LEONARD Y. ANDAYA
of the SAME TREE
Trade and Ethnicity in the Straits of Melaka

Leaves of the Same Tree

Leaves of the Same Tree
Trade and Ethnicity in the Straits of Melaka
Leonard Y. Andaya
University of Hawai‘i Press
Honolulu

© 2008 University of Hawai‘i Press
All rights reserved
Printed in the United States of America
13 12 11 10 09 08 6 5 4 3 2 1
Library of Congress Cataloging-in-Publication Data
Andaya, Leonard Y.
Leaves of the same tree : trade and ethnicity in the Straits of Melaka / Leonard Y.
Andaya.
p. cm.
Includes bibliographical references and index.
ISBN 978-0-8248-3189-9 (hardcover : alk. paper)
1. Ethnicity—Asia, Southeastern. 2. Ethnology—Asia, Southeastern. 3. Asia,
Southeastern—Commerce—History. 4. Malacca, Strait of—Commerce—History.
I. Title.
DS523.3.A5 2008
305.8009595’1—dc22
2007044638
University of Hawai‘i Press books are printed on acid-free
paper and meet the guidelines for permanence and durability
of the Council on Library Resources.
Designed by Paul Herr
Printed by The Maple-Vail Book Manufacturing Group

To Barbara
It’s been a truly wonderful journey

vii
Contents
List of Maps | viii
Acknowledgments | ix
Introduction | 1
Chapter 1: Malayu Antecedents | 18
Chapter 2: Emergence of Malayu | 49
Chapter 3: Ethnicization of the Minangkabau | 82
Chapter 4: From Malayu to Aceh | 108
Chapter 5: The Batak Malayu | 146
Chapter 6: The Orang Laut and the Malayu | 173
Chapter 7: The Orang Asli/Suku Terasing and the Malayu | 202
Conclusion: Framing the Southeast Asian
Past in Ethnic Terms | 235
Notes | 241
Abbreviations | 285
Select Bibliography | 287
Index | 315

viii
Maps
Southeast Asia | 2
East-West Trade | 3
Sea of Malayu | 23
Minangkabau | 92
Northern Sumatra | 130
Batak | 171
Riau and Lingga Archipelagoes | 179
Malay Peninsula | 233

ix
Acknowledgments
H
aving reached the“twilight”of my academic career, I have
a greater appreciation of any of my projects that reach
fruition. This particular book has been long in the mak-
ing, not only because of the demands of teaching but also because of having
ventured into academic disciplines and cultural areas that were less familiar
to me. In the process I have benefited immensely from the generosity of many
individuals who were willing to guide me through difficult material. It makes
one aware that in any academic enterprise, collegial cooperation is essential.
To the following colleagues, many of whom read sections of the manuscript,
I would like to express my sincere gratitude for their help: Taufik Abdullah,
Sander Adelaar, Jane Allen, Geoffrey Benjamin, Leonard Blussé, Robert Blust,
David Bulbeck, Cynthia Chou, Robert Dentan, Juli Edo, Kirk Endicott, Jeff
Hadler, Tsuyoshi Kato, Uli Kozok, Michael Laffan, Adri Lapian, Henk Maier,
John Miksic, Henk Niemeijer, Colin Nicholas,Wannasarn Noonsuk, Jon Oka-
mura, Nathan Porath, Jim Scott, Miriam Stark, and Wazir Johan Karim. Any
omission of a name is not deliberate, but simply a sign of failing memory.
The major part of the research was undertaken in Indonesia, Malaysia,
and the Netherlands in 1999–2000 through the Fulbright-Hays Program, and
I would like to acknowledge the generous funding that made all of this
possible. In the Netherlands in 2000 I was fortunate to have been a Fellow-
in-Residence at the Netherlands Institute for Advanced Study. I would like to
take this opportunity to express my sincere appreciation to the staff for the
opportunity to meet such intellectually stimulating colleagues from all parts
of the world and for all the wonderful support facilities in conducting research
in that country. The Netherlands has some of the best libraries and archives
in the world for the study of Indonesia, and I would like to express my thanks
to the staff at the National Archives of the Netherlands in The Hague and

x Acknowledgments
at the library of the Royal Institute of Southeast Asian and Caribbean Stud-
ies (KITLV), especially Sirtjo Koolhof, for their help during my research in
their institutions. In Indonesia I benefited from the kindness of members of
LIPI (Indonesian Institute of Sciences), particularly my friend and colleague,
Dr. Taufik Abdullah. I also would like to thank the helpful staffs in Malaysia
of the Malaysian-American Commission on Educational Exchange (MACEE)
and the Asia-Europe Institute at the University of Malaya, and particularly
their generous and gracious executive directors, Dr. Don McCloud and
Dr. Shaharil Thalib. Back home at the University of Hawai‘i, I would like to
express my appreciation to Yati Paseng, our Southeast Asian bibliographer,
who continues to make a researcher’s life a pleasant one. To the staff of the
University of Hawai‘i Press, particularly Pam Kelley, my sincere thanks for
the very helpful suggestions in the preparation and the completion of the
manuscript. I am also grateful to Jane J. Eckelman, who so patiently drew and
redrew maps to my specifications, and to the University of Hawai‘i Research
Relations Fund for their financial assistance.
Once again, as in my previous works, I owe so very much to my wife
and colleague, Barbara Watson Andaya, who has been so patient and long-
suffering. Having married another Southeast Asian historian has had many
benefits, of which having a captive reader is one. Throughout this project she
has helped me to think through many difficult problems in conceptualization,
and in the final stages she has patiently (encouraged by promises of Starbuck’s
coffee) waded through the manuscript identifying inconsistencies, lapses in
analysis, etc. It makes me realize that O.W.Wolters perhaps was not being
sexist but merely pragmatic when he jokingly told his (male) students:“Marry
a typist.”I actually type better than Barbara, but she obviously has many more
redeeming qualities, and for that I am immensely grateful.
To all of you, my heartfelt thanks.

1
Introduction
O
f some six thousand ethnolinguistic groups in the world,
about a thousand are found in Southeast Asia. This
immense ethnic diversity has piqued the curiosity of lin-
guists,anthropologists, and sociologists, but oddly, not that of historians until
recently. In general the latter have tended to apply ethnic names loosely, giv-
ing insufficient attention to the nature of ethnic identity and the constant
redefinition of groups, particularly in the precolonial period (i.e., before the
late nineteenth century). Historians can therefore profit from social science
insights regarding the shifting components that constitute an ethnic group
and the complexity of ethnicity as a concept. One such insight, from the
anthropologists Jean and John Comaroff, recognizes the ambiguous nature
of ethnicity. “Is [ethnicity] an object of analysis, something to be explained?”
they ask. “Or is it an explanatory device capable of illuminating significant
aspects of human existence?” They then proceed to demonstrate the mutual
and dialectic influences between ethnicity as an analytic framework and eth-
nicity as a conceptual subject.1
The Comaroffs are just two of many social
scientists who have sought to explicate some aspect of this slippery concept.
From this vast array of theoretical ideas, I have selected those that I feel have
direct relevance to historians who wish to use ethnicity as a way of under-
standing Southeast Asian history.
The value of problematizing ethnicity becomes apparent in the context of
trade, long the lifeblood of Southeast Asians and one of the dominant themes
in the region’s history. Southeast Asia sprawls across the Indian and Pacific
Oceans, and until perhaps the sixteenth century the only known sea passage
through the region was the Straits of Melaka. Located midway between the
major civilizations to the east and the west, the straits proved an ideal haven
for ships because it was protected from the strong monsoon winds by parallel

2 Introduction
mountain chains along the spines of Sumatra and the Malay Peninsula. It was
the “endpoint” of both the northeast monsoons that blew between January
and April and brought traders from the east, and the southwest monsoons of
July to November, which carried traders from the west. While traders awaited
favorable winds to return home, the communities located astride the straits
quickly seized the opportunities the situation provided.They established ports
for traders to repair their ships, replenish supplies, obtain local products, and
exchange goods with merchants from all parts of the world. Furthermore, the
interior of both landforms that bordered the straits produced valuable for-
est products, particularly camphor, benzoin, gaharuwood (eaglewood), and
dragon’s blood (a kind of kino)—all of which were highly prized in the inter-
national marketplace, particularly in China.
For more than two thousand years, this narrow waterway brought trad-
ers, religious scholars, diplomatic missions, and adventurers to the ports
bordering its shores. As a result of the economic opportunities provided by
the steady influx of people and goods, communities in the vicinity of this
waterway became increasingly involved in international trade. Much has been
written about the impact of international and domestic trade in the transfor-
mation of Southeast Asian societies, both materially and spiritually. In every
period it was trade that served as the stimulus for the movement of goods and
ideas across continents, and Southeast Asia’s ideal location midway between
Banda Is.

Introduction 3
major civilizations provided its leaders with the luxury of surveying, experi-
menting, and selecting those elements that were most appropriate to advance
their societies. Little noticed by historians has been the role of trade in the
process of ethnic formation. The continuing presence of foreign merchants
and visitors contributed to an intense awareness of self among local individu-
als and groups. To maximize advantage, small socioeconomic units ethnically
identified by their location and involved in small-scale exchange gradually
began to join others of like mind to form numerically larger and more exten-
sive community networks.
The vicinity of the Straits of Melaka is an ideal site to investigate the
relationship between trade and ethnic formation, especially in precolonial
Southeast Asia. Before the middle of the first millennium of the Common
Era, the favored passage through Southeast Asia combined sea and land
routes across the Isthmus of Kra and the northern Malay Peninsula. While
these northern routes continued to be used in later centuries, they became
secondary to the preferred sea route through the straits. Communities bor-
dering or in close proximity to the Straits of Melaka were therefore blessed
with a continuing flow of seaborne commerce, bringing benefits to those
most effective in adjusting to the opportunities presented. In the process of
adapting to change, certain communities in the straits area saw the value of
detaching themselves from a larger ethnic identity to form smaller and more

