PIL Has Eaten Its Own Doctrine

Public-interest litigation is the most celebrated jurisdictional innovation of the Indian constitutional courts since independence. It is also, on the current evidence, the most consequential failure of doctrinal discipline in those courts in the same period. The doctrine was made between 1976 and 1981, in a sequence of decisions that took the doorman off the door of the writ jurisdiction. The doorman was locus standi — the requirement that the petitioner have a personal stake in the controversy, an injury particular to herself, an interest the law would recognise as her own. Bhagwati and Krishna Iyer JJ, between them, dismantled the doorman. Forty years later the door is still open. What the doctrine has not produced is any settled account of who, in fact, gets in. The discretion that decides which causes are entertained, which orders are made, and which petitions are dismissed at the gates as 'busybody' applications, is now opaque, unreviewable, and indistinguishable from docket management. I want to argue that the doctrine must rediscover internal limits or be retired.

The argument, stated

The position I want to defend has three parts.

The first is descriptive. Indian PIL was built on a defensible structural premise — that the rigidities of standing doctrine had locked the constitutional courts out of vindicating the rights of citizens who could not, by reason of poverty or ignorance, organise their own writ petitions.² The structural premise has been replaced over four decades by a different premise — that the constitutional courts may, in the exercise of writ jurisdiction, take up any matter of public concern, on the petition of any person, with the relief shaped by the court rather than the pleading. The new premise is not what the old one became. It is a different premise that has been clothed in the rhetoric of the old one.

The second is doctrinal. The shift has not happened through any reasoned judgment of a constitution bench. It has happened through accumulation. The Supreme Court's own three-phase classification of PIL in State of Uttaranchal v Balwant Singh Chaufal — poverty and human rights, environment and collective goods, governance and probity⁵ — concedes the shift while not addressing it. The doctrine has migrated. The reasoning has not.

The third is normative. A jurisdiction whose entry rule is 'the court will entertain matters of public importance' is not a jurisdiction; it is a discretion. The Court has periodically said as much. It has said that PILs filed by petitioners 'pursuing a hidden agenda', acting as 'surrogates for phantom lobbies', seeking 'commercial gain', or using the process 'merely for publicity', will be dismissed and costs imposed.⁹ The negative criteria are clear. The positive criteria — what makes a cause judicially manageable, what makes a petitioner 'litigationally competent', what makes a public concern a writ — have never been articulated in any operational form. The result is that two PIL petitions raising structurally identical claims may receive opposite treatment from the same court in the same week, on the same docket, before different benches. The doctrine has eaten its own rule.

This is the argument. I owe the reader, before I press it, the strongest version of the case I am arguing against — but first, the briefer claim about why the question matters at all.

Standing rules are not procedural fussiness. They are the gate through which a court chooses the disputes it will adjudicate, and the gate matters because the court that adjudicates everything adjudicates nothing well. A standing rule that is wide on paper and narrow in practice is worse than a standing rule that is narrow on paper. The narrow rule on paper has the merit of advertising what it does. The wide rule on paper combined with selective application in practice has the structure of an unreviewable discretion in the institution that holds the gate. The Indian Supreme Court has, on the present evidence, exactly that structure.

This matters for three reasons. First, because PIL has migrated. The early PILs in Hussainara Khatoon, in Bandhua Mukti Morcha, in PUDR, were rights-vindication cases that the existing standing rules could not accommodate; the doctrine moved to accommodate the cases.²¹² The contemporary PIL docket is dominated by governance cases — natural-resource allocations, appointments to statutory commissions, election disclosure rules — in which the petitioner is not the rights-holder but a civic actor, and the relief is not the vindication of a right but the reform of a policy. The doctrine has been used to do work it was not built for, and the work is increasingly difficult to distinguish from the political. Second, because the lock-out from the doctrine is itself a power. Refusing to hear a PIL is an exercise of public power that decides who does not get to litigate the cause they claim. The selection is unreviewed and unappealable. Third, because the rhetorical celebration of PIL has insulated it from doctrinal scrutiny. Every assessment of the doctrine in mainstream scholarship still opens with the unarguable proposition that bonded labourers were freed, that prisoners languishing without trial were released, that environmental principles were laid down. The proposition is correct. It is not a defence of what the doctrine has since become.

