Chris Smaje. Handbook of Historical Sociology. Editor: Gerard Delanty & Engin F Isin. Sage Publication. 2003.
The key question posed by the classical sociologists which drew them to comparative historical analysis was how to explain the emergence of capitalism, or ‘modernity,’ to use the broader term much in vogue today. This they regarded as a historical development initially unique to Western Europe in the modern period. Their preferred answers were couched in terms of some particular combination of more generalizable structural phenomena, whether this be—in Marx’s case—the struggle between classes over access to the means of production or—for Weber—the developmental possibilities for rational action lodged in cosmological conceptions of the relation between the mundane and supermundane orders. Writing at much the same time, the pioneers of institutional history, such as Otto von Gierke, constructed more specific and continuist historical narratives of the collective institutions that framed the cherished ‘freedoms’ of Western European political culture, while somewhat romantically counterposing these institutions to the ill-starred progress of absolutism and individualism. The work of the early institutional historians paved the way for subsequent inquiry into the historical development of ideas underpinning political society and the way these ideas became manifested in actual institutional forms (Runciman, 1997), even if from a contemporary point of view the perspective of a figure like Gierke seems flawed by an essentialism symptomatic of cultural or ‘ethnic’ nationalism. Nevertheless, in this chapter I will argue that the emphasis placed by such historians upon enduring institutional forms is a necessary complement to the structural approach of the classical sociologists if a convincing comparative historical sociology is to be constructed today; particularly, I shall argue, if ‘institutions’ are interpreted liberally to refer to regularities in the ordering not only of political relations, but also of social relations more generally.
More recent writers have indeed continued to draw from the well of these two traditions of social-structural and political-institutional history, as, for example, in Perry Anderson’s influential account of the specifically European origins of capitalist modernity, which fused a Marxian emphasis on the social relations associated with surplus-extraction with an analysis of the post-imperial political structures of medieval Europe (Anderson, 1974a, 1974b). Asking how the ‘unique dynamism of the European theatre of international feudalism’ could be explained, Anderson averred that ‘no historian has yet claimed that industrial capitalism developed spontaneously anywhere else except in Europe and its American extension’ (1974b: 402). But in the succeeding quarter-century historians have come increasingly close to arguing precisely this, or at least that the priority of European industrialization is explicable in terms of contingencies affecting the modern world-system rather than any characteristics uniquely inherited by Europe from its past. In the contemporary historiography, medieval and early modern Europe emerges as simply one—by no means dominant—geopolitical constellation in an increasingly complex global commercial system. Thus, whereas classical comparative sociology was contrastive, seeking differences which could account for the putatively unique dynamism of Europe, the new approach has been characterized as a ‘connective historiography,’ showing how different parts of the world were integrated into a broader political and economic system through interactive processes whose provenance was not restricted to particular regions (Lieberman, 1997). At the same time, the historiography of colonial societies emerging from postcolonial studies has suggested that the notions of static cultural difference framing comparative sociology owe much to the fictions of colonial ideology. Both strands of scholarship—the connective and the postcolonial—have tried to wrest attention away from the cultural-historical specificities that have long commanded the interest of sociologists, such as caste ideology in India. These tend now to be seen at best as matters of no great import, curiosa better left to academic specialists, and at worst as mere orientalist gestures of differentiation.
These recent strands of revisionism are salutary in highlighting the limitations of unilinear historical explanation and the notions of cultural totality favoured by conventional contrastive sociology. Yet, I shall argue, they do so at the risk of an over-stated universalism that neglects questions of institutional and cultural-historical difference, thereby failing to appreciate the genuine achievements of comparative historical sociology and institutional history. Moreover, in certain respects they replicate the very failings they have uncovered in these earlier paradigms. For example, since Weber conceived only one historical path to full-blown capitalism that emerged in a specific connection between religious ideology and economic action in Europe, the revisionist response has generally been to downplay the connection between culture and economy and to emphasize the singularity of a common trajectory toward economic rationalization in the early modern world (Lieberman, 1997; Perlin, 1993). But this is simply to accept the thesis of a single historical path, rather than to explore the possibility of convergent economic developments through different kinds of cultural interaction. Similarly, in seeking to recuperate people as rational, ends-maximizing authors of their social institutions rather than as the mere vehicles or victims of ‘culture,’ the revisionist emphasis on a universal human ‘agency’ risks a reductionism which abstracts from the socially constitutive character of particular institutional and cultural-historical forms to inscribe people within a very particular (cultural) conception of what it means to be human.
My aim in this chapter, therefore, is to avow along with the ‘connective’ historians that the processes of economic rationalization taken to mark the onset of ‘modernity’ are not to be explained by European exceptionalism but emerged conjointly in a web of global political and economic connections, while attempting to show that this is compatible with local institutional configurations which, in the final analysis, differed from one another in quite fundamental ways. These configurations reflect specific ‘cultural traditions’ that did not simply secure a static and unchanging social order, but did (and still do) evince certain recurrent patternings. This, I will argue, is as true of Europe as anywhere else, so that in contesting the conventional story of European dynamism and non-European stasis, it is as germane to highlight the stasis of the former as the dynamism of the latter. In this respect, one might better speak of several mutually inflecting modernities, rather than a singular global process.
To develop this argument, the chapter focuses mainly upon Western Europe and the Indian subcontinent, pointing out—contrastively—some differences in the way that the relationship between persons and other persons (a domain conventionally understood in sociology under the rubrics of ‘politics’ and ‘kinship’) and the relationship between persons and things (conventionally understood as ‘economics’) have developed in these regions. By focusing upon ‘institutional’ history in this expanded sense with reference to two examples, the chapter shows how distinctive modes of status reckoning, characterized by the terms ‘caste’ in India and ‘race’ and ‘class’ in Europe, reflect the specific character and history of these domains in particular geopolitical contexts, while pointing to structuring principles which are homologous across domains. Yet, it is suggested, these institutional contrasts are not incompatible with the idea of a common economic order that came to connect the two regions from the early modern period onwards, not only because of global processes of capitalist development that diffused into local circuits of production and consumption—though this is assuredly one part of the story—but also because the separate institutional complexes were capable of generating independently similar economic orientations.
Before proceeding to this material, some further words are required on the ideas of politics, kinship and the economy as a means to pursue a comparative sociological enterprise. Much criticism has been directed at the tendency in classical sociology either to reify these domains into a set of structural antitheses (such that a concept as generic as ‘kinship’ can simply be opposed to ‘politics’ or the economy without paying adequate attention to the specific forms taken) or to reduce one to another (the tactic of much Marxist historiography, however hedged by concepts of ‘relative autonomy’). The work of the institutional historians reminds us that the history of specific institutions such as kinship structures or political formations matters, because these cannot be seen as epiphenomenal, mere forms of legitimation for ‘real’ causal forces such as economic interest or power. The distinction often enjoined between a (putatively ‘real’) ‘mechanics of power’ and a (putatively ‘ideological’) ‘poetics of power’ neglects its own ‘poetics,’ its own constitution from a particular ontology of the social (Howe, 1991). To push this insight to its logical conclusion would involve the reflexive recognition that analytic entities such as ‘kinship,’ ‘politics’ and the ‘economy’ are not given sui generis but are themselves the product of particular histories; nevertheless, I suggest that by exploring the homologous logics through which relationships among persons and things are in each case constructed and sustained within variegated and enduring institutional forms, the historical sociologist can employ them as a useful framework for comparison.
