Why the Better Question Lies in a Forgotten Hindu College Judgment on Caste, Representation and Equality

I. The Wrong Question and the Right One

Is Justice S. Muralidhar “casteist”? The question is deliberately uncomfortable, but the answer cannot responsibly be found in an adjective. Judicial disagreement is not proof of personal prejudice, and a judgment adverse to a Scheduled Caste litigant does not become a casteist judgment merely because of its result. A judge’s character cannot be read off an adverse outcome. The criticism that can legitimately be made of the Delhi High Court’s decision in Governing Body of Hindu College v. Dr Ratan Lal does not need that detour. It is a criticism of interpretation, not of motive, and it is stronger for being so.

The better question is this: what conception of representation did the Delhi High Court adopt when Justice Muralidhar, speaking for a Division Bench, reversed Justice Suresh Kumar Kait’s judgment concerning the selection of the Principal of Hindu College? More precisely, when a rule requires that a decision-making body include someone “representing” a list of historically disadvantaged groups, are those groups interchangeable? Justice Kait, sitting singly, answered no. The Division Bench of Justices S. Muralidhar and I.S. Mehta answered yes, largely on grounds of workability. That largely forgotten disagreement deserves to be recovered, because it anticipated a question now before the Supreme Court in the litigation over the UGC (Promotion of Equity in Higher Education Institutions) Regulations, 2026: when law recognises historically disadvantaged identities, must it recognise their differences too?

II. The Dispute

In March 2017, the University of Delhi prescribed the procedure for appointing college principals through a centralised online portal. Applications would pass through four stages: a University-constituted Pre-Screening Committee (PSC), a Screening Committee constituted by the college’s Governing Body, a Selection Committee, and finally an Apex Committee of the University.

Two provisions of that framework matter here. The Screening Committee, ordinarily the Governing Body Chairman and two members, was to include an additional academician “representing SC/ST/OBC/Minority/Women/Persons with Disability”, nominated by the Chairman, if a candidate from these categories had applied and no existing member belonged to “that category”. The Selection Committee carried a parallel clause, with the representative to be nominated by the Vice-Chancellor.

Hindu College advertised the post of Principal in October 2017, and twenty-two persons applied, among them the historian Dr Ratan Lal, a Scheduled Caste candidate. Eligibility for interview required a consolidated Academic Performance Indicator (API) score of 400, computed from the candidate’s self-assessment and verified against records. Dr Lal claimed 496. The PSC awarded him 95, allotting nothing for research papers in journals and 50 of the 200 claimed for other publications. On representation, the Screening Committee raised his score to 128, still far short of the threshold. Twelve candidates were ultimately interviewed, and the officiating Principal was appointed.

Dr Lal had meanwhile sought information under the Right to Information Act, 2005. The University transferred his request to the College, whose Principal replied that the information was not available with him either. A first appeal failed, and a second appeal to the Central Information Commission was pending when he approached the High Court.

III. The Single Judge: Representation as a Check on the Decision-Maker

In Dr Ratan Lal v. Hindu College (W.P.(C) 8865/2018, decided 27 November 2018), Justice Kait found the process wanting on several fronts:

• Scoring. He found no clarity on the criteria used to award API marks.
• The record. He noted that the complete record of the selection process had not been produced.
• Transparency. He criticised the delay in disclosure under the RTI Act.
• Favouritism. He concluded that the process was an “eye wash” favouring a pre-decided candidate, in violation of Article 14.

He also held the representational clause mandatory. Because Dr Lal was Scheduled Caste and no member of the Selection Committee belonged to that category, the committee was improperly constituted. He quashed the process and directed a fresh assessment by a reconstituted Screening Committee, adding that a representative of the Scheduled Caste category “shall be ensured” on the Selection Committee.

The last direction carries a constitutional idea worth stating precisely. It does not assume that an SC academician thinks in a particular way, or that one person can speak for a constitutional class. Its premise is institutional. A body exercising power over a candidate should not be structurally blind to the social location that triggered the safeguard in the first place. On this view, representation is a condition on the legitimacy of the decision-maker, not a preference for the candidate. The distinction is between presence and representation. A committee may be diverse and still lack the one perspective the rule was written to secure.

IV. The Division Bench: Restraint, Record and Reading

The Division Bench allowed the College’s appeal on 24 April 2019. Much of its reasoning stands on well-settled ground.

Judicial restraint in academic matters. Relying on University of Mysore v. C.D. Govinda Rao, Dalpat Abasaheb Solunke v. B.S. Mahajan and allied authority, the Bench reiterated that a writ court does not sit in appeal over expert selection committees. Interference is justified only for illegality, patent irregularity in the committee’s constitution or procedure, or proved mala fides.

The API scores. The Bench found that Dr Lal was not singled out. Other candidates’ claims were cut more steeply. One claim of 1,618 was reduced to 360, another of 1,387.5 fell to 115 before being revised to 202, and a third went from 702.5 to 74.5 before being restored to 535. The Bench also recorded the College’s explanation of discrepancies in Dr Lal’s form: a journal entry showing “00” pages, and books claimed that were published only after the application was made. It observed that he had not claimed points in the categories for which the Single Judge faulted the committee for awarding none. And if the Screening Committee had been hand-picked to exclude him, it need not have raised his score at all.

