L. Andaya

The trans-Sumatra trade and the ethnicization of the Batak

In: Bijdragen tot de Taal-, Land- en Volkenkunde 158 (2002), no: 3, Leiden, 367-409

This PDF-file was downloaded from


The Trans-Sumatra Trade and the

Ethnicization of the ‘Batak’

Considerations of historiography and ethnicity1

Early visitors to Southeast Asia were fascinated by rumours of a cannibal tribe

called the Batak in the interior of Sumatra. When John Anderson travelled

along the east coast and its interior areas in the early part of the nineteenth

century, he met a Batak who told him of having eaten human flesh seven

times, even mentioning his preference for particular parts of the body. Two

other Batak confirmed having also participated in this practice and ‘expressed

their anxiety to enjoy a similar feast upon some of the enemy, pointing to

the other side of the river. This they said was their principal inducement

for engaging in the service of the sultan.’2 Such reports simply reinforced

myths and partial truths which had circulated about these people since

Marco Polo’s oft-quoted story of a Sumatran people (presumably the Batak)

who consumed their ill (Latham 1978:255). European perceptions were also

influenced by stories commonly told in east coast Sumatra by ‘downstream’

(hilir) people that those ‘upstream’ {hulu), that is, in the interior, were hostile

and grotesque. A Portuguese chronicler even repeated downriver stories of

an inland group possessing tails ‘like unto sheep’ (B. Andaya 1995:542).

It has been suggested that lurid details of cannibalistic practices may

have been provided by the Batak themselves in an effort to prevent outsiders

from penetrating into their lands. From early times, therefore, cannibalism

became associated with Batak identity and had the desired effect of limiting

the intrusion of Europeans until the nineteenth century. But perhaps a more

1 My thanks to Barbara Watson Andaya, John Miksic, and Uli Kozok for reading earlier drafts

of this essay and for their most useful comments. I would also like to express my gratitude to Bob

Blust and Sander Adelaar for their helpful advice regarding linguistic evidence.

2 J. Anderson 1971:34. The ‘sultan’ was the Malayu ruler of Deli, who claimed many of Deli’s

hinterland Batak as his subjects.

LEONARD Y. ANDAYA obtained his PhD at Cornell University and is Professor of History at

the University of Hawaii at Manoa. A specialist in the history of Southeast Asia, in particular

Malaysia and Indonesia, he has published, among other titles, The heritage of Arung Palakka; A

history of South Sulawesi (Celebes) in the seventeenth century, The Hague: Nijhoff, 1982, and The

world of Maluku; Eastern Indonesia in the early modern period, Honolulu: University of Hawaii Press,

1993. Professor Andaya may be contacted at the Department of History, University of Hawaii at

Manoa, Honolulu, Hawaii, USA 96822. E-mail address:

368 Leonard Y. Andaya

important reason for the late entry of Europeans in Batak lands was the fact

that, from the beginning of sustained European involvement in the area in

the sixteenth century until the establishment of plantation and other export

industries in the nineteenth century, European orientation was toward the

sea and the coastal polities. With hindsight it is easy for historians to see that

the Batak were fortunate in avoiding the Europeans in these early centuries.

Yet European involvement often resulted in the keeping of records and the

accumulation of written materials which have been crucial in the reconstruction

of the history of many Southeast Asian societies.3 The lack of a European

presence in the Batak lands until the nineteenth century has meant that historians

have had very limited or no access to any contemporary European

accounts of the Batak in the pre-modern period.

The ethnonym ‘Batak’ is very likely an ancient name, but no one has been

able to give a satisfactory meaning of the term.4 Perhaps the very first time

that the name appears in written sources is in the Zhufan zhi, written by Zhao

Rugua, Inspector of Foreign Trade in Fujian, sometime in the mid-thirteenth

century. It mentions a dependency of San-fo-tsi (Srivijaya) called Ba-ta, which

may be a reference to ‘Batak’ (Hirth and Rockhill 1966:35,62,66).5 The next definite

identification of Batak comes from Tome Pires’ Suma Oriental, which was

written in Melaka sometime between 1512 and 1515. It mentions the kingdom

of Bata, bordered on one side by the kingdom of Pasai and the other by the

kingdom of Aru (Cortesao 1990, 1:145). From the sixteenth century onward,

references to the Batak as inhabitants of the interior of north Sumatra, and also

3 For the sixteenth and seventeenth centuries, the official records of the Portuguese and

Spanish overseas enterprise, plus the many accounts found in the collections of the Catholic

Orders in Portugal, Spain, France, and the Vatican, have been valuable for historians. For the seventeenth

and eighteenth centuries, the archives of the European trading companies have proved

useful. The most valuable are the voluminous records of the Dutch East India Company (VOC)

housed in the National Archives in The Hague. They date from the seventeenth and eighteenth

centuries and have been used by historians to reconstruct the early modern history of many parts

of Southeast Asia.

4 In the literature on the Batak, one of the most common explanations for this ethnonym is

that Muslims used it to refer to ‘pig-eaters’. Rita Kipp cites other possible derivations provided

by her informants: from the Sanskrit bhata or bhrta, meaning ‘mercenary, soldier, warrior, hireling,

servant’, because of their functions in the past; and ‘savage’ or ‘bumpkin’ (Kipp 1996:27).

It is tempting to define ‘Batak’ as ‘human beings’, which is a common definition of ethnonyms

of many indigenous groups around the world. The Batek on the Malay Peninsula, for example,

gloss their name as ‘human beings’. Despite the lexical similarity, unfortunately there is no

link between the two terms, because ‘Batek’ is from an Austro-Asiatic language, while ‘Batak’

is Austronesian. There is an Austronesian-speaking group called ‘Batak’ in Palawan in the

Philippines, but no meaning is known for the term.

5 Travellers, including Marco Polo at the end of the thirteenth century, refer to certain groups

who are cannibals in Sumatra without providing the names of such people. One should nevertheless

exercise caution in believing stories of ‘cannibalism’ because of the practice in medieval

Europe for travellers’ tales to depict ‘monstrous races’ in lands beyond their known world.

The Trans-Sumatra Trade and the Ethnicization of the ‘Batak’ 369

of certain kingdoms along the northeast coast, become more frequent.

Today, the Batak groups are listed as the Karo, the Simalungun, the Pakpak-

Dairi, the Toba, and the Angkola-Mandailing. It was the Europeans who first

placed these clusters of communities in and around Lake Toba who spoke a

similar dialect and shared customs under one rubric, the Toba. Following this

usage, I will apply the term ‘Toba’ in this essay to the communities living on

Samosir and the lands surrounding Lake Toba, including those of Silindung.

There is a growing tendency to use the word ‘Batak’ to refer solely to the

Toba, since many of the other groups prefer to be regarded as non-Batak

and as Mandailing, Karo, Simalungun, and so on, in the ongoing process of

redefinition of ethnic groups. In the nineteenth century, however, the term

‘Batak’ appears to have been applied to all these different groups.

In writing this essay, I have been very much aware of the uneven distribution

of source materials. Any systematic study of the Batak began with the

arrival of European missionaries in the nineteenth century. With the penetration

of the area by the Dutch colonial administration later in the century, more

studies were commissioned and travel reports published in governmental

and scholarly journals. The continuing presence of German and Dutch missionaries

and teachers in north Sumatra has assured an ongoing literature on

various aspects of Batak society, particularly its religious beliefs. In addition,

Indonesian government encouragement of local culture in the 1970s and ethnic

chauvinism and pride since the 1990s have fostered Indonesian and local

scholarship on Batak society. For the period before the nineteenth century,

there have been a few archaeological studies, particularly by E. Edwards

McKinnon and John Miksic, which have considerably advanced our understanding

of early settlements in the Batak areas. Nevertheless, much still

needs to be done to gain a more comprehensive understanding of northern

Sumatran communities for the first 1800 years AD.

With the unevenness of the sources in terms of both period and content,

I was confronted with a historiographical problem. Would it be possible to

reconstruct the history of an area on the basis of sources which pre- and

post-date the events themselves? Should a historian undertake such a task

as a legitimate historical enterprise? Both questions I have answered in the

affirmative, but with certain reservations. In the following pages I attempt to

provide a historical overview of economic and political events in the region

of the Straits of Melaka as a basis for suggesting a Batak response to such

events. This reconstruction is based on archaeological findings, as well as

nineteenth- and twentieth-century compilations of origin tales of the various

Batak marga.6 I have also drawn on a knowledge of the better-documented

6 In Batak social organization the marga is one of the basic kinship units and traces descent

to a single male ancestor. Membership of a marga is determined patrilineally, with children of

370 Leonard Y. Andaya

neighbouring communities of the Malayu7 (Malay), Minangkabau, and

Acehnese, as well as groups in the region confronted with similar conditions

as the Batak, in order to discuss the Batak situation. The result is a historical

reconstruction that combines available documentary evidence, historical

imagination, and thirty years’ experience in researching and writing about

societies in the region. I have tried to proceed with caution, and some of

the reconstructed scenarios may eventually prove wrong. Nevertheless, I

believe that this essay has advanced certain ideas that may be worth investigating

further, if new materials come to light, or if historiographical methods

become further refined in the future. In short, I hope that scholars will view

this venture as a genuine attempt to advance the study of a society whose

pre-modern history has been shrouded in mystery for far too long.

One of the analytical tools that I use is ethnicity. There has been a considerable

amount of literature written on ethnicity, principally by sociologists

and anthropologists. The aim of most of these studies has been to determine

the factors which contribute to the formation of ethnic identity. In the past

there were those who argued that each group recognized certain ‘primordial’

elements as the core of their identity, while others claimed that each ethnic

community is the outcome of specific historical circumstances and situations.

More and more, however, studies have taken the middle ground and

acknowledged the importance of ‘primordial’ sentiments, but argue that such

sentiments are in fact constantly undergoing change in response to specific


A factor noted in the formation of ethnic identity is the desire to maximize

the advantages of the group. Many have focused on the economic benefits to

both sexes belonging to the marga of their father. The marga can represent an ancient grouping,

as well as groups that have developed from the original unit. There is evidence that some of the

marga are of mixed origin and have been formed by in-migrants joining with the local population.

Gonda is not totally convinced of Van der Tuuk’s derivation of the term marga from the

Sanskrit varga, meaning ‘company, party, group1. In the Old Malayu inscription at Talang Tuwo in

Palembang from the seventh century, the Sanskrit term marga is used to mean ‘way’ (Gonda 1973:

129-30, 205). This derivation appears to have been retained in later centuries. In the Palembang-

Jambi area the term marga was used for a lineage group. When the Dutch in the early nineteenth

century asked a Palembang man what ‘marga’ meant, he replied: ‘One road, people of one inclination,

one relationship and the same origin1 (B. Andaya 1993:17). It is likely, therefore, that the

Batak marga stems from the Sanskrit term marga, meaning ‘way, road, path’.

7 Throughout this essay I have decided to use the alternative spelling ‘Malayu’, rather than

the current ‘Melayu’, in order to be consistent with archaeologists’ rendering of the name of

the earliest Sumatran kingdom as ‘Malayu’. The people of this kingdom would have thus been

orang Malayu, or the people of Malayu. Even after the demise of Malayu, the people who spoke

the Malayu language and adhered to a culture developed during the Srivijaya/Malayu period

would have been regarded as ‘Malayu’.

8 For a good introduction to the study of ethnicity, see Eriksen 1993. A clear discussion of the

different positions in the debate on ethnicity can be found in Cornell and Hartmann 1998. A useful

and thoughtful synthesis of the issues raised in the study of ethnicity can be found in Kipp

The Trans-Sumatra Trade and the Ethnicization of the ‘Batak’ 371

be gained from creating a particular ethnic unity. A view with less emphasis

on the material and more on the psychological advantages is Horowitz’s idea

of ‘group entitlement’. According to Horowitz, a group’s enhancement of status

and prestige in the eyes of others serves to bolster the individual’s own

sense of pride and self-worth (Horowitz 1985:185, 226-7). Basic to the notion

of ethnic identity is the fact that ethnic consciousness arises through contact

with others who are different. As Eriksen explains, ‘ethnicity is essentially an

aspect of a relationship, not a property of a group’ (Eriksen 1993:11-2). Once

difference is established, it is necessary to exploit this difference through the

establishment of ethnic markers or boundaries. Barth suggests that one focus

on ‘boundaries’, rather than the ‘cultural elements’ contained within such

boundaries (Barth 1969:11). In other words, how a group defines and continues

to maintain itself against another can be far more revealing of ethnic

identity than obvious outward signs such as dress, food, or even language.9

An ethnic group then creates legitimacy and group loyalty through the process

of ‘inventing traditions’ and ‘imagining communities’.10

While social scientists have been at the forefront of such studies, historians

are still to be convinced of the value of ‘ethnicity’ as a useful or even

valid historical pursuit. They may share the Comaroffs’ concern at the lack

of agreement on whether ethnicity is an analytic object, a conceptual subject,

or both (Comaroff and Comaroff 1992:49). The reluctance of historians to

engage the concept of ethnicity in their studies has resulted in an unreflective

acceptance of ethnic communities as somehow fixed forever in time. Yet

anthropological studies have demonstrated the fluidity and complexity of

ethnic identities, particularly in Southeast Asia. Edmund Leach’s classic 1954

study of the Kachin in Burma reveals the ease with which a Kachin could

become Shan and a Shan Kachin through a preference for one over another

form of social system (Leach 1954). Viewing the ethnic problem from a different

perspective, O’Connor argues that ecological adaptation, language, and

agricultural techniques are significant shifts which can explain the so-called

‘rise’ and ‘fall’ of ethnic groups (O’Connor 1995:987).

Among the insights of particular relevance for this essay are: (1) contact

1996:17-24. As mentioned, the literature on ethnicity is vast and the approaches greatly varied.

Historians have yet to contribute much to this literature, with the one major exception of Smith

1986 and Smith and Hutchinson 1996, both excellent sources for historians interested in ethnicity.

9 Nevertheless, Rita Kipp rightfully points out that the outsider still has the task of determining

which of the ‘differences’ – for example, language, dress, religion, or other – would be the

significant ethnic marker or ‘boundary’ (Kipp 1996:19).

10 The term ‘invention of traditions’ comes from Hobsbawm and Ranger 1983. Equally wellknown

is Benedict Anderson’s term ‘imagined communities’ from his book of the same name (B.

