WPS5109 Policy Research Working Paper 5109 Public Interest Litigation in India Overreaching or Underachieving? Varun Gauri The World Bank Development Research Group Human Development and Public Services Team November 2009 Policy Research Working Paper 5109 Abstract Public interest litigation has historically been an that complaints related to concerns having to do with innovative judicial procedure for enhancing the social separation of powers are better understood as criticisms of and economic rights of disadvantaged and marginalized the impact of judicial interventions on sector governance. groups in India. In recent years, however, a number of On the issue of inequality, the analysis finds that win criticisms of public interest litigation have emerged, rates for fundamental rights claims are significantly including concerns related to separation of powers, higher when the claimant is from an advantaged social judicial capacity, and inequality. These criticisms have group than when he or she is from a marginalized group, tended to abstraction, and the sheer number of cases which constitutes a social reversal, both from the original has complicated empirical assessments. This paper finds objective of public interest litigation and from the relative that public interest litigation cases constitute less than win rates in the 1980s. 1 percent of the overall case load. The paper argues This paper--a product of the Human Development and Public Services Team, Development Research Group--is part of a larger effort in the department to understand the relationship between the rule of law and development. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The author may be contacted at vgauri@ worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team Public Interest Litigation in India: Overreaching or Underachieving? Varun Gauri1 1 The author would like to thank Afroza Chowdhury, Kabir Duggal, Kaushik Krishnan, and Yamini Jaishankar for excellent research assistance. This paper has also benefited from very helpful comments from Vikram Raghavan, Arun Thiruvengadam, and the participants in a session of the Law and Social Sciences in South Asia conference in New Delhi. The views presented in this paper are those of the author alone and do not necessarily represent the views of the World Bank or its Executive Directors. The global reputation of Indian courts, and perhaps their national reputation as well, as judicial innovators and as defenders of the interests of the disadvantaged and downtrodden, rests largely on Public Interest Litigation (PIL), a new set of procedures for expanding access to justice that were developed some 30 years ago. Assessments of PIL in India range from the laudatory to the cynical, but recent scholarship has developed a widely held narrative that runs like this.2 PIL or "social action litigation," as some call it, originated in the late 1970s when the judiciary, aiming to recapture popular support after its complicity in Indira Gandhi's declaration of emergency rule, encouraged litigation concerning the interests of the poor and marginalized, and to do so loosened rules and traditions related to standing, case filing, the adversarial process, and judicial remedies. The Supreme Court issued a number of landmark social justice cases in the 1980s and early 1990s, including key rulings on the rights of prisoners, bonded laborers, pavement dwellers, and children. The frequency of PIL cases in the Supreme Court and the High Courts increased as claimants and their lawyers learned how to take advantage of the more liberal procedures associated with PIL. By the middle to late 1990s, the range of issues the courts were addressing had expanded to include complex environmental concerns, such as urban pollution and solid waste disposal, as well as explicitly political issues, such as official corruption and elections. At the same time, some claimants and their lawyers learned to "dress up" private disputes as PIL. Human rights activists began to grow disenchanted with courts' failure to enforce sweeping directives. Recently, many have questioned the appropriateness of judicial intervention in the legislative and executive spheres, as well as the constitutionality of the court's efforts to implement many of its expansive orders.3 Drawing on that narrative, concerns regarding the value and impact of PIL now take a number of forms. These will be described in more detail below, but in general terms, they can be categorized into two groups ­ questions related to the separation of powers, and a set of queries regarding judicial attitudes. The first group of concerns asks whether courts should be involved in environmental, social, and economic matters at all: Are not the legislative and executive branches better equipped to address these matters, and does not "judicial activism," precisely because the courts do not and cannot enforce many of their broad directives in these areas, erode the legitimacy of the courts? Are not PIL cases draining substantial resources from an already overburdened legal system in which ordinary civil cases can languish in courts for many years? Since many PIL cases 2 Elements of this broad narrative are supported in (Upendra Baxi, 1980);(R. Dhavan, Sudarshan, & Khurshid, 1985); (Sathe, 2002); (Cassels, 1989) (A. K. Thiruvengadam, 2009). 3 PIL has sparked concerns regarding judicial encroachment and the separation of powers that actually go back to the early days of the Indian Republic. Soon after Independence, the Congress government enacted the First Amendment, creating a schedule of laws deemed beyond judicial review, out of concern that the judiciary would find planned land reforms unconstitutional, a belief soon to be substantiated in a set of cases in which the Supreme Court did just that. This back and forth over land reform, and over the Ninth Schedule, continued for decades. Famously, the Court held that Parliament did not have the power to amend the Constitution if such amendments abridged the fundamental rights guaranteed in the Constitution (Golaknath v Punjab, 1967) and or altered the "basic structure" of the Constitution itself (Kesavananda Bharati v Kerala, 1973), to which the Indira Gandhi government responded by departing from the seniority tradition in judicial appointments and attempting to dissolve the Kesavananda majority. On "judicial sovereignty" in India, see (Mehta, 2007). 