Reading Article 15: Non-Discrimination and the Question of Inter-sectionality

Article 15(1) of the Constitution states:

The State shall not discriminate against any citizen on grounds only of religion, race, caste, sex, place of birth or any of them.”

Now consider the following hypothetical: a State instrumentality refuses to hire only Dalit women. It hires both non-Dalit women and Dalit men on parity with everyone else. So, it does not discriminate on the basis of caste (it hires Dalits) or sex (it hires women), but upon a combination of both of them. Is there an Article 15 violation?

This issue is commonly known as intersectionality. Intersectionality studies the way in which commonly constructed categories like race, sex, religion (precisely the categories listed in Article 15) do not reflect isolated, hermetically-sealed and individuated systems of dominance and suppression, but invariably tend to overlap and interact. Accounts of power-hierarchies therefore, that focus on a single axis (e.g., racism, or casteism) are inevitably incomplete.

The manner in which this incompleteness affects legal analysis comes to the fore in the American case of De Graffenreid v. General Motorsa decision of the Eighth Circuit Court of Appeals. Prior to the passage of the Civil Rights Act, 1964, General Motors did not hire black women. After the 1964 Act, it was compelled to. In 1970, during a recession, it initiated a series of seniority-based layoffs, and consequently, the black women who had been hired in 1964 lost their jobs. Their discrimination claim was rejected by the District Court a decision that was, in substance, upheld by the Appeals Court (we need not go into certain technical procedural issues here), the crucial observation being:

“… this lawsuit must be examined to see if it states a cause of action for race discrimination, sex discrimination, or alternatively either, but not a combination of both.”

The problem with this argument, as Kimberle Crenshaw points out in this excellent article, is that it implies that “the boundaries of sex and race discrimination doctrine are defined respectively by white women’s and Black men’s experiences. Under this view, Black women are protected only to the extent that their experiences coincide with those of either of the two groups.'”

That brings us to the question of how to read Article 15(1). The article prohibits discrimination on grounds only of… [x, y and z]… or any of them. The key interpretive issue here is the role of the phrase “any of them” as adding to – or qualifying – “only”. According to judicial decisions, “any of them” ensures that the State cannot get around its obligations by discriminating on the basis of a prohibited category combined with a non-prohibited category. For example, in my organisation, I have a promotion rule based on seniority, but I exclude women from it. I am thus discriminating on the basis of both sex and seniority. The former is a prohibited category. The latter is not. So, I am not discriminating only on the basis of sex, but I am discriminating on the basis of sex nonetheless, and fall within the ambit of Article 15(1).

This is a straightforward case; the problem arises precisely in intersectionality claims, however, because in the Dalit-woman case we discussed above, I am neither discriminating on the basis of caste or on the basis of sex, but an amalgamation of the two – and because of the amalgamation; in other words, my specifically-tailored discrimination towards Dalit women ensures that – unlike the case of sex + seniority – I escape the textual ambit of the statute by escaping the prohibited categories altogether.

Two posts ago, we discussed how, in the context of Naz Foundation, reading sexual orientation into “sex” was the only way to honour the Constitution’s core commitment to the respect and the autonomy principles: the first, in particular, mandated that no discrimination was permitted on the basis of those defining labels that a person is born with, and that constitute her public (and private) identity. The textual reading posited above violates this commitment, and for that reason it must be rejected, notwithstanding precedent to the contrary.

What is the alternative? It is to read the term “or any of them” as including a prohibition on intersectional discrimination. This might appear linguistically strained, at first sight, but surely it is – to use a term coined by the UK Supreme Court – an “intellectually defensible” reading. In other words, “or any of them” is used after “only” to mitigate two situations: discrimination on the basis of a prohibited category combined with a non-prohibited category, which therefore escapes the “only” prohibition; and simultaneous discrimination on the basis of two prohibited categories that also therefore escapes the “only” prohibition, albeit for very different reasons.

