Introduction
The problem of group rights in a liberal democratic state has attracted considerable scholarly attention in recent years. India is home to one of the oldest and most complex policy regimes of affirmative action. This dates back to the late nineteenth century and has considerable sub-national variations in form and extent.Footnote 1 At the time of independence, India's preferential treatment regime was reoriented towards Untouchable and tribal groups (officially known as Scheduled Castes and Scheduled Tribes). Since the 1990s, national affirmative action policies have expanded to include a new set of beneficiaries, namely the Other Backward Classes (OBCs, mainly lower-caste groups).Footnote 2 The expansion of affirmative action has been accompanied by a new-found prominence in the rhetoric of social justice.Footnote 3
This paper focusses on the landmark 1990 parliamentary debate on the government's decision to implement job quotas for the OBCs following the recommendations of the Mandal Commission report (henceforth Mandal debate).Footnote 4 This represented a decisive turning point in state policy of group preference in India, as job quotas instituted by the Indian Constitution of 1950 for the Scheduled Castes and Scheduled Tribes were extended for the first time at the national level to a large new group, the OBCs.Footnote 5 It also marked a defining moment in party political terms, coinciding with the decline of Congress dominance and the emergence of lower-caste-based parties as a significant force in Indian politics.Footnote 6 Most pertinent for our purposes, with the 1990 Mandal debate, the rhetoric of social justice acquired a new prominence in Indian politics and came to be identified with affirmative action for disadvantaged groups.Footnote 7
Drawing upon the results of debates in contemporary political theory on social justice, this paper will flush out conceptions of social justice implicit in political rhetoric and other arguments which were discussed in the 1990 parliamentary debate on job quotas for the lower castes. More specifically, it asks the following questions: How was social justice construed in arguments for and against OBC quotas? Was the appeal for social justice largely rhetorical, or were substantive notions of social justice at play in the case for OBC quotas? Did social justice clash with other values and, if so, how were conflicts between preferential treatment and values such as merit or social cohesion tackled in policy debate?
The approach to the study of political pronouncements adopted in this paper departs from the scholarship on caste and affirmative action in India in the following respects. An important set of writings focus on what might be termed the ‘politics of social justice’ – the growing political clout of lower-caste parties and leaders in Indian politics – and its implications for electoral and party politics as a whole.Footnote 8 Ideas of ‘social justice’ per se are accorded little significance in these important writings being largely construed as malleable instruments in the pursuit of electoral or factional power by politicians. In contrast, this paper undertakes a close conceptual analysis of the rhetoric of social justice, focussing on the internal structure of political arguments rather than the uses to which these were put.Footnote 9 Another important strand in existing literature focusses on the ideas of important lower-caste leaders such Jyotiba Phule, Dr B. R. Ambedkar and Ram Manohar Lohia.Footnote 10 The main aim here was to establish the significance of these individuals as radical political thinkers. Notions of social justice are often subsumed within anti-caste ideologies and have not therefore formed a separate focus of analysis. Again, in contrast, this paper probes the normative significance of caste for social justice. It foregrounds arguments in policy debate. Empirical studies of arguments of affirmative action in India have focussed on the pronouncements of higher courts, tracing the development of reasoning in the rich body of jurisprudence on India's preferential policies for the lower castes.Footnote 11 This paper grounds normative analysis of political arguments in a detailed study of pronouncements in a landmark parliamentary debate.
Legislative debates are set-piece occasions which bring the contending elements in public life on to one big stage thus offering an ideal site from which to access political arguments. Yet, systematic empirical studies of arguments espoused in such parliamentary debates are rare. A common view is that arguments advanced in policy debate are little more than smokescreens employed by politicians to disguise their own interests in power and as such lack causal efficacy. Further more, unlike the pronouncements of great figures or learned judges, the arguments professed in routine policy debate are assumed to be too crude and commonplace to merit scrutiny. This paper challenges these assumptions. It will show that in the course of arguing over policies, Indian politicians produced a rich and sophisticated body of public reasoning linking caste inequalities, social justice and job quotas. A detailed analysis of the role of social justice in political arguments furnishes materials for better explanations of policies of preferential treatment and the politics of lower-caste-based parties in India. It also illuminates important areas of tension in the political theory literature, such as the relationship of social equality with social justice and that of merit with affirmative action.
A couple of clarifications are necessary at the outset. First, this paper seeks to show that political rhetoric merits close analysis irrespective of whether politicians are genuinely committed to the principles they profess. My analysis is compatible with the motivations of politicians being purely strategic, purely normative or a mix of the two.Footnote 12 Second, the essay focusses on the structure of political arguments, restricting information about political actors and the historical context of policy debate to a minimum. To do so is not to claim, implausibly, that arguments espoused in legislative debate are independent of the wider context in which these are located – of course a range of factors external to political arguments influences their shape – but it does claim that focussing primarily and directly on the structure of political argument is a worthwhile exercise that yields important insights for explanatory and normative accounts.
The paper is divided into four sections. The first section unpacks the rhetoric of social justice in the 1990 Mandal debate. It will show that the opposing political positions on quotas for the OBCs were distinguished by competing conceptions of social justice. The second section critically scrutinises a key claim of the proponents of quotas, namely that the under-representation of lower castes in the bureaucracy was unjust, and that claims for OBC quotas in the bureaucracy drew substantially on standard social justice considerations. The third section examines different types of arguments for quotas. Despite appearances, several arguments for OBC quotas were compatible with the principle of merit. The concluding section highlights some implications of the analysis for explanations of lower-caste politics in India and political theory debates on social justice.
The political speeches chosen for discussion in this paper illustrate arguments characteristic of a particular party's discourse. The most commonly expressed opinion of a party is taken to represent its overall position in a debate, but where differing strands of opinion have been recorded individually by its members, these have also been included.
Competing Conceptions of Social Justice
Social justice was the central legitimating concept in political arguments about quotas for Backward Classes in the 1990 Mandal debate. Announcing the government's decision to implement the recommendations of the Mandal Commission report, Prime Minister V. P. Singh termed it ‘a momentous decision of social justice’,Footnote 13 and parties opposing the government's decision were careful to preface their criticisms with avowals of support for the cause of social justice.Footnote 14
What did an appeal to social justice entail? Social justice in the Mandal debate was mainly elaborated in terms of equality, by both the proponents and the opponents of reservations. Like a similar debate on affirmative action in the United States,Footnote 15 the reservations debate in India is largely an ‘intramural debate among partisans of equality’. Closer examination reveals that advocates and critics differed in terms of the kinds of equality they favoured, as well as in the specific policies these required.
