Review Society

Comparing Race and Caste?
Revisiting Louis Dumont’s Sociology of India

by Roland Lardinois , 25 May 2012

The work of anthropologist Louis Dumont, who would have been 100 years old on the 1st August 2011, remains hotly debated (see a recent analysis of his political thought on this site). The latest example is by American anthropologist Kamala Visweswaran, who reconsiders the notions of culture, race and caste in social sciences and addresses core issues in the sociology of India.

Reviewed: Kamala Visweswaran, Un/common Cultures. Racism and the Rearticulation of Cultural Difference, Durham, Duke University Press, 2010, New Delhi, Navayana, 2011, 341 p.

Kamala Visweswaran’s articles gathered in this book deal with one main question: how are racial ideologies, which have been totally disqualified for both political and scientific reasons after World War II, rearticulated in the beginning of the 21st century under the notion of immeasurable cultural differences between Western societies and the rest of the world? Anthropology is at the core of these debates because this discipline, since the pioneering work of Franz Boas, appropriated the description of differences in the name of cultural relativism that viewed itself as an antidote to racism. But today, writes the author, “culture has stood in for race or as a form of negative ideology.” [1]

To understand this shift, Kamala Visweswaran questions the history of anthropology by considering displacements and modes of operation of these notions among major authors whose works have left their imprint on the discipline. She develops her analysis in three domains, gender, caste and the “negro” question, and privileged the study of India. The main reason is that the last two themes are the subject of an article written by Louis Dumont “Caste, racism and stratification,” [2] first published in French in 1960, which Kamala Visweswaran situates in the history of African American sociology and the sociology of India. It is mainly these developments (chapters 2 to 5) that we discuss here.

Are There Castes Outside of India?

Louis Dumont’s article that deals with general sociology has a comparative aim. The issue is whether caste, as a social morphological characteristic, is peculiar to the Indian society or if there are castes outside India. Numerous American sociologists, known under the name “Caste school of race relations”, upheld the view that African Americans formed a sort of caste in North American society, according to the principle of color bar that is characterized by endogamy, spatial segregation of occupations and the hierarchy based on race between Whites and African Americans.

Dumont criticized these authors for one main reason. They authors selected some morphological characteristics of caste that they transposed in American society, ignoring the fact that these elements cannot be isolated from the system that gives them meaning. Certain aspects of caste can be compared with relations that link Whites and African Americans in the United States, but one cannot speak of castes in the sense of the caste system of India.

Furthermore, Dumont maintains, the theory of hierarchy gives meaning to a sociological characteristic of race relations in the United States which has been overlooked by most of anthropologists. Dumont emphasizes three main points. Firstly, “racial discrimination succeeded the slavery of the Negro people once the latter was abolished.” [3]

Secondly, “the distinction between master and slave was succeeded by discrimination by White against Black.” Finally, “the essence [my emphasis] of the distinction [between master and slave] was juridical; by suppressing it, the transformation of its racial attribute into a racist substance was encouraged.” Dumont specifies: “[...] racism fulfils an old function under a new form. It is as if it were representing in an egalitarian society a resurgence of what was differently and more directly and naturally expressed in a hierarchical society. Make distinction illegitimate, and you get discrimination; suppress the former modes of distinction and you have a racist ideology.” [4]

The analysis seemed ambiguous to many readers, and twenty years later Dumont had to clarify it: “I never said, as some seemed to believe, that hierarchy is better than equality, or in the present instance that slavery is preferable to racism.” [5] The fact remains that Dumont’s three points are debatable as much on the historical level as on the sociological one. First of all, racism takes various forms that vary in time and it cannot be reduced to a “modern phenomenon.” Then, in the United States, the opposition between White and Black pre-exists the abolition of slavery. Finally, the juridical character of the master slave relationship is nothing but “the essence of the system.” As Kamala Visweswaran highlights, the system of slavery is over-determined by a racial ideology and the hierarchical white-black distinction cannot be dissociated from the master slave relationship. This is the reason why the legal abolition of slavery does not ipso facto transform the white-black racial distinction. One could extend further Kamala Visweswaran’s analysis and note that Dumont makes, with respect to slavery, the same type of error in reasoning he reproaches American sociologists. He isolates certain characteristics of master-slave relations while ignoring the fact that slavery is an economic, social and political system that cannot be reduced to a hierarchical dimension, as he himself corrected later without modifying his analysis. [6]

