addicted to internal warfare. Many blood feuds
existed between individuals as well as villages. And “[w]hen opportunity
offered for a successful raid on an enemy’s village, it was quickly taken
advantage of and the heads of the victims were borne home in triumph as
coveted trophies.” For that very reason villages used to be much bigger, were
all protected by a wall of pointed bamboo stakes, and their main entrances
were carefully guarded.55
The Indian historian Jayanta Bhusan Bhattacharjee argues that it was the
Garo custom of collecting human skulls for religious purposes and social
distinction, which led to “interclannish feuds” and “internecine warfare”.
Bhattacharjee contends that “the tribes lived in a constant state of warfare.”
Members of the “slain’s clan” were bound to demand blood for blood and had
to kill either the murderer or one of his kind or slaves. War would then
continue until one of the rival clans was defeated or the elders from both clans
succeeded in bringing about reconciliation.56
My own informants, too, knew stories about fights among different Garo
groups. Some believed that these fights were at the very roots of the settlement
of Garos in the plains. On the basis of the aforementioned examples and
stories I collected, it seems safe to conclude that the so-called Garo country
was frequently unsettled by warfare during the nineteenth century. The
question remains whether these conflicts were caused by cultural incentives or
were politically informed. Historical data does not provide us unequivocal
answers to these questions. It is, however, clear that Garos had no central
organization. They were, and still are not, united under the leadership of one
king or other representative, or represented by one spokesman, whether in
India or in Bangladesh.
GAROS AND THEIR NEIGHBOURS

In the absence of central leadership or a central organization, on the basis
of which criteria did the colonial observers group them together and tell
them apart from their neighbours? Today we are used to the categorization
in population censuses, which suggests clear-cut boundaries between the
different groupings. The numerous essay collections of tribes reflect
contemporary ideas that each of these tribes constitutes a distinct ethnic
grouping or people.57 Again, historical sources reveal a great deal of confusion
and uncertainty among colonial observers about who belonged where and
who fit in which category. The following cases illustrate these problems of
inclusion and exclusion.
The northeastern part of the Indian sub-continent (including what is
now Bangladesh) has a great number of ethnic groups or peoples. The Khasis
are one of the many neighbours of the Garos. At present, Garos and Khasis
are recognized as two clearly distinct peoples. Early nineteenth-century
documents and the subsequent accounts reveal that they presented many a
researcher with problems of categorization.58 Buchanan seemed quite confident
about the boundaries between Garos and others, but possibly also included
people in the Garo category who, at least today, do not consider themselves
Garos. His contemporaries also pointed out that he failed to identify some
ethnic boundaries, for example, between Garos, Jaintias, and Khasis.59 In the
1880s, Horatio Bickerstaff Rowney argued that “[t]he older writers speak of
the Garos and the Cossyahs as one people; but this they are not, though there
is certainly much similarity between them.”60 Particularly one distinctive

aspect stands out for Garos and Khasis: language. It has long since been
pointed out by linguists and other researchers that the Khasi and Garo
languages are totally unrelated,61 and language seems to offer a rather hard
proof of distinctiveness. Nevertheless, in this respect the tribe or division
called Nuniyas presents us with a problem. Dalton clearly distinguished
between Nun(i)yas – whom he considered a Garo tribe – and Khasis, but
wrote about Nuniyas that they were “the fairest of these [Garo] tribes resembling
the Kasias in feature and complexion and in language”. To this he added that
“[t]he language of the Western Garos is unintelligible to the Nunyas.”62 This
example shows that language did not act as the one decisive criterion after all.
It is interesting to note that Nunyas were not included in later narratives.
According to Playfair, “Nuniya is the name applied to the Megams by the
Assamese of Boko in the Kamrup district.”63 Yet whether Nunya or Megam,
the issue of language remains the same. Megam have been equally difficult to
categorize. Even today there is some confusion about whether Megam are
really Khasi or whether they are Garo (see below).
Similar problems also arose with other neighbours of the Garos, such as
the Rabhas, Hajongs, Banais, and Koches. In a chapter about affinities,
Playfair mentioned two other sections of the Bodo group: Rabhas and Koches.
He noted differences of opinion among researchers about the names and
origins of these two groups. Playfair himself also struggled with the
distinguishing criteria. In the following fragment, he tries to establish the
degree of affinity on the basis of language and geographical proximity. This
turns out to be quite complicated:
From the fact that the Rabhas live in close proximity to the Akawés [Garo
subdivision], it would be natural to suppose that the Rabha language
resembled the standard Garo dialect. In like manner it would not be
surprising if the Tantekiya Koches spoke a language which was nearly
related to that of their immediate neighbours the Abengs [Garo subdivision].
This, however, is not the case, and it will generally be found that while both
the Rabha and the Koch languages differ very materially from the Awé and
Abeng dialects, they bear a quite remarkable resemblance to the dialects
spoken by the Atongs and the Rugas, two divisions of the Garos which
resemble each other very closely, but differ considerably from the remaining
divisions.64
In this fragment, Playfair refers to sub-divisions of Garos which are
linguistically similar to different tribes such as the Rabhas, and to other
Garo sub-divisions whose languages differ considerably with the language
spoken by these same Rabhas.
People called Megam present another exemplary case of how difficult it

really was to distinguish Garos from their neighbours and to decide which
linguistic or cultural groupings were just sub-divisions of Garos and which
were really members of a distinct tribe. Although Megams are considered
Garos these days, such was not always the case. This can be clearly illustrated
by Playfair’s observations. Apart from the twelve geographical divisions which
Playfair distinguished, he also mentioned the Megams or Lynngams to whom
he refers to as “another tribe”. Megams represented a “fusion of the Garo
and the Khasi” – a “hybrid race” – because “[i]n appearance and customs
they closely resemble the Garos, but their language has been classified by
Dr Grierson as Khasi, and is absolutely unintelligible to the ordinary Garo.”65
In 1963, Burling wrote that, although Megams share a number of
cultural characteristics with the Garos, they speak a dialect of Khasi, and
Garos generally consider them Khasis.66 Three decades later, however, Burling
reported that:
Finally, the Megam, who live in the westernmost part of the Khasi hills, just
to the east of the Garos, are also sometimes considered to be Garos. The
Megam speak a dialect of Khasi. This is a Mon-Khmer language totally
unrelated to Garo, and the few Megam whom I have met spoke much
poorer Garo than did most of Atong or even the Koch whom I knew. On
linguistic grounds the Megam must be classified with the Khasis rather than
with the Garos, but there is considerable intermarriage along the border,
and the Megam share so many features of Garo kinship and social
organization that they are sometimes described as another subgroup of
Garos.67
Milton Sangma and Julius Marak describe Megams as Garos, and during my
own research, I also found that Megams of Bangladesh consider themselves,
and are considered by others, as Garos and not Khasis.68 Even language,
which is often mentioned as an essential ingredient of social identity, offers
no unequivocal criterion for classification, or for group formation.69
CONCLUSION
Comparison between different nineteenth-century colonial accounts about
Garos on the issues of naming or labelling, units of analysis, and boundaries
reveal that colonial attempts to differentiate and systematize local people led
to a fixation of social boundaries which were much fuzzier in real life.
Colonial categorization brought only apparent clarity. A comparison of different
colonial accounts reveals that colonial categories did not correspond with
local identities and that etic perceptions of social boundaries differed from
emic perceptions of identity. We cannot, therefore, reconstruct emic perceptions

of identity on the basis of colonial documents alone.
I assume that late eighteenth- and nineteenth-century Garos were a
collection of groups rather than one distinct ethnic community. This would
also explain why colonial observers did not find it all that easy to decide who
were really Garos and who were not. Particularly the early observers found it
no easy task to understand how Garos were organized and managed. These
difficulties cannot only be attributed to a lack of information. After all, men
like Eliot, Buchanan, and Scott took their tasks seriously and spent a great
deal of time in the field. Their successors, who had much easier access to the
hills, did not necessarily study or understand Garos better. Nevertheless, the
most convincing arguments which support my assumption that colonial
observations of social boundaries did not correspond with local perceptions
are provided by Garos themselves. In the next chapter we will see that
personal accounts of Garo villagers confirm the picture that, until recently,
Garos did not form the same ethnically conscious people that they are today.
NOTES
1. F.K. Lehman, “Who are the Karen, and if so, Why? Karen Ethnohistory and a
Formal Theory of Ethnicity”, in Ethnic Adaptation and Identity. The Karen on the
Thai Frontier with Burma, edited by Charles F. Keyes (Philadelphia: Institute for
the Study of Human Issues, 1979), p. 216.
2. Compare this with observations on the Southeast Asian highlands. Peter Hinton,
for example, also noted that the “cultural variety of mainland South-East Asian
highlanders becomes a cliché amongst writers. … Such writers (who include
many anthropologists) then try to introduce order into this apparent chaos, by
dividing the highlanders into a number of ‘tribal’ or ‘ethnic’ groups, the members
of which share a distinctive language and a set of cultural traits.” Peter Hinton,
“Do the Karen Really Exist?”, in Highlanders of Thailand, edited by John
McKinnon and Wanat Bhruksasri (Singapore, Oxford, New York: Oxford
3. Charles F. Keyes, “Introduction”, in Keyes, ed., Ethnic Adaptation and Identity,
p. 10.
4. Playfair, The Garos, p. xxiii.
5. This is a series which has been published since 1985 on behalf of the
Anthropological Survey of India (ASI) with the objective of updating information
about a large number of Indian communities. K.S. Singh (general editor), People
of India (Calcutta: Seagull Books).
6. Jacobs, Julian, et al., The Nagas. Society, Culture and the Colonial Encounter
(London: Thames and Hudson, 1990), p. 17.
7. Sudipta Kaviraj, “The Imaginary Institution of India”, in Subaltern Studies VII,
edited by Partha Chatterjee and Gyan Pandey (Delhi, etc.: Oxford University
Press, 1993), pp. 25–26.
8. Cf. Burling, The Strong Women of Modhupur, p. 5; Richard M. Eaton, The Rise
of the Islam and the Bengal Frontier (Berkeley, etc.: University of California Press,
1993); Van Schendel, “Invention of the ‘Jummas’ ”, pp. 106–07.
9. The Garos are one of many different peoples or communities residing in these
hilly areas and adjoining plains in Assam, West Bengal, and Bangladesh. Others
(also referred to as tribes) include the Khasis, Jaintias, Rabhas, Banais, Koches,
Lalungs, Hajongs, Dalus, Boros, and Mikirs.
10. Alexander Mackenzie, The North-East Frontier of India (Delhi: Mittal Publications,
1989 (first published in 1884, entitled, History of the Relations of Government
with the Hill Tribes of the North-east Frontier of Bengal ), p. 245.
11. This area then contained the districts of Dacca, Mymensingh, and Rangpur. In
those days, Goalpara formed part of the Rangpur district. In 1822, it came
under the commissioner of northeast Rangpur and was eventually placed under
the jurisdiction of the commissioner of Assam. See Jayanta Bhusan Bhattacharjee,
The Garos and the English 1765–1874 (New Delhi: Radiant Publishers, 1978),
p. 1.
12. Bhattacharjee, Garos and the English, p. 23.
13 For very similar observations see for instance Conrad, “Lisu Identity in Northern
Thailand”, p. 193.
14. The lack of information about the interior Garo Hills was felt by the British
themselves. For example, in 1866 Colonel Hopkinson, agent to the northeast
frontier and commissioner of Assam, explains in a letter to the secretary of the
government of Bengal that because of the scarcity of information about the
“country occupied by Garrows who are called independent, lieutenant Williamson
has been unable to exact the position of their villages”. See letter from Hopkinson
to the secretary to the government of Bengal, in Judicial A Proceedings. Judicial
Branch, July 1866, pp. 29–30.
15. John M’Cosh states in the 1830s that “Above all jungly countries in India, that
of the Garrows is, perhaps, the most fatal for a European to visit.” See John
M’Cosh, Topography of Assam (Calcutta: G.H. Huttman, Bengal Military Orphan
Press, 1837), p. 165. About a decade later, John Butler, a British officer, explained
how the climate posed severe obstacles for Europeans to enter the interior of the
hills, because “no European constitution could endure a lengthened residence
among them.” See John Butler, A Sketch of Assam. With Some Account of the Hill
Tribes (London: Smith, Elder and Co., 1847), p. 179. In a letter dated 4 June
1853, to the Governor General of the Northeast Frontier A.J. Moffatt Mills,
Lieut. Col. F. Jenkins writes that: “Divided as these tribes are by a climate most
fatal to foreigners, we cannot exercise over them that coercion which otherwise
might as readily do as in the neighbouring Cossiah Hills.” See A.J. Moffatt
Mills, Report on the Province of Assam (Calcutta: Thos.Jones, Calcutta Gazette
Office, 1854).
16. See extracts from Scott’s report on the Garos, dated 20 August 1816, in Nirode
K. Barooah, David Scott in North-East India 1802–1831. A Study in British
Paternalism, Appendix D. (New Delhi: Mushiram Manoharlal, 1970), p. 247.
18. Julius Marak is one of the few researchers who attempts to explain the origin of
the name Garo. He summarizes seven possible explanations but subsequently
refutes all but one, and believes that Garo may be a corrupt form of the word
Garu or Garudas. The idea is that when the Garos migrated from Tibet they
were known as Garu Mandai, later came to be known as Garudas, and finally as
Garrow or Garo. Marak does suggest further research. Julius Marak, Garo
Customary Laws and Practices (A Sociological Study) (Calcutta: Firma KLM
Private limited, 1986), pp. 4–9.
19. There are many more examples of people using names for themselves which
simply mean human being, man, or hill man, etc. One example are the Lushais,
who, according to S. Barkataki, call themselves Mizos, which literally means
“hill man” (mi=man, zo=hill). See S. Barkataki, Tribes of Assam (New Delhi:
National Book Trust, 1969), p. 16.
20. To call oneself Achik would obviously not make much sense for people who live
in a country which is as flat as can be. However, if we take the widespread story
into account that the plains Garos came from the hills originally, then why
would they not refer to themselves as Achik? Was it too long ago? Does the story
of the migration from the hills into the plains not hold, or did they never call
themselves Achik in the first place?
21. Jacobs, et al., The Nagas, p. 20.
22. Dalton points out that Buchanan refers to the Lynteas as Achhik. According to
Dalton, Lyntea is derived from “langta”, which means “naked” and is a name
given by the Bengalis. See Edward Tuite Dalton, Descriptive Ethnology of Bengal
(Calcutta: Indian Studies Past & Present, 1973, originally published in 1872),
p. 73. There is a striking parallel with the people who were referred to as “Kuki”
– which is, according to Van Schendel, “a vague term for hill people who are
neither Chakma nor Marma”) – because other names for Kuki are Lan-ga, Lang-
ga, Lan-ka, Layn-ga, Lingta, and Linkta. See Van Schendel, ed., Francis Buchanan
in Southeast Bengal (1798), p. 201.
23 Burling, Strong Women of Modhupur, p. 3.
24. Note also Shubash Jengcham’s publication on the Garos of Bangladesh. Jengcham,
himself a Bangladeshi Garo, titled his book Bangladesher Garo Sampradai [The
Garo Community of Bangladesh] (Dhaka: Bangla Academy, 1994).
25. Eliot, “Observations on the Inhabitants of the Garrow Hills”, p. 17.
26. Ibid., p. 19.
27. Cf. Van Schendel, “The Invention of the ‘Jummas’ ”, p. 100. For a more
elaborate description of Buchanan and his works, see Van Schendel,
“Introduction”, in Van Schendel, ed., Francis Buchanan in Southeast Bengal
(1798), pp. ix–xxv.
28. Martin, ed., Eastern India, p. 689.

29. Buchanan’s informants were the chief of Raumari, the chief of Ramjongga (or
Amjongga) and his predecessor the chief of Damra, the chief Digman, and a
priest from the hills near Jira. Martin, ed., Eastern India, pp. 688–89.
30. Dalton, Descriptive Ethnology of Bengal, p. 65.
31. Playfair offers two possible explanations. He suggests that the name Garo is
derived from one of its sub-tribes, the Gara or Ganching, who were the first to
be approached by the British and Bengalis. Another possibility is that it originates
from a leader named Garu who is believed to have led the Garos from Tibet
several centuries ago. Playfair himself prefers the first explanation. See Playfair,
The Garos, pp. 7–8.
33. Italics mine. Martin, ed., Eastern India, pp. 689–90.
34. Cf. Jacobs, et al., Hill Peoples of Northeast India; Karlsson, Contested Belonging.
35. “I find, that the appellation of Mug is given by the people of this province to all
the Tribes, and nations, east from Bengal, who as differing from Hindoos, and
Mussulmans, are considered as having no Cast, and as therefore being highly
contemptible”. Van Schendel, ed., Francis Buchanan in Southeast Bengal, p. 28.
36. “A man of the tribe by the Bengalese called Joomea was brought to me. He says,
that many little Villages of his Nation … are scattered among the Hills east from
this …”. Van Schendel, Francis Buchanan in Southeast Bengal, p. 33.
38. Ibid., pp. 682–96.
39. Ibid., p. 690.
40. Scott as cited by Nirode K. Barooah, David Scott in North-East India 1802–
1831. A Study in British Paternalism, Appendix D (New Delhi: Mushiram
Manoharlal, 1970); M’Cosh, Topography of Assam; Dalton, Descriptive Ethnology
of Bengal (Calcutta: Indian Studies Past & Present, 1973, originally published in
1872); Moffatt Mills, Report on the Province of Assam; William Robinson, A
Descriptive Account of Assam: With a Sketch of the Local Geography, and a Concise
History of the Tea Plant of Assam: To Which is Added, A Short Account of the
Neighbouring Tribes, Exhibiting their History, Manners, and Customs (New Delhi:
Sanskaran Prakashak, 1975, original publication in 1841).
41. Extracts from Scott’s report in Barooah, David Scott in North-East India, p. 247.
In 1815, the attention of David Scott, the commissioner of Cooch Behar and
judge-magistrate of Rangpur, was drawn to the Garo frontier. In that year, Garos
carried out several attacks on inhabitants of two big land estates adjoining the
Garo Hills. Scott’s determination to put a stop to these raids led him to an
investigation into the relations between Garos and the zamindars [landlords]
along the foothills. In 1816, during a tour that lasted several months, he visited
the frontier areas of the hills. In August of that same year, he produced a report
about the relations between the Garos and zamindars. See Barooah, David Scott
in North-East India, pp. 39–41.
42. Extracts from Scott’s report in Barooah, David Scott in North-East India, p. 248.
43. Dalton remarks that the name Lyntea is derived from “langta” (naked), and was
given by Bengalis. Dalton’s Lyntea are the same people Buchanan called the
Achhiks. Dalton, Descriptive Ethnology of Bengal, pp. 65 and 73.
44. Ibid., pp. 64–75.
46. Playfair distinguishes the following geographical divisions: Akawés or Awés,
Chisaks, Duals, Machis, Matjanchis or Matabengs, Kochu, Atiagras, Abengs,
Chiboks, Dalu, Ganchings or Garas, and the Atongs. Ibid., pp. 59–62.
47. Other divisions not known in the hills are the Braks, Jariadongs, Somons,
Galnés, and Malongs. Ibid., p. 63.
48. Ibid., p. 60.
49. Scott as cited by Barooah, David Scott in North-East India, pp. 253–54.
50. See M’Cosh, Topography of Assam, p. 164. A decade later William Robinson
writes in almost similar words that “The Garos, like their neighbours the
Khassias, are divided into numerous petty tribes, each chief of which has his vote
in the assembled council, though no one of them is independent of others.”
Robbinson, Descriptive Account of Assam, p. 415.
51. Hunter, A Statistical Account of Assam. Volume II, p. 34.
52. Reynolds, C.S., “A Narrative of our Connexions with the Dusannee and Cheannee
Garrows, with a Short Account of Their Country”, Journal of the Asiatic Society
of Bengal 18, Part 1 (January–June 1849), p. 57.
53. John Butler, A Sketch of Assam. With Some Account of the Hill Tribes (London:
Smith, Elder and Co., 1847), p. 179.
54. Hunter based this story on an account of the Garo expedition of 1872–73 by the
deputy-commissioner of the district. See W.W. Hunter, A Statistical Account of
Assam. Vol. II (London: Trübner & Co., 1879), p. 158.
56. Bhattacharjee, The Garos and the English, pp. 6–7.
57. One of the many examples is the People of India Series, edited by K.S. Singh.
58. Burling and Karlsson also refer to the earlier confusion about the classification
of Garos and their neighbours. Burling wrote that “a number of small groups live
along the edges of the Garo Hills who are not now considered Garos, but who
share enough with the Garos that they may have been grouped with them by
early British visitors.” Karlsson, who studied the Rabhas of North Bengal,
described how Buchanan found that the Pani Koch were misnamed Garos by the
Bengalis, and how one forest officer first called his favourite labourers Garrows,
but in a 1931 article, suddenly renamed them Rabhas. Burling, Strong Women of
Modhupur, p. 14; Karlsson, Contested Belonging, p. 56.
59. See for example, Dalton, Descriptive Ethnology, p. 55.
60. Horatio Bickerstaff Rowney, The Wild Tribes of India (Delhi: Low Price
Publications, 1990, first published in 1882), p. 191. For a similar remark, see
R.W. Bastin, who wrote the following: “… but Buchanan Hamilton who
encountered the Garos in the present Assam valley uses the word Garrow to
apply indiscriminately to Khasis and Garos.” Bastin, Five Thanas of Partially
Exluded Area, p. 45.
61. Khasis speak a Mon-Khmer language, whereas Garos speak languages of the
Tibeto-Burman branch of Sino-Tibetan. See Burling, The Strong Women of
Modhupur, pp. 5–6 and 15.
62. Dalton, Descriptive Ethnology, p. 65.
64. Ibid., pp. 18–20.
65. Ibid., p. 62.
66. Robbins Burling, Rengsanggri. Family and Kinship in a Garo Village (Philadelphia:
University of Pennsylvania Press, 1963), pp. 358–59.
67. Burling, Strong Women of Modhupur, pp. 15–16.
68. Julius Marak distinguishes twelve sub-tribes among the Garos of the Garo Hills.
His divisions are very similar to Playfair’s. Unlike Playfair, however, he explicitly
mentions that there are not many cultural differences between the sub-tribes
(except in case of the Megams), and that the basis of each group is dialect. Marak
does not mention the Atiagra. Another difference with Playfair is that he
includes the Megams in his list of Garo sub-divisions. Milton Sangma suggests
that the Garos are divided into eleven groups (the A’kawes or A’wes, the Chisaks,
the Rugas, the Garas or Ganchings, the Atongs, and the Me’gams). See Marak,
Garo Customary Law and Practices, pp. 2–3; Milton S. Sangma, History and
Culture of the Garos (New Delhi: Books Today, 1981), p. 4.
69. Cf. De Vos, “Ethnic Pluralism”, p. 23.
5
‘DUAL WERE DUAL, KOCHU WERE
KOCHU’: GAROS DIVIDED
Although the cultural distinctiveness an ethnic identity provides is a necessary

condition for the existence of an ethnic group, it is not a sufficient condition.
… there must also be structural oppositions between groups for ethnic
boundaries to exist.1
Away from home and outside the Garo area, a Garo may well ask anyone who
looks Garo: “Na Mandi ma [are you Mandi]?” A positive reply surely leads to
smiling faces. This question shows more than inquisitiveness. It also reflects
the strong sense of Garo identity that is shared by most present-day Bangladeshi
Garos. Feeling Garo plays an important role in their lives and meeting a
fellow Garo on the way provides a sense of recognition, identification, and
security. Garos share a strong sense of belonging together, of constituting a
distinct ethnic community with a common history, culture, origin, and cause.
Contemporary Bangladeshi Garos also answer the definition of what De Vos’
calls an “in-group”:
A sense of common origin, of common beliefs and values, and of a common
feeling of survival – in brief, a “common cause” – has been important in
uniting people into self-defining in-groups. Growing up together in a social
unit and sharing common verbal and gestural language allows humans to
develop mutually understood accommodations, which radically diminish
situations of possible confrontation and conflict.2
88
Garos Divided 89
Although in practice, relations between hill Garos (from India) and plains
Garos (from former East Pakistan and now Bangladesh) have become much
more difficult to sustain since the partition of 1947, and the people on each
side of the border seem to have increasingly developed in two different
directions, Indian Garos “are also Garos”, and are considered a vital part of
the imagined community of Garos. These perceptions are consistent with
nineteenth-century colonial observations, which also portrayed all Garos –
whether they lived in hills or plains – as members of one race, nation or
tribe. However, in the previous chapter, it was suggested that colonial
categorization did not correspond with local perceptions of social groupings
and boundaries, and that nineteenth-century Garos did not constitute one
category of people or one self-defined in-group with a shared ethnic identity.
This chapter concentrates on emic recollections of the recent past. On the
basis of those memories, it is possible to suggest that Garo ethnicity is a
relatively recent phenomenon.
GARO DIVISIONS
In the early nineteenth century, Buchanan discerned five Garo tribes,
each living in their separate geographical area. Almost one century later,
Playfair distinguished as many as twelve geographical divisions among the
hill Garos, and five more among the Garos of Mymensingh.3 Oral history
provides the basis of this chapter. Personal recollections of elderly villagers
confirmed that Garos used to be divided into different groups, that these
groups played an important role in the organization of their social lives,
and that people were well aware of the groups they and others belonged
to. Our informants listed the following names: Chibok, Babul, Abeng,
Dual, Atong, Megam, Brak, Somon, Kochu, Ruga, Gara, Ganchi, Jaring
(or Jaring Adong), and Diggil.4 These divisions and namings closely
resemble Playfair’s categorization.
Current researchers of Indian Garos also confirm such segmentation.
They refer to Garo groups as geographical, linguistic, or cultural sub-groupings,
divisions or sub-divisions. For example, Chie Nakane contends that
“[t]raditionally the Garos distinguish among themselves a dozen groups
related to geographical area”,5 Julius Marak writes that the Garos are divided
into a number of sub-tribes, that “there is not much cultural difference
between these tribes,” and that the basis of each group is dialect. Burling
refers to the different groups both as linguistic and as geographical sub-
groupings.6 In his book about Rengsanggri, a Garo village in India’s Garo
Hills, he writes that: “The Garos divided themselves into a number of
geographical subgroups. Dialectical differences are important in defining

these, but there are other distinguishing cultural features as well.”7
The Garo anthropologist Milton Sangma writes that there were eleven
groups among the Garos. “Each group lived in splendid isolation and thereby
evolved a distinct culture and dialects of their own.” Sangma believes that in
the course of time, “all the eleven groups have almost become one, as a result
of the improvement of communications, transport facilities and other economic
projects.”8 Although historical data is not quite consistent with the idea of
“splendid isolation”, I too argue that in Bangladesh, different divisions have
become increasingly entangled with each other as a consequence of various
socio-economic, cultural, and political developments in the region.
Burling and Kibriaul Khaleque conducted research among Bangladeshi
Garos (of Modhupur forest). They mention a similar division for those Garos
living in present-day Bangladesh. Both authors refer to linguistic, cultural,
and geographic differences that separated the groups. Special reference is
made to the Atong, whose language is not mutually intelligible with the other
Garo languages or dialects. Yet, Burling and Khaleque themselves had limited
personal experiences with these different groups because most Modhupur
Garos identify as Abeng, one of the divisions.9
GARO GROUPS OR ‘DOL’

Divisions, sub-divisions, groups; these are all English terms. The question
remains how Garos themselves talk about these groupings. In his book,
written in Bengali, about the Garos of Bangladesh, Bangladeshi Garo author
Shubhash Jengcham mentions thirteen different dol.10 Nevertheless, the word
dol, which can be translated into the English word “group”, is a Bengali word.
Yet, contemporary Garo language does not seem to include any appropriate
word to refer to these particular different Garo groups. In the words of the
villager Ketu Patang: “Chiboks were Chiboks, Kochu were Kochu. We did
not have any special term word for them.”
This complicated our research and during our inquiries about different
(Garo) divisions, we had to resort to Bengali words such as jat, gotro, or dol.11
Each term led to confusion among our respondents, who were not quite sure
how to interpret our questions. For example, when the villager Rajendra was
asked about different jats among the Mandis, he answered that “there were
Mandis, Hajongs.” When we explained that we did not want to talk about
these jats, he asked whether we meant mahari (matrilineal descent group).
Only when we mentioned the specific names of the divisions (such as Dual,
Chibok, Kochu, or Abeng) did he understand what we were after: “Oh, that
Garos Divided 91
is not mahari. There were Dual, Chibok, Kochu, Abeng. I have only seen
those four jats.”
After this reply, we asked him if those groups were generally called jat
after all. To this he answered: “No, they have different languages, but the jat
is the same.” Today, the meaning of jat is similar to “ethnic group”. Garos
consider themselves a jat distinct from Bengalis, Hajongs, Chakmas, or any
other ethnic group.12 It is, however, difficult to assess whether contemporary
perceptions of identity are reflected in Garo representations of the past. What
does today’s use of jat (all Garos are a single jat) say about the past?
A second alternative, gotro, did not work well either, and Jengcham’s term
dol led to similar confusion. When Bupindra Toju was asked whether Atong
(one of the Garo divisions) used to be referred to as jat or dol he answered:
“They are also Mandi jat. They are not even a different dol. Why should they
be a different dol ? We had a relationship with them. They are Atong but
Mandi. For instance, the Sylheti Bangals speak a different language but they
are still Bangal.”
The only unequivocal manner to touch the subject of Garo segmentation
was by specifically using some of the group names. Our informants instantly
understood us when we asked: “What about Abeng, Atong, Dual, Chibok?”
Here, in the text, I have opted for dol, for want of a better alternative.
This word, also employed by Jengcham, comes closest to contemporary Garo
perceptions of these specific (Garo) divisions. And I need a distinct term to
explain this specific complex situation. Dol does not refer to just any
fragmentation, it refers to a specific divisionary system.
GARO ‘DOL’ TODAY

In this chapter, I discuss four different cases to illustrate Garo recollections of
these dol. Personal accounts also reveal that the relevance of the different dols
has diminished in recent decades.13 Today they no longer play the same role
in the lives of young Garos as they did in the lives of their grandparents. They
do not significantly influence, and certainly not hinder, social relations
between Garos. People from different group backgrounds wine and dine
together. They meet each other, even marry each other:
There is little sense of special loyalty to own’s own linguistic subgroup, and
differences in social organization or traditional religious belief are of little
importance. People can chuckle at the odd dialects spoken by people from
other areas, but the sense of being a Garo is much stronger than any sense
of being an A’beng, Chibok, or Awe.14
Although the daily significance of dol has diminished, three divisions or dol
still play some role in the lives of contemporary Garos. These are the Atong,
Megam, and Abeng.15 Atong and Megam live a little separated from other
groups and speak dialects or languages that clearly differ from today’s lingua
franca of Bangladeshi Garos: Abeng. Neither Atong nor Megam can be
understood by other Garos; their languages are clearly distinct. It is interesting
to find that the latter two are exactly the divisions which mostly confronted
colonial and post-colonial observers with problems of identification and
categorization.
Abeng, the third group, has come to stand for mainstream Bangladeshi
Garo. The Abeng not only constitute the largest dol, they are also most
popular. Often I heard that Abeng are “Mandi sa.agitang”, the real Mandis
(Garos of Bangladesh). I found instances of people claiming to be Abeng
when they knew they were not the offspring of Abeng parents. The opposite
never occurred. The popularity of Abeng may have to do with the role of their
language as lingua franca. Missionaries probably played an important role in
spreading Abeng among all Bangladeshi Garos (see also Chapter 7).
FRAGMENTED GAROS
Although Garos like to uphold an image of homogeneity, reality shows that
they are fragmented in many ways, according to various criteria. The first and
most obvious division is the separation into Bangladeshi and Indian Garos, a
distinction which largely corresponds with the division into hill people and
lowlanders (a minority of India’s Garos live in the plains of Assam, West
Bengal, and Tripura). Other factors which fragment contemporary Garos are
class, education, religion (and denomination), and habitat. We have already
seen how migration has led to a division between city and village Garos.
Access to higher education and white-colour jobs have created a new Garo
middle class. Christianity has brought about a division into Christian and
Sangsarek Garos and has further separated Garos into various Christian
denominations (see also Chapter 7). These segmentations have been bound
by time and place. Whereas denominational and educational divisions are
relatively new to Garos, geographical and linguistic differences are already the
basis of Buchanan’s separation of the Garo nation into five Garo tribes.
Kinship, another significant factor that organizes and categorizes Garos,
needs specific attention here. Although dol and kinship are two different
divisionary systems, they do not operate isolated from each other. Below, is a
brief discussion on how kinship organizes Garos and how it differs but also
intersects with dol.
Garos Divided 93
GARO KINSHIP DIVISIONS

Garo matrilineal kinship organization has been a favourite research subject of
many anthropologists.16 One important reason is that in the South Asian
context, matrilineal kinship organizations are quite exceptional.17 Another
motivation is the assumption that kinship has largely determined the social
structure of Garos. Burling argues that it provides the organizational principle
by which Garos relate to each other. He also writes that “[p]eople spend
much of their lives working with the members of their own households, but
these are embedded in a wider network of kinship, and the network is
constantly evoked as people address one another and refer to one another.”18
In Bangladesh we can witness a general tendency to loosen kinship rules
and kinship-related practices. Where marriage is concerned, kinship criteria
seem to be slowly giving way to criteria such as class, social status, and
educational background. I cannot tell how status and economic situation
influenced earlier matchmaking processes, but available data suggest that
kinship played a more significant role some decades ago.
Garos are often mistakenly called matriarchal.19 Such notions, which are
regularly heard in Bangladesh, render Garos more exotic and curious for
outsiders. Particularly Garo men feel uncomfortable with an image that
suggests female superiority, and which therefore contrasts all the more with
the position of Bengali patriarchal society, in which men clearly dominate
public life. Contrary to what many Bengalis think, Garos are not matriarchal,
and Garo women certainly do not rule all domains of Garo life. In fact, Garo
men dominate in most public spheres. This is certainly the case in matters of
politics and religion. Even though the matrilineal kinship organization certainly
attributes to a relatively free and equal position of Garo women – especially
when we compare their position with Bengali women – “matrilineal” primarily
means that children belong to the kinship group of their mother and carry
her kinship title.
In order to explain and describe the Garo kinship system, anthropologists
have resorted to terms such as “clan”, “sept”, “motherhood”, “moiety”, and
“lineage”. One complication is that concepts like these are not exactly equivalent
to local Garo kinship terminology.20 Another problem with kinship is its
particular “primordial” connotation. Kinship provides the organizational
principle of many social groupings and is primarily based on (real or imagined)
biological ties between individuals. Despite its primordial dimension, however,
kinship is primarily a cultural, and therefore dynamic, organization of human
relations. It provides structures and solutions to problems that societies find
themselves confronted with.21 Twentieth-century history of the Bangladeshi
Garos is a good example of how socio-economic, political, and cultural
changes can affect kinship system, kinship terminology, and related issues
such as gender relations, labour division and inheritance practices.22 Here I
only briefly discuss Garo kinship, touching on some relevant conceptual
changes and showing how it divides Garos and how it intersects with other
divisions such as dol.
Contemporary Garos are primarily divided into two exogamous groups:
Sangma and Marak.23 A Sangma has to marry a Marak and vice versa. Playfair
referred to these groups as katchis, whereas Khaleque and Marak, for example,
call them chatchis.24 I came across the word chatchi very frequently and found
that Bangladeshi Garos use it in different ways, often simply as guest or
relative.25 Both Sangma and Marak are sub-divided into dozens, perhaps even
hundreds, of smaller exogamous matrilineal kinship groups or matrilineages,
which Bangladeshi Garos either call ma’chongs or maharis. Examples of these
groups are Chisim, Chisak, Mrong, Drong, Rema, Toju, Nokrek, and Patang.
In Bangladesh, Garos use their ma’chong titles in a similar way Europeans use
surnames. It is interesting to note that this is quite different from the Indian
situation where one calls oneself Sangma or Marak.26 Contemporary Garos
are perhaps not always sure of their kinship terminology – I witnessed some
heated discussions about the exact meanings of mahari, ma’chong, and chatchi
– but every Garo knows perfectly well to which ma’chong he belongs. Divisions
into ma’chongs still play a very important role in the structuring and organization
of relations amongst the Bangladeshi Garo community.
Until recently, marriages used to be arranged by parents and other
relatives. There was a distinct preference for an alliance with the kinship
group of the father, preferably with his sister’s son. This explains why in
many Garo villages two kinship names dominate. In Bibalgree, for example,
many Garos were either Patang or Toju. The rationale behind this preference
was that men would always be assured of support from a close relative. After
all, his own son-in-law would be a close kin. These days, fewer and fewer
marriages are arranged. More and more young people are choosing their
own life partner. Such choices are certainly bound by rules, but kinship
considerations play a less dominant role in the spousal choice. Today’s
number one rule is that a Garo should always marry a Garo (see Chapter 6).
A second rule is to never marry someone from one’s own ma’chong, but
Sangma-Sangma or Marak-Marak matrimonies occur increasingly. Marriages
like these used to be called madong, which literally means “married to
mother”, and in earlier days, such couples had to flee away. Today these
types of unions are more or less accepted, although still perhaps not
without any difficulties.27 Marriages between a Chisim and Chisim or a
Garos Divided 95
Rema and Rema are still considered unacceptable, madong. Especially among
urban middle-class Garos, we can witness a tendency to name children after
their fathers. This is a significant break with Garo kinship rules. Today
parents defend that choice by explaining that the name is the father’s but
the ma’chong remains the mother’s.
Directly related to kinship is Garo inheritance. Here too we can witness
a number of very important changes. Until recently property was passed from
mothers to daughters. Usually, one daughter was appointed as the main
heiress, the nokna. These days more and more parents divide the property
among all their children. This violates Garo law as it was documented in the
course of this century and is still officially applied in Bangladeshi court.
Sisters can therefore officially file a suit if their mother decides to pass on
property to her son(s) as well. In order to circumvent these rules and regulations,
mothers sometimes sell their property to their sons for nominal sums.
But the question of how kinship is related to dol remains. Each different
Garo dol is divided into these kinship categories. We can find Chisim, Rema,
and Marak (ma’chongs) among the Abeng, Atong, Dual, and Chibok. Garos
often refer to members of their own ma’chong, whether from their own or a
different dol, as relatives with whom they have to maintain a good relationship.
For example, Abeng have relatives among the Chibok, with whom they
should maintain a relationship. In practice, this means that on the occasion
of burial ceremonies or weddings, relatives should come to pay their respect
to the deceased. Here kinship cuts across linguistic and geographical divisions.
This is evident for the present-day situation, in which dol has little significance
anyway. Unfortunately, the manner in which kinship and linguistic divisions
intersected earlier has not been sufficiently studied.
How can we understand our informants’ claim that they had no contacts
with other dol while at the same time they refer to “relatives” among those
other dol with whom they had to maintain a good relationship? Perhaps
Nakane’s analysis of the Garo kinship system can help us understand this.
According to Nakane, each mahari is segmented into many localized lineages.
And even though there is a strong feeling of cohesion among the same mahari
people, the “actual organic body” is established on the basis of these localized
lineage groups. Spouses are not chosen simply from any mahari from the
opposite “moiety” or chatchi (Sangma and Marak), but from a suitable
localized lineage group.28 This would also explain why Garos could stick to
their own dol, while having mahari relatives among other dol.
Nakane further explains how Garo kinship acts as an effective boundary
marker between Garos and their neighbours. Nakane writes that:
this localized lineage group emerges into broader mahari organisation and
again is linked to the widest range of kinship, chatchi. These vertical
unilineal lines across a moiety organisation through exogamous marriages
thus construct the Garo social organisation which marks off the Garo from
neighbouring tribes and peoples.29
Such an explanation gives rise to the question of how it is possible that
different dol have the same kinship system while carefully maintaining their
social (dol ) boundaries. Does this not mean that each dol must have the same
origin after all? In this respect, it interesting to note that among neighbouring
peoples such as Koches and Rabhas, ma’chong or mahari can be found that are
the same as those of the Garos, and that, as found by D.N. Majumdar, among
some sections of the Koches, there are “clans [ma’chong or mahari] identical
even in name”.30 The situation is obviously far more complex than Nakane’s
quotation suggests. More research needs to be done.
MEMORIES AND HISTORIES

Oral history was vital for this study. Both the scarcity of written documents
from the hands of others and the absence of written “views from within”
render personal accounts indispensable. The life stories which I collected
provided information on at least two analytical levels: the level of discourse
or presentation and the level of social reality. They both reflect how
contemporary Garos see themselves and others, and provide insight into
the recent history of Garos. This gives rise to the question of how to distil
historical “facts” from presentations, a question which is all the more
important since there is little scope for comparison with other (written)
accounts. Earlier I argued that contemporary Garos exert themselves to
present Garos as a close-knit community; they stress solidarity and harmony
and hide internal conflicts and divisions. Such attempts are also to be
expected in their representations of the past. Then what do their life stories
really tell us about the Garo past(s)? The answer is provided by the same
logic. If we expect Garos to stress harmony and unity, stories about internal
divisions and strife – which go much against the ideal-type image of Garos
– are all the more likely to bear some truth.
Bernard Cohn offers an interesting framework to understand the
complexity of history(s). He distinguishes two ideal-type pasts for traditional
societies. The traditional past grows out of mythological and sacred traditions.
It validates a present social position and provides a charter for the maintenance
of that position or the effort to improve it. The traditional past also relates a
particular group to an extensive network. The second type, the historical past,
Garos Divided 97
is a set of ideas about remembered experiences of a group of people in a local

region. It explains, advocates, or renders justification for social action in the
local social system.31 These two pasts also help us to unravel and understand
Garo narratives of their past(s). Stories about their migration from Tibet and
the subsequent diaspora (in Dolgoma) fit the first type. Narratives about the
different groups or dols, which suggest internal division and discord, conform
to Cohn’s historical past. In practice, Garo representations of their past(s) are
a mixture of both ideal types.
Below are four stories about the Garo dols. Each story emphasizes different
aspects of this particular group division, illustrates different aspects of its
relevance, and tells about how people lived and organized their social lives.
FOUR GARO STORIES OF GARO GROUPS

1. A STORY OF SIMILARITY AND SOLIDARITY
Nurudin32 lives with his wife, children, and grandchildren in the village
where I was staying. Although the younger generations in the household are
Christians, the old couple are still Sangsareks. Nurudin does not know his
age, but he remembers that he married his wife in 1942. He was born in a
village a little further away and moved to his wife’s house after marriage. His
attachment to the old Sangsarek religion is so strong that he has continued to
resist Christianity in an environment in which conversion and Christianity
have come to play an increasingly dominant role. When we asked Nurudin
about the early history of the Garos, he answered that his ambis and achus33
had told him the following story:
“They [Garos] lived in Tibet ‘Asong’ [country]. They came down to settle
in Dolgoma. But Dolgoma was surrounded by dense jungle, full of bears
and tigers. That’s why the Mandis came to the plains where there was less
jungle. They first settled in Bildura village and from that village they
spread.”
“Tibet was overpopulated and the lands were less fertile. They heard that
this area was very fertile and less populated. That’s why they left Tibet and
started to look for new lands. It’s like when the [Bengali] refugees came here
[in 1964] when they came to know that here were a lot of lands.”
In Nurudin’s story, Bildura was the place from where Garos left in different
directions. In many other stories, it is Dolgoma that features as the starting
point of Garo diaspora. At present, Dolgoma has a mythic ring to it. Several
informants told us that it was Dolgoma from where Garos split up in
different groups and dispersed over the Garo hills and adjoining plains.
Nurudin draws an interesting parallel with the mass influx of Bengali

refugees into northern Mymensingh in 1964. The sudden arrival of Muslim
Bengali migrants led to one of the most traumatic experiences for northern
Mymensingh’s Garos during the twentieth century. Garos often point to
1964 as the turning point between the good old days and the difficult
present. Many Garos tend to divide their personal history into good and bad
times, with the occurrences of 1964 as the turning point. Chapter 8 deals
with the 1964 disturbances in greater detail.
Nurudin also tends to describe the days before 1964 in terms of peace,
prosperity and unity. Everyone, whether Garos or others, maintained close
relationships. “Only the Muslims were different.” Nurudin’s tendency to
stress unity and harmony is clearly reflected by his representation of the
different Garo groups. When we asked him to describe relations between the
different groups, he gave the following answer:
“It was nice. Our origin was the same. Everything was the same. The only
difference was that we worshipped in different ways. There were also Braks
and Duals in this area. Babuls and Abeng are the same, and Chibok and
Kochu are the same. There was a good relationship between all groups.”
Nurudin’s story about the groups is clearly in line with his earlier account
about the common origin and shared history of the Garos. Although he does
not deny the existence of different groups, he stresses that differences between
the groups were minor and that relations between them were good. Such a
reply is not strange when we take Nurudin’s personal experiences into account.
After all, Nurudin himself had married a woman from a different group: “I
was Abeng and my wife Chibok. It was possible to marry anyone from among
the Mandis. Only marrying a Bengali was not possible. It had always been
like this. Only the Christians were confused about marriages with Sangsareks.”
According to Nurudin, Christian Garos were not quite sure whether they
should or should not marry Sangsareks and marriages with Bengalis were
unacceptable, but Garos could easily marry someone from another Garo
group. Others told us rather different stories, some villagers insisted that
Garos could only marry within their own group. Those interviews also reveal
how the intensity of inter-group contacts differed from group to group and
that the principles of exclusion and inclusion in fact allowed for differences of
degree. Some dol were perceived almost like oneself, whereas others were
considered very different.34 It seems that the relations between the different
dol were primarily determined by factors such as spatial distance, language,
culture, and religion. The closer two groups lived together, the more frequent
and intense the contacts, and the easier it seemed to cross the boundaries.
Garos Divided 99
This might also explain Nurudin’s “mixed” marriage: the Abeng and Chibok
were living closely together and were very similar.
Although Nurudin stressed that all groups interacted freely and that
inter-marriage was no problem between any of the groups, he continued that
“The Atong and Megam were different. They lived furthest away from here.
But even with them marriage was possible. The Brak were experts in medicine.
Like the Hajongs, they knew a lot of gun ban [the art of killing people with
the help of herbal potions].”
Here he acknowledges that the Atong and Megam, who lived furthest
away, indeed differed from other Garo groups, and that a third group, the
Brak, were quite intimidating: like the Hajongs, they knew how to brew
herbal potions that could kill people. Nurudin himself does not take any of
these slight differences into account; none of them stood in the way of Garo
unity and solidarity. Throughout the interview, he presented a picture of
unity, similarity, and solidarity. In Nurudin’s opinion, the only boundary of
real importance was between Garos and Bengalis.
The following accounts reveal a more complex society in which processes
of inclusion and exclusion were far less easy to pin down than he told us. I
cannot tell whether his account reflects his memories, or whether he merely
attempted to keep up the contemporary image of the Garos as a close-knit
group of people, sharing a common origin and history.
2. ‘WHY DO YOU CALL ME CHIBOK?’