4 Introduction
effective units, whereas others saw greater advantage in becoming affiliated
with a larger ethnic grouping.
Ethnic formation in the Straits of Melaka may have been stimulated fur-
ther by increasing contact with Europeans from the sixteenth century, the
century that has been called “a high point in the cycle of ethnic conscious-
ness” in Europe.2
With increased ethnic awareness, coupled with the desire to
classify and thus control, the Europeans assiduously listed local individuals
with whom they came into contact by their “ethnic group.” This was par-
ticularly evident in the ports, where European officials wished to control the
movement of certain rival or enemy groups. The results were predictable:
individuals tended to claim the most useful ethnic identity because there was
little to distinguish one group from another, and most could communicate
in Malayu, the trade lingua franca. When the Malayu3
kingdom of Johor was
given special privileges by the Dutch East India Company (VOC) for their
assistance in the seizure of Portuguese Melaka in 1641, there would have been
many who claimed to be Malayu from Johor. An opposite reaction occurred
when the Bugis of southwest Sulawesi were regarded as the enemies of the
VOC. They simply claimed to be Malayu, Javanese, or another more favored
ethnic community in order to be allowed to trade in Dutch ports and to travel
the seas free from VOC harassment.
Malayu ethnicity is an important theme in this study. In a situation of
increasingeconomiccompetitiontherewasapoliticizationof ethnicidentities,
or what Kahn has termed the “ethnicization” of groups.4
The emergence and
expansion of the Malayu resulting from a convergence of economic and politi-
cal interests encouraged at different times the formation of the Minangkabau,
the Acehnese, and to a certain extent the Batak ethnic identities. For such
groups, identifying cultural discontinuities within a common Malayu culture
was a necessary process in the erecting of ethnic boundaries.5
The Malayu
were also the stimulus for the formation of the new ethnic categories of Orang
Laut (Sea Peoples) and Orang Asli/Suku Terasing (Original Peoples/Isolated
Ethnic Groups, i.e., the forest and hill peoples). They performed valuable ser-
vices for the Malayu rulers as providers of ocean and jungle products and as
defenders of the routes through the various seas and forests. In return they
were richly rewarded economically and spiritually by the Malayu rulers, thus
encouraging the maintenance of this symbiotic exchange through the preser-
vation of separate lifestyles.Yet deliberate efforts by all groups in the straits to
erect ethnic boundaries to emphasize difference cannot disguise the fact that
they are “leaves of the same tree.”6
In this study I have attempted to capture the dynamism of the process of
ethnic formation with each individual group. Because of the unevenness in the
quality and quantity of materials available, it has not been possible to follow a

Introduction 5
single pattern of investigation nor to maintain a common time frame for all.
Instead, my primary concern has been to make the best use of the sources in
illuminating the process and thus demonstrating its vitality and significance in
the interpretation of Southeast Asian history. Too often the story of Southeast
Asia has been structured according to ethnic struggles, a presentist approach
that obscures the flexibility of ethnic identities in the past. I hope that this
work, focusing on trade and ethnic formation in a small area of Southeast Asia,
will encourage other historians to engage the issue of ethnicity to determine
the extent to which it informed the actions of Southeast Asians in the past.
Ethnicity as an Explanatory Device
The plethora of writings on ethnicity in the social sciences has led to a bewil-
dering variety of interpretations, raising some doubts regarding its useful-
ness as a concept. Yet scholars persist in attempting to understand ethnic-
ity because of the intensity of emotion that ethnic issues continue to evoke
among ordinary people. While some have argued that ethnicity is a modern
phenomenon, there is every reason to believe that group identity based on
shared beliefs, practices, and real and fictive ancestors would have been as
significant in the past.7
This is the proper task of the historian, who can bring
a different perspective to the studies of ethnicity long dominated by social
scientists. At the very least such an endeavor should encourage other histori-
ans to become aware of the problem of an unreflective acceptance of ethnic
communities as somehow fixed forever in time.
Anthropologists have demonstrated the fluidity and complexity of ethnic
identities, particularly in Southeast Asia. Edmund Leach’s classic 1954 study
of highland Burma reveals the ease with which a Kachin could become Shan
and a Shan a Kachin by means of a preference of one form of social sys-
tem over another. In viewing the Kachin as a complex product of its politi-
cal relations with neighboring distinctive communities, Leach encouraged a
new direction in the study of ethnicity.8
Since Leach’s work, social scientists
have examined the socially constructed and political nature of ethnicity, and
it has become clear that the colonial state and the modern nation-state have
been instrumental in the creation of ethnic categories and groups.9
Charles
Keyes has even argued that ethnicity has flourished as a result of national-
ist discourses.10
In the United States, the increasing politicization of ethnic
minorities has spawned an entire new field of ethnic studies and created new
identities based on geography (pan-Asian), as well as on culture and language
(Latino).11
Yet the interest in difference is a human quality, and there is every rea-
son to believe that ethnic ideas were also prominent in Southeast Asia’s past.

6 Introduction
Although people, and hence documents, may not have used such terms as
“ethnicity” or “nationalism,” there is no reason to believe that such notions
of group identities were absent. The anthropologist Richard O’Connor was
among the first to suggest that ecological adaptation, language, and agricul-
tural techniques are significant shifts that can explain the so-called “decline”
and “emergence” of ethnic groups in Southeast Asia.12
There are encouraging
signs that historians of Southeast Asia are finally engaging the issue of eth-
nicity. In a recent article, David K. Wyatt cautions against reading modern
ethnic identities into Thailand’s past.13
A similar critical reading of ethnicity
is addressed in Victor Lieberman’s 2003 study of Southeast Asia between the
ninth and nineteenth centuries.14
The persistence of ethnic issues suggests that
ethnicity should not be regarded simply as a precursor to nationalism of the
modern nation-state, but as a concept that was relevant in the past and may
help to illuminate the particular ways that events unfolded in Southeast Asia.
Although the much-quoted phrases “invention of traditions” and “imagined
communities” begin with the premise that this process was associated with
the creating of modern ethnic or nation-state nationalisms, this process was
also a feature of communities in precolonial Southeast Asia.15
The complexity of the subject demands a clarification of certain key
terms.“Ethnicity” is used throughout this work to refer to a way of conceptu-
alizing the world and acting in it by privileging group identity and interests.
Religion, class, and gender are other ways in which the past could be struc-
tured, but they are subordinated to and form components of ethnic identity.
The second key term is “ethnic group or community.” The historian
Anthony Smith believes that the French word ethnie best captures and com-
bines the distinction found in Greek between genos, applied to kinship-based
groups, and ethnos, a broader term used for groups sharing a culture. He lists
as attributes of an ethnie a collective name, a common myth of descent, a
shared history, a distinctive shared culture, association with a specific terri-
tory, and a sense of solidarity transmitted by the upper strata to the rest of the
community. The last point is particularly important because in times of crisis
all class, factional, regional, and other identities are submerged by the strength
of the group’s sense of solidarity.16
Smith’s ethnie attributes are relevant in the
formation of ethnic groups in the vicinity of the Straits of Melaka. In defining
a group, greatest emphasis was on a strong social network established through
real and fictive kinship ties, reinforced by shared myths and symbols associ-
ated with and often created by their leaders.
“Ethnic category” forms the third key term in this study. This refers to
a loose and generalized collectivity to which groups attach themselves or are
assigned by outsiders because of certain shared characteristics. While the
members of an ethnic category acknowledge some common cultural relation-

Introduction 7
ship, their interpersonal and intergroup relationships are limited. In central
Borneo, for example, such ethnic categories as Kayan, Kenyah, Kelabit, Penan,
etc., do not form social units or a distinct social system and may not even
share the same language and culture.17
A similar observation may be made of
the Orang Laut and Orang Asli/Suku Terasing ethnic categories in the Straits
of Melaka. Ethnic categories and ethnic groups are fluid concepts and can be
re-formed to include or exclude others.
Basic to the notion of ethnicity is that a group’s ethnic consciousness
arises through contact with others who are perceived as different. As Thomas
Eriksen explains, “ethnicity is essentially an aspect of a relationship, not a
property of a group.”18
Once difference is acknowledged, it is necessary to
exploit this difference through the establishment of ethnic markers. Com-
monly cited as ethnic markers are cultural elements, such as dress, clothing,
food, language, or even religious belief, but different ethnic groups may also
share the same cultural elements. For this reason Frederik Barth argues that
rather than focusing on the “contents,” one should identify the “boundaries”
erected by the group to distinguish itself from its neighbors. In his study of
the Pathans of Afghanistan,for example,he lists hospitality,councils of equals,
and seclusion of women as elements that make up the Pathan “boundary.”19
In a close reading of Barth’s study, however, Marcus Banks found evi-
dence that Pathans will in fact grudgingly claim a common ethnic unity
based on cultural features, or what Barth calls the “contents” of an ethnicity.
Among the shared features named by the Pathans are patrilineal descent from
a common ancestor, Islam, and custom, including language, oral literature,
and certain masculine attributes. Banks argues that both the Pathan-centric
and the Barthian-centric conceptions are closer to Barth’s “contents” than his
“boundaries,” since many of these features are shared by neighboring eth-
nic communities. Banks then makes the important observation that the only
principle that distinguishes the Pathans is their putative descent from a com-
mon ancestor.20
In 1998, responding to criticisms of his pioneering 1969 work on ethnic
boundaries, Barth modified his arguments. He acknowledged that in indi-
vidual lives, culture often consists of the blending of difference and of adap-
tation, rather than the erection of boundaries. For this reason he suggested
focusing on the process whereby variation of culture is identified and made
salient to form a shared understanding of the “cultural discontinuity” that
then forms the crucial boundary of an ethnic group.21
Such boundaries may
separate an ethnic group from another, or ethnic groups within an ethnic
category. Each new boundary-making exercise is accompanied by the pro-
cess of reinterpreting tradition to establish legitimacy for and loyalty to the
“new” community. As this study shows, ethnicity can be invoked to serve as a