These points are now widely recognized in contemporary writings, and the legacy of Marxism and similar approaches lies less in their determinism and more in the idea that different social domains articulate in a multi-valent fashion, such that an ever-emergent social order is produced through the complex relation of its parts. It then becomes necessary to explain how this relation between parts can embody the dynamism (or stasis) which produces temporally and spatially ‘emergent social orders’ of specific kinds, as in the classical question concerning the origins of capitalism or modernity. The contingent effects of creative human agency are often invoked in this connection. However, from the perspective of a contrastive historical sociology at least, ‘agency’ may not best be thought of as a transcendent or general human process of creativity ex nihilo, as is implied in a good deal of the ‘structure-agency’ literature (for example, Archer, 1995), even if this possibility is sometimes unduly neglected. Rather, it may be thought of as a process in which the structure of extant social orders is transformed through the elicitation of its immanent developmental possibilities by people acting as agents, but where the nature of this agency is constituted largely within the horizon of possibilities given by the prior ordering. Hence, the changes wrought upon social orders by human agency involve a process in which certain elements may be recursively elicited in ways that transform them, or even merely reproduce them, within a new ordering (Wagner, 1986), albeit in an unpredictable and emergent fashion. The level at which human agency or historical contingency may properly be lodged is in the unpredictable and emergent character of this process of elicitation.
Race and Caste
Let us now turn to the examples of Euro-American race formation and Indian caste formation with the preceding discussion in mind. Racial ideology has often been thought of as a recent pathology of modernity, in contrast to caste ideology, which has been viewed as an ancient and static cultural system (see Cox, 1987: 37). However, my contention will be that both can be made intelligible with reference to different cultural orderings of relationships between persons in the domains of kinship and politics, orderings which are deeply rooted and historically recurrent, but need not by virtue of this be thought of as securing static social forms. I shall go on to suggest that, notwithstanding these differences, both kinds of status order are compatible with the processes of economic rationalization generally taken to be indicative of modernity.
Race has been defined as an ‘antonym to politics’ (Hannaford, 1996: 13). That is, racial classifications involve defining sets of people as social wholes whereby one or more self-defined groups regard one or more others as partially or completely outside any reciprocal process of normative regulation of social relations, and therefore of mutual validation. Caste, on the other hand, has often been thought of as a ‘failure of politics.’ Here, normative regulation of social relations does obtain, but for some reason—suggestions have included the structural weakness of the regulating authority (Baechler, 1988) or its usurpation by a conquering power (Inden, 1976)—it is incomplete. Taking each case in turn, I want to suggest the utility of the first definition and the disutility of the second as starting points for the historical sociology of race and caste, respectively.
According to the preceding definition, race formation involves the construction of a political relationship which is conceived as an extra-political one. This apparent paradox is typically resolved by a somatization of the political relationship, so that racialized peoples are felt to differ by nature. A somatizing tendency of this kind can be discerned in the categories of social exclusion that began to emerge in medieval Europe from the tenth century, whereby Jews, heretics, lepers, homosexuals (‘sodomites’) and simoniacs (priests who had purchased ecclesiastical office) came to be conjoined by a hegemonic sense that political or religious non-conformity, moral failure and physical deformity transfigured one another, uniting these seemingly disparate groupings as a class outside normal social relations (Moore, 1987).
Scholars have tended to pass over this longer legacy of political somatization to locate race formation in the eighteenth or nineteenth centuries, when the various evolutionary schemata promulgated in the emerging biological and social sciences were used to proclaim an intrinsic incompatibility of peoples or species that was grounded in a now secular sense of natural difference. This certainly contrasted with the pre-modern emphasis on the assimilability of the ‘other,’ as in the traditional Christian view of Jews as ‘witnesses to the faith’ whose eventual conversion would signal the end of the Christian project, a view which began to be replaced in the high Middle Ages by the notion of Jewish guilt for the death of Christ and the poison of the Jewish presence within a healthy Christian body politic. Yet the construction of a somatized political boundary between particular types of people probably betokens a continuity of greater significance than the changes marked by new estimations of its porousness, since its prior existence creates the historical possibility of either its collapse or its reification. This is particularly so in the context of a European political cosmology predicated upon absolute and ranked domain distinctions between the supernatural, the human and the natural worlds which thereby universalized its criteria for defining the human, for as ‘tienne Balibar has argued,’ no such definition of the human is possible without what he terms ‘the infinite process of demarcation between the human, the more-than-human, and the less-than-human (or Supermen and Untermenschen)’ (1994: 197). Of course, the way in which such Untermenschen were conceived differed radically over time in different places, but as I hope to show later it has nevertheless broadly corresponded to a continuity in European political cosmologies which survived the putative ‘rupture’ of modernity and secularization.
Of greatest significance here, however, is the way in which racial boundaries came to be stabilized around geopolitical ones first in relation to the self-conceptions of Latin Christendom, and then in the context of European colonial expansion. Self-consciously corporate political communities emerged in medieval European kingdoms within a broader conception inherited from Rome of a Christian (Western) Europe as an integral geopolitical unit (Hay, 1968; Pagden, 1995; Reynolds, 1997). The process of incorporation was not, of course, an entirely smooth one, particularly in early modern times, when, among other things, the religious struggles of the Reformation unleashed a spate of killings between Catholic and Protestant. Yet while such atrocities were rationalized through a dehumanizing and proto-racial language of political somatization, they were arguably pursued not in the process of constructing racial boundaries, but, on the contrary, like the bitterness of a family feud, in the millenarian fervour of an attempt to restore a sense of a prior collective unity (Greengrass, 1999). In this respect, one might contrast these struggles, in which the enemy was taken seriously enough to be paid the compliment of being killed, with the casual dispossession or enslavement—that ‘fate worse than death’ described by Orlando Patterson (1982)—more commonly applied to non-European peoples in the course of European colonial expansion. Arguably, then, enslavement and dispossession in European colonial expansion were the mirror image of an emerging normative European political community (Eltis, 2000). What then needs to be explained in pursuing a historical sociology of race formation is how the flow of relatedness or similarity between people is averted, and a non-reciprocal sense of normative divergence in political relations formed. This question will be taken up in the following sections.