Ketu thought that he was somewhere around 65 years old. After his marriage,
he moved to his in-laws’ house. Since his wife did not own any land, Ketu had
to earn his money as a day labourer or sharecropper. Despite his own poor
background, Ketu remembered the days prior to 1964 as a good time: people
had lands to cultivate and peace ruled the area. Problems only started in
1964, when many Bengali Muslims settled in the region and most Garos
from the border areas fled to India. The year 1964 marked some drastic
changes in the relationships amongst the Garos:
“In 1964, the Mandi people got in trouble for the first time. These problems
were the same for all groups and everyone had to flee to India. During those
days, all Mandis became united. They realised that they were the same
people. Since then they have not cared about who is Atong, Megam. Before
that, there were no relationships with Megam and Atong.”
The disturbances in 1964, which led to a massive flight to India, were
experienced by all Garos from the border region and acted as an eye-opener
for everyone. Only then did all the Garos realize that they belonged to the
same people. Until that year, the situation had been quite different. Ketu
described the situation prior to 1964 as follows:
“There were Atong and Megam among the Mandis. They were also created
as Mandi, but their languages were different. We only maintained
relationships with those living around us. The Atong, Megam, they lived far
from here. That is why we did not maintain any contact with them. We
only stayed in touch with Dual, Chibok, Brak, who used to live very close
to us. Their languages were different, but not too different to understand.
Suppose we say ‘mi chajok ma’, the Chibok say ‘my chajok mo’.35 But none
of us married into other groups then.”
Particularly two factors that determine relations between different Garo
groups stand out: spatial distance and linguistic difference. Ketu’s own group
(Abeng) were in touch with the surrounding Dual, Chibok and Brak, but not
with the Megam and Atong, who lived far away. Interaction did not, however,
mean that people from different groups could marry each other. According to
Ketu, no one married into another group. In the following interview fragment,
he tries to provide an explanation for these marriage restrictions:
“When I was very young, I did not see anyone getting married to someone
from another group. Some of them used to live in separate areas: for
example, we lived in Haluaghat thana and the Atong lived in Durgapur
thana. Some groups lived nearby. But some of their languages were so
different that we could not understand them. It was not possible for us to
know what they were talking about, what they were thinking. Some of their
food was different as well. Before the Partition [1947], people from the
Garo Hills used to come here and some of them married our people. Their
language was also different, so I don’t know how they managed to marry
one of us, and I don’t know why marriage between other groups was not
possible. But today there are no restrictions about this.”
Language was evidently an important criterion for inclusion and exclusion,
even though, as most informants told us, everyone also knew Abeng. People
did not speak each other’s kushuk (language or dialect), but everyone also
knew Abeng.36 Burling writes that the Atong and Megam speak different
languages, but that other Garos normally understand one another. “People
from widely separated areas have to be just a bit careful if they are to avoid
misunderstandings, but the dialect differences pose no serious barrier to
communication.” Nevertheless, Burling also found that people are extremely
conscious of even the tiniest linguistic differences and easily create the
impression that they are talking about some major linguistic differences,
when these were in fact rather small, from a linguist’s point of view.37
Garos Divided 101
For our understanding of the relation between different Garo groups,

objective distinctions (if they are at all measurable) are not merely as interesting
and relevant as perceived differences. Even the slightest dissimilarities, when
they are considered important, may have a significant impact on the perception
of Self and Other. “Group identity can even be maintained by minor differences
in linguistic patterns and by styles of gesture.”38
Whereas language hindered interaction between Garo groups from
Haluaghat thana (where I conducted most of my research) and the Atong and
Megam, who lived much further to the east, it did not stand in the way of
inter-marriage between Garos from the hills and plains. Ketu could not
explain this difference. These marriages may be explained by kinship relations,
as we shall see later.
So far we have seen that groups living in Haluaghat hardly maintained
any contact with the Atong and Megam, who spoke different languages and
lived far away. Although no one married outside his own group, this did not
hinder other interaction between groups that lived in close proximity:
“Actually, there were no problems between the groups. We had been living
in the same area – of course not in the same villages – and sometimes we
drank chu (rice beer) together. During the drinking, when people got
drunk, they may have quarrelled with each other. Perhaps a drunken person
called a Chibok by his name Chibok. In such cases, that person would ask;
‘Why do you call me Chibok?’ And sometimes this led to disputes.”
Here, Ketu explicitly pointed out that the different groups did not inhabit the
same villages, but that people would meet and drink together. Interestingly,
it could be insulting to call each other by the name of his or her group, and
people were clearly conscious of each other’s group background. Hereafter we
shall also see that name-calling and stereotypes played an important role in
maintaining some social distance from other groups.
3. “WHO WANTS TO MARRY A BRAK WHO EATS COW DUNG?”

Suborno and I particularly liked to visit Nondita and her husband Rokesh.
Even on extremely hot and humid days, we found a cool place on their
perfectly clean and well-maintained homestead, and if they found the time,
they would sit and talk with us, Rokesh smoking his hookah (water pipe) and
Nondita chewing her pan. At times some of their children, grandchildren, or
neighbours would join our conversations. Rokesh had been married twice
before he wed Nondita. His first two wives had died. Rokesh, who must be
somewhere between 70 and 80 years old, is some twenty years older than his
present wife. His age and good memory made him a particularly interesting
informant. During the months that we stayed in the village, the couple told
us many stories. That age takes its toll became apparent when we met Rokesh
again one year later, during one of my short visits to Bangladesh, when his
memory had clearly started to let him down. A few years ago, Rokesh died,
but Nondita is still living in the same place.
The family had not always lived within the relatively safe boundaries of
the village. Until the disturbances of 1975–77, when a number of Garos
fought a guerrilla war against the Bangladesh army, their homestead was
located at some distance of the village, out in the open fields. After as many
as ten robberies, they decided to settle within the boundaries of Bibalgree.
These 1975–76 experiences, to which I shall come back in Chapter 8, added
up to a general sense of fear that had already started in the early 1960s. Today
Rokesh and Nondita clearly worry about the current situation of the Garos.
Like Nurudin and Ketu, they picture the days before the mass influx of
Bengalis (1960s) as an era free of theft and robbery. Everyone, including the
local Bengalis who had been living in the area for many generations [sthaniyos],
lived together in peace. Rokesh remarked regrettably:
“Now Ellen has come here, but if she had come before, she could have seen
how it was. Now this is a Bengali country. Now there is no village where the
Mandis are the majority, and rich. Had she come in those days, she would
have really believed that here were a lot of Mandis.”
Notwithstanding these positive recollections, neither Rokesh nor Nondita
were hesitant to speak of the division and separation that also marked those
earlier days. Rokesh and Nondita spoke alternately:
Rokesh: “Among the Mandis, the relationships were good. In this area,
there were only Mandis. The Megam and Atong were very far away from
here, in the east, on the other side of the river Simsang [near Durgapur]. In
Moheskola [bordering Sylhet], there were Diggil. They used to kill people.
They were Mandis, but they were different. There were other groups here
in this area, such as Babul, Dual, Chibok. We had a good relationship with
them, but they were a little bit different.”
Nondita: “Their wana krita was different.”39
Rokesh: “Now nobody cares who is Abeng, who is Chibok. But earlier, all
groups lived separately.”
Nondita: “The Chiboks lived in their village and the Abeng in theirs.”
Rokesh: “The Somons had a separate area too. The Chisim family in
Bhutiapara are Somon. … There were no quarrels between those groups.
We lived in peace with them, but intermarriage was not possible. There are
Garos Divided 103
some Braks in Kasibari and Somuniapara; they are Chisims. But now they
don’t speak their language.”
Rokesh and Nondita reported that every group lived in a different village, if
not in a separate area. The area where I conducted my fieldwork was primarily
inhabited by Babul, Chibok, and Dual. Other groups, like the Megam and
Atong, lived relatively far away, near Durgapur. Each group differed to some
greater or lesser extent from the other ones. And although people lived
together peacefully, intermarriage was not possible. In another part of the
interview, Rokesh said that, although uncommon, intermarriage was possible.
But the differences between the dol made such marriages very difficult.
Nondita added to this explanation that
“They said: ‘Who wants to marry a Brak who eats cow dung.’ And they also
said: ‘who would want to marry a Somon who eats tadpoles?’ We Chibok
were considered goats by them, so how could they marry us?”
Rokesh: “Nowadays it is not like this. There are no restrictions about inter-
marriage. But before it was different. We hated each other. The Babuls were
like slaves in our eyes. They also thought we were fools.”
Stereotypes and prejudices were recurrent phenomena in the narratives about

“the other groups”. We already saw how the Braks were supposedly well-
acquainted with herbal poisoning. Nondita knew this too: “Oh! they were
experts in ‘medicines’ such as gutok sam. That’s why some people were afraid
to go to their house.”
This story clearly illustrates that spatial distance and stereotypes added to
social distance between the different groups. I heard some particularly
interesting stories about the Atong.
4. ‘NOBODY COULD MARRY INTO A DIFFERENT GROUP’

One day Monen Morol (Monen leader),40 whose name is really Monendra
Ritchil, found a bible in which one of his uncles had written the date
7 February 1909. Monendra considers this the date of his birth. He was born
in a village near Haluaghat bazaar. His mother died of cholera when he was
still a baby so he and his older brother lived with their maternal grandmother.
His father stayed with them for some years, but then returned to his mother’s
village where he died some months later.41 After the death of his grandmother,
he moved to his ajong’s house (mother’s elder sister). His ajong and her
husband were not interested in sending him to school any longer and Monendra
had to work in the fields fulltime.
In 1342 (1935), his life changed suddenly when he was asked to marry
the granddaughter of a prosperous Garo businessman. Monendra accepted
and the marriage was arranged. Although Monendra’s parents had already
been Christians, conscious of his identity, he preferred to marry in the
traditional Garo way. He considered a Christian marriage ceremony as foreign.
The couple met for the very first time on the day of their wedding. Some
fifteen years later, Monendra took a second wife, a younger sister of his wife.
Monendra had altogether twenty children.42
His wife’s grandfather, who died some six or seven years after Monendra
had joined the family, was a successful businessman who involved Monendra
in his enterprise. They ran a profitable trade in buying and selling goods such
as rice, mustard seeds, cotton, jute, and tobacco. The partition of 1947 and
its aftermath finally led to the end of it around 1950. The family’s business
enterprise comprised much more than petty trade. They travelled widely to
buy, store, and sell their goods. Tobacco was, for instance, bought in Rangpur
and transported to Bhairab, a place not far from Dhaka. Here the bulk of it
was sold to Hindu businessmen from Sylhet who came to Bhairab with huge
boats to transport the tobacco back to Sylhet. Cotton was purchased in the
Garo Hills and stored in Dalu, on the border between Bengal and the Garo
Hills. Businessmen from Narayanganj, Tangail, and Bikrompur came by boat
to buy the cotton.
Monendra, who has travelled widely, is a rich and respected elder in the
Garo community, and has a great ability to recollect and narrate stories about
the past. This made him a very interesting and valuable informant to our
research. He was quite outspoken about the relationships between the different
groups. He had witnessed the division when he was young and argued that no
social relationships between the groups were maintained. Only through the
introduction of Christianity did this change: “Now all are one. That is the
one good thing that has come with becoming Christian. There were divisions
among us. Those are no longer there.”
About the different groups, he related the following:
“There are many groups among the Mandis, such as Dual, Chibok, Kochu,
Somon, Brak, Atong, Megam, Ruga, Gara, Babul. There were no social
relations between the groups. Inter-marriage was impossible. The Babul
maintained relationships with Babul only, the Chibok with Chibok. These
groups used to criticize each other. Now most of us are Christians. That’s
why we’ve forgotten about the groups. Earlier we used to say: ‘You are Dual,
you are Chibok, they are Somon.’ A Babul would never marry a Dual; a
Chibok would never marry a Dual. Only Chibok and Chibok, Dual and
Dual could marry. Mimang kam and other events were also different.”
Garos Divided 105
Monendra asserts that the different dol maintained no social relationships

with each other. For example, intermarriage was not accepted. This is not to
say that group members were unaware of the existence of other dol. They were
obviously very conscious of someone’s background, and the mutual feelings
about the other were marked by criticism. We have already seen that the other
examples also, revealed negative ideas and prejudices about other dol than the
own. Monendra said that: “Maybe the Chibok were not so clever and that is
the reason why other people teased them by saying that Chiboks are as stupid
as goats. But this was no reason for trouble. They didn’t take it seriously.”
Dol did not fight over such discord. Earlier, however, dol did fight each
other. According to Monendra, they went as far as cutting each other’s heads;
this caused the Garo diaspora. The Garos became dispersed, but the dol
stayed together:
“In a Brak village, there were only Braks. In a Kochu village, there were only
Kochu. Dual and Somon could never live together. You could only find
Atongs in the Baghmara area because only Atongs settled there. In the
Sibbara area, there are only Garas. The Matchies live in the upper Garo
Hills. The Babuls live in the hills. In Bangladesh, there are Babuls in
Donbanga, Bekamara, Gilaboi, Kakorkandi.”
Different groups lived in separate areas, or at least in different villages.
Nevertheless, despite Monendra’s outspoken ideas about the earlier segregation,
the absence of social relationships, linguistic differences, and different practices,
he asserted that all dol considered themselves Garo. He also indicates that
each could both be found in the hills and in the plains. Yet, from the
following statement it appears that the kinship system, at least to some extent,
cut through dol boundaries:43
“The relationships among the chatchis were very close. If someone died, all
members from that chatchi would come from far away. If someone didn’t
manage to come, he or she would fast until they came to know that the
body had been cremated. All chatchis joined mimang kam, even those from
far away would come. It didn’t matter whether they were Chibok, whether
they were Dual. Chatchis are chatchis. If a Chibok Chisim died, all Chisim
from Babul, Dual, Somon, and other groups came. These relationships were
very nice, no matter to which groups people belonged. Only with other
chatchis from different groups, it wasn’t very close. There wasn’t much
contact with them.”
In the event of funerals or important traditional celebrations, chatchis from
various groups would come to join the happening. Where does this leave the
earlier exegesis that dol maintained their boundaries so strictly, that there were
no contacts between the groups? How is it possible that group members have
chatchis in other groups when inter-marriage does not exist?
Confused as we were by Monendra’s latter explanation, we presented him
the following case: a Chibok Chisim could not marry a Babul Ritchil, but a
Chibok Chisim could marry a Chibok Ritchil. Monendra asserted that such
was indeed the case: “Earlier nobody could marry into a different group. That
wasn’t possible.”
CONCLUSIONS
Today Bangladeshi Garos form a distinct ethnic group. Its members share a
common set of traditions which include religious beliefs and practices (today
that is Christianity), language (Abeng), a sense of historical continuity, and of
a common ancestry and place of origin. Oral history reveals that not so long
ago the situation was quite different and that the different Garo divisions or
dol played a significant role in the lives of Garos. Our informants explained
that these dol differed from each other on the basis of religious and cultural
practices, language, dress, and eating habits. Some of these dol were more
closely related than others, but each had its own name, was located in its own
area or village, and differed to some extent from others on the basis of kushuk
and customs. The intensity of inter-group contact seemed directly related to
spatial distance; the larger the distance, the more restricted the contact.
On a number of issues, however, the opinions of the interviewees differed.
Disagreement arose on the matter of identity and interaction. Some villagers
argued that the division into different dol had practically no influence on
relationships between individuals. These informants stressed that Garo identity
was much more important than dol identities. Others argued that group
members stuck together and that inter-group boundaries were strictly
maintained. They sketched a situation of different groups or dol watching the
social boundaries carefully. This was, for example, reflected by a negative
attitude towards inter-marriage. Inter-dol marriages were limited to other dol
in the neighbourhood.
All our informants agreed on one thing: all dol were also Garos. Although
it is rather difficult to assess the depth of these feelings for the first decades
of the twentieth century, my evidence suggests a situation of fuzzy identities
and flexible social boundaries. People could transgress the social boundaries
of dol on account of kinship, and be rather strict about these dol boundaries
in other situations. At the same time, however, some feelings of same-ness
lingered among all people called Garos (Mandis) in those days. In the
course of the twentieth century, that sense of a shared identity grew
Garos Divided 107
increasingly stronger. These days, the boundaries between dol have lost their
importance in favour of a shared Garo identity. The Garos of Bangladesh
have become one ethnic community.
NOTES
1. Keyes, “Introduction”, in Keyes, ed., Ethnic Adaptation and Identity, p. 5.
2. De Vos, “Ethnic Pluralism”, p. 15.
3. Playfair, The Garos, pp. 61–64.
4. Although there was a great overlap between the names mentioned by different
informants, some villagers mentioned more groups than others, and some
presented two of the divisions as one and the same (like Babul and Abeng),
whereas others described these as two separate groups.
5. Nakane, Chie, Garo and Khasi. A Comparative Study in Matrilineal Systems
(Paris: Mouton and Co., 1967), pp. 19–20.
6. Marak, Garo Customary Laws and Practices, p. 2; Nakane, Garo and Khasi.
7. Burling, Rengsanggri, p. 358.
8. Sangma, History and Culture of the Garos, p. 134.
9. Burling, Strong Women of Modhupur, pp. 15–16; Kibriaul Khaleque, Social
Change among the Garo. A Study of a Plains Village in Bangladesh (Unpublished
M.A. thesis for the Australian National University, 1982), pp. 10–14.
10. Jengcham writes that Garo shomaj [society] is divided into different dol. He
mentions the Awe or Akawe, Abeng, Atong, Ruga, Chibok, Chisok, Dual,
Matchi, Kutchu, Atiagra, Matjang chi/Matabeng (who do nor reside in
Bangladesh), Gara Ganching, and Megam. Jengcham, Bangladesher Garo
Sampradai, pp. 10–14.
11. The Bengali-English Dictionary (published by Bangla Academy Dhaka) describes
jat as: 1) “one of the hereditary social classes among the Hindus’: a caste”; 2) “a
religious community”; 3) “a class or kind of anything”; 4) “a racial variety: a
group possessing common qualities, breed”. Gotro is explained as “a group of
families, all originally descended from one common ancestor; a clan; a tribe”.
Dol is defined as: “party, group, company, body, band, team, or gang”.
12. This corresponds with Burling’s findings about the meaning of jat for
contemporary Garos. He comments that Garos used jat in a way that is close
to our use of “ethnic group”. Garos form their own jat, and other jat are
Khasis, Hajongs, Dalu and Chakma. See Burling, Strong Women of Modhupur,
p. 165.
13. The anthropologist Erik de Maaker, who recently completed his study of death
rituals amongst Sangsarek Garos in West Garo Hills district, defines dol as “A
house of ten to twenty Houses, most of which are normally located in a single
ward. All these houses relate to the kram-drum of a single House, which is apical
to the dol” (p. xiv). I came across no such reference to dol. People were not even
well-acquainted with the word, let alone with such a specific definition. Erik de
Maaker, Negotiating Life: Garo Death Rituals and the Transformation of Society
(Ph.D. thesis, Leiden University, 2006).
14. Burling, Strong Women of Modhupur, p. 16.
15. Khaleque also mentions the Dual. I never noticed that that group stands out in
a more significant way than such groups as Chibok or Babul that featured in
most of my interviews.
16. For more elaborate accounts of the Garo kinship system, see Burling, Strong
Women of Modhupur; De Maaker, Negotiating Life; Kibriaul Khaleque, “Garo
Kinship Terminology: Idealization and Reality”, The Dhaka University Studies 39
(1983): 153–80; Kibriaul Khaleque, “The Operational Use of Anthropological
Concepts of Garo Matrilineal Descent Groups”, Social Science Review 2, no. 1
(1985): 67–85; Kibriaul Khaleque, Social Change among the Garo; Dhirendra
Narayan Majumdar, A Study of Culture Change in Two Garo Villages of Meghalaya
(Gauhati: Gauhati University Department of Publications, 1980); Marak, Garo
Customary Law and Practices; Nakane, Garo and Khasi; Sangma, History and
Culture of the Garos.
17. Garos are not the only matrilineal grouping in the region. Khasis are also
matrilineal. Colonial documents mention a number of other social groupings
who have witnessed a transformation from matrilineal to partrilineal kinship
organizations. Bastin, for example, writes that there is evidence “that matrilineal
descent and inheritance were common to all the original Bodo tribes”. See
Bastin, Five Thanas of Partially Excluded Area, pp. 63–64.
19. The Internet version of the Banglapedia (The National Encyclopedia of
Bangladesh) mentions that “the rate of literacy in the Garo community is
higher among the women than among the men. The reason is the matriarchal
system. This makes it difficult for a girl to find a husband with equal standing”.
<http://banglapedia.search.com.bd/HT/G_0043.htm> (accessed on 5 May 2006).
20. Khaleque, “The Operational Use”, p. 83.
21. For a detailed study of how Garo death rituals allow for the re-organization of
social relationships among Sangsarek Garos, see De Maaker, Negotiation Life.
22. Bina Agarwal has elaborately studied how socio-economic and political
developments influence landrights and thereby kinship regulations. Bina Agarwal,
A Field of One’s Own. Gender and Land Rights in South Asia (Cambridge:
Cambridge University Press, 1994).
23. A third group is Momin, but in Bangladesh they are very few in number.
According to Khaleque, Momin comprise approximately 2 per cent of the total
Garo population of Bangladesh. Both Khaleque and Marak also mention Areng
and Sira. According to Khaleque, Bangladeshi Garos do not know where either
one of these two are living. I myself did come across Areng, but in those cases it
was used in the sense of ma’chong. Khaleque, “The Operational Use”, pp. 73–75;
Marak, Garo Customary Law and Practices, p. 23.
Garos Divided 109
24. Khaleque, “The Operational Use”, pp. 73–75; Marak, Garo Customary Law and
Practices, p. 23; Playfair, The Garos, p. 64.
25. Here relative refers to kinsmen from both mother’s and father’s side and can even
be used for people with whom actual blood ties are not known. Suborno told me
that all Bangladeshi Garos consider themselves related to each other.
26. Note, for example, the names of the Indian researchers Julius Marak and Milton
Sangma.
27. Burling found that such marriages are less seriously condemned in Bangladesh
than in India. Burling, Strong Women of Modhupur, p. 155.
28. Nakane, Garo and Khasi, pp. 23–27.
29. Ibid., p. 27.
30. Majumdar, Culture Change in Two Garo Villages, p. 19.
31. Bernard Cohn, “The Pasts of an Indian Village”, in Time. Histories and Ethnologies,
edited by Diane Owen Hughes and Thomas R. Trautman (Ann Arbor: The
University of Michigan Press, 1995), pp. 21–30.
32. Where Garos are concerned, the process of naming also reflects self-perception
and boundary demarcation. That this Garo informant has a Muslim name is
rather exceptional. Most elderly Garos have a (Hindu) Bengali or a Garo name.
If their parents were Christians, they could also have a Christian name. Sometimes
people simply make up names (like Aterson, Waterson). Today we can witness a
tendency to give Garo children typical Garo names again.
33. Literally, ambi means grandmother and achu grandfather. The terms are also
used to refer in a respectful manner to elderly Garos in general.
34. Eriksen calls this an analog system of classification. Eriksen, Ethnicity and
Nationalism, p. 66.
35. Literally this means “Have you Eaten Rice?” [mi = cooked rice]. It can also be
understood in a broader sense: “Have you had your meal already?” The question
is often used as a greeting.
36. There are no separate Garo (read: Abeng) words for language and dialect. If
another group was said to speak a different kushuk, this did not necessarily imply
that their language was unintelligible to the informant’s group.
38. De Vos, “Ethnic Pluralism”, p. 23.
39. Wana krita refers to a ceremony in which a sacrifice to the spirits is made.
40. Literally, Morol means village leader. Villagers also use it for rich and influential
Garos.
41. After the death of a married woman, her ma’chong is expected to provide the
widower with a new wife from the same ma’chong. These attempts are not always
successful. Although the widower could stay and take a wife from outside, I also
heard of widowers leaving their wife’s homestead, leaving the children behind. In
such cases, the children will be raised or even adopted by one of their mother’s
sisters or their grandmother. The husband is attached to the ma’chong through
marriage only, but his children belong to it by birth. Cf. Burling, Rengsanggri,
pp. 140–52; Burling, Strong Women of Modhupur, pp. 144–45.
42. This has become a rare practice since the introduction of Christianity. However,
as Monendra did not officially become Christian until 1958, he did not face the
limitations of Christian doctrine. Theirs was the only “one husband – two wives”
household I encountered.
43. “This distribution of mahari seems to have certain localized areas …. Therefore,
marriage relationships are usually to be found between mahari of Sangma and
those of Marak which are localized in a particular area …. Thus, they form a
localized area. This socio-geographical area coincides to some extent with the
differences in dialect and culture of the seven groups which I mentioned …. At
least those areas I know of show marked endogamous units with mahari
distributions fairly corresponding with the cultural and geographical
classification.” Nakane, Garo and Khasi, p. 24.
6
NEGOTIABLE BOUNDARIES,
NEGOTIABLE IDENTITIES
Research on group formation and social identities has tended to regard

groups as mutually exclusive in a digital way: either one is a member of X
or one is not…. it may, perhaps, be more appropriate to think of identity
in general as an analog phenomenon than as a digital one. Conceptualized
in this way, degrees of sameness and difference, of inclusion and exclusion,
may be identified.1
This chapter examines the social boundaries between Garos and others, and
how Garos relate to those who transcend these boundaries. Whereas the
previous two chapters explored social boundaries from a historical perspective,
in this chapter the focus is primarily on the present.
We have seen that “primordial” characteristics play an important role in
the self-definition of Garos. The only way to become a Garo is to be born
as one, so it was argued.2 This chapter shows that both self-perception and
membership are more pliable and changeable than primordialist discourse
suggests (see Chapter 3). Especially where new membership is concerned,
Garos show much leniency with regard to the criterion of birth. Such
notions of flexibility and permeability lead to the question of how to define
who is really Garo and who is not. Here I use the term Garo for everyone
who considers himself Garo, who considered himself Garo in the past, and
who has been accepted as Garo by others. That reality is far more complex
111
should become evident in the following pages. Through a focus on the

ethnic boundaries, degrees of sameness and difference, and of inclusion and
exclusion, will come to the fore, and I hope to show that it is more correct
to consider Garo identity as an analog phenomenon rather than as a digital
one, as suggested in the afore-mentioned citation. Questions that arise are
how, in the perception of Garos, a person can become a Garo, and when a
Garo stops being one. How do Garos relate to those on the other side of the
boundary, and how do they relate to people crossing these boundaries? In
other words, how do Garos maintain and create boundaries between Self
and Other?
BOUNDARY TRANSCENDENCE
AND MIXED MARRIAGES
Earlier I argued that among contemporary Garos ethnic sentiments are strong
and act as an important binding factor. The feelings of sameness are stimulated
by the fact that they form a small minority in a country overwhelmingly
populated by ethnic Bengalis. Garos often perceive their minority position as
a threat to their existence. Past experiences have taught them that they are, for
that reason, at risk of being dispersed (see also Chapter 8). This also explains
the negative attitude they have towards Garos who transgress the ethnic
boundaries and “disappear” from the Garo community. Those Garos threaten
not only the Garo image of a close-knit community, they also threaten the
very existence of the Garos as an ethnic group in Bangladesh and contribute
to the fear that Garos will eventually dissolve into the larger Bengali society.
Nihar and Lipi from Dhaka expressed it at follows:
Nihar: “We are a very small community in Bangladesh. When I think about
the large population in this country, I feel that one day it will be very
difficult to find us.”
Lipi: “We are just like a candle in the middle of a fire. We may melt any
time.”
However, notwithstanding social norms and ideals regarding boundary
transgression, Garo individuals do cross the ethnic boundaries and vanish
into other communities. Existing disincentives against boundary-crossing
behaviour can complicate, but certainly not obstruct, the disappearance of
Garos from the ethnic community.
Here boundary is used as a metaphor for the imaginary line that separates
one ethnic group from another. It is, in a Barthian manner, considered as the
social product of inter-group relations, which has variable importance and
Negotiable Boundaries, Negotiable Identities 113
which may change over time.3 Ethnic boundaries do not need to be territorial
boundaries and they do not isolate groups from each other. Instead, there is
a continuous flow of contact, information and even people across them, and
a focus on them provides particular insights into interethnic relations and
identity formation.4 Close scrutiny of ethnic boundaries gives insight into
how people define their ethnic identity, how they include and exclude, and
how important their ethnic identities (and therefore ethnic communities or
groups) really are.
It is quite difficult to estimate how often Garos cross the ethnic boundary
and vanish into another (neighbouring) community. Everyone knows one or
more such stories, but the phenomenon is not something Garos are very
proud of and eager to address. It happened more than once that, only some
time after an interview had been held, I would find out from other Garos that
the particular family had “lost” a member. People were willing to discuss the
matter, but if we left the subject untouched, so would they. As I repeatedly
pointed out before, unity and cooperation are highly valued, and in their
presentation to outsiders, Garos tend to stress the solidarity within their
community rather than discord.
Garos who transcended ethnic boundaries through inter-marriage are
generally referred to as people who left the Garo community. Interestingly
enough, (e)migration, which often implies moving away physically from
other Garos, is not necessarily perceived as boundary-crossing. Even though
affirmation and reaffirmation of identity are necessary requirements for social
identities to survive, it is not so much geographical mobility that worries
Garos.5 No one minds a Bangladeshi Garo migrating to the United States, or
to any foreign country for that matter, not even if the person migrates as a
consequence of inter-marriage. Such a person is not referred to as someone
who has left the community, but rather as a very successful Garo who
provides those who stay behind with a sense of pride (and, if lucky, with
remittances). No one worries about the ethnic identity of such a successful
Garo and his offspring. The Garo woman who married a Frenchman and
followed him to France was merely considered a fortunate person. Her
relatives proudly told me about her.
More commonly, however, mixed marriage symbolizes the real threat to
Garo identity, and close scrutiny of (perceptions of ) inter-marriage and, for
similar reasons also adoption, provide an excellent way to examine inter-
group relations and identity formation.6 Both inter-marriage and adoption
into another ethnic community stand for the most intimate or close
connections that can be established with that other community. Especially so
in a context where marriage is foremost a community matter and where birth
and blood (race) weigh heavily in matters of inclusion and exclusion in

determining group membership and the guarding of boundaries.
Most studies on inter-ethnic or inter-cultural marriages explore the social
and psychological consequences of such relationships for the partners and
investigate their acceptance by the larger society.7 Pivotal in this chapter are
not the inter-ethnic marriages themselves but the ethnic boundaries that
people transcend when they marry outside their community. I shall not focus
on issues of tolerance or racism, or the well-being of the couple, but on the
perceptions of and reactions to mixed marriages, and on the ways in which
such matches are also legitimized by the couples involved, close kin, friends,
and the larger community. Those reflections and responses reveal much about
ethnic sentiments, boundary maintenance and how people distinguish the
Self from the Other. At the same time the marriages themselves (and ensuing
legitimizations) show that ethnic boundaries and identities are more open to
negotiation than Garos would like (others) to know.
Hence, in order to understand the phenomenon of mixed marriage, we
need to distinguish between community ideology and individual ambitions
and needs. Compared to their Bengali neighbours, Garos have more individual
freedom. This is especially notable for Garo women. This freedom is also
reflected in marriage matters. Garo parents and other (maternal) relatives
have an important say in the choice of spouse, but even the arranged marriage
of an heiress (nokna) and her partner will not take place without her consent.
At present, love matches are becoming more and more frequent.8 With an
ever-increasing number of Garos settling in the cities, at a distance from
parents and other influential relatives, and living, working and studying more
closely with members of other ethnic communities, the number of inter-
marriages is likely to increase. At the same time, powerful sanctions to
prevent mixed marriages (such as excommunication or disinheritance) are
limited. Moreover, if applied, they only seem to be practised temporarily. I
noticed that family connections are often re-established in the long run. Note
the following reaction of a Garo father to his daughter’s secret love marriage
with a Bengali. One of this girl’s sisters told the story:
“My father was very angry. He wanted to bring her back, but all people said
that it was too late. Many people gathered in our house to discuss the
matter. All relatives came. Some of them blamed my parents. They said that
they should have accepted an earlier marriage candidate who was a Garo.
My parents could not do anything, however, because Rahim [the Bengali
lover] was a very powerful person in Mymensingh. Many Garos knew him
because they had stayed in his hotel. Everyone suggested my parents to
avoid trouble with Rahim. My sister was afraid to visit our house. She came
after a long time, together with her husband. But that was only after my
mother had paid them a visit in Mymensingh.”
Although Garo women have relatively more freedom to decide about their
future spouse than Bengali women, gender does play an important role in
the assessment of (inter)marriages. Women, much more than men, are
considered the prime reproducers of the ethnic boundaries and their marriage
to someone of a different ethnic background is more often and more
strongly disapproved.9 Garo women are assigned a different role not only
because they are the ones who give birth, but also because of the matrilineal
kinship organization, which formally implies that women rather than men
pass on family titles, kinship connections, and property to their offspring.
In this light it is not surprising that the most widely used argument against
Garo women marrying (Muslim) Bengali men is that these men are really
only after the property of Garo women.
Marriages within the Garo group are also subject to an extensive set of
rules and suitable marriage partners are not only determined by kinship.
Factors such as status, social class, level of education, and denominational
(church) background are important criteria as well. A historical approach
reveals that the standards by which marriage candidates are judged have
changed.
Next I shall discuss six different mixed marriages. Not all inter-marriages
are perceived as equally threatening to the Garo community, and feelings of
resentment and disapproval really come with a few types only. We will see
how not only ethnicity influences the acceptability of a mixed marriage but
that other yardsticks, such as gender, religion, class, level of education, and
social status, also determine the suitability of a prospective marriage candidate.
None of these criteria are fixed or timeless, and all should be perceived as the
outcome of the (continuously changing) reflections on Self and Other. We
will see that, when it all boils down to real-life matters and daily choices, the
distinction between Self and Other is not quite as fixed and unchangeable as
dominant discourse suggests.
NEGOTIABLE BOUNDARIES, NEGOTIABLE IDENTITIES:

SIX MIXED MARRIAGES
1. TAMIKA CHIRAN MARRIES A BENGALI MUSLIM
Highest on the list of unwanted matches is the marriage of a Garo woman
with a Bengali Muslim man. Garo-Bengali marriage generates the strongest
resentment in the Garo community and is clearly considered the least desirable
option of all inter-marriages. It is not only the risk that these people run of
losing contact with their community, it is the fact that they marry into the
Bengali Muslim community of Bangladesh. Or in the words of a 30-year-old
Garo and village high school teacher:
“For us [the Garo-Bengali Muslim marriage] is very bad because when they
take someone away after marriage they don’t allow the person to maintain
a relationship. They become totally separated from our society. The ones
who marry a Bengali have no feelings for other Garos. According to our
tradition, we maintain a relationship with our parents and relatives after
marriage.”
One example of such union is the marriage of Jasimah Khanon. Her story
symbolizes the worst possible match: that of a Christian Garo woman with a
Muslim Bengali man. In many ways Jasimah’s life story embodies the idea of
what many Garos believe will happen when a Garo woman marries a Bengali
Muslim. Jasimah left her village in 1967 to attend a nursing training programme
in Mirzapur. There she met her future husband, a Bengali Muslim who
already had a Bengali Muslim wife. They got married and she converted to
Islam and took a new name. With the help of a private teacher, she studied
the Koran for five or six months. Since 1971, she has been living in Dhaka.
One day, Suborno and another Garo friend accompanied me when I visited
Jasimah in her apartment that she shares with her son. Her husband’s first
wife lives in a different place. Jasimah’s husband earns a good salary as a trade
union leader and she herself has a nice position in one of the big five-star
hotels of Dhaka. The family is obviously well-to-do.
She received us in a rather formal and distant manner, quite unlike my
experience with most Garo families I visited in Dhaka. I found it more
difficult than usual to establish a sense of mutual trust, and for that reason,
I could not ask all questions I would have liked to ask. At first glance, Jasimah
appeared well adapted to the Bengali way of life, wearing a sari and, more
conspicuously, a small nasal trinket. This type of jewellery is very common for
Bengali Hindu or Muslim women but is seldom worn by Christian or
Sangsarek Garos. It clearly functions as a (gendered) boundary marker.10
Jasimah told us how she had changed her Garo name (Tamika Chiran) for a
Bengali Muslim name. The conversation took place in Bengali. According to
Jasimah, she is still able to understand the Garo language but can no longer
speak it. She does not maintain any contact with the Garo community in
Dhaka but does stay in touch with her relatives. While her younger sister was
studying in college in Dhaka, she stayed with her. About her identity, Jasimah
told us that she feels Garo when she is visiting the village, but that she feels
Bengali in Dhaka.
In many ways, Jasimah exemplifies what Garos fear about inter-marriage

with Bengali Muslims: her seemingly perfect assimilation into the Muslim
Bengali community. Although Jasimah still understands the Garo language,
she can no longer speak it. She stays in touch with her relatives, but seeks no
contact with other Garos. When she is in Dhaka, she feels Bengali. To
complete this “fearful image”, her son grew up in an entirely Bengali way.
Garos consider him Bengali.
2. ‘ONLY BAD WOMEN MARRY BENGALIS’

Aversion to Garo-Bengali marriage is not a new phenomenon. But neither is
the fact that such marriages occurred in the past. Rumeli Patang (80 years
old) also married a Bengali Muslim. And this is where most similarities with
Jasimah’s marriage stop. Rumeli married her husband in 1947. She continued
to live in her own village where she was joined by her husband, who died a
few years ago. She is still living in the same house, together with her daughter
and son and their spouses and children.
Rumeli refers to her husband as a Bangal pi’sa, the child of a Bengali.
From the age of six or seven, he had been living in a Garo family where he was
raised as a son. Rumeli related the following story about him:
“The people called him Garo-Bangal. He was a Bangal pi’sa. They lived in
Kandapara village with the Mandis. Near Kandapara, there were many tanio
Bangals.11 They were there from the beginning. But the Bangals didn’t eat
in the Mandi houses. They maintained a good relationship with them, but
they didn’t touch any food of the Mandis. They didn’t even smoke the same
pipe. The father [of my husband] was a gambler. He had been very rich but
soon became poor. One of his sons was staying in a [Garo] house all the
time. He ate there, slept there. His father asked him to come back home
many times but he didn’t. The Bangals told him to get his son back by any
means. It was not nice that a Bangal was living with a Mandi family. His
father felt ashamed and became very angry. One day he came with a knife
and threatened his son. If he would not return, he would kill him. Still [my
husband] didn’t return, so Nadang Morol (Rangsa) [head of this Garo
household] told the father that they would keep him as their son and that
he should not kill him.”
It is generally believed that in those days, Garos and Bengalis lived more or
less peacefully together. At the same time, the relations between the two
communities were hierarchical and boundaries were strictly maintained.12
This is clearly illustrated by the fact that Bengalis did not wish to eat food
from the Garos or even share a smoking pipe with them. Often Garos told me
that they felt looked down upon by the Bengalis.
The father of Rumeli’s husband was extremely embarrassed about one of

his sons spending all his time with the Garos and forced him to return. The
story of the alleged killing is more likely to have been a rumour or a threat
than an actual intention, and Rumeli too does not believe the Bengali father
would have eventually gone as far as to kill his son. Nevertheless, it certainly
reflects how the whole case was taken very seriously. In the end, the Bengali
father gave in to the request of the Garo family to let the boy stay with them.
They would raise him as their own.
And so it happened. From then onwards, the Bengali boy Mumtaz Kabir
used the name Dhirendra Rangsa, a Garo name. The fact that this family was
childless probably accounted for their wish to raise the boy as their son.
Adoption of a child (even a Bengali) was and still is common in such cases.
This may also explain why a grand feast was demanded, for a feast was
generally required to announce and formalize an adoption. Rumeli reported
how her husband became “just like a Garo”; how he spoke the language
fluently, how he participated in the village meetings on behalf of the Rangsa
family, and how he knew the mantras of the midi amua.13
When the boy was old enough, he married a Garo girl, but she soon
died. He later remarried a girl from a village in the Garo Hills. After her
sudden death, Rumeli’s future husband started to roam around. That’s
when her parents met him. The story goes that they liked him because he
was known as very hardworking, and they decided to arrange a marriage
between him and their daughter. At that time Rumeli was only between ten
and twelve years old. Rumeli continued that “After I got married, many
people criticized me for this, but it was arranged by my parents, so I had no
choice. I didn’t manage to avoid it. The people used to tell me that only bad
women marry Bengalis.”
This reveals the complexity of the situation. Although her own family
made the deal, neighbouring Garos were not quite so approving of the match.
Rumeli gives the impression that she herself was not all too happy about it,
considering that she explained how she was left out of the decision-making
process. She herself was negative about inter-marriage. When we inquired
about the possibility of inter-marriage between Garos and other neighbouring
non-Bengali groups, she replied strongly: “No, that was not possible! They
were a different jat. They only married in their jat, and we in ours. Only now
people are becoming shameless and they take whomever they can get.”14
Rumeli’s account illustrates how ethnic boundaries are to a certain extent
negotiable. It also shows the possibility of flexible and individualized
interpretations and legitimization of those ethnic boundaries. While Rumeli
seems conscious of the fact that her marriage conflicted with commonly held
Garo norms and values, in the same breath, she points out that she considers
marriage outside the community as “shameless”.15 Despite here own marriage
to a man of Bengali descent, she does not rate herself among the shameless
women and she discusses inter-marriage as if it were a recent phenomenon.
This seems to indicate that Rumeli and her relatives did not consider her
husband, who was brought up by Garo parents, a real Bengali. Rumeli’s
husband had not only lived in a Garo family, he had been raised by them as
their own son. As an adoptive Garo, he was even in a position to marry three
Garo women one after the other. Interesting in Rumeli’s case is not only the
apparent gap between actual and ideal conduct in relation to boundary
maintenance, but also her complex, and at times contradictory, way to
negotiate the ethnic boundary and identity of the Garos.
3. ‘DANGEROUS LIAISONS’
Asim (27 years old), a Garo rickshaw puller, is living with the Bengali woman
Deepali (20 years old) and their one-and-a-half year old daughter in a very
small house with walls constructed of mud and bamboo and a corrugated tin
roof, in a slum-like area on the outskirts of one of the richest parts of Dhaka
city. Their marriage probably represents one of the worst possible marriages
for Bengali Muslims. This is not only because a Garo is upojati, or tribal, but
also because the law of the Koran does not allow Muslim women to marry
non-Muslims.16
In 1991, Asim came to Dhaka to look for work. He soon exchanged his
position as a guard for a job in a garment factory in Savar, where he met
Deepali. Deepali had come to Savar as a very young girl, and although her
mother remained in their village in Sripur thana, Deepali never went back
there even once and hardly has any memories of her village. Her sister who
was working as a bank employee in Savar, took Deepali to live with her.
Deepali found a job in the garment factory where Asim was working. This is
where they met and later decided to get married; the starting point of a very
troublesome relationship.
Asim and Deepali, whose name used to be Khadiya Begum, faced strong
resistance ever since they decided to get married. They are living together and
have a daughter, but they were not officially married in either a court or a
church. Particularly Deepali’s relatives and others from the Bengali Muslim
community strongly opposed their wish to get married. When they suspected
their sister’s involvement with Asim, Deepali’s sister and brother arranged a
marriage to a Bengali known as a smuggler of women, who would take
Deepali abroad.17 Deepali asked Asim to find her a hiding place. He found
her a place in a hostel for girls but was soon discovered by her relatives who
then informed the police:
“One day, the police came and took me to the police station. They told me
that if I would become Christian and marry a Christian, they would send
me to jail. They asked me to write down that I would never become a
Christian and to put my signature there. I did, and they let me go. But my
brother took me to Tongi where I have another brother. There they threatened
that they would kill me if I would become Christian. They didn’t allow me
to go out alone. I lived like this for a few weeks. One day – it was Eid – they
attended the prayer service. No one stayed at home so I fled and came back
to Savar to meet him [Asim]. The same day we left for the village.”
Their story does not end here. Asim and Deepali wanted to get married in
Asim’s village in a Christian ceremony. In order to do so, Deepali had to
convert. Deepali said how she felt accepted by the villagers, how she was able
to move freely around the village without being hassled, and how Asim’s
parents received them well. Deepali was asked to take a Garo name, which she
did. She took the name of her father-in-law’s ma’chong. Nevertheless, the
parish priest refused to convert her. Some people believed that this was
because Deepali was not yet sufficiently acquainted with the Christian religion
and needed some religious training first. Asim and Deepali themselves believed
that the priest had acted out of fear of retaliation from the local Bengali
Muslims. As there was no work for them in the village, the couple returned
to Dhaka where Asim took up a job as a rickshaw puller and Deepali found
work in a small handicrafts workshop.
“The problem was how to marry her. Everybody was afraid that if the
Bangals [of our thana] would come to know about it, it would become very
terrible. They would tell the police. I told her that, if any Bangal was to ask,
she had to tell them that she is a Christian. But her nose is pierced so it was
difficult to hide it.”
This story presents a mirror image of Jasimah’s marriage. This time the couple
is poor and has little status and no power. It is not the Garo community that
objects to the marriage, but the Bengali. Both partners suffer from intimidation
from the Muslim Bengali community. It is not the Garo who adapts to the
Bengali community, but the Bengali who adapts to the Garos.
Most of this account was based on the first of two interviews. That
particular interview was conducted by Suborno. I accompanied him the
second time and found the couple serving chu (Garo rice beer) to two Garo
neighbours. The couple told us that Deepali’s relatives had tracked them
down again, and that even her mother had come from the village to visit
them. They had started their preparations for a Christian wedding. Their
fear seemed to have diminished a great deal. They were likely to get married
after all, and relatives from both sides had, more or less, come to terms with
the inevitable.
4. NEETI MARRIES A BENGALI HINDU

