Notions of Rationality: A Dissenting Tradition in Articles 25 and 26

Indu Malhotra’s lone dissent in Sabarimala held, in two sentences, that the Court was not the arbiter of whether a religious practice was rational. The line is hers. The position is older — traceable through Khehar’s minority in <em>Shayara Bano</em> and Mukherjea’s opinion for the seven-judge bench in <em>Shirur Mutt</em>. A reconstruction of the doctrinal current the majority refused to follow.

The line is from paragraph 11.6 of her dissent: It is irrelevant whether the practise is rational or logical. Notions of rationality cannot be invoked in matters of religion by courts. It is two sentences. It sits in the middle of a section titled “Constitutional Morality in Matters of Religion in a Secular Polity” and it does what the rest of the dissent spends a hundred and fifty paragraphs justifying. The four judges who decided against her had argued that the practice of restricting women between ten and fifty from entering the Sabarimala temple failed every test the Constitution was built to apply. Indu Malhotra J. answered, in those two sentences, that they had picked the wrong tests.

The vote was 4:1. The case was Indian Young Lawyers Association v. State of Kerala, decided on 28 September 2018. The majority — Dipak Misra CJI, A.M. Khanwilkar, R.F. Nariman, and D.Y. Chandrachud JJ — held the practice violative of Articles 14, 15, 17, 25, and 26 and struck down Rule 3(b) of the Kerala Hindu Places of Public Worship (Authorisation of Entry) Rules, 1965. Malhotra J. would have dismissed the petition for want of standing and held the practice protected.

This essay is about Malhotra’s reading of Articles 25 and 26, what it asked the Court not to do, and why that asking sits in a longer line. The line runs through B.K. Mukherjea J.’s opinion in Shirur Mutt and J.S. Khehar CJI’s minority opinion in Shayara Bano. Not every voice in this line is a formal dissent. The unifying conviction is.

The case the Court was asked to decide

The Sabarimala temple in Kerala’s Pathanamthitta district is dedicated to Lord Ayyappa, who is worshipped at this shrine in the form of a Naishtik Brahmachari — a celibate ascetic bound by strict vows. Pilgrims undertake a forty-one-day vratham involving abstinence, isolation from women in the household, and barefoot travel through forested hills. The temple has, for at least several centuries on the evidence the Kerala High Court accepted in S. Mahendran v. The Secretary, Travancore Devaswom Board, restricted entry of women in the age group of ten to fifty during the notified pilgrimage period. The restriction was codified, after Independence, in Rule 3(b) of the 1965 Rules made under the Kerala Hindu Places of Public Worship (Authorisation of Entry) Act, 1965.

The Indian Young Lawyers Association — a registered body of lawyers, none of them devotees of Lord Ayyappa — filed a writ petition in 2006 under Article 32. They had learned of the practice from three newspaper columns. They asked the Supreme Court to declare Rule 3(b) ultra vires the parent Act and violative of Articles 14, 15, 17, 21, 25, and 26. The petition was eventually referred to a Constitution Bench.

What the majority held

The four majority opinions, separately authored, converged on the result by different paths. Misra CJI, writing for himself and Khanwilkar J, held that the devotees of Lord Ayyappa did not constitute a religious denomination within Article 26 — they failed the three-fold test laid down in S.P. Mittal v. Union of India of a common faith, common organisation, and a distinctive name. They were Hindus worshipping at a Hindu temple. The exclusion was therefore not protected by denominational autonomy. It was, instead, a religious practice testable against Part III; and tested against Part III, it failed.

Nariman J. agreed on the denominational question and added that the exclusion was not, in any event, an essential religious practice — the test the Court had been applying since Shirur Mutt. Chandrachud J. went furthest. He read the exclusion as a form of untouchability under Article 17, the constitutional provision that abolished caste-based ritual exclusion. He read “untouchability” in Article 17 to extend beyond caste to all stigma-based exclusion, including exclusion grounded in menstruation. He invoked constitutional morality as a principle that operated on religion itself, demanding that practices be tested against the Constitution’s commitment to dignity and equality.

The case for the majority, stated fairly, is straightforward. Article 25 protects religion subject to public order, morality, and health, and subject to the other provisions of Part III — which means subject to Articles 14, 15, and 17. Article 26(b) protects denominational autonomy, but only for groups that meet the denomination test, and the petitioners had not made that case. The exclusion was based on physiology, which is to say on sex, which is to say it failed Article 15. It was a stigma, which engaged Article 17. And constitutional morality, by 2018 a doctrine the Court had developed across Naz Foundation, Manoj Narula, and Navtej Johar, supplied the lens through which all of this was to be read.

