(This is a companion post with Tanmay Durani)
The directive issued by the Election Commission of India (“ECI”) mandating a Special Intensive Revision (“SIR”) of electoral rolls in Bihar has since been recast as a nationwide project. By a notification dated October 27, the ECI brought twelve states within its fold, including West Bengal.
This post argues that in managing the SIR exercise in West Bengal, the Supreme Court (“Court”) appeared to have assumed a constitutive role in shaping the adjudicatory process governing verification claims. In so doing, it pushed at - and arguably beyond - the limits of court administered executive action, raising concerns as to its competence, and more broadly, the separation of powers. A prior question - whether the SIR exercise is itself constitutionally permissible, having regard both to the absence of a specific statutory basis for mass re-verification and to the right of suffrage under Article 326 - remains live but is not addressed here (for a detailed treatment of the doctrinal arguments against the SIR’s legality, see Devadasan, “The Legality and Constitutionality of the SIR” (10 March 2026)).
The Making of an Adjudicatory Apparatus (By Court Orders)
The Court’s construction of the adjudicatory apparatus has been incremental, unfolding across a series of hearings between February and April 2025.
The trigger lay in what is called the Logical Discrepancy (“LD”) list - a category of voters flagged by the ECI’s software for inconsistencies in their enumeration forms. These inconsistencies include spelling variations in names, implausible parental age gaps, and instances where more than six individuals listed the same person as a parent. It bears noting, however, that a substantial share of these flags arose not from genuine irregularities but from the limitations of the enumeration system itself: reporting indicates that script-conversion errors, and AI-driven digitisation mismatches - particularly in the translation of Bengali names into English script - generated large volumes of apparent discrepancies that were products of data processing errors rather than evidence of fraudulent enrolment.
In aggregate, 1.36 crore voters were placed within this category, each requiring verification prior to inclusion in, or exclusion from, the final roll. The disenfranchisement risk was, accordingly, not hypothetical.
On 20 February 2025, in the context of an impasse between the state government and the ECI over the deployment of officers, the Court proposed that judicial officers of Additional District Judge (“ADJ”) rank adjudicate pending LD claims. On 24 February, citing that the existing pace of processing would stretch over an extended period, the Court then expanded the pool to include serving and retired judicial officers from neighbouring High Courts. On 10 March, yet another direction followed, to constitute appellate tribunals comprising sitting or former High Court judges, leaving questions of number, composition, and honoraria to be determined by the Chief Justice of the Calcutta High Court. Nineteen such tribunals were subsequently announced by the ECI.
The Apparatus, So Constituted
The result is worth pausing over. For the Court, in all practical senses, not only assumed control over the administration of LD verifications - which may have been defensible - but more significantly, over the constitution of the verification process itself - which is not.
The distinction between the two deserves elaboration, for it may not be self-evident. Administration, in the relevant sense, refers to the Court’s direction of how an existing process - already defined in its institutional contours by statute or delegated legislation - is to be carried forward. The Court may, in supervising executive action, issue mandates as to timelines, coverage, or the quality of compliance. What it does not do - and what, it is submitted, it cannot do - is design the process itself i.e. to determine who decides, by what procedure, subject to what appellate structure, and at what rate of remuneration. That function belongs to the legislature in the first instance, and within statutorily authorised spaces, to the executive. It is, in simpler terms, a function that the Constitution does not vest in the judiciary.
However, the apparatus - so constituted - comprises first-instance decision-makers (ADJ rank officers), an appellate tier (tribunals), a supervisory layer (the Chief Justice of the Calcutta High Court), alongside rules governing procedure and a payment structure. Worth taking note is that none of the foregoing arrangements trace back to the Representation of the People Act 1950 (“RP Act”), the Registration of Electors Rules 1960 (“Registration Rules”), or any other statutory instrument. What we have, therefore, is a de novo adjudicatory process conjured entirely through court orders, resting on nothing more textured than the “complete justice” mandate of Article 142.
The Misfit of Article 142
The “complete justice” mandate under Article 142 empowers the Court to grant relief where ordinary law proves insufficient. While the scope of its text has been read expansively, retaining what has been described as an “undefined and uncatalogued” elasticity (see Delhi Development Authority) - a baseline assumption nevertheless remains intact - i.e. the “complete justice” power operates to provide relief to parties within the “cause or matter” before the court.
For ease of reference, the relevant text of the provision is produced below:
“142. (1) The Supreme Court in the exercise of its jurisdiction may pass such decree or make such order as is necessary for doing complete justice in any cause or matter pending before it, and any decree so passed or order so made shall be enforceable throughout the territory of India...” emphasis supplied
In view of the above, two points may be made.