Neeti (40 years old) married a Bengali Hindu. She wed him at her own
initiative, some seventeen years ago, and took his caste name. Together with
their two daughters, she is currently living in a spacious modern apartment in
Dhaka. Two or three times a month her husband joins them for a few days.
She and her two younger brothers lived with her in-laws for many years. Her
husband makes a very good living as a general practitioner and out of the
factory which he and his relatives own. This family is obviously well-to-do.
Because Neeti and her husband wished their daughters to receive a proper
education and found no opportunity for this in their own town, they decided
it would be better for the daughters to live in Dhaka. Neeti supervises their
educational progress meticulously.18
Neeti met her husband during her nursing training. She relates how they
initially faced opposition to the marriage from her relatives, but that they
gave in when her husband converted to Christianity. “They didn’t like it. My
mother was very strong, but finally she accepted it. Then we got married in
the village. It was a Garo Christian marriage. About twenty-five people from
my husband’s side attended the wedding as well.”
An acquaintance of Neeti’s family later commented that she and her
husband had faced more opposition against their marriage than she told us.
Apparently, her mother had reacted so strongly against it that the couple had
decided to marry in court. Sometime later, the husband was asked by her
relatives to convert to Christianity. Only when he agreed could the couple
marry (again) in the village in “the Garo Christian way”. Evidently, Garos
who marry outside their community are well aware of the general opinions
held against such a match. From Neeti’s story, we can deduce several ways
inter-marriage is justified by the persons concerned, the next of kin, and the
Garo community at large. In the case of Neeti’s marriage, her husband’s
conversion seems to have broken the resistance.
Despite this, however, Neeti did marry into a Bengali Hindu family, and
although her husband converted to Christianity, his being a Christian does
not play an important role in the way she justifies the marriage. Instead she
stresses the point that marrying a Bengali Hindu is not quite the same as
marrying a Muslim and she argues that “the Hindus are more liberal towards
the Christians and the Muslims are more conservative about religion”. The
feeling that it is better to marry a Hindu than a Muslim is shared by many
Garos. In Neeti’s case, it is obviously a manner to justify her own marriage.
She does not disregard differences between Bengali Hindus and Garos. Initially,
she described these in terms of food habits. About the dissimilarity of mentality,
she commented that her in-laws are alright, but that “others are not so good.
Some of them are good, but some of them are bad. They have the caste
system, that’s not good.”
Garos are generally very disapproving of the caste system and its in-
built inequality. This might also have to do with the fact that not all that
long ago, Garos were the only non-Bengalis in Mymensingh who had not
converted to Hinduism, and who were consequently looked down upon by
all their neighbours.19
The characterization of her husband as a liberal and friendly person is
also interesting. She feels that his mentality is very good and points out that
“he is just like a Garo”. Questions which come to mind are how her
marriage to a non-Garo affects Neeti’s Garo identity, and perhaps even
more importantly, how it affects the self-perception of her daughters. Neeti
has no plans to ever return to her own village but will either continue to live
in Dhaka or move back to her husband’s town. But she thinks of herself as
a Garo and keeps in touch with her relatives in the village as well as with the
Garo community in Dhaka. Some twenty-five Garos are employed in their
factory. She certainly feels she has always been able to continue to “live like
a Garo”. Neeti also indicates that her husband gets along fine with Garos
(“he likes the Garos”). Perhaps most interesting is Neeti’s stand on the
future marriages of her daughters. She has already started looking for
suitable Garo boys. According to her, she can only accept candidates from
her own jati. The following interview fragment reveals that this consideration
is not rooted only in ethnic sentiments:
“My younger daughter is very nice. I have already decided that I will never
let her go with someone from another jati. I want to keep her in my own
house…. Actually, the most important thing is that we are spending so
much money on the education of our daughters. We don’t mind. But if they
will marry a ruri [Bengali],20 they will go away, so it will be a big loss for us.”
Neeti and her husband have no sons. According to Hindu tradition, explains
Neeti, her daughters would have to live with their future husbands. Who
would, in this case, take care of her and her husband when they are old?21 It
is hard to assess whether Neeti’s strong opposition against a Bengali candidate
should only be explained by rational considerations. Neeti certainly ventilated
some critical remarks about Bengalis, and there is no doubt that she herself
feels close to the Garos. Neeti, undoubtedly, identifies herself as a Garo. She
believes, however, that her daughters see themselves as Bengalis. That their
mother has a too simplistic idea about her daughters’ self-perception became
clear during an interview with Supta, the eldest daughter.
Neeti’s daughters were born and raised in their father’s extended family.
When Supta was 10 years old, they moved to Dhaka for the purpose of
further education. Both girls are now studying in a very renowned Christian
school and college. They have frequent contact with the Garo community in
Dhaka, particularly with their own relatives. They also visit their mother’s
village at times. Although Supta claims to be able to understand the Garo
language, she cannot speak it. She explains that there was never a necessity.
Everybody around her speaks Bengali and even the Garos in the village do.
With the exception of some friends among her relatives in Dhaka, all her
friends are Bengali. When Supta was asked to describe herself in ethnic terms,
she answered: “I think I am both. I think I am Garo and Bengali as well. I do
not care what others think of me. I never say that I am Garo or Bengali. I
always say that I am ‘mixed’.”
Supta has an independent outlook on life. She is against arranged marriage
and the second-rate position of women in Bengali society. She thinks that
women are merely considered a commodity, and she strongly protests against
their life-long dependency upon their fathers and husbands. She pictures
herself in future as a financially independent working woman who shares all
domestic work with her husband. This kind of thinking may have been
encouraged by discussions held at school, but the position of Garo women
certainly serves as a positive example. Supta refers to the Garo social system
as a positive example and does so in opposition to her “own society and its
practices” [“amar nijer shomaje”], by which she means the more liberal strata
of Bengali society. For instance, she believes it is a good thing that Garo
children receive their mother’s name. After all, it is the mother who gives
birth to the child. When Supta is asked why she did not take her mother’s
name, she answers that: “It’s our [Bengali] social system, we can’t take our
mother’s name.” A little later, however, she summarizes her status and position
as follows: “My father is a Bengali Hindu and my mother is a Garo. I can’t
deny either of them, so I say I am ‘mixed’.”
5. NOBODY MINDS ‘A MARRIAGE OF FLAT NOSES’

An example of a mixed marriage between a Garo Christian man and a non-
Bengali Christian woman is that of Johnson Raksam and Minoti Ching, a
Khyang from the Chittagong Hill Tracts, an area in southeastern Bangladesh,

quite far from Garo territory. Johnson (46 years old) married Minoti (about
40 years old) in 1978. Minoti had long been Christian when they met. Her
parents converted when she was young. The couple met in a hospital in the
Chittagong Hill Tracts, where they both received nursing training. In contrast
to the cases described above, their marriage did not generate any opposition
from either the Garo or the Khyang side. Neither close relatives nor others
opposed their personal decision to marry. The general Garo attitude towards
this type of intermarriage differs greatly from the ones described earlier.
Johnson explains this as follows:
“It doesn’t matter among the tribals. Nobody minds a marriage of flat noses.
Among Garos, Chakmas, Khyangs, marriage is alright…. I think there is
some relation between tribals. When we talk to someone who is a tribal, we
feel very close to each other.”
Even though, as Rumeli told us, half a century ago Garos were not to marry
members of neighbouring non-Bengali communities, marriages like these are
much better received today. Often the rationale is that Garos share their
mentality and looks with those other minorities. Johnson’s explanation is thus
more than an individual attempt to justify his choice to marry Minoti. His
explanation reflects the general opinion of contemporary Garos. That there is
much less resentment also has to do with the fact that both spouses are
Christians. This is also what Minoti told us: “I have been Christian since I
was young. That’s why there was no problem about the marriage.” Even
though both Johnson and his wife indicated that there are certain differences
between the Garo and Khyang communities (“they [Khyang] don’t behave
like Garos do”), they emphasize the similarities. They particularly stress
similar food habits and mentality; both Garos and Khyang eat dried fish, and
all are simple, open-minded people.
Another significant aspect which attributes to the acknowledgement of
their spousal tie is their close physical resemblance. Minoti relates how she
feels completely accepted in her husband’s village, and that some of the
villagers cannot see she is not Garo. In Dhaka, too, where the couple has been
living throughout their marriage, visitors often mistake her for a Garo.
Like many Garo migrants in Dhaka, the couple maintains close relations
with the Garo community. They also visit the husband’s village on a regular
basis. Contact with the Khyang community is much more limited, although
relatives do pay visits and stay with them. This may partly be explained by the
fact that there is no Khyang community in Dhaka and that Minoti’s village
is quite far from Dhaka. But it also seems that in this family, the emphasis is
on Garo identity. Johnson, who is very outspoken about his identity as a

Garo, plays a dominant role: “I think that I am a Garo because I was born as
a Garo. By birth I am a Garo. Christianity came from the West. I am a
Christian now, but I am not an American. I am not British. I am a Garo.
Garo is Garo.”
Minoti speaks of her identity in a slightly different way. Even though
neither one of them conceals the fact that she is Khyang, her self-perception
is more complex: “As a wife, I am Garo, but I am Khyang…. Sometimes I
forget that I am a Khyang. When someone talks to me in Garo, I try to
answer in Garo…. I feel that the Garos are also my relatives. They are my own
people, too.”
At home the family speaks Bengali, not only because this is their lingua
franca, but also because of the parents’ philosophy about the upbringing of
their two sons. Johnson stresses the importance of their proficiency in Bengali.
He had observed that Garo students with insufficient knowledge of the
language face communication problems with their Bengali classmates. Whereas
Minoti claims to be able to speak her husband’s dialect or language (Atong)
very well, Johnson does not the know Khyang language. Evidently, Johnson
does not view the Khyang language as a realistic option. This becomes even
more evident when he continues to tell about his sons’ upbringing:
“We raise them as Garos, and they also like to introduce themselves as
Garos. They are very proud to say that they are Garo. If someone calls them
Chakma or Khyang, they get angry with them…. Sometimes in school,
they get in trouble with other students because they are called ‘Chakma
Chakma’. Once, my elder son got very angry and hit them.”
Minoti adds that: “They don’t deny that their mother is a Khyang, but they
like to say that they are Garos; they are taking their father’s name.” Both
parents clearly believe that their sons have opted for the Garo identity. A talk
with the eldest son, Joel, gives an interesting picture. He makes a clear
distinction between his private feelings and his public presentations: “I am
both…. I am Garo from my father’s side and from my mother’s side I am
Khyang. There is no other way to say it than to say it like this.”
However, when Bengalis ask Joel about his background, he tells them he
is Garo. Nevertheless, when it comes down to it, Joel prefers to call himself
Christian: “Especially when they ask me from which jati I am I say that I am
Christian.” Joel was born in Dhaka and like Neeti’s daughter Supta, he is well
acquainted with Bengali society. All but one of his friends in Dhaka are
Bengali. Yet, no matter how well Joel is adjusted to Bengali society in daily
life, he has a strong awareness of the minority problems in Bangladesh. This
also influences his self-perception. He feels that he has to be very careful and
cannot always speak his mind. He talks about Bangladesh as the country of
the Bengalis. “It’s their country, so if they promise you something, or take
something from you, they don’t bother to return it. They think we are simple,
that we don’t have any power, so they aren’t concerned about us.”
A major difference with Supta is that both Joel’s parents belong to non-
Bengali minorities. And whereas Neeti’s daughters cannot be told apart from
Bengali girls, Joel’s appearance clearly reveals his non-Bengali tribal background.
He has been confronted with his “tribal looks” more than once. Like many
Garo students, he has frequently been asked about his food habits and he has
a funny story to tell about that: “They ask us if we eat snakes and frogs. I reply
that my Christian religion allows me to eat all those things. I also tell them
that snake is very expensive in Thailand and frog legs are even more expensive
in America. Only rich people can afford it.”
The story of Johnson, Minoti, and their two sons shows that the general
perception on inter-marriage is more positive when a tribal match is concerned.
This is particularly the case when matches between “look-alikes” are concerned:
“flat nose, flat nose, no problem”. This case study also shows the under-
communication of differences between the two ethnic communities and the
highlighting of similarities. The question remains how this match would have
been judged if Johnson had adapted to Minoti’s background, or had she been
Buddhist or Hindu. All evidence seems to suggest that the argument “flat
nose + flat nose = no problem”, would have weighed heavily in the assessment
of all such cases. In their particular marriage, however, it was not Johnson
who crossed the ethnic boundary, but Minoti, and the Garos as a group did
not suffer a loss with this marriage. Instead it is Minoti who stands with one
foot in the Garo community and the other in the Khyang community.
6. ‘ADVENTISTS WED ADVENTISTS’

All previous examples illustrate that Garos ought to marry Garos, and that,
when an inter-ethnic marriage takes place at all, it better not be with a
Bengali. Even if the Bengali spouse is a woman and/or a Christian, the
general attitude towards this type of marriage is one of disapproval. “A
Bengali is a Bengali” is what many people told me. The final case is an
exception to this rule: in the Seventh Day Adventist community of Bangladesh,
religion is judged as the most important criterion. “Adventists should only
wed Adventists.” For the Adventist Garos of Dhaka the eligibility of a
marriage candidate depends first and foremost on religious background. The
spouse-to-be must belong to the Adventist church. This means that there are
no actual restrictions or sanctions when it comes to marrying a Bengali

Adventist. Herein lies the basic difference between Adventist Garos and
others when it comes to view on inter-marriage with Bengalis. In theory,
Adventist Garos do not go against the rules if they marry a Bengali, as long
as he or she is a member of the Adventist church as well. We should keep in
mind that this may apply to Garo members of the Adventist church, but does
not change the idea of other, non-Adventist Garos about inter-marriage
between Adventist Garos with Bengalis.
Semanti (26 years old), Sumila (20 years old) and Preeti (16 years old)
share a tiny two-room apartment with their parents in Dhaka. Their father,
Monil Chisak (47 years old), works as a night guard and their mother,
Preetika Areng (40 years old), as a domestic servant for an expatriate family.
They are members of the Seventh Day Adventist church.
To the question of how they feel about inter-ethnic marriages, Preeti, the
youngest daughter, answered: “Nothing, we can marry any Adventist. It
doesn’t matter if he is Bengali or Garo.” Preeti’s response confirms the idea
that, for Adventist Garos, the importance of the marriage candidate to be
Adventist exceeds the importance of his or her being Garo. However, when it
comes to individual preferences, the situation is more complicated than can
be derived from Preeti’s reaction. Perhaps there are no sanctions against a
Garo-Bengali match, but this does not mean that in practice there are no
preferences for a Garo as marriage partner. Semanti, Preeti’s eldest sister,
explains her ideas about marriage:
“I will look for a Garo, but in our church there are few of us [Garos]. I will
need an Adventist Garo. But among the Adventist Garos, there is none of
my age who is fit for me, so if I want to marry a Garo, I will have to find
someone who isn’t Adventist but he must be converted. That’s also very
difficult.”
Semanti’s response indicates why the attitude of Adventist Garos towards a
Garo-Bengali match is much more lenient. That her future partner has to be
Adventist drastically limits the number of suitable Garo candidates. After all,
the Garo Adventist community is very small.22 A Garo with a different
denominational background would do, as long as he would be willing to
convert. But, as Semanti pointed out, that is certainly very difficult. Few
Garos are willing to join the Seventh Day Adventist church. Such change
would require a great many sacrifices.23
Chapter 7, which deals with the role that Christianity plays in the lives
of the Garos, demonstrates that denominational divisions influence how
Garos are organized and how they relate to one another. For none of the
denominations is religion more important than the ethnic background.

Everyone is first and foremost Garo. The Seventh Day Adventists are the one
exception to this rule.
CONCLUSIONS
Research on group formation and social identities has tended to regard
groups as mutually exclusive in a digital way: either one is a member of X or
one is not. Through a focus on social boundaries and on the perceptions that
Garos hold of those who transcend those boundaries, it is revealed that there
is no such thing as an unambiguously defined Self and one homogenous
Other. Instead there is a whole range of many Others, some closer to the Self
than others. Moreover, the whole set of rules and regulations, which should
prevent individuals from crossing ethnic boundaries is subject to negotiation
and manipulation, and it is possible to discern many degrees of sameness and
difference, and of inclusion and exclusion.
A case study of inter-marriage or mixed marriage provides an excellent
opportunity to examine (perceptions on) boundaries and boundary
transcendence. Like adoption, marriage into another community symbolizes
the most intimate connection to that community. Garos, who constitute a
tiny minority in a Bengali dominated society, often express their fear to “melt
like a candle in the fire”. Inter-marriage (into another community) is therefore
strongly opposed. Social reality however shows that boundaries are much less
fixed and impermeable than dominant ideology about boundary transcendence
suggests. Not even the strong ethnic identity of contemporary Garos and the
fear that they may dissolve in the large community of Bengalis one day can
prevent individuals from making their own choices.
This chapter also revealed that the acceptability of inter-ethnic marriage
depends on a complex constellation of variables and that the far most important
variable, ethnicity, is cross-cut by other criteria such as religion, social status,
and gender. A Garo should marry a Garo, especially in the case of a woman.
It is better to marry another “flat nose” than a Bengali, better a Hindu than
a Muslim, better a Christian than a Buddhist. Moreover, these criteria are
weighed differently by different people. Seventh Day Adventists form the
only exception to the rule that ethnicity is the most significant criterion.
Their first priority is religion. However, even in this small sub-group of
Garos, individual variation exists.
Opportunities to manipulate boundaries and identities are not endless.
Not only Garos but also their neighbours have clear ideas about inter-
marriage. Garo choices can be severely limited and complicated by norms,
values, believe systems, culture, and priorities held by others. Others, non-
Garos, may set much stronger limits to the boundary-crossing actions of
Garos. The case of Deepali and Asim, who had to face a strong opposition
from the Bengali Muslim side, made this painfully clear.
Inter-marriage initially influences the two spouses, their positions in
(Garo) society, their self-perceptions and the way they are perceived by
relatives, friends, and others. However, the implications for children born out
of such a union are much more far reaching, especially in relation to their self-
perceptions and identities. None of the Garo spouses (completely) lost a sense
of Garo-ness and their inter-ethnic marriage has not affected their own ethnic
identity to the extent that they no longer feel Garo at all. It is the self-
perception of children from such unions which is much more complex. The
question is whether the increasing freedom amongst Garos to choose ones
own life partner leads to more inter-ethnic marriages, and ultimately to the
end of ethnicity as it is still prevalent in Bangladesh.
NOTES
2. On the ideological level, the Garos discourse about boundary transgression is
very similar to Parsi discussions on the same issue. T.M. Luhrman, The Good
Parsi. The Fate of a Colonial Elite in a Postcolonial Society (Delhi, etc.: Oxford
3. Barth, Ethnic Groups and Boundaries.
4. Cf. Eriksen, Ethnicity and Nationalism, pp. 38–39.
5. Cf. Peterson Royce, Ethnic Identity, p. 7.
6. Mixed marriage originally referred to the marriage between a Catholic and a
Protestant. See Augustin Barbara, “Mixed Marriages. Some Key Questions”,
International Migration 32 (1994): 580, n.3. At present mixed marriages include
“inter-cultural” or “inter-ethnic marriage” as well. Here I make no distinction
between these different terms. Cf. Dienke Hondius, Gemengde huwelijken,
gemengde gevoelens. Aanvaarding en ontwijking van etnisch en religieus verschil
since 1945 (Den Haag: Sdu Uitgevers, 1999), pp. 5–6.
7. Cf. Yasmin Alibhai-Brown and Anne Montague, The Colour of Love. Mixed Race
Relationships (London: Virago, 1992); Susan Benson, Ambiguous Ethnicity.
Interracial Families in London (Cambridge: Cambridge University Press, 1981);
Hondius, Gemengde huwelijken, gemengde gevoelens; Walton R. Johnson and
D. Michael Warren, Inside the Mixed Marriage: Accounts of Changing Attitudes,
Patterns and Perceptions of Cross-Cultural and Interracial Marriages (Lanham:
University Press of America, 1993).
8. A love match or “love marriage” should not be understood as it is understood
in the Western culture. It primarily indicates that individuals choose their
own partners. After such a choice is made public, the decision is not easily
cancelled.
9. For a theoretical exposé about the relation between women, ethnicity (and
nation), and state, see Flora Anthias and Nira Yuval-Davis, “Introduction”, in
ibid., Women-Nation-State (Houndmills, etc.: Macmillan, 1989), pp. 1–16.
10. Cf. Jitka Kotalová, who describes how among Bengalis Muslims the piercing of
the girl’s nostril for the insertion of a nasal trinket constitutes the last major
beautifying operation for girls. This is an essential element of the socialization
process of girls. Jitka Kotalová, Belonging to Others. Cultural Construction of
Womanhood in a village in Bangladesh (Dhaka: University Press Limited, 1996),
p. 72.
11. Sthaniyo is Bengali for local. Garos only use the word to refer to local Bengalis
who have always lived in the area. Local pronunciations are tanio or taina.
12. Cf. Bastin, Five Thanas of Partially Excluded Area, p. 40.
13. Midi Amua, or mite amua is part of the traditional Garo Sangsarek tradition. It
means worshipping the supernatural beings (which includes gods and lesser
spirits). Burling, Rengsanggri, pp. 54–55.
14. See Chapter 5 for an explanation of jat.
15. Abeng, the Garo language spoken by most Bangladeshi Garos, has two different
words to indicate “shameless”: kracha.aja (kracha.a = shame) and migilgrak. The
first is generally used to refer to women who behave in an “unwomanly” way
(that is, as defined by Garos, for example, to whistle or to climb trees). However,
kracha.aja does not have a negative connotation as such and it is often used for
friendly, unconventional, modern women. These days, kracha.aja has a very
positive meaning. Garos believe that in order to be successful in life, they need
to be kracha.aja. Rumeli used migilgrak to denote shameless. Migilgrak literally
refers to people who do not lower their eyes after committing any immorality.
They do not show any repentance for their misconduct. Garos consider them
shameless in a very negative sense of the word. It can be applied to thieves,
prostitutes, cheaters, adulteresses. Sometimes, as Rumeli did, it is also applied to
women who are married to a Bengali. It is interesting to note that these days the
latter may also be referred to as kracha.aja. This appears particularly the case in
cities, where inter-marriage has become more common.
16. According to the law of the Koran, Muslim men can marry non-Muslim women
(provided they belong to a religion of the Book: Christian or Jewish), but
Muslim women must marry Muslims (in a mixed marriage, this implies conversion
of the spouse to Islam).
17. This man would marry a young woman, take her to India, and then leave her
after some time.
18. Neeti said that there is no Garo mother who takes as much care of her daughters
as she does. According to her, the reason Garo mothers in Dhaka do not concern
themselves enough with the education of their children is that “the Mandis don’t
mix with Bengalis so they don’t know how to bring up their children.” Neeti’s
close involvement with her daughters’ education strongly resembles the situation
which was described by Thérèse Blanchet in the part of her book where she
depicts Dhaka and Savar middle-class families and their preoccupation with the
educational accomplishments of their children. For a more elaborate discussion
on this subject, see Thérèse Blanchet, Lost Innocence, Stolen Childhood (Dhaka:
University Press Limited, 1996), Chapter 9.
19. This was not only said by elderly villagers, but also mentioned by Bastin, who
based himself on a gradation given by Stapleton. According to Stapleton, tribes
of northern Mymensingh were ordered in the following way, ranking from top
to bottom: Rajbansis, Meches, Rabhas, Hajongs, Hodis, Koches, Dalus, and
Garos. Bastin, Five Thanas of Partially Excluded Area, p. 64. It is interesting to
note that this situation has reversed completely. These days Christian Garos are
generally looked up to by neighbouring Hindu (and even poor Muslim)
neighbours in the villages.
20. The Garo word ruri is used to refer to both Hindu, Muslim and Christian
Bengalis.
21. It should be noted that although it is the norm among Bengalis in Bangladesh
for women to join the household of their fathers-in-law after marriage, there are
instances in which men join their brides’ parents’ households (usually when she
does not have any brothers). Bengalis as well as Garos refer to this type of
marriage as ghor jamai (which literally means “son-in-law of the house”). However,
ghor jamai is very unpopular among Bengali men for fear of losing face and
position. This explains why Bengalis look down upon the Garo marriage system
in which it is very common for a husband to join the household of his parents-
in-law. It is for this very reason that among young educated Garo men, there is
growing resistance towards ghor jamai these days. For the Bengali ghor jamai
system, see for instance, Sarah C. White, Arguing with the Crocodile. Gender and
Class in Bangladesh (Dhaka: University Press Limited, 1992), pp. 104 and 107.
22. According to another Adventist Garo informant, some 2,500 Garos are Seventh
Day Adventists in Bangladesh.
23. Seventh Day Adventists are not allowed to drink alcoholic beverages, tea, and
coffee, or to eat food like pork (which is a favourite dish of Garos).
7
GAROS AND CHRISTIANITY
“I would say that because of Christianity, we manage to remain separate

from the majority group of this country.”
This is how Semion, a student at Dhaka University, summarized the
significance of Christianity for contemporary Garos in Bangladesh.
Christianity does not only provide an important ethnic marker for Christian
Garos, who comprise at least ninety per cent of the Garos, but in many
ways also for Sangsarek and Hindu Garos.1 Other than this, it has a major
impact on their socio-economic, psychological, cultural, and political
situation; it influences their outlook on life and on the world, provides
them with a sense of belonging to a world that extends far beyond the
borders of Bangladesh, offers new educational and professional opportunities,
and regulates relationships within their society and with others. This chapter
explores the relation between (conversion to) Christianity and the (ethnic)
identity of Garos. The central question is how Christianity has influenced
their self-perception, sense of identity and group formation. How have
missionaries, the Christian religion, and Christian institutions and
organizations affected the (ethnic) identity of the Garos of Bangladesh?
The main purpose is not to write up yet another church history or
missionary success story, or to present an elaborate theory about the
susceptibility of Garos to Christianity, but to examine the social dimensions
of conversion to Christianity vis-à-vis the process of ethnogenesis. The
chapter is broadly divided into two sections. The first section deals with
132
Garos and Christianity 133
conversion. How and why has Christianity come to play such a prominent
role for the Garos? What does the process of conversion, for example, tell
us about how Garos have related to one another? The second section
concentrates on the relation between the church and the organization of
Garos. Here I touch on the relation between Garo identity and Christianity,
and on how Christianity has influenced the self-perception of Garos and
their ethnic identity. The central question is how Christianity has
(re)organized the social, economic and political life of the Garos. First I
shall briefly discuss how I define some relevant concepts to the chapter such
as Christianity, conversion, and religious identity.
CONVERSION AND IDENTITY

In order to understand the importance of Christianity for the Garos, we
first need to define Christianity and being Christian as a social phenomenon
rather than an intellectual answer to a changing social environment.2 This
does not mean that Christianity has no intellectual bearing on individual
Garo Christians. The point is, however, that being Christian is not simply
(or only) a matter of (intellectual) belief. It is also an essential aspect of
identity and self-perception of Garos, of their social organization, and of
their economic and political situation.3 Consequently, conversion, too,
requires a broad definition. Here the term is used to denote the process in
which Garos have embraced Christianity as their own religion, whether in
a religious or spiritual way, or in name only. Conversion is thus understood
as the ongoing process in which individuals or whole groups denounce one
religion and label themselves with a new one. For the Garos of Bangladesh,
this process has not completed.
Conversion can both be an individual and a social experience and it has
had a different relevance and different meaning for different persons in
different times. Suborno and I encountered Garo Christians who held on to
their Christian belief firmly, and others, who called themselves Christians
but who were still Sangsareks at heart. Some Garo Christians were third-
generation Christians, whereas others had only recently converted.4 Some
were Christians because Christianity was the “only true religion”, whereas
others only converted because they did not want to stay alone. The Garo
process of conversion has undoubtedly been “multi-causal” rather than
“mono-causal”.5 It has meant different things to different people and has
had many divergent social consequences.6
Semion indicated that Christianity is important for Garo ethnicity because
it acts as a significant boundary marker between Garos and others. The question
remains which came first: a shared Garo identity or Christianity? Was
Christianization the incentive to ethnic unification, or did ethnic sentiments

contribute to the conscious decision to convert to Christianity? Or were both
processes instigated by other developments? We shall see that for the Garos of
Bengal, this whole process of Christianization, boundary demarcation, and
ethnic unification has been complex and difficult to put one’s finger on. It is
particularly difficult to delineate the precise direction of causal relations. Different
individuals converted at different moments for different reasons. Initially it
could lead to exclusion from one’s own community, while later on, inclusion in
the ethnic community seemed the prime incentive to convert.
THE MISSIONARIES
Although different Christian churches entered the scene, here I mainly
concentrate on the two dominant denominations: the Baptists and the Roman
Catholics. These days, the Roman Catholics clearly outnumber the other
denominations. In the early 1990s, 67 per cent of the 67,576 Christian Garos
were Catholics, 19 per cent belonged to the Baptist Church (non-baptised
members included),7 and the remaining Garos were mostly members of the
Church of Bangladesh (Anglicans), the third largest Christian denomination
among the Christian Garos. 8 Catholics were not always the largest
denomination. During the first decades of missionary history, the Baptists
were the most influential. Although they still hold that position in the Indian
Garo Hills, in Bangladesh they lost it to the Roman Catholics.9
These denominations disagreed not only on matters of theology and
dogma, the national and cultural backgrounds of the missionaries also varied
to a large extent. Baptist missionaries arrived fresh from Australia, whereas
the Catholic mission constituted a potpourri of nationalities (French, American,
Italian, Polish, Belgian, Indian), who first received training elsewhere in
Bengal before they commenced their work with the Garos.
The two denominations were also differently organized. The Catholics,
with their archbishop residing in Dhaka, were much more hierarchically
structured than the Australian Baptists. All Catholic clergy were accountable
to the archbishop, the head of the church in East Bengal. The Australian
Baptists, who formed a flat organization, maintained friendly relations with
the British and American Baptist missionaries, but had their working fields
clearly demarcated. Australian missionaries paid occasional visits to their
American colleagues in the Garo Hills, but they did not interfere in each
other’s works. Hereafter I shall argue that the different Christian options
accounted for a greater appeal of Christianity to the Garos of Bengal than one
single denomination would have. These differences enhanced the space for
Garos to negotiate, and different denominations offered different opportunities.

If a convert was not pleased by, or got expelled from one church, he could
easily move to another.
GAROS AND CONVERSION

Although the subject of Christianization in South Asia has been explored in
several recent anthropological studies, historians have largely concentrated on
surface events, such as the beginnings of Christian missions, the foundation
of churches, and the number of converts.10 Such has also been the case for
Bengal and for Northeast India, where Christian missionaries have been
particularly successful.11 Most accounts deal with church history and
descriptions of missionary ventures. They do not explore the larger context in
which the process of Christianization has taken place, nor do they pay any
attention to matters of identity and group formation.12
Hence, with my initial aspirations to understand the susceptibility of
Garos to Christianity, I soon found myself drowning in a vast body of
literature about conversion. Scholars of conversion have offered numerous
explanations for the success of Christian missionaries all over the (colonial
and post-colonial) world. Some authors have conducted elaborate studies
that reveal the complexity of the matter.13 Nevertheless, for Bengal and
Northeast India, the subject is often approached from a rather one-sided
perspective. Available studies focus on the missionaries (their message and
their methods) or on the context (poverty, colonial subjugation), and overlook
the converts themselves. Such investigations approach converts as objects of
conversion rather than as active participants in the process.
This is also the case for the Christianization of the Garos. Although the
introduction of Christianity among the hill Garos has received a lot of
attention, and a number of authors have indeed attempted to explain the
susceptibility of these Garos to Christianity, none of these researchers have
approached the subject from a Garo perspective.14 Regarding the Garos of
East Bengal, their process of conversion has received even less attention, and
no comprehensive historical analysis of that process has been put forward.15
The bulk of primary and secondary publications are by missionaries and
clergy themselves. These reports, newsletters, articles, and monographs16
approach the subject from a missionary viewpoint. They concentrate on
missionary successes and hardships, and reveal the strategies they employed
and the image they held of the Garos.
It was, however, the encounter of two fragmented groups, Garos and
missionaries, that has resulted in the Christianization of the large majority of
Garos. There were the Garos, who, at the arrival of the first missionaries,
constituted a heterogeneous collection of different segments – some closely
related and others more distant – and there were the missionaries with
different denominational backgrounds, who came from different countries
(and cultural backgrounds) at different moments. Today we know that the
encounter has resulted in the Christianization of the majority of Garos: the
missionaries somehow succeeded in converting the Garos in large numbers.
At the time of their arrival, however, their success was evidently not ensured.
The very presence of missionaries could not explain their successes. These
were not the result of a simple, unilineal, and cumulative process, but the
outcome of fragmented Garo responses to fragmented missionary propositions
and offers, varying in time and place. In other words, the large-scale conversion
of Garos was the result of a complex, fragmented process in which missionaries
and their objectives – the Garos and their goals, needs, and considerations –
and the larger context in which this encounter occurred all played a role.
Thus, we need to distinguish different variables: missionaries (their message,
methods, behaviour, personality), the Garos (their reasons for conversion),
and the context (socio-economic, cultural, and political). I will now discuss
that process of conversion mainly from a Garo perspective, but first I briefly
discuss the early history of the missionaries in ‘Garoland’.
MISSIONARIES IN ‘GAROLAND’
There is a romantic tale about the start of missionary involvement with the
Garos of Bengal. Although the exact date of that first conversion is controversial,
the late 1860s or early 1870s mark the beginning of a very successful
missionary enterprise.
… to Durgapur in 1867 came three strangers. They too were soldiers, but
their errand was one of peace. One was a foreigner, a Swiss, by name
Ruprecht Bion, better known as the “Apostle of East Bengal.” … Two
Bengalis, Raj Kumar Tarka Bhusan and Gonga Charan, accompanied
Mr Bion. They came to Durgapur. And they went away. But they came
again, and one of them, Gonga Charan, was destined to play an important
part in the life of the Garo Christian Church. He had not as long to wait
for converts as had Carey. The exact date is not known, but everything
points to 1868, just one year later, that Radha Nath Bhomik, the first
Plains’ Garo, was won for Christ.17
Different missionary societies had been active in Bengal since the sixteenth
century. Their early history is one marked by a coming and going of individual
missionaries and missionary orders, who met little success and much hardship.18
The Bengali population showed little interest in their Christian teachings.

Even in 1993, the Christian community of Bangladesh counted only some
275,000 souls (0.3 per cent of the total population).19
In the second half of the nineteenth century, “[a] vast new field was
being opened in Garoland, the only place where there was hope of making
an appreciable number of converts.” 20 Missionaries jumped at the
opportunity to win souls. It was a Swiss missionary working for the British
Baptists, Ruprecht Bion, who initiated their work in northern Mymensingh.21
Soon after the opening up of the Garo field, however, Australian Baptist
missionaries took over from the British Baptists, who continued to
concentrate on Dacca district. In 1892, the first Mission House of Bengal’s
Garoland in Birisiri, near Durgapur, was opened by Reverend P.C. Nall and
his wife Mrs Annie Nall.22
Less than twenty years later, the Roman Catholics appeared on the scene.
In 1910, the Catholic Father Adolphe Français started to work amongst the
Garos, and in 1912 he established the first Catholic church in Tausalpara,
near Ranikhong, only some eight kilometres away from Birisiri.23 A very
similar pattern of mission work can be observed for Haluaghat. The Australian
Baptists started a mission and the Catholics soon set up theirs nearby. In
1898, the Baptists founded a church in Rangrapara, which has since developed
into a prosperous Baptist stronghold. Biroidakuni about three kilometres
from Rangrapara came to serve as an outstation of Ranikhong Catholic
Mission in 1917. After a decade, the Catholics established a resident mission
here. This station has developed into a flourishing mission too. The Baptists
and Catholics continued to dominate the Garo mission field. Both
denominations spread their wings all over the Garo area. Wherever you find
one church or mission, you will surely find the other denomination close by.
In 1901, the Anglican church (or Oxford church) founded a mission in
Haluaghat bazaar. Their church has always remained markedly smaller than
the other two denominations. Although the arrival of the Oxford mission in
Garoland did cause some distress among the Baptists, it was particularly the
relationship between the Baptists and Catholics that was marked by tension
and resentment. Hereafter I concentrate on these latter denominations.24
WINNING SOULS
The Garo pattern of conversion is complex and by no means a one-way
process. Different Garos converted for different reasons under different
circumstances at different moments. Some Garos accepted their new faith,
while others resisted it. Some Garos became devout Christians and gave up
many aspects of their earlier life styles. Others were not willing to give up
these same practices (such as participating in Sangsarek festivities or drinking
rice beer or chu). Different villagers provided different reasons for (resisting)
conversion. Today it is still not difficult to find extended families that consist
of both Sangsarek and Christian members.
The first Garo converts left no personal accounts. It is not easy to tell
what motivated them to join one of the Christian churches. Oral history and
missionary records suggest that many Garos (also) converted for material
reasons. Even missionaries who often prided themselves on having made
converts “of the heart” refer to such conversions. In 1921, the Catholic
missionary Matthew Kearns commented for example that Garos were not
drawn to the mission naturally, but that “[t]hey, like all pagans, look for
material advancement”.25 Such converts are often referred to as rice Christians.
Yet economic betterment alone cannot explain the missionary successes.
For example, it was not all that easy to convert to Christianity. Converts
could not only take, they also had to give. In 1899, the Bengali Baptist
missionary Babu Joy Nath Chaudhuri wrote that
The increase of Church members by baptism is not very large this year.
There are two chief reasons for this: One is that during these two years we
have often and often taught the Christians to give for the support of
teachers and preachers, and for other causes. So the heathen Garos see that
they cannot expect pecuniary gain by joining the Christian community. But
this cause of offence is good for the Church of God, as it is one of the means
of keeping away those who are not real believers. The other cause is our
inability to supply teachers to those places where people are inclined to
Christianity. The increase of the number of school teachers every year is too
small to cope with our demands.26
Moreover, conversion to Christianity had a great bearing on the social and
cultural lifestyles of the converts, who for instance had to distance themselves
from “heathen” practices (all rituals and festivities that were related to the
Sangsarek tradition) and to abstain from drinking chu or rice beer (the latter
requirement only applied to Baptists).27 Converts had to live Christian lives;
they had to pray regularly, marry in the church, and stay away from Sangsarek
festivities or rituals.28
Conversion was by no means an irreversible transformation. Sangsareks
opted for Christianity, but many Christians took to Sangsarek practices again.
Converts who could not resist the lure of Garo chu (rice beer) or Sangsarek
festivities and rituals were generally expelled from the church. Source material
reveals that many Garos converted to Christianity, but soon gave up and
returned to their earlier lifestyle. Joy Nath Chaudhuri commented that
last year many were baptized who did not repent, but were under temporary
excitement. These persons, whose hearts, of course, we could not know,
went back. Then, some fall back into the temptation of drinking. As the
Garos have no caste system, sometimes young men fall away by marrying
heathen wives, according to heathen custom.29
These days, many Christian Garos still forsake so-called Christian rules and
regulations. By “falling back into temptation”, they run the risk of losing their
church membership. The minutes of Rangrapara’s Baptist church reveal that,
in the 1990s, wrongdoing (such as drinking, Sangsarek marriages, or adultery)
was still punished by eviction from the church.30
One more important factor that must have complicated conversion was
the attitude of the Sangsareks. Early Christian converts often met with
hostility and anger from other Garos. In 1917, the Garo pastor of Birisiri
Church, Babu Kanai Sangma, said that “the Garos don’t easily become
Christians as they have to give up their drinking and their heathenish
amusements and often have to suffer persecution.”31 The antagonism between
Sangsareks and Christians lingered on far into the twentieth century. An
American priest who settled among the Garos of Modhupur in 1959 recalled
how Sangsarek Garos hindered his work and threatened to poison the
missionaries and their assistants if preaching was continued. And one Garo
interviewee related that “Many of the first converts became Christian in spite
of many difficulties. Society and relatives created many obstacles, and when
they said that they wanted to become Christian, they were threatened. There
are many examples of people who, after becoming Christian, were thrown
out by the Sangsareks.”32
Conversion also complicated contacts between Sangsareks and Christians
in other ways. Cecilia and her sister Renuka, members of a well-known third
generation Baptist family, recall how they used to be afraid of the Sangsareks
of nearby Bibalgree. Cecilia told us that:
“Our parents taught us not to enter their village because there were many
mimangs [spirits]. In every celebration, there would be a mimang. That is
why we were afraid. We did not go to their village when we were young.
Even in their drums there would be mimangs. Before they played, they had
to worship the drums. When we noticed that they were going to play, we
would flee. We thought the drums would bite us.”
It is interesting to note how these third-generation Christians continued to be
afraid of Sangsarek mimangs. Their Christian religion forbade them from
having anything to do with the mimangs, but could not take away the belief
in their existence. The interview fragment also shows how the religious divide
influenced social interaction between Christians and Sangsareks.
Nevertheless, compared to the many problems Bengali Hindus had to

face when they converted to Christianity, the consequences for Christian
Garo converts seem limited. Converted Hindus risked being thrown out
from their family and caste. We can read that missionaries often attributed
their successes with Garos to the absence of the caste system. In 1911,
Reverend Barnett argued that “having no organized religion, no zenana
system, and no caste or social prejudices, they presented, as a people,
unlimited scope for missionary enterprise.”33 Although missionaries and
Christian Garos did have to face anger and hostility, they were never
entirely expelled from their community. Sangsareks and Christians did not
end all contact. They continued to interact, except in situations that were
related to religious matters.
Missionary methods and attitudes also attributed to the successes of
Christian missionaries with Garos. Missionaries worked hard to win the
Garos for their religion and applied many methods to spread the gospel. They
set up schools and dispensaries and organized rice banks (collective saving
systems) to safeguard Garo peasants from the hands of moneylenders who
demanded exorbitant interest on each loan. Moreover, Garos were fascinated
by the fact that most missionaries were (Western) foreigners, who, also with
their disciplined and hard work, filled the Garos with awe. For example, they
had to travel widely (on foot, on horseback, and later sometimes by car). One
Garo woman explained why the missionaries impressed the Garos:
“Since they were foreigners, all Mandis were interested in them. They also
behaved very well with the people. They served them and took care of them.
Some people became close to them and were baptised into the religion of
the foreigners. After that, these Mandis preached among their own people.”34
Another villager told us that the missionaries’ car was the very first one he and
other villagers had ever seen. He related how “even the old men and women
who could not walk were carried there”. We can only imagine how impressive
that arrival must have been. Another elderly Garo woman told us how the
missionaries went about in her neighbourhood:
“At first they came to Biroi. They built a house and started to live there.
Very often they went into the villages just to meet the Mandis. Then they
started to preach. They told stories from the Bible, about Jiso Christo. In
the beginning, one or two people became Christians. Later these people
tried to convert others. They tried to increase their numbers. The missionaries
showed pictures of Jiso Christo on a big piece of cloth, magic. Seeing these
pictures, the people easily believed and became Christians. The missionaries
worked hard to convert the people. There were no roads, no cars, no
bicycles. They came here by horse, even the women. Sometimes they came
walking. The Baptists came from Birisiri.”
Christian converts themselves became actively involved in missionary activities.
Garo villagers told us how these first disciples took their Christian religion
much more seriously than Christian Garos do these days. One woman said
that “[t]he Christians of that time were very eager. They themselves did all the
work in the church.” Another informant remembered how her grandmother
prayed both in the morning and in the afternoon. She added that “this we do
not see anymore.”
The complexity of Garo conversion is also mirrored by its pattern. Garo
conversion to Christianity has been both an individual and a collective
phenomenon. The choice of Radha Nath Bowmick (the very first Christian
Garo of Bengal) was probably based on an individual decision. Villagers’
stories and missionary accounts presented other examples of similar individual
conversions. More often, however, people converted in groups, village, or
paras [village neighbourhoods]. The American priest who has been working
in Modhupur for some forty years asserted that “The people converted in
groups, one village after the other. Their personal opinions had no importance.
It was like a movement, really.”35
Garo informants explained that missionaries often tried to influence local
leaders first. If these leaders decided to convert to Christianity, other Garos
would follow, not out of conviction but because they were told to do so. Surjo
Gagra told Suborno that
“In the beginning, they [missionaries] went for the Garo leaders and other
influential people. When they had reached them, they told about the
Christian faith, gave examples and said: ‘You cannot go to heaven. And
your evil spirits, they are not God, the trees are not God, the sun and the
moon are not God, to do snake puja, all that is not God.’ That is what they
said. They said that there is only one God who is all, and in all, and all this
is his creation. (…) Among the leaders and influential people, they formed
preachers.”36
The Baptist Garo Reverend Sunil Dio (more than 60 years old) was involved
in such missionary activities. He remembered how Sangsareks converted
group by group and told us how groups of 200 to 400 Sangsareks converted
in one go. When the villagers of Bibalgree decided to convert midway
through the Pakistan era, they also did so collectively.
Nowhere did I find any evidence that Garo conversions which took place
during the first half of the twentieth century were politically motivated.
Today, however, religion is clearly politicized. In a context where politics,
religion, and ethnicity are closely linked, this does not come as a surprise.
During the Gulf War of 1991, Muslim Bengalis would yell slogans such as
“Muslim Hindu bhai bhai, Christiander rokto chai chai (Muslims and Hindus
are brothers, we want the blood of Christians).” But when Pakistan and India
face off in a cricket competition, Christians are replaced by Hindus in the
same slogan. The Garo man Poben Patang told us the following:
“During the ‘Atdam Satdam’ War [first Gulf War], Bengalis wanted to make
trouble with the Mandis. They shouted against the Christians. It was great
that the Christians defeated the Muslims. Otherwise, we could not have
stayed here. We would have had to flee. Because of this, they [Muslim
Bengalis] are dancing no longer.”
For contemporary Garos, Christianity stands for Western culture and influence,
which they consider a positive attribute. People feel that Christianity provides
them a link with Europe and the United States. The increasing politicization
and “ethnicization” of religion has clearly influenced Garo decisions to convert
to Christianity. This is also reflected by the conversion of the Garos of
Bibalgree. Poben Patang and many other inhabitants of Bibalgree converted
after the communal riots and the mass refuge of Garos in 1964.37 Until then
they had resisted conversion collectively. Moloya Toju explained the villagers’
decision to baptize as follows: “After 1964, we found that our community [of
Sangsareks] was becoming very small. All people were becoming Christians.
For that reason, we decided to convert to Christianity as well.”
Conversion to Christianity was motivated by various factors – ranging
from spiritual conviction and economic development to social and political
motivations. In the second half of the twentieth century, social, organizational,
and ethnic reasons became increasingly important. Later in this chapter, I
explain how Christianity has influenced the social organization of Garos and
how it is related to their ethnic identity. There is one question that requires
attention here. Why have the Garos, unlike all their non-Bengali neighbours,
not opted for Hinduism instead of Christianity?
WHY THE GAROS DID NOT BECOME HINDUS

