Misrepresenting caste and race

back to issue

ON 26 February 2007, the United Nations Committee on Elimination of Racial Discrimination (CERD) in charge of the International Convention on Elimination of all Forms of Racial Discrimination, held a meeting with a delegation from the Government of India (GOI). India had signed and ratified the convention in 1969 but has not yet given accession and succession. According to Article 1 of the Convention, the term ‘racial discrimination’ meant ‘any distinction, exclusion, restriction or preference based on race, colour, descent, or national or ethnic origin which has the purpose or effect of nullifying or impairing the recognition, enjoyment or exercise, on an equal footing, of human rights and fundamental freedoms in the political, economic, social, cultural or any other field of public life.’ The stand of the GOI has been that while it is committed to eliminating discrimination in all forms, it did not consider caste as part of ‘racial discrimination’. Two key claims of the GOI are that, ‘caste is not race’, and that ‘caste is not based on descent’.

The GOI delegation included the Solicitor General of India, the Permanent Representative of India to the UN, and anthropologist Dipankar Gupta of Jawaharlal Nehru University who argued against viewing caste as race. By framing the discussion as ‘caste is not race’, the GOI constructs a straw argument since no one in CERD claimed that ‘caste is race’, or even that the caste system was racial in origin. The real question is ‘How similar are the discriminations based upon caste and race?’ In reiterating the straw argument, Professor Gupta makes four claims that misrepresent race and caste, and falsely imbues the GOI position with a scholarly basis.

First, Dipankar Gupta argues that caste cannot be equated with race since ‘there is no phenotypical resemblance between members of the same castes.’ Such a view misrecognizes ‘race’ since it assumes that members of the same ‘race’ share phenotypical resemblance. However, it is now established among professional anthropologists and biologists that there is no concordance among human ‘races’, which implies that there is no single phenotypical trait that distinguishes all members of a so-called ‘race’ from members of another ‘race’. Members of any ‘race’ do not share either skin colour, or eye colour, or hair texture, or facial structure. Scientists agree that there is no abrupt change from one skin colour to another, and instead use the notion of ‘clines’ as opposed to ‘race’ to capture human variation. As Professor Alan Goodman, a biological anthropologist says, ‘Race is not based on biology, but race is rather an idea that we ascribe to biology.’ This is why the term ‘race’ is within quotes to signify that it is a socially constructed category, instituted by law and socially reproduced by popular prejudices. Professor Gupta’s position seems hopelessly outdated.

Second, Professor Gupta makes the startling claim that ‘caste is not about descent’ and hence cannot fall under Article 1. According to him, ‘descent means genealogical demonstrable characteristics,’ and ‘in the caste order people came of multiple descents. In fact, in the caste system people had to marry outside their lineage within the caste.’ The term ‘descent’ is at the core of anthropological studies of kinship, and is not restricted to ‘lineage’. A lineage is only one kind of descent group (the smallest) in which ancestry can be demonstrable since it spans living memory of a few generations. However, descent also includes other larger groups such as ‘clans’ (gotras in India) and ‘phratries’ in which a claim to a common ancestry is made but cannot be demonstrated. Marrying outside of one’s own lineage and clan are common practices in India and elsewhere, but caste is really about marrying within a group, as Gupta admits above. Ambedkar famously wrote in 1916 that the ‘superimposition of endogamy on exogamy produces caste.’ Castes are simply ‘large-scale descent groups’ as many anthropologists have pointed out. Castes are larger than clans and hence are very much based on ‘claimed’ ancestry, usually a mythical ancestor appearing in origin stories. Indeed, Gupta’s own work (Interrogating Caste) demonstrates this widespread existence of castes claiming a remote ancestor. Professor Gupta’s position is not substantiated logically, conceptually or empirically in scholarship.

Third, Professor Gupta cavalierly claims that ‘each caste equally discriminated against other castes.’ While this still leaves one to wonder how this distinguishes caste from race, Gupta’s position neglects decades of scholarship that has distinguished between institutional casteism or racism based upon power and individual or group prejudice. While it is quite feasible to argue that in a casteist (or racist) society, everyone can be prejudiced, it is simply not true that everyone’s prejudice has equal impact. Would Professor Gupta equate the daily humiliations, lynching, and rapes of dalits by all castes who wield power over them, with the presumed prejudice that dalits might hold against other castes? Discrimination requires attention to institutions, and not only subjective notions which Professor Gupta focuses on in his testimony. Thus, his evidence that ‘no caste accepted the notion that they were inferior’ is quite irrelevant since there are too many castes who not only think they are ‘superior’, but actually have the power to act upon their prejudice in systematic and violent ways. In a casteist society that stigmatizes particular castes and privileges others, the latter are raised to think that the resources of the country belong to them as a birthright and are willing to act violently to protect it.