The first leg: the doctrine had defensible content, and lost it

The defensible content of early PIL was structural. Bhagwati J in SP Gupta v Union of India set out the doctrinal architecture that has been quoted ever since: 'where a legal wrong or a legal injury is caused to a person or to a determinate class of persons by reason of violation of any constitutional or legal right ... and such person or determinate class of persons is by reason of poverty, helplessness or disability or socially or economically disadvantaged position, unable to approach the Court for relief, any member of the public can maintain an application'.³ The architecture has three load-bearing elements. There must be a legal right. The right-holder must be identifiable, even if as a class. The petitioner must be acting in good faith on behalf of someone who cannot act for themselves. These elements are not formalisms. They are the discipline of the doctrine. Each is doing real work.

The case-law that followed SP Gupta retained these elements in form and dispensed with them in substance. PUDR tested the doctrine when a group of social activists petitioned on behalf of construction workers who were being paid below the statutory minimum wage. The Court held the petition maintainable on a 'liberalised rule of standing'.⁴ So far the doctrine is whole; the workers were a determinate class, they were disabled by their economic position from approaching the Court, the petitioners were acting on their behalf. The doctrinal slippage came in the next decade. By the time the Court was permitting concerned citizens to challenge the allocation of telecom spectrum, the disqualification of candidates with criminal records, the use of public funds in advertising campaigns, and the appointment of the Central Vigilance Commissioner,¹ the three elements of SP Gupta's architecture had become rhetorical. The 'right' became 'the public interest in good governance'. The 'right-holder' became 'the citizenry'. The 'petitioner acting on behalf of someone who cannot act for themselves' became the petitioner, full stop. The structural discipline that SP Gupta imposed had been replaced by an open invitation.

I should be clear about what I am not saying. I am not saying the governance PIL is illegitimate. I am saying it is a different doctrinal form from the rights-vindication PIL, and the two cannot be defended by the same arguments. The rights-vindication PIL is justified by the inability of the right-holder to litigate. The governance PIL is justified by the importance of the matter and the willingness of the court to take it on. The first is a doctrine. The second is a discretion. The Court has not separated them, and the silent assimilation of the second into the rhetoric of the first is what has produced the present incoherence.

The second leg: discretion without internal limits is not a doctrine

The harder argument is that the absence of internal limits is, in the long run, fatal. Every legal doctrine must, on the standard account, have an entry rule (who may bring the action), an evidentiary rule (what must be shown), and a remedial rule (what relief follows). PIL has answers to all three, but the answers are circular. Who may bring the action? Anyone the court regards as a 'litigationally competent' petitioner — a category defined in the negative by reference to who the court will refuse. What must be shown? Anything the court regards as a 'matter of public importance'. What relief follows? Whatever the court regards as necessary in the exercise of its inherent jurisdiction, including continuing mandamus, expert committees, monitoring orders, and quasi-legislative directions.¹¹ Each of these answers refers to a discretion of the court. None of them refers to a rule.

The standard reply is that the criteria are common-law in character and will accrete with case-law. They have not. Four decades on, no reported judgment of the Supreme Court has set out an operational test for what makes a public concern a writ matter, what makes a petitioner competent in a non-rights-vindication PIL, or what makes a continuing-mandamus order appropriate rather than excessive. The negative criteria — busybody, surrogate, oblique motive, commercial gain — are not affirmative criteria. They tell you what disqualifies a petition; they do not tell you what qualifies one. A doctrine that has affirmative criteria of identical strength to its negative ones is workable. A doctrine that has only negative criteria, applied at the court's discretion, is not a doctrine. It is a power held by the bench to which the petition is assigned.