In the pronouncements of the supporters of the government's quota policy, ‘social’ justice implied, first, that the inequalities that quotas sought to rectify were not so much economic (concerning the distribution of wealth and income in society), but social, pertaining to social status.Footnote 16 Janata Dal representatives repeatedly asserted that quotas in the bureaucracy were not a means of reducing economic inequalities in society or of securing basic material needs. Ram Awadhesh Singh, a prominent advocate of OBC quotas, stated, ‘It is argued that is there not poverty among the upper castes, don't they need a livelihood. This is not a struggle for livelihood. This is a struggle for a share in state power. . . . This is not an economic question. This is a strange country. Here . . . izzat [respect] is not with wealth but with caste; respect is enhanced when power is attached to it.’Footnote 17 Improving the economic condition of beneficiaries was considered incidental to the aim of quotas.
Second, while social inequalities were associated with economic inequalities, so that the lower rungs of the caste hierarchy were characterised by both economic and social disadvantages, these were not viewed as a consequence of economic inequalities. Rather, the inequalities of status which quotas sought to remedy were seen as deriving from traditional caste hierarchies which had historically excluded ritually inferior groups from education and state power. Further, disparities in the representation of upper and lower castes in government jobs and higher-educational institutions provided the main evidence for the social inequalities that quotas sought to remedy, a point to which we shall return in the next section.
Third, social justice connoted that social groups, rather than individuals, were the units of justice calculation and benefit allocation. Quotas were intended for caste and communal groups included within the OBC category as a whole, irrespective of the differences of socio-economic circumstances within these groups. The assumption was that individuals were disadvantaged on account of their group membership and that the latter must thus be taken into account in rectifying such disadvantages. Proposals for excluding socially or economically advantaged individuals from the OBCs, advanced mostly by Congress and Bharatiya Janata Party (BJP) representatives, were rejected by the prime minister as ‘diluting’ the purpose of reservations.Footnote 18 ‘Social’ justice in the discourse of advocates of OBC quotas thus connoted that social disabilities rather than economic deprivation were being singled out for rectification, that these disabilities were the consequence of social hierarchies, rather than economic arrangements, and that social groups rather than individuals were the relevant units from the standpoint of justice. Construing social justice in these terms had important political consequences: the entire group of OBCs and only the Backward Classes could be identified as beneficiaries of quotas. For once quotas had as their primary aim the rectification of social inequalities deriving from ritual hierarchies, there was no need to exclude economically privileged OBCs or to include economically disadvantaged sections of upper castes within the ambit of job quotas.
The construal of social justice in terms of social equality and as pertaining to groups was most evident in the pronouncements of the ruling Janata Dal and Leftist parties, such as the Communist Party of India (CPI), that were the staunchest advocates of OBC quotas. Despite their avowals to the contrary, however, there continued to be a strong presumption in political discourse of a link between quotas and economic disabilities. A variety of proposals based on economic criteria were put forward with increasing frequency as the debate progressed, mainly by representatives from the opposition Congress party, as well as the government's ally, the BJP. These demanded modifications in the government's policy such as giving priority to the poor among the ‘backward castes’, stipulating means tests for ‘backward caste’ beneficiaries, and instituting quotas for economically weak sections among the upper castes. For instance, Congress spokesman S. S. Ahluwalia argued, ‘Equity and fairplay demand that job reservations and other special measures for the Backward Classes as a whole not be pre-empted by the richer or more privileged segments of Backward Classes. It is equally essential that the poor and deprived of all sections of our people also be given equitable opportunities of advancement.’Footnote 19
In proposals for giving priority to the poor among the OBCs, economic criteria were advocated to supplement caste membership, and there was no conflict between the two on the question of the identification of beneficiaries of quotas.Footnote 20 In proposals to exclude those from economically privileged backgrounds, economic criteria such as means tests were proposed as a standard, independent of, and potentially in conflict with, caste membership for the selection of beneficiaries. Implicit in most proposals that economic criteria be employed for the identification of beneficiaries of quotas was the claim that group membership was not always a reliable indicator of economic deprivation, and group-based quotas were unjust if this were the case.
What we find here is a distinct conception of social justice from that found in Janata Dal speeches.Footnote 21 Here, too, social justice was elaborated in terms of equality. Underlying criticisms of the government's reservation policy as unjust were principles such as basic needs, economic equality and non-discrimination.Footnote 22 The government's proposal was seen as unjust because it did not target those most in need within the OBCs and excluded the needy outside the OBCs: need was evaluated here in terms of economic criteria such as income rather than in terms of recognition or status. The government's policy was also criticised as unjust on the grounds that it would allow the already privileged to accumulate more and leave the position of those who had the least unchanged, thereby widening economic inequalities in society. Finally, the criticism of OBC reservations as unjust invoked the principle of non-discrimination. If economic disadvantage was relevant for identifying beneficiaries of quotas, the exclusion of those similarly disadvantaged simply on account of their (upper)caste backgrounds seemed discriminatory and therefore unjust. Taken together, the principles of basic needs, economic equality and non-discrimination demanded that quotas be directed to the economically deprived and be extended to all those who were similarly deprived, irrespective of their caste affiliation.
