The international boundary between India and Bangladesh came into being in 1947, but some of its segments have much older histories. The oldest segment lies below the mountains of Meghalaya and forms the northern border of the Sunamganj Zila of Bangladesh. This boundary runs east and west, cutting across many short rivers, whose names elude most maps, one being the Dhamalia river, which falls from Pandua, in India, and empties into the Surma River, near the town of Sunamganj, in Bangladesh. Borderlands of Mughal Bengal had once spanned the basins of the Dhamalia and other parallel rivers draining the mountains into plains below, but a definite geographical divide emerged in 1790, in the Sylhet district of British Bengal, in the form of a boundary line that served explicitly to restrict and regulate mobility between two political territories, defined as the homelands of two distinct cultures in the mountains and plains, respectively.
The rationale for inventing this boundary was an early precursor of the “two nation theory,” which eventually informed the partition of British India. At the same time, the birth of this boundary indicates that international borders are not homogenous, despite their appearance on maps as continuous lines. In addition, the local history of this boundary evokes many others in the old borderlands of mountains and plains spanning India’s northeast and Bangladesh, where state borders today have meanings quite distinct from the meanings enshrined in international law and in national sentiments.
Once an open terrain
National boundaries are now sacrosanct symbols of sovereignty, but people who move across them routinely experience these same boundaries as mere obstacles to mobility. This experience reflects a much older reality than national maps, because for most of human history, states had little power to regulate mobility across borders. For many centuries, social and cultural boundaries marked the supremacy of specific groups in particular places, without imposing restrictions on geographical mobility. Pre-modern territorial boundaries resembled island shores or edges of forest clearings more than gated city walls.
Inscriptions record the first boundaries in the basin of the Surma and Kushiara rivers, which flow through the Sylhet region (which now includes Sylhet, Sunamganj, Maulvi Bazar, and Habiganj zilas in Bangladesh), after they emerge from the Barak river in Cachar, in present-day Assam. In the first millennium CE, Kamarupa kings granted land to brahmins around the Surma and Kushiara, in places then called Srihatta and Khanda Kamarupa. These names indicated domains of royal patronage for Hindu elites whose religious rituals marked their local boundaries. Even today, in Cachar, rituals around the temple of Kapiliswar mark such boundaries. Here, high-caste Hindus worship Siva as Kapiliswar, while others venerate the site but not the deity. This ritual boundary suggests that an older settlement of Khasis had been incorporated by immigrant settlers for whom Hindu territory emerged as a physical space controlled by brahmins, their patrons, and their subordinates. Hindu boundaries thus emerged as frontiers, limits and edges of Hindu ritual and social order.
Over many centuries, numerous groups drew boundaries in similar ways in and around the Surma and Kushiara, forming disparate territories of social order. Because populations were small and land abundant, people had ample space to construct new territories, each with its own boundaries, its own elites and systems of subordination. A patchwork of territories had emerged by 1303, when Shah Jalal conquered local rajas and established Islam in Sylhet, creating a new Muslim cultural boundary. When the traveller Ibn Batuta met Shah Jalal, in 1346, the Sylhet landscape held diverse territories of Khasis, Garos, Hindus, Muslims and others.
Taming the wild
In 1612, Mughal armies created the first system of state authority in the region called the Sylhet Sarkar. This was the northeastern frontier of Bangla Suba (province), with boundaries marked by the power of a Mughal commander (faujdar) and by state rituals in which people paid homage and taxes to Mughal emperors. Thus, the Mughals created a boundary that was new for the people in this area.
Mughal authority also accentuated an existing boundary between settled farming communities and outsiders living in the jungle. From ancient times, ‘civilised’ groups gave ‘jungle people’ derogatory names to denote a wild nature. Such labelling valorised the subordination and expulsion of jungle people, but at the same time marked their autonomy within their own territories. Permanent farms, villages, towns, and cities became landmarks of civilisation, around which, jungles marked the frontiers of expansive agrarian societies. States advanced their supremacy by promoting agriculture, clearing jungle, and conquering, assimilating, and expelling jungle people.
The Mughals pursued this mission extensively, because farms, not jungle, provided state revenue, and farmers, not jungle people, bowed to Mughal authority. Boundaries of state authority fell at the jungle edge. Clearing jungles to make farms became a quintessential imperial project. In the 18th century, Nawab Murshid Quli Khan and his successors accelerated progress in northeastern Bengal by granting large tracts of jungle to men who would clear forest to make farms. Local farmers thus became pioneers pushing farms and state power together into the jungle, where jungle people often fought back, making agricultural expansion a violent process that progressed most rapidly where state power was most concentrated.
Because land was abundant, however, people on the margins of state power could move away to remain independent. Like many groups today designated “tribes”, Khasi people (also called “Khasia” in Bangla) lived in such spaces of mobility, and from ancient times, had scattered among river basins to engage in shifting rice cultivation. Ancient Khasis pioneered rice farming in Vietnam’s Red River delta; and when conquered there by Vietnamese, had moved up the Red River, into Yunnan (China), across Burma into Assam, Bengal and the Ganga river basin. Also in ancient times, Gangetic agrarian societies and states began expanding eastward, and each expansionist wave forced ‘jungle people’ such as the Khasis to submit, fight, assimilate, and move.