2 are patently frivolous and many others never enforced, is not PIL a device for the judiciary to expand its own powers and autonomy under the mantle of a popular social justice agenda? A separate set of questions involves the beneficiaries of PIL: Do PIL cases continue to benefit the poor and disadvantaged, or have not lifestyle issues and middle class concerns become predominant in PIL cases? Are not judges manifestly less disposed to the interests of the poor and marginalized than they were two decades ago, during the "heroic" years when PIL originated? These queries regarding PIL are fundamentally normative claims, and are based on principled understandings of the role of judges and courts in India's democracy. At the same time, the validity of some of them rests on facts, albeit complex ones. For instance, the challenge related to separation of powers raises questions about judicial capacity ­ critics charge that courts cannot monitor and supervise complex "polycentric disputes" (Fuller & Winston, 1978, p. 304), whereas others respond that they can, or at least as well as parliaments can (A. K. Thiruvengadam, 2009). The relative effectiveness of judicial supervision, if observed accurately and at scale, could help resolve this disagreement, at least for a subset of cases and in certain contexts. Similarly, whether or not PIL cases still address the concerns of the poor, and whether decisions are as supportive of their interests as in the past, are empirical questions. To date, the debate over PIL has largely been abstract (with some exceptions, to be described below). It has helped generate a set of normatively significant questions, but at this stage of the research cycle, empirical work may be more pressing. This paper contributes to that task by assessing PIL with empirical data. The next section of this paper analyzes the argument that PIL constitutes a case of judicial overreach. The contention that PIL weakens policy formulation and implementation in the legislative and executive branches is typically "dressed up" as a separation of powers concern, but a more apt framework involves an assessment of the impact of PIL on sectoral governance, which is fundamentally an empirical matter, not a doctrinal one. The following section describes the charge that PIL favors middle class interests rather than the concerns of the poor and marginalized. That section than presents estimates , based on original data taken from Supreme Court records an online legal database, to assess that claim. Public Interest Litigation: Is the Judiciary Overreaching? An old-fashioned view of legal rights holds that most social and economic matters do not involve genuine rights because they require positive actions, not merely restraint, and have no single, identifiable duty holder. Positive obligations, moreover, entail significant expenditures that are the purview of the other branches of government. Courts, therefore, should steer clear of the social, economic, and environmental concerns at the heart of PIL. More contemporary views (Holmes & Sunstein, 1999; Shue, 1996) hold that "for their fulfillment all rights require restraint, protection, and aid from the entity from 3 whom rights are claimed, and that a reasonably effective and well funded state is a sine qua non for all rights."(Gauri, 2004) Most of the criticisms of PIL in the Indian courts have not taken this somewhat old-fashioned form, perhaps because in a country where the scale of needs is so large it is hard to say that social and economic priorities are less commanding than civil and political ones. They have rather argued that the social and economic domain should be largely the prerogative of the other branches of government, which are better equipped to analyze, formulate, and implement complex policies, and that much of PIL is inappropriate judicial "activism" or "adventurism." For instance, in an assessment of the activities of the Supreme Court in the Delhi Vehicular Pollution and Municipal Solid Waste Management cases, Rajamani admonishes that "policy, environmental and social, must emerge from a socio political process and must be considered in a legitimate forum not a judicial one." (Rajamani, 2007) Citing cases in which courts formulated explicit guidelines, such as cases related to vehicular pollution, the management of the Central Bureau of Investigation, adoption by foreign nationals, custodial torture, and sexual harassment, Desai and Muralidhar note that "while in some cases, the Court has expressed its reluctance to step into the legislative field, in others it has laid down detailed guidelines and explicitly formulated policy" (Desai & Muralidhar, 2000). In their 2003 article, Rosencranz and Jackson welcome the environmental and health impact of the Supreme Court's 2001 decision requiring the Delhi government to convert its commercial vehicles to a fleet running on compressed natural gas (CNG), but then plead for leadership on the part of the regulatory and legislative authorities: "Some of the roadblocks to CNG implementation could have been avoided, or at least minimized, had the conversion been originally mandated through the normal legislative process." (Armin Rosencranz & Jackson, 2003, p. 21) Thiruvengadam documents a spate of similarly motivated criticism of PIL as an incursion into lawmaking from sitting and former judges on India's Supreme and High Courts, including comments from Justice Hidyatullah in 1984, Justice Srikrishna in 2005, and, perhaps most intemperately, Justice Kaju in 2008, who said PIL "has developed into an uncontrollable Frankenstein." (A. K. Thiruvengadam, 2009, p. 22) A motivation for some of this criticism is a suspicion that the courts have used their post-Emergency popularity, to which PIL has significantly contributed, to expand their own powers and shield themselves from scrutiny and accountability. To some, it appears as though the courts may be spending time on frivolous and ineffectual PIL cases at the expense of the real administration of justice, and choose to do so because PIL burnishes their popularity. Reported instances of frivolous PIL include prayers to rename India "Hindustan," rename the Arabian Sea "Sindhu Sagar," and replace the national anthem for one offered by the petitioner (and partly sung before the Chief Justice) (A. Thiruvengadam, 2007). At the same time, the systems of civil and criminal justice suffer enormous delays and arbitrary pre-trial detentions. These concerns are echoed widely enough that there is now visible a clear backlash against this perceived usurpation of powers by the courts, including a bill tabled in the Rajya Sabha in 1996 to regulate PIL, a 2007 statement by the Prime Minister 4 warning against judicial overreach (Shankar & Mehta, 2008), recent calls from the bench to set parameters for PIL (Times of India, December 12 2007), and efforts to establish the National Judicial Council, a body to investigate complaints against judges. Some of these complaints involve corruption: there have been allegations that some 20% of judges are corrupt (Rajeev Dhavan, 2002). Related complaints include the use of the law of privileges and contempt on the part of courts to shield themselves from criticism, resistance to efforts to require sitting judges to disclose their financial assets, and the uncomfortably close relationship between some