A Comparative Sociology of Minorities

This detour by the “negro” question seen against the yardstick of Louis Dumont’s comparative sociology leads Kamala Visweswaran to frontally question the sociology of India. Why do the most eminent works of this discipline not permit a thinking of discrimination based on caste? [7] The author anchors her argument in the debates that surrounded the World Conference Against Racism (WCAR) which took place in Durban in 2001. During this conference, representatives of Dalit groups [8] attempted to have caste discrimination recognized as homologous to race discrimination, and forged the notion of “casteism” to designate the modus operandi of caste discrimination as similar to that of racism. The Indian government, headed then by the rightwing nationalist Hindu party, the Bharatya Janata Party, firmly opposed this attempt, which also generated a lively resistance from Indian sociologists situated more on the left of the political field. Contrary to the opinion defended by Dalit organizations, the dominant view that emerged from the conference was that race relates to biology and caste to social system, and therefore that casteism was not at all comparable to racism.

Kamala Visweswaran shows that anthropologists since Franz Boas and his successors, like Melville Herskovits, Edward Sapir and Ruth Benedict, but also Claude Lévi-Strauss to whom she devotes a chapter, have rejected the notion of race as non-scientific, a point strongly defended by Ashley Montagu, and abandoned it to the biological sciences. The unforeseen consequence of this move is that anthropologists have left an empty space that the notion of culture comes to occupy, to the point of embracing everything and explaining nothing any more. In the two cases, says Kamala Visweswaran, this notion became “as essentialist and determinist as was the notion of race till now.” [9]

The work thus takes a very interesting turn when Kamala Visweswaran makes unknown links suddenly appear between three figures as different as W.E.B. Du Bois, the first significant African American sociologist, Max Weber who visited the United States, namely the South, in 1904, and finally B.R. Ambedkar [10] who was a student at Colombia University in New York when he presented a paper entitled “Castes in India. Their Mechanism, Genesis and Development,” at the seminar of the anthropologist Alexander Goldenweiser in May 1916. From the beginning of the 19th century, Du Bois imported the notion of caste to illustrate the nature of race relations in the slavery belt of the United States on the one hand; on the other, Weber [11] and Ambedkar read the works of Du Bois with whom they corresponded. These three sociologists have all seen links between the situations of three minority and dominated groups, African Americans, Untouchables but also Jews, with Ambedkar noting that the notion of ghetto can be applied to the segregation of Untouchables in India. [12] Reconsidering these forgotten groundings for another comparative sociology, Kamala Visweswaran states, should help us understand what these groups share in common. The title of the book, which is, at a first glance, enigmatic, thus finds its meaning. Un/common Cultures invites seeing links beween these groups and opposes social science studies that posit differences beyond comparison while relying on “incommensurable” cultural essentialism.

We are often tempted to engage with the author, particularly on Louis Dumont’s work, which she reduces to a one-sided structuralist exercise without questioning the historical social backgrounds in which his work developed. [13] Original analyses devoted to the question of gender in situation of requests for political asylum (chapter 7) could also be put in relation with the criticism that she develops of Dumont, as the notion of hierarchy, here again, casts a veil on the reality of women’s condition in India. [14] Finally, in upholding the cause of a “Subaltern Sociology” (subaltern groups including here Dalits, tribal populations, Muslims and women), Kamala Visweswaran joins a collective project that does nothing but reverse the dominant sociological point of view, which is also the view point of the dominant groups, and remains caught in a binary opposition just as biased as theirs.