First, the directions issued by the Court - to constitute a bureaucracy, vest supervisory authority in a Chief Justice, create an appellate tier, and fix honoraria - do not quite cohere with any orthodox understanding of “complete justice.” This is so, because the inherent power of the Court under Article 142 is, at base, remedial in nature - and not constitutive.
Second, even if we were to momentarily set that objection aside - the character of the LD verification exercise places it at a remove from the “cause or matter” before the Court. It is true that the Court has, in public law litigation, read the “cause or matter” language broadly, treating the constitutional subject-matter of the petition rather than the narrow inter-partes lis (see, in particular, Supreme Court Bar Association v. Union of India (1998), and, in the electoral context, Association for Democratic Reforms v. Union of India (2024), in which the Court exercised Article 142 powers to restructure disclosure obligations extending well beyond the immediate parties). That expansive reading, however, does not dissolve the distinction between fashioning a remedy for the cause and administering a parallel institutional process alongside it. The original petition concerns the legality of the SIR exercise in West Bengal. The Court’s directions do not, however, adjudicate that question. They organise the manner of its implementation. That is neither resolution of the lis nor incidental to it, and hence cannot - even on the broadest reading of “cause or matter” - be brought within the same ambit.
Thus, the Court’s move from remedy to design sits outside the ordinary scope of Article 142, which is confined in its operation to doing complete justice within the lis.
A Possible Continuing Mandamus Defence (Still a Misfit)
A possible line of defence would perhaps characterise these directions as an instance of the Court’s practice of issuing a writ of continuing mandamus deployed to secure compliance in the face of rights violations (here, potential mass disenfranchisement). Derived from the concept of structural injunctions, such directions are described by Professor Owen Fiss to operate as a formal medium through which the judiciary seeks to reorganise bureaucratic processes and bring them into conformity with constitutional requirements. Vineet Narain is the standard illustration, where the Court (under Article 142) employed a continuing mandamus to insulate investigative agencies from political interference and structure their effective functioning.
The intuitive appeal of this is not difficult to see. The state government and the ECI were deadlocked over the deployment of verification officers, the pace of processing was projected to extend well beyond any administratively acceptable timeline, and a significant volume of voter claims remained in limbo. In that context, judicial intervention to unblock the verification process carries the surface-level appearance of a court doing precisely what continuing mandamus is designed for.
The continuing mandamus analogy, however, would be inexact. For even at its most interventionist, the Court has worked through existing administrative structures to secure constitutionally compliant functioning. It has not, instead, erected de novo frameworks from the ground up. The Godavarman forest case is instructive here, given it likely marks the outer limit of what the Court has been prepared to do under a continuing mandamus. Therein, the Court assumed sustained control over questions of forest policy through continuing directions spanning decades - suspending tree felling, directing the auction of illegally felled timber, monitoring illegal mining - and functioned, in effect, as the de facto regulator of India’s forests. Yet even in that extreme exercise, the Court worked through existing statutory bodies: state governments, the Forest Advisory Committee, already-constituted regulatory authorities. It did not, notably, assume the constitutive functions of the forest administration. The idea, therefore, is not to take issue with supervision of failing processes in the interest of constitutional conformity (Professor Fiss’s formulation would accommodate that) - but with the assumption of functions that go further - i.e. the structuring and staffing of a parallel mechanism through which such conformity is to be achieved.
On Separation of Powers
A third line of critique sounds in separation of powers, for the Court appears to have done something that does not go well with the orthodox allocations of authority. In the Indian context, however, that observation would not alone take the argument very far. The Court has, for long, exercised a supervisory jurisdiction over executive action - that much is now a settled feature of our landscape, and need not be rehearsed at length.
Even so, limits remain within this otherwise permissive separation. When the Court issues directions or frames guidelines, they are addressed to pre-existing institutions which remain responsible for their implementation. The Court may supervise - by reviewing compliance and so on - but it does not assume the task of constituting the system through which such implementation is to be carried out. The Vishaka guidelines, for instance, imposed obligations on employers and state authorities, leaving undisturbed the institutional identity of those charged with giving effect to them. This permissiveness in the separation, furthermore, has typically rested on legislative vacancy or executive inaction - i.e. in circumstances where judicial initiative can be defended as gap-filling.
The present case does not fit this pattern. Three submissions follow.
