After Fifty Years of Political and Social Change: Caste Associations and Politics in India

Author(s): James Manor

Source: Pacific Affairs, Vol. 85, No. 2 (JUNE 2012), pp. 355-361
Published by: Pacific Affairs, University of British Columbia
Stable URL: https://www.jstor.org/stable/23266848
Accessed: 03-04-2020 05:46 UTC

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.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Pacific Affairs, University of British Columbia is collaborating with JSTOR to digitize,

preserve and extend access to Pacific Affairs

This content downloaded from 35.154.245.7 on Fri, 03 Apr 2020 05:46:22 UTC
All use subject to https://about.jstor.org/terms
 After Fifty Years of Political
and Social Change:
Caste Associations and
Politics in India

James Manor*

KEWORDS: Caste, caste associations, Indian politics, civil society,

hierarchy

DOI: http://dx.doi.org/10.5509/2012852355

nyone who seeks to understand contemporary India owes a

heavy debt to Lloyd and Susanne Rudolph. Their writings are
JL JL fundamentally important, and their extraordinary generosity
to colleagues has ensured that their impact extends well beyond their
published work. We are concerned here with one of their most influential
papers, which offered acute insights into a key topic at an astonishingly
early date. Their article on caste associations (on which they elaborated
in one of their classic books thereafter1) remains important even now, five
decades on. It enables us to see how and how much things have changed
over five decades. The questions that they asked and their analytical
approach remain valid now, even if the ground realities inevitably differ.
Their article has been overtaken not by superior analyses, but by social
and political changes. The present paper concentrates on those changes.
To understand them, we must start with two sets of basic points. First,
the term "caste" can mean three different things:"varna,""jati" and "jati
cluster." Varnas—the four large, traditional divisions of Hindu society—
seldom have much political importance, partly because they are so vast
and internally heterogeneous, and partly because (see below) they signify
very different things in different Indian regions. Jatis are much smaller
endogamous caste groups, within which people marry their children.
They mainly matter at the local level, especially in villages (where two

1 L.I. Rudolph and S.H. Rudolph, The Modernity of Tradition: Political Development in India
(University of Chicago Press, Chicago, 1967).

© Pacific Affairs: Volume 85, No. 2 June 2012

355

This content downloaded from 35.154.245.7 on Fri, 03 Apr 2020 05:46:22 UTC
All use subject to https://about.jstor.org/terms
 Pacific Affairs: Volume 85, No. 2 — June 2012

thirds of Indians still live). Jati-clusters have gained salience over the last
century or so, as similar jatis have aligned with one another to enhance
their numerical strength (and thus their political influence), mainly in
the space between the state level in India's federal system and the local
level. Most caste associations which are politically important represent
jati-clusters, and operate in that space. The impact of caste associations—
on politics or on castes—at the national and local levels is, by contrast,
quite modest.
Second, the main reason that caste associations have counted for
litde at the national level is that India has not one caste system but
many. Crudely speaking, each of the many linguistic regions has its own
distinctive caste system which differs, somewhat or markedly, from systems
in other regions. In 1956, most state boundaries were redrawn along
lines separating linguistic regions. Since then, India's most important
caste associations have represented jati-clusters within individual states,
which contain distinctive caste systems. Such associations are far more
important at that level than in New Delhi, where caste labels inspire
more confusion than solidarity. Jati-clusters from different states go by
different names and play somewhat different roles in their states, and
thus find it difficult to make common cause at the national level. To
complicate matters further, there are no indigenous members of two of
the four varnas (Kshtariyas and Vaisyas) in South India,2 and Brahmins
there represent less than 4 percent of the population and own little land.
They therefore have little in common with Brahmins from north central
India who have over five times more numerical strength and much more
landed wealth. So varnas find little political traction at the national level.
To reiterate: most caste associations which matter politically represent
jati-clusters and operate at the state level.
The Rudolphs wrote at a time when landowning castes dominated
social life in most villages, and leveraged that power to achieve
dominance over state governments and over the Congress Party as
well. Congress was then the dominant party in New Delhi and nearly
all states. Its dominance derived in part from its role in the struggle for
independence and the eminence of its leaders. But more important was
a formidable party organization: a cluster of political "machines" at the
state level with sinew and reach, which penetrated effectively downward
towards the grassroots by distributing patronage (goods, services, funds
and favours) to a diversity of interests. Some of those resources passed
through caste associations. The lion's share went to the landed castes
on which the party's political dominance largely rested. Other groups
received more modest political spoils or, in some cases, mere tokenism.