8 Introduction
stimulus and a justification for group action to maximize the group’s advan-
tage,as well as to counter a negative image or prevent absorption by a dominant
ethnic community. Membership in the group is determined by acknowledg-
ment of a shared field of interaction and communication. An ethnic group
can identify itself and be identified with an ethnic category, but most of its
interactions will be within an ethnic group or community.
A study of ethnicity usually begins with the old debate between the pri-
mordialists and those called situationalists, circumstantialists, instrumental-
ists, or constructivists. The former stance, often associated with Edward Shils
and Harold Isaacs,22
argues that individuals are born endowed with certain
fixed qualities that they share with a specific group of people. It is these “pri-
mordial” elements that serve to bond the members into an ethnic unity. The
situationalist position, which many social scientists adopt, criticizes the rigid-
ity implied in the primordialist argument and views ethnicity as a fluid con-
cept. It argues that the elements defining the group are constantly undergo-
ing change and rearrangement in response to shifting historical and cultural
circumstances.23
Most scholars writing on ethnicity today take a middle ground. They
agree that an ethnic group is fluid, is continually adjusting to shifting circum-
stances, and is multilayered, but they also recognize the significance of the
primordialist emphasis on some ineffable quality of group identity that defies
any situationalist explanation. It is this perceived “primordial” element that
has evoked such fervent, even fanatic response from individuals throughout
history. There is also a recognition of the agency of ethnic actors who are not
merely shaped by contexts, but who actively seek to construct their identity
from a host of variables. In the process of ethnic re-formation, the group
adjusts the “contents” and “boundary” to enable its members to be ideally
placed to benefit from new circumstances. The “middle” stance therefore
acknowledges the ongoing, active role of the group in redefining the cultural
elements constituting its identity, as well as the desire of a group to believe in
an essential core that distinguishes it from others.
The resulting “traditions” are not “invented” in the Hobsbawm sense
of being manufactured in order to “inculcate certain values and norms of
behaviour.”24
They are instead selected, reorganized, and reinterpreted from
a corpus of old and new symbols, myths, remembered events, etc., in light of
shifting circumstances. It is essential that members believe in an enduring core
that defines the group, despite the constantly shifting elements that make up
that “core.” Individuals seek commonalities that can be summoned to bind
them together as a group for maximum economic, social, or political advan-
tage. The enhancement of a group’s status and prestige in the eyes of others,
which Donald Horowitz describes as “group entitlement,” in turn serves to

Introduction 9
bolster the individual member’s own sense of pride and self-worth.25
The pro-
cess of ethnic formation enables the individual and the group to select from,
in Joanne Nagel’s memorable phrase,“a portfolio of ethnic identities.”26
The increasing globalization in all spheres of life and the resulting human
and capital mobility have all but transformed our traditional perceptions. The
porous borders, transnational activities of individuals, and the merging of
global economic forces have all produced a phenomenon Arjun Appadurai
has described as an “ethnoscape.” By this neologism, he means “the landscape
of persons who make up the shifting world in which we live: tourists, immi-
grants, refugees, exiles, guest-workers, and other moving groups and per-
sons.”27
Those inhabiting this ethnoscape interact with the more conventional
established networks of affiliations to create new possibilities of unities. The
cultural dynamics of deterritorialization thus enable individuals and groups
to imagine themselves from a wider set of possibilities than ever before.28
For
a historian working in the precolonial period in Southeast Asia, the situation
described by Appadurai is familiar. The Straits of Melaka served as a channel
of goods, ideas, and news from the outside world, thus igniting the imagina-
tions of individuals and groups living along its shores to new possibilities of
ethnic and other affiliations.
A common origin and a shared ancestor form meaningful ethnic markers
that legitimize the group and reaffirm its sacred links to the past.Acknowledg-
ing the spiritual potency of the idea of origins, John Armstrong and Anthony
Smith have both used the concept of a mythomoteur, defined as “the con-
stitutive myth of the ethnic polity,” which is based on the belief in a mythic
primordial past.29
Adherence to a mythomoteur, they argue, provides a power-
ful sense of a “common fate” among its members, thus defining them from
others.30
Although Smith distinguishes between a dynastic and a communal
mythomoteur, he nevertheless questions whether one should insist on such a
division. He asks,“Is it true that upper-class culture was generally of an utterly
different character from the many cultures of the peasantry,and that therefore
there could be no sense of shared identity between the classes in any area or
polity?”31
In the case of the Malayu in the precolonial period, sumptuary laws
may have been created to recognize difference but customary law and shared
cultural ideas clearly emphasized the communal purpose, thereby strengthen-
ing group unity. Precolonial Southeast Asian societies were characterized by
strong bonds between chiefs/rulers and their subjects, who were often kinfolk.
When a larger unity was required, the dynastic mythomoteur served to estab-
lish the social and political bonding for the newly extended boundaries of the
group.32
In the theories of ethnicity, the elite groups play a leading role in the
creation of a group’s cultural ideology. But the process is not all one-sided,

10 Introduction
and ordinary people are equally important in reinforcing these boundaries
by emphasizing differences, no matter how slight. How men and women wear
their hair or tie their sarongs, what types of food they eat, what language
they speak or even how they speak it, can all be important markers of ethnic
identity. For the common folk these are not “soft” boundaries33
but meaning-
ful ones that are reinforced through daily activities. By making these mun-
dane choices, people themselves strengthen the boundaries established on a
more reified level by their spiritual and temporal leaders. Tangible and easily
adopted, the boundaries erected by common people can be readily breached
to enable individuals and groups to strategically deploy one or more identities
in different circumstances to maximize advantage. The role of the elite and
the ordinary people in the process of ethnic formation thus allows for maxi-
mum flexibility in periods of rapid change. This is the situation that prevailed
among many of the communities living in the vicinity of the Straits of Melaka
in the precolonial period and explains the ease with which individuals and
groups moved from one ethnic community to another.
Language is one of the most cited elements in defining a group, and its
strength as a unifying force comes from its flexibility. This is clearly dem-
onstrated in an episode involving the main protagonist Hang Tuah and the
maidens from Indrapura in the popular Malayu tale the Hikayat Hang Tuah.34
When the maidens apologize that their use of the Malayu language lacks the
purity of that of the Melakans, Hang Tuah reassures them that the language of
Melaka itself is“mixed”(kacukan).35
During Aceh’s dominance as the center of
the Malay world in the seventeenth century, its form of the Malayu language
became the prestigious version even though it created difficulties in compre-
hension in parts of the wider Malayu-speaking world. A Muslim scholar from
Banjarmasin in the seventeenth century wrote a companion piece to a Malay
Islamic treatise from Aceh because he claimed that the latter contained too
many “Acehnisms.”36
These examples suggest that the Malayu language was
spoken in different ways in the seventeenth century. Even Melaka, regarded as
the center of Malayu culture in the fifteenth century, acknowledged the valid-
ity of the Malayu language spoken in Indrapura. Yet the dialectal differences
in no way diminished the importance of the Malayu language as an important
boundary marker in delineating a Malayu world that incorporated a diverse
population.
While the variation in the manner in which the Malayu language was
spoken and written was used to define specific ethnic communities, the
Malayu language was the boundary for the ethnic category. The variations of
the Malayu language suited the multiplicity of ethnic groups that used that
language as a basis of identity. The late nineteenth century, however, saw a
change in the attitude toward language use. In order to learn more about the

Introduction 11
area and to facilitate their control, European colonial powers commissioned
the recopying of local histories, law codes, belles lettres, and other cultural
works. The coincidence of a particular language that was used both for writ-
ten documents and for ordinary speech by the majority community often
became the colonial basis for ethnic identity. In time such ethnic boundar-
ies were self-fulfilling, with bilingual or even trilingual speakers claiming the
most advantageous language and ethnic group with which to be identified to
the colonial powers.
It may be scientifically indefensible to argue for distinctive ethnicities
because of the continuing intermingling and exchange of biological and cul-
tural elements among groups.37
Nevertheless, individuals and communities
have displayed a persistent desire to underscore difference and to define and
redefine themselves in order to promote their individual or group interests.
History is rife with examples of ethnonations and nation-states successfully
appealing to some sense of communal solidarity to defend a bounded entity.
There is a conviction that their “venerable traditions,” and hence their link
to the ancestral past, remain unchanged. Activity based on ethnic conscious-
ness, notwithstanding ethnicity’s variability and ongoing reinterpretations,
is an undeniable historical reality. The corpus of traditions allows variant
interpretations and a degree of ambiguity that facilitates the incorporation of
desired individuals or communities. Even the concept of hybridity, seemingly
counterintuitive to ideas of “origins,”can be harnessed to strengthen a group’s
identity. It is precisely this hybrid quality that enabled individuals to claim
Malayu ethnicity no matter how tenuous their claim to shared traditions.38
The ambiguity and multiple meanings that groups could extract from Malayu
origins and traditions made Malayu an extensive, expansive, and imperializ-
ing ethnicity.
There is a large menu of ethnic theories with a bewildering array of
approaches.Althoughsomelamentthelackof precisionandconsensusregard-
ing a definition of ethnicity, such “unsatisfactory” results are to be expected.
Human interactions are by nature unpredictable and dynamic, defying any
clear and definitive characterization. Yet it is possible to use ethnicity as an
important analytic tool to explain group relations in Southeast Asian history.
Ethnic Communities as Objects of Analysis
According to many oral traditions, the early communities in Southeast Asia
began as small, kin-based societies with clan elders as their natural leaders.
Such groups were generally known by a name they called themselves (end-
onym) and one or more names given to them by outsiders (exonym).The most
common form of self-identification was the local word for “human being”

12 Introduction
or “people,” in contradistinction presumably to animals, ethereal beings,
the forests, and all others that inhabited their universe. To distinguish them-
selves from other human communities, a group often added another form of
identification based on location, such as “people of the upriver,” “people of
the hills,” “people of the swamplands,” etc. These were appropriate and ade-
quate markers of ethnicity among economically interdependent groups living
within a limited geographic space.
In time the group’s numbers generally increased, the search for addi-
tional resources became necessary, and contact with the outside world grew
more frequent. The impingement of groups became common, and the need
for some type of mutually agreeable economic and political arrangement
encouraged the formation of a more active and intrusive form of governance.
The process is captured in local traditions, where a pre-existing community
seeks an arbiter in its affairs whose judgment would be accepted by the people.
This condition is met in the dynastic myth (Smith’s dynastic “mythomoteur”),
which associates the progenitor of the royal family with supernatural origins.
Around this sacred figure the various kinship communities coalesce to form
a single political entity. With the proliferation and expansion of such polities,
the authority of these sacred figures/rulers overlapped at the frontiers. These
frontiers thus formed the dynamic region of political arrangements termed
“mandala polities” by Wolters and “galactic polities” by Tambiah.39
According to this roughly similar conception, the mandala/galactic pol-
ity is the center of its universe, with satellite communities located around
it. A graphic image of the exercise of power in such polities is that of an
upturned lamp, whose light is intense in the center but gradually fades away
at the edges.40
What the image conveys is a situation of constant realignment
of groups, in which the overlapping edges of authority become the site for
contestation. The periphery retains a position of strength because it is able to
shift allegiances or maintain multiple allegiances in promoting its best inter-
ests. At these dynamic edges individuals and groups are able to claim multiple
ethnic identities, or to move in and out of ethnicities as the circumstances
warrant. The periphery, then, determines whether the “exemplary center”
survives or is replaced by another. For this reason, the center takes great care
to maintain strong bonds with influential families or individuals in the cru-
cial borderlands.
The common practice of bilateral kinship, which traces lineage through
both males and females, facilitated alliances among families in Southeast Asia.
There was no particular advantage in having male children; female children
were as valuable because they, too, could be strategically married to advance
the family’s economic, social, and political fortunes. Through such marriages,
certain powerful families had networks extending to more than one polity,

