Historiography of Caste: Some Critical Observations and Some Methodological Interventions

Historiography of Caste: Some Critical Observations and Some Methodological Interventions

Paper presented in the fourth Arvind Memorial Seminar

● Abhinav Sinha

In almost all the cases, the entire gamut of writings, research papers and various other kinds of essays on the caste-system, begin with some sentences or phrases that have been so overused as to be rendered into cliché, and since even after getting thoroughly worn out these clichés present the reality to a certain extent, as such I would also use a few similar sentences to begin with.

Caste/Varna is one of the main realities of the Indian social life. No historian, sociologist, anthropologist, or even a political economist, can afford to ignore this reality. Certainly, the influence of casteist mentality over the Indian social psyche goes deep. However while emphasizing upon the caste system and casteist mentality, many a times common people and even the academicians and political activists have this tendency of declaring it to be the only and the single most important aspect of the Indian life and society. While doing so, in essence, they do not actually put the problem of caste and casteist mindset on the agenda of resolution, rather turn it into a meta-reality that cannot be transcended. In fact, what is inherent in such conclusions is an ahistorical view towards the caste system. Somehow caste-system is turned into a system that does not have any beginning or end, a system that is perpetual and eternal. Undoubtedly, this is not the motive of those giving such kind of statements. However, objectively, such utterances lead to such conclusions only. If we do not adopt a historical view on the caste-system, a sense of defeat sets in, which presents the caste-system as invincible. By rejecting all other struggles, “identities” and class-struggles, such an outlook makes the caste system as an integral part of Indian life and people, it converts it into its organic characteristic and thereby it is made as a touchstone for defining Indian psyche. Recently, due to existence of such primitive and totalitarian consciousness (!) some intellectuals have declared the Indian people themselves as a ‘totalitarian community’! According to them, as the project of modernity remains unfinished, there exists an undercurrent of all sorts of totalitarian trends in the society ‘from below’ (that is among the common people), which manifest themselves in the form of casteism, Khap Panchayats, communalism, etc. Therefore, these intellectuals consider that the first priority is to complete the unfinished project of modernity in India, and until this project of modernity is carried to a decisive stage, the task of bringing in a revolutionary change in the whole socio-economic structure should more or less be suspended! They are not the only ones who think this way, there are many more intellectuals expressing such and similar views. These statements are usually governed by a pre-conceived notion; the preconceived notion that it is for capitalism to complete the tasks concerning the project of democracy and modernity and in case it does not do so, it becomes the main task of the progressive forces to complete these tasks, and so long as bourgeois democracy and modernity are not fully realized, proletarian tasks may be suspended. Whereas on one hand it is true that in every struggle of making capitalism more and more democratic, a revolutionary will take part always without fail, however, on the other hand she/he would do it precisely to make the soil more fertile for proletarian class-struggle, she/he does not put on hold the pure and concrete proletarian tasks until this process gets accomplished.

However, there are those intellectuals too, who take a diametrically opposite stand vis-à-vis the stand point of the aforesaid intellectuals. These other intellectuals consider the caste-system or at least the caste-system as we recognize it today, a construct of the colonial state. These academics feel that all the identities including caste were there all along in the Indian society before India was colonized, and they co-existed (harmoniously). The colonial state under its hegemonic design constructed caste, using its ethnographic state apparatus to oppress and crush the Indian masses. Armed with the logic of Western Enlightenment, they wanted to know India better, to rule it in a better way. The type of colonial understanding that emerged about India was the product of the fusion of brahminical  and other hegemonic groups with the ethnographic machinery of the colonial state, and this is what gave birth to the caste system in its contemporary form. There existed the fetish, born out of the Enlightenment mindset, of enumerating and categorizing things, due to which the Indian populace was also classified into “logical” categories, in which caste became the foremost category. The use of caste in the Census gave further impetus to this process.

Both the viewpoints neglect the historicity of the caste system. We will deliberate on both of these viewpoints further onward in this essay.

Our foremost aim in this essay is to humbly put forward a historical understanding of the genesis of the caste-system and the changes it has been undergoing through centuries. It is not our goal to present only a critical account of different trends of the historiography of caste, simply because that can be found in any standard textbook. Neither is our goal to demonstrate that the caste system has always been in flux, because that is also an established fact amongst serious academics. Historians of ancient and medieval India have repeatedly revealed it, that the caste system has undergone significant changes during different historical periods; historians of modern India have also shown how the colonial state as well as the nationalist politics has used the caste identity and in this process how they have brought changes in the hierarchical sequence of these identities and their interrelationships. Various sociologists have brought our attention towards the mobility persisting within the caste system. So if someone in our times claims that she/he has discovered the mobility existing within the caste system, is as if they have claimed to have discovered fire or wheel all over again! It has also been said that in different ages the socio-economic context or milieu is responsible for the changes occuring in the caste system, and it is through articulation with this alone that the changes take place in the internal structure of the caste system.

Thus, it is not our endeavor here to rediscover things that have already been discovered. One of the objectives that we have in this essay is to analyze this articulation more specifically. While arguing that socio-economic factors have been affecting and changing the caste-system, it should also be clarified that, what these socio-economic factors are, and what are the characteristic features of what we are, in general, terming as socio-economic milieu and context. In our opinion, it is the dominant production relations and the dominant mode of production of any period, with which the articulation of the caste system takes place. The second proposition, that we want to put forward in this essay, is that in this mutual interaction, in the final analysis, the aspect of development of the production relations and productive forces, and class struggle plays the main role. That is to say that in the mutual interaction between the caste system and the dominant mode of production prevailing in the society the material factor of the mode of production plays the predominant role. However, this in no way means that the caste system is being determined mechanically at each moment by the changes taking place in the mode of production and production relations. That is why we have clarified at the very outset that it is in the ultimate analysis that these changes play determining role. Then it does not also mean at all that caste and class are essentially one and the same, or that class is caste indeed. Definitely, any such concept is not really talking about any articulation, rather about the complete overlapping of two distinct phenomena, and evidence from the Indian history show that except at the stage of its inception, there has never been any stage in the entire history of caste, when there was any kind of complete  overlapping between caste and class. But subsequently the gap which was produced between the caste system and class division has continued in the history till date, and in different systems of production a correspondence between the two has existed whose form has been changing according to these very different production systems. The third point that we want to make in this essay is that the caste system, during every historical period, has been playing the role of a useful ideology for maintaining the hegemony of different ruling classes.

In this way, one must accept the peculiarity of the caste system, because in the history of other societies, we do not come across such an element of continuity in the ideologies according legitimation to the dominance and hegemony of the ruling classes. Generally, in other societies, with the arrival of a new ruling class, the main aspect in the new ideologies legitimizing the rule of the ruling class has been the aspect of change. But in the history of Indian social formation, despite various fundamental changes in the ideology of caste, the core element that determines and represents it, has remained the same. Of course, while the variables on which this ideology has been applied in different social formations have completely changed, and the execution of this ideology itself has undergone fundamental changes.

Later, we will consider the origin of the caste system, the changes that it has undergone in the historical epochs of ancient and medieval India, as well as the changes in the production relations that were the fundamental reasons behind these changes and then we will also underline some basic changes in the caste system in modern India, especially in the latter half of the colonial period and in the post-independence India, and on that basis would try to substantiate our aforesaid propositions.

• Interpretations of the Origin and Development of Varna/Caste System: Main Problems of Historiography

There is a lot of controversy among the historians regarding the development of the Varna system in its embryonic form during the last phase of the Ṛgvedic period (also known as the Early Vedic Period) and about its consolidation in the Later Vedic Period. There are several opinions prevalent among historians as to what were the main factors behind the emergence of varna system and also about the factors which played the main role in the emergence of caste (jati) later on. We will present the main views in brief, and also our opinion about them. We will also discuss later on the differences between varna and caste (jati). But the analysis of historiography must also be done in a historical manner, because the history of historiography is also indispensable for understanding the appropriate ideas, interpretations and propositions about history. Therefore, we will begin with the colonial period. The discussion about the ideas which were put forward by the native and foreign observers about the varna/caste system in the earlier periods is outside the scope of this paper. Moreover, at present such an analysis is also not needed, because systematic studies on the process of social differentiation of the Indian society broadly began during the colonial period only. In what follows, we will give a brief account of the main studies of the caste system and their interpretations during the colonial period.

• Main Interpretations during the Colonial Period

In a way, it were the colonial administrators and scholars who initiated a systematic study of the social structure of ancient India. The foremost among the initial representative works was “A Brief View of Caste System of North-western Provinces and Awadh” by J.C. Nesfield, which was published in 1855.  Nesfield, on the basis of his studies, proposed that the determination of occupations on the basis of heredity, is the basic foundation, on which the edifice of caste system stands. According to Nesfield, it were the earlier guilds of artisans and craftsmen in ancient India that got metamorphosed into various castes. The hierarchy among them was determined by the oldness or newness of the occupation. The newer an occupation was, the higher would be its position in the hierarchy. After this, several colonial administrators and the western scholars of that era tried to define and interpret the caste system. Among them French Indologist Charles Emilie Marie Senart played a significant role. Senart was the first person to make a distinction between varna and caste. He considered the motion of varna to be more akin to that of class, while caste was an autonomous entity to a certain extent. Later, however, the castes got assimilated into the varnas. Whereas the hierarchy of castes was a real phenomenon for him, he considered the hierarchical organization described in the varnasharma system to be unreal and conceptual. Senart thinks that the brahmins included the various Indo-European lineages in the varnashrama system and had given them a subordinate status, so that their own hegemony remained intact. However, this opinion of Senart was rejected by most of the historians. But the greatest contribution of Senart was that, he made a distinction between the varna and the caste system, which was to a large extent adopted in the later day historiography.

