Navigation – Plan du site

AccueilNouvelle série49Dialogues posthumes…The power of movement

Dialogues posthumes…

The power of movement

Philippe Sagants contribution to the history of migration in the Eastern Himalaya
La puissance du mouvement : contribution de Philippe Sagant à l’histoire des migrations dans l’est de l’Himalaya
Catherine Warner


Les recherches de Philippe Sagant ont mis en évidence l’importance de la mobilité dans la construction des sociétés et des cultures himalayennes. Montrant que les migrations ont façonné les relations entre les villages à travers le temps et l'espace, ses travaux ont historisé les communautés de l'Himalaya oriental, en faisant des acteurs des nouvelles modernités mondiales. Cet article s’appuie sur les travaux de Sagant pour offrir un aperçu des migrations dans l’est du Népal et à Darjeeling à la fin du xixe et au début du xxe siècle.

Haut de page

Entrées d’index

Géographique :

Himalaya oriental
Haut de page

Texte intégral

  • 1 Ethnographers of the eastern Himalaya included Caplan, 1970; Gorer, [1939] 1984; Jones and Jones, 1 (...)

1The Dozing Shaman, Philippe Sagant’s compilation of research articles in English translation, should be read as a foundational text for understanding migration history in the eastern Himalaya. The articles offer both nuanced observations of multi-generational migrations stretching back to the late nineteenth century as well as a methodology for exploring the spatialization of social and political relationships that created and sustained migration networks. Sagant embarked upon fieldwork during a period of academic professionalization in Himalayan ethnography. Following the increased access given to western fieldworkers in Nepal with the restoration of the Shah monarchy from 1951, scholars were animated by an urgent sense that an old world was dying which might yet be captured through participant observation. From roughly the 1930s to the 1960s, the last generation of colonial and amateur ethnographers and the first generation of Western-educated academics confronted the effects of nationalist sentiments, the world wars, and decolonization in the seemingly peripheral, old imperial “contact zone” of the eastern Himalaya.1 Attempting to capture the rapid social and political changes they observed, many scholars leaned heavily upon modernization theory. Sagant, however, used histories of mobility to place Limbus as actors within local, regional, and global histories built upon multiple continuities and ruptures since the late eighteenth century.

2Sagant’s insistence that mobility should be used to understand “l’évolution de la société” (1978: 93) informs many of the articles in The Dozing Shaman. During a period of several years in the late 1970s, in fact, Sagant placed migration at the center of his work, from which only one article, ‘Quand le Gurkha revient de guerre…’ was translated into English. It is in his article, ‘Ampleur et profondeur historique des migrations népalaises,’ however, that Sagant most clearly presented the implications of an ethnological and historical synthesis around histories of mobility. Here Sagant compared ethnologists’ observations of migrations in their fieldwork localities and their suppositions about the historical roots of such migrations with the work of historians, particularly M. C. Regmi. He further analyzed British Indian and Nepali census data as well as his own fieldwork notes to suggest preliminary conclusions and, most importantly, questions that remain salient for the study of Himalayan migration.

3By foregrounding relations of production, local encounters with the state and hybrid legal regimes, Sagant offered a methodology for drawing contextualized typologies of migration. For example, he categorized ‘temporary migration’ in relation to ‘seasonal migration’ and ‘emigration,’ not by an arbitrary length of time, but rather by the migrant’s break from village agrarian cycles. In villages of eastern Nepal, a temporary migrant would stay away long enough that his land was cultivated by someone else, his social obligations fulfilled by someone else. In contrast, seasonal migration only entailed an absence during the slack season in the agrarian cycle, and, moreover, was often organized through sardars or labor contractors rooted within the local systems of production. Finally, according to Sagant, the case of permanent migration produced ambiguity in social relations at the level of the village. Those who stayed behind did not know whether the migrant would return, and how he or she would fit into local political and cultural economies of affect, land, and labor in the case of return. The annual seasonal cycles of migration, an enduring feature of histories of mobility across South Asia, largely structured by the monsoon (Gommans, 1995; Kolff, 1990; Ramaswamy, 2016) interacted with historical change shaped by Gorkha and British expansion, state centralization, and the commercial integration of regional markets with global capitalist flows (Sagant, 1978). The juxtaposition of a regionally specific cyclical view of mobility with historical time further allowed Sagant to delve into the understandings of migration encapsulated in ritual and everyday life in eastern Nepal.

4Sagant’s framework could be applied to migration in South Asia more broadly, although the material Sagant employed in ‘Ampleur et profondeur historique des migrations népalaises’ focused on the movement of pahari groups of Nepalis (including Limbus) to, first, border areas, then further afield to mines and plantations, cantonments, and cities of northern and northeastern India from the late eighteenth century. In sum, his use of historical methods as an ethnographer allowed him a perspective on migration that anticipated by some years historians’ attempts to rethink binary views of mobility and migration, internal and cross-border movement, circulation and emigration, and forced and voluntary migration (Lucassen and Lucassen, 2014; Amrith, 2011). The following overview of Limbu migration from the eastern Himalaya draws upon Sagant’s work as evidence, methodology, theory, and history, a window onto stories of migrations and intergenerational struggles over place and identity. By examining Sagant’s notions of culturally specific forms of power and authority in Limbu society, historians of migration may find conceptual resources to fill in the gaps left by available archives which scarcely account for those who disappeared or evaded the gaze of the state.

