ADR v Union (the SIR Judgment): We will not protect you

This week, the Supreme Court delivered its judgment in ADR v Union of India upholding the legality of the ongoing Special Intensive Revision (‘SIR’) of electoral rolls. Earlier posts on this blog have analysed the Court’s flawed approach towards hearing the issues in the case , the key competing arguments , and how the SIR ties into the broader structure of democratic erosion . This post analyses the judgment itself.

Ordinarily, one might analyse a judgment’s reasoning to understand whether it was justified in its conclusions and outcome. However, as PART B of this blog demonstrates, the Court’s reasoning in ADR v Union rests on misstated doctrine, avoidance of binding precedents, and outright denial of logic and facts. These mistakes are too flagrant and obvious to suggest that the judgment’s analytical missteps stem from mistakes in judicial reasoning. Judges of the Supreme Court possess a high level of intelligence, and they know exactly what they are doing. The more plausible explanation here is that the Court is engaging in post-hoc rationalisation; that it reached its conclusion to uphold the SIR first, based on political and institutional considerations not explicitly stated in the judgment, and then subsequently prepared legal reasons to reject the Petitioners’ arguments. Given this, the analysis must also invert its gaze, first scrutinising the Court’s conclusion (PART A), and then critiquing the Court’s reasoning (PART B), which offers scant support for its conclusion.

PART A: Scrutinising the Outcome

On the outcome, the most important thing to understand is what the Court refused to do. Confronted with a process that has already disenfranchised millions of individuals, the Court: (i) declined to stop it; (ii) the Court declined to impose any curbs or introduce any additional safeguards; and (iii) the Court declined to even develop the doctrinal tools that might preserve voting rights going forward. Let us take each of these in turn.

Stopping the SIR

For the Court to have considered injuncting or invalidating the SIR, the Court would have to first acknowledge that something is amiss; that the right to vote of individuals is being infringed on or the Election Commission of India’s (‘ECI’) conduct warranted legal scrutiny. As previous posts have documented, there was never a formal stay-hearing or interim order which explained why the SIR should not have been injuncted until its legality was decided. In the final judgment, we have this:

“95. The post-exercise data placed on record does not disclose a level of disenfranchisement so widespread or systemic as to indicate a constitutional infirmity in the design of the exercise. While individual cases of exclusion may arise, they are addressable within the framework of claims, objections, and appeals that have been provided.”

In other words, there’s nothing to see here, let’s all move along. In Bihar, 6.5 million names were deleted from the rolls , in West Bengal 9 million names deleted . Even assuming some fraction of these, even a majority, were legitimate deletions, we are still talking about millions disenfranchised. But apparently this does not rise to the level that concerns the Supreme Court.

Building in safeguards

Next, could the Court have imposed any curbs on the SIR? A key contention of the petitioners was that conducting the SIR on the cusp of elections in Bihar, West Bengal, Tamil Nadu, Kerala, and Assam, unnecessarily rushed the process and led to exclusions. For example, in West Bengal, around 10 million voters could not because their names had been deleted, but because their hearings could not be completed on time. A simple safeguard for the Court to lay down would be to say no SIR can be conducted within six months of the expiry of an assembly term, avoiding the chaos seen in Bihar and West Bengal. But not once, in its 120+ page judgment does the Supreme Court address the question of the timing of the SIR. In fact, at Paragraph 139 it highlights the fact that under the SIR Guidelines, individuals are only excluded after hearings are concluded to demonstrate how the SIR process is compliant with the statutory scheme. The Court apparently forgot its own order concerning West Bengal stating that given impending elections, those whose hearings and appeals were not completed prior to polling could not vote.

Expanding doctrine

The last thing the Court could have done, even while allowing the SIR to proceed, was lay down rights protective doctrine to secure voting rights going forward. The Supreme Court of India has long turned this into an art form. In Olga Tellis , the Court held the right to livelihood to be an integral part of the right to life under Article 21 while permitting the eviction of the pavement dwellers who had approached in. In Anuradha Bhasin , it extended the protection of free speech to the internet while declining to restore internet access in Jammu and Kashmir. These doctrine-expanding, rights denying judgments have been criticised for some time, and rightly so.

Yet such judgments are not without certain beneficial qualities. For one, they allow High Courts to take up and run with the newly articulated doctrine. For example, High Courts have struck down internet suspensions relying in part on Anuradha Bhasin. And the expanded doctrine is itself a signal, a recognition by the Court as an institution that the citizens were right about the rights-protecting character of the law. That even where the Court declines to apply it to the facts of the case, for reasons bound up with the politics of the day and the Court’s own limited institutional capital, it is cognisant of how the values of constitutional democracy animate the interpretation of law.

