The Debasement of a Constitutional Institution: D.Y. Chandrachud's Dissent in K.S. Puttaswamy v. Union of India on Aadhaar, the Money Bill, and Bicameralism

On 26 September 2018 a five-judge Constitution Bench upheld the certification of the Aadhaar Act as a Money Bill under Article 110. Justice D.Y. Chandrachud dissented. A reading of the institutional axis of his 339-paragraph opinion: the judicial reviewability of the Speaker's certification, the substantive Article 110(1) test, and bicameral federalism as basic structure.

The Aadhaar Act of 2016 was introduced in the Lok Sabha as Bill No. 47 of 2016. It was certified by the Speaker as a Money Bill under Article 110 of the Constitution. Under the procedural regime that follows from such certification, the Bill could not be amended by the Rajya Sabha; the upper House had fourteen days to make recommendations which the Lok Sabha was free to accept or reject. The Bill, accordingly, became law without substantive consideration by the upper House of a bicameral legislature. The petitioners in K.S. Puttaswamy (Retd.) v. Union of India challenged this procedure as inconsistent with Article 110. The majority of the five-judge Constitution Bench, on 26 September 2018, upheld the certification. Justice D.Y. Chandrachud, writing alone in dissent, would have struck down the entire Aadhaar Act on this ground alone — before any consideration of its substantive provisions.

The proposition at the centre of the dissent’s procedural holding appears at sub-conclusion 2 of paragraph 339. The debasement of a constitutional institution cannot be countenanced by the Court. Democracy survives when constitutional institutions are vibrant.¹ The sentence is the dissent’s framing of what is at stake when an ordinary Bill is improperly certified as a Money Bill: not merely a procedural irregularity but a constitutional damage to the bicameral structure the framers embedded in the legislative process. The institution that suffers the debasement is the Rajya Sabha. The injury is to federal bicameralism, which the dissent holds to be part of the basic structure of the Constitution.

The Money Bill challenge is one of three institutional axes along which the dissent organises its analysis. The second is the rule-of-law section on the Government’s defiance of the Court’s interim orders between 2013 and 2017. The third is the substantive analysis of the Aadhaar architecture against the proportionality test laid down in Puttaswamy (I) (2017). Each axis is independently sufficient to support the dissent’s conclusion that the Aadhaar Act and project are unconstitutional. Read together, they form the framework of an institutional critique that operates at the level of constitutional design.

The procedural question and the architecture of Article 110

Article 110(1) defines a Money Bill as one that contains “only provisions” dealing with all or any of seven specified matters: taxation, government borrowing, the custody of the Consolidated and Contingency Funds, the appropriation of moneys, the declaration of charged expenditure, the receipt and audit of public moneys, and matters incidental to these. Article 110(3) provides that where any question arises as to whether a Bill is a Money Bill, the decision of the Speaker of the Lok Sabha shall be final.

The Attorney General’s position before the bench was straightforward. The phrase “shall be final”, on this argument, excluded judicial review. The Speaker’s certification was a parliamentary act, protected from judicial interference both by the textual finality of Article 110(3) and by the institutional immunity of parliamentary proceedings under Article 122. The petitioners’ contention that the Aadhaar Act extended beyond the sub-clauses (a) to (g) of Article 110(1) was, on the Attorney General’s reading, not a matter the Court could entertain.

The dissent rejects this reading in three substantive moves. The first concerns the textual meaning of “shall be final.” The dissent reads the finality clause as directed at two specific concerns: settling the question of certification as between the two Houses of Parliament, and settling it as between Parliament and the President. The clause does not, on the dissent’s reading, address the separate question of judicial review where a substantive constitutional violation is alleged. The same phrase appears elsewhere in the Constitution — Articles 103, 105, and 217 — and the dissent surveys these to demonstrate that “finality” in the constitutional vocabulary settles internal institutional questions without precluding judicial review on substantive constitutional grounds.

The second move concerns the scope of Article 122’s immunity for parliamentary proceedings. The dissent reads Raja Ram Pal v. Hon’ble Speaker, Lok Sabha (2007) and the Special Reference No. 1 of 1964 jurisprudence as having settled the proposition that Article 122’s immunity extends to irregularity of procedure but not to substantive illegality or constitutional violation. The Speaker’s certification of a Bill as a Money Bill, where the Bill in fact contains provisions outside sub-clauses (a) to (g) of Article 110(1), is not a mere irregularity. It is a substantive constitutional violation that places the Bill beyond the procedure the Constitution prescribed for ordinary legislation.

The third move is the overruling of Mohd Saeed Siddiqui v. State of U.P. (2014) and Yogendra Kumar Jaiswal v. State of Bihar (2016). The earlier three-judge bench in Mohd Saeed Siddiqui had read Article 212 (the state-legislature analogue of Article 122) together with Article 255 as excluding judicial review of a Money Bill certification. The dissent treats this as an error: Article 255 deals with the requirement of prior sanction or recommendation and has no application to the question of Money Bill certification. The two-judge bench in Yogendra Kumar Jaiswal had followed Mohd Saeed Siddiqui without independent analysis. Both decisions are overruled in the dissent’s account.²

The substantive question: does the Aadhaar Act fit Article 110(1)?

