The three opinions in today’s Aadhaar judgment run to a total of 1448 pages. It will require more than a single post to examine the judgment in its entirety. In the course of the coming days, therefore, we will do so in parts. In this first essay, I will provide an overview of the Majority judgment (authored by Justice Sikri, with whom the Chief Justice and Khanwilkar J. concurred). I will focus on the findings of the Court, and the reasoning that led to those findings. A critique of the reasoning (such as it is) shall be undertaken in later posts.
Proportionality: The Legal Standard for Testing Infringements of Rights
The constitutional challenge to the Aadhaar programme – and to the provisions of the Aadhaar Act – was mounted primarily on the touchstone of proportionality. Previous judgments have left the contours of the standard somewhat vague, and therefore, the issue was fiercely contested during the course of the hearings. The majority judgment attempts to provide some clarity on this point. It refers to the different shades of proportionality employed in different jurisdictions (unfortunately with hardly any reference to case law), and then adopts (without any further explanation) the articulation of proportionality that is provided by David Bilchitz, a South African constitutional law scholar. In addition to the requirement that there must exist a law (with adequate procedural safeguards), proportionality requires the following four prongs: first, a legitimate State aim; secondly, a rational nexus between the impugned measures and the aim; thirdly, that the impugned measure be the least restrictive method of achieving the aim (the “necessity” prong), and fourthly, that there must be a balance between the extent to which rights are infringed, and the overall public benefit (the “strict proportionality” prong). It was the third limb – necessity – which was most contested, and the Majority borrows Bilchitz’s formulation, restating it thus:
First, a range of possible alternatives to the measure employed by the Government must be identified. Secondly, the effectiveness of these measures must be determined individually; the test here is not whether each respective measure realises the governmental objective to the same extent, but rather whether it realises it in a ‘real and substantial manner’. Thirdly, the impact of the respective measures on the right at stake must be determined. Finally, an overall judgment must be made as to whether in light of the findings of the previous steps, there exists an alternative which is preferable. (paragraph 124)
The majority then holds that this is consistent with the meaning of proportionality, as articulated in the Puttaswamy judgment, and also in the Court’s prior judgments dealing with the issue.
In the following posts, we will discuss the Majority’s failure to apply this standard to its own substantive analysis. For now, however, as a constitutional matter, it is important to note that this does, at least, bring a measure of clarity to the standard that Indian Courts must apply when adjudicating rights infringements on grounds of proportionality (or “reasonableness”, which is the word found in our Constitution). It is important, especially, for future challenges in similar (but non-overlapping) contexts, such as – for example – the forthcoming Data Profiling Bill.
“Uniqueness”: The Overarching Factual Assumption
In order to break down the Majority’s analysis on the multiple issues that were up for adjudication, I suggest the following map towards navigating the judgment: under each issue, there are a number of factual assumptions that underly the analysis. Upon these factual assumptions, the Majority opinion then builds its legal argument, testing the constitutionality of the Aadhaar programme (or the statutory provision in question) against the factual background.
Underlying all the different legal findings is one overarching factual assumption: the Majority believes that biometric authentication provides us with a unique identification. We see this in the opening quotation – “It is better to be unique than the best. Because, being the best makes you the number one, but being unique makes you the only one” – which, I have since been informed, is a WhatsApp forward. But be that as it may. In paragraph 55 – under the heading “summing up the scheme” – the Court observes:
The whole architecture of Aadhaar is devised to give unique identityto the citizens of this country. No doubt, a person can have various documents on the basis of which that individual can establish her identify. It may be in the form of a passport, Permanent Account Number (PAN) card, ration card and so on. For the purpose of enrolment itself number of documents are prescribed which an individual can produce on the basis of which Aadhaar card can be issued. Thus, such documents, in a way, are also proof of identity. However, there is a fundamental difference between the Aadhaar card as a mean of identity and other documents through which identity can be established. Enrolment for Aadhaar card also requires giving of demographic information as well as biometric information which is in the form
of iris and fingerprints. This process eliminates any chance of duplication…
Wherever there would be a second attempt for enrolling for Aadhaar and for this purpose same person gives his biometric information, it would immediately get matched with the same biometric information already in the system and the second request would stand rejected. It is for this reason the Aadhaar card is known as Unique Identification (UID). Such an identity is unparalleled. (paragraph 55)
Notice carefully the language here. It is not that biometric authentication is more accurate or simply better than other forms of identification. Rather, it is perfect: it is unique, it “eliminates” any chance of duplication, and is “unparalleled.” As I shall go on to show, this factual assumption – of “unparalleled” uniqueness – is repeated several times in the judgment, and forms the backbone of the Majority’s reasoning.
