FOREST POLITIES AND TRIBUTARY STATES IN PRECOLONIAL INDIA
Of Emperors and Atavikas
“The Beloved of the Gods believes that one who does no wrong should be forgiven as far as it is possible to forgive him. And the Beloved of the Gods conciliates the forest tribes of his empire, but he warns them that he has power even in his remorse and he asks them to repent, lest they be killed.”
So spoke the great Mauryan emperor Ashoka in the 13th Major Rock Edict, in which he expresses his regret over the carnage of the battle of Kalinga in 261 BC. In the portion of the edict cited above, Ashoka was addressing the Atavikas – the forest-dwellers – who, with the development and expansion of the Mauryan Empire (321-185 BC), had come to play an increasingly salient role in processes of state formation in the subcontinent. However, the stern warning issued in the Rock Edict, in which obedience and acquiescence is demanded from the forest dwellers lest they be subjected to the military might of Ashoka’s empire, represents only one strand of a complex relationship that was woven as the Mauryan state expanded its frontiers. Another, equally significant strand can be found in the Second Separate Edict, where he exhorts his officers to strive to gain the trust and allegiance of ‘the unconquered peoples’ on the borders of his empire, as he ‘desires that they should have no trouble on his account, should trust in him, and should have in their dealings with him only happiness and no sorrow’ (Thapar, 1997: 258; see also Thapar, 1971: 418).
The fact that the balancing of consent and coercion in relation to the Atavikas had become an imperative of statecraft is expressive of a situation in which the expansion of agricultural settlements and trade in products like timber, minor forest produce and minerals had made the forests increasingly important as a source of revenue (Parasher-Sen, 1998: 178; Thapar, 2001: 10-11; Chattopadhyaya, 2004). Thus, Kautilya’s Arthashastra contains carefully crafted advice on how best to manage the clearing of forests and the settlement of peasant cultivators, and how to promote trade in forest products. Kautilya also recognized that the Atavikas could be both friends and foes in this process. In his words, they were ‘numerous, visible, brave, and could ruin a country’ (cited in Thapar, 2001: 11). Consequently, Kautilya also recommended that the emperor should harness the forest-dwellers to the state-making project by employing them as guards on the boundaries of his realms (Chattopadhyaya, 2004; Parasher-Sen, 1998: 183-4).
The two-sidedness of the Mauryan approach to the forest-dwellers can in turn be understood as originating in the fact that the actual workings of the imperial state, despite representing a quantum leap in political organization, deviated significantly from the ideals of centralized administration espoused in the Arthashastra:
Despite its size and administrative control, the Mauryan state does not appear to have attempted a restructuring of all the areas under its control. Possibly a distinction has to be made between what might be called the metropolitan state in such a system, which would be the core region or the area which initiates the conquest and control of the peripheral regions subservient to the metropolitan state but substantially continuing much as before (Thapar, 1978: 160).
The peripheries of the Mauryan state were also the forests, and the imperial centre was compelled to limit its interventions there to the tapping of resources, often in an indirect manner: ‘Thus, it is possible that the chiefs among these forest dwellers collected the forest produce demanded by the Mauryan administration and were the channels by which the administration obtained this tax in kind’ (Thapar, 2002: 198).
This scenario, in which forest-dwelling communities exist at a certain distance from, but by no means autonomously of or in isolation from the imperial state, is, I believe, a manifestation of a persistent liminality that defines the relationship between tribal groups and the ancient and medieval ‘state society’ (Chattopadhyaya, 1997) in the Indian subcontinent.
This liminality can be discerned as far back as during the initial processes of state formation in India that unfolded from the middle of the first millennium BC onwards. As the revenue collected from settled peasant cultivators came to constitute the pivotal material basis of socio-political organization, the contrast between ‘the … civilized and their domesticated landscape and the savages in their wild woods’ (Guha, 1999: 26) emerged as a central boundary-marking trope. The opposition between forest and village settlements was highlighted through the conceptual opposition between aryana and grama: ‘The forest was the unknown, the unpredictable, whereas the settlement was predictable and subject to human laws’ (Thapar, 2002: 56). Other oppositions invoked similar contrasts: the terms that denoted settled areas of cultivation, the ksetra, janapada, and the nadu, were ‘suggestive of spaces with greater human civilization’ compared to the terms that denoted the forest, such as vana, atavi, and jangala (Chattopadhyaya, 2004: 23). Underlying this distinction was also an intimation that the forest and its inhabitants posed a threat to the moral order of the kingdom that somehow had to be curbed. Yet, it is also the case that ‘even on recognizing the dichotomy of the vana and the ksetra, their complementarity immediately surfaces’ (Thapar, 2001: 16). This is evident in the fact that the forests appear as a site of hermitage in Vedic texts, and later on, in the Mahabharata, the forest also features as a royal hunting ground. Thus, the boundary between vana and ksetra was a porous one in ancient India; the former was liminal rather than antithetical in its relationship to the latter.