Although the Garos never converted to Hinduism collectively, they were
seriously influenced by the social (caste) restrictions which were practised by
their Hindu(ized) neighbours. Ideas of purity and pollution had a serious
impact on the relation between different non-Bengali peoples in the region,
such as the Hodies, Hajongs, Koches, Banais, and Dalus. For example, at the
time of R.W. Bastin’s report, the boundaries between the Garos and Hajongs,
one of the neighbouring non-Bengali communities, were well demarcated.38
Garos and Hajongs, who maintained friendly contacts, lived in separate

villages. Garo villagers told us many stories about Hajongs looking down
upon them, about Hajongs feeling superior and maintaining the necessary
social distance. Hajongs would never dine with Garos or accept food from
their hands. The Garo villager Nurudin told us about the Hajongs: “Their
religion was very strong, and they were very faithful to their religion. They
were also very proud. That is why they did not allow marriage with Mandis.
They considered us chonna jat [low jat].”
At present, Garos are the only non-Bengali ethnic group of northern
Mymensingh who have not converted to Hinduism en masse. Both the
census of 1931 and Bastin’s report referred to the Garos as the lowest of all
tribes.39 Nevertheless, this tribal order, which was based on the Hindu social
and religious system, would be reversed if other criteria had been used. After
all, “[n]o race or tribe of those mentioned above has made the same progress
in education and independence as the Garos.”40
At present we can witness a situation that is favourable to the Garos. In
terms of education, political influence, and the socio-economic situation,
they are faring better than any of their Hindu neighbours. Whereas some
five decades ago Hajongs would not take food from the hands of Garos or
allow Garos to touch their tube well, these days Garos are the most influential
and prosperous non-Bengali minority in northern Mymensingh. In
Haluaghat thana, minorities such as the Hodies, Koches, and Banais are
scattered. They are not organized like the Garos, and the families that we
visited around Bibalgree appeared poorer that their Garo neighbours. In the
next chapter, we shall see how the Hajongs had to flee the country en masse
shortly after partition.
It is therefore not surprising that one Hodie family from Biroidakuni
village decided to convert to Christianity. That they, in that process of
conversion, adopted Garo ma’chong names suggests that these Hodies converted
for social rather than (purely) religious reasons.
I was long puzzled by the question why Garos did not opt for Hinduism,
like all their non-Bengali neighbours, who had been living side by side with
Bengal’s Garos for many centuries.41 Although the 1921 population census
listed Garos partly as Hindus and partly as Animists, and the 1931 census
reported all Garos as Hindus, no other evidence supports these findings.42
Several authors provide different reasons. For example, according to the
Baptist missionary C.D. Baldwin, Hinduism never made much headway
among the Garos, as it would have involved a complete break with the Garo
matrilineal organization.43 However, if a break with such matrilineal
inheritance practices and kinship organization prevented Garos from
Hinduization, why did it not prevent others, such as the Koch, from
converting to Hinduism?44
Porter attributes the general lack of interest of Garos to convert to
Hinduism to the low percentage of literacy among their Hinduized neighbours.
“Most likely for this reason they have not been so open to the proselytizing
influence of the educated Hindus.”45
Burling offers yet another explanation. He argues that whereas other
minorities or plains tribes did not reside in well-defined territories but lived
in villages scattered among villages of Garos and Bengalis, Garos had a
separate territory (the Garo Hills) at their disposal. They have always been
oriented toward the hills, where the population is “pure Garo”, as the focus of
their own culture, and are therefore generally the least “plains-like”.46 This
cannot, however, explain the Hinduization of small groups of Garos, both in
northern Mymensingh and in Modhupur, in the early decades of the previous
century. In an interview with three Hindu Garos from Modhupur, we were
told that Garos converted to Hinduism before Christian missionaries set foot
in Modhupur. According to these Garos, there are still some 85 Hindu Garo
families in the area, totalling some 600 to 650 people.47 According to our
informants, there used to be many more before, but many Hindu Garos left
for Assam and others joined one of the Christian churches.
The 1931 census referred to recent attempts by Hindus to proselytize the
Garos, without much success. Here it is reported that a few Garos indeed
professed to be Hindus but reverted to old habits and merely adopted the
name of some spirit which they used for their traditional worship.48 These
early-twentieth century attempts to Hinduize Garos were not the first. In
fact, already in the early nineteenth century, Buchanan reported that a
process of Hinduization of Garos had long been going on when he visited
them.49 He wrote, for example, that “[m]any of the Garos have been in some
measure converted; but they are very apt to revert to their impure habits.”50
His reference to “Garo slaves” is also interesting. He wrote that
These are chiefly Garos, who had once been converted to the worship of
Vishnu; but who have lost caste, owing to their inability to restrain their
monstrous appetite for beef, and who are sent back among their impure
countrymen as a punishment for their transgression. The number I believe
is pretty considerable.51
The idea that Garos were never attracted to Hinduism in great numbers is not
in line with historical accounts. Garo attraction to Hinduism was in fact
reported as one of the reasons why Christian missionaries should be sent to
the Garos. In 1825, for example, W.R. Bayley, secretary to the government,
wrote that “nothing permanently good can be obtained by other means, and
that if we do not interfere on behalf of the poor Garos they will soon become
Hindoos or half-Hindoos, retaining and acquiring many of the bad parts of
both their present and improved creeds.”52
This leaves us with the question of what happened to all the people who
successfully converted to Hinduism and did not revert to their old “impure
habits”. In this respect it is interesting to read Bastin’s reference to the Koches
of the Assam province. He wrote that “[t]he name in Assam is no longer that
of a tribe, but rather of a Hindu caste into which all converts to Hinduism
from the different tribes Kachari, Garo, Hajang, Lalung, Mikir, etc., are
admitted on conversion.” And about the Pani Koches of northern
Mymensingh, he mentioned a “modern trend of opinion” which claims that
“these Koches are cognates of the Garos who have drifted some way towards
Hinduism.”53 In his description of the different non-Bengali peoples of the
northernmost part of Mymensingh, Bastin mentioned, for example, that the
Dalus, who were Hindus at the time, were really considered Garos by Garos:
“Very little is known about them but Garos assert that they are really Ruga
Garos. They are said to have ma’chongs corresponding with the Garo ma’chongs
but I have been unable to verify this …. They now adopt Bengalee dress and
refuse to eat with Garos.”54
Such findings suggest the creation of new social boundaries as a result of
Hinduization. Perhaps the very question of why Garos did not convert to
Hinduism is wrongly put and should we, on the basis of these observations
and my own findings about fuzzy identities and fluid boundaries, draw the
conclusion that contemporary Garos constitute of those people who never
(permanently) converted to Hinduism, rather than an ethnic group to whom
Hinduism (and perhaps even Islam) was never appealing?
At present the situation is rather lucid. Although social boundaries
between Garos and others are more permeable and negotiable than Garos
would like to admit, contemporary Garos do form a clearly distinct
grouping in Bangladesh. They try to maintain ethnic boundaries, and
Christianity has offered them a significant boundary marker in a country
overwhelmingly dominated by Muslim Bengalis. When I asked Bindu
Patang from Bibalgree why Garos did not become either Hindus or
Muslims, he answered strongly: “Oh! that is not possible, not possible!
[laughing] The Mandis did not want to become Hindus or Muslims!”
And another villager replied to the same question:
“How is it possible? The Hindus were very proud. They used to tell us that
we were a very low caste, that we were jungly and undisciplined people.
They would never accept us in their society. The Garos didn’t like the
Muslims either and there were not many in our area. They didn’t have any
interest in converting the Garos.”
ORGANIZING GAROS
Every present-day Garo (whether Christian, Sangsarek, or Hindu) has Christian
Garo relatives, neighbours, or friends. School-going Garo children attend
Christian schools, and it is widely accepted that these children become
Christians if they are not already. Christian NGOs, such as Caritas and World
Vision, play a significant role in the social and economic development of
Garos. They provide jobs, development projects, credit facilities, loans,
sponsorship programmes, scholarships, and so on. Many Garos would like a
job in one of these organizations. Christianity also provides Garos with a link
to the international community. It offers them contacts beyond the borders
of Bangladesh. If you want to get something done among the Garos today –
whether it is to start a development project, to build a community centre, or
to organize some cultural festivities – it is almost impossible to bypass
Christian leaders or organizations.
Here I argue that Christianization and the unification of Garos went
hand in hand. According to Burling, for example,
The older religion lacked any organized group of leaders who had a vested
interest in its preservation. There was no institutionalized structure of
power that could resist outside pressures. All men could perform sacrifices,
but none had a central position from which the value of the older religion
could be defended.55
It was this organizational void that the missionaries and their institutions
filled, thereby also contributing to the unification of the Garos. Christianity
offered an institutionalized power structure, which the traditional religion
lacked, and the missionaries stimulated the process of unification by
introducing an organizational basis, novel forms of leadership, a lingua franca
(Abeng), and so on. Moreover, Christianity offered Garos alternative ways to
organize, manifest, and express themselves outside the dominant political
arena of subsequently colonial, Pakistani, and Bangladeshi politics. Moreover,
as I shall explain hereafter, particularly the sub-division into a variety of
Christian denominations stimulated the process of unification of the lowland
Garos. Hence, while the missionaries are often blamed for disintegrating the
Garos, there are many reasons to suggest that the introduction of Christianity
contributed largely to the organization of one ethnic Garo community.
UNITY THROUGH DIVISION

Although it appears to be a paradox at first sight, Christian missionaries did
instigate unification by introducing new divisions. The availability of different
churches provided Garos with a negotiating position and offered dissenters
the option to join other churches. The following quotations from oral and
written accounts substantiate the argument that missionary competition and
denominational division contributed to the successful Christianization and
thereby ethnic unification of Garos. From the very start of the Catholic
mission in Garoland, competition and strife developed between the Baptists
and Catholics.56 Both Baptist and Catholic documents contain endless
complaints about the other denomination. Even the two (contesting) founding
stories of the Catholic church are based on that very competition. In 1910,
the Baptist missionaries of Birisiri referred to the arrival of the Catholics in
the following manner:
At the invitation of some few of our Garos who have gotten themselves into
disgrace with the Police through a theft case, two Roman Catholic Priests
have come to a village three miles from Birisiri …. Pray that our Christians
will remain true in face of this new trial?57
The Catholics have a different tale to tell. They claim that Garos invited them
to settle in their areas, out of disappointment with the Baptists, and as they
were interested in the “true doctrine”:
For years Protestant missionaries had been laboring among them, giving
them for the most part a decidedly unfavorable view of the Catholic
Church. Some of the neophytes, not fully satisfied, determined to investigate
for themselves. They walked forty miles to the nearest railway station,
bought tickets for Dacca, and there sought out the “Great Father” of the
Catholics. Coming at last into the presence of Bishop Hurth, they told him
of their desire to know more of the Catholic religion. “We are Garos,” they
told him. “We come from the hills in the north of Mymensingh. We have
been Baptists for many years, but full of doubts on the subject of our new
religion, as we are indignant at the conduct of our Padri, who deserted us
during the plague. We have come to Dacca hoping to find the true doctrine
taught by Jesus Christ.”58
Similar stories abound about the foundation of the Catholic church in
Biroidakuni, the parish which includes Bibalgree village. Poben Toju, from
Bibalgree, told us how the Catholics came to Biroidakuni:
“There was a rich man in Biroidakuni, Monka Pa. He was the father-in-law
of Kirod Chisim. They were Baptists, but they got into a fight with other
Baptists because they used to drink a lot. So the Baptists threw Monka Pa
out of the church. Then he went to the Catholic mission and invited them
to come to Biroi to start a Catholic church here. He gave them some land
to set up their mission. In this way the Catholic church started in this area.”
These quotations perfectly illustrate the negotiating position which resulted
from denominational strife. The availability of more than one church enabled
Garos to leave one church but to stay Christian. For these deserters, did the
benefits of formal Christian membership prevail over the Sangsarek religion
and tradition? The following fragment from the Baptist Our Bond describes
how the Catholics offered membership to those who were expelled from the
Baptist church:
The few whom we turned out of membership for drunkenness, adultery
and thieving, and who on that account have found a haven of refuge
amongst Roman Catholics, are following on to more wickedness; so the
prospects for the Roman Catholics taking many of our members are not at
all bright.59
At present, the Catholic church is the most popular among Garos. Baptists
often attribute Catholic successes to “sheep stealing” practices: Baptist
missionaries paved the way, and Catholics lured Christian converts into their
church. Another frequently mentioned reason is that Catholics, unlike Baptists,
allow their members to drink chu. A third important factor today is the
professional character of the Catholic church. The Garo Baptist Church more
depends on volunteer work and has fewer economic facilities at its disposal.
In the early twentieth century, the situation was rather different. In 1921, the
Catholic priest Matthew Kearns wrote that “we, on our arrival, contrasted
very poorly with the Baptists, who had a handsome church, a bank, a school
and a corps of teachers in almost every village.”60
Today, the Catholic church is both richer and has more members. But to
attribute these membership numbers to economic factors only does not
suffice. If that were the case, after all, Catholics would never have been able
to establish themselves in Garoland, comparatively poor as they were at the
time. Neither does it mean that Catholics alone were guilty of “sheep stealing”.
During my stay in Bibalgree I encountered quite a number of people who had
been member of more than one church in their lifetime. One Garo had
joined no fewer than five different denominations. Some informants converted
out of spite, others because of marriage, and our informant who swapped
churches at least five times referred to material benefits.
The following account is an example of how Baptists appealed to
Catholics. Here, the Catholic Chronicles of Biroidakuni Mission describe
an incident in which Catholic students left the mission to join the nearby
Baptists. The story goes as follows: one evening in November 1959, a
drama was being performed at Rangrapara Mission. Boys from the Catholic
boarding school were eager to watch the performance. When permission
was refused, they went anyhow. The next day, they were punished for their
disobedience. Sixteen boys, eight Garos and eight Hajongs, refused to
accept their punishment:
Because of this brother Robert told them they could not stay in St. Joseph’s
hostel. On 1 December, these boys brought two garis [carts] obtained from
the Baptists and after loading up their luggage they went directly from the
Biroidakuni Hostel to the Baptist Mission where they received lodging.61
Sixteen boys decided to leave one mission to join the other. The reason why
these boys were at the Catholic mission in the first place was to get an
education. This brings us to another, very significant aspect of missionary
involvement with the Garos: the introduction of education.
EDUCATING GAROS
In 1917, Sachse wrote that “Missionaries have been the chief agents in
spreading education among the primitive people at the foot of the hills.”62
Education was considered one of the most important instruments to convert
Garos, and missionary works and education became intimately linked. That
schools were not only proper instruments to convert but also essential to
keeping the new converts in the church had already been recognized at the
turn of the nineteenth century. In 1899, the Baptist Mission wrote that:
We have realized that, as the people are very ignorant and illiterate, unless
we teach them and educate them they will mostly be Christians in name
only and fall away very easily, and thus become a hindrance to the spread of
the gospel.63
In fact, the chief institution through which both Christianity and education
was spread were the village school.64 Although Garos were not quite so willing
to send their children to school at first, as time passed by, education became
more and more appealing to them. In 1931, it was reported that none of the
minorities of northern Mymensingh had made the same progress in education
as the Garos had.65
When Burling visited the Garo Hills in the 1950s, he found that education
was a good enough reason for parents to allow their children to become
Christian.66 Education turned out to influence self-perception and group
formation in many ways. In Chapter 8, we shall see how a number of

educated young Garo men involved themselves in social works and awareness
raising among other Garos.
Another important consequence of missionary education which has
attributed to the unification of Garos was the introduction of a lingua franca
among the Garos. We have seen that Garos used to be divided into different
groups or dol, each speaking its own language or kushuk. One 90-year-old
Atong remembered the differences between these dol very well. People dressed
differently, spoke different languages, and certainly did not marry each other,
until they were educated:
“There were no close relations before, but now we are close to all. All are
educated and becoming one. … In that time it was different, but later
everyone went to school, got to know others and started to marry them.
… In this way, many of us went to other areas and the relations improved.
Many got married where they went for studies, and stayed there.”
Schools became an important meeting ground for people from all over the
Garo area. These schools were also the place where they learned Abeng.
Through the introduction of education among Garos, Abeng was introduced
as the lingua franca of lowland Garos. Our Atong informant remembered
how his grandparents never learnt Abeng, but how that changed after the
introduction of education:
“Later, you know, Rajkishor master and Ruben from Birisiri came to our
area to work in the school. Since then we slowly started to learn Abeng and
Bengali. They stayed in the house of my parents-in-law. They were the first
Abeng who lived in our area.”
Today, Garos view the combination of education and Christianity as proof of
their development and modern identity. Christianity and education provide
them with a sense of pride and a distinct ethnic identity.
MISSIONARIES AND THE COLONIAL STATE

Many post-colonial anthropologists have described missionaries as mere
tools of the colonizer, and missionary activities as a symptom of colonial
control.67 “[B]ehind all that great facade of religious efforts was hidden the
instinct and the ambition of the imperialist and the coloniser to make other
people his political slaves.”68 Such an approach fails to take the role of
converts into account and overlooks that the marriage between colonizer
and missionary was not always a happy one.69 As for colonial involvement
with the Garo mission, it is important to distinguish between hill Garos
and the lowland Garos of East Bengal. In the case of the hill Garos, the
relationship between the missionaries and the government was complex.
Although it was a British administrator who conceived of the idea to
convert the Garos, already as early as the 1820s,70 the official government
policy of the time was to maintain neutrality. In the second half of the
nineteenth century, the government’s attitude changed considerably.
Christianization came to be viewed as a means to pacify the Garos. In the
1860s, the American Baptist mission received financial aid from the
government to further develop the educational system in the hills.71
Nowhere did I find an indication that the colonial government financed
missionary activities among lowland Garos. Perhaps this can be explained
by the fact that the colonial desire to “pacify” the hill Garos did not apply
to the “peaceful” Garo lowlanders. That fact that missionaries and the
colonial administration maintained a good relationship and shared similar
interests does not turn Christian missionaries into a mere extension of the
colonizer. Their primary goal was not to colonize but to Christianize. To
reach that goal they were, both metaphorically and literally, willing to go
very far. An example:
Our missionaries had 120 miles to cover but happily were able to do two-
thirds of it by rail. Then they walked ten miles knee-deep in mud and they
waddled in water up to their waists, crossing torrents swollen by rain, over
frail bamboo bridges which swayed and rocked over the muddy water. In
order to gain time, they had brought their bicycles with them, but these
were useless except on dry ground, and were very cumberstone when the
missionaries had to go over marshy trails.72
If we see conversion as an adaptation to changing circumstances, we
automatically arrive at the issue of power relations. Many Garos of colonial
India found themselves in the less powerful sections of society (see Chapter
8). The question remains to what extent this unequal power relationship was
transferred to missionary work.73 Although it is not possible to determine to
what extent the issue of power and powerlessness influenced the attraction of
Christianity to Garos, missionaries benefited from their relationship with the
colonial state. The next chapter demonstrates how this relation enabled
missionaries to act as mediators between Garos and the state, and how it
offered them a position as Garo patron or benefactor.
CONCLUSIONS
Christianity offered East Bengal’s Garos a way to maintain a separate
identity in times when contact with others intensified. In the words of
Burling, “Christianity provided a means by which they [Garos] could

maintain their separation from their neighbours, even while they were
becoming more closely involved with them.”74 In this chapter, I have
demonstrated that these observations indeed apply to Garos who converted
in the last few decades. These converts opted for Christianity primarily
because they wanted to stay with the other Garos. Nevertheless, as much as
Christianity offered Garos a way to distinguish themselves from their
neighbours in the longer run, initially it introduced hostility between the
new Christians and the Sangsareks. As time passed, and more and more
Garos converted, these relationships improved.
In the long run, Christianity proved an important factor in the process of
unification of the lowland Garos. Various reasons contributed to the
ethnogenesis of Garos. Other than that Christianity offered a modern identity
and provided links with the international world, it also offered Garos a shared
platform to organize and to define themselves as one distinct, ethnic group.
Through the introduction of education, a lingua franca, and a more
homogeneous Christian culture, Christianity (and its messengers) became a
significant factor in the ethnogenesis of Garos.
Chapter 8 further explores this process of ethnic unification. The special
focus lies on the role of the subsequent states in relation to the formation of
the Garos of Bangladesh. The chapter investigates the different developments
which took place in colonial and post-colonial Bengal, and analyses how
these interacted with the process of identity formation of the Garos.
NOTES
1. In 1995, the Catholic Mission of Modhupur Forest reported in their yearly
census that, out of a total of 14,141 Garos, 9,297 were Catholics, 2,056 were
Baptists, 917 were Anglican (Church of Bangladesh, or “Oxford Church”), and
164 were Seventh Day Adventists. Of the remaining 1,707 non-Christian Garos,
approximately 200 were Hindus and the remaining Garos were Sangsarek. Since
Sangsarek and Hindu influences are stronger in Modhupur than elsewhere in the
Garo areas, and since conversion to Christianity has continued since 1995, the
percentage of Christians is likely be higher in other parts of Bangladesh and has
probably also increased since then. For the figures of the Catholic Mission of
Modhupur, see Burling, Strong Women of Modhupur, p. 178. It should be noted
that, according to some Hindu Garo interviewees in Modhupur, there are at
present some 85 Hindu Garo families in the area, comprising 600 to 650 people.
2. Intellectualism attributes conversion to an intellectual need for more effective
explanations of the social world. See, for example, Robin Horton’s intellectualist
theory in Robin Horton, “On the Rationality of Conversion”, Part I, Africa 45,
no. 3 (1995).
3. Cf. Robert W. Hefner, “Of Faith and Commitment: Christian Conversion in

Muslim Java”, in Conversion to Christianity. Historical and Anthropological
Perspectives on a Great Transformation, edited by Robert W. Hefner (Berkeley,
etc.: University of California Press, 1993), p. 119.
4. During one of my visits to Bangladesh (in December 1997), the only Sangsarek
member of the household where I had lived proudly informed me of his recent
baptism. He was getting old and did not want to risk not being able to go to
heaven. Since he lived in a Baptist household, he converted to Baptism. I never
discovered whether he also quit drinking chu.
5. These terms were derived from Emefie Ikenga-Metuh. Robert W. Hefner,
“Introduction: World Building and the Rationality of Conversion”, in Hefner,
ed., Conversion to Christianity, p. 27.
6. Cf. Peter Wood, “Afterword: Boundaries and Horizons”, in Hefner, ed., Conversion
to Christianity, p. 305.
7. Baptism for official membership is required, and only adults can receive baptism.
8. This number is based on the figures that were provided by the two largest
denominations. For the Church of Bangladesh, staff members could only give
some estimation. Other, smaller denominations and the Garos of Sylhet (except
for the Baptists) are not included. See Catholic Directory of Bangladesh (Dhaka:
Bishop’s Conference of Bangladesh, 1992); Yearly Report of the GBC to the
ABMS, 1991–1992 (n.p., n.d.).
9. A lack of sufficient and reliable statistical data makes it impossible to say exactly
when that happened.
10. But see Susan Bayly, Saints, Goddesses and Kings. Muslims and Christians in South
Indian Society 1700–1900 (Cambridge, etc.: Cambridge University Press, 1989),
p. 3.
11. C. Becker, History of the Catholic missions in northeast India 1890–1915 (Calcutta:
Firma KLM [translated and edited edition], 1980); Edmund Goedert, The
Apostolate to the Garos (Dacca, manuscript, 1959); Edmund Goedert, Holy Cross
Priests in the Diocese of Dacca 1853–1981 (Notre Dame Indiana: Province
Archive Centre, 1983); Sebastian Karotemprel, Albizuri among the Lyngams (A
Brief History of the Catholic Mission among the Lyngams of Northeast India)
(Shillong: Vendrame Missiological Institute, 1985); Early History of the Catholic
Missions in Northeast India, 1598–1890, edited by F. Leicht and S. Karotemprel
(Calcutta: Firma KLM, 1989). Milton S. Sangma, History of American Baptist
Mission in North-East India (1836–1950). Part I & II (Delhi: Mittal, 1987);
Kanti Prasanda Sen Gupta, The Christian Missionaries in Bengal 1793–1833
(Calcutta: K.L. Mukhopadhyay, 1971); Gordon Soddy, Baptists in Bangladesh. A
Historical Sketch of More than One Hundred Years’ Work of the Baptist Missionary
Society in Bengal (Dhaka: National Council of Churches – Bangladesh, 1978);
George Kottuppallil, “Roman Catholic Work among the Garos 1911–1933”,
Indian Church History Review n.d., pp. 111–21.
12. Exceptions include Frederick S. Frederick S. Downs, North East India in the
Nineteenth and Twentieth Centuries. History of Christianity in India Volume V,
Part 5 (Bangalore: The Church History Association of India, 1992); Richard M.

Eaton, “Conversion to Christianity among the Nagas, 1876–1971”, The Indian
Economic and Social History Review 21, no. 1 (1984): 1–44.
13. See for example, Hefner, ed., Conversion to Christianity.
14. Frederick S. Downs, “Christian Conversion Movements among the Hill Tribes
of Northeast India in the Nineteenth and Twentieth History”, in Religion in
South Asia. Religious Conversion and Revival Movements in South Asia in Medieval
and Modern Times, edited by G.A. Oddie (New Delhi: Manohar, 1991, 2nd
revised edition), pp. 155–74; Downs, North East India; Parimal Chandra Kar,
Glimpses of the Garos. Chapters 6 and 7 (Tura: Garo Hills Book Emporium,
1982), pp. 89–140; Ferdaus A. Quarishi, Christianity in the North Eastern Hills
of South Asia. Social Impact and Political Implications (Dhaka: University Press
Limited, 1987); Sangma, History of American Baptist Mission (Delhi: Mittal
Publications, 1987), especially Chapter 6.
15. But see Burling, who wrote an anthropological account on the influence of
Christianity on the Garos of Modhupur Forest. Burling, Strong Women of
Modhupur, pp. 175–205.
16. See for example, C.D. Baldwin, God and the Garos (Sydney: Australian Baptist
Publishing House, 1934); Ian Emmett, Man of Faith and Flame. Bilu Babu of
Birisiri (Melbourne: Jenkin Boxton and Co, n.d.).
17. Baldwin, God and the Garos, p. 40. The Baptists Reverend P.C. Nall presented
another, perhaps more accurate, version of this first conversion and baptism of
a plains Garo. Nall based his account on the Baptist records and on an 1873
report of the Baptist missionary Ruprecht Bion. He writes that the earliest
preaching amongst Bengal’s Garos was by two Bengali evangelists in 1868.
Bengalis carried on this work until 1873, when Bion, of the English Baptist
Mission, toured the northern portion of our district with the purpose of exploiting
the Garo field. His reception was so hostile and the difficulties so great that Bion
concluded his report for 1873 with: “Thus have we been led to relinquish our
long-cherished hope of bringing the Gospel to these hill people.” Yet in 1875,
the first Garo convert named Radha Nath was baptised. See Rev. P.C. Nall, “The
Garo Mission”, Our Bond. Volume XIV (May 1908): 27.
18. See for example, Goedert, Holy Cross Priests in the Diocese of Dacca, pp. 1–5.
19. Bangladesh Bureau of Statistics, 1993 Statistical Yearbook of Bangladesh. 14th
edition (Dhaka: Bangladesh Bureau of Statistics, Statistics Division, 1994).
20. Goedert, Holy Cross Priests in the Diocese of Dacca, p. 16.
21. Bion, who had come to Bengal in 1847 and had joined the English Baptist
Missionary Society in 1851, became the Baptist pioneer in both Comilla and
Mymensingh.
22. Bilash Chandra Mukherji, “Progress among the Garos”, Missionary Herald (1940):
65.
23. Goedert, Apostolate to the Garos.
24. Other current Christian missionary ventures among the Garos of Bangladesh
are, for example, the Seventh Day Adventists, the Church of Fellowship, the
New Apostolic Church and Korean Baptists.
25. Matthew Kearns in The Bengalese 2, no. 8 (April 1921): 134.
26. Baptist Union of Victoria, Handbook Containing President’s Address and Reports
of the Union Home Mission, Fund, College, Foreign Mission & c. (Melbourne:
Watt & Co., Printers, 1900), p. 54.
27. There were many reasons why converts had to abstain from chu. One was that
chu was associated with drunkenness and social and economic decline. A second
reason was that chu was directly related to Sangsarek practices. They needed it
for their sacrifices and enjoyed it during their festivals. To give up chu meant a
clear break with former customs. Such dietary restrictions fit well in the South
Asian context, where they are directly related to conceptions of status. Cf.
Burling, Strong Women of Modhupur, pp. 176–77.
28. Catholics were generally more lenient towards traditional Garo culture. For
example, they did not prohibit chu. Burling attributes this different attitude
where Garo culture is concerned to the fact that no single national group
dominated the Catholic mission in the way Australians dominated the Baptists
in Mymensingh. Burling, Strong Women of Modhupur, p. 177.
29. Baptist Union of Victoria, Handbook (1899), p. 43.
30. Rangrapara Minutes for the period 12 March 1972 until 16 June 1991 mention
at least 120 persons being accused of, and often expelled for, adultery, illegal
marriage, consuming chu, joining another denomination, or marrying a Muslim
or Hindu. It is interesting that such matters receive the most attention. Hardly
any notice is paid to the socio-economic situation of the people or to religious
matters.
31. Our Bond XXIII (March 1917): 12.
32. This information is derived from an interview held by Suborno Chisim in May
1993. In May 1993, Suborno Chisim and Sumona Chisim conducted a number
of interviews for the Brothers of Taizé in Mymensingh to gather more information
on the Garos of Bangladesh, particularly the influence of Christianity on Garo
law. I want to thank the Taizé brothers for allowing me to use that material.
Hereafter I shall refer to both interviewers where I use fragments from the
interviews they conducted for this particular study.
33. Italics are mine. See Barnett in Missionary Herald (1911), p. 42.
35. Interview conducted by Suborno Chisim, May 1993.
36. Interview conducted by Suborno Chisim, April 1993.
37. In 1964, the Garos were confronted with a mass influx in their area of Bengali
refugees from India and Bengali settlers from other places in Bangladesh. At the
same time, they faced the hostile attitude of the Pakistani army, police, and East
Pakistan border security force (EPR). The central government did not interfere
until the large majority of northern Mymensingh’s Garos had fled to India.
Many Garos never returned. For more details, see Chapter 8 of this book.
38. See also Chapter 3, n.11, of this study.

39. Porter refers to the Baptist reverend V.J. White, who ordered the local non-
Bengali population as follows: Garos or Mandais (lowest), Dalus, Koch, Banais,
Hadis or Hotris, Hajangs with three sub-castes, Rabhas with seven sub-tribes,
Mechs, and Rajbansis. Porter, Census of India, 1931, p. 463. For Bastin’s comments
see Chapter 6, n.18 of this book.
40. Porter, The Census of India, 1931, p. 464.
41. Today the Bangladeshi Garos have two small and rather isolated communities
of Garo Hindus, one in Phulpur thana and one in Modhupur forest. These
communities are not growing.
42. Bastin, Five thanas of Partially Excluded Area, p. 39.
43. Baldwin, God and the Garos, p. 53.
44. Bastin, Five Thanas of Partially Excluded Area, pp. 53–55.
45. Porter, Census of India, 1931, p. 464.
46. Robbins Burling, “An Incipient Caste Organization in the Garo Hills”, in Man
in India 40, no. 4 (December 1960): 284–85; Burling, Renggsangri, p. 358.
47. Compare these figures with the 1995 census of the Catholic Mission in Modhupur,
which mentions some 200 Hindu Garos. Burling, Strong Women of Modhupur,
p. 178.
48. Porter, Census of India, 1931, p. 464.
50. Ibid., p. 676.
51. Ibid., p. 665. Buchanan further adds that “[a] great part of the slaves are
procured from the Nuniyas, who bring them from Assam. They are chiefly
Garos, who have lost caste by impure feeding and have been sold as a punishment
for their transgression.” (ibid., p. 693.)
52. W.R. Bayley, secretary to the government in a letter dated 27 April 1825, cited
in Mackenzie, North East Frontier of Bengal, p. 253.
53. Bastin, Five Thanas of Partially Excluded Area, p. 54.
54. Ibid., p. 53.
55. Burling, Strong Women of Modhupur, pp. 186–87.
56. Although differences also arose in matters of dogma and biblical interpretation,
I do not go into this here because I found nothing that suggests that such
intellectual differences played any role in the decision about conversion.
57. Our Bond 16, no. 2 (February 1910).
58. The Bengalese 4, no. 5 (May 1923): 7.
59. Our Bond 16, no. 5 (May 1910): 5.
60. Matthew Kearns in The Bengalese 2, no. 8 (April 1921): 134.
61. Chronicles of Biroidakuni Mission, 1 December 1959.
62. Sachse, Mymensingh, p. 139.
64. Such was a widely adopted method among Christian missionaries. See for
example, Richard Eaton, “Conversion to Christianity among the Nagas, 1876–
1971”, The Indian Economic and Social History Review 21, no. 1 (1984): 10.
65. Porter, The Census of India, 1931, p. 464.

66. Cf. Burling, Rengsanggri, p. 313.
67. Cf. Karlsson, Contested Belonging, p. 183.
68. Brahm Datt Bharti, “Political Roots of Christianity in India,” in Politics of
Conversion, edited by Devendra Swarup (Delhi: Deendayal Research Institute,
1986), p. 100. See also T.R. Vedantham, who expresses a very similar opinion
about the relation between the colonizer and the missionary in his article
“Church as a Tool of Imperialism in the Modern World”, in Swarup, ed., Politics
of Conversion, pp. 71–79.
69. Karlsson argues that government support was never whole-heartedly, and
throughout the colonial period restrictions were put on missionary activities.
Elizabeth Susan Alexander writes that, until the first decade of the nineteenth
century, it was feared by the English East India Company that missionary
activities would disrupt the social and political situation in company holdings.
In the early nineteenth century, their attitude towards missionary involvement
changed until the 1857 “mutiny” again raised doubts about the missionary
involvement in the colony. Missionary activities were held partly responsible for
the revolt, and the British Government did not wish to openly collaborate with
the missionaries. See Elizabeth Susan Alexander, The Attitudes of the British
Protestant Missionaries Towards Nationalism in India. With Special Reference to
Madras Presidency 1919–1927 (Delhi: Konark Publishers, 1994), pp. 1–20;
Karlsson, Contested Belonging, p. 187.
70. As early as 1822, the government administrator, David Scott, was corresponding
with the Serampore Mission about the appointment of a missionary for the
Garos. For a more extensive account on the early contacts between missionaries
and Garos: Sangma, History and Culture of the Garos, pp. 254–66.
71. For a more elaborate account of the ambiguous attitude of the British colonialists
towards Christianizing the Garos, see Bhattacharjee, The Garos and the English,
pp. 217–28.
72. Letter Legrand to Francais, 1917. Quoted in Raymond C. Clancy, The
Congregation of Holy Cross in East Bengal, 1853–1953. In Three Parts with a Brief
History of the Church in Bengal. Volume 1 (Holy Cross Foreign Mission Seminary,
Washington D.C.: unpublished manuscript, 1953), p. 193.
73. Dick Kooiman, Conversion and Social Equality in India. The London Missionary
Society in South Travancore in the 19th Century (New Delhi, Manohar, 1989),
p. 9.
8
GAROS AND THE STATE
INTRODUCTION
This chapter is about the role of the state in the ethnogenesis of the Garos
of Bangladesh. It explores how the successive states of East Bengal, East
Pakistan, and Bangladesh have affected the self-perception and organization
of the Garos. In order to study the dynamics of the relation between the
Garos and these states, the chapter deals with a lengthy period of some 250
years. Such an approach allows us to obtain an overview of how states
influence society and vice versa. It reveals that states are by no means static
or timeless entities. The discussion focuses on the dynamics of state-society
interaction rather than on its constants. Another, very practical reason for
the choice of such a long period is the availability (or rather, absence) of
historical documentation. The history of Garos in relation to the state is
marked by long silences. One moment Garos feature in historical accounts,
and the next moment they are invisible again.
I examine how both colonial and post-colonial states influenced the
region. Major differences are to be expected between the two. The colonial
state was primarily organized to the advantage of the British Raj. Its policies,
including those claimed to be in favour of (certain segments of ) the colonial
population, are often approached with doubt and suspicion. For example,
those measures that were adopted to protect so-called backward or tribal areas
and people are often disposed of as divide-and-rule tactics. Post-colonial
158
Garos and the State 159
states, on the other hand, are expected to act in favour of the people. Its
leaders have gone to great lengths to sustain an image of the state as “the
natural spokesman for the people of society”, that is, the nation.1 Here we
shall see that reality is more complex. Although post-colonial states shape and
reshape the social and political landscape – thereby inevitably influencing the
nation – they do not necessarily act in favour of the (whole) nation.2
In this chapter, I have adopted two different perspectives. The state
perspective focuses on how successive states have influenced the Garos.
Questions that come to the fore deal with the process of state formation,
revenue matters, boundary issues, and (the organization of ) state agencies.
The Garo perspective approaches the state through the eyes of the Garos.
How have Garos perceived the subsequent states and related to them? How
have they experienced their governments and agencies, and the policies
carried out by them?
Here, state is understood as “an organization, composed of numerous
agencies led and coordinated by the state’s leadership (executive authority),
that has the ability or authority to make and implement the binding rules
for all the people, as well as the parameters of rule making for other social
organizations in a given territory, using force, if necessary, to have its way”.3
In reality, the capability of states to determine how social life is ordered
differs significantly from state to state and from time to time, and state
leaders’ control over state agencies varies as well.4 It is also important to
realize that state and nation are two “very different entities with distinct
histories, constituents and ‘interests’ ”.5 This chapter explores how, if at all,
states have spoken for the Garos. To what extent have Garos had a place and
a voice in state policies? To what extent did successive states take account of
them? And finally, states and societies are not homogeneous, static entities.
There is more than one perspective on the state, and the dominant discourses
of subjugation and liberation do not necessarily correspond with those of
minorities like the Garos.
For my inquiry into the colonial period, I depend largely on secondary
accounts, which were produced by colonial administrators and Christian
missionaries. But I also use oral history, the principal source for the
investigations of the post-colonial era. Oral history breaks the silence and
reveals Garo perceptions of the state and allows for a shift of focus from a state
perspective (on the Garos) to a Garo perspective (on the state).
These two (colonial and post-colonial) periods are divided into four major
chronological sections, on the basis of some significant changes in the relation
between the state and the Garos. Each section deals with a number of shorter
periods and breaking-points. I start with Mughal rule in the eighteenth century
and the first decades of colonial hegemony over Bengal, followed by an account
of “high colonialism”, which stretches from 1833 to 1947. Next, I discuss the
era of Pakistani rule and domination. The chapter ends with a discussion of the
Garos and their relation with the state of Bangladesh.
MYMENSINGH’S BUFFER STATES6

We know very little about pre-colonial state policies and the Garos of
Bengal, and what we know depends on British documentation of the pre-
British era. For example, Buchanan suggested that the Assamese had never
been able to drive the landlords on the northern side of the Garo Hills from
their dominions, and that the same was possibly true for the subsequent
Mughal rulers, whose empire also bordered the hills on the eastern and
southern side of the Garo hills.7
Similarly, Gautam Bhadra, who studied the Kamrup-Goalpara region
adjacent to the northern side of the hills, contends that the Mughals were
never able to integrate peripheral zones within their state structure. He
bases his argument on the recurrence of peasant revolts in the area.8 Whether
by choice or not, the Mughals maintained a policy of non-interference
throughout their rule.
The frontier estates seemed to serve two purposes for the Mughal rulers:
they functioned as a buffer between the Mughal empire and the independent
Garo Hills,9 and the frontier landlords paid their Mughal overlords a small
tribute of elephants and aghur (a special kind of wood); they also paid for
certain small garrisons and the upkeep of the Dacca Artillery park.10
Thus the Mughals had adopted a policy of non-interference in northern
Mymensingh. They did not intervene in the management of this string of
frontier estates.11 The local landlords (also referred to as zamindars, choudries
or rajas) were more or less left to themselves. In order to protect their estates
against attacks of hill Garos, they were allowed to maintain small private
armies of borkondazes (armed retainers). These borkondazes did not only serve
the zamindars to protect the zamindari estates against outside attacks, but
also helped them to enforce their positions in their own estates (at the expense
of the peasantry).
Interestingly, despite their freedom, the zamindars often failed to fulfil
their duties and were imprisoned or tortured as a punishment.12 The question
remains whether these landlords were really unable to meet Mughal demands,
or whether we should interpret their behaviour as an act of defiance. Alexander
Mackenzie writes that the zamindars maintained their independence
throughout Mughal rule. They only paid a minor tribute to the Fouzdar, or
Mughal tax collector, while making enormous profits in the cotton trade at
the expense of the independent hill Garos.13 These Garos needed the cotton
trade just as much for the purchase of various goods,14 and they depended
totally on the markets along the foothills. The zamindars, who controlled
these markets, were free to extort high duties from these Garo traders from
the hills, who were subjected to the “worst types of exploitation”.15
Power relations in the region were obviously not settled. There were the
landlords with their local armies, the local tax collector, who carried out
transactions which were “mainly for the benefit of himself and the Choudries”,16
the Mughal overlords, who intimidated the zamindars by frequent
imprisonment and torture, and the hill Garos, who regularly raided the plains
to retaliate the zamindari exploitation.
BRITISH TAKEOVER
The British defeat of the Mughal emperor and his allies in 1757 did not
constitute a breaking point between two very different administrative systems,
and did not have immediate consequences for the local population. State
power in the region remained weak. The East India Company made its
appearance as a trading firm, and during their first years of administration in
Bengal, the British were more interested in trade than in territorial
acquisitions.17 This meant that the extent of the zamindari lands remained
unknown and no precise map was drawn up.18 The zamindars maintained
enough freedom to manage their estates and ryots (tenants) to their own
advantage.19 They continued to treat the frontier region as a buffer between
their empire and the independent Garo Hills. They left its actual governance
to the local landlords or zamindars, who retained much of their independence
for many years to come. It took the British another seventy years to firmly
establish British rule in northern Mymensingh.20 Until then, aided by its
physical isolation, the region continued to function as a buffer between the
colonial state and foreign invaders.21
Information about the local population in those days is scarce. These
people could hardly read or write, and they rarely attracted official attention.
To study state policies is one thing, to get a better insight into the daily
situation of the local population is quite another:
It is always easier to write about “high” politics at the centres of power and
the preoccupations of the early colonialists than to enter the twilight zone
of local politics under colonial power. In that zone most actors could
neither read nor write, unlike the colonial officials whose biased and often
ignorant reports are the main surviving source for us.22
What we do know is that at the time of the British takeover, the region had
its own agricultural structure, maintained its own regional identity – both
politically and culturally – and knew a great linguistic, ethnic, cultural, and
religious diversity. Apart from Bengalis, other peoples inhabiting the region
were the Hajongs, Garos, Koches, Banais, Dalus, Hodies, and Rajbansis.
Other than Bengali, which at the time was already used as the lingua franca
of the region, Hajong, Garo, and perhaps other languages were widely
spoken.23 The area had never been under any firm administrative rule.
Instead, it had been marked by a long history of changes between periods
of regional independence and absorption into larger states. The power over
the area, or parts of it, had been shifting hands for many centuries, and
several groups of people, including the Garos, had been involved in these
power struggles.24
THE END OF ZAMINDARI HEGEMONY

Slowly, the zamindars tried to increase their power and income externally (at
the expense the hill Garos), and internally, by increasing the pressure on the
local peasantry. These peasants had long been the major victims of instability.
They could expect claims on their production from different sides: the
zamindars (acting on behalf of themselves and the state), government officials
exacting illegal levies, and sannyasis and fakirs who gained their income from
moneylending.25 In the early nineteenth century, the local peasantry finally
took up arms and revolted against the zamindari exploitation. In 1824–25
and again in 1833, northern Mymensingh witnessed several peasant
insurrections, which entered history as the Pagol Panthi uprisings.26
From the start of these revolts, in which Garos and Hajongs played an
important role,27 both the zamindars and the peasants tried to mobilize state
support. It is interesting to note that the colonial government, which had
openly shown their disapproval of the illegal zamindari practices, supported
the zamindars to suppress the rebellious peasants.28 With government help
peasant resistance was soon broken, but so was the powerful position of the
local zamindars. Government interference meant a definite breaking point in
the relations between the colonial state, the local landlords, and the peasantry.
The balance of power had tilted in favour of the colonial state. It was not
until some seven decades after British succession to the Mughals that northern
Mymensingh was finally incorporated into the colonial state. From that
moment, the state increasingly involved itself with the local situation.
Nevertheless, the road towards state control was a bumpy one. It was not the
case that the colonial state completely dominated the region after 1833.
The establishment of colonial hegemony over northern Mymensingh did

not resolve the conflicts between the zamindars of the frontier estates and
the hill Garos. It has been argued that these relations deteriorated with the
increase of colonial influence, and that “[t]here is reason to believe that the
‘terrors of the British Musquetry’ encouraged the Chaudhuries to adopt a
more aggressive attitude towards the Garos than they had under the Mughals.”29
Zamindars extended their territory and income at the expense of the hill
people, who reacted with frequent raids into the plains.
This troublesome situation came to an end in 1869, when the British
brought zamindari influence in the hills to a definite end.30 The British laid
down an official boundary along the foot of the hills; the landlords could no
longer make any claims on a portion of the hills, and the collection of rent
and cess by the zamindars in the hills was stopped.31 Three years later, in
1872, the Garo Hills were officially incorporated into the British Empire.32
At this point, the colonial state truly established itself in this region. It not
only controlled northern Mymensingh, but also its bordering regions. Conflicts
between the zamindars and hill Garos stopped; Garo raids were brought to an
end, and northern Mymensingh’s landlords lost their last privileges and were
turned into “ordinary” zamindars.33
THE GAROS OF MYMENSINGH, 1765–1833

The Pagol Panthi movement was a class-based revolt, a case of peasants
rising up against the local landlords and the colonial state, with explicitly
political and economic goals. Different ethnic and religious groups
cooperated to resist economic exploitation. The movement attracted peasants
from various religious and ethnic backgrounds and there is no evidence of
any ethnic or religious animosity. Never again did the region witness similar
class-based uprisings. Instead, religion and ethnicity became increasingly
important as mobilizing factors.34 It is, however, entirely unclear how many
Garos participated in this movement and from where they came. Nowhere
did I find evidence of hill Garos joining the Pagol Panthis, of plains Garos
helping hill Garos to defend their territory, or of hill Garos closing ranks
and fighting off the zamindars collectively. Garos of the time possibly did
not share a strong sense of belonging together, of making up one distinct
category of people. They did not constitute an “imagined community”. It
seems that the Garo Hills only received their explicit meaning as a distinct
Garo domain after its annexation by the British, and that the significance
of the Garo Hills as proper Garo territory is directly related to colonial
expansion and state politics.
‘HIGH COLONIALISM’: STATE HEGEMONY IN

