Aushotosh Varshney PDF

Trustees of Princeton University
Ethnic Conflict and Civil Society: India and Beyond Author(s): Ashutosh Varshney Source: World Politics, Vol. 53, No. 3 (Apr., 2001), pp. 362-398 Published by: Cambridge University Press Stable URL: http://www.jstor.org/stable/25054154 . Accessed: 09/09/2013 05:19
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Cambridge University Press and Trustees of Princeton University are collaborating with JSTOR to digitize, preserve and extend access to World Politics.
http://www.jstor.org
This content downloaded from 180.149.53.102 on Mon, 9 Sep 2013 05:19:48 AM All use subject to JSTOR Terms and Conditions
ETHNIC CONFLICT AND CIVIL SOCIETY

India and Beyond
ByASHUTOSH VARSHNEY*
scholarly work has been done on the topics of civil society MUCH
but no systematic attempt has yet been made of my recent, India-based conclusions proj ect,2 supplemented materials, suggest that the links be by non-Indian are tween out for serious civil society and ethnic conflict crying attention. Does civic engagement between different ethnic communi ties also serve to contain ethnic conflict? Does interethnic engagement differ from intraethnic engagement from the perspective of ethnic con and ethnic to connect conflict, the two.1 The
role do civic organizations play in times of ethnic tensions are not and why? These and relevance, questions simply of academic to be systematically researched. Given that violence marks they have yet our multiethnic research well have societies, many may great practical ifwe can sort out some key relationships. meaning article argues that there is an integral link between This the struc ture of civic life in amultiethnic society, on the one hand, and the pr?s
flict? What
article was presented at the annual meeting of the American Political Science Association, at the Boston, 1998, and various versions were also presented in seminars and symposiums following universities and other institutions: Illinois Columbia, Harvard, Chicago, (Urbana-Champaign), (Ann Arbor), MIT, Notre Dame, Oxford, Texas (Austin), Toronto, Uppsala, Yale, theWorld Michigan on the this article is part was Bank, and the Ford Foundation (Delhi). Work larger project of which in International supported by the SSRC-MacArthur Program Security, the U.S. Institute of Peace, the Samuel P. Huntington and the Ford Foundation. For comments, Fund of Harvard University, criti Iwould like to thank Hans Blomkvist, Kanchan Chandra, Partha Chatterjee, cisms, and suggestions, Elise Giuliano, Donald Horowitz, Gary King, Arul Kohli, David Laitin, Scott Pradeep Chhibber, Bhikhu Parekh, Elizabeth Perry, Robert Putnam, Sanjay Mainwaring, Anthony Marx, James Morrow, van Srivasatava, Alfred Jack Snyder, James Scott, Manoj Reddy, Susanne Rudolph, Stepan, Steven reviewers of Evera, the lateMyron Weiner, Yogendra Yadav, Crawford Young, and three anonymous this journal. 1 This is not to say that community life within ethnic groups has not been studied as part of civil so to Be What ItMeans ciety. A striking recent example, though not the only one, isMichael Walzer, American? life can be called (New York Marsilio, 1992). The view that ethnic (or religious) community in Section I. civic is, of course, contested by many. The debate is summarized 2 in India (New Haven: Yale Ashutosh Varshney, Ethnic Conflict and Civic Life: Hindus andMuslims University Press, forthcoming).
* This
WorldPolitics 53 (April2001), 362-98
ETHNIC CONFLICT& CIVIL SOCIETY

ence or absence of ethnic violence, on the other. To First, illustrate interethnic different roles links, two intraethnic ethnic ethnic interconnected networks are made.
363
these and in
inter tensions, are of if communities but peace, agents organized with other com lines and the interconnections only along intraethnic or even nonexistent, are very weak munities is then ethnic violence conflict. networks are quite evance likely. The specific conditions under which this argument may not
arguments of civic engagement Because they build bridges
play very and manage
hold will be theoretically specified toward the end. Their empirical rel
can be ascertained civic networks, Second, broken down into two other tinction first is based on whether further research. only with and interethnic, both intraethnic types: organized civic interaction can also be
and quotidian. This dis is formal or not. I call the
associationalforms engagement. Business
clubs, film clubs, litical parties are examples forms of engage of the former. Everyday ment consist routine of simple, interactions of life, such as whether communities visit each other, eat together families from different regu gether
and the second of everyday forms of engagement associations, organizations, professional reading and cadre-based po sports clubs, NGOs, trade unions,
larly, jointly participate in festivals, and allow their children to play to

in the neighborhood. if robust, Both forms of engagement, their absence or weakness peace: contrariwise, opens up space promote for ethnic violence. forms turn Of the two, however, the associational con out to be sturdier when than everyday engagement, especially
fronted with attempts by politicians
to polarize people along ethnic
acts as a serious con lines. Vigorous associational life, if interethnic, even when is in their straint on politicians, ethnic polarization political cut across ethnic bound interest. The more the associational networks aries, the harder it is for politicians The article also briefly considers developed in place movement in India. Much in the to communities. polarize civic organizations how interethnic civic structure was moment a new form during of politics put the freedom emerged
of India's associational
a transformative 1920s, the British, when against
under the leadership ofMahatma Gandhi. After 1920 the movement had two aims: political independence from the British and social trans formation of India. Gandhi argued that independence would be empty
unless Muslim uplift, India's to Hindu attention addressed, drawing women's of untouchability, self-reliance, unity, the abolition so on. The associa and tribal uplift, labor welfare, prohibition, social evils were
tional structure of India before Gandhi had been minimal, but the Gandhian shift in the national movement laid the foundations of
364 Indias associational a host
WORLD civic order.
POLITICS
In the process, between the 1920s and came into 1940s, organizations being. we draw a distinction Historical therefore, reasoning, requires that between causation. The role of intercommu and underlying proximate of new has been crucial for peace at a proximate level. Taking the long view, however, the causal factor was a transformative shift in in national politics. Once put the civic place by the national movement, structures took on a life and of behav their the own, constraining logic run. ior of politicians in the short to medium as follows. The first section clarifies The article is organized three nal civic networks key me third section tion ciety basic terms whose to discover section meanings the relevance summarizes how are not self-evident: ethnicity, ethnic conflict. con The
flict, and civil society.The second section deals with the puzzle that led
of civil society for ethnic the puzzle was resolved in support and presents made.
the arguments that can link ethnic conflict and civil society.The fourth
presents empirical evidence of the arguments
The fifth section considers causation and endogeneity. The final sec
summarizes but also the implications a suggests possible about interethnic of the project for studies of civil so set of conditions under which the and intraethnic engagement is un
argument
likely to apply. I. Clarifying

The things terms
Concepts
and Terms
specify the meaning one is using.

There are two distinct
and "civil society" mean different "ethnic," "ethnic conflict," one needs to to different To preempt misunderstanding, people. inwhich ways of the term, construal the term "ethnic"
is interpreted. "ethnic" groups mean "racial" or sense term in is is under This the which the widely "linguistic" groups. in in For example, stood both India and elsewhere. popular discourse, Indian scholars, for politics and conflict based on religious groupings, In the narrower since the time of the British have used and politicians bureaucrats, term "communal," not "ethnic," the latter term primarily reserving the for
linguistically or racially distinct groups.

There all conflicts gion, a second, broader definition. As Horowitz is, however, argues, reli based on ascriptive group identities?race, language, or caste?can eth be called ethnic.3 In this umbrella usage, tribe,
forms
3 For an analysis of why, on the basis of a myth of common ancestry, ethnicity can take so many see Donald Horowitz, Ethnic Groups in Conflict (Berke (language, race, religion, dress, diction), now of California Press, 1985), 41-54. One might add that this definition, though by ley: University
365
nie conflicts range from (1) the Protestant-Catholic conflict inNorthern conflict in India to (2) black-white Ireland and the Hindu-Muslim conflict in the United States and South Africa, (3) the Tamil-Sinhala conflict in Sri Lanka, and (4) Shia-Sunni troubles in Pakistan. In the narrower construction of term, (1) is religious, (2) is racial, (3) is lin
and (4) is sectarian. In the past the term "ethnic" guistic-cum-religious, would often be reserved for the second and, at best, third conflicts but
would not be extended to the first and fourth.

of the broader usage reject such distinctions, arguing Proponents that the form ethnic conflict racial, takes?religious, linguistic, or relative not seem to alter its in tribal?does duration, intensity, core of on the in and cultural Their tractability. emphasis ascriptive
the conflict,
nonascriptive indeed have of spective tends gion class conflict
and they distinguish

and economic an economic basis, but
it primarily
conflict. is not that
from the largely

conflict may Irre
core of class
Ethnic
its core feature.
internal to define tends this
race, language, sect, or reli differentiation, the politics of an ethnic group. Contrariwise, on the whole to be economic, but if the class into class
which
one is born is also the class in which
one is trapped until
is true for large numbers of people?then class con on overtones. it is now well flict takes Horowitz, Following ascriptive not to ethnic understood that the latter characteristics systems apply death?and such as America the but to ranked ethnic systems, during general caste and India's sys during apartheid, period of slavery, South Africa tem. Ranked unranked eth ethnic systems merge and class; ethnicity in systems do not. one the The add, is also increasingly becoming larger meaning, might even not true in the social sciences, if that is of standard meaning yet use the term "ethnic" in this broader sense. In activism. I and will politics nic
otherwords, Imay distinguish between communal (that is, religious) and linguistic categories, but Iwill not differentiate between those that are
communal set to which is and ethnic. Ethnicity religion, race, simply the as subsets in this definition. sect and belong language, our second term, have a Does "ethnic conflict," uniquely acceptable ethnic some violence ethnic and ethnic conflict conflict. Such conflation is unhelpful. but it may not
meaning? On thewhole, the existing literature has failed to distinguish

between
In any ethnically plural society that allows free expression of political

demands, ismore or less inevitable,
not without If all ascriptive divisions can be the basis of ethnicity, can widely accepted, is problems. the landed gentry or women's groups be called ethnic? So long as we equate ascriptive identities with ethnic identities, there is no good answer to such questions.
366
WORLD POLITICS there are different ethnic groups

likely to be conflicts is violent or over resources,
necessarily lead to violence.4 When

that are free identity, The real issue organize, and patronage, to there are policies. is whether ethnic
conflict
If ethnic protest institutionalized channels. polity's in tionalized form?in in bureaucratic corri assemblies, parliaments, as on and nonviolent mobilization streets?it is the conflict but dors, a situation
via the waged an takes institu
not violence. Such institutionalized conflict must be distinguished from

in which takes violent forms, rioting breaks out on protest extreme the streets, and in its most form civil war ensues or pogroms are initiated some ethnic groups with the complicity of state au against thorities. Given how different for in these outcomes are, explanations stitutionalized conflict may not be the same as those for ethnic riots, on on the other. Ethnic and for pogroms and civil wars, as an insti be conceptualized peace should, for all practical purposes, tutionalized and resolution of ethnic demands and conflicts: channeling as an absence as an ar not absence The world would of violence, of conflict. a we could eliminate con be if ethnic and national guably happier place era does not flicts from our midst, but such a postethnic, postnational seem to be in the near term. Indeed, many in the offing postmodern even in richer societies, are on of conflicts, taking ethnic forms grounds authenticity of living styles and distinctiveness of expression.5 to critical sciences, the family, scrutiny. Ac "civil society" on the one between the one hand,
Though highly popular and much revived in recent years, the con
cept of civil society also needs to be subjected to conventional in the social notions cording refers to that space which (1) exists between hand, and the state, on the other, not (2) makes
interconnections
individuals or families possible, and (3) is independent of the state.

