Caste in the Census – some concerns
Written by Anirudh Wadhwa // September 8, 2010 // National Politics // 8 Comments
I have been following the recent move for including Caste as a relevant parameter in the 2011 National Census with much interest. In the face of near unanimity amongst all major political parties, there appears to have been hardly any debate on this issue. I found this unsurprising though a trifle disturbing. Inclusion of caste in the census, coming as it does for the first time since 1931, is an important issue and its many implications deserve closer scrutiny.
Let me advance, though somewhat cautiously, two main concerns that I have against this move.
Census is not the right mechanism for this study
Caste as a political category is unique. Unlike perceptible identifiers like colour or race – caste is a product of self-categorization and depends largely on people’s perception of themselves. For this reason, it is also not static. Since the 1931 census, there has been over two-three generations of fission, re-location, and adoption of new rituals amongst social groups. The Census is designed to collect observational data, and not information based on the perceptions of the respondents or self-categorisation. A demographic exercise which is targeted towards collection of observable data, cannot be transformed to a social science fieldwork. Besides, India is home to a humungous number of sub-castes with nomenclature variations across regions. Census officials are simply unsuited to note the various nuances which are present within the various castes. To give an example – the Gonds of North Karnataka have nothing to do with Gond tribals of Madhya Pradesh. Another example – in south Gujarat a person may describe himself as a Matiya at the village level, a Koli at the taluka level and a Kshatriya at the state level. There are many such local peculiarities. Professor Ghanshyam Singh of JNU has done some research on these issues and his paper can be accessed here. There is also a thorough article by Asha Krishnakumar in the Frontline (access here). This exercise is simply not possible at a national scale – it has to be localized and requires careful study by social experts.
There is yet another concern that needs to be kept in mind. There are many benefits that accrue to particular castes based on their treatment by the State. Reservation programmes in India provide benefits based on caste identity. These have the effect of exaggerating the difference between a favoured and an unfavoured caste, though they may be economically similarly situated. It is trite that data collected from the Census will form the basis for formulation and modification of existing State policy relating to Caste based programmes. Absent a corresponding threat of a sanction and given the huge institutional advantage in distorting the census figures, there is a big incentive to cheat.
Transforming social and political identifies
Official classification does not merely reflect social identities but has a role in transforming them. A process of categorization necessitates definition. It also entails order and rigidity. Individuals find themselves firmly fixed as members in various groups of a particular dimension and substance – as defined by the State. There is an excellent article by Kenneth Jones on “Religious Identity and the Indian Census” (access here) which details the impact of the British Census on religious identity in India. He argues that the census had a role, apart from merely providing statistics, in bringing religious identity to the forefront of national politics in British India. It is most likely that this will be repeated for caste identity as well.
It can of course be argued that caste-based classification is enshrined in the Constitution and has been institutionalized in the current system of reservations. If however we are committed to a caste-less society, this move will only perpetuate it. It will cement a caste based political agenda for at least the next decade or two. Ajay Maken circulated an impassioned letter to young MP’s across party lines arguing against the collection of caste data in the Census. The letter can be accessed here.
Conclusion
The Census presently collects data on SC’s and ST’s and has been doing so since 1951. However, collection of data relating to all castes is a different exercise. Presidential lists of SC’s and ST’s has been prepared by the President, and these categories have, at some level, become “objective”. This is not the caste with the other castes which remain amorphous, diffused and fluid.
As a mechanism for the collection of a nation-wide caste data the census is likely to – miss out on the many local nuances, provide an incentive to cheat to the participants and establish a caste based political agenda as the future of the country.
Perhaps we need to think more carefully about where this move is going to lead us.
8 Comments on "Caste in the Census – some concerns"
Dear Anirudh,
The points are extremely well made especially in that there must be greater debate and discussion before this decision is taken. Two points- I agree with the difficulties of caste classification. But you say caste is a question of self-perception. This seems counter-intuitive and wrong to me unless I’m missing something. Granted that there is sanskritisation and movement amongst castes but that doesn’t take away from the general point that caste is an ascriptive category, unlike economic status. Try really hard and one cannot change it though a certain caste can become upwardly mobile consciously or because of state benefits. That doesn’t however change the caste itself.
Second, I think this opens up a wider debate of where we see caste in modern India. Caste was stressed as a marker by the Indian state in positive discrimination with the ultimate hope that it would cease to be relevant to society when the baggage associated with particular caste names would go as economic growth took place. We may argue that we still haven’t reached that end and we should keep trying or that we must try the more straightforward approach of getting caste out of the picture by de-emphasising it in a phased manner and moving to purely economic classifications. I strongly feel that the latter is the correct approach in the long run and ultimately that is where we should head; i.e. caste must cease to have relevance. The caste census, in my mind, is thus a step in the wrong direction.