Turning now to caste, there has been a considerable scholarly reappraisal in recent years concerning its place in the historical sociology of the Indian subcontinent. Textbook definitions of a system of status closure based upon local ranked, endogamous and occupationally specialized corporate groups (jatis) within the broader varna categories of brahman (priest), rajanya/kshatriya (king/warrior), vaishya (‘people’) and shudra (‘servant’) have begun to look increasingly threadbare. So too has the influential theory of Louis Dumont (1980), which derived caste relations from an encompassing and organic ideology of ritual purity. The idea of caste as a static and conservative form of status closure is now generally understood to be a manifestation of the relatively recent rise to hegemony of a specific, universalizing Brahmanic ideology that has not always been dominant in the subcontinent, and not some total and perduring system definitive of Indian society in general. In recent studies, kingship has emerged as an alternative, more inclusive and muscular form of status reckoning than the minute discriminations of ritual rank associated with Brahmanism. It then becomes tempting to construe the brahman-king relation as a local manifestation of a general tension between sacred and secular authority in ‘pre-modern’ societies, akin perhaps to the conflicts between empire and papacy in medieval Europe. I will attempt to show, however, that the comparison is a misleading one, and that the king-brahman tension has a very specific ideological character which is partially constitutive of caste as it has traditionally been understood. It is sometimes argued, in fact, that the proliferation of endogamous, ranked corporate kin groups (jatis) which are often taken to be definitive of ‘caste’ can be found at historical moments when kingship weakened, for example with the decline of the Guptas after the seventh century ad (Chattopadhyaya, 1994), with the usurpation of ‘Hindu kingship’ by Muslim rulers from 1200 ad (Inden, 1976), and with the decline of the Mughals and the arrival of the British in the eighteenth and nineteenth centuries (S. Bayly, 1999). If nothing else, these different periodizations attest to the recurrence of something recognizably ‘caste-like’ at different historical junctures, and similar features can be traced yet further into the past (Thapar, 1978). This is not to suggest that caste is static but, on the contrary, that, in Romila Thapar’s words, it was ‘never confronted with the shadow of its decline’ (1978: 34) because its institutional structure permitted its constant elicitation in novel situations. In particular—and in marked contrast to the European case of race formation—caste ideology has proved able to combine normatively divergent and socio politically dissimilar polities into a shared, over-arching political structure through a logic of incorporation in which the parts retain their distinctive integrity. Whereas this is often put down to the ‘weakness’ of Indian kingship in its inability to subordinate local nodes of political authority, I argue that this does not betoken a ‘weakness’ in any empirical sense, but rather the very ideological basis of political authority which pulses between multiple centres and their peripheries.
Thus, in contrast to the question of race formation, a historical sociology of caste formation is required to explain how a flow of similarity between people is established which mitigates, but does not fully transcend, normative divergence in political relations. What is at issue in both cases is not the working out of two inviolable cultural logics, but the existence of logics with multiple meanings involving contradictions whose elicitation powers social change. Yet the recurrent features in each case would seem to point to certain dominant developmental possibilities, and it is these continuities that are worth tracing in the present context. The next two sections of this chapter, then, will examine how race and caste formation can be understood in relation to the ‘flow of similarity’ or relatedness between persons as it is manifested in two of the domains of human relationships singled out in the introduction, namely kinship and politics.
Race, Caste, and the Ordering of Kin Relations
When we examine kin relationships historically in our two areas of interest we find substantial historical continuities. The evidence suggests that the bilateral, ego-focused, ‘nuclear’ family without corporate affiliations has been a characteristic of kin relations since at least early medieval times in Western Europe (Goody, 1983; Laslett, 1983), a form especially amenable to ‘individualist’ notions of the biological human organism as the basic unit of society. In the Indian subcontinent, kin relations are subject to wide regional variation, but it is nevertheless possible to trace a similar antiquity to several such forms, including the preference for cross-cousin marriage in the south and the tendency towards expansionary, ranked agnatic lineages in the north (Uberoi, 1993). Again, this is not to suggest that kin structures are entirely static and unresponsive to changing circumstances, but that responsiveness tends to take place within an over-arching order of assumptions which has a certain historical resilience. In this section we examine these orders to help link the ideologies underlying political and economic formations to the social construction of the person in the context of race and caste formation.
Anthropological studies of kinship have recently moved away from the idea that systems of kin relations are built upon indivisible or ‘natural’ units constituted by certain human types which are inherently differentiated by gender and biological relatedness, as in labels like ‘mother,’ ‘father,’ ‘daughter’ and ‘son.’ Kinship can instead be modelled as a sense of similarity or analogy which is partitioned and differentiated in various ways to constitute a morally appropriate ordering or system of relationships, its ‘basic units’ constituted by the nature of this ordering, which establishes a flow of analogy between them. This approach was developed particularly in studies of South Pacific societies, and proved better able to model their complex orderings of kin reckoning, gender differentiation and exchange of objects than older models of descent and alliance (Strathern, 1985; Wagner, 1977; Weiner, 1985). One attribute of these systems is their totalizing or incorporative character. A status order is established within social networks which exchange persons (kinship) and objects (economy) not by excluding low-status parties from reciprocal relations entirely, since this would abrogate any social relation at all, but by attempts to augment the density of linkages to centres of prestige at the expense of inferiors. This supposes a ‘holistic’ conception of social relations of the kind that Dumont (1980) conceives at the root of Indian caste relations. In the Indian case it receives indigenous textual theorization at the cosmological level through ideas such as brahman, described in the Brhadaranyaka Upanishad (c.700 bc) as a unitary cosmic substance whose partitioning is homologous with the hierarchical order of caste.
Nevertheless, the idea of kinship as a morally appropriate flow logically implies the possibility of a morally inappropriate flow, which, in Wagner’s words, ‘will appear as a kind of contagion, a moral degeneracy spreading from one kinsman to another’ (1977: 624). Incestuous relationships can be regarded in this way as the ‘morally undesirable flow of similarity’ (1977: 624), a conception that establishes a principle of exogamy defining an internal boundary. In an Indian context, perhaps one might view some of the interdicts on inter-caste relationships in a complementary light as establishing external (endogamous) boundaries for the flow of similarity. Hence, we find a caste logic of careful discrimination between normatively divergent collective wholes which nevertheless figure as part of a more encompassing order, represented ideologically in the concept of jati dharma, the ‘natural’ human tendency to conform to a caste-appropriate code for conduct. This stands in marked contrast to the ‘natural’ human tendency towards unbridled individualism emphasized by Western philosophy in its constructions of a primal and anarchic ‘state of nature.’
A caste logic of this kind permits a range of contextually varied strategies of status enhancement which turn upon maximizing the appropriate relationships with status superiors and minimizing them with status inferiors, strategies which in different circumstances result in characteristic forms of marital alliance such as hypergamy, isogamy or cross-cousin marriage, all of which delimit the appropriate flow of relatedness within a larger circle of possible relationships. Inappropriate flow produces degraded types, which in the cosmology of indigenous texts such as the Manusmriti (c.200 bc) are conceived by reference to hypogamous unions between people of different varna categories, such as the despised Kandala, the ‘lowest of mortals,’ son of a shudra man and brahman woman (Tambiah, 1973). Of the range of possible relationships, this caste schema assigns purity to few of them and impurity to most of them, thereby constituting what Tambiah calls ‘a convenient intellectual device for generating various disapproved categories, assigning them degraded positions, and ideologically explaining and rationalizing why so many groups in the caste hierarchy are placed in low or downtrodden positions’ (1973: 207). In this respect, the caste order is compatible with an agrarian class structure characterized by demographically small landholding classes and a majority class of ‘peasant’ cultivators, organized perhaps as client groups clustering around their landholding patrons, or as junior lineages within a stratified locality society loosely integrated by centripetal ties to a higher-order territorial state. Indeed, critics of the idealism and exceptionalism attending traditional explanations for Indian caste relations have been quick to emphasize these aspects of caste ideology, since they resemble the status order found in many agrarian societies throughout the world (Meillassoux, 1973).