NORTHERN MYMENSINGH
In the late eighteenth and early nineteenth century, northern Mymensingh’s
population witnessed a significant change of power. The establishment of
colonial hegemony over the region brought about a great many changes for
the local population. The state increasingly invested in its state apparatus.
Both revenue income and expenditure on civil administration of Mymensingh
district increased substantially. Between 1795 and 1870, the net income from
revenue more than doubled, whereas state expenditures increased fourfold.35
Government expenditures were augmented most drastically during the second
half of the nineteenth century.36
State penetration went hand in hand with two other processes:
objectification and territorialization. Objectification means enumeration,
categorization, and quantification. The colonial state and its administration
developed an increasing hunger for facts and figures. In 1872, the first
systematic attempt to enumerate the district’s population was undertaken.
Until then, state knowledge of Mymensingh’s population had been fragmented
and limited. This first population census revealed that the rough estimations
in the Settlement Survey of 1850–56 were one-half lower than the results of
the population census.37
Territorialization refers to the demarcation of territory and the construction
of borders. At the time, the incorporation of the Garo Hills in British India,
and the construction of the district boundary between Mymensingh and the
newly established Garo Hills district may have had few practical consequences
for the local population of Mymensingh. Long-term consequences were,
however, manifold. What had become a district boundary would turn into an
international border in 1947. Moreover, with the Garo Hills district, the
British constructed a distinct Garo territory for the hill Garos, demarcated by
official boundaries, governed under one administrative umbrella. The
colonialists, who had perceived the Garos as one distinct nation, now also
provided them with their “own” well-demarcated territory.
Mymensingh’s plains Garos did not possess any officially recognized or
separately administrated territory. Whereas the state increased its hold over
northern Mymensingh, in practice and in knowledge, the state did not “see”
the Garos. The colonial administration seemed totally ignorant. H.J. Reynolds,
for example, the magistrate and collector of Mymensingh no less, wrote in
1868 that “[t]he Garrow Hills, however, are no part of the District, and any
notice of the Garrows would be out of place here, if it were not for the fact
that the villages lying within the District at the foot of the hills are largely
inhabited by a race of mongrel Garrows, known as Hajungs.”38 And since the
central government found that, in the words of Sachse, “the only historical
events of any interest connected with Mymensingh are the inroads of the
Gáros …, and the rebellion of the Sanyàsis,” we are left with a big historical
void where the plains Garos are concerned.39 Their administrative invisibility
only came to an end in the early twentieth century.
EXCLUSIVE POLICIES
In the early twentieth century, the colonial state did not merely see the Garos
of Bengal, it also developed a special policy for them. For the Garos, these last
decades of British domination were the highlight of high colonialism. In
1936, the British extended their all-India policies to exclude or partially
exclude certain backward or tribal areas to northern Mymensingh. The five
northernmost thanas of Mymensingh came under direct control of the
Governor of Bengal. They remained under Mymensingh district, but no
provincial or state act could be applied until approved by the governor, who
was thus allowed to directly interfere in the local situation. The roots of these
twentieth-century policies aimed to (partially) exclude certain specific tribal
areas from local or provincial legislature lay in the eighteenth century. They
were further developed in the nineteenth century, when most of the so-called
tribal areas were incorporated into the British Empire.40
The government decision to declare the five northernmost thanas of
Mymensingh (together with Dewanganj thana to the west) as Partially Excluded
Area had many short-term and long-term consequences for its inhabitants.
This decision has influenced the Garos of the East Bengal until this very day.
It did not only bring about several practical implementations, it also provided
the Garos of Bengal with a viable conception of a Garo homeland in Bengal.
These five thanas of partially excluded area (Garos refer to them as PEA) came
to symbolize their original homeland.
Three measures stood out in particular: the appointment of a special
officer, the establishment of the Aboriginal Welfare Association, and the
Revisional Survey of 1938–42. The special officer was endowed with several
tasks. For example, he was authorized to restore lands to the original owners,41
instructed to encourage the aboriginals to organize themselves, and asked to
make suggestions to protect the interests of these people. A second measure
was the formation of the Aboriginal Welfare Association, under the leadership
of the Baptist missionary V.J. White of Birisiri. The association was given the
special task to educate the “tribals”. On the suggestion of White, eleven
Aboriginal Welfare Boards were established under the association, all guided
by a local influential aboriginal, or missionary chairman.42 The Revisional
Survey resulted from the proposals made by the special officer. Since the last
survey of 1908–19,43 the situation had really changed. A lot of waste land had
been brought under cultivation but had never been officially allocated to the
original cultivators who often belonged to one of the region’s minorities. The
officer argued that without a new survey it would be impossible to establish
the right ownership and to cut down the excessive rents.44 It was the settlement
officer of this Revisional Survey, Bastin, who produced one of the most
valuable documents about the PEA.
It remains unclear why the colonial government decided to implement
these special policies in this particular part of Bengal. Other than Sherpur and
Shushong in Mymensingh district, no other regions in East Bengal acquired
the status of Partially Excluded. Why did the colonial state exclude this
portion of Mymensingh district while it had never been treated in a similar
exclusive fashion before? The decision is even more surprising when we take
into account that, with the exception of a small strip of land (varying in width
from five to eight kilometres) at the foot of the Garo Hills, the majority
population of these five thanas were Muslim Bengalis.45 There is no evidence
suggesting that the government devised their policies primarily to undertake
any special projects of economic or strategic interest, as was the case in the
Chittagong Hill Tracts.46 Bastin summarized the arguments in favour of
partial exclusion. The simplicity and lack of education of the aboriginals
supposedly placed them at a disadvantage in dealing with the cleverer and
better educated Bengalis. Secondly, the shortage of ready money made the
minorities victims of moneylenders and forced them to take land on the
tanka system.47 The third reason applied especially to the Garos, who formed
the majority of the so-called aboriginals. Their matrilineal system placed
them in an adverse position since, according to Bastin, it did not give men
any incentives to work.48 All these arguments fit nicely in the picture of the
colonial state as the protector of the “backward tribes”.
MISSIONARY MEDIATORS
The mentioned reasons alone cannot explain the attention paid to these
particular minorities in this particular “out-of-the-way corner” of Bengal.49
After all, the Garos or Hajongs were not the only so-called aboriginals of
Bengal, nor were they the only people who suffered from land loss or
indebtedness. Christian missionaries had already been working with the
Garos for some decades when northern Mymensingh was partially excluded.
The Garos had in fact become the greatest missionary success story of Eastern
Bengal. It is no great surprise that the governor based his decision to partially
exclude the region not only on a report that was provided by the district
magistrate, but also consulted the Australian Baptist missionaries of Birisiri

and the Roman Catholic missionaries of the Congregation of Holy Cross
working in the area.50
We can hear missionary voices resonate in all Bastin’s arguments. For
example, the missionaries worried about the impact of the matrilineal
kinship on the economic productivity and social interest of Garo men.51
Similarly, they were also concerned with the issue of indebtedness. This
problem was frequently mentioned in the early twentieth-century missionary
accounts. Many activities were developed to prevent Garos from falling into
the hands of the mahajans or moneylenders. This was a problem which
confronted the very first missionaries, who had already reported the problem
of indebtedness at the very start of their work. In 1903, one Baptist
missionary wrote that
Many of our folks [Garos] are yet under the cruel tyranny of the money
lenders. It is distressing, while on tour, to see our people’s rice being carted
off to these blood-suckers. Once a man gets in their clutches there is little
hope that he will, unless aided from outside, get free.52
The poor conditions of the Garos were also of major concern to the
missionaries. After all, “there was the obvious fact that a chronically destitute
Catholic is a poor advertisement for the Church in this land where wealth and
health are regarded generally as positive marks of divine favour.”53
British administrators and Christian missionaries obviously had some
shared interests in the region, especially since Bengal witnessed anti-colonial
agitations everywhere and a new gulf of peasant uprisings which were known
as the Tebhaga movement.54 The colonial administration was aware of possible
unrest among the local peasantry of northern Mymensingh. In his report,
Bastin writes:
The Settlement Staff were instructed to impress on the tenants that their
grievances were being looked into and were urged to pay their dues peaceably.
His Excellency Sir John Herbert who visited the area during the settlement
of rents also pointed out that action was being taken to remedy their
grievances and impressed on them the necessity of maintaining the peace.
The warning was the more necessary because Congress and Communist agitators
were busy in the area. Happily, the sturdy independence of the aboriginals
ensured that these agitators met with little success.55
It was to the advantage of the colonial administration that the anti-communist
pro-colonialist missionaries worked closely with the Garos. Successfully so,
because the Garos did not take part in any of the communist rebellions, and
the Indian nationalist movement could not establish a stronghold among the
Garos. For example, in 1930, the Baptist Our Bond proudly mentioned that
Swarajists were unsuccessful with the Garos:
It is evident that the Garo people have no doubts as to whether they want
Swaraj [Self-Rule] or not. The other day an old man said, “While the
English live we live; when they die, we die with them.” The youthful
courageous Gandhi followers have failed in all their attempts to hold hartals
[general strikes], and arouse the non-co-operative spirit among them. An
amusing tale is told of how, a few weeks ago, a number of young braves saw
the mighty host of Garos in all their war paint approaching them, they fled
for life and nothing more was heard of their four days’ hartal.56
Missionary activities were beneficial to the colonial administration and the
administrative “reorganization” was helpful to the missionaries. The latter not
only took responsibility for healthcare and education, they also developed
other activities to improve the economic situation of the local population. It
remains rather unclear how much other non-Garo groupings benefited from
the special facilities and provisions in the region. After all, these activities
seemed primarily designed to support missionary ambitions to convert people
to Christianity, and to serve the state to contain anti-colonial conflicts. In
June 1937, the archbishop wrote the following:
At a very early stage in the training of our new Christians, the missionaries
urge them to contribute to the upkeep of religion, but so many are
unbelievably poor that it is literally impossible for them to support their
local catechist and schoolteacher…. With the inauguration of the Home
Rule administration, a special officer will occupy himself in safeguarding
the interests of the aborigines. In future it is hoped that these poor people
will not be deprived of lands that their hard labor has reclaimed from the
jungle.57
Missionaries evidently believed that they would benefit from an economically
more well-to-do Garo community and hoped that these improvements would
result from the protective measures implemented by the government.
Garos themselves probably benefited from this missionary-cum-state
union. The introduction of education offered new opportunities to gain
higher status and jobs (as teachers, nurses, pastors), and a new Garo elite
emerged whose ambitions were clearly linked to missionary education. We
shall see how in the 1920s and 1930s, a social reform movement was organized
which undertook different kinds of activities to raise social awareness.
Christianity also helped the Garos in another way: Garos stayed out of the
violent peasant uprisings which struck northern Bengal in the 1940s. Unlike
the Hajongs, they did not take part in the armed communist struggles of the
1940s. Consequently, the Garos were spared the dire consequences which the
Hajongs had to suffer when their movement was crushed.
THE ACHIK SHONGHO SOCIAL MOVEMENT

In the 1920s and 1930s, a number of young educated Christian Garo men
organized a Garo reform movement, the Achik Shongho [Achik organization].
As they would go around the villages to make Garos conscious of their social
and economic position and to encourage them to send their children to
school, they would sing the following song: “Wake up, Achik children! Why
are you still lost in sleep? Get up, Garo people, Stand boldly on your own
land, Send your children to get education, virtue and insight.”58
The Achik Shongho had been set up in the early 1920s.59 When its
founder, the college student Lolit, died of tuberculosis a few years later, the
activities stopped. Not for long, however, because in 1935 the organization
was brought back to life and proceeded with its work.
Members of the Achik Shongho were obviously conscious of the social
and economic situation of the Garos. Although the organization was clearly
not instigated by Christian missionaries, it was a rather elitist undertaking of
young educated Christian Garo men. This hampered the recruitment of new
members (at the time, few Garos were educated) and might explain why the
Achik Shongho never developed into an all-Garo movement. Most of our
interviewees recalled the organization, but had never joined it themselves. For
example, Bindu Patang had heard about it – he had even joined some of their
meetings – but was not quite sure about the purpose of the Achik Shongho:
“I didn’t know much about their intentions. Perhaps they wanted to do
something for the Achiks from Meghalaya? They had many meetings, many
lectures. I also joined some of their meetings. They said that we were
becoming poorer and poorer. That we had to do something against it, and
maintain our unity.”
In 1947, the situation changed. The movement openly protested against the
incorporation of the five thanas in East Pakistan and actively fought for the
amalgamation of the PEA with India.
GAROS AND THE PARTITION

Partition propelled Garos into political action. Arun Gagra, a former general
secretary of the Achik Shongho, remembered that moment as follows:
“When they decided to divide India, we became involved in politics. We
had to change all our objectives. We wanted to stay together with the Achiks
from the Garo Hills. That was the main issue…. Then we submitted a claim
to the Boundary Commission against the decision of the British government.
All our activities were concentrated around this issue.”
“We organized ourselves for this, consulted with our missionaries and the
Aboriginal Protection Board. This PEA was administered socially and
culturally by the Aboriginal Welfare and Protection Board of the PEA of
Mymensingh. Then we filed a memorandum, that our PEA should be
amalgamated with Assam. We didn’t want to remain in Pakistan because we
have most social similarity with the hill people. In every way we are similar
to the people from Assam. So this area had to amalgamate with Assam.
Then we filed a memorandum to the Boundary Commission.”
The Achik Shongho decided to take action against these plans to transform
the district boundary between Assam and Mymensingh into an international
border between Pakistan and India. Its members succeeded in raising enough
money among the villagers to send a Garo delegation to Calcutta. Monendra
was one of the Garos who joined the delegation:
“We sent our demands to the commissioner of the Division Committee.
Radcliffe was in charge of this area. He was in Calcutta, so we went there
to meet the committee. Me, Arun, and Kitinath went there, but there was
a big riot in Calcutta city. It was in 1947, just before the 15th of August,
Independence Day. It was very difficult to leave the city…. There was an
advocate who dealt with the objections about the partition. We met him
and he listened very carefully. He tried his best but did not manage to settle
the matter. Radcliffe said that this was a very small area without a special
boundary, so it was not possible to attach the area to India.”
The story is confirmed by an unpublished report of members of the Boundary
Commission. Two of its Muslim members commented about the issue of the
Partially Excluded Area in Mymensingh that
A claim has been made on behalf of a minor non-Muslim organization that
the non-Muslim portion of the Partially Excluded Areas located on the
northern side of Mymensingh district in East Bengal should be excluded
from East Bengal and added to the Garo Hills area of Assam. The main
ground for this claim is that this area is inhabited by tribes who have not
much in common with the residents of the remaining part of East Bengal,
but have racial, social, and economic ties with the tribes inhabiting the
Garo Hills.60
Although the Garos did not watch passively as partition took place, they
never formed one homogeneous block against the Boundary Commission.
Firstly, the Modhupur Garos were not involved in the Achik Shongho.
Contacts with those Garos were limited because of the absence of a good
communication system. Moreover, an eventual amalgamation of the five
partially excluded thanas with India would not include them anyway. Secondly,
Garos from Birisiri – the stronghold of the Australian Baptist mission – did
not link up with the Achik Shongho. According to one interviewee, Garos
from the western part of the five thanas supported the Achik Shongho,
whereas the Garos from the eastern part supported the Aboriginal Welfare
Board. Thirdly, the actions were taken entirely by lowland Garos. There is no
evidence that hill Garos were actively involved in any way. And finally, many
Garo villagers remained unaware of the situation. They did not realize what
was going on. On 15 August, Independence Day, the Garos of Bengal became
citizens of the newly established Islamic Republic of Pakistan. The Boundary
Commission had found no reason to take the Garo claim into consideration.61
The efforts of the Achik Shongho were in vain.
‘UNDESIRABLE ELEMENTS’: GAROS IN EAST PAKISTAN

This section explores the transition of the colonial state into the Islamic
Republic of Pakistan as it was experienced by the Garos. It focuses on how the
new state manifested itself to the Garos and demonstrates how Garos dealt
with the changes that came with the partition. Whereas the previous sections
were based largely on written documents – often produced by state
representatives or by missionaries – we have now entered the era about which
Garos have their personal memories. Their stories are the primary source for
the post-partition history and fill what would otherwise have been another
silence. They allow us to explore the relations between the Garos and the state
from the perspective of the Garos themselves.
Previous chapters have shown that today Garo perceptions of Muslim
Bengalis are generally quite negative. It is difficult to establish how these
feelings (which have also resulted from their past experiences) consciously or
unconsciously shape Garo memories or representations of the past. Another
complication is that our informants concentrated on four dramatic post-
partition moments in Garo history: 1950, 1964, 1971 and 1975. During
these years, they experienced some particularly dramatic events. In spite of
our urgings, interviewees talked much less about other, ordinary happenings
or developments. The Garos also left us with some comparatively silent
periods which could not be compensated by occasional references to
administrative accounts, interviews, missionary reports, or other sources. The
following sections follow the periodization which patterned Garo accounts of
the recent past in interview after interview.
The year 1947 marked both the end of British colonial rule in India and
the birth of the Islamic Republic of Pakistan, with East Bengal as the eastern
wing. A pro-Islamic government replaced the colonial government and the
status of Islam changed significantly. After all, the very existence of Pakistan
was based on the shared Muslim identity of the large majority of its citizens.
Partition ushered in a new political era in South Asia and its subsequent
changes were manifold and certainly did not leave the Garos untouched.
The partition of 1947 was perhaps the most dramatic experience of the
twentieth century for South Asia. The break-up of British India in 1947 led
to the death of at least half a million people and to the immediate relocation
of many more.62 The total number of refugees moving between India and
Pakistan between 1947 and 1951 is estimated at 14.5 million.63 While in
the Punjab the transfer of people was short and swift, the influx of Muslim
refugees from Eastern India into East Pakistan and of Hindus from East
Pakistan into India did not stop after the turbulent year of partition, but
continued for well over two decades.64 This ongoing movement of people
has been one of many long-term effects and shows that partition should be
perceived as a process rather than an event or moment. Its consequences
were numerous and stretched far beyond the immediate circumstances of
1947. Moreover, the events of 1947 should be viewed not only as an
outcome of conflicting social identities, but also as a factor in the subsequent
processes of identity formation.
1947–50
Partition affected the lives of the Garos profoundly, as overnight they became
a tiny minority in a Muslim dominated country. According to the 1951
census of the almost 300,000 Garos, only some 40,000 Garos were living in
the lowlands of the northernmost portion of Mymensingh district. These
40,000 Garos had been separated from the Garos of India by the newly
established international border. Although this was at first a relatively easy
border to cross, in the course of the following years the consequences of
partition, embodied in a closed border (since 1952 passports and visas have
been required to cross the border legally), became more and more apparent to
the Garos of what was now East Pakistan.
The effects of partition manifested themselves gradually, and the Garos
developed the full awareness of its consequences only in the years to come. It
was the violent repression of the Hajong uprisings, the flight of many
Hajongs to India in 1950, and the subsequent settlement of Bengali refugees
in the border region that confronted the Garos with the new situation.
The international border between Mymensingh (East Pakistan) and the

Garo Hills (India) has received little attention in the partition literature.
Documentation about the first months after partition is scarce, and only a
few elderly Garos had any clear recollections of the events of 1947. Burling
writes that “[i]t was, at first, a quiet border, for it escaped the ghastly violence
and exchange of population that occurred in West Pakistan and western
India, and it remained relatively easy to cross.”65 Other available information
suggests a picture of a rather unsettled border. Although it had escaped the
“ghastly violence” in 1947, it was far from a peaceful border and, as we shall
see, remained unsettled for many years to come. Ian Emmett, for example,
argues that “[i]If the border was settled on a political level, however, it
certainly was not settled in actual fact.”66
Although few Garos realized what was going on at the time of the
partition, now a sense of fear started to take hold of them. The news about
violent outbursts and communal riots everywhere in the country also reached
them, and rumours went around that Bengalis from the south were coming
to attack. People started to prepare for flight and sent some food and
belongings across the border. Others had already left. Malabika, a mother of
one son at the time of the partition, was one of them: “In that time all of us
wanted to flee away, and we brought rice to the Garo Hills. Some of us
decided to stay here, and they got back their rice. But we fled and stayed there
for six months. Some others only stayed a few months.”
On why she wanted to flee, she said, “Hindustan, Pakistan. Everyone was
afraid and fled away. We heard that many people from the south were coming
to attack us. That is why we became scared.”
It was fear and rumours rather than actual violence that induced these
people to leave. Arun Gagra, a former active member of the Achik Shongho,
left Pakistan in 1949. He is still living in India. Suborno visited him in
India to talk with him. Arun explained why he had left Pakistan so soon
after its genesis:
“I was feeling insecure there. I realized that there would be no peace in that
country. It would be very difficult for the minorities and the non-Muslims.
Their attitude towards the non-Muslims was very bad. That is why I
decided to come here, like the other people, I mean, like the other leaders.
All Achiks thought that they would not manage to live in East Pakistan.
After partition they started to receive news about the trouble with Muslims,
and there were many rumours as well. That is why many migrated to
India. Not all at the same time, like in 1964, but slowly many Achiks left
for India.”67
Flight was one consequence of fear; neglect of properties was another. Kubinath
explained how the idea that one day all Garos were to leave Pakistan settled
in their minds and discouraged them from taking their work seriously.
Monendra confirmed this. He also told us that the lives of the Garos changed
after 1947, and that Garos stopped taking proper care of their landed property:
“They were no longer interested in cultivating their lands. They were always
talking about their future, and some Mandis sold their lands to leave for
India. Earlier, Mandis had been working very hard in their fields. There had
been joy everywhere. But after partition they almost gave up.”
The question is, what made the Garos so scared? Why did they have no faith
in the Pakistani state? That they wanted to stay with those people, with whom
they felt social similarity with was one thing. That they started to lose faith in
their future was another. It seems that the shift of power had influenced their
minds and circumstances profoundly. Although the following fragment should
not be taken too literally, it gives a good impression of the present-day
perception of this shift of power. Nurudin related that:
“There were no Muslims here [before partition], only a few families in the
south. Some of them came to our village. Whenever we saw them in our
village we hit them. When a Mandi was wearing Muslim clothes, we
shouted at him: ‘Are you a Muslim? Why do you wear Muslim clothes?’
Now it is the opposite. Now they are hitting the Mandis all the time. We are
being punished. In those days, when we hit them, they could not protest.
But now they come to protest, with a lot of people. When the British
Government decided to leave this country, we were very worried because
then the Muslims would get the power. We knew that would not be good
for us. When the Pakistan sarkar [government] came into power the Muslims
started to move into this area.”
The birth of Pakistan also meant the birth of a new state. Old power holders
were replaced by new ones. The Garos had not been part of the establishment
before and did not become part of it now. Nevertheless, the shift of power did
weaken their position.
First of all, the protection which was embodied in the partial exclusion of
the border area came to an end. Even though these protective policies were
continued officially, they were no longer felt in real life. Many of the special
facilities and regulations were abandoned in practice. One Garo expressed it
as follows:
“Also after partition there was an officer for the PEA, but his activities were
against the tribal people. When Muslim refugees came to Bangladesh from
India, the special officer rehabilitated them in our area…. The rehabilitation
started after 1950, just after the Hajong riot. In 1964, it became big.”
“When did the government abolish the partial exclusion?” “They did not.
Only its function disappeared.”
Secondly, with the departure of the British, the Christian community lost
its backbone. The relationship between the missionaries and the new state
changed substantially. The available documents paint a picture of careful
Christian missionaries, openly professing a pro-Pakistan stand, and at the
same time facing a suspicious and sometimes openly agitated state. Both
Baptist and Catholic sources show how Christian missionaries and churches
attempted to prove their pro-Pakistan stand. The following quotation shows
how a leading Garo Baptist, Bilu Babu, paid his respect to the newly established
Pakistan state:
The Australian Baptist Mission’s field in East Bengal came within the
borders of East Pakistan. It was typical of Bilu Babu, that when the
government decision was made, he proceeded to give honour to the state in
a thoroughly Christian way. Indeed, when the first celebration of
independence came, it was Bilu who led a great procession with flags flying
and crying “Pakistan Zindabad”.68
Despite the missionary effort to demonstrate a pro-Pakistan attitude,69 the

Pakistan Government adopted a much more suspicious attitude towards the
missionaries. In 1952, three criminal cases involved foreign Catholic
missionaries in Haluaghat and Durgapur, who were accused of forcing
Muslim converts (Garos) to return to Christianity.70 The District Magistrate
commented that
These cases are the result of the Christian Missionaries’ trying to get back
to their religion the few aboriginals who have recently embraced Islam, by
threat, intimidation and use of force. Surely we cannot tolerate this. The
Christian missionaries must forget their old Mission Raj in the Partially
Excluded Area.71
In many ways the Garos had lost missionary backbone, because the missionaries
had lost theirs.
A third factor that brought a sense of fear and insecurity was the
discontinuation of the Achik Shongho and the emigration of most of its
leaders. It remains unclear whether Garos themselves decided to quit the
Achik Shongho or whether the government forbade their activities. It is
clear that the members of the movement felt unsafe, that they felt intimidated
by the government and the Muslim League. One informant told us that the
Achik Shongho was banned by the government, which was afraid of an
anti-Pakistan political movement. Interestingly, he blamed the Hajongs for
the trouble:
“Because of their communist movement, the government sent a lot of

police and army into this area. We did not manage to organize ourselves
again. The government was Muslim League. They did not allow us to have
any meeting in this area. Only the Muslim League could organize meetings.”
Another informant, Jonathan, also referred to the intimidating attitude of the
government and the Muslim League. He said that “[t]he Pakistan Government
made trouble with us and the Muslim League worked against our activities.”
However, he also explained that the government never really forbade the
Achik Shongho. The ultimate reason for its discontinuation was that “educated
people and leaders of the Garos fled to India, so we no longer had enough
organizers for the Achik Shongho.” Garos were clearly too afraid to continue
their activities. They felt intimidated and scared.
These developments all took place in the context of what was perhaps the
most unsettling event of the time: the peasant uprisings of the Hajongs. This
was the fourth and probably most important cause for the sense of fear and
insecurity. In the period 1930–50, Bengal witnessed a gulf of peasant uprisings
that collectively entered the history books as the Tebhaga movement. In
northern Mymensingh, a special type of sharecroppers’ resistance developed:
the Tanka movement of the Hajongs. The Hajongs started their protests in
1946 and continued until 1950, when the peasant guerrillas were defeated
and driven across the border into India. Muslim refugees then settled on the
lands of the Hajongs, which were offered to them at attractive rates.72
The violent ending of this uprising confronted the Garos with their
new position in the newly established country. Firstly, the violence against
the Hajongs worried the Garos. They feared being attacked, and like
before, some Garos again anticipated possible flight by sending food to
the other side of the border. Secondly, the exodus of the Hajongs led to
an influx of Bengali Muslims. The government allotted the lands of the
Hajong fugitives to Bengali refugees. Thirdly, because of the tension,
there was a lot of police in the region. A new threat was introduced. “Poor
villagers were harassed by marauding Communists and bandits by night
and by demanding police in the day.”73 A fourth possible reason, which
was never mentioned by Garo interviewees themselves, was poverty.
Badruddin Umar writes that: “Since the first days of Pakistan, there was
serious food crisis in most districts of East Bengal and at times it assumed
the proportions of famines. Famine, flood, epidemic and oppressive
taxation made the lives of the peasants miserable.”74
These four factors may have also affected the situation of Garos. In
conclusion we can say that the consequences of partition were not directly felt
in 1947 – Garos had no face-to-face confrontations with outbursts of
communal violence – but these manifested themselves more gradually and

indirectly. The Garo Hills, however, offered a close refuge. In 1952, the
international border became more real because of the implementation of the
passport and visa system. From that moment, the partition of the Garos
became a real fact. At the same time, peasant uprisings had been brought to
an end, the old landholding system had been abolished, and the state seemed
to forget the Garos again.
1952–63
Although the sparse data – occasional interview fragments, missionary
documents, and administrative notes – suggest that the situation in the
border area calmed down, life clearly did not return to the way it had been
before partition. Whereas Garo immigration and Garo emigration used to be
balanced in the late 1930s and early 1940s,75 after 1947 the situation changed.
Garos continued to leave the area, to India and to the tea gardens in Sylhet,
but I found no indication of hill people settling in the plains of Bengal.76
The question remains why Garos continued to leave after the situation in
the border had quieted down. Did feelings of fear linger on, or did most
people merely leave for economic reasons? The Catholic Chronicles indicate,
after all, that many Garos were living in poverty or “hand up”.77 One answer
to the question was poverty, but another one was “oppression of Refugees and
Thieves”.78 Thus, fear and insecurity continued to be important themes in the
lives of Bengal’s Garos. Until 1964, the year in which Garos fled the country
en masse, people quietly left one by one, family by family. In the words of the
elderly Garo woman Munti: “Since the Mandis did not all leave at the same
time, it was never noticed.”
On 3 November 1963, the chronicles report the increasing sense of fear
among the Garos:
The Garos are very much concerned these days on [sic] the activities of
some non-Garo people who are allegedly entering homes and helping
themselves to anything they can find. There are even stories that molesters
have forced Garos out of their homes, and moved in. Talk of leaving this
country and going to Assam is very common among our Garos now but we,
of course, are assuring the people there is absolutely no good reason for
them running off in fear. This is Garo country and it is going to stay that
way. This is what we told the members of the meeting.79
The Catholic missionaries of Biroidakuni worried about the situation in
their parish. Fear was getting a stronger grip on the Garos and people were
asking themselves whether they should stay or go. Missionaries and Garo
leaders attempted to talk the people out of it. They tried to persuade them
to wait. On 7 November, the archbishop came in person to see how bad the
situation had become. At the time his pep talks seemed to help. Nevertheless,
more and more stories were spread about Garos being troubled by illegal
settlers or being threatened by refugees who told them that all Christians
should leave Pakistan or become Muslim.80 In the course of January 1964,
the situation exploded. On 15 January, the chronicles report that Garos
from four villages and fifty Banai families had fled the country. The exodus
of almost all Garos from the border region was triggered.81 Only Durgapur
thana was spared.
1964: THE GREAT EXODUS

At the very beginning of 1964, northern Mymensingh witnessed a sudden
influx of Bengalis. Bengali refugees from Assam, followed by Bengalis from
places like Gafargaon, Kishorganj, Trisal, and Nandail, invaded the area. The
inflow started in 1963, but dramatically increased in early 1964, when South
Asia witnessed new outbursts of communal violence.82 In the wake of these
riots, East Pakistan experienced the arrival of at least 800,000 Indian Muslims,
who came mainly from West Bengal, and another 540,000 Muslim refugees
from West Bengal, Tripura, and Assam in December of that year.83
The arrival of Bengali newcomers coincided with thievery and intimidation
of the local non-Muslim population and with illegal settlements on the lands
of Garos and their non-Muslim neighbours. Rumours rapidly spread
throughout the border area that more Bengalis would come to rape and kill.
Within one month, almost all the Garos from the border area fled the
country (with the exception of the people from Durgapur thana). Haluaghat
thana was seriously affected by the disturbances. Villagers’ stories reveal that
much was happening at the same time: the influx of Muslim Bengali refugees
from India, the arrival of landless Muslim Bengalis from other places in
Bangladesh, illegal settlements on the land of local non-Muslim people,
robbery, the spread of rumours, and intimidation (with a strong communal
flavour) of the newly arrived Bengalis, local Bengalis, and representatives of
the state (East Pakistan Rifles, Ansar, police).
The influx of Bengalis frightened the Garos because they tried to occupy
Garo lands and robbed villagers, especially at night. At the same time, the
houses of neighbouring Hajongs, Dalus, and Banais were set on fire. At night,
Garos could see houses burning everywhere around their own villages. The
Garo woman Munti, described the situation as follows: “They set houses
around our village on fire. There were many Dalus in the west. They started
with their houses. Many people were burnt, together with their cows and
goats. In the evening we saw houses burning everywhere.”
Although this kind of violence was not inflicted on Garos, Garos did
suffer from theft and intimidation. Bengalis entered their villages with knives
and spears, took whatever they could carry. People were not only harassed by
the newly arrived refugees. Interviewees insisted that it was not the refugees
from Assam, but Bengali settlers from areas such as Gafargaon who committed
the worst atrocities. Moreover, local Bengalis were also involved. These days
Garos still distinguish between different sections of Muslim Bengalis: locals
(sthaniyo’s), refugees and Bengalis from Gafargaon (the latter are considered
the worst). To the question which Bengalis tried to chase away the Garos,
Monendra replied:
“I knew them. Some of them were from this area. They stayed around this
house whole day and night. At night they were shouting ‘Allah Akbar’ . …
At night I did not recognize them, but in the daytime, when they came to
loot my house I saw that I knew them.”
The villagers became more and more frightened and confused. Attempts to
persuade them to stay failed. Monendra explained why:
“When they looted the houses of Hajongs and Banais and set their villages
on fire, the Mandis became afraid. They thought that the same would
happen to them. The attitude of the Muslims was like that. We asked the
Mandis not to leave but some villagers from Kakorkandi left for India.
While going through Bekamara village, the EPR’s [East Pakistan Rifles]
opened fire at them and some of them died. When the news about this
incident spread all over the area, it was no longer possible to stop the
people.”
The news about Garos leaving set off a chain reaction. When they started to
flee, more Bengalis arrived, more houses were looted and lands occupied, and
more Garos fled. When the news reached the villages around Biroidakuni
Mission that the EPR (East Pakistan Rifles) had shot Garos on their flight,
these villagers gathered on the mission compound and decided to leave.
Within one day, on 5 February, almost all the Garos from this particular area
left, under the eyes of their Muslim Bengali neighbours, who were left
undisturbed. Karim watched his Garo neighbours leave:
“They got a lot of trouble. They tried to take things with them on a bullock
cart. They had to do it quickly. Many of them were crying because they had
to go to the other side of the border. They were looking very vulnerable.
They were watching the homes that they had to leave behind. They told us
that they would have to eat leaves from the trees.”
Sunil witnessed the situation from the beginning to the end. He was one of
the very few Garos who stayed behind. His story demonstrates how, as soon
as the people had fled, the area was occupied by more Bengali settlers. Within
a day and a night, the border population had become almost totally Bengali:
“When the people fled, my family went to Biroi. I stayed here to watch our
house. In the evening it became so quiet that I did not dare to stay alone,
and I went to Biroi as well. The next day we found all surrounding houses
and villages occupied by refugees. Not a single house was left unoccupied.
The same day, Bengalis from Trisal, Nandail, Kishorganj and Gafargaon
arrived by buses and trucks. I don’t know how they got the news so quickly.
The whole day they continued to arrive.”
‘GOING IN CIRCLES’
When they arrived in India, the Garos were sent to refugee camps. Our
interviewees ended up in Dindini and Matia.84 Many of them never returned
to Pakistan. The circumstances in the camps were bad. Many people fell ill
and died. Local Garos who were working in the camps encouraged the Garo
refugees to stay, and both missionaries and the government took steps to
rehabilitate the refugees. They received financial support from Christian
organizations such as the World Baptist Alliance, Australian Baptist
Missionaries, Oxfam, and the World Christian Council. Nevertheless, many
did not want to stay in India. After two or three months, people started to
return slowly and hesitantly. The chronicles reported how people were going
about uncertainly. The mentioned that “there is a state of restlessness very
noticeable in those who have returned to stay and in those who went back to
the hills. They seem to be going in circles.”85
Before and during the rainy season, only a few Garos returned, and in
October of the same year, the chronicles reported that out of the 17,000
Garos who lived in Biroidakuni parish before the big exodus, some 10,000
people were still in India. One month later, in November, it was estimated
that half of the Catholic population of Biroidakuni had returned.86 The
interviews reveal that Garos also continued to come back in the following
year. After November 1964, however, the chronicles stopped reporting. No
one knows exactly how many refugees finally returned.
Why did so many Garos come back to Pakistan? After all, they had
been living with feelings of insecurity and fear for many years. Moreover,
missionaries and the Indian government offered them support and local
Garo leaders and government officers asked the refugees to stay. Nevertheless,
Garos decided to return to Pakistan for many reasons. One important
reason was the situation in the camps. People were afraid of the diseases:
they were afraid to die in the camps. Secondly, there was the news that the
situation in Mymensingh had improved significantly. Many Garos came
back because of their property or returned during the harvest to see to their
crops. Thirdly, in the words of a Garo woman, “It was our motherland, and
we had lands here. That is why we came back.” And fourthly, people
followed their relatives. Poben, who had received land from the Indian
Government decided to return for that very reason: “In Matia, I received
eight bighas of land from the Indian Government for rehabilitation, but my
father, mother, and other relatives had already managed to come back here.
That is why we decided to return as well.”
A fifth, quite remarkable reason was that the Pakistan Government asked
the Garos to return. The government literally called the refugees back on
loudspeakers which they had installed on the border. There was no such thing
as a consistent, homogeneous, unambiguous attitude of the state towards the
Garos during these years.
THE ATTITUDE OF THE STATE IN THE EARLY 1960S

Garos refer to the 1963–64 events as a conscious attempt of the Pakistan
Government to chase the minorities out of the country. Samuel talked about
“a plan to drive the tribals out of this area”, and, according to Kubinath, the
government wanted to evict the Garos after they had kicked out the Hajongs.
Monendra contended that the government was behind the anti-Garo actions
of the EPR and he believed that the government had instigated the lootings
and suppression of Garos and other minorities in the border area. He described
the visit of some high government officials before the great exodus:
“The DC [Deputy Commissioner] said that there were no signs of a riot:
cucumber and bean plants were in tact, so nothing had happened. I
wanted to object, but the SP [Superintendent of Police] showed me his
baton. He told me that if I wanted to stay in Pakistan, I had to keep quiet.
Otherwise I could go to India and eat stones in the hills…. When the DC
and SP came here, but did not listen to the people and instead declared
that nothing had happened, people suspected the government behind
everything, and they fled.”
Earlier we saw how a number of Garos distrusted the Pakistan state right from
its outset. They were very sceptical about the future of Garos in an Islamic
country. The intense reaction of Garos to the events in 1963 and 1964 clearly
corresponded with these feelings of scepticism and fear. The occurrences
confirmed the expectations which had for long been lingering on.
“After partition we knew that it would become difficult for non-Muslim

people to live in East Pakistan. The leaders of the Muslim League spoke
openly that everyone in East Pakistan had to become Muslim, or otherwise
leave the country. The EPR came to the border area. They entered Garo
villages arbitrarily and stole chicken, fruits from the trees, anything they
needed. Sometimes they intimidated Garos by saying that this is a Muslim
country.”
This quotation shows how fear was occasionally being stirred by state
representatives. In these same years, the population of the Chittagong Hill
Tracts faced a repressive attitude of the state. Interestingly, in 1964, the
special status of the Chittagong Hill Tracts, the only region in East Bengal
with the status of Excluded Area, was abolished, and the hills were opened up
to outsiders. Bengali families and refugees from India settled in the hills in
great numbers.87 This suggests that the 1964 events in northern Mymensingh
should be considered as part of an all-East-Pakistan policy to Islamize the
country by opening up the whole of East Pakistan to Muslim Bengalis and
refugees from India at the expense of the non-Muslim peasantry.
Interviews and the Catholic Chronicles reveal that it is important to
distinguish between state leaders and national political developments and
interests, on the one hand, and local state institutions and representatives and
their actions, on the other. The available data reveal that state agencies such
as the EPR, Ansar, and the police, played an active role in the suppression and
intimidation of the Garos. The EPR in particular, played an active role in the
subjugation of the non-Muslim population of northern Mymensingh. For
example, there was the incident in Bekamara where the EPRs fired at a line
of Garo refugees, wounding and even killing several people. EPR and Ansars
had a clearly pro-Muslim character. They intimidated non-Muslims and told
them that all minorities had to leave.88 The chronicles reported for example
the following incident:
Our Gobrakura master, Amorendra Tiggidi, sent a very discouraging letter
here this afternoon. He had sold thirteen maunds of rice in Haluaghat, and
had just about arrived at his house when some EPR’s and Ansars stopped
him, and told him that he was under arrest! They proceeded to search him
and took the 150 rupees he had just received from the sale of his rice. They
then told him that all the Christians would have to go to India.89
While such was the attitude of state representatives at the local level, leniency
at the central state level clearly allowed the situation to escalate. For example,
after his visit to the border area, the archbishop informed the government
about the critical situation in the border area. In his yearly Easter message in
March 1964, the archbishop wrote that:
I was aware of the danger long ago, and I warned the Government of what
was likely to happen if strict measures were not taken to stop these injustices.
Unfortunately, my warnings were not heeded. I have spent a great deal of
time during these months in the border area, trying to keep our people from
going away. You would not believe that such things could happen in such
a short time.90
The attitude of the Pakistan state was evidently detrimental to the position of
East Pakistan’s minorities. They had been aware of the problems in the border
but denied the seriousness of the situation and refused to take action. Thus,
whether or not the government was actively involved in the intimidation of
Garos and other non-Muslims, they did not find it necessary to act upon the
situation either. Interestingly, there was a sudden change in the state attitude
after the exodus. Police were sent into the area to protect the people who had
stayed behind, a special officer was appointed to transfer property to the
rightful owners as soon as they returned, and the government installed
loudspeakers on the border to call the refugees back. Sunil described the
situation as follows: “After a few months the government sent a special officer
who had to receive the Mandis who came back from India. The government
also sent police to protect the Mandis who had not fled away, or who had
come back already.”
The government attitude towards the Garos and other non-Muslims only
seemingly improved. State leaders had probably succumbed to international
pressure. India had to deal with tens of thousands of refugees, the archbishop
revealed the problems to the world’s Christian community, and organizations
like the World Christian Council and Oxfam supported Garo returnees. On
other levels, however, a more aggressive state attitude towards the non-
Muslim population of northern Mymensingh developed.
First, many Garo returnees found their houses and lands occupied and
other possessions stolen. The problems they encountered led to the
impoverishment of many. In need of cash money, they had to sell pieces of land
that they managed to get back in their possession. Rabindra’s story is illustrative:
“Some of us got their land and property back; some of us did not and had
to start from scratch. We needed to buy cows to plough the land and to
build houses. Many took loans from the mahajans. They had to pay high
rates of interest. For 100 taka, they had to pay eighteen to twenty maunds
of rice per year. Some did not manage to pay the money back and they had
to give land to the mahajans. In this way the economic situation of the
Garos started to deteriorate.”
A second problem was the acceptance of the “Enemy Property Ordinance” by
the government. This ordinance determined that the property of Indian
nationals and those residing in India came under the control of the Pakistan
Government. People who had found their lands declared “Enemy Property”
had to spend a great deal of money on court cases against this government
ordinance. The land of those who did not return continued to be occupied by
Bengalis, whether illegally or under the ordinance. It has been argued that the
law was strategically (mis)used against all non-Muslim inhabitants of East
Pakistan and later Bangladesh.91
A third development also suggested a less friendly attitude towards the
non-Muslim population of the border area: in 1965, foreign missionaries
were no longer allowed to work in the border area. In practice, this meant
that they could no longer work with the Garos of Mymensingh.
In conclusion, we can say that the role of the state in the events of 1964
and its aftermath was ambivalent and complex. State agencies such as the
EPR, Ansar, and the police took a clearly pro-Muslim stand. They not only
let Bengali immigrants have it their way, they were actively involved in the
intimidation and harassment of the local non-Muslim population. The central
government did nothing to prevent the situation from escalating. Their
attitude changed after the big exodus, probably under a lot of international
pressure. In many other ways, however, state policies remained repressive, and
the Pakistan state continued its Islamization projects. Contemporary Garos
often point to 1964 as the turning point between good and bad times,
between the days when they were left to themselves and the days of Bengali
Muslim domination. In reality, this process had already been going on for
many years. The events of 1964 had, however, a major impact on the Garos
of East Pakistan. According to Keta, it was only then that Garos put aside
their differences and realized that they were one and the same people.
GAROS OF BANGLADESH
After East Pakistan’s Sheikh Mujibur Rahman won the elections that would
have made him prime minister of Pakistan, the Pakistan army came into
action. On the night of 26 March 1971, a nine-month war began in which
between 300,000 and one million East Pakistanis died. In December, when
the Pakistan army surrendered, a free Bangladesh emerged. This liberation
war is generally perceived and presented as the Bengali struggle for
independence, with Bangladesh, the country of Bengalis, as its result. This
representation of the liberation war excludes people like the Garos, who also
joined the freedom struggle and had as many hopes and dreams about a
country freed from Pakistani domination as the Bengalis had. But in the
heroic stories about the struggle for freedom which legitimize the very existence
of Bangladesh, no role has been given to any other group than the Bengalis.
Peoples like the Garos, Koches, Hajongs, and Dalus have no place in this
nationalist discourse. In this section, I shall explore the meaning and
implications of the birth of Bangladesh for the Garos. I shall demonstrate
how, from the very beginning, the exclusion from this national discourse led
to further “othering” and exclusion of non-Bengali citizens.
It is interesting to note how the power holders in Bangladesh have
availed themselves alternately of ethnicity (Bengaliness) and religion (Islam)
as the most important ingredient of their national discourses.92 Whereas
Pakistan was based on the Islamic identity of the majority of its population,
Bangladesh was initially founded on the Bengali ethnic identity of the large
majority of Bangladeshis. Sheikh Mujib wanted to keep religion out of
politics, but stressed the Bengali identity of the population of Bangladesh.
The citizens of Bangladesh were accordingly referred to as Bengalis. Soon
after the liberation, however, the country witnessed growing anti-Indian
sentiments among different strata of the population, and the Islamic identity
of the population received more stress as the main characteristic of the
people of Bangladesh. When, shortly after the assassination of Sheikh
Mujib, Ziaur Rahman became the new president, citizens of Bangladesh
were referred to as Bangladeshis, with Islam as the central national symbol.
Although based on different mixtures of ingredients, neither one of the two
available nationalist discourses left any space for the non-Bengali, non-
Muslim inhabitants of Bangladesh. This section reveals not only some
factual omissions in these national discourses but, more importantly,
investigates the consequences of both exclusive nationalist discourses for
the identity and organization of the Garos of Bangladesh.
1971–75: DREAMS AND DISILLUSIONS

The initial phase of the liberation war in March 1971, closely resembled the
1964 situation. Garos fled the country soon after the crackdown in Dhaka.
With memories of 1964 fresh in their minds, they left before the border area
turned into a war zone. Anew, fear and rumours rather than violence induced
the Garos to leave, and again, all the villagers of Haluaghat thana left on the
same day. As in 1964, local Bengalis used the opportunity to loot non-
Bengalis. Kalan, who later joined the Mukti Bahini or freedom fighters, told
us that the plundering in Sherpur thana started as soon as the Punjabis93
arrived in Sherpur town: “When the Punjabis arrived in Sherpur town, the
Bangals started to loot the Garos and they took away everything; when the
Garos fled to India, they robbed them.”
Again Garos were confronted with aggression on the part of Bengali

Muslims in the area, and right after people fled, Bengalis occupied their
houses. Interestingly, the looting started as soon as the Pakistani army reached
the area. Apparently, the plunderers expected their actions to be condoned by
the state, as had been the case in 1964. They obviously did not fear for their
own lives at the time. Nevertheless, in two or three months the border area
turned into a real war zone, with the Indian army on one side of the border
and the Pakistan army on the other. Bengali neighbours joined the Garos in
the same refugee camps.
Soon after the war had started, many young Garo men joined the
Mukti Bahini. A recent inventory of freedom fighters from Haluaghat
thana lists 330 names, more than one third (118) of whom were Garos.94
Members from other minorities of northern Mymensingh, such as Koches,
Banais, and Hajongs, also joined the freedom fighters. Apparently, the
relations between Bengalis and non-Bengalis of the Mukti Bahini in this
part of East Pakistanis were much better than in the Chittagong Hill Tracts.
In an account of his personal experiences, a Chakma from the Chittagong
Hill Tracts wrote how Bengali leaders distrusted tribal participation soon
after the liberation war had started. The fact that tribal leaders extended
their full cooperation to the liberation force did not, apparently, take away
the distrust. Chakma recruits were not allowed to practise with real guns
but were given wooden rifles instead, no tribal leaders received a place
among the freedom fighters’ leadership, and in Agartala, where the Awami
League had opened an office to recruit new volunteers, Chakmas and other
peoples from the Hill Tracts were turned away.95
How can we explain this clearly different attitude towards the people
from the Hill Tracts? Sheikh Mujib’s Awami League had not won any seat in
the Chittagong Hill Tracts during the national elections of March 1971.
Secondly, a number of hill people joined the Pakistani army during the war
of 1971. Even today Bengalis sometimes refer to the Chakmas as razakaars (as
collaborators with the Pakistan army have been called).96
The situation was quite different in the north of the country. One Garo
informant told us how, at the very beginning of the war, the government-in-
exile approached the Garos to join the freedom fighters:
In that meeting, the Garos were requested to join the war. The Garo
leaders asked them what advantages they would be getting after the war.
The reply was that all demands would be fulfilled. After several meetings,
the Garos had still not managed to agree on the demands. They were
politically divided.
Apparently Garos had their own objectives: they wanted their participation to
result in some special benefits and, in this way, set themselves apart from the
Bengalis. They did not demand a fair share in the future power, but seemed
more focused on the local than national context and gave priority to a sense
of autonomy in their local area. One demand that was frequently repeated in
the discussions in the refugee camps was “autonomy”. By this they meant an
autonomous position of the PEA (the five thanas of Partially Excluded Area).
Nevertheless, when the war was over, the Garos still had not reached consensus
among themselves and had not entered into any agreement with the
government-in-exile.
GARO FREEDOM FIGHTERS