Introduction 13
with some family members at the periphery claiming multiple allegiances.
Bilateral kinship inheritance patterns made it imperative for individuals to
retain rights both in their own families and in those of their in-laws. Some-
times this involved belonging to two separate ethnic groups, as in the case of
the Batak, because land ownership and the rituals associated with its transfer
could only be effected by ethnic Batak.41
In some cases, the Batak adopted an
additional Malayu ethnic identity because of the advantage of being interme-
diaries between the Malayu coast and the Batak highlands.42
In short, precolonial Southeast Asia was not subject to international con-
ventions confining individuals within a fixed space and imposing on them a
specific legal identity. Ethnic identity was a fluid concept, and the decision to
adopt one or more ethnicities was the privilege of the individual. The man-
dala/galactic polity encouraged rather than opposed such practices because
people were a source of wealth. The relative paucity of people in Southeast
Asia until the twentieth century made rulers particularly anxious to retain
their subjects and to attract others. Indigenous documents exhort rulers to
perform good deeds to attract followers and thereby bring prosperity to the
land. In this regard, Southeast Asian groups were more concerned with the
maintenance of the porosity rather than the impermeability of their ethnic
boundaries.
In this study I have been guided by Joel Kahn’s astute observation that
one should not focus on the“principles”that unite a culture, but on the social
process operating under specific historical circumstances that produced that
culture.43
Implied in this statement is the futility of depicting any ethnic iden-
tity as fixed since the construction of ethnicity is an ongoing sociohistorical
process. For this reason I have focused on the process of ethnic formation
to highlight the contingent nature of ethnic identity and the fluidity of its
manifestation.
Process of Ethnic Formation in the Straits of Melaka
Each chapter relies on a historical narrative based on trade that helps explain
why, when, and where various ethnic groups and categories were formed or
re-formed in the distant and more recent past. The groups that have been
chosen as the basis of this study are those that are regarded as the “ancient”
inhabitants of the lands and seas bordering the Straits of Melaka. Although
Indians, Chinese, and the Bugis have played important roles in the history
of the straits, they are relatively recent settlers and are associated with home
areas outside the straits. Inclusion of these groups would also have required
attention to another major issue, that of diaspora, and thus complicate an
already complex subject.

14 Introduction
The Malayu were one of the earliest and most influential in the straits,
and their prominent role in international trade spurred the ethnicization of
other groups. As far as I can determine, as an ethnonym, “Malayu” referred
first to the communities living in southeast Sumatra and later came to include
those settled along both coasts and in the central and northern interior areas
of the island. From the fifteenth century the ethnonym was also applied to
those living on the Malay Peninsula who were descendants of Malayu immi-
grants from Sumatra. The name itself has been used at various times to refer
to a language, a culture, a regional group, a polity, and a local community. It
is not surprising, therefore, that it has spawned a wide variety of interpreta-
tions concerning its meaning and significance.44
Most of these discussions,
however, overlook an emerging culture in the northern portion of the Straits
of Melaka that formed the antecedents of Malayu culture. The settlements in
northern Sumatra and in the Isthmus of Kra and the Malay Peninsula were
part of an extensive network of communities, which I have termed the“Sea of
Malayu.”Chapter 1 explores this exchange network that extended from south-
ern India and Sri Lanka to northern Sumatra, the Isthmus of Kra, and the
northern Malay Peninsula, across to the Gulf of Siam and the Lower Mekong
of southern Cambodia, to the Cham areas of southern and central Vietnam.45
The long and profitable interaction within this common “sea” produced a
shared cultural idiom that helped shape Malayu identity.
Chapter 2 is a more specific examination of the Malayu culture that
developed in the early southeast Sumatran polities of Sriwijaya and Malayu
between the seventh and fourteenth centuries.While inscriptions and external
sources are limited, there is sufficient linguistic and archaeological evidence
to form the basis for a tentative reconstruction of the sociopolitical organiza-
tion and the nature of the economy of these polities, especially of Sriwijaya.
Certain features of the society can be detected, including the role of family in
government, a reliance on sea and forest peoples in assuring the collection of
products and protection of routes for international trade, the maritime and
riverine environment, the sacral quality of kingship, and the use of oaths as an
important political and economic tool.The term“Malayu”thus came to desig-
nate those communities that had incorporated many of the features identified
initially with Sriwijaya and its successor, the polity of Malayu. In Sumatra,
the expansion of the Malayu as an ethnic community and an economic force
served as a catalyst for the ethnicization of other groups.
The historical circumstances that gave rise to a separate Minangkabau
ethnic identity from the Malayu is the subject of chapter 3. In 1365 the Java-
nese court poem, the Desawarnana, included the Minangkabau highlands
and most of the areas on Sumatra as part of the bhumi Malayu, the “Malayu
world.” Inscriptions, artistic remains, and other archaeological finds indicate

Introduction 15
that there was a polity in the highlands whose royal settlement was called
Malayupura, “the Malayu City.” But sometime between the late fourteenth
and early sixteenth century, the local identity that had been subsumed by the
Malayu began to assert itself. Early sixteenth-century Portuguese documents
mention the Minangkabau by name and of their kings ruling in the highlands.
Only with the arrival of the VOC in the seventeenth century, however, are
there sufficient contemporary reports to trace the ethnicization process of the
Minangkabau. The economic opportunities provided by the removal of Aceh-
nese control and the increase in trade through the straits provided the impetus
for the formation of a separate Minangkabau ethnic identity. Through a con-
vergence of local beliefs in the supernatural powers of the Pagaruyung ruler
and theVOC decision to support his claims,a new Minangkabau ethnicity was
created that proved effective in rallying the people to act as one for economic
and political advantage.
The Malayu were associated with Sumatra until the rise of Melaka on
the Malay Peninsula in the fifteenth century. Melaka’s stunning success as an
international entrepot and center of Islamic scholarship raised the regional
status of the Malayu considerably. Melaka became synonymous with Malayu
and began to be regarded as the standard-bearer of Malayu culture. With the
capture of Melaka by the Portuguese in 1511, two competitors emerged to
claim the mantle of Melaka’s successor in the Malayu world: Johor and Aceh.
As shown in chapter 4,Aceh prevailed because of its strong economic and cul-
tural links to the great Muslim kingdoms in the Middle East and India. Dur-
ing the sixteenth and for much of the seventeenth century, Aceh established
new standards of Malayness based on Islam and on many court practices that
mirrored the foremost Muslim kingdoms at the time. As the leading Malayu
polity, Aceh’s new standards were applied along both Sumatran coasts, in the
west coast of the Malay Peninsula, and in Pahang on the east coast.
When Johor eventually emerged in the late seventeenth century to replace
Acehasthecenterof theMalayuworld,itadoptedthestrongerIslamicbehavior
instituted by Aceh but reverted to the court customs of the Melaka period. By
the late eighteenth century,Aceh’s rejection as the major Malayu center forced
it to emphasize a new ethnic identity centered on the interior and agriculture,
rather than on the coast and international trade. Unlike the coastal regions
of Aceh, where the Malayu language was dominant, the interior areas were
principally Acehnese-speaking. The new Acehnese identity was reinforced by
literary works written not in the Malayu but the Acehnese language. The new
Acehnese identity proved so successful that by the nineteenth century few
remembered Aceh as once being the leading center of the Malayu world.
Chapter 5 narrates the story of the ethnicization of the Batak. As with
the Minangkabau and the Acehnese, the Batak were formerly a part of the

16 Introduction
fourteenth-century Javanese depiction of the bhumi Malayu. Contrary to
widely held opinion, the Batak were never isolated from the outside world
because they were the principal suppliers of camphor and benzoin. These two
resins grow abundantly in north Sumatra in the Batak country surround-
ing Lake Toba and were in great demand in the international marketplace.
To meet this demand, the interior Batak communities organized themselves
for the collecting and transporting of valuable resins to the Malayu entrepots
on both Sumatran coasts. Until the destruction of Sriwijaya by the Cholas
in 1025, these products were brought to this leading entrepot on the south-
eastern coast. Subsequently, the Batak brought the resins to Kota Cina and
other polities on the northeastern coast of Sumatra, as well as to Barus, an
ancient entrepot located on the northwest coast. As a result of this long trade
relationship, there was a flow of ideas between the Malayu and the Batak. This
is clearly evident in the monuments and statues found at the archaeologi-
cal site of Padang Lawas, at the frontier of the Batak and the Malayu (later
Minangkabau) lands.
Beginning in the fifteenth century, the introduction of pepper cultiva-
tion in Sumatra provided yet another opportunity for the Batak to become
involved in international trade. The intensive labor required for the cultiva-
tion of pepper left little time for rice cultivation, and so rice became a valued
commodity in the pepper-producing areas of Sumatra. Many of the Batak
were thus encouraged to move out of their home areas around Lake Toba
to seek lands for the planting of rice. The spread of the Batak into different
areas led to separate developments and modifications of Batak cultural ideas
and the formation of various Batak subethnic communities known today as
Karo, Simalungun, Pakpak-Dairi, Toba, Angkola, and Mandailing. But in ear-
lier times the term “Batak” would have been used as an ethnic identity for
those who traced their origins to the area of Lake Toba and adhered to the
indigenous religion. The ancient belief system provided the myths and sym-
bols that defined and strengthened ideas of Batakness. Its priests and religious
teachers with their extensive network of marketplaces, worship centers, and
students forged a common Batak identity that proved useful in the competi-
tive economic environment of the Straits of Melaka.
The final two chapters discuss two ethnic categories, marginalized today
but once invaluable to the Malayu groups both in Sumatra and the Malay
Peninsula. Chapter 6 discusses the communities that form the ethnic category
known by the exonym “Orang Laut” (though the government in Malaysia has
arbitrarily submerged this identity under that of “Orang Asli”), and chapter
7 focuses on the Orang Asli (known as “Suku Terasing” in Indonesia). Their
current emasculated political and economic position has colored interpreta-
tions of their important role in Malayu polities in the past as collectors of