Where Malhotra diverges, and on what premise

Malhotra J. begins where none of the four majority opinions does — with standing. The petitioners are not devotees. They have never claimed to worship at Sabarimala. They learned of the practice from newspaper articles and brought a public interest petition to challenge it. The first thirty paragraphs of the dissent are concerned with whether such a petition is maintainable at all.

Her answer is no — not absolutely, but in this domain. Article 32, she notes, requires a violation of fundamental rights. The petitioners assert no violation of their own; they assert a violation of women devotees as a class, by a practice the women devotees themselves have not challenged. Public interest litigation, Malhotra writes, has its place, but its place is not the entertainment of “interlopers to question religious beliefs and practises” at the behest of those who do not share the faith. “The perils,” she observes in paragraph 7.3, “are even graver for religious minorities if such petitions are entertained.”¹

This is not a technicality. It is the doorway to the substantive argument. If the petitioners are inside the faith, then the question of whether the exclusion violates equality is an internal question, fought among co-religionists with the Court as referee. If the petitioners are outside the faith — if they are, in her words, “social activists” — then the question becomes one of judicial review of the religion itself. Articles 14 and 15, she argues at paragraph 7.4, presume comparators who are similarly situated; in matters of religion, similarly situated means belonging to the same faith. The petitioners did not.

From this premise the rest of the dissent follows. If the Court has no business adjudicating a religious practice at the suit of non-devotees, then the doctrines the majority deploys to do so — essential practices, constitutional morality, untouchability extended beyond caste — are answers to a question that ought not to have been asked.

The dissent’s reasoning architecture

Three structural moves carry the argument. The first is the denominational question. Malhotra surveys the evidence — the Sthal Purana of the Bhuthanatha Geetha, the 1893 and 1901 Memoir of the Survey of the Travancore and Cochin States by Ward and Conner, the testimony recorded by the Kerala High Court in S. Mahendran from the temple Thanthri, the Secretary of the Ayyappa Seva Sangham, and a senior member of the Pandalam Palace — and concludes that the Respondents have made out, at minimum, a “plausible case” that the worshippers of Lord Ayyappa at Sabarimala constitute a religious denomination, or a sect thereof, satisfying the three-fold test from S.P. Mittal. She emphasises that this is a mixed question of fact and law which ought properly to be decided on evidence by a civil court, not on affidavits in a writ petition by non-devotees.²

The second move concerns Article 26(b) itself. Read literally, Article 26 is subject only to public order, morality, and health — not to the other provisions of Part III. Malhotra accepts the qualification this Court placed on that text in Sri Venkataramana Devaru v. State of Mysore, which held that Article 26(b) is subject to Article 25(2)(b) and that the State can, by law, throw open Hindu religious institutions of a public character to all classes and sections of Hindus. She accepts, too, that the 1965 Act is such a law. But she reads the proviso to Section 3 of the Act — which carves out an exception for temples founded for the benefit of any religious denomination or section thereof — as a legislative recognition of the denominational right. Rule 3(b), she holds, is statutory recognition of a pre-existing custom and falls squarely within that proviso.

The third move is the most consequential, and it is what the pulled quote stands for. Even if the practice is testable under Articles 14, 15, and 25, the test cannot be one of rationality. The role of the Court, she writes in paragraph 10.1, is to afford protection to those practices which are regarded as “essential” or “integral” by the devotees or the religious community itself. The court is not an arbiter of the practice’s reasonableness. To assess whether the exclusion at Sabarimala is logical is to do, in religious matters, what no judge does in any other domain of conscience.

The line Malhotra inherits: from Mukherjea in Shirur Mutt

The proposition that courts ought not to second-guess what a religious community regards as essential is not new. It is older than Malhotra J. and it is older than Sabarimala. Its primary source is Justice B.K. Mukherjea’s judgment for a seven-judge bench in Commissioner, Hindu Religious Endowments, Madras v. Sri Lakshmindra Thirtha Swamiar of Sri Shirur Mutt (1954). Mukherjea’s opinion in Shirur Mutt is not a dissent. It is the unanimous opinion of the Court. But within the doctrinal line of Articles 25 and 26 it is, as Malhotra deploys it, the road the Court was supposed to take and from which it has at various points departed.

The relevant passage in Shirur Mutt holds that what constitutes an essential part of a religion is primarily to be ascertained with reference to the doctrines of the religion itself. Under Article 26(b), a religious denomination enjoys complete autonomy in deciding what rites and ceremonies are essential according to its tenets, and no outside authority has any jurisdiction to interfere with that decision. This is the autonomy reading. It treats the denomination as the authority on its own essence; it treats the Court as a guarantor of that authority, not its substitute.