The actual recruitment of volunteers took place in the refugee camps. Arthur
Drong, born in a village in Jinaighati thana, Sherpur district, joined the
freedom fighters at the age of seventeen. His younger brother came along.
Arthur was still a college student in Mymensingh when the war broke out.
When that happened, he went home, and in April or May he fled to India.
Arthur cited several reasons for joining the freedom fighters, but stressed that
emotions were his prime mover. He explained how, in the refugee camp, he
had felt the excitement in the air and how this thrill had made him join the
force: “We joined the freedom fighters for no other reasons than other people;
it was quite emotional.” A second reason was the dream of a better future in
a new country. Arthur blamed the Pakistan army for the exodus of 1964, and
their defeat would perhaps result in a better future for the Garos:
“We wanted an independent country. We were exploited by the Pakistanis.
We wanted to live in this country with the dignity of citizens of a free
country. In those days they [Pakistanis] did not recruit the adivasis in their
army or in the police force; they totally ignored the adivasis. Another thing
was that we wanted to prove our feelings for the country, that we also loved
the country. In that way we wanted to show other people that we were also
citizens of this country.”
The volunteers received a crash course from the Indian army in the Garo
Hills. Arthur joined a group of some 500 volunteers, including some 150
Garos, Koches, Hajongs, and Banais. After the training, they were divided
into two groups: one was sent to Netrakona and the other to Dalu, in India.
Arthur joined the second group. Of the 150 freedom fighters in his group, 50
were Garos, who formed their own platoon under the leadership of a Garo
commander. They were assigned tasks such as hit-and-run, border patrol, and
the attack of enemy camps. Arthur estimated the total number of Garos and
other non-Bengali volunteers at 1,500.
After the war, these freedom fighters never organized themselves in the
way their Bengali comrades did. No one knows exactly how many of them
joined the freedom fighters. This is particularly interesting since freedom
fighters became real heroes after the war was over, and until this day they
receive some special benefits, such as a 30 per cent reservation in the Civil
Service.97 Nevertheless, Garo freedom fighters made their first public
appearance only in 1996, during the twenty-five-year commemoration of the
liberation war. Later we shall see that many freedom fighters joined the
guerrilla force of Kader Siddiqui in 1975. Perhaps memories of the Kader
Bahini long overshadowed the memories of the liberation war? Or did they
remain silent because liberation did not bring what the Garos had anticipated?
DREAMS
Garos chose the side of the freedom fighters and dreamt of a country freed
from Pakistani domination. When in December 1971, the Indian army
administered the final blow to the Pakistanis, and Bangladesh became an
independent country, many Garos were eager to return to their villages. This
time they did not return only to their villages, houses, lands and relatives, but
to a new country that held the promise of a better life.
During their stay in India, they had been contemplating their position in
a country free from Pakistani domination. Would that country offer them
new opportunities, a better position? Would the new leaders approach the
non-Muslim non-Bengali citizens differently? Garo leaders gathered in several
meetings and discussed the future of the Garos, but never managed to
formulate a single plan of action. One group of Garos thought it safest to stay
in India. A second group wanted to strike a deal with the new government
and demand some special provisions, such as some kind of autonomy. A third
group was willing to give the government-to-be the benefit of the doubt.
They held high expectations for the future. That the Garos were not able to
agree on a common strategy also signifies the lack of a central leadership
among them. Until the very end of the liberation war, the Garos remained
divided. When the time to return had come, they had not entered any special
agreement with the new government of Bangladesh.
DISILLUSIONS
The refugees returned to their homes, only to find them destroyed. In
1964, their possessions had been stolen and their houses and lands occupied.
This time the devastation was far worse.98 Robindra described the situation
as follows:
“After the liberation we came back and found only jungle around our
houses. Only some frames of houses had survived. Corrugated iron from
the roofs and all other parts of the houses had been taken away a long time
before. The straw houses had been destroyed by the Pakistani army. When
we came back I had to clean the jungle, and there was no house to live in.
We lived under the trees until I built a small house.”
Relief started to pour into the country and NGOs started to mushroom; this
was a seemingly beneficial development for the Garos, who found important
supporters and employers in the Christian NGOs. Nevertheless, it was not
poverty that took away their dreams. In many ways the 1964 events had been
much more traumatic, because at that time Garos and their non-Muslim
neighbours had been victims of a conscious state policy. This time everyone
was the victim of a common enemy: the Pakistani army.
Nevertheless, it soon became clear that Sheikh Mujib favoured a democratic
country on the basis of secular principles and the ethnic Bengali identity, and
had no intentions to officially and ideologically turn Bangladesh into a multi-
ethnic country. In 1972, a Chakma delegation met Mujib to discuss the
demands of the inhabitants of the Chittagong Hill Tracts. Their demands
included autonomy and a special legislature for the Hill Tracts.99 Mujib
simply refused to accept any of the demands of the hill people and “angrily
threatened to drown the tribals in a flood of people from the plains….”100
During a visit to Rangamati in 1975, Sheikh Mujib addressed the crowd as
brethren. He told them to become Bengalis, suggesting they forget the
colonial past, and asked them to join mainstream Bengali culture. The crowd
then left the scene, to which Sheikh Mujib responded with the threat to send
the army and Bengali settlers into the hills.101
Northern Mymensingh never experienced a similar situation of repression
and warfare as the hill people of the Chittagong Hill Tracts would experience
in the period since 1975, but its non-Bengali population was soon to learn
about Sheikh Mujib’s attitude towards the ethnic minorities. Samuel reported
the following:
“Again we came back. There was no rehabilitation programme of the
government. Only Caritas and some other organizations helped the people.
Some Garos went to see Sheikh Mujibur Rahman. They told him: ‘We are
tribals so we need special care from the government.’ They presented him
millam-spie [Garo shield and sword] as a souvenir. Sheik Mujib said to
them: ‘All people here are Bengali.’ The Garos told him that they needed
special protection, but he refused. He told us that we are Bengalis, and said
‘You do not need any special privileges.’ ”
Whereas the Pakistan state applied a policy of Islamization that excluded all
other non-Muslim communities, the newly established state of Bangladesh
developed another exclusivist policy. In 1973, Sheikh Mujib stated that the
ethnic minorities would be promoted to the status of Bengali, and in 1974,
parliament passed a bill that declared Bangladesh as a “uni-cultural and uni-
linguistic nation-state”.102
THE GAROS AND KADER SIDDIQUI

It is in this context of disillusionment rather than as an expression of pro-
Mujib sentiments that we should explain the Garo participation in the Kader
Bahini [Kader’s Fighters], a force which held northern Mymensingh’s border
area in its grip from late 1975 to early 1977.
After the assassination of Sheikh Mujib, on 15 August 1975, Kader
Siddiqui, one of Mujib’s followers and a well-known freedom fighter, fled to
northern Mymensingh and started an insurrection along the northern border.
He recruited many young Garo men in his guerrilla force. According to my
interviewees, they constituted the majority of Kader’s guerrillas.103 In the
history of Bangladesh, this movement is nothing but a small footnote, but the
guerrilla war stirred northern Mymensingh for almost two years and greatly
influenced the lives of a large section of Mymensingh’s Garos. Garos have
clear recollections of the period 1975–77. Besides the interviews with Garos
– guerrilla fighters and others – I found a limited number of written references
to the movement.104 The Biroidakuni Chronicles provided a third type of
source for the following account.
Kader Siddiqui arrived in Haluaghat bazaar in October 1975 and one of
his first actions was to attack the police station. On 17 October, the chronicles
reported that “[a]bout an hour fight [sic] continued – some police are
wounded”.105 The Garo woman Cecilia, who lives in Bibalgree, described the
situation as follows:
“He [Kader] arrived in Haluaghat at night. There he fought with the police.
After that he took shelter with a Garo family in Gobrakura. The next day
BDRs [Bangladesh Rifles, the border force of Bangladesh] entered our
village, shouting: ‘You Garos offered shelter to Kader!’ I was at home for the
holidays. They came into the village and hit someone. We had no idea
about Kader then.”
With the attack on the thana of Haluaghat, a guerrilla war of a year and a half
started between the Kader Bahini and the Bangladesh Rifles. Soon after the
fighting started, many Garos fled to the border.106 People had never heard of
Kader before and did not understand what was going on.107 The Garos got
caught in the firing line again, literally and figuratively.
Kader Siddiqui is described as a Mujibite (follower of Sheikh Mujib).108
“Mujibite” is not, however a proper description of the Garo guerrillas. After
all, Sheikh Mujib’s attitude and ideas had disappointed many Garos.
Nevertheless, many young men joined the Kader Bahini and it seems that
Garos constituted the majority of Kader’s force. First, the liberation war
had taken its social and economic toll. It had caused great damage to the
country and its people. Many Garos (as well as Bengalis) had lost everything
because of the war. The people were becoming poorer day by day.109
Second, jobs were hard to get and Garos were not accepted in the army,
BDR, or police force. Third, there was the realization of former Garo
freedom fighters that their contributions had not brought them any benefits
individually or collectively. Frustration was an important motive to join
Kader Siddiqui. Finally there was the issue of the “autonomous PEA”.
Kader supposedly promised the Garos to fulfil all their demands, including
autonomy. Interestingly, Kader arrived at a time when Garos were in the
middle of organizing their own resistance:
“The [liberation] war put the Garos in a hopeful mood; they thought that
this would change their situation, but they found out that nothing changed.
In the meantime, the Chakmas had started their fight against the government.
This inspired the Garos to start an underground movement. A number of
Garo leaders held several meetings in Haluaghat to discuss it. They did not
manage to hide their activities from other Garos. Somehow, all Garos came
to know about it and there were a lot of rumours. It was such a pity that at
that same time Mujib was killed and the political situation changed overnight.
Kader Siddiqui and some followers came to this area and started to attack
BDR camps. It was no longer possible to organize a meeting to decide on
what to do next.”
CAUGHT IN THE FIRING LINE

Kader’s arrival in northern Mymensingh created confusion among the Garos.
Those who had initiated the Garo movement lost contact with each other.
There was no leader to take charge and the Garos had to make their decisions
individually. A number of them decided to join Kader Siddiqui. A the same
time there was the aggression directed against those who stayed behind:
“We thought that it would no longer be safe to stay in Bangladesh. And
things were going on that scared us. One Garo guy had been arrested by
the army and beaten up very badly when he came back from the bazaar.
They had left him at the side of the road, tied to a tree. He managed to
free himself and escape. After hearing this story we decided to flee and to
join Kader.”
For the Garos who did not join the Kader Bahini, the situation soon
became very difficult. Again the area turned into a real war zone. At night
the villagers found themselves in the middle of the crossfire between the
guerrilla fighters and the BDR. During the day, the police and the BDR
harassed and threatened the villagers. All Garos were suspected of Kader
Bahini sympathies and young men were being arrested. Then there was the
looting by guerrilla fighters, the BDR, and police. The few enthusiasts who
had joined Kader voluntarily were soon followed by many more young
Garos who felt there really was no other option.
The recruits received training and were located in several camps along
the Indian border. From there they made their attacks on BDR camps in
Bangladesh. Although they received rations from the Indian Border Security
Force (BSF) – who brought those in by trucks – the support was meagre
and soon the guerrilla fighters fell short of food and weapons. They
started to loot the villagers in northern Mymensingh, both Bengalis and
Garos. One ex-guerrilla fighter explained to us why the Garos robbed
their own people:
“It was very difficult to starve like this, carrying weapons with us. We came
here to ask the people to give us food, but since we had guns they thought
that we were robbing them. We did all those things because we believed in
Kader’s second revolution. If he would succeed, our basic needs would
surely be fulfilled.”
“The fight of marginalized groups often gets entangled in the larger politics
of nation-states, and leaves these people fighting for their rights with little
room to manoeuvre.”110 Such was the case for the hill people of the Chittagong
Hill Tracts, but equally applied to the Garos of northern Mymensingh. The
Kader Bahini was used in the power-play between India and Bangladesh.
India supported Kader’s actions against Bangladesh. The raids on the
Bangladesh border posts were launched with the help of the Indian BSF and
the guerrilla fighters were (insufficiently) fed with rations provided by the
BSF.111 Garos insisted that they were furnished weapons by the Indian
government. A similar thing was going on with the people of the Chittagong
Hills, who received arms from India in November 1975.112 In 1976, Talukder
Maniruzzaman wrote that “[t]here have been reports that New Delhi has
been helping pro-Mujib and pro-Soviet forces (who had crossed the border
into India after the 15 August coup) in making preparations for launching
another guerrilla war in Bangladesh.”113 That the Kader Bahini was totally
dependent on Indian support became very clear when Indira Gandhi lost the
elections in 1977; India immediately discontinued its support, and the Kader
Bahini came to an end.
‘WE ARE THE CHILDREN OF BANGLADESH’

“Chinga Bangla asongni, Achik pisarang, Chinga Achik pisarang” [We are
the Achik children of Bangladesh, we are the Achik children]
This song, which was probably written only one decade ago by a Bangladeshi
Garo, was frequently sung by the children of Bibalgree. It reflects the feeling
which is currently shared by many Garos that their homes are in Bangladesh,
that Bangladesh is the country where they belong. An interesting detail is that
the song refers to the Garo children as Achik, and not as Mandi children. This
suggests the inclusion of Indian Garos in the “imagined community” of
Garos. Here we will see that, in reality, Indian and Bangladeshi Garos are
increasingly growing apart.
The last section of this chapter briefly explores the post-Kader period. It
examines how Garos have become incorporated in the national state and how
this has resulted in two seemingly opposite developments: Bengalization and
ethnicization. Compared to the first three decades after the partition, the
post-Kader Bahini period has been relatively peaceful and stable. The state
has shown less overt aggression towards the Garos and, at the same time, the
Garos have become firmly embedded in the state.114
The Kader Bahini guerrilla fighting instigated a number of changes
which attributed to the formal participation of the Garos in the state. Other
recent developments which have significantly influenced the Garos of
Bangladesh since the second half of the twentieth century are: migration,
urbanization, education, political participation, the improvement of
infrastructure and means of communication, international aid and interference,
and other globalization processes.
In 1977, before the ending of the Kader Bahini, state leaders summoned
Garo leaders for several meetings to discuss a solution to the fighting. In these
meetings, they acquiesced to the Garo demands to employ Garos in the army,
BDR, and police force. Until then, Garos had been excluded from these
professions. One Garo leader reported that:
“The BDR general came to Joyramkura by helicopter. He wanted to meet
the Garo leaders. He said to us: ‘You are human beings, we are also human
beings. You are Bangladeshis, we are too.’ Jonathan babu then replied: ‘Sir,
you are talking nicely but we cannot digest it. You speak nicely, but behind
our backs you are cutting our roots.’ Then he asked us if we could provide
him with 300 people, and we said that we could, of course.”
A Bengali Catholic priest, who was posted in Biroidakuni Mission during the
Kader Bahini fighting and who had been threatened several times by the
guerrilla fighters, was also asked to participate in one of these gatherings. He
told us that:
“The leader [of the BDRs] said that he would recruit them, but that the
fathers had to sign. I can’t remember how many signatures I put. Many
Garo boys came to join the BDR. The condition was that they had
studied up to class VIII, have such and so height, such and so chest. Then
the general told me not to worry about their class level. If they would
manage to write their name, I could send them. We sent a lot of boys. Still
quite a few are there.”
In that same year, the government also founded the Tribal Academy, to
promote tribal culture and they made a legal and statutory provision for a
tribal platform, the Tribal Welfare Association (TWA).115 Although neither of
these institutions are faring very well, they are still in operation today.
The improvement of the infrastructure and means of communication
have made the world increasingly smaller. Cities like Mymensingh and even
Dhaka can now easily be reached within a couple of hours and contacts
between the migrants and their relatives in the villages are much easier to
maintain. At present, we can find Garos all over Bangladesh. Many migrants
are found in Dhaka, where they have various jobs, ranging from rickshaw
driver to household servant, and NGO-worker to Catholic nun.
Education has also contributed to the participation of Garos in the state.
A growing number of Garos are doing well in universities and colleges. They
also participate in the Civil Service Examinations. Recently, the first Garo
passed the Civil Service Examinations to enter the higher echelons in the
police force. At present there are Garo physicians, civil servants, army personnel,
nurses, businessmen, NGO workers, clergy, and teachers.
Particularly significant was the election of a Garo candidate, Promod
Mankin, in the national parliament in 1991. His party, the Awami League,
lost those elections, however, and the candidate ended in the opposition.
When the Awami League won the elections in 1996, the Garo candidate
himself lost the local elections and another local representative took place
in the national parliament. At present, the Awami League is again in the
opposition. Nevertheless, to have a political representative at the national
level is very important for the Garos in that it contributes to their sense of
pride, self-consciousness, and strength. During the last elections, numerous
Garos actively participated in the local campaigning, in favour of their
Garo MP.
Economically, Garos do not fare worse than their Bengali neighbours.

Through education, migration, urbanization, and participation in the market
economy, an educated middle class has developed. Although many Garos
have been forced to sell land, particularly after the 1964 and 1971 experiences,
to cover their material losses, we can also witness a reverse tendency of well-
off Dhaka dwellers buying up land from poor farmers, with the aim to return
to these lands after retirement. Thus, whereas a number of successful families
are faring better and better, numerous other families have to sell land, search
for work elsewhere, or work as day-labourers on the lands of others, like so
many of their Bengali fellow countrymen.
All these developments have affected the relations with the Bengali
population. Numerically insignificant in the border region, Garos have
produced several influential members. My hostess in Bibalgree, for example,
is the headmistress of a local high school. She is a respected member of the
local community, of Garos and Bengalis. This applied even more to the local
Garo candidate for the Awami League. His house was always crowded with
locals seeking his attention or support. Garos, like them, contribute to the
self-consciousness of the Garo community in general.
International developments have also significantly influenced the Garo
self-perception, their ethnic identity, and their relations with other Bangladeshi
minorities. We can witness the development of a pan-tribal consciousness and
the crystallization of an Indigenous Peoples identity. Internationalization of
the Garo identity is also stimulated by their employment in international
NGOs and Christian churches, and by the employment of a great number of
Garo domestic servants by the expatriate community in Dhaka. These
expatriates constitute the favourite audience during one of the regular tribal
cultural shows or other tribal festivities in Dhaka.
At the same time, however, whereas international networks of Garos with
others are intensifying and the relations between the governments of India
and Bangladesh have much improved since the last elections, connections
between Indian and Bangladeshi Garos seem to be diminishing. The gap
between them seems to widen. More than half a century after the partition,
the international border between the lowland Garos of Bangladesh and the
hill Garos of India increasingly signifies ethnic diversification. Although
Garos continue to cross the border, legally and illegally, to collect firewood,
to do business,116 to visit their relatives, to look for work, and for education,
they rarely speak positively about the Indian Garos and clearly feel different.
The same applies to Indian Garos, who do not hold very positive ideas of
Bangladeshi Garos either. During two brief visits to the Garos Hills in 2000
and in 2005, I noticed on a number of occasions that Indian Garos referred
to the Garos of Bangladesh as Garo-Bangals, and one person believed that the
Bangladeshi Garos are Muslim.
In Chapter 3, I referred to an urban Garo middle class which is actively
involved in a process of ethnic innovation clearly influenced by globalization
and modernity (for example, the Indigenous Peoples discourse and lobby,
human rights issues, and Christianity). At present, the state does not actively
undermine any of these activities. It is interesting to find that the Bengalization
(or “mainstreamization”, as Suborno called it recently) of the Garos goes
hand-in-hand with an active attitude of othering themselves. An international
network offers them a backbone, but an overtly distinct identity also renders
Garos more visible. Christianity and the Indigenous Peoples discourse provide
contemporary Garos with a modern identity. At the same time, the state has
also continued its exclusive policies through stressing the exotic character of
its tribal population by excluding them from modern discourses of nationalism,
by not granting them a place in the national history.
At present, Bangladesh is the home for Mymensingh’s Garos. The Garo
Hills of India are no longer looked upon as their true country, and their
feelings of belonging coincide with the borders of northern Mymensingh or
“Garoland”. However, feelings of belonging in Bangladesh and of being
recognized as equal Bangladeshi citizens by the majority are still two different
things. A few years ago, three of my Garo friends left for India to take part in
a conference and to visit friends and relatives. Although they enjoyed spending
time in the Garo Hills, they also noticed many differences between the Indian
and Bangladeshi Garos. They came to the realization that they were foreigners
in the Garo Hills and were happy about going home to Bangladesh. As soon
as they crossed the border people started to comment: “Oh look, foreigners.”
The three realized that even in Bangladesh they are considered strangers. One
friend described their feelings as follows: “We want to think that we belong
here. We want to feel Bangladeshi. But people think that we are foreigners.”
The example underlines that the “other peoples of Bengal” are still
considered outsiders, or immigrants, by many in Bangladesh.
CONCLUSIONS
The ethnogenesis of the Garos was by no means a sudden and emotional
response to a globalizing world, but the outcome of various long-term processes.
The subsequent states of East Bengal, East Pakistan, and Bangladesh have
played a significant role in this process of becoming the Garos of Bangladesh.
State formation, centralization, and international relations have seriously
influenced the lives of the Garos, their organization, self-perception,
relationships with other communities, and conceptions of homeland and

belonging. This history of the Garos in relation to the state is marked by
many long silences. One moment, the Garos feature in historical accounts;
the next, they are invisible again. It is as if we walked through Garo history
with giant steps, long silences interspersed with detailed descriptions. There
seems little to tell about the relation between Garos and the state during all
those moments of silence, but silences speak too: they tell us that, during
certain times, the state did not see the Garos. This is one important aspect of
the relations between the Garos and the state: on many an occasion, Garos
were invisible or taken for granted, unwanted, unimportant or merely unseen.
This chapter also showed that state and nation are not quite the same,
and that the successive states of East Bengal, East Pakistan, and Bangladesh
certainly did not always speak for the Garos, feel responsible for them in any
way, or identify with them. For example, the dominant Bengali discourse of
Pakistani subjugation and Bengali emancipation and liberation is not quite in
correspondence with Garo discourses of freedom and emancipation, and that
Garo remembrances of significant developments, dramatic moments, and
important turning points divert from contemporary nationalist views in
Bangladesh on both the colonial and post-colonial era. By and large, the
attitudes of the successive states towards the Garos can be characterized by the
catchwords exclusion and othering. In one way or another, they approached
Garos as a distinct group of people, and at times as an anomaly on their
national maps. They treated them differently and, now and again, resorted to
violence to exclude them from the nation.
Nevertheless, since the 1970s, Garos have become firmly embedded in
the state of Bangladesh. This has strengthened their ethnic awareness, but has
also further contributed to processes of othering by the state (through
folklorization, and the exclusion from a national history and identity). At
present, Christianity and the Indigenous Peoples discourse provide Garos
with new and modern means to distinguish themselves from their Bengali
neighbours. And, unlike a few decades ago, they are now proudly presenting
themselves as the Garo people of Bangladesh.
NOTES
1. See Migdal, Joel S., Strong Societies and Weak States. State-Society Relations and
State Capabilities in the Third World (Princeton, New Jersey: Princeton University
Press, 1988), p. 259.
2. For some excellent accounts of state-society (nation) incongruities, see Benedict
R. Anderson, “Old State, New Society: Indonesia’s New Order in Comparative
Historical Perspective”, in ibid., Language and Power: Exploring Political Cultures
in Indonesia (Ithaca: Cornell University Press, 1990), pp. 94–120; Ayesha Jalal,
“Societies, Cultures and Ideologies: Hybrids in Contrived Monoliths”, in ibid.,
Democracy and Authoritarianism in South Asia: A Comparative Perspective
(Cambridge: Cambridge University Press, 1995), pp. 201–46; and “State
Formation and Political Processes in India and Pakistan, 1947 to c.1971”, in
ibid., pp. 29–65.
3. Weber as interpreted by Migdal. Strong Societies and Weak States, p. 19.
4. Migdal, Strong Societies and Weak States, p. 9.
5. Anderson, “Old State, New Society”, p. 94.
6. Mymensingh derived its name from the old pargana or fiscal division of
Maimansingh. At the time of the company’s takeover, it was part of the niabat
that extended from the Garo Hills on the north to the Sundarbans on the south,
and from the Tippera Hills on the east to Jessore in the west. It was called niabat,
because it was ruled by a naib or deputy of the Nazim. The British took over in
the mid-eighteenth century. About 1787, the district of Mymensingh was formed.
It was placed under one collector with the revenue charge of Bhalua, which
included the districts of Tippera and Noakhali. In 1790, Bhalua was separated,
and its headquarters shifted to Mymensingh. See B.C. Allen et al., Gazetteer of
Bengal and North East India (Delhi: Mittal Publications, 1979, first published in
1909, pp. 319 and 223–24.
7. Buchanan writes that most tributary Rajas on the south side of the Brahmaputra
(and thus north of the Garo Hills) were of Garo origin. Milton Sangma believes
that the landlords of these large estates were Hindus, or possibly of mixed Hindu
and Garo blood. See Martin, ed., Eastern India, p. 684; Sangma, The History and
Culture of the Garos, p. 6.
8. Gautam Bhadra, “Two Frontier Uprisings in Mughal India”, in Ranajit Guha,
ed., Subaltern Studies II. Writings on South Asian History and Society (Delhi:
Oxford University Press, 1983), pp. 43–59.
9. Northern Mymensingh had been a frontier region for many centuries already. It
had formed the frontier of many kingdoms before the Mughals established their
rule in Bengal. This is for instance reflected by the remnants of many fortresses
that were built long before the British came, in order to defend Bengal against
the inroads of Garos, Koches and Hajongs. See for example, Jamini Mohan
Ghosh, “An Afghan Fortress in Mymensingh”, Bengal Past & Present, 27 (1924):
56–58. For a more detailed accounts of the history of northern Mymensingh, see
for instance, Allen et al., Gazetteer of Bengal and North East India, pp. 322–24;
Sachse, Mymensingh, pp. 22–33; Bastin, Five Thanas of Partially Excluded Area,
pp. 66–101.
10. Mackenzie, North-East Frontier of India, p. 245.
11. The whole of the northeastern frontier was divided into the estates Shushong
and Sherpur in Mymensingh district, and Koroibari, Kalumalupara, Mechpara,
and Habraghat in Rangpur. In the early nineteenth century these four estates
were separated from Rangpur and included in Goalpara.
12. Sachse writes that “In Muhammadan times the zamindars were frequently
imprisoned and tortured for falling in arrears, and under the early English
Collectors we often read of cases of their being in jail.” Sachse, Mymensingh,
p. 100.
13. Mackenzie, North-East Frontier of India, pp. 245–46.
14. Other products which the Garos traded for cotton were kine, hogs, goats, dogs,
cats, fowls, ducks, fish (dry and fresh), tortoises, rice, extracts of sugar cane,
tobacco, betel nut, hoes, spinning wheels, brass ware, ornaments, silk, and
cotton cloths. See Martin, ed., Eastern India, p. 686; Sangma, The History and
Culture of the Garos, pp. 6–7.
15. Bhattacharjee, The Garos and the English, pp. 19–20.
16. Mackenzie, North-East Frontier of India, p. 245.
17. See for instance, Bhattacharjee, The Garos and the English, pp. 24–26.
18. The following example illustrates the weak position of the central government at
the time: In 1793, Gird Garo (the area that united the plains of Sherpur with the
Koroibaria Hills) was included in Sherpur pargana on an annual rent of Rs. 12.
The government was not unaware that in the early 1820s, Gird Garo returned
the huge sum of 20,000 rupees to the local landlords. Bhattacharjee, The Garos
and the English, p. 76; Jamini Mohan Ghosh, “The Pagal Panthis of Mymensingh”,
Bengal Past and Present 28 (1924): 46; Mackenzie, North-East Frontier of India,
pp. 254–55; Willem van Schendel, “ ‘Madmen’ of Mymensingh: Peasant
Resistance and the Colonial Process in Eastern India, 1824–1833”, The Indian
Economic and Social History Review 22, no. 2 (1985): 143.
19. Bhattacharjee, The Garos and the English, pp. 23–52; Van Schendel, “ ‘Madmen’
of Mymensingh”, pp. 144–47.
20. This process of colonial expansion in Mymensingh has been extensively described
by Willem van Schendel. His article is both for its information and the theoretical
approach important for the part of this chapter that deals with the eighteenth-
and early nineteenth-century history of northern Mymensingh. Van Schendel,
“ ‘Madmen’ of Mymensingh”, pp. 139–73.
21. In the north, the area bordered the Garo Hills. On the south side, the Brahmaputra
River separated northern Mymensingh from the rest of Bengal. (This changed in
the late eighteenth century, when this huge river – once one of the largest rivers
of the world – started to change its course westwards.) Ghosh, “The Pagal
Panthis of Mymensingh”, p. 42. Sachse, Mymensingh, pp. 6–7; Van Schendel,
“ ‘Madmen’ of Mymensingh”, p. 147.
22. Van Schendel, “ ‘Madmen’ of Mymensingh”, p. 139.
23. Ibid., p. 151.
24. According to one story, half-way through the fourteenth century, the Shushong
estate was established by a Garo chief, who later sold it to a Brahmin. In the
sixteenth century, the Koch included a portion of the northern Mymensingh
tract in their kingdom. Later, the Afghans fled to the region after being driven
back by the Mughals. Their last independent king, Osman, fought the Mughals
from his fort at Boinagar. Because of its relatively isolated position, the region
also offered a safe retreat to adventurers and refugees from Assam and Bengal,
who involved themselves in, and added to the complexity of, the local power
conflicts. See for instance, Ghosh, “An Afghan Fortress”, pp. 56–58; Ghosh,
“The Pagal Panthis of Mymensingh”, p. 42; Sangma, History and Culture of the
Garos, p. 5; Van Schendel, “ ‘Madmen’ of Mymensingh”, p. 140.
25. Between the 1770s and the 1790s, particularly after the great famine of 1770,
raiding bands of Hindu and Muslim mendicants and missionaries disturbed the
area, plundered the peasantry, threatened the zamindars, and carried off their
loot. The zamindars developed an interest in their military skills. They employed
them as borkondazes (armed retainers). In this way, they had neutralized these
bands by the 1790s and at the same time consolidated their own position. See
for example, Bastin, Five Thanas of Partially Excluded Area, pp. 79–82; Sachse,
Mymensingh, pp. 28–31; Van Schendel, “ ‘Madmen” of Mymensingh”, pp. 143–
44 and 171.
26. The roots of the Pagol Panthi movement lie in the year 1775, when the Muslim
fakir Karam Shah Darbesh (also known as Karim or Karim Shah) set himself as
a reformer. He settled in Letarkanda (in present-day Durgapur thana), where he
assembled a great many followers around him. These followers became known as
fools or pagols, but called themselves Bhaishahib, or brethren. Hajongs and
Garos belonged to his most devoted supporters. When Karam Shah died in
1813, his son Tipu (also referred to as Tiphu or Tipoo) succeeded him. Tipu and
his mother Ma Shaheba derived a great deal of their power and popularity from
the general belief that they were gifted with supernatural powers. Tipu shifted
his headquarters to Sherpur, where the sect gradually transformed into an armed
band that slowly became a centre of discontent riots (tenants). The Pagol Panthi
sect played a central role in the peasant insurrections that started in 1824. See for
instance, Bastin, Five Thanas of Partially Excluded Area, pp. 79–84; Bhattacharjee,
The Garos and the English, pp. 75–80; Ghosh, “The Pagal Panthis”, pp. 42–53;
Bangladesh District Gazetteers. Mymensingh, p. 37; Mackenzie, North-East Frontier
of India, p. 254; Suprokash Roy, Bharoter-Bidroho O Gonogantric Shongram
(Calcutta: Book World, 1966), pp. 220–29; Sachse, Mymensingh, pp. 31–32;
Sangma, History and Culture of the Garos, pp. 15–16; Van Schendel, “ ‘Madmen’
of Mymensingh”.
27. They were devoted admirers of Karam Shah. The doctrine of truth strongly
appealed to them, and it was in fact their truthfulness that they were stigmatized
as pagauls, or madmen. See Ghosh, “Pagal Panthis of Mymensingh”, p. 43.
28. See Van Schendel, “ ‘Madmen’ of Mymensingh”, pp. 168–70.
29. Hamilton, as discussed by Bastin, Five Thanas Partially Excluded Area, p. 86;
Mackenzie, The North-East Frontier of India, p. 246. Additional causes of this
territorial expansion were the militant attitude of sannyasi and fakir against these
local landlords, conflicts between the zamindars, dynastic in fighting, and in the
early nineteenth century: a decrease in the agricultural production and the
Burma war exactions. In order to increase their income, zamindars increased

surplus extraction. Van Schendel, “ ‘Madmen’ of Mymensingh”, pp. 144–45.
30. In 1866, they established direct management in the hills. In 1867, Captain
Williamson established his quarters in Tura, and in 1869 the Garo Hills became
a separate district. In the winter of 1872–73, the last independent Garos were
brought under the British crown.
31. The Revenue Survey of Mymensingh (1857) demarcated the northern boundary
of Mymensingh along the foot of the Garo Hills. The Sherpur and Shushong
zamindars asserted that this line did not represent the northern boundary of
their estates, which also included portions of the Garo Hills. The Shushong
zamindar took the matter to the local revenue and civil courts and then to the
high court. The latter passed a decree in 1868 which was partially in favour of
the Shushong zamindar. Government appealed to this decision but before an
answer was given, the Government Act (Act XXII) of 1869 excluded the Garo
Hills from the jurisdiction of the Revenue Survey line of Mymensingh. The
Shushong zamindar had finally lost the battle. He did, however, receive financial
compensation for his loss. The 1869 Act, which came into effect in 1870,
asserted the boundary between Mymensingh and the Garo Hills. See Bastin,
Five Thanas of Partially Excluded Area, p. 91; Sachse, Mymensingh, p. 116; and
also General Adminstration Report of the Garo Hills District for the Year 1875–76
(Simla: Government Central Branch Press, 1876), and the Judicial A. Proceedings,
Judicial Department of Bengal, especially for the years 1870–73.
32. For the process of consolidation of British administration in the hills, see for
instance, Bhattacharjee, The Garos and the English, pp. 182–228; Parimal Chandra
Kar, British Annexation of the Garo Hills (Calcutta: Nababharat Publishers,
1970), pp. 42–77; Kar, Glimpses of the Garos, pp. 33–46; Sangma, History and
Culture of the Garos, pp. 122–33.
33. The relations between the hill Garos and the zamindars, as well as the British
interference in the situation, have been extensively researched and described. See
for instance, Mackenzie, North-East Frontier of India, pp. 245–69; Bhattacharjee,
The Garos and the English; Parimal Chandra Kar, British Annexation of the Garo
Hills (Calcutta: Nababharat Publishers, 1970); David R. Syiemlieh, British
Administration in Meghalaya. Policy and Pattern (New Delhi: Heritage Publishers,
1989).
34. Van Schendel, “ ‘Madmen’ of Mymensingh”.
35. Revenue income increased from 77,160 pounds to 161,617 pounds, and state
expenditures from 12,028 to 49,574. At the same time, the relative portion of
land revenue to the total income decreased. In 1795, land revenue amounted to
71,999 pounds, whereas in 1870, it amounted to 84,282 of the total income of
161,617 pounds. See W.W. Hunter, Statistical Account of Bengal. Volume V,
Districts of Dacca, Bakarganj, Farídpur, and Maimansingh (London: Trübner &
Co., 1875), pp. 462–64.
36. Total government expenses rose from 12,028 in 1795 to 14,521 in 1821, to
24,459 in 1860, to 49,574 in 1870. Highest expenditures were made on the

police and judicial departments. Other important posts were, for example,
education, jails, and the local survey department. See Hunter, A Statistical
Account of Bengal. Volume 5, pp. 462–64.
37. Hunter, A Statistical Account of Bengal. Volume 5, p. 392.
38. H.J. Reynolds, History and Statistics, Dacca Division (Calcutta: E.M. Lewis,
Calcutta Central Press Company Limited, 1868), p. 218.
39. Sachse, Mymensingh, p. 28.
40. G.S. Ghurye provides an elaborate account both colonial and post-colonial
policies of protecting the so-called aboriginal population of India. He explains
that, prior to the Government of India Act of 1935, the Government of India
Act of 1919 had already furnished the governor-general-in-council the possibility
to declare any territory in British India to be backward. The list which was
drawn up under the Government of India Act of 1935 was embodied in the
Government of India (Excluded and Partially Excluded Areas) Order of 1936.
Ghurye, The Scheduled Tribes, Chapter 3 and 4.
41. Under Chapter VII-A of the Bengal Tenancy Act.
42. Bastin, Five Thanas of Partially Excluded Area, pp. 123–24 and 207.
43. See F.A. Sachse, Final Report of the Survey and Settlement Operations in the
District of Mymensingh, 1908–1919 (Calcutta: Bengal Secretariat Book Depot,
n.d.).
45. The census of 1931, which was based on religion, mentions 62.9 per cent
Muslims, 32.0 per cent Hindus, 3.1 per cent Christians, and 2.0 per cent tribals.
This census, however, listed the Garos as Hindus. The 1941 census was based on
community. According to this census, of a total population of 483,081, Muslims
comprised 64.7 per cent; Hindus 24.9 per cent; Christians 0.3 per cent; and
Tribes 10.1 per cent. The Hajongs of Sribardi were included in the category of
Hindus. See Bastin, Five Thanas of Partially Excluded Area, p. 39.
46. The “enclavement” of the Chittagong Hill Tracts, for instance, offered the
British administrators the opportunity to directly exploit the Hill Tracts. Several
developments substantiate this view: firstly, the government claimed ownership
of all land in the Chittagong Hills, declared large parts of the hills reserved
forests, and gave high quality tracts to European entrepreneurs. Secondly, attempts
were made to replace shifting cultivation with settled plough cultivation because
the latter was believed to generate higher revenues. Thirdly, several groups of hill
people were resettled in other areas in order to strengthen the borders against
invasions. Fourthly, outsiders, such as Santals and Gurkhas, were encouraged to
settle in the hills and to establish settlements of plough cultivation. See Van
Schendel, “The Invention of the ‘Jummas’ ”, pp. 111–15.
47. Tanka meant that tenants were forced to pay their rents in produce instead of
cash. Since the landlord did not consider this as a permanent right, the land
could again be re-let to the bidder of the highest produce rent.

49. This expression is borrowed from Bastin, Five Thanas of Partially Excluded Area,
p. 124.
50. Bastin, ibid., pp. 44 and 123.
51. The Catholic Bengalese mentions, for instance, that “[b]oys and young men
cannot be expected to take an active interest in their homes which they know
will pass to their sisters and be enjoyed by an important husband.” And in June
1937, the archbishop himself writes that “[t]he missionaries believe that this
institution is an impoverishing weakness in the community.” The Bengalese 14,
no. 10 (November 1933): 7; ibid., 18, no. 6 (June 1937): 24.
52. Rev. P.C. Nall in Our Bond, no. IX (April 1903), p. 4.
53. The Bengalese also writes that the Bengali farmers “are seldom, if ever, out of the
clutches of the moneylender.” The Bengalese 14, no. 9 (October 1933): 13. Italics
mine.
54. These were organized by tenants who had entered into contracts that demanded
payment in kind instead of cash. The tenants were obliged to pay half of their
produce to the landowners. The peasant struggles aimed to reduce these rents to
one-third share. For the Tebhaga movement in Bengal, see Hamza Alavi, “Peasants
and Revolution”, in Kathleen Gough and Hari P. Sharma, eds., Imperialism and
Revolution in South Asia (New York and London: Monthly Review Press, 1973),
pp. 291–337; Sugata Bose, Agrarian Bengal. Economy, Social Structure and Politics,
1919–47 (Cambridge, etc.: Cambridge University Press, 1987), Chapter 8, pp.
252–73; Cooper, Sharecropping and Sharecroppers’ Struggles; Hashmi, Peasant
Utopia; Asok Majumdar, Peasant Protest in Indian Politics. Tebhaga Movement in
Bengal (New Delhi: New Publishers, 1993); Sunil Sen, Peasant Movements in
India. Mid-nineteenth and Twentieth Centuries (Calcutta and New Delhi: K.P.
Bagchi & Company, 1982); Sunil Sen, “The Dilemma”, in A.R. Dessai, ed.,
Peasant Struggles in India (Delhi, etc.: Oxford University Press, 1979), pp. 461–
68, and “Tebhaga Chai”, in ibid., pp. 442–52; For recent developments and
consequences of the sharecropping system in Bangladesh, see Anjan Kumar
Datta, “Enactment of ‘Tebhaga’ and Conditions of Sharecroppers: Some Findings
from Two Deltaic Villages of Bangladesh”, in Bengal. Communities, Development
and States, edited by Bandyopadhyay, Dasgupta and Van Schendel (Dhaka:
University Press Limited, 1995), pp. 319–55.
55. Bastin, Five Thanas of Partially Excluded Area, p. 121. Italics mine.
56. Our Bond, XXXVI, no. 1 (January 1930): 8.
57. The Bengalese 18, no. 6 (June 1937): 24.
58. The original Bengali text is as follows: “Uthore jagaro Achik shontan, Keno ghume
ocheton, Darao, darao Garo jati, Bhor diye paye apnar mati, Pathao, pathao
shontan o shontoti, Lobhite bidya dhormo jnan.” Transcription and translation by
Willem van Schendel.
59. There is some confusion among the informants regarding Lolit’s education,
whether he was a student at Serampore College in Calcutta or at Ananda
Mahan College in Mymensingh. Our informants could not tell us in what year
he died.
60. Mr Justice A.S.M. Akram and Mr Justice S.A. Rahman, “Report of Muslim
Members”, 28 July, 1947, in Reports of Members of the Boundary Commission
(Unpublished, India Office Library Archives, London).
61. “No part of the Province of Bengal can be tagged on to as an area outside the
Province. Secondly, the Partially Excluded Areas are really five thanas in the
Mymensingh district which are Muslim majority thanas. They are contiguous to
the main Muslim majority block, and therefore hardly any justification exists for
excluding any part of them from East Bengal.” Akram and Rahman, “Report of
Muslim Members”, 28 July 1947.
62. Estimates of the number killed have varied widely. According to Andrew J.
Major, half a million deaths would seem a safe guess. Andrew J. Major, “ ‘The
Chief Sufferers’: Abduction of Women during the Partition of the Punjab”,
South Asia, XVIII (Special Issue 1995): 57–58.
63. Leszek A. Kosinski and K. Maudood Elahi, “Introduction”, in Population
Redistribution and Development in South Asia, edited by Leszek A. Kosinski and
K. Maudood Elahi (Dordrecht, Boston and Lancaster: D. Reidel Publishing
Company, 1985), p. 5. Gyanesh Kudaisya argues that the actual figure of
displaced people may be much higher. See Gyanesh Kudaisya, “The Demographic
Upheaval of Partition: Refugees and Agricultural Resettlement in India 1947–
1967”, South Asia, XVIII (Special Issue 1995): 73 (footnote 1).
64. Cf. Leszek A. Kosinski, “Refugee Problems in Bangladesh”, in Kosinski and
Elahi, Population Redistribution and Development in South Asia, pp. 221–23;
Kudaisya, “The Demographic Upheaval of Partition”, p. 73.
66. Emmett, Man of Faith and Flame, p. 33.
67. Note that Anukul uses Achiks rather than Mandis to denote the Garos. Achik is
the name which the Garos of the Garo Hills use for themselves.
69. Similarly, on 23 March 1950, a Catholic procession was held in Mymensingh
town. Two thousand Garos had come to Mymensingh to join the procession:
“[t]he Garos – men, women, and children – formed a procession and, carrying
the banners of their sodalities [sic] and the Papal flag and that of Pakistan,
marched to the station reciting the rosary and singing.” Clancy, Congregations of
Holy Cross in East Bengal. Volume II, p. 341.
70. The archbishop of Dhaka came in person to see the chief secretary to the
government of East Bengal. He was of the opinion that the members of his
mission had been harassed unnecessarily. Letter to M.A. Majid, dated 29 March
1952, File E-6 1952. Home B Proceedings. Government of East Bengal, List 118
Bundle 70. I wish to thank Willem van Schendel for drawing my attention to
these cases.
71. Letter to Aziz Ahmed, dated 11 May 1952, File E-6 1952. Home B Proceedings.
Government of East Bengal, List 118 Bundle 70. Italics mine.
72. Cooper, Sharecropping and Sharecropper’s Struggles, p. 233; Hashmi, Peasant

Utopia, p. 237; Emmett also refers to Muslims settled on Hajong lands. He
writes that “their lands were forfeited to refugee Mohammedans”, Emmett, Man
of Faith and Flame, p. 33.
74. Badruddin Umar, “The Struggle of the Peasants. Class Struggles in East Pakistan
and the Emergency of Bangladesh, XXI”, Holiday, 1 January 1999, p. 5.
75. Bastin, Five Thanas of Partially Excluded Area, p. 37.
76. The Biroidakuni Chronicles for February and March 1956 mention the
migration of a great number of Garos to the tea gardens. On 10 February, 60
people, 12 families, left for the tea gardens. On 23 February, some 40 families
signed up. On 27 February, an unspecified number of families left. And on
27 March, another 40 people left for the tea gardens. We were also able to
locate and consult the chronicles for the Sylhet tea estates for the years 1952–
65. References to Garo immigrants were made on the following dates: 10 April
1954, 18 March 1963, 20 March 1963, and 17 June 1963 (on that date was
written that “over a hundred Garos arrived from Biroi”). I do not know why
the Biroidakuni chronicles made no more references to migration to Sylhet
later. The chronicles of the Sylhet tea estates continue to refer to new Garo
immigrants now and then. See Chronicles of the Catholic Missionaries. Sylhet
Tea Estates, Vol I–IV.
77. The chronicle of 28 March describes, for example, how one Father of the
Mission discovered that “many people are hand up in the villages.” Chronicles of
Biroidakuni Mission, 28 March 1956.
78. The chronicle of 27 March refers to the migration from two Garo villages
(Pekamara and Kalianikhanda), and specifies the reasons: the Garos of Pekamara
left because of “oppression of Refugees and Thieves”. Kalianikhanda people were
“just impoverished by floods”. Chronicles of Biroidakuni Mission, 27 March
1956.
79. Biroidakuni Chronicles, 3 November 1963.
80. On 21 January, the Chronicles report that “Our Garos all along the line are very
worried about their safety. The refugees are harassing them and threatening
them – telling them that all Christians must leave Pakistan, and if any Garos stay
in this country they will have to become Muslims!” See Biroidakuni Chronicles,
21 January 1964.
81. In his Easter message, the archbishop of Dhaka wrote that almost all Catholic
Garos had fled Mariamnagar, Baramari, Biroidakuni, and Bhalukapara. The
parishes of Ranikhong, Baluchora, and Jalchatra (in Modhupur) lost a smaller
number. See “Archbishop’s Easter Message”, Protibeshi 11 (29 March 1964): 3.
82. See for example, A.F.M. Kamaluddin, “Refugee Problems in Bangladesh”, in
Kosinski and Elahi, eds., Population Redistribution and Development, p. 222;
James J. Novak, Bangladesh. Reflections on the Water (Dhaka: University Press
Limited, 1994), p. 91.
83. The article situates the second influx in the year 1967. This is probably a
printing mistake, as the author of the article derives the numbers from the
Pakistan Observer of 2 September 1965. See Kamaluddin, “Refugee Problems in
Bangladesh”, p. 222.
84. For a detailed report on Bangladeshi Garo refugees in the Garo Hills, see
Sengjrang N. Sangma, Bangladeshi Immigrants in Meghalaya. Causes of Human
Movement and Impact on Garo Hills (Kolkata: Anshah Publishing House, 2005).
85. Biroidakuni Chronicles, 5 July 1964.
86. Ibid., 5 July, 25 October, and 10 to 30 November 1964.
87. See for example, Anti-Slavery Society, The Chittagong Hill Tracts. Militarization,
Oppression and the Hill Tribes (London: Anti-Slavery Society, 1984), p. 23.
88. See for example, Biroidakuni Chronicles, 25 and 26 January 1964.
89. Ibid., 26 January 1964.
90. “Archbishop’s Easter Message”, p. 3; see also Biroidakuni Chronicles, 28 January
1964.
91. The Enemy Property Ordinance was a result of the seventeen-day war between
India and Pakistan in 1964. The properties were to be returned after the war. Yet,
the state of war officially ended only with Bangladesh’s independence in 1971.
This meant that the property of Garo refugees were also declared Enemy
Property. In 1974, the Bangladesh Government confirmed it as the Vested and
Non-Resident Property (Administration) Act (XLVI). Officially no new names
were to be added after 1974, but in reality, the law continued to be applied until
1991. See Timm, The Adivasis of Bangladesh, p. 21.
92. Cf. Willem van Schendel, “Who Speaks for the Nation? Nationalist Rethoric
and the Challenge of Cultural Pluralism in Bangladesh”, in Erik-Jan Zürcher
and Willem van Schendel, eds., Identity Politics in Central Asia and the Muslim
World (London: I.B. Tauris, 2000). I do not mean that either one of the two
ingredients is exclusive per se. This is certainly not the case with the secular
Bengali identity as advocated by Mujib. And according to Lamia Karim, for
example, the majority of citizens have never accepted secularism as a central
value. See Lamia Karim, “Adivasi Peoples in Bangladesh”, p. 309.
93. Where the events of 1971 are concerned, many Garos refer to the Pakistani army
as the Punjabis.
94. Almost all Garos are registered under the name Sangma or Marak. This can be
explained by the fact that they were recruited and trained in India, in the Garo
Hills, where Garos use either Sangma or Marak as title, and not their chachi
name, as is the case in Bangladesh. See Mymensingh Jila Trust, 71-er Proshikkon
Shibirer Pramanyo Dolil [Authentic documents of the Training Camps of ’71]
(Mymensingh: Mymensingh Jila Trust, n.d.).
95. See A.B. Chakma’s autobiographic “Look Back From Exile. A Chakma
Experience”, in They are Now Burning Village After Village. Genocide in the
Chittagong Hill Tracts, Bangladesh, edited Wolfgang Mey (Copenhagen: IWGIA,
1984), pp. 42–45.
96. Chakma, “Look Back From Exile”, p. 42; Anti-Slavery Society, The Chittagong
Hill Tracts, p. 56.
97. Firoj Alam, “Selecting Civil Servants: Omission and Commission”, Daily Star
(17 July 1996). The article mentions a reservation in the Civil Service of 30 per
cent for freedom fighters, 10 per cent for women, 5 per cent for women who
were oppressed during the Liberation War, 5 per cent for tribal people, and 10
per cent for the region.
98. Cf. Burling, Strong Women of Modhupur, p. 69.
99. Anti-Slavery Society, The Chittagong Hill Tracts, p. 46; S.P. Talukdar, The
Chakmas. Life and Struggle (New Delhi: Gian Publishing House, 1988), pp.
50–51.
100. Talukdar, The Chakmas, p. 51.
101. Chakma, “Look Back from Exile”, pp. 58–59.
102. Karim, “Adivasi Peoples in Bangladesh”, p. 307.
103. Estimates ranged between 60 per cent and 85 per cent of Garos in the force.
104. Burling, Strong Women of Modhupur, pp. 71–73; Chandrika J. Gulati, Bangladesh:
Liberation to Fundamentalism (A Study of Volatile Indo-Bangladesh Relations)
(New Delhi: Commonwealth Publishers, 1988), p. 79; Talukder Maniruzzaman,
“Bangladesh in 1975: The Fall of the Mujib Regime and its Aftermath”, Asian
Survey 16, no. 2 (1976): 128–29; Talukder Maniruzzaman, “Bangladesh in
1976: Struggle for Survival as an Independent State”, Asian Survey 17, no. 2
(1977): 191–200; M. Rashiduzzaman, “Changing Political Patterns in
Bangladesh: Internal Constraints and External Affairs”, Asian Survey 17, no. 9
(1977): 807; Timm, Adivasis of Bangladesh, p. 21.
105. Biroidakuni Chronicles, 17 October 1975.
106. “BDR’s are trying their best to bring back all those Garos who are in the border.
Many people went to the border. People are not in peace for nobody knows
what is going to happen in future. We are also trying to help them to stay in
Bangladesh.” See Biroidakuni Chronicles, 28 November 1975.
107. On 18 November, a Catholic brother writes in the chronicles: “God knows
what is behind this fighting.” Biroidakuni Chronicles, 18 November 1975.
108. Term used to refer to the diehard followers of Sheikh Mujib, led by Kader
Siddiqui. See Maniruzzaman, “Bangladesh in 1976”, p. 192.
109. Biroidakuni Chronicles, 1 October 1975.
110. Karim, “Adivasi Peoples in Bangladesh”, p. 307.
111. Maniruzzaman, “Bangladesh in 1976”, p. 193.
112. Anti-Slavery Society, The Chittagong Hill Tracts, p. 57.
113. Maniruzzaman, “Bangladesh in 1975”, p. 128.
114. This is not to say that aggression of state representatives has disappeared. There
have been several instances where Garos have become the victim of threats and
aggression on account of state representatives. Particularly Garos of Modhupur
forest have been facing violence and manipulation from local representativtc of
the Forest Department and others.
115. Although the Tribal Welfare Association is meant for all minorities, in northern
Mymensingh the Garos dominate the organization. Garo members hold most
influential positions. One of its first activities was a legal aid project, which they
organized with the help of a German organization. A few years later, the
government ordered the Garos to stop the project and to hand over the
accounts.
116. Many Garos from the border area are involved in smuggling. Since they cannot
easily be distinguished from Indian Garos, whereas Bengalis are immediately
recognized as Bangladeshis, they monopolize the smuggling business in the
border region between Mymensingh and the Garo Hills.
9
SUMMARY AND CONCLUSION:
FROM TRIBES TO ETHNIC MINORITIES
The Garos are one of the many non-Bengali communities of Bangladesh.