Herbert H. Risley, the colonial administrator who started carrying out the Census in India, gave his own idea on the caste/varna system. According to him, the predominant factor in the evolution of the castes, was the racial factor. He used the nasal index (the length of nose) in order to distinguish between the Aryans and non-Aryans. The caste system got considerably consolidated after Risley started a caste-based Census, besides, it got ossified as well in its contemporary form. The influence of Risley’s racial interpretation continued till much later period, however, in the historiography of the post-independence India, the archaeological and literacy evidences have rejected this racial interpretation decisively.

After Risley the western scholar who left a marked influence on the studies of the caste-system, was the French sociologist Celestin Bougle who also collaborated with Emile Durkheim. The interpretation of the caste system which Bouglé gave, had a major influence on another French sociologist Louis Dumont’s thoughts which we will discuss later. Louise Dumont is considered to be the most authoritative scholar on the caste-system, although his ideas face intense criticism by the later historians and sociologists. For now let’s return to Bouglé’s thoughts. Celestin Bougle opined that caste-system can be identified by its three characteristic manifestations. Firstly, a hereditarily-determined occupation; secondly, hierarchy and thirdly, repulsion, i.e. the alienation of one caste from another. Bougle did not subscribe to the idea that it were the Brahmins who framed the caste-system. On the contrary, the caste-system came into being due to the socio-economic changes, the Brahmins gave it a legitimation only. The idea of purity and pollution was the main factor behind the hierarchy present in the system. Thus, Bougle completely rejected the racial interpretation of the caste-system given by Risley. Bougle’s study on the caste-system can be counted among the most serious and effective studies of his time. Bougle also accepted the idea of Senart that the varna-system is an idealized concept, while caste is a reality.

J. H. Hutton, whose book ‘Caste in India’ came into print in 1946, was the last among the foremost scholars of the caste system before 1947. Hutton considered the existing theories interpreting the caste-system inappropriate, as these did not properly grasp the reality of caste. He enumerated fifteen characteristic features of caste, prominent among them were environmental segregation, magical beliefs, totemism, idea of purity-pollution, the doctrine of Karma, clash of races, the prejudices concerning complexion of skin, and the tendency to exploit by dint of hierarchy. But there were numerous inconsistencies throughout Hutton’s theory. On one hand, he does not put any causal explanation about the emergence and development of the caste-system and on the other hand, for him caste becomes an aggregate of different social groups. Hutton was altogether unsuccessful in comprehending their interrelationships. Dumont, Pocock and all the later sociologists rejected Hutton’s theory. It was a kind of an eclectic theory that made a compilation of the different apparent manifestations of caste.

During the colonial period, some Indian scholars also made sociological studies of the caste-system. But they were somehow similar to the interpretations that we have discussed above. In 1911, S.N. Ketkar published his book ‘History of Caste in India’, in which he gave thoughts similar to those of Celestin Bougle and rejected the racial theory. In 1916, D. Ebetson published his book ‘Punjab Caste‘ which deals with the castes of Punjab. In it he stressed on the role of tribes in the emergence of castes. But the main interpretative frameworks which existed before independence were mentioned above.

Before proceeding it is important to clarify here that Ronald Inden, Nicholas Dirks and many Subaltern Historians such as Partha Chatterjee have put forward the view about the studies of the colonial administrators that they invent or imagine the caste system. It was the colonial ruling class which established the caste system in its ossified form. In order to break Indian people’s resistance, the colonial state also used knowledge and culture apart from economic and political means. According to them, the use of knowledge and culture was even more important than the economic and political factors. As per their view, caste becomes a construct of the colonialists. This entire viewpoint faces two problems. On the one hand, if your agree to it, that the caste system is a construct of the colonialists, a specimen of colonial knowledge, which was prepared to establish their dominance over the Indian people, then you become uncritical towards the pre-colonial India without saying so. Attributing each and every wrong to the Enlightenment rationality and modernity, you declare everything including imperialism, communalism, caste system, etc as colonial constructs and knowingly or unknowingly glorify the pre-British India. For example, Nicholas Dirks admits that caste existed before the arrival of colonialism in India but it was just one among various other social identities. But colonialism constructed caste as the only effective identity and classified the whole Indian population accordingly. Doing so, the Occident successfully degraded the Orient, made it appear as an inferior civilization, and projected the entire Indian population as backward and primitive. Caste was presented as an natural peculiarity of the Indian people and was condemned. But on this whole outlook it can be said that while on the one hand the colonialism did indeed play an important role in ossifying the caste system and it increased the rigidity of caste divide, it is also true that even after the establishment of colonialism there were multiple identities in the Indian society. For example, the linguistic and tribal identities, which were also used as instruments of identity politics.

Secondly, political and economic hegemony was not at all secondary in the project of colonial domination; on the contrary, the efforts that the colonialists made to understand the Indian society, in order to be able to rule it, were made precisely to make the political and economic domination possible and more effective. It was no conspiracy. In fact, the colonialists really believed that to rule India in a more effective manner, it must be understood properly. The process already began with William Jones establishing the Asiatic Society in 1784 and it continued thereafter. We may indeed argue that the colonialists tasted both success and failure in this endeavour of theirs, and they were not fully successful in understating India “in the proper way”! But to term their failure as a conscious conspiracy and a construct is to forcibly impose anti-modernity and anti-Enlightenment ideas of postmodernist, postcolonial theory and Orientalism on the Indian history. Susan Bayly, in her book ‘Caste and Politics in Eighteenth Century India’, has criticized this line of thought of Nicholas Dirks from her point of view (which we can definitely criticize), and has argued that Brahminism and its hegemony were not a product of colonialism, though they were certainly strengthened by it. The brahmins played a significant role in construction of this colonial knowledge, and the collaboration of the colonial state and native elites could be discerned throughout this entire process. The collaboration between the colonial state and the native elites and feudal classes was neither an imagination, nor a construct, but was a stark reality.

So, it is a futile effort to present the caste-related studies of the colonialists as a machination of the rationality of Enlightenment, and to show “Oriental innocence” (Ashish Nandy) as a ‘passive victim’. The celebration of the pre-colonial past by historians of Subaltern Studies and the academics motivated by Orientalism of Edward Said and post-modernism in the name of opposing modernity and Enlightenment, is a flight of imagination and a mental construct of these historians. Sumit Sarkar in his book ‘Beyond Nationalist Frames’ has shown that this cultural critique of colonialism, ultimately aligns itself with the revivalism of the extreme Right, though it superficially terms communalism also as a colonial construct (which is more accurate in this context as compared to caste). This whole logic is a circular and a self-defeating one.

• Post-Independence Sociological Studies: Disregard of History and Essentialization of the Caste System

Suvira Jaisawal while commenting on post-Independence sociological studies in her book Caste: Origin, Function and Dimension of Change, states that these studies, in a way ignore the aspect of history. The whole stress goes into the study of the intricacies of the contemporary nature of caste, but they do not venture to delve into its origin or at least do not do so sincerely. To a great extent, this analysis seems to be correct. Since, while studying the caste system these sociologists ignore its evolution and origin, and see it in its contemporaneity only, they arrive at extremely divergent and incomplete conclusions. Undoubtedly, these studies provide several insights regarding the contemporary caste system. But, while they are unable to use these insights, historians use them.

Among these sociologists, the most renowned was Louis Dumont, whose book Homo Heirarchicus has a Biblical eminence for sociologists studying caste system, irrespective of whether they are in concord or discord with it. One of the reasons is that, Dumont’s interpretation is chiseled with great sophistry. No sharp contradiction is apparent in it. Different concepts have been made to fit in a precisely sculpted structure. As the name of the book suggests, it is about those people or communities, who do not follow the principle of equality. According to Dumont, the Occidental man has faith on the principle of equality by virtue of his individualism (Homo equalis or Homo economicus). But every society needs hierarchy. Dumont says, the moment you imbibe a value, you are in effect accepting a hierarchy. The greatest peculiarity of the Hindu society lies in that, its hierarchy is harmonious. This hierarchy, namely the caste system, has nothing to do with material and economic factors. The element that determines the caste system and even builds it up, is the ritualistic hierarchy. This ritualistic doctrine is the basic structure (as Levi Strauss means it) that is determining the reality here. Brahminical ritualistic ideology constructs the social reality in the Hindu society. The most fundamental element of this ideology is to build up an entire social hierarchy based on the logic of purity and pollution with the Brahmin at its apex, and the untouchables, at its bottom. Every caste is defined on the basis of its relationship with other castes, and consequently we get a complete structure of castes organized in a hierarchical manner. Dumont has answer also for the question about the origin of the idea of purity and pollution! He contends that this idea is that structure of fundamental values that builds reality, and it is pre-given. Such a set of values exists in every society. Hierarchy is an essential value, and every society needs it. In this sense, the caste system endows the Hindu society with such a hierarchical structure, which is uncompetitive, harmonious, unchangeable, and makes the society stable. Dumont repeatedly places these peculiarities vis-à-vis the Western society, and in a way subtly asks the question, what have the values of equality and individualism given to the Occidental Civilization? Thus, Dumont, in the words of Gerald Berreman, adopts a brahminical view of caste. It is in a way equivalent to justifying the caste-system. Dumont fails to explain the fact, in any way, though he is obliged to admit it, that with the development of industries and capitalism, caste restrictions on occupation and commensal prejudices have been weakening steadily, as demonstrated by G S Ghurye and E K Gough; the only characteristic feature that persists is endogamy. Dumont thinks that these political, social and economic changes have no bearing on the caste system, rather they get absorbed within the caste system. Dumont does not draw any conclusion from these changes. For him the Hindu society, along with its caste system and hierarchy, becomes an ideal, unchanging society. Obviously, we need not spend many words to refute Dumont’s thesis.

