On the concept of tribe André Beteille* The Constitution of India

gives recognition to a category of people designated as the Scheduled Tribes, and makes special
provisions for their political representation and their economic and social welfare. The
Scheduled Tribes and the Scheduled Castes constitute the two principal components of the
backward classes towards which the Government of India has adopted a policy of protective
discrimination. This policy has various implications, and questions are raised from time to
time about the exact basis of the social identity of the communities whose members are its
beneficiaries. These matters come up before the courts, and some lawyers at least seem to as-
sume that there must be a clear answer to the ques-tion as to what constitutes a tribe within the know-
ledge of the sociologist or the social anthropologist. Anthropologists have since the time of Lewis
Henry Morgan argued about the definition of tribe, but very little account has been taken of the
tribal communities of India in these general debates over definition. This is unfortunate, not
only because of the legal and constitutional sig-nificance of the problem, but also because of the
size and variety of the tribal population of India. This population comprises about 6.5 per cent
of the total, and stood at nearly 40 million at the 1971 census. There are said to be over
400 tribes in India, and they cover the widest range of variation in terms of race, religion and
language, as well as economic and political organization. It is sometimes said that in putting
the policy of protective discrimination into action officials of the Government of India are
interested not in tribes as such but in only those communities that have been identified and listed as
the Sched-uled Tribes. But this is to evade the issue, for it can hardly be maintained that the
list itself is a random collection of communities put together with no regard for any rational
criteria of classi-fication. As a matter of fact, the criteria used have been several and varied; they
have not always been related to each other; and they have been left implicit more often than
made explicit. The list of Scheduled Tribes now in use has been constructed by the efforts of
various people, beginning more than a hundred years ago. The British Indian civil servants of the
nineteenth century often combined the duties of adminis-tration with the pleasures of
ethnography. Their labours produced a large corpus of reports, man-uals and gazetteers devoted
to the enumeration and description of the peoples of India. The decennial census provided
these ethnographer-administrators with opportunities for constructing a kind of ethnographic map
of India, which they hoped to use in the administration of the country. They proceeded by trial
and error, rather than according to any clearly formulated system of classification. The earlier
accounts of the peoples of India show a preoccupation with the identification and description of
the various tribes and castes in the population, without any clear criterion of distinction between
the two kinds of units. What are now acknowledged as castes are freely de-scribed as tribes, and
vice versa. Even today it is not always easy to tell whether a particular group should be described
as a caste or a tribe. Yet it might seem that if only we followed the anthro-pologist's definition of
'tribe' nothing would be easier than to distinguish between the two. In practice the distinction
between 'tribe' and 'caste' continues to bedevil the student of Indian society, * ISSJ correspondent
in Delhi. Int. Soc.Sci. J., Vol. XXXII, No. 4, 1980
826 and so we must ask whether this does not reveal a deficiency in the conception itself of
tribe. Anthropological convention has treated the tribe as a 'completely organized society', 1 i.e. a
self-perpetuating system having within its bound-aries all the resources necessary for the continued
maintenance of a particular mode of collective existence. The tribe is in this sense a whole
society and a whole culture, unlike the moiety, the phratry, the clan, the lineage or the
family. Each tribe has its own territory, which means that it is politically autonomous, whether or
not it is pol-itically organized. It also has its own language or dialect which is the mark of its
distinctive culture. It has been tacitly assumed that, as in the case of the nation-states of Europe,
common territory and common language go hand in hand. Until recently social anthropologists
have been inclined to take the boundaries of the tribe for granted, focusing their attention on its
internal structure. It is as if a tribe could be understood on its own terms without taking into
account other tribes or other societies of a different kind. The work of Evans-Pritchard among the
Nuer showed that a tribe is a tribe only in opposition to other tribes. 2 The experience of India
(and other Asian societies) would seem to indicate that a tribe might usefully be viewed as a
tribe only in oppo-sition to a social order of another kind. Nineteenth-century scholars viewed
tribal societies in the light of evolutionary theory. This •was true not only of anthropologists like
Lewis Henry Morgan but also of historians like Fustel de Coulanges. Morgan sought to
demonstrate the stages of social evolution by the comparison of contemporary primitive
societies. Fustel recon-structed the transformation of Greek and Roman society from a primitive
to an advanced type. In all of this the tribe represented a type of social organization as well as a