Conflict, labor, and migration: 1770 to 1840

5Despite the absence of census data, Sagant, M. C. Regmi, and other scholars have suggested that large-scale conflict-driven migrations resulted from the territorial expansion of Gorkha and the East India Company and related political turmoil from about 1769 to 1814. In the Mewa valley and (present-day) Taplejung district, where Sagant conducted research on Limbu society, political authority was contested between multiple actors including Sikkim, Gorkha, and Limbu headmen, and Company authorities. From the seventeenth century, the Limbus of the eastern Himalayas were part of the Sen kingdom of Vijaypur, with Limbu notables acting as chief ministers. Internal discord between the Vijaypur ruling family and the chief ministers had weakened the polity with a number of notables seeking exile and support in Bengal under the East India Company and in Tibet. The flight of regional elites allowed Gorkha to push eastward into the region (Hamilton, 1819: 133–134). The Gorkha offensive bogged down for several years near the Arun River, encountering fierce attacks by Limbu forces. Many Kirantis (people of eastern Nepal’s mountains who did not adhere to a caste-based social system) sought refuge in Sikkim (established as a kingdom in the mid-seventeenth century), and its adjoining plains.

  • 2 India Office Library. Hodgson Collection. MSS EUR. D. 537. “Free Translation of the Limbu History” (...)

6A brief history written in Limbu by Jobhansingh Phago Limbu in 1843, and translated by Iman Singh Chemjong over a century later, detailed the intensity and success of guerrilla resistance in Limbuan and lamented the divisive and dishonorable tactics the Gorkha army used to capture the region.2 A negotiated settlement between many Limbu chiefs (especially those located around Chainpur) and Gorkha in 1773 ended the war. Some Limbus joined Nepal’s expansive royal army, even traveling west to secure the independent polities of Lamjung and Tanahun, for which they were awarded land grants or jagirs. Apparently, Gorkha commanders capitalized on the fearsome and barbaric reputation of the easterners to scare locals into abandoning their farmsteads in (present-day) West Nepal. In contrast, the Limbu chiefs of Yangwarak (part of present-day Taplejung district) allied with Sikkim to continue hostilities against Gorkha and their Limbu allies of Chainpur. Unable to provision enough troops to recapture and hold Chainpur, the rebel forces retreated and Gorkha followed, pushing eastward into Ilam. From this period, the Limbus were divided between those who had sided with Gorkha and those who had sided with Sikkim. The Limbu villages that Sagant studied fell within the latter division, and were known as areas of historically entrenched resistance to Gorkha rule.

7Fighting between Sikkimese and Gorkha forces lasted intermittently for decades, with Limbus ranged on both sides—the conflict stretched from Bhojpur, Nepal to Tumlong, Sikkim. At the urging of Chinese and Tibetan representatives, some Limbus and Lepchas also actively opposed Nepal during the Nepal-China War from 1791 to 1793. Both sides treated political prisoners and local villagers ruthlessly. Many thousands fled the region to India, Sikkim, Bhutan, and Assam, or in the opposite direction to Limbuan (claimed by Nepal). In response to population movements, the Gorkha state alternated between confiscating lands of fugitives and issuing proclamations declaring amnesty and confirmation of land rights for others (Regmi, 1965: 87–88). As the Gorkha treaty of 1774 explicated in declaring its right to govern the Limbus, seeking sharan or refuge with a political power for protection during periods of conflict entailed longer-lasting obligations of political dependence (ibid.: 151). Part of the easternmost territory captured by Nepal in this period was later returned to Sikkim after the Anglo-Nepal war of 1814 to 1816, in recognition of the effective resistance of local militias, comprised of many Lepchas and Kirantis, during the war (Vajracharya and Shrestha, 1978). Thus, although political refuge, sharan, was understood as producing permanent bonds of political dependence, the reality in frontier regions worked against such notions.

  • 3 Gorer notes that the position of mandals was hereditary in the Dzongu region he studied and had bee (...)

8While it is clear that conflict shaped migration patterns during the initial period of Gorkha and the East India Company’s territorial aggrandizement, the practice of extending asylum to politically useful settlers was grounded in patron-client relations and patterns of land use in the region. In the late eighteenth century, mobile and diverse local societies bounded by more fixed states and political interests (Tibet, Sikkim, Bhutan, Mughal functionaries, the East India Company, Gorkha, the Sen principalities) often were loosely governed via intermediaries. The headmen (mandals in Sikkim, chaudharis in the Morang, perhaps the precursors of subbas in Limbuan) offered tax payments and labor services directly to states or powerful landlords recognized by states, such as kazis in Sikkim, in return for recognition of their positions, including rights to labor and mediation of disputes.3 Territorial expansion could also be carried out through intermediaries. For example, prior to the Anglo-Nepal War, a Gorkha subah (military commander), Jayanti Khatri, was stationed at Nagrikot (in present-day India), where he had offered settlers from the west, Majh Kirat, as well as Lepchas and other local cultivators, a tax-free period to use the land. With the retreat westward of Gorkha control due to the outbreak of war, the more recent settlers—from Majh Kirat and parts of Limbuan—fled to Ilam, Nepal’s easternmost army outpost. In contrast to the new settlers, indigenous people, such as the Lepchas, resisted Gorkha domination during the war of 1814 to 1816. As with the Limbus of Yangwarak since the late eighteenth century, the Lepcha resistance relied upon Sikkim’s support. The alliance apparently affirmed Sikkim’s rights to the Lepchas’ labor—Lepcha headmen working for Sikkim captured and relocated many of the shifting cultivators of the Mechi headwaters within its own territory after the war, although some eventually fled. Rights to labor and political control of a territory were closely interwoven, meaning that labor and conflict-driven migration were not easily distinguishable categories in this period.

  • 4 A glimpse of how patron-client networks functioned in the early nineteenth century is provided by t (...)
  • 5 India Office Library. Board’s Collections. F/4/1813. 1839–1840, 74755. Affairs of Nepaul. India Pol (...)