Javeed Alam has remarked somewhere rightly indeed, that most of such sociological ideologies are in reality designed to enter into a shadow-boxing with Marxism and the materialist dialectical historical methodology. In fact, Dumont does criticize Marx for predicting the elimination of caste with the arrival and development of railways and large-scale industries. Actually, Marx was talking about the disintegration of caste-based hereditary division of labour, and in this aspect Marx’s prediction has been proved more or less correct. Dumont thinks that since the Indian social structure is unchangeable, eternal, hence its history cannot be written. This point of view aligns markedly close with the old colonial viewpoint, to which Edward John Thomson, father of E P Thomson, has given a remarkably wonderful expression. Thomson said, India is a country singularly bereft of history. On this idea of Dumont, Irfan Habib has aptly written:

            “If such is to be the history of India, to fit a contemporary western sociologist’s image of the caste system, is it not more likely that there is something wrong with this image rather than with Indian history? It may, in fact, well be that there is a good historical explanation for Dumont’s excessively narrow view of caste. During the last hundred years and more, the hereditary division of labour has been greatly shaken, if not shattered. As a result, this aspect has increasingly receded into the background within the surviving domain of caste. The purely religious and personal aspects have, however, been less affected. (One can see that this is by no means specific to India: religious ideology survives long after the society for which the particular religion has served as a rationalization has disappeared)” (Irfan Habib,1995.Caste in Indian History, ‘Essays in Indian History’, Page 164, Tulika Books, New Delhi)

A whole lot of sociologists have studied the caste system after Dumont. They have drawn attention towards the use of casteist consciousness by the affluent elite classes born in every caste in post-Independence India, and have shown the way the caste equations are being used in electoral politics. Two aspects can be discerned as we go through these studies, that remain today as the characteristic features of caste politics. One is that, in every caste, dalits also included, there has emerged an affluent class which, in order to garner votes or to have usufruct of the resources, or to establish its monopoly over the access to them, invokes the caste-consciousness of the plebeians of their own caste. This aspect can be prominently seen in the politics of BSP, SP, RJD and parties of their ilk, and all the electoral candidates, even of the BJP and the Congress, who use their caste identity at the grass-root level, and frame caste-based equations. Eventually, when the election results are out, the different caste elites enter into mutual bargaining, deals, and negotiations, and on the basis of these exchanges, the ruling alliance is put together. In other words, in its mutual rivalry, the ruling class makes use of the caste equation. The other aspect which is the more significant, is that the electoral parties which claim to represent all the castes, the dalit caste included, are the electoral parties of elites of these castes, and these elite classes of the different castes join hands to oppress the masses and to keep the people divided and foment caste consciousness among them. Notwithstanding these important insights, the greatest shortcoming of these sociological studies is that they do not pay serious attention to the history of caste system. Leaving aside some cursory mention, the understanding of these people about the emergence of caste system and its subsequent development is inappropriate. This is the reason why they cannot give any explanation of the changes that take place in the phenomenon of the caste. Their total attention is focused on the study of the dynamics of the contemporary phenomenon of caste. But the irony is that, a balanced understanding even about this dynamics can be reached only when, one has a clear view on the emergence and development of the caste system.

It is the lack of a historical vision that does not allow the whole lot of sociologists to comprehend the dynamics of the caste-system and often the sociologists see the caste system as a static system, which consequently becomes the identity of the Hindu/Indian society, and its fundamental characteristic or logic. Something which has always been there and will be there forever. Many a times, such theorization goes to the extent of justifying the caste system, as is done by P. A. Sorokin. Sorokin has made the peristence of the caste system through ages, that is, its sustainability, the basis for its justification. His logic goes like this, the reason that the caste system still exists is that, it gives the people of the society a satisfactory hierarchy. Here also one can notice the inherent preconceived notion, that the caste system is an unchanging phenomenon that has been providing the Hindu society with a semblance of stability. In a similar vein, Nirmal Bose has also considered the the caste system to be an unchanging factor which provides stability. He thinks that, in the society the caste system saves people from getting uprooted, since it ensures them, their right over their occupations. Monopoly over occupation gives people a sense of security.

In order to look for the reasons behind the trend that is there in these sociological studies, of viewing the caste system as a static one, we cannot refer to this entirely diverse lot of sociologists. We must understand that this lacuna is actually the lacuna of the very academic discipline of sociology. The discipline of sociology was designed precisely to disprove the dialectical and historical materialistic outlook of Marxism. For instance, the sociological method of viewing the hierarchy as an indispensible necessity of every society, gives a legitimacy to the caste system also, and puts a question-mark on the goal of an egalitarian society itself, as propounded by Marxism. Afterwards, on the face of the riposte made by Marxism, the branch of sociology has also undergone through a number of changes and there have appeared a number of Marxist sociologists, who placed even Marx along with Weber and Durkheim as the founding father of the discipline of sociology. The basic prejudice or preconceived notion of sociology is a positivist prejudice, whose roots can be seen in the ideas of Auguste Comte. In this essay we cannot write a critique of the entire discipline of sociology, but this much is clear that the discrepancy present in the sociological studies of the caste system has its roots in the absence, rather a kind of conscious negation, of a historical outlook in this entire discipline. As a result, studies made, divorcing contemporaneity completely from history, gives us some valuable fragmentary insights, but fail to provide us with any consistent approach or methodology of explaining the caste system.

Other than these sociological interpretations, the study on caste system done by G. S. Ghurye also made a significant contribution. On the whole, Ghurye put stress on the racial origin of the caste-system. Besides him, there were some other sociologists also, such as N. K. Dutt, D. N. Majumdar and R. P. Chandra who supported this idea of racial origin. These people are of the opinion that, the Aryans invaded the Indian subcontinent at its north-western area, and subjugated the people of Dravidian origin. To keep these subjugated people under a structural subordination, the Brahmins constructed the theory of purity/pollution. With this theory at the base, the caste hierarchy was designed according to relative purity/pollution in comparison to the Brahmins, and thus came the caste system into being. But as Suvira Jaiswal has argued, there are no evidence to substantiate this theory. Sociologists have also debated a lot over the difference between caste and varna. Max Weber saw varna as a phenomenon akin to the European ‘estate’. Trautman declared caste to be a real phenomenon while varna was a phenomenon similar to the ‘estate’. There are sociologists who are of the opinion that varna system gives a bookish description of the caste system, which provides an idealized categorization. Castes are a real phenomenon, which, as they were born, got successively ensconced within these varnas. That is why we can witness different localized patterns of co-option of castes into the varnas, while the latter have a pan-Indian character. But one thing is common everywhere. The scale, or definition of purity of every caste or the unit of its measurement is the highest purity of the Brahmins. Which means that all the castes get their places within the caste system (hierarchy) depending on their relative distance from the Brahmins. The difference determined between caste and varna by the sociologists is also only and only the difference decided on the basis of the contemporary caste system. Nobody disagrees with the fact that these notions are different. But the way the sociologists, without developing any understanding of the evolution and development of these categories, have presented the varna system as ‘book view of caste’ and the jatis as ‘field view of caste’ is totally ahistoric. Ancient history reveals it, that at those beginning phases, jati and varna were used synonymously. But when the word varna vyavastha was used, the implication was that the classic, idealized system of the four varnas was being discussed, which was mentioned for the first time in the ‘Purushasukta’ of the later part of Ṛgveda, according to which the Vedic society was divided into four varnas – Brahman, Rajanya, Vis and Śudra. Using the word jati meant that we were talking of those tribal groups which were assimilated into the Vedic society, and depending on different influencing factors, were considered as a part of one or the other of the four varnas. But so long castes were yet to emerge, the words Jati and Varna were used synonymously. We witness use of the word jati for the first time in the period prior to circa 200 BC. Suvira Jaiswal considers that it was the period when the large-scale proliferation of castes was yet to be a wide-spread phenomenon, and the use of the word jati in the literature of the period immediately after the Vedic period, especially during the time of Buddha, was not itself a sign of a full-fledged caste system coming into existence. In effect, the word jati was still used to mean varna only. Historians are divided in their opinions about how the transition from varna towards jati took place, and to have a fair understanding, we must observe briefly the historiography of ancient India.

• Origin and Development of the Caste System: Problems of Historiography

Suvira Jaiswal tells that both the words varna and jati are used in ‘Ashtadhyayi’ of Panini. Panini belonged to the period around circa 200 BC. In ‘Bṛhatsamhita’ of Varnamihira also jati and varna were used synonymously. But in ‘Yajnyavalkyasmriti’ there is one instance where jati and varna come with different connotations, but, several times they are used synonymously also. Clearly, till 200 BC the development of the system of castes did not reach a decisive stage.