Introduction 17
sea and forest products and as guardians of the sea and jungle routes. The
Orang Laut’s knowledge of the seas and their navigational skills made them
an indispensable part of the Malayu ruler’s naval forces. Malayu traditions
themselves acknowledge the debt owed to the Orang Asli and the Orang Laut,
and even highlight the significant marital arrangements contracted between
these two groups and the Malayu rulers to strengthen their mutually benefi-
cial relationship.
The distinct, complementary economic role of the Orang Laut and
the Orang Asli/Suku Terasing to that of the Malayu was a major reason for
a respected partnership in earlier times. Their ethnicization was therefore a
deliberate effort to preserve a way of life that guaranteed their advantage and
eventual survival from the intrusions of their numerically dominant Malayu
neighbors. By the beginning of the nineteenth century, with the shift in eco-
nomic wealth away from sea and forest products, the Orang Asli/Suku Teras-
ing and the Orang Laut lost their value to the Malayu. In a relatively short
space of time, an exonym once bestowed in respect and proudly ethnicized by
its members became a stigma. The result was a predictable rise of mutual sus-
picion and of violence committed mainly by the Malayu against the sea and
forest peoples. Through the revitalization and resymbolizing of the “Orang
Asli”name, the group has been able to promote its political interests in Malay-
sia and acquire greater recognition from the outside world. No such progress,
however, has been made in the position of the Suku Terasing in Sumatra.
Ethnic formation is an ongoing process, with trade being the principal
stimulus for change in Southeast Asia in earlier centuries. With this under-
standing of the nature of ethnicity and of the process of ethnic formation, it
is necessary to rethink views of “ethnic” politics in history. Ethnicity can be
a means of explaining difference, a basis for group action, and a mechanism
contributing to the successful functioning of the mandala/galactic polities in
precolonial Southeast Asia. Fortunes of groups change, and the stories of the
Orang Asli/Suku Terasing and the Orang Laut are useful reminders that some
groups exercise greater agency than others in the formulation of ethnic iden-
tities. By acknowledging both ethnicity’s explanatory value and its dynamic
characteristics, historians should be able to examine this concept with greater
precision and offer a more nuanced view of its role in Southeast Asian pasts.

18
Chapter 1
Malayu Antecedents
I
n many history books the story of the Malayu1
begins with
the fifteenth-century kingdom of Melaka and occasionally
with the seventh- to eleventh-century kingdom of Sriwi-
jaya. The first can be justified in terms of the history of the Malayu on the
Malay Peninsula, while the second is based on growing evidence of the early
development of Malayu culture in southeast Sumatra. But the story can be
pushed back even further as a result of the latest linguistic and archaeological
research. In reaching back into the past, the outlines of ethnic groups as we
know them today become blurred and indistinct. For this reason it would be
presumptuous to assume that there was a clear Malayu ethnicity in prehistoric
times that has continued basically unchanged into the present.2
It is not my
intention here to “establish” the antiquity of the Malayu people, but simply to
try to understand how such a group could have emerged from an ancient past
where ethnicity was an indeterminate and perhaps even irrelevant category.
In the seventh century CE there was a group of people speaking the same
language living in a place known as Malayu, who could have called themselves
at appropriate times “people of the land Malayu.” Through comparative lin-
guistics it is even possible to trace the Malayu language to a proto-language
that developed in west Borneo. But the ability of the linguist, the archaeolo-
gist, and the historian to see connections between groups of people in no
way implies that such links were perceptible to the people themselves. A more
reasonable assumption is that long social and economic intercourse among
communities helped create a form of identity—not necessarily expressed
in ethnic terms—based on a common language of communication and of
shared experiences. In this chapter I argue that the ancient ongoing inter-
course among particular communities straddling the Indian subcontinent,
the Southeast Asian isthmus and the northern Malay Peninsula, and the
shores of the South China Sea created a “voyaging corridor”3
and a pattern

Malayu Antecedents 19
of interaction that became the basis for a common identity. It is impossible
to know whether such an identity was actually formalized or even referred to
by name. Although early travelers and modern scholars have given specific
names to such complexes for convenience or heuristic purposes, one should
not assume that the participants of the complexes themselves perceived an
overarching identity. As discussed in the introduction, specific circumstances
would have determined choices of identity. My decision to refer to this voy-
aging corridor as the “Sea of Malayu” is based on the nature of the relation-
ships and the prominence played by groups who later became identified as
“Malayu.” Linked by sailing ships, a “sea” of communities came to be charac-
terized by the most dominant of the participants. In the first half of the first
millennium CE, communities in the Lower Mekong, which the Chinese called
“Funan,”4
were most likely the dominant partner. Funan’s language and cus-
toms may have then become the norm for the lesser partners in the common
sea. Other traders, like those now called“Arabs,”may have dealt with different
communities and hence called the network by the name of that group. In the
tenth century, for example, Arab geographers referred to the network (which
extended to east Africa) as the “Cham Sea.”5
The evidence points to the Malayu as the major group within this sea
extending from India to Vietnam and the most likely successor to Funan. By
the end of the seventh century, Sriwijaya had arisen in Palembang in south-
east Sumatra as a major polity. It was inhabited by people who wrote stone
inscriptions in the Malayu language but who are not mentioned specifically as
“Malayu” people. When Sriwijaya was succeeded by Malayu, a polity located
in present-day Jambi, it would therefore have been likely that local inhabit-
ants would have been called orang Malayu, or “people of Malayu.” The stone
inscriptions supply convincing evidence that Malayu was one of the major
languages of both Sriwijaya and Malayu. A way of life developed by the orang
Malayu would then have become the basis for the association of the group with
certain cultural features. It is important to reiterate, however, that the name
“Malayu” and what it meant in southeast Sumatra would have undergone
a number of permutations over the centuries.6
What the “ancestors” of the
Malayu may have called themselves is a mystery, and even the name “Malayu”
has never been convincingly explained. In other words, it is only possible here
to speak of the antecedents of those who later came to identify themselves as
“Malayu”in southeast Sumatra, without claiming a direct link between the two.
Austronesian Speakers and the Nusantao Communities
The most widely accepted reconstruction of ancestral Malayu origins is the
Bellwood-Blust synthesis. It dates the initial settlement of proto-Austronesian

20 Chapter 1
speakers in Taiwan between 4000 and 3000 BCE. Proto-Austronesian either
developed in Taiwan or in a subsequent move to the northern Philippines
c. 3000 BCE. With the dispersal of these peoples throughout the rest of the
Philippines, proto–Malayo-Polynesian emerged about 2500 BCE. By about
2000 BCE the proto–Malayo-Polynesian language began to break up as migra-
tion resumed to the southern Philippines,Sulawesi,Maluku,and Borneo,with
settlement in western Borneo dated between 1500 and 500 BCE.7
In the proto-Austronesian family tree reconstructed by linguists, a sub-
group called Malayo-Chamic forms part of the Western Malayo–Polynesian
languages. Working from this basis, Graham Thurgood lists two branches of
Malayo-Chamic: one is Malayic languages, from which derived proto-Malayu
and the various Malayu dialects, including Minangkabau; and the other is
proto-Chamic, which gave rise to coastal Cham and Acehnese.8
Linguists
believe that the homeland of Malayo-Chamic was in western Borneo and that
several hundred years BCE there was a move outward through the Tambelan
and Riau islands to the Malay Peninsula. From the Malay Peninsula, one group
crossed over to southeast Sumatra and became the ancestors of the Malayu
speakers, while another group proceeded to the coasts of Vietnam and became
the ancestors of the Cham language speakers.9
Sometime before 1000 CE
a northerly Cham group left centralVietnam and became the Acehnese speak-
ers of northern Sumatra.10
From very early on, therefore, the Acehnese in
northern Sumatra formed a different branch of the Malayo-Chamic subgroup
from the Malayic speakers in the southern part of Sumatra. Resulting from a
back-migration to Sumatra, the Acehnese language contains clear borrow-
ings from interaction with non-Austronesian speakers.11
Contact between the
Acehnese and the Chams may have been maintained through the centuries.12
Early Malayo-Polynesian communities developed in a subtropical coastal
and riverine environment where the economy was based on cereal, tubers,
and domesticated animals. In the process of adapting to specific ecological
niches, their descendants began to embrace differing lifestyles. Some foraged
the rain forests and the seas for products in great demand in the interna-
tional marketplace; others engaged in various forms of irrigated and rain-fed
cultivation of cereals, fruits, and tubers; while still others specialized in the
exploitation of the sago palm.13
Archaeological records for island Southeast
Asia indicate that during these migrations the best coastal sites were occupied
first. Only when or if there were no suitable coasts to settle did migrants move
into the interior.A feature Peter Bellwood terms“founder rank enhancement”
played an important part in this process. Because founders of new settlements
and their descendants were elevated to almost godlike status, there was strong
motivation for members of a junior branch to seek an empty area and estab-
lish a new senior line with priority over resources.14

Malayu Antecedents 21
Less well known is a theory advanced by Wilhelm Solheim over a num-
ber of years. This ambitious conception incorporates the story of the Austro-
nesian speakers into a wider network of “Nusantao” communities. Instead
of positing a monodirectional Austronesian movement, Solheim proposes a
multidirectional flow from the different“lobes”that formed the Nusantao net-
work. He believes that the Nusantao “homeland” (calculated simply in terms
of the earliest dates known for the existence of a group) is in the Early Central
Lobe in eastern coastal Vietnam and dates it to c. 8000 BCE, much earlier than
Bellwood’s reconstruction for the ancestors of the Austronesian speakers. He
suggests that in c. 5000 BCE the people in the Late Central Lobe involved in
this network began moving by water and developed a trade communication
network. It was these maritime trading people who developed Austronesian
as a lingua franca from pre- and proto-Austronesian to facilitate communica-
tion among the communities forming this network.As the Nusantao network
expanded out of Taiwan, it was Malayo-Polynesian languages rather than
Austronesian that developed with it. Solheim emphasizes that the expansion
of Malayo-Polynesian was not the result of migrations but of the interaction
occurring within the network. He also emphasizes the important role of mar-
itime people in the dispersal of the Nusantao community.15
In discussing these two major theories regarding the antecedents of the
Malayu, it is important to stress that “Austronesian” and “Nusantao” are not
synonymous. The former is linguistic, the latter cultural, and neither refers to
a genetic group. Solheim, however, uses a gene marker identified among the
Southeast Asians but not found in China as an argument for rejecting the view
that the origin of the Austronesian speakers is in southern China. He believes
that ancestors of the Southeast Asians had been living in the region since 5500
BP, after the retreat of flood waters following the end of the Ice Age some eight
thousand years ago. Solheim also disagrees with linguists regarding the route
taken by the Austronesian speakers from southern China to Taiwan, the Phil-
ippines, and then down to Southeast Asia and out into the Pacific. Instead,
he suggests that a trade language in the form of Austronesian developed in
coastal south China, northern Vietnam, Taiwan, and northeast Luzon, and
evolved through ongoing contact among the Nusantao communities. The
notion of interacting communities moving in multiple directions allows for
local variations and adaptations to specific geographic conditions.16
Although Solheim’s dates are generally regarded as being too early, the
appeal of his model is the idea that the spread of a culture, including a lingua
franca, evolved as a by-product of the trade and communications network
of a large number of different communities in a widely dispersed area. In
the historical period the Malayu language and culture were developed and
sustained in very much the same fashion. Linguistic reconstruction of the