Bengali societies evolved on the eastern frontiers of Gangetic expansion, in landscapes inhabited by numerous non-Bengali peoples, who hunted, farmed, and fished in the jungles without settling down permanently. The people living in jungle habitats had distinctive languages, religions, and social practices, including matrilineal kinship, which marked them as primitive aliens for agrarian folk who invested in permanent cultivation, under state authority. Over the centuries, as agrarian states expanded across the lowlands, many Khasis and other jungle inhabitants moved up into forest highlands and mountains, where they formed independent domains.
Empire in cowry country
When the English East India Company took Bengal from the nawabs, in 1757, the old northeastern frontiers of Mughal Bengal posed many problems. Khasis held most land north of the Surma and ruled mountains above. Jaintia Khasi rajas held mountains and lowlands north and east of Sylhet town. Cachar rajas held the lower Barak valley. Tripura rajas ruled southern uplands and adjacent plains. In the lowlands, the English increased taxation as much as they could, but Sylhet district remained poor revenue territory, covered with forest.
The old Mughal northeastern frontier also posed a peculiar monetary problem, because cowry shells were the only coin. For centuries, these tiny shells from the Maldives were the cheapest coin all around the Indian Ocean region, and Bengal was a famous cowry market. But in Bengal, only the northeast had no metallic coins in its markets, only cowries. People here imported almost nothing from downstream, except cowries, which merchants brought upriver on boats that returned downstream with rice, fish and mountain products. By 1780, limestone was the most important mountain product, and it came only from quarries in the high northern mountains, in what is today Meghalaya. Khasi rajas around Pandua controlled quarries from which limestone came down the Dhamalia on boats to Sunamganj, a market town on the Surma built in the 18th century and named for its limestone trade.
Cowry currency thus described a monetary boundary. The Mughals and Nawabs would spend all their Sylhet revenues inside this boundary, but the English needed to convert cowries into rupees to serve wider imperial ambitions. In 1780, the English Collector in Sylhet sought to accomplish this conversion, without exporting cowries, by using the East India Company’s cowry revenue to finance his private business ventures. He used state tax revenues to buy limestone in Pandua, which he sold downstream for rupees that he then used to remit the Sylhet revenues to the Company treasury in Calcutta. He focused on limestone, shipped from Pandua to Sunamganj, and then downriver to Dhaka and Calcutta.
In the 1780s, mountain trades boomed in mountains and lowlands, between Pandua to Sunamganj, where, at the time, most land, in the Collector’s words, was “covered with an impenetrable jungle and so infested by elephants, tigers, and other wild beasts that … clearing and cultivation [was] attended with great difficulty and expense”. Elephants provided most income for the few forest zamindars, who paid the English revenue. Khasi rajas in the mountains exercised sporadic authority in the jungles below, and the English could no more conquer the mountain Khasis than could the Mughals, for, as the Collector said, “you might as well attack the inhabitants of the moon.”
In the 1780s, natural calamity made the jungles between Pandua and Sunamganj more attractive for lowland farmers and investors. Massive floods in 1784 and 1787 disrupted life drastically in the plains and mountains, and famines ensued. More investors focused their attention on land above the flood line, covered with jungle, even as merchant activity expanded along routes to the mountains. Pressed for funds to finance imperial wars, the English became more willing to use force to increase tax revenues. As a result, a motley, violent and chaotic mixture of British imperialism and Bengali enterprise invaded the land north of the Surma, where prime land for new farms lay in forests filled with people who never submitted to the Mughals, nawabs or British.
Conflict in the borderlands
In the 1780s, various kinds of boundaries defined localities north of the Surma. Ethnic boundaries surrounded Khasi settlements. State boundaries marked land where people paid taxes. Social boundaries enclosed farming communities. All these mingled inside the old jagir of Omaid Rezah, where generations of Khasis and Bengalis had formed mixed settlements of people called Bengali Khasis, who lived in forests, owned farmland, and traded and married among mountains Khasis as well as among Bengali communities.
North of the Surma, Omaid Rezah’s authority declined rapidly in the 1780s. The English demanded more tax than he could pay and then divided his jagir among his heirs and creditors. Merchant power also increased in his old jagir. New farmers and investors moved into the forest. The English gave and took away land rights according to people’s ability to pay taxes. In this context, serious conflict ensued.
Tension erupted first around Pandua, where the English maintained a small force to protect merchants. In 1783, Khasi mountain warriors seized Pandua and the passes around. Sporadic warfare continued for seven years, between Company armies and Khasi rajas around Pandua. In 1788, conflict began in lowland forest villages north of Sunamganj, during flood-induced famines. In early 1789, two lowland Bengali Khasi warrior rajas, Ganga Singh and Aboo Singh, captured numerous villages and controlled several river routes. In the summer, rebel Khasis and Bengali Khasis controlled “137 Bengalee villages”, Ganga Singh escaped to the hills, and Aboo Singh attacked the Pandua fort, killing its commander.