That the decision in ADR v Union declines to take even this step speaks volumes. There is no doctrinal expansion of the right to vote, and no articulation of any limit on the ECI’s powers that future challenges to disenfranchisement might invoke. There is only total deference, on law and on facts, to the Commission’s arguments. There can be two possible reasons for this. First, the Court views its institutional capital as so limited that it cannot even articulate democracy protecting language even in cases where it gives the Union exactly what it wants. Second, that the Court is in total agreement that the Constitution and the law have no role in protecting citizens voting rights from the actions of a Commission whose members are unilaterally appointed by the Union Government. In either case, its message to the citizens who challenged the SIR was unambiguous, we will not protect you.

PART B: Critical Analysis of Judicial Reasoning

The judgments reasoning, which ought to provide the Court’s justifications for arriving at the above discussed conclusions, instead ranges from the ambiguous, to the incorrect, to the openly illogical. The Court was confronted with four key questions: (i) whether the ECI was empowered to conduct the SIR; (ii) whether the ECI had to comply with the statutory scheme of the Registration of Electors Rules, 1960 (‘Registration Rules’) when conducting the SIR; (iii) whether the SIR was a proportionate exercise of power; and (iv) whether the ECI can examine the citizenship of individuals when revising rolls. On each of these issues, the Court rules in favour of the ECI.

ECI’s Powers vis-à-vis statutory compliance

In answering the first two questions, the Court goes to great lengths to argue that Art 324 gives the ECI independent and inherent powers to issue directions “of a general and regulatory character” to govern elections, even where Parliament has legislated (¶¶18, 28). While the ECI is a constitutional body, such powers are clearly legislative and in tension with both the separation of powers doctrine and the express text of Art 327 which authorises Parliament to regulate elections by law. Even the Court seems to recognise this, acknowledging that the ECI’s powers cannot be exercised contrary to statute (¶22). Coming to the SIR, this would suggest that when the ECI embarked on the intensive revision, it could not have contravened the Registration Rules adopted by Parliament for the revision of electoral rolls, especially given that r 25(2) expressly states that the Registration Rules apply to intensive revision of rolls.

But the Court does not reach this conclusion. It instead finds that the statutory power conferred on the ECI under s 21(3) of the Representation of the People Act, 1950 (‘RPA’) allows the Commission to depart from the Registration Rules. It notes that s 21(2)(b) RPA requires the rolls to be revised in “the prescribed manner” while s 21(3) allows for a special revision in any manner the ECI thinks fits. Based on this distinction, the Court holds that when conducting a special revision, the ECI is not bound by the Registration Rules (¶¶35-36). The Court finds that acting under s 21(3), the Commission has a “wide and uncircumscribed discretion, not hedged in by procedural limitations” (¶42).

But no statute can responsibly be read this way. Such a reading presumes that Parliament granted another constitutional body a free pass to act as it pleases on a subject that Parliament itself has adopted statutory rules. Where Parliament has delegated power in this manner to the executive, courts have been quick to interpret these provisions restrictively or invalidate them altogether . But in ADR v Union, the Court does the opposite, giving s 21(3) its broadest possible interpretation to free the ECI from any constraint. Further, under the Court’s interpretation, the ECI acting under its inherent constitutional powers under Art 324 cannot contravene the Registration Rules, but acting under its statutory power under s 21(3) can override the Registration Rules.

The Court implicitly recognises that this reasoning is tenuous at best, because after holding that s 21(3) frees the ECI from the constraint of the Registration Rules, it nonetheless proceeds to examine whether the ECI complied with these Rules (Section E3 of the judgment). Surely if s 21(3) confers on the ECI “uncircumscribed discretion,” no question of complying with the Registration Rules ought to arise. But at paragraph 53, it observes that the ECI’s actions under s 21(3) must avoid transgressing any “express prohibitions contained in the Act or the Rules,” suggesting that the Registration Rules impose some constraints on the SIR Process. But the Court refuses to articulate what these constraints might be, except for holding that the Commission has not breached them. Thus, in a case that asked the Court what the limits on the ECI’s discretionary powers were, the Court fails to lay down any principled reasoning for what such limits may look like.

Proportionality analysis is per incuriam

The test of proportionality is now relatively well settled in Indian constitutional law. To survive it, an impugned measure must: (i) pursue a legitimate aim; (ii) be necessary to achieve that aim; and (iii) be proportionate to the aim sought to be achieved. The term “necessary” has been interpreted by a Constitution Bench of the Court to mean that there should be no lesser restrictive measure the government could have adopted that would have achieved the same aim. The term “proportionate” is understood as a proper balance between the legitimate aim and limitation of the constitutional right.