members of the judiciary and the Bar. (U Baxi, 2006) Roy goes so far as to assert that judicial accountability is so low that "we live in a sort of judicial dictatorship." (Roy, 2007) A few comments about separation of powers are in order. First, policy formulation by the courts or its agents is, to some extent, inevitable. Judicial review of any sort requires ongoing commentary on laws and policies, including guidelines regarding their proper content. Because dispute resolution entails an elaboration and application of the normative structures of society as the necessary ground for the dispute resolver's decision, judges inevitably involve themselves in rule making, which is a form of lawmaking whether in common law or civil law jurisdictions. (Stone Sweet, 1999) Courts have not traditionally been significant actors in the area of social and economic policy; and resistance to public interest litigation and the court directives it prompts in these areas may stem more from the novelty of the phenomenon than from anything like a real "judicial dictatorship." Reluctance on the part of the Indian judiciary to be held accountable for performance and probity is certainly problematic ­ from the point of view of democratic theory it limits the power of the people to review public action. The expansion of judicial power in the area of social and economic concerns, on the other hand, catalyzes legislative and executive activity more often than it paralyzes it. That is because, as an empirical matter the world over, public interest litigation typically spurs judicial dialogue with the other branches: rarely do courts issue all or nothing demands, backed with common law contempt power or its civil law counterparts,4 in a way that requires the state to restructure its policy framework. "Courts' decisions do not so much stop or hijack the policy debate as inject the language of rights into it and add another forum for debate." (Brinks & Gauri, 2008, p. 304) As Fredman puts it, PIL allows the judicial forum to become, potentially, a space for democratic deliberation among equal citizens, rather than a place of interest group bargaining, which prevails in the legislature. (Fredman, 2008, p. 149) In addition, an important use of public interest litigation is to make public and scrutinize hidden or obfuscatory information, including cost of potential social programs, which the state and corporate entities on occasion have reasons to exaggerate or hide. In India, PIL during droughts in Rajasthan and Orissa in 2001 disclosed the extent of unreleased government grain stocks, and subsequent PIL disclosed that state governments could in fact afford to widen several statutory food and nutrition programs, including the midday meals scheme in schools, despite official protests to the contrary. In the Delhi vehicular pollution debate, the Delhi Health Minister claimed that air pollution did not 4 These would include the astreinte in France; the amparo in Mexico and Venezuela; the tutela in Colombia; and the mandado de segurança in Brazil. 5 increase the risks of heart or lung disease, the Delhi government said that the timely installation of CNG stations would be impossible, the Ministry of Petroleum and Natural Gas argued that CNG bus conversion would not be sustainable in the long run, producers of commercial vehicles stated that the conversion to CNG was not economically cost- effective, and other argued that CNG is explosive. The court, largely by empowering certain technical committees, played a significant role in helping to ascertain accurate information on these issues. It was, moreover, not an instance of judicial fiat but rather a judicial-executive branch collaboration: "Government experts essentially became advisors to the Court as it drove policy implementation forward." (Bell, 2004, p. 35) The argument that PIL constitutes judicial overreach, resulting in poor or inefficient decision making, is not really a separation of powers claim. The balance of power among government organs, as Madison conceived it, was not primarily about a strict separation of powers but "the partial interpenetration of relatively autonomous and balanced powers." (O'Donnell, 2003) In other words, the separation of powers was not conceived as a design for the promotion of efficient decision making by preventing undue encroachment from one institution upon the prerogatives of another, but rather a check on the ability of any group or faction to dominate government from its enclave in a specific organization. The doctrine of separation of powers seeks to accomplish this precisely by opening certain governmental tasks to competing competences and concurrent powers of review. Despite occasionally hyperbolic claims on the part of critics, Indian judges and their professional social classes are not using the courts as a staging ground to threaten the Indian state. There have been specific rulings, such as Kesavananda Bharati or Advocates-on-Record, or the ruling on the Jharkhand legislative procedures, in which courts assumed powers not delineated in the Constitution. Even in those cases, it arguable that in so doing the Court restored a constitutional balance because the executive and legislature had themselves been engaging in extra-constitutional activities. These criticisms regarding separation of powers are better cast as concerns related to the impact of judicial intervention on sectoral governance. Does judicial involvement through PIL improve state performance in a given sector? Is forest policy, for example, more equitable, efficient, and effective as a result of court involvement? That is an empirical question, but most treatments of the issue do not take the empirical challenge seriously. (A. Rosencranz & Lélé, 2008) believe that the Supreme Court's intervention following the TN Godavaraman vs Union of India case (1996) "hurts the process of governance," but adduce little evidence about the capacity and authority of central and state executive agencies prior to and after the court's assumption of powers. Writing in 2003 on the Delhi vehicular pollution case, Rosencranz and Jackson speculated that strengthening the pollution control boards (PCBs), rather than Supreme Court action, "would seem to provide the most effective long-term solutions [to air pollution in India]" and worried that "the Court's action seems likely to impede capacity building in the pollution control agencies, and thereby to compromise the development of sustained environmental management in India." (Armin Rosencranz & Jackson, 2003, p. 21, 23) This is a fundamentally an empirical claim, and one can examine whether PCBs are weaker now than they were before the Court got involved in the Delhi pollution case. A cursory review suggests that it is not obvious that they are weaker ­ the budget of the 6 Central Pollution Control Board has nearly tripled since the