In contrast with the case in favour of OBC quotas, social justice was elaborated here in terms of economic, rather than social equality. This meant, to begin with, that the disadvantages that quotas sought to rectify concerned inequalities, mainly in the distribution of economic resources and their consequences for the opportunities and life chances of individuals, rather than inequalities of status. In the pronouncements of Congress and BJP representatives, jobs in the bureaucracy were viewed less in terms of their implications for status and power, and more in terms of their economic aspects. Second, unlike in social equality arguments, the emphasis was not so much on historical discrimination as on current distributive inequalities. While it was accepted that these often overlapped with an inferior position in the ritual hierarchy, the implicit contention was that historical discrimination was neither a necessary nor a sufficient condition for economic deprivation. Third, unlike in the interpretation of social justice in terms of social equality in Janata Dal discourse, the units for the assessment of disadvantage or ‘backwardness’ were individuals, not groups. It was assumed that the individual or the family (assessed as a unitary category), rather than the group, was the appropriate unit of justice calculation and thereby of benefit allocation.Footnote 23
The main bone of contention here was the caste basis of the quota policy. Criticisms of the government's policy as ‘casteist’ attacked both the employment of caste groups as the units of justice calculations and a caste's position in the ritual hierarchy as an indicator of disadvantage.Footnote 24 A caste-based reservation policy was unjust, as it suffered from the shortcomings of both over-inclusion and under-inclusion: justice required that economically advantaged members of the OBCs be excluded and economically deprived individuals from non-OBC groups be included within the purview of quotas. In the use of caste position as an indicator of disadvantage, the Mandal debate reflected a decline in the importance of the religious disabilities of a caste as criteria of ‘backwardness’. Instead, the secular profile of a caste, involving criteria such as literacy, poverty and representation in administrations and legislatures, had gained ascendancy in descriptions of ‘backwardness’.
Under-representation, Social Justice and Democracy
So far, I have contrasted two conceptions of social justice in the Mandal debate, one which focussed on status inequalities and another which emphasized economic inequalities. It might however be contended that my characterization of social equality arguments as a type of social justice claim is too quick. Theorists of social justice have held that social equality, the ideal of a society free from hierarchies, is ‘freestanding and independent of justice.’Footnote 25 Analysts of Indian politics have contended that quotas have been advocated in recent times simply as a means of rectifying inequalities in group representation, that group equality has been interpreted in ways that have little to do with disadvantage or discrimination.Footnote 26 The relationship of social equality and social justice needs closer scrutiny.
Nearly every political speech advocating OBC quotas cited the under-representation of lower castes in the bureaucracy in support of its case. The statistics cited referred to the representation of group members as a percentage of the total number of administrative positions, or of positions at a particular level of the administration in proportion to the demographic weight of the group and relative to other groups, notably the upper castes.Footnote 27 Now, while ‘the under-representation of certain categories of people is often so stark that its injustice seems beyond question’,Footnote 28 the claim that disproportional group presence in the administration represents an injustice can easily be challenged.
First, a group might be under-represented in the bureaucracy because the population in the group which has the education and skills for bureaucratic jobs is low compared with groups which are better represented in the bureaucracy.Footnote 29 Under-representation here would result even in the absence of direct discrimination against the group. Second, the belief that a group's representation in the bureaucracy must approximate its demographic percentage seems to assume that people equally desire bureaucratic jobs. Members of different social groups, however, may differ in their disposition to seek bureaucratic jobs because all cultures do not place an equal value on such jobs. In such a situation, disproportionalities in group presence in the bureaucracy might reflect the diversity in preferences between social groups, rather than injustice.Footnote 30 Third, it might be argued that as the bureaucracy comprises mainly middle-class jobs, the social composition of the bureaucracy is going to mirror the demographic profile ‘only if all major social groups are distributed equally, in proportional terms, along the social stratification system’.Footnote 31 Therefore, disproportional group representation in the bureaucracy would occur even in the absence of any discrimination against members of the group.
Each of these objections, however, can be countered from a standpoint of social justice. First, the lack of a pool of qualified candidates from a group, as well as the unequal distribution of groups along the stratification system, could indicate the absence of fair equality of opportunity in education and employment. Fair equality of opportunity requires, in John Rawls' well-known formulation, ‘not merely that public offices and social positions be open in the formal sense, but that all should have a fair chance to attain them. . . . In all parts of society, there are to be roughly the same prospects of culture and achievement for those similarly motivated and endowed’.Footnote 32 On this more demanding version of equality of opportunity, the under-representation of a group in the bureaucracy could constitute evidence of injustice even in the absence of direct discrimination, for it would suggest inequalities in opportunities for education and training along group lines. Second, there is the issue of adaptive preferences: people adapt their preferences to ‘excessive limitations in opportunities or unjust background conditions’.Footnote 33 Members of culturally inferior or economically disadvantaged groups might not aspire for prestigious positions considered unattainable; they often lack the motivation and confidence to seek such positions. In other words, differences in preferences along group lines, as scholars have noted, might themselves be the product of the interaction of cultural and economic injustices, of ‘injustices of recognition and injustices of distribution’.Footnote 34 From a social justice standpoint, thus, the under-representation of a group in the bureaucracy while not constitutive of injustice is likely to be indicative of it.Footnote 35
This was the type of move that advocates of OBC quotas in the Mandal debate were making. Under-representation of lower castes, it was asserted, was a feature not only of bureaucratic jobs but also of higher education – the statistics cited sought to make the case that the higher one went up the educational ladder, the lower was the representation of OBCs. This in turn was thought to indicate massive disparities in educational opportunities between lower and upper castes. Cultural differences in preferences did exist which prevented lower castes from seeking higher education and administrative jobs which meant that these groups predominated in agricultural and manual labour occupations. These, however, were undesired differences, the result of stunted aspirations in a context in which access to education and state power was denied to lower castes over several generations. Finally, while arguments for OBC quotas emphasized inequalities in status and power, they also emphasized the fact that there was considerable overlap of membership at the lower levels of caste and class hierarchies. Lower castes were predominantly lower class as a result, it was held, of historical discrimination.Footnote 36 Caste thus was seen to have both class and cultural effects which resulted in inequalities in opportunities for members of lower castes. The contention was not that the under-representation of a group constituted an injustice in and of itself – the claim was that justice required the proportionate presence of all groups in every arena.Footnote 37 While statistics of disproportionate group representation served as a powerful rhetorical shorthand here, the underlying normative claim was about systemic socio-economic disadvantages which corresponded to the traditional caste hierarchy.
