Read an Excerpt
Democracy against Development
Lower-Caste Politics and Political Modernity in Postcolonial India
By JEFFREY WITSOE THE UNIVERSITY OF CHICAGO PRESS
Copyright © 2013 The University of Chicago
All rights reserved.
ISBN: 978-0-226-06316-4
CHAPTER 1
State Formation, Caste Formation, and the Emergence of a Lower-Caste Politics
In a penetrating analysis of the symbolism of the Imperial Assemblage of 1877 that was held to proclaim Queen Victoria empress of India, Bernard Cohn (1983) demonstrated that the ritual reflected two distinct "sociologies of colonial rule" that had emerged by the later part of the nineteenth century. An older theory, modeled on feudal Europe, saw India as "a feudal society consisting of lords, chiefs and peasants" with a "native aristocracy" that needed to be respected and, if necessary, even produced, in order to have allies through which to rule effectively (658). A second theory, emerging with force after the rebellion of 1857 and the direct incorporation of India into the empire, saw India as essentially consisting of communities of religions, regional ethnicities, tribal groups and, above all, castes. Such a theory necessitated both the production of knowledge about these communities—the notion being that a lack of knowledge about India was at least partly responsible for the rebellion—and the need to work through the " 'representative men,' leaders who were thought to speak for, and who could shape responses from, their communities" (658). Both "representative men" and "native aristocrats" had been constructed by the imperatives of colonial rule, and both were present in the representational politics of the Imperial Assemblage.
These two sociologies of rule correspond to two different phases of colonial governance (roughly pre-1857 and post-1857) that coexisted rather uncomfortably in the late colonial period, but also to very different notions of caste that continued to shape India's political world after independence. Caste became increasingly imagined as translocal communities that could be democratically mobilized into "vote blocks" represented by caste-leader politicians, but most Indians also continued to experience caste as inseparable from the realities of local dominance by caste lineages who continued to see themselves, even if with increasing difficulty, as lords of the land. Colonialism, then, didn't just shape representations of caste; it also influenced the ways in which caste operated within the agrarian economy, where the "feudal" model of colonial rule still held sway in zamindari-settled areas until independence. In addition, emergent caste identities and caste networks played important roles within colonially constructed governmental institutions. I shall argue, therefore, that the colonial influence on caste must be situated in relation to overlapping, and often incongruous, processes of state formation that have had a lasting impact. The aim of the chapter is not to provide a comprehensive account of Bihar's modern political history—a task accomplished elsewhere—but to examine the ways in which the politics of caste documented in subsequent chapters reflects the contradictions of a specifically postcolonial democracy.
I draw here on Corrigan and Sayer's (1985) well-known study of state formation in England as cultural revolution. As they point out, processes of state formation do not just produce institutions and power relations but also identities and cultural forms. Such cultural transformation—playing out in India according to distinctly colonial strategies of rule—accompanied the governance of caste. And as Gilbert Joseph and Daniel Nugent (1994) emphasized in their theorization of the Mexican Revolution, processes of state formation are best thought of as being shaped by a dialectical interaction between the hegemonic culture and "cultures of resistance." While processes of colonial state formation shaped caste identities, processes of "caste formation" that occurred through a plethora of caste movements in the late colonial period profoundly shaped the functioning, if not the form, of the state. As a Yadav caste activist in 1914 saw it, "The world is full of change!" (quoted in Pinch 1996, 88). But this change was not revolutionary in a decisive way since colonialism ensured that earlier structures of power continued and the emergence of radical discourses have often masked underlying continuities.
The chapter then turns to the ways in which these processes of state/ caste formation continued in the postcolonial period. Independence added new sociologies of rule. India became a constitutional democracy with universal franchise, and Nehruvian planned development promised rapid socioeconomic transformation. But these "sociologies of rule" were superimposed over the colonial ones that Cohn identified and that continued to shape political life—caste as distinct communities with representatives, the realities of caste dominance within local sites, and caste as networks providing access to state institutions. The result was that neither planned development nor the legal provision of liberal rights resulted in decisive structural transformations.
