A caste census is a Pandora’s Box that India must open anyway

In fact, the 2011 Census had included an enumeration of castes, but the data was never released (for reasons discussed below). Bihar conducted a state-level caste survey in 2022. The governments of Maharashtra and Tamil Nadu, among others, have been vociferous in their support of this step. With the acquiescence of the  Cabinet Committee on Political Affairs this April, all major political interests are now aligned in favour of a caste census.

Also Read: Himanshu: India’s caste census must serve its purpose

Caste demography represents a massive and critical gap in India’s self-knowledge. There is some information on the most marginalized groups: Scheduled Castes and Scheduled Tribes, who comprise roughly 16.6% and 8.6% respectively of the population. Together, they make up about one-quarter of the country’s people. Another 20% are people who do not identify as Hindu: Muslims (14.2%), Christians (2.3%), Sikhs (1.7%), Buddhists (0.7%), Jains (0.4%) and others, according to the 2011 Census. 

That leaves about 55% of the Indian population as undifferentiated Hindus, including Forward and Dominant Castes and Other Backward Classes (OBCs). Therefore, the purpose of a caste census is primarily to disaggregate and understand the composition of this 55%. 

While there is some interest in learning more about the demography of Forward and Dominant Castes, the most important objective of a caste census is to estimate the size and composition of the OBC population. This is surely India’s largest social group. But how large is it? No one knows. The Mandal Commission (1980) placed it at 52% of the total population. The National Sample Survey Organization calculated it to be 32-36% in 2000 and 41% in 2006. Other government and non- government surveys place the number in the 30-35% range. The Bihar caste survey counted 63% of the state’s population as OBC.

Also Read: Caste census? Okay, but we must handle it with care

Why is there so much imprecision and uncertainty about India’s OBC population? There are several answers to this question. For one, self-identification of social identity can be tricky: people may have mixed identities, or they may believe they have an identity that does not match official categories, or they may choose not to identify as ‘Backward’ in any way. 

Second, the official categories themselves aren’t fixed. The Union ministry of social justice and empowerment, which is in charge of maintaining caste lists, moves castes in and out of the OBC list based on dynamic factors (like education and economic conditions). 

Moreover, the OBC category is itself very diverse: castes that may be lagging in some regions may be dominant landowning groups or thriving small business communities in others. So much so, that there exists a quasi-official category called ‘upper OBCs’ that includes caste groups like Yadav, Kurmi, Koeri and Bania (in Bihar).

Also Read: We need a reformation: Caste salience must fall for India’s social capital to rise

One of the main reasons for the caste census data of 2011 not being released, as reported, is that people self-identified in tens of thousands of castes, so it became impossible to impose some sort of order on the data. This is also what happened when the British Indian government began conducting the census exercise in 1867-72. 

The original intent was to capture caste data under the so-called Chaturvarna system: Brahmin, Kshatriya, Vaishya and Shudra. That turned out to be impossible. By the 1901 Census, there were 1,641 castes and sub-castes listed; by 1931, the last census with caste information, that list had grown to 4,147. C.F. McGrath, who was the commissioner of Independent India’s first Census in 1951, had earlier concluded that the “meaningless division into the four castes alleged to have been made by Manu should be put aside.”

Also Read: The Bihar caste survey marks a big step towards a reality check

The only solution to this ‘problem’ of India’s teeming diversity is to reduce it; to impose order by using a preset rather than open-ended caste list. This is what Bihar’s survey did and is presumably what the national Caste Census of 2027 plans to do. The obvious problem with this is that by reducing choice, millions of people may feel un- or mis-represented and therefore opt out of the system.

This points us to a deeper problem with this approach. A census does not merely record reality, but alters it by imposing order (by using categories and lists). A census therefore shapes the world it seeks to describe and reactions to it then reshape the census and the world it describes. One possibility with a pre-set caste list is that unlisted castes will begin to disappear as recognized social identities. Another possibility is that resistance and political action will shape future caste lists.

Also Read: A national caste census looks all but inevitable

A census is a powerful tool with wide-ranging and long-lasting consequences. The British Indian censuses were arguably the most important social experiments ever carried out in the subcontinent. The categories and categorical definitions they created—especially for religion but also for caste and tribe—have come to be accepted as real and permanent. This space does not permit any further exposition. Interested readers may look up my 2019 book The Truth About Us.

There is no doubt about India’s need for a Caste Census that provides robust information on the composition of 55% of its population. Core policies on welfare, education and employment should ideally be based on this data, but have been shooting in the dark at shape-shifting targets so far. But let us not play down the challenges of doing this well and recognize that, even if done well, the Caste Census will inevitably change the composition and politics of caste in ways that are impossible to predict now.

The author is a professor of geography, environment and urban studies and director of global studies at Temple University.