35 | Caste and Race Science: Hutton, Guha and 1931 Census of India

By Gaurav Kalyani, Nishitha Mandava and Shivakumar Jolad

Published on: 31 August 2026

Introduction

The 1931 census produced voluminous work on caste and then prevalent racial ethnography. Then chapter in the main census report written by Census Commissioner J. H. Hutton was supplemented by a volume called “The Racial Affinities of the Peoples of India” by B. S. Guha, an anthropologist working for the Zoological Survey of India. Together, these two documents mark both the high point and the beginning of the unravelling of the colonial project, begun by Herbert Risley in 1901, of explaining Indian caste through race science. 

Hutton’s Commentary on Origins of Caste 

Hutton at length reviews the various theories that explain the existence of the caste system. The first explanation considered is the traditional account associated with the Code of Manu. According to this view, society was originally divided into four varnas, and the large number of castes subsequently arose through different kinds of marriages between these groups. Restrictions on marriage, food, and occupation were central to maintaining these divisions. The author rejects this as an explanation of the actual origins of caste, arguing that the four-varna system described in Manu is a later, artificial systematisation rather than an account of how caste originally developed.

The second theory was advanced by Nesfield and Dahlmann, which held that caste developed along the lines of occupation and occupational guilds, which we have discussed in our previous essay. The third was put forth by Ibbeston, who suggested that tribal organisation, occupational guilds and religious institutions were features of castes rather than their origins. 

The fourth explanation, associated with Senart, traces caste to the gens or family organisation and family worship. The author rejects this too. He distinguishes caste from gotra: a caste does not generally claim descent from a single common ancestor, whereas gotra does involve claims of common ancestry. Moreover, several gotras can exist within the same caste, which, in his view, makes it difficult to argue that caste originated simply from family organisation or family worship (Hutton, 1933).

The fifth explanation is Risley’s theory of race and hypergamy. The author argues that this cannot adequately explain the restrictions on food and marriage that are central to caste. He also rejects the idea that caste and race correspond neatly. At the same time, however, he does not dismiss race altogether. He argues that racial distinctions and prejudice may have played an important role in crystallising and perpetuating caste, even if they did not constitute its original cause (Hutton, 1933).

(Prehistoric skulls used in the 1931 Census's discussion of the racial composition of India: (a) Skull from Adichanallur; (b) Skull from Nal, Baluchistan. Source: Hutton, 1933)

The report uses prehistoric cranial remains from sites including Adichanallur, Nal, Sialkot, Bayana and Mohenjo-daro as material for reconstructing the racial composition and population history of the Indian subcontinent. 

Hutton proceeds to explain his views by drawing on communities that, in his view, had remained relatively little changed by later historical developments. He focuses particularly on communities in the Naga Hills. There he observes villages organised around particular occupations. 

These communities exchanged goods but maintained distinctions between occupations. He also notes an important difference that these occupational divisions generally did not involve prohibitions on intermarriage or commensality. He therefore had to look elsewhere to explain those characteristic features of caste. 

He argued that the underlying beliefs and taboos associated with caste may have very ancient, pre-Aryan origins. When communities practicing different occupations came into contact with one another, existing beliefs about pollution, magical power, food and contact could generate restrictions around particular crafts and groups. These restrictions could gradually become more rigid and develop into divisions between communities. 

Within this theory, the Indo-European/Aryan invasion played an important role, but not as the source of caste. According to his understanding, the basic ingredients for caste, particularly occupational and ritual taboos, already existed before the Indo-European invaders. The invaders subsequently crystallised these pre-existing distinctions into a more rigid social hierarchy, partly through their concern with racial identity and social superiority. 

Hutton proposes that the encounter between patrilineal Indo-European groups and indigenous groups that he characterised as matrilineal, could have produced particular forms of social hierarchy and marriage restrictions. He then connected this argument to practices such as purdah, suggesting that restrictions on women's movement could have emerged partly from the need to control relations between these different social and kinship systems.

In short, Hutton's argument was essentially that caste did not originate at one identifiable moment or from one source. He rejected explanations that reduce it to the four varnas, occupation, tribe, family, or race alone. Instead, he proposed a layered history in which ancient occupational and ritual taboos and hierarchies were later crystallised through the Indo-European migrations. 

It is worth noting here that Provincial census reports, published as separate volumes for each administrative unit, also often carried their own theoretical excursions on caste beyond the tables required of them. An essay in the United Provinces report draws on the newly excavated Indus Valley cities as evidence of a continuous, sophisticated urban civilisation underlying the province's social order. 

Citing contemporary anthropological functionalism, quoting a Royal Commission's finding that “religion, law, tribal authority, tribal customs and the economic life are inseparably bound up together,” the author defends caste as an integrated, internally coherent “polity” rather than a set of oppressive rules (Census of India, 1931, United Provinces of Agra and Oudh: 541–542). Interestingly, this was an anonymous essay written by, in its own words, “an old friend of India who prefers to remain anonymous.” 