Among the historians of ancient India, both Iravati Karve and Romila Thapar (notwithstanding having different opinions on numerous occasions) agree that the origin of caste system should actually be explored in the Harappan civilization before the arrival of the Aryans. Romila Thapar is of the opinion that, some basic elements of the caste system such as groups divided on the basis of heredity which controlled the institution of marriage, the idea of purity/pollution, and the elements of the jajmani system, were all incipient in the Harappan civilization itself. Romila Thapar concludes that the Great Bath of Mohen Jo-daro was actually meant for some ritual connected with purity/pollution. But this seems to be more like a flight of imagination based on a blend of fractured factums and evidence. Aryans are exonerated from the crime of introducing the caste system and varna system, and the caste-system becomes a natural endowment of the Indian subcontinent. That is, there is something (which is) completely Indian in the caste system. This becomes a prominent feature of the Indian way of life and system of ideas. Similar notions were forwarded earlier also. It is certainly not the motive of Romila Thapar to make an Indianized essentialization of the caste system, but on the objective plane, her thesis supports this conclusion. And the most significant thing is that, it has no evidence in its support, rather there are several contra-evidences.

If we make a perusal of the emergence of the caste system in the history of ancient India, we observe that it is inseparably linked with the emergence of classes, state, and patriarchy in the society. A consistent understanding of this history is essential because without it, the historicity of caste and the mindset connected with the caste system cannot be understood, and to us also the casteist mindset and the caste system will become a natural trait of the Indian people. A dialectical and historical materialistic interpretation of ancient Indian history, can be considered to begin with Damodar Dharmanand Kosambi. According to Kosambi one can find evidence of the beginning of the varna system at the end of the Ṛgvedic period. But, the system of castes does not grow simultaneously with it. When the Vedic civilization spread eastward from the north-western frontiers, caste emerged along with the assimilation of new tribes into the Vedic society. We will present our views on this interpretation in detail in the coming pages. Morton Klass also studies the origin of the caste system. Klass comes to the conclusion that castes originated right in the prehistoric era with the beginning of agriculture. The tribes having access to cultivable lands turned into high castes, whereas the tribes coming into this region from other areas became the lower castes. These castes voluntarily accepted their subordinate status vis-a-vis the other castes that already had the access to arable land and practised agriculture. But we can find no evidence in history to support this theory. The notion working behind this theory is that the caste system came into being with the beginning of surplus production i.e. with the beginning of agriculture. But surplus production cannot on its own create the caste system unless a Brahminical ideology also is present there. This Brahminical ideology was the ideological apparatus to institutionalize class division in the form of the system of varnas. This is the reason why caste system emerged in the north-east long after the stage of surplus production was reached and classes came into existence, when the Brahminical ideology gave this division of classes, its casteist form. Moreover, Morton Klass‘s theory of the transition from clans/tribes to castes can explain the emergence of those castes only who are engaged in production. In his schema, the origin of the brahmin caste  itself, remains unexplained. Besides, Morton Klass is also incorrect when he opines that caste system emerged almost simultaneously in the entire Indian subcontinent. Historical evidence now reveal it clearly that caste system spread in the southern and eastern India afterwards, and it acquired an form vastly different from the caste system of the north and north-western India.

Besides this, there is also a theory of the Dravidian origin of the evolution of castes, according to which, the Dravidian civilization had some elements which gave birth to the caste system. One such theory puts stress on the concept of tinai, in ancient south Indian Sangam literature. According to this, tinai is a word used to connote a region. Five tinais are mentioned which were occupied by different communities. The socio-economic conditions in these tinais were altogether different. In some places agrarian society was coming into existence, while in others, elements of the pastoral society still existed. Fishing was the mainstay of the economy in the tinais of the coastal areas. When fusion started between these societies, then people of tinais with advanced production relations started to construct higher castes. But this theory cannot properly explain the origins of the caste system. This is due to the fact that the tinais mention five different geographical-ecological regions, and the communities inhabiting these areas did not belong to a society divided into classes. The society whose characteristic feature is the caste system, is in reality a unified society with definite property relations.

Another reason that gave birth to the theory of Dravidian origin, is the theory of untouchability of the sacred communities in the Dravidian civilization. According to this theory, a holy man is actually a carrier of all sorts of impurities, and these deadly impurities resident in him are contagious. But here the relation of the pure and the polluted is just opposite to the one found in the caste system. Historical evidence have now demonstrated that the doctrine of purity/pollution can originate in many nomadic and pastoral societies, where often, according to a sociologist named Bruce Lincoln, rise priest and warrior classes. In this era of magical world outlook one class performs its role by sacrificing animals for enhancing the cattle wealth through rituals, while the other class performs the role of leadership in the process of capturing the cattle wealth of other tribes by attacking them. Other remaining classes formed the common plebeian masses. The first class forms the class of priests, and often constructs the doctrines of purity/pollution. But this class cannot by itself become the cause of the origin of the caste system. Thus, the theory of Dravidian origin also is a scheme only for which no historical evidence exists.

Kosambi’s theories on the emergence of the varna system are significant. Many ideas of his theory were later found to be inappropriate. However, his methodology presents a consistent interpretation of the existing evidence and makes on its basis, extremely logical simulations about the unknown aspects. According to Kosambi, an Aryan community had already settled in the Indian subcontinent before the coming of the Vedic Aryans. Chances are there that this group got assimilated with the remaining elements of the Harappan Civilization. When the Vedic Aryans came, the people of this group clashed with them. In Ṛgveda these very people have been called dasyu or däsa. A few positive comments have also been made about some powerful chiefs of these tribes/clans of dasyus or däsas. The term asura has been used for them. But it seems that, at that time the word asura was used to mean a deity. Because we see that it has also been used for Indra, who was the chief this-wordly (ih-laukik) deity of the Vedic Aryans. For the deities of the other-world (parlok) the word deva was used. It has been said about these dasyu/däsas that their complexion (varna) was nigrescent or dark which shows that they had undergone intermingling with the residual elements of the Harappan civilization, and this is quite possible that they mixed with the other aboriginal people as well. Many references of the clashes of the dasyus/ däsas with the Aryans are found in the Ṛgveda. Eventually, these Vedic Aryans vanquished the däsas. The meanings of the words ‘asura’ and ‘däsa’ changed with the defeat of the däsas. Since the word ‘asura’ was used for the däsa chieftains, so later the word ‘sura’ came to be used for the Aryan chieftains/gods. When the däsas/dasyus were completely brought under the subjugation of the Vedic Aryans, the modern meaning of the word ‘däsa’ i.e. a slave, came in use. These subjugated dasyus/ däsas got transformed into the Śudra caste. According to D. D. Kosambi, new production relations came into existence along with the Śudra caste coming into being and with the Vedic Civilization reaching the Gangetic plains. With the expansion of agriculture and beginning of the use of iron, the stage of surplus production was attained. During the introduction of this stage, new tribes were getting assimilated in the society of Vedic Aryans. According to Kosambi, with this, castes based on principle of endogamy came into being. Romila Thapar opines the same but in a slightly different manner. According to her, the vanquished tribes became the lower castes, whereas the victors became the upper castes.

According to Kosambi, the reference of the system of four varnas that we find in the ‘Purushasukta’ of the tenth mandala of Ṛgveda at almost the close of the earlier Vedic age, was in reality manifesting class-division only. According to him, the varna system in that primitive stage of production was indeed a symptom of class division, and what we are calling by the name varna in this stage, was actually class and nothing else. There is ample amount of historical evidence in support of this argument of Kosambi. For instance, the system of four varnas that is described in ‘Purushasukta’, does not yet mention endogamy and hereditary division of labour. That is, none of the basic characteristic features by which we identify the caste system today, were in existence yet. Ramsharan Sharma has also confirmed it.

Kosambi has considered the birth of slave labour also, as one of the origins of the emergence of class division in the later half of the period of Vedic society. Definitely, in the Indian subcontinent slave-labour has never been used to that scale in productive activities, as the scale on which it was used in the ancient Greek or Roman civilizations. But the logic put forward by Kosambi in this context, and which seems to be correct, is that the emergence of slave labour, in a primitive tribal or a nomadic pastoral society has a significance in itself, and it makes no difference, that to what extent it was used in production activities. The moot point is that, whatever be the extent to which slave labour is put into use, it is a symptom of disintegration of communal relations. The coming into existence of the Śudra varna in the later half of the Ṛgvedic period and especially in the post-Vedic period, their use as slaves, the collusion of Brahmins and the Kṣatriyas to oppress and exploit the Vaiśyas to a certain extent, and to oppress and exploit the śudras to the hilt, were the signs that class society had arrived. But we must present sufficient arguments to show that, at this primitive stage of production, there was basically and mainly, an overlapping present between varna and class.