This argument, of course, is based entirely upon the text and structure of the Constitution. I have not gone into Article 15 case law, primarily because – for the reasons adduced above – I think that the text and structure provide a satisfactory answer to what might appear – at first blush – the problem of intertextuality arising out of a plain reading of the text.

This might appear to be a mere semantic quibble; no doubt, if a case of this sort was to actually come before the Court, it would hold that an Article 15 discrimination case was made out, textually or not. Nonetheless, I think it reveals an interesting facet about our intuitions about discrimination, that are reflected in our constitutional text – we tend to think of discrimination as taking place through individual a priori categories (race, class, sex). Intersectionality tells us that this reflexive belief might exclude a set of extremely important experiences of discrimination; and that therefore, this deeper and richer sense of what discrimination is ought to be reflected not only in our changed attitudes, but in our laws as well.

PUCL v Union of India: The Supreme Court and Negative Voting

(This is an immediate response to the judgment; I hope to have a couple of guest posts on this in the coming week)

Yesterday, the Supreme Court, in PUCL v. Union of India, upheld the constitutional right of citizens to cast a negative vote in elections. The exact political ramifications of this judgment will probably become clear once the Election Commission comes up with the modalities of its implementation; from a constitutional point of view, this judgment is important as it further underscores the centrality of Article 19(1)(a) to election disputes.

In PUCL v. Union of India, the constitutional validity of Rules 41(2), (3) and 49-O of the Conduct of Election Rules, 1961, was impugned. Both sides agreed on the fact that thecombined effect of these rules was that persons who did not vote in elections were recorded (by the presiding officer) as having not voted. The petitioners argued that this was a violation of the right to secret balloting, protected by Articles 19(1)(a) and 21 of the Constitution.

Relying upon Kuldip Nayar v. Union of Indiathe State raised a preliminary objection on the ground that since voting was not a fundamental or constitutional right, but only a “statutory right” brought into existence by the Representation of Peoples Act, this wasn’t an Article 32 fundamental rights petition in the first place. Rejecting this contention, the Court distinguished between the “right to vote” and the “freedom of voting as a species of the freedom of expression“. [Paragraph 19] This is what explained the Court’s earlier decisions in PUCL v. Union of India and Association for Democratic Reforms v. Union of India, where  the right to know the antecedents of politicians had been brought within the ambit of Article 19(1)(a) as part of the “right to know”. [Paragraph 20]

The Court then found that in a system of direct elections, secrecy was essential in order to ensure the effectiveness of the vote – that is, elections could not be free and fair unless secrecy was maintained (that much is logical; naturally, the only way of preventing bribery, coercion and post-election reprisals is secrecy).  Further, since the freedom to vote naturally included the freedom not to vote, it would be arbitrary to extend secrecy to one and not the other. [Paragraph 31]  It buttressed its argument by invoking Indira Nehru Gandhi v. Raj Narain and Kihoto Hollohan v. Zachilhu for the unexceptionable proposition that an effective democracy functioning through periodic fair and free elections is part of the basic structure of the Constitution. [Paragraph 45] In addition, the act of not voting was as much a positive exercise of free expression under Article 19(1)(a) as was voting itself, and so deserved similar levels of protection [Paragraph 49].  The Court therefore directed the Election Commission to introduce a “None of the Above [NOTA]” option into the Electronic Voting Machines. [Paragraph 61]

The judgment of the Court is important, as it clarifies the constitutional status of voting. What does it mean to say that the right to vote is only statutory, but the act of voting is an exercise of free speech protected by Article 19(1)(a)? Only this: the right to vote is statutory insofar as the modalities of voting are regulated by statute (the Representation of Peoples Act); questions of who can vote, when and in what manner, what restrictions and rules political parties must abide by, and so on – these are matters undoubtedly determined by statute, and subject to the control of the legislature. The act of voting, however, is – at least in theory – the most important act of expression through which the citizen participates in a representative democracy. So while the right to vote remains only a statutory right, parliament nonetheless may not erect any formal or substantial barriers that render voting ineffective or nugatory. Consider the following hypothetical: parliament tomorrow repeals the Representation of Peoples Act, and substitutes it… with nothing (effectively creating itself as a permanent oligarchy). Can one seriously suggest that that act would be constitutional? It is in this sense that the freedom to vote is – in its abstract sense – a constitutional and a fundamental right, the contours and lineations of which are to be worked out by parliament through statute.