Under the Mauryas, there arguably occurs and intensification of the relationship between forest polities and the ancient state as the forest increasingly becomes ‘the recipient of the discipline and norms associated with those regarded as civilized. Above all, it was being viewed as a resource to be exploited’ (Thapar, 2001: 10). Yet, the fact that the resources of the forest could only be exploited indirectly and required a delicate balancing of consent and coercion in relation to its inhabitants is significant, because it signifies the partiality of subordination and the negotiated character of imperial incursions into the domains of the Atavikas. This can in turn only be understood if we relate it to the way in which state-making and state-craft in ancient and medieval India was characterized by ‘sharing out sovereignty along different layers of the subcontinental polity’ (Bose and Jayal, 2002: 205).
Precolonial States and the Tributary Constraint
The state, as Neil Smith (1990: 41) reminded us, ‘makes its historical appearance as a means of political control’ when a permanent surplus is being produced and class society crystallizes. Now, the sharing of sovereignty that is so characteristic of states in ancient and medieval India flows from the ways in which these states were embedded in the extraction of surplus from direct producers. On this reading, then, the sharing of sovereignty is ultimately a manifestation of what John Haldon (1993: 156) has called ‘the tributary constraint’.
Tributary modes of production, Haldon argues, were the predominant mode of production in precapitalist societies that had initiated the development of a class structure polarized between peasant producers whose labour generated the social surplus and those non-producing groups – basically state agents and landlords – who appropriate this surplus. This is of course a very wide definition, encompassing vast swathes of time and territory, but it is nonetheless useful in focusing our attention on ‘the political economy of appropriation’ (ibid.: 104) and its ramifications for processes of state formation. The crucial aspect of surplus appropriation under tributary modes of production was that it was organized in the context of what Berktay (1991: 243) calls ‘a hierarchy of interlocking land rights’, usually consisting of the king’s or emperor’s overlordship, the rights of aristocratic and bureaucratic fief-holders and landowners, the rights of the village community, and ultimately the rights of the individual peasant household. In this context, and given the nature of transport and communications technology in ancient and medieval times, the rulers of states could only secure the extraction and redistribution of surplus from peasant producers through ‘fief distribution’ (ibid.: 245) – that is, the monarchical ruler was compelled to either appoint followers and allies as fief-holders in various parts of the kingdom, or to reach an agreement with the pre-existing lordship that had been vanquished whereby the latter agreed to the revenue demands of the former ‘in return for recognition of an irreducible minimum of their traditional status and privileges’. In most cases, the ‘medieval ruling class’ was an amalgamation of these two solutions to the revenue quandary (ibid.: 245). And irrespective of its origins, this ruling class came to function as a centripetal force, as it was ‘forever trying … to extend its control over (part of) the revenue, the land itself – and therefore also trying to weaken royal overlordship or to wrest it entirely out of the hands of the most ‘statist’ faction of the ruling elite’ (ibid.: 260).
It is precisely this constellation that is the primary determinant of what Haldon (1993: 156) calls ‘the tributary constraint’. Due to its dependence on a complex lattice-work of relations with fief-holders and landowners, the power of the tributary state ‘to extract surplus in the form of tribute or feudal rent depends entirely upon its power to limit the economic and political strength of the other classes, but more precisely other fractions of the ruling class’ (ibid.: 157). There is thus an inherent and perpetual contradiction in tributary states between ‘the ruler or ruling elite and those who actually appropriate surplus on their part’ (ibid.: 157) in terms of the organization of the appropriation and redistribution of surplus, and a tributary state faces crisis when its ability to control the groups who appropriate surplus. It is for this reason that governance in the context of a tributary state does not centre so much on the extension of sovereignty outwards from a singular and central point of power and authority situated within a clearly and unequivocally bounded territorial space, but rather hinges on the ability (or lack thereof) of rulers to ‘manage the ebb and flow of this internal tension that was part and parcel of the historical dynamics of all traditional agrarian subject-peasant … societies’ (Berktay, 1991: 160; see also Banaji, 2010: 18-26).