What happened next is the part Malhotra spends a section addressing. In Durgah Committee, Ajmer v. Syed Hussain Ali (1961), Justice P.B. Gajendragadkar, writing for a Constitution Bench, struck what he called “a note of caution” — practices which had sprung from “merely superstitious beliefs” might be “extraneous and unessential accretions to religion itself” and could be denied protection under Article 26. The same judge repeated the point in Tilkayat Shri Govindlalji Maharaj v. State of Rajasthan (1964), holding that a court might reject a claimed practice as based on “irrational considerations.”

H.M. Seervai, whose Constitutional Law of India Malhotra cites at length in paragraphs 10.2 to 10.7, treats these observations as obiter and as inconsistent with the seven-judge holding in Shirur Mutt. Seervai’s objection is not procedural. It is conceptual. “What is ‘superstition’ to one section of the public,” he writes in the passage Malhotra quotes, “may be a matter of fundamental religious belief to another.” The Gajendragadkar line, Seervai argues, substitutes the view of the court for the view of the denomination on what is essentially a matter of religion.³

This is the inheritance Malhotra claims. The dominant doctrinal current in Articles 25 and 26 — the Mukherjea current — leaves the determination of essentiality to the religion. A minority current — the Gajendragadkar current — lets the Court strike down what it considers irrational. Malhotra reads the majority in Sabarimala as an aggressive extension of the second current. Her dissent is a return to the first.

The line Malhotra inherits: from Khehar in Shayara Bano

Closer in time, the inheritance runs through Chief Justice J.S. Khehar’s opinion in Shayara Bano v. Union of India (2017). Shayara Bano struck down the practice of talaq-e-biddat (triple talaq) among Sunni Muslims by 3:2. Khehar CJI, writing also for Abdul Nazeer J., would have upheld the practice as part of personal law protected under Article 25(1). His was the minority opinion.

The passage Malhotra quotes in paragraph 10.6 of her dissent is from Khehar’s opinion at paragraph 389. It is worth setting alongside her pulled quote:

It is not difficult to comprehend what kind of challenges would be raised by rationalist assailing practises of different faiths on diverse grounds, based on all kinds of enlightened sensibilities. We have to be guarded lest we find our conscience traversing into every nook and corner of religious practises, and Personal Law. Can a court, based on a righteous endeavour, declare that a matter of faith be replaced, or be completely done away with? … it is not for a court to make a choice of something which it considers as forward-looking or non-fundamentalist. It is not for a court to determine whether religious practises were prudent or progressive or regressive. Religion and Personal Law, must be perceived, as it is accepted by the followers of the faith.Khehar CJI, dissenting in Shayara Bano v. Union of India, (2017) 9 SCC 1, para 389

The thread between the two opinions is exact. Khehar in Shayara Bano was outvoted on a Muslim practice; Malhotra in Sabarimala the following year was outvoted on a Hindu practice. The doctrinal position in both cases is the same. The Constitution’s commitment to religious freedom is not, in this reading, a commitment to those religious practices that survive a rationality audit. It is a commitment to religious practices as the religion holds them. The audit is what the Constitution was supposed to prevent.

Malhotra cites Khehar’s minority position not as authority but as confirmation. She does not say so explicitly. She does not need to. The juxtaposition is the argument.

The fault line: rationality, untouchability, and constitutional morality

Three doctrinal disagreements separate the majority and the dissent so completely that both cannot be right.

The first concerns the role of rationality. The majority holds that a practice which excludes a class of persons on a physiological characteristic fails the rationality requirement of Article 14, the non-discrimination requirement of Article 15, and — for Chandrachud J — the dignity requirement of Article 21. Malhotra holds, with Mukherjea and Khehar behind her, that rationality is not the test for religious practice. Either the practice is essential to the religion or it is not; if it is, the Court protects it.

The second concerns Article 17. Chandrachud J’s reading expands “untouchability” beyond caste to include stigma-based ritual exclusion of any kind, with menstrual exclusion as a paradigm case. Malhotra responds, in Section 14 of her dissent, with three arguments. The Constituent Assembly debates, particularly the statements of V.I. Muniswamy Pillai and Monomohan Das on Draft Article 11, show that the framers used “untouchability” in its historical sense — caste-based exclusion of Harijans. The leading commentaries by Seervai and M.P. Jain read the term the same way. And the line of Article 17 cases the Court itself had decided before Sabarimala — a line treating untouchability as the caste-based disability the Untouchability (Offences) Act 1955 was framed to address — had not extended the provision beyond that historical core. “Not a single precedent,” she writes in paragraph 14.7, “has been shown to interpret Article 17 in the manner contended by the Petitioners.”

The disagreement on this point is not about whether stigma is bad. Both sides accept that it is. The disagreement is about whether the framers wrote Article 17 to deal with stigma in general or with one historically specific form of it. Malhotra reads them to have done the second. Chandrachud reads them to have done the first while reserving the answer to the future. Both readings are arguable on the debates; only one is consistent with the line of cases the Court itself decided in the seventy years between the Assembly and the judgment.