Finally, Professor Gupta empties caste of all power (and discrimination) by portraying caste as a matter of cultural traditions claiming that ‘people in the caste system were proud of who they were and their traditions and position in the country.’ It is almost as if he is being far too accommodative of those ‘upper castes’ whose ‘caste pride’ and ‘position in the country’ is based on the humiliation of other castes. His position that ‘caste members did not want to escape their caste’ also makes a mockery of historical attempts by individuals from stigmatized castes who prefer to ‘hide their caste origins’ in the face of contempt of so-called ‘upper’ castes. It also mocks groups who have claimed new identities over time by leaving Hinduism altogether (for example, neo-Buddhists, Christians, and others). For all those who Professor Gupta sees as revelling in ‘caste pride’, there are many more who are weary of caste identities, and resist its inscription upon their bodies. In denying this, he also denies the patriarchal nature of caste.

Balmurli Natrajan


Why caste discrimination is not racial discrimination

IN my view the allegation that caste is a form of racial discrimination is not just an academic misjudgment but has unfortunate policy consequences as well. It is for that reason that I felt it was important to set the record straight even though there is a kind of seductive charm in finding easy parallels between caste and race. But unless we can see beyond these superficialities, the cause of combating caste can be prejudiced, and the clock turned back on the advances made so far.*

To begin with, both caste and race are social constructs. That there is something physically demonstrable, even phenotypically, has long been disproved. Even the myth of the fair skin Aryans charging down the mountains in waves to crush the dark Dravidians is now in disrepute. For a long time many scholars were votaries of this Aryan invasion view even though the evidence for it was always scanty and dubious.

However, the similarities between segregation under racial apartheid and untouchability in Hindu India prompted many certified specialists to wonder whether casteism could be seen as another form of racial prejudice. In my earlier work on this subject I had shown that caste identities and prejudices are manifested differently from racial ones, and I will go over the ground once again very quickly.

Caste identities get stronger the more local one gets. In other words, nobody is a Brahman or a Kshatriya or a Vaishya. They are either Kanyakubj Brahmans, or Rarhi Brahmans, or Rajputs, or Jats, or Agarwals or Guptas or Soods. These too are pretty broad categories. Identities that operate on the ground usually have a very limited range, sometimes no more that 200 miles. I have recently come across an instance when Koeris in Jaunpur district of East UP did not know about the existence of the Kurmi caste who were in large numbers barely 60 kilometres away.

Race identities, on the other hand, do best when the sweep is a wide one. It does not matter in apartheid societies whether a ‘white’ person comes from Holland, England or Germany. Such an individual would be accorded a superior status against those considered to be ‘black’, regardless of which part of the world they come from.

Further, even in antebellum America a black could be a nanny or a cook. In a traditional Hindu setting it would be unthinkable for a person of a supposedly ‘low caste’ occupying such a position in the home of a privileged person. In fact, commensal restrictions were so strict in the past that many castes refused to take food even from Brahmans. I have documented all of these instances in my book Interrogating Caste (Penguin: Delhi, 2000). In fact, most of these facts are not new and have been known for some time, except that few paid them the attention they deserve.

Racial prejudice does not make exceptions for those who are children of mixed marriages. In the United States, till as late as the early 1960s, the one-drop rule prevailed in the designation of a person as ‘black’. According to this principle, even 1/64th black blood would disqualify individuals from being considered as ‘whites’. This prompted Gene Lees to ask in the Jazz Newsletter if black blood was so strong that it could neutralize generations of white breeding. In inter-caste marriages the child does not carry the parents’ caste identity in equal proportions, but belongs to a totally different category – the outcaste. In fact, Yagnavalkyasmriti justifies the abhorrence it advocates against untouchables by claiming that they are children of mixed caste unions.

Members of so-called low caste communities do not share these upper caste textual views at all. In my work again, I have elaborated the origin tales of low castes and shown how every one of them claims an exalted status that is equal to, if not better than, the best. This is how it is everywhere in the world. There is no community that admits that it is essentially ‘bad’ or ‘impure’. The belief that low caste people participate in their own subjugation and acquiesce to their reviled status is a position that only certain Brahmanical texts recommend but stands refuted on the ground.

The question then is: why do certain castes function under such degrading and humiliating circumstances? The answer is a simple one, but it eluded us as the exotic aspects of caste so overwhelmed our senses. Caste hierarchies were maintained not because those in the Hindu fold agreed unanimously on the hierarchy, but largely because the stratification on the ground was upheld by the power and wealth of village oligarchs who functioned best in a closed natural economy. Once this rural economy began to crumble, castes that were hitherto seen as ‘low’, or even ‘defiling’, stood up and claimed a higher status, but without giving up caste.

This is where ‘caste patriotism’ comes in. Look at the matrimonial columns of any leading Indian daily. Marriage preferences are listed caste wise and this is true for scheduled castes as well. There need be no puzzlement on this especially if we keep in mind the fact that while others may consider a caste to be low, members of that community never bought into that view.