Many all of the existing definitions also suggest two more in be associations that at space organized requirements: tend to the cultural, and of citi needs the social, economic, political not and that the associations be modern and voluntaristic, zens; to the first be trade unions would ascriptive. According requirement, though that the civic
part of civil society, but informal neighborhood associations would not.

4 Indeed, such conflict may be inherent in all pluralistic political systems, authoritarian or demo an cratic. Compared with authoritarian systems, a democratic polity is simply more likely to witness ethnic groups into long open expression of such conflicts. Authoritarian polities may lock disaffected a coercive contain a society, but periods of political silence, giving the appearance of well-governed ment of such conflicts also runs the risk of an eventual and accumulated outburst when an authoritar ian system begins to liberalize or lose its legitimacy. 5 Ronald Inglehart, Modernization and Postmodernization (Princeton: Princeton University Press, 1997).
ETHNIC CONFLICT 6c CIVIL SOCIETY

Following associations Should the second would
367
clubs and parent-teacher requirement, philately be civic, but a black church or an association of Jews
active on behalf of Israelwould not.

we agree with the latter two proposals? Can nonassociational or to civic also be called associations, space part of civil society? Must a constitute "modern" kind?voluntaristic and part of civil society, be of rather than ascriptive and based on ethnic affiliation? crosscutting, are to The modernist civil of attributed origins society originally recent theoretical In formulations.6 years, Hegels nineteenth-century however, it has often been suggested that the revival of amodernist no
tion of civil society derives from debates in Eastern Europe and the
translation English lic Because Sphere.1 who have explored of Habermas the concept it in recent s Structural of civil times.8 Pub Transformation of the so society has been important the analytic work
to the field of political philosophy, it ismostly political philosophers

In comparison,
on civil society in the more empirical fields of the social sciences has
not been as voluminous,9 though the need for it should Only by systematic empirical nonassociational forms of civic tions exist As nitions and forms as more attributed than investigation life can we determine society be quite of the associational whether clear. and
the func literature defi
to civil theoretical
in the normative
an illustration of civil whose
simply of the modernist consider on civil
propositions. biases of the conventional have
society,
the theoretical
Gellner, influential.
society writings argues Gellner, "Modularity," a traditional defines "segmentalism" society.10 By modularity,
of Ernest arguments as well as been plentiful "makes civil society," whereas he means
6 For the early history of the idea, see Adam Seligman, The Idea of Civil Society (Princeton: Prince Press, 1992). University 7 The Structural Transformation J?rgen Habermas, of the Public Sphere: An Inquiry into the Category of trans. Thomas a Burger and Frederic Lawrence (Cambridge: MIT Press, 1989). For Bourgeois Society, debate built around the publication of the English translation, see Craig Calhoun, ed., Habermas and the Public Sphere (Cambridge: MIT Press, 1994). 8 See also Charles Taylor, "Modes of Civil Society," Public Culture 4 (Fall 1990); Michael Walzer, "The Idea of Civil Society," Dissent 38 (Spring 1991); and Jean Cohen and Andrew Arato, Civil Society and Political Theory (Cambridge: MIT Press, 1992). 9 The debate generated by Putnam's work is finally leading to empirically based scholarship. See, of theWeimar inter alia, Sheri Berman, "Civil Society and the Collapse Republic," World Politics 49 on social (April 1997); the special issue capital o? American Behavioral Scientist 39 (April 1997); and for Development and Michael Woolcock, "Social Capital: Implications Deepa Narayan Theory, Re search and Policy," World Bank Research Observer 15 (August 2000). 10 Ernest Gellner, "The Importance of Being Modular," in John Hall, ed., Civil Society: Theory, His 1995). This article is a good summary of a large number of tory, Comparison (Cambridge: Blackwell, on civil in both the reflective and the activist mode. Many of these Gellner's writings society, written some essays, have been put together in Gellner, Conditions writings, including polemical of Liberty: Civil Society and Its Rivals (New York: Penguin Press, 1994). ton
368
WORLD POLITICS
or the ability to transcend traditional and associ ascriptive occupations a ations. Given and modern education and given secular, multipurpose,
also the objective availability of plentiful as well as changing profes

from man can move in times, modern opportunities posttraditional one to another, one to another, one association occupation place to another. In contrast, traditional man s occupation and place were de sional
termined by birth. A carpenter in traditional society,whether he liked it

or not, would and all his kinsmen would be carpenters. be a carpenter, not He would in associations; be involved also and if he were, generally an the association would most be ascriptive guild of carpenters. likely An agrarian society, argues Gellner, might be able to avoid the tyranny is because of the state. That the power of the state could not reach all nature of pro of a traditional segments society, given the decentralized duction the low level of communication and the structure, technology, not self-sufficient character of each But that does segment. relatively mean that such a society would be "civil," for instead of the "tyranny of con it would the state," society, "tyranny of cousins." Civil experience is not only modern but also based on strictly voluntary cludes Gellner, associations between the state and family, not on ethnic or religious considerations.11 Such number claims of studies can be First, a remarkably challenged. large empirically associations show that ethnic and religious combine all Christians have to be members of a and choice. Not
ascription in a given church of a black church. town, nor all blacks members it that ethnic associations has also been documented Moreover, widely can as in demo such "modern" functions, participating perform many cratic group kinsmen A politics, to enter setting newer up funds professions,
to encourage members of the ethnic of ethnic and facilitating migration
and modern education.12 occupations can be raised with to the respect requirement objection of the developing that associations be formal. In much world, especially in the countryside do not exist. and small towns, formal associations or activities does not mean, that civic interconnections That however, similar are absent. If what is crucial to the notion of civil society is that families
into modern
and individuals connect with other families and individuals? beyond

11 For a similar argument, see Edward Shils, The Virtue of Civility (Indianapolis, Ind.: Liberty Fund,
see uses of and Susanne For pioneering work on the modernist ethnicity, Lloyd Rudolph of Chicago The Press, 1967); and Myron (Chicago: University Rudolph, Modernity of Tradition Sons of the Soil (Princeton: Princeton University Press, 1978). Weiner,
1997). 12

their homes?and interference must talk about matters it seems
369
of the state, then take place only in "modern" or in the tradi whether such engagement takes place in associations on the tional sites of social get-togethers depends degree of urbaniza as on as tion and economic well the nature of the political development, to formal Cities tend have associations; system. villages make do with sites and meetings. informal Further, systems may political specify which groups
of public relevance without the too to insist that this far rigid associations. Empirically speaking,
or may have access to formal civic spaces and establish ones may not. and which pro ganizations Nineteenth-century Europe access to a whole vided the propertied classes with range of political instruments for and institutional of interest articulation; trade unions were slower to arrive. Some in the commentary of the spirit of these remarks is conveyed Habermas's distinction between the "lifeworkT'and "sys by Transformation the distinction indicated In its Public original of the Sphere. a radical between the sig rupture and that of interaction made possible
workers
generated tem" in The Structural
formulation, interaction nificance of everyday to Habermas, and organizations. The latter, according by institutions was associated with amodern interaction made public sphere. Everyday life, but organized sphere in much interaction made history.13 The new history of pop
ular struggles launched by those not formally admitted to the public

and America?women, Europe the limited peasants, workers, minorities?suggests utility of the origi more recent in nal Habermas distinction.14 his Indeed, positions, his earlier, radical distinction.15 Street Habermas has all but dropped corner can now be viewed as a serious civic form activity civic sites are not available?whether ized and institutional The social North groups. particular is not of course, point, that formal associations ifmore organ or generally do not matter. of nineteenth-century
to some
One of the arguments of this paper is that they do. But at least in the
and cultural America, are different from those of Europe and settings that if not more generally, of activity rather the purposes
13 in Harry Boyte, "The Pragmatic Ends of Popular Poli See the brief but thoughtful discussion tics," in Calhoun (fn. 7). 14 Pen The Making of the English Working Class (Harmondworth: Starting with E. P. Thompson, now exist. For a quick review of how they relate to guin Press, 1968), many such historical works by see in Nineteenth Politics Habermas, Century Mary Ryan, "Gender and Public Access: Women's in the Nine and Geoff EUy, "Nations, Publics, and Political Cultures: Placing Habermas America," teenth Century," both in Calhoun (fn. 7). 15 on the Public "Further Reflections (fn. 7). Sphere," in Calhoun J?rgen Habermas,
370
WORLD POLITICS
than the forms of organization should be the critical test of civic life.
Tradition to a tyranny of cousins, and is not necessarily equal capitalist does not always make civic interaction At best, modernity possible. on are ideal such dualities types or based normatively preferred visions. the cousins challenging are violated.16 Similarly,
tradition often permits speaking, Empirically norms when of reciprocity and ethics existing people joining tivities ciety in America
even capitalist modernity may be highly unsocial and atomizing,
if
instead of soap operas on TV, stay home and watch PTAs and other civic Both informal group ac organizations.17 so and ascriptive associations should be considered part of civil so
as connect build trust, encourage individuals, long they on matters and views facilitate the of of public reciprocity, exchange concern?economic, cultural, and social. That political, they may have or an different for is conflict very consequences peace entirely different matter. The latter ter for governance is an argument about what not and peace, whether type of civil society is bet civil society per se is en
dowed with benign possibilities. II. Why Civil Society? Feature of Ethnic Conflict it
A Puzzling
Civil society is a new variable for the study of ethnic conflict. How
emerged therefore, and why. Sooner as a causal requires or later, factor in research on Hindu-Muslim of what conflict my project a brief scholars explanation of ethnic
relations, sets out to do
are struck a by puzzling some ethnic diversity, (re regularity?that despite empirical places to remain whereas towns, villages) nations, manage gions, peaceful, some soci others experience of violence. patterns enduring Similarly, an in record of ethnic peace impressive suddenly explode as well. that the observer often the scholar Vari and ways very surprise an unresolved ations across time and space on the whole constitute eties with
puzzle in the field of ethnicity and nationalism.

