Jump to ContentJump to Main Navigation
Building LegitimacyExploring State-Society Relations in Northeast India$

M. Sajjad Hassan

Print publication date: 2008

Print ISBN-13: 9780195692976

Published to Oxford Scholarship Online: October 2012

DOI: 10.1093/acprof:oso/9780195692976.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2017. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 30 March 2017

(p.266) Appendix I Methodological Note

(p.266) Appendix I Methodological Note

Building Legitimacy
Oxford University Press

Research Design

Studies on violence and disorder in Northeast India suffer from the single case bias—they either take the entire region as their unit of analysis, or they limit their study to a single state. There has been, generally, a reluctance to compare conflicts and state failure systematically within the region, across its different units. This is despite the significant variance in levels of violence and political order between states within the region.1 As it turns out, states in the region differ not only in terms of the level of political order, but also in terms of the key characteristics and capabilities of state institutions and in their relationships to society. Painting the region with a single broad brush, then, hides much of the finer difference between states that could exist at the meso and micro levels, and which may say a lot about the political outcomes of conflict and breakdown. And seeking to provide generalizations about politics in the region based on studies of specific cases may not be very helpful, as inferences drawn from them may actually have little travelling capacity beyond the particular state and society in question. As has been observed, such ‘idiographic’ studies also provide little ground for disproving alternative generalizations about the phenomenon being studied (Lijphart 1971: 691).

I think there are crucial lessons to be learnt about politics and breakdown in Northeast India by undertaking a comparison of the politics in the region. Comparison has helped me highlight the causal factors that could be determining the divergence in political outcomes between states in the Northeast, while holding constant other factors (p.267) that impinge on the region as a whole. It is true that comparison involves simplification and a reduction of the social reality to a set of manageable variables. This also means that the phenomenon might have to be taken out of its specific context. As a result, the inference so derived may lose its relevance to ground reality. As a way out, so as to be able to provide at least a partial generalization of the reasons behind the violence and collapse in Northeast India and still keep the analysis firmly grounded in the specific contexts of the case, I have used the method of comparative case study research, conducting a ‘controlled comparison’ of a predetermined set of variables across my cases.2 This has, I hope, helped me to strike a balance between the depth and thickness of understanding of purely ‘idiographic’ studies and the inclusiveness of the ‘nomothetic’.

Case Selection

Much of the strength of the inferences derived from comparative research, however, depends on the choice of cases (Geddes 2003: 129). The choice of the cases itself depends on the incidence of the phenomenon being studied and the research strategy chosen. The sharp variation in the incidence of violence and disorder in Northeast India means that the strategy best suited for this exercise is the ‘method of difference’, that is, selecting cases with variation on the dependent variable. The best choice here is of cases that are most dissimilar on the dependent variable and, taken together, represent the full range of possible outcomes.

For our study of violence and disorder in Northeast India, we must take the state as the unit of analysis for the simple reason that under India's federal polity, states are the sub-national units of political and economic governance. Also, most local-level data on politics and economy in India are available at the level of the state. For a comparative study of violence at the sub-national level, then, the state is the most viable unit of analysis, allowing for comparison of trends concerning public policy and economy, as well as issues that lie in the realm of state-building and institutional capacity, and collective identity formation and mobilization.

Of the seven states in the region, Assam, Manipur, Nagaland, and Tripura show the greatest propensity to violence. Of these, Manipur stands out as the case where violence levels have not only been sustained (p.268) over time, but have also showed signs of becoming institutionalized in ways that have made attempts at resolution difficult. Manipur's is perhaps the worst case scenario of political disorder in terms of the incidence of violence and conflicts, its spread, complexity, and scale, although Assam, Tripura, and Nagaland too have demonstrated similar characteristics at different points in time. The other state that stands out in the region is Mizoram. Organized violence has mostly been absent in Mizoram. That state has also successfully managed crises and has avoided collapse. Thus, while Mizoram presents the counterfactual to the general situation of violence and breakdown in the Northeast, Manipur, with its severe breakdown and spiral of violence, could be said to represent the crisis of the region at its most aggravated.

