There is no single and accepted criterion by which to distinguish a tribe from a caste. Andre Beteille (1977: 7) has claimed that the concept of ‘tribe’ can be understood clearly not on the basis of its existing definitions but by finding out the attributes and analyzing the specific conditions in India which are distinctive of groups conventionally regarded as tribes. For example, social anthropologists like Nadel have described tribe as “a society with a political, a linguistic, and a cultural boundary.” This means that a tribe is a society the members of which have a common government, share common language, and hold a common culture or beliefs and practices. Though many tribal societies have a clear cultural and linguistic boundaries, if not the political one, but there are several tribal societies which lack government and the centralized authority in the ordinary sense of the term. Likewise, cultural homogeneity in a tribe is also elusive because in this age no iron wall exists where one ‘culture’ comes to an end and another begins. A common dialect, however, is possessed by the tribes. The prevalent definition of tribe is, therefore, inadequate.
Scholars like Ghurye, Naik, Bailey and Verrier Elwin, etc. have used different criteria for distinguishing between caste and tribe. Some of these criteria used are: religion, geographical isolation, language, economic backwardness, and political organization.
It is said that the religion of the tribal people is Animism and that of the caste people is Hinduism. Hutton (1963) and Bailey (1960: 263) believe that tribal people are not Hindus but are animists. The basic tenets of Animism are: phenomena of sleep, dream and death, and belief in possession, in spirits and ghosts, and in magic. On the other hand, the principle characteristics of Hinduism are: dharma, bhakti, and rebirth. It will be wrong to say that the Hindus, particularly the lower castes, do not believe in spirits and ghosts or in magic, dreams, etc. Similarly, there are many tribals who worship Hindu gods and goddesses, celebrate Hindu festivals and fairs, and observe Hindu customs, traditions and rituals. It is, therefore, not easy to distinguish between Animism and Hinduism. Ahuja (1965), Verrier Elwin (1943), and Risley (1908) also maintain that the distinction between Hinduism and Animism is artificial and meaningless. This means that because we find admixture of Hindu religious elements and values in tribal religion and tribal values in Hindu religion, religion as a single criterion cannot be used to distinguish between a tribe and a caste. Ghurye, Naik and Bailey have also rejected this criterion.
Using geographical isolation as a criterion of distinction, it is said that the tribals live in geographically isolated regions like hills and mountains, but Hindus live in plain regions. Due to lesser contacts with the civilized neighbours, tribals are more uncivilized than the Hindus. It may be true by and large that tribals live in hills away from the lines of communication but we have examples which show that many caste Hindus also live in isolated regions and many tribals live in plains. This means that in addition to a purely geographical isolation, we demand other criteria also to distinguish a tribe from a caste.
The third criterion is language difference between a tribe and a caste. It is suggested that each tribe has its own language but not a caste; for example, Gonds speak Gondi language, Bhils speak Bhili or Vagdi language, Santhals speak Santhali language, and so on. But since there are tribes which do not have their own languages but speak a dialect of one of the main Indian languages, as in South India, therefore purely cultural criterion of language also is not a scientific criterion for distinguishing between a tribe and a caste.
Economic backwardness too is not a correct criterion for distinction between a tribe and a caste. To maintain that tribals are backward and primitive but caste Hindus are not is not a correct statement. It is true that many tribes even today are economically backward; they have low income, use primitive methods in cultivation and in some cases still use barter system in exchange, but there are many tribes (for example, Meena) which are economically advanced. At the same time, there are many castes which are as much economically backward as many tribes. Bailey (1960: 9) also rejects this criterion by holding that in so far as the phrase ‘economically backward’ refers to a standard of living rather than to a type of economic relationship, it is sociologically unsatisfactory. He has suggested that instead of taking the totality of behaviour, we should narrow the enquiry (in differentiating between a tribe and a caste) by concentrating on particular fields of behaviour in a given society. He, thus, used politico-economic system or ‘economic structure’, as he calls it, for differentiating between a tribe and a caste in his study of Konds (tribe) and Oriyas (caste) in Orissa. In the analysis of the politicoeconomic organization, he concentrated on two factors: (i) control over land, and (ii) right to resources of land. He maintained that in both the tribal and caste societies, we find ‘landowners’ who have direct access to land, and ‘dependents’ who are dependent on the landowners for achieving their share of land’s resources. But analyzing the economic organization of a village territory (inhabited by castes) and a clan territory (inhabited by tribes), he found that a village is divided into economically specialized interdependent castes arranged hierarchically, whereas though a clan territory is also composed of groups but these are not hierarchically arranged and nor they are interdependent through economic organization. In other words, in a tribal society, a larger proportion of people has a direct access to land while in the case of a caste-based society, the larger population of people achieves the right to land through a dependent relationship.
Thus, according to Bailey (Ibid: 264-65), a tribe is organized on a ‘segmentary system’ and a caste is organized on an ‘organic system’. He writes: “The only solution (to differentiate between tribe and caste) is to postulate a continuum, at one end of which is a society whose political system is entirely of the segmentary egalitarian type and which contains no dependents whatsoever, and at the other end of which is a society in which segmentary political relations exist only between a very small proportion of the total society, and most people act in the system in the role of dependents. The political system of this society can be compared with an organic system.” But he holds that at what point of continuum a tribe ceases and a caste begins is impossible to say.
In India, the situation is even more complicated because there is hardly any tribe which exists as a separate society. No tribe in India has a completely separate political boundary. Big tribes like Bhils, Santhals, Oraon, etc. are territorially dispersed. Further, almost all tribes have been absorbed in varying
degrees into the wider society. Economically too, the tribal economy is not different from the regional or national economy. Thus, tribes which answer to the anthropologists’ conception of the ideal type are rarely to be found. Andre Beteille (1969) says that what we find today in India are tribes in transition. But we do regard certain communities as tribal and have included them in the recognized list of tribes, called Scheduled Tribes.