Anderson 1983). These scholars focused on the manner in which new, or even not particularly

new, nations invented traditions or found commonalities in order to emphasize their shared

identity and hence unity.



60 km

Map 1. Location of camphor and benzoin forests (from Perret 1995)

The Trans-Sumatra Trade and the Ethnicization of the ‘Batak’ 373

with another group is essential to ethnic consciousness; (2) the group is created

to promote its advantage; and (3) certain ethnic markers are emphasized,

‘invented’, and ‘imagined’ to provide the primordial sentiments for group

solidarity. These insights are useful in assessing historical inter-group relations

within Sumatra, where borderlands provide the opportunity for individuals to

move in and out of ethnicities. Evidence of ethnic shifts from Batak to Malayu

and vice versa has been noted by both Milner (1982) and Perret (1995); less

well documented but equally revealing have been the historical ethnic shifts

between the Batak and the Minangkabau, and the Batak and the Acehnese.

Before examining these ethnic shifts, a significant question that must be

asked is why there should have been a need for a larger ethnic identity in the

first place (Kahn 1993:15). In an effort to seek an answer, I have attempted to

describe the process of ‘ethnicization’ of the Batak. I use this term to indicate

a deliberate decision by the Batak to emphasize their ethnicity for a particular

advantage. On the basis of origin tales and linguistic evidence, I have assumed

that the Batak occupied the area around Lake Toba in the interior of northern

Sumatra in the first millennium AD (Bellwood 1997:122).n International

trade, I argue, was a major catalyst in the movement of Batak from the Toba

highlands towards both coasts, though personal and environmental reasons

also contributed to the out-migration. The interior redistribution centres

and the international marketplaces on the coasts exposed the Batak to new

peoples, new ideas, and new products. In searching for economic advantage

in the highly competitive market environment, they sought support among

their kinfolk, both real and fictive, by ethnicizing their Batak identity. The

last part of the essay then suggests which boundaries were erected by the

ethnicized ‘Batak’ as part of a strategy to maximize economic advantage and

emphasize their unique self-worth.

The camphor and benzoin trade

The camphor (Dryobalanops aromatica Gaetn.f.) and benzoin {Styrax benzoin,

Dryander) trade provided the first, though indirect, evidence of Batak parti-

11 There is no archaeological evidence to reconstruct early habitation of this area, and so I

am basing my assumption on linguistic evidence. According to linguists, much of the spread of

Western Malayo-Polynesian languages occurred after 1500-1000 BC and included the Malayic

speakers. There was an earlier spread of Western Malayo-Polynesian languages which included

those of the Batak and the Gayo of northern Sumatra. Linguists rightfully warn against equating

language with language speakers, since an earlier population could adopt the language of a newcomer.

Unless more conclusive evidence is presented on the ethnicity of the group that occupied

the Toba highlands, I will assume that the inhabitants were ancestors of the group that came to be

identified in later centuries as the Batak. I am grateful to K.A. Adelaar for his informed comments

on this subject.

374 Leonard Y. Andaya

cipation in international commerce. These forest resins were among the products

in greatest demand at the major port-cities in the Straits of Melaka from

the early fifth century, and in Srivijaya between the seventh and eleventh

century. Camphor and benzoin trees grow in the areas of northern Sumatra

now occupied by the Batak (Wolters 1969:111-2,124-5,230-1).12 Camphor was

a highly prized luxury item and so valued in China that it was placed on

a par with gold (Donkin 1999:127).13 Benzoin was regarded as a substitute

for myrrh (Commiphora tnukul Engl.) in southern China by the sixth century,

and later came to replace it as a permanent, valuable commodity in China,

Western Asia, and Europe (Wolters 1969:111). In addition to their muchvaunted

medical qualities as a cure for a host of illnesses and complaints14,

camphor and benzoin were difficult to obtain, which further contributed to

the high prices they could command in the marketplace.

The camphor tree is one of the largest of the dipterocarps in western

Indonesia, reaching a height of between sixty and seventy metres. It grows

at altitudes of 60 to more than 365 metres above sea level on well-drained

soils and often on steep ridges. These conditions are met in the Batak lands

between Singkel and Air Bangis in northwest Sumatra. Benzoin trees grow in

the same areas and under similar conditions. They are found in clumps from

the north of Padang Sidempuan to the area around Tarutung, as well as in

three locations from the mountain valley of the Lai Cinendang, a tributary of

the Singkil River, northward to Sidikalang (see map 1). Camphor crystallizes

in the wood of the tree from an oleoresin present in the tree itself and accumulates

irregularly in the cavities of the trunk. Only after twelve years does the

12 The resin comes from a variety of species. The Styrax paralleloneurum produces a betterquality

benzoin, but the most frequently mentioned in pharmaceutical and botanical literature is

the Styrax benzoin (Katz 1998:243-5).

13 Though no comparative prices are available for this period, a nineteenth-century report

estimates that between a half and 15 kati (280 grams to 8.38 kilograms) could be collected per

tree, and one picul (56 kilograms) of camphor would cost 4000 guilders, a considerable sum in

the nineteenth century (Zeijlstra 1913:826).

14 Among the Chinese, camphor was used against all types of pain and against typhoid, intestinal

discomfort, nasal polyps, rheumatism, eye disease, and so on (Ptak 1998:138). According to

a ninth-century Nestorian physician to six caliphs, in the Arab lands camphor was regarded as

one of the five basic aromatics. It was also used in medicines for gum and eye infections, as an

astringent, and as a prophylactic against the disease-bearing warm winds’. Among the Persians

it was used as a cure for headaches, colds, and bulimia, and was an important ingredient, with

rosewater and sandalwood, in a solution washed on walls during plagues or epidemics (Stephan

1998:234-9). The Sumatrans and Europeans treated camphor as a medicine, using it for ‘strains,

swellings, and rheumatic pains’ (Marsden 1966:153). Benzoin was used in China as an incense to

expel demons and attract benevolent spirits. There is an extensive description of its value from

the tenth century, where it is prescribed as a remedy for a variety of conditions, from ‘warding off

poisonous cholera’ to preventing involuntary emissions by males’ (Wolters 1969:118-9). In Arabia,

Persia, and parts of India it was used as an incense ‘to expel troublesome insects, and obviate the

pernicious effects of unwholesome air or noxious exhalations […]’ (Marsden 1966:155).

The Trans-Sumatra Trade and the Ethnicization of the ‘Batak’ 375

tree produce the camphor, with the oldest trees supplying the greatest quantity

and others yielding nothing at all (Burkill 1966, 1:876-81). Camphor was

presumably collected by Batak men under a special leader known in later centuries

as pawang, whose spiritual prowess was employed in locating the elusive

commodity. Nevertheless, even with the aid of religious practitioners and

adherence to strict taboos, including the use of a special camphor language,

expeditions were not always successful. Writing in the late eighteenth century,

William Marsden claimed that not even 10% of all trees cut down yielded

any crystallized resin or camphor oil (Marsden 1966:150). Benzoin trees were

tapped for their resin after seven years, but stopped producing after about

ten to twelve years. While it may have been easier to collect, the finest quality

could only be obtained in the first three years of tapping. After that the quality

deteriorated, hence its market value lowered (Marsden 1966:154-5,184).

O.W. Wolters has shown that camphor and benzoin were appearing in

China, India and the Middle East by the early sixth century, though not in any

sizeable quantities. But by the eighth century camphor was being included

in the tribute to the Chinese emperor from non-Indonesian rulers, indicating

the growing value of the product in China. It also implies that there was very

likely an increase in the export of camphor from Indonesia (Wolters 1969:230-

1, 233, 235-7). The export of benzoin to China may have begun as early as the

fifth century, though some believe that it began as late as the eighth or even

the ninth century (Katz 1998:259). The increased demand for camphor and

benzoin was met by Srivijaya, a kingdom founded in the late seventh century

on the Musi River in Palembang (Wolters 1969:246-9; Coedes and Damais

1992). Through a series of campaigns Srivijaya overcame its competitors and

became the dominant entrepot in the area.

A Srivijayan inscription placed at Ligor (Nakhon Si Thammarat) in AD

775 indicates an expansion of Srivijayan power across the Straits of Melaka.

A consequence of, and perhaps even an important motivation for, this expansion

would have been the control of camphor supplies from the Isthmus and

the Malay Peninsula. In the annals of the Liang dynasty, which ruled China

from 502 to 556, there is a reference to camphor coming from both Funan and

Langyaxiu. It is believed that the latter is somewhere on the eastern side of

the Malay Peninsula, while the civilization of Funan was centred in the south

of modern Cambodia. Funan must have imported and redistributed the camphor,

since it did not produce the Dryobalanops aromatica variety brought into

China (Ptak 1998:137). Srivijaya’s incursion into the Malay Peninsula would

have prevented the further export of camphor to ports on the Mekong Delta.

By the latter part of the eighth century, therefore, Srivijaya may have succeeded

in monopolizing the sale of camphor and benzoin in the region.

A major source of Srivijayan camphor and benzoin was the forests in

northwest Sumatra. The supply route from these forests to Srivijaya went





• Tarutun



Hutan°Pan >V_1> Muara Sipong

vZ^ Ps

Pariamanjf Q Uw Singkamk V



Map 2. Areas to the south of Lake Toba

The Trans-Sumatra Trade and the Ethnicization of the ‘Batak’ 377

to Padang Lawas via Sipirok and the valley of the Batang Toru (see map 2).

Padang Lawas appears to have been a collecting centre. From here there was

a route leading directly to Barus, as well as two alternative routes southward.

One of the southern routes went via Padang Sidempuan to the valley of the

Batang Angkola, while the other passed near Sibuhuan in Padang Lawas

across the mountains into the Angkola valley near Si Abu. From the Angkola

valley the route continued southward through Bonan Dolok to Penyabungan

and Hutanopari in the Batang Gadis valley. It then crossed the mountains at

Muara Sipongi to Rao.

From Rao one could go directly to Muara Takus in the valley of the Batang

Mahat, a tributary of the Kampar Kanan. But the more frequently used route

passed through the valley of the Batang Sumpur, a tributary of the Sungei

Rokan Kiri, and then through Tanjung Medan and Lubuk Sikaping via Bonjol

into Minangkabau territory. The Batak most likely transferred the products

to the Minangkabau, who then completed the journey through their own

lands downriver to the Malayu in Srivijaya. There were again two alternative

routes leading from Bonjol to Buo, from which place it was possible to reach

the headwaters of the Batang Hari, which is the major river through Jambi

(Edwards McKinnon 1984, 2:340-2). From the Batang Hari the goods could

be sold to the Malayu downriver and then transported by sea to Srivijaya.

Another possibility was to use the tributaries linked by land routes leading

from the Jambi River to the Musi River in Palembang. One such route

followed the tributary Tembesi River, which flowed down along the Jambi-

Palembang border. From Ulu (upriver) Tembesi it was only eight days’ travel

to Palembang and about twelve to Jambi (B. Andaya 1993:102).

The method used to transport the camphor and benzoin in earlier centuries

is not mentioned explicitly in the sources. From available evidence it

appears that cargo was carried by men on their backs travelling on foot along

narrow footpaths. Miksic describes a series of footpaths which ran from the

interior along the hills to both the east and west coasts. Such trails were found

on the summits of the Batak highlands, as well as along the upper reaches

of rivers such as the Panai and Bila (Miksic 1979:97,106). Even as late as the

mid-nineteenth century the Dutch linguist Van der Tuuk recalled an evening

when he hosted half a dozen Toba Batak in Barus who had transported their

cargo of benzoin on their backs (Nieuwenhuys 1962:46). Though horses are

mentioned as an item of trade, it is difficult to find evidence of horses being

used to transport export products. Marsden writes that there were numerous

horses in the Batak lands and that the Batak supplied many to Bengkulen.

Nevertheless, they kept their finest for ritual purposes and apparently as

special delicacies for their festivals: ‘Horse-flesh’, according to Marsden, ‘they

esteem their most exquisite meat, and for this purpose feed them upon grain,

and pay great attention to their keep’ (Marsden 1966:381). Such precious ani378

Leonard Y. Andaya

mals would most likely not have been used as beasts of burden.

For nearly four centuries Srivijaya controlled the trade in forest products

in the region. Its success as a major entrepot to traders from around the world

aroused the envy of other major kingdoms seeking economic dominance in

the area. In 1025 the southern Indian kingdom of the Colas launched an attack

and subdued Srivijaya and its dependencies along the Straits of Melaka.15

Although Srivijaya recovered and reconstituted the kingdom on the Batang

Hari River in Jambi, the name Srivijaya disappeared from the records and

was replaced in the eleventh century by that of an entity known as ‘Malayu’.

Following the Cola invasion, the temporary weakness of Srivijaya and its

Jambi successor, Malayu, as well as the increasing volume of Indian Ocean

trade, enabled several polities to emerge as suppliers of camphor and benzoin.

Nevertheless, Srivijaya continued to maintain its overlordship into

the thirteenth century. Although its secondary centres and feeder ports had

always had some direct trade with foreign merchants, after the late eleventh

century this privilege emerged as a regular pattern. This development was

tolerated as long as the vassal areas did not challenge Srivijaya’s orientation

away from the trans-shipment trade to the direct export trade in Indian

Ocean commodities (Soo 1998:306-8). Two of the most important of these

alternative ports were Barus and Kota Cina.

Barus and Kota Cina

The location of the Tamil inscription dated 1088 from Lobu Tua near Barus

is the strongest evidence so far for Barus’ return to prominence after the

late seventh century. The inscription was erected by a Tamil merchant

guild, the Ayyavole-500 (The Five Hundred of the Thousand Directions’),

which enjoyed the patronage of the Cola dynasty in Tamil Nadu, the Tamil

homeland in southern India. By the end of the eleventh century the guild in

India had begun to include several ethnolinguistic groups among its ranks

and had become established in a number of coastal towns. The Lobu Tua

inscription refers to the guild ‘having met at the velapuram in Varocu, also

called the […] pattinam […]’. ‘Varocu’ is the name for Barus, but there is a

difference of opinion about the meaning of the terms velapuram and pattinam.

Subbarayalu (1998:30-3) believes that the former refers to the harbour, while

the latter describes the town itself. Christie (1998:257), on the other hand,

interprets ‘pattinam’ as designating Barus as a commercial centre of the first

15 Edwards McKinnon (1996:88) suggests that the Tamil merchant guild may have been the

instigator of Cola intervention in Srivijaya territories, with a view to gaining economic advantage

in the increasingly profitable international trade flowing through the Straits of Melaka.