How does one account for such variations? has been The standard research strategy, with some exceptions,18 to seek the commonalities
16 Press, 1978). James Scott, Moral Economy of the Peasant (New Haven: Yale University 17 of Civic For an argument along these lines, see Robert Putnam, "The Strange Disappearance America," American Prospect 8 (Winter 1996); and idem, "Bowling Alone," Journal ofDemocracy 6 1995). (January 18 the towering exceptions are Horowitz (fn. 3);Weiner (fn. 12); andW. Crawford Young, Among ofWisconsin The Politics of Cultural Pluralism (Madison: University Press, 1976).

across the many cases of violence. Although this approach will
371
con
tinue to enlighten us, it can give us only the building blocks of a

not a full-blown theory, theory of ethnic conflict. The this is often misunderstood, ing proposition simple, on the basis of commonalities stating.19 Suppose logic underly re and worth we find that
interethnic economic rivalry (a), polarized party politics (b), and seg regated neighborhoods (c) explain ethnic violence (X). But can we be sure that our judgments are right? if (a), (b), and (c) also exist What
in
sity of (a), (b), and (c) inX; or there is an underlying factor or con textual element that makes (a), (b), and (c) conflictual in one case but not in the other; or there is yet another factor (d) that differentiates
peace cover flictual This from violence. because It will, precisely ones. argument be a factor that we did not dis however, cases were not studied with the con peaceful of studying at what question: of analysis variance level must leads to another variance itself
peaceful
cases
(Y)?
In that
case,
violence
is caused
by the
inten
for the necessity
important methodological be studied? Should the unit towns, or villages? What
help us identify the spatial and temporal trends in violence and allow
us to choose therefore, the appropriate level for analyzing considered all reported Hindu-Muslim crucial. constitute a remarkably small portion of communal ri variance. riots project, in the country The
methodologists
be nations, states, regions, can call a large-N analysis
between 1950 and 1995.20 For purposes of identifying larger trends,

two results were First, villages
oting. Between 1950 and 1995 rural India,where amajority of Indians

in communal still Uve, accounted for a mere 3.6 percent of the deaths an urban turns out to be violence violence. Hindu-Muslim primarily riots urban India, too, Hindu-Muslim Second, within phenomenon. are account for a concentrated. cities21 highly locally Eight hugely dis in the country: share of communal violence nearly 46 proportionate
percent of all deaths inHindu-Muslim

19 For an elaborate
violence (Table 1, column 4).
argument for the need for variance in social science research, see Gary King, and Sydney Verba, Designing Social Inquiry (Princeton: Princeton University Press, Keohane, commonalities 1993). For why discovering may matter even in a world where variance exists, see on Ronald Rogowski, Social Inquiry, American Political Science Review 89 (June Symposium Designing Robert
1995).
data set was put together in collaboration with Steven I. Wilkinson of Duke University. cities, as Table 1 (column 4) shows, are Ahmedabad, Bombay, Aligarh, Hyderabad, Meerut, Baroda, Calcutta, and Delhi. The last two are not normally viewed as riot prone. But because they have had so many small riots and had some large ones in the 1950s, they are unable to escape the list of worst cities in a long-run perspective is un In a 1970-95 time series, however, Calcutta (1950-95). to likely figure and Delhi may also disappear.
20 The 21 The
Hindu-Muslim
1 Table in 28 Indian Riots
Cities
(1950-95)
Minimum Minimum Minimum Minimum
of 15Deaths in 3 Riots
over 2 Five Year Periodsb Bombay
of20 Deaths in4 Riots

over 3 Five Year Periodsc Bombay
of25 Deaths in 5 Riots

over 4 Five Year Periodsd Bombay
of50 Deaths in 10 Riots

over 5 Five Year Periodse Bombay
Total Deaths
1950-95 1,137 1,119 312 265 160 198 194 194 149 109 108 93 81 63 59 56 49 47 45 42 37 37 32 30 30 29 23 18
Ahmadabad Hyderabad
Meerut
Ahmadabad Hyderabad
Meerut
Ahmadabad Hyderabad
Meerut
Ahmadabad Hyderabad
Meerut
Aligarh Jamshedpur Bhiwandi

Surat Moradabad Baroda
Aligarh Jamshedpur Bhiwandi

Moradabad Baroda
Aligarh Jamshedpur
Aligarh
Baroda
Baroda
Bhopal Delhi
Kanpur Calcutta
Bhopal Delhi
Kanpur Calcutta
Bhopal Delhi
Kanpur Calcutta
Delhi
Calcutta
Jabalpur
Bangalore Bangalore Bangalore
Jalgaon
Sitamarhi Indore Varanasi
Jalgaon
Indore Varanasi
Jalgaon
Indore Varanasi
Allahabad Nagpur Jaipur

Aurangabad Srinagar
Allahabad Nagpur Jaipur

Aurangabad Srinagar
Allahabad Nagpur
Aurangabad Srinagar
Ranchi
Malegaon Malegaon
Godhra
aTotal number of deaths from riots for all of India, 1950-95 = 7,173, of which 3.57 per
cent of deaths took place in rural India. b Total number of deaths from riots in these cent of deaths from riots throughout from riots India cities = 4,706. is 66 per This approximately in urban riots dur of all deaths
and 69 percent cities =
ing these periods. c Total number d Total
of deaths
in these
4,359.
This
is about
61 percent 54 percent
of deaths
from riots throughout India and 64 percent of all deaths in urban riots during these periods.
number of deaths from riots in these cities = 3,887. This = is about of deaths
from riots throughout India and 58 percent of all deaths in urban riots during these periods. e
Total number of deaths from riots in these cities 3,263. This from riots throughout India and 49 percent of all deaths in urban is 45.5 percent of deaths riots during these periods.
ETHNIC As
CONFLICT
& CIVIL
SOCIETY
373
a mere a group, however, 18 percent of these eight cities represent total of the about 5 India's urban population (and percent country's of the 82 percent and rural). Put otherwise, both urban population, has not been "riot prone." urban population Given
in urban India, the large-N local concentrations such high as India's the unit of analysis. town/city clearly establishes analysis is city specific, not state specific, with state (and violence Hindu-Muslim
national) politics providing the context within which the localmechan

isms linked with violence are activated. To understand the causes of these local mechanisms. violence, we must investigate from six cities?three the selected this reasoning, project Following cities and three peaceful ones?and the list of eight riot-prone arranged vio communal each pair had a city where them in three pairs. Thus, it is rare or entirely absent. To ensure and a city where lence is endemic roughly similar Hindu-Muslim against comparing apples and oranges, control in the minimum constituted in the city populations percentages communal based on popula and Calicut?was each pair. The first pair?Aligarh and Lucknow? The second tion percentages only. pair?Hyderabad one to two of controls added percentages, previous Muslim population third pair? similarities. The cultural rule and a second of reasonable Ahmedabad came and Surat?was from the North pairs state of Gujarat, sharing violence. All communal over four million Why and was the most tightly controlled. The third came from and South. The first two the same
and culture but not endemic history, language, of these cities, at this time, have a population
of above half amillion, and the biggest, Hyderabad,

people. in demographic political proportions behavior, similarity
is ametropolis of
as the min
chosen
imum control in each pair? Both in India's popular political discourse

in theories about Muslim the size of the com
munity is considered to be highly significant. Many politicians, espe cially those belonging to the Hindu nationalist Bharatiya Janata Party who have often subscribed to the idea of "Muslim disloyalty" to (BJP), India, have argued that the demographic distribution of Muslims
makes them critical to the electoral outcomes. Muslims constitute more
than 20 percent of the electorate in 197 out of a total of 545 parliamen

30 in India. In a first-past-the-post system, where tary constituencies to win a seat in vote is often contests, multiparty percent of the enough make the Muslims these percentages electorally highly significant.22 The
22 Lloyd Rudolph Press, 1987), 196. and Susanne In Pursuit (Chicago: University of Chicago
Rudolph,
ofLakshmi
374
WORLD
POLITICS
in a given constituency, the numbers of Muslims argue politi higher cians of the BJP, the greater the inclination of mainstream par political to their sectional/communal ties to pander demands and the lower the ac to build to Hindus. therefore, for Muslims Thus, bridges on to this argument, of their based Muslims, appeasement cording a cause of communal in numbers is the and conflict large democracy, in India.23 violence incentive,
That Muslim
not an argument lim politicians reversed. The
demography has political consequences

confined also make to the Hindu nationalist a claim, demographic of Muslims the numbers to
is, however,
BJP. Leading Mus but with the causation in a city or town, Muslim source they
higher who
argue, the greater the political threat felt by the leaders of the Hindu
community, on on the part react with hostility they legitimate is the anxieties
about politics and identity.An unjustified, even self-serving opposition

of Hindu leaders, argue, of communal
hostilities.24 Both extremes of the political spectrum thus rely heavily

for their explanations. demography as well, are shared al These arguments by social scientists popular is different. their reasoning It has been argued, for example, though or a a has Muslim that when plurality, Mus city/constituency majority over centrist in lims typically prefer Muslims-only confessional parties centrist when their tercommunal Muslims support parties parties.25 is small in a town/constituency. share of the population/electorate a to seek the it rational Smaller numbers make security of large and or constituencies?where similar demo
mainstream party. powerful Can one find cases?cities
graphic distributions lead to very different forms of political behavior?

from a larger sample of such cases, the project did precisely Selecting a three pairs of cities where that. As described above, it compared rough in politi variance in coexists with similarity proportions demographic or violence. cal outcomes?peace
III. Resolving
The
the Puzzle: The Role
of Civil
Society
the two of civic engagement between local networks preexisting as communities stand out the single most proximate expla important such net nation for the difference between peace and violence. Where
23 in Sunday, July 22,1990. L. K. Advani, leader of the BJP, interviewed 24 a leader, has often made this argument Syed Shahabuddin, prominent Muslim cussions, and political speeches. 25 and Rudolph (fh. 22), 195. Rudolph
in lectures, dis

works
375
are of engagement and conflicts and exist, tensions regulated are to where endemic communal identities lead missing, managed; they can in turn be and ghastly violence. As already stated, these networks broken down everyday communal, to withstand forms two parts: associational forms of engagement and of engagement. Both if inter forms of engagement, forms peace, but the capacity of the associational promote into national-level "exogenous shocks"?such as Indias parti
tion in 1947 or the demolition of the Baburi mosque in December 1992 in the presence of more than three hundred thousand Hindu mil
itants?is What substantially higher. are the mechanisms that link civic networks and ethnic con
flict? And why is associational engagement a sturdier bulwark of peace

than everyday engagement? One can identify two mechanisms that connect civil society and eth nic conflict. First, by promoting communication of between members different civic networks often make neighbor communities, religious hood-level together Called to the peace possible. Routine and form organizations committees allows engagement in times of tension. of members to come people Such organiza of both commu
tions, though only temporary, turned out to be highly significant.