The English then launched a war on two fronts: in mountains around Pandua and in jungles behind Sunamganj. By early 1790, Company troops had conquered most people below the mountains, and open warfare ended soon after a British commander ordered the massacre of Bengali Khasis around Ganga Singh’s home village. But by then, the Company had lost Pandua irretrievably to mountain Khasi rajas.
In November 1791, the 35th Sepoy Battalion left Sylhet, its mission only half-accomplished. The new state boundary drawn between British Bengal and mountain Khasi domains became a reality based on Khasi victories in the mountains and British victories below. The new border ran along the base of the mountains and bisected the route from Sunamganj to Pandua. It marked the northern limit of British Bengal, which only then extended indisputably to the mountains, and equally indisputably, did not include Pandua.
A new kind of boundary
Marking boundaries firmly was not normal imperial practice at the time, so it required extensive justification, recorded in official correspondence. First and foremost, the boundary secured Company territory against threats to British authority posed by unregulated mobility between mountains and lowlands. The English drew this boundary to restrict mobility by defining the northern mountains as independent Khasi territory. Henceforth, mountain Khasis officially became aliens in the plains, where all the land became Bengali territory.
As per the official culture of the Company’s governance, the new boundary separated the “races” of Khasis and Bengalis. It imposed restrictions on “intercourse and intermarriages” that produced what one official called the “degenerate Race called Bengalee Cosseahs”. As one collector explained, problems addressed by the boundary did not arise from the inherent character of the Khasi ‘race’, but rather from Khasia and Bengali miscegenation; hence, from interracial alliances that threatened the Company’s territorial order by mixing lowland popular culture with the wild, unruly culture of the mountains, where people did not respect state authority.
The new state border also became a boundary of ‘free trade’. Collectors prohibited European merchants from operating inside Khasi territory; and they also prohibited the northern mountain Khasis from trading in Sylhet, while non-Khasi merchants from Jaintia and Cachar were “considered as quiet and inoffensive”, and thus continued to be allowed free access to markets in Company territory, contingent on good behaviour.
In addition, the border defined eligibility to own landed property in ethnic terms. The English prohibited Khasis from owning land on the grounds that their land had been acquired illegally, and on the assumption that Khasis could never abandon mountain methods for establishing property rights, deemed alien in the plains. Collectors then expropriated the lowland Khasi landed property rights, already established under Company law, to make way for Bengalis. The land market thrived as land buyers acquired state-defined land rights to support the laborious process of clearing jungle and creating new farms.
The impact of the new border thus fell most heavily on Khasis below the mountains. Before 1790, shifting cultivation, permanent farms, hunting, trading and fighting were carried out in the open borderlands spanning mountains and plains, in mixed Khasia and Bengali environs. After 1790, these mixes faded away under the impact of British authority and new waves of Bengali colonisation. Government auctions of Khasi land began in 1792, and a year later, the collector reported that he had completed sales of all land expropriated from rebellious Khasi and Bengali villagers, except a “trifling remainder of Cosseah land.”
Clearing jungles to make farms took a long time, however. High taxes slowed the process and encouraged the new colonists to establish estates that were, so that the landowners could get enough land cleared quickly by tenants to pay taxes and reap some profit. Extensive forest zamindar estates remained covered with forests. Expanses of open jungle also remained outside the reach of private property, leaving Khasis some room to manoeuvre. New small estates multiplied quickly, however. By 1797, land formerly owned by Bengali Khasis had been occupied by Bengalis “very willing and eager to enter into regular engagements to pay revenue” to secure their property rights. The Collector justified his official erasure of Khasi rights to this land by saying that Khasis had held it by force and not only failed to bring it into cultivation but had “scared away farmers”, until the Company “secured the area and confined Cosseahs to the hills”.
The old Mughal northeastern frontiers witnessed a rapid increase in privately owned landed property. In 1789, 4000 Sylhet taxpayers had owned land under Company law; in 1795, the number was 26,000, and a year later, it rose to 27,000. Landed estates were mostly very small, and they were so numerous that by 1800, about 25 percent of the population of Sylhet district may have lived in landowning families, in villages that averaged a mere 67 people and four landed estates (taluks), each representing about 18 people, few more than one landowning family, its dependents and servants.
The 1790 northern boundary of the Bengal Presidency, north of the Surma River, gave an old Bengali boundary a new geographical form. Inside the boundaries of Bengal, the state defined local elites, who owned bounded plots of landed property. The state’s external boundary became a part of everyday life in villages, where social boundaries marked the Khasis’ subordinate, outsider status.
Thus, local histories impart to the former imperial boundary that now separates Bangladesh and India meanings quite distinct from those that emerged after 1947. This boundary defined Bengalis and Khasias as peoples with separate histories, homelands, and cultural identities, which mingle in the local history of the borderlands. Here, each defines the other, and the memory of Bengali Khasis north of the Surma indicates a distinctive borderland cultural past outside the reach of the national imagination.
Note: An extensively documented version of this essay is forthcoming in the Journal of the Asiatic Society of Bangladesh.