Having established that the Speaker’s certification is judicially reviewable, the dissent turns to the substantive Article 110(1) test. The test, as Article 110(1) makes explicit by its use of the word “only”, is restrictive. A Bill containing both provisions that fall within sub-clauses (a) to (g) and provisions that fall outside them is not a Money Bill. The dissent’s analysis is provision by provision.

Section 7 of the Aadhaar Act — the provision linking subsidies, benefits, and services to Aadhaar authentication — is the only provision the Union of India seriously argued fell within sub-clause (e) of Article 110(1), which deals with the declaration of expenditure charged on the Consolidated Fund. The dissent reads Section 7 narrowly. The provision does not declare any expenditure to be charged on the Consolidated Fund. It provides that, in the case of services, benefits, or subsidies for which expenses are incurred from the Consolidated Fund, the receipt of an Aadhaar number may be made a condition. The receipt-of-Aadhaar condition is not itself a declaration of expenditure. On the dissent’s reading, Section 7 does not satisfy sub-clause (e).

The dissent then walks through the other provisions of the Act. The framework for enrolment under Sections 3 and 8. The collection of demographic and biometric information under Section 2(g), (j), (k) and (t). The creation of the Unique Identification Authority of India as a statutory body. The procedures for authentication. The disclosure regime under Section 33. The cognisance bar under Section 47. The permission for private use under Section 57. The retrospective validation under Section 59. None of these provisions, the dissent observes, falls within any of the sub-clauses (a) to (g) of Article 110(1).

The Union of India’s alternative argument was based on the doctrine of pith and substance. The submission was that even if individual provisions of the Aadhaar Act fall outside Article 110(1), the Act in its essential character is a Money Bill because its dominant purpose is the targeted delivery of subsidies funded from the Consolidated Fund. The dissent rejects this submission as a misapplication of the doctrine. The doctrine of pith and substance, the dissent observes, operates in the field of legislative competence — the question of which legislature has the authority to enact a law on a given subject as between the three Lists of the Seventh Schedule. The doctrine has no application to Article 110, which uses the word “only” to make exhaustive what would otherwise be inclusive. The dissent’s conclusion is that the Aadhaar Act fails the Article 110(1) test and could not validly have been certified as a Money Bill.

Why the Money Bill question is a basic-structure question

The dissent’s argument does not stop at the textual reading of Article 110. The third movement of Section E is the proposition that the bicameral structure of Parliament is part of the basic structure of the Constitution. The argument runs as follows. The framers chose bicameralism deliberately. The Rajya Sabha is not a procedural appendage to the Lok Sabha; it is an institution of federal representation, with members chosen by the legislative assemblies of the States and with a constitutional role in the legislative process distinct from that of the Lok Sabha. The Money Bill procedure under Articles 109 and 110 is a carefully circumscribed exception to the ordinary rule of bicameralism: the Rajya Sabha’s role is limited because the subject-matter of Money Bills is closely tied to the financial responsibility that the lower House bears as the popularly elected chamber.

The improper certification of a non-Money Bill as a Money Bill, the dissent argues, removes the Rajya Sabha’s constitutional role from a piece of legislation to which that role properly attaches. This is not merely a procedural irregularity; it is a denudation of the upper House’s constitutional function. The dissent’s position is that the cumulative effect of such denudations, if left without judicial remedy, would alter the basic constitutional architecture of bicameral federalism. The Court’s power to review the Speaker’s certification is therefore not merely a power of correction in a particular case; it is a structural safeguard against the displacement of one of the constitutional institutions whose vibrancy the dissent treats as essential to constitutional democracy.

The rule of law and what the Government did with the interim orders

Section K of the dissent addresses what is, in institutional terms, a separate constitutional concern. The Court had repeatedly ordered, in interim proceedings between 2013 and 2017, that Aadhaar enrolment would remain voluntary and that no person could be denied any service or benefit for want of an Aadhaar number. The interim orders had been issued on 23 September 2013, 11 August 2015, 15 October 2015, 14 September 2016, and 27 March 2017 — a sustained line of judicial intervention recognising that the constitutional question was unresolved and protecting individuals from coercive linkage in the interim.

The Government’s response, the dissent records, did not honour these orders. Between 2013 and 2018, successive notifications and statutory provisions had made Aadhaar mandatory for bank accounts, mobile SIM cards, school admissions, foodgrain rations, pension disbursements, scholarships, and a host of other engagements with the State. Section 139AA of the Income Tax Act, introduced in 2017, made Aadhaar mandatory for filing income-tax returns and for obtaining a PAN card. The 2017 amendments to the Prevention of Money Laundering Rules required Aadhaar linkage with all account-based relationships.

The dissent’s reading is that this pattern of executive and legislative action defied the spirit of the Court’s interim orders. The constitutional duty that the dissent identifies is that the Government, faced with interim orders that bind it, must either comply or seek variation. To legislate around the orders without seeking variation, the dissent argues, is to test the institutional foundations of the rule of law. Defiance of judicial orders (both interim and final) be it by the government or by citizens negates the basis of the rule of law. Institutions of governance, the dissent observes, are bound by a sense of constitutional morality which requires them to abide by judicial orders.