This conclusion follows inexorably from Article 19(1)(a) and from the structure of the Constitution itself. On this blog, we have often discussed how, in a series of cases, the Court has located the philosophy of Indian free speech in a functioning liberal democracy where speech plays the important role of raising political awareness, communicating views to government, and so on; would this mean anything if the basic mechanism that defines a representative democracy – periodic change in government through elections – was compromised or made ineffective (for a similar argument deriving the right to free speech from the right to petition government for grievances, see Charles Black, Structure and Relationship in Constitutional Law)? In addition, Part XV  of the Constitution is devoted entirely to the conduct of elections, including non-discrimination rules (Article 325); and Article 326 expressly states that elections to the House and the Assemblies are to be on the basis of adult suffrage; what would be the point of all this if the government was free to abolish elections altogether? And lastly, representative democracy – as discussed above – is a basic feature of the Constitution; potentially, the government might someday come up with an alternative way of doing representative democracy that dispenses with elections. Until that time, however, effective elections – and consequently, an effective vote – remain a constitutional right and a fundamental right.

It is submitted, therefore, that the Supreme Court’s decision is correct, and that it’s analysis of the relationship between elections, the right to vote, and Article 19(1)(a) is substantially correct as well. One minor objection might be made: it was enough for the Court to have said – as, in substance, it did – that the freedom envisioned by Article 19(1)(a) is the freedom to have a say in government through the mechanism of the vote, and that naturally must include the option to not vote as well; but it was a mistake to further ground the right by referring to potential negative consequences of revealed identity. It is easy enough to imagine how, if one’s vote is revealed after an election, the winning (or losing) party might exact reprisals – which, in turn, will affect how one votes in the election. Yet it is difficult to see how by not voting at all (for anyone), and having that fact as public knowledge, would lead to repression or reprisal. That remains, however, a minor quibble.

And lastly, as a closing aside, it is worthwhile to note while the challenge was made under Articles 19(1)(a) and 21, the final holding turned only upon Article 19(1)(a) [see paragraph 61]. In an age where it is often said that Article 21 has been transformed into a bottomless receptacle for the judiciary to pour in its own ideas of the right and the good, the Court’s conscious resistance to Article 21’s siren call is to be commended.

Textualism vs Structural Analysis – or why the Court ought to Uphold Naz Foundation

In the previous post, we noted that the concept of the basic structure, in both India and Germany, takes its shape and form not from any one or multiple clauses of the Constitution, but from “overarching principles” that explain and justify the Constitution as a whole. This method of analysis, which we may call “structural analysis” (as opposed to “textual analysis”) has received some amount of judicial and scholarly attention in the United States.  In his dissenting opinion in Panama Refining Co v. Ryan, Justice Cardozo observed:

“… the meaning of a statute is to be looked for, not in any single section, but in all the parts together and in their relation to the end in view.”

One year later, in Duparquet Co v. Evans, this time writing for the Majority, he added:

“There is need to keep in view also the structure of the statute, and the relation, physical and logical, between its several parts.” 

This theme was taken up by the famous constitutional scholar, Charles Black (who, incidentally, authored the petitioners’ briefs in Brown v Board of Education). In a series of three lectures, brought together into a book titled Structure and Relationship in Constitutional Law, Black developed the basic thesis that constitutional analysis involves “inference from the structures and relationships created by the Constitution in all its parts or in some principal part.” In the book, he discussed a number of hypotheticals in which famous American cases like McCulloch v. MarylandCrandall v. Nevada, and Gitlow v. New York  would – he argued – have been decided in the same way even if the specific textual provisions that they relied upon did not exist – simply because of the inexorable logic of constitutional structure. In Crandall v. Nevada, for instance, Nevada’s imposition of a one-dollar tax upon leaving the State was held unconstitutional on the ground that the American constitutional structure, from the Preamble to the Supremacy Clause, clearly envisaged a unified people living in a unified nation, for which reason travel between states was a question of right, not of privilege.