This view can certainly help locate caste formation within a broader comparative sociology, but it risks underplaying some more specific ideological features which might help explain the persistence of caste in non-agrarian contexts, and the peculiar inability of territorial states in India to overcome kin-based local political orders. On the latter point, Fox (1971) suggests that what he calls the ‘web of attributional genealogical linkage’ binding the locale to the region vitiated larger-scale state formation, a point developed by Quigley (1993) in arguing that the strength of locality kinship reflected the weakness of regional kingship in ecological circumstances which were not propitious for the latter. However, rather than imputing an independence to kinship and kingship, it may be more germane to examine the specific character of their interaction in what Subrahmanyam (1986) has called the ‘contact zone’ or ‘middle ground’ between state and civil society. The emphasis here is upon the proliferation of low-level intermediaries such as revenue farmers between the state and the direct producers in a context where no single ruling centre monopolized the use of force, and where the political intermediaries of the ‘contact zone’ cannot easily be assigned a class status in accord with conventional depictions of the lineage or tributary modes of production. This, I argue, is because the particularization of political ‘part-communities’ within a totalized status order involved here reflects the specifically ideological character of state formation as much as the dynamics of the agrarian class order. We shall pursue this point in the following section.
The model of kinship as a flow of morally appropriate similarity can be applied to Western Europe as well as the South Pacific or India (Strathern, 1992), but not without registering certain significant differences. Most importantly in the European case, there has been no totalizing or unitary conception of the field within which social relations occur, but rather a series of domain distinctions through which conceptions of incommensurate ‘natural’ difference are sustained (Smaje, 2000; B. Williams, 1995). In this situation, the exclusion rather than the inclusion of dyadic relations becomes a viable, if not indeed preferred, strategy of status closure, capable of defining a category of effectively non-human persons to whom no social position is attached. This possibility is rendered easier by the fact that in Europe the analogy between flows of persons and flows of things has been weak, so that it is possible to engage in ‘economic’ relationships with others substantially lacking in implications for social status. Thus, as in India, the flow of morally appropriate similarity can be subverted to produce a series of degraded types, but the transgression involves a different generative principle (Smaje, 2000).
Both in India and in Europe, then, the logic of kinship as the flow of similarity created the counter-possibility of morally degraded types, but the implications of this differ. In keeping with the non-corporate, bilateral, ego-focused character of its kin constructs, the modality of relatedness in Europe was of the individual as citizen, or as non-citizen/slave bearing no moral entitlements. The problem then was to combine these atomized individuals into a political community. This was generally achieved by legitimating the state with respect to the nation—the latter conceived as an organic, pre-political community (though one that could none the less operate at various geopolitical levels)—in contrast to the encom-passment of divergent political communities in Indian political systems. It is to these differences in the nature of the political formation that we now turn.
Race, Caste, and the Ordering of Political Relations
Sociological explanations for the different historical shape taken by particular states have generally been couched in terms of a structural or ‘realist’ universalism, whether in relation to the dynamics of different forms of power (Mann, 1988), the oscillation of authority between decaying centres and emerging rimlands (Ertman, 1997; Lieberman, 1997) or the constraints of particular modes of production (Haldon, 1993). There has been little concession within this enterprise to the ‘idealist’ argument that the ideology of states, their particular modes of legitimation, have any constitutive effect upon historical outcomes (and, reflexively, upon the intellectual frameworks of social science itself). Yet I want to suggest that the nature of the political cosmologies invoked to legitimate state power is material to the shape that that power is able to take, and that this can manifest itself among other things as an independent force in the social relations of production (cf. Smaje, 2000: 35-79). In this section, then, I argue after Weber that different cosmologies evince different possibilities for political appropriation in the way that they construct and connect the mundane and the supermundane orders, with significant consequences for the relationship between the state and extra-state political institutions—‘civil society’—and thereby for the shape of the social relations of production, the institutional apparatus of the social order and the structuring of social status.
It has been argued that division into the ‘four orders’ of king/warrior, priest, landlord and commoner often figures in the structuring of political status in agrarian societies, as represented in the Brahmanic varna classification, or the social schemata of eleventh-century European clerics (Ertman, 1997; Milner, 1994). However, it needs to be asked whether such taxonomies should be regarded in terms of classthat is, as representative of actual social groups self-consciously organized around their interests—or as categories, that is, as mental constructs of the social world which may only have potentially informed class formation at particular junctures. In what follows, I abstract two ideal-typical political formations in relation to Europe and India, respectively, the former a ‘sovereign polity,’ the latter a ‘galactic’ or ‘segmentary’ polity. Each represents a particular dynamic of legitimation in relation to cosmological principles of some historical persistence. Whereas the sovereign polity tends to collapse the four orders into a tension between king and commoner, or ruler and ruled, the galactic polity involves their full ideological articulation. The contrast is necessarily somewhat overdrawn. Arguably, there are examples of sovereign type polities in Indian history (Pollock, 1993) and galactic-type polities in European history (Bendix, 1978). Nevertheless, I do want to suggest that the distinction is useful, and that the internal contradictions of each form are capable of developing counter-types which do not thereby undermine the original formulation. It is worth adding that in taking social taxonomies as categories and not classes, this approach does not reify categories such as ‘the Brahmans’ as an actual group of people. This, ironically, often occurs in materialist approaches, which make much of the complicity of ‘idealist’ sociology with the self-conceptions of political elites (for example, Meillassoux, 1973).
The Sovereign Polity
The Peace of Westphalia, concluded in 1648, is conventionally considered the birth of the European system of modern territorial polities, each possessing a comprehensive, supreme and exclusive sovereignty. Whereas this is sometimes viewed as the irruption of a new order—modernity—into the medieval realm of holistic political values in which political authority was partitioned between overlapping associations, subtler treatments have recognized the continuities between modern and pre-modern conceptions of political authority undergirded by the reconfigured apparatus of juristic thought (Gierke, 1934; cf. Runciman, 1997). More importantly for present purposes, I hope to show in the following discussion that the modern conception of sovereignty involves a similar ideological architecture to that informing the medieval ideology of Christian kingship, suggesting that continuities exist in the conception of political society across the supposed watershed of modern and pre-modern thought.
The roots of European political thought are often traced to Hellenic or Indo-European kingship, but these might be more safely glossed as ‘charismatic’ without geopolitical specificity. The charismatic king acts both sacerdotally in connecting his realm to the sacred, and politically as a generous benefactor and distributor of tribute. Herein lies the potential for a separation into two moral persons—the priest-king and the warrior-king which doubtless parallels the personae of actual historical monarchs, for example in the contrasting styles of the Frankish kings Childeric III and Pippin III. However, it is noteworthy that the powerful Pippin first usurped Childeric and then bolstered his sacerdotal claims by invoking papal legitimation. This establishes, if only metaphorically, the important point that in Europe the two personalities of the king have usually been conjoined whereas, as we shall see, the tension between them has been definitive of Indian kingship.