The rule-of-law section connects to the Money Bill section through a shared institutional concern. In both, the dissent identifies a pattern of executive and legislative conduct that operates around the constitutional procedures and judicial supervision the framers prescribed. The Money Bill certification removes the upper House from the legislative process; the defiance of interim orders removes the Court’s supervisory role from the period before final adjudication. Both are, in the dissent’s reading, debasements of constitutional institutions — forms of constitutional damage that the Court has a duty to recognise and resist.³

The parallel ground: privacy, biometrics, and what the architecture does

The dissent does not rest on the procedural ground alone. Section F of the opinion develops the substantive privacy and dignity analysis. The starting point is that biometric data is constitutionally distinct from other categories of personal information. The dissent’s formulation at sub-conclusion 7 is the doctrinal anchor:

While citizens have privacy interests in personal or private information collected about them, the unique nature of biometric data distinguishes it from other personal data, compounding concerns regarding privacy protections safeguarding biometric information. Once a biometric system is compromised, it is compromised forever. Therefore, it is imperative that concerns about protecting privacy must be addressed while developing a biometric system.D.Y. Chandrachud J., sub-conclusion 7 of para 339

From this premise the dissent draws specific consequences. The standards for collection, retention, and use of biometric data must be substantially more demanding than for ordinary personal data. The Aadhaar Act contains none of these heightened protections in adequate form: no defined options for those who decline authentication, no robust informed-consent regime at enrolment, no individual right to access one’s own biometric record under Section 28(5), no independent regulatory oversight, no graduated data-minimisation principles, and a foreign-source-code dependency for the de-duplication technology at the heart of the architecture.

Applying the proportionality test from Puttaswamy (I), the dissent accepts the legitimate state aim of targeted welfare delivery but holds that the means chosen fail the necessity prong (less intrusive alternatives were available) and the balancing prong (the impact on privacy, dignity, autonomy, and informational self-determination outweighs the welfare-delivery benefit, particularly given documented authentication failures leading to exclusion of eligible beneficiaries). Section 7’s overbreadth, Section 47’s cognisance bar, Section 57’s permission for private use, Section 33’s disclosure regime, and Section 59’s retrospective validation are each held to be independently unconstitutional.⁴

Identification, identity, and the closing claim

The dissent’s philosophical close, at sub-conclusion 21 of paragraph 339, completes the institutional analysis with an ontological one. Identity is necessarily a plural concept. The Constitution also recognizes a multitude of identities through the plethora of rights that it safeguards. The technology deployed in the Aadhaar scheme reduces different constitutional identities into a single identity of a 12-digit number. The distinction the dissent draws is between identification, which is the technical-administrative function of verifying that the person presenting themselves is the person to whom the State assigned a number, and identity, which is the constitutional concept of how a person is recognised in law across the plurality of their religious, linguistic, cultural, occupational, associational, and expressive engagements with the polity. The dissent’s objection is that an architecture which collapses constitutional identity into administrative identification, and which conditions practical citizenship on the production of that identification, alters the constitutional position of the individual against the State at the conceptual level.

The disposition at sub-conclusion 23 follows. The Aadhaar Act, the Rules and Regulations under it, and the pre-Act framework are declared unconstitutional. Under Article 142, the data already collected is to be preserved for one year. If Parliament fails to enact a fresh, conforming statute within that period, the data is to be destroyed.⁵

What the dissent records

The majority’s reading of the Aadhaar Act survived. The Act stands, with the reading-downs the majority introduced. The Money Bill certification was upheld, though the Money Bill question itself was subsequently referred by a different bench in Rojer Mathew v. South Indian Bank (2019) to a larger bench, which is currently constituted and pending. The institutional question the dissent raised is therefore part of the live constitutional record — not settled by Puttaswamy (Aadhaar), but open to a future bench whose composition is yet to be announced.⁶

What the dissent puts on that record is a worked institutional analysis with three components. The first is the constitutional case for judicial review of the Speaker’s certification under Article 110(3). The argument is textual (the meaning of “shall be final”), doctrinal (the limits of Article 122 immunity), and structural (the basic-structure implication of bicameral federalism). The second is the substantive Article 110(1) reading of which provisions can and cannot form part of a Money Bill, with the operative principle that the word “only” makes the sub-clauses exhaustive rather than inclusive. The third is the rule-of-law section’s framing of constitutional morality: that institutions of governance, including the executive and the legislature, are bound to respect both interim and final judicial orders, and that legislating around interim orders without seeking their variation is itself a constitutional concern.

The dissent’s closing sentence at the end of Section A had placed the institutional stakes at their highest. The case, the dissent had written, would determine the future of freedom. The framing was not rhetorical. The institutional architecture the dissent defends — bicameralism, judicial review of parliamentary procedure, the binding force of interim orders, the heightened standards required for biometric data, and the plural conception of constitutional identity — is the architecture within which the relationship between the State and the citizen operates. To debase any of its components is, on the dissent’s reading, to damage the constitutional fabric. The 339 paragraphs of the dissent are the substantive record of why.