year of the Court's order in 2002, and a number of efforts are underway to strengthen them and fill staffing vacancies in central and state PCBs.5 Another problem with criticisms like these is that they compare an ideal or hypothetical legislative intervention to a real judicial one when it is often the real-world failings of the other branches that prompted litigation in the first place. Thiruvengadam describes the deliberative failings of India's Parliament, noting that of the total 36 Bills passed in 2008, "16 were passed in less than 20 minutes, most without any debate whatsoever." (A. K. Thiruvengadam, 2009, p. 32) Why do analysts tend to describe issues of sectoral governance with the language of the separation of powers? The motivation stems in part from a belief, sometimes inarticulate, that governance should look similar the world over. In this case, courts, in order to be courts properly understood, must limit their tasks to interpreting laws, rather than writing or enforcing them. But it is a mistake to speak of "courts" as such. The task of judicial institutions depends on the way they interact with the other institutions of their society. It is less useful to assess judicial activity against a preconceived institutional design than to evaluate, using "normative benchmarks," the (positive or negative) contribution of courts to the key tasks of governance in any specific sector. (Trebilcock & Daniels, 2008) In the same way that careful studies of the institutional foundations of economic growth in East Asia have challenged the rule of law orthodoxy, showing that successful market-sustaining institutions need not take the specific form that courts, corporate boards, and bureaucratic agencies have taken in, say, the Untied States or the United Kingdom (Ginsburg, 2003), studies of public interest litigation should recognize that courts may play a variety of roles in different settings. There is less institutional convergence in the world than believed, and it is important "not to confuse institutional function and institutional form" (emphasis in original). (Rodrik, 2003) What, then, are the normative benchmarks that should be used to assess the contribution, or lack thereof, courts toward sectoral governance? Those depend on the sector, of course ­ they would look different in health than in forestry. But, generalizing, one can identify three key elements of governance for the broad category of tasks in government service delivery: the capacity and authority of the organizations charged with delivery or oversight, the availability of information and transparency regarding service delivery, and state accountability for performance. An empirically minded assessment of PIL in India, then, would take the form of a series of case studies based on those normative benchmarks. The case studies would focus on these questions: 1. Did the capacity and authority of institutions tasked with addressing the social problems increase or decline as a result of PIL? 5 The sanctioned budget was Rs 1592.58 lakhs 2001-02, and 4500.00 lakhs for 2007-08. These budgets are available at http://www.cpcb.nic.in/upload/AnnualReports/AnnualReport_6_annualreport2001-02.pdf and http://www.cpcb.nic.in/upload/AnnualReports/AnnualReport_34_final-report-06-07-A.pdf 7 2. Was accurate information on sectoral concerns more widely available before or after judicial intervention? 3. Were mechanisms of accountability, including legal and hierarchical oversight, markets and the power of actors to pursue their own interests, and social assessments of the motives of public officials, functioning more effectively before or after judicial intervention? Does Public Interest Litigation Benefit Marginalized Individuals and Groups? The purpose of Indian public interest litigation, like most social causes, was likely over-determined; but most historians agree that it was partly an effort on the part of the courts to speak to the poverty, social exclusion, and powerlessness that the majority of citizens in India continue to suffer. Several commentators suspect, however, that this objective for PIL has not been realized, or indeed has been lost. These criticisms take two forms. The first focuses on beneficiary inequality: the concern that the middle classes have more organizational and financial resources than the poor, which facilitate easier access to courts, and results in more benefits from PIL for them than the poor. In some countries, the Bar develops a pro bono practice that, to some extent, mitigates this inequality of access, but this has not emerged in a significant way in India. Compounding this problem, NGOs and social movements in India are wary of courts, and largely do not utilize litigation strategies to achieve their objectives (Krishnan, 2003; Shankar & Mehta, 2008). The second is policy area inequality: this is the concern that judges, because of their social class and ideological dispositions, are more alert to the concerns of the middle classes and the wealthy. This is evident, critics argue, in the demonstrable urban and middle-class bias of their rulings. Baxi believes that the interests of global economic elites now color the thinking of the Indian judiciary, resulting in rulings that benefit those global elites, such as decisions in the cases involving WTO accession, Union Carbide's liability for toxic emissions in Bhopal, and the construction of the Narmada dam. (U Baxi, 2006) Similarly, Rajagopal believes that the Indian courts have adopted the statist and development mindsets associated with the Indian state itself, and that the universal privileging of civil and political rights over socioeconomic rights has affected the thinking of the Indian judiciary (Rajagopal, 2007). In a related vein, Ramanathan writes that "increasingly, the constituency on whose behalf the enhancement of judicial power has been strengthened began to emerge as the casualty of the exercise of that power." (Ramanathan, 2002) These claims tend to be based on a set of exemplary cases, rather than a systematic review of PIL case law and implementation. It is also important to realize that judges, and the courts that they constitute, are creatures of their environment, and that patterns of judicial recruitment and appointment will almost always ensure that courts reflect the dominant political trends of their countries (Dahl, 1957). Some argue that the Indian courts are an exception to this rule, having obtained as much, if not more, autonomy regarding judicial appointments than any other courts in the world. But that 8 constitutional autonomy is constrained by the Indian civil service traditions under which courts operate, as well as the social dependence of judges on the Bar and on political and economic elites. "Leaders of the Indian Bar," writes Baxi, "whether elected or otherwise, have always wielded historically disproportionate influence