The claim that disproportionate representation in the bureaucracy was unjust, sometimes took another form. Here, it was urged that the under-representation of the lower castes was undemocratic as it signified a denial of the equal right of each individual to participate in government, and was therefore unjust. Indrajit Gupta, a veteran parliamentarian from the CPI quoted approvingly from the Mandal Commission report:
In a democratic set-up, every individual and community has a legitimate right and aspiration to participate in ruling this country. Any situation which results in a near-denial of this right to nearly 52 per cent of the country's population needs to be urgently rectified. . . . 52 per cent of this population which we are talking about at present, is supposed to enjoy only 4.5 per cent of the top grade jobs. Is this not gross injustice?Footnote 38
Again, the general claim, that disproportional representation in the administration signifies a denial of the equal right of each individual to participate in government, is problematic. First, as scholars have shown, the principle of political equality is compatible with a range of institutional mechanisms; these may seek to equalize either opportunities for political influence or power over political outcomes, and the two may not coincide.Footnote 39 More generally, egalitarian procedures such as equal voting rights regularly produce unequal outcomes in terms, for instance, of the social composition of political assemblies. Group under-representation as such does not necessarily prove that that the norm of political equality or some form of procedural equality has been violated. A second reason for scepticism towards the claim that disproportional group representation violates the equal right to participate in government relates to the distinction, emphasized most notably by Hanna Pitkin and Anne Phillips, between participation and representation.Footnote 40 The notion of equality of presence is implicit in the idea of participation, as everyone can, in principle, participate. The same, however, is not true of representation, which rests on the principle of the few representing the many. While proportionate representation of all social groups may be justified, it does not automatically follow from an equal right to participate in politics.Footnote 41
A third objection to the claim that the under-representation of a group in the bureaucracy violates an equal right to participate in politics stems from the fact that the bureaucracy is not a representative political institution. Throughout the 1990 Mandal debate, the bureaucracy was characterized in terms which rendered it indistinguishable from political assemblies – as a ‘decision-making institution’, ‘institution of governance’. These characterisations facilitated the application of norms of political representation to bureaucratic recruitment and undermined, as we shall see, claims of merit. It might however be argued that group representation in the bureaucracy cannot be judged by the same criteria as political representation; indeed, we ought not to be concerned with the social composition of the bureaucracy at all because bureaucrats are not expected to perform a representative function, unlike those elected to political assemblies.Footnote 42
In response to these objections, it should be noted, first, that the claim that the under-representation of the lower castes violated their equal right to participate in politics was a variant of the argument from fair equality of opportunity discussed above. Disproportionate representation was seen as evidence that inequalities in opportunities persisted. As the arena in which opportunities were unequal had to do with political decision-making, and the effect of the inequalities was to deprive the majority of the population of the opportunity to participate in the exercise of political power, under-representation was simultaneously undemocratic. Second, it is important to recall that the equal right to participate in government plays an important symbolic role in democratic societies, that of affirming the equal status of all citizens: it has become ‘an essential expression of the basic equality between members of each state’.Footnote 43 When the under-representation of certain groups in arenas of state power occurs against a historical background of their exclusion on account of low status, it seems to imply, as Anne Phillips has noted, that members of these groups are inferior to others, less fit to govern than the rest.Footnote 44 In the claim that the under-representation of the lower castes in the administration constituted a denial of the equal right to participate in politics, part of what was at issue was that the latter served to exemplify the ideal of equal status, and it was this symbolic sense of political equality that under-representation was thought to offend. CPI parliamentarian Indrajit Gupta quoted the following passage from the Mandal Commission report:
[A]n essential part of the battle against social backwardness is to be fought in the minds of the backward people themselves. In India, the Government service has always been looked upon as a symbol of prestige and power. By increasing the representation of OBCs in Government services, we give them an immediate feeling of participation in the governance of this country. . . . Even when no tangible benefits may flow to the community at large, the feeling that now it has its own man in the ‘corridors of power’ acts as a morale booster.Footnote 45
Democracy and social justice were thus closely entwined in arguments advocating OBC quotas. Democratic values of popular participation and political equality were constitutive of notions of justice. A just society was in part a society in which the people participated in the making of political decisions, and no one was considered inferior on account of their social background. That unequal opportunities of the ‘backward sections’ meant that the majority of the population faced barriers to institutions of political power, and were deprived of recognition of their equal status, enhanced the injustice of under-representation. Social justice and democracy were also related analogically in arguments for OBC quotas.Footnote 46 As in the social justice argument, the claim in the democratic argument was not that the under-representation of a social group was undemocratic per se, more that in a context in which lower castes had been excluded from institutions of government for centuries on grounds of inferiority, their current under-representation in the bureaucracy was undemocratic.Footnote 47
In under-representation claims, the normative work done by democracy pertained primarily to the denial of equal opportunities to exercise political power and equal status and dignity: the substantive representation of the interests of lower castes in government was not the central issue. It was because the underlying concern in under-representation claims was less with representation in the sense of the inclusion of a distinctive point of view and more with reversing the exclusion of lower castes from arenas of state power and the presumption that members of these groups were inferior, that the bureaucracy was cast as simply another political institution, with group presence in the administration treated as identical with representation in political assemblies.
In the case in favour of OBC quotas, thus, under-representation remained an indicator of disadvantage. Quotas in post-independence India have only been justified as a means of redressing disadvantage and not for enabling power-sharing between different groups, as was the case in the colonial period. Defenders of OBC quotas departed from standard social justice arguments in maintaining that the caste hierarchy rather than the economic structure of society was the fundamental cause of inequalities of opportunities and, in holding that social justice, required the greater equalisation of status, respect and power and not just material goods.Footnote 48 Nevertheless, in relying substantially on notions of fair equality of opportunity, arguments for OBC quotas were akin to familiar social justice claims.
Social Justice and Arguments for Quotas
Even if it is accepted that the under-representation of lower castes in the administration constitutes an injustice that needs to be rectified, this does not necessarily establish the case for quotas. Policies of equal opportunity for all individuals in employment may be regarded as sufficient, or other forms of preferential treatment such as financial assistance may be the preferred remedy. How were quotas construed as a requirement of justice in the Mandal debate?Footnote 49
Adapting Galanter's classification, three main kinds of justice arguments for quotas can be discerned: compensatory justice, non-discrimination and general welfare.Footnote 50 In the Mandal debate, compensatory justice considerations were rarely deployed as an independent argument for quotas; these were usually invoked to reinforce arguments from systematic group disadvantage. Quotas were mainly advocated as a requirement of social justice in the Mandal debate, in individual fairness and general welfare types of arguments. Importantly, despite appearances, several arguments in favour of preferential treatment in government jobs in the Mandal debate were compatible with the principle of merit.