22 Chapter 1
migration of the Austronesian speakers does not emphasize the trade aspect,
but for Solheim trade was the major feature and basis for the creation of this
“Nusantao maritime trading and communication network.” While Bellwood
explained the spread of the Austronesian migration by the phenomenon of
rank enhancement, Solheim points to the long-standing existence of many
maritime populations who became part of this extensive trade network.
Nusantao culture was not associated with a single ethnic group, but with a
style of life and a trade language comprehensible throughout an interactive
region. This particular aspect of Solheim’s model is helpful in understanding
the formation of an early network of communities I call the “Sea of Malayu.”
The Sea of Malayu
The first reference to a “Sea of Malayu” is from an Arabic document dated
c.1000,whichnotedthattravelers“reachingtheSeaof Malayu,wereapproaching
the area of China.”17
Eredia, writing in 1613 from Melaka, also uses the phrase
“Sea of Malayu,” but he identifies it with that “land-enclosed sea between the
mainland of Ujontana [Malay Peninsula] and the Golden Chersonese [Suma-
tra].”18
By privileging land over water, Eredia believed that the“Sea of Malayu”
referred simply to the Straits of Melaka. For the Malayu, who were shaped by
their orientation to the sea and the riverine environment in which they lived,
stretches of land were viewed as barriers that fortunately could be breached
through short land passages.19
The people were named after a particular river
or stretch of river, stream, or coast. In this maritime world, rivers and seas
formed unities, while land formed the link between bodies of water. Based on
this particular way of viewing waterways and identifying people, the Malayu
would have conceived of their sea as a far larger unity than that proposed by
Eredia. Although the Arabic document is not specific, the general reference to
a Sea of Malayu approaching the area of China is an accurate description of
the extensive network viewed as one sea stretching from India to Vietnam.
The sea itself I have called “Malayu” after the people most prominently
associated with this particular body of water. But the “Sea of Malayu” that I
am proposing in this study is a community of settlements conjoined through
extensive and intensive economic and cultural interactions. From the late
seventh century the people of Malayu would have played a role, even a lead-
ing role, in such a network. The evidence suggests that there existed a single
continuous “sea” linking southern India and Sri Lanka to the Bay of Bengal,
Sumatra, the Straits of Melaka, the Malay Peninsula, the Gulf of Siam, the
South China Sea, the Lower Mekong, and central Vietnam. The pivotal point
in the network of Sea of Malayu communities was the Straits of Melaka. In
the days of sailing ships, the straits were conveniently located for traders

24 Chapter 1
at the“beginning”and“end”of the seasonal monsoon winds. Between Novem-
ber and February the northeast monsoon winds brought ships from East
Asia, and between June and August traders from India, the Middle East, and
Europe rode the southwest monsoon winds to the straits and to points further
east. In between these two dominant patterns, the winds moved in a clock-
wise direction, enabling traders from the various parts of Southeast Asia to
reach the major entrepots located in or near the Straits of Melaka. Because
the straits provided protection from the force of the monsoon winds, ports
on both shores of the straits have historically competed for the status as the
leading entrepot in the region.
Evidence for the vitality of these early exchanges is provided by recent
research on the Indo-Pacific bead trade, which has demonstrated that South-
east Asia and India were already important trade partners prior to the Com-
mon Era, often regarded as the beginning of Indianization in Southeast Asia.
High-quality Indian carnelian and agate beads dated to the last centuries BCE
have been found in central Thailand in sites such as Ban Don Tha Phet, in
peninsular Thailand at Khao Sam Kaeo, in coastal Vietnamese sites of the Sa
Huynh culture,and in the Tabon caves on Palawan in the Philippines.Bérénice
Bellina attributes beads of high quality workmanship and distinctive styles to
Indian artisans fulfilling orders from Southeast Asian elite.By contrast, beads
dating from the early centuries of the first millennium CE are of much lesser
quality and have been traced to Southeast Asian production centers. These
were probably intended for the lower levels of society or for trade with inte-
rior groups.20
The sophistication,wealth,and self-confidence that SoutheastAsian elites
shared is apparent in discoveries of similar ornaments and prestige goods,
such as Dong Son drums, objects found at Sa Huynh, and bronze knobbed
ware.21
These findings suggest a depth of a common culture and a trade net-
work that persisted into the second millennium CE. Archaeologists date the
Ban Don Tha Phet site to the end of the third or the second century BCE. In
addition to beads, a significant find was bicephalous ear ornaments made of
nephrite (a variety of jade). Such jade ornaments are associated with the Sa
Huynh sites in central Vietnam, a cultural area where the Cham civilization
later emerged.22
It therefore appears likely that in the first millennium BCE
communities between central Vietnam and at the head of the Gulf of Siam
formed part of an exchange network extending from India to China through
the transpeninsular routes. At this site were found bronze ritual vessels and a
carved carnelian lion, both of which have symbolic functions in Indian Bud-
dhism, as well as glass beads and semiprecious stone beads. These finds indi-
cate that there was early Buddhist activity in Thailand and perhaps elsewhere
in Southeast Asia before the Common Era.23

Malayu Antecedents 25
Kuala Selinsing in Perak in the northern Malay Peninsula was another
significant prehistoric site. It is thought to have been occupied from at least
the second century BCE or even earlier, but its contact with India may have
come later. Despite the long occupation of this site, Kuala Selinsing was not a
major port but, as Leong Sau Heng puts it, “a feeder point,” one of a number
of “small local supply centres serving the entrepots and important regional
collecting centres.” The recovery of glass and stone beads, some half-finished,
led Leong to conclude that there was a local bead-making industry, an observa-
tion substantiated by Peter Francis through glass analysis. Evidence of Indian
influence is limited to a small carnelian seal inscribed with a south Indian
script and a gold ring with an Indianized motif.24
Other early sites were incorporated into the international and regional
trade network of the Sea of Malayu, notably Khao Sam Keo (beginning of
the fourth century CE) and Khuan Luk Pat (“Hill of Beads,” c. third to sixth
century–seventh century CE), located in Khlong Thom in Krabi province, the
terminus of a transpeninsular route.The latter was replaced by Kuala Selinsing
as the main producer of beads perhaps from the sixth to the tenth centuries
CE.25
In a recent study, David Bulbeck has also emphasized the importance
of the Andaman Islands in the network of seafaring populations that helped
open the sea lanes for trade between India and Southeast Asia. He notes that
Andamese traditional decorations focus on “Sa Huynh Kalanay” geometric
decorations that show strong similarities with the pottery designs at Kuala
Selinsing.26
While Chinese sources describe Indianizing kingdoms in Southeast Asia
in the early centuries CE, archaeological studies have yet to yield evidence
for such settlements predating the fifth century. The absence of archaeologi-
cal records for pre–fifth-century settlements accords with Monica Smith’s
contention that substantive Indian contacts with Southeast Asia only date
from the rise of the Gupta dynasty in India in the fourth century CE.27
One
is therefore faced with a curious situation in which Chinese records describe
Indianizing settlements in the region, while Indian documents merely men-
tion names without any geographic or historical information. Furthermore,
archaeological evidence is limited to Chinese ceramics, which can only offer
limited insights into the local communities.
Early Indian works provide only a generalized reference to Southeast
Asia. In the Buddhist Jataka tales originating before the Common Era, the
term Suvannabhumi (“the Gold Country”) is an epithet for the lands east of
the Bay of Bengal, meaning Southeast Asia. The epic Ramayana, whose com-
position would have begun before the Common Era, mentions Suvarnadvipa
(“the Gold Islands”) to refer to Southeast Asia and later specifically to Suma-
tra. Of later provenance is the Tamil narrative poem Pattinapalai, composed

26 Chapter 1
in the early centuries CE but not later than the beginning of the third century.
It describes the trade between southern India and Kalagam, usually identified
with Kedah. The Mahaniddesa, believed to contain information from the sec-
ond and third centuries, mentions Yavadipa, or the island of Java.28
At the beginning of the Common Era, southern India became a major
focus of Indian–Southeast Asian trade. Tamil culture was flourishing, Brah-
manic Hinduism was displacing Buddhism, new agricultural lands were
opened, and urban settlements were increasing. All of these developments
provided the basis for a lively and lucrative exchange with Southeast Asia in
the second and third centuries from the southern Indian ports of Arikamedu,
Kayal, and Kamara. Both Mantai in Sri Lanka and Arikamedu in southern
India were the most likely sources of Roman and Persian artifacts from the
subcontinent that moved across the Bay of Bengal, across the transisthmian/
transpeninsular route, to Oc Eo in the Lower Mekong.29
A Greek text titled The Periplus of the Erythrean Sea is a compilation of
knowledge available in the second half of the first century CE. It refers to ships
sailing from the southeast coast of India to “Chryse,” which is believed to be
Southeast Asia or perhaps even the Malay Peninsula.30
Also mentioned are
land routes of the silk trade, from which Wolters inferred that seaborne com-
merce between India and Southeast Asia at the time was very limited. Chinese
sources indicate that sometime between the third and fifth centuries CE the
sea route between India and China came to be used more frequently. In 413
the Chinese pilgrim Faxian returned all the way from Sri Lanka to China by
sea, and a few years later Gunavarman, a Kashmiri prince, went to China via
the same route. Under the Song dynasty in China (960–1279), Chinese over-
seas trading activity grew rapidly, particularly to Southeast Asia. The increas-
ing popularity of this all-sea route had important repercussions for some of
the early polities along the Straits of Melaka. Chinese sources mention the
existence of a western Indonesian polity called Ko-ying or Chia-ying in the
first half of the third century CE. Their source for this information came from
an area in the southern Mekong known to the Chinese as “Funan.”31
Funan, perhaps a Chinese rendering of the local term bnam/phnom
(mountain), consisted of a number of communities with a shared culture,
whose links to one another varied in nature and intensity at different times.32
An earlier suggestion that the inhabitants were Austronesian speakers was
apparently based on circumstantial evidence. It could be argued that the port
of Oc Eo, as an international port on a well-established international trade
route,wouldhavebeenthetemporaryhomeof Austronesianspeakersinvolved
in this trade. Chinese descriptions indicate that there was some Austronesian
presence in Funan and along the coast to the south,33
and an Austronesian
language (Malayu?) could have been a trade lingua franca. Based on recent