In pre-Christian thinking, the kingdom was usually conceived as a direct mimesis of the sacred order, whereas the Christianization of Europe introduced the now familiar dualism between the sacred and profane orders, at least in its Catholic rather than Orthodox version (Nicol, 1991). The separation was breached by the ‘royal’ personage of Christ himself as both man and god, and the claim to stand in Christ’s place as his vicar was central to the assumption of political legitimacy for Christian kings. Just as Christ’s mediation between the two orders was singular and unique, so was that of the Christian king. There could be no other legitimate point of supermundane authority within the king’s sphere of jurisdiction. Hence, the political theology of Christian kingship was centripetal, strongly oriented to a single political centre. This is not to say that Christian kingship was necessarily ‘strong’ in actuality. The political and military weakness of certain medieval royal houses in the face of seigneurial assertion is notorious. Nevertheless, kings remained at the political apex of society in theory at least, and this unique authority gave them practical advantages—though by no means decisive ones—in the contest for political power. Despite all their reverses, the royal dynasties of Christian Europe remained very stable in contrast, for example, to the proliferating sovereignties of India, and cases where nobles pretended to the throne or kings subordinated themselves as vassals remained aberrant in contemporary political discourse (Baechler, 1988; van Caenegem, 1991).
During the High Middle Ages, the efforts of the reformist popes to assert their supremacy as secular rulers emerged as another rival to monarchical power. Yet this struggle precisely expresses the ultimate monism of Christian political authority, for while a truce between secular and ecclesiastical power was possible through an uneasy division of responsibilities, in the end there could only be one legitimate ruler. In the political tussles between popes and secular rulers, such as the eleventh-century ‘Investiture Contest,’ intellectual partisans marshalled baroque theological arguments to support both the hierocratic position of the papacy and the ‘dualist’ case for royal supremacy, but as Olive Dickason has pointed out, the contest was not between separate secular and ecclesiastical power so much as between what she calls ‘two versions of theocracy’ (1989: 146), not least because both positions shared a conception of the polity as a single corporate body, the argument turning only on what kind of ‘king’ stood by right at its head.
More of an obstacle, perhaps, to monadic conceptions of political authority was the proliferation of territorial sovereignties in Christian Europe, which were never fully to be reincorporated into a single polity modelled after the Roman imperium. Indeed, much of the contest between the popes and kings had to do with attempts by the latter to delimit the former’s sphere of intervention within their territorial jurisdictions. Appropriating the aura of Christian emperorship to their own more local ambit, they successfully established themselves as ‘emperors in their own country,’ to use the juristic phrase. A conception of a kingdom as an indivisible territory associated with a particular ‘people’ and independent of the vicissitudes of any given king had long existed in Europe, but by the thirteenth century it was becoming a relatively pervasive reality. Nevertheless, the disjunction between de facto kingdoms and the universal empire established a tension which was to have longer-term consequences in constituting Europe as a political meta-community.
There is no need to state a position in the debate on the modern or archaic origins of nationalism in order to suggest, as an extensive literature attests, that the territorial European monarchies of the high medieval period enjoyed a popular legitimacy grounded in conceptions of a ‘natural’ connection between ruler, ruled and land, and sustained by a ramifying network of parochial communities under the auspices of a clergy largely subservient to territorial monarchy (Guenée, 1985; Hastings, 1997; Reynolds, 1997). To borrow a phrase of Peter Laslett’s from a somewhat different context, the geopolitical pattern was a ‘reticulation rather than a particulation’ (1983: 57), in which the interlinked ties of political allegiance were ultimately drawn centripetally into the monarchy at their hub. This, of course, looks like something of an idealization corresponding to the ‘elite’ perspective of the hub or epicentre. But it is an important part of my argument to suggest that non-elite political visions, even ones radically opposed to the existing order, were typically based upon the same political geography. This is so partly in the sense that popular political activism often appealed to the unrealized ideals of the extant legal, political and monarchical system—or, in cases of revolutionary regicide, reconstructed a still-unitary ‘supermonarchical’ sovereignty in post revolutionary times, most notably in terms of some idea of the sovereignty of ‘the people’ (Bendix, 1978; Burke, 1994; Sharp, 1985). But the reason for this lies in the more fundamental political cosmology referred to earlier—the unitary mediation between the mundane and the supermundane orders which, as Roy Wagner (1986) has ingeniously shown, can be represented in both ‘hierarchical’ and ‘egalitarian’ terms in historical tension with one another (cf. Smaje, 2000).
One manifestation of medieval egalitarianism was lay Christian asceticism, in which the emphasis on an apostolic poverty oriented to the supermundane world stood as an implicit criticism of the worldly church. The Church was able to incorporate some forms of asceticism such as the Franciscan movement to its advantage in the struggle to assert its autonomy from the lay political sphere, but others such as the Cathars were ruthlessly suppressed as heresy. Heretics were represented as ‘unnatural’ deviants who stood outside the corporate unity of the church, and therefore of society, needing extirpation in order to guarantee the health of the social body. Robert Moore (1987, 1997) has argued forcefully that the concomitant formation of a ‘persecuting society’ from the eleventh century was fundamentally a political project of elites, corresponding to the emergence of a vigorous new civilization in northwest Europe based upon a literate executive cadre. This civilization attempted to make its highly unified politico-religious cosmology hegemonic, the process involving complex struggles in which prejudice was mobilized both to suppress the formation of an alternative literati among Jews and as a conservative tactic of critique in the face of eroding status hierarchies and the increasing abstraction of social relationships through the greater use of money. In Moore’s words,
the rhetoric of crusade, like that of heresy and anti-semitism, and against sodomy, prostitution and usury and indeed when it served their turn the cry for law and order, had a useful place among the battery of considerations with which courtiers might present to their masters, and the masters to their subjects, the continuing necessity for higher taxes, wider powers and a more insistent invasion of local communities and overriding of local custom. (1997: 595)
However, Moore rather understates the involvement of non-elites in the dynamic of persecution, which tapped popular asceticism and millennianism (Cohn, 1957; Rubin, 1991). As the preceding quotation attests, the ‘othering’ and somatization of deviance involved a hegemonic project in the proper Gramscian sense of bilateral equilibria between classes which resulted in a common political imaginary. In this respect, popular egalitarianism involved a corporate and corporeal essentialism of blood and belonging that was no less exclusionary than its hierarchical antithesis, an exemplar of the ‘universalism’ discussed by Balibar that necessarily seals the limits of the human. It was not inevitably so, perhaps. We have evidence of a popular materialist pantheism which, in breaking with any sense of the unitary mediation of the mundane and supermundane orders—and, often enough, denying Christ’s divinity in the process—culminated in the anti-corporate political theology of the radical Reformation (Ginzburg, 1980; Hill, 1972). However, it is to be doubted how widespread these sensibilities really were. The Protestant doctrine of grace, adumbrated as early as the fourteenth century by Wyclif, was deeply ambiguous in its implications for dissolving corporate political identity, and—in learned debates at Constance and Valladolid—proved a superior tool for the proto-racist geopolitical reification of normative divergence, though whether this was replicated in popular unorthodoxy is harder to say (Ginzburg, 1980; cf. Reay 1997). Although the political thought that emerged from the Renaissance in some ways sharpened the tension between hierarchical and egalitarian conceptions of the political order, its dominant development was to establish the rights of European states against their local populations, against each other and against the rest of the world in a descending order of mutual validation. This was certainly the direction taken by humanist thinkers like Grotius and Bacon who defined early modern political thought, but it was built upon an older theme in which geopolitical communities were defined inclusively and successively from the level of the individual to the parish or locality and thence to higher levels such as the kingdom, empire or European Christendom (Hay, 1968; Reynolds, 1997). The process involved the definition of ‘others,’ internal as well as external, who were excluded from the political formation. It also left as its residue more organic conceptions of political unity which were later to inform racist nationalism. But the construction of quasi ‘natural’ or ‘racial’ distinctions was not dependent upon organicist conceptions alone. Popular participation in the definition of sovereign political communities lies behind not only earlier examples of political exclusion within the emerging geopolitical entities of Europe, but also the lack of substantial opposition among Europeans of any stripe to the enslavement of non-Europeans in the Americas until the late eighteenth century (Eltis, 2000), aside perhaps from early—and questionable—evidence of biracial proletarian class formation (Allen, 1997).