An appreciation of this fact brings into view another significant difference between caste and race. In race societies there is the widely acknowledged phenomenon of ‘passing’. This might seem reprehensible to many black intellectuals, but it has been written about in great detail by a large number of American scholars. Blacks would on occasions strive to ‘pass off’ as whites. It is also true that light skinned black people get a better deal even in contemporary western societies than dark skinned black people. In India, on the other hand, we have come across tragic instances, in recent times, when families conspired to kill their own children who dared to marry outside their caste. This outrage is not limited to so-called ‘upper castes’ but has occurred among the scheduled castes too.

This is why in caste societies, no matter how low a particular community may have been considered once, the fight today is not to deny one’s background, or to ‘pass off’ and merge with a dominant group, but to claim that this background had always been misunderstood, misjudged and misrecognized. Those who were traditional tanners or leather workers, for example, are not aspiring to marry into Brahman families but are pressing for a group elevation of their status. Many claim to be Brahmans of a certain kind, but not of the kind that are around and do not want to merge with them. Several artisan castes, like the Lohar, Patharwat, or Brazier believe they are Vishwakarma Brahmans but will not marry a Saraswat, Chitpawan or Gaud Brahman. This resistance against merging and losing one’s identity in a larger, and more established, formation is equally true of those who claim, or aspire, to be Kshatriya or Vaishya as well. I have already mentioned how scheduled caste people search specifically for matches from similar caste backgrounds, as can be easily gleaned from matrimonial columns in national dailies. Predictably, nobody wants to be a Sudra, and given what we have said so far, there should be nothing surprising about that!

All of this should alert us to any easy equation between caste and race, or between caste discrimination and racial discrimination, or even between caste identity and race identity. Descent and race may have some connection as the child supposedly partakes equally of the racial traits of both parents. The situation changes when we turn to mixed caste marriages as the child belongs to the caste of neither parent. Yet in terms of descent, the person is still a descendant of both parents. Thus while the child may, as in this case, have no caste, yet s/he remains a member of the descent group and, probably, also of the corporate group. This truth should not be lost sight of.

I should at this point make clear that descent is not the same as clan, or gotra. Descent kicks in only when the genealogical ties are demonstrable. The relationship between members of the same clan cannot be genealogically demonstrated. They are putative but not real ties. Hence jati origin tales cannot demonstrably link people to the same ancestor and, for the same reason, the Hindu gotra does not become a descent group. To take the point further, may I also remind my readers that no caste, including the subaltern castes, would accept a marriage within the same gotra. In fact, some of the most heinous instances of killing one’s own kind have happened because the young couple belonged to the same gotra and this could not be tolerated by families on both sides.

As was mentioned earlier, caste identities are fine-grained, multi-faceted, and extremely local in their realization. This is what prompted a famous scholar to comment that in caste there is an obsession with ‘minor differences’. It is not as if Brahmans or Kshatriyas, howsoever loosely conceived, are on one side and the lower castes on the other. This kind of divide would mimic a racial division. Caste intolerance and prejudice goes down the line. The better off and the more powerful one is the greater is the possibility of actually exercising this prejudice socially. The so-called ‘backwards’ have been some of the worst perpetrators of caste atrocities. Swami Achyutanand importuned the British authorities in the late 1920s to protect the ‘untouchables’ from these backward classes, who were often worse than the Brahmans. Thevar caste prejudice is feared in Tamilnadu as much as Gujar or Kurmi arrogance is in Uttar Pradesh. Further, as I.P. Desai had once noted, there was ‘untouchability among untouchables’ as well.

The fight against casteism should then be conducted in a fashion that is different from the way the struggle against racism is waged. This is why from the very beginning, under the stewardship of Dr. Ambedkar, the policy of independent India was to extirpate casteism, and it was felt that ‘reservation’ would aid this process. Dominant forms of ‘affirmative action’ in America, on the other hand, seek primarily to represent different races and colours but not to wipe out race. In India we thought on a much bigger scale and our reservation programme was set to eradicate caste and not stop short in terms of representation. Unfortunately, there are many today who value representation highly, without realizing how far they are moving away from the vision of Dr. Ambedkar.

Our struggle against caste prejudices and discrimination is far from over. Let us not muddy the waters further and make our job infinitely more difficult by making facile linkages between caste and race. As I have tried to show, any lowering of our intellectual and political guard on this issue will not only obfuscate matters but might encourage inappropriate policy interventions.

Finally, I would like to end with a quote from J.B.S. Haldane whose reflections on race are very instructive to all those who seek to extend the scope of racism. According to Haldane:

‘As for the word race, it has so many different meanings, as to be useless in scientific discussion, to very useful for getting members of the same nation to hate one another (emphasis added).’

Dipankar Gupta


* This is a summary of my speech at the CERD conference in Geneva on 26/2/07 as the proceedings were recorded. For more details contact UN Human Rights Commission, Geneva.