This aspect was elucidated by the excellent historian of ancient India Ramsharan Sharma. Sharma makes it clear that a stratification/categorization was in existence, there in the Ṛgvedic age, but that could not be given the name ‘class’ yet. Slave labour was also present in the form of female slave labour only, who were not only engaged in domestic labour, but many a times they were used to replenish the depleted number of women in the victor tribes; i.e they got assimilated into the victorious tribe/caste. But neither was there surplus large enough yet, that these categories could transform into classes, nor did they acquire the traits of varna or caste,  such as endogamy, hereditary occupation (division of labour), and rigid hierarchy. In the form of slave labour, there were Śudras, who were none other than the subordinated dasyus/däsas. Their children sired by the higher varnas used to be absorbed in the Vedic society without any discrimination. The social categorization between the four varnas that came into being in the later half of the Ṛgvedic period, was not yet a varna/caste system as such, rather it was a manifestation of the embryonic class-division in the society. Ramsharan Sharma called it ‘small scale non-monetary peasant  society’, in which inequity in distribution had already started, but powerful elements of tribal society (nomadic pastoral society) were still present. Around Circa 1000 BC to 700 BC, with the beginning of use of iron, the Gangetic plain was cleared off forests, use of iron plough was started, that enhanced productivity, and the amount of surplus production crossed the threshold, creating conditions conducive for the formation of class and state. Another historian B. N. S. Yadav, submitted some new evidence in support of Ramsharan Sharma’s interpretation. He showed that this process of consolidation of class-society continued during the period extending from the 7th century BC to the 1st century AD. In this very period, new tribes got assimilated in the varna-based society and new castes came into being as a result of it. In this period another phenomenon also appeared on the scene. The hold of the Kṣatriyas and the Brahmins on the Śudras got weakened to a certain degree and the latter gradually started getting transformed into a dependent agrarian population in which previously, the vaiśyas were the majority. The vaiśyas who still pursued agricultural activities, were on the decline on the ritualistic plane and many of them started descending into the śudra varna. The rest of them went on to take trade as their occupation. Thus, there was a fall in the population of the vaiśyas and they made trade their principal occupation.

What was the fundamental cause behind this change that appeared in the varna/caste system? The principal reason behind these changes was the emergence of a new mode of production and new production relations. We have evidence of land grants from the first century AD. Brahmans were the principal beneficiaries of these land grants. However, they were not the exclusive beneficiaries and sometimes it were the kṣatriyas while in the other cases it could be the vaiśyas as well. The brahmin-kṣatriya alliance had the main sway in the state authority. During the Maurya period, this feature was clearly visible in the state power. The main function of Brahmins was still priestly activities but with the emergence of feudalism in its embryonic form and with Brahmins becoming the recipients of land grants, changes appeared in their character. They were now also emerging as landlords. The character of the kṣatriya varna was already that of warriors and landlords. The brahmin-kṣatriya alliance still assumed the role of ruling class. However, during seven hundred years from the fourth century to eleventh century AD to mature, when feudal production relations kept developing, there appeared fundamental changes in the roles of the four varnas. We would discuss more about it afterwards.

Suvira Jaiswal agrees with the description of the feudal mode of production as given by Ramsharan Sharma and B.N.S Yadav. According to her, the objection raised by Harbans Mukhia, that the then prevalent social formation could not be called feudal because serfdom did not have any significant presence, as inconsequential. Indian feudalism did not need serfs as a separate class. The subordinate status of the śudras and the untouchable castes fulfilled this need. Many times, the śudras became sharecroppers. Actually the partial overlapping that can still be seen to this day between the landless labourers and the lower castes has its roots in the times of feudalism itself. Jaiswal argues that ignoring the class functions of the caste system would be tantamount to ignoring its economic and political aspects. And if these fundamental economic and political aspects of the caste system are neglected, then nothing remains of it other than endogamy and hereditary division of labour. In such a case, caste system would become an ahistoric part of the Indian life, history and society, without any beginning or end, and hence also a natural element of Indian life, history and society. It is known to us that many ideologues and organizations who talk about dalit liberation, say similar things on this question and unwittingly naturalize the caste system. This leads towards the idealization and, in a way, legitimization of the caste system. According to Suvira Jaisawal, in the context of Indian society before the arrival of colonialism, we can find numerous evidence showing that whenever there was a relation of correspondence between the caste system and class division, the caste hierarchy got reinforced and became more rigid; on the other hand, wherever and whenever the ritualistic hierarchy present among the castes stood in opposition to the dynamics of class division, a process of fusion and fission was engendered within the caste system, which brought in significant changes in the caste hierarchy in a gradual process.

Suvira Jaisawal has criticized Kosambi, Ramsharan Sharma and Irfan Habib for making an external factor, viz, assimilation of new tribes into the folds of the Vedic society, responsible for the emergence of castes within the varna system. Whereas it is true on the one hand that the eastward expansion of the Vedic Civilization and the assimilation of new tribes within it gave birth to the castes, concurrently it is also true that if the elements of caste division (namely, the hereditary division of labour and varna division on the basis of the elements of endogamy ) did not already exist within the varna system  then the mere induction of new tribes will not by themselves give rise to new castes. According to Suvira Jaisawal, this belief that the pre-Vedic tribes used to follow endogamy while there was no such culture among the Vedic Aryans is false. She has given evidence to the contrary that with the emergence of patriarchy, the tradition of clan endogamy was on the way out, and with the imposition of subordinate status on women, the seeds of caste endogamy were sown. Moreover, we can find evidence of existence of such pre-Vedic tribes, where the tradition of endogamy was still absent. Therefore, it cannot be argued that castes based on the practice of endogamy emerged only with the assimilation of new tribes within the fold of the Vedic society. On the other hand, it was in the Vedic society along with the origin of the caste of śudras only that the process of treating certain forms of manual labour as inferior had begun. In such a scenario, when the tribes having expertise in the new kinds of productive labour were included in the Vedic society, they were included in the form of different castes and at the same time the hereditary division of labour also began. This was the reason why the entire tribe did not get transformed into a single caste. Rather what happened was that the upper priest class got assimilated with the Brahmins and other classes with the other varnas of the Vedic society. A lot of people from several tribes also got assimilated with the kṣatriya varna. In a nutshell, it can be said that the ground for castes based on endogamy and hereditary division of labour had already been existing in the Vedic society and that is why the assimilation of the new tribes into the Vedic society could become as a factor in the origin of castes. The assimilation of other tribes into the vedic varna system continued right up to the later half of the middle ages. This could not be in itself the main force behind the creation of castes. In this context the position taken by Subira Jaisawal appears to be more balanced. In all these developments, it was the internal process of class division within the Vedic society which was mainly responsible. The inclusion of the external tribes into the Vedic varna system was continued till the latter half of medieval era. It on its own could not have become the reason for the emergence of caste. Suvira Jaiswal’s stand on this subject appears to be more balanced.

If we look into the history of the period from the end of the Vedic period to the beginning of the period of ancient republics, one thing clearly emerges out. Origin of the varna system and the coming of castes into existence was an extensive and complex historical process. Several aspects of that period still remain untouched and do not have enough evidence related to them. But this much is certain that the varna system was constantly dynamic right from its inception. Even the form which the caste system assumed after the emergence of castes was also dynamic. The prime mover behind their dynamism was the changes that occurred in the mode of production and the production-relations. The varna-class overlap is clearly visible at the time of emergence of the class society. However, this overlapping could not last very long and it was bound to be ultimately transformed into a relation of correspondence.

The reason behind this is that the varna system at the moment of its inception was the ideological legitimization of the existing class relations, but it was an ideological legitimization which was peculiar in itself. In all the societies of the world, with the emergence of class rule, there evolved ideologies to legitimize the rule of the ruling class. But in India this ideology had not only taken a religious form, but got ossified into a ritualistic form. Obviously, when a ruling class under its rule uses its ideology to ossify the prevalent structure of class divisions in the society ritualistically then that ideological legitimization fails to keep itself in conformity with the motion of development of production relations and mode of production. In such a situation a gap will arise in the old ideological legitimization or the ideologically ossified form of previous class divisions and the new class divisions. Surely, this gap does not mean that there will be co-relation or correspondence between the class divisions prevailing in the society and its ideological ritualistic legitimization. What it means is that whenever a radical change in the class divisions takes place, there will be tremors in the old ritualistic structure and it will need some corresponding adjustments.

Such changes abound in the entire history of caste system and caste ideology. And these changes have taken place spatially as well as temporally. That is to say in the same era the caste hierarchies have been different in different regions. For instance, by the time the Vedic Civilization reached the societies of southern and the eastern India, the agrarian economy was already considerably developed and the status of the agrarian castes within the caste system too underwent changes. Consequently, we do not find kṣatriya and vaiśya varnas in these regions. We will discuss these later. But at present it is sufficient to point it out that one can find radical changes and diversities in the caste-system, spatially as well temporally. There is just one feature in the varna/caste-system that persists. What is it? It is that the ritualistic caste divisions which take place on the basis of class structure of any region depends on the brahamanical ideology, which in turn based on the doctrine of purity/pollution. However, the consequent caste hierarchy which arises out of it, varies in different regions based on the prevailing production relations and the production system. This becomes still clearer if we look at the changes which have taken place in the entire varna/caste system and the status of different varnas/castes along with the changes in the production relations.