A territorial expression to religious or political identity had emerged in Europe, then, by the High Middle Ages, and these identities were often expressed in somatic terms. The appropriation of Christ’s mediating role to underwrite political legitimacy was manifested in a sense of Europe itself as the divine centre of Christendom which encompassed all lesser communities (Reynolds, 1997; R. Williams, 1991). This theocratic warrant for political rule did not survive the secularization of political philosophy between the sixteenth and eighteenth centuries, but the essentials of its structure were maintained. This can be seen, for example, in Thomas Hobbes’s (1968 ) bluntly ‘realist’ rationale for absolutism or monadic sovereignty as a necessary counterweight to the disutility of multiple sovereignties. In John Locke’s Two Treatises of Government (1993 ), Hobbes’s political problematic of unitary or multiple sovereignties was dissolved by articulating the doctrine that rulers and ruled held each other in mutual subjection. However, this agonistic conception of an emergent political order put the Hobbesian problematic of sovereignty into abeyance not by transcending it but by reconfiguring the multiple sovereignties of Hobbes’s state of nature in the more positive light of a responsible right of resistance to oppressive rule by a subject citizenry, so that sovereignty is disintegrated into a kind of field or matrix suffusing the whole social body. This allowed Locke to draw boundaries around political communities not on the basis of mere subjection to government, but according to the qualities of social and political life through which that subjection was conveyed. In the context of Europe this could underpin liberal checks upon absolutism, but it also enabled strong distinctions to be made between Europe and the Americas as political meta-communities. Locke’s political theory, with its emphasis on a sovereignty suffused through the social body, was predisposed to the closure of political communities around a territorial space congruent with the cultural attributes supposedly necessary to the citizen’s proper exercise of that sovereignty—in other words, to the ethnogenesis of territorial nation-states within the broader field of a European political meta-community.
This situation did not exist only in Locke’s political imagination. After the Peace of Westphalia, it was the de facto condition of Western Europe. For Locke and his contemporaries, however, it was not the case that all the people who fell within European boundaries naturally assumed the moral cast of citizens. Rather, they had to be actively shaped into an appropriate citizenry through a pervasive disciplinary regimen which inculcated techniques of self-discipline and an increasing levelling of status distinctions between the individual and the political formation. The political formation was therefore actively and continuously constituted by working upon the dispositions of the people it enclosed and counteracting, as it were, the centrifugal forces tending to disturb their integration within the central values of the nation-state. Political theory of this kind paved the way for a nationalism which emphasized the necessary cultural attributes for membership of the national political community. If it was not to succumb to a racism which invested these attributes with the imprint of an inherent biology until the nineteenth century, the character of this and later ethnonationalisms was surely dependent upon the potential for organic political closure which emerged within this earlier formation. That even so rationalist and universalist a thinker as Locke could coolly countenance the slaying of people ‘not under the ties of the common law of reason’ as ‘beasts of prey… dangerous and noxious creatures’ (1993 : 123) betokens the generally limited opening for the normatively divergent ‘other’ within European political thought. This ever-present face of colonial racism emerges directly from the secularization of monadic sovereignty in Europe.
The Galactic Polity
Of the ideal-typifications that scholars have offered for Buddhist and Hindu political formations, a recurrent theme is of a political authority which—unlike the sovereign state is shared and divided, replicated several or many times at different levels of political inclusion, so that the polity is conceived as a cluster of political centres, sometimes drawn into the orbit of a major exemplary centre, sometimes with substantial autonomy between centres so that power ‘pulsates’ both centripetally and centrifugally within the polity, while the pattern is repeated at lower levels of geopolitical organization. This would apply, broadly speaking, to Stein’s (1980) concept of the ‘segmentary state’ or Tambiah’s (1977) analysis of the ‘galactic polity.’ Both authors limit their analyses mainly to specific empirical cases, but their work arguably has wider applicability to Hinduized kingship. In this section, I want to draw out the implications of this type of political formation by contrast with the preceding analysis of the sovereign state.
Scholarly explanations for the form of the galactic polity have, predictably, divided between the familiar alternatives of materialism and idealism, notwithstanding ingenious but perhaps ultimately evasive attempts such as Tambiah’s (1977) to mediate between the two. Materialist explanations emphasize the empirical weakness of South Asian kingship, its inability to prevail over more local forces, due to the limitations upon the direct extraction of surplus value imposed by rice agriculture (Tambiah, 1977) or the capacity of non-dependent peasantries to escape economic exploitation by flight to other polities (Chibber, 1998; Quigley, 1993) resulting in enduringly ‘feudal’ relations oriented to weak political centres (Baechler, 1988). None of these explanations is especially convincing. It is difficult to read off the sovereign and galactic polities as forms determined in any sense by their respective dependence on wheat and rice agriculture, and the issue of peasant flight begs the question of why political alternatives were so readily available to a peasantry which is exogenously assumed to be non-dependent, especially when this is placed alongside the geographical expansion of the peasantry in medieval Europe, which occurred within the ambit of centralized political control (Duby 1972). It may in fact be the case that the advantageous position of the Indian peasantry stemmed precisely from the independent proliferation of political centres. Such a speculation is strengthened by the recognition that local political orders in India were not generally autonomous ideologically, as one might expect to be the case if a community fell under the suzerainty of a weak ‘alien’ centre, but were thoroughly penetrated by that ideology to the extent that local status distinctions precisely replicated the larger political cosmology. This might be taken as evidence that the particulation of political authority in India was written into the very conception of that authority.
Certainly, this is the implication of a good deal of writing on the political cosmology of the Indian subcontinent, particularly as manifested in its textual traditions, where, to quote Diane Eck, instead of a ‘discourse of exclusivity and uniqueness, more typical of the monotheistic traditions of the west,’ Indian thought evinces a ‘system of meanings in which significance is marked not by uniqueness, but by plurality and duplication. Those things that are deeply important are important enough to be widely repeated’ (1998: 166-8). Perhaps, then, the impetus to find materialist explanations for Indian political ‘weakness’ stems from a particular cultural standpoint in which political ‘strength’ is always equated with the idea of the supreme, comprehensive and exclusive rule associated with the sovereign state.