The Trans-Sumatra Trade and the Ethnicization of the ‘Batak’ 379

rank, and ‘velapurarrC as referring to the enclave of Lobu Tua as a trading settlement

of secondary rank.16 Permission was required for admission to the

city, and prices in the trade in aromatics (kasturi) were calculated in gold.17

As an international port, Barus would have had a mixed population, though

its core inhabitants may have been Batak. Direct overland routes from the

nearby camphor forests directly to Barus helped assure the city’s reputation

as a reliable supplier of that prized commodity. Camphor from Barus could

command such high prices that Batak collectors working on the right bank

of the Singkel River in the sixteenth century did not sell their product at the

nearby port of Singkel, but took it to the more distant port of Barus (Miksic


Ptak (1998:139-40) believes that, though Barus was frequented by Indians

and other traders from the west, it was not a major port for the export of

camphor to China. Song and Yuan texts, that is, information from the tenth

to the fourteenth century, do not indicate a regular trade contact between

west-coast Sumatra and the southern Chinese ports of Guangdong, Fujian

and Zhejiang.18 The strong Chinese trade in camphor and benzoin was

most likely focused on another port located on the northeast coast bearing

the revealing name Kota Cina (‘Chinese Stockade’).19 Chinese traders were

more familiar with Sumatra’s northeast coast and the Straits of Melaka20 and

would presumably have gone to Kota Cina, rather than to Barus itself, to

16 Joustra explains that ‘lobu’ means ‘abandoned settlement’ (Joustra 1910:28). ‘Lobu Tua’,

meaning ‘the old abandoned settlement’, could have been the name of an earlier centre which

later moved to the town of Barus.

17 In Sanskrit the word ‘kasturi’ refers to musk. Since musk does not occur in the Barus area,

Subbarayalu has suggested that the term may have been used to refer symbolically to aromatics

in general (Subbarayalu 1998:31-2; Edwards McKinnon 1996:91).

18 This may account for Edwards McKinnon’s speculation, based on Chinese ceramic evidence

at Lobu Tua, that the site was abandoned at about the time of the foundation of Kota Cina

(Edwards McKinnon 1996:89).

19 The name originates from a common practice among’the Chinese to create a fortified

enclosure to protect themselves and their goods while awaiting a shift in monsoon winds before

resuming their journey to India (Miksic 1996:292).

20 Pulau Kompei on Aru Bay is another important place on the northeast Sumatran coast

which produced trade ceramics in the twelfth and thirteenth centuries. This is probably the site of

the Kompei mentioned in Chinese sources as having sent a mission to China in AD 662. Wolters

has suggested that ‘P’o-lo’, which sent a mission to China in the seventh century, was located

in northeast Sumatra. On the same coast flourished Panai between the tenth and fourteenth

centuries, and Aru from the late thirteenth to the early seventeenth century. Milner et al. suggest

that Aru and Deli were different names for the same place. According to Tengku Luckman,

the kingdom of Serdang then split off from the from the old Deli kingdom in the seventeenth

century. Between the seventeenth and nineteenth centuries Asahan, on the same coast, became

a prominent kingdom and an outlet for products from the Batak interior (Nik Hassan Shuhaimi

1984:110; Wolters 1969:187, 193, 220; Milner, Edwards McKinnon, and Tengku Luckman 1978:

18-9; Tengku Luckman 1986:39; Hirosue 1988:40-1).

380 Leonard Y. Andaya

obtain forest resins. The existence of Song and Yuan sherds in interior sites

in Kota Bangun and Deli Tua appears to support this contention. Moreover,

there would have been the added attraction of gold from the nearby mines

in such areas as the Bohorok and Pengkuruan Rivers, some fifty kilometres

west of present-day Medan (Nik Hassan Shuhaimi 1984:109-10).

Although Miksic stresses the Chinese component of the settlement,

Edwards McKinnon argues that Kota Cina was predominantly a Tamil

trading settlement established by merchants like those responsible for the

Lobu Tua inscription in Barus. The existence of permanent religious structures,

including a Siva sanctuary and a Buddhist vihara, is indicative of the

economic importance of the Tamil community for whom they were built

(Edwards McKinnon 1987:86-7). Nevertheless, the Chinese were also a major

presence in the city, judging by the ‘tens of thousands of Chinese porcelain

sherds’ from between the twelfth and fourteenth centuries found on the site

(Miksic 2000:111). Kota Cina was inhabited between the late eleventh and

the fourteenth century, and grew from a small village into a large settlement

of some 10,000 inhabitants by the middle of the twelfth century (Edwards

McKinnon 1996:89; Miksic 1996:292). The ruined site was mentioned by John

Anderson on his trip to east-coast Sumatra in the early nineteenth century

and was only ‘rediscovered’ in 1972 (J. Anderson 1971:294). Located some

three to four miles from the port of Belawan Deli, near the confluence of the

Belawan River (known also as Hamparan Perak or Buluh Cina) and the Deli

River, it was once accessible to sea-going ships (Edwards McKinnon 1984,


The rise of Kota Cina should be viewed in the context of Tamil trading

activity in Sumatra in this period. So far there are three known Tamil settlements

in Kota Cina, Lhok Cut (Aceh), and Lobu Tua, and possible settlements

at Neusu (Aceh, thirteenth century), Bahal 1 (Tapanuli Selatan in the

Padang Lawas area), Buo (West Sumatra), and Kota Kandis on the Batang

Hari in Jambi (Edwards McKinnon 1996:87). It is noteworthy that the Tamilinspired

Buo inscription, the bronze imagery, and a possible temple foundation

at Kota Kandis on the Batang Hari are located on a major route between

the resin forests in the Batak lands and Srivijaya/Malayu. Other Tamil

inscriptions reinforce the view of a fairly extensive Tamil trade involvement

in Sumatra. A provisional reading of the Tamil inscription found at Neusu

appears to refer to trade regulations, while, the nearby site of Lhok Cut is

believed to be the remains of an eleventh-century port. Two further Tamil

inscriptions dating from the second half of the thirteenth century have been

found. The first is a late thirteenth-century inscription found at Batu (or

Bandar) Bapahat, near Suruaso, in the Minangkabau highlands. Though no

transcription or translation has been made, nor any archaeological context

provided, the inscription may relate to the Minangkabau trade in camphor

The Trans-Sumatra Trade and the Ethnicization of the ‘Batak’ 381

and gold.21 The second inscription is from Porlak Dolok near Paringginan in

the Padang Lawas area and dates from either 1258 or 1265. From what can be

inferred from a very damaged text, the inscription commemorates an offering

made by the ruler as a meritorious act (Christie 1998:259-63). The sustained

Tamil economic activity in north and west Sumatra from the eleventh

to the fourteenth century provided the economic stimulus for the increasing

participation of the Batak communities in the camphor and benzoin trade.

These products continued to be transported southward to the entrepots in

Malayu, but by the late eleventh century most of the supplies were going to

Barus and Kota Cina.

The founding of Kota Cina was not an isolated event but was part of the

historical oscillation in the Straits between a single dominant entrepot and a

number of smaller dispersed ports exporting the products of their immediate

interior. Based on recent archaeological explorations in Singapore, Miksic

believes that Kota Cina may have been simply one of a number of similartype

settlements along the Straits of Melaka, which came to include Singapore

(circa 1300) and Melaka (beginning of the fifteenth century) (Miksic 2000:111-

2). Contemporary with Kota Cina was a similar port at Pengkalan Bujang,

across the Straits in Kedah, to the north of the Merbok River. The area of

South Kedah was a site for two important centres based at Kampung Sungai

Mas from the ninth century and at Pengkalan Bujang from the end of the

eleventh century to approximately the beginning of the fourteenth century

0acq-Hergoualc’h 1992:300). Though Jacq-Hergoualc’h considers these two

sites to have been entrepot ports, Leong believes they were mainly a place for

loading and offloading ships, whose cargoes were then redistributed on the

Peninsula (Leong 1990:29). It is apparent that Kota Cina, too, served principally

as a depot for the supply of fresh water and Sumatran forest products.

Though Kota Cina may have been the dominant port on the northeast coast,

there were other possible outlets for Batak goods in this period.22

The economic opportunities offered by Barus and Kota Cina as alternative

sources of camphor and benzoin encouraged the Batak to move toward both

the east and west coasts in order to profit more directly from international

trade. A trans-insular route, though difficult because of the rough and broken

21 The main Minangkabau gold-producing areas are located in Tanah Datar. According to

Dobbin, the main route to the east coast from the valley of the Sinamar around Buo and the

Sumpur around Sumpur Kudus was by water or land to the headwaters of the Indragiri River

and then overland to the headwaters of a tributary of the Kampar Kiri (Dobbin 1983:60-1).

Satyawati (1977:9) suggests that Adityavarman moved his centre to the Minangkabau highlands

in order to control the gold and camphor trade via the Kampar and Batang Hari Rivers.

22 Soo (1998:296) mentions Kampar and Lamuri, but other possible ports were Pulau Kompei,

on Aru Bay, and Panai. Nevertheless, archaeological evidence seems to support the belief that

Kota Cina was the dominant port during its existence.

382 Leonard Y. Andaya

terrain, provided a safer alternative to the sea voyage from the west coast

around Aceh into the Straits. There was therefore an increase in the numbers

of Batak beginning to settle along the new trade routes.

Expansion of the Batak world

The Toba area is said to have been populated by people migrating from

the legendary first Batak village, Sianjur Mulamula, situated on the slopes

of the sacred Pusuk Buhit on the western shore of Lake Toba. Pusuk Buhit

is considered to be the birthplace of their common ancestor, Si Raja Batak,

and the home of the most powerful deities. From here groups left and settled

the series of valleys along the west coast of Lake Toba and then the

southern shores of the lake (Toba-Holbung) in search of rice-growing lands

similar to those found in their homeland. They later fanned out to the island

of Samosir, to the highlands west of the lake (Humbang), to the Silindung

valley, and then westward to the coast (see map 2) (Situmorang 1993:41-2).

In subsequent periods emigration from the Toba lands continued to occur

in response to economic conditions. The process is known among the Toba

Batak as marserak, which originally denoted migration within the territories

of one’s marga or into lands not yet occupied by other marga.23

According to marga origin tales, the point of dispersal was in the Toba

homeland (specifically the island of Samosir and the areas to the west and

south of Lake Toba) and the Pakpak region west of the lake (see map 3).24

Perret points out, however, that most European commentators place the origin

of the Batak peoples somewhere south of the Lake, where the German

mission was strongest (Perret 1995:56, 60). Their reports, Perret infers, may

have influenced later marga origin tales which acknowledge the Toba lands

as the point of origin of their group. As I hope to show, however, the circumstantial

evidence suggests that the Toba area may indeed have been a major

centre for later out-migrating Batak to both coasts and southward to the

present-day Minangkabau homeland.

As a result of the economic opportunities provided by Kota Cina and

other east-coast Sumatran ports between the eleventh and fourteenth centu-

23 The meaning of marserak has n ow expanded to refer to economic a n d social mobility. Other

words are currently in u s e to describe different types of migration (Purba and Purba 1997:22-5).

It must be emphasized here that reasons for emigration of individuals and groups vary considerably.

Economic opportunities, such as n ew trade possibilities, have always been a major pull

factor in migration.

24 This statement is based on genealogical stories contained in a number of sources, including

Sangti 1977; Hoetagaloeng 1926; De Boer 1922; Keuning 1953/54; Wilier 1846; Van Dijk 1895; and

J.H. Neumann 1926.

The Trans-Sumatra Trade and the Ethnicization of the ‘Batak’ 383

ries, Batak groups moved eastward from the Lake Toba and Pakpak regions

using a number of routes. Perret has drawn a useful map showing the spread

of various Karo marga from their homeland in the current Pakpak districts to

the present-day Karo region (see map 4). What is noteworthy is that the area

of the Karo homeland in the Pakpak districts is in close proximity to the camphor

and benzoin forests.25 The thriving trade in forest products encouraged

the establishment of settlements along the major routes which led from the

camphor and benzoin forests through passes in trie Bukit Barisan mountains

and finally down the rivers to Kota Cina. The shortest route from the Karo

highlands to Kota Cina was via the Cingkem pass and then either down the

Serdang River (known in Karo as Lau Tawang) or the Deli River (in Karo, Lau

Petani) to the coast (see map 5). But the easiest route from the highlands was

via the Buaya pass, which followed the upper course of the Ular River (in Karo,

Lau Buaya) to the area of Seribudolok on the border between the present-day

Karo and Simalungun lands. In the nineteenth century the most important

market for the Karo and Simalungun continued to be on this well-frequented

trade route (Westenberg 1905:603). A focus of many of these routes, as well

as the paths leading to the Alas and Gayo lands, was the village of Seberaya,

strategically located within a network of trails leading from the camphor- and

benzoin-producing forests, across the Karo plateau, down to Kota Cina and

the east coast (Edwards McKinnon 1996:69,1987:11, 22-4; Miksic 1979:254).

South of Lake Toba one of the earliest trans-insular routes led from Sibolga

on the west coast, through a low pass in the mountains, to Gunung Tua and

Portibi in the Padang Lawas region. Many of the sites from the eleventh to

the fourteenth centuries are located inland, their main function involving

trade with the highland groups (Bronson et al. 1973:77). Miksic points out

that ceremonial sites, such as those at Padang Lawas and Muara Takus (on

the upper Kampar River), were often located near the border between the

highlands and the coastal plains and ‘may reflect some function in regulating

intercourse between highland and lowland groups’ (Miksic 1979:97, 103).26

From Padang Lawas the major route southward passed through a number of

valleys and towns to Rao. From Rao it was possible to go directly to Muara

Takus via a tributary of the Kampar River, but the more used route seems to

have been to Buo and then out to the Batang Hari River. These routes encour-

25 See Perret 1995:37, map ‘Karo migrations according to tradition’. Sinaga also cites evidence

that the Karo trace their roots to the Pakpak area, which in turn acknowledges an origin in Toba

(Sinaga 1996:46-7).

26 In support of this claim, Edwards McKinnon suggests that the name of the village Portibi’

(Batak for ‘region or quarter) may derive from the Sanskrit pertiwi, referring to a centre of power.