Malayu Antecedents 27
archaeological evidence found at Angkor Borei, however, a far more likely
possibility is that most of the people were Mon-Khmer speakers. At its height,
Funan was said to have extended its influence to settlements on the Isthmus
of Kra and the Malay Peninsula.
As active participants in the Sea of Malayu, these areas would have been
part of a family of communities that exchanged goods and ideas and even
shared ambitions. It is no surprise that a powerful ruler of Funan extended his
political influence westward as far as the northern Malay Peninsula, or that an
ambitious Tambralinga ruler intervened in the politics of Angkor (see below).
These are only two striking examples recorded in history, but they would have
been commonplace and part of family politics in the Sea of Malayu. The well-
developed trade network contributed to an increasing sense of interlinked
political and cultural relationships among the communities. The art historian
Stanley O’Connor describes it as a feeling of a “neighborhood.” “How else
would one explain,” he asks, “the almost parallel development of the monu-
mental Visnu images wearing the long robe in three such widely separated
locations as Dong Si Mahapot, in Prachinburi, at the head of the gulf in east-
ern Thailand, the Mekong delta sites explored by Louis Malleret, and the pen-
insula?” O’Connor is convinced of a“family resemblance” in the architectural
styles and other features used in the service of Buddhism or Hinduism.34
These early sources thus suggest that there was increasing contact between
India, Southeast Asia, and China by the middle of the first millennium BCE.
The land route was favored until the third century CE, when more travelers
began using the sea route. In this early evidence, perhaps of greatest inter-
est to historians is the role of Buddhism in tracing the early trade contacts
between these three regions. The impact of Buddhism in long-distance mari-
time trade in the first millennium CE has long been intimated through stories
from the Mahavamsa and the Sasanavamsappadipika, describing Emperor
Asoka’s decision to send Buddhist missionaries to Suvannabhumi.35
Sona and
Uttara were two of the missionaries sent to Suvannabhumi soon after the
Third Buddhist Council in mid–third century BCE. Although there has been
a tendency to view the two as legendary figures, recent studies on the link
between Buddhism and international trade demonstrate that such a mission
may have indeed occurred.
In the early years of the Common Era, Buddhism shifted its focus from
being a pioneer in agricultural expansion to a promoter of commerce. Bud-
dhist emphasis on accumulation of wealth and its approval of interest earned
on investments made it a favored religion among traders. Links between
traders and Buddhist monasteries grew stronger, and Buddhist symbols
were widely used on pottery, terra-cotta seals, and a variety of other objects.
Monastic establishments in India became economic centers and promoted a

28 Chapter 1
Buddhist trade diaspora that extended to Southeast Asia. Different forms of
Buddhism continued to play a significant role in structuring Southeast Asian
beliefs, statecraft, and trade networks well into the early modern period and
beyond. In the first millennium CE, Buddhism provided an alternative to the
Hindu/Brahmanic model and helped to reinforce trading networks in the
region.36
In a recent study,Tansen Sen documents the commercial role of Bud-
dhist monasteries in China as well as India in funding maritime mercantile
enterprises, including overseas trading ventures. Monks provided both physi-
cal and spiritual care for the merchants, in return for which the merchants
assisted monks in their travels, brought Buddhist items for their patrons, and
financially contributed to the maintenance of Buddhist institutions.37
The Lower Mekong sites provide further evidence of the link between
Buddhism and trade. Buddha images dating between the third and fifth cen-
turies CE were found in Funan and the Cham areas of southern Vietnam. In
Champa, particularly at Tra Kieu, a major Cham center in central Vietnam,
clay Buddhist votive tablets date from the seventh century. John Guy argues
that because Southeast Asian rulers regarded themselves as part of a religious
world that naturally extended to India, the trade in Buddha imagery would
have been as lively as that in spices, aromatic woods, and other desired prod-
ucts from Southeast Asia.38
In offering new religious ideas as well as artifacts,
Buddhism helped strengthen the common cultural bond among communi-
ties already linked by trade.
From the fifth century CE, a rival Vaisnavite trade network developed.
The popularity of both a devotional (bhakti) sect of Vaisnavism and Bud-
dhism would explain why inscriptions and statuary found in pre-Angkorian
sites, in Funan, the Malay Peninsula, and Sumatra are overwhelmingly Vais-
navite or Buddhist. Although Siva lingas are found at these Vaisnavite devo-
tional sites, Pierre-Yves Manguin believes that Siva was regarded as a lesser
divinity. Vaisnavite influence may have accompanied traders from Sogdania
and Bactria in central Asia who settled in Dunsun, somewhere on the Isthmus
of Kra, in the third century CE. Artistic styles and funerary practices dated to
the fifth and seventh centuries reflect the ongoing impact of the Iranian world
on the region.39
Although it is possible to demonstrate contact between India and South-
east Asia as early as the beginning of the first millennium BCE, Smith believes
this was sporadic and initiated by Southeast Asians themselves with their
superior sailing technology. The presence of iron, beads, and a black polish
ceramic known as Rouletted Ware has been cited by many scholars as evidence
of large-scale trade between India and Southeast Asia, but Smith is more cau-
tious. She cites the possibility of local manufacture of iron and beads, and
the possibility of Rouletted Ware being traded much later than the date of

Malayu Antecedents 29
manufacture. There was no compelling reason, she argues, for sustained trad-
ing contact because there was little to be gained. Prior to the fourth century
CE, India had little to offer Southeast Asia economically or politically, and
Southeast Asia’s few requirements could be met in the region itself.
There was, however, a qualitative change in the relationship between the
two areas beginning in the fourth century CE, which is attributed to the rise
and the expansion of the Gupta dynasty in the central Gangetic valley. Dur-
ing the consolidation of power, the Guptas created a political structure and
administrative practices that became a model for other polities in the region.
Among Gupta practices was the use of copper plate to maintain land records
and temple donations,a shift from Buddhism to pre-BuddhistVedic tradition,
and the revival of Sanskrit as the main language of inscriptions, land grants,
seals, and coins. It is about this time that one begins to find in Southeast
Asia evidence of borrowing of Gupta models in iconography, language, and
religion, which are grafted onto indigenous ideas. Only then, Smith suggests,
should one speak of “Indianization” to describe the relationship between
India and Southeast Asia.40
In assessing the evidence thus far presented, certain ideas have been
advanced. First of all, the prevailing Bellwood-Blust synthesis argues that
the general movement of Austronesian speakers, the ancestors of the Malayu
speakers, was southward from Taiwan through the Philippines, down the
Makassar Straits, then to the west as far as central Vietnam and to the east
through eastern Indonesia and out into the Pacific. Of those that went west-
ward from the Makassar Straits, one group settled in west Borneo and became
the ancestors of a subgroup of Malayo-Polynesian called by linguists Malayo-
Chamic. Sometime in the last few hundred years BCE or at the turn of the
Common Era, there was an emigration of Malayo-Chamic speakers out of
Borneo to the coastal areas of the Malay Peninsula. From here one group went
to east coast Sumatra and became the ancestors of the Malayu speakers, while
another sailed to coastal central Vietnam to form the Chamic speakers. Sol-
heim, on the other hand, attributes the existence of people speaking related
languages and sharing common cultures not to migration but to long social
and economic interaction within a network of trade-linked communities,
which he terms “Nusantao.”
A second important idea is that although the “Nusantao”/Malayo-Poly-
nesian speakers settled principally in insular Southeast Asia because of their
maritime orientation, early Indian trade contact with Southeast Asia appears
to have been stronger on the mainland. This suggests that the early Buddhist-
and Vaisnavite-inspired contact, which was later strengthened by the growing

30 Chapter 1
trade relations with the increasingly powerful polities in the subcontinent
of India, probably used one of a number of transpeninsular/transisthmian
routes across mainland Southeast Asia.
The Transpeninsular/Transisthmian Routes
The early Buddhist and Indo-Pacific bead trade from India and Sri Lanka
between the last half of the first millennium BCE and the first half CE appar-
ently used the transpeninsular route located in the Isthmus of Kra and the
northern portions of the Malay Peninsula. The route continued to be used,
though less frequently in later centuries. From the late seventh century, Arab
and Persian ships trading to Southeast Asia and China departed from differ-
ent ports on the Persian Gulf with cargoes of cloth, metal work, carpets, iron
ore, and bullion. They could follow two possible routes: The first began in the
ports on the coast of Oman, then went directly across the Indian Ocean to
Quilon in southern Malabar; the other went along the coast from Hormuz to
northern India, and then to southern Malabar. From southern Malabar, ships
could continue coastal sailing around the Bay of Bengal to the eastern shores,
or they could go south to Sri Lanka, known as Sarandib to the Arabs, and
then directly to the Andaman and Nicobar islands before reaching “Kalah,”
believed by some to be a generalized name for any port on the west coast of
the Isthmus of Kra or the Malay Peninsula. From “Kalah” there were two pos-
sible alternatives: the transpeninsular route to the Gulf of Siam, then to the
Lower Mekong, central Vietnam, and finally China; and the sea route which
went south through the Straits of Melaka to Sumatra, Java, and China.
Those using the all-sea route proceeded southward through the straits,
stopped to replenish their water supplies at Pulau Tioman, an island off Pa-
hang on the east coast of the Malay Peninsula, and then crossed the South
China Sea to ports in Champa in central Vietnam. From the Cham areas the
ships sailed northward to Canton in southern China, either through Hanoi or
via the dangerous Paracel reefs. It has been estimated that in the mid-tenth
century the total sailing time between Muscat and Canton, excluding stops,
was 120 days. In China the sale of the cargo and loading of the new shipment
of goods could be completed in time to catch the northeast monsoon winds,
which blow more strongly and with a more consistent tailwind from China
to the Straits of Melaka. By relying on these winds, Arab and Persian traders
could make the round trip once every year.41
Until the technology for open-sea sailing became widely employed in the
first century CE, ships tended to sail within sight of the coastline. But even
when mariners mastered open-sea sailing, ships continued to hug the coast
because of the profits to be made by buying and selling from one port to the