This account builds on analysis of economic change in India as a "passive revolution" (Chatterjee 1986, Kaviraj 1988) that had to be painstakingly negotiated by elites with divergent interests and imposed from above through planned development. But I argue that the upper-caste control of state institutions—a legacy of long histories of state formation—and democratic mobilizations made such a project a practical impossibility. The green revolution occurred just as what became known as the "backward-caste" movement was taking off. Landlords were faced with increasingly politicized backward-caste tenants who could not be reduced to laborers, preventing landlord-driven development, while landlords used their control over state institutions to sabotage the growth of an alternate peasant-driven agriculture. Any passive revolution in Bihar, therefore, had to involve a restructuring of local power and the relationships between state institutions and local power. And the force of this gradual restructuring was caste-based democratization. I therefore argue that the project of state-directed development and democratic practice need to be analyzed as two distinct forces in a single process—a "passive revolution" of capital and a "silent" democratic revolution (Jaffrelot 2003). I will seek to show the ways in which there has been a back-and-forth process between efforts to utilize state-directed development in order to consolidate "historical blocs" from above and caste-based democratic mobilizations that have disrupted this project, utilizing democratic politics in an attempt to actualize transformations from below.
Caste, the State, and Local Control
Historical research demonstrating the changing forms of caste over time, and the crucial role of processes of state formation in these changes, discredits both the Orientalist conception of India as essentially religious in nature and the influential theory of the French anthropologist Louis Dumont (1980) who asserted that caste hierarchy was based on a Brahmanical opposition of purity and pollution, with the Brahman's "encompassing" of the king being the defining feature of the "caste system" (Bayly 1999, Cohn 1983, Dirks 2001). Nicholas Dirks (1987), for instance, demonstrates that in a "little kingdom" in southern India in the precolonial period, caste existed at the local level as lineages whose status (reflected, above all, in temple honors) was derived from proximity with the king. While Brahmans and temples were important for legitimizing kingship, it was the king who was always the first devotee of the deity, and "ritual and political forms were fundamentally the same" (4). Networks of proximity with "the state"—even if, as Dirks points out, Western conceptions of the state cannot be applied here—were constitutive of what came to be known as "caste."
In north India, the Mughal Empire drew heavily from a Rajput tradition of kingship in legitimating and spreading its rule—including strategic marriages with prominent Rajput families and the reliance on Rajput military recruits and officers—a process that facilitated the increasing importance of a kshatriya (kingly/warrior) ethic (Bayly 1999, 34–38). Such an ethic was of crucial importance at the village and regional level, where the lineages of castes that claimed a lordly kshatriya status were the local controllers and revenue collectors for the Mughals (Cohn 1983). "By the later seventeenth century, there were many areas of north and central India where the most fertile lands had become subject to lords based in mud-walled forts who used their bonds of marriage with fellow Rajput, Maithil, Bhumihar or Kanyakubja to command a flow of resources and deference from non-elite tillers and dependent laborers" (Bayly 1999, 35–37).
The case of Rajnagar, the village that will be the subject of subsequent chapters, demonstrates how caste dominance worked during this period. It was within such a context of a Mughal/Rajput state that the lohtamia Rajputs migrated to Rajnagar and, later to surrounding villages (eight lineages claim descent from a common ancestor), in the seventeenth century, occupying an elevated fort (Rajnagar garh) surrounded by two moats, the remains of which are still visible today. As Cohn describes, the regional context was crucial since the lordly lineage not only had to maintain control over its territory, but also had to unite in order to protect itself from would-be conquerors from outside and to maintain leverage vis-à-vis higher-level kings and emperors. The lohtamia Rajputs had to contend with other Rajput lineages—I was told by lineage elders that there was a history of territorial conflicts with the lineage of the legendary leader of the 1857 rebellion, Kuer Singh—with local kings (in this case the maharaja of Dumraon) and, of course, with Mughal revenue demands. But while the legacy of dominant caste lineages is the product of a history that stretches back more than three centuries, colonial rule altered the experience of local dominance in subtle but important ways.
In colonial Bengal, which included present day Bihar, land tenure was regulated through the legal framework of the permanent settlement, enacted by Cornwallis in 1793. Following a feudal "sociology of rule" (Cohn 1983), the permanent settlement was explicitly designed to "create a loyal elite based on landed property" (Dirks 2001, 111), a class of zamindar landlords who had the legal right—a right that could be bought and sold—of revenue collection, exercising a great deal of control over their tenants (Guha 1996, Mitra 1985). Most zamindars had been tax collectors for Mughal revenue extraction but now attained a significant degree of independence, a development that contemporary observers from the old regime fiercely criticized. The desire of the East India Company in India to maximize profits discouraged the costly expansion of an internal security apparatus. Rather, zamindars were allowed to exercise a sufficient degree of control over their respective territories to extract the routine tax revenue from cultivators. Zamindari families also enjoyed privileged relations with the colonial state. From the mid-eighteenth century, the East India Company had encouraged recruitment of members of zamindari households into the Bengal Army, and "in the hope of keeping its soldiers loyal to their 'salt' ... the Bengal Army had adopted elaborate techniques of selection and regimental ritual ... designed to establish an explicitly high-caste identity for the army's regular regiments" (Bayly 1999, 202).