My basic contention is that the liminality of forest-dwelling communities in relation to the monarchical states of the Indian subcontinent has to be understood in relation to the political economy of the tributary mode of production from which these states emerged and were sustained. Although the initial socioeconomic and sociopolitical organization of the Atavikas – these were, largely, kin-ordered modes of production organized politically as chiefdoms (see Ratnagar, 2003) – was distinct from that of the Hindu kingdoms, they were not the autonomous Other of these forms of state. Rather, they were partially integrated in these state societies, and the partiality of their integration was in turn a consequence of the intrinsic sharing of sovereignty in tributary states and the interstitial spaces between subordination and autochthony generated as a result of this. Now, the specific articulation and organization of shared sovereignty has assumed many different manifestations across the time period we are dealing with here, and so has the exact nature of the relationship between tribal forest polities and monarchical states in precolonial India.
Tribes and Tributary States From the Guptas to the Marathas
The imperial reign of the Guptas (300-700 AD) was characterized by yet another round of expansion of settled agriculture as rulings elites sought to widen their revenue base. Consequently, this era witnessed ‘the emergence of a peasant society in erstwhile forest regions’ (Thapar, 2001: 13). This was obviously not without consequence for the forest polities: forest clearances and the construction of road networks entailed incursions into their domains, and in some cases tribal groups were incorporated into caste structures. Nevertheless, this was not ‘a simple, linear process exploiting the forest dweller’ (ibid.: 14). More complex dynamics were at work, which is evidenced by the fact that land grantees in forest regions are likely to have married into the families of tribal chieftains. Moreover, expansion into forest domains brought about a substantial degree of cultural syncretism as Atavika practices ‘became part of Puranic Hinduism’ (ibid.: 13).
Fief distribution under the Guptas in turn established a foundation for the regional states that emerged in the wake of imperial decline in the late seventh and early eight centuries AD (Thapar, 2002: 293). These kingdoms extended their institutional reach by vanquishing and subjugating tributary princes known as Samantas, and then reinstating them on the condition that they pay tribute to the monarch (Kulke, 1995a/b; Chattopadhyaya, 1994, 1995, 1997). However, the resulting institution, the Samantacakara – that is, a “circle” of tributary princes – ensured the persistence of a centripetal impulse in the process of state formation (Chattopadhyaya, 1997: 6).
Regional state formation clearly entailed the expansion of the reach of state power in early medieval India also entailed ‘the extension of statehood into the tribal hinterland and its stepwise integration into the process of state formation increased considerably’ (Kulke, 1995b: 1). There can be little doubt that this entailed a degree of conflict and subjugation – after all, land grants and new agricultural settlements entailed dispossession of forest dwellers – but it would be far to simplistic to reduce the narrative to one of subordination and impoverishment. Whereas I will elaborate on this point in my discussion of the development of Bhil-Rajput during the formation of Mewar state, it is instructive at this point to note two features of the complex political processes that unfolded which suggest that the subordination of tribal forest-dwellers to the Hindu kingdoms of early medieval India was at best partial in nature.
Firstly, as Kulke (1978) points out in the context of Orissa, the small Hindu kingdoms that emerged between 500 and 1000 AD sought legitimacy for their rule among the forest polities by integrating tribal deities into their political culture. This was necessary as the Rajas depended on tribal support to bolster the security of internal communications and borders, and needed their land to extend settled agriculture: ‘Royal patronage of autochthonous deities seems to have been an essential supposition for the consolidation of political power and its legitimation in the Hindu tribal zone of Orissa’ (ibid.: 33; see also Schnepel, 1995). Conversely, this was also a period in which state formation took place among tribal groups in central and eastern India (Singh, 1985: Chapter 2). Surajit Sinha’s study of the Gond, Munda and Bhumij kingdoms shows that this process was rooted in social stratification in terms of ‘differential landholding and the territorial extent of political dominance’. Furthermore, rather than revolving around assertions of Otherness in opposition to Hindu kingdoms, Sinha argues that the new tribal elite appropriated the model of state developed by the Rajputs in northwestern India in 900-1000 AD: they claimed kshatriya status for themselves, promoted brahminical values, entered into marriage alliances with Rajput kshatriya families, solicited the ritual services of brahmins and so on (see also Kulke, 1976). This scenario militates against the construction of an analytical optic predicated upon a binary opposition between the ‘state-space’ of the Hindu kingdoms and the ‘non-state space’ of the tribal forest polities (see Scott, 2009). Rather, what we are confronted with is a composite field of force characterized by crosscurrents of fluidity and liminality, as well the co-existence of accommodation and conflict between tribal groups and kingly rulers.
