From Varna to Jati (Part 2) – Y.Naveen Babu

 This is the second part of Martyr Comrade Naveen Babu’s MPhil dissertation. For the first part click here.


The relationship between Varna and jati is an important aspect for understanding the changes in ancient Indian society. Most of the writings that deal with this relationship are ambiguous and full of contradictions. As we will see later, many scholars hold different perspectives on Varna and jati simultaneously. The different perspectives in understanding Varna and the relationship between Varna and jati may be broadly classified as follows: (1)

1. The Theory of Mixed Unions: According to this theory the society was originally divided into four Varnas and the numerous jatis emerged out of the inter-mixing of various Varnas or varnasamkara.
2. The Theory of Dual Reality: According to this theory Varna provides a universal framework and jati refers to empirical phenomenon (reality).
3. No difference between Varna and jati, both are one and the same.
4. Varna is an irrelevant and confusing category,jati is the only relevant category.
5. Varna and jati are different categories and belong to two different modes of production.


For the first time the word Varna was mentioned in the Rg Veda. Throughout Rg Veda (except in the PurushaSukta) it was used in the sense of colour and referred to Aryans (fair in complexion) and dasas or dasyus (dark in complexion). It is in the PurushaSukta that a mention is made of the origins of brahmanas, kshatriyas, vaisyas and sudras from different parts of the Purusha’s body. Brahmanas came out of the mouth, kshatriyas out of the hands, vaisyas out of the thighs and sudras out of the feet.(2) Interestingly, brahmanas, kshatriyas, vaisyas and sudras are not referred to as Varnas in the PurushaSukta. The PurushaSukta only mentions about the origin of these four groups but not of four Varnas.(3)

It is in the later writings (that) these four groups are referred to as four Varnas. With this the application of the word Varna has changed from colour — referring to Aryans and dasas or dasyus — to the division of society into four groups. Henceforth, the word Varna is used in the latter sense by all later writers. The post-Rg Vedic writers unanimously talk of the Indian society in terms of four Varnas. These Varnas are arranged hierarchically: brahmanas are at the top, followed by kshatriyas, vaisyas are below kshatriyas and sudras are at the bottom. The law books prescribe functions, privileges and duties to the four Varnas.

But in the later-Vedic society, there were groups other than the four Varnas. Moreover, new groups were emerging either by incorporating tribal communities into the mainstream society or by the internal divisions of the old society. The law makers of the time were faced with uneasiness to account for the theory of four Varnas in reality. In order to link the ‘sacred’ theory of Varnas with the existing reality (i.e., jati), they developed another theory — the theory of mixed unions or varnasamkara. According to this theory, there were originally only four varnas, but due to the intermixing of various varnas in the later period (kali age) the intermediary and lower jatis came into existence. The status of these jatis depended upon the status and the nature of mixture between the father and the mother.

We find many proponents of this view among contemporary scholars. According to Kane, the criticism against the theory of mixed unions:‘is true only partially… The element of hypothesis and speculation lies only in the theory of a particular sub-caste having sprung from the union of two persons belonging to two particular varnas or castes.’(4) Tambiah goes further and upholds the theory of mixed-unions or varnasamkara by using taxonomical classificatory concept of ‘key’. He feels that the theory of mixed unions, based on approved or disapproved unions, enables us ‘to comprehend a whole universe of numerous castes, all in principle capable of being ranked and interrelated into a single scheme.’5 Tambiah begins by showing how Varna and jati are interrelated through the mixed-unions but ironically ends up saying that the theory of mixed unions is ‘fictional and non-historical.’6 Dumont while maintaining that Varna and jati are two distinct categories, implicitly agrees with the theory of mixed unions.7 According to Risley, ‘a man may marry a woman of another tribe, but the offspring of such unions do not become members of either the paternal or maternal groups, but belong to a distinct endogamous aggregate, the name of which often denotes the precise cross by which it was started.’8 Nevertheless, Risley argues that the classical writers have clubbed all other processes of jati formation into the theory of mixed-unions.9 For R.S. Sharma, varnasamkara indicates kali age where vaisyas and sudras refused to pay taxes and perform the functions allotted to them. As a result of this, a crisis has emerged and the functionin of the society has become difficult. This resulted in the formation of feudalism.10 N.K. Dutt, Bougle, Ketkar, Ghurye, Karve, SuviraJaiswal and V.N. Jha are some of the scholars who accepted the theory of mixedunions.11

Now let us see whether the theory of mixed-unions is logically and empirically valid or not. Scholars like Senart12 and Trautmann13 rejected the theory of mixed-unions as ‘unconvincing’. Fick believes that brahmanas, in order to further their interests, have introduced the theory of mixed-unions. He points out that the names of these mixed-unions suggests the names of lands or peoples (for example, Magadha, Nisada, Vaideha, Ambashtha, Malla, Licchavi, etc.) or professions (suta or cart-driver, vena or maker of reeds, nata or dancer, kaivarta or fi shermen, etc.) they followed.14

A jati needs a certain minimum number of members to continue as a group. To say that a jati is formed out of the mixed-unions of two varnas or groups means that: a) the offspring of this union are substantial in number to form a jati; b) the offspring are not accepted either by the father’s Varna/group or by the mother’s Varna/group, but are grown up independently as suggested by Risley.

If a substantial number of offspring arise out of the mixed-unions, then it suggests that the two varnas or groups are in continuous conjugal interaction. This goes against the basic principle that Varna is an endogamous group.15 If the governing principle of any institution is violated on a large scale continuously, then we cannot uphold that principle as the governing principle.

Kane mentions 62 occupational groups ‘which had probably become castes or were in the process of becoming castes, before the close of the Vedic period.’16 The number of mixed jatis or samkarajatis mentioned by Dharma Sutra writers varies form one writer to another.