In the Padang Lawas area there are two villages named Portibi: Portibi Jae (Downriver Portibi)

and Portibi Julu (Upriver Portibi), which may have been associated with groups representing the

uplands and the lowlands (Edwards McKinnon 1984,1:30-1).




60 km

Map 3. Early Toba migrations according to traditions collected by Vergouwen

(from Perret 1995)

The Trans-Sumatra Trade and the Ethnicization of the ‘Batak’ 385

Binjei Medan

y Kutacane










‘/ j ^Gunungrintih

, ^ v , – • Barusjahe



OCEAN 15 km

Map 4. Karo migrations according to tradition (from Perret 1995)

aged the migration of peoples from the area of Lake Toba southward into the

region that later came to be associated with the Angkola-Mandailing groups

(J.B. Neumann 1885, 2:17-8).

Migration from the Toba highlands to areas south of Lake Toba extended

into regions of the Malayu and the Minangkabau. It may have begun sometime

in the eighth century, with increased Srivijayan demand for camphor and

benzoin. According to some Malayu traditions from Kampar, the area of Rao

was once Batak but was later seized by certain Minangkabau chieftains. In

addition, the lands directly east of Rao were regarded as Batak. There is also

a story of an attack in the past on Muara Takus by Batak based in Kuamang,

which today is occupied by Malayu. In the nineteenth century a Dutchman

reported seeing in the neighbourhood of Kota Gelugur, on the Kampar River,

a stone inscribed in Batak characters. He explained that the inscribed stone

was intended as a commemorative tablet in honour of the first village heads,

assumed to be Batak in origin. Certain unique traits suggest that the people

of the area may have originated from Mandailing. J.B. Neumann believes that

until the middle of the thirteenth century the Batak occupied the northern

half of the Pasaman Mountains (known in Batak as Dolok Pasoman), which

386 Leonard Y. Andaya





Map 5. Areas to the north and east of Lake Toba

were the source of the Rokan, Siak, and the Kampar Rivers. These mountains,

he argues, marked the southernmost border of the Batak lands. In support of

this argument, he explains in a footnote that the word ‘Pasoman’ indicates ‘the

end of a world’ (J.B. Neumann 1885, 2:17-8). The fourteenth-century Lubuk

Layang inscription found on the border of South Tapanuli, near Padang

Lawas, dates from the time of the Minangkabau ruler Adityavarman and is

believed to have marked a frontier post set up to guard against attacks from

the presumably Batak kingdom of Panai (Satyawati 1977:6).

Ideas of a single Batak ethnicity were strengthened by the fact that many

of those who moved into new lands had a common origin. On the basis of

genealogies collected in Portibi and Mandailing in the early nineteenth century,

Wilier concluded that these areas were settled by migrants from the

Toba homeland. Only after they had been in the area for a long time did a

new noble lineage arrive claiming to be linked to the legendary rulers of

Minangkabau (Wilier 1846:262, 344-5, 400-2, 405). Other origin tales collected

by Batara Sangti indicate that the Lubis and the Nasution, two of the largest

The Trans-Sumatra Trade and the Ethnicization of the ‘Batak’ 387

marga in Angkola-Mandailing, stem from ancestors in the Lake Toba region

(Sangti 1977:129-30).27 The Lubis marga itself acknowledges that its founding

ancestor, Namora Pande Bosi, ‘the great iron-smith’, originally came from

Toba. Also claiming an origin in Toba is the Rangkuti, one of the oldest marga

in Mandailing. They believe that their ancestors were from the marga Parapat,

part of the Borbor group, whose datu are particularly feared for the potency

of their black magic. This may account for the Rangkuti’s fame as the home

of powerful datu (Ypes 1944:141-2). Smaller marga in Mandailing, such as the

Pulungan, Parinduri, Rangkuti, and Borotan, all acknowledge a Toba origin.

According to J. Keuning, two of the largest marga, the Mandailing Godang

and Mandailing Julu, trace their ancestors to Toba lands (Keuning 1953/54:

160-1; Vergouwen 1964:12).28

This movement of Batak people may have occurred at the time of the

most intensive use of the camphor-benzoin routes to Srivijaya/Malayu and

Kota Cina between the eighth and fourteenth centuries.29 Once these groups

became established in their new lands, others were encouraged to join them

in response to economic conditions that rose and fell in accordance with the

rhythm of international trade in the Straits of Melaka.30 The rise of pepper

as an export commodity proved to be a new factor contributing to Batak

emigration from the well-populated areas around Lake Toba. In about the

fifteenth century black pepper (Piper nigrum, Linn.) found a mass market in

China, where it was used in the preparation and preservation of food, and

by the seventeenth century China may have been importing between ten

27 In the current climate of strong ethnic identification and pride in ethnic difference, some

may take issue with these findings, since Batara Sangti himself is a Toba Batak.

28 Mhd. Arbain Lubis, a modern local historian, rejects any idea of a Toba origin for t he

Nasution marga, but argues that the ancestral figure, Si Beroar, was indigenous to Mandailing

(Lubis 1993:193-6). This view represents a common trend among various groups who stress

their difference with the Toba as a way of emphasizing their non-Batak identity. Batara Sangti,

a Toba Batak, cites genealogies to show that the Lubis and Nasution, two of the largest marga

in Angkola-Mandailing, originated from the Toba area (Sangti 1977:129-30). There will be those

who reject such claims because they represent views of a partial source.

29 After the Cola invasion of Srivijaya in 1025, the centre of activity shifted northward to

Jambi, to the old settlement known as Malayu. While the Srivijayan site on the Musi continued

to exist, it was the Malayu kingdom, with capitals both on the coast and in the interior, which

attracted the attention of foreign merchants. In the late thirteenth century, the Javanese kingdom

of Singosari under King Kertanagara attempted to assert its overlordship in the upper reaches

of the Batang Hari. The rivalry between the rulers of Java and Sumatra eventually led to the

movement of the interior Malayu kingdom even further inland to the mountains of the Bukit

Barisan. This then gave rise to the Malayu kingdom in the highlands of Minangkabau under

Adityavarman in the fourteenth century (L. Andaya 2001b).

30 A similar response to economic opportunities is recorded among the Iban groups of

Sarawak. Iban migration is a well-known phenomenon which continues to the present day. They,

like the Batak, moved into empty lands or into sparsely populated areas, quickly absorbing or

dominating the local inhabitants (Pringle 1970:249-51).

388 Leonard Y. Andaya

and twelve thousand picul (1 picul = 60.5 kg) of pepper annually. Europe also

became a major market for pepper, and by 1500 was importing about twelve

hundred tonnes yearly. To meet this burgeoning demand the Sumatran kingdoms

of Aceh, Palembang, and Jambi increased their pepper production.

Some of the Batak may have been enticed to move to the hinterland of these

kingdoms to participate in pepper planting.31 Aceh, at the northern tip of

the island, began to transform some of its interior areas into pepper lands,

and Sultan Iskandar Muda (1607-36) expanded pepper cultivation down

both coasts. He conquered other pepper-producing areas across the Straits,

in Kedah and Perak, to monopolize their production (B. Andaya 1993:43-6;

Lombard 1967:66).

The cultivation of pepper was labour-intensive and required almost continual

attention. Once the men had cleared the forests and planted the pepper,

the women and children were responsible for putting in support plants,

training the pepper vines around them, and weeding the root areas of the

pepper vine. The first pepper harvest came after the fourth year, with a large

and a minor harvest annually thereafter. The pepper-growers were therefore

kept busy picking, cleaning, drying, and bagging the fruit for much of the

year. It was estimated that it took a woman an entire day to sift a picul of pepper

berries. Because of the labour involved in growing pepper, most families

could not plant rice at the same time (B. Andaya 1993:70).

As the powerful rulers of Aceh, Palembang, and Jambi required more

and more of their subjects to plant pepper, rice production in these areas

declined. Rice had to be imported to feed the families now occupied fulltime

in the pepper fields. The surplus rice from the extensive wet-rice

(sawah) fields of the Minangkabau and the Batak in the interior of central

and north Sumatra became the favoured source of supply. Rice, which was

ordinarily scarce in Aceh, was available in great abundance under Sultan

Iskandar Muda (Lombard 1967:73). A major source of Aceh’s supply was the

east-coast polities of Tamiang, Deli, and Asahan, which he seized in order to

gain control of the rice grown in their hinterlands mainly by Batak. By the

mid-seventeenth century, Aceh was importing about 400 tonnes of rice from

Deli alone (Hirosue 1994:21). In the late seventeenth century a Chinese who

lived for ten years among the Batak in the hinterland of Deli described the

over-abundance of rice which the numerous inhabitants enjoyed annually (F.

31 Bugis slaves were used to plant pepper in Jambi and Palembang in the seventeenth century

because many local people refused to remain involved in the strenuous task of pepper cultivation

(B. Andaya 1993:96-7). Batak migrants willing to plant pepper would have been welcome in these

Sumatran kingdoms. Even in the early nineteenth century, when the peak of the pepper trade had

already passed, Anderson noted large numbers of Batak engaged in pepper production in the

interior of Deli. He observed that in the pepper season the river at the ford in Sunggal ‘is almost

impassable for the multitudes of people who flock there with produce’ 0- Anderson 1971:258).

The Trans-Sumatra Trade and the Ethnicization of the ‘Batak’ 389

de Haan 1877:647-8).

The lands in the Lake Toba region were well known as a major source of

food, in particular rice and various types of root crops. When the missionaries

Burton and Ward visited the Silindung valley in 1824, they remarked that

rice and sweet potatoes were widely grown: ‘The former is produced both on

the hills and in the vallies in great abundance, and forms a principal article of

their barter with the bay. On the hills it is grown by the dry process, according

to the common practice with mountain rice; in the valleys irrigation is

employed with some ingenuity. The sweet potatoe grows luxuriantly in

every part of the country, but occupies chiefly the sides of the hills.’ (Burton

and Ward 1827:510.) In the Karo lands sawah fields irrigated by small streams

were laid out mainly in the dusun (the Karo plains from the foothills to the

east coast); whereas in the highlands they were located in the ravines. Dryrice

(ladang) cultivation was more typical in the highlands. The Simalungun

areas grew ladang east of the Karei River, and sawah in the ravines. The Purba

district and some pockets adjoining Lake Toba were planted in sawah, but

ladang cultivation was more common (Westenberg 1905:579-80). In the lands

south of Lake Toba, rice surpluses arose as a result of the extensive cultivation

of sawah in the fertile valleys of the lowlands of Mandailing Godang

(Groot Mandailing), and ladang in the highlands of Mandailing Julu (Klein

Mandailing) (Wilier 1846:370, 373). The sawah fields in the Padang Lawas

region, particularly those in Ulu Barumun, were also noted for their productivity

(Joustra 1910:286, 293, 302-3). Much of the extra labour required to

bring these new lands under cultivation would have come from the populous

areas in the Lake Toba region with their experienced food producers,

thus giving rise to another movement of people from the Lake area to lands

in Karo, Simalungun, and Angkola-Mandailing.

While the international demand for camphor, benzoin, and pepper provided

a major stimulus for Batak migration (marserak), other factors contributed

to the process. They were status enhancement through the founding of

new villages, desire for land, family disputes, the desire for safety from enemies,

and the need to find new land for a growing population (Vergouwen

1964).32 Other more cultural motives for continuing Toba Batak migrations

mentioned by modern scholars are the desire for a long life and numerous

descendants (hagabeon); prosperity and well-being (hamoraon); social status

(hasangapon); ability to exercise authority {sahala harajaon); and skill in gaining

respect (sahala hasangapon) (Purba and Purba 1997:21).

As a result of the extension of the Batak world into new areas, modifications

in the existing marga system occurred. Individuals became members

32 See also ‘Nota over de Landsgroten van Deli’ (unpublished manuscript owned by Tengku

Sinar Luckman, with no indication of original source), p. 15.

390 Leonard Y. Andaya

of new marga through migration, adoption, and birth from ‘incestuous’ relationships

(that is, marriage between members of the same marga) (Ypes 1932:

v). The lands now occupied by the Karo, the Simalungun, and the Angkola-

Mandailing offer more examples of newly formed marga than the Toba areas.

The Toba have extensive genealogies tracing groups to the primeval ancestor,

Si Raja Batak, whereas Simalungung genealogies, for example, rarely go

beyond three generations (Clauss 1982:44). When Van der Tuuk was translating

the Old Testament into Toba Batak in the mid-nineteenth century, he

found that what interested the Toba most were the long biblical genealogies

(Nieuwenhuys 1962:47). In the following century Keuning (1948:15-6) also

noted the great Toba interest in and knowledge of the links among the marga.

People would explain how the marga came to form a main marga, which were

the oldest, middle, and youngest, and how marga came to give rise to even

larger marga, culminating in the moieties of the Lonrung and the Sumba.

The tendency for the Batak, other than the Toba, to downplay genealogical

depth may reflect the relative newness of their marga and therefore the need

to emphasize other more useful linkages than that of an ancestral lineage.

The Karo today usually characterize their society by and base their

identity on the idea of the Merga Silima, or ‘the Five Marga’.33 They are the

marga Karo-Karo, Peranginangin, Ginting, Tarigan, and Sembiring, which all

claim an origin from lands to the west. J.H. Neumann (1926:2-3) suggested

that the ‘original’ inhabitants were a small marga, Karo Sekali, on the basis

of their name, which he translated as ‘genuine or true Karo’ (echte Karo), but

that idea has been challenged.34 Unlike the Toba, with their extended patrilineally

based genealogies going back to a common mythical ancestor, the

Karo emphasize the matrimonial bonds among the five major clans and the

alliances created in the formation of new marga under a local mother marga

(Kipp 1996:34; Singarimbun 1975:71-6; Sinaga 1996:283).35 Equally striking

is Singarimbun’s claim that the ‘Karo do not possess any myth of the origin

of their own society’, nor a ‘ritual center’. The Karo clans, he argues, are not

descent groups, ‘have no history of common origin’, and ‘do not regard themselves

as agnatically related to one another’ (Singarimbun 1975:70, 72).

Simalungun society is very much like that of the Karo in stressing the

equality of the four basic marga of the Saragih, Purba, Damanik and Sinaga,

and ignoring the importance of long genealogical links to the founder of

33 Merga is the Karo term, but I have used marga throughout this essay to avoid confusion.

34 Rita Kipp first raised doubts about Neumann’s interpretation, which identified this marga

as the first or original Karo, because it was found in only one ward in a village (Kipp 1996:44).