• Changes in the Status of Different varnas/castes with the Changes in the Mode of Production and Production Relations

Suvira Jaisawal has drawn our attention towards the changes in the status of the Brahmin varna/caste in the caste system. It could be clearly seen that the changes taking place in the production relations and class structure were the main cause behind these changes as well. Romila Thapar has shown that in a nomadic pastoral society the main source of income of the Brahmins was in form of gifts presented to them. This source was declared as the only permitted source of income even in the contemporary religious samhitas (codes). However, with the transition to agriculture, land grants replaced gifts of things. This practice of land grants transformed the Brahmins, who were earlier priests only, into landlords also. This brought in a significant change in the status of Brahmins. When we move onwards from the Vedic period to the history of the janapadas and then to the Mauryan period, we see Brahmins assuming the positions of the rulers also. Many such states developed whose rulers happened to be Brahmin. Now the functions of kṣatriyas, who were earlier believed to be inferior to Brahmins, were no longer treated as prohibited or lowly for Brahmins. On the contrary the status of such Brahmins was elevated in the caste hierarchy. What is surprising is that by the early medieval era those Brahmins began to be treated as inferior who used to take alms or do priestly work, and the status of those Brahmins got rose in the rank who had become rulers-administrators or landlords. Why did these changes take place? Clearly, the transition from a pre-feudal social formation to a feudal social formation, brought in fundamental changes in the status of the Brahmins. Besides, a lot of new castes came into being within the Brahmin caste. The emergence of the caste of brahm-kṣatriya, as mentioned by Suvira Jaisawal, can have three probable sources; first, matrimonial relations between the Brahmins and kṣatriyas; second, the function of kṣatriyas viz., governance-administration, being adopted by Brahmins, and third, the prior existence of the root of such a caste (brahm-kṣatriya) in the form of the Puru clan.

The way iin which the status of brahmins in the caste hierarchy and their functions as determined by caste ideology underwent changes, we can observe similar changes among the kṣatriyas as wells. New castes emerged from within the kṣatriyas which had diverse sources. For instance, we have now sufficient historical sources regarding the formation of Rajput caste which show that this caste did not possess the status of kṣatriyas varna from the beginning. This caste was formed by the fusion between the Indianized foreign elements that conquered other tribes and established their rule and the members coming from other varnas and some native tribes. This was a warrior landowning caste formed by the amalgamation of the elements coming from different sources. This caste established matrimonial alliances with the kṣatriyas and other upper castes as well which elevated their ritualistic status. In this entire process the people of this community adopted the name of rajputra which subsequently turned into Rajput.

In south India there existed no such warrior tribes. There the emerging land-owning peasant castes performed the functions of the warrior tribes. Consequently, no kṣatriya varna appeared there. When the process of state formation among the agriculture-based tribes reached a decisive stage, big regional states came into being. The kings of these states came from the peasnat communities only. And then the Brahmins from north India were in a way imported into these states. These Brahmin elements also got fused with the priestly elements within those tribes and they formed the Brahmin castes in south India. The ruling peasant castes were assimilated in the varna system as śudras by these Brahmins. However, the status of śudras here was not the same as that in north and north-western India. They were included in the śudras varna as castes because by then, śudras had become the main peasant caste in the core regions where caste system had emerged. The status of the śudras in south India was much better because they were not only an agrarian caste, but they were the ruling class as well. Thus, for instance, one such caste, vellala in south India has been referred to as the patrons/protectors of Brahmins. Since, Brahmins had the ritualistic “power”, therefore, no other caste could perform their functions. But the character of the conventional power of the kṣatriya was not other-worldly, but this-worldly, and hence the tasks which were traditionally reserved for them could be carried out by any other caste. In south India, this task was carried out by Vellala caste which enjoyed quite a high status in the south Indian caste hierarchy. Here those who were dependent, exploited and having slave-like status were termed as asat śudras. It was easier for brahmins to put forth such a proposition because long ago a distinction had been made between ‘hīna’ and ‘ahīna’ śudras in Brahman Samhitas. There were some śudras whose pollution could not be rectified, whereas there were others whose pollution was not contagious and could be remedied. It was on this basis that the Vellalas were termed as Sat śudras whose position was quite high up in the caste system while the adi-dravid castes were termed as Asat śudras whose position became similar to the serfs and extremely poor artisan castes, much like that of the śudras in the Vedic period in north and north-western India.

In Eastern India, too, such peasant castes came into being that reached the position of the ruling class. There too, no separate vaiśya & kṣatriya varnas came into being. Therefore, in Eastern and Southern India, we come across only two varnas–brahmin and śudra. In the coming centuries new castes were born within these very varnas—sometimes with the assimilation of new tribes and at other times, owing to the process of disintegration and fusion among the already existing castes. In this way, vaidyas and kayashtas came into existence in Bengal.

            Ramsharan Sharma has shown how cultivation, which was originally an occupation of the vaiśyas, became the principal occupation of the śudras. According to him, as the feudal practice of land grants started, the migration of brahmins to new areas led to the assimilation of new tribes into the varna system. These new tribes were assimilated in the śudra varna and agriculture became their main occupation. However, according to Suvira Jaiswal, with the advent of feudal mode of production, manual agricultural labour gradually became an ignoble occupation. And with this, the new peasant castes were inducted into the Vedic society as śudras and not as vaiśyas. Besides, those vaiśyas also who remained attached with agricultural occupation gradually turned into śudras. Those vaiśyas, who took to trading on the basis of accumulated agricultural surplus, succeeded in retaining their vaiśya status. Thus, with the emergence of the feudal mode of production, and the concurrent induction of new tribes into the Vedic society, the pattern of traditionally-determined occupation for vaiśyas and śudras changed. Earlier the vaiśyas were mainly engaged in farming, and a section of the poor śudras too were attached to the land as dependent cultivators. Both Ramsharan Sharma and Suvira Jaisawal have shown, how the connotation of the word ‘Gṛhapati’ was originally used to mean the chief of a tribal clan, but went through a gradual change and came to be understood as the head of a peasant family in the era of Buddha. By following the gradual evolution of this term, we can get a complete description of how the division of labour, between the vaiśyas and śudras (agriculture and trade) evolved.

Suvira Jaiswal also describes how the four varnas appeared in Maharashtra and Gujarat and how the new tribes got assimilated in all the four varnas. The reason behind it was that the spread of brahminical society, culture, and ideology had already begun before the rise of feudalism in those regions, i.e. between 500 BC & 200 AD. The change that came into the status of different varna/jatis brought in significant changes in the entire caste hierarchy as well. There are sufficient evidence to substantiate that the changes that took place in mode of production and production relations have time and again exerted pressure to usher change in the varna/caste system from within. A gap between caste and class always remained, but only a blind can claim that there is no clear correspondence between them. There have been times when this gap appears wider, and there have been times when it appears less. At a particular moment in the dialectics of production relations and development of productive forces, untouchability was born. It is imperative to understand that process too.

• Development of Untouchability : The Highest Stage of Development of Relations of Feudal Exploitation

With the emergence of asat śudras in southern and eastern India and with the transformation of the śudras into mainly peasant castes in northern and north-western India, the Untouchables (achūt) came into existence as the most subjugated, most oppressed and exploited section of the society, who later came to known as dalits. We have already mentioned that the relegious codes had made a distinction between the hīna and ahīna śudras long ago. For example, chandal caste was counted as śudra in the varna system, but it was placed in the category of hīna śudras. On the one hand untouchability came into existence among those who were at the lowest rung among the śudras, while on the other hand, when some forms of manual labour were declared to be of extremely inferior kind during the process of the development of feudal production relations, then the element of untouchability was appended to the castiest ideology of purity/pollution. We can see that the idea purity/pollution has been present in the brahminical ideology as a variable. That is why many castes were declared to be untouchables much later. For instance, nowhere in the Vedic sources, occupations connected with leather work, or the caste of tanners and cobblers (charmakar) who did these jobs, were declared lowly or inferior. Just the opposite, it was customary to carry various materials required for the Vedic rituals, only in leather bags. It was in the 8^th and the 9^th century that the charmakaras were declared untouchables.

According to the thoughts of Bhimrao Ambedkar regarding the origin of untouchability, it was a conscious and deliberate act of the brahmins to make some castes untouchable; especially those who had been involved in resistance, still indulged in beef-eating and also adopted the Buddhist religion. But Vivekananda Jha has refuted this line of argument with evidence. Jha has demonstrated that the rise of untouchability had no relation with beef-eating and adopting the Buddhist religion. It was closely connected with the development of the feudal mode of production, which in order to make the exploitation and oppression of the exploited and the oppressed castes structural, gave this exploitation and oppression the extreme expression of untouchability. Some other scholars have also worked to explore the origin of untouchability, for example G.L. Hart who opines that untouchability was a product of the ancient Tamil society; N. K. Dutta considers the attitude of the Dravida communities towards the non-Dravidian communities to be the origin of untouchability; the German scholar Fürer-Haimendorf sees the development of urban civilization as the reason behind untoucability. However, Vivekanand Jha’s work on this subject is considered to be the finest. He has shown that it was not the notion of purity and pollution which made certain tasks so inferior that people performing these tasks were declared untouchables; rather, the exploitation of some classes became so naked and barbaric, that the concept of pollution was attached to their occupation and the people in these occupations were declared untouchables. As it is its wont, the brahminical ideology has given the class division and exploitation a ritualistic form. Needless to reiterate, we are not talking about overlapping of class and caste here, but religious ritualistic legitimation and ossification of the relations of exploitation and oppression that are inherent in the entire socio-economic formation. In this entire structure, as we have already mentioned, a relation of correspondence exists between caste and class.