Ap.Dh.S. mentions only candala, paulkasa and vaina. Gautama names five anuloma castes, six pratiloma, one and eight others according to the view of some. Baudhayana adds to those mentioned by Gautama a few more viz. rathakara, svapaka, vaina and kukkuta. Vasistha names even a smaller number than Gautama and Baudhayana. It is Manu (X) and VisnuDh.S (XVI) that for the fi rst time dilate upon the vocations of the mixed castes. Manu refers to 6 anuloma, 6 pratiloma and 20 doubly mixed castes and states the avocations of about 23; Yaj. names only 13 castes (other than the four varnas). Usana names about 40 and gives their peculiar vocations.17

Kane further adds:

A Smrti verse quoted by Visvarupa on Yaj. I. 95 says that there are six anulomas, 24 doubly mixed castes (due to the union of the six anulomaswiththe four varnas), 6 pratilomas and 24 doubly mixed castes (due to the union of 6 pratilomas with four varnas) i.e., in all 60 and further mixtures of these among themselves give rise to innumerable sub-castes. Similarly, Visnu Dh. S. 16.7 says that the further mixed castes arising from the union of mixed castes are numberless. This shows that before the time of the Visnudharmasutra (i.e., at least about 2000 years ago) numberless castes and sub-castes had been formed and the writers of dharmasastra practically gave up in despair the task of deriving them, even though mediately, from the primary varnas.18

The theory of mixed unions suggests that an innumerable number of jatis have emerged out of the mixed-unions, starting with the four varnas and continuing with the groups (jatis) formed out of these unions. If this is the case, it clearly violates the principle of endogamy, to the fullest extent possible, on which both jati and Varna are supposed to be based. This also suggests that instead of endogamy, the mixed-unions should have become the norm. But the Dharma Sutra writers are particular about maintaining endogamy strictly. ‘The smrtis ordain that it is one of the principal duties of the king to punish people if they transgress the rules prescribed for varnas and to punish men and women if guilty of varnasamkara.’19 Thus, the Dharma Sutra writers are faced with a contradiction in dealing with the theory of mixed unions. On the one hand, they condemn the varnasamkara, while on the other hand they go into the minute details of ranking the jatis emerging out of mixed unions. Moreover, they suggest that upward mobility of one’s own jati position is possible by continuously marrying into higher jati for five generations or more.20

Empirical evidence might provide some insight into the validity of the theory of mixed unions. The north Indian kinship system is characterized by hypergamy. That means, a woman of low status marries a man of high status. As a consequence of hypergamy, especially those who are at the bottom of a particular jati which practices hypergamy do not strictly follow endogamy. These men marry women belonging to lower jatis usually next to them in the hierarchy. In this case, mixed unions are taking place. But this has not necessarily resulted in the formation of new jatis. Both the women and the offspring(s) are taken into the man’s jati and the offspring gets the status of the father.21

Another case where mixed unions have been taking place without leading to the formation of new jatis is that of Nayars. The offspring of Nayar women (considered as sudras) and Nambudiri men (brahmanas) belong to Nayars and get the status of the mother. These two cases show that mixed unions need not necessarily lead to the formation of new jatis. As we have stated earlier, the theory of mixed unions is adopted by classical writers only to account for the relationship between Varna and jati. This also shows that even though the theory of four varnas did not represent reality, the Dharma Sutra writers did not do away with it (which has attained a mythical status by now), but tried to somehow explain the existing reality in terms of Varna divisions. It further indicates the ideological aspects involved in the theory of four varnas and the theory of mixed unions.22 It is ironical that this view is continued in the contemporary period by so many scholars.


There is a widely prevalent viewpoint among the social scientists that Varna represents a universal framework, whereas jati indicates the empirical phenomenon.23 According to this view the Indian society is divided into four varnas and all the numerous jatis we encounter in our day-to-day life can be fi tted into one of these four varnas.

The main proponent of this viewpoint, Srinivas points out a number of difficulties that arise by accepting Varna as a model to understand Indian society. He rightly points out that ‘the caste system of even a small region is extraordinarily complex and it does not fit into the Varna-frame except at one or two points.’24 He further adds:

The Varna-frame is too rigid to fit the points of inter-caste relations today, and it may be assumed that it was always so rigid. According to Varna, caste appears as an immutable system where the place of each caste is clearly fixed for all time. But if the system as it actually operates is taken into consideration, the position of several castes is far from clear. This is due to the fact that the caste system always permitted a certain amount of mobility… Varna also conceals the considerable diversity, which exists between the caste system of one region and another.25

Besides this, Varna also results in pre-occupation with attributional or ritual factors in caste ranking at the expense of economic and political factors.26 Despite so many inadequacies of the Varna-model to explain Indian reality, which Srinivas himself points out, he accepts Varna as a theoretical framework. ‘Varna has provided a common social language which holds good, or is thought to hold good, for India as a whole.’27 He also uses Varna categories to explain the process of Sanskritization. Mandelbaum also points out the shortcomings of Varna framework but upholds it on the basis that most villagers are familiar with Varna and adopt the Varna model for Sanskritization. Moreover, varnas provide a neat outline of social relations, which enables the villager to place any new jati within this scheme of hierarchy.28 K.N. Sharma argues that Varna and jati belong to two levels of reality, Varna denotes ‘guna’ or ‘style of life’, and jati denotes birth. He feels that there is nothing new in the fi eldwork experience of two realities because even the ancient scholars were faced with the same problem and accepted Varna and jati as two layers of reality.29 According to Trautmann, the relation of ‘Varna to caste is that of the sacred and enduring to the empirical and ephemeral.’30 Dumont agrees with Srinivas’ views on Varna as a universal model. Dumont maintains that Varna and jati are two distinct systems but traces the homology between these two systems, both of which are structural and culminate in the brahmanas. He goes further and says that ‘far from being completely heterogeneous, the concepts of Varna and jati have interacted, and certain features of the osmosis between the two may be noticed.’31 In fact, Varna model is so important for Dumont, it is from this model he develops the notions of the subordination of power to priesthood, and then applies this notion to jati system. He writes, ‘the theory of castes resorts implicitly or obliquely to the varnas to complete its treatment of power.’32