Being states from the same region of India, the choice of Manipur and Mizoram allows for exercising control, thus helping make appropriate comparisons—their political, economic, and social isolation; late colonization; and comparable efforts at post-Independence state- and nation-building from the national capital. The two also face similar socio-political and economic environments, and a national leadership that has used comparable sets of policies and instruments to respond to the political and economic dynamics in the two cases. Societies in the two states also show similarity in terms of ethnic fractionalization, linguistic multiplicity, and ongoing attempts—some successful, others not so—with constructing collective identities. Use of these control variables helps direct our attention rather to the dissimilarities in institutional arrangements in the two cases, specifically to difference in state capability and in state–society relations, determining the dependent variable—breakdown and collapse or its avoidance.

Could I have chosen my cases in any other way? It could be argued that a comparison of Mizoram and Nagaland would allow for better control, especially on account of Nagaland's all tribal and all-Christian population, and its late experience with the state system, characteristics that it shares with Mizoram. However, the two cases hint at, at least a priori, less variance on my principal causal variable—state capability—than do Mizoram and Manipur. Perhaps a comparison with the politics of Nagaland could provide useful insights into another facet of the politics of the Northeast—identity construction and mobilization. (p.269) Whilst both Mizoram and Nagaland are made up of a plurality of groups and communities, all autonomous of each other to begin with, the drive to construct and formalize a Mizo identity has been more successful than has been the attempt to forge a Naga ‘nation’. But we will leave that comparison to some future date. A comparison of Mizoram and Tripura could also be productive. Levels of violence in Tripura have been high and attempts by the political leadership there to respond to tribal demands have been more focused. But the similarities here in the control variables is even less, with the immigrants issue dominating everything else in the politics of Tripura. Ultimately, a ‘method of difference’ research design implies that it is cases that are the most dissimilar on the dependent variable—incidence of violence and disorder in this instance—that must be chosen for the study. Manipur, with its widespread violence, the picture of severe collapse, and of the helplessness of its leaders to do anything about it, is a natural candidate as is Mizoram where violence is largely absent. In sum, the choice of Manipur and Mizoram for the comparison of conflict and breakdown in the Northeast provides a fine balance of maximizing the variance in the operative variables and minimizing differences in the control variables. Moreover, given my logistic and time constraints, it would have been difficult for me to compare more than two states with any degree of adequacy.3

The methodological task I then set for myself was to conduct a qualitative comparison of violence and breakdown between Manipur and Mizoram. My principal causal variable for the investigation is state capability, which determines not only the extent of violence, but also the state's ability to act as the central entity in society, dominating the provision of security, public services, and rule systems for all citizens. My intervening variables, which have a direct bearing on my principal causal variable, are: the historical process of state-making in the two cases; the nature of social structure; political, economic and social institutions structuring relationships between groups in the two cases; political organization, the roles and interests of their leaders, and the strategies they use to create and mobilize collective identities; and, finally, the capacity of the agencies of the state vis-à-vis that of non-state actors to provide key political goods. Any comparative explorations of political outcomes must factor in both internal and external factors. The study, therefore, encompasses all three levels of (p.270) analysis: survey of macro-economic data, and review of policies and strategies at the national level, and the study of macro structures of conflicts and nationalism, combined with meso-level analyses of institutions and social structures in specific cases, and with context-specific micro focus on interests and motives of actors, as well as of changes and variations in institutional capabilities across space and time.