Ramsharan Sharma, has clearly shown that the casteist restrictions and stereotypes pertaining to commensality, matrimonial alliances and untouchability too, have undergone a process of evolution and development. Suvira Jaisawal and D.D. Kosambi also have shown that there is indeed a history of the development of the idea of purity/pollution. The task of framing and propounding these ideas was done by the brahmins, both as a part of the ruling class as well as its ideologues. The function of these ideas was to provide permanence to the dominant relations of exploitation by ritualistically ossifying them. Whenever the old ritualistic structure became suffocatingly restrictive for the changes taking place in the class-equations, necessary adjustments and modifications were done in this structure. In this entire process, by the medieval period, among brahmins too, such divisions were created that some brahman castes were pauperized. In particular, there was a decline in the material and ritualistic status of those brahmins who used to live on alms and donations (dān-dakshina). Declan Quigley has mentioned the case of untouchable brahmins in his book ‘The Interpretation of Caste’. Thus, the status of the entire brahmin population too was not fixed and impervious to any change.

Vivekananda Jha has mentioned four stages in the origin and development of the untouchable castes, for which historical evidence are available. The first stage was the Vedic period. There is no mention of untouchability in the Ṛgvedic period. Even in the later Vedic period the Chandalas are mentioned as hīna śudras and a sense of repulsion is expressed towards them but there is no mention of untouchabiltiy in clear terms. The second stage was from 700 BC to 200 AD. Some castes clearly emerged as untouchable castes in this period. This is the period when slave-labour was extensively used in the economy, and the first century AD saw the rise of feudal mode of production. The third stage was from 200 AD to 600 AD. In this period, some new tribal groups were inducted in the Aryan Vedic society as untouchable castes. And the fourth stage was from 600 AD to 1200 AD which is the high period of feudalism, and this is when untouchability appears on a large scale as a phenomenon. B.N.S. Yadav has drawn attention towards the fact that villages gained significance with the development of feudal economy, and there came a system of stable and static, which did not permit any mobility to the oppressed and exploited castes, especially to the artisanal castes. For Yadav, this factor also gave impetus to untouchability since it further degraded the lowest sections of the population.

●

While Buddhism and Jainism challenged the hegemony of the brahmins, they failed to pose any serious challenge to the varna/caste system; rather, these religions strengthened the varna/caste system in certain respects. Irfan Habib writes that Buddhism and Jainism have rejected the religious legitimation of the caste system, but have accepted the caste system as a reality of the society. This seems to be correct because the prejudices that exist in these religions against slaves, farmers under debt, and along with them against women, is explicitly clear. When the vaiśya trading castes with their rising economic might opposed the brahmin hegemonism and entered into the fold of Jainism, elements of the caste system also in a way penetrated Jainism, because the vaiśya castes there too continued to follow the rigid conventions of caste-based occupations and endogamy. It would not be incorrect to say that today Jainism has to a large extent been transformed into an appendage of Hinduism. Irfan Habib also remarks that the emphasis on the principle of karma and non-violence by Buddhism in fact proved to be an anathema for the untouchable population, because while laying stress on these values, the occupations which were declared as lowly were generally the occupations of the untouchable castes. Buddhism also largely became irrelevant with the emergence of Vaishnava and Shaiva sects in the Hinduism and also due to the fact that it showed even more enthusiasm in prohibiting cow-slaughter. It was not due to the reason that Hinduism had re-established its claim on the notion of purity, as claims Louise Dumont; rather due to the fact that Hinduism had once more got into step with the production relations of the changing times. Seen in this way Hinduism is a remarkably flexible religion, and as Weber has said, it is actually not a religion at all in the classical sense (however, this idea is incorrect as, according to Weber whereas a religion thrives on dogma, doxa prevails in Hinduism); Ambedker, in a way was right to remark that the core value of Hinduism is the caste system. In fact, this caste system too enhances the flexibility of Hinduism. The ideology of caste has given a useful instrument to the ruling classes through all the ages. It is such a flexible ideology, which, in all ages and especially in the pre-capitalist societies, provides the ruling classes with an instrument to consolidate their rule. It gives religious legitimation to the naked and barbaric exploitation of the ruling classes, and assumes the form of ritualistic ossification. Definitely, due to this ideology there persists a difference between caste and class. But until all the economic and political registers of caste essentially disappear (as it happened with the rise of the capitalist mode of production), a profound correspondence remains between caste and class. At least the history of India stands as a testimony to this fact. The caste ideology remains autonomous from the system of class in a certain sense. And it is essential for the caste ideology to exist in that way, if it wishes to remain really effective.

If the caste ideology were to reflect the class division, then it would lose all its divinity and aura. We should not forget that caste ideology is a religious ideology, which obtains its authority from religion, through occupational and matrimonial restrictions, and on the basis of purity/pollution, to justify its hierarchy. Obviously, if we comprehend this, then it becomes easier for us to realize that caste can never perfectly overlap with class. They can have a relation of correspondence only. But definitely, caste ideology from the time of its inception to this day has been providing an enormously powerful instrument to the ruling class in different forms. On the one hand it keeps the poor toiling masses under structural subordination, and at the same time it keeps them divided among themselves in so many castes. But the caste ideology performs this task in different ways, keeping itself in conformity with different modes of production.

It is this utility of the caste ideology that made it tolerable to the rulers of the Delhi Sultanate and the Mughal Empire, or rather we should say that it made itself desirable to them. Irfan Habib has shown that the Muslim rulers have not only kept themselves away from tampering with the caste system, they never even mouthed a couple of bad words against it. The only Muslim observer who has made a mild criticism of the caste system was a scientist namely, Al-Beruni. But if we leave this exception, then the Muslim rulers per-se have never objected against caste oppression and repression. On the contrary, when the Arabs conquered Sind, the commander of the army sanctioned the terrible casteist oppression of the Jatt population. Islam criticizes the Hinduism only for idol worship and polytheism. But it views the caste system with jealousy! The Quran only mentions the distinction between a slave and a free man; had it not been so, the religious leaders and administrators might have tried to co-opt this system in their own way! And in practice, the caste system has successfully made inroads into the Islamic society. The people from the dalits and the lower castes who adopted Islam came to be known as kamins, which means inferior and lowly. All of this does not mean that the caste system possesses some deadly but divine weapon that pollutes everything that comes into its contact, but itself never perishes. It only means that, in all ages the casteist ideology has presented itself to the rulers who came to India, as a readymade, extremely flexible, and useful tool for the legitimation its exploitation. In such a case, why should any ruling class shy away from putting it into use? This is the reason why the caste system remained intact as a useful ideology providing religious ritualistic legitimization to class exploitation throughout the medieval era.

•  Historicity and Contemporaneity of Caste in Modern India: A Brief Note.

With the start of the colonial era, the caste system went through a few significant changes. The principal factors behind these changes can be observed on different levels.

At one level the contemporary form of the caste system and caste hierarchy itself was consolidated with certain changes. For instance, in 1793 when Permanent Settlement was implemented, it provided a base for the exploitation and oppression of the landless dalit castes. At the same time, the Ryotwari land settlement made one section of peasant castes, which was already showing upward mobility, owners of land. Mahalwari settlement in a way passed the control over the land to the chief of the village community. The land reforms brought by the British did not make any appreciable change in the casteist hierarchy and equations prevalent in different areas. If anything came out of it, it was that, that a thorough arrangement was made to keep the dalit population in a perpetual state of structural oppression, exploitation and repression even in the future. In some places their oppressors were the old upper castes viz. the brahmin and the kṣatriya castes (e.g. in the United Province and Northern India) and in others they belonged to the emerging peasant castes which although had the status of śudras in the ritualistic hierarchy, but economically and politically their condition had improved.

Yet another level at which the British had influenced the caste system was development of industries to a certain extent and their role in bringing in the railways. Marx had foretold that the hereditary division of labour, which prevailed in the caste system, would begin to break with the development of railways and industries. Broadly this formulation proved to be correct. The British did not develop the industries on very large scale. In a way the old industries were destroyed and some new industrial centres had developed whose task was to supply the raw material.  But among the proletariat which had grown in the industrial centres such as Calcutta, Bombay, Surat, Ahamedabad etc. the rigid hereditary division of labour was obviously not possible within it. It is true that this proletariat was largely composed by dalit and people from lower castes. But there happened to be a rigid occupational divide among these castes themselves. The process of disintegration of this rigid hereditary division of labour had begun in the British period. Surely, after independence and with the capitalist development, this process unfolded with much rapidity. However it is an undisputable fact that its seed were sown in the colonial era itself.

The third level at which the British colonial state left a profound impact on the caste system is the one which we have already discussed above. The colonial state reconstructed the whole concept of the caste system. The belief of Nicholas Dirks and other followers of Subaltern Studies like him, that caste is an Orientalist construct of the colonial state, would be a kind of subjectivism. No state can ever make a construct of any such divide from the above, unless that division has a history of its own. It must certainly be accepted that the fetish of the the British ethnographic state to count, enumerate, classify and systematize the castes did indeed shake the division and hierarchy in the castes once and made it rigid in a new way. Historians like Arjan Appadurai, Bernard. S. Cohn, Susan Bayly, and Nicholas Dirks have written profusely on this whole process. The criticism of people like Dirks by Susan Bayly, Sumit Guha and Richard Eaton is correct that he fails to see the the collaboration between the colonial state and the native elites, including the brahmins also, which led to the reconstruction of the caste system in its modern form. Nor the Subaltern historians are able to understand that the theory of construction of caste by the colonial state for the oppression of the Indian people is like a conspiracy theory which fails to explain that in reality the archives of colonial knowledge, that the colonial state had been building up, was its own necessity, i.e., the necessity of ruling in more effectively. This whole exercise was not for the project of cultural domination rather definite political and economic factors were at work behind it.

