Religion, caste and 150 years of census

Salil Misra 

It will not be inaccurate to say that official modern knowledge about Indian society and people began to be systematically built up since the second half of the 19th century. The state created certain categories which soon acquired great power and legitimacy. Those categories and their power and legitimacy have shown great persistence and tenacity over the years. They were created under an alien colonial rule but have also cast their shadow in Independent India. This knowledge about India and Indian people was generated and disseminated through various institutions. The census was an extremely important one among them. On the 150th anniversary of the Indian census, it is apposite to ask the question: how and in what ways did the census create new knowledge on India that has persisted over the years? 

Census, as the project of collecting information on the people by the State, is a modern idea. Pre-modern State systems did not have the resources, the need or the inclination to gather detailed information about the societies they ruled over. Census was not just a way of gathering information; it was also an instrument of control. As with so much else, this project, too, started in modern Europe. The British started their formal census from the beginning of the 19th century. It was driven by a great concern about the number of poor people and the possible decrease in population. The British census showed no desire to collect information on the religious life of the people, or the numbers of followers of religious denominations.  

Quite by contrast, when the British started the census operations in India from the second half of the 19th century, they showed a great obsession with the social profile of the people in terms of religion and caste. Religion was treated as the fundamental organising principle through which people were to find entry into the knowledge registers of the British government. While collecting religious information, the census officials employed numbers and percentage of the total population, relative increase and decrease in numbers across decades, the concentration of a given religious community in a particular area, among other things. In operational terms, this meant the exact numbers of Hindu and Muslim communities, their numerical status vis-a-vis each other in a given area, increase or decrease in numbers — both relative and absolute — across decades, and the identification of regions as Hindu or Muslim-dominated. A new terminology entered the political language — majority and minority. It is important to recognise that all these issues have played such an important role in promoting communal animosities, both during the British period and in Independent India. 

While identifying religious communities, an important and contentious question arose during the early censuses: who is a Hindu? This may arouse no intellectual curiosity today, but it was a mind-boggling riddle in its times. Were ‘untouchables’ to be included within Hinduism? A textual, varna-based criterion excluded them from the Hindu fold. What about the many tribal communities whose ritualistic practices resembled those of formal Hinduism, but who were outside the doctrinal umbrella of formal Hinduism? And most importantly, if ‘caste’ was a part of Hindu religion and was also rooted in specific occupations, how to enumerate a non-Hindu practitioner of an occupational activity, putatively, within Hinduism? In simpler words, how to enumerate a Muslim cobbler? Was he a Muslim by virtue of following Islam, or was he a part of the cobbler caste, like many of his fellow cobblers? These were not just academic questions. These were serious dilemmas which needed to be resolved, if the census operations were to proceed smoothly. Often the manner in which these dilemmas were resolved shaped the contours and boundaries of Hinduism. Many of these boundaries were not very old; they were created during the course of census operations. To take just one example, the Census of 1891 reported a large community of around 10,000 people in Bombay city who called themselves followers of Kabirpanthi (followers of Kabir) faith. Their ritualistic practices were such that they could not be easily identified as either Hindu or Muslim. Their religion was enlisted as Kabirpanthi. However, at the next census held in 1901, it was perhaps considered necessary to know which big religious tradition they belonged to — Hinduism or Islam. And so the erstwhile community of Kabirpanthis got dissolved and split into Hindus and Muslims. Kabirpanthi as a religious category disappeared not just from the census but also from the socio-religious lives of the people. Some of them became ‘Hindus’ and some ‘Muslims’. 

Some clear implications flowed from the way in which this exercise was conducted. It is true that from the very beginning of the British rule, religion was treated as the primary unit of social division. If the cultural and social lives of the Indian people had to be understood, it could best be done through the prism of religion. But in all this, religion ceased to be purely a matter of faith or rituals; it became a ‘community’ instead. Religion became a community mapped, counted, and above all, seen in relation to other religious communities. This was an important transition from faith to community. It was also recognised by the officials of the census operations. 

This new configuration of religion, in which the census played a crucial role, created very new fears and anxieties among members of different religious communities. It created a sharp religious consciousness, which also entered modern politics. Claims and counter-claims emanated from this consciousness. Some Muslim elite in Bengal presented a memorandum to Viceroy Lord Minto that any constitutional arrangement for Muslims should be made not on the basis of their numerical strength, but their political importance. This was the first attempt made by the Muslim elite to try to ensure against any possible disadvantage that might emanate from their minority status. The result of this memorandum was formation of the All India Muslim League in 1906, with official blessings. Three years later, the British proceeded to grant separate electorates to Muslims, according to which Muslim legislators were to be elected exclusively by Muslims voters. 