This method is, of course, the exact antithesis of textualism. Textualism focuses on explicating the precise meaning of individual words at issue in important constitutional cases. It is a method championed by the American Supreme Court justice Antonin Scalia through various opinions from the bench, and in his book, A Matter of Interpretation. Broadly, Justice Scalia makes two arguments for preferring textualism over structural analysis: first, that structure affords much greater leeway for judicial discretion than does textualism; and secondly, textualism – in particular, Scalian textualism – that focuses on the original public meaning of disputed words – privileges the popular conception of these words, and is thus more faithful to democracy.

As to the first, the factual assertion is suspect, to say the least. District of Columbia v. Hellera case about the American Second Amendment right to bear arms, provides a classic example of textualism’s own interpretive uncertainties: in that case, using unimpeachable textualist methodology, marshaling a vast array of historical sources (much of which overlapped), Justices Scalia and Stevens nonetheless managed to come to exactly the opposite conclusion about the scope of the Second Amendment. The broader point – as Dworkin argues repeatedly, and as Professor William Eskridge examines in some detail here – is that it is a mistake to think that “meaning” exists external to and beyond the interpreter, that it simply exists to be discovered by an impartial interpreter; a more accurate way of understanding meaning is to acknowledge, in the spirit of hermeneutics, that it is constructed by interpreters who bring to the enquiry their own set of fore-understandings. Once this is acknowledged, the idea that texts operate as passive depositories of meaning that by virtue of themselves, limit and constrain the interpretive enquiry, dissolves.

Further, it is a popular – yet invidious – mistake – to equate structural analysis to a free-wheeling moral enquiry that “ignores text”; no structuralist would deny that the constitutional text is the point of departure, that it informs any constitutional analysis – and indeed, structure itself supervenes over text; as Black himself stated:

… the structure and relations concerned are themselves created by the text, and inference drawn from them must surely be controlled by the text.”

Justice Scalia’s second argument – about democracy – holds even less water. As Professor Akhil Amar points out, the American Constitution was not ratified clause-by-clause, but as a whole – the decision was essentially in the form of “take-it-or-leave-it”. In such a scenario, it would actually be more faithful to popular democracy to interpret the Constitution as a coherent whole, with individual clauses taking their meaning from a sense of the whole, rather than the other way around.

In India, of course, the Constitution was never ratified, so the Scalia-Amar disputation is perhaps academic. Nonetheless, the speeches in the Assembly Debates (for instance, Nehru’s famous Objectives Resolution) do seem to reveal that our framers were well aware that they were drafting a document animated by certain fundamental purposes, and that individual clauses were designed to fulfill those purposes. To this we can add Dworkin’s argument, discussed in the last post, that principled consistency in law-making is an essential pre-requisite for political legitimacy.

These scattered observations are not meant – in any way – to serve as a full-blown defence of structural analysis in the Indian constitutional context; they are meant only to serve as a point of departure, and at the very least, establish it as a credible (and perhaps intuitively more desirable) alternative to textualism (and what often comes to be – but by no means necessarily is – its corollary, originalism). Let us now consider what structural analysis would  look like in practice.

As we all know, the Delhi High Court in Naz Foundation v. NCT decriminalised homosexuality, holding that insofar as S. 377 of the Indian Penal Code made sexual intercourse between consenting adults a crime, it was unconstitutional. This decision was appealed, heard by a two-judge bench in 2012, and the decision is due within the next three months. The Delhi High Court grounded its judgment in many constitutional provisions; specifically, let us consider its holding on Art. 15(1), which states:

“The State shall not discriminate against any citizen on grounds only of religion, race, caste, sex, place of birth or any of them.”