The third concerns constitutional morality. The majority deploys constitutional morality as a corrective lens — a standard above and beyond the text of the rights, by which religious practices may be tested. Malhotra accepts the concept and inverts its application. Constitutional morality, in her formulation at paragraph 11.2, in a secular polity “would comprehend the freedom of every individual, group, sect, or denomination to practise their religion in accordance with their beliefs, and practises.” It does not authorise the Court to substitute its judgment for the denomination’s. It requires the harmonisation of competing fundamental rights, with neither side erased. It is, in her hands, a defensive doctrine. In the majority’s, it is an offensive one.

One of these readings of constitutional morality is wrong, in the sense that one of them is doing work the doctrine cannot bear. The argument for Malhotra is that constitutional morality cannot, without circularity, be used to determine what the Constitution’s commitment to religion actually means; that would make the doctrine the master of the right it was supposed to interpret. The argument for the majority is that without such a doctrine, the Court is left without resources to address practices that the formal tests of Articles 14 and 15 cannot reach. Both costs are real. Malhotra’s cost is that some practices survive that, on a broader view, should not. The majority’s cost is that the Court becomes an arbiter of which religious practices deserve to continue, with all that this implies for minorities.

How the dissent has travelled

The majority’s judgment did not, in operational terms, take effect. Implementation in Kerala met sustained protest. The State, which had earlier shifted its affidavit position more than once, eventually told the Court it could not guarantee the safety of women of the notified age group who attempted entry. Within weeks of the judgment, review petitions were filed.

In Kantaru Rajeevaru v. Indian Young Lawyers Association (2020), a five-judge bench heard the review and, by 3:2, did not recall the 2018 judgment but referred a set of broader questions — the relationship between Articles 25 and 26, the scope of essential religious practices doctrine, the place of constitutional morality — to a larger bench of seven, then nine, judges. Malhotra J. was a member of the majority that made the reference. The reference order quoted her dissent. The larger bench has heard arguments but has not yet delivered judgment.⁴ The 2018 judgment thus formally stands and is, in practice, partly suspended.

The dissent has travelled in other ways. The Bombay High Court relied on it in Nikhil Soni v. Union of India-line challenges to religious practices, treating Malhotra’s standing argument as authoritative on PIL challenges by non-devotees. The proposition that Sabarimala’s constitutional-morality reasoning awaits revision by the larger bench has appeared in subsequent Supreme Court orders dealing with religious autonomy claims. In academic commentary, Malhotra’s framing of constitutional morality — as a doctrine of harmonisation rather than override — has become the leading rival reading to the Chandrachud formulation.

A separate consequence is institutional. Malhotra J. retired in March 2021 having sat on five-judge benches in some of the most contested cases of the period — Sabarimala, the Ayodhya appeal, the Aadhaar review, the Triple Talaq review. Her Sabarimala dissent is the opinion most closely associated with her tenure. It is also, in the line of religion-and-state opinions traced here, the one that placed a sitting judge most squarely against the prevailing reading of Article 25 in 2018.

The thread, and what remains of it

The thread that runs through Mukherjea in Shirur Mutt, Khehar in Shayara Bano, and Malhotra in Sabarimala is not a thread of result. Shirur Mutt upheld the religious denomination’s claim; Shayara Bano went against it; Sabarimala went against it. The thread is a thread of method. It is the proposition that when the Court is asked to decide whether a religious practice is protected, the determination of what counts as religious is not a determination it makes in its own voice. The religion speaks first. The Court decides where the religion’s voice ends and where other constitutional commitments begin.

The objection to this position, fairly stated, is that it leaves the protection of equality dependent on the religion’s own willingness to extend it. The objection has force, and Malhotra does not fully answer it. Her response, to the extent there is one, is that Article 25(2)(b) and Article 17 are the constitutional instruments by which the State — not the Court — may compel change, through legislation responsive to social reform and through the prohibition of caste-based untouchability. The instruments exist. They have not been used to override Sabarimala because the legislature has not chosen to use them. The Court’s decision to do the work the legislature has not done is, on this reading, the constitutional difficulty.

The doctrinal position is not new with Malhotra. The line she inherits goes back to a seven-judge bench in 1954 and runs through a Chief Justice writing in dissent in 2017. What is new is that in 2018 the position lost 4:1 on facts that made the loss politically defining — a Hindu temple, a women’s rights petition, a Court reading constitutional morality with confidence. The pending nine-judge reference in Kantaru Rajeevaru will decide whether that confidence was justified. Until it does, the law on Articles 25 and 26 is what the 2018 majority said it is, unsettled by the reference but not displaced.