Between the thirteenth and the early eighteenth centuries, state formation in India was propelled by Muslim conquest states seeking to centralize state administration to an unprecedented extent – first under the Delhi Sultanate (1229-1398) and then under the aegis of the Mughal Empire (1556-1707). The process reached its apogee under the Mughals, who according to M. Athar Ali (1978: 40) – a leading representative of the Aligarh school of Mughal history – orchestrated the systematization of administrative structures, a renewed emphasis on the idea of a social contract between ruler and ruled, and a restructuring of relations between kings and nobles to ensure aristocratic stability in such a way as to forge a ‘large, stable, long-lasting political structure’. Irfan Habib (1999: 364) similarly argues that the Mughal state was unprecedented in covering ‘a whole subcontinent, united under a highly centralized administration’. The institutional moorings of this achievement could be located in the mansabdar and jagirdar systems, which established the basis for ‘a centralized apparatus through which absolute monarchy could function’ (ibid.: 264).
This interpretation of state formation under the Mughals has come in for substantial challenges. C. A. Bayly (1983, 1998) has arguably been foremost among those who have criticized the view of the Mughal state as a tightly sutured Leviathan. Mughal expansion, he argues, resulted in the creation of a form of state in which there was a multiplicity of co-sharers in monarchical power, holding overlapping rights and obligations in such a way that there was not a singular locus of state sovereignty. Arguing along similar lines, Muzaffar Alam (1986:5) characterizes the Mughal state in a very different way than Ali and Habib: ‘The empire signified a coordinating agency between conflicting communities and the various indigenous sociopolitical systems at different levels’ (see also Subrahmanyam, 1992). Now, this is not an argument that is meant to belittle the significance of the Mughal state. It is undoubtedly the case, as Perlin (1981: 281) has argued, that the Mughl era witnessed ‘a remarkable general development in the apparatus of state rule and of the administrative ‘technology’ of appropriation’. However, the tributary constraint persisted and so did ‘the dialectic between centrifugal and centripetal tendencies’ (Haldon, 1993: 219) in the centre-periphery dynamics of the empire (see also Singh, 1988a; Hasan, 2004).
This also meant that the relationship between tribal communities and the state was still characterized by liminality and partial subordination under the Mughals. Chetan Singh (1995: 221) has called attention to the fact that ‘there was much that lay beyond the agrarian economy of the Mughal heartland … The Mughal ruling class did not enjoy an unchallenged monopoly over geographical space and natural resources even within the political confines of the empire’. Large portions of the empire remained forested and uncultivated, and in western, central and northern India, these areas were inhabited by tribal groups who derived their livelihood from a wide variety of sources such as agriculture, forestry and pastoralism (Singh 1988b, 1995). Like their predecessors, the Mughals extracted a share of the surplus wealth generated in these communities through intermediaries, in the form of forest produce and pastoral products. For example, Archana Prasad (1999: 363) writes of the relationship between the Mughals and the Gonds: ‘The nominal annual tribute which they paid to the Mughal emperor signified their loyalty to him and gave them a sufficient autonomous control over their own territories’ (see also Prasad, 2003). To the extent that a change occurred in the relationship between forest polities and the imperial state under the Mughals, it was primarily quantitative in nature. Agricultural expansion created a larger market for forest produce, and as a result of this ‘peripheral communities which had earlier been only marginally affected by the Mughal state were gradually drawn into playing an increasingly important role in its economy, as well as its political fortunes’ (Singh, 1988b: 48).
Much the same argument can also be made about the relationship between tribal communities and the Marathas, who emerged as the pivotal rulers in western, central and eastern India in the eighteenth century (Gordon, 1993). The Maratha polity, like its Mughal precursor, exhibited the simultaneous workings of centripetal and centrifugal forces in a structure characterized by what Perlin (1985: 475) calls ‘quasi-autonomies’ – that is, the presence of ‘sectors of autonomy structurally generated by, and linked up with, the larger order’ – and this also characterized the relationship between Maratha rulers and tribal communities. For example, when Bastar came under the suzerainty of the Bhonsle Marathas of Nagpur, subordination was organized primarily ‘on the basis of tribute payments’ (Sundar, 2007: 51). Beyond this, tribal communities in the region ‘retained their identity, their cults and their own form of political organization, at the same time as they adopted some of the administrative practices and royal rituals of the Mughal and Maratha states, including the patronage of temples and Brahmans’ (ibid.: 51).
Although there is much that changes throughout the historical process of state formation in India, and despite the fact that the state can be seen to expand its infrastructural power over time, the constraints that are intrinsic to ‘states dominated by tributary production relations’ (Haldon, 1993: 203) persisted throughout, generating a sharing of sovereignty that made it possible for tribal forest polities to sustain a position that was neither wholly autochthonous nor wholly subordinated to regional states and subcontinental empires. In my next post I will explore this dynamic in the context of the Bhil heartland of western India.