Mandal II: Marc Galanter Revisited

Marc Galanter’s classic, Competing Equalities: Law and the Backward Classes in India was published in 1984, a few years before the Mandal I controversy erupted in 1990. Many years have gone by, and in the context of the debate on Mandal II, one is tempted to know what would have been his scholarly analysis of the recent events. While an interview with him could bring out his insights on the issue, I decided to revisit one of his latest articles for the same. His article, ‘The Long Half-life of Reservations’ appears in the edited volume, India’s Living Constitution: Ideas, Practices and Controversies, (Zoya Hasan, Permanent Black, 2002). Thanks to Mr.Abi, a regular reader of our blog, I am giving the link to this article here.
Galanter starts with a debate on the terminology to describe the policy. To some extent, he justifies his choice of “compensatory discrimination” in place of ‘affirmative action’. He is mildly critical of those who unabashedly borrow the American term, ‘affirmative action’, which began to be used in the early 1980s. He says: “On first encounter, we might dismiss this new terminology as yet another instance of uncritical adoption of American terms and of a wider and unreciprocated receptivity to American law. However, perhaps there is more here. Conveying at least an oblique reference to the American experience, it signals a tendency to regard Indian developments not as intractably unique, but as a complexly particularized instance of fundamental issues about disadvantaged segments of the population in ‘meritocratic’ democracies.” Gallanter refers to the fact that scholars from the U.S. and South Africa consider the Indian experience as a source of lessons that might be relevant in other settings.
The Supreme Court’s interim order of March 29 has been criticized for its uncritical acclaim for certain judgments of the U.S.Supreme Court. Galanter’s view only strengthens that criticism.
That apart, Galanter tries to grapple with the scope of the OBCs in this way:
Is the OBC category meant to include a wide middle band of castes and communities situated ‘above the Scheduled Castes and Scheduled Tribes, and excluding only prominently well-advantaged groups at the top? Was it intended to catch those whose advantages were far below the average, comparable in circumstance to the SCs and STs? Was the backwardness of the OBCs to be relative to the average or to the top? Mandal, he says, opts for the expansive view, and this has now been ratified by the govt. and the SC. (One wonders a further amplification of what he meant could help better understanding).
Gallanter’s understanding of Mandal I is expressed succinctly here: “Mandal employs caste in both senses. It uses castes as the units whose backwardness is going to be measured (along with jati-like formations among non-Hindus) and it proposes to use caste standing as the measuring rod, for determination of the backwardness of social units. The alternative to castes as units is to measure the backwardness of families or occupations or individuals; a substitute to caste as a measuring rod is to base eligibility on such criteria as poverty/ illiteracy. Obviously, he considered these alternatives imperfect.
Having said that, however, Galanter appears sympathetic to the Mandal II petitioners’ ground of attack against the CEI(RA) Act: lack of data. He says: “The appearance of precision in applying the Mandal Commission’s criteria is dissipated by a serious methodological flaw. These caste groups have not been counted in the census since 1931; the extrapolation of community population figures for half a century, on the assumption that all communities experienced equal growth rates, renders suspect many of the Commission’s findings about relative conditions”.
In the Mandal II debate, a question very often asked by the pro-reservation group is that whether the petitioners believe the OBC population in the country to be well below the 27 per cent. The petitioners so far have not answered this question; but Gallanter appears to have one. His articulation of the problem shows that the petitioners might be in fact concerned with the unsubstantiated, yet genuine fears that the scheme can be so designed to allocate seats over and above (italics his) those gained by the beneficiaries in open competition. Mandal clearly opts for the over and above method, rather than the guaranteed minimum (according to this, if a smaller number of beneficiaries are selected in open competition, the results are adjusted to add additional members of the beneficiary group upto the level of reservation.) “Curiously, so far as I know, the permissibility of these over and above reservations has not been the subject of constitutional challenge in recent years, although a generation ago several courts found them objectionable”, he says. He cites Mandal report as saying: “The chief merit of reservation is not that it will introduce egalitarianism amongs OBC when the rest of the Indian society is seized by all sorts of inequalities. But reservation will certainly erode the hold of higher castes on the services and enable OBCs in general to have a sense of participation in running the affairs of their country.”
Gallanter expresses his agreement with Mandal that compensatory discrimination cannot do everything. “Compensatory preference may be a viable policy for breaking barriers and forced draft inclusion of excluded groups;p it cannot secure a chimerical ‘equality of result’ throughout India’s social order.
A serious programme of compensatory preference, he says, must include measures for self-assessment and a design for it to dismantle itself. There can be no dispute with this view, only the timing of dismantling may be debated. His conclusion: “For all its failures in implementation, India’s policy of compensatory discrimination has been remarkable in its scope and generosity. It is time to upgrade its effectiveness while acknowledging the limits of what can be accomplished with it. “
Join the discussion

This site uses Akismet to reduce spam. Learn how your comment data is processed.