Neumann’s views, however, seem to have been adopted by Batak authors themselves; see, for

example, Sangti 1977:129-30.

35 See also Sinaga 1996:284-7 for a description of h ow immigrants from the Toba and Pakpak

areas became part of newly formed Karo marga.

The Trans-Sumatra Trade and the Ethnicization of the ‘Batak’ 391

the marga. The marga do not play a very important role in Simalungun, and

there is an absence of any tradition of common marga territory, property, or

ceremonies (Tarigan 1972:47; Joustra 1910:184). These features of Karo and

Simalungun society appear to be much more in keeping with the nature of

rapidly evolving frontier societies where long-standing traditions have less

relevance than developments in the more recent past. With less venerable

traditions to consider, such societies were more likely to experiment and to

adopt new forms and ideas. A continuing important source for such innovation

among the Batak, particularly in the newly settled communities, was the

Indian subcontinent.

Indian influence and Batak identity

The Tamils were a formative influence on Batak society. Although a ninthcentury

inscription on the Malay Peninsula mentions the presence at Takuapa

of members of the Manikkiramam, a Tamil merchant guild, it was only after

the successful Cola invasion of Srivijayan territories in 1024-5, perhaps at the

behest of Tamil traders, that there was a noticeable increase in Tamil economic

activity in the region (Nilakanta Sastri 1949:25-30; Miksic 1998:120-1). In the

1088 Lobu Tua inscription described above, mention is made of local armed

men, oarsmen, agents, and merchants serving the Tamil guild. Through daily

intercourse between the Tamils and the local inhabitants in this thriving settlement,

ideas would have been exchanged (Subbarayalu 1998:31-3). Another

direct consequence of the Cola invasion was the emergence of Kota Cina.

Edwards McKinnon, the foremost expert on this historical site, has stated

unequivocally: ‘I now see Kota Cina as a predominantly Tamil trading settlement

established by a community of merchants such as the Ainnurruvar

[also known as the Ayyavole] who left an inscription at Lobu Tua’ (Edwards

McKinnon 1987:87).

In response to the rise of Kota Cina, there was a movement of some of

the Tamil population from Barus towards the east coast. Edwards McKinnon

found that the Sembiring marga of the Karo established itself at strategic

points along the routes leading from the west to the east coasts, and that

two of the villages, Deli Tua and Hamparan Perak, were located within easy

reach of Kota Cina (Edwards McKinnon 1987:90-1). The Sembiring marga is

believed to have had direct ties with Tamil traders. The name ‘Sembiring’,

meaning ‘the black one’, is often cited as a major clue. The names of certain

sub-marga – Colia, Berahmana, Pandia, Meliala, Depari, Muham, Pelawi and

Tekan – are clearly of south Indian origin (Edwards McKinnon 1987:85-6;

Parkin 1978:82, 94 fn 47; Singarimbun 1975:78-80; J.H. Neumann 1926:16-7).

In further support of a southern Indian origin of the Sembiring marga, some

392 Leonard Y. Andaya

scholars have cited a mode of disposing of the dead believed to have been

borrowed from the Tamils. This practice involves secondary cremation and

setting the ashes adrift (the pekualuh ceremony) and is found only in the Dairi

lands in the west and among the Karo (N. Siahaan 1964:114-5; Parkin 1978:94,

fn 47; Singarimbun 1975:75). There may also have been some Tamil influence

on Karo ideas on village structure. Urung, the Karo term for a village federation,

is believed to refer to a form of organization found in medieval Tamil

society (Edwards McKinnon 1996:93).

Another source of Indian ideas, particularly in the realm of magic and

religion, were the Indianized Malayu communities. Their influence is especially

evident in the Padang Lawas complex, perhaps the second-largest

archaeological site in Indonesia, encompassing an area with a radius of fifteen

kilometres. Judging from inscriptions found here, Padang Lawas played

an important role in the region from the mid-eleventh to the end of the fifteenth

century. Between 1935 and 1938 Schnitger found some twenty temples

here, as well as a Heruka figure. From the inscriptions and an analysis of the

statuary, he concluded that the devotees were adherents of Vajrayana Tantric

Buddhism, Sivaism, and a syncretic Siva-Buddhism. In one of the temples

found at Parmutung, Schnitger identified what he believed to be an image

of a queen of Panai who founded the temple and who was consecrated as a

Bhairavi (Schnitger 1964:93-4; Parkin 1978:84).

Many authors believe that the presence of Tantrism in the Padang Lawas

complex was due to Indian influence from Malayu/Minangkabau36 via east

Java. In support of this argument, they cite the famous fourteenth-century

Adityavarman statue in the form of the god Bhairava, one of the important

deities in Kalacakra or Left-Handed Tantric Buddhism, found at Rambahan

on the Batang Hari. The inspiration for this statue can be traced directly to

the Singasari court of east Java, where Adityavarman spent some years and

left an inscription in 1343. The model was a similar statue dated 1292 of the

Bhairava seated on a dais surrounded by skulls and wearing a crown, earrings,

and a necklace of skulls (Parkin 1978:254-64; Heine-Geldern 1972:326;

De Casparis 1985:246; Fontein 1990:162-3). Tantric influence appears to have

continued under Adityavarman’s son, Anangavarman, who identified himself

as Heruka, a demon figure. At Kampung Lubuk Layang in Rao, in the

Pasaman district, a headless weatherworn statue broken in two was found

displaying Hindu, possibly Tantric, elements similar to the guardian statues

36 Although Adityavarman is generally regarded as the first Minangkabau ruler, he began his

career as ruler of Malayu. Once he established his base in the Minangkabau homeland, he called

himself Kanakamedinindra, or ‘Lord of the Gold Land’ – a reference to the island of Sumatra.

This shows that he sought to be remembered as the heir of the Srivijayan rulers who first

reigned in Palembang and later moved to Jambi, where the kingdom became known as Malayu.

Adityavarman never mentions the name ‘Minangkabau’ in his inscriptions (Satyawati 1977:9).

The Trans-Sumatra Trade and the Ethnicization of the ‘Batak’ 393

in Padang Lawas (Satyawati 1977:2, 6; Bronson et al. 1973:19).

There is also support for the argument that Indian influence may have

reached Padang Lawas from the north. Parkin, for example, argues that

many Sivaite ideas were brought by Indians themselves through communities

such as those found in Lobu Tua and Kota Cina.37 A team of archaeologists

visiting the site in 1973 concluded that it had no clear relationship with

Java (Bronson et al. 1973:19, 61, 64, 77; Satyawati 1977:2). Their preliminary

findings would suggest that the Padang Lawas complex was a result of

Indian influence coming from the port cities in northern Sumatra rather than

from Java and southern Sumatra. A third possibility is that Padang Lawas

received Indianized ideas from both directions and formed a cultural frontier

between the Minangkabau and the Batak.

Religion and the high priests in the service of trade

Whatever the ultimate source of Indian religious inspiration in Padang

Lawas, the evidence suggests that Indian magico-religious ideas were

eagerly sought by the Batak in order to strengthen their belief systems in

the ongoing struggle to improve their spiritual and material well-being. The

indigenous Batak religion, known as Perbegu or Pemena38, was not supplanted

by religious concepts from India, but came to co-exist with them.

It was therefore possible for the Batak to retain their own beliefs while also

adopting Mahayana Buddhist, Sivaite, and Tantric rituals.

Parkin explains that Perbegu can be viewed as ‘a cult of the human soul,

which in a living person is known as “tondi” and for a dead person is generally

called “begu”‘ (Parkin 1978:6).39 Tondi is sometimes translated as ‘soul stuff

and is found in smaller quantities in animals and plants. It is present in every

part of the human being, including the hair, fingernails, sweat, tears, urine,

excrement, shadow, and even in the name of a person. The most powerful

tondi resides in the placenta and the amniotic fluid at birth, and hence great

care is taken to dispose of these with the utmost secrecy. Ritual cannibalism

37 Three more recent works which include a detailed discussion of the impact of Indian ideas

on Batak indigenous religion are Parkin 1978, Pedersen 1967, and Rae 1994. In the present essay

I have simply focused on Tantrism as an important part of Indian religious ideas that appears

to have been particularly relevant in the southward expansion of Batak society towards the

Minangkabau lands.

38 The old religion is referred to by Christian Batak as Perbegu, or worship of ancestral spirits.

Because of the perceived derogatory nature of this name, adherents prefer the term Pemena,

meaning ‘the First [Religion]’.

39 The word varies from one Batak language to the other. For example, tondi is Toba, tendi

Karo, and tenduy Simalungun. In the following discussion the Toba terms are used.

394 Leonard Y. Andaya

provided the opportunity to strengthen one’s tondi at the expense of the victim

by consuming those parts of the body that are potent with tondi, such as the

blood, heart, palms of the hands, and soles of the feet.40 When a person dies,

the tondi becomes begu (ancestral spirit).41 The most powerful begu, and hence

the one subject to the most frequent appeals, is the sombaon, the spirit of an

ancestor who founded great communities and had at least seven generations

of descendants (Pedersen 1967:19-26; Rae 1994:18-20).42 Through public feasts

at which homage is paid, a begu is transformed into a sumangot, then a sombaon

(Sherman 1990:82).43 The ultimate test of potency was the possession of sahala,

which can be succinctly translated as ‘manifestation of supernatural power’.44

Sahala is manifested in successful economic and other ventures, numerous

progeny, influential relatives, skill in oratory, or bravery in battle. Respect

{hasangapon) accompanies one possessed of sahala, while to refuse to obey and

venerate such a person is to court disaster (Castles 1972:13-4).

From early times religion was closely linked to trade among the Batak.

Religious edifices were erected along trade routes to protect the trader from

adverse human and natural forces and thus assure the economic success of

his venture. Edwards McKinnon observed that from Padang Lawas southward

was a line of candi or temples marking a route from Tapanuli down to

the Minangkabau lands. More candi were found along rivers that were used

to gain access to the east coast. The Padang Lawas or Panai complex arose

due to its strategic location at the crossroads of several riverine and overland

routes.45 The ancient kingdom of Panai, sufficiently important to have warranted

an attack by Cola forces in 1025, benefited from its links to the interior

areas through the important trans-insular portage in the Panai-Barumun

river valley (Edwards McKinnon 1984,1:31-3, 330; Miksic 1979:97).

40 Early Western observers with little or no knowledge of Batak beliefs attributed the preference

for these particular parts of the human body simply to a matter of individual taste.

41 Joustra, however, subscribed to the view of others, who argue that the last breath of a person

becomes the begu. This is based on the belief that the breath cannot be destroyed, that what

is spoken is immortal because it is the wind (Joustra 1902:416).

42 Warneck (1906) describes sombaon as the highest stage that the spirit of the dead can


43 Sombaon is a general term for earth spirits or deities; Ypes believed that it referred also to

the dwelling-places of these beings (Ypes 1932:196).

44 Sahala is in essence the same as the idea of mana in Pacific Island societies. These communities

share a common Austronesian past, and the concept is one which can be traced to the

Austronesian language. For a discussion of mana, see Shore 1989:137-43.

45 Jacq-Hergoualc’h also noted the numerous temples in South Kedah, an area long associated

with Indian traders. These religious edifices were located at the ports and along the rivers

leading to the ports. He believes they were erected by a merchant or group of merchants seeking

the favour of the gods. He also noted the similarity in architectural styles between the temples

in South Kedah and those of Padang Lawas, which he attributes to the use of an Indian model

(Jacq-Hergoualc’h 1992:299, 304-5, 309).

The Trans-Sumatra Trade and the Ethnicization of the ‘Batak’ 395

At the Padang Lawas site, as well as in the Tamil settlements at Lobu Tua

and Kota Cina, temples were prominent. With the withdrawal of the Tamil

population and/or its absorption into the Batak community, perhaps after

the demise of Kota Cina in the fourteenth century, the candi were replaced by

tombs erected in honour of important Batak ancestors (sombaon). Westenberg

noted in 1891 that ‘Malay’ (most likely Batak, who moved easily between two

worlds; perhaps more properly called ‘Malayu Batak’) horse traders were

going to the Karo plateau from the east coast to make offerings at the tombs

of the Sibayak [lords] of Kabanjahe and Barusjahe. On the outward journey

betel was offered, but on the homebound journey, after successful transactions,

a goat or a white chicken was sacrificed (Westenberg 1892:227). These

ancestral tombs proved popular sites of spiritual power.

The religious institution that had the greatest economic impact on the

Batak was that of the high priests.46 Though it originated in the Toba lands, it

spread rapidly to the new areas where Toba migrants had settled. Situmorang

suggests that the Toba Batak believed in a sahala-harajaon, or ‘spiritual power

of governing’, which derived from the gods and was transmitted patrilineally

through the original founders of the three major Toba marga – the Borbor,

the Lontung, and the Sumba.47 It was this sahala-hamjaon which legitimized

the rule of high priests bearing the title Jongi Manaor among the Borbor,

Ompu Palti Raja among the Lontung, and Sisingamangaraja (preceded by

Sorimangaraja) among the Sumba (Situmorang 1987:221-4).48 Although they

were equal in stature within their respective marga, the Sisingamangaraja

was the best known to Europeans. The Ompu Palti Raja, unlike the

Sisingamangaraja, did not claim a divine origin, or authority beyond his own

jurisdiction among the Lontung. The Jongi Manaor’s pretensions were also

far more modest than those of the Sisingamangaraja; he claimed to have his

own areas, independent of either of the other two high priests (Situmorang

46 I have opted for the term ‘high priest’, rather than the more commonly used ‘priest-king’.

‘High priest1 appears more appropriate to the function of these figures in Batak society and

accords with Kozok’s belief that only the last Singamangaraja, the twelfth (1875-1907), referred

to himself as king. In his letters he claimed to be ‘Ruler of the Batak Clans’ and even ‘Ruler of

Sumatra’ (Kozok 2000b:274-6).

47 According to Keuning, Borbor initially formed part of Lontung. As a result of expansion

into areas both of the Lontung and the Sumba, the Borbor came to be regarded as a separate,

major marga (Keuning 1948:16).

48 In a more recent work, Situmorang asserts that Sorimangaraja was the title of the high

priests prior to the creation of the Sisingamangaraja institution in the sixteenth century (Situmorang

1993:218). This date, which is widely cited in the literature, has been arrived at by the

questionable method of counting backward assuming a certain number of years per sundut or

generation. Oral traditions (including those surrounding the origins of the Sisingamangaraja)

tend to telescope years and often refer to events which occurred far earlier. The Sorimangaraja

may have preceded the Sisingamangaraja, but when that occurred cannot be determined with

any certainty.