Khare, following the footsteps of Dumont and Tambiah, looks into the interrelationship between Varna and jati from a ‘symbolic’ view point. He starts with the single aim of proving the importance of Varna in understanding contemporary Indian society. The result is the imposition of his scheme of Varna on the empirical facts, even though there are many ‘complications’ and ‘logical strains.’ He writes: ‘proceed towards a contemporary jati, with its constituent social groups and their interrelationships, and the concrete and the factual receive increasing attention. However, once the Varnas are given (emphasis added) primary attention, the ideal and the symbolic take over.’33 He further writes that ‘the Kanya- Kubja Brahman confronts jati as concretely as he handles goods and services in a marketplace classified by money.’34 Nevertheless, ‘A Kanya-Kubja Brahman… requires both jati (practical) and Varna (symbolic) specifications of his status.’35 He further feels that, Varna is a culturally necessary “key” for a jati to fi nd its place within the system, but one which does not fulfi l all the empirical conditions an overarching taxon should have within a perfect taxonomy. Hence, beyond a point, jatis themselves must carry all major clinching classifiers… The Varna system set up what taxonomists call a tree, but it is not a perfect tree (where all derivative nodes are labelled), much less a perfect paradigm or a perfect taxonomy.36 Further, ‘Complications in the jati-Varna classification thus appear, among other reasons, because of incomplete but necessary classifiers…’37 Necessary for whom? Khare himself points out that Varna is necessary for those who occupy the top position.38 Despite so many complications encountered by him to bring together Varna and jati, Khare insists that varnas ‘must help jatis fi nd their relative signifi cance.’39 Jaiswal says that in modern times ‘Varnas are broad categories subsuming within them a large number of jatis in a rather loose fashion.’40 Nevertheless she argues that in ancient period both Varna and jatisignied the same thing. Romila Thapar also considers Varna as a theoretical framework and jati as a more evident and concrete phenomenon.41 She argues that in the Vedic society, which she characterizes as a lineage society, Varna developed with the emergence of stratification. Thapar suggests that Kshatriya and vaishya emerged out of the Jana, whereas brahman and sudra were derived from the earlier Harappan culture. The integration of these two sets of dichotomous groups gives rise to four Varnas in which terms the later Vedic society was sought to explain.42 When the Vedic society has transformed from lineage to state, Varna has also undergone a change. This change is reflected in the duality between ritual status (Varna) and actual status (jati).43 Thapar thinks that in the transition from lineage to state, Varna as a theory helped this process by integrating the old with the new elements. In this sense, Varna takes up an intermediate position between stratified (lineage) society and class (state) society. Nevertheless, Varna has continued in state society as a theoretical framework.44

Besides these scholars, Kano, Ketkar, Ghurye, Risley, Pocock, R.S. Sharma and many others support the view that Varna is a theoretical framework and jati is an empirical phenomenon.45 Is it logically possible for a number of jatis to constitute a Varna? If we accept that both Varna and jati are characterized by endogamy, then it is not possible to say that a number of jatis constitute a Varna, because we cannot have an endogamous group within another endogamous group.

An endogamous group is constituted by many exogamous groups. A person can marry outside his/her exogamous group into any one of the exogamous groups within the endogamous group. When we say that a group is endogamous it means marrying outside this group is not possible unless one violates the norm. Thus, it is not possible to have an endogamous group within another endogamous group.46 Secondly, we do not have these four Varnas throughout India.47 In all the four states of South India, in Maharastra and in eastern India the Kshatriya and vaisyaVarnas are conspicuous by their absence. A theory which is logically inconsistent and which fails to take into account the absence of some of its basic categories in the major parts of the country cannot be said to be a convenient theory to understand Indian society. It seems that the scholars who support the four Varna theory, presuppose the necessity of this theory rather than looking for a viable alternative.



Sharma divides the ancient Indian society into four phases: tribal society (Rg Vedic period, c.1500 B.C–c.1000 B.C; characterized by pastoral economy), Chiefdom (later Vedic period, c.l000 B.C.–c.500 B.C.; characterized by small-scale non-monetary peasant society), vaisya-sudra social formation (post-Vedic period, c.500 B.C.–c.300 A.D.; characterized by classes) and feudalism (beginning from c.300 A.D.).48 In the Rg Vedic period, the society was not organized either ‘along Varna lines or class lines but along tribal lines.’49 Sharma feels that in a predominantly pastoral society the surplus accumulation is not possible at a large scale. Thus, Rg Vedic society was a ‘tribal, pastoral, seminomadic and largely egalitarian society.’50 Rg Vedic society was based on ‘gift economy respected by custom in the beginning and sanctioned by force at a later stage.’51 Fellow tribesmen made offerings to the king in cattle, dairy products and foodgrains, which were later redistributed at periodical sacrifices organized by the tribal princes.52 The major source of wealth in Rg Vedic society was cattle, and a wealthy person was called gomat. The king was referred to as gopa or gopati.53 Agricultural activities were less in Vedic period. Barley was produced in some quantity.

Nevertheless, Rg Vedic society was mainly a pastoral society. Another very important source of wealth in the Rg Vedic society was spoils of war. ‘War in a predominantly tribal society of the Rg Veda was a logical and natural economic funtion…’54 Rg Vedic tribes were constantly at war with each other and spoils of the war were distributed among the tribesmen. The distribution was of course not equal. Priestly and warrior groups managed to corner the major share of the booty, since the surplus given to the chiefs by tribesmen as gifts, which is supposed to be redistributed on religious occasions, was not redistributed equally. As a consequence of this, the chiefs and priests accumulated large amount of wealth, creating inequalities in the society. Even though Sharma fi nds that some sections of the Rg Vedic society were in possession of more wealth than others: the distribution of the spoils was unequall thus benefi ting priests and warriors more than others. He prefers to characterize the Rg Vedic society as a rank society rather than a class society. In the Rg Vedic society, he fi nds the presence of domestic slaves mainly consisting of women. Besides this we do not fi nd slaves on a large scale in this period. Rg Vedic society was divided into two main groups — Aryans and dasyus or dasas. ‘Although the word Varna is applied to the Aryan and Dasa in the Rg Veda, it does not indicate any division of labour, which becomes the basis of the broad social classes of later times. Arya- and Dasa-Varnas represent two large tribal groups which were in the process of disintegration into social classes.’55

