Towards an interdisciplinary, transregional study of the state in South Asia

The anthropological literature on the politics of recognition in South Asia has burgeoned in recent years. Many scholars have focused on the cultural politics that emerge in relation to the Indian state's policies of affirmative action through reservations for Scheduled Tribes, Scheduled Castes, and Other Backward Classes, and their implications for political subject formation.Footnote ⁶ Several of these works build substantially on the broader anthropology of the state, which has seen much of its formative work conducted in South Asia.Footnote ⁷ This literature has focused either on how common people experience the state in their everyday lives, or on how state institutions are produced and maintained through a focus on the lives of bureaucrats and other state actors. However, the anthropology of the state in South Asia has not yet adequately explored the specific historical processes through which state and sub-state structures have been forged, how administrative boundaries have been drawn, and how such administrative choices and their implementation at the sub-state level have affected—and been affected by—the formation of political consciousness at the individual level. The legacies of the postcolonial Subaltern Studies project, which promoted a bifurcated, relatively ahistorical view of state power and subaltern resistance, coupled with more recent Foucauldian approaches that emphasize governmentality as a diffuse and historically unmediated source of power, have constrained empirical enquiries. In this instance, we suggest they have foreshortened analysis of the relationship between the particularities of administrative structure and the articulation of identity at specific geo-historical locations, focusing instead on an analytically abstracted ‘state’.Footnote ⁸ Yet it is such structural choices about the shape and apparatus of state units which to a great extent determine how the politics of recognition plays out in specific locales, and how ethnicity is experienced and expressed. By the same token, we suggest that the particular administrative form of each state and sub-state unit emerges in part in response to the affective content of locally-specific ethnic configurations.

Political scientists have paid considerably more attention to state structures. Literature on federalism in multi-ethnic societies has focused particularly on the question of whether the boundaries of federal sub-units should be drawn in ways that recognize ethnicity. This question is primarily animated by a concern with how institutions should be designed in order to minimize ethnic conflict, and it has been at the heart of the recent constitutional negotiations in Nepal. One group of scholars, drawing on the experience of the former Soviet Union and Yugoslavia, argue that ethnic sub-units are likely to promote conflict and, at an extreme, act as the building blocks of secessionist movements.Footnote ⁹ This view has been contested by others who show that the accommodation of ethnic conflict via some form of territorial autonomy is likely to diminish rather than increase the risk of ethnic conflict.Footnote ¹⁰ India has been a common reference point for this latter group of scholars who point to the linguistic reorganization of state boundaries in the 1950s and 1960s as a successful accommodative strategy.

Beyond the question of the positioning of internal boundaries of federal systems in multi-ethnic settings, a second order discussion amongst political scientists has focused on whether culturally or ethnically defined units should be granted differential or ‘special’ rights compared with ‘non-ethnic’ units or those units that are sub-divisions of a national majority population.Footnote ¹¹ Those who argue in favour of such asymmetrical arrangements commonly draw on a normative commitment to a politics of recognition, in which the acknowledgement of difference is seen as a critical move in the achievement of universal equality.Footnote ¹² In India, asymmetrical provisions have been constitutionally mandated for states in the country's Northeast, and to some extent for Jammu and Kashmir. In the Northeast, these provisions have granted restricted rights of land ownership, reservations of seats in state assemblies, delimitation of electoral constituencies to favour particular groups (constitutionally mandated forms of over-representation for certain ‘indigenous’ communities beyond their proportion of the population), and the respect of customary law. Alfred Stepan, Juan Linz and Yogendra Yadav argue that such asymmetry is foundational to what they call the Indian ‘state-nation’—as opposed to ‘nation-state’—which has allowed the Indian federal system to accommodate the multiple identities held by Indians.Footnote ¹³ Yet this political science literature has typically paid less attention to the question of how different modes of state organization—even within the same country—affect the formation of political subjectivities.

In order to fully understand how ethnicity shapes and is shaped by state restructuring processes, we suggest that it is vital to recognize the differences between territorial (such as redrawing state boundaries) and non-territorial (such as affirmative action) forms of recognition, yet situate them within a single analytical framework. In post-2006 Nepal, one anchor for mobilization has been the demand for identity-based federalism—in other words, explicit territorial recognition of ethnic difference at the constitutional level. Another has been the demand for affirmative action—a set of policies to address socio-economic inequality through what has often been called ‘special rights’. These two demands are often conflated within political discourse, with arguments for affirmative action embedded in those for identity-based federalism, as if special rights are inherently linked to territorial recognition through the model of self-determination.

The conflation of these two forms of recognition in the public sphere is somewhat ironic given that, as we have seen, scholarly work often treats the question of subject formation through the politics of recognition as a separate issue from that of federal state structuring as a mode of ethnic accommodation. By bringing these disparate perspectives into conversation we can address a shared question: what is the relationship between specific kinds of state and sub-state structures, and the emergence of specific kinds of ethnic subjects? A combined approach helps to clarify on the one hand how regimes of recognition are sometimes embedded in the territorial structures of the state, and on the other, to show how legitimate agendas for ethnic recognition can be analytically and practically de-linked from those for state restructuring in contentious political contexts.

The remainder of this paper looks at how processes of state restructuring and political subject formation intersect in India, and the implications of these dynamics for Nepal. Within India, we describe two main approaches to state creation: those which have embedded preferential rights for designated communities on a territorial basis, and those which have recognized societal diversity without conferring preferential rights to groups on a territorial basis. We first provide an overview of debates about these issues in India's Northeast, where the question of territorial recognition has perhaps been more contested than anywhere else in the country. We then consider the history of the relationship between ethnic subject formation and state structuring in three in-depth case studies drawing on fieldwork conducted by both authors individually. First we discuss Darjeeling, a hill district of West Bengal that is populated largely by Indian citizens of Nepali heritage who have alternately demanded statehood for ‘Gorkhaland’ as a single ethnic unit, and recognition as up to 14 ethnically distinct Scheduled Tribes (a form of non-territorial recognition). We then discuss Sikkim, a state in which officially recognized ‘subjects’—who may or may not be members of designated Scheduled Tribes—receive preferential rights (a form of territorial recognition). Finally, we consider Jharkhand, a new state created in a region where there had historically been calls for a ‘tribal’ state but where the granting of statehood has not enshrined preferential rights for local Scheduled Tribe communities.