peace local and consisting
nities, they policed neighborhoods, killed rumors, provided information

between and facilitated communication administration, in times of tension. communities Such neighborhood organizations were difficult to form in cities where interaction did not cross everyday
Muslims religious lines, orwhere Hindus and

Sustained neighborhoods. of emergence appropriate, The second mechanism prior interaction or
lived in highly segregated

cordiality facilitated the
crisis-managing organizations. also allows us to sort out why associational are sturdier than forms of engagement everyday forms in dealing with ethnic tensions. If vibrant organizations cultural, serving the economic, commu for and social needs of the two communities the exist, support ex nal peace tends not only to be strong but also to be more solidly pressed. possible, ject of quotidian survive because with those forms Everyday but associations of engagement may make can often serve interests Intercommunal the business interests associational that are not forms the ob
interactions. connect
business
they of Muslims,
not because
of neighborhood
organizations of many Hindus between warmth
Hindu
not
andMuslim
families. Though valuable in itself, the latter does
constitute the bedrock for strong civic organizations. necessarily That this is so is, at one level, a profound all, we know paradox. After is that at the village level in India, face-to-face, engagement everyday
376
WORLD POLITICS
are the norm, and formal associations Yet rural virtually nonexistent.26 was in the home to about 80 percent of India's population India, which not 1950s and contains of still two-thirds the has been country, early even site of communal the primary violence. By contrast, in life urban ciational flourishes India, containing cities, accounts for the overwhelming majority of deaths though about asso one vio
third of India's population today and only 20 percent in the early 1950s,
in communal
lence between 1950 and 1995. Why should this be so? Figure 1 presents a formal resolution of the
paradox. size and civic links dia the relationship between constant. civic the level of engagement grammatically, holding Moving we can see to from circle 1 is associational 4, necessary engagement why It depicts
in cities ifwe answer the following question: how many links will have to be made ifwe wish to connect each individualwith every other indi
as we move
to cities? Let N represent from villages the number or must be in a of persons and K the of links that number city, village if everybody is to be connected with made else. everyone The four circles in the diagram increase the size of the local setting. vidual In circle
= only two individuals (N 2); to connect them, we need only one link = = (K 1). In circle 2 there are three individuals (N 3);we need at least
three a links (K = 3) to connect one with them all, we all. This need circle can represent a small
1, our diagrammatic
representation
for a small village,
there are
town. Circle 3, inwhich we have four individuals (N

city. To connect six links diagrammatic substitute for a metropolis,
= 5), and at least ten links (K 10) are needed to connect each of them to everybody else. Thus, for a given level of civic density (in this case, each
connected tionship to everyone else), K rises faster can be written as than N. This whole rela
(K 6). In circle 4, our = there are five individuals (N
4), can be called
person
K =N (N-l)/2.
means to from villages that as we move essentially towns to cities, we need many more links to connect to be less than the increase in population. Cities tend naturally people some is inevitable. interconnected; degree of anonymity can now understand inti associations We what do, when villagelike This formula towns and from macy is no longer possible. Since each association end up reducing N in cities people, organizations
26 M. N. Srinivas, Remembered Village (Berkeley: University
can represent a lot of a lower K and making

Press, 1979).
of California

N Circle = 1 2,K=1 N Circle = 3,K 2 = 3
377
Circle = 4,K
3 = 6
Circle N =
4 5,K=10
= number K = number
of persons of links
Figure 1 inVillages Everyday Engagement
and Cities
viable. That with
is why everyday may be effective engagement smaller Ns, but not in cities, with higher Ns. Therefore, in cities as in tain the same level of civic engagement villages, interaction. rather than informal and everyday associations, The
in villages, to main one needs
en It above is deductive. everyday explanation explains why in it rural but has and urban different very gagement meanings settings, com or us not still does tell how exacdy associations prevent mitigate an to which is munal do. That conflict when they empirical question,
we now turn.
378
WORLD POLITICS
a better not civic networks, when intercommunal, Organized only do of the communal shocks?like exogenous job withstanding partitions, civil wars, and desecration of holy places; local they also constrain in their strategic behavior. Politicians who seek to polarize politicians Hindu and Muslims for the sake of electoral advantage can tear at the
fabric of everyday engagement through the organized might of crimi nals and gangs. All violent cities in the project showed evidence of a
nexus of politicians peace, to and criminals.27 often neighborhood erogeneous organized are most gangs readily disturbed from het causing migration communally as homogenous neighborhoods, people Organized
communally
moved away in search of physical safety. Without

gangs,
the involvement of
of killings large-scale rioting afforded and without the by politicians, unlikely, protection cannot such criminals the clutches of law. Brass has rightly escape an institutionalized riot system.28 called this arrangement In peaceful
and tens and hundreds
an institutionalized cities, however, peace system exists. are when such as trade forces created Countervailing organizations associations of businessmen, doctors and unions, traders, teachers, at least some cadre-based from and lawyers, political parties29 (different are commu the ones that have an interest in communal polarization) lose from a communal that would nally integrated. Organizations split
fight for their turf, alerting not only their members but also the public
large far more practical synergy ganizations, at are to the violence. Local administrations dangers of communal in such circumstances. for all Civic effective organizations, purposes, emerges making rumors escalate become between the ears and arms of the administration. A of the state and local civic or the local wings situation and pre it easier to police the emerging
venting
it from degenerating
and
into riots and killings. Unlike

often
violent
cities, where
polarizing politicians either do not succeed or eventually give up trying

to provoke and engineer diagrammatically. communal violence. Figure 2 represents the argument
spread, quickly ful cities nip rumors,
and skirmishes, strategically planted into riots, the relationships of synergy in peace in the bud. In the end, small clashes, and tensions
27 can be proven social scientifically, not legally. The latter requires establishing These connections and gangs as groups. individual culpability, not obvious links between politicians 28 Paul Brass, Theft of an Idol (Princeton: Princeton University Press, 1997). 29 In a democratic system, political parties would be part of civil society, for not all of them may be tend to become ap linked to the state. In one-party systems, however, parties, even when cadre-based, India is a multiparty democracy. pendages of the state, losing their civil society functions.
379
Intraethnic Engagement
Ethnic Violence
Shocks Exogenous Tensions Rumors
Interethnic Engagement
Ethnic Peace
Figure 2 Civic Life and Ethnic
Conflict
This argument, it should be clarified, is probabilistic, not lawlike. It

indicates no the odds but should not be taken to mean that there can be exceptions to the generalization. lawlike generalizations Indeed, pending about ethnic violence for example, further empirical may not be a state bent on
investigation, at all. Upsetting possible
the probabilities, hell. My to
inciting ethnic pogroms and deploying itsmilitary may indeed succeed

a veritable in creating ethnic more to riots than applicable that deals with the more be therefore, would argument, or civil wars. A civil of theory pogroms form of ethnic violence: riots.
wars or pogroms would have to be analytically distinguished from one

common
Indeed, perhaps the best way to understand the relationship between civic life and political shocks is via an analogy from meteorology. If the
civic edifice is interethnic ethnic can absorb and associational, that register earthquakes of smaller intensity there quite can is a good chance it on the Richter high the edifice down
scale (a partition, a desecration of a holy place); if it is interethnic and

quotidian,
(defeat of an ethnic political party in elections, police brutality in a par ticular city) ;but if engagement is only intraethnic, not interethnic,
small tremors (unconfirmed rumors, victories and defeats in sports) can
earthquakes
bring
380
unleash tions violence. torrents across ethnic
WORLD POLITICS
of violence. boundaries A multiethnic is very vulnerable society with to ethnic few connec and
disorders
IV. Evidence
To establish causality,
That
Civil
Society Matters
of process by step, tracing looking was ap
a modified or
technique
plied to each pair of cities. The technique of process tracingworks back

from one the outcome?peace violence?step to iden
tifywhat led towhat. It can be shown that process tracing, as applied to

establish case, may not conclusively argument causality.30 The the desirability summed up in Section of variance, II, is also ap can us not to process give history but why case-based plicable tracing was a modification to the tech necessarily causality. Therefore, applied about
nique. In each pair, we looked for similar stimuli that led to different outcomes in the two cities and then identified the mechanisms by
which same outcomes. Civil trigger produced society divergent we had studied as a causal factor from such If comparisons. emerged interconnections between Hindus and Mus cities, where only violent or we were not have dis would lims minimal absent in the first place, the covered what intercommunal based on variance causality. civic can links thus can do. A controlled into a turn process tracing
comparison for establishing method
Similar Provocations,
The process outlined Civic links between such riots. To
Different
above was
Responses
to all three in the project. pairs combined with the use of
applied the two communities,
from escalating into kept tensions was established, let me concentrate on the first in a stylized all cities together only pair of cities. Presenting involved. fashion will not give a good sense of the process links by local administrations, explain how this sequence
The first pair consists of Aligarh and Calicut. The former is a riot prone city in the North Indian state of Uttar Pradesh (UP), and the lat
ter a
a single riot in this century;Aligarh figures in the list of the eight most
cities (Table 1, column riot-prone cent Muslim, with the remaining
30 For a debate 1997). 31 Calicut
peaceful
city in the South
Indian
state of Kerala.
Calicut
has not had 36-38
4). Both population
cities are roughly overwhelmingly
per Hindu.31
on
why
process
tracing will
not
easily
establish
causality,
see APSA-CP
(Winter
also has a small Christian
population.
381
Between 1989 and 1992, when the Hindu nationalist agitation to de stroy the Baburi mosque inAyodhya (hereafter referred to as theAyo dhya agitation) led to unprecedented violence inmuch of India, both
cities experienced and small clashes. But the final rumors, tensions, were very different. was outcomes In Calicut the local administration able to maintain law and order. Unfounded rumors Similarly, circulated in the ru a site city that pigs had been thrown into mosques. mors that Muslims had attacked the famous gready venerated by Hindus in the state. Such there were
Guruvayur temple, rumors often led to riots
in several cities in India and frequently did so inAligarh. In Calicut the