Scholars who have taken seriously the particularity of the way that authority is constituted in Indian political thought have produced impressive analyses of the tensions between different figures in the Indian political firmament, such as the ascetic renouncer, the priest, the king, the bandit and the clown (Burghart, 1983; Dumont, 1980; Heesterman, 1985; Shulman, 1985). For whereas in the European sovereign polity all of these figures were subordinated to the king as the true and full embodiment of divine (political) authority-notwithstanding a certain ambiguity about who the king actually was—in India each of them could mount an autonomous claim for his own supremacy, though once again it is better to think of them as categories which various protagonists could appropriate to themselves, rather than as given kinds of person. The key tension here is between the king and the brahman, who each try to secure their authority by connecting with their supermundane ‘alter egos’; the king as warrior (kshatriya) to the priest-king (rajanya), and the brahman as priest to the ascetic renouncer. Yet neither is able to achieve complete transcendence, and they fall back into a dependence upon one another that flows from their mundane personae. David Shulman characterizes this interdependence as an ‘unending circulatory process,’ in which control ‘is always threatening to break down as both the major figures, saddled with impossible ambivalence toward each other and toward themselves, cling reluctantly to the thin lifeline of their common… distaste for disorder’ (1985: 88). None the less the point is that here we confront two genuinely different forms of political legitimation, in contrast to the European case, where all forms of legitimacy terminate in the conception of the king.
Various strategies were available to Indian kings in overcoming their defective authority, some of which—such as temple patronage and land-grants to brahmans—were quite successful. Nevertheless, royal authority was ultimately circumscribed by the fact that the king was a partial embodiment, representing just a part of a more encompassing deity or cosmic order to which people had other kinds of access, in contrast to the appropriation of divine authority in Europe, where the sovereign was the representative of God (Inden, 1995; cf. Smaje, 2000: 106-27). Moreover, whereas the ideal king was characterized as a secular benefactor in both cases, in Europe this was an expression of a more fundamental paradigm—the gift relationship—in which dyadic exchange is constitutive of a set of solidary relations and a status order, indeed of nothing less than ‘society’ itself (Mauss, 1979 ). In India, on the other hand, exchange of this type was circumscribed by another conception of giving, in which one’s impurity or evil is expunged by objectifying it into a ‘gift’ (dan) which is passed unilaterally to a status inferior (Parry, 1986). The ‘society’ that emerges from such a system is one that is particulated and not reticulated, and one of its consequences is the replication of rituals of status contest which are definitive of caste at every level of status.
The foregoing discussion is, of course, couched at the level of a somewhat idealized political theory. Yet one can discern this pattern of dislocation between centralized political authority and semi-autonomous locality societies as a persistent feature in many Indian polities, just as in Europe a monadic conception of sovereignty has persisted. Medieval India had witnessed a long and in some instances violent theological struggle out of which emerged a devotional and theistic Hinduism centred on temple complexes, particularly in the south. These were able to encompass local cults—though not always without significant tensions—and offer up their local legitimacy to a higher level via royal patronage. Still, the resulting connections remained delicate and provisional and, as a large ethnohistorical literature attests, kings were rarely able to cloak or dominate other foci of political authority very successfully, or for very long. Moreover, once again political opposition from lower levels was manifested through the same over-arching political imaginary (cf. Schnepel, 1995; Tooker, 1996). Nothing quite like the ‘persecuting society’ of Europe therefore manifested itself in India since the emphasis was always upon the sub-sumption rather than the excision of ‘difference,’ as, for example, in the efforts even of Brahmanic texts to show that religious heterodoxy was inherent to the cosmic order, rather than an affront to revealed truth (Shulman, 1984; Thapar, 1992).
Thus, the galactic-type polity evinces a different kind of contradiction to that of the sovereign polity, the former involving a limitation to centralized political authority and the latter a limitation to political universalism in terms of the incorporation of normative divergence. Incorporation of normative divergence is indeed precisely the tactic effected through caste formation in India. And, although the changing elicitation of political legitimacy has arguably followed a less determinate historical trajectory of secularization there than in Europe, caste formation has nevertheless flowered at historical junctures when political power was in flux, as, for example, during the rise of Muslim polities in medieval North India, the decline of Mughal suzerainty in the eighteenth century and the rise of British hegemony in the nineteenth (S. Bayly, 1999; Inden, 1976). Certain features recur in these very different examples: the tension between king, brahman and renouncer as political models, with their different possibilities for cementing political hegemony; the logic of pollution, which can be used to regulate the boundaries of clean and unclean caste status; and a logic of incorporation where discrepant local political orders can be subsumed into larger ones without any need to lose their autonomy via some Lockean process of disciplinary citizenship. Susan Bayly (1999) has convincingly argued that it was only in the British colonial period that the formal distinctions of caste became so general as to constitute a more-or-less all-India sociopolitical framework, despite the ancient roots of its composite features. But this is not the same as saying that caste ideology was sprung anew on hitherto unwitting peoples. The subcontinent had long been integrated around overlapping networks of political power, sacred space and trade, so that its manifest pluralisms of custom and political style can be inscribed with the larger ambit of a determinate meta-society. It can thus usefully be regarded, like Europe, as a ‘culture area’ in the old anthropological sense. This is not because of some mystical cultural connection uniting particular peoples but because cultural conceptions, promulgated through decidedly non-mystical means, are constitutive of perduring social orders in the face not only of their objectification in inherently conservative institutional structures, but also of the relatively limited human capacity for radical intellectual rupture in its conception of the relationships between persons.
As we have already seen, comparative historical sociologists have conventionally argued that the emergence of industrial capitalism in Western Europe was the culmination of a unique historical trajectory, a ‘European miracle’ with deep historical roots. Scholars have singled out a variety of possible explanatory factors, including: the logic of this-worldly ascetic accumulation underpinning the potential for the complete alienation of the products of human labour (Weber, 1930); the peculiarities of a class structure in which peasantries were sufficiently strong to break seigneurial power but too weak to establish free tenures (Brenner, 1985); the development of a comprehensive law sustaining private property rights, including property rights in labour (R. Williams, 1991); and the peculiarities of a socio-economic system whereby the lower to middle strata were at once thoroughly integrated into higher-level political authority and yet enjoyed an unparalleled local autonomy in political and economic affairs—the ‘multiple acephalous federation’ described by Michael Mann (1988: 11). This situation is contrasted, in the orthodox argument, with pre-capitalist tribute-taking societies where the state acts as a predatory, surplus-extracting mechanism from a peasant society with which it evinces no organic connection, instead standing astride otherwise autonomous locality societies. Although it is conceded that such states can develop these mechanisms into a sophisticated and quasi-modern apparatus—as, for example, among the Mughal rulers of early modern India—it is argued that they are ultimately unable to overcome the class contradiction between expanding state revenue exaction and the exertion of peasant interests. In the case of the Mughals this situation culminated in their eighteenth-century demise at the hands of military alliances between peasant cultivators and zamindars (landlords) squeezed beyond endurance by revenue extraction, precipitating a political instability which served British colonial interests (Habib, 1963; Haldon, 1993).
Yet a recent revisionist historiography has called into question such neat polarities. Taking the example of India, it has been argued that the apparent political ‘chaos’ of the eighteenth century conceals the retrenchment of more streamlined and economically ‘progressive’ polities (Alam, 1997; Perlin, 1994). One consequence of this has been a renewed emphasis upon the relative weakness of the central state apparatus, but the more important point for present purposes is the emphasis on economic dynamism rather than political stagnation that emerges from the revisionist historiography, which has shown how local economic decision-making in India, as in Europe, had already by the eighteenth century been incorporated into an increasingly global and certainly ‘acephalous’ system of commodity production. This system broke any substantial peasant autonomy remaining from resistance to local political incorporation in both regions (Perlin, 1983). At the same time, Indian mercantile structures had reached a level of sophistication and rationalization in no way inferior to European ones, which indeed began their colonial penetration of the Indian subcontinent rather lamely as small-time intermediaries in the larger circuits of the South Asian trading world (Chauduri, 1994). In this context, arguments for a European ‘miracle’ deeply rooted in its distinctive past look decidedly shaky, especially when it must further be conceded that it is not ‘Europe’ that underwent capitalist industrialization, nor even some entity within it such as England or the Netherlands. Rather, particular parts of these countries industrialized in the face of the non-industrialization of other immediately neighbouring parts. These parts need not be thought of as belonging inherently to a given socio-political formation, as an implicit nationalism often leads one to suppose. Rather, the trading world of the eighteenth century comprised a set of powerful mercantile centres connected to extensive hinterlands of rural industrial production in both Europe and Asia.