The final portion of the paper explores how contemporary constitutional debates in Nepal have sought to address shifting ethnic assertions and their relationship to structures of a future federal state; a debate that is informed by the experiences of neighbouring India. We suggest that an emergent discourse in Nepal has begun to harden a distinction between ethnicity and identity. This is a strategic move to assert that multi-ethnic identities focused around region, class and marginalization—rather than ‘ethnicity’ per se—should be the basis for fixing territorial boundaries. Such a compromise is similar to the transformation that occurred in the movement for a new Indian state of Jharkhand, where, through a process of strategic accommodation, what was initially a ‘tribal’ agenda eventually became focused around a regional, rather than exclusively ethnic, conception of identity. We show that, as in Jharkhand, such shifts away from ‘ethnic’ to broader identity formulations may emerge out of pragmatic political strategies in areas with complex demographies where there is opposition from those cast as ethnic ‘others’ to the establishment of states along potentially exclusionary lines. This formulation of the basis of ‘stateness’ stands in contrast to the ‘ethnic homelands model’ adopted in India's Northeast.Footnote ¹⁴ Yet notwithstanding this putative shift from ethnicity to broader conceptions of identity by political actors seeking to appeal to wider constituencies, we will see that ethnicity remains not only politically salient, but an emotionally powerful category of self-definition. This helps to explain why ethnic assertions remain so prominent in Nepal today, even though political organizations increasingly shift towards the rhetoric of ‘identity-based’, rather than ‘ethnic’ solidarity.

Approaches to ethnicity

Before proceeding further, we must situate our work in relation to the major disciplinary approaches to ethnicity in recent years. For some time, both anthropology and political science had largely consigned ethnicity to the past—whether understood as a remnant of the colonial ethnographic project, which once understood to be constructed rather than essential would lose its analytical value in shaping socio-cultural inquiries;Footnote ¹⁵ or as something that would wither away as the modernist project came to fruition.Footnote ¹⁶ With the first decade of the twenty-first century behind us, we can now say with certainty that ethnicity has never disappeared in many parts of the world, either at the level of political discourse or in cultural practice. Rather, ethnicity is asserted in ever more complex ways, many of which are at the heart of current debates in India and Nepal. In recent years, influential scholars from across the social sciences have turned to new analytical frameworks in an effort to explain the ongoing—and changing—prevalence of ethnicity as a category of self-identification and political mobilization. Examples include Jean and John Comaroff's Ethnicity Inc, Kanchan Chandra's Constructivist Theories of Ethnic Politics, Andreas Wimmer's Ethnic Boundary Making and James Scott's The Art of Not Being Governed.Footnote ¹⁷ The latter is a figure who bridges political science and anthropology, and other scholars from both disciplines have also turned towards each other in an effort to develop applied approaches to understanding contemporary ethnicity in all its complexity. For instance, Ravi Kanbur, Prem Kumar Rajaram, and Ashutosh Varshney write on the value of interdisciplinary approaches to ethnicity in a special issue of World Development entitled ‘Ethnic Diversity and Ethnic Strife’.Footnote ¹⁸ Our contribution builds upon such work both by initiating an interdisciplinary conversation as well as developing a framework to understand the relationship between processes of state restructuring and ethnicity formation. At the same time we seek to understand the range of social and political consequences that arise from different models of and for institutionalizing the relationship between the state and ethnicity. In this we are particularly interested in the dialectical relationship between identity formation and different possible state regimes of recognition.

There are arguably two ways in which ethnicity has been understood over time within the scholarly literature of both anthropology and political science. The first is premised upon a relational notion, in which ethnicity is defined in the context of inter-group or inter-personal relations, and in the presence of ethnic ‘others’ or as a result of ethnic ‘othering’ by political entrepreneurs. Frederik Barth's seminal arguments emphasized the fluid boundaries between ethnic groups. Boundary-focused notions have influenced theories of ethnicity and nationalism in the Indian context by political scientists such as Paul Brass and Kanchan Chandra, as well as anthropologists like Stanley Tambiah and Arjun Appadurai.Footnote ¹⁹ A concern with inter-group relations also arises in work that takes as its starting point Charles Taylor's discussion of the ‘politics of recognition’ in which the need for recognition of difference arises from the psychological consequences of non-recognition at an inter-personal or inter-group level.Footnote ²⁰

The second strand of work on ethnicity focuses on what the Comaroffs call the ‘substantive content’ of ethnic consciousness, and seeks to understand how the affective reality of ethnic identification is forged and shapes life experiences.Footnote ²¹ Such work focuses on how the experience of ethnic identification is produced through cultural practice, for instance, through the ritual expression of deep-seated attachment to territory (although one does not necessarily need to live in that place itself to feel that way) through the propitiation of territorial deities, or through public performances that demonstrate the contents of ethnic consciousness to outside observers.Footnote ²² Sivaramakrishnan and Cederlof have described a related form of territorial claim as ‘ecological nationalism’ in which attachments to nature and place are understood through ethnic and sometimes regional lenses.Footnote ²³ In this second formulation, ethnicity is conceived as arising in significant part from within groups, as something intrinsic to their connection to particular territories or landscapes, rather than exclusively from boundary encounters with those who become ethnic ‘others’.