The doctrinal vacuum has consequences beyond the inelegant. It produces the docket-management problem I named at the start. When the entry rule is 'the court will entertain matters of public importance', the court is institutionally invited to allocate its scarce attention to matters by reference to political salience, public profile, and bench composition. The criteria are not articulated, but they operate. The PIL that the press will cover gets heard. The PIL that turns on a constitutional issue with no political profile gets dismissed for lack of standing or, more commonly, for being a 'publicity interest litigation'. The dismissal does not require reasoned reconciliation with the cases that were heard the same morning on indistinguishable facts. A doctrine that produces this kind of decision pattern is not in the business of vindicating rights. It is in the business of selecting the political controversies the court will manage.

The contrast with the older standing doctrine sharpens the point. The traditional rule — that the petitioner had to show a personal legal injury — produced its own injustices. The framers of the PIL revolution were right to identify those injustices and to act against them. What they did not do, and what their successors have not done, is to replace the discarded rule with an alternative rule of comparable specificity. The replacement is not a rule. It is a discretion. Discretion can be exercised well; the early PIL judges did. Discretion can also be exercised badly; later benches have done so. The structure that the doctrine offers is the same in either case, which is the indictment.

The third leg: PIL has produced PIL disasters, and the doctrine has not absorbed them

The argument that PIL must rediscover internal limits is not made plausible by the principle alone; it is made plausible by the record. Shyam Divan's account of PIL — the most institutionally credible one in the contemporary literature, because Divan himself is among the architects of the field — concedes that the doctrine has, on occasion, produced 'disastrous judgments re-criminalising homosexuals, depriving millions of Mumbai's citizens access to green and recreational spaces, and foisting on a nation a river inter-linking project that could severely undermine the subcontinent's ecological integrity'.¹ Each of those phrases names a specific PIL line.

The Naz Foundation reversal in Suresh Kumar Koushal v Naz Foundation⁶ was made possible by the PIL doctrine. Religious and conservative organisations, none of whom were criminalised by Section 377, had standing under the PIL doctrine to seek the reversal of a fundamental-rights victory that the actually-affected community had won. The doctrine treated them as 'concerned citizens' with a stake in the public interest. The two-judge bench reversed Naz; Navtej Singh Johar later restored it. The point is not that PIL produced a wrong answer that was later corrected. The point is that PIL produced a forum in which a fundamental-rights win could be unwound by petitioners who, on the rights-vindication account, had no business being in the forum at all. The structural defect was the standing rule, not the substantive holding.

The Mumbai mill-lands cases⁷ are the second class. The Supreme Court, supervising redevelopment formulae for textile-mill lands, produced outcomes that disadvantaged exactly the constituency — the working-class residents of central Mumbai — whose claim to recreational and green spaces was strongest. The PIL framework converted what should have been a planning question, contested before the Maharashtra Town Planning authority on the record, into a series of court-supervised formulae arrived at on bench-balanced equities. Divan, whose work has been substantially shaped by the environmental PIL tradition, has flagged this as a failure mode of the doctrine.¹ The failure mode is structural, not personal.

The river-linking direction⁸ is the third class. The Court, on a PIL petition, directed the executive to consider an inter-basin transfer scheme of intercontinental ecological consequence on a record consisting principally of expert affidavits filed by interested parties. The direction was not a holding of unconstitutionality; it was a quasi-legislative instruction to a future executive on a question the petitioner, by no possible reading of standing, had any greater interest in than any other Indian. The Court that issued the direction did not, by virtue of doing so, acquire the technical competence to evaluate it. The PIL framework permitted, without internal challenge, the conversion of a major policy question into a Court-supervised process.

These are not three small failures. They are three PIL lines that touch fundamental rights, urban planning, and inter-state environmental policy respectively, and they were made possible by the same doctrinal infirmity — that PIL has no operational entry rule beyond the court's discretion to admit or refuse. A doctrine whose admitted failure modes touch this range of questions has produced enough record to support a structural critique.