Sharma feels that dasyu and dasa represent two tribal names, which were later used indiscriminately to refer to the pre-Aryan people and the earlier wave of Indo-Aryan tribes.56 According to Sharma, both dasyus and dasas were part of the Indo-Aryan people. The Indo-Aryans came to India in successive waves and dasyus (Iranian Dahyus) were one of the earliest waves to reach India. Their way of life has undergone a major change by interacting with aboriginals and when the later wave of Aryans came, whom dasyus opposed, they were treated as low. The same is true with dasas (Iranian Dahaes tribe) who came after dasyus but still retained contacts with the original Aryan groups and thus were more acceptable to Aryans. Sharma elsewhere refers to dasyus as the people with a different language and life-style from Aryans.57 If dasyus were an earlier wave of Indo- Aryans, one fails to understand how their language is different from the later wave of Indo-Aryans. We know that Aryans came as destroyers of the earlier Indus civilization and established their hegemony over here. If we accept the proposition that dasyus were an earlier wave of Indo- Aryans, then it means that dasyus were completely assimilated by the natives to such an extent that dasyus changed not only their life-style but also their language. This goes against the established view about Aryan invaders. Sharma’s proposition that dasas were also an earlier wave of Indo-Aryans is also fraught with similar flaws. In Rg Veda, dasyus and dasas were used as synonyms and interchangeb1y at many places. If we follow Sharma’s arguments, this is highly improbable because dasas were close and friendly with Aryans, whereas dasyus were enemies of Aryans. How can both enemies and friends be clubbed together and referred as one? Moreover, the distinction between Aryans and dasas or dasyus continued throughout Rg Vedic period. This is significant because, whereas the social differentiation within the Aryan tribes has not clearly distinguished, the distinction between Aryans and/or dasyus was strictly maintained throughout Rg Veda. This distinction becomes even more important if we accept Sharma’s proposition that dasas and dasyus were part of Indo-Aryans.

The later Vedic texts divide the society into four social orders or statuses based on occupation — brahmana, kshatriya, vaisya and sudra. ‘These cannot be regarded as four social classes in the sense that some of them owned land, cattle, pasture grounds and implements and the others were deprived of them.’58 Nevertheless, there are clear indications of the rising inequalities. ‘In a way, the fi rst two orders constituted the ruling class, and tried to establish their authority over the vaisyas who formed the producing peasant class with the sudras as a servile domestic adjunct, which was small in number at this stage.’59

By the end of Atharva veda, sudras had become a servile class. Sharma considers sudras as a tribe having close affinities with Aryans.60 According to him, sudras are a later wave of Aryans who came to India at the end of the Rg Vedic period and were defeated by the Vedic Aryans.61 In later times, sudras refer to both degraded Aryans and aboriginal tribes.62 The origin myth (that the four varnas originated from various parts of Purusha’s body) served as a useful fi ction to assimilate the heterogeneous elements into the Aryan fold.63 By the end of the Rg Vedic period, the defeated and dispossessed sections of the Aryan and non-Aryan communities were reduced to the position of sudras. Sudras at this period enjoyed several religious rights enjoyed by other upper Varnas. Sharma thinks that this has something to do with the nature of the economy. At this stage, ‘the peasants did not produce much over and above the needs of their daily subsistence’ to pay taxes and maintain a non-producing class.64 In the post-Vedic period, when the middle Gangetic basin was cleared, when iron was used for agriculture, agriculture has become the main activity of the society. At this stage, sudras were clearly distinguished from others and made into a servile group. ‘The sudras were excluded from Vedic sacrifices and investiture with the sacred thread, which were considered to be the ritualistic hallmark of an arya or twice born. The sudra was saddled with economic, politico-1ega1, social and religious disabilities. All this could be justified on the basis of his mythical origin from the feet of the creator.’65 It is in the Mauryan period that the condition of sudras has completely reduced to that of slaves. Sudras were forced to work on agricultural land under the direct control of the state. The post-Mauryan period was faced with the bitter ‘social conflicts and tensions, which was perhaps aggravated by the intervention of the non-brahminical foreign elements and the increasing importance of artisans. Probably as a result of this conflict, the disappearance of the strong state power of the Mauryas, and the rise of new arts and crafts, we notice signs of change in the position of sudras.’66 The development of crafts and the refusal of the two lower varnas to perform their functions in the kali age have necessitated new changes. As a result, in the “Gupta period the sudras gained some religious and civic rights, and in many respects were placed at par with the vaisyas.’67

Nevertheless, sudras remained as a servile class with some changes in their position by paying some part of their produce as peasants and artisans to the state. The Rg Vedic Aryans have transformed into later Vedic brahmana, kshatriya and vaisayavarvas. But what happened to the Rg Vedic dasas or dasyus? Sharma fails to account for the conspicuous absence of dasas and dasyus in later Vedic texts. Even if we assume that dasas or dasyus were converted to sudras, one fails to understand why they were called sudras but not dasas or dasyus, since dasas or dasyus also consist of defeated and dispossessed people. The non-continuation of these categories in the later Vedic texts becomes significant. Sharma’s arguments on the development of sudras shows that in the later Vedic period, they enjoyed religious rights, but lost them in the post-Vedic period and ultimately gained them again in Gupta and post-Gupta period. Another interesting development is that the vaisyas who enjoyed dvija status and who were treated on par with brahmanas and kshatriyas in religious matters were slowly, in later periods, degraded to the position of sudras. This is an interesting aspect, considering the fact that the vaisyas, who were peasants and numerically more, were reduced to the status of the servile group; and sudras, the servile group was elevated to the position of vaisyas. No serious explanation is provided by Sharma on what were the factors that necessitated this kind of change. Sharma failed to give due consideration to the changing notions of Varna categories while talking about the position of various varnas in different time periods. Vaisya in the inter-Vedic period denotes a peasant whereas in the post- Vedic period refers to a trader. Likewise, sudra in the later-Vedic period refers to all servile groups below the three dvijavarnas, whereas in the post- Vedic period it refers exclusively to peasants and artisans. In the post-Vedic period, an unnamed ‘fifth Varna’ developed, consisting of all the menial workers. This shows that the position of peasants remained more or less the same in both later-Vedic and post-Vedic periods and the same is true with menial workers. The only change was in the names used to refer these groups at different time periods. This further shows that overemphasis on the Varna categories (which no longer refer to the real groups in the society) rather than on the real position of the groups in the production process leads to the misinterpretation of the reality. Sharma also fails to explain how these Varna categories continued in different stages of development. SuviraJaiswal argues that Varna and jati signify the same phenomenon and can be used interchangeably. They constitute a single system.68 She agrees with R.S. Sharma’s view that Rg Vedic society had differentiations of ranks but not of classes.69