Justice and Group Fairness: Compensation as Atonement
Janata Dal spokesmen sometimes suggested that it was because past injustices against the lower castes had assumed the form of exclusion from institutions of state power that justice now required that these groups be included in the bureaucracy through quotas.Footnote 51 Quotas were the compensation owed to the victims of past injustices and were the preferred form of reparation because they effected the transfer of unjustly acquired benefits – government positions – from the wrongdoers (upper castes) to the victims (lower castes), thereby making good this loss in some measure.
An interesting feature of compensatory justice arguments in India has been that Hindu beliefs regarding sin have been invoked to make the claim that the current generation of upper castes ought to bear the costs for the oppression inflicted by their forefathers. One way in which quotas and other forms of preferential treatment have been defended is as a form of atonement by the current generation for the misdeeds of their ancestors. This offers a way out of a major problem of compensatory justice arguments, notably how members of the current generation can be made to bear the costs of such provisions, when they were not the agents responsible for historical injustices.Footnote 52
The idiom of atonement for past sins was less prominent in the Mandal debate than in earlier debates on preferential policies.Footnote 53 Partly, this reflected a change in the register of political claims – the case for quotas as compensation was no longer addressed to upper-caste Hindus in the form of appeal for self-reform or paternalistic, philanthropic concern for underprivileged lower castes. Now, the demand for quotas as compensation was couched in a language of rightful entitlement. Also, compensatory justice considerations appeared less powerful in the case of OBCs than that of Dalits. Most importantly, however, the case for OBC quotas relied predominantly on social justice rather than compensatory considerations.Footnote 54 Social justice was invoked in arguments of both individual fairness and general welfare types, some of which challenged merit-based objections to quotas.
Social Justice and Merit
Preferential treatment is commonly thought to conflict with the principle of merit, namely that jobs be assigned to the best qualified individual. Arguments for why jobs should be allocated on merit broadly divide into two camps: individual justice arguments, which claim that the best qualified applicant deserves the job,Footnote 55 and utilitarian arguments, which hold that considerations of utility, in terms of the efficient performance of institutions for instance, demand that the best qualified person be hired.Footnote 56 The speeches of the advocates of OBC quotas in the Mandal debate often attacked merit. Closer examination, however, reveals that the principle of merit was often invoked in support of OBC quotas.
One line of argument sought to redefine job requirements and thereby what would count as a relevant qualification for a job.Footnote 57 Requirements for bureaucratic positions were redefined in line with preferred social outcomes, such as making the administration more responsive or better informed about the needs of disadvantaged groups. Once job requirements were recast in these ways, membership of a disadvantaged group could become a relevant qualification for positions in the bureaucracy. Members of such groups could be seen as better qualified for these jobs than non-members because they served the social function of the job better, and quotas for Backward Classes would no longer compromise merit but enable the identification and hiring of the best qualified individuals for the job. Prime Minister V. P. Singh argued:
[W]hat is the criticism about the administration or bureaucracy today? Not that it is not knowledgeable or that it is mediocre . . . [but] that it does not care. A mother is a mother, not because she is an intellectual . . . [but] because she cares; and that is what people look for, in the administration. And if that be so, we will have to re-define what is administrative merit . . . to put a social content into it . . . . [T]hose who have gone through suffering, those who know the pinch . . . if they are in the administration, they will be more responsive when the oppressed and the weaker sections are in the bureaucracy.Footnote 58
Here, merit-based objections to quotas were countered through a redefinition of the goals of the bureaucracy and correspondingly of the merit required for administrative positions. Against the contention that quotas would diminish the efficiency of the administration and undermine the quality of services provided by the bureaucracy, it was urged that preferential hiring of members of the Backward Classes who understood the needs of disadvantaged groups better would improve the performance of the bureaucracy.
Another common merit argument in support of OBC quotas in Janata Dal discourse relied on considerations of desert. ‘Merit’ which was measured by performance in standard tests, it was argued, was the product of favourable educational and socio-economic circumstances. Standard merit tests were not a good measure of the abilities or the efforts of these candidates: they over-represented the potential of candidates from privileged backgrounds and under-represented the potential of members of Backward Classes. Quotas sought to compensate for inequalities in performance, which were the product of unfair social circumstances, rather than differences in ability. This argument did not contest the notion that jobs ought to be awarded to those who deserved them on account of their abilities and efforts. Rather, it was argued that gifted individuals from lower-caste backgrounds had historically been prevented from realising positions commensurate with their talents by the caste system.Footnote 59 Furthermore, massive and inherited inequalities in economic, social and educational opportunities persisted along caste lines, preventing individuals from the Backward Classes from achieving results which corresponded to their abilities and efforts, so that their performance did not reflect their true deserts. This implied that a Backward Class candidate with a lower grade might deserve a job in preference to a general-category candidate with a slightly higher grade on grounds of merit, as the former's grade predicted a better job performance than the latter's. It also implied that similar grades might be indicative of a difference in qualifications: identical grades could suggest that a candidate from a disadvantaged group was better qualified for the job than one from a privileged group as she ‘had to overcome more severe obstacles to acquire those credentials’.Footnote 60 Janata Dal spokesman Jaipal Reddy held:
Can merit be . . . assessed in a socio-educational vacuum? This question was put by a Supreme Court judge in Vasant Kumar's case and I would quote: ‘Is not a child of the Scheduled Castes, Scheduled Tribes or other Backward Classes who has been brought up in an atmosphere of penury, illiteracy and anti-culture, who is looked down upon by tradition and society, who has no books and magazines to read at home no radio to listen, no TV to watch, no one to help him with his homework, who goes to the nearest local board school and college, whose parents are either illiterate or so ignorant and ill-informed that he cannot even hope to seek their advice on any matter of importance . . . has not this child got merit if he, with all his disadvantages, is able to secure the qualifying 40 per cent or 50 per cent of the marks at a competitive examination where the children of the upper classes who have all the advantages . . . and who have perhaps been specially coached for the examination may secure 70, 80 or even 90 per cent of the marks? Surely, a child who has been able to jump so many hurdles may be expected to do better and better as he progresses in life?’Footnote 61
‘Treating unequals equally is the greatest injustice’ was a frequent refrain of Prime Minister V. P. Singh in the Mandal debate. Such claims relied on the principle of fair equality of opportunity discussed earlier and defended quotas as a corrective to policies of equal opportunity in a context characterised by socio-economic inequalities.Footnote 62
It is important to note that this line of argument provided support for preferential treatment as a means of rectifying unfairness in the competition for desired jobs on account of inequalities not just in basic goods but also in secondary goods, such as higher education. As Gutmann and Thompson have noted, inequalities in education and employment are correlated and persist in a cyclical way across generations, and the requirement for many prestigious, highly paid jobs in society is success in higher education.Footnote 63 While advocates of OBC quotas often cited low levels of income and primary education and lack of housing and other basic goods among lower castes,Footnote 64 the case for preferential treatment on grounds of non-discrimination primarily focussed on breaking the cycle of inequalities in educational and employment opportunities associated with caste-based discrimination.