396 Leonard Y. Andaya

1993:77-8). The high priests’ success in promoting trade and agriculture was

an important measure of their sahala.

There is a fair amount of literature on the Sisingamangaraja, but little on

the Jongi Manaor or the Ompu Palti Raja. One can assume, however, that

many of the distinctive features attributed to the Sisingamangaraja would

have been ascribed to the other two categories of high priest. One of the most

extensive accounts of the origins of the first Sisingamangaraja comes from a

Batak text collected by CM. Pleyte. In this legend the deity Batara Guru causes

a jambu fruit to fall to the ground. It is found and eaten by the wife of the chief

of the village of Bakkara, and she becomes pregnant. After three years pass

with the baby still unborn, a spirit informs the mother that another four years

will elapse before the birth can occur. She will know when it is time because

there will be earthquakes, lightning and a heavy rainstorm, spirits will fill the

village square, and tigers and panthers will tear at one another. These things

occur, and the Sisingamangaraja is born with a black, hairy tongue. The afterbirth

is buried under the house, but lightning strikes at that very spot and

transports the afterbirth to heaven.49 Batara Guru’s messenger then brings

to the child manuscripts with astrological charts for augury purposes and

matters concerning planting and weaving, the calendar, the laws, and a handbook

of spells. The Sisingamangaraja confirms his supernatural origins by

openly declaring, ‘I am a descendant of the gods’ (Pleyte 1903:3, 6-7,15,17).50

Other legends were later added to reaffirm the Sisingamangaraja’s supernatural

attributes. In 1870 C. de Haan was told that the Sisingamangaraja could go

seven months without food and three months without sleep because the gods

supplied his every need (C. de Haan 1875:30).

The divine origins of the Sisingamangaraja made him an ideal intermediary

between the gods and the human community. He could make

peace, create laws, and expose both truth and lies – qualities that made

him unsurpassed in settling disputes. If a war continued unabated, he sent

a staff as a sign that a ceasefire should be declared and the parties submit

to his mediation (Tideman 1936:25-6; Meerwaldt 1899:530). He intervened

in disputes not only among the Batak, but also between the Batak and the

outside world (Cummings 1994:63-4). Early European observers believed

that these high priests exercised very little authority because there were no

visible signs of political power. Heine-Geldern, for example, acknowledged

that the Sisingamangaraja was effective in settling quarrels and mediating

49 As mentioned previously, the afterbirth is regarded as one of the most important sources

of a person’s tondi. The story of the removal of the afterbirth to the heavens emphasizes the

Sisingamangaraja’s divine origins.

50 There are variations on the story, but the general outline is the same. For a very detailed

account of the miraculous birth and life of the first Sisingamangaraja, see Tobing 1967:23-47.

The Trans-Sumatra Trade and the Ethnicization of the ‘Batak’ 397

peace between warring parties, but concluded, ‘otherwise his political power

was weak’ (Heine-Geldern 1953:376). What he failed to realize was that the

Sisingamangaraja and the other high priest figures exercised effective control

not so much through the use of force as through the threat of supernatural

sanctions implied in their words, letters, and widely recognized spiritual

powers (Drakard 1999; L. Andaya 2000).51

Although precolonial Batak society has been characterized by Castles as

being ‘stateless’, there was a hierarchy of institutions under these high priests

which provided a form of supra-village unity. The basic social unit was the

huta, or village, with a varying number of huta forming a horja, and a number

of horja constituting a bins.52 Religious leadership was provided by the parbaringin,

with the chief official of the bius (known variously as raja bius, raja

oloan, or raja na ualu) being chosen by the heads of the horja.53 At the apex of

this hierarchy stood the Sisingamangaraja, who instituted bius markets and

legitimized officials through letters of appointment. Among the responsibilities

of the bius was the hosting of the ‘large market’ {onan na godang or onan

bius), where the ‘great council’ (rapot bolon) mediated disputes and made

binding decisions on important public issues (Kubitscheck 1997:193; Sangti

1977:303; N. Siahaan 1964:112; Castles 1975:74; Tobing 1967:17-8; Situmorang

1993:40-4, 100-2).54

Situmorang traces the origins of the bius to the need for management of the

irrigation system, and hence the organization of agriculture and the implementation

of laws. The bius is usually described as a ‘sacrifice community’

because the culmination of its activities is the annual agricultural ritual and

sacrifice, at which the parbaringin officiated. In addition to ensuring the fertility

of the crops, this sacrifice provides an occasion for community integration

and renewal of commitment to its customs and traditions. Perhaps the most

important agricultural function of the bius was the promotion through the

51 Heine-Geldern points out, however, that the Sisingamangarajas had employed force in the

past. The first had led a war against the Lotung marga, another against the Padris, and a third

against the Dutch (Heine-Geldern 1953:374). However, these rulers were obeyed not so much for

their military as for their spiritual prowess.

52 Sangti says that some twenty huta would then form a horja, and seven horja would make u p

a bius (Sangti 1977:293-4). However, most other commentators give varying figures.

53 Situmorang further divides the bius into three categories, with the most developed being

the bius under the parbaringin. He characterizes the others as ‘developing’ and ‘backward’ bius

(Situmorang 1993:42-3).

54 So great was the reverence for the Sisingamangaraja institution that even after the last

Sisingamangaraja had disappeared in the nineteenth century, the Batak continued to respond

to rumours of his continued presence. In the 1920s a man emerged in Karoland who claimed

that the Sisingamangaraja had commanded everyone to slaughter a white chicken. The response

was immediate and widespread, causing an unprecedented rise in the price of white chickens.

In Angkola, people began to eat a certain type of fish because it was rumoured that the

Sisingamangaraja had ordered it to ward off evil (Castles 1975:74).

398 Leonard Y. Andaya

year of feasts and rituals devoted to the rice-growing cycle and the appeasement

of spirits (Korn 1953:36,126; Sherman 1990:80-5).55 The network of bins

organizations throughout the land provided a supra-village structure based

on a blend of economic, political and religious authority.

The Sisingamangaraja was revered for his powers in ensuring the material

welfare of the people through the promotion of agriculture, creating

harmony among the Batak groups through mediation, and maintenance

of the marketplace. In agriculture he was credited with the ability to bring

rains, locate wells, maintain the irrigation system, enforce the acceptance

of his allocation of rice lands, and ensure the efficacy of agricultural rituals

(Tideman 1936:25-6; Meerwaldt 1899:530; Situmorang 1993:42-3). The young

Sisingamangaraja was said to have been capable of causing rice plants to

grow with their stalks in the ground and their roots in the air. His control

over the growth of rice and various types of ubi or root crops and his ability

to cause rain and to locate well water were attributes expected of one

with direct links to the agricultural deities. Before the rice-planting season

began, the Sisingamangaraja conducted rituals invoking the ancestral spirits

to ensure a good harvest and hence prosperity for their descendants. In Toba

proper – though apparently not in Silindung56 – his appointed officials, the

parbaringin, presided over the sacrifices in the important agricultural rites.

Although there is very little information about the other two high priests,

the Ompu Palti Raja and the Jonggi Manaor, nineteenth- and twentieth-century

sources indicate that they continued to be highly revered for their ability

to summon rain and control rice growth (Hirosue 1994:20, 22; James 1902:

137; Van Dijk 1895:300-1). Conducting the agricultural ritual was considered

an essential task of the parbaringin to assure the ongoing prosperity of the

inhabitants, the animals, and the crops. As late as 1938 the Dutch received

delegations of parbaringin seeking the revocation of a colonial measure introduced

earlier in the century which forbade the performance of this ritual. It

was this prohibition, they asserted, which had resulted in problems in their

community (Korn 1953:32-3).

The esteem and respect for high priests among the Batak may have

increased even further when rice became an important Batak export commodity.

The rise of the pepper trade in the fifteenth century led to an

increasing demand for rice by communities engaged in pepper production

in Sumatra and the Malay Peninsula. It may have been around this time that

the Batak intensified rice planting in existing fields to meet this need. Rice is

55 Sherman, studying the ritual functions of the bius, concluded that it might be compared to

ancestral cults of the earth found elsewhere in Southeast Asia (Sherman 1990:82).

56 The Silindung constitute one of t he major marga in t he Toba area, which may account for their

ability to remain outside the Sisingamangaraja sphere of influence (Ginting forthcoming:291).

The Trans-Sumatra Trade and the Ethnicization of the ‘Batak’ 399

a fragile plant requiring intensive preparation and great care. Moreover, during

its growth it is vulnerable to unexpected weather changes, diseases, and

pests, which may destroy the entire crop. As a result, traditional rice-growing

societies everywhere have resorted to appeals to supernatural forces to

prevent the loss of a crop and to ensure a bountiful harvest. The Batak were

no different, and Sir Thomas Stamford Raffles commented on their belief that

the Sisingamangaraja could ‘blight the paddy, or restore the luxuriance of a

faded crop’ (Raffles 1991:436).

A second important function of the Sisingamangaraja was promoting harmony

among the Batak groups through his mediation. In this role he was able

to effect wide agreement on standard rice measures, as well as the assurance

that the sanctity of the marketplace would be observed. When the missionaries

Burton and Ward travelled to the Toba lands, they commented on the

influence of the Sisingamangaraja, who was considered by the inhabitants

to be ‘bertuati, or ‘invested with supernatural power’. His representatives,

whom Burton and Ward believed to be village chiefs from the surrounding

districts (Burton and Ward 1827:514), were known as parbaringin in the

Sumba districts. They were appointed by the Sisingamangaraja and had the

important responsibility of maintaining the viability of the markets (Castles


By the nineteenth century it was possible to distinguish a heartland and

an extended network of communities forming a single Batak cultural unity,

promoted and strengthened by the activities of the high priests. Although

the latter had arisen among the Toba, their influence extended to the other

areas where the Batak had settled. The Ompu Palti Raja was the high priest

with the greatest influence among those in the Simalungun lands involved

in the trade between Lake Toba and the east coast, while the Jonggi Manaor’s

area of jurisdiction was in the lands between the interior and Barus. Of these

three, the Sisingamangaraja exercised the greatest influence among the Batak

communities in general. Representatives bore their insignia and exercised

authority on their behalf because of the awe and veneration with which the

Batak regarded these high priests (Hirosue 1994:22). As the Batak became

increasingly involved in international trade, these magico-religious figures

became the foci and the facilitators of the production and delivery of rice

and forest products from the interior to the coasts. The expansion of their

functions contributed to the evolution of a supra-village authority and to a

growing sense among the people of belonging to a single ethnic group under

the leadership of the high priests.

400 Leonard Y. Andaya

Ethnicization of the Batak

As the Batak moved toward both coasts and southward from Lake Toba in

response to economic opportunities, they came into direct competition with

the Malayu, the Minangkabau, and the Acehnese. In face of this development,

the institution of the high priests was invoked to promote ethnic unity.

The acknowledgement of the Sisingamangaraja as the overarching spiritual

authority over all Batak may have been a deliberate economic decision by the

Batak in order to compete effectively against the newly ethnicized Malayu,

Minangkabau and Acehnese.57 With the appointment of parbaringin, a hierarchy

was created whose major responsibility was the maintenance of agriculture

and the marketplace. If not the threat of supernatural sanctions then the

promise of economic advantage made the institution of the Sisingamangaraja

appealing to the Batak.

A European report from the early nineteenth century confirms the elevated

status and veneration enjoyed by the Si Singamangaraja among the Batak.

In a letter to Marsden dated 27 February 1820, Raffles wrote that among the

Batak he was ‘something like an ecclesiastical Emperor or Chief, who is universally

acknowledged, and referred to in all case of public calamity, etc. His

title is Si Singah Maha Rajah, and he resides at Bakara in the Toba district. He

is descended from the Menangkabau race, and is of an antiquity which none

disputes. My informants say certainly above thirty descents, or 900 years. He

does not live in any very great state, but is particular in his observances; he

neither eats hog nor drinks tuah [palm-wine]. They believe him possessed

of supernatural powers.’ (Raffles 1991:435-6.) In this letter Raffles claims

that the Sisingamangaraja was ‘universally’ acknowledged. Although it is

more likely that he had direct influence only over the Sumba group of marga

among the Toba Batak, stories of his supernatural powers would have been

sufficient to convince many other Batak to heed his words or those of the persons

delegated to represent him. In this way the Batak in the southern Lake

Toba region, who were the Sisingamangaraja’s principal adherents, would

have been joined by Batak elsewhere in forming a group responsive to his

wishes. While he did not possess any means for physical coercion, he had a

reputation for magico-spiritual powers which in earlier centuries proved far

57 I argue in other essays that there was a conscious decision by the Malayu rulers of Melaka

and Johor, the Minangkabau rulers of Pagaruyung, and the Acehnese rulers to appeal to a politicized

ethnic identity for economic reasons in the period between the sixteenth and eighteenth

centuries. Between the sixteenth and the late seventeenth century, Aceh saw itself as a Malayu

kingdom and was the dominant economic, political, and cultural entity in the Malayu world.

Only from the eighteenth century did a separate Acehnese identity emerge in recognition of the

success of Johor in becoming acknowledged as the centre of Malayu culture. See L. Andaya 2000,

2001a, and 2001b.

The Trans-Sumatra Trade and the Ethnicization of the ‘Batak’ 401

more intimidating. Instead of a political structure with the accoutrements

of state power, the Sisingamangaraja and other high priests created a unity

among many Batak groups on the basis of their sacred reputation, system of

marketplaces, and coterie of magico-religious officials operating in a borderless


Batak ethnic consciousness was reinforced by the creation of pustaha, or

bark books. Written in a language and a script unlike anything possessed by

their neighbours, the pustaha were regarded as distinctly ‘Batak’. Although

employing an old Indian Pallava-derived script, there is no record of when

pustaha first began to be written. Kozok has shown that the Batak script continues

to display an affinity with the Pallava and Old Javanese (Kawi) scripts,

whereas modern Javanese has diverged quite significantly from the original

Pallava (Kozok 1999:65). Batak writing may have originated with the creation

of the pustaha, but remained relatively unchanged over the centuries, perhaps

because of the sacred contents. The pustaha contained astrological tables and

magic formulae and were intended for magico-religious purposes.58

The survival of a Batak language using a modified Pallava script to

transmit sacred and other tribal knowledge is noteworthy. From the seventh

until at least the fourteenth century, the dominant intellectual and political

languages in Sumatra were Sanskrit and Malayu. Their influence was particularly

strong, and evidence of their presence has been noted in the discussion

of the archaeological finds at Padang Lawas. Yet despite these cultural

incursions, the Batak were not overwhelmed by the expansion of the Malayu

language and culture into northern Sumatra (Teeuw 1959:148-51; Collins

1996:9). The survival and persistence of the pustaha tradition may have

been the result of a deliberate political choice at a time when the Batak were

becoming increasingly involved in economic rivalry with neighbouring communities.