and the press helped the administration peace committees rumors. The storm of the the biggest agitation, Ayodhya and one that left hundreds dead in several cities, partition cut and left it unharmed. quash the since India's skirted Cali
By contrast, blinded by aHindu nationalist fervor during theAyod

into horrendous the city of Aligarh violence. hya agitation, plunged rumors investi Unlike Calicut's which neutralized after newspapers, local false them, Aligarh's newspapers gating inflammatory printed wrote in lurid hoods. Two of the largest-circulation Hindi newspapers
nurses, doctors, and staff of the Aligarh Muslim Uni versity (AMU)hospital killing Hindu patients in cold blood.32 Some
Hindus the university campus,33 but nobody was murdered rumors were believed, in the AMU how The hospital.34 ever. And gangs of Hindu on a criminals went Some of spree. killing outside were indeed killed
detail ofMuslim
them stopped a train just outside the city, dragged Muslims out, and brutally murdered them. The press underreported their acts of killing.
Although Gruesome these newspapers violence rocked were later reprimanded for unprofessional
behavior by the Press Council,
the damage had already been done.
the city for several days, leading to nearly more seventy deaths and many injuries. s local mechanisms of peace were remarkably As in the past, Aligarh an exogenous to the task of with shock?in this dealing inadequate
case, the Ayodhya agitation. The criminals who engaged in killings could not be brought to justice. Not only were they protected by politi
32Aaj, December 33 "For an Aligarh 11,1990. 10,1990; Amar Ujala, December of Peace," interview with District Magistrate
A.K. Mishra,
Frontline,
December
22,1990,22-23. 34 several interviews with AMU vice-chancellor M. Naseem Farooqui, Delhi, July 15,1994; Author AMU professors, August review of all such 1994. For a thoughtful 1994; and local journalists, August see Namita ka khabar ban jana newspapers, reports appearing in local Hindi Singh, "Sampradayitka 1991. nahin, kahbron ka sampradayik ban jaana khatarnak hai," Vartaman Sahitya, September
382
cians, but they also had peace committees
WORLD POLITICS
remarkable could were journalistic not be formed often started connections?Muslim at the city level in Ali and then exploited by the press sim to all
criminals with the Urdu press and Hindu

Effective
thugs with the Hindi press.
garh, for itwas difficult even to get theHindu nationalists andMuslim

politicians political together. organizations. the situation Rumors Instead with of investigating Calicut. Two rumors, were
ply printed them with abandon.

Contrast well as those points common how
accounts given by administrators of Calicut between 1989 and 1992 (as was kept. First, politicians of all parties helped establish peace in the
as in communities, Second, city, instead of polarizing city-level Aligarh. were critical to management of tensions.35 They pro peace committees a forum at which to the administration, vided information became all were welcome to speak out and express in addition, committees their anger, gave citizens formed a sense of par smaller peace in posted there since the mid-1980s) about the peace
ticipation to local actors, and provided links all the way down to the
neighborhood committees. By are contrast, level, where, those
peace
that do emerge
from below
Aligarh have often tended to be intrareligious, not interreligious. They

at the from a level to protect coreligionists neighborhood organized not facilitate communities communica and do attack other by possible tion with those other communities. increase the Rather, they simply in of risk and harden the attitudes of those who participate perception them. The community. from what trust but munity, other members The of these process be there would forms take turns policing their a very different kind of consciousness were since if the committees interreligious, committees within or one s own staying com up in
by definition intrareligious committees are based not on interreligious

rather on a lack of such trust. Moving rumors that no one can verify hearing disprove,
the middle
of the night for weeks
together, collecting firearms and
to ensure that retaliation is swift in the event of small weapons committees fuel and reflect a activities of intrareligious attack?these not a consciousness that builds bridges. communal consciousness,
35 inTrivandrum with Amitabh interviews Author 1991-94, July Kant, district collector, Calicut, Shankar Reddy, police commissioner, 1991-94, July 22,1995; Calicut, 20,1995; po Siby Matthews, lice commissioner, Calicut, 1988-91, July 21,1995; K. Jayakumar, collector, Calicut, July 21,1995; Ra of the Muslim Politicians 1986-88, Calicut, July 21,1995. League and jeevan, police commissioner, in peace committees. The political leaders interviewed were Dr. their participation BJP confirmed K. Sreedharan since 1991, July 23,1995; of Legislative Assembly Muneer, Muslim League member Pillai, president BJP,Calicut district BJP committee, Calicut, July 25,1995.
ETHNIC CONFLICT& CIVIL SOCIETY The Variety of Civic Networks
383
Why did the two cities respond so differently?Why did politicians of all kinds cooperate inCalicut but not inAligarh? Most of all,why did even those politicians of Calicut who stood to benefit from Hindu Muslim polarization, like the Hindu nationalists of the BJP, avoid
working to inflame communal be passions and instead cooperate in peace
making efforts? The BJP leader of Calicut admits thatHindu-Muslim

would polarization lead larger numbers munal parties, of Hindus in his it would interest because party's political to vote for the BJP instead of noncom do currently. But he is also convinced that it
as most
would not be wise for his party to systematically initiate the polarizing process, because itmight then be blamed for undermining the local
peace. If, however, the radical Islamic groups were to launch a violent
would be campaign, itwould doubtless benefit the party and the BJP happy to respond in kind.36
ac to engage in the BJP is unwilling why polarizing one needs to survey the texture of civic life there. in Calicut, runs so in Calicut civic integration Hindu-Muslim (and, many deep as a a state in is would the that whole) argue, polarization highly risky To understand tivities decades-long Hindu-Muslim
strategy. If a party can be clearly linked to activities destroying the
it will be there is a good chance peace, reverse is true in the where punished by the electorate. The Aligarh, utter weakness of crosscutting links opens up space for communal politicians Consider play havoc. in the two first the forms of citizen engagement quotidian to survey results, of and cities. According 83 Hindus percent nearly to
Muslims in Calicut often eat together in social settings; only 54 per cent inAligarh do.37About 90 percent of Hindu andMuslim families in Calicut report that their children play together; in Aligarh amere
42 percent report 60 percent that to be the case. Close to 84 percent Hindus of Hindu and Mus
andMuslims
garh only
in the Calicut
survey visit each other regularly; inAli

at that. The in
do, and not often
lims of Calicut simply socialize more often and enjoy itmuch of the
interactions on all of these only are compara Aligarh interactions would be neighborhoods.
Hindu-Muslim time, whereas statistics tively thin. Aligarh's much lower ifwe had concentrated
on the violent
We
see from the few peaceful but integrated neighborhoods
that poli
36 Author interview, Sreedharan Pillai, president, BJP district committee, Calicut, July 25,1995. 37 are from the survey conducted in Calicut Unless otherwise reported, the statistics here and below see the and Aligarh. For the methodology, appendix.
384
tics has not destroyed
WORLD POLITICS
civic trends interaction elsewhere in all parts in the town. of the town, as some
of the neighborhoods
hegemonic however,
have managed
to keep their distance from the

of respondents It should be noted, over the age
political that an overwhelming
proportion
of sixty reported that their neighborhoods inAligarh had been much more integrated in the 1930s and 1940s.38 But in the 1930s, as politi
cians munally criminal What started using thugs homogenous nexus. about the associational forms of engagement? Much of "joiners." Associations like spread violence, migration localities. Neighborhood-level to began to com was intimacy
simply unable towithstand the depredations of the emerging politician
sAmerica, is a place Calicut of all Tocqueville kinds?business, labor, professional, social, theater, film, sports, art, to the From traders associations, the reading?abound. ubiquitous
Lions and Rotary Clubs found in almost all towns in India, to the oth
erwise the association, reading clubs, the head-loaders (porters) even an as and like art-lovers association, rickshaw-pullers something sociation?citizens of Calicut excel in joining clubs and associations. is the extent of interreligious interaction in nondenomina rare
Religiously based organizations also exist, as they do inAligarh; what is

distinctive tional organizations.39 trade. The in 1995, was city, with dominated a population by merchants
Consider the economic life of Calicut, which

merchandise thousand
is based primarily on
seven hundred traders.40 About and
of about
38 was older than sixty, which allowed us to gather recollections of the Forty percent of the sample 1930s and 1940s. 39 in such large numbers. Since It may be asked why people in Calicut join interreligious associations violence and peace constitute the explanandum (the dependent variable) in this analysis and civic net works, the explanans (the independent variable), I only ask whether causality is correctly ascribed to a case of it constitutes civic networks or, alternatively, whether endogeneity. The question of why is analytically different. To answer in Calicut but not inAligarh people join interreligious associations it requires a research design different from the one that investigates why violence or peace obtains in in the other. the two places, for the explanandum is violence in one case and associational membership that Calicut citizens have greater faith in the "rational That said, it is quite plausible to hypothesize to state by change the behavior of the legaT functioning of the state, and therefore, instead of seeking can exercise on it are confident through associations. capturing state power, they they enough pressure It may also be that Calicut citizens identify less with caste and religion today than do the citizens of there is no doubt that caste played an enormously though historically important role in gen Aligarh, see a recent account of the caste basis of such erating struggles for social justice there. For struggles, 1900-1948 Caste, Community and theNation: Malabar, (Cambridge: Cambridge Uni Dileep Menon, more than prevention versity Press, 1995). Finally, integrated civic networks conceivably achieve much of communal riots. They may, for example, be related to the better provision of social services in Cali cut (and Kerala), but such outcomes are not the main object of analysis in this paper. Only communal violence is. 40 Calicut has no industry in all. hundred workers except tiles. It is small in size, with nine factories and about twenty-five
ETHNIC
CONFLICT
& CIVIL
SOCIETY
385
one hundred thousand people depended partially or wholly on trade,

are not available, exact numbers estimates that indicate and, although was ten the city had between and twelve thousand traders.41 It the rare These trader who did not join a trade association. associations?rang to those who in food/grains from of traders who deal ing organizations deal in bullion?were, as many in turn, members as eleven of the Federation of Traders
Associations
In 1995 holders:
(Vyapari Vyavasayi Ekopana Samithi).

out of twenty-six trade associations reg
isteredwith the federation had Hindu, Muslim,
(andChristian) office
was from one if the president of the association community, associations the general secretary was from one of the others.42 These con in refuse to align with electoral any particular political parties our tests: "We dont want to enter will because be bro unity politics resolved." so conflicts, in our association, if any, get was the of such that many Moreover, engagement depth were concluded contracts. transactions without formal "Our rela any are on trust. businessmen entirely based Pay tionships with Muslim have debates
ken. We
ments as large as 10 to 15 lakhs ($30,00O-$35,000)