Such unofficial, signed-but-anonymous essays sat uneasily beside the statistical apparatus of the census proper, which was presented as a neutral demographic exercise. 

Anthropometric Survey by Dr. B.S Guha 

The 1931 census results were published across several volumes. Among them, two are essential to the study of caste and ethnography. The first was Chapter XII, “Caste, Tribe and Race,” of the General Report (Census of India 1931, Volume I – India, Part I), written by Hutton himself, consisting of appendices more than hundred pages (Hutton, 1933). 

The second was a standalone ethnographic supplement, Census of India 1931, Volume I, Part III—Ethnographical, published in Simla in 1935, two years after the main report because of the extensive work involved. It was divided into two separate sections. Part A, “The Racial Affinities of the Peoples of India,” was written entirely by B. S. Guha. Part B, “Ethnographic Notes by Various Authors,” brought together contributions from different provinces and was edited by Hutton (Hutton and Guha, 1935). 

Part A 

To establish an empirical foundation for physical anthropology in India, the Census Department commissioned Dr. B.S. Guha of the Zoological Survey of India to conduct a comprehensive anthropometric survey during the 1931 Census. Guha was the founder Director of the Anthropological Survey of India. 

Following the footsteps of Risley, Guha measured the stature, cephalic index (head shape), nasal index (nose shape), and orbito-nasal features. He conducted anthropometric fieldwork across the subcontinent, measuring 3,771 individuals drawn from 51 population groups on eighteen physical characteristics, supplemented by somatoscopic observation of skin, eye, and hair colour, and by the calculation of “coefficients of racial likeness” between groups (Padmanabha, 1978). 

Guha’s analysis identified six primary racial elements in the Indian population:

1. The Negrito Element: Represented by woolly hair, short stature, and broad noses, found in vestigial pockets among the Kadar and Pulayan of Southern India, as well as the native Andamanese.

2. The Proto-Australoid Element: Characterised by dark skin, long heads (dolichocephalic), wavy/curly hair, and broad, flat noses. This element formed the foundational genetic substrate of the tribal populations of Central and Southern India (e.g., Santhal, Bhil, Gond, Chenchu).

3. The Mediterranean Element: Characterised by medium stature, dark olive skin, and long heads with narrow noses. This element constituted the primary demographic base of the agricultural and urban populations across Northern and Peninsular India.

4. The Alpine, Dinaric, and Armenoid Elements: Characterised by broad heads (brachycephalic). This element was prominently traced along the western and eastern belts of India, stretching through Gujarat, Maharashtra, Coorg, Bengal, and parts of Orissa.

5. The Indo-Aryan / Nordic Element: Characterised by tall stature, fair skin, long heads, and narrow, prominent noses. This element was concentrated among the upper castes of North-Western India, particularly in the Punjab, Kashmir, and Rajputana.

6. The Mongoloid Element: Characterised by epicanthic eye folds, flat facial profiles, and yellowish skin tones, dominant along the Himalayan belt, Assam, Indigenous Chittagong, and Burma. 

Hutton, in his preface, called Guha's survey “easily the most important contribution to the physical anthropology of India since Sir Herbert Risley's” (Hutton and Guha, 1935). Guha would go on, a decade later, to become the first Director of the newly founded Anthropological Survey of India in 1945, making the 1931 ethnographic survey the direct institutional ancestor of that body (Padmanabha, 1978).

Part B

Unlike Part A, Part B is less a single structured argument than a province-by-province anthology of ethnographic notes on individual castes and tribes. It was contributed by census superintendents, missionaries, and administrator-ethnographers, many of them reworked from material already published in provincial census volumes. 

Contributions were organised loosely around a standard questionnaire, covering physical type, social organisation, marriage and kinship rules, religious practice, and occupation, printed as an appendix to the volume. Hutton circulated it to correspondents across British India and the princely states (Hutton and Guha, 1935). 

The resulting work is a patchwork of sorts consisting of observations, descriptions, and local accounts. It comprises notes on the Angami and Sema Nagas (drawing on Hutton's own earlier monographs), on the primitive tribes of Travancore, on the Baiga of the Central Provinces, and dozens of others, of uneven length and rigour. Collectively they form one of the most comprehensive ethnographic surveys of India's tribal and low-caste populations attempted for decades after (Hutton and Guha, 1935, Part B).

It is worth noting that this was only the second such dedicated ethnographic volume the Indian census had produced, the first being Risley's own 1901 supplement, and it would also prove to be the last (Hutton and Guha, 1935). No subsequent census, colonial or independent, has commissioned a comparable nationwide physical-anthropological survey, in part because the racial paradigm underlying Guha's work fell into disrepute after 1945, and in part because independent India had little appetite for state-sponsored racial classification of its citizens.