As the later Vedic society expanded, differentiation grew within each Varna. Jaiswal considers the brahma ksatra phenomenon (brahmanas becoming kings) as a transitional category where more preference is given to kshatriya status.70 Jaiswal agrees with the view that the non-emergence of Kshatriya and vaisyavarnas in south and other parts of India is due to the fact that in these areas there was “no conquering elite which might seek to preserve its identity through putative kshatriya status” and by forging kinship relations horizontally through widespread marriage networks, rather than vertically in the absence of traditional local roots… In the north, the four-tired Varna system has developed through the fi ssion and fusion of later Vedic tribes in which the brahmana, the kshatriya and the defeated sudra were clearly identifi able and the vaisya was a residual category including artisans, herdsmen, peasant, etc., that is independent producers.71 Jaiswal supports R.S. Sharma’s views on the development of sudras in various periods of ancient Indian history. She however feels that the change in attitude towards peasant communities in the Gupta period and post-Gupta period needs elaboration. Jaiswal shows that the occupation of vaisyas has changed from agriculture and crafts to trade and commerce from the beginning of the Christian era. ‘This shift in the concept of the vaisya Varna was primarily responsible for the characterization of the peasant communities in the Gupta and post-Gupta periods as sudras. At the root of this change lay the contempt for manual labour and the depression of the peasantry.’72 Vaisya Varna did not develop in the southern and other parts of India because in these areas the spread of brahminical culture has taken place when trade was declining.73 Jaiswal characterizes the early medieval India as consisting of three broad strata — the brahmana, the kshatriya or Rajput and the sudra in the north and the brahmana, the sat-sudra and the asat-sudra in the south and the east.74 The ranking of groups was no longer on ‘twice-born’ and ‘once-born’ but on the basis of the groups ‘which were created by the approved unions and hence were “pure” and those, which originated from disapproved unions hence were “impure”.’ In the former category were included not only the four principal Varnas but all those non-brahmana castes of high social status, which castes were described as sat-sudra or uttamasamkaras.’75

The regional variations in the Varna system during the feudal age shows that ‘secular factors had placed the Varna theory under a severe strain and four-Varna hierarchy was transformed into a hierarchy of numerous endogamous groups coming from diverse sources. But the basic principle, the intertwining of the pure and the dominant, remained unchanged.’76 Jaiswal, while upholding the Varna model in principle, discards vaisyavarna on the pretext that it is a ‘residual’ category. It is surprising to note that Jaiswal characterizes peasants, artisans, herdsmen, etc., as a ‘residual’ category.



Both Senart and Bougie reject the theory of four varnas as an ideological creation of brahmanas. Senart cautions the reader about the brahminical bias of the classical texts. He points out the self-interest of brahmanas in maintaining the theory of four Varnas.77 Senart distinguishes between Varna and jati. He feels that jati alone is relevant in understanding the reality and rejects Varna as an artificial system which was carefully thought out and adopted to the conditions in which it does not have any roots.78 Nevertheless, he suggests that these two orders ‘may combine and complete one another.’ Varna model by claiming domination to brahman class preserved a rigidity concerning religious scruples and further provided legitimacy to the notions of hierarchy and purity.79 According to Bougie, the ‘contemporary observations tend to show that the theory of four castes, the chaturvarna has never been more than an ideal, blending a simplified and as it were, shortened picture of the reality with a reiteration of frequently violated prescriptions. It would be useless to look at the caste of the present as the descendants of the four traditional castes…’80



We have already discussed Dipankar Gupta’s views on Varna and jati in the Introduction. We agree with him on two points: that Varna and jati belong to two different modes of production; and there is a necessity to study the material bases of both Varna and jati. But we disagree with his periodization and characterization of Varna and jati social formations. There are other scholars like Ambedkar who are not concerned about the relationship between Varna and jati, but deal with either one of them. By pointing to the inconsistencies in the classical texts, Ambedkar tries to prove that sudras were originally kshatriyas but later reduced to the low position because of their antagonism with brahmanas.81 He argued that brahmanas refused to perform upanayana (initiation ceremony) to ‘sudras’ thus reducing them to the low position.82 Ambedkar proves the kshatriya origin of sudras by tracing their genealogy to Sudas, a Rg Vedic kshatriya.83 The paradox in Ambedkar’s work is that while criticizing that brahmanas are biased and enemies of sudras and thus suggesting that one cannot rely on their works, he proves the kshatriya origins of sudras mainly based on brahminical writings. Nevertheless, Ambedkar provides some very beautiful insights into the ideological aspects of the theory of four Varnas. He points out that even though there are more than one Cosmogony in Rg Vedic dealing with the origin of man, the lat- er Vedic (brahman) writers have consciously adopted only Purushasukta where the origin of four groups or varnas was mentioned.84 Even in the Purushasukta, Ambedkar argues, where the origin of different species is dealt with, when it comes to the origin of man it mentions the origin of social groups or classes.85 He also contrasts the ‘unique’ nature of Purushasuktawith the ancient European Cosmogonies and points out that no other system has encouraged rigid class structure as that of the Purushasukta 86- Varna and the Pastoral Social Formation