We send bills, but come there law. Payments no breach of trust."43 Aligarh also has promissory in thirty days. We work association are no notes
are sometimes due.

valid in the courts brokers. There of is
through
a traders
1980s it had about six thousand members. In the 1970s it had even ac
a fair number on the busi of Muslim who members, emerged quired ness map after the Gulf to The association, however, began migration.
(Vyapar Mandai).
In the late
engage in infighting overwhether

litical party, into the argument being
it should support andwork for a po

that supporting a party favorable to
traderswould benefit all of them. In the 1980s the association finally

split two bodies: a "secular" organization and a "nonsecular" one,
with the nonsecular faction joining the BJPand theMuslims

the "secular" faction.44 Unlike sector and trade-based is among
turning to
also has a significant industrial Calicut, Aligarh lock the largest producers of locks in India. The
41 These numbers and the information below are based on extensive interviews with the president of Trade Associations and general (Kerala Vyapari Vyavasayi secretary of the Kerala Federation Samithi hereafter Samithi). The Samithi is a powerful all-state body, based in all towns of Ekopana run office. It is rare to find a Kerala. The Samithi keeps records and statistics and has a professionally inNorth India. traders association run so professionally 42 Data supplied by the Samithi, Calicut branch, July 1995. 43 rice dealer, Calicut, July 25,1995. interview with V. Ramakrishna Author Erady, wholesale 44 Author interview with Mohammed Aligarh, Au Sufiyan, former president, Vyapar Mandai, gust 1995.
386 manufacturing developed lock is mostly an economic
WORLD POLITICS
small scale. Moreover, between have different units special
ize in different parts of themanufacturing process. Yet Aligarh has not

symbiosis as no Hindus been and Muslims.
It is impossible to estimate the number of people working

conducted.45
inAli
know We surveys garlas industry, come from both from ethnographic that the workers work, however, as owners. Muslim and Hindu firm do the We also know communities, no intercommunal that there is virtually The informal dependence. was credit market, dominated lenders normally (mahajan)9 by Hindu on which some Muslim manu the only Hindu-run economic activity to rotat facturers used traditionally the few Over last decades depend. are intra-Muslim credit societies have these But ing emerged.46 societies that build trust within are not constitute If the businessmen communities, integrated, what a not across about them.47 Since the workers?
of the city than the larger proportion they numerically in trade unions could, in links formed businessmen, interreligious prin more an than make for of such links among the busi absence up ciple, nessmen. But trade unions hardly exist in Aligarh. offices of Decrepit trade unions, with no staff and little data, trade unions greet researchers who study labor activities. By contrast, are linked to two thrive in Calicut. The unions largest major national trade-union federations: is associated with the Commu CITU, which the local branches of national
nist Party (Marxist), and INTUC, whose political patron is the Congress
Party.48 Both wins of these elections. unions are intercommunal. Calicut does have a
political party of theMuslims,

general is neither however, vibrant. It is the weakest and large vote
theMuslim
League, which
regularly
It also sponsors a trade union, the STU. The STU, as as the local units of CITU or INTUC nor as large and smallest elections character of the three. Muslim for the Muslim of CITU does not League, stop workers but them by they
tend typically to join INTUCor CITU for protection of their labor rights.
The Marxist and atheistic from
in assembly
joining CITUs unions, if they think that CITUwill do a better job of

45 It pays to underreport how much labor an industrial unit employs, for under Indian law the small, informal sector does not have to pay pension and other benefits to its workers. Official statistics are thus entirely useless. Foucault s concept of "popular illegality,n as one keen observer puts it, has caught the fascination of Aligarh's lock manufacturers. Elizabeth A. Mann, Boundaries and Identities: Muslims, Work and Status inAligarh (Delhi and Newbury Park: Sage Publications, 1992), 83. 46 Ibid., 101-2. 47 Ibid., 84-85. 48 are almost to Exact numbers of unionized members and their religious distribution impossible on the interviews are the best one can do. The come on by. Estimates based description below is based interviews with labor leaders in Calicut, especially a long and detailed interview with M. Sadiri Koya, state secretary, INTUC,August 4,1993.

with A Hindu
387
fighting for their rights andwages. In the process, they come in contact
Muslim links are formed, and a Hindu intercommunal workers, division of the workforce does not take place. or most site for unionization?the "head loaders" unlikely for it shows the associational over hundreds porters in Indian abilities and thou are bazaars Distributed units, loaders
worth mentioning, porters?is and success of Calicut workers. sands were of shops and
small business head
rarely unionized. But they are inCalicut (and inKerala). In 1995 there
nearly ten thousand There in Calicut?about in Aligarh, 60 percent but there
Hindu
and 40 percent Muslim. Most were part of INTUCand CITU

are head loaders in the bazaar
trade unions.
no associations. they have
A final and highly distinctive aspect of associational life in Calicut

an array city has had is nothing societies. There of film clubs, theater, and science popular are South India. unusual about film clubs?they throughout popular in bringing But societies interested theater and science to the masses are rather uncommon. have been reading clubs. Even more uncommon concerns its social and educational activities. The
The
literacy rate in Kerala today is the highest

a unique Kerala institution, from most accompanied communities
in India. "Reading
Kerala's would remarkable
rooms,"
rise in literacy and formed deep social networks between the 1930s and
the 1950s. Young people several times every week to read newspapers and cultural get together and political
books. The fascinating story of the birth of reading clubs has recently been told byMenon:
Between 1901 [and] 1931, the rise in the numbers of literate was phenomenal. The growing numbers of schools and the rise in literacy found expressions in the numbers of reading rooms thatwere established both in the countryside and
in the towns_One of the novelties in the organization of reading rooms was
the [communitarian] drinking of tea, as one person read the newspapers and the others listened_Tea and coffee lubricated discussions on the veracity of the news and of political questions, and a new culture emerged around the reading
rooms. It was premised upon sobriety and knowledge rather than drunken com
panionship transcending consciousness which characterised the toddy shops. The importance of tea and coffee lay in the fact that they were recently intro duced beverages and did not fit into any taboos regarding what could be shared
between castes. Tea shops and reading rooms all over Malabar provided formal [a]
common place for people to meet and to drink together regardless of caste [and
community]_The reading rooms emerged as central to both attempts
at organization by the left wing of the Congress

49 Menon(fh. 39), 145-49.
aswell as local initiatives.49
388
The noting. Muslims even Calicut twenty did cumulative In our Calicut
WORLD POLITICS
movement is worth of the reading-room as many as 95 of Hindus and percent sample, to statistic that is be reading newspapers?a likely cities of the richer of over countries of the world. thousand, most Hindus has in population and magazines!50 often travels a seven hundred outcome
reported than higher today, with newspapers
in most
By contrast, while in the Muslim
the Aligarh sample read newspapers, less than 30 percent ofMuslims

so. Information As links with mouth. only To the Hindu by word of community are nonexistent, it takes community rumors and make them stick. are worlds in associational apart. So many and neighbor of engage out of the as and Hindus. the two com and all
a few to spread nasty people sum up, the civic lives of the two cities and Hindus are interlocked
Muslims tension ment. Routine munities;
hood relationships inCalicut that peace committees during periods of

are an extension simply of the preexisting reservoir of social trust 51A considerable interactions and everyday between communication facilitates familiarity rumors are squelched through better local networks is formed Muslims between
sociational
communication; in those trust civic
of this helps the local administration keep peace. InAligarh, however,

the average Hindu economic, social, and Muslim do not meet mutual educational?where settings? can be forged. police and civil The dif
even the support of such networks, competent Lacking as riots unfold. administrators look on helplessly, The other pairs in the project experienced similar processes. ferent
in the absence of resided neither outcomes, however, religious identities nor in the presence of tensions, provocative rumors, and small net of the intercommunal clashes. Decisive, rather, was the presence Intracommunal of engagement. or violence. contain, stop, networks, by contrast, did not
works
50 And the state of Kerala has "a library or a reading room within walking distance of every citizen." in Kerala: An Overview" K. A. Isaac, "Library Movement Control and Bibliographic (Paper presented at the International of Kerala Studies, Trivandrum, India, August 1994). Congress 51 It may be suggested that this finding is close to being a tautology: a city is not riot prone because it iswell integrated. This claim, however, would not be plausible for two reasons. First, a conventional common sense of the field, suggests that for peace, multieth explanation, which has long defined the nic societies require consociational is an argument about segregation arrangements. Consociationalism at the mass level and at the elite level, not integration at either level. My argument is very bargaining have often fought violently to apurifyw their communities different. Second, religious fundamentalists have often sought to undermine of influences from other religions in society. Islamic fundamentalists of neigh combined the practice of Islam with the incorporation Sufi Islam, which has traditionally seen as a source of integrated lives and belief systems have often been boring influences. Communally version of the debate, see H. D. tension and conflict rather than peace. For the North American Press, 1997). Forbes, Ethnic Conflict (New Haven: Yale University
ETHNIC CONFLICT& CIVIL SOCIETY V. Endogeneity

Before questions we
389
and the Underlying

about how
Causation
the argument accept must be explored. First,
two more civic engagement, can one be sure that the causa
tion did not flow in the other direction? Did communal violence de
in civic networks towns, or did the stroy the Hindu-Muslim riot-prone we of such networks from violence presence prevent occurring? Might can not have a case of at here? Second, process tracing best endogeneity establish short-run Is causation different from the causality. underlying proximate or absence variable causation? Are there historical of civic networks? What of the short-run that explain the vitality turn the emerges independent a variable to networks?into analysis?civic forces ifwe
be explained historically? The city of Surat, the third historically peaceful city in the project,
helps run us address primacy the problem of endogeneity and establishes the short of civic networks. In Surat (Gujarat) a nasty riot occurred
after the destruction of the Ayodhya mosque,
the first such riot in
of violence, how seventy years. An overwhelming nearly proportion to the slums; all 192 deaths took in the ever, was confined shanty place some arson and towns. The old looting but city, by contrast, witnessed to the same stimuli, the preexisting social net Subjected the city. works accounted for the variance within an industrial boom in the last twenty years, Surat has experienced cities of more the of India. becoming small-industry capital Among than one of the Surat has registered highest popula people, rates since 1980. tion and from outside from within Migrants growth in the shanty towns. into the city and settled the state have poured a million no deaths.
Working
the Factory returning
in small industrial units and unprotected by the labor laws of