The counter-argument, addressed honestly

The strongest reply to all this is also the most institutionally serious. It runs as follows. PIL has done historic work. Bonded labourers were liberated by the thousand in Bandhua Mukti Morcha. Undertrial prisoners in Bihar's jails were released after Hussainara Khatoon.²¹² Children's rights, women's rights in the workplace (the Vishaka guidelines were a PIL outcome), prison reform, custodial-death compensation, environmental jurisprudence in its entirety — the precautionary principle, polluter-pays, sustainable development, the public-trust doctrine — were laid down through PILs. Parliament has, in many of these areas, since enacted statutes that ratified the Court's directions. The doctrine has, in this account, been a generative force in the legal system. The price of having an open-ended jurisdiction is occasional bad cases. The benefit is that the rights of the inarticulate, the diffuse, and the underrepresented have a forum.

I take this seriously. I take it seriously enough that I will not argue for the retirement of PIL in its rights-vindication form. The rights-vindication PIL is a defensible, doctrinally specific institution, and I do not want to lose it. The argument is narrower. It is that the rights-vindication PIL — the PIL that vindicates the rights of identifiable right-holders who cannot litigate for themselves — has been doctrinally collapsed into the governance PIL, which is a different beast and which must be either rebuilt on its own foundations or retired. The reply that 'PIL has done historic work' is a reply to a position I do not hold. The relevant question is not whether PIL has done historic work but whether the doctrine can be relied upon to do that work consistently, and whether the doctrinal vehicle that did the historic work can be distinguished from the vehicle that has produced the governance failures.

The further reply — that the doctrine has internal corrective mechanisms, including the imposition of costs on frivolous petitioners, the dismissal of 'busybody' applications, the appointment of amicus curiae to take over PILs from suspect petitioners — does not, in the end, do the work the doctrine needs. These mechanisms are themselves discretionary. They do not constrain the court; they augment its power to manage its own docket. A doctrine cannot be saved by reference to the discretionary mechanisms that further enlarge its discretionary character. That is a circle.

What real limits would look like

If the doctrine is to be saved, it has to be limited. Three structural reforms suggest themselves, each defensible on its own and stronger together.

First, the rights-vindication PIL and the governance PIL should be doctrinally separated. The Supreme Court could, by judgment of a constitution bench, hold that the SP Gupta standing test — legal injury to an identifiable class unable to litigate — is the entry rule for the rights-vindication PIL, and that the governance PIL operates under a distinct, narrower rule which requires the petitioner to show, at the threshold, that the matter is non-justiciable in any other forum, that the relief sought is judicially manageable, and that the petition is not a substitute for political contestation. The two doctrines have different normative bases and should not share an entry rule.

Second, the relief framework needs articulated criteria. Continuing mandamus, expert committees, and quasi-legislative directions are doctrinal innovations of real value. They are also doctrinal innovations that have been deployed in cases that do not warrant them. The Court should hold, by judgment of a constitution bench, that continuing mandamus is available only where (i) there is a specific statutory or constitutional duty whose performance the executive has failed to discharge, (ii) the failure has produced a measurable rights-injury to an identifiable class, and (iii) periodic court supervision is necessary to remedy that injury and cannot be replaced by a one-shot direction. Anything that fails this test should be a writ of mandamus, declared and disposed of, not a continuing one. The river-linking direction, on this test, would not have survived.

Third, the PIL judgment must be appealable on the merits within the Court — by review or by reference to a larger bench at the petition of any non-party interest substantially affected by the relief. This is not a perfect remedy and it has its own pathologies. It is, however, a structural answer to the present situation, in which a PIL judgment is binding on every Indian whether or not she had any opportunity to argue against it before the bench that produced it.