The Court held that the word “sex” includes “sexual orientation”, and that therefore, discrimination on the basis of sexual orientation is impermissible.

Did the “public meaning” of the word “sex” in 1950 include sexual orientation? Perhaps not – that is, if the question makes any sense in a country where the majority didn’t even speak English. Did Nehru, Ambedkar, Patel and the rest contemplate that they were prohibiting discrimination on the basis of sexual orientation? Probably not – but then again, they probably didn’t think they were permitting it either – in all likelihood, the issue didn’t cross their minds. What would they have said if the question had been put to them? We don’t know – perhaps they would have gone with the dominant prevailing opinion that homosexuality was a form of disease (but perhaps not); what would they conceivably say if we could bring them back from the grave, put copies of the Wolfenden Committee Report, reams of medical evidence and the lead opinion of Lawrence v Texas before them – and ask them for their opinion? Again, we don’t know.  Textualism, originalism, original intent – these theories simply give us little to no purchase on the issue. Yet even if they did, their relevance would be limited at best – because as we have argued, the enquiry is not about determining the most accurate possible meaning of a given text.

So much for what we don’t know. Here is what we do know: Article 15 prohibits discrimination on a number of grounds: religion, race, caste, sex and place of birth. With the possible and partial exception of religion, what unites these features is that they are all essential aspects of any individual’s private and public identity (by public personality, I mean a series of labels used by the society to identify her and differentiate her from others) that she is born into and is powerless to choose or change. To this we can add Article 16(2) (prohibition of discrimination in employment on similar categories); Article 17 (prohibition of untouchability – discrimination on the basis of birth); and Article 18 (abolition of titles – advantages (a form of discrimination), normally on the basis of birth). Let us – summarily – call this the “non-disrcimination principle”.

Now let us consider Articles 19 and 25. Article 19 guarantees the freedom of speech and expression, assembly, association, movement, residence and profession; each of these freedoms, it can hardly be disputed, are fundamental for two reasons: first, they are essential expressions of individual (and, for that matter, communitarian) personality; in the words of Justice Kennedy in Planned Parenthood v Casey:

“At the heart of liberty is the right to define one’s own concept of existence, of meaning, of the universe, and of the mystery of human life…. people have organized intimate relationships and made choices that define their views of themselves and their places in society.”

Secondly, as Dworkin puts it, these freedoms ensure that every citizen of the polity is provided the opportunity to contribute towards shaping the moral, cultural and political environment that she finds herself in – and that that, in turn, is the very essence of government according equal respect and concern to all its citizens. To this we add Article 25, that guarantees the freedom of conscience and religion (and further, the entire scheme from Articles 26 to 30); and indeed, arguable the two most important freedoms within this set (speech and conscience) are not limited by public interest concerns. Let us summarily call this the “autonomy principle”.

We are now in a position to understand why not only the Delhi High Court’s reading of “sexual orientation” into “sex” was not only correct, but the only possible correct decision. Our Constitution is structurally committed to a two-pronged principled attitude towards individuals: freedom in those matters that are related to the most fundamental expression of one’s humanity and personality (autonomy principle); and no discrimination on the basis of aspects of private and public identity that a person is born with and into (non-discrimination principle). And these principles stem not from any one provision, but a combination of Articles 14, 15, 16, 17, 18, 19, 25 and 26 to 30.

Can anyone seriously deny that sexuality is integrally – and centrally – consistent with both these principles? Constitutionally, therefore, there is no warrant for the Supreme Court to interfere with the judgment of the Delhi High Court; a contrary opinion would imply that our Constitution is committed to the non-discrimination and respect principles (as discussed above) – but in an entirely insupportable, capricious, arbitrary and unprincipled fashion, withholds that commitment from homosexuals. That certainly cannot be the Constitution we live under, or the Constitution to which we owe our allegiance.