Etymologically, the word Varna means ‘colour’. Throughout Rg Veda (except in the PurushaSukta), Varna is used in this sense only. There were two Varnas in the Rg Vedic period — Aryans (fair coloured) and dasas or dasyus (dark coloured). These Varna distinctions are maintained throughout Rg Veda period. In order to understand how the word Varna developed, to understand its application to different sections of the Rg Vedic society, and to ascertain its importance in the later development of the society it is necessary to understand the socio-economic conditions that gave rise to the emergence of Varna. There are two kinds of social differentiation during the Rg Vedic period. The first is the differentiation between Aryan and Dasa or Dasyu tribes; the second is the differentiation within the Aryan tribe —brahmanas, kshatriyas and vis (commoners). It is important to distinguish between these two types of differentiation for understanding ancient Indian social history. The mixing of these two kinds of differentiations result in the four Varna theory which has obscured the reality to an irreparable extent. These two kinds of differentiation arose out of a particular historical context and they played a significant role in shaping the later history of India. The distinctions of Varna arose when two cultural and linguistic groups with different skin colour, following different ways of living came into violent contact, where one group of tribes subjugated the other group of tribes. The differentiations within the Aryan tribe arose because of unequal distribution of wealth and specialization of occupations. It is this later differentiation (i.e., the differentiation within the Aryan tribes), which sets forward the development of society enabling the transformation from pastoral to agrarian social formation. As the society transforms from pastoral to agrarian economy, the distinctions of Varna (i.e., the distinctions between Aryans and Dasas or Dasyus) become redundant and a new kind of social differentiation begins, based on the internal differentiation of Aryan tribes.



Aryans, a cultural and linguistic group, who migrated to India from central Asia in two waves, the fi rst in the beginning of the second millennium B.C. and the second at the end, destroyed the Indus cities which were agrarian-based and settled down in Punjab, the land of fi ve rivers. Aryans were a semi-nomadic pastoral people. Aryans subjugated the local tribes, whom they called Dasas or Dasyus. In a pastoral society, where war booty was one of the main sources of acquiring wealth, different tribal groups fought with each other for cattle and other wealth. In the process, the winners subjugated those tribes — Aryans and non-Aryans, which were defeated. Probably these defeated tribes were not used as labour force because the pastoral society does not need so much of labour force. Protection of cattle is the task of the warriors and there were women domestic slaves to do the household work, which included milking, cleaning and feeding cows.2 It is only at a later stage, when agriculture became a major economic activity, that the necessity arose for labour force on a large scale and the tribes, which were subjugated earlier, were used as a labour force.(3) This is clearly shown by the post-Vedic usage of the word dasa in the sense of slave.



It has been suggested that dasas and dasyus were earlier waves of Aryans who were degraded for not following Vedic rituals.(4) The Rg Vedic hymns make it clear that the Aryans and dasas are two opposing camps fighting each other.(5) Prayers are offered to India by Aryans to subdue and destroy dasas. In Rg Veda (1.51.8), Indra is requested to differentiate between Aryans and dasas.(6) ‘This does not mean that there was difference between the two in bodily appearance only; on the contrary, the antithesis between the arya who is referred to as barhismat and the dasyu who is styled avrata clearly shows that ‘the emphasis was rather on the difference of their cults.’(7) The dasyus are described as avrata (not obeying the ordinances of the gods), or akratu (who perform no sacrifice), mrdharvacah (whose speech is indistinct or soft), anash (snub-nosed or dumb).(8) This clearly indicates that Aryans and dasas or dasyus are two distinct cultural and linguistic groups with differences in skin colour. The interchanging use of dasas and dasyus in various passages of Rg Veda indicate enmity from the Aryans.(9) Even though the Aryan tribes fought amongst themselves, they maintained a distinction between Aryan tribes and dasa tribes.(10) This distinction continued throughout the Rg Vedic period, and we can fi nd the traces of this distinction also in later-Vedic and post-Vedic periods where dasas become ‘slaves’. In the later part of Rg Veda, dasas were brought into friendly relations. Not all the dasa or dasyu tribes were defeated or subjugated. There were some powerful dasa tribes for whom brahamans performed Vedic sacrifices. For example, in Rg Veda (Vol III. 46.32) we read, ‘the singer took a hundred (cows or other gifts) from the dasa Balbutta and from Taruksa.’(11) In Rg Veda, there is no religious discrimination of dasas because they are not yet part of the Aryan tribes and as we have stated earlier dasas form a different cult.(12) Therefore, in the earliest period we find the word Varna associated only with dasa and with arya. Though the words brahmana and kshatriya occur frequently in the Rg Veda, the word Varna is not used in connection with them. Even where the words brahma, rajanya, vaisya and sudra occur, the word Varna is not used. Hence, one may reasonably say that the only watertight groups that are positively or expressly vouchsafed by the Rg Veda are arya and dasa or dasyu.(13)



This brings us to the internal differentiation of Aryans tribes. When they came to India, Aryans were broadly divided into priests, warriors and commoners. This is evident from the existence of such divisions among the Iranian Aryans. Vedic Aryans migrated from Iran. But the division of Aryans into brahmana, kshatriya and vis (commoners) had not crystallized into closed endogamous groups or jatis at this stage.(14) They indicate the broad divisions of  society based on certain specializations. In Rg Veda, there are references to other occupations, but the main functional groups at this period were priests and warriors. All other occupations were of secondary importance. Intermarriages between different divisions of Aryans were common and there was no restrictions regarding partaking of food amongst the divisions.(15) There are number of citations in Rg Veda which prove that the divisions — brahmana, kshatriya and vis — have not become closed endogamous groups or jatis. In fact, in many cases, both brahmanas and kshatriyas hail from the same family. Vasistha is addressed as brahman whereas ‘he is said to have born of Urvasi from Mitra and Varuna.’(16) Similarly, in Rg. IX.96.6 (Brahma Devanam), the word brahma does not certainly mean ‘brahmana by birth’ nor does Vipranam mean ‘brahmanas by birth’. In that verse, one who is super eminent among a group is specified, just as the buffalo among animals, the hawk among carnivorous birds.(17) In the story of Devapi and Santanu, the sons of Rstisena, the younger brother Santanu became king as Devapi was not willing to become king. The result was a famine due to Santanu’s transgression and so Devapi performed a sacrifice to induce rainfall. This shows that out of the two brothers one became a king and the other a purohita, so king and purohita did not depend on birth.