Merit-based objections to quotas were also countered in the Mandal debate through criticisms of the criteria of merit that informed standard tests and popular conceptions of merit. Thus, it was argued that selection procedures did not provide a level playing field, as criteria such as fluency in English or the importance attached to interview performance placed members of Backward Classes, who were mostly rural and non-English-speaking, at a disadvantage. While couched as an attack on merit, the underlying argument did not reject the principle of merit. Rather, it implied that many of the criteria attached formally or informally to administrative jobs in which Backward Class candidates found themselves at a disadvantage were not, strictly speaking, relevant to the job, in the sense that the job could be carried out equally well without them. So modifying these criteria in order to facilitate the entry of more Backward Class candidates would not detract from the principle of merit.Footnote 65
Speeches in favour of OBC quotas in the Mandal debate also attacked the historical associations of merit-based arguments. It was held, for instance, that the claim that Indians were not capable of self-government had served as one of the main ideological props of British colonial rule.Footnote 66 Lower castes had been denied education and political power for centuries on grounds that they lacked ‘merit’ – the requisite capacities and skills. While such claims sought to discredit merit-based objections to quotas, they did so by attacking the unmeritocratic elements (racial superiority, hereditary privilege) of systems which claimed to be meritocratic – and, in doing so, affirmed the principle of merit.
So far, I have sought to show that contrary to appearances, several arguments in favour of OBC quotas in the Mandal debate invoked principles of meritocracy. This, however, is not to suggest that all arguments for quotas in the Mandal debate sought only to protect merit. In general welfare arguments discussed below, quotas implied departures from policies of hiring the best qualified in order to promote desired social goals. Other arguments rejected the principle of merit more directly. For instance, it was argued that jobs were a training ground for skills and hence ought to be awarded to those who were lacking in skills. Janata Dal MP Hukumdeo Narayan Yadav declared:
You talk of qualifications and competition for getting jobs. This is like telling a person to learn swimming before he is allowed to jump into the swimming pool. If that is so, where will he learn the swimming? . . . Dr Lohia's philosophy was that whether such a person is competent or not first he should be given a job. In due course of time, he will become competent for holding such a job . . . . The people belonging to upper castes, have behind them thousands years’ old ‘Sanskaras’. We have got no such ‘Sanskaras’.Footnote 67
This understanding of the social purposes served by bureaucratic jobs implied that the least qualified had the best claims on positions and was at odds with the meritocratic view of hiring the best qualified person for the job. Implicit in such arguments was the assumption that given a fair chance, anyone could acquire the requisite qualifications: there could be no differentials of qualification arising from differences in talents and efforts between individuals because there were no differences of ability that were not a consequence of unjust social arrangements. This view was indicative of the strong egalitarian tenor of arguments for quotas in the Mandal debate, where social justice was identified with promoting the claims of the disadvantaged and not with ‘protecting the claims of the most qualified’.Footnote 68 While the requirements of these two can clearly conflict, the arguments of the advocates of OBC quotas also illustrate that the normative underpinnings of policies of group preference and hiring by merit are closer than appearances suggest.
While a detailed comparison with other political debates is outside the scope of this paper, it is important to note that the advocacy of quotas in terms of the principles underlying policies of merit in the 1990 Mandal debate marked a shift from earlier political arguments. Whereas at the time of making the Constitution, for example, quotas were largely seen as detracting from merit and equality of opportunity, albeit admissible for promoting desired social goals, within an egalitarian conception of social justice professed by Janata Dal politicians, quotas were now rendered consonant with principles of merit and equality of opportunity.Footnote 69 The case in favour of quotas was strengthened, as these appeared no longer as exceptions to desired norms but as mechanisms necessary for their realisation. At the same time, however, to the extent that it emphasized inequalities of opportunities faced by individuals from disadvantaged social backgrounds, the case in favour of OBC quotas became more vulnerable to criticisms of unfairness, for the inclusion of individuals from privileged OBC families and the exclusion of upper-caste poor.