They feel very different from other Bangladeshis and make it a point to
maintain their separate identity. This book presented a history of this Garo
community. It analysed the evolution of their distinct ethnic identity and
tried to unravel the complex processes that contributed to their ethnogenesis.
Although this study is heavily indebted to theoretical and empirical
contributions from anthropologists, sociologists and other social scientists, it
is primarily a historical endeavour. The book thus breaks away from the
majority of tribal studies, which are generally dominated by anthropologists,
and often lack a historical perspective. Historians of South Asia, on the other
hand, seldom focus on so-called tribal communities. Particularly in the case
of Bangladesh, historical accounts give no voice to tribal minorities as agents
in the (national) history. Their experiences and histories have been marginalized
in the dominant accounts, because they disturb the nationalist presumptions
which inform the bulk of writing on Bangladesh today.
My research on the Garos clearly demonstrates that South Asian tribal
studies have much to gain from a historical perspective. It shows that only a
long-term perspective allows us to uncover the changeability and flexibility of
social boundaries and identities, and to understand contemporary identity
formation and inter-group relations in the region. The Garo case offered
209
ample opportunity to analyse complex processes of categorization, group

formation, identification, and othering in Bangladesh. It also revealed that
Garos themselves have been active participants in the making of their own
history (and that of others).1 Hence, my prime concern was to historize
anthropological categories, to shift the perspective from anthropological objects
to historical subjects, and to give voice to people who have long been missing
from (national) histories.
The Garos also belong to the category of so-called tribes. This has in
various ways influenced their ethnogenesis. Attached to the concept of tribe
are many preconceived notions of primitivism, isolation, and general
backwardness, which do no justice to the experienced day-to-day realities of
the people belonging to this tribal category, and which have impacted the
very process of identity formation and boundary articulation throughout the
region. In other words, tribalist discourse does not only reflect ways of
thinking but also influences acting. India and East Pakistan, and later also
Bangladesh, developed into different directions and we can only understand
processes of marginalization and othering of Bangladeshi “tribes” through a
careful contextualization of these processes. The Garo case clearly revealed
how tribalist discourse is manifested both on an academic level and in daily
life experiences in Bangladesh. It showed how it fosters paternalist attitudes
and exposed how dominant perceptions of tribes in Bangladesh (and previously
in East Bengal and East Pakistan) have had devastating effects on the daily
lives of so-called tribal people. The idea that they are somehow lagging
behind “mainstream” Bangladeshis in civilization is connected with associations
of hills, forests and other marginal places. Colonial and post-colonial ruling
elites have considered the “tribal” inhabitants of these places as being in need
of “uplift” and improvement, but never as full-fledged citizens.
IMAGES
In the first part of the book, I argued that the tribal category is an outcome
of the interaction between colonial attempts to objectify India and a variety
of different local responses. Contemporary debates about caste and race
criticize the presumed fixed and essentialized make-up of both concepts.
Similar debates never developed about tribes, even though the consequences
of the tribalist discourse are numerous. One critical aspect of the tribalist
discourse is the suggestion that people who are designated as tribal or
indigenous somehow share basic political, social, economic, and cultural
characteristics. This has enabled administrators, politicians, and academics
alike to group all tribes together into one single category. But the essence of
From Tribes to Ethnic Minorities 211
what tribal or indigenous might mean remains highly elusive and only
broadly defined. I argued that this makes the category a catch-all for any
group considered to be distant from the “observing Self ”, and perhaps
explains more about the categorizer than about the categorized, but also that
it carries the danger of becoming a social reality.
Ethnic relations with the majority population have for long been highly
asymmetrical in terms of access to political power and economic resources.
This inequality is also reflected in Garo narratives of Self, which are obviously
informed by the tribalist discourse. Private as well as public narratives of
Garo-ness include tribalist characteristics and arguments. Nevertheless, despite
the intricate and unequal relation between tribalist discourse and emic
narratives of Self, Garos do not fully comply with the dominant discourse
either. They juggle with tribalist arguments and bend them into positive
attributes and strategically advantageous reasoning, which serve their own
agendas (whether these are merely related to self-respect or some larger
political goal). This is, for example, clearly visible in the ways in which Garos
link up to the contemporary discourse on Indigenous Peoples.
BOUNDARIES
In part two, I tried to show that the colonial attempts to categorize local
people brought only apparent clarity. A comparison of different colonial
accounts reveals that colonial categories did not correspond with local
identities. Here I also argued that the distinct ethnic community of Garos
is the outcome of complicated and long-term processes, in which both
colonial as well as local attempts to differentiate and categorize have played
an important role. In order to understand the full complexity of social
categorization and identity formation, we need to adopt both an etic and
emic perspective. After all, etic perceptions do not necessarily correspond
with emic perceptions of Self.
Colonial and local perceptions of communities and boundaries differed,
and colonial observers determined social boundaries which did not have the
same significance or meaning for the people concerned. From the very first
colonial observations, the Garos were considered one group, clearly different
from other neighbouring “races”, “nations”, or tribes. Variation amongst
Garos was made secondary to differences with others, and questions about
local perceptions of Self and Other were not asked. The confusion and
inconsistencies which marked these early nineteenth-century accounts never
led to a critical reconsideration of the term tribe and of the observed
category of Garos.
At the same time, however, a critical reading and comparison of these

same late eighteenth- and nineteenth-century reports suggests that the
Garos were a loose collection of different groups rather than one distinct
ethnic community. The notion of fuzzy communities, which was introduced
to describe pre-modern ideas about identity, seemed to capture nineteenth-
and early-twentieth century Garo conceptions of identity much better than
colonial perceptions of fixed categories and identities. The most persuasive
arguments that support this suggestion came from the Garos themselves.
The personal recollections of elderly Garo villagers revealed a kaleidoscopic
co-existence of several social groupings – rather than of one close-knit
community – who related to each other in looser or stronger ways, hardly
ever or on a frequent basis.
While etic approaches are clearly not sufficient to understand social
boundaries and categories, emic narratives of social boundaries are not
necessarily (more) true. This became clear when we took a closer look at Garo
perceptions of boundary transcendence. The minority position of Garos in
Bangladesh has generated a strong awareness of their vulnerability. This is
clearly reflected by their negative attitude towards inter(ethnic) marriages.
Nevertheless, in reality there is a clear lacuna between ideology and practice.
The whole set of rules and regulations that should prevent people from
marrying across ethnic boundaries are continuously subject to manipulation
and negotiation and the social boundaries between Garos and others are more
flexible and permeable than Garos would like others (and themselves) to
believe. Neither the strong ethnic identity of Bangladeshi Garos, nor their
fear of dissolving (“like a candle in a fire”) in the large community of
Bengalis, prevents individuals from making choices that would typically
disrupt social unity. Moreover, the practice of inter-marriage also shows that
Garos distinguish many Others instead of one. Garo-ness and Otherness are
evidently always, at least to some extent, open to discussion, and at different
moments in time, different criteria determine who is “in” and who is “out”.
BECOMING
Garos, as a distinct ethnic community, are the outcome of the interaction
between various socio-economic, political, and cultural processes. In part
three of this study I analysed the process of becoming the Garos of Bangladesh.
Particularly two factors, the introduction of Christianity and the role of the
state(s), have informed the ethnogenesis of the Garos of Bangladesh.
The large-scale conversion of Garos was the result of a complex, fragmented
process. At different moments, different Garos converted for different reasons
– ranging from spiritual conviction and economic development, to social and

political motivations. Although it is particularly difficult to determine the
direction of causal relations between Christianization and unification of
Garos, the two processes seem clearly linked. This is, for example, revealed by
the very pattern of conversion. Whereas early converts were liable to exclusion
from their own communities, in the course of the twentieth century, a shared
sense of ethnic identity became an increasingly important incentive to convert;
many Garos opted for Christianity because they wanted to stay with other
Garos. Christianity also brought novel forms of leadership and offered Garos
alternative ways to organize, manifest, and express themselves outside the
dominant political arenas. Although Christian missionaries are often held
responsible for more disintegration amongst the Garos, my sources indicated
that it rather was the other way around. Denominational divisions cross-cut
local fragmentation, and inter-denominational strife and competition offered
Garos space to negotiate their needs and wishes, without having to denounce
Christian religion (and identity) altogether. Today, Christianity, as a social
identity, enables Garos to maintain a distinct identity in a predominantly
Muslim country. It provides a shared platform to unite and to define themselves
as one single category of people, and to be included in a larger international
Christian world. Christianity has united, and the wish to stay united has
further stimulated Christianization.
Many post-colonial anthropologists have described missionaries as mere
tools of the colonizer, and missionary activities as a symptom of colonial
control. Although both the state and the different missionary denominations
had their own objectives, they did share a number of interests where
Mymensingh’s Garos were concerned. Missionaries acted as patrons of Garos
as well as mediators between Garos and the state. They facilitated the unification
of Garos but also their exclusion from the larger society. They brought
education to northern Mymensingh, provided healthcare and other social
services, and introduced a “modern world religion”, but they encouraged the
decision of the colonial state to partially exclude northern Mymensingh from
the rest of Bengal.
This decision and many other state formation processes have profoundly
affected the lives, organization, and self-perception of the Garos. State projects
such as objectification, territorialization, and nationalism included many but
often excluded the Garos and other non-Bengali, non-Muslim minorities.
While, for example, the emergence of Bangladesh symbolizes liberation and
emancipation for many Bengalis, such an interpretation of the history of
Bangladesh does no justice to many other non-Bengali Bangladeshi citizens,
for whom the independence of Bangladesh did not bring an end to suppression
and to their status as second-rate citizens. The second, competing, dominant

discourse of national identity, which is based on a shared Islamic identity, and
which is propagated by the state as the national identity of Bangladesh at this
very moment, excludes non-Muslims from the nation.
States are related to territory, and territorial policies are linked to
geographical demarcation and processes of inclusion and exclusion. In
combination with nation-ness, which is the most universally legitimate
value in the political life of our time,2 territorialization has been most
harmful to those who did not fit mainstream nationalist ideologies and
identities. With the establishment of the Garo Hills district, the colonial
state created a separate Garo area (located in contemporary India). Possibly,
the area only received its explicit significance as a distinct Garo realm
because of the colonial expansion and state policies. With the partial
exclusion of the five northernmost thanas of Mymensingh (PEA), the idea
of a Garo homeland in Bengal was created. It even induced a number of
Garos to partake in the guerrilla warfare that followed upon the assassination
of Sheikh Mujibur Rahman in 1975, and which has lived on in the minds
of many until this very day.
The post-colonial states of East Pakistan and later Bangladesh underwent
many transformations. Their heterogeneous character was also reflected by
their often ambivalent and complex attitude towards Bengal’s non-Bengali
minorities. On the whole, however, we can view a continuous tendency of
othering and exclusion. Although periods of invisibility were followed by
visibility, and silence was followed by voice, neither did any state attempt to
accept non-Bengali minorities as truly equal citizens, nor did they grant them
equal positions and voice. This also explains why in Bangladesh, and in many
other parts of South Asia for that matter, claims concerning indigeneity –
which make little sense historically and are politically highly sensitive – are
becoming widely used methods to demand special rights and provisions. The
Indigenous Peoples discourse provides so-called tribes an international podium
and audience, and the vehicles to demand a right to self-determination.
In October 2001, the people of Bangladesh voted the Bangladesh
Nationalist Party (BNP) back into power. Together with the Jamaat-e-Islami
(the Islam Unity Council) they form a pro-Islamic government. After a
period in which the secular Bengali identity, rather than the Islamic identity,
was promoted as the national identity of Bangladesh, at present, “the constant
propaganda that the State is an Islamic State encourages and generates a belief
in the minds of the ordinary Muslims that non-Muslims have no place in the
State except perhaps as subservient people and are not entitled to any rights
in the State.”3 In the 1960s, pro-Islamic sentiments allowed for the suppression
of Garos and other minorities, and even led to their expulsion from East
Pakistan. Yet history does not repeat itself. More than forty years have passed
since. In the meantime, Garos have developed into a self-conscious community.
They have given priority to educating their children, and thousands of them
are studying in colleges and universities, and have found jobs in national
and international businesses and governmental and non-governmental
organizations. They, more than ever before, demand a voice. They are,
however, a tiny minority still, and very vulnerable indeed. “The minorities in
East Bengal cannot feel a sense of security unless the majority community as
a whole are made to believe that the minorities are the citizens of the State as
much as the majorities are and are entitled to the same privileges and
protection of law with equal rights.”4
RETHINKING TRIBE
A historical examination of the Bangladeshi Garos does away with the apparent
timeless quality of so-called tribes. It demonstrates that the categorization of
South Asia’s population into tribes and others, which also suggests a simplistic
dichotomization between a primitive, unchangeable Other, and a sophisticated,
developed Self, needs critical evaluation. Some basic incongruities have
remained with the introduction of the Indigenous Peoples concept. The
Indigenous Peoples discourse also suggests basic similarities where there are
none, and presumes that we can somehow make a distinction between
“people of the world” and “people of nature”. Such a straightforward
categorization clouds our understanding of the complex historical and present-
day realities in South Asia.
The importance of this work is not confined to a better understanding of
how smaller ethnic groups form, survive or disappear in modern society. It
also leads to a more comprehensive insight into how social life in Bangladesh
is organized and how it has transformed over time. It challenges existing ideas
about the relationship between states and citizens in this part of the world,
and of notions such as development, democracy, legitimacy, and citizenship.
“After all, it is in their treatment of smaller groups that societies show their
priorities, their norms for the application of power and governance, their
quest for participation and consent, and their blueprint for the future.”5 The
question whether Garos eat frogs refers as much to misconceptions (Garos do
not eat frogs) as it does to deep-rooted notions of primitivism (eating frogs as
an uncivilized habit). It is a crystal-clear illustration of how labels illustrate as
well as inform the articulation of boundaries, and serve to include some in the
category of Self, and to exclude others. In Bangladesh, to ask if one eats frogs,
instantly reflects such inclusion and exclusion. Perhaps the appropriate answer
to the question should be: so what if Garos do eat frogs?
NOTES
1. Cf. Gravers, “Karen Making of a Nation”, p. 268.
2. Benedict Anderson, Imagined Communities. Reflections of the Origin and Spread
of Nationalism (London and New York: Verso, 1998, revised and extended
edition), p. 3.
3. Kathinka Sinha-Kerkhoff, Tyranny of Partition: Hindus in Bangladesh and Muslims
in India (New Delhi: Gyan Publications, 2006), p. 136.
4. Ibid.
5. Van Schendel and Bal, “Beyond the ‘Tribal Mind-Set’ ”, p. 139.
REFERENCES
I. Unpublished Sources
a. Government Records
General Adminstration Reports of the Garo Hills District (Simla: Government Central
Branch Press, 1876)
Government of Bengal Proceedings: Home B. Proceedings
Government of Bengal Proceedings: Judicial A. Proceedings
Reports of Members of the Boundary Commission (Unpublished, India Office Library
Archives, London)
b. Missionary Records and Publications

Chronicles of the Catholic Missionaries
Missionary Herald
Our Bond
Protibeshi
Yearly Report of the GBC to the ABMS
II. Published Sources

Agarwal, Bina. A Field of One’s Own. Gender and Land Rights in South Asia (Cambridge:
Ahmed, Imtiaz, Abhijit Dasgupta and Kathinka Sinha-Kerkhoff, eds. State, Society
and Displaced People in South Asia (Dhaka: University Press Limited, 2004).
Alam, Firoj. “Selecting Civil Servants: Omission and Commission”, Daily Star,
17 July 1996.
Alavi, Hamza. “Peasants and Revolution”, in Imperialism and Revolution in South
Asia, edited by Kathleen Gough and Hari P. Sharma (New York and London:
Monthly Review Press, 1973), pp. 291–337.
Alexander, Elizabeth Susan. The Attitudes of the British Protestant Missionaries Towards
Nationalism in India. With Special Reference to Madras Presidency 1919–1927
(Delhi: Konark Publishers, 1994).
Alibhai-Brown, Yasmin and Anne Montague. The Colour of Love. Mixed Race
Relationships (London: Virago, 1992).
Allen, B.C., et al. Gazetteer of Bengal and North East India (Delhi: Mittal Publications,
1979, first published in 1909).
Ambasta, Ashesh. Capitalist Restructuring and Formation of Adivasi Proletarians.
217
10 TheyAsk_EatFrogs Ref By: ROS 217 Size: 6" x 9" J/No: 06-15540 4/16/07, 10:43 AM Fonts: AGaramond
Agrarian Transition in Thane District (Western India) c. 1817–1990) (Unpublished

Ph.D. thesis: Institute of Social Studies, The Hague, 1998).
Anderson, Benedict. Imagined Communities. Reflections of the Origin and Spread of
Nationalism (London and New York: Verso, 1998, revised and extended edition).
Anderson, Benedict R. Language and Power: Exploring Political Cultures in Indonesia
(Ithaca: Cornell University Press, 1990).
———. “Old State, New Society: Indonesia’s New Order in Comparative Historical
Perspective”, in Language and Power: Exploring Political Cultures in Indonesia,
edited by Benedict R. Anderson (Ithaca: Cornell University Press, 1990), pp.
94–120.
Anthias, Flora and Nira Yuval-Davis. Women-Nation-State (Houndmills, etc.:
Macmillan, 1989).
Anti-Slavery Society. The Chittagong Hill Tracts. Militarization, oppression and the hill
tribes (London: Anti-Slavery Society, 1984).
Bal, Ellen. “An Untold Story of the Partition: The Garos of Northern Mymensingh
(Bangladesh), the Process of the 1947 Partition, and the Articulation of Identities”,
in State, Society and Displaced People in South Asia, edited by Imtiaz Ahmed,
Abhijit Dasgupta and Kathinka Sinha-Kerkhoff (Dhaka: University Press Limited,
2004), pp. 245–79.
———. “Becoming the Garos of Bangladesh: Ethnic Unification and Identity
Formation in Bangladesh”, Anthropology Review (Institute of Anthropology Hanoi,
Vietnam; appeared in Vietnamese) 5 (2003): 56–66.
Bal, Ellen, Yasuhiro Takami and Johannes Sangren. Manderangni Jagring. Images of
the Garos of Bangladesh (Dhaka: UPL, 1999).
Baldwin, C.D. God and the Garos (Sydney: Australian Baptist Publishing House,
1934).
Bandyopadhyay, Sekhar. “Caste in the Perception of the Raj. A Note on the Evolution
of Colonial Sociology of Bengal”, Bengal Past and Present, 104 (1985): 56–80.
———. Caste, Politics and the Raj. Bengal 1872–1937 (Calcutta and New Delhi: K.P.
Bagchi & Company, 1990).
———. “Development, Differentation and Caste: The Namasudra Movement in
Bengal, 1872–1947”, in Bengal. Communities, Development and States, edited by
Bandyopadhyay, Dasgupta and Van Schendel, pp. 90–119.
Bandyopadhyay, Sekhar, Abhijit Dasgupta and Willem van Schendel, eds. Bengal.
Communities, Development and States (New Delhi: Manohar, 1994).
Bangladesh Bureau of Statistics. 1993 Statistical Yearbook of Bangladesh. 14th edition
(Dhaka: Bangladesh Bureau of Statistics, Statistics Division, 1994).
Bangladesh District Gazetteers, Mymensingh (Dacca: Bangladesh Government Press,
1978), p. 58.
Baptist Union of Victoria. Handbook Containing President’s Address and Reports of the
Union Home Mission, Fund, College, Foreign Mission & c. (Melbourne: Watt &
Co., Printers, 1892, 1899, 1900).
Barbara, Augustin. “Mixed Marriages. Some Key Questions”, International Migration
32 (1994): 571–86.
References 219
Barkataki, S. Tribes of Assam (New Delhi: National Book Trust, 1969).

Barnes, R.H., Andrew Gray and Benedict Kingsbury, eds. Indigenous Peoples of Asia
(Michigan: Association for Asian Studies, 1995).
Barooah, Nirode K. David Scott in North-East India 1802–1831. A Study in British
Paternalism (New Delhi: Mushiram Manoharlal, 1970).
Barrier, N. Gerald, ed. The Census in British India. New Perspectives (New Delhi:
Manohar, 1981).
Barth, Fredrik, ed. Ethnic Groups and Boundaries. The Social Organization of Cultural
Differences (Bergen and Oslo: Universitets Vorlaget/London: George Allan &
Unwin, 1970).
Bastin, R.W. Final Report of the Settlement Operations in Five Thanas of Partially
Excluded Area of Mymensingh 1938–42 (Dacca: East Bengal Government Press,
1954).
Basu, Swaraj. “The Colonial State and the Indigenous Society: The Creation of the
Rajbansi Identity in Bengal”, in Bandyopadhyay, Dasgupta and Van Schendel,
op. cit., pp. 43–64.
Bates, Crispin. “ ‘Lost Innocents and the Loss of Innocence’: Interpreting Adivasi
Movements in South Asia”, in Barnes, Gray and Kingsbury, op. cit., pp. 102–
19.
———. “Race, Caste, and Tribe in Central India”, in The Concept of Race in South
Asia, edited by Peter Robb (Delhi, etc.: Oxford Press, 1995), pp. 219–59.
Baumann, Gerd. The Multicultural Riddle. Rethinking National, Ethnic, and Religious
Identities (New York and London: Routledge, 1999).
Baud, Michiel, et al. Etniciteit als Strategie in Latijns-Amerika en de Caraïben
(Amsterdam: Amsterdam University Press, 1994).
Bayly, Susan. “Caste and ‘Race’ in the Colonial Ethnography of India”, in Robb, ed.,
op. cit., pp. 165–218.
———. Caste, Society and Politics in India from the Eighteenth Century to the Modern
Age. The New Cambridge History of India IV, no. 3 (Cambridge: Cambridge
———. Saints, Goddesses and Kings. Muslims and Christians in South Indian Society
1700–1900 (Cambridge etc.: Cambridge University Press, 1989).
Becker, C. History of the Catholic Missions in Northeast India 1890–1915 (Calcutta:
Firma KLM under the auspices of Vendrame Missiological Institute, Sacred
Heart College, Shillong, 1980).
Benson, Susan. Ambiguous Ethnicity. Interracial Families in London (Cambridge:
Bessaignet, Pierre, ed. Social Research in East Pakistan (Dacca: Asiatic Society of
Pakistan, 2nd revised edition, 1964), pp. 172–236.
Bessaignet, Pierre. “Tribes of the Northern Borders of East Pakistan”, in ibid.
Béteille, André. “The Definition of Tribe”, in Tribe, Caste and Religion in India,
edited by Romesh Thapar (Delhi: Macmillan, 1977), pp. 7–14.
———. “The Idea of Indigenous People”, Current Anthropology 39, no. 2 (1998):
187–91.
Bhadra, Gautam. “Two Frontier Uprisings in Mughal India”, in Subaltern Studies II.
Writings on South Asian History and Society, edited by Ranajit Guha (Delhi:
Oxford University Press, 1983), pp. 43–59.
Bhattacharjee, Jayanta Bhusan. The Garos and the English 1765–1874 (New Delhi:
Radiant Publishers, 1978).
Blanchet, Thérèse. Lost Innocence, Stolen Childhood (Dhaka: University Press Limited,
1996).
Bose, Sugata. Agrarian Bengal. Economy, Social Structure and Politics, 1919–1947
(Cambridge, etc.: Cambridge University Press, 1987).
Bosma, Ulbe and Gijsbert Oonk. “Bombay Batavia. Parsi and Eurasian Variations on
the Middlemen Theme”, in Mediators between State and Society, edited by Nico
Randeraad (Hilversum: Verloren, 1998), pp. 17–41.
Breman, Jan, ed. Imperial Monkey Business. Racial Supremacy in Social Darwinist
Theory and Colonial Practice (Amsterdam: VU University Press, 1990).
Breman, Jan. Of Peasants, Migrants and Paupers. Rural Labour Circulation and Capitalist
Production in West India (Delhi, etc.: Oxford University Press, 1985).
Buchanan, Francis, in Montgomery Martin, The History, Antiquities, Topography, and
Statistics of Eastern India; Comprising the Districts of Behar, Shahabad, Bhagalpoor,
Goruckpoor, Dinajepoor, Puraniya, Ronggopoor, and Assam. Volume V, Rangpur
and Assam (Delhi: Cosmo Publications, 1976, first published in 1838).
Burling, Robbins. “An Incipient Caste Organization in the Garo Hills”, in Man in
India 40, no. 4 (December 1960): 283–99.
———. Hill Farms and Paddy Fields. Life in Mainland Southeast Asia (Englewood
Cliffs: Prentice Hall, 1965).
———. Rengsanggri. Family and Kinship in a Garo Village (Philadelphia: University of
Pennsylvania Press, 1963).
———. The Strong Women of Modhupur (Dhaka: University Press Limited, 1997).
Butler, John. A Sketch of Assam. With Some Account of the Hill Tribes (London: Smith,
Elder and Co., 1847).
Caplan, Lionel, ed. Studies in Religious Fundamentalism (London: Macmillan,
1992).
Caplan, Lionel. “Popular Conceptions of Fundamentalism”, in ibid., pp. 1–24.
———. Warrior Gentleman. ‘Gurkhas’ in the Western Imagination (Providence and
Oxford: Berghahn Books, 1995).
Carey, Rev. William. The Garo Jungle Book (Tura: Tura Book Room, 1993, first
published in 1919).
Carroll, Lucy. “Colonial Perceptions of Indian Society and the Emergence of Caste(s)
Associations”, Journal of Asian Studies 37, no. 2 (1978): 233–50.
Catholic Directory of Bangladesh (Dhaka: Bishop’s Conference of Bangladesh, 1992).
Census of India, 1901. Volume I, India, Ethnographic Appendices (by H.H. Risley)
(Calcutta: Government Printing, 1903).
———, 1911. Volume I, India, Part I, Report (by E.A. Gait) (Calcutta: Government
Printing, 1913).
References 221
———, 1921. Volume V, Bengal, Part I, Report (by W.H. Thompson) (Calcutta:
Bengal Secretariat Book Depot, 1923).
———, 1931. Volume V, Bengal and Sikkim, Part I (by A.E. Porter) (Calcutta: Central
Publication Branch, 1933).
Chakma, A.B. “Look Back From Exile. A Chakma Experience”, in They are Now
Burning Village After Village. Genocide in the Chittagong Hill Tracts, Bangladesh,
edited by Wolfgang Mey (Copenhagen: IWGIA, 1984).
Chakrabarty, Dipesh. “Modernity and Ethnicity in India”, South Asia XVII, Special
Issue (1994): 143–55.
Chatterjee, Partha and Gyan Pandey, eds. Subaltern Studies VII (Delhi, etc.: Oxford
Chaudhuri, Binay Bhushan. “ ‘The Myth of the tribe?’ The Question Reconsidered”,
The Calcutta Historical Journal XVI, no. 1 (1994): 125–56.
———. “Tribal Society in Transition: Eastern India 1757–1920”, in India’s Colonial
Encounter. Essays in Memory of Eric Stokes, edited by Mushirul Hasan and
Narayani Gupta (New Delhi: Manohar, 1993), pp. 65–120.
Chowdhury, Anwarullah, ed. Pains and Pleasures of Fieldwork (Dhaka: National
Institute of Local Government, 1985).
Clancy, Raymond C. The Congregation of Holy Cross in East Bengal, 1853–1953. In
Three Parts with a Brief History of the Church in Bengal. Volume 1 (Holy Cross
Foreign Mission Seminary, Washington D.C.: Unpublished manuscript, 1953).
———. The Congregation of Holy Cross in the Diocese of Dacca 1888–1927 (Holy
Cross College Brookland Station, Washington D.C.: Unpublished manuscript,
1948).
———. Congregations of Holy Cross in East Bengal. Volume II (Holy Cross Foreign
Mission Seminary, Washington D.C.: Unpublished manuscript, 1953).
Cohn, Bernard S., ed. An Anthropologist among Historians and Other Essays (Delhi,
etc.: Oxford University Press, 1987).
Cohn, Bernard S. “The Census, Social Structure and Objectification in South Asia”,
in Cohn, ed., ibid., pp. 224–54.
———. “Notes on the History of the Study of Indian Society and Culture”, in
Structure and Change in Indian Society, edited by Milton Singer and Bernard S.
Cohn (Chicago: Aldine Publishing Company, 1968), pp. 3–28.
———. “The Pasts of an Indian Village”, in Time. Histories and Ethnologies, edited by
Diane Owen Hughes and Thomas R. Trautman (Ann Arbor: The University of
Michigan Press, 1995), pp. 21–30.
Conlon, Frank F. “The Census of India as a Source for the Historical Study of
Religion and Caste”, in The Census in British India. New Perspectives, edited by
N. Gerald Barrier (New Delhi: Manohar, 1981), pp. 103–17.
Conrad, Yves. “Lisu Identity in Northern Thailand: A Problematique for
Anthropology”, in Hill Tribes Today. Problems in Change, edited by John
McKinnon and Bernard Vienne (Bangkok: White Lotus-Orstrom, 1989), pp.
191–219.
Cooper, Adrienne. Sharecropping and Sharecroppers’ Struggles in Bengal 1930–1950

(Calcutta and New Delhi: KP Bagchi & Company, 1988).
Cornell, Stephen and Douglas Hartmann. Ethnicity and Race. Making Identities in a
Changing World (Thousand Oaks, London and New Delhi: Pine Forge Press,
1998).
Dalton, Edward Tuite. Descriptive Ethnology of Bengal (Calcutta: Indian Studies Past
& Present, 1973, originally published in 1872).
Das, Suranjan. Communal Riots in Bengal 1905–1947 (Delhi, etc.: Oxford University
Press, 1991).
Das Gupta, Pranab Kumar. Life and Culture of a Matrilineal Tribe of Meghalaya (New
Delhi: Inter-India Publications, 1984).
Datt Bharti, Brahm. “Political Roots of Christianity in India”, in Politics of Conversion,
edited by Devendra Swarup (Delhi: Deendayal Research Institute, 1986), pp.
81–100.
Datta, Anjan Kumar. “Enactment of ‘Tebhaga’ and Conditions of Sharecroppers:
Some Findings from Two Deltaic Villages of Bangladesh”, in Bengal. Communities,
Development and States, edited by Bandyopadhyay, Dasgupta and Van Schendel
(Dhaka: University Press Limited, 1995), pp. 319–55.
De Maaker, Erik. Negotiating Life: Garo Death Rituals and the Transformation of
Society (Ph.D. thesis, Leiden University, 2006).
De Vos, George A. “Ethnic Pluralism: Conflict and Accomodation”, in Ethnic
Identity. Creation, Conflict and Accommodation, edited by Romanucci-Ross, Lola
and George A. De Vos (Walnut Creek/London/New Delhi: Altamira Press,
1995, 3rd edition), pp. 15–47.
Dena, Lal. Christian Missions and Colonialism. A Study of Missionary Movement in
Northeast India with Particular Reference to Manipur and Lushai Hills, 1894–
1947 (Shillong: Vendrame Institute, 1988).
Dessai, A.R., ed. Peasant Struggles in India (Delhi, etc.: Oxford University Press,
1979).
Devalle, Susana B.C. Discourses of Ethnicity. Culture and Protest in Jharkhand (New
Delhi, etc.: Sage Publications, 1992).
Dirks, Nicholas B. Castes of Mind: Colonialism and the Making of Modern India
(Princeton and Oxford: Princeton University Press, 2001).
Downs, Frederick S. “Christian Conversion Movements among the Hill Tribes of
Northeast India in the Nineteenth and Twentieth History”, in Religion in South
Asia. Religious Conversion and Revival Movements in South Asia in Medieval and
Modern Times, edited by G.A. Oddie (New Delhi: Manohar, 1991, 2nd revised
edition), pp. 155–74.
———. North East India in the Nineteenth and Twentieth Centuries. History of
Christianity in India Volume V, Part 5 (Bangalore, The Church History Association
of India, 1992).
Dudley Jenkins, Laura. “Another ‘People of India’ Project: Colonial and National
Anthropology”, The Journal of Asian Studies 62, no. 4 (November 2003): 1143–
70.
References 223
Eaton, Richard M. “Comparative History as World History: Religious Conversion in

Modern India”, Journal of World History 8, no. 2 (1997): 243–71.
———. “Conversion to Christianity among the Nagas, 1876–1971”, The Indian
Economic and Social History Review 21, no. 1 (1984): 1–44.
———. The Rise of the Islam and the Bengal Frontier (Berkeley, etc.: University of
California Press, 1993).
Eidheim, Harald. Aspects of the Lappish Minority Situation (Oslo: Scandinavian
———. “When Ethnic Identity is a Social Stigma”, in Ethnic Groups and Boundaries.
The Social Organization of Cultural Difference, edited by Fredrik Barth (Bergen
and Oslo: Scandinavian University Press), pp. 39–57.
Eliot, John. “Observations on the Inhabitants of the Garrow Hills. Made during a
Public Deputation in the Years 1788 and 1789”, Asiatic Researches 3 (New Delhi:
Cosmo Publications, 1979, first published in 1794), pp. 17–37.
Emmett, Ian. Man of Faith and Flame. Bilu Babu of Birisiri (Melbourne: Jenkin
Boxton and Co, n.d.).
Eriksen, Thomas Hylland. Ethnicity and Nationalism. Anthropological Perspectives
(London/Sterling, Virginia: Pluto Press, 2002, 2nd edition).
———. Us and Them in Modern Societies. Ethnicity and Nationalism in Mauritius,
Trinidad and Beyond (Oslo: Scandinavian University Press, 1992).
Fried, M.H. The Notion of Tribe (Menlo Park, Cal.: Cummings Publishing Company,
1975).
Fruzzetti, Lina M. The Gift of a Virgin. Second Indian Impression with a New
Introduction on ‘Some Contemporary Issues in Context’ (Delhi, etc.: Oxford
Fruzzetti, Lina M. and Ákos Östör. “Bad Blood in Bengal: Category and Affect in the
Study of Kinship, Caste, and Marriage”, in Concepts of Person. Kinship, Caste,
and Marriage in India, edited by Östör, Lina Fruzzetti and Steve Barnett (Delhi,
Gain, Philip, ed. Bangladesh. Land, Forest and Forest People (Dhaka: Society for
Environment and Human Development, 1995).
Gain, Philip. The Last Forests of Bangladesh (Dhaka: Society for Environment and
Human Development, 1998).
Galanter, Marc. Competing Equalities. Law and the Backward Classes in India (Delhi,
Geertz, Clifford. The Interpretation of Cultures. Selected Essays by Clifford Geertz (New
York: Basic Books, 1973).
Ghosh, Jamini Mohan. “An Afghan Fortress in Mymensingh”, Bengal Past & Present,
27 (1924): 56–58.
———. “The Pagal Panthis of Mymensingh”, Bengal Past & Present, 28 (1924): 42–
53.
Ghurye, G.S. The Scheduled Tribes (Bombay: Popular Prakash, 1963, 3rd edition).
Goedert, Edmund N. The Apostolate to the Garos (Dacca: Unpublished manuscipt,
1959).
———. Holy Cross Priests in the Diocese of Dacca 1853–1981 (Notre Dame Indiana:
Province Archive Center, 1983).
Gomes, Stephen G. “The Paharias in Bangladesh: A Case Study of Assimilation and
Identification with Policy Implications”, in Tribal Cultures in Bangladesh, edited
by Mahmud Shah Qureshi (Rajshahi: Institute of Bangladesh Studies, 1984),
pp. 135–57.
———. The Paharias: A Glimpse of Tribal Life in North Western Bangladesh (Dhaka:
Caritas, 1988).
Gough, Kathleen and Hari P. Sharma, eds. Imperialism and Revolution in South Asia
(New York and London: Monthly Review Press, 1973).
Gravers, Mikael. “The Karen Making of a Nation”, in Asian Forms of the Nation,
edited by Stein Tonnesson and Hans Antlöv (Richmond: Curzon, 1996), pp.
237–69.
Gray, Andrew. “The Indigenous Movement in Asia”, in Indigenous Peoples of Asia,
edited by Barnes, R.H., Andrew Gray and Benedict Kingsbury (Michigan:
Association for Asian Studies, 1995), pp. 35–58.
Guha, Ranajit, ed. Subaltern Studies II. Writings on South Asian History and Society
(Delhi: Oxford University Press, 1983).
Guha, Sumit. Environment and Ethnicity in India, 1200–1991 (Cambridge: Cambridge
Gulati, Chandrika J. Bangladesh: Liberation to Fundamentalism. (A Study of Volatile
Indo-Bangladesh Relations) (New Delhi: Commonwealth Publishers, 1988).
Hall, Stuart. “Old and New Idenities, Old and New Ethnicities”, in Culture,
Globalization and the World-System, edited by A.D. King (Basingstoke, England:
Macmillan, 1991), pp. 41–68.
Hannerz, Ulf. “Culture between Center and Periphery. Toward a Macroanthropology”,
Ethnos 54 (1989): 200–16.
Hardiman, David. The Coming of the Devi: Adivasi Assertion in Western India (Delhi,
Hasan, Mushirul and Narayani Gupta, eds. India’s Colonial Encounter. Essays in
Memory of Eric Stokes (New Delhi: Manohar, 1993).
Hashmi, Taj ul-Islam. Peasant Utopia. The Communalization of Class Politics in East
Bengal, 1920–1947 (Dhaka: University Press Limited, 1994).
Headland, Thomas N., Kenneth L. Pike and Marvin Harris. Emics and Etics. The
Insider/Outsider Debate (Newbury Park, etc.: Sage Publications, 1990).
Heesterman, J.C. The Inner Conflict of Tradition. Essays in Indian Ritual, Kingship,
and Society (Chicago and London: The University of Chicago Press, 1985).
Hefner, Robert W., ed. Conversion to Christianity. Historical and Anthropological
Perspectives on a Great Transformation (Berkeley, etc.: University of California
Press, 1993).
Hefner, Robert W. “Of Faith and Commitment: Christian Conversion in Muslim
Java”, in Hefner, ed., ibid., pp. 99–123.
———. “Introduction: World Building and the Rationality of Conversion”, in Hefner,
ed., ibid., pp. 3–43.
References 225
Hinton, Peter. “Do the Karen Really Exist?”, in Highlanders of Thailand, edited by
John McKinnon and Wanat Bhruksasri (Singapore, Oxford, New York: Oxford
University Press, 1986), pp. 155–68.
Hobsbawm, Eric and Terence Ranger, ed. The Invention of Tradition (Cambridge,
etc.: Cambridge University Press, 1983).
Hodson, T.C. India. Census Ethnography, 1901–1931 (New Delhi: Usha, reprint
Hoekstra, Hanneke. The Orthodoxy of the Heart. Faith, Fryslân, and Feminism in the
Novels of Ypk fan der Fear (Groningen: FFYRUG, 1998).
Hondius, Dienke. Gemengde huwelijken, gemengde gevoelens. Aanvaarding en ontwijking
van etnisch en religieus verschil since 1945 (Den Haag: Sdu Uitgevers, 1999).
Horton, Robin. “On the Rationality of Conversion”, Part I, Africa 45, no. 3 (1995).
Hunter, W.W. A Statistical Account of Assam. Volume II (London: Trübner & Co,
1879).
———. Statistical Account of Bengal. Volume V, Districts of Dacca, Bakarganj, Farídpur,
and Maimansingh (London: Trübner & Co., 1875).
Inden, Ronald. Imagining India (Oxford: Basil Blackwell, 1990).
Islam Khan. Nurul, Bangladesh District Gazetteers. Mymensingh (Dacca: Bangladesh
Government Press, 1978).
IWGIA. The Indigenous World 1998–1999 (Copenhagen: IWGIA, 1999).
Jacobs, Julian, et al. The Nagas. Society, Culture and the Colonial Encounter (London:
Thames and Hudson, 1990).
Jalal, Ayesha. Democracy and Authoritarianism in South Asia: A Comparative Perspective
(Cambridge: Cambridge University Press, 1995).
———. “Societies, Cultures and Ideologies: Hybrids in Contrived Monoliths”, in
ibid., pp. 201–46.
———. “State Formation and Political Processes in India and Pakistan, 1947 to
c. 1971”, in ibid., pp. 29–65.
Jasal, Smita and Eyal Ben-Ari, eds. The Partition Motif in Contemporary Conflicts
(New Delhi, Thousand Oaks, London: Sage Publications, 2007).
Jengcham, Shubash. Bangladesher Garo Sampradai (Dhaka: Bangla Academy, 1994).
Jenkins, Richard, Rethinking Ethnicity. Arguments and Explorations (London, etc.:
Sage Publications, 1997).
Johnson, Walton R. and D. Michael Warren. Inside the Mixed Marriage: Accounts of
Changing Attitudes, Patterns and Perceptions of Cross-Cultural and Interracial
Marriages (Lanham: University Press of America, 1993).
Kamaluddin, A.F.M. “Refugee Problems in Bangladesh”, in Population
Redistribution and Development in South Asia, edited by L.A. Kosinski
and K.M. Elahi (Dordrecht, Boston and Lancaster: D. Reidel Publishing
Company, 1985).
Kar, Parimal Chandra. British Annexation of the Garo Hills (Calcutta: Nababharat
Publishers, 1970).
———. Glimpses of the Garos. Chapter 6 and 7 (Tura: Garo Hills Book Emporium,
1982).
Karim, Lamia. “Pushed to the Margins: Adivasi Peoples in Bangladesh and the Case
of Kalpana Chakma”, Contemporary South Asia 7, no. 3 (1998): 301–16.
Karlsson, Bengt K. “Anthropology and the ‘Indigenous Slot’: Claims to Debates
about Indigenous Peoples’ Status in India”, Critique of Anthropology 23, no. 4
(2003): 403–23.
Karlsson, B.G. Contested Belonging. An Indigenous People’s Struggle for Forest and
Identity in Sub-Himalayan Bengal (Lund: Lund University, 1997).
Karotemprel, Sebastian. Albizuri among the Lyngams (A Brief History of the Catholic
Mission among the Lyngams of Northeast India) (Shillong: Vendrame Missiological
Institute, 1985).
Kaviraj, Sudipta. “The Imaginary Institution of India”, in Subaltern Studies VII,
edited by Partha Chatterjee and Gyan Pandey (Delhi, etc.: Oxford University
Press, 1993), 1–39.
Kertzer, David I. Family Life in Central Italy, 1880–1910. Sharecropping, Wage Labor,
and Coresidence (New Brunswick: Rutgers University Press, 1984).
Keyes, Charles, ed. Ethnic Adaptation and Identity. The Karen on the Thai Frontier
with Burma (Philadelphia: Institute for the Study of Human Issues, 1979).
———. “Introduction”, in Keyes, ed., ibid., pp. 1–24.
———. “Presidential Address: ‘The Peoples of Asia’ – Science and Politics in the
Classification of Ethnic Groups in Thailand, China and Vietnam”, The Journal
of Asian Studies 61, no. 4 (November 2002): 1163–203
Khaleque, Kibriaul. “Garo Kinship Terminology: Idealization and Reality”, The
Dhaka University Studies 39 (1983): 153–80.
———. “My Fieldwork Experience in a Garo Village of Bangladesh”, in Chowdhury,
ed., op. cit., pp. 207–23.
———. “The Operational Use of Anthropological Concepts of Garo Matrilineal
Descent Groups”, Social Science Review 2, no. 1 (1985): 67–85.
———. Social Change among the Garo. A Study of a Plains Village in Bangladesh
(Unpublished M.A. thesis for the Australian National University, 1982).
Kooiman, Dick. Conversion and Social Equality in India. The London Missionary
Society in South Travancore in the 19th Century (New Delhi, Manohar, 1989).
Kosinski, Leszek A. and K. Maudood Elahi, eds. Population Redistribution and
Development in South Asia (Dordrecht, Boston and Lancaster: D. Reidel Publishing
Company, 1985).
Kosinski, Leszek A. and K. Maudood Elahi. “Introduction”, in Kosinski and Elahi,
eds., ibid.
Kosinski, Leszek A. “Refugee Problems in Bangladesh”, in Kosinski and Elahi, eds.,
ibid.
Kotalová, Jitka. Belonging to Others. Cultural Construction of Womanhood in a village
in Bangladesh (Dhaka: University Press Limited, 1996).
Kottuppallil, George. “Roman Catholic Work among the Garos 1911–1933”, in
Indian Church History Review (n.d.), pp. 111–21.
Kudaisya, Gyanesh. “The Demographic Upheaval of Partition: Refugees and
References 227
Agricultural Resettlement in India 1947–1967”, South Asia XVIII (Special Issue