over the Justices." (U Baxi, 2006) In addition to the embedding of judges in their social and political milieu, strategic factors also limit judicial autonomy: courts in every country depend on the other branches of government for their political survival, so they will, by and large, take care that their decisions garner sufficient support from key political actors (Epstein & Knight, 1998; Gauri & Brinks, 2008b). In short, because ideological orientations in the larger Indian political economy have drifted rightward over the past two decades or so, it is to be expected that the courts have followed. Assessments of the extent to which PIL supports the poor, then, need to develop an implicit benchmark of what can be expected of courts in the specific contexts in which they operate.6 There are a set of empirical questions on PIL answers to which might guide these assessments: 4. What share of PIL cases have claimants who are not poor, and has this share changed over time? 5. How many PIL cases have NGOs or CSOs as counsel or supporters, and has this changed over time? This is important because NGO- or CSO-led cases typically involve collective claims that have a greater likelihood of generalizing benefits to an entire class of similarly situated people, such as the poor or scheduled castes, scheduled tribes, and other backward classes (SC/ST/OBC). 6. What is the win-loss rate for PILs, and for PILs of different types? Do these win- loss rates vary with the social class of the claimants? To fill in some of the information needed for a more complete assessment of PIL in India, four different samples of PIL and Fundamental Rights cases at the Supreme Court level were examined: (i) cases that, according to the Supreme Court registrar's office, the Court has itself classified as PIL from 1988-2007 (some 2800 "cases" overall); (ii) all Supreme Court cases in the Manupatra database that involved Fundamental Rights and that addressed concerns regarding women and children rights, whether or not explicitly admitted as PILs (86 cases); (iii) all Supreme Court cases in the Manupatra database that involved Fundamental Rights and were related to issues regarding SC/ST/OBCs, whether or not explicitly admitted as PILs (180 cases); (iv) all Supreme Court cases in the Manupatra database that the Supreme Court explicitly called a PIL (44 cases). 6 For an extended discussion of inequality in social and economic rights litigation, see (Brinks & Gauri, 2008) 9 For the set of cases involving women and children's rights, these search terms, among others, were used: "women" and "children" and "abuse"; "violence" and "women" and "neglected children"; "juvenile justice" and "Article 21"; "trafficking" and "women" and "Article 21"; "labour" "children" and "Article 14" ; "custodial violence" and "children"; "education" "fundamental right" and "Article 14"; "gender and fundamental right"; "prison rights and children"; "rape" and "Article 21"; "emancipation" and "women and children." The SC/ST/OBC search used the terms "scheduled caste" or "scheduled tribe" or "other backward classes" in conjunction with cases that addressed Fundamental Rights and constitutional concerns. Note that the data from the Supreme Court were a listing of orders related to PIL cases, and not necessarily a list of cases per se. But an examination of the names of the cases revealed that of the 2,800 separate orders, instances of duplicate or once repeating nearly identical plaintiff and defendant names occurred fourteen times, triplicates once, and quadruplicates once. Still, because of uncertainty regarding the basis for the listings, this paper refers to this list as constituting "cases" rather then cases per se. The Manupatra database captures about 80-85% of Supreme Court cases.7 While this results in something less than complete compendium of Court decisions and orders in the selected topic areas, there is no evidence that the database censors certain categories of cases in a systematic fashion. There is no evidence, in other words, of significant sample bias. Figure 1 shows the number of PIL "cases" instituted per year, whether brought to the court for admission ("admission matters" in the Court's terms) or argued on the merits ("regular matters"), from 1997-2007, based on data from the Supreme Court itself. The figure shows that there has been a slight upward trend in PIL matters over the last ten years. It also shows that there are some 260 "cases" instituted per year, on average. This compares to about 60,000 "cases" per year overall, based on data publicly available in the Supreme Court's "Court News" publication. So, on average, some 0.4% of "cases" before the Court involve PILs. This suggests that PIL does not drain significant Court resources from the administration of day to day justice, contrary to the claims of some critics; but it also suggests that the outsize reputation of the Supreme Court's PIL work belies its modest scale. And, contrary to popular conceptions, the large majority of the 260 or so PIL "cases" instituted in the Supreme Court each year are brought through formal channels; of the tens of thousands of letters and handwritten petitions that the Supreme Court receives from ordinary people each year, only a handful are converted into cases.8 The Court instituted a classification system for PILs in 1988. This might be used to determine whether the concerns cited in PILs have changed over the years, and if the "cases" that appear before the Court have shifted to the concerns of the middle classes. Table 1 shows the breakdown by "case" type and year. Unfortunately, the large majority of "cases" fall in the category of "other," so the classification scheme is not useful for tracking changes in PIL composition over time. Noteworthy, however, is the increase in "cases" related to election commissions, and the relatively low share of "cases" involving 7 Reported by Nick Robinson, based on personal communication with Manupatra staff, January 14, 2009. Also see the similar figure in (Shankar & Mehta, 2008) 8 Interviews with officers of the PIL section, Supreme Court, New Delhi, January 13, 2009. 