Social Justice and the Empowerment of the ‘Backward’
OBC quotas were also advocated on social justice grounds in the Mandal debate as a means to the realisation of a more just society in the future and, as such, in consequentialist justice arguments. What was the vision of a just society, and how would quotas contribute to the realisation of this goal? When quotas for Untouchables and Tribals were introduced in the Indian Constitution in the late 1940s, consequentialist justice considerations had been prominent. A just society was envisaged, broadly speaking, as one in which there would be fewer socio-economic disparities between groups.Footnote 70 The mechanism whereby quotas would further this goal was not spelt out in detail but was broadly related to improvement of the material conditions of beneficiaries. In Janata Dal discourse in the 1990 Mandal debate, by contrast, as we have seen, a just society was envisaged as one in which there would be fewer inequalities not only of economic resources but also of status and power. The desired goal was defined as a society which was democratic as it was just, where the equal worth of all citizens was affirmed, and the distribution of power was more equal. Furthermore, the mechanism whereby quotas were thought to contribute to the creation of a more just society was not so much by improving the economic well-being of the lower castes but by giving them political power. Elaborating on the rationale of quotas, Prime Minister V. P. Singh said:
[I]t is not so much the issue of economic condition but of the power structure – the power structure which has crystallized in the social structure . . . Bureaucracy is a very important component of this whole power structure . . . [I]f you take the strength of the whole of the Government employees as a proportion of the population, it will be 1 per cent or 2 per cent. We are under no illusion that this . . . will resolve the economic problems of the whole section of 52 per cent . . . We consciously want to give them a position in the decision-making of the country, a share in the power structure.Footnote 71
Precisely how the transfer of political power would bring about social transformation was not specified. It appeared to offer at one stroke, symbolic benefits of respect and agency, enhanced employment and educational opportunities and the prospect of better policies. What was clear was that political power for the underprivileged was envisaged as the main agency and motor for the creation of a just society and was advocated as the solution for all injustices in society. Oppression – the deprivation of political power – was seen as the means through which all forms of injustice, social denigration and economic exploitation alike, operated. The key to the removal of injustices was placing the oppressed – lower castes, the poor – in positions of political power. Responding to criticisms of quotas for the OBCs, Prime Minister V. P. Singh proposed additional quotas be instituted for the poor: ‘Without giving power, a poor man cannot remove his poverty . . . . Poverty is not only an economic problem, poverty is a political problem. Give political power and the poor will make their own way.’Footnote 72 The emphasis on the centrality of political power was not a new theme: Janata Dal discourse was here drawing upon the ideological work of low-caste leaders, parties and movements dating back over a century. What was new in the 1990 Mandal debate was that this theme had moved from the periphery to the centre ground of policy-making.
The animating democratic ideal here was that of participative democracy: the disadvantaged had to be granted control over decisions, by ensuring their presence in political institutions. This meant, first, that the hitherto dominant paternalist tenor of welfarist arguments was weakened. Unlike in earlier quota arguments, ‘backward sections’ were not cast in the role of passive objects of benevolent state action, recipients of philanthropic initiatives directed at them, but were to be agents of their own betterment. Second, the participative ideal carried with it a mandate for an extensive application of democratic norms; it demanded that these be applied to all institutions of power, not just popular assemblies.Footnote 73 Hence the bureaucracy, as an institution of state decision-making, could be treated as akin to elected bodies, bureaucratic recruitment assessed by criteria of democratic participation rather than merit.
The ideological moves we have identified in Janata Dal discourse in the Mandal debate strengthened the case in favour of quotas in the bureaucracy. Quotas were no longer seen as only one means among others and ineffectual in themselves for ameliorating the vast socio-economic disparities in society. Rather, quotas became the pre-eminent instrument of social justice because political power for the ‘backward’ was now central both to how the desired goal of a just society was defined and to the process through which it was to be brought about. By affording representation to the OBCs in a key decision-making institution, quotas were already accomplishing a large part of their purpose. How this power was subsequently used, whether it contributed to the reduction of socio-economic inequalities were separate issues; the case for quotas did not wholly depend on these.
This marked an important shift in general welfare arguments for quotas in at least two respects. First, the literal empowerment of the disadvantaged became fundamental to the characterization and realisation of a just society. Second, social justice was extricated from the embrace of national unity which had characterized earlier Congress arguments on affirmative action. These respects are brought into sharp relief when Janata Dal discourse is counter-posed with arguments advanced by representatives from parties more equivocal in their support for OBC quotas. In one argument, prominent in the speeches of parliamentarians from the Left party CPI (M), the desired goal was defined in terms of a society with fewer disparities in the distribution of socio-economic goods – income, education, health and the like. This too was a consequentialist justice argument. However, political power was not of paramount importance here. Quotas in the bureaucracy were valued not so much for ensuring the representation of disadvantaged groups but as thereby contributing to the creation of a society in which there was less poverty and economic inequality. Greater representation of the OBCs in the bureaucracy would create a sense of empowerment among the disadvantaged and as such was a good thing. However, quotas were just one among numerous measures for tackling massive distributive inequalities in society and, unlike in Janata Dal discourse, were not seen as having any special efficacy. Inasmuch as quotas had little impact on the more fundamental material inequalities, they were a limited instrument of social justice. CPI (M) leader Somnath Chatterjee held:
[I]t is essential that there has to be an all round economic development directed towards alleviating principally the economic distress, social backwardness, of these poor people in this country. . . . [T]his mere reservation will not solve the problem. It will be a good step, it will give them an impression that they will be participating more effectively in the governance of the country.Footnote 74
Another set of general welfare arguments for quotas, most prominent in the speeches of Congress and BJP representatives, emphasized the interests of the nation. The desired goal was defined in terms of national unity and development. It was argued for instance that by improving the conditions of the ‘backward’, quotas would harness their energies for the nation's development.Footnote 75 This was not fundamentally a justice argument: the creation of a more just society, while desired here, was valued not so much for itself as for its contribution to national unity and development. The rationale of national good for quotas was most cogently articulated by Congress Rajya Sabha MP Rajmohan Gandhi:
After we attained Independence . . . the Government has rightly concentrated, on this section and the whole House has agreed that this section is largely backward and has suffered for such a long time . . . . But the question is this: Are we supporting and sheltering that section only for its own sake or are we supporting and sheltering that section so that it can find itself equal in relation to the other sections? These are two ways of looking at it. The third way of looking at it is to realise that we are sheltering and supporting that section which is a part because we want to strengthen the whole. Today, at this time, when our nation's unity is assailed from many quarters, it is all the more important that we are absolutely clear that we should strengthen and support a part because we want to preserve the unity and strength of the whole.Footnote 76
This had been the dominant argument in favour of quotas when these were introduced in the Constitution in the late 1940s. Quotas, it had been argued, would enable the ‘upliftment’ of ‘backward sections’ and thereby their closer integration with the rest of the population, as well as national progress – India could not be a developed nation if large sections of its population remained ‘backward’.Footnote 77 By contrast, in Janata Dal discourse in the 1990 Mandal debate, the consequences of OBC quotas for national integration and development were marginal. This marked a major shift in the justification of preferential treatment in India. It meant that social justice was now a goal which was valued in its own right, the pre-eminent appeal in political argument. Considerations of national unity and development could be overridden in the interests of social justice. Correspondingly, we find few worries among the proponents of OBC quotas that these would consolidate caste divisions, detract from a sense of belonging to a common nation or fuel social conflict. Such concerns had dominated debates on quotas in the late 1940s and 1950s and continued to be voiced by Congress and BJP representatives in the Mandal debate. Janata Dal representatives by contrast demonstrated a greater willingness to tolerate costs for political stability and social cohesion that a commitment to social transformation might entail. A few speeches did seek to reconcile social justice and national unity; this however was not a central concern.Footnote 78
While the ideological shifts in general welfare arguments in Janata Dal discourse made the case for OBC quotas in the bureaucracy stronger in some respects as noted above, these also carried some costs. Given that the goal of a just society also included the reduction of socio-economic disparities, the policy of OBC quotas was vulnerable for its neglect of the welfare of the economically disadvantaged. Importantly, with the uncoupling of social justice and national unity in Janata Dal discourse, a powerful account of how preferential treatment served the common good was lost. An alternative account of how a more just society would benefit all citizens was rarely elaborated by defenders of quotas, although materials can be discerned in several speeches.Footnote 79 As such, the advocacy of OBC quotas remained susceptible to the charge that it was the instrument of a sectional interest seeking power, its appeal to social justice largely rhetorical.