The Basic Structure Doctrine: Notes from Germany

The basic structure doctrine is too well-known to merit detailed explanation: briefly, ever since Kesavananda Bharati v. State of Kerala, it has been judicially determined that the power of Parliament to amend the Constitution does not extend to the power to “damage or destroy” its basic structure. A corollary of this argument is that certain aspects of the Constitution are unchangeable, immutable and so bound up with the fabric of the Constitution itself, that as long as the Constitution exists, they too must necessarily exist. While the Court has steadfastly refused to provide an exhaustive list of unamendable “basic feature”, it has variously held – inter alia – democracy, republicanism, secularism and judicial review to be part of the basic structure. At the risk of gross oversimplification, let us note the following important implications:

(1) India now follows the doctrine of strict constitutional supremacy, where even a supermajoritarian amendment procedure (the closest analogue to a constituent assembly) is limited in its power to change the Constitution.

(2) The basic structure is not to be found in any one provision, or combination of provisions, but is found in abstract principles that explain and justify the architecture of the Constitution as a whole (for example, a combination of Articles 14, 15, 19, 25, 26, 29 and 30) would, together, constitute the principle of secularism that cannot be reduced to either one or many of them in conjunction. As the name itself suggests, you arrive at basic features not through an examination of individual constitutional clauses, but through a structural analysis.

(3) The constitutional status of amendments is subordinate to the text of the original Constitution, because no part of the original Constitution can be subject to a basic structure challenge.

(4) The basic structure doctrine presupposes a certain unity and principled integrity – if only as a legal fiction –  that binds and holds together the original Constitution as a single, coherent document.

We can best evaluate some of these assumptions by examining another constitutional democracy that has a basic structure doctrine as well, but with a few interesting – and important – differences.  Article 79(3) of the German Basic Law states:

Amendments to this Basic Law affecting the division of the Federation into Länder, their participation on principle in the legislative process, or the principles laid down in Articles 1 and 20 shall be inadmissible.”

Articles 1 to 20 contain our equivalent of Part III – human dignity, personal freedom, equality before law, freedom of expression, and so on.

In The Southwest State Case 1 BVerfGE 14 (1951), the constitutionality of the First Reorganisation Law, which sought to extend the term of state legislatures pursuant to Article 118 of the Basic Law, was challenged. The Court held that this violated both democratic government (from Articles 20 and 28 of the Basic Law) as well as Federalism (Articles 20, 19 and 28); consequently, it read Article 118 in a manner that did not extend to the extension of state legislative terms; crucially, however, the Court stated, in a lengthy paragraph that deserves to be quoted in full:

“An individual constitutional provision cannot be considered as an isolated clause and interpreted alone.  A constitution has an inner unity, and the meaning of any one part is linked to that of other provisions.  A constitution has an inner unity, and the meaning of any one part is linked to that of other provisions.  Taken as a whole, a constitution reflects certain overarching principles and fundamental decisions to which individual provisions are subordinate. Artcle 79 (3) [the Eternity Clause] makes it clear that the Basic Law makes this assumption.  Thus, this Court agrees with the statement of the Bavarian Constitutional Court:  “That a constitutional provision itself may be null and void is not conceptually impossible just because it is a part of the constitution.  There are constitutional principles that are so fundamental and so much an expression of a law that has precedence even over the constitution that they also bind the framers of the constitution, and other constitutional provisions that do not rank so high may be null and void because they contravene these principles.” From this rule of interpretation it follows that any constitutional provision must be interpreted in such a way that it is compatible with those elementary principles and with the basic decisions of the framers of the constitution.” (c.f. Kommers & Miller, The Constitutional Jurisprudence of the Federal Republic of Germany)

Notice, immediately, the fundamental difference between German and Indian constitutional jurisprudence: it is accepted in both jurisdictions that the Constitution has a set of “overarching principles” that are grounded in – yet irreducible to, and conceptually prior to – individual provisions; and that these principles are derived from structural, holistic analysis of the Constitution as a unified document, projecting a coherent moral and political vision. Yet Germany takes this argument to its logical conclusion – if basic constitutional principles are not reducible to individual provisions, then conceptually, it is possible that an individual provision is not in consonance with that basic constitutional identity; this leads to the seemingly bizarre – yet logically quite defensible – corollary that “a constitutional provision can be unconstitutional.”