9Ties between headmen and subjects often weathered conflict-driven migrations. Many cultivators either followed their patrons into exile or continued to recognize and pay taxes to their patrons who had fled into exile.4 Influential refugees with the ability to maintain and attract dependents were highly prized in the unsettled politics of the late eighteenth to mid-nineteenth centuries. Nepal, for example, was eager to pull refugees from Sikkim even after the war of 1814 to 1816. For example, following the assassination of the Lepcha Prime Minister in Sikkim, his nephew Yukla Thup Kaji was given refuge in Ilam, Nepal, in 1826. The Kaji (or “minister”) brought his extended family along with various Lepcha and Limbu functionaries, headmen, and cultivators, numbering about 800 people. They were awarded a jagir with generous terms regarding taxes and labor services—Yukla Thup Kaji claimed that the jagir covered land historically held by his lineage. Yukla Thup Kaji was enjoined to attract and settle cultivators from Sikkim and Muglan (India) on available land in the region; it was further hoped that he would prove useful for extending Gorkha interests in Sikkim (Vajracharya and Shrestha, 1978: 40–41). Nepal’s political patronage, however, was undermined by the refugee Kaji’s contacts with the East India Company. Yukla Thup Kaji’s people had been encouraged to settle under Gorkha protection by the Company’s Agent for the Northeastern Frontier in order to ease tensions in Sikkim, and the refugees remained hopeful of settling around Darjeeling, which the Company annexed in 1835.5 In sum, the patron-client ties were maintained through shifting and temporary alliances with neighboring states.

  • 6 West Bengal State Archives. Proceedings of Lieutenant-Governor of Bengal. Political Department. Bri (...)
  • 7 The Company invaded Sikkim in 1860–1861, again partly in response to conflict over subjects as well (...)

10By the mid-nineteenth century, the relationship between headmen and tax-paying subjects had become more dependent upon state sanction. In other words, previously non-state communities practicing mobile livelihoods (transhumance, shifting cultivation, hunting) had effectively controlled land and labor. In the nineteenth century, imperial intervention swayed the balance of power more decisively in favor of centralized and authoritarian states. Established in the late 1830s, the East India Company’s administration in Darjeeling tried to attract cultivators and laborers from neighboring states. Its success in doing so was critical to the founding of the hill station and later the tea plantation economy. Tension escalated between Sikkim and the East India Company administration over the flight of Sikkim’s subjects (often referred to as “slaves” by the Company) to the annexed territory. Sikkim’s ruler, Tsugphud Namgyal, considered the Company’s refusal to return his subjects impolitic and noted Sikkim had a mutual agreement with Bhutan for exchange of dependents.6 Disputes over subjects and access to Tibet via Sikkim led to further conflict and the Company’s annexation of Sikkim’s tarai in 1850, giving it direct access to Darjeeling from Bengal.7 Historian Saul Mullard has argued persuasively that subjects from Sikkim fled to Darjeeling in significant numbers (as much as one-fourth of the total population by 1865) to escape increasingly oppressive terms of labor, taxation, and regulation on large landed estates. As Sikkimese subjects emigrated, Nepali migrants (especially from the eastern hills) took their places (Mullard, 2017: 22–23).

  • 8 Superintendent Campbell established wage rates based on colonial ethnic categories for services suc (...)

11During the same period, the first Superintendent of Darjeeling, Archibald Campbell, managed to attract laborers from Gorkha and Sikkim by exploiting pre-existing patron-client relationships as well as working through new intermediaries to supply laborers.8 As territorial boundaries were fixed, flows of laborers began to be described in ethnic terms. Thus, “Limbu” migration to Darjeeling can be dated to the earliest years of the hill station, even before the establishment and spread of tea plantations in the 1850s. The 1901 Census of Bengal recorded 23,200 self-identified Limbu speakers living in Bengal and 45,320 speakers of Kiranti languages, although these numbers must be considered a very rough estimate of rates of migration to the Darjeeling area. In both Sikkim and Darjeeling in the early nineteenth century, the monarchical and colonial states, respectively, relied heavily on patron-client relationships to move labor where needed, effectively taking some control out of the hands of community leaders while weakening the political salience of such relationships.

The Limbu household, sovereignty, and mobility from 1850

12The political-economic changes of the latter half of the nineteenth century ushered in a new world of mobility, the contours and implications of which slowly unfolded into the early decades of the twentieth century. This story has been described in terms of Himalayan state-making, the colonial appropriation and re-categorization of space, the extension of commercial capital to new regions, and the development of colonial ethnography as a tool of governance (Chatterjee, 2001; English, 1985; Sharma, 2011; Sivaramakrishnan, 1999). Sagant brings out the contours of this shift in the spatialization of power through the lens of changing Limbu “institutions,” which he understood through “[Limbu self] presentation and reality, myth and everyday life” ([1996] 2008: 3).

  • 9 This view contrasts with Caplan, 1970: 23–24; Regmi, 1965: 93–94.
  • 10 The administration preferred to collect rent in cash rather than labor, livestock, or grain as had (...)