10 bonded labor and criminal justice, two of the concerns crucial in the early justifications for PIL. Have increasing numbers of cases involved middle class concerns? It is difficult to say for PILs more generally because an identifiable record of PIL cases is not available. To address this question, the paper used the set of Fundamental Rights cases described above. Fundamental Rights cases, at least those related to women and children's rights and SC/ST/OBC concerns, may in fact underestimate the orientation of PILs to the middle classes because they exclude issues like urban quality of life that have been the concern of recent prominent litigation. Figure 2 shows the number of cases in categories related women and children's rights or to SC/ST/OBC issues, or which the SC explicitly called a PIL, as reported in Manupatra, in which the claimant was likely to have been, on the basis of a review of the written opinion, a member of the "advantaged classes." For purposes of the coding, a member of the advantaged classes included a professional (doctors, teachers, members of the armed services, etc), a landowner or business person, or someone otherwise in the global middle class (including formal sector workers and civil service employees not designated as workers or laborers). Those not in the advantaged classes included those belonging to a scheduled caste or scheduled tribe or a member of the other backward classes (unless otherwise designated a member of an advantaged class), peasants, laborers, and those detained in the criminal justice system. Where groups or publicly minded individuals made claims in courts, and where there was no obvious plaintiff, the coders attempted to identify the general social class of the individuals whose interests were being advocated or defended. In 48 of the 310 cases in the sample (some 23%), the social class of the claimant(s) could not be discerned from the written opinion; or, and this was the more frequent problem, the social class was diffuse. Examples of this included cases concerning environmental protection of rivers and forest, challenges to the constitutionality of judicial action, and challenges to the statutory definition of rape. For these cases, the calculations and figures below considered the observation to have a missing value for social class. For every case, an effort was made to identify the individual or class of individuals on whose behalf the case was instituted ­ in some instances the lawyer or some other publicly minded individual is listed as the plaintiff/complainant in official records, but the claimant was coded was the person or class of persons whose interests the case was seeking to advance (e.g., child bonded laborers). Figure 2 plots the number of cases involving Fundamental Rights in the categories described above, and the number of cases in which the claimant was a member of the advantaged classes. Simple OLS estimates are used to identify trend lines, which are inserted in the figure. The trend lines show that the number of claimants from advantaged classes has increased at almost the same rate as the overall number of Fundamental Rights cases in these categories. This suggests that class bias concerns that stem from the organizational advantages of the advantaged class are not as pronounced as some have feared. 11 Another way to examine this problem of inequality in organizational resources is to look the share of Fundamental Rights cases in which the rights violation originated in the BIMARU states. Figure 3 plots the share of cases originating in the BIMARU states. (In 14% of the cases, the location of the rights violation was national or otherwise impossible to pinpoint.) Figure 3 shows that although that number of cases originating in rights violations in the BIMARU states has been increasing, it remains below the share of the national population residing in those states (some 40%). This suggests that inequalities related to middle class legal mobilization, though real according to this measure, are declining. Finally, the data allow a third way to examine the issue of middle class legal mobilization advantage ­ one can identify for each year the share of cases initiated by or involving an NGO or cooperative group. Cases that involve an NGO or cooperative are more likely to involve the concerns of the poor than the middle class, and are also more likely to result in generalized benefit than cases brought on behalf of individual private claimants by private lawyers (Gauri & Brinks, 2008a). Overall, 15% of cases were filed by cooperatives, groups, or NGOs, while 80% were filed by individual litigants (4% of cases were unclear). Figure 4 shows that the annual number of cases filed by a cooperative/group and reported in Manupatra has been more or less flat since the 1970s, with a median number of about one per year, and has not grown as fast as cases involving Fundamental Rights filed by individuals. This suggests that although the absolute number of Fundamental Rights claims brought on behalf of disadvantaged classes has been increasing, as shown in Figure 2, this increase is not due to greater mobilization by NGOs/CSOs; rather, it appears to have been the result of increased mobilization by private individuals. To examine policy area inequality, which is related to the content of judicial rulings on the cases that reach them, we examined win-loss rates, both in general and for subsets of claims from advantaged classes, disadvantaged classes, members of SC/ST/OBC, and the middle and upper castes. Figures 5 and 6 show that claimants in cases involving women's and children's rights were more likely to win than claimants in cases involving SC/ST/OBC matters. Overall, the win rate for claimants in Fundamental Rights cases involving women and children's rights was 84%, compared to 51% for cases involving SC/ST/OBC, and 72% for the explicit PIL cases. In addition, the trend line for the win rate of claimants in SC/ST/OBC cases was sloped downward. That suggests that judges may now be less favorably disposed to SC/ST/OBC claims than they were in the past. But it is hard to tell whether this indicative trend is a function of the change in the composition of the claimants or a change in judicial attitudes; in other words, it is possible that judges are as favorably disposed to SC/ST/OBC concerns as they were in the past, and that win rates in SC/ST/OBC cases have declined only because most claimants in these cases are now members of the advantaged classes. So the cases were further disaggregated on the basis of the social class of the claimant. Overall, 67% of claimants in SC/ST/OBC cases were themselves members of SC/ST/OBC; but this falls to 60% for cases admitted after 2000, indicating a recent increase in claims related to 12 SC/ST/OBC by members of the middle and upper castes. For women's and children's rights cases, overall 72% of claimants were members of the advantaged classes. Figure 7 graphs the win rates in all Fundamental Rights cases for claimants who were and were not members of advantaged classes. It shows that the average annual win- rate for claimants from advantaged classes was below the win rate of claimants who were not from advantaged classes until the late 1980s. Now claimants from advantaged classes have higher win rates than claimants not from advantaged classes. For example, advantaged class claimants had a 73% probability of winning a Fundamental Rights claim for cases in which an order or decision was rendered from years 2000-2008, whereas the win rate for claimants not from advantaged classes for the same years was 47%. For the 1990s, rates