The debate about whether the decennial Census should collect data on caste from individuals who fall into the administrative categories of ‘General’ and ‘Other Backward Classes’ (OBCs) has been argued by public intellectuals, politicians, and government administrators for decades. As the Census currently only collects data on ‘Scheduled Castes’ (SCs) and ‘Scheduled Tribes’ (STs), it fails to provide comprehensive data on India’s graded caste hierarchy. In the run-up to the 2011 Census, the political leadership agreed to include a full caste count in the Census. It later prevented a caste-wise enumeration in the Census. The suppression of caste-wise data took place then because of two interconnected dynamics which are likely to reoccur unless they are collectively challenged.
The importance of caste data
First, caste elites generally believe that caste no longer matters in shaping opportunities and outcomes in the 21st century. This caste blindness, or castelessness, obscures caste privileges and conceals sources of multi-generational structural advantage. Many caste elites view the collection of caste data about anyone but the most disadvantaged as unnecessary and a misuse of public resources. This perspective both serves their own interests and ignores the relational nature of caste — that is, the same societal institutions, systems, and cultural norms that have led to historic and ongoing subjugation of oppressed castes have simultaneously empowered others. To understand the full scope of disadvantage, we must also examine the full scope of privilege and advantage.
The suppression also occurred as a result of the machinery of government. Organisations tasked with designing Census questions and overseeing data collection, similar to every other key institution in society, have caste-based inequalities entrenched within them. The bureaucracy blocked the inclusion of a full caste count in the Census 2011 on methodological grounds. It argued that a caste count would be “administratively difficult and cumbersome,” “jeopardise the whole exercise,” and “compromise the basic integrity of the Census”. The official language used by the Congress-led government in 2011 was identical to the language used in the affidavit filed in the Supreme Court on September 23, 2021 by the present BJP-led government. The presentation of (supposedly) insurmountable methodological and logistical challenges is particularly effective as an excuse because it silences non-experts. Caste elites have a numerical and cultural stranglehold over the upper bureaucracy, despite more than 70 years of Central government reservations. In 2019, out of the 82 Secretaries to the Government of India, only four were SCs or STs. Among 457 serving secretaries, joint secretaries, and additional secretaries, merely 12% were SCs and OBCs; similarly, group 1/A of the Central Civil Services (i.e., the top tier of the bureaucracy) has still not fulfilled its reservation quotas for SCs and OBCs. Following the suppression of the caste count in Census 2011, the executive bureaucracy reconfigured the Below Poverty Line survey and renamed it the 2011 Socio-Economic Caste Census, which had little resemblance to the original demands by caste census advocates and produced unusable caste data.
The purpose for collecting caste-wise data in the decennial Census is to understand the contours of inequality. These data are crucial to understand how caste intersects with class, gender, and regionality to structure access to resources. The collected caste data should be publicly available for use. In this regard, the caste data would continue the existing practice of the Office of the Registrar General of India to make Census data publicly available. The Census has the legal standing, public trust, operational expertise, and resources to collect, analyse, and make public caste data. Caste data must be collected as part of this constitutionally required exercise. Having the caste Census as part of another state project, or overseen by nodal agencies other than the ORGI, as happened 10 years ago, will relegate it to parts of the bureaucracy with insufficient expertise in a nationwide data collection operation.
While counting (or not counting) caste is political, the decision should not be reduced to immediate political contingencies i.e., the expansion of reservation policies, the caste-based mobilisation by political parties, etc. In the absence of detailed caste data, we fail to name and confront major structural and foundational problems of society; leave space for opportunistic politicians to exploit each caste; and miss the opportunity to craft reasoned, data-driven, and inclusive public policies.
Yet, important concerns remain. Some progressive and anti-caste scholars fear that a full-caste count will further entrench caste identities. A caste census will require all households to think about, acknowledge, and speak about caste identities. Yet, historically outcast groups have already had to provide caste data in all postcolonial Censuses to implement reservations. A full caste-wise enumeration will help to make visible privileges and resources that have become over time disassociated with caste, despite historical, sociological, and economic evidence to the contrary. Updated data on the entire caste system, including its intersections with other identities, will provide a more complete picture of exclusion and inequality in India.
Another concern is that groups will misuse the caste data. But misuse of caste data already takes place. Private groups with access to money and power regularly collect caste data for their needs. Political parties map the caste and religious composition of neighborhoods, cities, and villages to mobilise votes. Collecting caste data in the decennial Census removes this private power by making caste data publicly available to all.
While methodological and logistical challenges are real, they are surmountable. Demographers in government agencies and universities have extensive experience working through these challenges. Sample surveys such as the India Human Development Survey have collected caste-wise data. Census bureaus in the U.S., Brazil, and South Africa, as well as in other countries with long histories of white supremacy, collect detailed data on race and class to understand the current scope of inequality and develop justice-oriented policies. In addition, research on the failed caste count suggests the importance of careful planning to prevent groups from being made invisible in the data, such as Dalit Muslims, Dalit Christians, inter-caste and inter-religious households (particularly those that cut across the line of ‘untouchability’ or communal divide), and LGBTQ+ individuals. Related to the discussion of castelessness, if a ‘no caste’ option is included in the Census, the caste count will likely undercount well-to-do caste elites. Given the purpose of the caste count, omissions of marginalised groups and elites require specific attention while designing the survey instrument, training enumerators, educating the public, and analysing collected caste-wise data. Hence, the entire process requires external oversight if the data are to be usable and to minimise potential harm. As the process unfolds, a public oversight group should work to ensure that major operational and methodological decisions align with the data collection’s purpose: to understand the scope of caste-based inequities and address structural inequalities. Anti-caste organisations and public intellectuals, who have devoted their life’s work to challenging caste hierarchy, must provide oversight and input. Their perspectives and lived experiences of fighting caste oppression are the best safeguards to ensure that the collected data will be used for liberatory purposes.
Trina Vithayathil is Chair and Associate Professor, Global Studies, Providence College, and Kalaiyarasan A. is a Fulbright-Nehru postdoctoral fellow at the Watson Institute for International and Public Affairs, Brown University , U.S., and Assistant Professor at the Madras Institute of Development Studies, Chennai. Views are personal