The Numbers Do Matter: Religion Data of Census 2011: I

With this post we begin a new series on important aspects of the Religion Data of Census 2011, which was released in the last week of August 2015.

The data indicates that the share of Muslims in the population of Bharat has increased by 0.8 percentage points; this is similar to the increase they have registered for the last several decades. It also indicates that the aggregate share of Christians in the population has remained nearly unchanged during the decade, though they have registered large increases in some of the northeastern States and in several other pockets of the country; this is also similar to the pattern of the last three or four decades (but the official numbers for Christian population do raise questions if one looks at the ground situation in states like Andhra Pradesh).

But these headline figures hide several very significant changes that indicate the beginning of a new dynamics of religious demographic change emerging in the country. Particularly, the data shows that in several States of Bharat, including some which have seen great contraction in the share of Hindus in the past, the Hindu share has begun to stabilize and even increase. The phenomenon is visible most remarkably in several districts of the heartland States of Bihar and Uttar Pradesh. This obviously is not a consequence of the plateauing of the growth rates of all communities that the demographers have been awaiting; such plateauing, if it were happening, would have been seen in a demographically ‘advanced’ State like Kerala, but there the gap between the growth of Muslims and others has only widened.

There are also very significant changes happening in the share of Sikhs, Buddhists and Jains, etc., in different parts of the country. In this series, we propose to look at all these issues, which have important social and political implications, in detail. But we begin with a note on why it is important to take the data on Communities and Castes seriously and have open and informed discussion on these issues.

Religion and Caste have been the two great determinants of Bharatiya history, politics and social organisation. These two parameters are so important that no field level administrator can possibly do without knowing in detail about the relative numbers and strength of various communities and castes within his area of responsibility, and no one would be taken as a serious political worker in any part of Bharat unless he has this information on his fingertips.

But in elite Bharatiya society, there is certain coyness about raising the issue of community and caste in polite or academic discussion. These are unavoidable facts of Bharatiya public life; we cannot possibly run our politics and administration without taking these into account. But we want to avoid talking about these; we persist in treating the castes and communities as some kind of social vices to be discussed privately between politicians, administrators and media-persons.

The most glaring example of our hypocrisy in these matters is perhaps that of the Sachchar Committee, a high-powered committee set-up by the UPA Government to report on the status of Muslims in the country. The Report of the Committee is entirely about counting of numbers of Muslims and others; it counts the share of Muslims in the population of different parts of the country and the relative share they have in education at different levels, in employment in different sectors and services, in bank deposits and loans, and so on. The emphasis of the Committee on obtaining numbers of different communities in various facets of public life was such that it wanted even the armed forces to compile numbers on communal lines. But in the early part of its report the Committee makes the lofty assertion about the irrelevance of the numbers saying, “Since the growth of the Muslims population has been above average, and is likely to remain so for some more time, the question often asked is whether, and if so, when, will the Muslim population become the largest group? The counter position is that how does it matter which population is the largest?”

This is the level of hypocrisy we practise. We, all of us who are active in public affairs, keep counting numbers of different communities and castes; we keep minutely measuring their relative space in public life, their relative privilege or lack of it. But we also keep saying that the numbers do not matter. That is why we stopped collecting the data on caste in our decennial censuses after Independence; numerical strength of different castes and sub-castes has thus have become a matter of generally biased guesswork based on the Census of 1931, the last time when the caste numbers were actually counted. Efforts to carry out a caste census during the UPA regime were strongly resisted by many sections, even from within the UPA, and ultimately the process was so rigged that no meaningful data on the number of different caste groups is likely to emerge from the Caste Census, which is said to have been completed now.

Fortunately, the Census has not stopped counting the data on religious communities. Though, efforts to influence the counting of the data and delay its publication are made every decade. After the count of 2001, the data was published somewhat promptly and in greater detail than usual; that led to the Registrar General losing his job. This time, the publication of the data had to await ‘political clearance’ for nearly two years after the data was rumoured to be compiled and ready; many significant pieces of the data were available with the press for at least a year before its formal release.

But even when the data is published and released, it is rarely analysed in the kind of detail and intensity that it deserves. When we first compiled the religion data of Bharatiya censuses from 1881 to 1991 up to the district level in our book, The Religious Demography of Bharat, (authored by A. P. Joshi, M. D. Srinivas and J. K. Bajaj and published by the Centre for Policy Studies, Chennai 2003), the then Registrar General privately remarked that the Census had been collecting this data for a hundred years but it was for the first time that it had been taken seriously. We were, however, reviled by the social-science community for daring to look at a subject that they had made taboo.

Notwithstanding the reluctance of the social scientists to deal with the issue, it remains important. The country has been partitioned in living memory on the basis of religious composition of certain regions; in 1947, religion tables of the latest census were perhaps the only data used for drawing the dividing line across the map of Bharat; nothing else, not even natural geographical features like the rivers and mountains, mattered. And, since Partition and Independence, much of the energy of the armed forces of Bharat has been expended in protecting the country against further erosion of the territory of Bharat on religious lines.

Community and caste continue to determine the polity of Independent and partitioned Bharat. The Governments at the Centre and in the States routinely get elected on the basis of community and caste equations, and their policies and postures are often decided by calculations of interests on these lines. Media analysis and commentary on most issues of public interest often follows caste and community considerations, particularly the latter.

Such interests and considerations are not always invalid or illegitimate. But, if the caste and community have such salience in the public life of Bharat, then it is obviously necessary to continuously collect all possible demographic, sociological and other data on these aspects of the population. However, there continues to be great reticence in collecting or dealing with such data in an open and systematic manner.

Continuing with such reticence, Census 2011 has released only the bare numerical data on religion, giving the number of persons of different communities, without correlating the data with other socio-economic parameters like literacy, age distribution, work participation rates, etc., as was done, for the first time, with the religion data of Census 2001. The First Report on Religion published as part of Census 2001, gave all this information; the title of this report also gave rise to the hope that the Census might publish further reports correlating the religion data with other significant parameters like, say, migration. But, as we have said, the Registrar General was relieved of his job soon after the First Report was published. Census 2001 released no further data on religion; and, now Census 2011 has reverted to the earlier practice of giving only the bare numbers of different communities without any socio-economic correlations.

However, the limited data that has been published is of great interest. It shows very significant changes in the relative numbers of different communities that have taken place in the course of 2001-2011. It is possible to draw several sociological inferences and to discern crucial socio-political indicators and historical trends by looking at this data in detail. There has been some discussion in the media of some aspects of the religion data of Census 2011. Beginning with this post, we look at these trends and indicators systematically at the national level as well as at the level of the States and Districts, etc. We hope that the detailed presentation of this data shall generate open and serious discussion on the issues involved and their implications for Bharat.

— Dr. J. K. Bajaj

Disclaimer: This post first appeared on the Centre for Policy Studies site at

This article represents the opinions of the Author, and the Author is responsible for ensuring the factual veracity of the content. HinduPost will not be responsible for the accuracy, completeness, suitability, or validity of any information, contained herein.

About the Author

Jatinder K Bajaj
Dr. J. K. Bajaj, Director, Centre for Policy Studies (