I owe this formulation to discussions with M.N. Srinivas.

356

This content downloaded from 35.154.245.7 on Fri, 03 Apr 2020 05:46:22 UTC
All use subject to https://about.jstor.org/terms
 Caste Associations and Politics in India

They also wrote at a time when voluntary associations and organized

interests were still beginning to crystallize and to engage with the political
process in pursuit of power and resources. The Rudolphs explained
that since caste was such a widely understood source of solidarity, caste
associations naturally loomed large among such interests in that early
phase. Associations which served the interests of landed castes were
better organized and resourced and had greater access to political
leaders than those representing less exalted castes, but the latter had
also begun to emerge, fitfully.
The ensuing years brought change, but as usual in India, it occurred
gradually.3 A dozen years after the Rudolphs' paper appeared, caste
associations still had considerable potential as key elements in the political
game. This can be illustrated by the use which an imaginative state-level
politician made of them in pursuit of progressive goals. In 1972, Devaraj
Urs was named by Indira Gandhi as Chief Minister of Karnataka. He was
the first person to hold that office who did not come from one of that
state's two dominant landed jati-clusters. Most leaders from those groups
united against him. To remain in power, he had to attract support from
less prosperous castes who outnumbered the dominant castes, but who
were (then, unlike now) poorly mobilized. To galvanize them, he formed
caste associations for them, and used illicitly obtained money to fund the
new associations. This made sense because a political awakening among
poorer groups had gone far enough to ensure that many would respond
to his "top-down" efforts to form associations, but it had not reached a
stage where they could form associations for themselves, or where many
other types of voluntary organizations had emerged to compete with
these new associations.

This strategy also enabled Urs to select the leaders for these
associations. This guaranteed their loyalty, and he drew many of them
into politics as state legislators—once again, bankrolling them with funds

3 As the Rudolphs' fully understand, Congress began losing state elections in the late
1960s, and then its leader Indira Gandhi destroyed the organization (which she perversely
saw as a threat) by abandoning intra-party democracy and radically centralizing power within
it. Rival parties took advantage of this and of a political awakening among all sections of
society, to make gains. By 1983, every Indian state had had at least one spell of non-Congress
government. Nearly all of those parties lacked strong, penetrative organizations, including
the Bharatiya Janata Party. Its organization is strong in several respects, but it cannot
penetrate effectively downward towards the grassroots in most rural areas where the majority
of voters reside. For details, see J. Manor, "In Part a Myth: The BJP's Organisational Strength,"
in Coalition Politics and Hindu Nationalism, eds. K. Adeney and L. Saez (Routledge, London,
2005), 55-74. After Indira Gandhi, the Congress failed in nearly all states to regenerate its
own organization. So in recent years, politicians in almost all parties have been forced to
reach voters mainly by other means, especially through official programs which are managed
by the bureaucracy and often protected from patronage bosses who wish to divert resources
to their networks. Such programs qualify as "post-clientelist" initiatives which have become
very important in recent years.

357

This content downloaded from 35.154.245.7 on Fri, 03 Apr 2020 05:46:22 UTC
All use subject to https://about.jstor.org/terms
 Pacific Affairs: Volume 85, No. 2 — June 2012

raised by corrupt means. Enough of these associations and new leaders

succeeded to provide Urs with a strong party organization and a solid
base among disadvantaged groups. He could thus resist the backlash
from the dominant landed castes, gain re-election, and introduce
significant pro-poor policies.4 (This reminds us that on rare occasions,
corrupt "fund-raising" has served progressive purposes.)
Before long, however, further changes made this and similar
strategies infeasible, as caste associations lost their prominence.
Many factors contributed to this. Education, government policies and
economic development spurred substantial occupational and economic
differentiation within castes of all descriptions. There was thus less
commonality of interest within castes, and solidarity within associations
became harder to achieve. Certain sub-sections of castes or small cliques
captured most of the benefits which many caste associations obtained;
and some associations fragmented. These things eroded both the
effectiveness of caste associations and their appeal to many members of
the castes that they claim to serve. A great welter of voluntary associations
and organized interests have emerged, many of which crowd out and/
or cut across caste associations. Other avenues have opened up (within
both the public and private sectors) which offer individuals within castes
better opportunities to advance their interests than do caste associations.
Ordinary people, whatever their caste, have become far more politically
aware, so that at least since 1990, politicians have felt confident about
appealing directly to them, without relying on intermediate institutions
like caste associations, even when their appeals are caste-based.5 They
reach out by promising and sometimes implementing initiatives to
provide selected castes with more reservations in government schools
and employment, and other policies that benefit specific castes, blocs of
castes, or economic strata.
So when politicians like Chief Minister Nitish Kumar in Bihar today
seek support from disadvantaged groups, they offer special concessions
to carefully selected social groups: jatis which have received less than
their share of benefits that passed to a jati-cluster, or other sub-groups
within social collectivities. These leaders proceed on the well-founded
assumption that even poorer people are sufficiently aware that they
will recognize these direct overtures and respond. Nitish Kumar was re