Malayu Antecedents 31
next. An early sea route went from the east coast of India along the shores
of the Bay of Bengal, down present-day peninsular Burma and Thailand, or
the isthmian region, and then southward to the northern part of the Malay
Peninsula. From the Kra Isthmian and northern Malay Peninsular ports, ships
could continue through the Straits of Melaka to the Gulf of Siam, or they
could unload their goods and have them transshipped via overland routes.
Wolters believes that the Straits of Melaka were not normally used by ships
coming from the west in the first and second centuries CE.42
Use of the transpeninsular routes increased in times of political turmoil
in the straits. The shortest was just sixty-five kilometers at the Isthmus of Kra,
but there were others between the Isthmus of Kra and Kedah that could be
crossed with little difficulty. One was from Kedah to Songkhla, and another
from Trang split into three different branches leading to Phattalung, Nakhon
Si Thammarat, and Bandon on the Gulf of Siam. The route from Takuapa on
the west coast led across the isthmus to Chaiya, but because of political cir-
cumstances this route may have been abandoned in the mid-eleventh century
for one further south in Kedah.43
At various times the competing powers in
the region used different routes across the Isthmus of Kra and the Malay Pen-
insula. Paul Wheatley has identified eleven routes stretching from the Isthmus
of Kra to the southern end of the Malay Peninsula.44
Some of the routes were more difficult than others and involved a vari-
ety of transport: boats, rafts, carts, pack elephants, horses, and bullocks.
Depending on the season and the route used, crossing the isthmus or the
peninsula could take anywhere from a week to about a month, though indi-
viduals without much baggage or cargo could make the journey even faster.
Goods shipped using the Martaban/Moulmein route went by Kokarit, then by
caravan to the Three Pagodas Pass and the Kwai River. The goods were then
reloaded onto boats or rafts, which carried them to ports on the Gulf of Siam.
The Tavoy route along the Kwai River to Kanchanaburi and on to Ayutthaya
was shorter but far more difficult. Because traders had to cross a series of
steep mountains and deep valleys before arriving at the Kwai River, goods
were transported by elephants or porters. Through the centuries, however,
the problems of transport through some formidable landscapes were gradu-
ally overcome. On these routes were found post guards, rest houses, and small
temples dedicated to deities. Every means of transport, from porters to pack
animals and bullock carts, could be rented, and foreign traders resident in the
terminal ports served as interpreters and provided information on business,
types of transport, the roads, lodging, and even alternative routes in times of
war.45
It would have been to the benefit of the local authorities on both ends
of the route to maintain the security of these passages to assure the flow of
trade goods to their lands. Evidence from the Chinese Buddhist pilgrim Yijing

32 Chapter 1
indicates that Sriwijaya may have become involved in the affairs of Kedah
toward the end of the seventh century, at the time of Sriwijayan expansion.
The eighth century Ligor inscription at Nakhon Si Thammarat confirms this
involvement.46
Both Kedah and Ligor were termini of transpeninsular routes
and were obviously still of sufficient importance to warrant the attention of
the rising Sriwijayan power.
The alternative to the land routes was the all-sea route, which in earlier
centuries also had its problems. Sailing the eight hundred kilometers through
the Straits of Melaka took about a month, and fickle wind conditions would
often cause delays. But the major deterrent to using this route was not so
much the length of the journey as the dangers to seaborne commerce. The
Orang Laut, or sea people, inhabiting the islands and coasts at the southern
entrance to the straits were notorious for preying on passing ships. Even if a
ship survived such attacks, it still had to navigate the treacherous shoals, sand-
banks, and submerged islands in the waters to the south of Singapore. For
safety and convenience, traders, diplomats, and other officials in earlier cen-
turies therefore preferred to use the land route. Even during the later period
when the all-sea route was generally favored, any political upheaval in the
Straits of Melaka with the resulting increase in piratical activities forced trad-
ers to use the transpeninsular routes.
In the first millennium CE the typical trader sailed between the Red Sea
and China in one long continuous voyage. From the turn of the millennium,
however, there was a change to less costly, shorter trips dividing the long tra-
jectory into segments. According to K.N.Chaudhuri, the first segment was
from the Red Sea and Persian Gulf to Gujarat and the Malabar coast, the sec-
ond was from the Indian coastal provinces to the Indonesian archipelago, and
the third from Southeast Asia to China. This segmentation was accompanied
by the rise of “great urban emporia” providing neutral ports that provided
merchants with all necessary facilities.47
Leong confirms that most of the
sites in the Malay Peninsula between 1000 BCE and 1000 CE were not major
emporia but small trading settlements serving as collecting centers for special
local products. Notable are the prehistoric sites on the Selangor coast and
in Terengganu located near areas rich in alluvial tin or gold, or along rivers
leading to such areas. In addition to providing local produce, these sites had
the added advantage of being in natural harbors with access to provisions for
revictualing trading ships. There were a few that operated as entrepots, but
most served as redistribution centers for regional trade in the Southeast Asian
area.48
Ships arriving on the Isthmus of Kra or the Malay Peninsula from the
west could unload their goods and reload a new cargo at the same dock, thus
making the entire journey across the Bay of Bengal and back in less than

Malayu Antecedents 33
six months.49
It was in the period of the segmenting of the trade routes that
the eastern termini on the transpeninsular routes, particularly those on the
western shores of the Gulf of Siam, grew prosperous. They profited from
their ideal position as the midpoint of the segmented east-west trade, fac-
ing directly opposite the major entrepots in the Lower Mekong and in cen-
tral Vietnam. While the western termini of the transpeninsular routes may
not have developed into major entrepots, as Leong argues, they proved to be
ideal shelters from the heavy monsoon rains in the Bay of Bengal between
May and October. Ships could anchor in a series of good natural harbors
at Martaban, Ye, Tawai (Tavoy), Mergui/Tenasserim, Kraburi, and Phang
Nga/Phuket. These ports provided storage facilities and were well organized
for the unloading and loading of goods, while the surrounding countryside
offered wood, good drinking water, meat, fruit, and rice to provision the ships
for their onward journeys. Teak was also plentiful for ships in need of repair.
Another attraction was the tin, silver, lead, rubies, sapphires, benzoin, and lac
that were available in the Tenasserim–Isthmus of Kra area. The Takola and
Ligor inscriptions written in Tamil indicate that the Tamil commercial guilds
were certainly using this transpeninsular crossing regularly, perhaps as part of
the trade route to Oc Eo.50
Though the frequency and importance of the overland routes for inter-
national trade are in dispute, there is nevertheless a consensus that the routes
continued to be used. The advances in shipping technology would have most
definitely encouraged a greater use of the sea route, but others may have pre-
ferred to continue using the passage across land for other reasons. In addition
to those already advanced, another was to avoid the exactions of powerful
indigenous and foreign port polities on the shores of the straits. The trans-
peninsular routes would have also had their economic attractions. If Manguin
is correct in assuming that the transpeninsular route was used principally for
regional trade, then the east coast termini on the Gulf of Siam would have
played a role as redistributing centers to areas in mainland Southeast Asia.
A number of Southeast Asian communities came to participate in an eco-
nomic network extending from northern Sumatra, the Isthmus of Kra, and
the northern Malay Peninsula to the Gulf of Siam, the Lower Mekong, and
central Vietnam.
Southeast Asian Components of the Sea of Malayu
Through archaeological and early historical evidence it is possible to describe
the process by which trade fostered a communal identity linked to a region
that I have called the “Sea of Malayu.” In the Southeast Asian part of this
sea, the earliest polities are described in Chinese sources. The first is Dunsun,

34 Chapter 1
perhaps a Mon name, meaning “Five Cities,” which is described as a depen-
dent of Funan with some five hundred families from India. The settlement
had two fo-tu (interpreted as either “stupa” or “Buddhist”), and a thousand
Indian Brahmans, who spent their days studying the sacred canon and prac-
ticing piety. The people, so the texts report, offered their daughters to these
Brahmans, who therefore remained in the polity. The location of Dunsun is
disputed, but Wheatley is convinced that it was somewhere on the Gulf of
Siam and that it extended across the breadth of the isthmian tract. The evi-
dence he cites is a Chinese description of the polity being situated at an“ocean
stepping-stone,” where traders come from east and west. Wheatley interprets
the Chinese phrase as a reference to a place where one crosses from one sea to
another, an apt description of a transisthmian/transpeninsular route linking
the Bay of Bengal and the South China Sea.51
Michel Jacq-Hergoualc’h believes
that Dunsun was a short-lived polity serving as a regional transit center for a
trading network between India and civilizations in Cambodia and Vietnam.52
The next important polity mentioned in Chinese sources is Panpan,
which existed at the end of the fourth century CE and sent an embassy to
China in the early fifth century.Various Chinese sources locate Panpan south-
west of Lin-yi (in central Vietnam) on a bay with “To-ho-lo” adjoining it
to the north and “Lang-ya-hsiu” to the south. To-ho-lo has been identified
as Dvaravati and Lang-ya-hsiu as Langkasuka, thus placing Panpan on the
Isthmus of Kra in the Bay of Bandon. According to the Tang dynastic history,
“the people all learn the brahmanical writings and greatly reverence the law
of the Buddha.” A later fuller account reflects the coexistence of these two
religions in Panpan, where Buddhist monks and nuns study the canon in ten
monasteries and many Brahmans with royal favor are “in search of wealth.”
The people live mostly by the water and within wooden palisades. Another
Chinese source mentions that a Brahman called Kaundinya settled in Panpan
(at the end of the fourth or the beginning of the fifth century) before going
to Funan to become its ruler.53
Evidence thus points to the continuance of an
earlier link between the Lower Mekong delta area with the Isthmus of Kra and
the Malay Peninsula.
Panpan’s northern boundaries could have reached as far as Chumphon
on the Gulf of Siam, and its southern boundaries to perhaps the vicinity of
Songkhla, thus incorporating the region of Sathing Phra and Phatthalung.
But Panpan’s control was only on the east coast and did not extend to the west
coast of the Isthmus of Kra. Buddhist works linked to the art of Dvaravati
of the seventh and eighth centuries have been found along with a number
of Vaisnavite and Saivite remains from the fifth to the eighth centuries. This
supports Chinese accounts of the coexistence of Buddhism and Hinduism,
a common occurrence in Indianizing communities in Southeast Asia. The