13In Sagant’s ethnography, migration became increasingly woven into the political and economic context, as observed in the reach of the Nepali state into Limbu institutions via the subba’s prerogatives and the complexities of kipat (community-owned) land tenure. In contrast to other scholars, Sagant regarded kipat tenure as primarily political, with origins in neither pre-modern communal patterns nor lineage histories of first settlement and land clearance, but rather the agreement between the Gorkhas and Limbus of the 1770s.9 His observations about kipat tenure and political power in Limbuan described an arrangement that was only possible following the widespread sedentarization of a relatively more mobile population. Kipat tenure—a legal category created by the Gorkhas, associated with numerous non-caste based groups in the past, with taxation fixed on homesteads rather than the extent and quality of cultivated land, and including rights of access to pastures, forests, water, and minerals—would logically have developed among groups who combined agriculture with mobile livelihoods (Subba, 1999: 37–39; Regmi, 1965: 94; Forbes, 1996). From its initial foothold in the region, the Gorkha administration pushed to develop taxable (and thus fixed) cultivation, especially of irrigated paddy crops, while securing its interests in stabilizing a strategically volatile region bordering Sikkim, Tibet, and Bengal. Kipat lands were legally converted to raikar, individually owned property guaranteed by the Nepali state, and often vested in the hands of caste Hindu immigrants to Limbuan. Wastelands, often customarily used by the whole community, were privatized. Kipat was increasingly regulated and fixed titles and boundaries were increasingly demanded; kipat eroded far more quickly in the southern reaches of Limbuan closer to Gorkha centers of administration (Regmi, 1965: 97–98). As early as 1841, the Gorkha military administration of Ilam allowed and even encouraged tax-paying cultivators from the area to migrate seasonally to the new hill station of Darjeeling to earn wages in cash.10 Thus, while converting communally-held lands to more taxable forms of property, the Gorkha administration realized early on that in some cases out-migration of former communal rights-holders would be necessary to stabilize the emerging property regime and infuse cash into the local economy.

14Sagant’s work highlights the interconnections among new patterns of migration, new forms and extension of credit, and the changing role of intermediaries or headmen. In the Taplejung district of northern Limbuan, Sagant observed in detail the pressure for ready cash to which the intricate practice of mortgaging kipat land answered alongside the stratification of Limbu society that gave rise to the subbas’ power. The subbas relied upon and exploited their connections to the Nepali state, as well as their intermediary status in managing the kipat system, rather than achieving power through the community—for example, as strong heads of households with many dependents, and as leaders of Limbu councils, chumlung ([1996] 2008: 283–314). The subbas also presided over local courts, amal, with the legal backing of the government from the 1820s to 1960, where they mediated cases that arose according to religiously sanctioned social norms. The system appears to have combined Nepali legal norms with Limbu cultural and religious ideals. Punishments usually involved fines which the subbas kept for themselves. A subba’s subordinates (those who paid him taxes) were typically required to attend the annual Dasain celebration at the subba’s house to demonstrate continued allegiance to his authority and submit taxes (Forbes, 1999: 114–138). The headmen’s fines, often finely calibrated to a family’s earning potential, could lead to indebtedness. For example, Sagant recorded the case of a seemingly personal quarrel between a father and son during which the son struck the father. The subba charged a steep fine because the family was known to migrate yearly to Sikkim to carry tangerines and thus had access to cash (Sagant, [1996] 2008: 23–36). Men from a number of kipat-holding families migrated to India in order to try to earn money and redeem the mortgages their elders had taken, often in order to pay debts owed to the headmen for social infractions. However, migrations could increase rather than reduce indebtedness through mortgages which might be taken to finance migrations or support migrants who returned empty-handed and needed to purchase grain. The inalienability of the kipat holdings meant that the same piece of land could be mortgaged numerous times, with usufruct transferring to the mortgagor. Frequently, the original owner became a tenant on the land. Money and power continued to accrue to headmen and moneylenders (often but not always high caste immigrants or Limbu subbas), increasing social stratification and the rate of emigration (ibid.: 161–212).

  • 11 Gros and Schlemmer (2016) also stress the significance of this development in Sagant’s thinking as (...)

15Despite the growing oppressiveness of the subba’s office, it is the household that Sagant located as the enduring political center of Limbu society through his reading of ritual and notions of honor and shame. Careful to note that historical evidence about the household before Gorkha domination is limited, Sagant relied upon ritual and oral accounts to argue that public rituals were far less significant to Limbus than those centered on the domestic space. He suggested that the subba’s power mirrored that of the head of the household, rather than the other way around. The head of the household figured in ritual as a warrior and a hunter, whose ‘life force’ was raised through the actions, ritual and mundane, of the other members of his household and successful propitiation of Nahangma (Sagant, [1996] 2008: 9).11 In a later article, Sagant evaluated the relevance of Dasain as a ritual symbolizing the domination of the Nepali state over Limbu society. Nahangma, despite her importance to Limbu ritual, was conspicuously absent in the Dasain ritual, even though many other regional and lineage gods have been incorporated on the seventh day of the festival throughout Nepal. Rather, Yuma, another god, remembered as having quarreled with Nahangma, was incorporated into Dasain. While Durga lent legitimacy to the subba’s power, in Sagant’s estimation, Nahangma did the same for the head of the household, and their ritual worlds did not overlap. The household, in Sagant’s formulation, appears as an alternative source of sovereignty, with its own gods, its own terms of spatiality and regulation of labor. Household heads were distinguished primarily from dependents, or yogs, people who had fallen into their service through various kinds of ritual or monetary debts. Into the 1920s, some of the powerful households numbered as many as forty people, including lineage members and clients or yogs. Under the influence of Nepali law, throughout the twentieth century, households shrank to smaller conjugal units. Householders who wanted to undermine the subba’s authority absented themselves on the ninth day of Dasain. Successful resistance would elevate a new subba and fragment the geographical area under the former subba’s control (Sagant, 1996).

16Although households changed substantially throughout the twentieth century, oral histories concerning the migration and settlement of lineages in particular places remain relevant into the present for purposes of identity formation and ethnic politics (Balikci, 2008: 65–84; Shneiderman, 2015: 61–97). Such narratives often include some reflection of a transition from forest-based livelihoods to land clearance, farming, and more permanent settlement. Ian Fitzpatrick’s recent ethnography of a village in Taplejung district, for example, records the oral histories of the settlement of Mamangkhe village. Residents claimed the village was settled some 15 to 16 generations ago, and explained that previously the Limbu stayed in the forest (perhaps a way of accounting for migrant origins or a mobile past). The oral histories offered by Mamangkhe residents suggest a widely shared value placed on stone and wood building materials over bamboo and thatch, fixed and multi-generational location over shifting residences, and perhaps a sense that “civilization” comes from outside via high caste intermediaries (Fitzpatrick, 2011: 63). Such narratives, while not providing historical timelines or verifiable accounts of historical processes, serve to reinforce notions that privileged the spatially fixed household in relation to state and lineage in the eastern Himalaya.