In Rg IX.112.3, a poet exclaims, ‘I am a reciter of hymns. My father is a physician and my mother grinds (coin) with stones. We desire to obtain wealth in various actions.’ In Rg. Ill 44.5, the poet wistfully asks Indra, ‘O, Indra, fond of Soma, would you make me the protector of people, or would you make me a king, would you make me sage, that has drunk of Soma, would you impart to me endless wealth?’ This shows that the same person could be a rishi or a noble or a king.(18) In Rg Veda, the word brahmana generally means ‘prayer’ or ‘hymn’. But in Atharva Veda (II.15.4.), brahma seems to mean ‘the class of Brahmanas’(19). The word kshatriyas in Rg. Veda means ‘valour’ or ‘power’ but in the later-Vedic period it implied ‘a class of warriors’. But at this period they have not formed a closed hereditary group. The word vis frequently occurs in Rg Veda referring to ‘people or group of people’.(20) The word vis is sometimes contra-distinguished with jana. Kane feels that since vis is qualified as Panchajanya, there is hardly any difference between jana and vis.(21) As the text indicates, it is preferable to distinguish between jana and vis because the jana is a wider group which includes brahmana, kshatriya and vis (commoners). ‘It is possible that sometimes vis is loosely used to refer to all people as in the case of Panchajanya.’



D.D. Kosambi assesses the impact of Aryans on the development of ancient Indian society in the following terms: The Aryans trampled down so many isolated primitive groups, and their beliefs, so as to create the preconditions for the formation of a new type of society from the remains. They were not themselves consciously nor magnanimously bent upon the creation of that society. They acted in their own destructive rapacious manner, for immediate gain. The chief contribution of the Aryans is, therefore the introduction of new relations of production, on a scale vast enough to make a substantial difference of quality. Many people previously separated were involved by force in new types of social organization. The basis was a new availability to all of skills, tools, production techniques that had remained local secrets till then. This meant flexibility in adoption, universality of improvisation. It meant new barter, hence new commodity production. The result was the opening up of new regions to cultivation by methods, which the more or less ingrown local populations had not dreamt, of using… The violent methods whereby these innovations were introduced effected more and greater improvements than did trade, warfare, or ritual killing.(22) He further writes that ‘the barriers so torn down could never be effectively re-erected because the Aryans left a priceless means of intercourse, a simpler language distributed over a vast region.’(23) Moreover the continuous shifting of the Aryan settlements enabled a wider region to come under the Aryan influence.(24) Archaeological evidence from the Rg Veda period shows that people did not know iron, but copper and bronze were used in small quantities mainly for weapons.(25) Rg Vedic society was predominantly pastoral. Cattle was the main source of wealth, so much so that cattle were considered to be synonymous with wealth (rayi), and a wealthy person was called gomat.(26) Agricultural activities were less in Rg Vedic period. There are twenty-one references to agricultural activities in the Rg Veda, most of them occurring in the latter part of the Rg Veda. Barley (yava) was produced during this period.(27) Thapar suggests that in the Rg Vedic period the pastoralists may well have controlled the agricultural niches without being economically dependent on them, particularly if the cultivated areas were worked by people other than those who belonged to ‘the pastoral clans’.(28) Accumulation of cattle was done by breeding as well as by capturing other herds. Cattle raids, thus, formed one of the basic economic activities in pastoral societies. The winner of the cows is called gojit, an epithet for hero.(29) War in a predominantly tribal society of the Rg Veda was a logical and natural economic function. …The Rg Vedic tribes, being primarily herdsmen who lived on beef and dairy products, fought one another and outsiders for the sake of cattle… Other animals such as horses, goats and sheep were also prized, particularly horses, which may have been mainly in possession of princes, tribal chiefs and elders. The spoils may also have consisted of the personal effects of the defeated parties, e.g., the dresses, weapons, etc. Land and crops did not form the bone of contention. Women, who are rightly called the producers of producers in a tribal context, were of course an important object for which wars were fought.(30) In Rg Vedic period, the domestic slaves were mostly women. Slaves formed a part of the property. Nevertheless, chattel slavery did not develop in India. As we have stated earlier, this might be because the pastoral society did not need a large labour force.


How did the appropriation and distribution of surplus take place in Rg Vedic society? Who had the ownership rights? What was the social organization of the Rg Vedic society? The appropriation of wealth takes place in two ways. War spoils was one of the main sources of appropriating wealth. Rg Vedic tribes fought each other for cattle. lndra was prayed to retrieve and recover the cattle of his patron from adversaries. We find 27. frequent references like ‘protectors of the good! You (two) killed Arya foes and dasa foes’ (Rg. VI.60.6), ‘Oh lndra and Varuna! You killed dasa foes and also Aryan foes and helped Sudas with your protection’ (Rg. VII.8 3.1).(31)

The spoils belonged to the tribal members but the redistribution was not necessarily equal. Another way of appropriating wealth was through the presentations or gifts given by the tribal members to the chief. We have discussed how this gift economy had come into existence in the introductory chapter. The chief and the warrior groups as the fighters and winners of the war retained the major part of the booty. Priests also claimed a substantial part of the booty, for it is they who performed the sacrifices for success in the war. The remaining part was redistributed to the common people at a sacrificial ceremony. One of the main duties of the chief is giving gift (dana) which indicates the redistributive function of the chief. The surplus, which was collected as war booty and in the form of gifts and presentations to the chief were similarly redistributed to the warrior groups and brahmanas. These sacrifices were used to reinforce the dominant position of the chiefs, warriors and the priests.(32) Thapar suggests that the Vedic jana (tribe) consisted of a number of vis (clans). These clans in the beginning may have been more egalitarian but by the time of Rg Veda they were bifurcated into the vis and the rajanya. The rajanya constituted of ruling families or senior lineages and it was from this group that the raja was chosen. Thus, the Vedic society was bifurcated into senior lineages, from which the raja hailed; and the junior lineages that worked on the lands settled by raja and gave presentations to rajanyas, who then redistributed them among a limited group consisting of the rajanyas, brahmanas and bards, and spent part of the presentations at the Yajna rituals.(33)