1995).
Kuper, Adam. Anthropology and Anthropologists. The Modern British School (Londen
and New York: Routledge, 1991).
———. The Invention of Primitive Society. Transformations of an Illusion (London and
New York: Routledge, 1991).
Lal, Dena. Christian Missions and Colonialism. A Study of Missionary Movement in
Northeast India with particular reference to Manipur and Lushai Hills, 1894–1947
(Shillong: Vendrame Institute, 1988).
Lehman, F.K. “Who are the Karen, and Is So, Why? Karen Ethnohistory and a
Formal Theory of Ethnicity”, in Keyes, ed., op. cit., pp. 215–54.
Leicht, F. and S. Karotemprel, eds. Early History of the Catholic Missions in Northeast
India, 1598–1890 (Calcutta: Firma KLM, 1989).
Luhrman, T.M. The Good Parsi. The Fate of a Colonial Elite in a Postcolonial Society
(Delhi, etc.: Oxford University Press, 1996).
Mackenzie, Alexander. The North-East Frontier of India (Delhi: Mittal Publications,
1989, first published in 1884, entitled History of the Relations of Government
with the Hill Tribes of the North-east Frontier of Bengal).
Major, Andrew J. “ ‘The Chief Sufferers’: Abduction of Women during the Partition
of the Punjab”, South Asia XVIII (Special Issue 1995): 57–72.
Majumdar, Asok. Peasant Protest in Indian Politics. Tebhaga Movement in Bengal (New
Delhi: New Publishers, 1993).
Majumdar, Dhirendra Narayan. A Study of Culture Change in Two Garo Villages
of Meghalaya (Gauhati: Gauhati University Department of Publications,
1980).
Maniruzzaman, Talukder. “Bangladesh in 1975: The Fall of the Mujib Regime and its
Aftermath”, Asian Survey 16, no. 2 (1976): 119–29.
———. “Bangladesh in 1976: Struggle for Survival as an Independent State”, Asian
Survey 17, no. 2 (1977): 191–200.
Manorama, Ruth. “The Situation of Dalit Women”, Indigenous Affairs 2 (1995):
4–7.
Marak, Julius. Garo Customary Laws and Practices (A Sociological Study) (Calcutta:
Firma KLM Private limited, 1986).
Marak, Sergius Siram. “The Indigenous People of Bangladesh Cry in the Wilderness”,
Chiring, International Year for the World’s Indigenous People Issue (2 October
1993).
McKinnon, John and Wanat Bhruksasri, ed. Highlanders of Thailand (Singapore,
Oxford, New York: Oxford University Press, 1986).
McKinnon, John and Bernard Vienne. Hill Tribes Today. Problems in Change (Bangkok:
White Lotus-Orstrom, 1989)
M’Cosh, John. Topography of Assam (Calcutta: G.H. Huttman, Bengal Military
Orphan Press, 1837).
Memmi, Albert. The Colonizer and the Colonized (London: Earthscan Publications,
1990, translated from Portrait du Colonisé précédé du Portrait du Colonisateur,

1957).
Mey, Wolfgang, ed. They are Now Burning Village After Village. Genocide in the
Chittagong Hill Tracts, Bangladesh (Copenhagen: IWGIA, 1984).
Migdal, Joel S. Strong Societies and Weak States. State-Society Relations and State
Capabilities in the Third World (Princeton, New Jersey: Princeton University
Press, 1988).
Mills, Sara. Discourse (London and New York: Routledge, 1997).
Moffatt Mills, A.J. Report on the Province of Assam (Calcutta: Jones, Calcutta Gazette
Office, 1854).
Murray Li, Tania, ed. Transforming the Indonesian Uplands. Marginality, Power and
Production (Australia, etc.: Harwoord Academic Publishers, 1999).
Murray Li, Tania. “Introduction”, in Li, ed., ibid., pp. xiii–xxiv.
———. “Marginality, Power and Production: Analysing Upland Transformations”, in
Li, ed., ibid., pp. 1–44.
———. “Relational Histories and the Production of Difference on Sulawesi’s Upland
Frontier”, Journal of Asian Studies 60, no. 1 (February 2001): 41–66.
Mymensingh Jila Trust. 71-er Proshikkon Shibirer Pramanyo Dolil [Authentic documents
of the Training Camps of ’71] (Mymensingh: Mymensingh Jila Trust, n.d.).
Nagengast, Carole and Michael Kearney. “Mixtec Ethnicity: Social Identity, Political
Consciousness, and Political Activism”, Latin American Research Review 25, no.
2 (1990): 61–91.
Nakane, Chie. Garo and Khasi. A Comparative Study in Matrilineal Systems (Paris:
Mouton and Co., 1967).
Nandy, Ashis. The Intimate Enemy. Loss and Recovery of Self Under Colonialism (Delhi,
Neff Gurney, Joan. “Female Researchers in Male-Dominated Settings. Implications
for Short-Term Versus Long-term Research”, in Experiencing Fieldwork. An
Inside View of Qualitative Research, edited by William B. Shaffir and Robert A.
Stebbins (Newbury Park, etc.: Sage Publications), pp. 53–61.
Nokmandi Prakashana. A Census of Garo Housemaids and Others in Dhaka (Dhaka:
Nokmandi Prakashana, 1994).
Novak, James J. Bangladesh. Reflections on the Water (Dhaka: University Press Limited,
1994).
Nawaz, Ali. “The Tribal and Other Links of Bangladesh”, The New Nation,
21 February 1988.
Oddie, Geoffrey A., ed. Religion in South Asia. Religious Conversion and Revival
Movements in South Asia in Medieval and Modern Times (New Delhi: Manohar,
1991, 2nd revised edition).
Oddie, Geoffrey A. “ ‘Orientalism’ and British Protestant Missionary Constructions
of India in the Nineteenth Century”, South Asia 17, no. 2 (1994): 27–42.
Östör, Ákos, Lina Fruzzetti and Steve Barnett, eds. Concepts of Person. Kinship, Caste,
and Marriage in India (Delhi, etc.: Oxford University Press, 1983).
References 229
Owen Hughes, Diane and Thomas R. Trautman, eds. Time. Histories and Ethnologies
(Ann Arbor: The University of Michigan Press, 1995).
Paraka Garo. Cultural Centre, Mande-rang-ni Chiti, Easter Issue (1996).
Peterson Royce, Anya. Ethnic Identity. Strategies of Diversity (Bloomington: Indiana
Pfeffer, Georg and Deepak Kumar Behera, eds. Contemporary Societies: Tribal Studies.
Volume V: Concept of Tribal Society (New Delhi: Concept Publishing Company,
2002).
Pianazzi, A. In Garoland (Tura, Assam: Catholic Mission, 1935).
Playfair, A. The Garos (Gauhati and Calcutta: United Publishers, 1975, first published
in 1909).
Porter, A.E. Census of India, 1931, Volume V. Bengal and Sikkim Part I (Calcutta:
Central Publication Branch, 1933).
Prakash, Gyan. “Science ‘Gone Native’ in Colonial India”, Representations 40 (1992):
153–178.
Qureshi, Mahmud Shah, ed., Tribal Cultures in Bangladesh (Rajshahi: Institute of
Bangladesh Studies, Rajshahi University, 1984).
Quarishi, Ferdaus A. Christianity in the North Eastern Hills of South Asia. Social
Impact and Political Implications (Dhaka: University Press Limited, 1987).
Radcliffe, Sarah A. “The Geographies of Indigenous Self-representation in Ecuador.
Hybridity, Gender and Resistance”, in European Review of Latin American and
Caribbean Studies, 63 (December 1997): 9–28.
Rashiduzzaman, M. “Bangladesh in 1977: Dilemmas of the Military Rulers”, Asian
Survey 18, no. 2 (1978): 126–34.
———. “Changing Political Patterns in Bangladesh: Internal Constraints and External
Affairs”, Asian Survey 17, no. 9 (1977).
Ravi, Srilata, Mario Rutten and Beng-Lan Goh, eds. Asia in Europe, Europe in Asia
(IIAS: Leiden/ISEAS: Singapore).
Rema, Peter. “The Aborigines of Greater Mymensingh”, Angshidar (Partnership).
1993 International Year for World’s Indigenous People (1993), p. 50.
Reynolds, C.S. “A Narrative of Our Connexions with the Dusannee and Cheannee
Garrows, with a Short Account of Their Country”, Journal of the Asiatic Society
of Bengal 18, Part 1 (January–June 1849): 45–61.
Reynolds, H.J. History and Statistics, Dacca Division (Calcutta: E.M. Lewis, Calcutta
Central Press Company Limited, 1868).
Risley, H.H. The Tribes and Castes of Bengal. Ethnographic Glossary, Volumes I & II
(Calcutta: Firma Mukhopadhyay, 1981, first published in 1891).
Robb, Peter, ed. The Concept of Race in South Asia (Delhi, etc.: Oxford Press, 1995).
Robinson, William. A Descriptive Account of Assam: With a Sketch of the Local
Geography, and a Concise History of the Tea Plant of Assam: To which is Added,
A Short Account of the Neighbouring Tribes, Exhibiting their History, Manners,
and Customs (New Delhi: Sanskaran Prakashak, 1975, original publication in
1841).
Romanucci-Ross, Lola and George A. De Vos, eds. Ethnic Identity. Creation, Conflict
and Accommodation (Walnut Creek/London/New Delhi: Altamira Press, 1995,
3rd edition).
Rowney, Horatio Bickerstaff. The Wild Tribes of India (Delhi: Low Price Publications,
Roy, Suprokash. Bharoter-Bidroho O Gonogantric Shongram (Calcutta: Book World,
1966).
Royce, Anya Peterson. Ethnic Identity. Strategies of Diversity (Bloomington: Indiana
Rozario, Santi. Purity and Communal Boundaries. Women and Social Change in a
Bangladeshi Village (Sydney: Allen & Unwin, 1992).
Sachse, F.A. Final Report of the Survey and Settlement Operations in the District of
Mymensingh, 1908–1919 (Calcutta: Bengal Secretariat Book Depot, 1920).
———. Mymensingh, Bengal District Gazetteers (Calcutta: Bengal Secretariat Depôt,
1917).
Sahlins, M. Tribesmen (Englewood Cliffs, N.J.: Prentice Hall, 1968).
Said, Edward W. Orientalism (London, etc.: Penguin Books, 1995).
Sangma, Milton S. History of American Baptist Mission in North-East India (1836–
1950). Part I & II (Delhi: Mittal, 1987).
———. History and Culture of the Garos (New Delhi: Books Today, 1979).
Sangma, Sengjrang N. Bangladeshi Immigrants in Meghalaya. Causes of Human
Movement and Impact on Garo Hills (Kolkata: Anshah Publishing House,
2005).
Sattar, Abdus. The Sowing of Seeds. The Sociology of Primitive Sex (Dacca: Abdeylebros
& Co, 1978).
———. In the Sylvan Shadows (Dhaka: Bangla Academy, 1983, first published in
1971).
———. Tribal Culture in Bangladesh (Dacca: Muktadhara, 1975).
Schipper, Mineke. De boomstam en de krokodil. Kwesties van ras, cultuur en wetenschap
(Amsterdam: Van Gennep, 1995).
Schulte Nordholt, Henk. “Locating Southeast Asia: Postcolonial Paradigms and
Predicaments”, in Asia in Europe, Europe in Asia, edited by Srilata Ravi, Mario
Rutten and Beng-Lan Goh (IIAS: Leiden/ISEAS: Singapore), pp. 36–56.
Scott, James C. Seeing like a State. How Certain Schemes to Improve the Human
Condition Have Failed (New Haven and London: Yale University Press), 1998.
Sen, Sunil. “The Dilemma”, in Peasant Struggles in India, edited by A.R. Dessai
(Delhi, etc.: Oxford University Press, 1979), pp. 461–68.
———. Peasant Movements in India. Mid-nineteenth and Twentieth Centuries (Calcutta
and New Delhi: K.P. Bagchi & Company, 1982).
———. “Tebhaga Chai,”, in Dessai, ed., op. cit., pp. 442–52.
Sen, Sipra. The Tribes of Meghalaya (Delhi: Mittal Publications, 1985), pp. 14–62.
Sen Gupta, Kanti Prasanda. The Christian Missionaries in Bengal 1793–1833 (Calcutta:
K.L. Mukhopadhyay, 1971).
References 231
Shaffir, William B. and Robert A. Stebbins, eds. Experiencing Fieldwork. An Inside

View of Qualitative Research (Newbury Park, etc.: Sage Publications, 1991).
Shakespear, L.W. History of the Assam Rifles (Calcutta: KLM, 1977, first published in
1929).
Silberstein, Lawrence J., ed. Jewish Fundamentalism in Comparative Perspective. Religion,
Ideology and the Crisis of Modernity (New York: New York University Press,
1993).
Silberstein, Lawrence J. “Religion, Ideology, Modernity: Theoretical Issues in the
Study of Jewish Fundamentalism”, in Silberstein, ed., ibid., pp. 3–26.
Sinha-Kerkhoff, Kathinka. Tyranny of Partition: Hindus in Bangladesh and Muslims in
India (New Delhi: Gyan Publications, 2006).
Sinha-Kerkhoff, Kathinka and Ellen Bal. “De-partitioning Society: Contesting Borders
of the Mind in India and Bangladesh”, in The Partition Motif in Contemporary
Conflicts, edited by Smita Jasal and Eyal Ben-Ari (New Delhi, Thousand Oaks,
London: Sage Publications, 2007), pp. 75–98.
Singer, Milton, and Bernard S. Cohn, eds. Structure and Change in Indian Society
(Chicago: Aldine Publishing Company, 1968).
Singh, K.S., ed. People of India. Meghalaya, Volume 32 (Calcutta: Seagull Books,
1994).
Skaria, Ajay. “Shades of Wildness. Tribe, Caste, and Gender in Western India”,
Modern Asian Studies 56, no. 3 (1997): 726–45.
Soddy, Gordon. Baptists in Bangladesh. A Historical Sketch of more than One Hundred
Years’ Work of the Baptist Missionary Society in Bengal (Dhaka: National Council
of Churches – Bangladesh, 1978).
Swarup, Devendra, ed. Politics of Conversion (Delhi: Deendayal Research Institute,
1986).
Syiemlieh, David R. British Administration in Meghalaya. Policy and Pattern New
Delhi: Heritage Publishers, 1989).
Tajfel, Henri. Human Groups and Social Categories. Studies in Social Psychology
(Cambridge: Cambridge University Press, 1981).
Talukdar, S.P. The Chakmas. Life and Struggle (New Delhi: Gian Publishing House,
1988).
Thapar, Romesh, ed. Tribe, Caste and Religion in India (Delhi: Macmillan, 1977).
Thompson, Paul. The Voice of the Past (Oxford: Oxford University Press, 1989).
Timm, R.W. The Adivasis of Bangladesh (London: Minority Rights Group,
1991).
Tripura, Prashanta. “The Colonial Foundation of Pahari Identity”, Journal of Social
Studies 58 (1992).
Umar, Badruddin. “The Struggle of the Peasants. Class Struggles in East Pakistan and
the Emergency of Bangladesh, XXI”, Holiday, 1 January 1999.
Van Schendel, Willem, ed. Francis Buchanan in Southeast Bengal (1798). His Journey
to Chittagong, the Chittagong Hill Tracts, Noakhali and Comilla (Dhaka: University
Press Limited, 1992).
Van Schendel, Willem. The Bengal Borderland. Beyond State and Nation in South Asia
(London: Anthem Press, 2005).
———. “Geographies of Knowing, Geographies of Ignorance: Jumping Scale in
Southeast Asia”, Environment and Planning D: Society and Space 20 (2002):
647–68.
———. “The Invention of the ‘Jummas’: State Formation and Ethnicity in Southeastern
Bangladesh”, Modern Asian Studies 26, no. 1 (1992): 95–128.
———. “ ‘Madmen’ of Mymensingh: Peasant Resistance and the Colonial Process in
Eastern India, 1824–1833”, The Indian Economic and Social History Review 22,
no. 2 (1985): 139–73.
———. Peasant Mobility. The Odds of Life in Rural Bangladesh (New Delhi: Manohar,
1982).
Van Schendel, Willem and Ellen Bal, eds. Banglar Bohujati: Bangali Chhara Banglar
Onnyanyojatir Proshongo (Calcutta: International Centre for Bengal Studies,
1998).
Van Schendel, Willem and Ellen Bal. “Bhumika: Name Ki Eshe Jay?”, in Van
Schendel and Bal , eds., ibid., pp. 9–24.
———. “Beyond the ‘Tribal’ Mind-Set: Studying Non-Bengali Peoples in Bangladesh
and West Bengal”, in Contemporary Societies: Tribal Studies. Volume V: Concept of
Tribal Society, edited by Georg Pfeffer and Deepak Kumar Behera (New Delhi:
Concept Publishing Company, 2002), pp. 121–39.
Vedantham, T.R. “Church as a Tool of Imperialism in the Modern World”, in Politics
of Conversion, edited by Devendra Swarup (Delhi: Deendayal Research Institute,
1986), pp. 71–79.
Veer, Peter van der. Religious Nationalism. Hindus and Muslims in India (Berkeley, Los
Angeles and London: University of California Press, 1994).
Vermeulen, Hans and Cora Govers. The Anthropology of Ethnicity. Beyond “Ethnic
Groups and Boundaries” (Amsterdam: Het Spinhuis, 1996).
Waddell, L.A. “The Tribes of the Brahmaputra Valley: A Contribution on their
Physical Types and Affinities”, Journal of the Asiatic Society, Part III. 1900.
Anthropology, no. 1 (1900).
White, Sarah C. Arguing with the Crocodile. Gender and Class in Bangladesh (Dhaka:
University Press Limited, 1992).
William Carey. The Garo Jungle Book (Tura: Tura Book Room, 1993, first published
in 1919).
Wolf, Eric R. Europe and the People Without History (Berkeley, etc.: University of
California Press, 1990).
Wood, Peter. “Afterword: Boundaries and Horizons”, in Hefner, ed., op. cit., pp.
305–21.
Zürcher, Erik-Jan and Willem van Schendel, eds. Identity Politics in Central Asia and
the Muslim World (London: I.B. Tauris, 2000).
Index 233
INDEX
A Atiagra, 87
Abeng, 15, 64, 74, 77, 89, 92, 95, 146 Atongs, 10, 87, 89, 90, 91, 92, 95, 103
Abengyas, 73 Australian Baptist Mission, 171
aboriginal tribes, 9, 23 Australian Baptists, 134, 167
Aboriginal Welfare Association, 165 Australian Baptist Missionaries, 180
Aboriginal Welfare Board, 171 Awami League, 186, 194, 195
aborigines, 9, 10, 35
Achik, 6, 64, 73, 74, 84 B
Achik Shongho Social Movement, 169, Babu Joy Nath Chaudhuri, 138
170, 171, 175, 176 Babul, 89, 103
Achikrang, 64 Badruddin Umar, 176
adivasi minority, 60 Baldwin, C.D., 143, 154
adivasis, 2, 10, 13, 37, 53 Banais, 80, 143, 162, 187
administrative researchers, 31 Bangladesh, 9, 10
agriculture independence of, 188
dependence on, 4 relief efforts for, 189
Amerindian population, 9 significance of emergence of, 213
Ansars, 184 uni-cultural stance, 190
pro-Muslim, 182 Bangladesh Garo Student Organization
Anthropological Survey of India, 82 (Bagachas), 60
Areng, 108 Bangladesh Nationalist Party, 214
Aguri, 47 Bangladesh Population Census (1991), 24
Ambasta, 47, 48 Bangladesh Rifles, 190
anachronism, 32 Bangladeshi Garos, 6, 50, 72, 88, 92,
Anglican church, 137 106, 193, 215
anthropological research, 5 Bangladeshis
anthropologists, 32, 37 distinguished from Bengalis, 6
anti-colonial agitations, 167 Banglapedia
arranged marriages, 64 Internet version, 1, 39, 108
Aryan race, 35 Baptist church, 138, 148
Assam, 170 Baptist mission, 151
Bengali refugees from, 178 Baptists, 137
Assamese of Boko, 80 complaints of, 147
233
11 TheyAsk_EatFrogs Index By: ROS 233 Size: 6" x 9" J/No: 06-15540 5/4/07, 5:30 PM Fonts: AGaramond
234 Index
Barooah, Nirode K., 85 British

Barth effect of departure, 175
Ethnic Groups and Boundaries, 4 British colonialism
Bastin, R.W., 65, 145, 156, 166, 167 nature of, 31
Battacharjee, Jayanta Bhusan, 78 British colonialists
Baumann, Gerd, 24 classification of population, 28
Bayley, W.R., 144, 145 British India
Bayly, Susan, 31, 35, 45, 46, 153 break up of, 172
Bengal Borderland, 6 British Indians, 35
Bengali, 6, 10 British East India Company, 70
influx of, 176 British Occupation, 79
middle class, 58 Buchanan, 73, 74, 82
Bengali Baptist missionary, 138 use of “tribe”, 75
Bengali culture, 6 Buchanan, Francis, 71
Bengali dressing style Burling, 83, 84, 86, 90, 93, 100
adoption by Garos women, 28 visit to Garo Hills, 146, 149, 154
Bengali journalist, 53 Burling, Robbins, 63, 72
Bengali Muslims, 60 Butler, John, 71, 86
Bengali population
little interest in Christianity, 137 C
Bengali refugees caste classification, 34
influx of, 155 Catholic Chronicles, 177, 182
Partition (1947), 172 Catholic church, 137, 148
Bengali women, 66 Catholic Garo priest, 61
Bengalis, 91 Catholic Mission of Modhupur Forest,
Bengalization, 193 152
Bessaignet, Pierre, 40 Catholic missionaries
Beteille, Andre, 48 Biroidakuni, 177
betel leaf plantations, 11 Catholic students, 149
Bibalgree, 52, 94, 102, 139, 143, 148 Catholics, 148
Bilduri, 97 complaints of, 147
Birisiri, 171 Chakma Chakma, 57
Birisiri Church, 138 Chakma delegation
Biroidakuni, 147 meeting with Mujib, 189
Birodukani Chronicles, 205 Chakmas, 66, 91
Biroidakuni Mission, 148 chatchis, 95, 105
Bodo group, 80 Chibok, 89, 90, 95, 99, 103, 104
borkondazes, 160 Chie Nakane, 89
Boundary Commission, 170 children
boundary demarcation, 109 education of, 146
Brahman Hindus, 36 Chiring, 60
Brak, 89 Chisak, 94
Index 235
Chisim, 94, 95 Commissioner of Cooch Behar, 71

Chisim, Suborno, 16, 17 Communists, 176
see also Suborno conversions
Chittagong, 11, 55 definition, 133
Chittagong Hill Tracts, 11, 66, 182, Christianity, to, 135
186, 189, 192, 202 converts
Christian community abstaining from chu, 155
Bangladesh, in, 137 Cossyahs, 79
Christian Garos, 98 Counch, 74
Christian lifestyles, 58 cultural variety, 67
Christian missionaries, 15, 213 culture
effect of British departure, 175 importance of, 4
widely adopted method, 156
Christian NGOs, 189 D
Christianity, 14, 54, 104, 141, 148 Dacca, 83
boundary marker, as a, 145 Dacca Artillery Park, 160
conversion into, 138, 142, 151, 168 dalits, 23
ethnogenesis of Garos, 152 Dalton, 73, 75, 86
role in Garos society, 92 Dalton, Edward Tuite, 71
role of, 137, 138, 213 Dalus, 162, 185
Christian schools, 146 Damakchigirif, 78
chu (rice beer), 138 De Vos, George, 64
Church of Bangladesh, 134 debates
Civil Officer, 68 caste and race, on, 210
Civil Service, 188 definitions
Civil Service Examinations, 194 state, 159
Cohn, Bernard, 96 territorialization, 164
colonial administrators, 30 tribe, 7, 8
colonial history deforestation, 59
reconstruction of, 29 Dhaka, 11, 16, 55
colonial observers expatriates in, 59
India, on, 43 Diggil, 89
colonial period, 159 Dhaka Garo community, 12
colonialism Dhaka University, 12
scientific observations during, 31 Dipesh Chakrabarty, 33
state hegemony, 164, 165 Dirks, Nicholas, 3
colonalists division
attempts at categorizing local unity through, 147
population, 211 documentary, 23
preoccupation of, 36 Garos of Bangladesh, on, 62
Commissioner for Scheduled Castes dol, 90, 91, 92, 95, 106, 150
and Tribes, 8 Dolgoma, 97
236 Index
domestic servants, 195 Garo community, 12

Drong, 94 Garo culture, 5
Dual, 89, 95, 103, 104 Garo diaspora, 105
Dugol, 75 Garo divisions, 89–96
Durgapur thana, 178 Garo dols, 97
Garo emigration, 177
E Garo Hills, 6, 40, 70, 71, 76, 97, 144,
East Bengal 149, 161, 177, 187, 195, 196, 214
Garos of, 165 incorporation into British India, 164
East Pakistan, 9, 182, 214 international border, effect of, 173
Garos, 171 proper Garo territory, 163
Indian Muslim refugees, 178 Garo Hindus, 156
East Pakistan Rifles, 178 Garo identity, 12
Eidhem, Harald, 64 Garo immigration, 177
Eliot, 71, 82 Garo inheritance, 95
Eliot, John, 5 Garo kinship, 95
emic discourse, 51 Garo kinship divisions, 93, 94
Enemy Property Ordinance, 183, 206 Garo language, 54
environmental issues, 61 absence of written, 70
Eriksen, Thomas, 4, 65, 66 Garo narratives of Self, 211
Ethnic Groups and Boundaries, 4 Garo Newsletter, 59
ethnic pluralism, 64 Garo political system, 77
ethnic relations, 51 Garo Progothi Shongho, 56, 57
ethnicity Garo recollections
anthropological studies of, 4, 5 dol, of, 91
ethnography Garo systems of classification, 54
Garos, 29 Garo sub-divisions, 80
Europe Garo tribes, 74, 75, 76, 77
scientific practice in, 33 Garo women
dressing, 28, 44
F Garo-ness, 5
fakirs, 162 etic and emic ideas, 67
food crisis, 176 public narratives, 58–62
freedom fighters Garoland, 136, 137, 148
Garos, 187 Garorang, 64
Garos, 10, 36, 166, 186, 187
G arrival in India, 180, 181
Gain, Philip, 66 Bangladesh, of, 6, 11–15, 39–43,
Ganchi, 89 184–19
Ganchings, 87 Bangladeshi, 72, see also Bangladeshi
Garo, 87, 89, 162 Garos
conceptions of identity, 212 central organization, lack of, 79
narratives of Self, 18 communication lines, 52
Index 237
distinct ethnic community, 212 Hefner, Robert W., 153

economic progress of, 195 herbal potions, 99
education, 59, 149, 150 hill tribes, 67
effect of Partition (1947), 172, 173 hilly border areas, 70
embedded in Bangladesh, 197 Hindu castes, 43
ethnic identity, 62, 63 Hindu Garo community, 152
ethnogenesis of, 158 Hindu society, 34
flight to India, 99, 173, 174, 179 Hinduism, 142, 143, 144
heterogenous understanding of Self, 51 conversion into, 145
importance of Christianity, 92 Hindus, 30
influence of states, 159 conversion into Christianity, 140
loss of culture, 57, 58 Hindu-Muslim divide, 43
making their own history, 210 historiography
matrilineal society, 14, 166 Garos, 29
migration, 55 Hinton, Peter, 82
organizing, 146, 147 historical perspective
perceptions of state, 159 history (20th century), 70
physical appearance, 53 importance of, 209, 210
preserving identity, 60 Hodies, 143, 162
pride, 60, 61 homelessness, 61
prime division, 22 Hopkinson, 83
process of naming, 109 household servants
publications on, 29 survey of, 66
return to Pakistan, 180, 181, 183 human rights organizations, 39
trading by, 199 Hunter, W.W., 77
Garos category, 79 hybrid race, 81
Garos language, 80
Garrow Hills, see also Garo Hills, 72 I
Goalpara, 83 identity constitution, 64
Government of Bengal (1884), 31 Inden, Ronald, 34
Government of India (1884), 31 Independent Commission on
Greater Mymensingh district, 11 International Humanitarian Issues,
gun ban, 99 49
gutok sam, 103 India
initial Orientalist view on, 43
H objectifying, 33–39
Hajongs, 80, 91, 143, 162, 164, 166, support for Kader Bahini, 193
168, 176, 185, 186, 187 Indian Border Security Force, 192
Haluaghat thana, 52, 101, 143 Indian ethnology, 37
hardship Indian historiography, 37
battle against, 63 Indian Medical Service, 31
harijans, 23 Indian society
Hawaraghat, 73, 74 colonialist perceptions, 28
238 Index
indigenous Kader Bahini, 190, 191, 192, 193

debate about notion of, 48 before end of, 193
Indigenous Peoples, 2, 9, 10, 38, 39, 43 Kader Siddiqui, 188, 190, 191
contemporary discourse on, 211 arrival in northern Mymensingh, 191
introduction of concept, 215 Kader’s Fighters, 190
Indigenous Peoples discourse, 61 Kaibortos, 47
Indira Gandhi kaliparaj, 48
loss in elections, 193 Kalumalupara, 74
inter-dol marriages, 106 Kamrup district, 80
inter-group contact Kamrup-Goalpara region, 160
intensity of, 106 Karam Shah, 200
inter-marriages, 102, 103, 104, 105 Karlsson, 86, 157
see also marriages, mixed Kasia, 76
international organizations, 39 Kaviraj, Sudipta, 82
internationalization Kearney, Michael, 64
Garo identity, 195 Kearns, Matthew, 138, 148
Islamic Republic of Pakistan Keyes, Charles, 67
establishment of, 171, 172 Khaleque, 90, 94, 108
Islamization projects, 184 Khaleque, Kibriaul, 90
isolation Khasis, 79
living in, 41, 42 matrilineal, 108
spoken language, 87
J Khasis language, 80
Jacobs, Julian, 68, 82 Kibriaul Khaleque, 63
Jaintias, 79 kinship, 92, 95
Jamaat-e-Islami, 214 kinship system, 105, 108
James Scott, 33 Kocha, 74
Jaring, 89 Koches, 10, 80, 96, 143, 162, 185
jat, 53, 64, 91, 107 Kochu, 89, 90, 104
jati, 53, 64 kriti, 65
Jayanta Bhusan Bhattacharjee, 78 Kuper, Adam, 44
Jengcham, 107 kushuk, 100, 150
Jenkins, F., 75
Jeypore, 78 L
Jhum, 11, 12 Lalung, 145
Jhumia, 11 landless labourers, 14
Jinaighati thana, 187 landlessness, 61
Joomea, 85 langta, 84
jumma, 10, 11 language, 100
Jummas, 84 landlords
repressive, 14
K Leach, Edmund, 3
Kachari, 145 Lehman, F.K., 82
Index 239
liberation war, 185, 188 matrilineal society, 93

Lookee Dooar, 78 M’Cosh, John, 75, 77, 83
lost culture Mechpara, 74
ideas of, 54, 55 Megam, 81, 89, 92, 101, 103
Lowland Garos, 6, 151, 171 Me’gams, 87
Lushais, 84 Meghalaya, 6
Lyntea, 73, 84 middle class, 57
migration, 55
M Mikir, 145
Maaker, Erik de, 107 mimangs, 139
Machi, 77 mimang kam, 105
ma’chong, 94, 96 Mirza Nathan, 1
Mackenzie, Alexander, 83 Mission House of Bengal’s Garoland,
Mahmud Shah Qureshi, 42 137
McKinnon, John, 82 missionaries, 136, 137
madong, 94 methods and attitudes, 140
mahari, 95, 96 tools of colonizer, as, 150
distribution of, 110 missionary activities, 168
maharis, 94 missionary mediators, 166, 167
mainstreamization, 196 mite, 25
Mande, 73 mixed marriages, 99
Mandi, 6, 72 modern Indian, 9
Mandi-rang ni Chiti, 59, 66 modern traditions, 63
Mandirang, 64 modernity
Mandis, 98 narrative of, 59, 60
unification of, 99 Modhupur, 63, 72, 139, 141, 144
Marak, 94 Modhupur Garos, 170
Marak, Julius, 81, 84, 87, 89 Modhupur forest, 13
marginality Momin, 108
understanding of, 47 Mon-Khmer language, 81, 87
marginalized position, 7 Moulvi Bazar, 24
market economy Mrong, 94
dependency on, 13 Mughal emperor, 161
marriages, 94 Mughal rule, 159, 160
arranged, 94 Mughal rulers, 70
mixed, 99, 112–28 Mughal tax collector, 161
restrictions, 100 Mujibite, 191
married women Mukti Bahini, 185, 186
death of, 109 Muslim Bengali migrants, 98
Matabengs, 77 Muslim Bengalis, 166
matrilineal landless, 178
definition of, 14 Muslim League, 175, 176
matrilineal grouping, 108 Muslim peasantry, 13
240 Index
Muslims, 30 attitude towards Garos, 181–84

perceptions about, 98 attitude towards missionaries, 175
Myanmar, 3 Islamic identity of, 185
Mymensingh, 6, 12, 13, 41, 70, 83, 89, perceptions about, 174
143, 163, 170, 190, 198 Pani Koches, 145
buffer states, 160, 161 Paraka, 59
exclusive policies concerning, 165 Partha Chatterjee, 32
influx of Bengalis, 178 Partially Excluded Area, 165, 166, 170,
international border, 173 187
northern, 164, 198 Partition (1947), 100, 169, 172
population of, 164 Patang, 6, 7, 94
Mymensingh Garos, 196 peasant uprisings, 176
People of India series, 68, 86
N people of nature, 61, 62
Nagengast, Carole, 64 people’s rights, 61
Nakane, 95 persistent image
narratives of self, 50–66 history of, 27–44
national encyclopaedia, 1 Playfair A., 41, 68, 73, 74, 78, 80
Nesfield, John, 47 political system, 77, 78
Nokmandi Survey, 66 polygamy, 110
nokna, 14, 95 prejudices, 103
non-Bengali minorities, 7 Primitive Tribes, 36, 38
non-Governmental organizations public perceptions, 62
(NGOs), 12, 56 purdah, 14, 65
Northeast India purity of blood, 65
scholars from, 68
Northern Indians, 35 R
Norway, 64 Rabhas, 80, 96
Nuniyas, 73, 75, 80 race
discourse of, 35, 36
O social category, as a, 2
oral history racial purity, 53
role of, 15 racial theories, 37
Orientalism, 32 Rajas of Kamrup, 76
Orientalist discourse, 32 Rajmahal Hills, 10
Orientalist perception, 30 Rajbansis, 162
Orientalists, 34 Ramjongga, 85
Oxfam, 180, 183 Rangpur, 70, 73, 83, 85
Raumari, 85
P religion
Pagol Panthi movement, 200 role of, 34
pahari, 10 religious conversions
Pakistan, 176 Christianity, 138
Index 241
Rema, 94, 95 Sira, 108

Rema, Peter, 66 situationalist approach
Rengsanggri, 89 culture and identity, 4
representations of Self, 62 slash-and-burn cultivation, 12, 59
Reynolds, C.S., 71, 86 smuggling, 208
Revenue Survey of Mymensingh, 201 social category
Revisional Survey, 165, 166 tribe, 2
Risley, H.H., 33, 47 social construct
Robbinson, William, 75 caste as a, 34, 35
Roman Catholic missionaries, 167 Somons, 89, 102
Rowney, Horatio Bickerstaff, 79, 86 South Asia
Rugas, 87, 89 tribe, 2
South Asian tribal studies, 209
S Southern Indians, 35
Sachse, F.A., 42 state
Salgetal, 58 definition of, 159
Sangma, 94 role of, 158
Sangma, Milton, 81, 87, 90 stereotyping, 103
Sangsarek, 15, 54, 92 Suborno, 50, 51, 155, 173
conversion into Christianity, 141 Sumit Guha, 41
Sangsarek festivals, 58 Sunamganj, 24
celebration of, 138 Suranjan Das, 13
Sangsarek Garos, 107 suppression
Sangsarek religion, 148 battle against, 63
Sangsareks, 25, 98, 139 Susangga, 74
sannyasis, 162 Swarajists, 168
sanskriti, 65
Santal, 10 T
Sattar, Abdus, 40 Talukder Manuruzzaman, 192
Scheduled Tribes of India, 8, 9, 37 Tangail district, 13
Schipper, Mineke, 63 tea plantations, 11
schools Tebhaga movement, 167
importance of, 150 territorialization
Scott, 71, 75, 77, 82, 85 definition of, 164
report in Barooah, 85, 86 Toju, 94
Scott, David, 71, 157 Topography of Assam, 83
Scott, James C., 46 Tribal Academy, 194
segmentation tribal image
process of, 32 dealing with, 60
self perception, 109 Tribal Welfare Association (TWA), 194
Seventh Day Adventists, 152 tribalist discourse, 7–11, 29, 43, 63
Sheikh Mujib, 185, 189, 190 tribes
shomaj, 107 category of 36, 37
242 Index
colonial construct, 8 W
common depiction of, 2 Waddell, L.A., 31, 32, 84
definition of, 7, 8 wana krita, 102
dominant notion of, 51 Wanat Bhruksasri, 82
Tripuras, 10 West Garo Hills, 107
Western culture, 142
wet-rice cultivation, 62
U
White, V.J., 165
United Nations Year for Indigenous
World Baptist Alliance, 180
Peoples, 9
World Christian Council, 180, 183
untouchables, 23
upland-lowland dichotomies, 47
Z
upojatis, 10, 37
zamindari influence, 163
zamindars, 14, 60, 70, 85, 160, 161, 199
V conflicts involving, 163
Van Schendel, Willem, 2, 6, 84 end of hegemony, 162
volunteers illegal practices, 162
Garo freedom fighters, 187 Ziaur Rahman, 185
ABOUT THE AUTHOR
Ellen Bal holds a Ph.D. in history and at present teaches Anthropology at the
VU University of Amsterdam. She has published various articles and books,
among them Manderangni Jagring: Images of the Garos of Bangladesh (Dhaka:
University Press Limited, 1999), with Yasuhiro Takami and Johannes Sandgren.
She has also been involved in research on the Indian diaspora in Surinam and
the Netherlands and has edited the Autobiography of an Indian Indentured
Labourer: Munshi Rahman Khan (with Kathinka Sinha-Kerkhoff and Alok
Deo Singh).
12 TheyAsk_EatFrogs Author By: ROS 243 Size: 6" x 9" J/No: 06-15540 5/4/07, 5:09 PM Fonts: AGaramond
IIAS/ISEAS Series on Asia
Series Editors: Wim Stokhof and Paul van der Velde
The IIAS/ISEAS Series on Asia takes a multidisciplinary approach to issues

of inter-regional and multilateral importance for Asia in a global context. The
series aims to stimulate dialogue amongst scholars and civil society groups at
the local, regional and international levels.
Titles in this Series
Geoffrey Benjamin and Cynthia Chou, eds., Tribal Communities in the Malay
World: Historical, Cultural and Social Perspectives (2002).
Srilata Ravi, Mario Rutten and Beng-Lan Goh, eds., Asia in Europe, Europe in
Asia (2004).
Wim Stokhof, Paul van der Velde and Yeo Lay Hwee, eds., The Eurasian
Space: Far More Than Two Continents (2004).
Cynthia Chou and Vincent Houben, eds., Southeast Asian Studies: Debates
and New Directions (2006).
Ellen Bal, They Ask if We Eat Frogs: Garo Ethnicity in Bangladesh (2007).
12 TheyAsk_EatFrogs Author By: ROS 244 Size: 6" x 9" J/No: 06-15540 5/4/07, 5:09 PM Fonts: AGaramond
View publication stats

Theyaskifweeatfrogs GaroEthnicityinBangladesh

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Theyaskifweeatfrogs GaroEthnicityinBangladesh

Uploaded by

Copyright:

Available Formats

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

They ask if we eat frogs: Garo ethnicity in Bangladesh

Article · January 2007

The user has requested enhancement of the downloaded file.

The Institute of Southeast Asian Studies (ISEAS) was established as an autonomous

First published in Europe in 2007 by

All rights reserved. No part of this publication may be reproduced, stored in a

© 2007 Institute of Southeast Asian Studies, Singapore.

ISEAS Library Cataloguing-in-Publication Data

ISBN 978-981-230-446-9 (hard cover)

Typeset in Singapore by Superskill Graphics Pte Ltd

2 The History of a Persistent Image 27

3 ‘The Importance of Being Garo’: Garo Narratives of Self 50

4 Peoples without History? 67

5 ‘Dual were Dual, Kochu were Kochu’: Garos Divided 88

6 Negotiable Boundaries, Negotiable Identities 111

7 Garos and Christianity 132

8 Garos and the State 158

9 Summary and Conclusion: From Tribes to Ethnic Minorities 209

About the Author 243

occasional friends and relatives I brought along. I am particularly indebted to

I feel particularly indebted and thankful to my research companion

Abeng Division of Garos and language spoken by Garos in

Gari Car, cart

Mujibite Follower of Sheikh Mujibur Rahman

Tebhaga Lit. three shares; movement by sharecroppers for two-

FIXED CATEGORIES, FROZEN IMAGES

This is a common depiction of a so-called tribe in South Asia. Hundreds of

In his seminal work entitled Castes of Mind, Nicholas Dirks argues

TOWARDS A SITUATIONALIST APPROACH OF TRIBE

WHY A HISTORY OF THE GAROS OF BANGLADESH?

TRIBALIST DISCOURSE AND ITS TERMINOLOGY

THE SCHEDULED TRIBES OF INDIA

INDIGENOUS PEOPLES AND ‘ABORIGINES’

In Bangladesh, it is often argued, without much historical evidence, that

UPOJATI, PAHARI, AND JUMMA

A similar example is Jumma. Jhum refers to a particular type of cultivation,

THE GAROS OF BANGLADESH

BREAKING THE SILENCE

of the twentieth-century history of the Garos, their stories derived their

THE RESEARCHER(S) AND THE RESEARCHED

professor. With me they obviously cared much less about “respectful

ORGANIZATION OF THE BOOK

7. Here, discourses are understood as “complexes of signs and practices which

13. Murray Li, “Marginality, Power and Production”, p. 1.

Bangali Chhara Banglar Onnyanyojatir Proshongo, edited by Willem van Schendel

What a bore.” Gerd Baumann, The Multicultural Riddle. Rethinking National,

It is a familiar experience that the ordinary untravelled European on first

Different observers, with different backgrounds, personalities, objectives, and

wrote at the turn of the century: “Unfortunately for science, however, no

CASTE AS A SOCIAL CONSTRUCT

THE GAROS OF BANGLADESH

thinking of many Bangladeshis today, is also a blueprint for descriptions of

PEOPLE WITHOUT HISTORY

In 1984, Mahmud Shah Qureshi introduced the most comprehensive

THE PRIMITIVE OTHER

mirror. Primitive man constituted the antithesis of these pioneer

TRIBALIST DISCOURSE AND THE

6. Cohn, “Study of Indian Society and Culture”, pp. 6–8.

19. Skaria, “Shades of Wildness”, pp. 727–29.

32. Bayly, “Caste and ‘Race’ ”, pp. 165–67.

Li (Australia, etc.: Harwood Academic Publishers, 1999), p. xvii. See also

51. For example, the Independent Commission on International Humanitarian