stage in social evolution. The evolutionary perspective has been re-vived in the writings of
Marshall Sahlins and in Godelier's critique of Sahlins. 3 Godelier goes back to the writings of
Morgan to argue that we can understand the tribe as a type of social organ-ization only if we
view it as a stage in social evolution. Now, we may follow Morgan and show how savagery is
replaced by barbarism which in turn is replaced by civilization; or we may use a more
differentiated sequence of stages, identifying the tribal type of organization with a particular
stage in the sequence. But the real problem, it seems to me, lies not in identifying the
evolutionary stage to which the tribal type of organization corresponds, but in coming to terms
with the coexistence of the tribal and other types of organ-ization within the same social and
historical context. And it is precisely here that the evol-utionists of both the past and the
present fail to give satisfaction, in so far as evolutionary theory cannot be a substitute for
historical understanding. The trouble with nineteenth-century evol-utionists was that they too
readily believed that the development of a more complex or a more advanced type of society
led automatically to the effacement of the tribal type. It is a truism that tribe has preceded state
and civilization on the broad scale of social evolution. That is not the problem. The problem is
that tribes have for centuries and millennia continued to exist on the lap of state and civilization,
and to be marked by their impress. The evolutionist is preoccupied with the succession of
types; our problem is how to deal with the coexistence of types in the multi-structural
formations that are a characteristic feature of so many Asian societies. Godelier has drawn
attention to the two schemes of classification presented by Sahlins in his discussion of tribes. In
the first scheme, pre-sented in 1961, there are four types of organization corresponding to four
stages of evolution: the band, the tribe, the chiefdom and the state. In a longer essay, published
seven years later, the scheme is somewhat simplified, and, instead of four, we have three types,
namely, the band, the tribe and the state. Godelier finds fault with Sahlins for collapsing the
two middle terms of his first scheme (tribe and chiefdom) into one, and so making his later
conception of tribe somewhat more elastic than his earlier. In his first essay Sahlins had considered
a segmentary structure to be the defining feature of the tribe as a type of society. The
significance of segmentary political systems was brought to light by British social anthropologists
who had worked in Africa. The initial effect of the publication o African Political Systems
was to highlight the differences between centralized and segmentary societies, characterized by
Fortes and Evans-Pritchard as societies of Group A and Group B. However it soon became
apparent that the dis-tinction between the tribe as segmentary system
On the concept of tribe and the tribe as chiefdom is relative rather than absolute. In the interval
between the first and the second essays by Sahlins, Gluckman had pub-lished his authoritative
work in which he had argued that 'the difference between tribes organized under chiefs and those
which lack chiefs is not as great as it appears to be'. 4 Even such a redoubtable ethnographer as
Malinowski had, it would seem, mistaken for a chiefdom a system that was basi-cally
segmentary in character.6 In the hundred years since Morgan anthro-pologists have learnt to
distinguish analytically between the band, the segmentary system and the chiefdom. But they have
continued, by and large, to apply the same term 'tribe' to all the three. It is to some extent a matter
of convenience whether we emphasize the continuities between the three modes of tribal
organization or their disconti-nuities. For those who are interested in examining as a historical
process the interactions between tribe and state (or between tribe and civilization) there are
obvious advantages in starting with the continuities between the various modes of tribal
organization. The several hundred units that comprise the Scheduled Tribes of India cover all the
modes of tribal organization from the band to the chiefdom. This was so as far back as the
early nineteenth century when the tribal areas began to be systematically opened up by the colonial
admin-istration. Indeed, up to that period one might with some justice speak of 'tribal states' in
addi-tion to tribal bands, segmentary tribes and tribal chiefdoms. Historically, whether a given
tribe was to be reckoned as a tribal chiefdom or qualified as a tribal state depended very much on
the fluctu-ating fortunes of the larger polity of which it was a part or to which it was related, and
not simply on its own evolutionary potential. At the beginning of the nineteenth century the mix of
the different modes of tribal organ-ization among those who comprise the Scheduled Tribes of
today was different from the present mix. Bands of hunters and gatherers such as still exist among
the Andaman Islanders, or, on the main-land, among the Birhors, were more common then than
now. The segmentary mode of tribal organ-ization was also more common in Orissa, Madhya
Pradesh, Bihar and in the frontier areas. But there were chiefdoms as well, in addition to the tribal
states referred to above. 827 The concept of tribe thus faces a double •problem in the context of
Indian society.There is first of all the problem of discriminating among related and overlapping