  • 12 For Sikkim, Balikci provides an interesting discussion of how incorporation into a Lhopo lineage me (...)

17The lineage, rather than describing biological and patrilineal descent, however, could operate as a device to incorporate newcomers.12 Historian and activist Iman Singh Chemjong’s scholarship suggests a widely shared value placed on the incorporation of non-kin into Limbu society. In discussing the foundation of Limbu community in the eastern Himalaya, he highlighted the decision made by the legendary ten ancestral Limbu leaders (migrants from the direction of present-day Burma via the Morang) to admit all the remaining members of the royal families they had defeated as well as their subjects into the Limbu community, with the guarantee that all would be treated alike. Further, when the Limbus attacked their neighbors and captured prisoners, instead of making them into slaves, they were admonished to incorporate them into the dash Limbu vamsha, the ten Limbu lineages, erasing former group affiliations (1975: 8–11, 61–62). Lionel Caplan’s ethnography from Ilam district, observed in 1964–1965, indicated that incorporation into the lineage remained a possibility until quite recently. He found that a clan unique to the Indreni settlement, with no known members elsewhere, was considered to be the only lineage to have fully incorporated outsiders as kin within recent memory. Another lineage had also accepted outsiders as kin as recently as the late nineteenth century—when a court case over the inheritance of kipat lands was brought by an heir from the incorporated lineage, he was denied land, perhaps in part because he had been living in India for several decades. The scarcity of land in the Indreni settlement meant that Limbu men who lived in the village but could not trace their lineage to kipat-holding patrilines in the last several generations were denied kipat ownership. Approximately 13% of the Limbu population did not own kipat property (Caplan, 1970: 24–25, 44–47). The construction of lineage was clearly tied to the regulation of land and labor, or property and production.

  • 13 The agreement also cast the Nepali settlers as a threat to the identity of Sikkim, which was explic (...)

18The contrast with southern Sikkim in the nineteenth century is instructive. Saul Mullard’s research shows that the earliest waves of migrants from eastern Nepal had found a place due to the labor shortage caused by outmigration of Sikkimese subjects from landed estates. Whereas previously tax-paying subjects were identified in tax registers according to their lineage, a quarter of such lineages were deleted in the register of 1865 and new ones without local precedent were added. A handful of relatively less established Sikkimese landlords who could not have easily withstood a loss of subjects on their estates began to facilitate the immigration of Nepali settlers and, consequently, the extension of terraced agriculture. Sikkim’s aristocracy divided into pro- and anti-immigrant factions and, in 1876, a number of the latter signed an agreement with the king’s approval to ban new immigration and reform legal procedures on the estates in order to retain subjects. The agreement did not work in preventing Nepali immigration, but it highlighted the social and political ruptures created by the new patterns of migration. From the late nineteenth century, the bulk of migration could no longer be accommodated via extension of fictive kinship or even, perhaps, reworking of oral traditions to accommodate newcomers.13 Thus, in both eastern Nepal and Sikkim, the fluidity of lineage identity seems to have been inversely related to the availability of land and both lost a great deal of elasticity in the late nineteenth century.

The making of the Nepali migrant and the colonial economy into the twentieth century

  • 14 Every year about 200–300 Limbus of eastern Nepal were recruited. Prior to recruitment into the Indi (...)

19Sagant’s and Caplan’s case studies of land disputes in Taplejung and Ilam, respectively, offer evidence that throughout the twentieth century the tightening of credit markets around kipat lands paralleled the increased policing of lineage and kinship norms and boundaries. It is not surprising that the more vulnerable family members were the ones most likely to migrate. Sagant listed a number of situations that could lead to loss of face and social marginalization, such as failing to meet bride payments, marrying against intricate rules of exogamy, losing at factional politics, and committing adultery ([1996] 2008: 26–46). The weakening of the subbas’ powers in the Panchayat era (1960 to 1990) meant that orphans, boys born without the payment of ritual debts, children of bondsmen, and other customary dependents were able to escape to Assam and Sikkim, perhaps as sawyers or porters of tangerines (ibid.: 143). Fitzpatrick noted, with regard to far northeastern Taplejung in the 2000s, that Sikkim had long served as a destination for couples who violated the rules of lineal exogamy, although some eventually returned (2011: 70–71). Since the beginning of the twentieth century, recruitment into the Gurkha regiments in the Indian Army provided opportunities to sons without adequate inheritance in land or lineal prestige. World Wars One and Two spread the geographical reach of recruitment further throughout eastern Nepal and stepped up the rate of recruitment and length of service.14 Military recruitment, however, only comprised a small portion of all types of temporary migration, which included contract work as porters, sawyers, road diggers, stone crushers, construction laborers, and many other forms of manual labor (Sagant, 1978; Mazaudon and Michailovsky, 1977). Tighter boundaries of kinship and increased migration meant that belonging was reshaped spatially and affectively according to market constraints.