were 68% and 47%, respectively. But in the years prior to 1990, claimants not from advantaged classes enjoyed higher success rates than those from advantaged classes. The differences for the 1990s and 2000s are significantly different from each other, based on a simple chi-square test and a simple probit estimation (see Tables 2a and 2b). Similarly, when one divides the claimant into those that are identified as members of SC/ST/OBC and those that are not, the same pattern emerges: Figure 8 shows that claimants who are not SC/ST/OBC now have a higher win rate than those who are. Even in the subset of cases involving SC/ST/OBC concerns, claimants who were not from SC/ST/OBC began to have higher average annual win rates than those who were starting around 1990 (see Figure 9).9 These findings are consistent with the claim that judicial receptivity in the Supreme Court to Fundamental Rights claims made on behalf of poor and excluded individuals has declined in recent years. There are other explanations, however. The decline in the win rates for marginalized individuals could be attributed to the fact that cases brought on their behalf are weaker, on the merits, than they used to be, perhaps as a result of changes in statutes, decisions to litigate more challenging cases, or weaker legal representation. Each of these possibilities warrants careful scrutiny. Still, the data demonstrate not only a decline in the win rate for marginalized individuals but a simultaneous increase in the win rate for advantaged individuals. Though not impossible, it is unlikely that that the quality of legal representation has simultaneously increased for the advantaged and decreased for the marginalized, and sufficiently to explain the significant reversal in win rates. Similarly, it is conceivable but not likely that advantaged clients started to select a less challenging set of cases at the same time that marginalized claimants did the reverse. The data here constitute a prima facie validation of the concern that judicial attitudes are less favorably inclined to the claims of the poor than they used to be, either as the exclusive result of new judicial interpretations or, more likely, in conjunction with changes in the political and legislative climate. Conclusions A number of criticisms of PIL have been voiced in recent years, including concerns related to separation of powers, judicial capacity, and inequality. While critics 9 The social class of the claimant was not discernible in 1% of the cases related to SC/ST/OBC, 34% of cases related to women's and children's rights, and 45% of explicit PIL cases. 13 have been persuasive when pointing to particular cases, the sheer number of cases, as well as the variation in tendencies over time and among court benches, have made reaching a general conclusion difficult. This paper has argued that complaints related to separation of powers concerns are better understood as criticisms of the impact of judicial interventions on sectoral governance, and that structured case studies of sectoral governance are necessary to assess those criticisms. On the issue of inequality, this paper contributes to an overall assessment by systematically examining the relative magnitude, case composition, and geographical origins of, as well as legal representation and the claimant's social class in, PIL and Fundamental Rights cases that reached the Indian Supreme Court. The analysis of PIL "cases" shows that they do not appear to consume a significant share of the resources of the Supreme Court; they constitute less than 1% of the overall case load. The subject matter of PIL cases and orders remains difficult to discern because over 70% of them are classified as "other," which is problematic from the point of view of judicial transparency. Concerns regarding inequality appear to be validated by some of the quantitative data on Fundamental Rights cases. On the one hand, although the number of Fundamental Rights cases related to women and children's rights and to the concerns of scheduled caste/scheduled tribe/other backward classes, as well as cases explicitly called a "PIL" in the Court's written opinions, appear to have increased; and the rate of increase has been similar for claimants belong to both marginalized and advantaged population segments. The share of cases from BIMARU states also appears to be climbing. On the other hand, a very low share of the cases are brought by cooperative entities, such as NGOs, which is a useful predictor of the likelihood that benefits will generalize to the larger population. Most striking, win rates for Fundamental Rights claims are now significantly lower when the claimant is from an advantaged social group than when he or she is from a marginalized group. That constitutes a social reversal both from the original objective of public interest litigation and from the relative win rates in the 1980s. 14 Table 1 Fundamental Rights cases in the Indian Supreme Court, % of Cases Submitted by Subject Matter Subject 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Child Labour Matters 0.4 0.49 0.73 Including Neglected Children Air Pollution Matters, i.e. 2.58 2.8 2.58 2.34 5.32 8.3 5.78 2.44 2.89 2.19 2.88 Industrial, vehicular, Power Stations etc. Water Pollution: Industrial, 1.94 6.4 2.58 4.3 2.84 0.72 2.22 1.46 18.18 2.19 1.44 Domestic, Sewage, Rivers and Sea Noise Pollution: Industry & 0.65 1.2 0.43 0.35 1.81 1.33 0.83 1.46 Vehicular Ecological Imbalance: 8.39 3.2 3.43 2.73 3.19 4.44 5.85 7.85 16.06 12.98 Protection and Conservation of Forests throughout the Country, Protection of wild life, Ban on Felling of Trees and Falling of Under Ground Water Level Bonded Labour Matters 2.58 0.4 1.29 4.33 0.48 Matters relating to Custody 3.87 2 1.72 1.56 0.35 0.44 0.36 0.96 Harassment, Jails, Complaint of Harassment, Custodial Death, Speedy Trial, Premature Release, Inaction by Police etc. Matters relating to Harassment 0.65 0.4 1.42 0.36 0.89 0.49 0.36 0.48 of SC/ST/OBC and Women Matters relating to 3.23 3.6 1.72 5.47 2.84 1.33 0.98 2.07 2.92 1.44 Unauthorized Constructions including Encroachments, Sealing, Demolitions, Urban Planning Matters relating to Election 1.29 1.6 1.72 0.39 1.42 3.25 0.44 7.8 3.72 3.28 5.77 Commissions Scam Matters 1.94 0.43 0.39 0.36 0.98 0.83 0.36 0.48 Others 72.9 78 84.12 82.81 82.27 80.87 83.11 79.51 63.64 70.07 72.6 Natural and Man Made 0.48 Disasters Including Riots Numbers are in Percentages 15 Table 2a: Win rates in selected Fundamental Rights before the Indian Supreme Court, Claimants from Advantaged and Disadvantaged social classes Social class 1961-1989 1990-1999* 2000-2008* Advantaged 57.9% 68.1% 73.3% Disadvantaged 71.4% 47.1% 47.2% * probability of 2 < 0.05 Table 2b: Probit estimation of win rates for advantaged classes in Fundamental Rights cases before the Indian Supreme court, by decade 1961-1989 1990-1999 2000-2008 