In this context, the origins of European colonial hegemony lie not in any particular technical or fiscal superiority, but rather in the peculiar advantages conferred by the institutional structure of its overseas trading companies, which—drawing upon the impressive resource-generating potential of their joint-stock organization—were able to suppress autonomous European commercial activity and subordinate it to a rigorous political hierarchy (Chauduri, 1994; Guha, 1999; Richards, 1993). This degree of control proved a sharp weapon in undermining the more refractory political and economic forces of their Indian rivals, and in the longer run proved decisive in their usurpation. Whether the structure of the trading companies can simply be read off from the centripetal characteristics of European polities, or whether it is better understood as one pragmatically developed in the course of the long-term intra-European rivalries over far-flung seaborne colonial empires, is something that cannot detain us here. The point is that the eventual dominance of the global trading system by Europe and its American offshoots has to do not with its own intrinsic processes of economic development, as the orthodox position suggests, but with the particular circumstances of its insertion into that system. Ironically, in this context, it was not the acephalousness but the singularity of European commercial colonial command structures which enabled them to achieve control. Only after the balance had been fate-fully tipped towards the West did such arrangements come to seem anachronistic according to emerging doctrines of economic liberalism which did not in themselves constitute the decisive rupture with pre-capitalist economic organization.
There were, then, substantial similarities in economic agency between early modern Europe and South Asia. However, I now want to suggest that the entry of both regions into this unified modern trading world is compatible with prior developmental tendencies within their respective political cosmologies, and need not be taken as some irruption into a homogeneous global modernity which makes those cosmologies marginal or irrelevant.
It is conventional to consider the emergence of capitalism in Europe as a consequence of institutional rather than ideological features in a pre-capitalist world whose Christian moral framework was antithetical to the rational accumulation of money as an end in itself. In this view, when the inevitable ideological transformation occurred, it came in the form of the thorough overhaul of Christian morality worked by the early rationalist prophets of modernity, who were able to synthesize urban commercialism and popular piety against the trappings of the ancien régime in a proto-nationalist context whereby property rights in personal labour had already developed as one element in a contractual nexus between ruler and ruled. This offered modest protection to the labouring peoples of Europe from the worst excesses of ‘rational’ labour exploitation such as chattel slavery (Eltis, 2000). However, there were also older roots to the commercial imperative, which more restrained medieval thinkers such as Thomas Aquinas had sought to reconcile with the contemporary critique of commerce. The emphasis of the church upon the fellowship of Christians over and against the more immediate ties of kin and community, and its devaluation of the trappings of the present world as against the truly significant spiritual world to come, both conspired to establish a logic of generalized and disinterested exchange in which persons were sharply separated from things (Parry, 1986). The emphasis of classical republicanism upon ethical relationships within an undivided community also leant weight to the same logic from a more secular direction when its influence was felt in the republican and commercial citadels of post-Renaissance Europe (Pagden, 1990). Thus, an ideological apparatus was available to support the emerging institutional order of European colonialism and capitalism through reconfiguring an older emphasis upon individual spirituality and domestic economic autarky into the idea of the nation itself as a gigantic community of kin where spiritual values were represented in a political ‘manifest destiny’ moving in lockstep with economic expansion.
When attention turns to early modern India, it is harder to find the same emphasis upon individualism, person-thing distinctions and a logic of market utility. Perhaps the situation bears greater resemblance to the kind of non-monetary ‘gift’ society classically described by Marcel Mauss (1979 ), in which exchange relations are entered not between individuals but between groups that figure as irreducible ‘moral persons’; in which the things given embody social attributes of the person; and in which the logic of the polluting and inferiorizing religious gift or dan establishes internal social boundaries antithetical to the notion of collective fellowship.
These attributes have tended to be seen as representative of a form prior to, and incompatible with, capitalist commodity logic, but this evolutionary assumption is open to serious question. For example, David Rudner (1994) has shown that a ‘rational’ logic of accumulation among commercial castes in India can be lodged at the collective level of the caste itself, and the idea that a community of trust unified at the societal level is a necessary precondition for capital accumulation is then revealed as an unnecessarily restrictive assumption grounded in specifically European ideological struggles from which our contemporary conceptions of modernity and society were born. Moreover, even if it is the case that in India the interpenetration of persons and things hedges transactions with greater status ambiguity, this need not be incompatible with the ramification of exchange relationships. Higher-status trading castes might maximize transactions in ‘subtle’ substances like money, and minimize transactions in ‘gross’ substances. The latter might include the exchange of persons in kin alliances, something rigorously restricted within endogamous marriage circles; other ‘gross’ substances might include such items as vegetables and cattle, whose exchange fell to inferior trading castes (Grover, 1994; Rudner, 1994). There was therefore considerable scope for a ramifying process of capital accumulation based upon caste interdependencies around the margins of state agrarian revenue exaction, particularly as the grip of great imperial dynasties such as the Mughals weakened, allowing for the proliferation of local polities in which rulers drew upon the kshatriya ideology of the king as a recklessly generous benefactor whose insatiable and none-too-fastidious tastes fostered economic activity and ordered the difficult relationships between castes (C. Bayly, 1986; S. Bayly, 1999; Washbrook, 1988). The royal endowment of temples, and endowment by wealthy merchants aping royal virtue, was another mechanism of economic development built upon ancient foundations, although scholarly opinion is divided as to the potential of temple endowment to circumvent the limitations of agrarian revenue exaction (Chibber, 1998; Washbrook, 1988).
What I hope to have suggested in the foregoing analysis is that the emerging global trading world of the early modern period involved both Indian and European protagonists who, for a time, were on a roughly equal footing as participants in a process of capital accumulation. This was compatible with a certain combination of indigenous political and economic traditions in both cases. It is unnecessary either to de-emphasize caste relations and the refractory nature of Indian polities or to trumpet the arrival of a ‘modernity’ with a uniquely European provenance in order to establish this. Capitalism, understood as the ramification of social relations of commodity production, is compatible with both Indian caste society and the European nation-state, and the reasons for the priority of large-scale industrialization in certain parts of Europe under the impetus of capitalist relations of production are not to be found in any gross civilizational differences between it and other parts of the world. Nevertheless, such differences do exist and they inhere not only through material accumulations in given institutions but also in an enduring, though far from static, structure of ideas invested in institutional forms. In the contemporary political climate, where notions of culture and tradition so often serve the ends of repressive and authoritarian political programmes, there are doubtless good reasons for a historiography of multivalence, polysemy or ambivalence. Here, though, I have tried to argue that the questions of difference raised by the comparative historical sociology of institutions should not be prematurely foreclosed.