20Reintegration for returning migrants was challenging, even for Gurkhas who had earned a prestigious pension. Ex-soldiers distributed gifts to ease their way back into local society and appease the appropriate authorities (elders, moneylenders, and subbas), after which they had to shoulder the expenses of marriage (including bride price), set up a new house, and ideally redeem the mortgages on their families’ kipat. Court cases often resulted from the presence of multiple claimants. Reintegration within the village was so challenging that half of all servicemen Sagant surveyed, even those who attempted it, had to settle elsewhere, typically in the tarai, Assam, or their wives’ villages in Limbuan where they might become mortgagors of land like other migrants ([1996] 2008, 278–312). In contrast to former soldiers, women who were marginalized from the family were rarely allowed to return. Sagant’s discussion of childhood and marriage by abduction suggests that generational conflict concomitant with adolescents’ sexual independence previously was managed through seasonal mobility. It was socially expected that young people would live and work away from the control of elders while residing in temporary shelters to pasture livestock in upland areas or form work groups of their own to participate in seasonal agriculture. Especially in the case of young men, the expected period of adolescent mobility was replaced by new opportunities for longer distance migration in the twentieth century (Sagant, [1996] 2008: 50–89). As bride price and marriage by abduction became less common (and, in the latter case, illegal), women’s sexual and reproductive labor seems to have been subjected to greater surveillance and regulation (Jones and Jones, 1976: 101–120). The problem of reintegration for many migrants was their still marginal position vis-a-vis the same power structures that had forced them out in the first place.

21While migrants left eastern Nepal to improve their position vis-a-vis the credit market engulfing kipat tenures in Limbuan, migration itself brought them into other systems of debt, credit and obligation in British India. Throughout this period of rapid change, states (Nepal, Sikkim, and the British Government of India) attempted to restrict land ownership in mountain regions (including Limbuan, northern Sikkim, and the Government Estate in Kalimpong) to particular ethnic groups, placing pressure on these groups to police boundaries in new ways (O’Malley, [1907] 1985: 109, 144). At the same time, migrants from the hills, regardless of caste or ethnic background, were increasingly categorized simply as ‘Nepali’ in British India in the latter half of the nineteenth century. Nepalis, according to British colonial knowledge, were hard working cultivators who used the plough or laborers who readily took up waged work ranging from carpentry to road building to tea plantation. More rarely, administrators admitted that from the official perspective ‘Nepalis’ were not owed the same rights to land or social protection as other groups under colonial jurisdiction (Government of Bengal, 1873: 79). Nepal at mid-century under Prime Minister Jang Bahadur tended to discourage the emigration of military recruits, but was more favorable to the outmigration of laborers to Darjeeling. Nepal’s policy on migration, albeit inconsistent, reversed under Prime Minister Bir Shamsher (1885–1901) to favor migration for the military rather than for plantations, mines, and public works. Not all migrants to Darjeeling were landless laborers—some settled, with variable success, on Government Estates at Kurseong and Kalimpong.

  • 15 In 1901, the total plantation workforce was roughly 64,000, representing about 25% of the district (...)

22By the turn of the twentieth century, a system of sardari management and labor recruitment funded by Marwari moneylenders in Darjeeling had become entrenched on the tea plantations. The labor leader or sardar mediated between the creditor (who was not officially part of the plantation) and the laborers who initially took advances, sometimes to cover the costs of migration or other debts, which were later deducted from their pay. The office of sardar could be hereditary. For some Limbu migrants it may have resembled the office of the subba, but it was not based upon shared community or ethnic ties. Women made up a larger portion of the plantation labor force (about 60% in 1910) but received significantly lower pay than men. Their relative immobility on the plantations, as well as their isolation from kinship ties and other networks outside of the plantations, tended to fix them in place and cheapen their labor—they could not threaten to move on as male workers often did. By the early twentieth century, the ‘labour question’ became more pressing on plantations, with managers lamenting that many workers simply treated Darjeeling as a staging ground for onward migration to Bhutan, Sikkim, the tarai, Assam, and even Thailand, rather than a final destination. Finding the cost of living in Darjeeling too high, many migrants stayed on less than six months and either returned to Nepal or proceeded to a new destination. A number of plantation workers, especially ‘the male labour’ searched out higher wages for service in the town of Darjeeling, the railways, and the Public Works Department.15 From the early decades of the twentieth century, migrants increasingly extended their networks, searching for land to cultivate and more lucrative wage labor in frontier areas of South Asia and abroad. Migrants negotiated multiple forms of belonging, mediated by households, contractual labor leaders, and state bureaucracies, without modern citizenship fully replacing older, overlapping forms of political subjecthood.


* *

23This paper has discussed a prominent ethnographer’s work in terms of what it can reveal about migration history in the eastern Himalaya. Philippe Sagant interpreted change in Limbuan during an era of national integration in Nepal and increased transnational connections, in part, by emphasizing migration as a culturally specific performance and spatialization of power under intensified capitalist expansion. Even as Sagant sought to capture a diachronic view of a cultural region that was not well documented archivally, he did not retreat from shouldering the responsibility of placing the actors he found there historically. Rather than thinking about migration as something that occurs as a sign of an isolated community’s entanglements with modernity, Sagant recognized migration as an enduring phenomena that shaped communities without erasing ritual and local understandings of power. His work points to a methodology for migration history that moves beyond descriptive case studies or aggregations of incomplete statistics produced under the gaze of the state in South Asia. On one hand, Sagant’s articles occasionally present the Gorkha state as an oppressive, colonizing power which threatens to contaminate independent Limbu culture. On the other hand, Sagant’s meticulous care in working out how each observed cultural and material practice fits with or jars against oral narratives and self-representations, his attempts to reconcile disparate perspectives of a lived community, and his attention to how structures of inequality are lived and embodied spatially offer a perspective that seems particularly fitting to migration history.

24Recent research that conceptualizes traditions and processes of migration in the Himalaya has moved in the same direction. Various contributions to Huber and Blackburn’s volume, Origins and Migrations in the Extended Eastern Himalaya, emphasize the importance of migration for forming community over time. Despite the prevalence of origin stories that typically detail the spatial movements of a group’s ancestors and their entanglements with gods to reach a particular territory and found a genealogically connected community, various scholars stress that these narratives create rather than reflect history (2012). A post-colonial trend in writing about Himalayan migration also promises to question how narrative should be used for understanding migration and identity (Subba and Sinha, 2016). These trends suggest that migration histories will expand to encompass new voices and perspectives centered on migrants’ understandings of power and place. Re-reading Philippe Sagant’s work offers hope, too, that local histories can re-open questions about imperialism and global processes with a sense of the urgency with which ordinary people found spaces and made meaning in their own times, sometimes against all odds.

