In a year from now, the much awaited and long-delayed Census exercise will begin, first in the western Himalayan regions like Jammu and Kashmir, Ladakh, Himachal Pradesh and Uttarakhand, from October 1, 2026, and later in all other parts
of India, from March 1, 2027. Every Census yields adequate and authentic data essential for determining various welfare policies.
But the Census was to take place originally in 2021. Although the government has not disclosed the reason for the delay even beyond the Covid-19 pandemic, one can speculate what it could be. There are whispers that destabilising the well-set convention of carrying out the Census in the first year of every decade may be linked to the potential increase in the number of elected representatives in Parliament. In view of the Constitution Amendments, 84 of 2002 and 87 of 2003, the number of MPs will be substantially higher. Had the 2021 Census taken place when it was due, the delimitation process would have started — as laid down in the Constitution amendment — after 2031 when the next Census would be due. But now, the delimitation exercise appears to be on the cards even before the general elections of 2029. So the apparent delay in the Census, it can be said, has effectively led to the hastening of the delimitation exercise. The sole basis for it will be the Census that will unfold from October 2026.
The responsibility of conducting the Census rests with the Central government as per Provision 69 of the Seventh Schedule created with reference to Article 246 of the Constitution. The Article provides for a “Central List”; and the one word Provision 69 says is “Census”. Clearly, the government is all powerful in this matter; therefore, the country has to be watchful of the manner in which the Census is designed and carried out. In this context, three important questions need to be asked.
The first is related to the structure of the caste Census. In June this year, the government declared that caste enumeration will take place in the national Census. The last caste enumeration was done in 1931; the map of India was much different then. On the basis of that enumeration, it was possible to determine the population of the communities which we now recognise as Denotified and Nomadic Tribes. I should add that the DNTs are not ‘tribes’ as understood by the term, scheduled tribes or jan-jatis. The DNTs are the communities that got segregated from the rest of Indian society — both caste and tribe — on account of the infamous 1871 Criminal Tribes Act brought in by the colonial administration when Lord Mayo was the governor-general. These communities were erroneously branded as “criminal”, kept confined in soft prisons known as “settlements”, subjected to harsh and unpaid labour, such as constructing roads and railway lines, and treated as “born criminals”. This meant that every new-born in the community had to carry the stigma of criminality. Successive laws during the colonial times brought more and more communities under their purview such that — projected on the basis of the 1931 Census figures — their population around 1991 was estimated to be about five crore or more. The notifications were withdrawn in the early 1950s. Except for constituting a National Commission for Denotified, Nomadic and Semi-Nomadic Tribes in 2014, precious little has been done to reduce their plight. The question for the Census authorities is this: will they include features in the Census structure that can help the country know the precise number of the DNT population? A count on the basis of caste will not help; the Census schedules would have to introduce a specific question related to the DNT identity. If that is not done, the country shall never be able to tackle the question of human rights and welfare for DNTs.
The second question is even more prickly. It relates to the data on religion to be collected. In my work with nomadic communities, I have often come across families observing two distinct religions. Kalbelias, Madaris, Garudis, Bahurupias and Gadiya Lohars often say that some in the same family are Muslims and some Hindus. Such bi-religious communities are found particularly in the border areas of Gujarat, Rajasthan and Kashmir. A comparable situation prevails on the border areas of the Northeast. The question on religion asked during the Census compels respondents to name only one religion, or none, giving bi-religious or multi-religious individuals the slip. The Census must add a layer of question on religion which will prevent misrepresenting of religious identity. An even more complicated aspect of the religion count pertains to the tribal communities. While the tribal population has remained close to 9% of the total population of the country, currently some 13 crore people almost invariably get clubbed together with Hindus and, in some instances, with Christians. The 2011 Census had reported that out of a total of 121 crore people, 96.62 crore were Hindu, 17.22 crore Muslim, 8.39 crore Christian, 2.08 crore Sikh. 0.8 crore Buddhist, 0.4 crore Jain, while 0.8 crore belonged to “others”. Those who had not stated their religion numbered 28 lakh. When we add the last two categories, the total is just a little over one crore. India has an estimated 13 crore tribals. From all the wealth of anthropological studies, one knows that they do not follow temple-based religions. Will they then be pushed into categories like ‘Hindu’ or ‘Christian’ at the whim of the enumerators? The Census needs to be mindful of the fundamental right to follow any religion laid down in the Constitution.
The third question is about language. In the last Census, 1,369 “mother tongues” claimed by citizens were ‘grouped’ under 121 ‘languages’. This was done most unscientifically. The Census believes that in order to be a ‘language’, a language has to be spoken by at least 10,000 persons. No variety of linguistics can accept this weird logic. Similarly, many languages were shown to be sub-sets of a neighbouring, large language. For instance, Bhojpuri, claimed by over five crore persons as their language, was shown as a subset of Hindi. There were 55 such other cases in relation to just Hindi alone. Why does the Census want to deny citizens their linguistic identity? It will have to represent the cultural facts and linguistic realities as they are without such largescale doctoring.
An exercise well-set since 1880, and on which people still have faith, cannot be allowed to be used to bolster the numbers of one religion and one language. If that is done, the Census Commissioner will look no different from the Election Commissioner. In effect,
the already problematic process of delimitation might end up becoming even more so.
G.N. Devy is a cultural activist