The relational approach to ethnicity epitomized by Barth was of course a corrective to earlier anthropological approaches which presumed a one-on-one correlation between culture and each bounded group. But now after several decades of theorizing focused on the relational aspects of ethnicity, we see the need to ask again what the content of particular ethnic formations looks like in order to understand why ethnicity remains so emotively powerful, even in contexts where those who identify with it are well aware of its constructed nature. Our intention is not to return to the old argument about whether ethnicity is primordial or constructed. Rather we recognize fully that ethnic identity is historically and politically constructed, but believe that this is just the starting premise. The question is how is it produced at the intersection of state policy, administrative boundaries and grassroots practice. We assert the need to engage with the content of ethnic consciousness that lies between boundaries—both administrative and psychological—as well as understanding how those boundaries are themselves produced.

State structures and ethnicity in India

In the section that follows, we examine the intersection of processes of state (re-)structuring and the formulation of ethnic identities in India. India's 1950 constitution largely avoided the creation of federal sub-units along identity lines, but the contours of India's states have been substantially reorganized over time to recognize different facets of identity. As Rajesh Dev writes, the postcolonial Indian state has attempted to overcome the ‘assimilationist individualism’ inherent in a liberal conception of citizenship by enshrining a ‘differentiated citizenship’ and creating states on an ethno-linguistic basis.Footnote ²⁴ Yet not all states created on the basis of group identities in India have also been granted differential rights on a territorial basis.

In India's Northeastern region (see Map 1), amongst states created from Assam between the 1960s and 1980s, regimes of self-governance have been combined with substantial forms of positive discrimination to protect autochthonous or ‘indigenous’ communities from processes of demographic and economic change, and to enshrine recognition of their cultural autonomy. By contrast, in other parts of India federal restructuring has not involved the granting of differential rights to communities presumed to be autochthonous. Linguistic reorganization in the 1950s and 1960s created states for speakers of different languages but not in ways that officially embedded preferential rights for such communities, except by virtue of their demographic majority within new administrative jurisdictions. The linguistic states of south and west India—home to Tamil, Telugu, Malayalam, Kannada, Marathi and Gujarati speakers—have the same constitutional rights in terms of self-governance as the predominantly Hindi-speaking states of north and central India. Yet becoming a separate state provided institutional recognition and protection to each major linguistic community with the rights to oversee education, language, recruitment to local government jobs and so on. Newer states such as Chhattisgarh, Jharkhand and Uttarakhand—created in the year 2000 from predominantly Hindi-speaking regions of north and central India—have also not seen the embedding of preferential rights for any community as part of the process of state formation.Footnote ²⁵

Map 1. States of Northeast India and surrounding countries

India's Northeast has a distinctive ethnic demography which may make it more comparable with Nepal than other parts of India. Within India, the Northeast is often treated as a place of exception and left aside from discussions of politics elsewhere in ‘mainstream’ India. The region is geographically remote from India's centres of power, sitting on the borders of Bangladesh, Myanmar and China. It has the largest concentration of Scheduled Tribe communities of any part of India, a significant Christian population (the majority in three states), a mix of hills and plains areas, and a diverse linguistic fabric. Territorialized regimes of positive discrimination for certain groups were envisaged for the region in the Indian constitution under the Sixth Schedule which created a set of cascading ‘autonomous’ institutions below the level of the state. Under the Sixth Schedule, certain ‘tribal areas’ of Assam, Meghalaya, Tripura and Mizoram were given provisions for their own autonomous district councils. On paper, these councils have far-reaching rights over law-making with respect to areas such as land-use, forest management (except ‘reserved’ or ‘protected’ forests), the establishment of village or town committees, property inheritance, marriage, and other social customs.Footnote ²⁶ The Regional Council may also oversee the establishment of village councils or courts to try cases between Scheduled Tribes within the area, assess and collect land revenue, impose taxes, and regulate money-lending and trading by non-tribals.Footnote ²⁷ These measures were substantially expanded from the 1960s onwards as the Indian state sought to accommodate separatist movements amongst Nagas and Mizos.Footnote ²⁸ Provisions which offer a greater degree of autonomy than other states of the Indian Union and protect the status of groups recognized locally as Scheduled Tribes have been constitutionally embedded in some of the states created from erstwhile Assam under clauses of Article 371 of the Indian constitution. The overwhelming majority of seats in the state assemblies of Arunachal Pradesh, Mizoram, Nagaland and Meghalaya, for instance, are reserved for Scheduled Tribes and special rights are accorded to Nagas and Mizos to manage cultural and legal practices (see Table 1). Beyond these special measures accorded to particular ethnic communities on a territorial basis, the Northeastern states are also recognized as ‘special category’ states by the Planning Commission. This provides these small states access to more generous financial assistance from the central government, notably a higher proportion of grants to loans. Table 1 provides an overview of the forms of special rights that are accorded on a territorial basis to Northeastern states.

Table 1. Special Rights in India's Northeastern States

Source: Authors' summary based on provisions of the Constitution of India.

Notes: * Previously North Cachar Hills Autonomous District Council.

** Since 1995, these Autonomous District Councils have had additional powers under the Sixth Schedule to make laws in areas including agriculture, education, health, and for Dima Hasao and Karbi Anglong, industries and communications. See Schedule 6 of the Indian Constitution.

Critics like Sanjib Baruah suggest that the cumulative effect of the territorialized forms of recognition deployed in the Northeast has been to create an ‘ethnic homeland model’ that stands at odds with the ‘actually existing political economy of the region’.Footnote ²⁹ Not only do ethnic identities correspond imperfectly with state boundaries, but territorialized special rights have created effective dual citizenship regimes. Baruah writes:

One result of the special apparatus of ethnic federalism in the Northeast has been the cascading of group claims for recognition, sometimes pursued using violent strategies against ethnic ‘others’.Footnote ³¹ Such claims have ranged from calls by groups for recognition as Scheduled Tribes within a particular state in order to gain access to state resources reserved for local Scheduled Tribes; to the extension of the Sixth Schedule to new areas; to calls for fully-fledged statehood; to recidivist claims by militant groups to parts of the territory of neighbouring states. By contrast, in most other areas outside the Northeast, state formation—even where it has recognized distinct communities, such as linguistic groups—has not involved the granting of differential rights to ‘local’ communities on a territorial basis. In the case studies that follow we explore these different modes of state structuring and their impact on political subjectivity.