Advantaged class -0.37 0.54* 0.69* (0.29) (0.25) (0.32) Constant 0.57* -0.07 -0.07 (0.21) (0.15) (0.21) Observations 80 114 66 Pseudo R2 0.02 0.03 0.05 Standard errors in parentheses; * significant at p < 0.05 16 Figure 1 PIL Cases in the Indian Supreme Court Number of Cases Instituted per Year 350 300 250 No. of Cases 200 150 100 50 0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Year Figure 2 Fundamental Rights cases Number of Cases in Which Complainants Belong to Advantaged Classes 25 20 Number of Cases 15 10 5 0 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 Year Advantaged Classes All 17 Figure 3 Share of Fundamental Rights cases from BIMARU states 100 80 Percentage 60 40 20 0 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 Year Figure 4 Fundamental Rights Cases Filed by Cooperative/Group 25 20 No. of Cases 15 10 5 0 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 Year Filed by Cooperative/Group Total 18 Figure 5 Cases Involving Women and Children's Rights Percentage of Cases in Favor of Complainants 100 80 Percentage 60 40 20 0 1960 1970 1980 1990 2000 2010 Year Figure 6 Cases Involving SC/ST/OBC Percentage of Cases in Favor of Complainants 100 80 Percentage 60 40 20 0 1950 1960 1970 1980 1990 2000 2010 Year 19 Figure 7 Fundamental Rights Cases Percentage of Cases in Favour of Advantaged Classes vs. Others 100 80 Percentage 60 40 20 0 1950 1960 1970 1980 1990 2000 2010 Year Advantaged Classes Not Advantaged Classes Figure 8 Fundamental Rights Cases Percentage of Cases in Favor of ST/SC/OBC Complainants vs. Others 100 80 Percentage 60 40 20 0 1950 1960 1970 1980 1990 2000 2010 Year Not SC/ST/OBC SC/ST/OBC 20 Figure 9 Cases Involving SC/ST/OBC Percentage of Cases in Favor of SC/ST/OBC Complainants vs Others 100 80 Percentage 60 40 20 0 1950 1960 1970 1980 1990 2000 2010 Year Not SC/ST/OBC SC/ST/OBC 21 References cited Baxi, U. (1980) The Indian Supreme Court and Politics. Lucknow: Eastern Book Company. Baxi, U. (2006) "Structural Admustment of Judicial Activism, Inaugural Lecture," West Bengal National Academy of Juridical Sciences. Bell, R. G., Kuldeep Mathur, Urvashi Narain, and David Simpson. (2004) "Clearing the Air: How Delhi Broke the Logjam on Air Quality Reforms," Environment, 46: 22-39. Brinks, D. M., & Gauri, V. (2008) "A New Policy Landscape: Legalizing Social and Economic Rights in the Developing World," In Courting Social Justice, edited by V. Gauri & D. M. Brinks. New York: Cambridge University Press. Cassels, J. (1989) "Judicial Activism and Public Interest Litigation in India: Attempting the Impossible?," The American Journal of Comparative Law: 495-519. Dahl, R. (1957) "Decision-Making in a Democracy: The Supreme Court as a National Policy-Maker," Journal of Public Law, VI: 279-295. Desai, A. H., & Muralidhar, S. (2000) "Public Interest Litigation: Potential and Problems," In Supreme but not Infallible: Essays in Honour of the Supreme Court of India, edited by B. N. Kirpal, A. H. Desai, G. Subramanium, R. Dhavan & R. Ramachandran. New Delhi: Oxford University Press. Dhavan, R. (2002) "Judicial Corruption," In The Hindu (online). Dhavan, R., Sudarshan, R., & Khurshid, S. (1985) "Judges and the judicial power: essays in honour of Justice VR Krishna Iyer," In: Sweet & Maxwell Bombay: NM Tripathi, London. Epstein, L., & Knight, J. (1998) The Choices Justices Make. Washington, D.C.: Congressional Quarterly Press. Fredman, S. (2008) Human rights transformed: positive rights and positive duties: Oxford University Press. Fuller, L. L., & Winston, K. I. (1978) "The forms and limits of adjudication," Harvard Law Review: 353-409. Gauri, V. (2004) "Social Rights and Economics: Claims to Health Care and Education in Developing Countries," World Development, 32: 465-477. Gauri, V., & Brinks, D. M. (2008a) "The Elements of Legalization, and the Social and Economic Rights Triangle," In Courting Social Justice, edited by D. B. Varun Gauri. New York: Cambridge University Press. Gauri, V., & Brinks, D. M. (2008b) "Introduction: The Elements of Legalization and the Triangular Shape of Social and Economic Rights," In Courting Social Justice: Judicial Enforcement of Social and Economic Rights in the Developing World, edited by V. Gauri & D. M. Brinks. New York: Cambridge University Press. Ginsburg, T. (2003) Judicial Review in New Democracies: Constitutional Courts in Asian Cases. Cambridge: Cambridge U. Press. Holmes, S., & Sunstein, C. R. (1999) The cost of rights : why liberty depends on taxes 1st ed. New York: W.W. Norton. Krishnan, J. K. (2003) "Social Policy Advocacy and the Role of the Courts in India," American Asian Review, 21: 91-124. Mehta, P. B. (2007) "The Rise of Judicial Sovereignty," Journal of Democracy, 18: 70. 22 O'Donnell, G. (2003) "Horizontal Accountability: The Legal Institutionalization of Mistrust," In Democratic Accountability in Latin America, edited by S. M. a. C. Welna. Oxford: Oxford University Press. Rajagopal, B. (2007) "Pro-Human Rights but Anti-Poor?: A Critical Evaluation of the Indian Supremem Court from a Social Movement Perspective," Human Rights Review 18: 157-187. Rajamani, L. (2007) "Public Interest Environmental Litigation in India: Exploring Issues of Access, Participation, Equity, Effectiveness and Sustainability," Journal of Environmental Law, 19: 293-321. Ramanathan, U. (2002) "Of Judicial Power," Frontline, 19. Rodrik, D. (2003) "Growth strategies," In NBER Working Paper Series. Cambridge: National Bureau of Economic Research. Rosencranz, A., & Jackson, M. (2003) "The Delhi Pollution Case: The Supreme Court of India and the Limits of Judicial Power," Columbia Journal of Environmental Law, 28: 1- 23. Rosencranz, A., & Lélé, S. (2008) "Supreme Court and India's Forests," ECONOMIC AND POLITICAL WEEKLY, 43: 11. Roy, A. (2007) "Scandal In the Palace," In Z-space: The Spirit of Resistance Lives. Sathe, S. P. (2002) Judicial Activism in India. Delhi: Oxford University Press. Shankar, S., & Mehta, P. B. (2008) "Courts and Socioeconomic Rights in India," In Courting Social Justice: Judicial Enforcement of Social and Economic Rights in the Developing World, edited by V. Gauri & D. M. Brinks. New York: Cambridge University Press. Shue, H. (1996) Basic rights : subsistence, affluence, and U.S. foreign policy 2nd ed. Princeton, N.J.: Princeton University Press. Stone Sweet, A. (1999) "Judicialization and the Construction of Governance," Comparative Political Studies, 32: 147-184. Thiruvengadam, A. (2007) "Recent PIL cases decided by the Supreme Court," In Law and Other Things. Thiruvengadam, A. K. (2009) "Evaluating contemporary criticisms of 'Public Interest Litigation': A progressive conception of the role of a Judge," In Paper presented at the 2009 inaugural LASSNET Conference. New Delhi. Trebilcock, M. J., & Daniels, R. J. (2008) Rule of Law Reform and Development: Charting the Fragile Path of Progress. Northampton, MA: Edward Elgar. 23