Act, most of these migrants to the slums and shantytowns work only long hours, exceptionally are eat. There to and sleep
few institutionalized settings for building civic ties. When the mosque came down inAyodhya inDecember
slums were
1992, the
In the old city, how and violence. the site of awful brutality were associations committees formed. The business ever, peace quickly are in in the old city, live primarily of Surat, whose members especially
tegrated. These Hindus andMuslims, who had lived side by side for years and had participated in the old city's business and social life,were
set up to lower tensions. They neighborhood together committees and deployed their own resources and organizations watch rumors and As with the administration. in checking communicating able result, the local administration was more effective in the old city than to come
a in
390 the industrial and criminals What networks
WORLD POLITICS
where shantytowns, were free to commit were entirely missing acts of savagery and violence. causation? Have civic the Hindu-Muslim civic networks in peaceful towns, directing their Hindu
about
the long-run always been robust
Muslim
nous come moment that civic networks
politics and making

networks?quotidian in the short to medium were politically in the nations
it possible for them to withstand

conducted in the cities and associational?determined
exoge
the out
shocks? Historical
research
demonstrates
run, but in the long run intercommunal 1920s were a transformative constructed. The because it was then that mass politics
politics
emerged in India under the leadership ofMahatma Gandhi. Politics before Gandhi had been highly elitist, with the Congress Party a
lawyers' club that made its constitutional arguments for more rights
with the British in the Queens English. Gandhi seized control of the movement in 1920 and quietly revolu tionized it by arguing that the British were unlikely to give independ
ence to India until the Indian masses were involved in the national
movement. Gandhi
(swarafjy one other against cial
talked of two intertwined battles of independence

an adversary, the colonial power, and Indias social evils. He was interested
an external against an internal enemy,
not only in political independence from the British but also in the so
transformation without meaningful of India, arguing that the former could not be on three social ob the latter. He first concentrated
jectives:Hindu-Muslim
unity, abolition of untouchability, swadeshi (buy
Indian, wear Indian, think Indian). To these were later added other pro women swelfare, labor wel tribal welfare, jects of social transformation: on. so In the process millions of his followers created and fare, prohibition, a large number of organizations between the 1920s and the 1940s. Be
fore Gandhi the civic structure of India had been quotidian. After the
Gandhian The moment biggest in the national movement it became the Congress associational. Party, which organization, of course, was
led themovement politically and developed cadres all over India during
about social reconstruction also created a argument set of organizations, the voluntary second agencies. The Congress Party was that dealt with and education, political, organizations primarily self of the tribals and "untouchables," women's issues, the welfare were movement concerned the and reliance, immediately homespun the 1920s.52 The
with their social projects.

52 It should, however, be pointed out that in Calicut and the neighboring areas, it is the left wing of the Communist the Congress Party, later splitting from the parent organization and becoming Party of India (CPI), that engaged in the most systematic association building.

The The unity civic order movement
391
in different that emerged was not identical places. had greater success in putting together Hindu-Muslim a Hindu-Muslim in towns where had not already cleavage since the 1880s. caste If local politics emphasized the Hindus among the Congress some or Party
emerged in local politics. Indias towns had been having elections for
local governments other cleavages?for divisions Shia-Sunni example, among cleavage the Muslims?then
and Gandhian socialworkers found it easier to bring Hindus andMus lims together in the local civic life. If, however, Hindu-Muslim differ
ences were could not the dominant build axis of local politics, organizations the national with the movement success. integrated same
Though
in place,
originally a child of politics, these organizations, once firmly
on how relative autonomy from politics. Depending acquired or communal to create very different integrated they were, they began sum in To civic net the role of intercommunal pressures up, politics. works has been sense, however, crucial for peace at a proximate level. In a historical a space for them was created mass by forms of politics a twofold con suggests a trans it turns out that
that emerged all over India in the 1920s.

For problems of endogeneity, this reasoning is applied, clusion. If a historical perspective in formative shift national ideological politics, evils and to reorient tematic the fight effort.
organizational civic order, instituted by the national movement, became organizational a constraint on the behavior of politicians. the thrust of the na Given tional movement, of politicians' the civic constraint on politics was especially
to address social seeking was the cause of a sys for independence, In the short to medium the run, however,
serious
if building or destroying bridges between Hindus andMuslims was the

object strategies.
VI. Concluding
Are the conclusions
Observations
of this paper India specific or have they resonance on civil sets of concluding observations?one elsewhere? Two society in order. and one on ethnic conflict?are Putnam has used term First, the term focus of the "networks my "social capital" for civic networks.53 My in of engagement" differs from Putnam's civic ties, and intraethnic is on interethnic
use
two ways.
53 Robert Putnam, Making Democracy Work: Civic Traditions in Italy (Princeton: Princeton Univer sity Press, 1993). It should be noted, however, that since writing Making Democracy Work, Putnam has the dis civic networks. Putnam acknowledges introduced the notions of bridging and nonbridging tinction further in Bowling Alone (New York: Simon and Schuster, 2000).
392
not a civic ties per
WORLD POLITICS
se. Communal or ethnic and ethnic organizations, can be shown focusing to generate on
single religious great deal of trust
a group only, are If they their members. such among plentiful, can endow a a Putnam's with definition, organizations, by place high in my materials, these organizations degree of social capital. However, are often not riots but of preventing Hindu-Muslim only incapable are also associated for ethnic but whether trust with violence social based civic on the escalation is not whether and civic not of communal ethnic violence. What capital life or social
matters exists
ties cut across ethnic
differently, critical.
interethnic^
Stated groups. is networks intraethnic, work includes rightly individuals and fami
Second, while both formal and
in Putnam's engagement interactions informal between
between the two forms should also be noted. For lies, the difference ethnic peace, everyday engagement between ethnic groups may be bet ter than no interaction at all, but it is also different from qualitatively interethnic the more formal, organized engage engagement. Everyday a or on ment may be to maintain small scale peace enough (villages small towns), but it is no substitute for interethnic associations Size reduces the efficacy (cities and metropolises). settings formal associations.54 interactions, privileging for the literature on ethnic also have implications My findings in larger of informal con
flict. Although
of ethnic countries,
disaggregated statistics on local or regional dispersions

have not been first be noted systematically that the data collected that we for many do have?for
violence it should
for the United States or Northern Ireland55?show roughly example, ethnic vi the same larger pattern that exists in India. On the whole, or not to be olence concentrated tends locally regionally, highly
spread evenly geographically
across the length and breadth of the
of ethnic relations, more charac breakdown country. A countrywide teristic of civil wars, is rare: we tend to form exaggerated impressions and not the quiet continu of ethnic violence, partly because violence
a third way inwhich this research differs from Putnam's reasoning also suggests Making De the existence of social capital differentiates mocracy Work. In Putnam's formulation, good governance from bad. The relationship between social capital and communal violence, however, yields a different If my argument is right, civic networks determine the presence or absence of riots, but formulation. are in the long run. Putnam's study appears to emphasize the independent they politically constructed role of social capital in both the short run and the long run. 55 see and Underlying and Arnold "The Pr?cipitants For the U.S., Silverman, Stanley Lieberson Ire Conditions of Race Riots," American Sociological Review 30 (December 1965); and for Northern of Political Violence in Northern Ireland," in John land, see Michael Poole, "Geographical Location and A. C. Hepburn, Darby, Nicholas Dodge, (Belfast: Appletree Press, 1990). eds., Political Violence: Ireland in Comparative Perspective 54 This

ation of routine to have life iswhat pockets attracts of violence the attention coexisting of media. with large common
peace.
393 It is more stretches of
If we systematically investigate the links between civil society and

there is a good chance we can get a good theory that can conflict, or reason for this intuition is explain these local regional variations. The can be built communities networks of nationally, quite simple. Though ethnic internationally, mains that most ness units associations and business but most and, era, also "virtually," the fact re or civic life locally. Busi community people experience across local or trade unions may well be confederated in this electronic of the time most or labor leaders may also have national arenas of op are businessmen and workers who experience them associational life locally. The
eration, members
of such organizations or
type and depth of these local networks?whether

communities tween have together pull are associational communities for explaining apart, whether or informal?are
they bring ethnic

the interactions the variables of ethnic be that vio
the potential lence and peace. Though indications
the observable
patterns
Protestant variance bourne
done, some potentially powerful are available. A few of Catholic studies existing post-1969 in Northern intranational violence Ireland have dealt with such research has not been in violence. John Darby, for example, All three has studied three local
communities
in Greater Belfast?Kileen/Banduff,
and Dunville.56 the first two have
the Upper Ash

have mixed since the late
Estates, but populations, 1960s, whereas
communities
seen a lot of violence
the third has been
were schools, and political parties but Dunville had some distinctive In contrast nities, bowling hockey, mixed to the segregated Dunville had mixed clubs, as well
found that churches, quiet. Darby in all three communities, segregated features not shared by the other two. in the first clubs, two commu clubs, and field and my soccer
groups voluntary and lions rotary for
swimming, single
cricket, athletics, boxing, table tennis, and golf. There was also a vigorous with club. These results are quite consistent parents in the U.S. are also of interest, from these but
as clubs
Indian findings.
Studies of racial violence in a dif that can were
ferent and potentially highly challenging way. There is?to

my knowledge?no city-level explain good theory emerging in racial violence variance in the
the best of
studies
1960s. Why
Ireland
Newark
56
(New Jersey), Detroit
(Michigan), Los Angeles

of Conflict inNorthern
(California),
(Dublin: Gill and
John Darby, Intimidation 1986). MacMillan,
and the Control
394
WORLD POLITICS
which together accounted for a very large proportion of all deaths in the 1960s riots, so violent? And why did Southern cities, though po
not have riots?57 The engaged, African between Americans equalities nor the location the timing explained litically have been studies show that economic in and white and Americans Silvermans neither alternative
work explanations provided. comes am to what I for close India: arguing reasonably they empha size local integration, in the participation especially African American no scholar has in local government structures.58 But to my knowledge civic associations?labor whether unions, PTAs, churches, vestigated on the whole in the peace and so on?were integrated racially better ful cities.59 If they were not?and here race relations in a comparative of segregated multiethnic Ufe?for civic sites (unions, societies where lies the innovative sense?we might schools, of American potential need an initial distinc business
of riots, Lieberson
but no firm
tion in our theory between (1)multiethnic

churches,
societies that have a history

associations,
and so on)?for
example, theUnited States and South Africa?and

ethnic
(2)
India instance, civic engagement may be a key vehicle of peace in the latter, but, given civic sites in countries like the relative absence of common black-white the United interracial
civic groups have led an intermixed or intercommunal and Sri Lanka. Interracial
for there may not have been any space historically States, to about the associational pre engagement, leading puzzles cise nature of mechanisms and that led to peace in a different historical social more setting. accurate
If we think about the above distinction further, itmay actually be