As Pollock so succinctly explained, ‘Vernacular literary languages

do not “emerge” like buds or butterflies, they are made’ (Pollock 1998:7).59 A

Batak world was thus inscribed and circumscribed by the pustaha, which not

only played a magico-religious role but also became an important marker of

Batak identity.

Often in the introduction to pustaha, a chain of transmission of knowledge

from the legendary founder to the current writer is listed. Teachers

58 In a d d i t i on to t h e pustaha, there were other forms of writing, such as letters, pulas (a t y p e of

threatening letter), and laments, though the latter two forms tended also to have a strong magicoreligious

intent (Kozok 2000a:43-4).

59 I have based my arguments on Pollock’s stimulating discussion of the process of vernacularization

in India. Of particular value and relevance for the Batak situation is his argument that

there is a division of labour in languages, in which Sanskrit retains its position as ‘the public

literary expression of political will’, while the vernacular is restricted to ‘business’ or practical

aspects. He terms this language division ‘hyperglossia’ (Pollock 1998:11-2).

402 Leonard Y. Andaya

and pupils from different regions travelled together through the Batak areas

because their services were sought everywhere (Voorhoeve 1927:10, 13).

When the intrepid Italian traveller Elio Modigliani journeyed through the

Toba Batak area in 1890, he befriended the great datu, Guru Somalaing, from

whom he obtained a text by the ‘wandering datu’ of the Simanjuntak marga

intended for his pupils belonging to the Siagian marga. The itinerant character

of these datu is emphasized in another text collected by Modigliani, where

one of the great masters is called ‘Singa Mortandang’, or ‘wandering lion’

(Voorhoeve 1979/80:62, 78, 82). It was also commonplace for pupils to travel

long distances to study with famous datu (Kozok 1999:17).

Through long and intensive study, the datu acquired an incomparable

knowledge of the future, the characteristics of plants, and the wisdom contained

in the writings of the ancestors. The wandering datu was described as

not simply a religious practitioner, but also ‘a man of science who embodies

all current available historical, medical, theological and economic knowledge’

(J.H. Neumann 1910:2). Through his knowledge of the contents of the

pustaha, he became the primary source of the old tales, legends, and traditions

from which the Batak gained an understanding of their rituals 0.H.

Neumann 1910:2, 10).60 This latter function still survives among the Batak

today. Ginting describes a Karo guru, the Karo equivalent of a datu, who can

‘recite in a sing-song tone the old legends and myths which are important

in the performance of a ritual so that the participants understand its background

and can therefore experience the ritual more intensely’ (Ginting 1991:

86-7). The datu also used his knowledge of plants and the spirit world to

concoct the various medicines for treating and preventing illnesses, conduct

special rituals to ward off evil or recall a spirit which had wandered away

from a body, and prescribe potions to assist in affairs of the heart and give

self-confidence (Wilier 1846:295-6; Ginting 1991:86-7).

Because of the datu’s ability to assure the well-being of the community in

so many different ways, he gained the confidence and support of the people.

He thus became an influential advocate and an ideal conduit for information

and directives of the high priest. His wandering life-style and the practice of

accepting pupils from all over the Batak lands contributed to a network that

transcended territorial and marga divisions. Also strengthening the sense of

a unified Batak world was the pustaha tradition. Voorhoeve, in his intensive

study of pustaha, concluded that the sacred language of the texts derives

from a sub-Toba dialect spread by wandering datu, who were immune to

inter-marga and inter-village conflicts in precolonial times (Voorhoeve 1973:

60 Ginting reminds us, however, that not all guru [or datu] achieved the same level of competence.

Those with exceptional skill won a reputation as guru mbelin, or ‘great guru’ (Ginting 1991:

94, 96).

The Trans-Sumatra Trade and the Ethnicization of the ‘Batak’ 403

39). The spread of the pustaha tradition helped create a shared sacred language

and a common store of magico-religious lore. Prior to the twentieth

century, Perbegu/Pemena, or the old Batak religion, was a core element of

Batak identity. The key to the ethnicization of the Batak was provided by the

components of Perbegu/Pemena: the high priests, the datu, and the pustaha.


The people who are collectively known as Batak today were historically

never isolated from the developments occurring in the region. From very

early times they were incorporated into regional trade networks because

they were major suppliers of camphor and benzoin – two of the most highly

valued Southeast Asian commodities in the international trade from at least

the eighth up to the nineteenth century (Burkill 1966, 1:878-9). The involvement

of the Batak in international trade made them responsive to political

and economic shifts that had a direct impact on their livelihood. When

Srivijaya was conquered by the rival Cola dynasty in 1025, the Batak sought

other outlets for their products. The rise of Kota Cina on the east coast and

the re-emergence of Barus on the west coast as ports for the export of camphor

and benzoin drew the Batak towards both coasts. Though Kota Cina

itself disappeared sometime in the fourteenth century, other east-coast kingdoms

came to provide an outlet for the export of Batak forest products and

rice in later centuries.

While Srivijaya was still the dominant entrepot in the Straits, the Batak

used routes from the camphor and benzoin forests located to the northwest

and southeast of Lake Toba southward to Padang Lawas, then on to the

Batang Hari, and eventually to Srivijaya on the Musi River in Palembang.

After 1025 Kota Cina and Barus joined Srivijaya and Malayu as exporting

centres for these resins, and much of the camphor and benzoin supplies was

diverted eastward and westward towards the coasts. From the eighth to the

fourteenth centuries, Batak groups sought to profit from international trade

by following these routes and settling in proximity of these export centres.

Another major economic stimulus to Batak migrations was the growing

demand for rice among pepper growers in Sumatra and the Malay Peninsula,

beginning in the fifteenth century. To meet this new demand, there were

migrations from the Toba region to new lands south and east of Lake Toba

in search of rice lands.

Crucial to the success of Batak involvement in international trade were

their religious institutions. Candi and ancestral tombs were judiciously

placed along major trade routes to assure spiritual protection and success

for Batak traders. With the increasing tempo of trade and the dispersal of

404 Leonard Y. Andaya

Batak communities from the Lake Toba region, a need arose for some form of

supra-village control. This was provided by the institution of the high priest,

which originated in the Toba lands but gained support in the other Batak

areas. Through their claims of supernatural power, access to agricultural deities,

and creation of a network of officials and markets, the high priests were

instrumental in the promotion of Batak trade until their demise in the early

twentieth century. The activities of the datu helped to ensure continued support

for the high priests among the Batak in the pre-modern period.

As different ethnic groups became increasingly competitive in international

trade, particularly in the period between the fifteenth and eighteenth

centuries, every avenue was explored to gain an advantage over the others.

One response was the ethnicization of identity, or in other words, a conscious

decision to emphasize ethnicity to maximize their advantage. The Batak

became ‘ethnicized’ by stressing commonality in their acknowledged origins

in the Toba highlands, their recognition of the authority of the high priests,

and their reliance on the knowledge and spiritual powers of the datu and

their pustaha. In the early modern period being ‘Batak’ became both a political

and an economic option, resulting in the removal of huta and marga barriers

in the formation of a common Batak ethnicity.


Andaya, Barbara Watson, 1993, To live as brothers; Southeast Sumatra in the seventeenth

and eighteenth centuries. Honolulu: University of Hawaii Press.

-, 1995, ‘Upstreams and downstreams in early modern Sumatra’, The Historian 57:


Andaya, Leonard Y, 2000, ‘Unravelling Minangkabau ethnicity’, Itinerario; European

Journal of Overseas History 24-2:20-43.

-, 2001a, ‘Aceh’s contribution to standards of Malayness’, Archipel 61:29-68.

-, 2001b, ‘The search for the origins of “Malayu”‘, Journal of Southeast Asian Studies


Anderson, Benedict, 1983, Imagined communities. London / New York: Verso.

Anderson, John, 1971, Mission to the east coast of Sumatra in 1823. Kuala Lumpur:

Oxford University Press. [Oxford University Reprints.]

Barth, Fredrik, 1969, Introduction’, in: Fredrik Barth (ed.), Ethnic groups and boundaries;

The social organization of culture difference, pp. 9-38. Boston: Little, Brown.

Bellwood, Peter, 1997, Prehistory of the Indo-Malaysian archipelago. Honolulu: University

of Hawaii Press. [First impression 1985.]

Boer, D.W.N. de, 1922, ‘Een en ander over de herkomst en uitzwerming der Bataks’,

Koloniaal Tijdschrift 11:86-95.

Bronson, Bennett, Besoeki, Machi Suhadi, and Jan Wisseman (eds), 1973, Laporan penelitian

arkeologi di Sumatera. Jakarta: Pusat Penelitian Arkeologi Nasional.

Burkill, I.H., 1966, A dictionary of economic products of the Malay Peninsula. Kuala

Lumpur: Ministry of Agriculture and Co-operatives, 2 vols.

The Trans-Sumatra Trade and the Ethnidzation of the ‘Batak’ 405

Burton and Ward, 1827, ‘Report of a journey into the Batak country, in the interior of

Sumatra, in the year 1824′, Transactions of the Royal Asiatic Society 1:485-513.

Casparis, J.G. de, 1985, ‘Srivijaya and Malayu’, in: SPAFA Final Report, SEAMEO

Project in Archaeology and Fine Arts, Jakarta/Padang/Prapat/Medan, Indonesia,

September 16-30, pp. 245-53. N.p.: n.n.

Castles, Lance, 1972, The political life of a Sumatran residency; Tapanuli 1915-1940. [PhD

thesis, Yale University, New Haven.]

-, 1975, ‘Statelessness and stateforming tendencies among the Bataks before colonial

rule’, in: Anthony Reid and Lance Castles (eds), Pre-colonial state systems in

Southeast Asia, pp. 67-76. Kuala Lumpur: Malaysian Branch of the Royal Asiatic


Christie, Jan Wisseman, 1998, ‘The medieval Tamil-language inscriptions in Southeast

Asia and China’, journal of Southeast Asian Studies 29-2:239-68.

Clauss, Wolfgang, 1982, Economic and social change among the Simalungun Batak of North

Sumatra. Saarbrucken: Breitenbach.

Coedes, George, and Louis-Charles Damais, 1992, Sriwijaya; History, religion and

language of an early Malay polity. Kuala Lumpur: Malaysian Branch of the Royal

Asiatic Society 20.

Collins, James, 1996, Malay, world language of the ages; A sketch of its history. Kuala

Lumpur: Dewan Bahasa dan Pustaka.

Comaroff, Jean, and John Comaroff, 1992, Ethnography and the historical imagination.

Boulder: Westview Press.

Cornell, Stephen, and Douglas Hartmann, 1998, Ethnicity and race; Making identities in

a changing world. Thousand Oaks: Pine Forge Press.

Cortesao, Armando (ed.), 1990, The Suma Oriental of Tome Pires. New Delhi / Madras:

Asian Educational Services, 2 vols. [First impression 1944.]

Cummings, William, 1994, ‘Cultural interaction in a Sumatran state; Deli 1814-1872’.

[MA Thesis, University of Hawaii, Honolulu.]

Dijk, P.A.L.E. van, 1895, ‘Eenige aanteekeningen omtrent de verschillende stammen

(margas) en de stamverdeeling bij de Battaks; Het priesterhoofd Si Singa

Mangaradja, syn ontstaan en syne afkomst’, Tijdschrift voor Indische Taal-, Land- en

Volkenkunde 38:294-315.

Dobbin, Christine, 1983, Islamic revivalism in a changing peasant economy; Central

Sumatra, 1784-1847. London/Malmo: Curzon Press.

Donkin, R.A., 1999, Dragon’s brain perfume; An historical geography of camphor. Leiden:


Drakard, Jane, 1999, A kingdom of words; Language and power in Sumatra. Kuala

Lumpur: Oxford University Press.

Edwards McKinnon, E., 1984, Kota Cina; Its context and meaning in the trade of Southeast

Asia in the twelfth to fourteenth centuries, 2 vols. [PhD thesis, Cornell University.]

-, 1987, ‘New light on the Indianization of the Karo Batak’, in: Rainer Carle (ed.),

Culture and societies of North Sumatra, pp. 81-110. Berlin: Reimer.

-, 1996, ‘Mediaeval Tamil involvement in northern Sumatra, cll-cl4 (the gold and

resin trade)’, Journal of the Malaysian Branch of the Royal Asiatic Society 69-1:85-99.

Eriksen, Thomas Hyland, 1993, Ethnicity and nationalism; Anthropological perspective.

London / East Haven, CT: Pluto Press.

Fontein, Jan, 1990, The sculpture of Indonesia. Washington: National Gallery of Art.

406 Leonard Y. Andaya

Ginting, Juara Rv 1991, ‘Pak Surdam; A Karo Batak guru’, in: Achim Sibeth (ed.), The

Batak, pp. 85-99. London: Thames and Hudson.

– forthcoming, ‘Inter-group relations in North Sumatra’, in: Geoffrey Benjamin and

Cynthia Chou (eds), Tribal communities; Historical, cultural and social perspectives, pp.


Gonda, ]., 1973, Sanskrit in Indonesia. New Delhi: International Academy of Indian

Culture. [First impression 1952.]

Haan, C. de, 1875, ‘Verslag van eene reis in de Bataklanden’, Verhandelingen van het

Bataviaasch Genootschap 38:1-53.

Haan, F. de, 1877, ‘Een oud bericht aangaande de Batta’s’, Tijdschrift voor Indische Taal-,

Land- en Volkenkunde 39:647-8.

Heine-Geldern, Robert, 1953, ‘Le pays de P’i-K’ien; Le roi au grand cou et le Singa

Mangaradja’, Bulletin de I’Ecole Frangaise d’Extreme-Orient 49-2:361-404.

