By Gail Omvedt

``TO HOLD that distinctions of caste are really distinctions of race and to treat different castes as though they were so many different races is a gross perversion of the facts. What affinity is there between the Brahman of the Punjab and the Brahman of Madras? What affinity is there between the Untouchable of Bengal and the Untouchable of Madras? The Brahman of the Punjab is racially the same stock as the Chamar of the Panjab, and the Brahman of Madras is the same race as the Pariah of Madras. Caste system does not demarcate racial division.''

With these words, in Annihilation of Caste Ambedkar in 1936 strongly disavowed the ``Aryan theory'' of caste origins and identities. But his argument pointed to another problem: what lay behind the Punjabi and ``Madrasi'' identities he so strongly asserted? The use of the term ``race'' is scientifically inaccurate - and even if it were possible to locate some ``racial'' aspects of Punjabi-Tamil differences, what could be said about Telugu-Tamil, Bengali-Oriya or numerous other major groups?

The primary basis of the identities Ambedkar was pointing to is, of course, language. Linguistic communities have constituted perhaps the most important unit of action and thinking of most Indians beyond the narrow confines of family, caste and village. From the 12th to 17th centuries, partly expressed in bhakti movements and other cultural forms, partly in economic linkages, more tenuously in political ones, social communities based on Tamil-speakers, Telugu-speakers, Gujarati, Marathi and so forth became consolidated. The rhythm and pace were different in different parts but it is arguable that had colonial rule not intervened, India might have developed nation-states within an overall ``subcontinental'' identity similar to those of Europe.

This, of course, did not happen. European imperialism changed the history of the world for both the coloniser and the colonised. Within India, a new framework of action and identity developed: communication based on English, intensified economic linkages and a pan-Indian nationalist organising and struggles that confronted the pan-Indian Raj. ``India'' took on new force both as a weapon against imperialism and an increasingly articulated social reality. Yet, even in the 19th and 20th centuries, both Indian elites and representatives of the Dalit-Bahujan subalterns wrote and acted as much in their linguistic communities, speaking and organising as ``Marathas,'' ``Tamils'' or whatever, as in the context of an overall ``Indian'' nation.

The assertion that such language-based communities have represented the real ``national'' identities of India has perhaps been the major contribution of the Dravidian movement. Discussion of the antiquity and glory of the Tamil language and of a Tamil identity and culture, as distinct from (and oppressed by) a north Indian-based, Sanskritised ``Hindu/Hindi'' tradition, began in the 19th century itself. But it was during the 1930s that this got constructed as Tamil/Dravidian nationalism.

From the late 1930s also, communist conceptualisations of ``nation'' began to develop, marked by the writings of Stalin himself. ``Nations'' were seen as based on language and culture, not on race; as having the right to self-determination. But just as the abstract right to self-determination was never interpreted to mean support for the independence of nationalities in the USSR or China, so it was also played down in India. India was conceptualised as a ``multinational country,'' but one in which the anti-imperialist struggle was central. This provided much scope for communist work in terms of Marathi, Tamil, Malayali or other nationalities, for supporting the linguistic reorganisation of States and the linguistic-cultural rights of the masses. But the communist movement as a whole, as Periyar and others bitterly noted, was ready to (mis) apply its concept of nationalities to support a ``Muslim Pakistan'' but not to support an independent Tamil Nadu.

The basic problem with the usual Marxist and many later sociological formulations is that they left no scope for conceptualising social-cultural communities larger than the ``nationality.'' If Tamil Nadu and Andhra Pradesh and Maharashtra were ``nationalities,'' then what was India?

Concepts formed in the era of the nation-state could legitimise the struggle of nationalities to claim an independent state, but they could only conceive of the relations between states as ``international relations.'' A ``multinational country'' was an extremely abstract, if not empty, formulation. It has resulted in violent criticism by Hindutva forces, accusing the Left of being imitation Western secularist who never really understood the meaning of Indian existence, the glories of a 5000- year-old culture. At the same time, ironically, the leftists, from the Nehruvian socialists to the communists, remained vulnerable to Hindutva ideology itself, accepting the basic themes that ``Hinduism'' was the majority religion of India - and hence the main cultural basis for Indian identity - and that the core of Hinduism lay in the Vedas, in Sanskrit and in the Aryan heritage.

The argument that beyond all the diversities in India, even those of language communities, have lain common traditions, linguistic similarities and widespread cultural linkages could never be adequately answered in a framework that included only ``nations'' and ``nation- states.''

Similarly, it is not the secular Left, but rather the tradition of the Dalit-non-Brahman movements which could effectively challenge Hindutva at the cultural level. And the powerful and community-based polemics of Satyashodhak leaders like Phule or Self-Respecters like Periyar have spoken not only of Tamil and Marathi realities but also of ancient, much broader cultural- historical traditions; their movements have had all-India ramifications and implications, if not an organisational expression.

From a near-21st century perspective, there is something unreal about the ``nationality'' debates. It is clearly a historical fact that powerfully articulated communities based on linguistic- cultural identities have been a salient social reality in India.

The existence of a broader, encompassing ``India'' is also a reality; it is not for nothing that soldiers fighting and dying to recapture the snow-bound peaks in the far north evoke emotions everywhere. The question is that of the best ways to conceptualise and analyse these diverse identities, one regional and language-based and the other pan-Indian.

The concepts of ``nationality,'' ``nation'' and ``nation-state'' have arisen out of the political, economic and cultural developments connected with industrialisation and Enlightenment, and are too impoverished to capture the complexities of world culture today. Though works like Samuel Huntington's Clash of Civilisations may be flawed, it might be argued that the best concept to capture the reality of India is not that of ``nation'' but that of ``civilisation.'' India has been a civilisational entity of great power and age, like Europe encompassing various languages and nationalities but retaining its uniqueness.

The development of such ``nationalities'' into independent nation-states has been a historical fact only in some regions of the world at some periods in history. The political structures appropriate to a 21st century reality are, perhaps, still to emerge. A Tamil community, firmly based in India but maintaining global linkages, providing firm support to the Dravidian parties but never really pushing to the point of a demand for independence, perhaps shows what Kashmir and ``kashmiriyat'' might have become without external interference.

(The writer is Visiting Professor, University of Pune.)