It is also important to note – without further comment – that this issue was contested (and not, as the Majority makes it out to be, a given). The issue of false positives and false negatives were raised, and the existence of a duplicates was demonstrated as well (as a perusal of the oral record will show).
This factual assumption of uniqueness ties in with the Majority’s view that Aadhaar is a document of “empowerment.” In the next paragraph, the Majority notes that the uniqueness of the ID ensures both that subsidies are delivered in a targeted manner, and also that “when an individual knows that no other person can clone her, it assumes greater significance.” (paragraph 56) This, in turn, assumes significance in the Majority’s assessment of the constitutional validity of Section 7, which makes Aadhaar mandatory for availing subsidies.
The Majority begins its first substantive analysis of the constitutional challenge at page 219 of its judgment (paragraph 129). This is on the issue of surveillance and data protection (which, for some reason, are discussed together). At the outset, it is important to draw a distinction between two kinds of arguments that could be used to uphold the Aadhaar programme against a surveillance-based challenge. The first could be to agree that there is surveillance, but that it is constitutionally justified in this case. The second is to uphold the programme on the ground that there is no constitutionally significant surveillance. Interestingly, the Majority chooses to follow the second path: that is, this is not a judgment that defends and upholds surveillance, but one that avoids the legal question by making a factual finding that the Aadhaar project is incapable of being turned into a surveillance engine.
Surveillance: Factual Assumptions
Indeed, the Majority makes multiple factual findings on this score. First, it holds that the chain of authority that is responsible for collecting Aadhaar data is “secure” (paragraph 151) (interestingly, the Majority does not appear to address Mr Shyam Divan’s point that 49,000 such private operators were blacklisted by the UIDAI itself). Secondly, it holds that “minimal data” is collected (only iris and fingerprints), that UIDAI is purpose-blind, and – most importantly – that “merging of silos is prohibited” (paragraph 152) (there is no observation with respect to the demonstrated “merging of silos” in the State Resident Data Hubs). It goes on to hold that during authentication, “the nature of the transaction” is not known (paragraph 152), and there can be no “storage and replay of biometrics” (paragraph 152). The Majority then notes:
… we are of the view that it is very difficult to create profile of a person simply on the basis of biometric and demographic information stored in CIDR. (paragraph 153)
This is very important, because of two reasons. The first is, of course, that this is a sweeping factual assertion, and the correctness of the Majority opinion stands or falls with the validity of this (and many other) factual assumptions. But equally importantly, it also lays down a legal limit – that is, profiling issurveillance, and is constitutionally suspect. This has impact both for future constitutional challenges involving profiling, and for future cases involving the use of Aadhaar itself.
The Majority then discusses whether the Aadhaar programme is compatible with principles of data protection. An analysis of this section of the judgment is somewhat difficult, because the Majority quotes extensively from European and American precedent, refuses to lay down any standard, and then goes on to analyse the provisions of the Aadhaar Act anyway – but on what touchstone, it is unclear (this is a problem that plagues the Majority judgment at more than one place). In any case, the Majority holds that Aadhaar is compatible with “data minimisation” (by limiting itself to certain basic demographic and biometric information). This is potentially important for future challenges where the principle data minimisation is evidently not complied with.