State structures, ethnicity, and identity in Darjeeling

One of India's oldest yet still unmet demands for a separate state is the call for Gorkhaland in Darjeeling. This northernmost district of West Bengal is adjacent to Sikkim, the erstwhile Buddhist kingdom which became India's smallest state after incorporation (or ‘annexation’ or ‘occupation’, depending upon whom you ask) into India in 1975. Both Darjeeling and Sikkim share a similar population demographic, comprising roughly 80 per cent Nepali-speaking Indian citizens of Nepali heritage. Yet their trajectories of political mobilization along ethnic lines diverge, and a call for a unified Nepali-speaking state that would incorporate both Sikkim and Darjeeling has never gained much political traction. In this section, we suggest that a careful look at the ebb and flow of political subject formation in Darjeeling in contrast to comparable processes in Sikkim over time reveals much about the impact of specific strategies of federal incorporation and territorial recognition on political consciousness in general, and ethnicity in particular.

First proposed by the Hillmen's Association in 1907 (then an alliance between the Nepali, Bhutia and Lepcha communities), the idea of a ‘separate administrative set-up’ for the Nepali-speaking population of northern Bengal gained new purchase post-independence, and by the 1980s led to a violent agitation.Footnote ³² This movement mobilized Darjeeling residents from a broad array of ethnic and linguistic backgrounds around the idea of a shared ‘pan-Nepali’ or ‘Gorkhali’ identity,Footnote ³³ which was also cast in territorial terms as a ‘hill’ identity distinct from that of the plains dwellers who dominated West Bengal state politics. In 1989, a tripartite agreement between the Gorkhaland National Liberation Front, the West Bengal state government, and the Government of India put a temporary end to the agitation with the establishment of the Darjeeling Gorkha Hill Council. This is one of several instances in which the Indian state has granted limited provisions for territorial autonomy, without either acceding to demands for statehood or implementing the Sixth Schedule.Footnote ³⁴

But within a few years this arrangement began to seem unsatisfactory to many Darjeeling residents. The Darjeeling Gorkha Hill Council was granted little financial autonomy which resulted in poor implementation of many of the infrastructural development projects with which it had been charged, and in-fighting emerged between the central leaders of the Gorkhaland movement. Several groups of Nepali heritage in Darjeeling began to demand tribal recognition instead, moving away from the unified Gorkhaland demand.Footnote ³⁵ In the post-Mandal climate of the early 1990s, members of several communities demanded and received Other Backward Class status, which quickly became viewed as a stepping stone towards tribal recognition. By the early 2000s, members of 14 different communities of Nepali heritage were campaigning for recognition as Scheduled Tribes from both the state of West Bengal and the central government.Footnote ³⁶

For nearly a decade from the mid-1990s to the mid-2000s, individuals who had earlier supported the Gorkhaland movement—for a state in which ‘Gorkhali’ would be recognized as the operative ethnic category—shifted their political loyalties to ethnic organizations representing only one or other of the constituent ethnic communities comprising the Gorkhaland alliance. Through political rallies, cultural performances, letter writing and much on-the-ground diplomacy in Calcutta and Delhi, two of these groups succeeded in securing Scheduled Tribe status in 2003: the Tamang and Limbu. This upped the ante for the remaining groups, who expanded their campaigns in the middle part of the decade, in a manner which often led to small scale inter-group violence as well as disaffection between members of individual communities themselves who disagreed about the cultural basis for ethnic solidarity.Footnote ³⁷

The Darjeeling Gorkha Hill Council chairman, Subash Ghisingh, frowned upon such group-by-group ethnic mobilization, and sought to rejuvenate the Gorkhaland movement by demanding Sixth Schedule status for the region instead. Yet he was outflanked by the Gorkha Janamukti Morcha, a new party launched by former Darjeeling Gorkha Hill Council council member, Bimal Gurung. The Gorkha Janamukti Morcha claimed that acceptance of the Sixth Schedule would seal the fate of Gorkhaland with a relatively meaningless form of territorial recognition whilst closing off future avenues to full statehood. In 2008, the party relaunched an agitation for full-fledged statehood (adding additional non-hill areas of the Duars to its demand), wresting control from Ghisingh. A new Gorkhaland Territorial Administration was established by the freshly elected Trinamool Congress state government in West Bengal in 2012, with Gurung at the helm. However, in the wake of the central government's July 2013 announcement that it would proceed with the creation of a new state of Telangana from Andhra Pradesh, Gurung resigned from the Gorkhaland Territorial Administration chairmanship in order to continue agitating for Gorkhaland. As stated in a March 2012 interview, Gurung and his party had always viewed the Administration as a halfway house to full statehood: ‘So many states have been created in India since Independence. Why should only the Gorkhas not be allowed to have their state? Nothing short of statehood is a complete solution to the problems of the Gorkhas, be it identity or development. The geopolitical situation of the region logically demands a separate state’.Footnote ³⁸

Whether or not Gurung's position enjoys full popular support is hard to ascertain, since like the Gorkhaland National Liberation Front before him, the Gorkha Janamukti Morcha leader has a knack for silencing opposition through what are often talked about on the Darjeeling streets as ‘strong arm tactics’. However ethno-historical research on the formation of identities in Darjeeling does demonstrate that ethnicity has been an operative concept for group mobilization since the early twentieth century,Footnote ³⁹ whether understood as the ethnicity of individual groups of Nepali heritage, or as a pan-Nepali ‘Gorkhali’ ethnicity whose diverse constituents are the erstwhile citizens of Gorkhaland. The archives of organizations representing different groups of Nepali heritage—many of which date back to the 1920s and 1930s—demonstrate that the ideals of unnati (improvement) and utthan (upliftment) on the basis of ethnic identity were enshrined as the objective of several group-specific organizations in Darjeeling long before the 1950 Indian constitution attached entitlements to the categories of Scheduled Caste, Scheduled Tribe or Other Backward Class; or before the onset of (neo)liberal economic and social policies introduced what are often described as entirely new forms of identity-based mobilization around radically different modes of ethnic subjectivity.