The Hindus also showed similar anxieties. Many of them were gripped by a genuine, though unrealistic, anxiety regarding their declining numbers and the possibility of them being reduced to a minority status. In Punjab, Lala Lal Chand wrote a series of articles in the journal Panjabee, titled “Self-Abnegation in Politics”. The articles pointed out the bleak future that awaited the Hindus as a result of their declining numbers (either because of conversions, or changing criteria of enumeration across different censuses), the British concessions to Muslims, and the Congress’ apathy towards the Hindus of Punjab. Very similar concerns were expressed in Bengal and other regions also. Soon, this became a major narrative in politics. Quite clearly, the politics of communalism followed in the 20th century did emanate from some of the implications of the census enumerations. 

The consolidation of religious communal consciousness was only one aspect of the census operations. The impact on caste was much more decisive and comprehensive. The census did not simply create a caste consciousness, but a new social category of caste. It is important to recognise ‘caste’ not as a continuation from India’s long past, but an altogether new social category  created since the second half of the 19th century. 

When the British started the census at its nascent stage, they were familiar with two important categories of varna and jati. The varna, or the four-fold social division, came from the classical Sanskrit texts. They were also aware of jati as an operative social stratification on the ground. The trouble, however, was that varna was a completely textual construction, not available anywhere on the ground in its pure form. It was more textual rather than real. Jati, on the other hand, was real. But the reality of jati was so local and diverse that it could not be transformed into a pan-Indian aggregate. Therefore, left to itself, neither category — varna or jati — was suitable as a classificatory principle for census enumeration. The dilemma was compounded by the modern academic discipline of anthropology, which gave the census officials yet another category: race. This was understood in 19th century Europe as a product of both history and biology. It was argued that the category of race could also be applied to Indian conditions. 

As a result, from the available conceptual raw material of varna, jati and race, the British officials, ethnographers and census enumerators carved out a new social category: caste. Elements from all the three were imputed on it. Caste could be shown to be pan-Indian (much like varna), operative on the ground (much like jati), and fixed, exclusive and immutable (much like race). This was nothing short of a great feat of conceptual engineering. But at the level of social application, the category of caste kept being employed in the censuses in loose and unselective ways. For instance, the 1901 Census list of castes in Bengal mentioned not just ‘Madrasis’ and Marwaris, but also Chinese and Japanese! 

In fairness to British officials, they remained uneasy about the appropriateness of this category for India’s operative social stratification. Caste was too important a category to be dropped, but also too much of a manufactured category to be able to authentically capture the social realities. This dilemma showed up in the ambivalent attitude towards caste at the Census of 1931, when they recognised its importance but also realised that it was too fluid a category to be applied with any degree of precision. Eventually, they decided to remove caste from the Census of 1941. But by then, a dominant social category had been created, with an entire political apparatus — organisation, mobilisation, demands, representation — built around it.  

Politics in Independent India has persisted with the colonial category of caste. Such strong stakes have built around it that it is virtually impossible to imagine Indian politics without it. It is also not very easy to convince ourselves that caste could be seen as an invented tradition. But if we do, it should not be difficult to recognise that caste, in its present form, is a ‘gift’ by the British, particularly by their early census operations.

— The writer is Professor of History  at Ambedkar University, Delhi 

On Census of 1931

As on the occasion of each successive census since 1901, a certain amount of criticism has been directed at the census for taking any note at all of the fact of caste. It has been alleged that the mere act of labelling a person as belonging to 

a caste tends to perpetuate the system, and on this excuse a campaign against any record of caste was attempted in 1931 by those who objected to any such returns being made. It is, however, difficult to see why the record of a fact that already exists should tend to stabilise that existence... It’s fair to conclude that there is a tendency for the limitation of caste to be loosened... It is possible that in another 10 years, it may be feasible to substitute some other criterion. 

From Census of India, 1931, Vol. I, Part I, by JH Hutton, Census Commissioner

On Census of 1941

There has been much misunderstanding 

of the change this time [from religion] to community and what it implies.... In the past, the sorting of religion has been accompanied by a degree of caste sorting, complete till 1931... In 1941, caste sorting on an all-India scale was dropped.

The religious question itself was unsatisfactory. Unfortunately... the answers given or attributed to it were being used unconsciously as answers 

to a question on community or origin, a most unscientific position which it was desirable to end...  a religious return was being used as a community one.

From Census of India, Vol.I, Part I, by MWM Yeatts, Census Commissioner

Join Whatsapp Channel of The Tribune for latest updates.