LLM Group 1 Constitutional and Administrative Law Administrative Process and Judicial Control 2015 Question Paper with Solutions

Where these grounds sit

The grounds of judicial review are conventionally arranged under the three heads Lord Diplock gave them in Council of Civil Service Unions v. Minister for the Civil Service, [1985] AC 374, the GCHQ case: illegality, irrationality and procedural impropriety. He added that proportionality might in time be adopted as a fourth. Indian courts use this arrangement constantly, and the three heads set by this question straddle it: mala fides is a species of illegality, bias is procedural impropriety, an oppressive decision is irrationality, and proportionality is the fourth head Lord Diplock foresaw.

It is worth saying at the outset what unites them. Judicial review is not an appeal. The court is not asked whether the decision was right; it is asked whether the decision maker had the power, exercised it honestly, heard those affected, and reached a result within the range that the power permits. Each of the three grounds below is a way of showing that one of those conditions failed, and each is therefore about the legality of the process and the limits of the power, not about the merits.

Why it usually fails is illustrated by E.P. Royappa v. State of Tamil Nadu, (1974) 4 SCC 3. A Chief Secretary alleged that his transfer to a newly created post was a punishment engineered by the Chief Minister. The Court accepted that mala fides would vitiate the transfer, but held the burden not discharged, observing that mala fides is a serious allegation, that it must be pleaded with particulars and proved, and that a court will not infer it from suspicion. That case is more famous for Bhagwati J.'s reformulation of Article 14 as a guarantee against arbitrariness, and the two halves belong together: because malice in fact is so hard to prove, the law developed a route to the same result through arbitrariness, which needs no proof of motive.

Malice in law is what actually does the work, and Express Newspapers (P) Ltd. v. Union of India, (1986) 1 SCC 133, shows it. Re-entry upon a lease and a notice to demolish the Express Building were challenged as retaliation for the newspaper's coverage of the Emergency. The Court found the exercise of power to be a colourable one, taken at the instance of persons who wanted the newspaper silenced, and struck it down. The reasoning is the point: an authority acting on an extraneous consideration is acting outside the power, and the vice is objective, established from the file rather than from the mind of the officer.

The test India applies is the reasonable likelihood test, and Ranjit Thakur v. Union of India, (1987) 4 SCC 611, states it. A soldier who had complained against a superior officer was tried by a court martial in which that same officer participated, and was sentenced to rigorous imprisonment and dismissal. The Supreme Court quashed the proceedings, holding that the test is not whether the judge was in fact biased but whether a reasonable person in the position of the party would have a reasonable apprehension of bias, because the apprehension of the litigant, if reasonable, is what matters. The Court added that a disproportionate sentence is itself evidence of the mind that imposed it, which links this head to the third.

(b) Oppressive decision

"Oppressive" is not a loose synonym for harsh. It is a term of art drawn from the law of byelaws, and its source is Kruse v. Johnson, [1898] 2 QB 91. Lord Russell of Killowen there held that a byelaw made by an elected public body would be benevolently construed, but would be struck down as unreasonable if it were partial and unequal in its operation between classes, manifestly unjust, made in bad faith, or involved such oppressive or gratuitous interference with the rights of those subject to it as could find no justification in the minds of reasonable men. That last phrase is the ground, and it is the ancestor of the modern irrationality head.

In India the test Kruse supplies has been received directly, and its classic application is to subordinate legislation. A byelaw or rule that prohibits where the statute authorised only regulation, or that imposes a burden bearing no relation to its object, is struck down as oppressive. In Dwarka Prasad Laxmi Narain v. State of U.P., AIR 1954 SC 224, a control order left the grant, refusal and cancellation of coal licences to the unrestrained will of a single licensing officer, with no principle to guide him and, in some cases, no appeal. The Supreme Court struck the provision down, holding that a clause committing to the absolute discretion of one individual the power to grant, withhold or cancel licences in every case, without any standard or rule to control it, cannot be regarded as a reasonable restriction. The domestic authority most often cited is Indian Express Newspapers (Bombay) (P) Ltd. v. Union of India, (1985) 1 SCC 641, where the Supreme Court held that subordinate legislation may be questioned on the ground that it is manifestly arbitrary or unreasonable, adding that it does not enjoy the same immunity as an Act of Parliament, and that the court may examine whether the delegate has exceeded the standard laid down or acted on irrelevant considerations.

Om Kumar v. Union of India, (2001) 2 SCC 386, is the decision that fixed its Indian scope, and its distinction must be reproduced accurately. The Court held that where fundamental rights are restricted, the court applies proportionality as a primary reviewer, deciding for itself whether the restriction is excessive. Where an administrative decision, in particular a punishment in service law, is challenged, the court applies Wednesbury and interferes only if the penalty is so disproportionate as to shock the conscience, acting as a secondary reviewer. Proportionality therefore replaced the second limb of Wednesbury, irrationality, only in the fundamental rights field; it did not sweep Wednesbury away.