Data Collection

Data for this study were collected in the course of fieldwork I undertook in Manipur and Mizoram—and briefly in Shillong, Guwahati, and New Delhi—between May 2004 and February 2005. In my original research design I had planned to administer a survey questionnaire to investigate the structure of civil society in the two states, examining associational activity within and across different groups. By exploring the extent, nature, participation, agenda, and activity of ‘modern’ as well as ‘ascriptive’ organizations in the two states, I had hoped to explain the variance in the region's political outcomes. As the first results of my survey confirmed, exploring the structure of civic life meant falling into the trap that most similar voyages had been victims to: ‘endogeneity’, that is, the fact that the presumed causes of the phenomenon were actually the consequences.4 As I pushed on with my initial exploration of the nature and extent of inter-community engagements in Manipur as against Mizoram, I began to recognize the pitfalls of differentiating between cause and effect, that is, between associational life that may lead to violence and collapse, and violence (among other factors) affecting the nature and extent of associational life itself. The segmented habitation pattern in Manipur could be as much the effect of population polarization caused by past violence (the Naga—Kuki clashes, for instance) as the cause of the poor cross-group engagements in society precluding any associational activity across groups, making those societies vulnerable to further violence and breakdown. And as the Mizo case showed, where community identities are themselves dynamic and nebulous, and where associational life is overwhelmingly based on ethnic lines, there are the additional problems of, first, determining whether the engagement is inter or intra-community in nature, and, second, measuring the extent of this supposedly civic engagement.

(p.271) I realized a better manner of engaging with inter-community processes leading to my dependent variable would be to engage with the more complex set of dynamics around identity construction and mobilization, grounded firmly in the analysis of the history of state-making and its impact on how groups relate to each other and to state institutions. This entailed visiting census and anthropological offices to collect historical and current statistics on habitation and settlement patterns of communities, and changes therein, and public (and sometimes private) archives to survey records of administrative notes, memos, communications, and ethnographic accounts and gazetteers. Subjects I was interested in were the penetration of the colonial state system in Manipur and Mizoram; how colonial administrators established the rule systems; what patterns of relationships they maintained with local ‘strongmen’; and how they managed social relationships.

I accessed government departments to obtain information on their plans, policies, projects, and priorities, as well as on the impact of their interventions in terms of providing public goods: security, development, economic management, and management of intercommunity relations. These visits also helped me better understand the working of different agencies of the state, their relationship to non-state actors, and the variety of ways in which leaders respond to contestations and crises. The agencies I was particularly interested in were the police and home departments, finance and planning departments, the development departments—local administration, rural development, education and health, and that in charge of essential supplies. Together, these departments provide the essential complement of functions of the state. This research involved interacting with a variety of state actors, from chief ministers down to middle- and lower-rung officials in the bureaucracy, to conduct in-depth interviews, and to access budget and plan documents, inquiry reports, programme evaluation studies, and orders, as well as published documents of government departments. Much of the macro-economic data and those on social well-being are available from the websites of national and state-level agencies. I have made use of these freely.

Survey of newspaper clippings and reports representing the working of government agencies and civil society organizations, and those involving inter-community dynamics, helped me to put government programmes and projects in their context, and also perform a (p.272) confirmatory test of the veracity of claims made by state agents. Newspapers also provided me with a wealth of data on political contestations and mobilizing strategies of political actors. My attempt here was to cover important periods in the political history of Manipur and Mizoram—such as outbreaks of violence, periods of ethnic clashes, progress of peace talks, and to focus on periods when political contestations and mobilizations are generally high—such as during election campaigns. This survey was limited to English language newspapers published from the state capitals, and to regional and national newspapers from Guwahati, Kolkata, and New Delhi.

A key aspect of my research was trying to understand the nature of the society and its relationships to the state. This part of my work involved studying organizations in ‘civil society’, the many public organizations, particularly tribal authorities, community groups, youth, and church-based organizations, as well as groups that have taken to arms, to understand their composition and constitution, their activities, and, most importantly, their ability to influence political outcomes, provide services to their constituents, and demand obedience. This investigation was informed by in-depth unstructured interviews with senior leaders of key social organizations of the main ethnic groups in the provinces. My explorations also took me to offices of different political parties in the two states—both national and regional—to understand their ideologies, their organization, and their mobilizing strategies, and how they linked up with state and social actors to channel resources and influence political outcomes. Also fruitful for getting a grip over political contestations and the working of the state was surveying records of debates in state assemblies and the national Parliament.