As far back as Golak Nath, however, the Indian Supreme Court rejected this argument. We have seen, however, that as a matter of logic, it cannot be faulted. What, then, justifies the contrary observations of the Court? Historically, we know that questions of legitimacy in the early years of the basic structure doctrine made the Court particularly wary of grounding the basic features in anything outside specific constitutional provisions (see, for instance, Mathew J.’s Holmes-ian observation in Indira Nehru Gandhi v. Raj Narain, denying that the basic structure was some “brooding omnipresence in the sky“). So the argument could be that the basic structure cannot – by definition – ignore any constitutional provision; rather, it is the combination of all the individual clauses that provide us with our architectonic basic structure.

The problem with this argument is that it stretches a legal fiction to the breaking point. There is no reason to imagine that all 395 articles of our Constitution actually come together in a seamless, structural unity – that could simply not be true as a matter of fact; in any event, that is a matter for investigation rather than a priori stipulation.

A valid question might be raised at this point: why structural unity at all? Why must we pick between the unpalatable alternatives of shoehorning 395 articles into some artificial moral whole, and holding provisions of the constitution unconstitutional? Indeed, in this powerful piece, Professor John Manning argues that imposing structural unity upon a Constitution (or a statute) could often amount to an interpretive betrayal by failing to respect the numerous compromises and bargains that went into producing the patchwork statute that eventually passes the all-important parliamentary vote.

Dworkin gives us one reason to strive for structural unity: recognising that the ultimate question here is one of political legitimacy, he argues in Law’s Empire that government can only claim citizens’ allegiance if it accords to each one of them “equal concern and respect”; and equal concern and respect is possible only if there is principled consistency in the laws that it frames and implements (for the full argument, see Ch. VII, Law’s Empire [“Integrity”] and the Preface to Freedom’s Law). We therefore have reason to depart from fact and accept the fiction of imputing a principled unity to our Constitution – because that is the only thing that justifies it as a political document that can claim legitimate political authority.

Accepting this to be true – and the basic structure doctrine does seem to need to accept some variant of the Dworkinian argument to justify its own existence – we are faced again with the untidy dilemma of the Indian Constitution. If both basic structure exists, and no provision of the Constitution is unconstitutional, we must accept a (in my mind, very unsatisfactory legal fiction) of a definitionally principled and unified Constitution.

There is, in addition, a further untidy corollary: as we pointed out in the beginning, this conclusion means that constitutional amendments are subject to the basic structure test, but constitutional provisions are not; that, in turn, can only be justified by according a higher place to the provisions of the original Constitution; above, we have argued that there is no logical warrant for doing so. Is there a justification in political morality? The only conceivable one might be that the original Constitution is treated as the authentic voice of “We the People“, whereas amendments – even though they require a supermajoritarian procedure – are, in the absence of referenda, one step removed from the people. Yet this argument is another one of those legal fictions that strains the imagination, especially given that the Indian Constitution (unlike, for instance, the American and the Brazilian) was not subject to popular ratification (no matter how sketchy or limited – it just wasn’t). Consequently, amendments – if anything – are more representative of “We the People” than the original Constitution! Unsatisfactory as it is, though, this seems to be the only justification that can even remotely whole water.

The basic structure doctrine has, of course, had a checquered and controversial life; it also rests upon – as I have tried to show in this post – a series of problematic and potentially contradictory assumptions. If only for constitutional clarity, it would be desirable if, in a future case, the Court was to develop a strong defence of the intellectual foundations of the doctrine.