Haut de page


Archival collections consulted

India Office Library
Board’s Collections
Hodgson Papers

West Bengal State Archives
Bengal Proceedings

National Archives of India, Delhi
Foreign Department

Published works

Amrith, Sunil
Migration and Diaspora in Modern Asia (Cambridge, Cambridge University Press).

Balikci, Anna
Lamas, Shamans, and Ancestors: Village Religion in Sikkim (Leiden, Brill).

Caplan, Lionel
Land and Social Change in East Nepal: A Study of Hindu-Tribal Relations (London, Routledge & Kegan Paul).

Chatterjee, Piya
A Time for Tea: Women, Labor, and Post/Colonial Politics on an Indian Plantation (Durham, Duke University Press).

Chemjong, Iman Singh
Kiratakalina Vijayapurako Sankshipta Itihasa (Darjeeling, Nepali Sahitya Parishad).

English, Richard
1985 Himalayan state formation and the impact of British rule in the nineteenth century,
Mountain Research and Development, 5 (1): 61–78; DOI: 10.2307/3673223.

Fitzpatrick, Ian
Cardamom and Class: A Limbu Village and its Extensions in East Nepal (Kathmandu, Vajra Publications).

Forbes, Ann
1996 The discourse and practice of kipat,
Kailash, XVIII (1–2): 39–80.

1999 Mapping power: Disputing claims to kipat lands in northeastern Nepal, American Ethnologist, 26 (1): 114–138; DOI: 10.1525/ae.1999.26.1.114.

Gorer, Geoffrey
[1939] 1984 
The Lepchas of Sikkim (Delhi, Cultural Publishing House).

Gommans, Jos
The Rise of the Indo-Afghan Empire, c.1710–1780 (Leiden, Brill).

Government of Bengal
Annual Report on Administration of the Bengal Presidency, 1860–61 (Calcutta, Bengal Secretariat Office).

1873 Report on the Administration of Bengal, 1872–73 (Calcutta, Bengal Secretariat).

Gros, Stéphane and Schlemmer, Grégoire
2016 Obituary | Philippe Sagant (1936–2015):
Thoughts on disappearing worlds, Himalaya, the Journal of the Association for Nepal and Himalayan Studies, 35 (2): 171–176.

Hamilton [Buchanan], Francis
An Account of the Kingdom of Nepal, and of the Territories Annexed to this Dominion by the House of Gorkha (Edinburgh, A. Constable and Company, 1819).

Huber, Toni and Stuart, Blackburn (eds.)
Origins and Migrations in the Extended Eastern Himalayas (Leiden, Brill).

Jones, Rex L. and Jones, Shirley Kurz
The Himalayan Woman: A Study of Limbu Women in Marriage and Divorce (Palo Alto, CA, Mayfield).

Kolff, Dirk
Naukar, Rajput, and Sepoy: The Ethnohistory of the Military Labour Market in Hindustan, 1450–1850 (Cambridge, Cambridge University Press).

Lucassen, Jan and Lucassen, Leo
Globalising Migration History: The Eurasian Experience (Leiden, Brill).

Mazaudon, Martine and Michailovsky, Boyd
1977 On the road, interview with Narad Bahadur Phembasong,
Pangloss Collection, online:

Morris, John
Living with Lepchas: A Book about the Sikkim Himalayas (London, W. Heinemann).

Mullard, Saul
2017 Regulating Sikkimese society: The fifteen-clause domestic settlement (
nang ’dum) of 1876, in J. Bischoff and S. Mullard (eds.), Social Regulation: Case Studies from Tibetan History (Leiden, Brill): 10–48.

Nebesky-Wojkowitz, René de
Where the Gods are Mountains: Three Years Among the People of the Himalayas (London, Weidenfeld and Nicolson).

O’Malley, Lewis Sydney Stewart
[1907] 1985 
Darjeeling (New Delhi, Logos).

Ramaswamy, Vijaya
Migrations in Medieval and Early Colonial India (Abingdon, Routledge).

Ramirez, Philippe
2013 Ethnic conversions and transethnic descent groups in the Assam-Meghalaya borderlands,
Asian Ethnology, 72 (2): 279–297.

Regmi, Mahesh C.
Land Tenure and Taxation in Nepal, vol. III: The Jagir, Rakam, and Kipat Tenure Systems (Berkeley, Institute of International Studies, University of California Press).

Sagant, Philippe
1978Ampleur et profondeur historique des migrations népalaises, L’Ethnographie, 77–78: 93–119.

1996Dasaĩ et le double pouvoir chez les Yakthumba, in G. Krauskopff and M. Lecomte-Tilouine (eds.), Célébrer le pouvoir: Dasaĩ, une fête royale au Népal (Paris, CNRS éditions/Éditions de la MSH): 283–314.

[1996] 2008 The Dozing Shaman: The Limbus of Eastern Nepal, translated by N. B. Scott (Delhi, Oxford University Press).

Sharma, Jayeeta
Empire’s Garden: Assam and the Making of India (Durham, Duke University Press).

Shneiderman, Sara
Rituals of Ethnicity: Thangmi Identities Between Nepal and India (Philadelphia, University of Pennsylvania Press).

Sivaramakrishnan, Kalyan
Modern Forests: Statemaking and Environmental Change in Colonial Eastern India (Stanford, Stanford University Press).

Subba, Tanka
Politics of Culture: A Study of Three Kirata Communities in the Eastern Himalayas (Chennai, Orient Longman).