Understanding social reality also meant my exploring how individual citizens relate to their surroundings—especially to agencies of the state—what their hopes and aspirations were, how public perceptions have been forged, and what motivates people to act the way they do in their understanding of the inter-community dimension. This entailed my travel to districts inhabited by different communities and interacting with a variety of informants—at both elite and subaltern levels. In Manipur I wanted to capture dynamics that undergird both the alienation of peripheral communities as well as inter-community contestations within. While the central Valley region (p.273) is inhabited overwhelmingly by the Meteis, the Hills in the north are dominated by the Nagas. Kukis and their related tribes dominate the south, bordering Mizoram. I visited Tamenglong district in the north, home to the Zeliangrong Naga community, and also the district popularly know as the ‘most backward’ in the state, constantly on the boil due to the ‘stepmotherly treatment’ its inhabitants feel they receive at the hands of the state government. Tamenglong was also the district that experienced large-scale violence between Nagas and Kukis during the clashes in 1992–6, which triggered extensive internal migration of people, mostly Kukis, to Churachandpur district in the south. To understand intra-community dynamics within the Kuki-Chin group, I visited Churachandpur also to understand why the district, inhabited by the same groups that inhabit Mizoram and where they have forged such unity, has suffered repeated conflicts and conflagration.

In Mizoram ‘Mizos’ inhabit the central regions around Aizawl and Lungleh—its principal towns—while minority communities dominate the fringes: Maras in Saiha and Pawis in Longtlai, both in the south; Chakmas on the borders with Bangladesh in the in the east; Brus on the borders with Tripura in north-west; and Hmars in the north. While majority of Hmars and Pawis are happy being ‘Mizo’, Maras, Chakmas, and Brus have maintained their separate identity. To get a flavour of how non-Mizos look at Mizo mobilization and how they relate to the state, I thought it essential to study in detail one example from either category. Mara being a dominant non-Mizo community with heightened sense of separatism, albeit peaceful, was an instructive case to study. On the other hand, the separatist Hmars posed a curious challenge: while the majority counted themselves as Mizo, a section of the Hmars had rebelled against Mizo domination. And yet they had returned to the negotiating table having settled for a peaceful end to the armed movement. The Mara study required interacting with Mara public organizations and political leaders to understand elite interests and their mobilizing of popular sentiments. As for the Hmars, besides engaging with Hmar social organizations and politicians, I visited Bilkawthlir and Vairengte villages. The first is the biggest Hmar village in the Sinlung Hills Development Council, a special development area for the Hmars. Most people have retained their Hmar ethnicity and cultural pride. The latter, a mixed habitation town (with Hmar (p.274) majority) on the borders with Assam, was a useful laboratory to see how different communities related to each other.

All along I have used secondary sources extensively as sources of information and insights, as well as tools to interpret and triangulate data already collected. I have drawn from published works such as ethnographic accounts and historical studies to understand the nature of social structure, and the processes and outcomes of elite contests, and for historical interpretations of the processes of state-making and society formation. I have also used online sources of information available on numerous websites.

Sources of Data


  • Manipur State Archives, Imphal

  • Mizoram State Archives, Aizawl

  • National Archives, New Delhi

  • British Library, India Office Collection, London


  1. 1. In Manipur:

    • Imphal Free Press, Imphal.

    • Sangai Express, Imphal.

    • Manipur Mail, Imphal.

    • Resistance, Imphal

  2. 2. In Mizoram:

    • Highlander, Aizawl

    • Newslink, Aizawl,

  3. 3. National/Regional:

    • North East Tribune, Guwahati Telegraph, Kolkata and Guwahati

    • Ananda Bazaar Patrika, Kolkata

    • The Times of India, New Delhi,

    • Hindustan Times, New Delhi

    • Pioneer, New Delhi

    • The Hindu, Chennai

    • The Indian Express, New Delhi

Government Departments

  1. 1. Interviews with (state government officials)

    (p.275) Senior officials of the Home/Police Departments, commanders of Army formations deployed in the states; heads of Finance, Planning, Rural Development, Local Administration, Food and Civil Supplies, Health, Education, and Personnel Departments.

