Terming the caste-based survey to be “prima facie unconstitutional”, a division bench of Patna High Court imposed an interim stay on the state government’s move. The stay is being seen as a major setback for the Nitish Kumar-led Bihar government which kickstarted the survey on January 7 this year. The High Court’s intervention, however, is a step in the right direction. The court pointed out the possibility that the caste-based survey may violate Constitutional provisions, and is plagued with logistical inconsistencies. This, however, doesn’t downplay the need for having caste-based data for targeted affirmative programmes. It also raises obvious questions about the inaction of the Central government in this regard. As far as the exercise of conducting caste-based census is concerned, it is a Constitutional prerogative of the Central government, with states having almost no role in it. The power is conferred to the Centre under entry 69 of the List 1 of the Indian Constitution. Furthermore, Section 3 of the Census Act, 1948, empowers the Centre to single-handedly “declare its intention” to conduct a census in the whole or any part of the country. What the Bihar government was doing — or at least it claimed so — was to conduct a survey, and not census. This rather deceptive venture of the state government got exposed when the Patna High Court made comparisons between the dictionary definitions of the two terms. While a census entails the collection of accurate facts and verifiable details, the survey collects and interprets subjective opinions and perceptions of people. It is usually aimed at a specific community or group of people. Against this background, the High Court noted that the “caste survey was conducted in the garb of a census.” Needless to say, this evidently camouflaging act of the Bihar government had to be stalled and cross-checked. Apart from the violation of Constitutional and legal provisions, the conduction of caste-survey by different states may lead to irregularities and even conflicts among the states. The crucial element of uniformity might get lost as there will be subtle differences between the method and design of each state. The most threatening consequence of the conduct of the caste survey, however, will be the breach of privacy — as argued by the petitioners and acknowledged by the court. The High Court termed it a “matter of grave concern” because the safety framework for the collected data in caste survey is still not in place. The sharing of collected data with different stakeholders can open a can of worms when it comes to privacy breach. The Data Protection Bill is still lingering in limbo. Responding to the Advocate General of Bihar’s argument that the nature of data collected under the caste survey is already shared in the public domain, the court exclaimed, “If the details are available in the public domain, we fail to understand why such a massive exercise, expending public resources has to be undertaken.” Though the stay pertains to the caste survey in Bihar, the questions raised by the court will have a wider impact on other states as well. Notably, several states including Maharashtra, Chhattisgarh and Odisha are in the process of conducting similar exercises. The real irony is that the state governments are facing the compulsion of furnishing caste data before the Supreme Court and some High Courts but are not empowered to collect data in the form of census. The courts have, on several occasions, highlighted the need for such data to devise and formulate new affirmative action policies. It then becomes the duty of the Central government to facilitate the collection of such data. It can do so primarily in two different ways. Firstly, it can use its Constitutional powers to conduct a nationwide caste census. Secondly, it can use its residuary power under Article 248 to make laws on caste census. Through legal changes, the Centre can empower states to carry out caste census in a well-calibrated manner. This, in turn, requires strong political will and effective ignorance of short-term electoral gains or losses.