4 For more detail, see E. Raghavan and J. Manor, Broadening and Deepening Democracy:
Political Innovation in Kamataka (Roudedge, New Delhi and London, 2010), part 1.
It was in 1990 that Prime Minister V.P. Singh sought to mobilize support among
disadvantaged caste groups by deciding to implement a commission's recommendations that
a proportion of government jobs and places in government educational institutions should
be reserved for members of the "backward castes" that stand in the traditional hierarchy
just above Dalits (Scheduled Castes, ex-untouchables) for whom reservations had long been
provided.

358

This content downloaded from 35.154.245.7 on Fri, 03 Apr 2020 05:46:22 UTC
All use subject to https://about.jstor.org/terms
 Caste Associations and Politics in India

elected in 2010, partly because he had offered new benefits to certain

jatis and certain sections of the Muslim minority that had been ignored
by his rivals.
It would be wrong to conclude, however, that even ragtag caste
associations are utterly irrelevant today. Some, especially those which
have emerged among severely disadvantaged groups, play important
roles. In recent field research in rural south India, this writer found that
caste associations that seek to serve Dalits (ex-untouchables) are riven
with factional infighting, and can do little to mobilize that cluster of jatis
as a political force.6 But those associations have extended their presence
into most rural arenas over the last two decades so that, despite internal
weaknesses, they now collect evidence of abuses against Dalits and transmit
it to the media and to sympathetic officials at the state level who punish
abusers, under a 1989 law to combat atrocities against those castes.7

ooooooo

Thus far, this discussion has not addressed two monu

that have occurred in India since the Rudolphs' pap
conclude, let us consider these, and the oddities that t
the role and importance of caste associations in re
first change occurred within society: the decline in th
hierarchies over the thinking and actions of rural dwel
of India. The second is political: the marked dispersal
1989, away from the once dominant Prime Minister's
other institutions and forces in the political system.
Curiously, neither of these changes has had much im
associations. Why? The explanations bring us back to
seemed a tiresome point: the link between caste as
different levels in the political system.
Evidence from diverse regions demonstrates tha
caste hierarchies in rural areas has been waning—unev
enough to qualify as a national trend.8 In Dipankar Gup

6 I am grateful to Chandan Gowda for stressing that internal divisi

the emergence of a coherent Dalit "movement" in many states, and
wealth of insights into these issues.
7 The law is the Scheduled Caste and Scheduled Tribe (Prevention of Atrocities)
Act 1989. As its tide indicates, it is to protect both Dalits (Scheduled Castes) and Adivasis
(Scheduled Tribes who stand outside regional caste systems). The law is enforced with vigour
across South India and in several other regions, but some state governments make less
energetic use of it.
8 See for example, G.K. Karanth, "Caste in Contemporary Rural India" in Caste: Its
Twentieth Century Avatari, ed. M.N. Srinivas (Penguin, New Delhi, 1996), 106; A. Mayer,
"Caste in an Indian Village: Change and Continuity" in Caste Today, ed. C.J. Fuller (Oxford
University Press, Delhi, 1997), 32-64; G.K. Karanth, Change and Continuity in Agrarian Relations
(Concept, New Delhi, 1995); S.R. Charsley and G.K Karanth, eds., Challenging Untouchability:
Dalit Initiative and Experience from Karnataka (Altamira Press, London, 1998); D. Gupta, Caste

359

This content downloaded from 35.154.245.7 on Fri, 03 Apr 2020 05:46:22 UTC
All use subject to https://about.jstor.org/terms
 Pacific Affairs: Volume 85, No. 2 — June 2012