Conclusions
Political rhetoric is rarely the subject of close scrutiny and is usually regarded as little more than a tool of political expediency. This paper has undertaken a detailed reconstruction of political rhetoric as argument. Focussing on a landmark legislative debate, it has shown that Indian politicians, in the course of arguing over preferential policies, produced a rich and sophisticated body of reasoning on social justice and quotas. This involved complex negotiations between individual fairness considerations and group-based mechanisms such as quotas and between standard liberal commitments to equality of opportunities and egalitarian liberal concerns with inequalities of results.Footnote 80 In conclusion, and by way of illustration, I now want to suggest some ways in which the articulation of norms implicit in political rhetoric undertaken here helps illuminate discussions of social justice in Indian politics and in political theory.
In relation to Indian politics, we saw, to start with, that the contention over OBC quotas was not simply about politicians jockeying for power: there were substantive and distinct conceptions of social justice at play in the Mandal debate. The competition for power between opposing parties on OBC quotas was structured around distinct conceptions of social justice and different views about its importance relative to other political ideals, such as national unity and development. That policy debate involved differing values suggests that political conflict would tend to recur: OBC quotas are a key political issue today as they were in 1990, even as the political context and main actors have changed. Second, our analysis also offers a better grip on the evolution of quota policy since 1990. The broad direction of change has been one of expansion – notably OBC quotas were extended in 2008 to institutions of higher education.Footnote 81 We saw that social equality arguments in Janata Dal discourse offered a much stronger basis for OBC quotas than had been available previously, undermining merit and national unity objections. At the same time, these relied heavily on the class effects of caste discrimination and remained vulnerable to criticisms on grounds of distributive injustice. That the enlargement of quotas has been accompanied by attempts to institute economic criteria for the selection of beneficiaries is not surprising and is likely to continue. Most notably, the Supreme Court judgement in 1992 upheld quotas for OBCs but asked for the exclusion of economically well-off OBCs (the so-called creamy layer).Footnote 82 The ‘creamy layer’ criterion is a key policy development: OBC policies now use ascriptive groups while seeking to exclude well-off individuals – and one that attempts to bridge the tensions in the case for OBC quotas between caste/class and group/individual criteria. Third, my analysis furnishes materials which shed light on some puzzling features of lower-caste-based party politics in India. Lower caste-based political parties have become an established feature of the polity since the 1990s, despite the fact that the parties themselves have often been short-lived and prone to fragmentation.Footnote 83 We saw that in the course of advocating OBC quotas, Janata Dal representatives recast social justice in terms of social equality and accorded it primacy in the legitimating vocabulary. This represented the demarcation of a distinct ideological domain for centre-left caste-based politics that differed from other parties that emphasized distributive inequalities or national unity considerations. At the same time, our analysis also suggests possible reasons why the lower-caste centre-left parties have been less successful in becoming national, all-India parties. The vision of social justice espoused by the Janata Dal focussed overwhelmingly on caste, whose components vary regionally. Little attempt was made to provide an account of how overcoming caste oppression was in the interest of the nation as a whole or of citizens in general.Footnote 84 As such, the ideological resources of the kind fashioned in the Mandal debate provided a basis for the resilience of lower-caste-based centre-left parties but not to their expansion into all-India parties. More empirical work is required to establish these claims. Nevertheless, my analysis suggests that ideological explanations offer rich potential to advance the understanding of lower-caste politics.
Social justice has also been a central preoccupation in contemporary political theory. Among advocates of social justice, the focus has been on distributive equality, i.e. the distribution of economic resources among individuals in a society, which in turn has often been contrasted with social equality or equality of status.Footnote 85 The findings of this paper support the strand of egalitarian scholarship that emphasizes continuities between distributive and social equality and underlines the significance of the latter for social justice.Footnote 86 First, as advocates of OBC quotas highlighted, inequalities of status, like distributive inequalities, can undermine equality of opportunity in education and employment. Second, the characterization of social equality as ‘group-based’ needs qualification. Social equality arguments in the Mandal debate derived their normative force chiefly from the denial of equal opportunities and equal worth, to individuals; these did not affirm the normative significance of groups as such. Third, social equality advocates in Indian debates held that the distribution of respect and power in society are also concerns of social justice, not just the distribution of material goods. The extent to which the distribution of the democratic goods of respect and power are matters for social justice is a question that recent egalitarian scholarship has begun to address, although a full treatment is still awaited. At the same time, the findings of this paper also suggest that it is useful to retain the conceptual distinction between social and distributive equality, to hold these as distinct, albeit related, ideals. Proponents of social equality in the Mandal debate differed from those who emphasized distributive inequalities in maintaining that the fundamental cause of unjust inequalities was the caste hierarchy, not so much the economic structure of society. In emphasizing that social justice required the greater equalization of status, respect and power as much as material goods. Among other factors, differences in the significance accorded to social equality is likely to distinguish different philosophical and real-world theories of social justice. The elaboration of these claims requires further normative work, but the identification of theoretical gaps is a further important contribution of a study of political rhetoric as argument.
