Affirmative action is the deliberate use of positive measures to improve the share of goods, offices and standing held by groups that carry the accumulated disadvantage of past discrimination. Its defining feature is that it goes beyond prohibition. A law forbidding an employer to refuse a Dalit applicant removes one obstacle; affirmative action asks what must be done for a group whose obstacles were removed only after generations of exclusion had already shaped who is qualified, who applies and who is believed to belong.
Sociologically, the concept rests on the claim that discrimination is not only an act but a structure — that it leaves behind unequal endowments of land, literacy, networks and confidence which reproduce themselves long after the act ceases. Formal equality treats like cases alike; affirmative action holds that where starting points differ, identical treatment preserves inequality rather than ending it. In Indian constitutional language this is the difference between formal and substantive equality.
Three distinct instruments
Debate is often confused because three different things travel under one name. Anti-discrimination law penalises unequal treatment: it is negative, individual and remedial. Outreach and development measures expand the pool of eligible candidates without altering selection standards — scholarships, hostels, coaching, targeted recruitment advertising, remedial teaching. Only the third instrument, preferential treatment, alters the selection itself, either softly by weighting group membership among many criteria or hard by reserving a fixed proportion of places for members of designated groups.
Quotas are thus a species, not the genus. Much of the transatlantic disagreement about affirmative action is a disagreement about which instrument is legitimate, not about whether historic disadvantage exists.
Four justifications
The compensatory argument treats the benefit as reparation for a wrong: the group was harmed, so the group is owed. Its difficulty is that recipients are rarely the individuals harmed and payers rarely the individuals who harmed.
The distributive argument avoids history and looks at the present: severe under-representation in offices, professions and universities is itself an injustice, and correcting it is a matter of fair shares of social goods.
The diversity argument is instrumental. A bench, a bureaucracy, a faculty or a police force that mirrors the society it serves decides better, is trusted more and enlarges the range of experience brought to public problems. This is the dominant justification in American jurisprudence.
The anti-subordination argument, closest to Ambedkar, is the most sociological. Caste is not merely unequal distribution but a system of ranked status that assigns dishonour. Placing members of stigmatised groups in positions of authority attacks the ranking itself, altering who is imagined to be capable of command. On this view affirmative action is a lever against graded inequality, not a welfare transfer.
The Indian architecture
India's is the world's oldest and most extensive constitutionally mandated scheme, with roots in the communal representation of the colonial period, the Poona Pact of 1932 and the reservations pioneered in Mysore, Bombay and Madras. The Constitution permits the state to make special provision for socially and educationally backward classes and for Scheduled Castes and Tribes in education (Articles 15 and 46) and in public employment (Article 16), reserves seats in the Lok Sabha and state assemblies for SCs and STs (Articles 330 and 332), and abolishes untouchability outright (Article 17). Reservation for Scheduled Castes and Scheduled Tribes was fixed roughly in proportion to population; the Mandal Commission report, implemented in 1990, extended reservation to Other Backward Classes in central services, and the Supreme Court in Indra Sawhney upheld it while excluding the creamy layer and indicating a ceiling of half the available posts. Later extensions — reservation for women in panchayats through the 73rd Amendment, promotion quotas, and the economically weaker sections quota of 2019 — have progressively loosened the original link between reservation and caste-based social stigma.
Comparison sharpens the point. In the United States affirmative action grew from executive orders and civil rights litigation rather than constitutional command, and rigid quotas were held impermissible in Bakke, leaving race as one factor among many in a holistic review — an approach itself curtailed by the Supreme Court in 2023. Indian reservation is therefore both firmer in legal foundation and blunter in instrument.
Criticisms and sociological replies
Four objections recur. The merit objection holds that selection should track ability alone. The reply is that observed merit is itself an outcome — examination performance is heavily conditioned by the school, the language of instruction and the household's cultural capital — and that merit is defined by whoever sets the test.
The stigma objection argues that preference marks its beneficiaries as incapable. The reply notes that stigma preceded reservation, and that visible success in office erodes prejudice more effectively than absence does.
The creamy layer and intra-group inequality objections are the strongest. Benefits concentrate among the better placed strata within a beneficiary category, so that dominant OBC communities and urban Dalit households capture a disproportionate share while the most deprived jatis remain untouched. Income ceilings, the separate treatment of the most backward classes and the sub-categorisation of reserved quotas are responses to this, as are arguments for extending affirmative action into the private sector where most employment now lies.
Finally, the displacement objection: reservation redistributes scarce public posts without expanding opportunity, and can become a substitute for land reform, school quality and job creation. That is less an argument against affirmative action than a reminder of its limits.
An Indian illustration
The reservation of one third of panchayat seats for women, and of sarpanch posts for Scheduled Castes and Tribes, offers a natural experiment. Early observation described sarpanch pati arrangements in which male relatives exercised the real authority — apparent proof of futility. Longitudinal work found something else: repeated exposure to women and Dalit leaders reduced villagers' bias in evaluating their performance, raised the aspirations reported for daughters, and made later election of unreserved candidates from these groups more likely. The lesson is that affirmative action operates on beliefs and eligibility over time, not only on the immediate allocation of a post.
For the UPSC answer
Open by separating anti-discrimination law, outreach and preferential treatment, and note that reservation is only the hardest of the three instruments. Anchor the Indian discussion in Articles 15, 16, 17, 330 and 332, the Mandal moment and Indra Sawhney with its creamy layer exclusion and fifty per cent indication, and use the comparison with the American refusal of quotas to show that affirmative action is not one policy but a family. Handle criticism honestly: concede intra-group capture and answer the merit objection by showing that merit is socially produced. Close with the anti-subordination justification, which connects the policy to Ambedkar and to the sociology of caste rather than to welfare economics.
References & further reading
- Galanter, M. (1984). Competing Equalities: Law and the Backward Classes in India. University of California Press.
- Béteille, A. (1992). The Backward Classes in Contemporary India. Oxford University Press.
- Weisskopf, T. E. (2004). Affirmative Action in the United States and India: A Comparative Perspective. Routledge.
- Deshpande, A. (2013). Affirmative Action in India. Oxford University Press.
- Jaffrelot, C. (2003). India's Silent Revolution: The Rise of the Lower Castes in North India. Permanent Black.
- Sowell, T. (2004). Affirmative Action Around the World: An Empirical Study. Yale University Press.