First, in orthodox terms, the design of adjudicatory processes lies primarily within the legislative domain, and in limited statutorily authorised spaces, with the executive. The constitutional scheme contemplates no freestanding judicial power to undertake comparable design. This observation reflects a deeper principle traceable to the basic structure doctrine in Kesavananda (given sharper expression in Minerva Mills; and more recently in Asif Hameed) that no organ of government can, in the exercise of its constitutionally defined powers, effectively arrogate to itself a function that the Constitution assigns to another. Thus, the Court’s power cannot extend to the creation of adjudicatory institutions and processes - for it is a function that is, by its very nature, ‘constituent’ rather than ‘judicial’.
Second, it might be argued that the ECI could itself have devised such a structure under its residuary powers in Article 324, and so the Court’s directions merely facilitate that exercise. On this view, the directions could also have been located within the logic of continuing mandamus, with Article 324 supplying the requisite pathway for its operation. That argument does not travel far either. The Court’s jurisprudence has consistently held that Article 324 operates in “unoccupied areas” — i.e. where statutory law is silent and only to the extent necessary to conduct free and fair elections (see Mohinder Singh Gill, AC Jose, Kuldip Nayar). Devadasan relies on AC Jose to make an analogous argument against the ECI: that the Commission cannot authorise novel procedures through its SIR notifications and then invoke Article 324’s residuary powers on the basis that the existing statutory scheme does not cover what it has itself newly introduced. The logic applied there to the ECI applies, if anything, with greater force to the Court - for where the ECI might at least claim some institutional proximity to electoral administration, the Court can assert no comparable warrant for assuming the ECI’s constitutive functions.
To this end, the RP Act and the Registration Rules together form a self-contained code governing the preparation, revision, and adjudication of electoral rolls. Under the RP Act, responsibility is vested in the Electoral Registration Officer (“ERO”), assisted by designated officers (ss 13B, 13C), to determine claims of inclusion, exclusion, and correction (ss 22, 23). The Registration Rules, in turn, lay down the procedure through notice, inquiry, and hearing (rr 13–21), while retaining decision-making authority in the ERO (r 20). It may further be noted that s 28 of the RP Act vests rule-making power in the ECI only with the approval of the Central Government - a provision that forecloses the suggestion that the ECI enjoys a unilateral statutory power to innovate upon the adjudicatory structure so prescribed. Thus, both the identity of the decision-maker and the manner of adjudication are statutorily fixed, leaving no “unoccupied area” for Article 324 to operate in.
Any attempt therefore, whether by the ECI or the Court, to introduce alternative decision-makers, appellate tiers, or supervisory control would enter a field already occupied by statute. Devadasan makes a similar point in the context of the ECI’s own reliance on s 21(3) and Article 324. He argues that the ECI had no gap to fill under Article 324, and s 21(3) could not authorise a departure from set procedure. If that argument is correct (as it is submitted it is) - then the Court’s intervention faces an additional difficulty - for not only did it lack constitutional authority of its own to design the adjudicatory process, but the space it purported to occupy on the ECI’s behalf was one that the ECI itself could not have lawfully claimed in the first place.. The “facilitation” defence thus collapses at both ends.
Third, even if it is assumed that Article 324 affords the ECI some latitude, it does not follow that the Court may exercise that latitude in its stead. A useful point of contrast is the directions under a continuing mandamus (as discussed earlier), which even at its most interventionist, operated within existing structures of administration so as to supervise constitutional compliance without mandating de novo institutional creation or designing adjudicatory processes. In the present case, by contrast, the Court did not direct the ECI to exercise its powers - it exercised those powers itself, in the ECI’s place.
To Conclude
Three lines of critique have been traced above. All point in the same direction: the Court did not have the authority to assume a constitutive function in shaping the adjudicatory process governing verification claims. Article 142 does not reach directions of this kind (1). Continuing mandamus, even at its most expansive, has never extended to the creation of adjudicatory institutions (2). And the separation of powers analysis establishes that constituting such a process is a function that neither the Court nor the ECI could perform in the manner attempted (3).
The evident counter to all of this would be one of practical necessity. Over 1.36 crore voter claims hung in the balance. Urgency explains the Court’s impulse. The turn to a parallel adjudicatory apparatus, then, likely reflects a judgement that the statutory scheme could not have managed a verification exercise of this scale within time. That judgement may even be right as a matter of administrative realism. It does not, however, supply authority. The Court’s power to act is not coextensive with the urgency of acting. And this is precisely why the Article 142 and separation of powers are important to articulate. For if such a logic is left unexamined, it furnishes a template for the judicial redesign of statutory processes whenever they are perceived to fail. That, it is argued, is a latitude the Constitution does not contemplate.
No comments:
Post a Comment