On the first limb, the Court in ADR v Union finds that the exclusion of ineligible voters from the rolls was a legitimate aim (¶63). Next, the court articulates the test of “necessity” as follows:

“78. At the outset, we deem it appropriate to clarify that the enquiry under this limb is not whether the Court would have adopted a different or more optimal method, but whether the measure chosen by the component authority is so disproportionate or manifestly excessive that it cannot be sustained within constitutional bounds. The doctrine of proportionality does not mandate the adoption of the least restrictive measure in the abstract; rather, it requires that the measure adopted must not be so palpably arbitrary when viewed against the objective sought to be achieved.” (emphasis added)

This is not the necessity test. As Constitution Bench decisions, which are binding on the ADR v Union bench sitting in Division, have held, the ‘necessity’ inquiry asks whether the government could have adopted an alternative measure which restricted rights to a lesser degree while still achieving its legitimate aim. What the Court has done here is to substitute that inquiry with a restatement of Wednesbury unreasonableness (whether a decision is so unreasonable that no reasonable authority would take it) and pass it off as the necessity test. The two are not interchangeable. The Wednesbury test is far more deferential to the State because a measure may have any number of less restrictive alternatives without being “palpably arbitrary” or “manifestly excessive.” The mischaracterisation of doctrine and avoidance of binding precent is so blatant that the only conceivable explanation is that the Court has knowingly misstated the proportionality test to spare the SIR the scrutiny its own precedents demand.

Just why the misstatement of the proportionality test matters becomes abundantly clear when one examines the Court’s answer to whether voters already on the electoral roll prior to the SIR ought to enjoy a presumption of validity.

Presumption of validity

A key contention of the petitioners was that the SIR violated the precedent of Lal Babu Hussein v ERO . That decision, passed by a three-Judge Bench decision of the Court and thus binding on the Division Bench in ADR v Union, held that the inclusion of an individual in the electoral roll, being an official act performed by the ECI itself, carries a presumption of validity and regularity which cannot be displaced absent a notice, enquiry, and hearing. At first glance, the SIR process directly contradicts this reasoning by requiring every citizen, even those already on the electoral rolls prior to the SIR, to fill out an enumeration form to avoid exclusion. In other words, the SIR treats every voter on the rolls as presumptively invalid.

One might ask, how can the ECI create fresh rolls from scratch (which is what an intensive revision requires) without disregarding the existing rolls. After all, the raison d’etre for the SIR is that the existing rolls are unreliable. The answer is straightforward. The ECI could have conducted its house-to-house enumeration, invited all citizens to submit enumeration forms and documentation, and matched these documents against existing rolls – treating as presumptively invalid only those voters on the rolls who did not fill out the forms/documentation, calling them for a hearing. This process would have been consonant with Lal Babu Hussein, because BLOs must visit a house three times before marking a voter absent. Thus, the presumption of validity of a voter already on the rolls would only be disturbed after the State has tried to reach the voter three times, as opposed to the SIR process where the presumption of validity was overturned by executive fiat and the onus placed on citizens to re-register.

This is the obvious less restrictive measure. If the Court had applied the proportionality test as it was legally bound to do, the mere availability of this alternative ought to have been fatal to the SIR. But instead, the Court misstates the necessity limb of the test, excusing it from having to consider alternative measures. While it avoids binding precedents on the proportionality question by misstating doctrine, it reasoning to circumvent the precedent in Lal Babu Hussein is equally problematic.

First, the Court finds that if electoral rolls were treated as presumptively valid, the ECI would be “substantially disabled” from verifying voters on the list (¶117). This is not true, as the alternative outlined above demonstrates, the ECI could very well verify voters, it just could not delete them without an enquiry and hearing. Looked at another way, the Court is uncomfortable with the ECI not having the power to unilaterally delete voters without reasons. Second, the Court finds that the presumption of validity of existing rolls cannot be used “as a shield to obstruct the exercise of [the ECI’s] constitutional powers” (¶121). Again, the presumption is only a shield against the arbitrary exercise of these powers. Third, the Court simply denies that requiring all voters to submit enumeration forms or be excluded inverts the presumption (¶119). Except, as noted above, this is exactly what the SIR does. Requiring every person to submit a form at the risk of exclusion treats their existing place on the voter rolls as presumptively invalid, with the onus being placed on citizens to be included.

Conclusions on Citizenship

It is on the question of citizenship that the mask is well and truly off. The Court holds that the ECI may examine a person’s citizenship while revising the rolls, because s 16 RPA bars non-citizens from voting. The Court states that while the decision to exclude a person from the electoral roll does not amount to a declaration that they are a non-citizen, the ECI must refer all such individual to the Central Government for a determination of citizenship under the Citizenship Act, 1955 (¶186-h).