Declan Quigley has rightly termed this approach as Idealist. Quigley says that the outcome of the ideas of people like Nicholas Dirks, Ronald Inden, etc is that caste becomes a mental construct, a linguistic jugglery. This point of view a moral ‘crusade’ born out of a kind of imperialist guilt-conscience, which holds imperialism guilty of those crimes, which it simply did not commit. But these of kind of ideologies which work behind this entire exercise end up strengthening imperialism itself. Because in the present era, imperialism is in a direct alliance with the revivalist Fascist forces. They have also the same argument that it was the British who created caste and before that, in Hinduism, we had a division of labour which was based on karma only, not on birth.

It is evident that with the development of capitalism and large scale industries and with the further development of urbanization, the two aspects of caste system are moving towards an end. First, the hereditary division of labour. Determining the occupation or job on the basis of birth is now a thing of the past. The caste character still manifests itself in some occupations in the field of self-employment, for example, washermen, barbers, etc. But this is no more a rigid division of labour, which cannot be transcended. Moreover, commensal prejudices too have been broken to a large extent, because it cannot be continued in the same way in the new kinds of villages, and in the cities and towns their complete disappearance is inevitable. We may say that these two registers of caste have weakened to such as extent that in the near future they will become extinct. These two aspects are not congruent with the capitalist mode of production and production relations, therefore, with the advent of capitalism they were bound to meet this fate. We will not say it in the words of Irfan Habib that the social and economic registers of caste are fading away. But surely the two aspects of caste which we mentioned above, namely, commensal prejudices and hereditary division of labour, are heading towards the end.

There is yet another aspect which is still intact and that is the practice of caste endogamy. It is so because it does not have any conflict with the capitalist mode of production. Actually it is better for capitalism, and is in conformity with it. Even the persistence of patriarchy in a new form in capitalism is due to this very reason. And both these factors reinforce each other; that is to say, the patriarchy reinforces the capitalist system based on caste endogamy and the capitalist caste reinforces capitalist patriarchy. And these two join hands together to allow the capitalist system and the bourgeoisie to streamline its machinery of oppression and exploitation. In one aspect capitalism stands apart from all other pre-capitalist systems. It does not look for any other-worldly power to obtain the legitimation of its rule. It gets the legitimation of its rule from the ‘consent’ of the masses. This is what Gramsci names as hegemony. The rule of the capitalist class is based on hegemony and not on the direct domination. In this system the capitalist class manufactures ‘consent’ for its rule. In such a system the ideology of caste cannot be the ideology that provides legitimation to the ruling class and its rule as it used to do earlier. In fact, no religious ideology is any more able to perform this tak because the legitimation of the rule in its entirety is, by its own nature, no more other-worldly, but has become this-worldly. However, the question of caste system is not linked with the state only. Over the centuries the casteist mentality and ideology, with the various changes it has undergone, has been made to permeate every pore of the Indian psyche. The core of the casteist metality and ideology is the hierarchy determined on the basis of purity/pollution, and not a particular caste hierarchy, that prevailed during a particular historical era. This casteist ideology works in subtle forms and it does not always require invocation by the ruling classes. No capitalist ruling class can draw its legitimation from the caste ideology, but can use the caste ideology in two ways. One, to keep sections of the exploited working masses divided on casteist lines, and along with it, as a intrument to construct hegemony in its favour. We can see the naked run of this entire process during the bourgeois elections. Besides, as we have mentioned elsewhere, different factions of the ruling class in their mutual rivalry use caste equations, albeit rulers of every caste without fail, stand united against the people.

The capitalist development of agriculture has brought in many significant changes in the caste structure during the last fifty years. We can see these changes in the upsurge of the middle peasant castes.  Over the whole region from South India to Uttar Pradesh, Bihar, Haryana and Punjab right up to Gujarat, it is a well-recognized phenomenon. Most of these middle castes are backward castes whose ritualistic status is that of Śudra. However, in their own areas they have become economically and politically powerful, dominating castes. All other castes, which include the brahmins and thakurs remain under their domination. We may call these castes kulak castes as well. Sociologists like Gloria Raheja, McKim Marriott etc have made considerable efforts to theoretically formulate this entire phenomenon. Raheja while making a study of a village Pahansu of Uttar Pradesh tells that in this village Gujjars are the dominant caste with all the other castes surrounding it. Here Raheja presents the theory of ‘centrality of the dominant caste’ and tells that it is the Gujjar caste that gives gifts and donations to all the other castes, but apart from kanyadaan (a ritual of donating girl performed during wedding) it does not accept any donation. Giving donations is symbolic of their elevated status. The relation of the Gujjars with other castes is the relation of authority and power, but no such mutual hierarchy is seen in the rest of the castes.

There is yet another phenomenon which we can consider as an outcome of the domination of the capitalist mode of production. It is the decline in the status of the brahmins living on alms and donations. In some places, their status has become just like that of the dalits. In our opinion, the reason behind it is that, in a capitalist society only exchange gets recognition or it is the practice offering gifts among people of equivalent status (of course, we all know that this also is a kind of exchange only, and nothing else!). With the emergence of capitalist social formation decline in the material and ritualistic status of brahmins who live on alms and donations is quite normal and it can be understood.

• Epilogue

All sorts of phenomena can be enumerated which have occured in the caste system with the emergence of the capitalist mode of production; the capitalist system of production would definitely not put an end to the caste system. The caste system provides it with continuity of property relations in the form of caste endogamy and also a powerful political instrument to divide the masses. With capitalist development and emergence of a massive class of proletariat, the aspect of gap in the correspondence between class and caste has increased considerably. This correspondence becomes visible only with incisive study. For instance, in the present times, this correspondence between class and caste can be seen more strongly among the class of landless peasants. But the population of the other backward castes and the middle castes has rapidly grown in the entire proletariat. But the weakening of the correspondence between caste and class has created an opportunity for capitalism to use caste ideology. While on the one hand, conditions of spontaneous breaking up of caste bonds in the working class arise; on the other hand, the ruling classes also get an opportunity to divide the proletariat on caste lines. Had this gap been small and had 80 to 90 percent of the proletariat come from the dalit castes, the scope of use of the caste ideology to divide it would have been less.

Therefore, the caste ideology is providing a powerful weapon to capitalism to divide the proletariat and, through caste endogamy,  maintain the continuity of the sacred bourgeois property. Therefore, it would be foolish to expect capitalism to put an end to the caste system. But at the same time it is also essential to understand that the caste system has not remained the same from its inception; it has been continuously changing, and the principal factors behind these changes have been the changes in production relations, mode of production, and class contradictions. It is also evident that the caste system has come into existence along with class, state and patriarchy and has become an instrument for their legitimation. Therefore, till class, state and patriarchy exist in any form, the caste contradictions, ideology and mentality too will continue to exist. Only a struggle for a classless society can be a struggle for a casteless society. This certainly does not mean that the question of caste should be pushed under the carpet till the time, the struggle for a classless society reaches completion. On the contrary, it means that from this day itself the proletariat in its struggle against capitalism, has to wage a war against all these ideologies, identities which break it, divide it and disintegrate its resistance. Without a relentless, untiring propaganda against caste and casteism the proletariat cannot be organized against capitalism and without the establishment of a socialist state under the leadership of the proletariat and without marching forward to a classless communist society, caste and casteism can never be destroyed.

●

Certainly, it was not a comprehensive and complete account of historiography of caste, nor is it proper to expect this from an paper. More than just presenting historical facts, our objective was to reject every kind of  reification of the caste/varna system (be it done by the post-modernists, Orientalists etc., be it done by the state, or then, done by the religious authorities, or else, by those who themselves practice identity politics on the basis of caste), every kind of its valorization, every kind of idealization, essentialization, and naturalization; to understand the caste system in its historicity and dynamism; to comprehend the essential character of this historicity and dynamism, that is to say, to understand the dynamics of production relations, mode of production, and  class contradictions; and to grasp the fact that if the varna/caste system which, through its origin and development over some thousands of years, determined by its socio-economic context and background has reached this juncture, then the same would happen in future as well.

To say that, ‘caste determines everything’ would be reductionism to the same extent, as it is to say that ‘only economic factors determine everything’, and Marx and Engels have rejected determinism of all shades in the characterization of a social phenomenon and have advocated a dialectical and historical materialist method. If it is understood that the caste/varna system has a beginning, then we can think about the projects to put an end to it in a more meaningful way. Without understanding it in its historicity, we will be either a victim of defeatism or pessimism, or else, of a pseudo-optimism which is always more dangerous than pessimism. The only objective of this essay of ours was to present in all humility, a historical understanding of the caste system, and if we have been able to present even a hazy portrait, we will consider ourselves successful.

(Paper presented in the Fourth Arvind Memorial Seminar on ‘Caste Question and Marxism’, March 12-16, 2013, Chandigarh)

Historiography of Caste: Some Critical Observations and Some Methodological Interventions