modes of tribal organiz-ation. There is the equally vexatious problem of drawing clear lines of
demarcation between tribal and non-tribal society. In a sense the first problem is only an aspect of
the second. In North America, Australia, Melanesia, Polynesia and, to a large extent, even in
Africa south of the Sahara, the relationship between tribe and civilization (or between tribe and
state) has been of a very different order from this re-lationship in India and in Asian countries
gen-erally. It is for this reason that a concept of tribe based on the experience of Australia or North
America—the richest fields of classical anthro-pology—can do little justice to the realities of the
Indian situation. In India evolutionary schemes which outline the succession of different types of
social organization must yield to the actual his-torical analysis of coexisting social formations
varying widely in scale and complexity. In North America, in Melanesia and in Australia the
encounter between tribe and civiliz-ation was sharp and sudden; and it has a dramatic, not to say
traumatic, character. The conditions of the encounter were such as to bring- out the contrast and
the cleavage between tribe and civi-lization, rather than the overlap and the continuity between
them. Australia provides the best example: here tribe and civilization represented the two
extremes on the scales of technology, social organ-ization and ideology. Race, language and
culture divided the tribal from the non-tribal population so sharply that there never could be, as
in India, any doubt about their respective identities. In India the encounters between tribe and
civilization have taken place under historical con-ditions of a radically different sort. The
coexistence of tribe and civilization, and their mutual inter-action go back to the beginnings of
recorded history and earlier. Tribes have existed at the margins of Hindu civilization from time
immem-orial, and these margins have always been vague, uncertain and fluctuating. 6 Hindu
civilization ac-knowledged the distinction between tribe and caste in the distinction between
two kinds of communities, jana and/aft', the one confined to the isolation of hills and forests, the
other settled in villages and towns with a more elaborate division
828 of labour. The transformation of tribes into castes has been documented by a large number of anthro-
pologists and historians; undoubtedly, the opposite process also took place, though it cannot be as easily
documented. The tribe as a mode of organization has always differed from the caste-based mode of
organization. But considered as individual units, tribes are not always easy to distinguish from castes,
particularly at the margins where the two modes of organization meet. The native termin-ology itself
reflects this ambiguity. For instance, in the Bengali language the term for caste is jati and the term for tribe
is upajati; but upajati might also denote subcaste. The distinctive condition of the tribe in India has been
its isolation, mainly in the interior hills and forests, but also in the frontier regions. By and large the tribal
communities are those which were either left behind in these ecological niches or pushed back into them
in course of the expansion of state and civilization. Their material culture and their social organization have
largely been related to the ecological niches in which they have lived their isolated lives. Notes 1 This is
Morgan's phrase, repeated in a recent text by Godelier referred in note 3, below. 2 E. E. Evans-Pritchard,
The Nuer, Clarendon Press, 1940; see also M. Fortes and E. E. Evans-Pritchard (eds.), African Political
Systems, Oxford University Press, 1940. 3 Marshall D. Sahlins, 'The Segmentary Lineage: an
Organization of Predatory Expansion', American Anthropologist, Vol. 63, 1961, p. 322-45; Marshall
D. Sahlins, Tribesmen, Prentice Hall, 1968; Maurice Godelier, Per-The isolation of the tribal communities
is and always has been a matter of degree. Some tribes have been more isolated than others, but at least
in the interior areas, where the bulk of the tribal population is to be found, none has been completely free
from the impress of the wider civilization. Their isolation, whether self-imposed or imposed by others,
blocked the growth of their material culture, but it also enabled them to retain their distinctive modes
of speech. Today the most important single indicator of the distinc-tion between tribe and caste is
language. The castes speak one or another of the major literary languages; each tribe has its own distinctive
dialect which might differ fundamentally from the preva-lent regional language. But even this test of identity
does not always work. There are many tribes in western India, including the Bhils, one of the most populous
in the country, who do not have any language of their own, having at an unknown date adopted the
language of the region. It is no ac-cident that from time immemorial the Bhils have also been associated,
both materially and sym-bolically, with some of the most important states in the history of western India.
spectlves In Marxist Anthropology, Cambridge University Press, 1977. 4 Max Gluckman, Politics, Law
and Ritual in Tribal Society, p. 85, Basil Blackwell, 1965. B J. P. Singh Uberoi, The Politics of the Kula
Ring, Manchester University Press, 1962. 6 There is a vast literature on the subject, but the best single
account is in N. K. Bose, The Structure of Hindu Society, Orient Longman, , 1975.