However, the Majority also finds that on five counts, the provisions are not compatible with principles of data protection. The first is that of metadata storage. The Majority holds that Regulation 26 – which deals with storage of metadata – must be read down to include only “process metadata” (i.e., limited to the time of authentication, the requesting entity, and the yes/no answer), and nothing further (paragraph 202) (later in the judgment, the Majority refers to this as “metabase”). Secondly, the Majority also holds that the seven year storage period is excessive, and limits it to six months (paragraph 205). Thirdly, the Majority holds that Section 33 of the Aadhaar Act – which allows for an individual’s information to be shared on the orders of a district judge – must include a right to a hearing. Fourthly, it holds that Section 33(2) – which allows for sharing of information on “national security grounds”, a decision to be made by a Joint Secretary – is unconstitutional (however, this is only on procedural grounds, and can be rectified if a judicial member is added (para 349). And fifthly, Section 57 of the Aadhaar Act – which allows private parties to mandate Aadhaar – is unconstitutional to that extent. (paragraph 219)
Privacy, Section 7, and All That
This Section of the judgment is somewhat difficult to analyse, because the Majority appears to conflate a number of separate and distinct arguments into one unwieldy hotch-potch: privacy, dignity, Section 7 (i.e., compulsory authentication for availing of subsidies), Article 14-based arguments on exclusion, and Article 21-based arguments on exclusion. Let us, however, make an attempt.
After setting out the parties’ submissions, the Majority observes that “only those matters over which there would be a reasonable expectation of privacy are protected by Article 21.” (para 260) This will be discussed in more detail in a subsequent post, but I will pause here to note that it is a significant doctrinal error. The Majority relies upon Puttaswamy for this proposition, but Puttaswamy categorically rejected the “reasonable expectation of privacy” standard.
In any case, the Majority then applies the proportionality standard to the challenge. It takes the “legitimate State aim” as one of ensuring targeted delivery of subsidies to “deserving beneficiaries”, through accurate identification. Next, at the stage of ascertaining the rational nexus between the measure and the goal, the Majority – crucially – falls back upon its overarching factual assumption of uniqueness:
“… giving unique identity of each resident of the country is a special feature of this scheme, more so, when it comes with the feature stated above, namely, no person can have more than one Aadhaar number; Aadhaar number given to a particular person cannot be reassigned again to any individual even if that is cancelled and there is hardly any possibility to have fake identity.” (paragraph 277)
And this uniqueness is important because:
We have seen rampant corruption at various levels in implementation of benevolent and welfare schemes meant for different classes of persons. It has resulted in depriving the actual beneficiaries to receive those subsidies, benefits and services which get frittered away though on papers, it is shown that they are received by the persons for whom they are meant. There have been cases of duplicate and bogus ration cards, BPL cards, LPG connections etc. Some persons with multiple identities getting those benefits manifold. Aadhaar scheme has been successful, to a great extent, in curbing the aforesaid malpractices. (paragraph 278)
This, then, forms the basis of the “rational nexus” prong. However, there is something particularly disingenuous about the factual assumption smuggled in here. That becomes clear when you tack back a few hundred paragraphs, all the way back to paragraph 72, which is under an innocuous heading called “Introductory Remarks”. Here, the Majority notes that the Union of India’s claims of savings have been contested by the Petitioners. It then makes the following observation:
But the argument based on alleged inaccurate claims of savings by the Authority/Union of India in respect of certain programmes, like saving of USD 11 billion per annum due to the Aadhaar project, as well as savings in the implementation of the MGNREGA scheme, LPG subsidy, PDS savings need not detain us for long. Such rebuttals raised by the petitioners may have relevance insofar as working of the Act is concerned. That by itself cannot be a ground to invalidate the statute. (para 72)
We shall return to this point in subsequent posts, when we discuss the Majority’s strange tendency to dodge every argument that is inconvenient to its case. For now, what is important to note is this: in its overarching factual assumption about uniqueness, the Majority ignores the contrary factual arguments. However, in its factual assumptions about whether Section 7 meets the proportionality standard, the Majority does one better – it acknowledges that there exist contrary factual arguments, but then refuses to consider them altogether!
In fact, how much rests on this factual assumption is made clear when the Majority considers the third prong of the proportionality standard – that of “necessity”. Here, the Majority says:
Insofar as third component is concerned, most of it stands answered while in the discussion that has ensued in respect of component No. 1 and 2. The manner in which malpractices have been committed in the past leaves us to hold that apart from the system of unique identity in Aadhaar and authentication of the real beneficiaries, there is no alternative measure with lesser degree of limitation which can achieve the same purpose. In fact, on repeated query by this Court, even the petitioners could not suggest any such method. (paragraph 280)
What you have here, therefore, is a classic example of the assumed factual premise transforming itself into the conclusion. Because Aadhaar is unique (Factual Assumption No. 1) and because it is working to plug leakages with respect to targeting beneficiaries (Factual Assumption No. 2), therefore it is necessary, and the least restrictive alternative. On the first assumption, the contrary evidence is ignored, and on the second, its existence is acknowledged, but it is deemed irrelevant.
