Caste Counts

**Rohini Mehta

INTRODUCTION

A caste census is an effort to aid and enable the effective construction of affirmative action policies in India. However, like most topics in the caste and reservation jurisprudence, this aspect of statistical compilation of caste-based data has been met with controversy and resentment, even. A fear emanates primarily from upper caste scholarship regarding such census, since it would be a formal reflection of the sheer number of lower caste communities, legitimising demands for continued and expanded reservation. After much debate, and under immense political pressure, the 2011 Socio-Economic Caste Census (‘SECC’) collected caste-related data, but later did not release it, citing ‘technical flaws’. 

In February, 2019, the caste census controversy was stirred up yet again when the Bihar Legislative Assembly passed a resolution to advocate for a caste-based census in 2021. Due to delays from the Centre, the State took it upon itself to conduct a ‘caste-based survey’ in 2023. This survey was immediately challenged by petitioners like the NGO ‘Youth for Equality’, where the Patna High Court affirmed the validity of the survey. This decision is challenged before the Supreme Court and remains pending. This article will aim to address one of the primary constitutional arguments against the census: the alleged infringement of the right to privacy. Part II will attempt to address the question of whether caste, in the context of a census, constitutes ‘personal information’ as discussed under K.S Puttuswamy & Anr. v. Union of India & Ors. (‘Puttuswamy’). Part III will proceed to address specific arguments raised by the petitioners using the principles laid down under the K.S Puttuswamy & Anr. (Aadhar) v. Union of India, (‘Aadhar’).   

PUTTUSWAMY, PERSONAL INFORMATION & CASTE

While the Puttuswamy judgment dealt with laying down the principle and fundamental right to privacy doctrinally, the Aadhar judgment focused on the application and limitation of this right in context of the Aadhar Act. The Puttuswamy judgment discussed the concept of ‘informational self-determination’, which is the control over dissemination of one’s personal information. While the bench had various opinions, Gautam Bhatia writes that a singular consensus that can be gleaned from the judgment is the emphasis on ‘informed consent’, wherein an individual has the right to know the use of their data and correct that use. The bench  rightfully refrained from defining ‘personal information’, and instead used situations to explain what it may constitute. Examples of medical information, sexual orientation, DNA profiles and fingerprints were taken. The Court referred to international jurisprudence as well, such as the United States v. Miller, wherein personal information was held to include information that ‘reveals intimate details of the lifestyle and personal choices of the individual’.

The Court also referred to Section 8(j) of the Right to Information Act, 2005 (‘RTI Act’), which contains an exemption from disclosure of personal information, as that information which has no relationship to public activity or interest and would cause ‘unwarranted invasion’. Keeping this in context, the question of whether ‘caste’ can qualify as personal information to the extent that it attracts the complete protection of Article 21 is slightly complex.

No concrete judicial opinion on this question of caste as personal information exists, with only a handful of conflicting High Court decisions regarding disclosure of caste certificates under the RTI Act. In Satpal v. Central Information Commission, the Delhi High Court opined that the personal information, like a caste certificate submitted by an employee to an employer are expected to be kept confidential. However, this was in the context of the analysis of a fiduciary relationship exemption, rather than an analysis of the caste question itself. On the other hand, in Tamil Nadu Public Service Commission v. P. Muthian, wherein the respondent sought information regarding the number seats and list of candidates of the Most Backward Community category, his request was denied on the ground that caste information has no relevance to public activities. The Madras High Court responded sternly, calling the behaviour of the Public Information ‘tactical and mechanical’, directing the immediate furnishing of documents requested to the respondent.

It is important to identify a distinction between collection of caste data as violation of the right to privacy, versus such infringement argument being targeted towards methodology and subsequent use. This is due to the contextualised manner in which the right to privacy and its violations are assessed. Interestingly, most arguments target the very activity of a caste census itself rather than any particular modality, positing that caste is ‘personal information’. For example, Pratap Bhanu Mehta writes, without referring explicitly to privacy, that counting castes diminishes “our freedom, agency and dignity”. Several Dalit scholars have reacted strongly to such arguments, claiming that they are mere facades for the real reason upper castes want to suppress caste censuses, which is the fear of data reflecting a need for stronger affirmative action policies. It is argued that there is evidently a ‘politics of convenience’ at play, wherein caste cannot be categorised as ‘intimate’ information, confined to a private realm. In India, caste is social currency, wherein caste based hierarchies and activities are very much in the public domain. 

It may also be useful to draw a parallel to ‘racial privacy’ (or rather, lack thereof) in the United States, wherein legislation requires racial classification through the decennial census. In fact, a referendum to make race private in California failed, wherein people believed that such classification was required for affirmative action policies. Anita Allen also argues against the expanded use of racial privacy, since it would be impractical and ironically, take away from the social goals it seeks to protect. In this way, race and caste, while different institutions, can be studied parallelly to construct appropriate policies with the correct amount of ‘caste privacy’ in appropriate domains, given the inherent paradox. 

YOUTH FOR EQUALITY V. STATE OF BIHAR: PRIVACY & MODALITY

The petitioners put forth two main arguments in respect of the right to privacy being violated – firstly, that a caste is ‘imposed’ on everyone due to the absence of an inapplicability option. It was further argued that a right of self-determination under Article 21 is violated since every individual undertaking the survey must identify with a certain caste. Reliance was placed on Shafin Jahan v. Asokan K.M, which deals with privacy in relation to the intimacies of marriage and choice in religion.