-, 1972, ‘The archaeology and art of Sumatra’, in: Edwin M. Loeb (ed.), Sumatra; Its

history and people, pp. 305-32. Singapore: Oxford University Press.

Hirosue, Masashi, 1988, Prophets and followers in Batak millenarian responses to the

colonial order; Parmalim, Na Siak Bagi and Parhudamdam, 1890-1930. [PhD thesis,

Australian National University, Canberra.]

-, 1994, ‘Port polities and hinterlands in north Sumatra’, Paper presented at the 13th

Conference of the International Association of Historians of Asia, Tokyo, 5-9 September.

Hirth, Friedrich, and W.W. Rockhill, 1966, Chau Ju-kua; His work on the Chinese and Arab

trade in the twelfth and thirteenth centuries, entitled chu-fan-chi. New York: Paragon.

[First impression 1911.]

Hobsbawm, Eric, and Terence Ranger (eds), 1983, The invention of tradition. Cambridge:

Cambridge University Press.

Hoetagaloeng, W.M., 1926, Poestaha taringot taromba ni Bangso Batak. Laguboti: Zendings-


Horowitz, Donald, 1985, Ethnic groups in conflict. Berkeley: University of California


Jacq-Hergoualc’h, Michel, 1992, La civilisation de ports-entrepots du sud Kedah (Malaysia)

V-XIV6 siecle. Paris: l’Harmattan.

James, K.A., 1902, ‘De geboorte van Singa Maharadja en het ontstaan van de Koeria

[District] Ilir in de onderafdeeling Baros’, Tijdschrift voor Indische Taal-, Land- en

Volkenkunde 45:134-43.

Joustra, M., 1902, ‘Het leven en zeden en gewoonten der Bataks’, Mededeelingen vanwege

het Nederlandsch Zendelinggenootschap 46:385-426.

-, 1910, Batakspiegel. Leiden: Van Doesburgh. [Uitgaven van het Bataksch Instituut 3.]

Kahn, Joel S., 1993, Constituting the Minangkabau; Peasants, culture and modernity in

colonial Indonesia. Providence/Oxford: Oxford University Press.

Katz, Esther, 1998, Sexploitation du benjoin dans les hautes terres batak; Situation

actuelle’, in: Claude Guillot (ed.), Histoire de Barus, pp. 243-64. Paris: Association

Archipel. [Cahier d’Archipel 30.]

Keuning, ]., 1948, Verwantschapsrecht en volksordening en huwelijksrecht en erfrecht in het

koeriagebied van Tapanoeli. [PhD thesis, Leiden University.]

-, 1953/54, ‘Toba-Bataks en Mandailing-Bataks; Hun culturele samenhang en daadwerkelijk

antagonisme’, Indonesia 7:156-73.

The Trans-Sumatra Trade and the Ethnicization of the ‘Batak’ 407

Kipp, Rita Smith, 1996, Dissociated identities; Ethnicity, religion, and class in an Indonesian

society. Ann Arbor: University of Michigan Press. [First impression 1993.]

Korn, V.E., 1953, ‘Bataksche offerande’, Bijdragen tot de Taal-, Land- en Volkenkunde 109-

1:1, 32-51, 97-127.

Kozok, Uli, 1999, Warisan leluhur; Sastra lama dan aksara Batak. Jakarta: ficole Franchise

d’Extreme-Orient / Kepustakaan Populer Gramedia.

-, 2000a, ‘On writing the not-to-be-read; Literature and literacy in a pre-colonial

“tribal” society’, Bijdragen tot de Taal-, Land- en Volkenkunde 156-1:33-55.

-, 2000b, ‘The seals of the last Singamangaraja’, Indonesia and the Malay World 28-82:


Kubitscheck, Hans Dieter, 1997, ‘Horja and bius; “Pseudo-tribalism” in Toba Batak

society, North Sumatra’, in: Sri Kuhnt-Saptodewo, Volker Grabowsky, and Martin

Grossheim (eds), Nationalism and cultural revival in Southeast Asia; Perspectives from

the centre and the region, pp. 191-9. Wiesbaden: Harrassowitz.

Latham, Ronald (ed.), 1978, The travels of Marco Polo. Middlesex: Penguin.

Leach, E.R., 1954, Political systems of highland Burma. Cambridge: Harvard University


Leong, Sau Heng, 1990, ‘Collecting centres, feeder points and entrepots in the Malay

Peninsula, c. 1000 B.C. – A.D. 1400′, in: J. Kathirithamby-Wells and John Villiers

(eds), The Southeast Asian port and polity; Rise and demise, pp. 17-38. Singapore: Singapore

University Press.

Lombard, Denys, 1967, Le sultanat d’Atjeh au temps d’Iskandar Muda, 1607-1636. Paris:

Ecole Frangaise d’Extreme-Orient.

Lubis, Mhd. Arbain, 1993, Sejarah marga-marga asli ditanah Mandailing. Medan: Departemen

Pendidikan dan Kebudayaan.

Marsden, William, 1966, The history of Sumatra. Kuala Lumpur: Oxford University

Press. [Oxford in Asia Historical Reprints. First impression 1783.]

Meerwaldt, J.H., 1899, ‘Aanteekeningen betreffende de Bataklanden’, Tijdschrift voor

Indische Taal-, Land- en Volkenkunde 37:513-50.

Miksic, John, 1979, Archaeology, trade and society in Northeast Sumatra. [PhD thesis,

Cornell University.]

-, 1996, ‘Archaeology, ceramics, and coins’, Journal of the Economic and Social History

of the Orient 39-3:287-97.

-, 1998, ‘The Cola attacks’, The Encyclopedia of Malaysia, Early history 4, pp. 120-1. Singapore:

Didier Millet.

-, 2000, ‘Heterogenetic cities in premodern Southeast Asia’, World Archaeology 32-1:


Milner, A.C., 1982, Kerajaan; Malay political culture on the eve of colonial rule. Tucson:

University of Arizona Press.

Milner, A.C., E. Edwards McKinnon, and Tengku Luckman Sinar, 1978, ‘A note on Aru

and Kota Cina’, Indonesia 26:1-42.

Neumann, J.B., 1885/86, ‘Het Pane- en Bila-stroomgebied op het eiland Sumatra

(studien over Batahs en Batahsche landen)’, Tijdschrift van het Nederlandsch Aardrijkskundig

Genootschap 2-2:1-134, 3-1:1-99.

Neumann, J.H., 1910, ‘De Bataksche goeroe’, Mededeelingen vanwege het Nederlandsch

Zendelinggenootschap 54:1-18.

-, 1926, ‘Bijdrage tot de geschiedenis der Karo-Batakstammen’, Bijdragen tot de Taal-,

Land- en Volkenkunde 82:1-36.

408 Leonard Y. Andaya

Nieuwenhuys, R. (ed.), 1962, Herman Neubronner van der Tuuk; De pen in gal gedoopt;

Een keuze uit brieven en documenten. Amsterdam: Van Oorschot.

Nik Hassan Shuhaimi b. Nik Abd. Rahman, 1984, Art, archaeology and the early kingdoms

in the Malay Peninsula and Sumatra, c. 400-1400 A.D. [PhD thesis, University

of London.]

Nilakanta Sastri, K.A., 1949, ‘Takuapa and its Tamil inscription’, Journal of the Malayan

Branch of the Royal Asiatic Society 22:25-30.

O’Connor, Richard A., 1995, ‘Agricultural change and ethnic succession in Southeast

Asian states; A case for regional anthropology’, Journal of Asian Studies 54-4:968-96.

Parkin, Harry, 1978, Batak fruit of Hindu thought. Madras: Christian Literature Society.

Pedersen, Paul B., 1967, Religion as the basis of social change among the Bataks of North

Sumatra. [PhD thesis, Claremont College.]

Perret, Daniel, 1995, La formation d’un paysage ethnique; Batak et Malais de Sumatra Nord-

Est. Paris: Ecole Francaise d’Extreme-Orient.

Pleyte, CM., 1903, ‘Singa Mangaradja, de heilige koning der Bataks’, Bijdragen tot de

Taal-, Land- en Volkenkunde 55:1-45.

Pollock, Sheldon, 1998, ‘The cosmopolitan vernacular’, Journal of Asian Studies 57-1:


Pringle, Robert, 1970, Rajahs and rebels; The Ibans of Sarawak under Brooke rule, 1841-

1941. Ithaca: Cornell University Press.

Ptak, Roderich, 1998, ‘Possible Chinese references to the Barus area (Tang to Ming)’,

in: Claude Guillot (ed.), Histoire de Barus, pp. 119-43. Paris: Association Archipel.

[Cahier d’Archipel 30.]

Purba, O.H.S., and Elvis R Purba, 1997, Migrasi spontan Batak Toba (marserak); Sebab,

motip dan akibat perpindahan penduduk dari dataran tinggi Toba. Medan: MONORA.

Rae, Simon, 1994, Breath becomes the wind; Old and new in Karo religion. Dunedin: University

of Otago Press.

Raffles, Lady Sophia, 1991, Memoir of the life and public services of Sir Thomas Stamford

Raffles. Singapore: Oxford University Press. [First impression 1830.]

Sangti, Batara, 1977, Sejarah Batak. Balige: Karl Sianipar.

Satyawati Suleiman, 1977, ‘The archaeology and history of west Sumatra’, Bulletin of

the Research Centre of Archaeology of Indonesia 12:1-25.

Schnitger, F.M., 1964, Forgotten kingdoms in Sumatra. Leiden: Brill.

Sherman, D. George, 1990, Rice, rupees, and ritual; Economy and society among the Samosir

Batak of Sumatra. Stanford: Stanford University Press.

Shore, Bradd, 1989, ‘Mana and tapu’, in: Alan Howard and Robert Borofsky (eds), Developments

in Polynesian ethnology, pp. 137-73. Honolulu: University of Hawaii Press.

Siahaan, E.K., et al, 1976/7, Monografi kebudayaan Tapanuli Utara. Medan: n.n.

Siahaan, N., 1964, Sedjarah kebudajaan Batak. Medan: Napitupulu.

Sibeth, Achim, 1991, The Batak. London: Thames and Hudson.

Sinaga, Richard, 1996, Leluhur marga-marga Batak dalam sejarah, silsilah dan legenda.

Jakarta: Dian Utama.

Singarimbun, Masri, 1975, Kinship, descent and alliance among the Karo Batak. Berkeley:

University of California Press.

Situmorang, Sitor, 1987, ‘The position of the Sisingamangarajas from Bakkara in

relation to the three main marga-groups: Borbor, Lontung, and Sumba’, in: Rainer

Carle (ed.), Cultures and societies of North Sumatra, pp. 221-33. Berlin: Reimer.

-, 1993, Toba Na Sae. Jakarta: Sinar Harapan.

The Trans-Sumatra Trade and the Ethnicization of the ‘Batak’ 409

Smith, Anthony D., 1986, The ethnic origins of nations. Oxford: Blackwell.

Smith, Anthony D., and John Hutchinson (eds), 1996, Ethnicity. Oxford: Oxford University


Soo, Kee-long, 1998, ‘Dissolving hegemony or changing trade pattern? Images of

Srivijaya in the Chinese sources of the twelfth and thirteenth centuries’, Journal of

Southeast Asian Studies 29-2:295-308.

Stephan, Nouha, 1998, ‘Le camphre dans les sources arabes et persanes; Production

et usages’, in: Claude Guillot (ed.), Histoire de Barus, pp. 224-41. Paris: Association

Archipel. [Cahier d’Archipel 30.]

Subbarayalu, Y., 1998, ‘The Tamil merchant-guild inscription at Barus; A rediscovery’,

in: Claude Guillot (ed.), Histoire de Barus, pp. 25-33. Paris: Association Archipel.

[Cahier d’Archipel 30.]

Tarigan, H. Guntur, 1972, ‘The structure and organization of Simalungun society’,

Sumatra Research Bulletin 2-1:42-58.

Teeuw, A., 1959, ‘The history of the Malay language’, Bijdragen tot de Taal-, Land- en

Volkenkunde 115-2:138-56.

Tengku Luckman Sinar, 1986, Sari sejarah Serdang. Jakarta: Departemen Pendidikan

dan Kebudayaan.

Tideman, }., 1936, Hindoe-invloed in noordelijk Batakland. Amsterdam: De Valk.

Tobing, Adniel L., 1967, Si Singamangaradja I-XII. Medan: Marpaung. [Sixth edition.]

Vergouwen, J.C., 1964, Social organization and customary law of the Toba-Batak of Northern

Sumatra. The Hague: Nijhoff. [Translation of Dutch edition of 1933.]

Voorhoeve, Petrus, 1927, Overzicht van de volksverhalen der Bataks. [PhD thesis, Leiden


-, 1973, ‘Some remarks on Karo-Batak literature’, Sumatra Research Bulletin 3-1:39-42.

-, 1979/80, ‘Elio Modigliani’s Batak books’, Archivio per VAntropologia e la Etnologia


Warneck, Johannes, 1906, Toba-Batak-Deutsches Wbrterbuch. Batavia: Landsdrukkerij.

Westenberg, C.J., 1892, ‘Aanteekeningen omtrent de godsdienstige begrippen der

Karo-Batak’, Bijdragen tot de Taal-, Land- en Volkenkunde 41:218-53.

-, 1905, ‘Bataksche rijkjes Dolok en Poerba’, Tijdschrift van het Aardrijkskundig Genootschap


Wilier, T., 1846, ‘Verzameling der Battahsche wetten en instellingen in Mandheling en

Pertibi; Gevolg van een overzigt van land en volk in die streken’, Tijdschrift van

Nederlandsch-Indie 8-2:145-424.

Wolters, O.W., 1969, Early Indonesian commerce. Ithaca: Cornell University Press.

Ypes, W.K.H., 1932, Bijdrage tot de kenis van de stamverwantschap, de inheemsche rechtsgemeenschappen

en het grondrecht der Toba- en Dairibataks. Leiden: Adatrechtstichting.

-, 1944, ‘De Bataks en de tempelbouwvallen in Padang Lawas in Mandailing’, Cultureel

Indie 6:135-43.

Zeijlstra, H.H., 1913, ‘Boschproducten’, in: K.W. van Gorkom (ed.), Oost-Indische cultures,

Vol. 3, pp. 791-874. Amsterdam: De Bussy.


1 Comment

  1. Hesperonesia said,


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: