The Strange Demise of Normativity

It must be said that for the most part, normative theories of a proper judicial role have not fared well in the recent past. In particular in India, judges give scant evidence of even being conscious of these theories, and they are not seen as binding in any sense. There is certainly a lively debate in India about judicial overreach.Footnote ¹ Such a debate presupposes exactly the type of normative debate to which this discussion alludes. After all, on what grounds is a judicial intervention characterized as overreach? The challenge, however, is that there is significant skepticism about the formal distinctions as well as the principles that usually were deployed in such theories.

One way to think about proper judicial behavior is to invoke a set of formal considerations. On this view, there are certain formal restraints on judicial behavior. Judges are bound by the text of the constitution. In traditional legal conceptions, courts are disciplined by a set of formal techniques, which also form the basis of authority for their decisions. Therefore, courts routinely justify their rulings on the basis of a range of techniques: appeal to the text of a constitution, precedent, and so forth. Many judgments still have this formal structure and they seem to buttress the courts’ authority by reference to these techniques. They will cite constitutional texts, cite precedent, make reference to formal distinctions between law and policy, allude to separation of powers, and point to different sources of law. However, these are not so much normative constraints on judicial behavior as they are forms in which judges express their opinions.

A judge, for instance, may appeal to a constitutional text, but the truth is that constitutionalism is not so much an appeal to a higher law that binds. Rather, it is a practice that is constantly being created and re-created through the actions of concrete agents, including judges. A constitution is not a text that binds with a transparent meaning. Instead, it poses a challenge about how we handle slippages in systems of meaning, resolve ambiguities, and overcome silences – all through acts of choice. Constitutional texts are indeterminate all the way down and, in any case, judges are the creators and arbiters of meaning; they are not bound by meanings given independently of their interpretation. In the case of Indian courts, this was already evident in the late 1960s. It is best exemplified by the legacy of Justice Gajendragadkar, one of India's most respected judges, who handed down major social-reform judgments. The late P. K. Tripathi once commented that:

Gajendragadkar's decimation of the constitutional text, Tripathi observed, accompanied major substantive innovations, and it was these innovations that assumed greater credibility in the long term. Such examples are pervasive. For instance, the Indian Supreme Court has read all types of interpretations into the “right to life,” in which the meaning of the terms right and life are now entirely a function of judicial interpretation. Judges are not bound by semantics; they exercise sovereignty over meaning.

These instances can be multiplied: the distinction between law and policy and between making and interpreting law and treating precedents as a binding feature of the rule of law seem, for the most part to be dead. It is difficult for judges and lawyers to openly admit this for two reasons. The first, of course, is perhaps sociological: the very identity of legal reasoning, the training in law schools, and the liturgy of argumentation in court rest on alluding to these formal techniques. The more substantive second reason, however, is that many of these formal techniques were thought to be essential to court judgments that have rule-of-law characteristics. For these judgments to be “law,” there must be a formal way to express fidelity to the idea of law, to the properties that bind judges. These properties – for example, citation of texts and wrestling with precedents – must be instantiated in forms of legal justification. However, there seems to an emerging consensus that it is difficult to make sense of judicial behavior in these formal terms. Indian judges are more openly willing to acknowledge – perhaps more than scholars – that underneath the “panoramic compass for rationalization,” there is a straightforward pragmatism about the law.Footnote ³ Legal reasoning, quite simply, is more openly instrumental. It exists to serve certain broad needs and expectations of a society, and no set of formal restraints can deter how judges serve those needs and expectations. The promiscuity one perceives is a court attempting to serve society; the promiscuity being warranted by the needs of that society. The test of the legitimacy of what the court does is not its fidelity to law; rather, it is its social acceptance. What does this mean, and how would judges take it into account? The discussion returns to these questions later in this chapter.

This skepticism still may leave some residual disquiet. In American jurisprudence particularly, there is a vast body of work on the topic of “by what judges should be bound?” One reason that we have so much investment in finding meta-theories of constitutional interpretation is precisely because we want to hold onto the supposition that the authority of the law and constitution is grounded in something other than the choices exercised by judges. Should the authority of judges rest on being conduits for the congealed intentions of the founders? Should it, as Dworkin suggested, rest on their ability to act as interpreters of an indubitable moral principle that lies behind the constitution?Footnote ⁴ There is a presumption that law cannot be an empty vessel, subject only to the untrammeled will of the lawmaker. In a way, both positivism and legal realism had the advantage of dispensing with the question of what “constrains” judges. However, in legal theory, the fear that the “emperor has no clothes” persists. We are still trying to find a principle that both constrains judges and provides a yardstick of accountability. However, the blunt truth is that this horse, at least in the Indian case, bolted the door long ago; little constrains judges other than their own judgments.

The Rise of Institutional Theories

The second line of normative critique of judicial roles takes its cue less from normative than from institutional considerations. This shifts the grounds of the debate by asking: Are courts better at performing the tasks that they do perform? On one view, the choice of institution must be based largely on empirical grounds.Footnote ⁵ Whether courts can perform these functions better is a debatable matter. The success of a court often is projected by taking examples of their success and comparing it with executive failure, but it is difficult to make the case that, as an institution, it performs better. It is important to judicial authority that the courts project the claim that they can do a better job in those instances when, in fact, they do a better job. However, there is an inner tension in the articulation of this type of pragmatic approach that courts perform. On the one hand, the arguments that suggest that judiciaries may be better at performing some roles than others (e.g., protecting rights versus governing) often already have a prior normative theory of what would count as performing better. On the other hand, if it were a strictly empirical theory, the test of what works better simply might be whatever a democratic political order is willing to tolerate. “Better” here simply means what is broadly acceptable to the political culture of a society.

There is a different institutional theory of the judiciary's role based on a normative idea of legitimacy. Waldron, for instance, was critical of the idea of judicial review, based on the idea that it is incompatible with the dignity of legislation and representative democracy more generally.Footnote ⁶ This argument was subject to various critiques,Footnote ⁷ but it had the singular merit of trying to place the Supreme Court's role in the context of democratic theory. Why should society repose so much authority in “unelected” judges at the expense of representative institutions? In India, also, one of the grounds for criticizing judicial overreach has been just this. Do judiciaries, in some ways, undermine democracy?

This argument also has a parallel sociological version. Recently, Hirschl provided a powerful version in the comparative context.Footnote ⁸ The claim is that there is a broad movement in liberal democracies toward enhancing the power of unelected institutions. Some scholars attribute this to the complexities of the modern administrative state,Footnote ⁹ but others argue that this is one form in which middle and upper classes continue to exercise influence in the face of being challenged by democratic upsurges from below. I am not entirely persuaded that the reliance on courts (and other nonelected institutions) as institutions of governance is largely a consequence of the fact that the electoral process has seen power shift away from traditional elites. I am less persuaded by this simply because there is no evidence that a shift in structures of representation actually threatens the economic power of traditional elites. Democracy has proven to be more a mode of acculturating rather than overturning new aspirants to existing structures of power. Democracies are considerably more conservative than their defenders acknowledge.

However, this argument has one implicit standard for critiquing judicial power: the idea that judicial power must be assessed from the perspective of those whose class interests it serves. There should be a presumptive suspicion over judicial power because it empowers elites in two ways. On the one hand, the exercise of this power is not independent of the class presumptions – that is, the values and prejudices of judges as a class. On the other hand, the repertoire of tools required to have access to judiciaries – lawyers, legal knowledge, the ability to mobilize experts, and the power to create hegemonic discourses – is predominantly vested in elites. Although judiciaries may cloak their enhanced power in a narrative that they are acting at the behest of the poor, and although they provide relief to the causes of the poor in some cases, it is difficult to argue that judiciaries have been potent instruments of addressing structural inequality or empowering the poor. They may appropriate the rhetoric of the poor and position themselves as “The People's Court,”Footnote ¹⁰ but they unwittingly serve the interests of the powerful. This argument is not easy to assess; it draws strength from the fact that courts have never been instruments of producing deep structural transformation in societies. They often have been potent instruments of ending unequal treatment, opening doors in the face of invidious discrimination,Footnote ¹¹ but they are far less successful in empowering the poor.

One way that this can be demonstrated in the Indian context is to examine precisely those tools that courts craft to empower the poor. For example, it has been argued that the great innovation of public-interest litigation has become primarily a preserve of middle-class legal activism.Footnote ¹² The Supreme Court also, by and large, has not questioned a middle-class discourse on development and the framing ideology that the executive often uses to intervene in the livelihood and property of poor people. In fact, it can be argued that had the judiciary exercised more stringent standards on the executive in its exercise of eminent domain, India's land conflicts would have been less acute. In the early days of eminent-domain–related litigation, for example, the courts were driven by the thought that the state should not exercise eminent domain without adequately compensating landowners. However, as some have argued, the Court paradoxically made it possible to dispossess the poor of their land.Footnote ¹³ Neither did the courts exercise great scrutiny until very recently on the state's invocation of public purpose. It is perhaps unfair to place on the judiciary the burden of economic transformation. However, it would be difficult to convince those who are suspicious of the rise of judicial power that the courts are not dominated by middle-class conceptions and interests.

Of course, these two versions of the democratic narrative – that is, courts as a replacement for representative government, and courts as an instrument of deeper empowerment of the dispossessed – can work at cross purposes. It can be argued that courts are deferential to economic inequality because they are concerned to not be seen as intervening too often in economic decisions of the executive. Therefore, democrats must live with the idea that it is quite possible that judicial restraint, as a form of deference to representative government, ultimately may condone serious economic inequality. Conversely, interference on the grounds of a substantive conception of equal democratic rights indeed may subvert the authority of representative institutions. There is no easy answer to this question: which side one comes down on may depend ultimately on what one fears the most. However, implicitly, a court seeking to enhance its own legitimacy will balance these two considerations. This balancing, however, is not of principles as much as a determination of what the court can do without undermining its authority. It must expand rights to provide a sense that this is a court for the poor; it also must relax rules on standing to ensure access to justice on many consequential issues. At the same time, however, it cannot do this in such a way that society thinks electoral democracy is being undermined. The type of argument that Khosla makes about the gap between the rights that the Supreme Court promulgates and the remedies it fashions perhaps can be explained by exactly this tension.Footnote ¹⁴ It is both providing enough to be a locus of hope but also restraining itself in its actual effects so as not to provoke a backlash.

I return to the question of democratic legitimacy of judicial review. There may be one way to approach this question that is consistent with the arguments being advanced herein. The question, “What is the degree of legitimate judicial intervention?,” may be answered behaviorally. Judicial intervention is legitimate to the extent that it does not provoke, formally or informally, a democratic backlash. What counts as a backlash, and how it could be measured, is a tricky question. In part, this is because the form in which this backlash may be experienced is a function of institutional design or of contingent political configurations. A judicial system may be more emboldened to the extent that it judges that its patterns of intervention will not provoke a backlash sufficient to undermine it authority. In India, the argument was made that the increase in judicial power coincided with the advent of coalition government, in which (except in cases pertaining to reservations) it was unlikely that the political class could mobilize against judicial decision making. It is not a coincidence, on this view, that the first act of a full majority government in twenty years was to enact a Judicial Appointments Bill to increase the executive's role in judicial appointments.Footnote ¹⁵ Representative institutions can take on judiciaries to different degrees in different contexts. Again, the idea is that there is no antecedent clear theory about which branch of government should exercise which power. It is disingenuous to think that such a theory exists. Rather, there is a set of institutions created by a constitution. Over time, these institutions will jostle for power, in relation to one another but also with regard to their standing in the eyes of a wider democratic public sphere. The real question is not what judges should or should not do; it is what they think positions them in this shifting contest over legitimacy.

Public Reason?

Surely, there must be something to discipline judicial reasoning. One possible line of reasoning that seriously considers the relationship between law and democracy tries to provide an answer. This is the Habermaisan idea that what judges should do is “not appeal to reasons that they believe to be right, but to reasons that they sincerely believe that other reasonable people would and should accept as justification.”Footnote ¹⁶ This theory argues that what courts are promulgating is not their own normative predilections, or fidelity to a technical conception of law, but rather the advancement of public reason.

There is a distinction to be made between transient legislative majorities and the will of the people. Transient majorities may subvert popular sovereignty. The rise of judicial power, at least in a country such as India, cannot be separated from a perceived sense of crisis in liberal democracy expressed in an ever-growing gap between legitimacy and representation. Judicial review, in some ways, steps in to preserve our status as free and equal citizens. This often involves protecting rights and often also protecting the institutional features of democracy. In short, courts preserve the constitutive rules that make citizens free and equal and allow them to exercise popular sovereignty. In the most general form, courts step in to bridge the gap between representation and legitimacy. Representation refers to a process of popular authorization. Legitimacy concerns the reasons that people have for accepting the political relationships in which they find themselves. Again, to simplify, the normative ideal is: “Are the relations in which citizens stand to each other as citizens acceptable to them? Are the policies they find themselves governed by those that they would have chosen as free and equal individuals?” The gap between legitimacy and representation arises when we think that practices of popular authorization do not necessarily or often routinely produce policies, or the promulgation of values that we would have chosen as free and equal individuals.

Judicial review in the classic sense is a safeguard to protect the fundamental identity of the state and its basic political form. In liberal democracies, it will protect the identity of a political system as a liberal democracy (e.g., protect rights). However, once we acknowledge a gap between representation and legitimacy, the question arises: Why cannot the scope of judicial power extend beyond traditional areas such as preservation of rights? What role can courts perform in bridging the gap between legitimacy and representation in other areas as well, including policy formulation? Hence, we see the road to promiscuity.

There is something attractive about the idea that courts should be exemplars of public reason. This overcomes the arid formalism of law, on the one hand, and the imperiousness of normative theories on the other. However, a conception of public reason must wrestle with a challenge. First, which reasons would be acceptable to all in the face of deep disagreement? These theories often risk smuggling in a thick account of normativity in their conception of what counts as a good reason, or they are reduced to the idea that judges are looking for something that would be acceptable to reasonable people (i.e., something that will not provoke widespread dissent among reasonable people). The test of legitimacy, then, becomes a form of political or social acceptability.

In summary, it seems unavoidable that judges can do anything but intuit what they think social acceptability might be. The questions are: How do they do this? Is the discovery of acceptability simply after the fact? How does this sense that they are engaging in public reason both empower and constrain their authority? Or is public reason simply another intuitive concept?

Public Legitimacy

This question becomes particularly pertinent when courts play a larger role in society that surpasses the narrow conceptions of rule of law. It is widely agreed that the Indian Supreme Court has played an important role in the preservation of Indian democracy. Its promulgation of the idea of the “basic structure” of a constitution, that no constitutional amendment can abrogate, is widely regarded as an excellent example of a judicial innovation that expressed and maintained India's commitment to democracy.Footnote ¹⁷ The idea of the core political identity of the constitution may be rough at the edges, but there is no doubt that it has a strong conception of democracy and secularism at its center. Courts in India also have greatly expanded the ambit of rights through a creative interpretation of Article 21, and they have made several social and economic rights justiciable. The remedies that the Supreme Court crafts often may not be commensurate with its ambitious rhetoric, and the extent of real change that these decisions produce can be debated. However, there is no doubt that the Court has been a vehicle for the expansion of Indian democracy's commitment to justice. Finally, courts increasingly have become institutions of governance. They have advanced the cause of democracy by holding the executive accountable on a range of issues, from corruption to procedural impropriety. They increasingly have forced change in laws that many believed gave politicians undue protection. In summary, the Supreme Court may have overstepped its bounds, but it has deepened democracy by protecting India's political identity, expanding the scope of democratic justice, and producing a modicum of accountability.

In recent years, however, the judiciary has acquired more prominence in Indian democracy as an institution of accountability. Courts perform many routine functions, but it is not an exaggeration that in public consciousness, there is – in different time periods – a particular set of grounds on which courts gain legitimacy and prominence in ways that enhance their power. The role that a court performs in a network of institutions changes considerably and often is determined by broader political circumstances. This role is not determined by a simplistic reading of the text of the constitution or even by a consistent judicial philosophy. It often is determined by what the court sees as the nature of the threats to democracy. It carves out an institutional niche based in part on a sense of gaps in broader democratic institutions. In filling these gaps, however, a court must make a delicate judgment on the nature of its intervention, that such an intervention must have some type of democratic legitimacy. One easy test of democratic legitimacy is whether a court's decisions are overturned by the legislature or whether a backlash is provoked. The backlash may take many forms, but most prominent is the desire to control the judiciary. The extent to which this can be done, of course, will be determined by the nature of majorities in parliament. A parliament dominated by a single party with large majorities will find it easier to amend the constitution or pass a law to overturn a judicial intervention. On issues in which there is almost complete political consensus, the court's decisions will be overturned easily. This is the case with reservations cases in India, in which any attempts by the Supreme Court to roll back, limit, or clarify the terms of reservations (e.g., the Inamdar case)Footnote ¹⁸ have been swiftly countered by constitutional amendments. It is not an accident that the maximum number of constitutional amendments in recent years has been occasioned by reservations decisions of the Supreme Court.Footnote ¹⁹ In some cases, fearing a democratic backlash, the Court refused to implement its own past rulings on reservations that had not been overturned by the legislature. A striking instance of this is the Court providing an exception to its own 51 percent ceiling on reservations for the State of Tamil Nadu, where reservations are as high as 69 percent.Footnote ²⁰

In most cases, however, what the Supreme Court faces is not a direct “empirical” test of the democratic validity of its decisions. Rather, it faces something more complex that requires a democratic judgment of a different type. It must be seen, in the public perception, as not abdicating its responsibility. In fact, judges often speak about the need to discard purist or theoretical conceptions of the proper role of judges in the face of dire social need. However, the exercise of power must be sufficiently “popular” such that politicians perceive that there will be a political penalty involved in overturning a court intervention.

However, this judgment of democratic legitimacy is often both unavoidable and tricky. The political penalties involved in overturning court interventions are not only a product of the nature of majorities in parliament. They also are a product of the comparative credibility of different institutions. For instance, the Indian Supreme Court – much against the articles of the constitution – had acquired powers to appoint judges to the higher judiciary in a way that minimized the role of any other branch of government (through what are commonly known as the “three Judges Cases”).Footnote ²¹ For almost two decades, this ruling was not overturned or challenged because there was a sense that doing so would be perceived as political interference with the judiciary. In 2014, a Judicial Appointment Bill was passed that created a new mode of appointment to the higher judiciary. Although judges will have a predominant voice in this selection, the bill makes provisions for the law minister and two other appointed representatives to be part of the selection committee. What made a near-unanimous passage of this bill possible were charges of favoritism and corruption within the judiciary. In summary, the outcome was determined by the perceived comparative democratic credibility of the two institutions: the more the judiciary's credibility on appointments diminished, the easier it was for an executive to reinsert itself.

However, the crucial point is that in carving out a role for itself, the Indian Supreme Court is looking outward to a conception of public legitimacy – as it were – rather than inward to the text of the law or upward to a self-evident provision in the constitution. Of course, this does not suggest that it dispenses with the law or the constitution; rather, it must deploy them in ways that it believes will command democratic legitimacy. Its formal tools (e.g., reference to the text, statute, or precedent) do not determine what the Court does. They comprise the form in which judicial arguments are expressed, in which the substance is driven primarily by a sense of democratic and social purpose. Perhaps this explains why Indian courts are capacious in the sources that they cite and in their relative lack of reticence in acknowledging different sources of law, including international law. To paraphrase Deng Xiaoping: “What does it matter if the cat is black or white as long as it catches mice?”

Operating within the horizons of democratic legitimacy poses challenges. The moment that one invokes concepts such as “public interest,” as in public interest litigation (PIL) cases, the questions become: How do courts discipline their reasoning on what counts as the public interest? What balance of considerations determines the public interest? What does it mean to adjudicate the public interest in the face of competing versions of the idea? Which canons of reasoning are appropriate to such a concept? Answering these questions would require another chapter, but the pertinent point is this: inevitably, court interventions are shaped by judgments about public legitimacy. Yet, catering to this idea often involves dispensing with classic rule-of-law characteristics. The next section illustrates this by examining the Supreme Court's recent accountability jurisprudence.

The “Accountability” Court

As suggested previously, the grounds on which the Supreme Court has carved out its legitimacy have varied over time. It often has been a function of the comparative credibility of other branches of government and of the nature of the threats it perceives to the polity. To simplify, one account of its evolution might be as follows. During the 1950s and 1960s, the Supreme Court operated in a political environment in which the executive had large reservoirs of political legitimacy based on the legacy of the nationalist movement, a mandate for far-reaching social change. Under these circumstances, the Court's main role and niche was as a bastion of procedural safeguards. What often is perceived as the Court's conservatism during the 1950s – that is, its fidelity to procedural issues, particularly in cases involving the right to property – can be reinterpreted as an attempt by the Court to carve out its own authority on the grounds of institutionalizing procedural safeguards in a young democracy.Footnote ²² The executive responded by using its legislative majorities to amend the constitution, but it did not need to attack the judiciary as an institution. The Court did just enough to show that it mattered as an institution but did not stand in the way of the legislature once it had spoken through constitutional amendments.

During the late 1960s and early 1970s, the legitimacy of the executive was eroding in an era of great political tumult. The Supreme Court perceived two unprecedented types of threats to the constitution. The first was the genuine fear that democracy might come under threat and the Court needed to draw a red line over the extent to which transient majorities could alter the constitution. Although occasioned by property-rights cases, both Golak Nath and Kesavanada tried in different ways to draw this line: the first case by enunciating strongly a set of rights that no parliament could abrogate; the second case by invoking something more architectural, a basic structure that parliament could not overturn.Footnote ²³ The Court positioned itself as a protector of democracy. The second threat was that the executive would interfere more directly with the functioning of the judiciary. Whether this was occasioned by the judiciary's own enunciation of a rights doctrine being a threat to the government's economic agenda or it was in keeping with the encroaching authoritarianism of the time is a debatable matter.

The Supreme Court did not acquit itself well during the Emergency; however, post-Emergency, it rehabilitated itself in three different ways. First, it began to expand the scope of rights to include social and economic rights. In a sense, the Court built on the political populism of antipoverty schemes to nudge the state in the direction of being a welfare state. Formally, the Indian state has always been preoccupied with poverty. During the 1950s and 1960s, there was still a sense that India's antipoverty aspirations had to be constrained by its means. In the 1970s, Indira Gandhi herself made “Garibi Hatao” a political slogan, thereby acknowledging that there was no irrevocable constraint on removing poverty. Thus, the policy became more receptive to the idea of social rights. The Supreme Court made the directive principles justiciable through Article 21. Second, it innovated through the creation of PILs that relaxed rules on standing and made the Court more accessible in matters relating to public interest. Third, it slowly began to assert its supremacy over judicial appointments in the name of judicial independence. Arguably, none of the three moves was simply a function of reading the law or the constitution. It is true in any judicial culture that courts subsequently may discover meaning in a statute or constitutional text that hitherto had remained undiscovered. It is equally true that external circumstances – for example, shifts in popular expectations of the state and the actual conduct of the executive – give the courts a new role.

Beginning in the early part of this century, the Supreme Court tried to personify a new zeitgeist: accountability. As political corruption became the central democratic anxiety, the Court began to address this concern more centrally, as discussed later in this chapter. First, however, three caveats are in order. First, the brief history sketched here should not be read too literally or teleologically, as if there were a neat progression from procedure and traditional rights to social and economic rights to governance. The actual history of the Supreme Court's conduct is more complicated. In the case of civil rights, for example, the Court has not always provided the strongest of protections. It has upheld the constitutionality of the Armed Forces (Special Powers) Act of 1958, legislation that gives extraordinary power and discretion to military personnel in peacetime.Footnote ²⁴ The Court has been less than clear in its protection of artistic expression, such as in its refusal to permit the publication of a book about the Hindu philosopher Basaveshwara.Footnote ²⁵ The Court recently made a shocking decision that overturned the decriminalization of homosexuality by a lower court.Footnote ²⁶ It could be argued that even in the case of Fundamental Rights, the Court has operated with a sense of democratic acceptability: these happen to be areas in which, to put it simplistically, the political culture is more ambivalent than the logic of these rights or that the requirements of human dignity would warrant. This illustrates the point that the Court has tried to grapple with narratives of democratic anxiety, and its role and function have changed in response to those anxieties.

The second caveat, in the Indian case, is that the Supreme Court is not a monolithic institution. Because of the short tenure of chief justices, great turnover in benches, and inordinate effects of individual judges in small benches, there can be substantial volatility in the interpretation of law.Footnote ²⁷ Some scholars argue that to have a jurisprudence that is more considered and coherent, the Court must have a more stable system of constituting benches. This is a debatable matter. It can be argued that in a polity beset by differences, it is precisely this perceived lack of stability that can be attractive. It is precisely the fact that the Supreme Court is not always predictable, that its positions cannot be deduced from consistent ideological predilections or the requirements of legal formalism, that makes many litigants think they have a chance at relief from the Court, even against legal odds. A court captured by a more stable and coherent body of jurisprudence, paradoxically, may inspire less confidence because it might project the impression of being captured by a particular worldview. The Indian Supreme Court has given all sides a reason to play the game: that is part of its legitimacy.

The third caveat is that even when the Supreme Court formally expanded the scope of rights and provided a new basis for its own power, it was careful to do so in a way that did not impinge seriously on the fundamental workings of electoral democracy. This is best illustrated by the case of social and economic rights. There is the “strong courts, weak rights” paradox of Indian courts in contrast with the “weak courts, strong rights” observed elsewhere around the world.Footnote ²⁸ The Indian Supreme Court in particular is strong on the promulgation of values and new rights; however, the expressive character of what it does is much stronger than the actual implications of what it does. As Khosla brilliantly argued with regard to several rights that the Court has promulgated – including the right to shelter, the right to food, the right to education – what the Court, in fact, promulgates is weak in terms of not challenging the executive or legislature (at least in budgetary terms).Footnote ²⁹ In Olga Tellis, for example, the Court pronounced a right to shelterFootnote ³⁰; however, this right turned out to be limited merely to protection against being evicted for a few weeks, and it did not impose much cost to the state. Therefore, the Supreme Court “courts” public opinion by a type of expressive populism. Charitably, however, this can be read as furthering a social dialogue on particular values. The demands that the Court imposes are weak; therefore, the Court has promoted a dialogue on rights and values, without seriously threatening executive power.

These three caveats reinforce the point that jurisprudence has become as much about a form of democratic negotiation as it is about legal doctrine or first principles. It is about a Supreme Court positioning itself in a democracy; the direction in which it goes will be influenced by the nature of that democracy.

The Court as an Instrument of Accountability

As the nature of democratic anxieties shifted, the Supreme Court also began to shift in its role and functions. Corruption always has been a major issue in Indian politics, but beginning in the late 1990s, the issue acquired greater prominence. The spectacular growth of the Indian economy in the early part of this century increased the scale of rents available to the state manifold. In Vineet Narain (1998), the grounds for judicial intervention in corruption cases were established.Footnote ³¹ Justice Verma cited, with approval, seven principles of “Standards in Public Life” recommended by the Lord Nolan Committee in the United Kingdom: selflessness, integrity, objectivity, accountability, openness, honesty, and leadership. He then proceeded to note:

This then was the basis of the Supreme Court invoking the doctrine of continuing mandamus, which involved measures such as directly supervising corruption investigations. Nevertheless, from the late 1990s to the early part of the twenty-first century, the Court had little success in positioning itself as a major player in anticorruption cases. This was due to two reasons. First, the number of actual prosecutions of high-profile politicians was minuscule; indeed, the argument often was made that the Court is against corruption but not necessarily against corrupt politicians. Admittedly, the Court had little role in determining the nature of evidence presented. By and large, however, it was true that the political class was not threatened by the Court. Second, even if the lower courts delivered guilty pleas, politicians could continue their political career as long as the case was in appeal.

This situation changed after 2010. There were a series of major scams, with the comptroller and auditor general (a constitutional authority) alleging large-scale arbitrariness in the allocation of natural resources, such as spectrum and coal mines. This led to the creation of an anticorruption popular movement, which arguably was the central axis of the 2014 elections. Again, it took this democratic “crescendo” for the Supreme Court to position itself in the new democratic zeitgeist, which had three visible manifestations. First, the Court now became less sympathetic to the argument that the political career of legislators should not be jeopardized as long as a conviction was in appeal. It was the first time that the political class was truly threatened by a Court decision. In Lily Thomas, the Court held that a legislator would be disqualified if convicted by a lower court.Footnote ³³ The government tried to overturn this judgment by promulgating an ordinance, but it was unsuccessful. Second, and perhaps coincidental, convictions of major politicians – particularly chief ministers – increased. As a result of the decision in Lily Thomas, these convictions now had real political consequences. Third, the Court subjected government allocations of natural resources to greater scrutiny.

There are technical details in all of these cases, but what is striking is that they did not give the Supreme Court pause in ways they might have only a few years ago. Indeed, the changing nature of politics and democratic discourse had changed the ways in which the Court perceived political risk. For instance, the Court's position in Lily Thomas upheld the claim that legislators should be disqualified if convicted, even if an appeal was pending. In previous cases, the Court had been more nuanced in its approach to the question of disqualification. To see how nuanced the Court was, consider the following question and its answer: What is the best case that can be made for stating that a Member of Parliament (MP) must not be disqualified immediately after conviction by a lower court? That he should be given, at most, ninety days to appeal, to have a higher court look at the conviction, and be disqualified immediately if the appeal is not admitted. Why might this safeguard be necessary?

In the case of a sitting MP, the ramifications of the judgment extend to government as a whole and possibly to a democratic verdict. Consider a scenario in which a government has a thin majority of one or two. A lower court convicts an MP and the government falls as a result. Of course, if the MP is guilty, the court has no choice but to convict, regardless of the consequences. However, it is not entirely unreasonable to think that – given how significant the consequences might be for government – it would be better to have a superior judge at least ascertain that the lower-court judgment was not mala fide or deeply erroneous. The argument is not that MPs should not be disqualified, but surely it is not unreasonable to seek a safeguard against one mischievous or incompetent judge causing a major disruption in government. Given the high rates of appeals that overturn lower-court judgments, this is not an unreasonable concern. The idea is not to protect the wrongdoing of individual MPs; rather, it is to prevent the possibility of the House of the People and government being waylaid by a judicial error. Exactly such a possibility had been hinted at in an older constitutional judgment in the case of K. Prabhakaran.Footnote ³⁴ Here, the Court had suggested that it was within legislative competence to provide safeguards for the integrity of the House of the People as a whole.

This background must be recalled in the discussion of another point often raised. Why should a sitting legislator and a potential candidate be treated differently? They should not. However, it is not unreasonable for a legislature to simply provide for an extra safeguard, given the potential political consequences. Again, in K. Prabhakaran, the Supreme Court had left open the possibility that the legislators, for certain purposes, may be singled out as a special class, provided that there was reasonable public justification for doing so. These considerations may not be decisive, but they are not without weight.

This is where a background reading of democratic anxiety makes a difference. The democratic mood had become more punitive, with a growing sense that fine distinctions in the law were becoming ruses to protect corrupt politicians. In K. Prabhakaran, the Supreme Court gave more weight to possible disruptions to the House than might be caused by disqualifications. In Lily Thomas, it gave more weight to the consideration that politicians were not being held accountable. The relative weight assigned to these two anxieties cannot be understood simply by invoking doctrine; it was a function of the Court's democratic judgment.

In some ways, what the Court was doing was recognizing that the “Rule of Law” in the larger sense (with uppercase letters) was being undermined by the “rule of law.” The arsenal of legal doctrines, technicalities, and possibilities was being used in ways to convince people that politicians were not being held to account. Hence, it used what was, given its own history, a blunter ruling. The same tension between the “Rule of Law” understood as a broad regime of accountability that Justice Verma had outlined using the Nolan Report and the “rule of law” was manifest in two of the Court's most important recent decisions: the Telecom caseFootnote ³⁵ and the Coal case.Footnote ³⁶

In the Telecom case, the Supreme Court delivered a major order invalidating the government's award of spectrum licenses to several major telecom companies. These licenses were cancelled after the comptroller and auditor general's report alleged that their award caused a major loss to The Exchequer. Had these licenses been awarded through an auction, far greater revenue would have accrued to the government. The Court ruled that allocation of natural resources was a public good and therefore was subject to scrutiny by the judiciary. More controversially, it also ruled that natural resources should be allocated by auction, treading on what many regarded as a possible policy area.Footnote ³⁷ The decision was a serious intervention. For the first time, the Court made it clear that the judiciary would scrutinize the allocation of natural resources, not only on procedural grounds but also by some criteria of the public good. Only time will reveal the true significance of this decision. Decisions acquire significance in light of subsequent history. Vineet Narain was a legal landmark, but its actual effect on governance was slight. Decisions such as the three judges cases were made on dubious legal grounds, but they salvaged the autonomy of the Supreme Court as an institution. Even the famed Kesavananda, a complex and contradictory mélange, acquired significance through later developments. For the time being, however, it appears that the Supreme Court has established a powerful idea that whether an allocation is in the public interest will be judged by courts and not by the executive.

In the Telecom case, the high-minded and independent assertion of important constitutional principles is praiseworthy and the observations are admirably pointed. Some of the Supreme Court's observations will reform governance for the better by forcing the executive to justify its decisions on a more considered basis. In this sense, the Court is instigating a larger movement from executive discretion to an exercise of public reason. However, the orders that the Court delivered also raise legal puzzles. Furthermore, there is an “odor” of politics surrounding some observations. Can we use this judgment as an opportunity to draw correct institutional lessons? How the institutional dynamics will play out is an open question. Otherwise, the more things seem to change, the more they will remain the same.

The immediate risk is that the Central Bureau of Investigation (CBI) has been made more powerful, which is a serious challenge. In Vineet Narain, the Supreme Court rightly identified two bottlenecks in the system: the CBI's functioning and the sanction for prosecution. The Court has tried to fix the second bottleneck by stipulating a three-month limit. This is a sensible idea, but its actual effects will depend on other reforms. If this problem is fixed without reform of the CBI, there will be major perverse consequences. We are rightly fixated on the problem that the guilty go free; we should be equally concerned with the fact that there is little protection of honest public officials. It is not a state secret in India that the one mechanism by which officials who take a stand often are threatened is a CBI inquiry. With the filtering mechanism of a sanction to prosecution now eliminated and locus standi on complainants relaxed, it is possible that the state will acquire one more instrument to threaten those who stand in its way rather than more efficiently prosecute the guilty. This will be easier in a climate in which all government servants are presumed guilty by the larger public. It is facile to state that the honest have nothing to fear; after all, they can obtain justice in the courts, eventually. The political overtones to so many recent investigations should make citizens wary of the CBI. The degree of psychological coercion that the state and the CBI can deploy is immense. There but for the grace of God goes every honest civil servant. After Vineet Narain, there was a euphoric delusion that an external monitoring of a CBI investigation was possible. In principle, the Central Vigilance Commissioner can do that, but it has never been clear what this means operationally. Sins of omission and commission are difficult to detect. Under the illusion of monitoring, more injustice can be legitimized. Therefore, unless the culture of the CBI is reformed and made accountable, the good consequences of this judgment will be short-lived.

The second governance implication is that there is no question that the judiciary can scrutinize any policy for arbitrariness and unconstitutionality; it can do so even for constitutional amendments. Furthermore, the Supreme Court has made a nuanced case for why the first-come-first-served policy for the allotment of spectrum – at least in 2007 – was arbitrary. However, this debate over judicial review has been hampered by the unhelpful view of practical reason. There is an assumption in much of the discourse that for a decision to not be arbitrary, it must be derived from a rule. One sign of this is the way “discretion” is considered. Discretion has become a dirty word; it is automatically associated with arbitrariness. However, discretion is the power or right to make official decisions using reason and judgment to choose among acceptable alternatives. These decisions must be justified through public reason; however, giving reasons is not the same as making the decision fall under a general rule. We are called on to exercise judgment precisely because there are tradeoffs to be made and externalities to be taken into account. Elected governments must make these calls, not unelected judiciaries. There is a tendency in Indian courts – across different domains – to suspect the exercise of judgment. Therefore, their solutions often are to propose rules, so that no discretion remains in the conventional sense of the term. We inherently suspect the exercise of judgment and overcompensate for it by peddling the illusion that rules are the only effective way to check arbitrariness. Between the binary of arbitrary discretion and rigid rules lies the space for public reason – and few are willing to occupy it.

In the Telecom case, both the government and the judiciary committed this fallacy. Government has the discretion to decide policy. It could have concluded that revenue maximization is not a worthy objective; giving the spectrum at no cost might be conducive for consumers or developing technology. However, it should have given a clear public justification for its decisions rather than relying on technical cover of precedence and so forth. The government's defense of its policy was shockingly feeble; the reasons adduced were vague and general. The Supreme Court was correct to see through it. However, this point is more important: democracy is not decision making only by rules or precedence; it is government by public justification. Barring a few exceptions, the culture of government is not attuned to clearly recording its reasoning for decisions and politically defending them in public. The judgment is a severe indictment of professionalism in government across the board judged by the yardstick of public reason.

The Supreme Court was not convinced that the government had adequately justified its decisions. However, it went to the other extreme by suggesting the idea that auctions are virtually the only legitimate way of disposing of natural resources. Although often true, the validity of this proposition depends on a number of factors: context, market conditions, and the tradeoff between different objectives. These can be a matter of debate and judgment. It almost seems to remove legitimate government discretion – understood in the correct sense of the term. Its overreach is not that it held a policy decision to judicial scrutiny; rather, the overreach is that it has used a generalized suspicion of the executive to prescribe a policy, with that also framed in terms of seemingly general rules. In the case of the telecom sector, at this juncture, it probably does not make any difference; it is the right thing to do. In the meantime, it has given – by implication – license to question first-come-first-served under the next government, when it ostensibly had more justification. How this will be used as a precedent remains to be seen. The Supreme Court, quite rightly, insisted that any procedures for allocation should be transparent and nondiscriminatory. In cases in which there are multiple and often competing objectives, translating this into practice will be a challenge. Courts should provide reasonable latitude on what counts as a fair procedure relative to specific objectives. Government without the right degree of discretion is impotent, which is to be feared as much as arbitrariness. However, if new standards can be set on public justification, much will have been achieved, and there is evidence that this may be achieved. There is no question that after this judgment, no government will be able to arbitrarily dispose of natural resources. The Supreme Court sent a signal that it will hold the executive to account. What standard it will hold it to remains an open question.

However, this strong signal at an aggregate level came at a cost in terms of “rule of law” (in lowercase letters). In its implications for political institutions, the two Telecom orders are distinctly disconcerting. Simply stated, they seem to have put a “seal of approval” on the evisceration of two cardinal principles of democratic government: ministerial responsibility and Cabinet responsibility. The form in which this happened renders the Supreme Court somewhat less than convincing and internally inconsistent. Consider, for example, the following rather odd observation:

This is an astonishing character certificate to the prime minister, but it is legally beside the point. How could anyone make a determination about what someone else, without a doubt, would have done in the future? A supposition about character, not a consideration of official responsibility, is the grounds for exoneration. What makes this logic utterly specious are the facts in the second Telecom order cancelling licenses. The Court indicted the telecom minister, Mr. Raja, for not following the prime minister's advice, but it did not follow through on the implication of its own reasoning. In this instance, the prime minister and the Empowered Group of Ministers (EGoM) knew that a wrong policy was being pursued and did not act on it – so much for the assurance that the prime minister without a doubt would have acted if only he knew. There is an old joke that if you refute yourself, it does not count as a refutation; the Supreme Court seems to be following a similar logic. The prime minister advised the minister but he did not do anything to stop the policy from taking effect. In the first order, the prime minister was exonerated because he did not know but would have acted if he had known. In the second order, he knew but merely urged Mr. Raja to do something else. His knowledge, which should have at least signaled that he was capable of failing to act, is used as evidence for exoneration.

Two things are odd about this situation. The first is the presumption that the prime minister could not possibly abdicate his responsibility. One should not prejudge the matter as one should not prejudge anyone's guilt until the full facts are available. However, the way the judgments are written, it appears as if the prime minister, simply by virtue of who he is, could not have been guilty of any omission. This is an amazing personification of an office. Although it is a different type of case, even Judge Saini's order on the then-finance minister, Mr. Chidambaram, has this quality.Footnote ³⁹ It does not contest the fact that Mr. Chidambaram knew and concurred with the policy. Contrary to the spirit of the Supreme Court, it finds that the policy decisions per se were not illegal or arbitrary. Much is being made of the fact that in the context of a criminal proceeding, one should make a distinction between possible criminal liability and possible moral or political responsibility. However, this distinction is beside the point. The Court was not being asked whether to judge Mr. Chidambaram as guilty; it was merely being asked whether there were grounds for further investigation. Given the Court's own pointed description of how much Mr. Chidambaram knew about the policy and how he was in a position to stop it, the refusal to further review the matter appears to be inconsistent with its own logic. The standard of proof required, even to merely inquire further, by definition would be impossible to meet. Nothing in this argument implies that the prime minister or Mr. Chidambaram did anything wrong. It merely points out that the Court's reasoning behind the refusal to inquire further seems less than convincing.

However, the fundamental institutional mistake in the Telecom case seems to be to eliminate any idea of ministerial or Cabinet responsibility. In the first order, officials of the prime minister's office are chided for not placing the full facts and legal issues before the prime minister. It is an interesting legal question as to whether this should constitute grounds for an appropriate investigation and action against those bureaucrats. Did they – and there seem to be many – merely make a mistake or was it by design? Surely, ministerial responsibility enters the picture, or is every official in the hierarchy now entitled to “get off the hook” because they were not fully appraised? It is true that the petition had not specifically alleged any wrongdoing by the prime minister. However, the Court rebukes the government for serious constitutional arbitrariness in the Telecom decision. It could have taken the view that there was nothing illegal about the policy but, having passionately indicted the policy, the Court failed to ask any probing questions of all those in a position to stop it. Having drawn a mighty sword on behalf of accountability, the Court than let it fall entirely on bureaucrats, corporates, and the telecom minister Mr. Raja alone. Given the rebuke to the Telecom Regulatory Authority of India, Mr. Raja, and civil servants, it is surprising that it does not show the slightest interest in the core question: Given that the prime minister himself seemed to know that a wrong policy was being acted on, given that an EGoM could have stopped it, why did the Court not act? In a parliamentary system of government, is there absolutely no conception of collective responsibility remaining? Is the prime minister's office simply like that of a private citizen's – all that he can do is write a letter? The Court seems to be rewriting the principles of parliamentary democracy – that is, a minister is not responsible for actions taken in his department. The Cabinet is not responsible, even when it has full knowledge and it empowered the decisions taken by one of the ministers. This can have potentially major governance implications because key actors are not being held responsible for decisions. However, it is difficult to fathom the revolutionary character of what the Supreme Court has enunciated. At least Judge Saini's order in the lower court was premised on stating that no wrong was done in the policy. However, the Supreme Court, having stated that the policy was arbitrary, implies that ministers have no responsibility for the fundamental decisions of their department, and the Cabinet has no collective responsibility for any policy decisions. Never has there been such a subtle subversion of the basic principles of parliamentary government in India, despite that fact that “accountability” was achieved, in a manner of speaking.

The other issue that the Supreme Court did not address was a nuanced consideration of the rights of the investors. To send a message, the Court cancelled the licenses of all telecom operators, without providing individual hearings. A similar move was made in a more recent decision when the Court cancelled the allocation of coal blocks, stretching as far back as 1993!Footnote ⁴⁰ What is jarring in the Coal decision is not only that ordinary citizens will realize that there is no statute of limitations in India. The short-term economic consequences also are not an issue; the Court cannot be instrumental about the rule of law. The real issue is that if the state makes a mistake and, twenty years later, a private party that was a beneficiary of that particular contract with the state is caused to suffer because of mistake, what protection does it have? How is a private citizen supposed to know whether the state did its procedural due diligence? How does a citizen know whether the authorized committee met for three minutes or had a full discussion, whether all companies were properly vetted, and so on? Again, corruption is an easy issue: contracts in which there is corruption must be annulled. However, as in the Telecom case, the Court detected a massive lapse without fixing anyone's responsibility. The CBI becomes involved only if it can prove real corruption. Who bears the cost of the state's lapse? There is little protection for good-faith investors. Corruption has clouded our thinking to such an extent that these deeper underlying issues are being given short shrift. There is a vast range of areas in which this is at issue. For example, in land acquisition, if the state does not conduct a social-impact or an environmental-impact assessment appropriately, should the costs be borne only by the private party that accepted a contract in good faith? If corruption privatizes the public exchequer, this type of regulatory structure provides no private-party protection against state recklessness. If the Supreme Court were serious, it should have sent a real message to the state; the only message was that the state easily can pass all costs for its lapses to private citizens. Indeed, it could be argued that the Court implicitly used the suspicion of crony capitalism to deny justice to individual litigants.

What are the general features of these decisions, and what do they imply? It is clear that the Supreme Court produced accountability. Its main aim was to force the government to act according to public reason. It rode on a wave of popular discontent against corruption but, in the process, it managed to eliminate all nuances, all protections for individual litigants. In a strange twist, the Court detected massive wrongdoing in the state but did not fix responsibility. How is this outcome explained? One way to look at it is that the Court is engaging in democratic positioning; it is answering the clamor for accountability. It sets a benchmark for future decisions of the state. At the same time, however, the Court is careful not to venture where the current executive is destabilized, even though its own logic warrants fixing blame. Furthermore, the Court is able to penalize private players, not because they have been proven guilty but rather because fear of crony capitalism is the dominant political narrative. The Supreme Court is an actor in a democratic negotiation; it is not a purveyor of the simple idea of the rule of law.

The Federal Demand in Sri Lanka

Although a small number of individuals and organisations suggested federal-type arrangements in what became Sri Lanka when political reforms were discussed in the early 1900s and in the period leading to independence from British colonial rule, federalism was not the preferred option of Tamil political leaders even at the time of independence in 1948. The preferred option of the main Tamil political party at the time, the All Ceylon Tamil Congress (ACTC), and its leader, G. G. Ponnambalam, was to share power at the centre. Ponnambalam and the ACTC decided to accept a Cabinet portfolio in the first post-independence Cabinet of Ministers, thereby ensuring participation at the centre of government – which, in turn, they thought, would ensure that Tamil concerns and aspirations could be raised and addressed at the heart of where power was located.Footnote ⁴

Only when the Ponnambalam approach was perceived to have failed did Tamil support move to an approach favoured by S. J. V. Chelvanayakam, which stressed the importance of decentralisation and autonomy in Tamil-majority areas of the country. Chelvanayakam broke away from the ACTC and formed the Federal Party; for the first time, the Tamil political leadership espoused federalism as it basic political demand. Chelvanayakam, who was a committed Gandhian, believed in peaceful, nonviolent, democratic means for the pursuit of federalism. He also was willing to consider compromise alternatives to federalism, as he did in 1957 and 1965 when he entered into two well-known agreements with the then-prime ministers of Ceylon: S. W. R. D. Bandaranaike in the Bandaranaike–Chelvanayakam Pact of 1957; and Dudley Senanayake, the so-called Dudley–Chelva Pact of 1965.Footnote ⁵ Both agreements envisaged decentralisation of power to the north and the east, with the possibility of the two areas working together, and were to be enacted by legislation rather than by constitutional amendment. The agreements fell short of basic features of federalism and, on both occasions, opposition from within and outside of the ruling governments forced the prime ministers to withdraw their support for the agreements. It must be noted that during the 1950s and 1960s, Tamil frustrations continued on other issues as well. Laws dealing with citizenship and language that were discriminatory against the Tamils were enacted and legal challenges to such initiatives on the basis that they violated the much-touted minority-protection provision of the constitution, Section 29, failed due to the legalistic and positivist approach of the judiciary. In general, there was growing frustration at the rise of Sinhalese majoritarianism in addition to the failure of Ceylonese governments to respond positively to demands for regional autonomy. Tamil youth and nationalist elements became impatient with the Chelvanayakam approach, which – they argued – had failed to deliver positive results.

The parliamentary elections of 1970 gave a sweeping victory to the United Front, a coalition of the Sinhala nationalist Sri Lanka Freedom Party and the Marxist parties. The coalition obtained a two-thirds majority in parliament and decided to initiate a process of constitutional reform to replace the Soulbury Constitution that had been introduced before independence and had served as the country's post-independence constitution since 1948. A Constituent Assembly consisting of all members of the House of Representatives elected in the 1970 elections was convened and assigned the responsibility of drafting and adopting a new constitution. The fact that the United Front government possessed a two-thirds majority in parliament made it unwilling to accommodate opposition proposals. The Minister of Constitutional Affairs, Colvin R. de Silva, initiated proposals in the form of basic resolutions that were debated and then voted on. He proposed in Basic Resolution No. 2 that the new constitution should include a provision expressly declaring that “Sri Lanka is a unitary state.” The Soulbury Constitution, although clearly unitary in character, contained no such express self-description.

The First Republican Constitution of 1972 made things worse by introducing several other substantive constitutional provisions. Not only did it abolish many of the minority safeguards, including Section 29 of the Soulbury Constitution; it also entrenched majoritarianism in the supreme law of the land. The secular character of the state was severely undermined by the provision that gave Buddhism the foremost place. The language of the majority, Sinhalese, was made the sole official language. The cumulative effect of these new features in the constitution was to further alienate an already frustrated Tamil political leadership and polity. The introduction of the Constitution of 1972 was a major landmark in the process of national disintegration.Footnote ⁶ It is not surprising that the Tamil militant movement commenced in the mid-1970s, exploiting the growing sentiment within the Tamil community that power sharing and decentralisation options pursued through peaceful democratic means for more than twenty-five years had not only yielded no results but that also during this period the situation for the Tamils in fact had become worse.

The Second Republican Constitution of 1978 did not make the situation better for the Tamil people. The framers of the constitution, who believed in a strong, centralised government to promote economic development, introduced an executive presidential system to establish a strong executive. The article declaring Sri Lanka to be a unitary state was reproduced and made more secure because it was part of a series of articles that were granted special protection. These were the so-called entrenched articles, the amendment of which required – in addition to the usual two-thirds majority affirmative vote in parliament – the approval of the people at a referendum. Almost ten years later, following attacks and violence against the Tamil community in 1983 and increasing Indian pressure on the government, the government of President J. R. Jayewardene reluctantly introduced a system of provincial councils, or devolution within a unitary state, through the Thirteenth Amendment to the Constitution.Footnote ⁷

Perhaps because the Thirteenth Amendment sought to introduce devolution within the framework of a unitary state, the devolution was not substantial and secure. It provided for a veneer of devolution while retaining vast powers in the centre. The amendment ultimately failed to grant complete control over any subject to a provincial council and also made it easy for the centre to retake power. There was no clear division of powers between the central parliament and the provincial councils. Furthermore, the fact that the Thirteenth Amendment was incorporated into a constitution that provided for a centralised political structure with a powerful executive president contributed to the retention of power at the centre and to the undermining of effective devolution of power.

Several other constitutional features, many of which are found in federal countries, were conspicuously absent in the Sri Lankan Constitution. There was a constitutional prohibition on judicial review of legislation and there was no independent public service or second chamber to facilitate provincial representation at the centre. The absence of comprehensive judicial review of legislation, combined with the wide immunity given to the president,Footnote ⁸ prevented the president from being made a party to legal proceedings for acts or omissions. It also permitted important sections of the Thirteenth Amendment, including those dealing with police powers and land, to remain unimplemented.

It therefore was not surprising that moderate Tamil political parties totally rejected the existing constitutional framework and called for “substantial devolution of power,” or federalism. The LTTE by this time had waged an armed struggle for a separate state based on the principles of self-determination, nationhood, and a Tamil homeland in the northeast of the country.

The Federal Demand in Nepal

The demand for federalism arose in Nepal as a response to the dominance of a privileged elite group and the economic, social, and political exclusion of a large section of the population. Since its unification as a state in 1768, in one sense Nepal has been a country of disparate minorities, and the diversity of population is far greater than in Sri Lanka.Footnote ⁹ Nepal has more than 120 ethnic groups that speak more than 100 languages.Footnote ¹⁰ The dominant group constituted about 30 percent of the population; therefore, strictly speaking, it did not constitute a majority. However, the centralisation of power in the Kathmandu-based dominant Bahun–Chhetri elite excluded large sections of society, including women, Dalits, ethnic groups or “janajathis,” people from remote and inaccessible regions, and otherwise marginalized groups.Footnote ¹¹

The movement for democracy that developed in the second half of the twentieth century was focused mainly on the powers of the monarchy; multiparty, parliamentary democracy as opposed to the panchayat system; and the degree of influence of Hinduism in Nepali society. The democratisation process reached a significant landmark with the promulgation of Nepal's fifth constitution in November 1990.Footnote ¹² The new constitution, drafted by a commission of constitutional experts, declared that the power of government was vested in the people; accepted the principle of universal franchise (although there was disagreement on who was entitled to be treated as citizens); imposed restrictions on the power of the monarchy; introduced a Constitutional Council, an independent body to appoint persons to important public positions (a feature subsequently copied in Sri Lanka); and included several other provisions that were broadly consistent with basic norms of liberal constitutionalism. Although the constitution recognised that Nepal was a multiethnic and multilinguistic kingdom, it also reaffirmed the Hindu character of the state and declared that the king must be an adherent of Aryan culture and the Hindu religion as the symbol of the Nepalese nation and the unity of the Nepalese people.Footnote ¹³ The dominant language, Nepali (and the Devanāgiri script), was recognised as the sole official language. The constitution prohibited political parties based on ethnicity or religion, and restrictions on Fundamental Rights could be imposed if the exercise of such rights disturbed “harmonious relations subsisting among various castes and communities.” The cumulative effect of many of these provisions was to orient the nature of the state in favour of the majority religion and the dominant language and culture, whereas the unitary state made it easier for the Kathmandu-based dominant elite to continue to wield power to the exclusion of other caste and ethnic groups.Footnote ¹⁴

The Constitution of 1990 therefore was seen as doing little to address structural inequality and the aspirations of the majority of Nepal's population, the excluded minorities. It was not surprising that the Maoist insurgency that developed in the mid-1990s grew in the mid- and far west of the country and other remote areas where poverty was widespread and access to basic infrastructure and the resources of the state was limited.Footnote ¹⁵ The Maoists used the feelings of alienation and exploitation of people from excluded groups and skillfully modified their demands to include those that were responsive to the Dalits, Janajathis, Madheshis, and other vulnerable groups.Footnote ¹⁶ The Maoists were soon seen as the force for change even though the People's War they had launched was accompanied by acts of terrorism, intimidation, and brutality. Among the Maoists' demands were a new constitution drafted by the representatives of the people rather than a commission of elite experts; a secular state; equal rights for women; the abolition of untouchability and caste-based discrimination; the equal status of all languages; mother-tongue education; and decentralisation and local autonomy.

As the conflict intensified after 2001, King Gyanendra declared states of emergency, reconsolidated the powers of the monarchy, and dismissed the democratic government of Nepal. As the Maoists and the democratic parties came together to challenge the king, a collaboration that culminated in the Peoples Movement (i.e., Jana Andolan II), various agreements between them highlighted the need for “full democracy” to address problems relating to class, gender, and region through the “restructuring of the state.” These commitments were further strengthened in the CPA of November 2006 that brought an end to the conflict and contained various commitments with respect to both process and substance for a new constitution for a new Nepal.Footnote ¹⁷

Clause 3.5 of the CPA declared an agreement to “eliminate the existing centralized, unitary state system and introduce inclusive, democratic, and progressive restructuring of the state in order to address problems relating to women, Dalits, indigenous and janajathi communities, Madheshis, oppressed, neglected and minority communities, and backward regions, by ending the prevailing discrimination based on class, caste, language, gender, culture, religion and region.” Whereas opposition to a unitary state and a commitment to state restructuring were made clear in the CPA and other statements made during its preparation, the commitment to federalism was not made explicit until later. Amendments to the Interim Constitution of 2007 and 2008 introduced commitments that the state shall be restructured into a “progressive, democratic, federal system and that accepting the aspirations of indigenous ethnic groups and the people of backward and other regions, and the people of the Madhesh, for autonomous provinces, Nepal shall be a Federal, Democratic Republic.” The Interim Constitution that was adopted in 2007 to provide for an election to a Constituent Assembly and the transition to the new Nepal already had committed Nepal to be a secular state.

Sri Lanka and Nepal: The Contexts Compared

In both Sri Lanka and Nepal, the demand was fueled by (1) majoritarianism (of the Sinhalese Buddhist majority in Sri Lanka) or the dominance of a large privileged group (the Bahun – Chhetris in Nepal); and (2) a strong perception of discrimination and a denial of equality among the nondominant groups and the desire for recognition, respect, and dignity on the part of such groups. Although in Sri Lanka there was a campaign for federalism and an attempt to explore a federal solution during peace negotiations between the government and the LTTE, federalism per se was never accepted by the country at any stage. In Nepal, however, the CPA, the Interim Constitution, and (finally) the main political parties all agreed to an Interim Constitution that committed Nepal to be a federal, secular, inclusive, democratic republic. The debate advanced to a stage in which there was even agreement that the provinces should be based on identity and viability. Ultimately, however, consensus was not reached on the balance between these two criteria and the details for the demarcation of provinces.

In both countries, however, there also was widespread opposition to federalism. Opponents argued that far from promoting unity in diversity or offering a solution to ethnic conflict and exclusion, federalism would only exacerbate ethnic conflict and polarise and divide ethnic groups that, in turn, would endanger the territorial integrity and national unity of the country. Whereas many opponents of federalism were motivated by chauvinism or a belief that the dominant group was entitled to its position of privilege, some opponents of federalism presented arguments that were more reasonable and therefore required a response from the advocates of federalism.

In both countries, the “minorities-within-a minority” critique of federalism evoked many sympathizers, as did the fear that a division into provinces, over time, would result in demographic shifts that would lead to ethnic polarisation. The “minorities-within-a-minority” critique suggests that as attempts are made to address demands for autonomy of minorities in a particular region, the interests of smaller minorities within that region may be ignored.Footnote ¹⁸ In Sri Lanka, the Muslims, for example – the third largest group in the country with a significant population in the east – were apprehensive that constitutional responses to the Tamil homeland claim to the northeast of the island ultimately would be detrimental to their interests. The critique was used in Nepal where, given its diversity, in none of the proposed provinces would a particular ethnic group constitute a majority. The argument was made by critics of federalism that federalism was not appropriate for Nepal or that it would only favour the larger minority ethnic groups and be detrimental to the interests of the smaller minorities.

Sri Lanka

When Chelvanayakam established the Federal Party in 1949, the party in the Tamil language was called the Ilankai Thamil Arasu Katchi, which in English means Tamil State Party of Lanka. The word arasu (state) was ambiguous because it could stand for either a state within or outside of a federal entity. There were no obvious terms in both the Tamil and Sinhalese languages for the word federalism; more recently, terms have been developed that are not completely satisfactory because they could be open to misinterpretation.Footnote ¹⁹ The main opponents of federalism, the Sinhala Buddhist nationalists, accused the Federal Party of deception and deviousness in using a less-threatening title for the non-Tamil-speaking populations while pandering to Tamil nationalists by implying commitment to an independent Tamil state. From the beginning of the federal debate in Sri Lanka, mistrust and suspicion made it difficult for an objective and more rational discussion to develop on the merits and demerits of the federal idea.

The conduct of the Federal Party throughout its existence, however, belied the fear that the party actually was seeking separation. As stated previously, the party on two significant occasions, in a spirit of compromise, agreed to proposals based on decentralisation, which the Sinhalese political leadership subsequently repudiated. During the parliamentary elections of 1970, when Tamil leaders contested several electorates on a separatist platform, Chelvanayakam and the party were categorical in their rejection of secession. The party manifesto declared:

Significantly, the manifesto included a categorical repudiation of separation, as follows:

The candidates who adopted a prosecessionist line were heavily defeated by the Federal Party, which swept the polls in the north and the east. This clearly demonstrated that the Tamil people, at least until 1970, desired a certain degree of autonomy and self-government but had no interest in secession. The situation became more complicated after the adoption of the “autochthonous” republican Constitution of 1972, when the next phase of the Tamil campaign for equality, dignity, and recognition commenced.Footnote ²⁰

Soon after the adoption of the 1972 Constitution, the Federal Party, the ACTC, and the Ceylon Workers Congress came together to establish the Tamil United Front, which later was renamed the Tamil United Liberation Front (TULF). The Vaddukodai Resolution was adopted by the TULF at its first convention held in May 1976. The resolution was based on four key ideas: (1) the recognition of the Tamils as a nation and their historical possession of the northern and eastern provinces; (2) the creation of a free, sovereign, secular, socialist state of Tamil Eelam based on the right to self-determination inherent in every nation; (3) the will of the Tamil nation to exist as a separate self-governing entity; and (4) a call to the Tamil nation and its youth to join the struggle for a sovereign state.Footnote ²¹

The dominant discourse of the main Tamil political movement shifted from a moderate federal discourse to a more assertive discourse based on nationhood and self-determination. This was a reaction to the failure of the Sinhalese political leadership to respond to more moderate demands for power sharing, decentralisation, and (finally) the adoption of the new Constitution of 1972, which further entrenched Sinhala Buddhist majoritarianism and expressly repudiated federalism.Footnote ²²

The Tamil militant movement gained momentum in the late 1970s and 1980s. The tragic events of July 1983, when Tamils in the south of the island – including the capital of Colombo – were attacked and killed and their property looted and destroyed. There was little or no protection from the law-enforcement agencies, described by many as an anti-Tamil pogrom, which led to increased pressure from India on the Jayewardene government to engage in serious negotiations with the Tamil political leadership. At a meeting convened by India in Thimpu, Bhutan, in 1985 – where discussions were held between the Government of Sri Lanka and both the TULF and the Tamil militant organisations – all of the Tamil groups endorsed what was to become known as the Thimpu Principles: that is, recognition of the Tamils as a distinct nationality; recognition of an identified Tamil homeland and the guarantee of its territorial integrity; based on these principles, recognition of the inalienable right of self-determination of the Tamil nation; recognition of the right to full citizenship; and other fundamental democratic rights of all Tamils, who look on the island as their country. The principles were summarily rejected by the Government of Sri Lankan as undermining the sovereignty and territorial integrity of Sri Lanka.

The nature of the discourse and the debate from then onward was different and far more complicated. In the period 1949–1972, the debate was about equality of status with respect to language, equality with respect to dignity, power sharing and decentralisation, or a degree of self-government in areas where the Tamils had constituted a majority population for generations. The Sinhalese majority and their political leaders rejected these claims and, in 1972, they ignored appeals from moderate Tamil leaders that the best way to protect the unity and territorial integrity of the island was by the adoption of a federal constitution. From 1985 onward, the discourse and the gap between the two sides widened considerably. The Sinhalese political leadership was compelled to become more amenable to decentralisation or, as demonstrated by the introduction of the Thirteenth Amendment to the constitution in 1987, to devolution within a unitary state. However, the Tamil political leadership was committed to the Thimpu Principles, the recognition of which in constitutional terms went beyond conventional federal-type constitutional arrangements. The saga of the Sinhalese responding with “too little, too late” continued.

An attempt to finally break the shackles of the constitutional commitment to a unitary state was made when President Chandrika Kumaratunga was elected president in 1995 and her government sought to introduce a new constitution. Her strategy was to introduce enhanced devolution of power and remove the provision that declared Sri Lanka a unitary state – but without explicitly declaring the country to be federal. The proposals sought to overcome the by now well-documented defects in the Thirteenth Amendment by removing the pro-centre bias in the provisions addressing the division of powers, abolishing the concurrent list of subjects and functions that was ultimately under the control of the central parliament, reassigning some of those powers to the provinces, and strengthening the revenue-raising power of the provinces.Footnote ²³ However, there was opposition to many of these pro-devolution features from within Kumaratunga's own party,Footnote ²⁴ whereas the United National Party – the main opposition party – was ambivalent about its support. It finally cited the unwillingness of the government to deliver on its campaign commitment to abolish the executive presidency as the excuse for opposing the new constitution. Sinhalese Buddhist nationalist forces both within and outside of her party accused her government of seeking to introduce federalism by stealth. Federalism, they argued, would divide the country, promote ethnic consciousness and polarisation, discriminate against the minorities within proposed provincial entities, and impliedly accept the concepts of a Tamil homeland and nation. These forces also argued that given the LTTE's maximalist position on Tamil national independence and their lack of respect for political pluralism and constitutional democracy, it would only exploit the enhanced devolution arrangements and the weaker central government to further its secessionist aims.

A further weakness in the Kumaratunga constitution-making initiative was that it was not conducted in a participatory manner but rather through the mechanism of a Select Committee of Parliament. The process, therefore, was essentially closed from public scrutiny. Furthermore, the Tamil militant group, that had by this time eliminated its rivals and assumed a dominance that included control over parts of the northern part of the country, was not involved in the process. Kumaratunga's brave attempt to reach a consensus among the moderates of all communities by addressing what she perceived as the reasonable aspirations of the Tamil people failed. The failure was due to her inability to reach consensus both within her ruling coalition and with the main political opposition, as well as to the fact that it sought to marginalise the LTTE – which had, by that time, become an actor that could not be ignored.

The next occasion in which federalism surfaced as a basis for a solution to the island's ethnic conflict was when parliamentary elections in 2001 resulted in the opposition gaining power in the legislature, thereby creating a cohabitation government with Kumaratunga as president and the former leader of the opposition, Ranil Wickremasinghe, becoming the new prime minister. Under the new government, primarily at the insistence of the new prime minister, the strategy changed. The new strategy was to engage with the LTTE and seek agreement on political and constitutional reform for conflict resolution. Norway entered the process as a facilitator. A cessation-of-hostilities agreement was signed by the Government of Sri Lanka and the LTTE and was followed by Norwegian-facilitated negotiations. A significant breakthrough was reached in the third round of negotiations held in Oslo in December 2002, when the LTTE, which remained committed to the Thimpu Principles and the government, which remained bound to maximum devolution within a unitary state, agreed to “explore a solution founded on the principle of internal self-determination in areas of historical habitation of the Tamil-speaking people, based on the federal structure, within a united Sri Lanka.”Footnote ²⁵

There also was agreement to initiate discussions on power sharing between the centre and the region, as well as within the centre and the region, and on human-rights protection. The Oslo Statement of December 2002 also stated that the solution should be acceptable to all communities. This was a clear reference to the island's third-largest group with a substantial presence in the east, the Muslims, who were apprehensive that a negotiated settlement between the two largest groups would be at the expense of their dignity and security. The significance of the Oslo Statement was that it clarified the ambiguity of the Thimpu Principles. The references to internal self-determination and a united Sri Lanka were crucial in allaying the consistent and perennial fear of the Sinhalese – that federalism was a steppingstone to secession. For the Government of Sri Lanka, responding positively to the federal idea and internal self-determination was not difficult given the groundwork of the constitutional-reform project of 1995–2000, in which the limitations of the unitary state and the Thirteenth Amendment had been widely discussed and debated in the public arena. Although the initiative to introduce a new constitution might have failed, the public awareness and education campaigns of both the government and civil society certainly had an impact. People were more sensitised to the challenges of constitutional design for a plural society, the need for a political solution to a problem that was essentially political in nature, and the inadequacy of a solely militarist response.

The promise of the Oslo agreement was not without problems, however. Soon afterwards, there were statements from the LTTE that raised doubts about its commitment to the Oslo formulation, and the negotiations became more difficult with both sides accusing the other of violating the cessation-of-hostilities agreement. The conduct of the LTTE in continuing to silence alternative voices within the Tamil community raised serious doubts about its commitment to pluralism, human rights, democracy, and power sharing – values that were inextricably linked to the federal idea. The negotiations broke down right before the parties were to discuss two matters: a human-rights memorandum of understanding (a major lacuna in the negotiations up to that point) and the “roadmap” to implement the Oslo agreement.

The negotiations resumed after concerted international pressure, and the LTTE insisted that the issue of the interim arrangements for the governance of the northeast be the focus before negotiations on a final political settlement based on the Oslo agreement continued. This resulted in the LTTE releasing a set of proposals for an Interim Self-Governing Authority (ISGA) in October 2003. The proposals were maximalist in nature and provoked an outcry among Sinhalese opponents of devolution of power as well as among the Muslims in the east. The most vociferous critics of the proposals were conservative Sinhalese lawyers who were deeply skeptical about the bona fides of the LTTE, committed to Sinhalese majoritarianism and the preservation of the unitary state, and therefore opposed in principle to the Oslo agreement.

The proposals, however, also were amenable to a more powerful critique from a constitutionalist and federalist perspective. The ISGA proposals were entirely about self-rule with no mention of shared rule. The proposals were extremely weak with respect to basic constitutional principles of the rule of law, separation of powers, and protection of the rights of minorities. Proposals that went far beyond powers exercised by regions in federal countries to give the ISGA control of and the right to control access to the marine and offshore resources of the adjacent seas were cited by critics of the negotiation process to raise doubts about both the bona fides of the LTTE and its commitment to a federal solution. It was ironic but not surprising that when the dominant Tamil nationalist group was given a chance at constitution-making produced proposals that would have resulted in a unitarist, centralised, majoritarian northeast with woefully inadequate human-rights and minority-rights protection mechanisms – this despite the fact that it had campaigned for years against the unitary, majoritarian constitutional framework of Sri Lanka. Another negative aspect of the ISGA proposals was that the text suggested that the LTTE was contemplating a two-nation confederal model, rather than a more conventional or even asymmetrical federal arrangement, designed to promote unity in diversity.

By the next presidential election in 2005, the perception that the LTTE had out-negotiated the Kumaratunga–Wickremesinghe cohabitation government, that they were not interested in a federal-type solution within a united Sri Lanka, and that they therefore could not be trusted was widespread. This mistrust combined with dislike for Wickremesinghe's liberal economic policies and provided candidate Mahinda Rajapakse a platform for an effective challenge in the election. Realising that outgoing President Kumaratunga's support for him was lukewarm and that she controlled their party's machinery, Rajapakse depended on two small Sinhalese nationalist parties, the Janatha Vimukthi Peramuna (JVP) and the Jathika Hela Urumaya (JHU), for organisational support. These parties insisted on various commitments in the election manifesto that were hawkish, unashamedly Sinhala nationalist in tone, and contained a promise to preserve the unitary status of the island's constitution. The final “nail in the coffin” was when the LTTE called for and forcibly implemented a boycott of the presidential elections in the north, thereby depriving Wickremesinghe of a significant number of Tamil votes. The fact that the LTTE helped to ensure the hawkish Rajapakse's victory in the election raised a host of new doubts about its commitment to a negotiated political settlement and a constitutional compromise based on federalism.

The election of President Rajapakse in 2005 effectively ended the ten-year effort to solve Sri Lanka's ethnic conflict through negotiation, constitutional reform, and compromise. In both initiatives, the first under President Kumaratunga and the second under the Kumaratunga–Wickremasinghe cohabitation government, the governments were willing to explore quasi-federal or federal-style solutions with strong safeguards to prevent secession and protect the territorial integrity of the country. Many would argue that this willingness was “too little, too late.” The fact that there was a powerful body of opinion within the Sinhalese community that strongly opposed both initiatives indicated that many possessed a majoritarian mindset that could not comprehend or have empathy for power sharing or a political arrangement that accepted the equality of groups, group rights, and recognition of their dignity.Footnote ²⁶ Concepts such as plurinational or multinational states, internal self-determination, and possibly even federalism remained for them academic or theoretical concepts that had little practical relevance.

The decade of exploring peace was followed by almost a decade of war. The LTTE was defeated militarily in 2009, at an enormous price in terms of lives – both military and nonmilitary – human rights, and interethnic goodwill and coexistence. The militarisation of the north and east continues today. There is little if any discussion about addressing the underlying causes of the conflict, which many in power will argue do not exist. They argue that the cause of the conflict was terrorism and that it is legitimate for a majority community to be privileged.Footnote ²⁷ In fact, powerful elements within the government want to dilute the powers of the ineffectual provincial councils established under the Thirteenth Amendment, the inadequacy of which was the basic assumption of the constitutional-reform initiatives of the decade of peace.Footnote ²⁸ Today in Sri Lanka there is no discussion about federalism.

Nepal

The debate on the merits and demerits of the federal idea, and its relevance for a country of diverse minorities like Nepal, intensified as the participatory constitution-making process commenced in 2008. It is widely believed that a main reason for the failure of the Constituent Assembly to adopt a final constitution – despite several extensions of its original deadline of two years – was the inability to reach consensus on the details of the federal system. The CPA committed the parties to ending the unitary, centralised state and to progressive restructuring: amendments to the Interim Constitution in different provisions declared Nepal a federal, democratic republic; stated that Nepal is an indivisible, secular, inclusive, federal, democratic republic; and also stated that Nepal shall be a federal, democratic republic. The reality, of course, is that notwithstanding these provisions in the Interim Constitution, between 2008 and 2014, Nepal was not federal because it did not have even a second tier of government; however, the provisions are aspirational in character and meant to ensure that the final constitution is indeed federal. Today in Nepal, as a second Constituent Assembly attempts to complete the constitution-making process, there is still a consensus that Nepal should be a federal republic; however, there is no consensus on the design of the federal system.Footnote ²⁹

The CPA and the post-1990 political developments clearly demonstrated the need for political and constitutional reform that would end the dominance of the Kathmandu-based, high-caste elite and empower the diverse minority ethnic groups in Nepal. Federalism and state restructuring thus were inextricably linked with not merely devolution of power but also with the empowerment of excluded and marginalised groups. In subsequent years, the federalism debate in Nepal was complicated by the linkages with inclusion and empowerment. Even with respect to the vexed questions of the name, number, and boundaries of the proposed provinces, calculations were made regarding what influence the dominant Bahun–Chhettri group would have in the proposed provinces. Generally, the smaller the number of provinces that were proposed, the greater was the likelihood that the elite would remain influential. The proponents of a ten- or fourteen-province federal model argued that under such a framework, the dominance of the elite would be significantly reduced.Footnote ³⁰

The debate on federalism became more complicated after the Government of Nepal ratified International Labour Organization Convention 169, a convention on the rights of indigenous peoples, in 2006. Many of Nepal's ethnic minorities asserted indigenous status and made claims based on the convention.Footnote ³¹ Some of the dominant groups also claimed to be indigenous, because the definition of the term indigenous in the law focused on language, traditional customs, distinct cultural identity, social structure, and history. Some of the proposals on federalism that emerged from this rather confusing situation combined ethnicity as the main basis for demarcation of a province, preferential rights for the largest group in the province, and included references to terminology used in ILO 169 and related international documents. This mix of concepts and claims confused the debate on federalism. Because these proposals included a strong emphasis on ethnicity, they often were described as proposals for ethnic federalism and evoked fear and opposition among many Nepalis. Many Nepalis, particularly from the elite and smaller ethnic groups, stated that they were willing to accept federalism but not ethnic federalism.

The first Constituent Assembly assigned the task of developing the details of Nepal's federal structure to one of the thematic committees appointed to facilitate its work, the Committee on Restructuring of the State and the Distribution of State Powers (CRSDSP). In its draft report published in 2010, the CRSDSP highlighted several reasons for the federalisation of Nepal, including recognition of its sociocultural diversity including ethnicity, language, and culture; the need to enhance representation, participation, and ownership of as well as affinity for and proximity to the state; enhancement of the democratisation of society and the search for equal and inclusive democracy free from discrimination; and the provision for more equitable distribution of services, facilities, and development.Footnote ³²

The CRSDSP, by a majority decision, proposed that the basis for the federalisation process should be twofold: identity and capability. Identity was defined to include ethnicity, community, language, culture, and historical habitation in certain regions. Capability was defined to include economic capacity and interrelationships, natural resources, infrastructure development, and administrative convenience. The CRSDSP report proposed that Nepal should be divided into fourteen provinces and it specified their names and boundaries. The report proposed a three-tiered governmental structure – federal, provincial, and local – but envisaged “special structures within a province to deal with concentrations of populations from particular ethnic groups that may require autonomous arrangements within a province.” The report proposed that the provinces should have the right to self-determination, provided such right did not undermine the sovereignty, territorial integrity, and unity of the country. The minority dissenting report proposed a six-province federal model.

The majority report was criticised on the grounds that it gave too much weight to identity at the expense of capability. Two of the proposed provinces would have only 0.2 and 0.4 percent of the national population and would find it difficult to be viable in economic terms. The publication of the report generated a debate on the basic issue dividing Nepal and the political parties in the Constituent Assembly: that is, the balance to be struck between identity and viability in designing the federal structure. Proponents of giving identity greater weight invariably propose a larger number of provincial units, whereas those who focus more on viability tend to propose a smaller number. Supporters of recognising identity as the preeminent factor tend to dismiss the focus on viability, declaring that Nepal itself may not be a viable entity if assessed by the criteria used by critics and that intergovernmental transfers and equalisation payments are inevitable in Third-World federations. Supporters of viability argue that it would be irrational to have large disparities in size, population, and resources among provinces and that “weak” provinces that are dependent on the centre would have little effective political power and autonomy, thereby undermining the federal scheme and its rationale.

Given the deadlock on the issue of federalism, the main political parties agreed to form a High Level State Restructuring Commission (HLSRC) in November 2011. The nine members or experts were nominated by these parties and tended to reflect their respective nominating party's views. They were unable to reach a consensus. In this case, as well, two separate reports – a six-member majority report and a three-member minority report – were submitted to the government. The majority report proposed eleven provinces: ten on the basis of territory and a nonterritorial entity for the Dalits. The minority report proposed six provinces. Whereas some of the smaller provinces envisaged by the CRSDSP were dropped, some aspects of the majority report – including the delineation of boundaries to promote greater ethnic homogeneity at the provincial level – suggested that the majority gave even greater significance to identity. The proposal of a nonterritorial entity to address the concerns of the Dalit community also highlighted the recognition of identity by the majority members of the HLSRC. The Dalit community, which is dispersed throughout the country, was divided on whether such an entity would be effective in addressing the aspirations of their community. This issue is one example of how in the context of Nepal's exceptional range of diversity, there is a close interrelationship between federalism and inclusion. For some groups, principles or mechanisms that would fall under the category of federalism might be advantageous, whereas for other groups, principles or mechanisms that would more accurately fall under the category of inclusion might be preferable.

Confusion in terminology in the debate on federalism in Nepal was related to the dispute about the naming of the provinces. Advocates of using the name of the largest ethnic group in the proposed province were referred to as proponents of single-identity provinces. Critics argued that because all provinces were plural and unlikely to have the largest group constituting a majority, they should have multiple identities in their nomenclature. A third group sought to seek a compromise on the issue by proposing a mixed identity for the naming of provinces – a “double-barrel” name with one focusing on the ethnicity of the largest group in the province and the other name using a significant geographical term to represent other groups in the province. Single-identity provinces were considered part of ethnic federalism, whereas multiple- or mixed-identity–based provinces were part of a more reasonable or moderate form of federalism. Thus, the issue of the “name, number, and boundaries” of the provinces emerged as the most contentious issue in the federalism debate. In the frantic attempt to reach consensus in the weeks leading up to May 27, 2012 – the final, extended deadline for the Constituent Assembly to adopt a constitution – the pragmatic suggestion that the naming of the provinces be left to the provincial legislatures was welcomed as a compromise solution.

The debate on federalism in Nepal therefore has been more detailed than the debate in Sri Lanka. This is perhaps because there has been a broader consensus on the need for federalisation in Nepal than in Sri Lanka. The debate has been confused, however, by concepts and terminology that often are contested or capable of different interpretations. For example, a common response of many Nepalis when questioned about federalism was that they were open to federalism but not ethnic federalism. However, most Nepalis had very different understandings about what was meant by the concept of “ethnic federalism.” To some, it was a federalism that focused on ethnicity as one of the bases for the demarcation of provinces; to others, it was where ethnicity was the dominant aspect – not only with respect to the demarcation of boundaries but also where the dominant ethnic group in a province was assured the position of provincial premier, the preferential use of land, and other features that privileged ethnicity. There were differences of emphasis and nuance between federalism that recognised ethnicity as one criterion in the demarcation of provinces and federalism that privileged ethnicity in various ways, in addition to it being the main criterion in establishing provinces. It is the latter form of strong ethnic federalism that seems to be unacceptable to a large body of opinion within Nepal.

The dissolution of the Constituent Assembly on May 27, 2012, created a crisis of constitutionalism in Nepal.Footnote ³³ The Interim Constitution of 2007 did not envisage such a scenario, assuming that the Constituent Assembly would draft and adopt a new constitution, the promulgation of which would see the repeal of the interim document. The Interim Constitution could not be amended to respond to the new political reality because the Legislature Parliament, which had the sole power to amend the constitution, had ceased to exist.Footnote ³⁴ The collapse of the process also generated apprehension and suspicion among those for whom the constitution was important for the introduction of radical change; the fear was that the conservative elite that felt threatened by change had orchestrated the collapse. The major political forces negotiated a political agreement in early 2013 that the way forward was to conduct fresh elections to a second Constituent Assembly and that some residual powers vested in the president be used to “amend” the constitution to facilitate such a process. The elections that were held on November 19, 2013, were widely recognised as free and fair despite attempts by a breakaway faction of the Maoists to sabotage them. The election resulted in a shift from the Maoists, which was the largest party in the 2008 elections, to a shift in favour of the Nepali Congress Party (which obtained the largest number of seats) and the United Marxist Leninist Party (UML) – both of which are seen as reluctant federalists.

The post-election period witnessed a debate about what the election results meant on constitutional issues. Whereas conservative commentators were quick to interpret the decline in support of the Maoists as a vote against ethnic federalism (however defined), others suggested that the decline in support resulted from other political factors, such as the party seen as the party of change losing some of that lustre as it was co-opted into the political mainstream and became distant and alienated from its mass base. Another weakness in the argument that the results were a vote against federalism is the fact that both the Nepali Congress and the UML in their manifestos had commitments to a three-tiered federal model, which – to varying degrees – accepted identity and capability as comprising the basis for the demarcation of the provinces. The manifestos affirmed the plural character of the provinces, thereby requiring a multiple-identity approach to the controversy on naming the provinces but accepting that a provincial institution should make the final decision on the matter. Both parties also highlighted the importance of building on the agreements made under the first Constituent Assembly, recognising the importance of moving the constitution-making process forward without revisiting some of the difficult issues that the first Constituent Assembly grappled with for several years.

Nepal therefore needs to resume and complete its constitution-making process as a matter of priority. The process that it deliberately embarked on was participatory and inclusive. Such processes are more complicated and challenging than more closed processes led by commissions of experts. An inclusive participatory process also generates greater expectations, particularly on the part of excluded groups that have struggled for years for dignity and recognition. Adopting a constitution that fails to respond to such expectations of change – commitments made in the CPA or entrenched through amendments to the Interim Constitution – will be seen as a betrayal by a large segment of the population. For better or for worse, the change agenda – important for a majority of the population – is through the adoption of a federal, secular, inclusive, democratic republic. The fact that Nepal – which was a unitary, Hindu monarchy until 2006 – has reached consensus on these broad constitutional features is – when viewed in comparative perspective – a remarkable achievement. The challenges are in the details and in ensuring a culture of accommodation on the part of the elite that is apprehensive of change and that has legitimate concerns about a strong ethnic federalism. This also includes the minority ethnic groups using the discourse of self-determination and single-identity–based, strong ethnic federalism. Many of these groups remain apprehensive that the commitments of the CPA and the achievements of the first Constituent Assembly will be repudiated or undermined by well-entrenched conservative elites that is determined not to change. Unlike in Sri Lanka, therefore, the debate on federalism continues and has proceeded much farther and in greater detail.

Constitutional Theory in a Context of Constitutional Stasis

The constitutional reform debate in Sri Lanka that spans a century to the early 1900s has three striking features.Footnote ⁴ The first feature is the richness of intellectual imagination in ideas for institutional reform that have been proposed for enduring problems of constitutional democracy and ethno-cultural pluralism.Footnote ⁵ The second is the failure of those ideas to find traction and implementation in the realm of the political, whereby ethnicised majoritarianism, partisan political interests of the government in power, and executive convenience have either dominated or hijacked every major attempt at constitutional change, especially in the post-independence era.Footnote ⁶ The third feature is the near complete absence of theory in this debate. Reformist arguments for the institutional restructuring of the state in appreciation of societal pluralism or in furtherance of constitutional democracy have been richly informed by constitutional comparativism.Footnote ⁷ They also no doubt have been influenced by various analytical and normative perspectives, including those provided by liberalism and liberal constitutionalism,Footnote ⁸ socialism, republicanism, and even ethno-symbolic explorations of the precolonial past.Footnote ⁹ Yet, in the main, these institutionalist responses have not been adequately theorised and contextualised to the empirical conditions of the Sri Lankan case, and they have followed the dictum articulated by Neelan Tiruchelvam that ‘The quest for a political resolution [of the ethno-national conflict] within a united Sri Lanka must…relate to the substantive issues relating to the exercise of political power rather than more abstract formulations of political identity.’Footnote ¹⁰

This absence of theory has had several consequences, especially for liberal constitutionalism. The lack of descriptive theory has meant that the fullest implications of key political dynamics and empirical factors, including sociological and historical implications of phenomena such as nationalism and ethnicity, have not been considered in proposing institutional solutions.Footnote ¹¹ Liberals, therefore, have consistently underestimated the power of the past and, by extension, the deep legitimacy of majoritarian ethnic nationalism – however repugnant to the liberal values of choice, tolerance, and pluralism that nationalism might be. Their general critique of ethnic nationalism on civic-rationalist grounds also has been seen as a selective critique of Sinhala-Buddhist nationalism because the liberal commitment to minority accommodation favours the Tamil claim to territorial autonomy, even though Tamil nationalism is as ethnic-communal and should be as unacceptable to liberal values as its counterpart. Flowing from this, the failure to contextualise the normative principles underpinning institutional-reform proposals, by means of historical, sociological, or other theoretical arguments, has entailed their easy rejection by nativist nationalists on the grounds that especially liberal democratic norms are ethnocentric Western values – which have as much place in contemporary Sri Lanka as colonial rule.

Conversely, the mechanical comparativism that has characterised institutional reformism so far, together with its undiscriminating adoption of norms developed on the basis of radically different empirical conditions elsewhere (perhaps because they are norms that have a dominant following internationally), also has meant that the Sri Lankan debates added little to comparative constitutional law or global constitutionalism. In particular, the distinctive features of the Sri Lankan case as a multinational polity require the reconceptualisation of several key assumptions and organisational principles of the modern ‘nation-state’ at a deeper and more general level than mere reform of the state's institutional framework.

The Structure and Scope of the Argument

These critical considerations constitute the entry point of this chapter into the Sri Lankan constitutional-reform debate. Only a small selection of the issues raised by these questions can be addressed here. The first substantive concern is the issue of ‘national pluralism’ – that is, the existence of more than one group claiming to be a ‘nation’ within the territorial and historical space of the state, as well as the issues for constitutional law and theory that arise in the structural accommodation of this distinctive type of polity. Second, this is an intervention into the debate among those who subscribe broadly to the principles of democratic constitutionalism; it is not directed at those who reject those values on nationalistic or other grounds of cultural authenticity. Accordingly, my theoretical critiques also are directed at the conceptual assumptions of reformist political opinion in Sri Lanka rather than those who reject reform outright. Third, it is a contribution of applied constitutional theory and comparative constitutionalism for Sri Lanka, using the critical theses, principles, and structures associated with the model known as the ‘plurinational state’. The primary purpose is to encourage theoretical debate and clarity so that choices about institutional forms are analytically contextualised and normatively focussed (contra the Tiruchelvam dictum); it is not concerned with the political viability of plurinational ideas in the current political context.

The substantive and methodological issues involved in introducing a new model of constitutionalism to any given political context are complex and multifarious, and all of these matters cannot be addressed here. Accordingly, the chapter does not deal with the following relevant issues. The plurinational logic of constitutional accommodation seems fundamentally incongruent with the centralised state tradition that Sri Lanka has had since the instantiation of the modern state through British colonialism in the nineteenth century. However, in fact, it can be strongly historically contextualised in Sri Lanka by reference to the highly devolutionary and asymmetrical precolonial ‘galactic’ form of the state.Footnote ¹² Likewise, plurinational constitutionalism so far has been concerned with civic–societal models of nationalism, whereas in Sri Lanka, we must contend with ethnic forms of the nation and nationalism. The challenge of the civic–ethnic dichotomy also can be overcome, both theoretically and empirically. The Western plurinational state is firmly a liberal democratic model of state, whereas in Sri Lanka, there is an established procedural democracy – the substantive values of which are derived not from political liberalism but rather from other cultural values, in particular, ethnic nationalism. The plurinational state in these circumstances must and can be theorised by reference to a broader conception of democracy than political liberalism. All of these issues require serious attention and much theoretical work, which must be pursued elsewhere.

Similarly, this chapter is devoted to the underlying theoretical issues arising from national pluralism for constitutional law, and this focus has meant that I am unable to fully canvass the institutional options through which a plurinational settlement might be actualised. I seek forbearance for this choice, partly because, as observed previously, the Sri Lankan constitutional-reform debate is saturated with discussions about institutional form but is striking for its poverty of theory. This is not a mere scholarly complaint. As previously shown, the absence of theory – by failing to adequately contextualise institutional reform within political, cultural, and historical realities – has had the regrettable practical consequence of contributing to the failure of liberal reformism in Sri Lanka.

Before discussing the merits of plurinational constitutionalism and its relevance to Sri Lanka, an elucidation of the main positions of ideology and ethno-political interest in the postwar constitutional-reform debate is necessary. The second section delineates four major positions according to a typology based on the traditional unitary–federal dichotomy (which, for the first time in nearly four decades, following the military defeat of the Tamil Tigers, does not include an explicitly secessionist position). The two dominant positions with regard to the constitutional form of the state on the unitary–federal axis are associated with the two major ethnic-nationalisms; the unitary state and federalist positions can be appreciated only through an understanding of the ethno-cultural, historical, and territorial claims asserted by these two nationalisms. Both the unitary and federal perspectives also are supported in Sri Lankan constitutional discourse on grounds other than ethno-nationalism; however, because they are politically less dominant, they are included in the typology as ‘variants’ to the dominant positions. Nonetheless, the conceptual assumptions of the variant positions, representing different traditions of democratic modernity, are the main critical concerns of this chapter.

The third section argues that all of the positions with regard to constitutional order reflected in this typology – and especially the two variant positions, however conceptually opposed to one another they may seem – are united by a common subscription to the type of normative order associated with the traditional Westphalian nation-state. Furthermore, this model of internal constitutional organisation is inadequate to the task of addressing the type of pluralism that characterises Sri Lanka's polity. In the context of the historic failure of postcolonial nation-state building – that is, the failure to build a modern ‘Sri Lankan’ nation-state that transcends traditional ethnic identities – the main question of this part of the argument is the utility of continued subscription to this model. Other models of state have emerged that may provide better answers in terms of both norms and structures to the central constitutional problem of Sri Lanka.

Based on this critique, the fourth section outlines an alternative approach drawing on the body of constitutional theory known as ‘plurinational constitutionalism’, which has relevance and use to the Sri Lankan case in three ways. First, it offers a new analytical understanding of the sociological nature of the Sri Lankan polity as one characterised by national pluralism rather than one of mere ethno-cultural pluralism. Second, it articulates the normative propositions that must inform constitutional approaches to national pluralism, in particular to expose the conceptual inadequacy of the monistic conception of the modern nation-state and the need to disaggregate the ‘nation’ from the ‘state’. Third, plurinational constitutionalism allows us to outline the principles of structural organisation, including the principles of autonomy, recognition, representation, and reciprocity, that are necessary to the constitutional accommodation of national pluralism according to a specifically plurinational logic within a united state. The discussion closes with concluding remarks in the last section.

The Unitary State Perspective

The two approaches to constitutional form discussed in this section – Sinhala-Buddhist nationalism and the Jacobin Variant – both support the maintenance of Sri Lanka as a unitary state, albeit on the basis of quite different analytical and normative justifications.

The Federal State Perspective

The two approaches represented by Tamil nationalism and the Liberal Variant both argue for the federalisation of the Sri Lankan state but on different grounds. The former demands federal autonomy as an ethnic claim to self-government; the latter advocates federalism as a general check on majoritarianism.

The Monistic-Demos Thesis and Host-State Societal Dominance

A key ground of normative critique presented by plurinational constitutionalism against the Westphalian nation-state model, and political liberalism's attachment to it, concerns the ‘monistic-demos thesis’ – that is, the notion that operates as both postulate and presumption that the nation is synonymous with the state.Footnote ⁵⁵ In other words, substate nations, by their very existence, challenge the monistic presupposition of traditional liberalism (or ‘Liberalism I’ in Taylor's term) that there is or can be only one demos within the state.Footnote ⁵⁶ A related contention is that regarding ‘host-state societal dominance’ – that is, that the conceptualisation of the nation in both unitary terms and in exclusive association with the state not only prevents the fullest constitutional recognition of national pluralism but also serves to privilege, in effect, a majority or otherwise dominant cultural identity to the disadvantage of minority nations – and in violation of fundamental principles that traditional liberalism claims to defend. Implicit in this challenge is the question of whether liberalism as an ideology has the theoretical and normative capacity to respond to the realities of national pluralism. As discussed herein, the monistic-demos thesis is the central conceptual basis on which the Jacobin and Liberal positions approach constitutional reform in Sri Lanka; consequently, there is a tendency in both of these positions to give insufficient regard to the potential for cultural dominance that is inherent in the nation-state model itself, regardless of the ethnocratic character of the Sri Lankan state. This also is reflected in the tendency to perceive ethno-nationalism as an exclusively or mainly substate problem.

The plurinational critique of traditional liberalism points to the imperviousness with which the existence of plural demoi within the state and their attendant claims has been treated, rendering substate nations ‘voiceless and faceless’.Footnote ⁵⁷ As Requejo noted in relation to the work of liberal scholars such as Rawls and Habermas, national pluralism ‘is a question that is not so much badly resolved as completely unaddressed by the premises, concepts and normative questions of these theorists’.Footnote ⁵⁸ Tierney goes further in pointing out a more insidious consequence of this empirical and, consequently, theoretical deficiency in traditional liberalism:

The assumption that the collective identificatory function of nationhood rests only with the statal nation denies the possibility of multiple conceptions of national identity that are commonly held by citizens of plurinational polities. By the failure to acknowledge this important dimension of individual identity and autonomy, traditional liberalism denies to individual members of substate nations – for whom the substate societal space is an important means of political self-expression – such cardinal commitments of political liberalism as choice, equality, and justice. Conversely, the pretence that the state national society is a culturally neutral entity held together by purely normative values (sometimes accompanied by the disparaging implication that substate nations are not similarly modern, progressive, and inclusive entitiesFootnote ⁶⁰) hides the reality that the statal identity more often than not is associated with a dominant societal or cultural influence within the plurinational polity. The failure to apprehend the homogenising consequences of this approach, it is contended by plurinationalist critics, signals a failure on the part of traditional liberal theory to fulfil fundamental liberal precepts.Footnote ⁶¹ Moreover, the tendency among some traditional liberal theorists to treat the reemergence of substate nationalism as premodern ethnic particularism – and therefore as something to be discouraged – fails to recognise the nationalist dynamics that animate the state itself. The elision of state and nation in the theory of traditional liberalism, as much as in everyday political parlance, conceals the fact that statist discourse also is a form of nationalism – albeit one that seeks to undermine the validity of substate nationalisms at the same time as refusing to acknowledge the intensely nationalist nature of its own discourse and praxis.Footnote ⁶²

Since the early 1990s, these conceptual inadequacies of traditional liberalism have given rise to a new school of liberal political philosophy that seeks to reposit the normative values of liberalism in ways that take proper account of both nationalism and national pluralism. According to Norman, this theoretical work addressing the interstice between nationalism and traditional liberalism came in two waves. The first attempted to prove ‘that it was not impossible to be a liberal and a nationalist at the same time’ and the second used the insights produced by that work in proceeding to critique traditional liberalism and its conception of the state according to the monistic-demos thesis.Footnote ⁶³ The ‘liberal nationalist’ theoretical accounts developed in this second wave, in particular, constitute the body of political philosophy that Taylor termed ‘Liberalism II’, associated prominently with the work of Taylor himself and Kymlicka, among others.Footnote ⁶⁴

In contrast to the mixture of complacency and hostility that marks traditional liberal accounts, the methodology of liberal nationalist theorists places emphasis on the sociological reality of nationalism in and as a form of ‘normal politics’.Footnote ⁶⁵ They recognise that substate nations are deliberative spaces for the conduct of politics, which play an essential intermediary role in the relationship between the citizen and the state. The substate nation rather than the statal national society often is the foremost vehicle of identity for the individual member of the substate nation. Accordingly, ‘the value which he finds in the democratic process can be more fully explained by appreciating these ties, and by understanding the preferences felt by this citizen for the location of his right of individual self-determination within the broader condition of collective self-determination for his primary demos’.Footnote ⁶⁶ The most important analytical proposition of liberal nationalist theory, therefore, is that the meaningful realisation of liberal democratic commitments in respect of citizenship in plurinational polities requires the empirical reality of national pluralism to be fully acknowledged and accommodated constitutionally.

With regard to the Sri Lankan case, there are three points to note in these respects. In practical terms, it would seem as if the issue of the host state's cultural dominance needs no further stress, given the ethnocratic character of the state that makes the ethnic dominance of the Sinhala-Buddhist nationalism over it transparent. Nonetheless, at a more abstract level, we need to emphasise the absence of cultural neutrality in the nation-state model itself, given the way in which both Jacobins and Liberals deploy it as a heuristic exemplar. This suggests that the standard model of modern nation-state is inadequate and inappropriate as a state form in multinational polities, and that no amount of federalist or devolutionary institutional reform within this paradigm meets the deeper challenge of national pluralism.

Related to this, constitutional theory and practice must treat ethnicity seriously as well as positively – and not as some unpalatable primordial remnant that hopefully will disappear with the march of time and progress. This is not so much an exercise in making a virtue out of necessity as a realistic appraisal of the ‘societal context’Footnote ⁶⁷ of constitutional law in Sri Lanka, in which the resilience of ethnicity as the primary referent of identity materialises from the total collapse of modernist assumptions with regard to nation-building in the six decades since independence. In other words, there seems to be little point in the dogmatic adherence to modernist shibboleths about the nation in the face of the ethnic reality. At the theoretical level, the plurinational critique has shown the conceptually problematic nature of the modernist nation-state in the context of national pluralism to be both deeper and more general than the specificities of the Sri Lankan case.Footnote ⁶⁸

Finally, it is important to recall that some exponents of the Liberal view have grappled with the issues of democracy and multiple identities in the context of balancing civic and ethnic conceptions of the nation within the constitutional structure of the Sri Lankan state. The importance of the civic-statal demos was underscored by Edrisinha when he observed that there is a vital need, ‘given the dominance of ethno-nationalism in the past three decades to forge a supra-ethnic authentic Sri Lankan national identity’.Footnote ⁶⁹ His argument continues:

This can be readily endorsed, subject to the critical caveat that in our scheme, both state and substate are national spaces rather than the hierarchy suggested by Edrisinha's formulation of ‘national and regional’. Although we should reject the monistic-demos thesis implicit in this formulation and some of the more simplistic descriptive conclusions with regard to the ethnos to which other traditional liberals are led as a result of applying the civic–ethnic classification to nationalisms, the democratic normative vision of the demos that they endorse cannot be dismissed. The argument rather is that democratic consent as well as cultural inheritance define the nationFootnote ⁷¹ or, more precisely in the plurinational context, the nations. How the divide between civic and ethnic conceptions of the nation might be bridged is beyond the scope of this discussion. However, it is possible to do so theoretically by challenging the civic–ethnic dichotomy itselfFootnote ⁷² and empirically by drawing a distinction between nations (as organisational cultures) and nationalisms (as ideological movements),Footnote ⁷³ which serves to differentiate the historical and sociological nature of ethnic nations from the ideological claims of contemporary nationalists. This differentiation enables us to reveal the pluralistic potential of ethnic nations in history, against the intolerance of contemporary ethno-nationalist ideology. This in turn allows us to recast ethnic nations within a plurinational framework that promotes multiple nationality allegiances to the substate and state levels, while acknowledging that the substate nation is the ‘primary demos’ – in our case, the ‘primary ethnos’ – through which the individual citizen's relationship with the plurinational host state is mediated.Footnote ⁷⁴

The Disaggregation of Nation and State

From the preceding discussion, it is clear that a polity defined by national pluralism is not only in a distinctive politico-sociological category but also that it calls for a fundamental normative reconceptualisation of the Westphalian conception of the sovereign nation-state. From the perspective of constitutional law and theory, an essential feature of the plurinational challenge may be summarised in Tierney's words:

Although not theoretically elaborated, a similar observation was made from within the Liberal view in the Sri Lankan constitutional-reform debate. Relying on the work of Lapidoth and Buchanan, Bastiampillai noted:

From the analytical characterisation of the Sri Lankan polity as one of national pluralism that was established previously, this proposition with regard to the nation-state must be extended to the Sri Lankan case, which then reframes the entire constitutional-reform debate about structures as well as norms. As indicated by Bastiampillai, in a plurinational context, sovereignty is one of the major norms associated with the nation-state that would require a new form and explanation going far beyond the monist and positivist understanding of the doctrine (which Jacobins, especially, treat as a “sacred cow”). Although this chapter is not the place to address this issue in detail, the ‘relational’ concept of sovereignty is especially useful here. The essence of this approach is set out by Croce: ‘In the relationship between the ruler and the ruled, sovereignty belongs to neither but to the relationship itself’.Footnote ⁷⁷ This implies, on the one hand, that sovereignty is essentially a political relationship between citizens and state, and on the other hand, the importance of the correspondence between that political relationship and the legal structures of the state. Plurinational constitutionalists extend this relational conceptualisation of sovereignty to the plurality of peoples or nations within the plurinational state.Footnote ⁷⁸

Whereas the scholarly work in relation to the Western plurinational state demonstrates that the separation of nation and state is a theoretically viable epistemological shift, and thus liberates us from the constraints of Westphalian orthodoxies, the comparative application of the plurinational-state model to the empirically distinct Sri Lankan case requires, in some respects, a different approach to the normative consequences that are intended to flow from the disaggregation of nation and state. Central to this proposition in Western plurinational constitutionalism is the historical reconceptualisation of the plurinational polity as a ‘union state’Footnote ⁷⁹ or a ‘coming-together-federation’Footnote ⁸⁰ on the basis of which further claims are made for the recognition, representation, and autonomy of substate nations constituting the union on a footing of equality.

The same methodological concern with the history and historiography of state-formation takes us in Sri Lanka into the precolonial era and the dominant state form that prevailed then. There is much that is promising in the cosmo-topographical model, theorised in historical anthropology as the ‘galactic polity’ or the ‘mandala-state’, in helping to historically contextualise constitutional responses to national pluralism in the present. This historical exploration also takes us into the realm of constitutional metaphysics and to an ontology of the state that is fundamentally different from that of the West. The contractarian ideal at the heart of the union state is anchored in Enlightenment rationalism, whereas the pluralist potential of the precolonial state form derives from the principles of Indic cosmology. Although this provided for an extraordinarily heterogeneous conception of society, and devolutionary, asymmetrical, and ‘pulsating’ administrative practices, this political order also was governed by principles of hierarchy and encompassment that maintained cohesion. In contemporary terms, this appears to demand an emphasis on asymmetry as opposed to equality in the constitutional treatment of substate nations.

The disaggregation of nation and state also raises the issue about the form and content of the statal ‘Sri Lankan’ national identity within a putative plurinational dispensation. A ‘thin’ conception of the state-nation would render it a minimalist juristic identity, virtually devoid of any ‘national’ content, with the substate level assuming primacy over the identificatory and functional roles of nationhood. Although a hypothetical possibility, the political and historical implausibility of this model suggests that it would be a misleading line of constitutional enquiry, not least because of its formal likeness with the confederal postures associated with the erstwhile secessionist movement.Footnote ⁸¹ In the Western context, illustrated in the debates over the Scottish constitutional referendum, the liberal-democratic host state is committed in policy and principle to respect the democratic wishes of the substate nation, either to effect a separation or to fundamentally renegotiate the terms of the union. Needless to say, such options do not form part of the empirical context on which constitutional theory of any solidity can be built in the Sri Lankan case; arguments for the reform of the unitary state are strengthened if they are based on a clear a priori commitment to the unity of the state. To this extent, the theoretical limits of substate autonomy are predetermined in the Sri Lankan case in a way that they are not in the Western contexts. This then suggests that we need a more substantial, ‘thick’ conception of Sri Lankan identity to underpin or overarch the radical pluralisation and devolution involved in a plurinational constitution.

More specifically, due to the failure of postcolonial nation-building, the sharply defined divisions determined by ethnic nationalisms, the injustice of the ethnocratic state, the substate mistrust of central institutions, and the history of violent ethno-national conflict, the radical autonomy of a plurinational constitution in the Sri Lankan case well might lead to ethnic division and disintegration unless it is counterbalanced by stronger guarantees for the integrity of the statewide constitutional order than the weak incentives contemplated by liberal plurinationalism. Such an eventuality would constitute a resounding failure of the thesis advanced in this chapter that both unity and autonomy can be secured by a plurinational constitution in Sri Lanka. At the same time, it is important to marshal those residual attachments to a united Sri Lanka and Sri Lankan identity that have survived protracted ethnic antagonism and conflict. As even R. Sampanthan, the most senior Tamil nationalist politician, stated recently:

A fresh approach therefore is needed and, above all, as Edrisinha emphasised, this involves a critical need to inculcate an inclusive, shared, overarching national society at the level of the state that balances the constitutionalisation of substate national identity and autonomy.Footnote ⁸³ As Kearney pointed out in the UK case, ‘there has been a “British” history over and above our “multi-national history” and therefore the central question of national and institutional pluralism is not so much “four nations or one” as “four nations and one”’.Footnote ⁸⁴ Transposed to the Sri Lankan case, this approach can be articulated as recognising an overarching Sri Lankan national identity at the state level as well as the Sinhala and Tamil nations at the substate level, together with the non-national diversity represented by the smaller ethnic and religious minorities. Moreover, such a statal nation – which is substantially to be conceived in civic-societal and constitutional terms rather than ethnic-communal and ascriptive terms (although shared myths and memories also may have a place) – is a valuable opportunity to incorporate an element of civic-constitutionalist values into a plurinational system that is otherwise primarily concerned with ethnic forms of the nation. In this regard, the concept of the ‘state-nation’ has major relevance as a model for a plurinational Sri Lankan statal nation.Footnote ⁸⁵ Although permitting plurinational-type substate autonomy, it provides for a robust yet noncoercive framework for the preservation of the unity of the whole. It does so by providing an incentive for the modernisation of the statal nation, not by a resumption of teleological, monistic nation-buildingFootnote ⁸⁶ but rather according to a specific politico-institutional logic that is meant to implant (or ‘craft’) a pattern of multiple but complementary collective identities across the plurinational polity.

Guiding Principles in a Plurinational Constitution

Building on the preceding theoretical discussion, we are now in a position to enunciate more prescriptively the contours of the major principles that should inform the design of a plurinational constitutional system. I rely here substantially on Tierney's work.Footnote ⁸⁷ The principles are set down in deliberately broad and general terms, seeking to articulate the normative core of a plurinational system but without overly constraining design options within that framework.

First is the principle of self-determination or autonomy. This principle involves the right of each substate nation within the plurinational state to determine its own political and constitutional future and self-development. In liberal theory, this may or may not involve a right of secession,Footnote ⁸⁸ but the principle is more concerned with ensuring the national status of a substate nation to be reflected in extensive autonomy in the constitutional arrangements of the plurinational state. Tierney observes that this principle accords with MacCormick's notion of ‘self-rule’ in which ‘the members of a nation are as such in principle entitled to effective organs of political self-government within the world order of sovereign or post-sovereign states; but these need not provide for self-government in the form of a sovereign state’.Footnote ⁸⁹ The principle of self-determination in the sense used here is more a tenet of political morality than the rule in international law; therefore, it applies not only to a substate nation seeking specific legal recognition but also to the host state to respect the aspirations to recognition of such groups.

Second, the principle of representation again recognises that the ‘primary agenda’ of substate entities is not secession but rather full and fair representation in the constitutional and political processes of the host state. This involves power sharing and representation in central government, particularly the legislature and the judiciary and also, critically, in procedures for constitutional change. This principle therefore seeks to reflect the interdependent and cooperative nature of the plurinational state, which not only recognises national pluralism in the self-rule dimension but also the shared-rule dimension in the governance of the whole.

Third, the principle of recognition has both a symbolic and a practical character. Recognition of the plurinational character of the state in symbolic commitments and institutions has a practical effect in that they ‘set the tone for the way in which the constitution is in general interpreted and applied, and will determine whether a vision of a plurality of nations, interacting equal to equal, is in fact a constitutional reality’.Footnote ⁹⁰

Fourth, the principle of reciprocity stresses the tenet of political morality that in plurinational states in which substate nations enjoy the kind of respect ensured by the first three principles, the latter in turn owe certain obligations to the host state. Tierney identified three such duties: to reciprocally recognise the ‘national status’ of other entities within the state, to work in ‘good faith to consolidate the state as a common polity’,Footnote ⁹¹ and to respect the rights and interests of all citizens arising out of the common citizenship of the state as a whole. There is an important caveat attached to this principle: ‘nationalist movements with an overtly “separatist” programme may not accept these responsibilities, but in turn they must accept that the normative force of any claims they might make of the host state in terms of representation and recognition may, accordingly, be substantially weakened'.Footnote ⁹²

Fifth, the principle of democracy relates to the situation of the individual within a substate nation, in which an overriding concern is to ensure that its individual members in the exercise of their democratic rights determine the collective political direction of the group and not the converse. This underscores an important point made previously, that the substate nation in the plurinational scheme is not only an ethno-cultural entity but also a territorial unit of government. It therefore must respect the democratically expressed wishes of resident citizens and not only the interests of members of the ethnic nation.

Representation in the Rajya Sabha

The distinction between coming-together and holding-together federalism provides both a descriptive account of the formation of federal polities and a set of normative principles about the sovereign autonomy of States and their relationship with central government. The representation of the States in the composition of the legislative branch of Union government is one device used to preserve States’ interests at the federal level and to protect their autonomy. The parliament in India is composed of two houses: the Council of States and the House of the People.Footnote ¹⁵ Article 80 of the constitution provides that the Council of States shall have not more than 238 ‘representatives of the States and of the Union Territories’.Footnote ¹⁶ Article 80(4) further provides that such representatives shall be ‘elected by the elected members of the Legislative Assembly of the State in accordance with the system of proportional representation by means of the single transferable vote’.Footnote ¹⁷ Article 84 specifies two minimum requirements for a member of the Council of States: Indian citizenship and a minimum age of thirty years.Footnote ¹⁸ The parliament is empowered to specify further qualifications of candidates for such an election through ordinary law,Footnote ¹⁹ and it is these specifications that effectively determine the extent of effective representation of the States and have been the focus of Court challenges.

The Representation of People Act of 1951, a law made under Article 84(c) of the constitution, originally provided in Section 3 that ‘A person shall not be qualified to be chosen as a representative of any State or Union territory in the Council of States unless he is an elector for a Parliamentary Constituency in that State or territory’.Footnote ²⁰ To be ‘an elector’ on the electoral rolls of a territorial constituency maintained by the Election Commission of India, one must be ‘ordinarily resident in a constituency’.Footnote ²¹ The Representation of People (Amendment) Act of 2003 amended this section to substitute the concluding phrase, ‘in that State or territory’, with the phrase, ‘in India.’Footnote ²² Hence, the amendment effectively removed the requirement that candidates for the Council of States must be ordinarily resident in that State as long as they can show ordinary residence in India.

In Kuldip Nayarv. Union of India,Footnote ²³ the petitioners challenged the 2003 amendment to the Representation of People Act on the grounds that it destroyed the basic feature of federalism, which motivates the creation of a bicameral legislature with different models of representation. They urged the Court to read in a State ‘domicile’ requirement into Article 80(4) and to declare the statutory amendment unconstitutional. In its response, the Union relied on the absence of a State domicile requirement in Articles 80 and 84 and on parliament's power to legislate on all ‘other qualifications’ required for a member of the Council of States.Footnote ²⁴

The five-judge constitutional bench in Nayar, speaking through Chief Justice Sabharwal, reviewed the institutional history of the bicameral structure of the Houses of Parliament prior to the Constitution of India 1950 to conclude that:

Hence, the Court concluded that ‘residence/domicile is an incident of federalism which is capable of being regulated by the Parliament’Footnote ²⁷ by laws enacted under Article 84. Chief Justice Sabharwal's distinction between ‘strict federalism’ and Indian federalism is one that rests on the recognition that the Indian Constitution was dissimilar from the U.S. Constitution with respect to the composition of the upper house. However, he did not pay attention to the more relevant and critical academic literature on the history of Senate election reform and the impact of the Seventeenth Amendment to the U.S. Constitution, which replaced election by State legislatures with direct election to the Senate.Footnote ²⁸ His focus was on the particular role and function assigned to the Council of States in the Indian Constitution.

Chief Justice Sabharwal noted that except for its special legislative function under Article 249, where it may resolve to grant the Union legislature temporary superiority over the State legislature on a specific subject matter, the Council of States did not perform any other federal function. Hence, he concluded that India adopted the bicameral legislature not to advance federalism by creating a chamber to ‘champion local interests’Footnote ²⁹ but rather as a ‘revising chamber’ that enhanced the quality and extent of deliberation.Footnote ³⁰ Although he justified the Rajya Sabha composition rule as an expression of bicameralism, he did not reject the idea that the Indian Constitution is a federal constitution. He surveyed the Constituent Assembly Debates and Supreme Court precedent to locate ‘[t]he basic principle of Federalism [to be] that the legislative and executive authority is partitioned between the Centre and the States not by any law to be made by the Centre but the Constitution itself’.Footnote ³¹ Furthermore, he attempted to reconcile the principle of federalism with the new amendments to the composition of the Rajya Sabha by proposing that in India, the ‘principle of federalism is not territory related’Footnote ³² and that it ‘is no part of Federal principle that the representatives of the States must belong to that State’.Footnote ³³ He stressed that this nonterritorial federal principle was a basic feature of the Indian Constitution and the amendment to the Representation of People Act of 2003 did not damage or destroy this constitutional principle.Footnote ³⁴ In other words, the ‘Constitution does not cease to be a federal constitution simply because a Rajya Sabha Member does not “ordinarily reside” in the State from which he is elected’.Footnote ³⁵ The use of the phrase 'nonterritorial federalism' in this case bears little resemblance to the academic use of it.Footnote ³⁶ Nonterritorial federalism refers to the constitutional strategies devised to reconcile ethnic, linguistic, or religious diversities that are not territorially dispersed. Chief Justice Sabherwal's reference to nonterritorial federalism simply refers to his conclusion that a representative of the State in the Council of States need not reside in the State.

A more convincing rationale for the decision in the case would be one that distinguished between nation-state and state-nation federal relations that could articulate why representation of the States in the Council of States may have a different normative status in a coming-together rather than a holding-together federation. In a holding-together federation, the State is not an autonomous independent unit and a bearer of sovereign rights that accedes to the federation only on the premise that their sovereign autonomy will not be eroded. Instead, bicameralism in the Indian Constitution arguably is motivated by the need to preserve ‘difference’ in composition and deliberation in the different legislative chambers by representing politically salient linguistic and cultural diversity.Footnote ³⁷ The change in the mode of representation of the Council of States in Nayar does not vitiate the demos-enabling features and the principles of political asymmetry embedded in the federal design of the Union legislature. Hence, state-nation federalism better justifies the conclusion arrived at in Nayar than the account of nonterritorial federalism offered by the Court.

Redrawing State Boundaries

Stepan, Linz, and Yadav posited that when the history and formation of the nation is better understood to be the result of a holding-together rather than a coming-together process, state-nation federal arrangements tend to be adopted. In a holding-together federation, there often is a need to reorganize State boundaries or to create enclaves within States that enjoy a special relationship with the State and Union government. Often, the capacity to alter State boundaries has been criticized as an illustration of why India is not a ‘proper’ federation.Footnote ³⁸ Instead, as Stepan et al. argue, if we understand this flexibility of State boundaries to be a valuable feature of a state-nation's capacity to accommodate and reflect politically salient cultural, linguistic, and ethnic demands, it no longer seems constitutionally or normatively deviant.

In the past three decades, two types of cases in which the Union alters State boundaries deserve close attention: reorganization of existing States (in this section) and the admission of new States (in the following section). The process of internal reorganization of State boundaries has been motivated by the need to accommodate four contingencies: linguistic-identity–based political-autonomy claims; tribal and ethnic claims in northeast and central India; new territory added to the Union by conquest or accession; and emerging claims of underdevelopment and political neglect.Footnote ³⁹ Often, these contingencies combine to present a complicated case for division of existing States into one or more resulting States.Footnote ⁴⁰

In Pradeep Chaudhary vs. Union of India,Footnote ⁴¹ the petitioners challenged the Uttar Pradesh Reorganization Act of 2000, which created the State of Uttaranchal. Under the proviso to Article 3, the president must refer the Bill to create a new State ‘to the Legislature of that State for expressing its views thereon’.Footnote ⁴² The Schedule to the Bill creating the new state of Uttaranchal referred to the Uttar Pradesh State legislature included Haridwar ‘city’ but not the entire Haridwar ‘district’. Once the State legislature approved the Bill by resolution, the parliament amended the Schedule to include the entire ‘Haridwar’ district. Petitioners claimed that the president had to refer the amended Bill back to the State legislature to satisfy the conditions set out in Article 3. The Court rejected this argument and held that ‘substantive compliance’ with the proviso was sufficient and ‘even in a case where substantive amendment is carried out, the amended Parliamentary Bill need not be referred to the State Legislature again for obtaining its fresh views’.Footnote ⁴³

The predominance of the Union legislature in determining State boundaries is not a novel development in Indian constitutional law. In this case, the two-judge bench affirms the earlier constitutional-bench decision on this question.Footnote ⁴⁴ However, the Court did not recognize and reaffirm the unique model of ‘state-nation’ federalism to justify this extraordinary Union power in the Indian Constitution. The failure to justify this Union power to reorganize territorial boundaries with substantive constitutional or political reasons has led to persistent criticism and renewed political demands to construe the State resolution as a constitutional fetter on the Union power. In Babulal Parate, a constitutional bench of the Supreme Court confronted the political jostling around whether the city of Bombay should belong to Maharashtra or Gujarat. The president had referred a Bill to the State Assembly, which proposed a three-way split among Gujarat, Maharashtra, and a Union Territory of Bombay. After the State Assembly approved this Bill, it was amended by parliament to include Bombay within the State of Maharashtra – a not-so-insignificant change! Despite the significant unilateral alteration by the Union parliament, the Supreme Court concluded that this was not a sufficient reason to strike down the Union law because the States had no rights under the Indian Constitution.

The Court's appreciation of the unique character of ‘state-nation’ federalism is critical to understanding and justifying the Union's preeminence while designing state boundaries. In 2013, there was divisive and rancorous political and constitutional argument about the creation of the new State of Telangana from the existing State of Andhra Pradesh. It was argued that no new State should be created if the Andhra Pradesh legislative assembly rejects or refuses to pass a supporting resolution. However, as discussed previously, the absence of a resolution by the State Assembly will not be a constitutional barrier to the creation of the Telangana State unless the Supreme Court changes its interpretation of Article 3.

Hence, the constitutional argument against the creation of the Telangana State is not about what the law is but rather what it should be. There is nothing in the text of the proviso to Article 3 to indicate that the parliament must accept or act on the views of the State legislature. The Supreme Court's justification for such a view should be understood as the result of the embrace of a particular type of federalism: a holding-together federalism as a part of a state-nation political arrangement. In Babulal Parate, the Supreme Court observed that:

As Stepan, Linz, and Yadav showed, in a federation of this type, the Union's capacity to shape State boundaries to respond to claims for political autonomy based on linguistic, ethnic, religious, or tribal identities arguably has strengthened the capacity of the Indian federation to endure in the past sixty years. Any attempt, by the president or the Supreme Court, to constrain this Union power with new constitutional or political limitations may have a lasting impact on the future of the Indian federation. However, it appears that the state-nation federalism justification for the Union power to create new States does not take the problem of political partisanship seriously enough. The third section of this chapter considers this aspect of the federal problem. The final part of this section discusses State-nation federalism, which examines the capacity of the Union to create an asymmetric federalism in India.

Asymmetric Federalism

The second case in which the Union power to alter State boundaries has come up before the Supreme Court relates to the power to admit new States into the Union. In the 1960s, the Supreme Court clarified that the cession of the territory of the Union required a constitutional amendment under Article 368 but said nothing about the accession of territory by the addition of new States.Footnote ⁴⁶ Subsequently, the Union added the new State of Sikkim to the Union of India under Article 2 and, through the Constitution (Thirty-Sixth Amendment) Act of 1975, preserved certain historical features of the erstwhile princely state in Article 371-F. In RC Poudyal v. Union of India,Footnote ⁴⁷ the addition of the new State under Article 2 and the reservation of seats for certain ethnic and religious groups through Article 371-F was challenged on the grounds that this damaged or destroyed the basic features of the Constitution – namely, secularism, democracy, and republicanism.Footnote ⁴⁸ The basic structure arguments in this case are reviewed elsewhereFootnote ⁴⁹; this chapter focuses on the arguments around Article 2 in this case.