Thapar considers vis as a clan. She refutes Sharma’s connotation of vis as a peasant.(34) As we have already mentioned earlier, the word vis refers to people or group of people’.(35) We cannot translate the word vis as a peasant because in the Rg Vedic society agriculture has not developed to the extent where peasant becomes a separate and independent identity. It is quite possible that in some tribal societies, some lineages or clans become dominant and control the process of production as well as administrative functions of the tribe. But the Rg Vedic society has not developed on these lines. We have seen that Aryans, before they came to India, were divided broadly into brahmanas, kshatriyas and vis (commoners not clans or peasants).(36) But according to Thapar’s account the Vedic jana (tribe) is bifurcated into rajanya and vis. She suggests that branmana-sudra dichotomy, which has its origins in the Harappan Culture, is fused into the dual division of Kshatriya and vis, thus forming the four Varnas.(37) If one accepts Thapar’s arguments that the Rg Vedic society is a lineage society then it follows that brahmanas, kshatriyas and vis represent different lineages, whether senior or junior. But the evidence we gather from Rg Veda shows that even though the society is broadly divided into these groups, there is no closed formation of these groups and every member of the tribe could take up any of the specializations, provided he has the skills. In fact, in many cases, brahmanas and kshatriyas come from the same family. It is true that in the later period these groups, i.e., the brahmanas and kshatriyas, became closed groups. Entry into these groups became restricted. This did not mean that in the beginning they were so. The differentiation within the Aryan tribes during the Rg Vedic period took place on the line of specialized functions like priests (brahmanas) and warriors (kshatriyas) rather than on the lines of lineages. Kinship might have played the same role in the social organization of the Rg Vedic society, but certainly not into arrangement of groups, senior and junior lineages, as Thapar suggests. Because of lack of historical material, it is very difficult to talk about the social organization of feted Aryan tribes during Rg Vedic period with certainty. Nevertheless, with whatever minimum references we have to social organization in Rg Veda we can construct a fair picture of the social organization. This is very important because our understanding of the social organization of the Aryan tribes in this period is going to affect not only the characterization of Rg Vedic society but also post-Vedic societies. Two facts are important to our understanding of the social organization of Aryan tribes in the early Rg Vedic period. First, the functional groups brahmana and kshatriya were not closed groups; any person from the tribe could become either a brahmana or a kshatriya. Secondly, clan as a unit owned the cattle or other property. This is evident from the etymological meaning of the word gotra — a “cowpen,” which later came to be known as a clan.(38) Aryan tribes consisted of clans, gotras, and owned cattle and other property collectively. Some members of the clan had become priests who conducted sacrifices for the success in war and for the welfare of the tribe. When a tribe was at war with another tribe, all the able and valorous persons participated in the war. Initially, the booty of war might have been distributed equally among all the clans of the tribe. But in the later period, the booty of the war was distributed differentially according to individuals/‘groups’ participation in the war. Thus, the warriors got the major share. Next came the priests who performed sacrifices for the success in war. The shares that each warrior and priest got from the booty belonged to his clan. All the members of the clan had rights over it, or in other words, it became the collective property of the clan.(39) As the society developed and the functions of the priest and the warrior became specialized, the priest had to undergo the training of how to perform sacrifices and the warrior had to master the craft of fighting. These groups became distinct (but not closed) and the old arrangement of clan ownership, when only few members of the clan became either priests or warriors, was not in the interest of these specialized groups. This led to the breakdown of the clan ownership and the rise of individual family ownership. This process started by the end of the Rg Vedic period but it had clearly emerged only in the later-Vedic period. This is evident because: (a) even though gotra has remained as an exogamous clan, it no longer implied a cowpen, thus suggesting that the collective ownership of cattle by the clan had declined, and (b) we have references to poor brahmanas in the late Rg Vedic period. This is significant because, if the collective ownership had continued, as a member of the clan, the brahmana would have had some property. Thus, at the end of the Rg Vedic period, collective ownership by clan was giving way to individual family ownership and the brahmanas and kshatriyas were becoming independent groups. This breakdown of collective ownership by clan was facilitated with the increase of settled agriculture. Had the society remained semi-nomadic and pastoral, the individual family ownership would have become difficult, because a single family, without slaves, cannot look after the huge herds. With the development of agriculture, specialized occupations like carpentry, pottery, etc., have also developed. The development of brahmanas and kshatriyas as specialized groups corresponds to this period.


The Rg Vedic society is characterized by rank and stratified society instead of a class society.(40) The absence of surplus in a pastoral economy, tribal society did not create conditions for class differentiation. There could be differentiation of rank, as can be inferred from the titles of tribal chiefs such as jansya, gopa, vispati, visampati, ganasya raja, gananatnganapati, gramani and probably grhopati. Certain vipras were considered worthy of attending the sabha (sabheya), but the phenomenon of the upper classes living on the labour of tribesmen was just beginning to emerge; it did not prevail to any considerable degree.(41) It is implicitly assumed here that the production of surplus was possible only in an agrarian society and classes could emerge only in a society where state and agriculture have developed. However, even in the Rg Vedic period, we find the existence of inequalities, the unequal redistribution of booty and presentations, and the fact that kshatriyas and brahamans were accumulating more wealth than others. These factors/conditions were not considered by scholars like Sharma, Thapar and others as necessary conditions of a class society. They fail to consider the fact that spoils of war indicates the extraction of surplus from the defeated groups, who are subjugated and forced to look for alternative means of subsistence. It is this factor, which has played an important role in the transformation of the pastoral society into an agrarian one. Those tribes who lost their cattle and pasture lands in the wars had to look for alternative means of subsistence. These defeated tribes migrated towards the east and slowly took up agriculture. On the other hand, internal accumulation of wealth by some groups among the victors forced other powerless groups among them to conditions of servitude. These people, like the defeated tribes, were also forced to look for alternative means of subsistence. Thus, a set of favourable conditions for the development of agriculture was set forth by the end of the Rg Vedic period.

(To be continued in part 3)

Leave a Reply

Your email address will not be published. Required fields are marked *