I predict that the exercise will be given up halfway as being impossible, futile and mostly useless… precisely the reasons the British gave for not enumerating caste from 1931 onwards. If anything, we will also probably find out that caste means completely different things to different people and that the biggest factor in perpetuating the caste system is the present scheme of caste based reservation.
I also think that the census creates perverse incentives for people to give wrong information since people can hide their real castes and give the names of castes which benefit from reservations. There’s already a thriving market for fake caste certificates and I think this census is a boon for those who wish to fraudulently claim caste based benefits.
Dear Arghya
Caste still remains an ascriptive category – however it has changed, in some measure, from being an other-ascriptive category to a self-ascriptive one. The description of caste as an other-ascriptive category is true in closed and static societies. This is challenged by changing practices and geographic dispersion amongst social groups. My point being – unlike race or colour which is observable and in that sense objective – caste is a result of what one makes of it. On may perhaps draw a parallel to one’s lingual identity; a person’s grandmother’s first language might be X – but through two generations’ of social/geographic mobility such person may now describe her first language as Y.
Dear Alok
The point about perverse incentives goes even further. Caste is now established as perhaps the most important identity for securing state benefits. Further amends to existing systems are very likely to be based on the data collected through the Census (and if not, then this dilutes the argument for carrying out this exercise in the first place). There is a perverse incentive for groups to en masse misreport their social and economic status in a way which would further their own interests. This is not mere conjecture alone – the article by Kenneth Jones (above) provides some interesting examples of such group behaviour (in the context of their religious identity) during the times of the British census.
Further, I think, there are two points:
1. Is the caste enumeration is happening by asking jati or by asking social group (SC/ST/OBC)?
2. Is the data going to be based on self-reporting? or will there be any cross-validation?
@Anirudh: I believe, your point on impossibility of undertaking such an exercise is a weak argument, because you underestimate the capacity of the Office of the Registrar General of India. However, if you are arguing that a caste-based census, given such a short notice, is impossible to conduct, then I would share your views.
@Alok: I am not quite sure if you correct on the reasons for discontinuing the caste census. Correct me if I am wrong, but the caste census was also carried out in 1941. The financial pressures of the the second world war meant that the exercise was conducted in a half-hearted manner and therefore, the data was not reliable. (I don’t know what happened in 1951). Sorry, I can’t seem to find a reference on this right now.
Aaditya, as regarding your second point, the caste census was last carried out fully in 1931.
I am pretty sure that I have read in a reputable publication that it was discontinued partly because of pressure from nationalist leaders. Secondly. if the second world war was the main reason for its discontinuation, then there’s no reason why it could not have been resumed in 1951 and subsequent decades.
r
It seems to be one of the grey areas. I think it was Galanter that I read this in. A frontline article hints at the point that I trying to make:
The 1941 Census, the last to be conducted by the British, was constrained by the War-time economies, and did not tabulate any data except the basic population totals and community totals.
As “abandonment of caste enumeration seemed a step toward the deestablishment of caste” (Galanter 1984: 260), after India attained independence, enumeration of caste in censuses was dispensed with, as part of national policy, since 1951.
URL: http://www.hinduonnet.com/fline/fl1519/15191010.htm
The ORGI has a comprehensive note on this topic:
Caste, Tribe or Race
The question on Caste, Tribe or Race was asked from each individual right from 1872, though the type of information collected was different in different censuses. In 1881, caste if Hindi; sect; if of othe religion, were recorded. In 1891 Main Caste and Sub-division of caste or race was recorded. In 1901 and 1911 censuses, the caste of Hindus and Jains; tribe or race of those of other religions were recorded. In 1921, 1931 and 1941 censuses, caste, tribe or race of all the individuals enumerated was recorded. In the 1931 Census, tabulation of figures for individual castes was limited to :
(i) Exterior castes;
(ii) Primitive castes; and
(iii) All other castes with the exception of:
(a) those whose members fell short of four per thousand of the total population; and
(b) those for which separate figures were deemed to be unnecessary by the local Government.
Pursuant of the policy of the Govt. Of India to discourage community distinction based on Caste, the 1951 Census marked a complete departure from the traditional recording of Race, Tribe or Caste and the only relevant question on caste or tribe incorporated in the Census Schedule was to enquire if the person enumerated was a member of any ‘Scheduled Caste’, or any ‘Scheduled Tribe’ or any other ‘Backward class’ or if he was an ‘Anglo Indian’.
In 1961 and 1971 Censuses the information was collected only for each Scheduled Caste and Scheduled Tribe.
URL: http://censusindia.gov.in/Data_Products/Library/Indian_perceptive_link/Census_Terms_link/censusterms.html
Wadhwa-
Great article. Excellently argued.
Today’s Hindu contains an article echoing your gvery thoughts. In fact you seem to be in very esteemed company with your point about caste being a moving flexible appellation.
http://www.thehindu.com/opinion/editorial/article627576.ece