As the power and resources of the colonial state grew, zamindars and their rent-collecting agents (thikadars) not only retained, but also strengthened, their power at the local level. Anand Yang writes, "Endowed with political standing in their localities and by virtue of their positions as local allies of the colonial state, landholders thus enhanced their roles as local controllers" (1998, 68), resulting in what Yang (1989) refers to as a "limited Raj." "With the British leaning heavily on local allies and imperial considerations weighing decisively in favor of minimizing the economic costs of rule, a strong administrative machinery became less of a requisite for colonial government" (Yang 1989, 229). In fact, the only functioning bureaucracy at the local level was operated by the large zamindars, with the local police (chaukidar) and records keeper (patwari) effectively serving as servants of the zamindar, despite repeated attempts to integrate them into colonial administration. At the local level, zamindars were the state.
The zamindars, who enjoyed legally sanctioned territorial control under British Raj, were from locally dominant lineages or, in the case of absentee zamindars, made their own alliances with these lineages. In Rajnagar, for instance, lohtamia Rajputs were given zamindari rights over the village with the Permanent Settlement. Over time, aided by what lineage elders described as a "close relationship" (nazdik rishta) with the British, zamindari rights were acquired over eight other revenue villages (usually comprising several residential villages) through auctions, with some located nearby but others located at considerable distance (with revenue collection farmed out to intermediaries).
Caste divisions served to reinforce class relations within agrarian contexts, primarily through the distinctions between "twice-born" (dwijas) wearers of the sacred thread, peasant castes that did not don the sacred thread and had to settle for inferior tenancy arrangements and provide free or discounted labor and services, and "untouchable" castes whose ritual impurity perpetuated a class of landless and often bonded laborers. These caste divisions underpinned the entire zamindari system. In addition, the steady expansion of population and of land area under cultivation, and the felling of forests, decreased the mobility of many groups, making flight and resistance more difficult in settled areas such as Shahabad. The East India Company largely succeeded in disarming previously armed groups and establishing a monopoly of the use of violence at the regional level (but not within villages). This meant that armed conflict between landed lineages decreased precipitously, weakening the need to secure loyalty from subordinate groups. This concentrated power at the level of the village—with the possibility of expanding territory through conquest eliminated, and the fixed revenue demands of Permanent Settlement encouraging the realization of profits through coercive extraction (especially since zamindari rights had to be purchased). "The chief consequence of these developments with regard to caste was a widespread hardening of boundaries between the superior landed groups and those deemed to be low and 'impure' in caste terms" (Bayly 1999, 203).
The Colonial Governance of Caste
After the 1857 rebellion and the direct integration of India into the empire, it became clear to many colonial officials that the feudal model of rule had become inadequate and, as part of a process of expanding the state's presence, a sociology of rule based on "communities" gained influence (Cohn 1983). This emphasis on communities reflected a colonial version of civil society that, unlike the feudal mode, was conceived as having the capacity to change over time (Cohn 1983, 633; Dirks 2001,16). Colonial governance increasingly supplemented limited rule with the systematic collection of ethnographic and statistical data on caste, and its use within administration. In the process, formerly fluid and localized caste lineages were categorized into homogenous ethnic identities that became central to the workings of colonial institutions. This was especially the case in the judicial system and police force, where certain castes were deemed "criminal" as "part of a larger discourse in which caste determined the occupational and social character of all its constituent members" (Dirks 2001, 181). In the Indian army, the old Mughal practice of recruiting members of aristocratic families was replaced with a policy of recruiting the "martial races" (Bayly 1999, 202–203; Dirks 2001, 177–180). And, perhaps most influentially, the colonial census enumerated broadly defined caste categories, "objectifying" caste as numerical data that could be interpreted, compared, and politically utilized by both the colonial state as well as various local actors (Appadurai 1996, 114–139; Bayly 1999; Cohn 1987a; Dirks 2001, 198–228). The first Indian census was held in 1872 and the decennial census began in 1881. The census became the most visible project for systematically collecting information on caste. The superintendent of census operations in 1921, wrote: "We pigeonholed everyone by caste.... We deplore the caste system and its effects on social and economic problems, but we are largely responsible for the system we deplore.... Government's passion for labels and pigeonholes has led to a crystallization of the caste system."
(Continues...)
Excerpted from Democracy against Development by JEFFREY WITSOE. Copyright © 2013 The University of Chicago. Excerpted by permission of THE UNIVERSITY OF CHICAGO PRESS.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