Article 2 of the constitution provides that ‘Parliament may by law admit into the Union…new States on such terms and conditions as it deems fit’.Footnote ⁵⁰ The central questions in Poudyal are the extent to which parliament's power to admit new States is constrained by basic principles of Indian constitutionalism and the need to maintain parity between States. The Court concluded that parliament must adhere to the foundational principles of the constitution (i.e., secularism, republicanism, and democracy) but need not ensure complete equality among the States. It is surprising that at no point in this judgment did the Court offer a justification for this view of Indian federalism that is tolerant of unequal States in the Union. It avoided a substantive account of State asymmetry and preferred to justify this conclusion to arise from due deference to parliament's power under Article 2. A substantive constitutional justification for Sikkim's place in the Indian Union would be one that gives special attention to state-nation federal arrangements. However, there seems to be considerable academic disagreement on whether Sikkim is representative of Indian federalism.

The academic view on the significance of Sikkim's status in the Union under the constitution is rather sharply divided. Tillin argued that we should see Sikkim and other North Eastern States as ‘peripheral’ units whose relationships with the Union and the other States are distinct from those of the main constituent units. Hence, Tillin concluded that these are not serious asymmetric federal arrangements characteristic of Indian federalism.Footnote ⁵¹ More recently, Saxena argued that Tillin's view of Sikkim and other North Eastern states as peripheral units misunderstands the radiating effect of constitutional and political asymmetry beyond these examples to contribute to a robust plural and ‘postmodern’ democracy.Footnote ⁵² By giving equal attention to political and constitutional asymmetry, Saxena argued that these arrangements well may indicate the future political and constitutional contours of Indian federalism. This debate in the political science literature ignored the decision on the special status of Sikkim in Poudyal, which effectively locates a constitutional principle of asymmetric federalism in its interpretation of Article 2. Arguably, this is a principle with a more general application beyond Parliament's power to admit new states. The discussion now considers another asymmetric federal arrangement.

The second type of asymmetric division of power arrangement adopted under the Indian Constitution relates to sub-State political entities. Such entities are vested with political autonomy so that they enjoy greater political autonomy than other regions within the State, as well as a special relationship with the Union government. This sub-State asymmetry is constitutionally distinct from the federal territorial division of power because these arrangements were established through the Fifth and Sixth Schedules of the Constitution, special constitutional amendments for regions such as Vidarbha and Telangana, and nonconstitutional statutory arrangements as in the Gorkhaland Autonomous Hill Councils. These constitutional and legal arrangements have been seldom tested before the courts. However, in Pu Myllai Hlychho v. State of Mizoram,Footnote ⁵³ the role of the governor in constituting the Sixth Schedule Councils was challenged before the Court.

The provisions of the Sixth Schedule to the Constitution evolved a separate scheme for the administration of the tribal areas in Assam, Meghalaya, Mizoram, and Tripura through the institution of District Councils or Regional Councils. The Mara Autonomous District Council (MADC) set up under Paragraphs 2(1) and 20 of the Sixth Schedule of the Constitution had nineteen elected and four nominated members. The Governor of Mizoram was empowered under Paragraph 2(1) and Paragraph 20BB to nominate four members of MADC. After the general elections to the MADC in 2000, the governor appointed Pu Myllai Hlychho and three others as nominees to the MADC. In 2001, the governor issued a notification terminating the nomination of Hlychho and others and, by another notification, nominated four other members as nominees in their place. Hlychho challenged the termination and new nomination by way of a writ petition before the Gauhati High Court, which upheld the validity of both notifications. Hlychho and others filed appeals before the Supreme Court. The main issue was the nature of the discretion to be exercised by the governor of a State while nominating and removing members of a council under Paragraphs 2 and 20BB of the Sixth Schedule.

Typically, the executive powers of the State are vested in the governor who exercises his functions on the aid and advice of the Council of Ministers headed by the Chief Minister, ‘except in so far as he is by or under this Constitution, required to exercise his functions or any of them in his discretion’.Footnote ⁵⁴ Hlychho argued that Paragraph 20BB the Sixth Schedule of the Constitution read with the Object and Reasons of Constitution Amendment Act of 1988 required the governor to discharge a dual role: the ordinary constitutional role as the head of the State executive and a special role as the guardian of minorities in the Sixth Schedule areas. The constitutional bench failed to appreciate the arguments of the petitioner that rested on the special asymmetric character of the political arrangements in the Sixth Schedule areas and held that the governor was bound by the aid and advice of the Council of Ministers. Justice Balakrishnan concluded that the dismissal of the four members from the MADC by order of the governor, after consultation with the Council of Ministers, did not in any way constrain the exercise of discretion by the governor.

Unlike in Poudyal, in Hlychho, the Court failed to appreciate and develop the special political and constitutional character of asymmetric sub-State political entities.Footnote ⁵⁵ In both of these cases, the political branches of government have been adroit at developing asymmetrical federal and sub-State arrangements to respond to political challenges at maintaining a state-nation federal polity. However, the courts have failed to identify, recognize, and develop a constitutional jurisprudence that comprehends state-nation federalism. This section shows how an appreciation of the distinctive character of state-nation federalism will liberate the Indian courts to offer, in some cases, better remedies and in others more persuasive justification for their decisions in the federalism cases before them. The next section turns to another aspect of Indian federalism in which the Indian courts have responded with more alacrity and insight: the problem of partisan federalism.

Proclamation of Regional Emergencies

The Constitution of India grants the President of India an exceptional power to suspend a State government and legislature if the president is convinced that ‘the government of a State cannot be carried on in accordance with the provisions of this Constitution’.Footnote ⁶³ The president's exercise of this power may be prompted by the report of the governor of a State. The frequent use of Article 356 after the collapse of the single-party monopoly of the Congress Party in 1967 resulted in searching judicial review of the decisions of both of these high constitutional authorities: the governor and the president. The first significant case in which the politically egregious use of these proclamations was challenged was the State of Rajasthan v. Union of India.Footnote ⁶⁴ The Janata Party for the first time had won the general elections to the Union Government and broken the Congress monopoly. Promptly, the new government issued a letter or directive to six Congress Party–ruled State governments to resign or face the risk of a presidential proclamation of regional emergency under Article 356. The States challenged this directive before the Supreme Court, which was confronted with a case in which evidently partisan considerations sought to override federal constitutional arrangements that provided for autonomous independent government in the Union and the States.

The seven-judge bench issued plurality opinions, which for the most part steered clear of parsing partisan political considerations from constitutional justifications for the use of Article 356. The difficulty of such an exercise is best set out by Justice Bhagwati, who deferred to the executive's view on the relevance of partisan considerations in the proclamation of emergency. He observed that:

However, this permissive approach in State of Rajasthan allows partisan political reasons to masquerade as constitutional justifications under Article 356 and effectively opened the gates to the abuse of Article 356. In the next fifteen years, in almost all instances in which an Article 356 proclamation of emergency came to be used, the party in power at the Union government was different from the party in power at the State government. Nine judges of the Supreme Court were called on to revisit the place of partisan considerations in Article 356 proclamations in SR Bommai v. Union of India.Footnote ⁶⁶ In this case, the Supreme Court was confronted with several legal issues, including the applicability of basic-structure review to protect federalism and the scope and extent of administrative law judicial review of high constitutional authorities. Because these questions have been addressed elsewhere,Footnote ⁶⁷ this chapter focuses on the capacity of the Court to address the problem of partisan federalism.

The Supreme Court's reasoning on this issue is best understood by analyzing the relief granted by the court in Bommai. With respect to Karnataka, Meghalaya, and Nagaland, the Court invalidated proclamations issued by the president on the grounds that these proclamations were issued without exhausting or in contravention of constitutional ‘options’ that could have enabled the respective newly constituted governments to justify their formation. When proclamations were issued on the grounds that the secular fabric of these States had been compromised – by the involvement of several Members of Legislative Assemblies in the demolition of Babri Masjid and the subsequent riots that broke out in the Bharatiya Janata Party–ruled States of Himachal Pradesh, Madhya Pradesh, and Rajasthan – the Court upheld these proclamations. Although in all six States opposition parties were in power or had the potential to secure power, the Court effectively distinguished constitutionally valid reasons for a presidential proclamation from invalid partisan reasons.

The majority judgment of Justices P. B. Sawant and Kuldip Singh squarely grasped the core analytical issue posed by partisan federalism and recognized Indian democracy's multiparty character:Footnote ⁶⁸

The judges proceeded to observe that the provision had been used in more than ninety instances and almost invariably against opposition-party governments. It was the task of the judiciary, they argued, to intervene in such cases and preserve a pluralist Indian democracy. By locating the use of presidential proclamations at the fault lines between the multiparty system and the federal constitutional arrangement, this opinion rightly identifies the judicial role in the preservation of the federal constitutional arrangement by negating the place of partisan considerations in federal decision making. However, is it always possible for judges to identify such partisan considerations? Justices B. P. Jeevan Reddy and Aggarwal in their concurring opinion in Bommai clarify the judicial role by distinguishing between the constitutional and political considerations that motivate presidential proclamations under Article 356:

The Supreme Court's willingness to distinguish between valid constitutional reasons and invalid partisan reasons to invoke the presidential proclamation under Article 356 clarifies the question of constitutional validity and emphasizes the normative purpose of federalism as a constitutional doctrine in India: to be demos-enabling in character. However, it is not always possible to distinguish between demos-enabling and demos-constraining outcomes in complicated and fast-developing political situations.

A constitutional bench of the court in Rameshwar Prasad v. Union of India was confronted with such a situation.Footnote ⁷⁰ The Governor of Bihar, appointed by the Union government, issued a notification dissolving the State Legislative Assembly, even before its first meeting, on the grounds that attempts were being made by a political party in opposition to the Union government to cobble a majority by illegal means in a hung assembly with no clear political majority. The governor concluded that if this situation were allowed to persist, it would amount to a derailment of the democratic constitutional process. The president approved this notification. The petitioners filed a petition alleging that the governor had misused his power to prevent the formation of a government led by a party opposed to the one in power at the Union government for partisan political reasons. Several legal arguments were raised in this case, which have been responded to more fully elsewhere.Footnote ⁷¹ This discussion focuses primarily on the Court's capacity to distinguish between constitutionally valid reasons and politically partisan reasons.

In Rameshwar Prasad, much turned on the governor's report to the president, which concluded that there was a breakdown of constitutional machinery. The different conclusions reached by the majority and minority opinion in this case illustrate the perils of adjudication in such cases. Justice Y. K. Sabharwal, speaking for the majority, was skeptical of the governor's ability to make a substantive constitutional judgment about the proclamation of emergency. He observed that:

While not expressly doubting the political motivations of the governor in this case, the majority opinion clearly understood the report in this light. By circumscribing the range of valid constitutional reasons that the governor may consider when sending a report under Article 356, the majority sought to eliminate partisan considerations from such a decision. Moreover, the majority, when in doubt, sought to enhance the demos-enabling aspects of federalism by allowing the electorate to determine the political fortunes of parties.

In his vigorous dissent, Justice Pasayat stated the matter more precisely:

Justice Pasayat concluded that the governor was right to prevent corrupt legislative maneuvers to secure a majority because he was expected to act to protect the constitutional values of democracy. The dissent in this case makes clear that the juridification of this arena of political action well may eliminate some partisan political considerations but yet not yield neutral constitutional principles. Hence, the dissent upholds the governor's decision as a legitimate constitutional restraint on the conduct of politics at the State level.

Despite the doubts expressed by the minority, the Supreme Court's judicial review of presidential proclamations under Article 356 effectively has constrained, if not eliminated, partisan federalism in this area of Indian constitutional practice. The invocation of regional emergencies has been reduced rather dramatically since Bommai. However, the capacity of the Court to clearly identify when Governors are motivated by partisan considerations in complex factual circumstances is in some doubt. Whereas in Rameshwar Prasad, the Court was divided on this question, in more recent cases it has turned its focus on the appointment of governors rather than the exercise of its functions.

Appointment and Role of Governors

In 2004, the president removed the governors of the States of Uttar Pradesh, Gujarat, Haryana, and Goa, all of whom were not appointed by the Congress party then in power at the Union government. Article 156 provides that the governor shall ordinarily hold office for a term of five years, subject to the general principle that the governor shall hold office at the pleasure of the President of India.Footnote ⁷⁴ In BP Singhal v. Union of India,Footnote ⁷⁵ the petitioner, who was a member of a political organization, argued that the doctrine of pleasure does not give rise to unfettered discretion to dismiss a governor because he is a high constitutional authority and not an employee, servant, or agent of the Union government. The attorney general, representing the Union government, argued that 'in a democracy, political parties are formed on shared beliefs and they contest election with a declared agenda. If a party which comes to power with a particular social and economic agenda, finds that a Governor is out of sync with its policies, then it should be able to remove such a Governor'.Footnote ⁷⁶ This argument invited the Court to reconsider the extent to which partisan considerations may determine the appointment and conduct of a key constitutional authority who mediates between the power of the Union and the States.

Justice Raveendran, speaking for the bench, situated the question of the federal role of the governor in the complex political landscape in India. The nature of Indian democracy had evolved, Justice Raveendran observed, from one in which the same political party held power at the federal and State level to an era of coalition politics, multiple parties, and power sharing. In such circumstances, with political parties often having varied and shifting ideologies, the task of a governor is not to implement policies or popular mandates. As the constitutional head of the State, his role and responsibility are nonpartisan:Footnote ⁷⁷

By recognizing the perils of partisan federalism, the Supreme Court reinterpreted the doctrine of pleasure, insofar as it applies to governors, to clearly identify them as nonpartisan constitutional functionaries. The protection of tenure created by the Court gives the office of governor the normative role that can fundamentally reshape the federal character of Indian politics and constitutional law. By focusing on the appointment and role of the governor rather than the manner of exercise of executive power, the Court signals its intention to root out partisan considerations from becoming embedded in Indian constitutional practice. Despite the clarity of the Court in Singhal on the unconstitutionality of partisan considerations in the appointment and dismissal of governors, the recently elected Bharatiya Janata Party Union government has proceeded to nudge or sack governors who were appointed by the previous government. The capacity of the Court to uphold its precedent and root out partisan considerations from Indian federalism will be tested in these cases.Footnote ⁷⁸

Creation of New States

The previous section discussed the significant power of the Union of India to reorganize State boundaries to illustrate the distinctions between state-nation and nation-state federal arrangements. This chapter concludes with an account of the controversies surrounding the creation of India's twenty-ninth State in 2014 – namely, Telangana. This section revisits the Telangana controversy to highlight and illustrate the particular difficulties in distinguishing between partisan political considerations masquerading as State interests and the political salience that must be accorded to the preferences of particular political identities that claim territorial autonomy. In the case of Telangana, many commentators argued that the Congress Party and the Telangana Rashtriya Samiti may well be endorsing the Telangana State to secure partisan electoral gains. Therefore, should the President of India or the courts intervene to craft neutral constitutional rules that prevent the federal constitutional arrangements from being exploited for partisan political considerations?

The diverse range of political mobilization that gives rise to State reorganization claims in India alerts us to the analytical problems of distinguishing between legitimate political mobilization and partisan mobilization. Although linguistic State reorganization in the early decades of the Republic is now construed as legitimate constitutional redrawing of boundaries, there is no doubt that several regional party formations benefited from this process: the Dravidian parties in Tamil Nadu and the Shiv Sena in Maharashtra are prominent examples. The Akali parties benefited from the creation of a Sikh-majority Punjab state and, as Tillin persuasively showed, the Bharatiya Janata Party's political interests were critical to the formation of the States of Uttaranchal, Jharkhand, and Chattisgarh.Footnote ⁷⁹ Therefore, States invariably have emerged from legitimate demands for political autonomy anchored by political parties that either motivated or benefited from the creation of new States.

Is the movement for the creation of a Telangana state an exception to this historical pattern? There is one issue on which the Justice B. N. Srikrishna Committee Report is emphatic and clear: that the political demand for a Telangana state is perceived as legitimate due to the persistent underdevelopment of these regions.Footnote ⁸⁰ The Gentlemen's Agreement settled in 1956 devised statutory means to eliminate under-representation and underdevelopment in the Telangana region.Footnote ⁸¹ Article 371-D was introduced in 1973 to formalize the Six Point Formula through a nonterritorial asymmetric arrangement to reserve jobs and educational opportunities for people from the region. The failure of these statutory and constitutional arrangements led to the present demand for a new State. Although the Telangana Rashtriya Samiti has benefited handsomely from its leadership of the political movement that led to the formation of a new State, there is debate over whether the President of India or the Supreme Court should change the balance of power between the Union and the States in the process of State reorganization by insisting on an affirmative State resolution under Article 3.

There are three compelling reasons not to do so. First, as discussed previously, India's successful holding-together-federalism model helped to craft an enduring state-nation by allowing the Union to redraw State boundaries. Second, although we have crafted neutral constitutional rules to check partisan federalism in several cases, such as the proclamation of regional emergencies, it is difficult if not impossible to do so in the complicated and contested political environment that accompanies State reorganization. Third, there is no limited set of constitutional principles that ex-ante justify the formation of States because the primary justification for State formation is of an ex-post political character. In these circumstances, it is best left to the political process to craft a resolution to competing group claims for political autonomy and statehood rather than the president or the Supreme Court to second-guess this process through constitutional rules.