However, religion is a chosen belief, and the intimacies of marriage can be placed squarely in the private realm, as already highlighted in the Puttuswamy judgment. There is a considerable difference between these aspects and caste as an institution. Further, there is no established ‘right to self-determination’ in terms of caste. The Puttuswamy judgment only refers to such ‘self-determination’ in relation to gender identity, and choices made purely in the ‘privacy of the mind’. Such a right in relation to caste would be highly problematic and illogical, given that caste is hereditary, objectively determined by society and such a right could allow upper castes to suddenly ‘identify’ as marginalised.

Secondly, the submissions of the petitioners also attempt to disprove the limitations on the right to privacy by asserting that there is no legitimate objective or proportionality by emphasising on their legislative competency submissions. The Aadhar judgment dealt particularly with the Unique Identification Authority of India program, involving issuing unique identity numbers to each citizen using biometric and demographic data. The Court used the three-fold test, involving the ascertainment of a legitimacy, need and proportionality (in terms of lack of less stringent, equally effective alternatives and suitability) to uphold the Aadhar Act. 

Interestingly, while the petitioners target specific modalities in the Bihar caste survey like the lack of a “No” option under the caste head and the collection of only paternal caste, they are delivered in a manner that generalises the Bihar caste survey as representative of all caste censuses. For example, at multiple points the petitioner has argued that the activity of caste census has been avoided as a deliberate Central government policy decision, and that such an effort to enumerate caste itself “smacks of arbitrariness”. The way the Patna High Court responds to such arguments is also concerning, with heavy emphasis placed on the fact that the survey was “completely voluntary” and not a single instance of objection was encountered at an 80 percent completion stage. These arguments and the subsequent response rings alarm bells for the constitutional fate of caste censuses in general, since they lend credence to the notion that such a census would automatically be violative of constitutional rights. 

While it is possible that Bihar caste survey may fail the Puttuswamy proportionality test due to its specific modalities and the issue of legislative competency affecting the fulfilment of the first prong of legitimacy, it is extremely important to recognise that these failures should not cast a shadow on every other caste census. In Indra Sawhney v. Union of India, the bench made clear that a process to investigate ‘backwardness’ cannot be declared unconstitutional simply because caste was an identifying or major parameter. The Bihar caste survey has seventeen heads, such as education, employment, income, vehicle ownership, residential status, housing and technological accessibility. 

However, the petitioner’s argument that such caste censuses should be conducted by independent commissions and not the State government due to the risk of electoral politics, holds water. While caste information is important for welfare policy, it is also extremely potent information, requiring high levels of safeguards, both technologically and methodologically. Since the survey was not conducted under the Census Act, 1948, which contains guidelines and standards for confidentiality, it is unclear what safeguards will apply to the data collected. A similar challenge was faced by the Socio Economic and Caste Census (2011), which was done under Article 73 of the Constitution, by the Union Ministry of Rural Development and Urban Development, leading to a poorly designed questionnaire, and subsequent failure. 

Additionally, the problem with caste data has been identified as it being in the hands of the police, who have previously shown tendencies to buckle down on areas with marginalised castes. Further, drawing from the principles of informational ‘self-determination’, the lack of a “No” option could indeed be a problem for not just the Bihar caste survey, but any future caste census. To avoid invalidation, caste censuses should include an inapplicability option. From a policy perspective, Satish Deshpande predicts that the “No” option will be availed by an infinitesimally small number of people. This coupled with the fact that there are multiple heads as indicators of backwardness, does not defeat the object of such a census. 

Therefore, a careful demarcation between the constitutional examination of modalities and the overall policy must be kept in mind. The Bihar caste survey case will prove to be landmark in the sense that it is an attempt at a policy that has been avoided and hushed institutionally. The Apex Court must realise that there are two distinct questions at play – firstly, whether a caste census by itself is violative of the right to privacy and secondly, whether the Bihar caste survey is violative of such right. The decisions in this regard and their ramifications will go a long way in deciding the future of India’s affirmative action.

CONCLUSION

The petition against the Bihar caste survey cannot be examined in isolation and must be seen in conjunction with inordinate delays in conducting a decennial census by the Centre, as well as a historical avoidance of this exercise. This article does not attempt to claim that the Bihar caste survey is remotely a perfect policy, but only to reveal the possible role it could play in future constitutional law discourses regarding the validity of caste censuses. The discussion on such a survey being landmark, could very easily turn into a black and white ‘pro census versus anti census’ one that completely invalidates the idea of this exercise. However, this is an opportunity for courts to truly discuss the nuances of what methodologies could make such a census violative of constitutional provisions, thereby leading to improved policy. The need for such a census has been repeatedly echoed and supported by many, especially members of marginalised communities. Dalit scholars also write about the issue of institutional ‘castelessness’ as a tool used by upper castes to shun the realities of caste and challenge its enumeration by using liberal facades. Therefore, regardless of the specific constitutional outcome of the Bihar caste survey, it is a signal to present and future governments to construct an appropriate caste census policy. 

**Rohini Mehta is a BA LLB Hons student at the National University of Juridical Sciences.

Disclaimer: The views expressed in this blog do not necessarily align with the views of the Vidhi Centre for Legal Policy.