Mala fides. Dr Lal’s counsel could not point to any specific averment of mala fides in the writ petition. The allegation that the woman academic on the Screening Committee was related to an advisor of the Governing Body surfaced only in written submissions in appeal. Applying Purushottam Kumar Jha and Ratnagiri Gas, the Bench held that malice could not rest on insinuation, least of all against a multi-member, multi-stage process. The College also pointed out that, despite a bar on Dr Lal holding administrative posts arising from earlier disciplinary proceedings, it had issued him a no-objection certificate to apply.

The record. The Bench accepted the College’s position that the information sought was not in its possession, and that the applications and post-screening scores had in fact been produced before the Single Judge.

On all of this, the Division Bench’s conclusions are defensible and in places compelling. Judicial review of expertise is not judicial substitution for expertise. The controversy lies elsewhere, in paragraphs 47 and 48.

V. Reading the Clause Again

The Bench read the Screening Committee clause to mean that where a candidate belongs to one of the listed categories, an academician from any of those categories may be nominated to represent “those interests”. It held that the Single Judge erred in requiring the nominee to share the candidate’s category. Its illustration was disability: if an applicant is hearing-impaired, the academician need not be hearing-impaired, and a contrary reading “would make the provision unworkable”. Since an academician from one of the listed categories had been nominated, the Screening Committee was properly constituted. Three difficulties follow.

The text. The clause is triggered when no existing member “does not belong to that category”. The singular is naturally read as referring to the category of the candidate who has applied. If any listed category would do, the proviso’s final limb would more naturally say “any of those categories”. The Selection Committee clause uses the same construction. The Division Bench did not engage with this wording. That does not make its reading impossible, since “that category” could refer back to the list treated as a single class, but it makes the reading contestable rather than self-evident.

The illustration. The disability example proves less than it claims. The listed category is “Persons with Disability”, not “persons with hearing impairment”. A same-category reading requires an academician with a disability, not one with the candidate’s particular impairment. The Bench’s example refutes an ultra-granular rule that nobody had proposed, and then treats that refutation as licence for full interchangeability across categories. Between exact identity and complete fungibility lies a wide middle ground, and the Single Judge’s reading sits within it.

The strongest case for the Division Bench. The better defence of the Bench’s reading is structural, and the judgment only gestures at it. The clause provides for an academician, singular. The Screening Committee is a small body with a quorum of three. In a field of twenty-two applicants, candidates from several listed categories may well apply at once. A strict same-category reading could require up to six additional members, potentially outnumbering the core committee and changing its character. That is a genuine workability concern.

It does not, however, compel the conclusion that the categories are interchangeable. A purposive middle reading is available: where the candidate pool includes members of more than one category, the nominating authority should, so far as practicable, choose an academician whose own location covers the categories represented. For instance, an SC woman academician would cover both SC and women candidates. Where that is impossible, the authority should record its reasons for the choice it makes. Such a reading keeps the committee small without treating a woman academic as an adequate proxy for a Scheduled Caste perspective, or vice versa.

Would it have changed the result? Honesty requires acknowledging what the clause could and could not have done for Dr Lal. His exclusion turned on an API score that the Division Bench found to be fairly assessed. And his counsel had himself treated the Selection Committee’s composition, the subject of the Single Judge’s specific direction, as irrelevant to his case. A court could therefore hold that the Screening Committee was irregularly constituted and still decline to quash a completed appointment for want of prejudice. The interpretive question matters less for Dr Lal’s case than for every selection conducted under the same rule afterwards. The Division Bench’s reading has not been displaced, so it shapes how these safeguards operate across the University.

VI. Why the Categories Are Not Interchangeable

The Bench’s reading treats the slash-separated list as a single class of “diversity”. The Constitution does not. Each category in the clause has a distinct constitutional and statutory source:

• Women: Article 15(3).
• Socially and educationally backward classes, SCs and STs: Article 15(4), and Article 16(4) in public employment.
• Minorities: Articles 29 and 30.
• Persons with disabilities: principally the Rights of Persons with Disabilities Act, 2016.
• Scheduled Castes specifically: Article 17’s abolition of untouchability, a provision unique among fundamental rights in naming a social practice and declaring its enforcement an offence.

These provisions respond to different histories of exclusion. They carry different remedial logics, operate through different institutions (the National Commission for Scheduled Castes under Article 338 has no general analogue for women or minorities), and are measured through different data. The fact that an administrative clause lists them together tells us that the University considered them all worthy of representation. It does not tell us that one can stand in for another.