Yet, we do not need to agree with the author on all points to follow her stimulating analysis of the sociology of India, certain forms of which she sees as caught up between two stands, “one apparently internationalist in scope, the other nationalist in orientation.” [15] There is food for thought for a practice of social sciences without national borders.

First published in Translated from French by Rohini Karnic with the support of the Centre de Sciences Humaines, New Delhi.

by Roland Lardinois, 25 May 2012

To quote this article :

Roland Lardinois, « Comparing Race and Caste?. Revisiting Louis Dumont’s Sociology of India », Books and Ideas , 25 May 2012. ISSN : 2105-3030. URL :

Nota Bene:

If you want to discuss this essay further, you can send a proposal to the editorial team (redaction at We will get back to you as soon as possible.


[1Kamala Visweswaran, Un/common culture. Racism and the Rearticulation of Cultural Difference, Durham, Duke University Press, 2010, p. 7.

[2Louis Dumont, “Caste, Racism and ‘Stratification.’ Reflections of a Social Anthropologist,” in Homo hierarchicus. The Caste System and Its Implications, New Delhi, Oxford University Press, 1980, Appendix A, p. 247-266.

[3Louis Dumont, “Caste, racism and stratification”, op. cit., p. 262-263 for three points referred to by Kamala Visweswaran.


[5Louis Dumont, “A Comparative Approach to Modern Ideology and to the Place within it of Economic Thought,” in Louis Dumont, From Mandeville to Marx. The Genesis and Triumph of Economic Ideology, The University of Chicago Press, Chicago and London, 1977, p. 12, cited by Kamala Visweswaran. This clarification conceded, Louis Dumont nevertheless affirms: “I must confess my irenic preference for [hierarchy],” in Essays on Individualism. Modern ideology in Anthropological Perspective, The University of Chicago Press, Chicago and London, 1986, p. 267.

[6Louis Dumont’s views had already been criticized by the anthropologist Gerald Berremann, “Race, Caste and Other Invidious Distinctions in Social Stratification”, Race and Class, 1972, 13, p. 385-414.

[7Kamala Visweswaran, Un/common Cultures, op. cit., p. 131-132.

[8The process of labeling the lower castes (Chandala, Harijan, Untouchable, Scheduled Caste and lastly Dalit) has always been a political stake in India. Militant groups always claim to speak in the name of the whole lower castes, trying to unify them at least from a symbolic viewpoint. In spite of having a more militant than descriptive dimension (or, if one prefers, for this very reason), the term Dalit imposed itself not only in the press and the media but also in the academic world. Yet, if people agree on the necessity of upholding economic, social and cultural emancipation of these groups, they diverge on the ways it should be implemented.

[9Kamala Visweswaran, Un/common Cultures, op. cit., p. 63.

[10See Jules Naudet, “Ambedkar and the Critique of Caste Society,” Books and Ideas, 5 November 2010.

[11Kamala Visweswaran notes that Weber’s work on Hinduism makes numerous comparisons between caste and racial groups, in India and the United States; see the entries “United States” and “race” in the index of Max Weber, Hindouisme et Bouddhisme, translated by Isabelle Kalinowski with the collaboration of Roland Lardinois, Paris Champs, Flammarion, 2003.

[12Near Madurai, in the state of Tamil Nadu, the government destroyed a wall that separated the area of the Untouchables from the rest of the village (they were also prohibited from entry the village temple). See The Hindu, 11 November 2011.

[13For a socio-historical analysis of the work of Louis Dumont, see Roland Lardinois, L’invention de l’Inde. Entre ésotérisme et science, Paris, CNRS éditions, 2007 (English translation, Scholars and Prophets. The Genesis of French Sociology of India, New Delhi, Social Science Press, 2012 forthcoming).

[14See Prem Chowdhry, Contentious Marriages, Eloping Couples. Gender, Caste, and Patriarchy in Northern India, New Delhi, Oxford University Press, 2007.

[15Kamala Visweswaran, Un/common Cultures, op. cit., p. 131-132.

Our partners

© - Any replication forbidden without the explicit consent of the editors. - Mentions légales - webdesign : Abel Poucet