None of this requires constitutional amendment. None of this requires legislation. It requires the Supreme Court, in a future PIL case raising the doctrinal question directly, to constitute a constitution bench under Article 145(3) and to do the doctrinal work that the doctrine has not done for itself in forty years. The Court has shown, in Balwant Singh Chaufal, that it is capable of describing its own evolution; it has not yet been willing to discipline that evolution.⁵

These three reforms would not, by themselves, restore SP Gupta's discipline. They would, however, force the doctrine to articulate the criteria on which it has been operating in the dark. The articulation would itself be a discipline. Where the criteria the Court were prepared to defend on the record turned out to be defensible, the doctrine would gain legitimacy. Where the criteria turned out to be indefensible, the doctrine would lose them, and the Court would, by virtue of having written them down, find them harder to use selectively. The point of writing rules is not that rules constrain perfectly; the point is that rules constrain in ways that discretion does not. A discretion exercised in private at the Chief Justice's roster is a different beast from a rule announced in a reported judgment of the Court. The PIL doctrine has the first. It has not had the second since SP Gupta, and SP Gupta was a different doctrine.

The reforms would also unsettle a particular interest group within the PIL bar — the petitioners' counsel who have built practices around the doctrine's openness, and the State counsel who have built their oppositional strategies around its irregularities. Both groups have, on the current arrangement, a working accommodation with the doctrine's discretion. Both would lose, on a structural reform, some of what makes the practice profitable. This is not a reason to refuse the reform. It is a reason to expect that the reform will not come from the bar. It will have to come from the bench, and the bench has not yet shown the institutional appetite to do it.

If the structural reforms do not happen, the case for retirement is real. A jurisdiction that admits matters on standards that cannot be articulated, produces relief that cannot be appealed, and corrects its own failures only by occasional reversal in later judgments by different benches, is not a defensible jurisdiction in a constitutional democracy. The standing rule it displaced was at least visible. The doctrine that displaced it is not.

The retirement option is not, on this account, an act of judicial conservatism. It is the recognition that the rights-vindication work the early PIL judges did is still doable, but it is doable through a properly disciplined writ jurisdiction under Articles 32 and 226 in their orthodox form. The early PILs would, on a strict reading of SP Gupta, have all been admissible. What would not be admissible is the contemporary governance PIL, and the loss of the contemporary governance PIL is, on the record, a smaller loss than the doctrinal incoherence is producing. The activist civic claims that the governance PIL now carries would have to find their forum elsewhere — in legislative advocacy, in administrative-law challenges with conventional standing, in writs that produce one-shot directions rather than continuing supervision. Some of those claims would be lost. Many would be relocated to forums better suited to them.

The doctrinal architecture of Articles 32 and 226 is, as Gopal Subramanium has put it, predicated on the protection of 'an existing legal right of the applicant, or an existing duty of the respondent'.¹⁰ The text is narrower than the doctrine has become. The text has not been amended. The doctrine has migrated away from it. The case for retirement is that, in the absence of internal limits, the doctrine is not a development of the constitutional text; it is a parallel jurisdiction the Court has assigned to itself, justified by the rhetoric of social transformation, and operated on the institutional discretion of whoever holds the roster.

I do not think retirement is the right answer. I think the right answer is structural reform of the kind I have set out. But the case for retirement is the disciplining backdrop to the case for reform. If the Court is unwilling to do the doctrinal work, the doctrine should not survive. The early PIL judges built it because the existing doctrine could not accommodate Hussainara Khatoon and PUDR. The contemporary PIL judges have inherited it without taking on the discipline its founders, in different times, did exercise. The doctrine cannot live on the achievements of its first decade in perpetuity. Either it must do new doctrinal work that recovers its discipline, or it must be returned to the writ jurisdiction it grew out of. The status quo — celebration without limits, discretion without rules, occasional failures absorbed by the general goodwill of the institution — has run its course. PIL has eaten its own doctrine. The question is whether it can spit some of it back up, or whether it must be permitted to digest the rest.

Get the week's biggest decisions, distilled.

A weekly digest for lawyers, students and the legally curious. Every Friday morning. Free.