The point here is that whereas ethnicity has been used as a basis for political mobilization in a variety of different ways over the last century in Darjeeling, the general concept has remained consistently important as a means of voicing collective aspiration. Yet the specific ways in which those aspirations have been expressed have shifted over time in relation to broader cultural and political dynamics, as well as the perceived willingness of the central state to offer either territorial (Sixth Schedule or statehood) or non-territorial (Scheduled Tribe classification) recognition, at particular political-historical conjunctures.

State structures, ethnicity, and subjecthood in Sikkim

The elusiveness of statehood in Darjeeling has shaped political and ethnic consciousness in a manner quite different from its neighbouring state of Sikkim. There, after initial resistance to annexation in the late 1970s led primarily by Bhutia elites close to Sikkim's erstwhile royal family,Footnote ⁴⁰ ethnic claims have remained relatively muted at the national level vis-à-vis the Government of India, whilst becoming an important tool in power relations amongst different political actors at the sub-state level within Sikkim.Footnote ⁴¹ Although Sikkim shares Darjeeling's large Nepali-speaking demographic, it differs in several essential aspects.

At the point of annexation, the Government of Sikkim was allowed to maintain the legal category of ‘Sikkimese subject’ as a means of distinguishing historical residents of the state from new immigrants.Footnote ⁴² Enforced through the distribution of ‘domicile certificates’ for verified subjects, this system enables the Government of Sikkim to limit in-migration and carefully monitor the distribution of entitlements. This is a territorialized form of recognition, but nonetheless one which differs from the Sixth Schedule as employed elsewhere in the Northeast,Footnote ⁴³ since it is not defined exclusively by Scheduled Tribe status. Rather, it is a category defined by historical residence, as members of any group who can document their residence in Sikkim before 1975 are eligible. Maintaining these distinctions among historical residents and newcomers is particularly important due to the generous financial subsidies that the state of Sikkim and its documented subjects receive from the central government in recognition of the border state's strategic importance, disproportionate to its size and population. Darjeeling residents frequently comment upon the flush resources they perceive their cousins (often literally, since many kin networks extend across state boundaries) in Sikkim to benefit from. Such comments refer both to the powerful, direct relationship the Government of Sikkim maintains with the central government—in contrast to Darjeeling's experience of being always one step removed, due to the state government of West Bengal's mediating role—and the individual benefits of subject status in Sikkim. Both are forms of recognition that create a sense of security for Indian citizens of Nepali heritage in Sikkim which individuals of comparable ethnic backgrounds in Darjeeling do not enjoy. It is for this reason, many Darjeeling residents assert, that a separate state of Gorkhaland is necessary: they can see the benefits that statehood has brought to their ethnic compatriots in Sikkim, and desire the same for themselves. By the same token, the Gorkhaland movement has not always been supported by Sikkimese political elites, since many feared that violent agitation next door by members of the same broad ethno-linguistic category might jeopardize their special relationship with the central government.

This is despite the fact that the majority of Sikkim's political elites are none other than Indians of Nepali heritage. The state's two chief ministers since 1979 have both been from this background (Nar Bahadur Bhandari and Pawan Kumar Chamling), as has most of the leadership of the dominant political parties. Such Sikkimese state politicians of Nepali heritage do however often serve as informal communication channels between Darjeeling-based activists and the central government—whether in relation to demands for Gorkhaland, the Sixth Schedule, or tribal recognition. For example, ethnic activists campaigning for Scheduled Tribe status on behalf of several communities of Nepali heritage in Darjeeling, have systematically sought to mobilize community members resident in Sikkim in order to forward their applications for tribal recognition to the Centre via Sikkim's official governmental channels—successfully in the case of the Tamang and Limbu.Footnote ⁴⁴ This political configuration is quietly disapproved of by Sikkim's Bhutia and Lepcha communities, the former being one of India's few socio-economically elite groups to maintain Scheduled Tribe status.Footnote ⁴⁵ Along with the Lepcha community, the Bhutia claim indigeneity to Sikkim, and in private decry the political capture of the state by Nepali ‘migrants’—although all holders of political office must possess Sikkim subject status, and often come from families who have been resident in the area for several generations.

This scenario has led the Lepcha community to demand the status of ‘Most Primitive Tribe’, a classificatory category unique to Sikkim, but not unlike the ‘Indigenous Tribe’ category that Karlsson describes in Meghalaya.Footnote ⁴⁶ The purpose of such designations, however, are to claim power within the extant boundaries of the states in question, rather than to bolster claims for the creation of new states, as in Darjeeling next door. The Sikkim legislative assembly already reserves 37.5 per cent of its seats for Scheduled Tribes—they are thus over-represented in the assembly (only 20.6 per cent of the population were Scheduled Tribe according to the 2001 census). This stands in contrast to West Bengal, which reserves only 5 per cent of seats in its legislative assembly in line with their proportion of the state's population (whilst the Scheduled Tribe population in Darjeeling is approximately 11 per cent as of the 2001 census). Nonetheless in Sikkim, Scheduled Tribe is seen as a category of diminishing value since more groups have acceded to it.

Although ethnicity is understood as a strategically important category of identification in Sikkim, it is not an active area of negotiation between the state and central government because Sikkimese subjects already have direct channels of communication with, and subsidies from, the central government. This latter fact often leads to tension between members of the same ethnic communities resident in Darjeeling and Sikkim respectively, who may agree about much of the substantive content of ethnic consciousness, but disagree about how that content may be most effectively mobilized to yield the locally specific political results they desire. In turn, such divergent political agendas themselves come to influence the way in which ethnic consciousness is expressed.

The comparison between Darjeeling and Sikkim demonstrates how the different positionalities of these two locales within India's larger federal structure have shaped expressions of ethnic identity within their boundaries. Similar demographics have yielded very different mobilizations in relation to ethnicity. Darjeeling's ethnic activists view Sikkim's political leaders from the same ethnic backgrounds as complacent, whilst Sikkim's subjects fear encroachment on their special status by those from Darjeeling (news items abound about false Sikkim domicile certificates confiscated from Darjeeling residents, as well as migrants from Nepal).