It should also be noted – for the record – that the Court’s statement that the “petitioners could not suggest any [alternative] method” is – to put it bluntly – quite simply false. “Alternative methods” – including smart cards – were argued in open Court (a look at the oral record will confirm this), and they form part of both the written submissions, as well as the rejoinder submissions, of Mr. K.V. Viswanathan.
This then leads the Court to the final – “balancing” – prong of proportionality. The Court decides to frame this in an interesting way. On the one hand – it observes – is the individual right to privacy. On the other hand, are the rights to food, health, shelter etc.:
Axiomatically both the rights are founded on human dignity. At the same time, in the given context, two facets are in conflict with each other. The question here would be, when a person seeks to get the benefits of welfare schemes to which she is entitled to as a part of right to live life with dignity, whether her sacrifice to the right to privacy, is so invasive that it creates imbalance? (paragraph 285)
The Court then repeats the “reasonable expectation of privacy” standard to find that the infringement on individual privacy is “minimal” (paragraph 295) (thus confusing “minimal information” and “minimal infringement”, which we shall discuss in our subsequent post on doctrine), whereas, on the other hand:
As already pointed out above, the Aadhaar Act truly seeks to secure to the poor and deprived persons an opportunity to live their life and exercise their liberty. By ensuring targeted delivery through digital identification, it not only provides them a nationally recognized identity but also attempts to ensure the delivery of benefits, service and subsidies with the aid of public exchequer/Consolidated Fund of India. National Security Food Act, 2013 passed by the Parliament seeks to address the issue of food, security at the household level. (paragraph 298)
It is, by now, probably tedious to point out that we have wheeled back to the two factual assumptions discussed above. Therefore:
As against the above larger public interest, the invasion into the privacy rights of these beneficiaries is minimal. By no means it can be said that it has disproportionate effect on the right holder. (paragraph 308)
I want to point out a rather important conceptual slip here. When the Court first frames its “balancing test” – see above – recall that it frames it as a clash between two rights: the right of an individual to privacy, and her right to food, healthcare etc. Towards the end, however, the Court has shifted from the language of a clash of rights, to a clash between “rights” and “public interest”. Why is this? The answer is very simple. If it remained a clash between two individual rights, then there was no justification for Section 7 making Aadhaar mandatory. The clash could just as easily be resolved by making it voluntary. The State cannot, after all, tell a person that she is being forced to give up her privacy so that she exercise her right to food – that, conceptually, makes no sense. What does make sense is to say that the individual must give up her right to privacy in the “larger public interest.” But when the argument ends, the Court returns to the language of clashing rights:
Thus, even when two aspects of the fundamental rights of the same individual, which appear to be in conflict with each other, is done, we find that the Aadhaar Act has struck a fair balance between the right of privacy of the individual with right to life of the same individual as a beneficiary. (paragraph 308)
These slips are not accidental. There is something particularly distasteful of saying openly to a person, “give up your right to privacy if you want food.” Even the Majority in this case cannot bring itself to say this. So what it does is this sleight of hand, shifting between “clash of rights” for rhetorical purchase, and slipping in “larger public interest”, which actually does the heavy lifting of upholding Section 7.
The Majority then comes to the specific argument of exclusion. Here, at paragraph 314, we find the following lines:
TO DICTATE FURTHER
Re. : Studies on exclusion
Re. : Finger prints of disabled, old persons etc. See other mode of identity
I think that this should be treated with the same respect that the Majority appears to have treated all of us, and will not analyse this any further. The only thing to note is that the Majority takes on record the Attorney-General’s submission that nobody will be denied food if authentication fails, and that a Circular has been passed to this effect. That, according to the Majority, “takes care of the problem.” (paragraph 315) This shall be discussed further in the doctrinal post.