to say that not societies as a whole, have a history groups, associ of segregation. In India, where political parties, unions, business are ations, film clubs, and voluntary by and large ethnically agencies between the Scheduled has marked relations quite mixed, segregation were who "untouchable" for and the centuries, Castes, "upper castes." there have been no civic untouchables or associational could come sites where the Simi together. find common and Jews could eventually Catholics, larly, Protestants, on in whole could not.60 civic sites the the U.S., but blacks and whites and the former
Historically, upper castes
57 in the United States," inNathan Glazer and Ken Young, eds., Donald Horowitz, "Racial Violence Ethnic Pluralism and Public Policy (Lexington, Mass.: Lexington Books, 1983). 58 Lieberson and Silverman (fh. 55). 59 us an an excellent chance to It missed the The Kerner Commission give explanation. Report had cities only, not the peaceful ones. chance because it studied the riot-afflicted 60 Now Nathan Glazer, We Are All Multiculturalists Press, 1996). (Cambridge: Harvard University
395
a mechanism of peace proposed recently by Fearon "Self-policing," to such and Laitin, may well be relevant In the segregated settings.61 or intra in this article, it means intraethnic, terminology developed communal, policing. by civic organizations country remains an ethnic association, or by elders, by such as black churches, intraethnic policing may lead to the same result as interethnic does in India. Cross engagement If exercised research must to be learned. take such alternative possibilities seriously. Much
Appenddc: Research The National Level
Materials
The large-N analysis of Hindu-Muslim
riots is primarily based on a
a span of years forty-six reading of the daily Times of India, covering case in other journals were In of doubts, (1950-95). reports appearing source be as the but the Times of India was chosen checked, primary cause it is the only newspaper the entire period that (1) covers
(1950-95);
(2) had a truly national coverage of Hindu-Muslim
vio
some other newspapers, often refused to run lence; and (3) had, unlike commu most in this stories the potentially inflammatory period about
nal violence without

have some
double-checking
them. The Times of India did
of bias, but these problems were resolvable. problems not the Moreover, newspaper was read interpretively, literally. News not sometimes m/rareli do between paper reports carefully distinguish on on the one the and ?irreligious violence, hand, gious violence, as a clash riots are simply presented other. At other times communal between correctly, in two communities. can represent And the term Christian-Hindu even if "communal," applied in clashes the (as Northeast),
or Christian-Muslim
An
violence (as inKerala), orHindu-Sikh
clashes (as
was thus necessary, reading of the reports Punjab). interpretive of the based on a detailed groups, variety of religious understanding issues found in different parts of India. Unless festivals, and contentious the labeling of the riot in the newspaper was supported by the descrip an to which tion of the symbols and issues involved, read interpretive was as a was a communal not riot Hindu-Muslim coded ing applied, in collaboration with I. Steven riot. This data base was put together
Wilkinson
of Duke University.
"Explaining Interethnic Cooperation," American Political Science
61 James Fearon and David Laitin, Review 90 (December 1996).
396
WORLD
POLITICS
The City Level

Apart were 20-25 tional) drawn from documentary at two conducted interviews level in each and archival levels?elite for the cities, interviews I conducted and cross-sectional. research educa religious, in a methodologi
at the elite city. To
bureaucratic, (political, cross-section the survey
cally defensible way, a stratified sample of 100-140 households was also

was used as a city. To reach the poor, literacy principle is a good proxy for poverty, deprivation, basis for stratification. Illiteracy were in India. Five or six neighborhoods and "subalternity" then se lected in each city?two Hindu dominated the (one violence prone, for each other ond peaceful), two Muslim and finally, dominated one where peaceful), cities neighborhoods peaceful the violence-prone category.) neighborhood and three borhoods. A ensure search about The team of two research assistants?one trained a research city, yielding were Muslim candor, respondents and Hindu assistants, respondents in each 700 cross-section elites. interviews one Muslim?was Hindu, team of twelve in six cities. To interviewed by Muslim We thus 125 re were illiterate, or two (one violence prone and a sec "mixed" neighborhoods. (In tensions recently surfaced replaced were selected on the basis of twenty in all neigh of in
Respondents
literacy. If 50 percent ofMuslims terviewees (ten Hindu,

illiterate Hindus.
and 30 percent of Hindus

sample repeated
in the
the neighborhood procedure was
tenMuslim)
This
included five illiterateMuslims
by Hindus. in six cities and nearly
got inter
views with
to The first was survey was used for two different purposes. toward politics, and Muslim attitudes administration, po study Hindu to in and and the lice, religion, history identify everyday particular in neighborhoods. forms of engagement the two communities between The social second science purpose research to some standard respond on ethnic conflicts. Unlike works was to criticisms of on the func
of legislatures, executives, tioning forms institutionalized deal with
which and bureaucracies, typically of elite politics, research on commu
and nationalism tends to be part of mass politics and nalism, ethnicity, now a runs up criticism, made by post against especially by popular is critics. The that even while modernist talking about the complaint our sources on communalism end up being and nationalism masses, highly elitist, interviews with gious elites, conduct consult government reports. We a select group of reli educational and leaders, political we And read officers. and bureaucrats, newspapers, police or "official." We

which
397
at least in poor countries, do not necessarily the opin represent masses. use ions of the poor In the of "official records" on particular, violence records has been vehemently criticized. It is argued that of are unreliable, when the state, both colonial and especially as an at itself be least involved, may instigator of partially, and/or violence. divisions these objections and to collect "unofficial transcripts," I
communal ficial
postcolonial, communal To on
deal with
turnedmy survey,first of all, into away of collecting brief oral histories

specified ods because collect was questions. This those who sample oral histories an unconventional rarely collect use of survey meth oral histories and those Forty percent
who
of the sample (280 respondents out of 700 separate households

cities) was sixty plus in age: the aim was to retrieve
rarely sample
their respondents.
in six
of
local memories
Hindu-Muslim realtions in the 1930s and 1940s. Accounts of Hindu Muslim relations in the 1930s and 1940s are plentiful for the national
or provincial level but not for the town level. Since the city was the unit
of analysis in the project, localmaterials for the 1930s and 1940s had to
be created?in part, orally. as a stratification I also used Moreover, by using literacy principle, as al the survey to collect the so-called "subaltern narratives." Illiteracy,
a in India: those who are il ready argued, is good proxy for subalternity to literate also tend since such a large part of India is be very poor. And me access to a to illiterate, a sample stratified according literacy allowed
large number of illiterate people, especially in theMuslim

who rarely get to "hear heard interviewed the voices by the newspapers of the focus subaltern." on one
community,
in a sys voices are sam
or researchers The town subaltern or one
tematic way. To put it differently, through a sampling technique, I

sought typically by those who wants to reach the subaltern pling?sampling roughly village. stratified If one
in a multitown
the same proportion subaltern struck voices to create
project, of the subaltern as authentic a
as in the
population?was
by far the most methodologically
defensible way of
as possible, situa
To make being representative. was considerable prior rapport tion for the interviewees.
nonthreatening
into statistics only where such The survey data have been converted conversion ismore meaningful: in assessing for example, the degree and or attitudes nature of to between communities engagement everyday ward Wherever laws, and administration. history, personal were more maries useful, statistics have not been used. To archival textual sum
were used: (1) sum up, the kinds of research materials following on which research for historical historians have not yet periods
398
written and oral records
WORLD POLITICS
cannot be created; (2) documentary research
for contemporary issues; (3) purposive and focused interviews with the elite in all six cities; (4) stratified survey research for the cross section, including the illiterate poor; and (5) a reading of each day s Times of India between 1950 and 1995 to figure out the long-run and large-N
distribution of communal violence over forty-six years.

Aushotosh Varshney PDF

Diunggah oleh

Informasi Dokumen

Judul Asli

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

Aushotosh Varshney PDF

Diunggah oleh

Hak Cipta:

Format Tersedia

Trustees of Princeton University

ETHNIC CONFLICT AND CIVIL SOCIETY

WorldPolitics 53 (April2001), 362-98

ETHNIC CONFLICT& CIVIL SOCIETY

arguments of civic engagement Because they build bridges

play very and manage

and quotidian. This dis is formal or not. I call the

associationalforms engagement. Business

larly, jointly participate in festivals, and allow their children to play to

fronted with attempts by politicians

to polarize people along ethnic

a transformative 1920s, the British, when against

364 Indias associational a host

WORLD civic order.

likely to apply. I. Clarifying

specify the meaning one is using.

linguistically or racially distinct groups.

ETHNIC CONFLICT& CIVIL SOCIETY

would not be extended to the first and fourth.

and they distinguish

from the largely

its core feature.

internal to define tends this

one is born is also the class in which

one is trapped until

meaning? On thewhole, the existing literature has failed to distinguish

In any ethnically plural society that allows free expression of political

WORLD POLITICS there are different ethnic groups

necessarily lead to violence.4 When

via the waged an takes institu

not violence. Such institutionalized conflict must be distinguished from

individuals or families possible, and (3) is independent of the state.

part of civil society, but informal neighborhood associations would not.

ETHNIC CONFLICT 6c CIVIL SOCIETY

active on behalf of Israelwould not.

to the field of political philosophy, it ismostly political philosophers

the func literature defi

an illustration of civil whose

simply of the modernist consider on civil

propositions. biases of the conventional have

also the objective availability of plentiful as well as changing profes

termined by birth. A carpenter in traditional society,whether he liked it

to encourage members of the ethnic of ethnic and facilitating migration

and individuals connect with other families and individuals? beyond

ETHNIC CONFLICT& CIVIL SOCIETY

generated tem" in The Structural

ular struggles launched by those not formally admitted to the public

even capitalist modernity may be highly unsocial and atomizing,

relations, sets out to do

puzzle in the field of ethnicity and nationalism.

ETHNIC CONFLICT& CIVIL SOCIETY

tinue to enlighten us, it can give us only the building blocks of a

for the necessity

important methodological be studied? Should the unit towns, or villages? What

be nations, states, regions, can call a large-N analysis

between 1950 and 1995.20 For purposes of identifying larger trends,

oting. Between 1950 and 1995 rural India,where amajority of Indians

percent of all deaths inHindu-Muslim

violence (Table 1, column 4).

1 Table in 28 Indian Riots

of20 Deaths in4 Riots

of25 Deaths in 5 Riots

of50 Deaths in 10 Riots