The politics of recognition is pervasive in both contexts, but the specifics of those politics vary greatly according to individual and community location (both geographical and political) within a larger federal structure. This suggests that we cannot understand the contemporary power of ethnicity in such contexts only by invoking the onset of global neoliberal policies that have given rise to new forms of ethnicity and identity, nor even with reference to the juggernaut of liberalization at the national level in India, and the post-Mandal climate that has produced new demands for recognition from the central state. Rather, we must look to the highly localized features of ethnic consciousness that have emerged over time in relation to long-standing territorial engagements, perhaps only most recently the establishment of administrative boundaries at state and district levels. Such a perspective will generate a better understanding of how demands for recognition emerge out of context-specific matrices of power that are strongly shaped by the historical trajectories of sub-national federal boundaries and their attendant political configurations.

Jharkhand: statehood without preferential rights

Although Jharkhand has a sizeable population of Scheduled Tribes and had seen a long-running popular movement for a tribal state, it is difficult to see the state that was formed there in 2000 as an ‘ethnic homeland’. It thus stands in contrast to the situation of many states in the Northeast, and the outcome desired by some proponents of Gorkhaland. The long-term negotiation of the statehood demand for Jharkhand unsettled the link between ethnic and territorial claims as political parties appealing to both tribal and non-tribal groups over time came to support the demand for a state.Footnote ⁴⁷

The demand for a separate Jharkhand developed over several decades from a call for a state in which the rights of indigenous, tribal communities would be enshrined, to a demand for statehood which was supported by national political parties attempting to bind migrants and longer-resident non-tribal populations to the idea of a regional identity. The first movement for statehood in the region, led by the Adivasi Mahasabha (and later Jharkhand Party), predated India's independence. In the early period, it demanded a ‘tribal’ state, which would also incorporate tribally dominated districts of neighbouring states. Parties that mobilized in the name of Jharkhand from the 1950s to 1980 targeted ‘exploitative outsiders’ (or dikus in local parlance) who were seen as benefiting by acquiring land from indebted tribals and cornering the benefits of employment in local industry.Footnote ⁴⁸ Yet successive central governments refused to create a state in Jharkhand on these grounds. From the 1980s onwards, pro-Jharkhand organizations such as the Jharkhand Mukti Morcha made concerted attempts to encourage non-tribal groups that had been long-resident in the region to identify with the Jharkhand demand.Footnote ⁴⁹ This was part of a reframing that became necessary for the movement as a result of the changing demography and decreasing percentage of the population officially classified as ‘Scheduled Tribes’ (in 2001 they accounted for just 26 per cent of the population).Footnote ⁵⁰ The Jharkhand Mukti Morcha were in competition with the Hindu nationalist Bharatiya Janata Party, which itself began to support a demand for statehood as it sought to establish a foothold in the region.

The Bharatiya Janata Party's state president, Inder Singh Namdhari (1988–1990), a Sikh politician of Punjabi origin who was elected from a Jharkhand constituency, coined the name ‘Vananchal’ as a means of distancing the Bharatiya Janata Party from the historical demand for statehood.Footnote ⁵¹ He explicitly acknowledged that, in promoting the idea of Vananchal as distinct from Jharkhand,Footnote ⁵² the party was attempting to move away from the idea of a new state being a tribal homeland. He explained: ‘Because tribals weren't a majority in the region, it was necessary and feasible to create a regional rather than racial identity’.Footnote ⁵³ Once a Bharatiya Janata Party government came to power in New Delhi and announced its intention to create a new state in the region, the local party agreed to adopt the name of Jharkhand rather than Vananchal for the new state, providing limited recognition to the historical movement for statehood.

The kind of state that has been created in Jharkhand is quite different to the territorialized regimes of positive discrimination described above in Northeast India. Instead, affirmative action policies (primarily via ‘reservations’, but also the Fifth Schedule in tribal majority districtsFootnote ⁵⁴) have sat alongside, but not been embedded in, territorial self-governance via the granting of statehood. Nevertheless after statehood was granted in 2000, the idea of Jharkhand as a supposedly ‘tribal’ state did not disappear, and the question of whether tribal communities, or ‘local’ Jharkhandis, should have ‘special’ or preferential rights in the new state remained a contentious subject. Attempts by the first Chief Minister of the new state to introduce a preferential regime for ethnic ‘insiders’ via a domicile policy that would have reserved jobs in the local administration for ‘local Jharkhandis’—similar to the extant system in Sikkim—were rolled back following violent protests in the state capital and an adverse ruling by the state's High Court. The extent of preferential treatment that should be accorded to local adivasi communities is still contested. This complicated the delimitation of constituencies in the state, and delayed elections to local panchayati raj institutions within ‘scheduled areas’ because the proportion of seats that should be reserved for Scheduled Tribes was called into question. Furthermore, despite the fact that each Chief Minister of the state has been an adivasi, more substantial empowerment of poorer adivasi communities has been harder to detect. Some observers suggest that the creation of the new state has done little to address the needs of the region's poorest tribal residents, instead serving the interests of an elite political class.Footnote ⁵⁵

Thus in India, we can see that local political subjectivities have evolved over time in relationship with the changing territorial structures of state and sub-state units. In Northeast India, statehood and territorialized regimes of positive discrimination combine to consolidate access to state resources for groups that are ‘recognized’ as legitimately ‘local’ by the state. This has given rise to claims for Scheduled Tribe status by groups not currently so recognized, to movements for sub-categorization within the Scheduled Tribe category in order to claim special entitlement within an increasingly populated pool of Scheduled Tribes, or movements for the introduction of new, hyper-tribal categories such as Most Primitive Tribe in Sikkim or Indigenous Tribe in Meghalaya.Footnote ⁵⁶ In contrast, outside the Northeast, the creation of new states—even in areas with large Scheduled Tribe populations—has not led to the combination of territorial recognition in the form of statehood with other forms of positive discrimination for Scheduled Tribes applied on a territorial basis. The political difficulties inherent in attempting to create ethnic homelands in demographically diverse regions have led some political entrepreneurs from the 1980s onwards to create broader multi-ethnic conceptions of regional identity, as political coalitions come together to support campaigns for statehood via electoral politics. This was the case in Jharkhand, and to some extent is emerging as a pattern in Gorkhaland too, as the recent self-immolation of a self-declared ‘Bihari Gorkhali’ attests.Footnote ⁵⁷ But these new articulations do not replace the affiliations that individuals have with constituent ethnic communities, nor lessen the frustration amongst many Scheduled Tribe residents of Jharkhand that a state created in their name has not transformed a situation of marginalization.