"caste" tends increasingly to denote "difference" more than "hierarchy."9

That is as fundamental a change as can be imagined. Indeed, along with
the emergence of a consolidated democracy that is deeply rooted in
society, it is one of the two most important things to occur in India since
independence.
It must be stressed that the waning of hierarchies has not weakened
"caste," if by that we mean the institution of jati (endogamous caste
group). It is the most resilient pre-existing social institution in Asia,
Africa and Latin America, and it remains a robust material reality.10
The important point to stress in the present discussion is that this
extraordinary change has mainly affected social and political dynamics at
the village level, where caste associations have little or no importance. It
has, in many localities, dramatically affected relations between jatis.11 But
the declining power of hierarchies has had little discernible impact on
two things that might have triggered change at the state level: social and
political dynamics within jati-clusters, and people's voting behaviour.12 So
it has induced very little change at the state level, which is where caste
associations mainly operate.
The dispersal of power away from the PMO has occurred because since
1989, no single party has managed to win a majority in the dominant
lower house of India's Parliament. That has forced prime ministers to
head minority or coalition governments, both of which prevent them
from governing in an overweening manner as Indira and Rajiv Gandhi
did until that year. Most of the power which has flowed away from the
PMO has passed to other institutions and forces at the national level,
which now check one another. As a consequence, on only one or possibly
two occasions since 1989 has an Indian prime minister abused his power.
That is far fewer abuses than India witnessed under the Gandhis before

1989, and fewer than the United Kingdom witnessed under either
Margaret Thatcher or Tony Blair. But the point to stress in the present
discussion is that the main impact of the dispersal of power has been felt
at the national level, where caste associations find little traction.

in Question: Identity or Hierarchy? (Sage, New Delhi, London and Thousand Oaks, 2004); S.S.
Jodhka, "Caste and Untouchability in Rural Punjab," Economic and Political Weekly, May 11,
2002,1813-23; and S.S. Jodhka and P. Louis, "Caste Tensions in Punjab: Talhan and Beyond,"
Economic and Political Weekly, July 12, 2003, 2923-36.
I D. Gupta, Interrogating Caste: Understanding Hierarchy and Difference in Indian Society
(Penguin, New Delhi, 2000).
10 For more detail, see the section on "the materiality of caste" in J. Manor, "Prologue" in
Caste in Indian Politics, second edition, ed. R. Kothari (Orient Blackswan, New Delhi 2010),
xxi-xxiv.

II For a preliminary discussion of these issues, see J. Manor, "Accommodation and

Conflict as Caste Hierarchies Wane: Field Notes from Andhra Pradesh and Karnataka,"
Seminar (forthcoming May 2012).
12 See the detailed discussion of "caste and voting behaviour," in Manor, "Prologue xxviii
xxxiii.

360

This content downloaded from 35.154.245.7 on Fri, 03 Apr 2020 05:46:22 UTC
All use subject to https://about.jstor.org/terms
 Caste Associations and Politics in India

Some power has, however, passed downward to governments and

parties at the state level in this federal system, where caste associations,
usually claiming to represent jati-clusters, exert some influence. The
empowerment of state governments has enhanced such associations'
influence, modestly, in some states. However, another consequence of
the dispersal of power to the state capitals has been the emergence of
assertive, radically centralizing chief ministers in (at this writing) six large
states containing just over 45 percent of India's population.13 In such
states, the chief ministers exercise something close to personal rule, and
caste associations—like all voluntary associations—have been rendered
largely powerless. So on balance across India, the redistribution of
power in the political system since 1989 has done little to change the
importance of caste associations.
The comments above are intended to supplement the Rudolphs'
remarkable early study of caste associations, which remains valid to this
day. For India's rural majority, one "caste" institution, jati (endogamous
caste group), remains strong as a crucial arena within which social
interactions occur in rural areas, despite occupational and economic
differentiation within jatis and the waning of hierarchical relations
between jatis. But caste associations have been substantially overtaken
by change, most notably by that differentiation and by the emergence
of alternative voluntary associations and alternative channels to gain
benefits and to achieve influence within the political system.

University of London, London, United Kingdom, March 2012

13 It is important not to overstate this, since the situation fluctuates. But at this writing,
the states are Andhra Pradesh, Gujarat, Orissa, Tamil Nadu, Uttar Pradesh and West Bengal.
Note, however, that in Uttar Pradesh a state election which may change power dynamics is
underway. In Andhra Pradesh, chief ministers' personal dominance has diminished since the
death of the centralizing Y.S. Rajashekhar Reddy in 2009—although the centralized system
that he and his predecessor, Chandrababu Naidu, established is still partly intact.

361

This content downloaded from 35.154.245.7 on Fri, 03 Apr 2020 05:46:22 UTC
All use subject to https://about.jstor.org/terms