Indian equality jurisprudence has long rejected formal sameness as the measure of equality. State of Kerala v. N.M. Thomas treated remedial classification as a facet of Article 16(1) itself, and Indra Sawhney v. Union of India built reservation doctrine around the specific character of backwardness. More recently, Lt. Col. Nitisha v. Union of India recognised that facially neutral structures can reproduce disadvantage. Most directly on point, the seven-judge Bench in State of Punjab v. Davinder Singh upheld sub-classification within the Scheduled Castes, overruling E.V. Chinnaiah. The Court accepted that even a single constitutional category can contain groups whose disadvantage differs in kind and degree.

If the law now recognises differences within the Scheduled Castes, it becomes harder to defend a reading under which the differences between Scheduled Castes, women and minorities disappear altogether for the purpose of committee composition. The objection to the Division Bench’s reading is therefore not that it was insufficiently generous to one litigant. It is that it adopted a conception of representation that sits uneasily with the direction of constitutional doctrine.

The limits of the argument should be stated too. Identity-conscious measures remain subject to constitutional discipline. Reading them too rigidly can harden categories into permanent compartments, and can make institutions unable to function. The task is calibration, not maximalism.

VII. The Same Problem at Scale: The 2026 UGC Regulations

The problem has now reached the Supreme Court in a different form. The 2026 Regulations trace their origin to Abeda Salim Tadvi v. Union of India, the 2019 petition filed by the mothers of Payal Tadvi and Rohith Vemula, in which the Court had directed the UGC to notify new anti-discrimination regulations. Clause 3(c) of the Regulations, notified on 13 January 2026, defined “caste-based discrimination” as discrimination against members of the Scheduled Castes, Scheduled Tribes and Other Backward Classes. Petitioners in Mritunjay Tiwari v. Union of India argued that this excluded persons from non-reserved categories, contrasting it with the broader definition of “discrimination” in Clause 3(e).

On 29 January 2026, a Bench of Chief Justice Surya Kant and Justice Joymalya Bagchi found some provisions prima facie ambiguous and open to misuse. It framed four questions of law:

1. Whether Clause 3(c) bears a rational nexus to the Regulations’ purpose, given that no distinct procedure had been prescribed for caste-based discrimination as against the inclusive definition of discrimination in Clause 3(e).
2. Whether the Regulations adequately protect Extremely Backward Castes within the SC, ST and OBC classifications.
3. Whether “segregation” in Clause 7(d), even on transparent criteria, would amount to a “separate yet equal” classification.
4. Whether the omission of ragging was a regressive step.

The Court kept the Regulations in abeyance, and, invoking Article 142, directed that the UGC’s 2012 Equity Regulations continue in force until further orders. During the hearing, the Chief Justice reportedly clarified that his concern was not the general category but the need for an effective mechanism to protect marginalised candidates.

Two features of this litigation connect it to Hindu College. The first is the internal-differentiation question. The second question, on Extremely Backward Castes, is the Hindu College problem turned inward. It rejects both crude caste-neutrality and crude category essentialism, and asks whether a remedial instrument sees the differences within the categories it protects.

The second is the interim technique. The Court did not leave campuses without an anti-discrimination framework while it examined the new one. That separates two propositions that public debate tends to merge: questioning the design of a particular protective instrument is not the same as abandoning protection.

These remain petitioners’ contentions and prima facie observations, not adjudicated conclusions. On 20 August 2026, the Solicitor General told a three-judge Bench that the Regulations were under active consideration. The Court expressed the hope that a decision would be taken before the next hearing, now listed for 5 November 2026.

 

VIII. Conclusion: Answering the Question

So, is Justice Muralidhar “casteist”? Nothing in the two decisions supports that charge. Nothing in the Division Bench’s judgment suggests hostility towards Scheduled Castes, and much of it is a careful application of settled law on academic restraint and mala fides. Calling its author “casteist” would weaken the most serious criticism of Hindu College, not strengthen it, because it converts a disagreement about doctrine into an allegation about character. A judicial judgment should be criticised through its text, its reasoning, the evidence before it and its consequences, not through speculation about the motives of the judge who wrote it.

The criticism that survives that discipline is narrower and more durable. In paragraphs 47 and 48, the Division Bench read a representational safeguard as a generic diversity requirement. It justified that reading with an illustration that did not test it, and did not engage with the one phrase in the clause that pointed the other way. Specificity need not mean perfect identity; but neither does inclusion mean interchangeability.

Justice Kait’s judgment may have overreached on the scores and on mala fides. But on representation he saw something the formal reasoning of the appeal missed: a safeguard exists to govern the institution, and it is satisfied only when the perspective it was designed to secure is actually in the room.

That insight now matters well beyond a single college. The scale has changed, from the principalship of one Delhi college to the anti-discrimination architecture of every higher education institution in the country, but the constitutional dilemma has not. The questions the Supreme Court has framed about the definition of caste-based discrimination and the position of Extremely Backward Castes are, at their core, the Hindu College questions asked again: whom does a safeguard protect, and can one form of disadvantage stand in for another?

That is why the question with which this article began should be set aside, and the better question taken up in its place. The label “casteist” ends an argument; the question of what representation requires begins one. The next chapter of this story has not yet been written. When it is, the forgotten disagreement between Justice Kait and Justice Muralidhar deserves to be read as part of it.