It is, however, impossible not to note paragraph 317:
In fairness to the petitioners, it is worth mentioning that they have referred to the research carried out by some individuals and even NGOs which have been relied upon to demonstrate that there are number of instances leading to the exclusion i.e. the benefits are allegedly denied on the ground of failure of authentication. The respondents have refuted such studies. These become disputed question of facts. It will be difficult to invalidate provisions of Parliamentary legislations on the basis of such material, more particularly, when their credence has not been tested. (paragraph 317)
First, in 2018, after thirty years of PIL jurisprudence based upon newspaper articles, this is an astonishing statement. but secondly, it is important to note this, because in a previous part of the judgment, the Majority relies extensively on a “power-point presentation” that was made in Court by the Chairperson of UIDAI. It is a rather basic rule of trial that evidence submitted in court, that is not on affidavit, and that has not been put through the rigours of cross-examination, is no evidence at all. But for the Majority, “disputed questions of fact” cannot be adjudicated upon only when they lead to inconvenient legal conclusions. This is classic constitutionalism of convenience.
One final point: the Majority reads Section 7 narrowly to exclude benefits that are “earned” such as pensions, and all initiatives that are not “subsidies” such as school admissions (paragraph 322).
The Majority devotes a separate section to children, stating that their enrolment may take place only under parental consent, and that they have a right to opt-out when they attain majority. Crucially, the Majority also notes:
We also clarify that no child shall be denied benefit of any of these schemes if, for some reasons, she is not able to produce the Aadhaar number and the benefit shall be given by verifying the identity on the basis of any other documents. We may record that a statement to this effect was also made by Mr. K.K. Venugopal, learned Attorney General for India, at the Bar. (para 332)
The Majority then examines some of the specific provisions of the Act. Section 57 – as we have seen above – is struck down insofar as it allows private parties to make Aadhaar mandatory. There appears to be some confusion over the scope of this part of the ruling, with private players arguing that Parliament can authorise this through law. The judgment, however, is clear on this point:
That portion of Section 57 of the Aadhaar Act which enables body corporate and individual to seek authentication is held to be unconstitutional.
There is no case to be made, therefore, that Parliament can now allow this “by law.”
On Section 47, however – which only allows the UIDAI to file an FIR for a breach of the law – we have this priceless gem:
Notwithstanding the above, we are of the opinion that it would be in the fitness of things if Section 47 is amended by allowing individual/victim whose right is violated, to file a complaint and initiate the proceedings. We hope that this aspect shall be addressed at the appropriate level and if considered fit, Section 47 would be suitably amended. (paragraph 353)
Now, what on earth does this mean? Is Section 47 constitutional, or is it unconstitutional? What is “the fitness of things”? Should a Constitutional Court be concerned with expressing the “hope” that this aspect would be addressed “if considered fit”? Is the Court unaware that language matters?
The penultimate part of the judgment involves the analysis of making Aadhaar mandatory for bank accounts, phones, and for PAN Card and the payment of taxes. The analysis here follows the factual assumptions already outlined above, so I will not repeat in detail. Suffice it to say that the Court holds that Section 139AA of the Income Tax Act is proportionate since it helps preventing duplicate PANs and tax fraud, while bank linking and phone linking are unconstitutional, since in neither case has the State sufficiently justified what Aadhaar linking will achieve. In short, therefore, anyone who pays Income Tax must get Aadhaar. It is rather ironic to note, however, that the Court comes down hard on the government for “ritualistic incantation” of “black money” in the case of banks, but that “ritualistic incantation” of uniqueness and savings is what forms the heart of its decision to uphold Section 7.
The Majority judgment is 568 pages long. For all that, however, as I have tried to show, its basis is a series of factual assumptions. These factual assumptions are so crucial to the judgment, that – far more than any legal analysis or application of legal standards – the judgment has to stand or fall with them. But forget, for a moment, the question of whether those factual assumptions are correct or not. If there is one thing that characterises this Majority opinion on one of the most important constitutional questions of our generation (apart from WhatsApp opening lines and “TO BE DICTATED”), it is that on these foundational issues, the contrary arguments – far from being addressed adequately – have simply been ignored. That, perhaps, will be the enduring legacy of this judgment.