Debating ethnicity and state structure in Nepal

Nepal faces a similar set of challenges as it seeks to design the territorial architecture of its federal system. Long-standing demands for self-determination have gained traction as regional and ethnic interest groups have emerged as key political forces since 2006.Footnote ⁵⁸ In late 2007 and early 2008 (before the constituent assembly elections were held), the interim government signed a series of agreements with madhesi Footnote ⁵⁹ and janajati Footnote ⁶⁰ organizations guaranteeing that provisions for some form of territorial autonomy would be included in the as-yet-undetermined framework for federal restructuring. The government's 2007 ratification of the International Labour Organization's Convention on the Rights of Indigenous Peoples (ILO 169), added to expectations that the structure of the new federal state would in some substantive way deploy mechanisms of territorial recognition to guarantee the rights of indigenous communities. This expectation arose in part out of the convention's emphasis on the state's responsibility to protect the ‘rights of ownership and possession of the peoples concerned over the lands which they traditionally occupy’.Footnote ⁶¹ Indigenous activists mobilized around this idea by popularizing the concept of agra adhikar, or prior rights. This, in turn, prompted a backlash from members of erstwhile dominant communities, who viewed any such legal protection of rights for certain communities identified as ‘indigenous’ as an assault on the rights (and long-standing privileges) of what political scientist Mahendra Lawoti has called the ‘Caste Hindu Hill Elites’.Footnote ⁶²

This is the crux of Nepal's current debate: how can the state at once offer special entitlements for marginalized groups, whilst ensuring equality and universal access to resources for all?Footnote ⁶³ One set of political actors, led by the Maoists and supported by many smaller parties that represent regional and ethnic interests, advocates an ‘identity-based federalism’. In August 2012, Maoist chairman Pushpa Kamal Dahal (aka Prachanda) announced the formation of a Federal Democratic Republican Alliance to advocate for an identity-based model of federalism. They propose that provincial boundary lines within federal Nepal should be drawn in a manner that recognizes deep-seated ethnic attachments to specific parts of the country by carefully shaping electoral constituencies, and naming new states in reference to the primary ethnic group in each area. Both the State Restructuring Committee of the Constituent Assembly, which submitted its report in January 2010, and the expert High-Level State Restructuring Commission, which submitted its report in January 2012, advocated such models, albeit in different specific geographical terms, with recommendations for 14 and 11 federal states respectively. The 11-state model includes 10 territorial states, and one ‘non-territorial’ Dalit state (see Map 2). The functional modalities of the proposed ‘non-territorial’ state remain unclear, yet this is an interesting attempt to reconcile the tensions between territorial and non-territorial forms of recognition discussed above.Footnote ⁶⁴ The State Restructuring Committee also submitted a dissenting minority opinion, accompanied by a proposal for a six-state model. All of these proposals also included provisions for at least 22 smaller sub-state ‘autonomous regions’ to accommodate less populous ethnic communities who could not be accorded a state of their own. At the time of writing, none of these plans has been implemented, and new Assembly elections are awaited.

Map 2. Nepal: current and proposed internal borders* Note: *None of the proposed models has yet been adopted, and several other proposals also exist. Source: By kind permission of Mapman.co.uk © (2013).

However, these proposals have often been characterized in a negative light as ‘ethnic federalism’ by the Maoists’ political opposition—comprising the Nepali Congress, Unified Marxist Leninists, and several other right-of-centre groups—who suggest that any such territorial acknowledgment of ethnic identity will lead inevitably to conflict and the dissolution of the Nepali state. Instead, they have proposed ‘geographical federalism’, by which the country would be divided into a smaller number of larger states, with boundaries determined by geographical features rather than ethnic demographies (for instance the six-state model proposed by the dissenting members of the State Restructuring Committee, as described above). Such proposals are also referred to as ‘multiple identity-based federalism’—because multiple ethnic communities would be recognized as equally indigenous residents of every state—in contrast to the ‘single identity-based federalism’ advocated by the Maoists and their allies. These political groups point to the definition of ‘identity’ developed by the State Restructuring Committee which states that ethnicity is only one of five facets of identity; the others being language, culture, history and geography. In this formulation, ethnicity and identity are not to be treated as synonymous.

The Nepali Maoists’ support for an identity-based federalism may seem counter-intuitive from a global comparative perspective, from which Maoists might be expected to build solidarities around class, rather than ethnic, consciousness. However, in Nepal, class and ethnic formations have been deeply intertwined over time, both in terms of the political trajectories of communist parties and ethnic associations since the 1950s, and in terms of individual life histories which often bridge both forms of political consciousness and mobilization.Footnote ⁶⁵ These links go back perhaps even further to the legal codification of ethnicity effected by the 1854 Muluki Ain, or national legal code, which attempted to classify all of Nepal's communities within a structure of caste hierarchy that enabled labour extraction by the Hindu state.Footnote ⁶⁶ It also attached both the terms of labour and land ownership—through the system of kipat, or ethnically-based collective land tenure—fundamentally to specific ethnic identities.Footnote ⁶⁷

Current demands for identity-based federalism emerge out of this historical matrix, in which the affective dimensions of ethnic identification have been so long linked to specific territories—by none other than the state, which now resists such linkages. As multiple authors have demonstrated, Nepal's contemporary ethnic landscape has been produced over time through processes of ethnicization that brought diverse linguistic and cultural communities within the ambit of state control.Footnote ⁶⁸ The current debates over state restructuring may be seen as the newest episode in this longue durée historical process, rather than as a departure from it. Such a perspective enables a critical interrogation of the relationship between state regimes of recognition and the emergence of ethnic subjects in political moments like the current one, without discrediting contemporary ethnic actors as disingenuous, or acting only out of politically expedient bad faith.