Subba, Tanka and Sinha, Awadhesh
Nepali Diaspora in a Globalised Era (London, Routledge).

Vajracharya, Dhanvajra and Shrestha, Tekbahadur
1978 Sikkimka Kaji Yukla Thupko Nepalma Sharan [The asylum of Sikkim’s Kaji Yukla Thup in Nepal],
CNAS Journal, 5 (1): 37–50.

Viehbeck, Markus (ed.)
Transcultural Encounters in the Himalayan Borderlands Kalimpong as a ‘Contact Zone’ (Heidelberg, Heidelberg University Publishing).

Warner, Catherine
2014 Flighty subjects: Sovereignty, shifting cultivators, and the State in Darjeeling, 1830–1856,
Himalaya, the Journal of the Association for Nepal and Himalayan Studies, 34 (1): 23–35.

Haut de page


1 Ethnographers of the eastern Himalaya included Caplan, 1970; Gorer, [1939] 1984; Jones and Jones, 1976; Morris, 1938; Nebesky-Wojkowitz, 1956. For a discussion of the eastern Himalaya as a “contact” zone where transcultural encounters and knowledge about the Himalayas were facilitated by borderland intermediaries from the late nineteenth century, see Viehbeck, 2017.

2 India Office Library. Hodgson Collection. MSS EUR. D. 537. “Free Translation of the Limbu History” (1843) by Iman Singh Chemjong, 1962. Photocopied and preserved in Centre d’Études Népalaises. (I wish to thank Boyd Michailovsky for sharing this document with me.)

3 Gorer notes that the position of mandals was hereditary in the Dzongu region he studied and had been held in some families for 11 to 12 generations. Thus, he speculated that the role of the mandal predated the Namgyal monarchy of Sikkim ([1939] 1984: 127).

4 A glimpse of how patron-client networks functioned in the early nineteenth century is provided by the records of the East India Company’s deputation to settle the border in the 1830s. The Company had accorded itself this authority in the Treaty of Titalya, 1817, that turned over Nepal’s recent conquests to Sikkim. The treaty envisioned that conflicts would arise in terms of people rather than territory, abrogating to the Company the right to mediate “disputes or questions that may arise between [Sikkim’s] subjects and those of Nepaul, or any other neighbouring State.” Lepcha informants told the Company’s deputation that they controlled the local territory around the Mechi River watershed; however, they also admitted having worked as low-level officials for Nepal, Sikkim, and semi-independent Lepcha Kazis. For more details, see Warner, 2014.

5 India Office Library. Board’s Collections. F/4/1813. 1839–1840, 74755. Affairs of Nepaul. India Political Department.

6 West Bengal State Archives. Proceedings of Lieutenant-Governor of Bengal. Political Department. Brief Report on the Political Relations between the British Government and the Sikhim State by Dr. A. Campbell.

7 The Company invaded Sikkim in 1860–1861, again partly in response to conflict over subjects as well as trade, and established a new treaty in 1861 designating Sikkim as a British protectorate. Government of Bengal, 1861: 125–131.

8 Superintendent Campbell established wage rates based on colonial ethnic categories for services such as portering, road building and construction. Limbus were paid less than Lepchas. National Archives of India, New Delhi. Foreign Department. 14 June 1850. No 504. From Campbell to Lushington.

9 This view contrasts with Caplan, 1970: 23–24; Regmi, 1965: 93–94.

10 The administration preferred to collect rent in cash rather than labor, livestock, or grain as had previously been the norm. India Office Library. Board’s Collections. F/4/1968, 1841–42, 87212. Affairs of Sikhim & Darjeeling. India Political Department.

11 Gros and Schlemmer (2016) also stress the significance of this development in Sagant’s thinking as he used the household as an entry point for understanding Limbu ideologies and practices of power before Gorkha expansion.

12 For Sikkim, Balikci provides an interesting discussion of how incorporation into a Lhopo lineage meant accepting the ancestors as well as the gods and rituals of a particular lineage (2008: 73–79). For the Assam-Meghalaya borderlands, see Ramirez, 2013.

13 The agreement also cast the Nepali settlers as a threat to the identity of Sikkim, which was explicitly described as founded upon an agreement by Lhopos, Lepchas, and Limbus to live under Sikkim’s monarchy and Buddhism. Mullard, 2017: 10–48.

14 Every year about 200–300 Limbus of eastern Nepal were recruited. Prior to recruitment into the Indian Army, Limbus had joined the military police in Assam and Burma. Caplan, 1970: 109; Sagant, [1996] 2008: 281–283.

15 In 1901, the total plantation workforce was roughly 64,000, representing about 25% of the district population in Darjeeling. WBSA. Bengal Proceedings, Revenue Department. July 1910. Nos 17–20. File 8–M/27 1–2. Indebtedness of tea-garden employees and other inhabitants of the Darjeeling District.

Haut de page

Pour citer cet article

Référence électronique

Catherine Warner, « The power of movement », Ateliers d'anthropologie [En ligne], 49 | 2021, mis en ligne le 11 janvier 2021, consulté le 25 janvier 2021. URL : ; DOI :

Haut de page


Catherine Warner

Independant scholar, PhD History, University of Washington[at]

Haut de page

Droits d’auteur

Licence Creative Commons
Ateliers d'anthropologie – Revue éditée par le Laboratoire d'ethnologie et de sociologie comparative  est mis à disposition selon les termes de la licence Creative Commons Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International.

Haut de page
  • Logo LESC - Laboratoire d’Ethnologie et de Sociologie Comparative
  • Logo Maison René Ginouvès - Archéologie & Ethnologie
  • Logo DOAJ - Directory of Open Access Journals
  • OpenEdition Journals
Search OpenEdition Search

You will be redirected to OpenEdition Search