Modern Critiques and Political implications

Later historians and anthropologists have been considerably less charitable toward the 1931 exercise than its own officers were. Historians of science have shown that caste became a “racial laboratory” almost by accident of timing. European ideas about the Aryan invasion, developed by philologists such as William Jones and Max Müller from the discovery of common roots between Sanskrit and European languages, gave colonial administrators a ready-made story in which lighter-skinned Aryan invaders had subordinated darker Dravidian populations and crystallised the result into varna and jati. Risley supplemented it with “biological” proof by ranking castes according to the nasal index, treating a low index as evidence of high social rank and Aryan ancestry (Weaver 2022).

Bernard Cohn in his forceful critique argued that colonial census operations along with describing the Indian society also “objectified” it, converting fluid, relational, locally negotiated identities into fixed and comparable units (Cohn, 1987). His argument finds ample support in Hutton's own report, whose pages record officers wearily fielding a “pestiferous deluge” of status claims that the census itself had incentivised (Hutton, 1933).

Guha's racial classification, meanwhile, enjoyed a long institutional afterlife within Indian physical anthropology through the Anthropological Survey of India, even as the biological race concept underpinning it was steadily abandoned by anthropology worldwide after the Second World War (Padmanabha, 1978).

Modern genetics has also dismantled the empirical foundation of the entire race-caste correlation. Ancient DNA studies published in the last decade show that the population history of the subcontinent is one of complex, layered migration and mixture stretching back many thousands of years, producing a north-south genetic gradient rather than a set of discrete races. 

Steppe pastoralist ancestry associated with Indo-European languages did spread into the north-west after the decline of the Indus Valley cities, but the resulting pattern is clinal, not racial, and it does not map neatly onto the sharp lines that Risley and, to a lesser extent, Guha tried to draw between castes. As one recent review put it, the Aryan invasion theory has “never been substantiated despite centuries of debate,” yet its vocabulary of Aryan, Dravidian and Mongoloid “races” persists in popular and even some scientific discourse in India today (Weaver 2022). 

The consequences of this history are not confined to academic debate. Historians attribute the caste censuses as one of the chief drivers for the caste-based political mobilisation in the twentieth century (Fuller, 2016). Hutton himself noticed this happening in real time: he observed in 1933 that many caste groups were consolidating into broader “portfolio” units, such as various artisan castes combining under a single label, partly to raise social status and partly because political strength under the new, more democratic institutions rewarded larger, more organised blocs (Hutton 1933b). 

The racial framing of caste also outlived colonial rule in less benign ways. Independent India's leading statistical and anthropological institutions, founded by figures who had worked within the Risley-Guha tradition, continued to use race categories in population research for decades after 1947, even as B. R. Ambedkar's constitution formally outlawed caste discrimination (Weaver 2022). 

The old three-way Aryan-Dravidian-Mongoloid scheme has also fed a distinct kind of prejudice against people from India's north-east, who are still sometimes racially othered (Weaver 2022). Internationally, the same history shaped debates about whether caste should be treated as a form of racial discrimination. Dalit activists drew explicit parallels with the Black civil rights movement and pressed the issue at the 2001 UN World Conference against Racism, only for the Indian government to insist officially that caste and race were unrelated categories with no bearing on international law (Weaver 2022).

Ninety-five years on, as India returns to caste enumeration in its 2027 census, planners are once again reaching for the 1931 list as a starting point, a reminder that colonial categories, however contested at their origin, rarely simply disappear; they are inherited, contested afresh, and pressed into new political service.

(Authors: Gaurav Kalyani works as Research Associate for the India State Stories Project at the Center for Legislative Education and Research, FLAME University, Pune;

Nishitha Mandava, is an independent researcher and research consultant for Center for Legislative Education and Research, FLAME University, Pune;

Shivakumar Jolad works as Associate Professor (Public Policy), and is the Chair of Center for Legislative Education and Research and Director India State Stories, FLAME University, Pune)

(Author Contributions: Gaurav and Nishitha did research, primary writing and editing. Shivakumar contributed to conceptualization, research and editing)

Bibliography

Fuller, C. J. (2016). Colonial anthropology and the decline of the Raj: Caste, religion and political change in India in the early twentieth century. Journal of the Royal Asiatic Society, 26(3), 463–486.

Hutton, J. H. (1933). Census of India 1931: Volume I — India, Part I — Report (Chapter XII: “Caste, Tribe and Race” & Appendix I: “Exterior Castes”, pp. 425–476). Government of India Press.

Hutton, J. H., & Guha, B. S. (1935). Census of India 1931: Volume I — India, Part III — Ethnographical (Part A: “The Racial Affinities of the Peoples of India” by B. S. Guha). Government of India Press.

Weaver, L. J. (2022). The laboratory of scientific racism: India and the origins of anthropology. Annual Review of Anthropology, 51, 67–83.