Disentangling the prospects of and mechanisms for delivering territorial and non-territorial recognition can help to defuse the notion that recognizing ethnicity as a legitimate basis for statehood necessarily leads to conflict or new forms of exclusion. The significance of ethnic identity—at the individual and group level—can be recognized via territorial restructuring that follows loose ethnic boundaries, uses ethnic state names and changes patterns of political representation by creating new electoral arenas. But this need not always involve the embedding of territorially-based special rights for particular ethnic groups, especially where it is difficult to create states that encompass homogenous populations or where there is substantial demographic mobility. In India, where, the lists of Scheduled Castes and Scheduled Tribes are determined on a state-by-state basis at the first instance, and not all groups recognized in one state are recognized in others, people who migrate internally are not entitled to the same rights across the country. This has led to protests in various areas, for instance in Assam where groups recognized as Scheduled Tribes in Jharkhand who migrate to work have demanded the same recognition in their new area of settlement as they had at home.Footnote ⁶⁹ This raises broader questions about the definition of citizenship: how can one attain full political citizenship at the national level if one's special rights are constrained by residence in a particular place?

Territorial recognition is an important means of recognizing a community's genuine attachment to a place, as widely expressed in ritual forms that demonstrate the deep relationship between the content of ethnic identity and specific territories. But the assertion of a close linkage between indigenous bodies and indigenous territory reflects the complex entanglement of contemporary political assertions in the international arena with histories of classification, both colonial and internal. The idea of such absolute linkages between indigeneity and territory has emerged in part through international indigenous movements, and its legal conventions like ILO 169. Such instruments help to reinforce old ideas of indigeneity,Footnote ⁷⁰ which centre on the linkage between indigenous bodies and indigenous territory. These ideas are conceptually and historically related to the colonial paradigms for ethnic classification out of which India's constitutional mechanisms for territorial recognition of Scheduled Tribes emerge. An overemphasis on the linkage between claims for recognition and territory may in the long run be counterproductive, especially if such claims are not coupled with other forms of affirmative action that are effectively implemented at the central level.

By the same token, the fact that ethnic people move around is not in itself a valid basis upon which to challenge their claim to association with a specific territory. Rather, many contemporary people may possess what anthropologist James Clifford has called ‘a portable sense of the indigenous’ (which we might also extend to ethnicity, or indeed identity wholesale).Footnote ⁷¹ In other words, you do not need to live in a particular place to maintain a strong symbolic attachment to it. And it is that symbolic attachment that ethnic state names in Nepal could help to recognize. Yet the very portability of identity means that embedding preferential rights for specific groups only in the states that bear their name is unlikely to benefit all members of any group. Moreover, it might create new insiders and outsiders in the manner that Sanjib Baruah describes for the Indian Northeast.

In Nepal, we can see that even the as yet uncertain potential for federal states based around ethnicity has led to new social mobilizations around identity, particularly for members of dominant groups. Organizations such as the Chetri Samaj (Chetri Society), and the Chure Bhawar Rastriya Ekta Party (Chure Bhawar United National Party) are examples of relatively new movements mobilized around erstwhile dominant identities—the Chetri caste and the regional identity of paharis (inhabitants of Nepal's middle hills), respectively. Both of these parties are reported to have strong local organizations in some parts of rural Nepal, and succeeded in a recent push to have the high caste, historically dominant Brahmin and Chetri groups classified as ‘indigenous’ in a recent Social Inclusion bill (although at the time of writing the bill has not yet been passed). The strong mobilization around this issue may be understood at least in part as a direct result of fear of exclusion from prospective paradigms of territorial recognition. This might also help to explain a quiet retreat from the platform of ‘preferential rights’ on the part of indigenous activists since early 2012. This shift might be understood not as a capitulation to dominant forces, but rather as an astute acknowledgment that territorial recognition is not the only way to achieve the goals at hand—but rather one component of broader agendas for the transformation of inequality and access to state resources.

It appears that many scholars and activists who have been at the forefront of the indigenous people's movement over the last two decades in Nepal have in fact made a strategic decision to shift from the language of ethnicity to that of identity. This is in part as a means of expanding their political platform to include those local residents who do not usually recognize themselves as belonging to the ‘ethnic’ groups claiming statehood in particular locales. In August 2012 a well-positioned group of self-identified indigenous intellectuals declared their intention to form a new political party, which rather than being called an adivasi janajati—or indigenous nationalities—party, is called the Social Democratic Plurinational Party.Footnote ⁷² The formal announcement of this party's planned launch in Kathmandu drew an audience of several hundred, but few of the speakers or signatories to the petition to register the party belonged to non-janajati groups. Yet the janajati speakers who dominated the event described the need for a party within which all identities could be politically recognized, in order to combat the generalized terms of marginalization. One month later, 500 mainstream party activists from janajati backgrounds within all of the major parties resigned their party affiliations to form another new party, the Federal Socialist Party-Nepal. Early press reports emphasized the extent to which party leader Ashok Rai (former Communist Party of Nepal-Unified Marxist Leninist vice chairman) was called upon to ‘disprove that his party is solely an ethnic party’.Footnote ⁷³ It remains to be seen how much momentum either of these nascent parties can generate. Despite being established by individuals who have built careers around the notion of ethnic empowerment, their public political rhetoric is now shifting towards promoting a more inclusive conceptualization of identity.