  2. 2. Reports and data from the following state government departments:

    State Assembly, Planning, Finance, Economics and Statistics, Local Government, Police, Food and Civil Supplies.

  3. 3. Reports and data from the following national/regional agencies:

    • Ministry of Home Affairs, New Delhi

    • Ministry of Finance, New Delhi

    • Ministry of Development of Northeastern Region, New Delhi

    • Election Commission of India, New Delhi

    • Northeast Eastern Council (NEC), Shillong

    • Guwahati High Court, Guwahati

Political Parties

Interviews with senior members of the following parties.

  1. 1. In Manipur:

    Congress party, Communist Party of india (CPi), MPP, FPM, NNP.

  2. 2. In Mizoram:

    Congress party, MNF, PC, Zoram National Party (ZNP), MDF. Also accessed accounts of their agendas, their election manifestos, constitutions, and programmes.

Social Organizations

Interviewed senior leaders of the following organizations:

  1. 1. In Manipur:

    AMSU, AMUCO, and UCM, among the Meteis; the ANSAM, Rongmei Council (RC), UNC, and ZU; besides church-based ones such as Manipur Baptists Council (MBC), and Tangkhul Baptist Church of Imphal (TBCI) among the Nagas, and the Kuki Baptist Church (KBC), Kuki Inpi, and KSO among the Kukis. It also involved interactions with ex-cadres of armed organizations such as the National Socialist Council of Nagalim (NSCN-IM), Kuki National Army, and UNLF; with human rights organizations and with cultural organizations as the Sahitya Academy (Academy of Letters).

  2. (p.276) 2. In Mizoram:

    YMA, MZP (Mizo Students' Union), Presbyterian Synod, the Baptist Church, the Catholic Church, Young Chakma Association (YCA), YMA, Hmar Students' Association (HSA), Mizo Academy of Letters, Human Rights and Law Network of India (HRLNI), Peoples' Union for Civil Liberties (PUCL), Mizoram Journalists' Association (MJA), All Mizoram Bar Association (AMBA), Village Council Presidents' Association (VCPA), and ex-cadres of Mizo National Front and the HPC.

Online Data Sources

Comptroller and Auditor General of India:


Crisis States Research Centre:


Election Commission of India:


E-pao (news portal from Manipur):


Government of Manipur, official website:


Government of Mizoram, official website:


Guwahati High Court


Institute of Conflict Management, New Delhi:


Development Research Centre on Citizenship, Participation and Accountability, Institute of Development Studies:


Institute of Peace and Conflict Studies, New Delhi:


Kanglaonline (news portal from Manipur):


Ministry of Development of Northeastern Region:


Ministry of Finance, Government of India:


Ministry of Home Affairs, Government of India:


Ministry of Law, Legal Affairs, and Justice, Government of India:


National Crime Records Bureau, Government of India:


North East Tribune, Guwahati, Online edition:


Planning Commission, Government of India:



(1.) The notable exception to this is Sajal Nag's (2002) account of the trajectories of political violence in three states in the Northeast—Manipur, Mizoram, and Nagaland. But rather than attempting a systematic comparison of the three cases using a common set of variables, the volume turns out more as a compilation of discrete single case studies.

(2.) For a discussion on the comparative case study research design, see King et al. (1994: 45). Alexander George (1979: 61–2) has argued that for comparative case studies to be able to provide explanatory or descriptive inferences, they must be made more systematic, something that can be achieved by observing discipline in the collection of data. He calls this controlled method of conducting comparison ‘structured, focused comparison’.

(3.) This number of cases also satisfies the rule of case choice that says that one must have at least one more case than the number of hypothesized causes for reasonable estimates of relationships to be made (Geddes 2003: 135). My principal causal variable is state capability.

(4.) The classic example of this is, of course, Robert Putnam's (1993) work on associational life in Italy. Similar criticism has been levelled at Ashutosh Varshney's (2002) account of the variance in communal violence in India, although he has tried to demonstrate that violence—the dependent variable—in his case studies followed and did not precede breakdown of associational life—the independent variable.