Ranjit Thakur supplies the service law illustration, the sentence of rigorous imprisonment and dismissal for refusing to eat food being held to be so strikingly disproportionate as to be perverse, and Coimbatore District Central Cooperative Bank v. Coimbatore District Central Cooperative Bank Employees Association, (2007) 4 SCC 669, restates the Om Kumar division and reviews the authorities.

The most recent application worth citing is the electoral bonds decision, Association for Democratic Reforms v. Union of India, (2024) 5 SCC 1, 2024 INSC 113, decided on 15 February 2024, where the Court struck down the scheme and used a double proportionality analysis to reconcile the voter's right to information with the donor's claim to informational privacy, asking whether the scheme was the least restrictive means of curbing black money in political funding and holding that it was not. In 2025 the Court continued to speak of proportionality as a general administrative law standard, observing in Adarsh Sahkari Grih Nirman Swawlambi Society Ltd. v. State of Jharkhand, 2025 INSC 1389, decided on 5 December 2025, that an executive instruction adding a requirement the statute does not impose is illegal and that unnecessary and excessive requirements must be set aside.

The three grounds compared

The limits of these grounds

Each ground carries a built in restraint and an answer that ignores it is one sided. Mala fides must be specifically pleaded and particularised, and the Supreme Court has repeatedly deprecated its casual pleading, because an unproved allegation of dishonesty against a public servant is itself an injustice. The rule against bias is qualified by the doctrine of necessity, under which a disqualified decision maker may act if there is no one else competent to act, since the alternative is that no decision is taken at all. The oppressiveness head is deliberately narrow, because a court that strikes down every decision it disagrees with has converted review into appeal. Proportionality is the most intrusive of the four, and its critics say that the fourth stage, the balancing, transfers to judges a choice that belongs to the elected branch.

For full marks, cover: for habeas corpus, the meaning, the constitutional source in Articles 32 and 226, who may apply and why the ordinary rule of standing is relaxed, the procedure and the reversed burden, the grounds on which detention is quashed, preventive detention under Article 22 and its safeguards, and the two cases that define the writ's Indian history, ADM Jabalpur and Rudul Sah; for droit administratif, that it is a separate body of law administered by separate courts, its origin in the separation of powers after 1790, the structure of the Conseil d'Etat and the tribunal administratif, the heads of the recours pour excès de pouvoir, Dicey's attack on it and why his account is now accepted to be wrong, and what India has taken from it.

The rules of standing are relaxed for this writ more than for any other. A detained person is by definition unable to come to court, so an application may be made by a relative, a friend, or a stranger acting pro bono publico, and the Supreme Court has treated a letter or a postcard as a petition. In Sunil Batra (II) v. Delhi Administration, (1980) 3 SCC 488, a letter from one convict to a judge, complaining that a warder had tortured a fellow prisoner named Prem Chand by driving a baton into his anus to extort money from his visitors, was converted into a habeas corpus petition. The Court held that the writ is available not only to secure release but to protect a prisoner against illegality within lawful custody, which is a substantial extension of the writ's classical function.

Procedure and burden are what give the writ its force. Once the petitioner shows a prima facie case, the court issues a rule nisi and the burden shifts to the detaining authority to establish lawful authority for the detention. The court examines the legality as at the date of the return, so an initially bad detention cannot be validated by a later good order and a detention lawful at its inception may become unlawful. Delay does not defeat the writ, and the ordinary rule that alternative remedies must first be exhausted has little application, because liberty admits of no waiting.

ADM Jabalpur v. Shivkant Shukla, (1976) 2 SCC 521, is the case the writ is measured against, and it must be stated with its facts. During the Emergency the President suspended the right to move any court for enforcement of Articles 14, 21 and 22. High Courts nevertheless entertained habeas corpus petitions from persons detained under the Maintenance of Internal Security Act. By four to one the Supreme Court held that during the suspension no person had locus standi to move for habeas corpus, and that Article 21 was the sole repository of the right to life and liberty, so that its suspension left no remedy even against a detention admitted to be without authority. Khanna J. dissented, holding that the sanctity of life and liberty did not depend on the Constitution alone and that the State had no power to deprive a person of life or liberty without the authority of law even in an Emergency. He was superseded for the Chief Justiceship, and resigned. The majority was expressly overruled in Puttaswamy, (2017) 10 SCC 1, where the Court described it as a serious aberration, and Article 359 was amended by the Forty Fourth Amendment, 1978, so that Articles 20 and 21 can no longer be suspended.

(b) Droit administratif

Droit administratif is the body of rules, developed by the French administrative courts, which governs the relations between the administration and the citizen and which is different in content from the private law applied between citizens. Its distinguishing feature is not that it is written down, for it is largely judge made, but that it is administered by a separate hierarchy of courts at whose head stands the Conseil d'Etat, with tribunaux administratifs at first instance and cours administratives d'appel between.

Its origin lies in a particular reading of the separation of powers. The revolutionary law of 16 to 24 August 1790 forbade the ordinary courts from interfering with the administration, the memory of the pre revolutionary parlements obstructing royal reform being fresh. The consequence was not that the administration escaped law; it was that a distinct body of law had to be built for it, and the Conseil d'Etat, created in 1799 as an advisory body to the executive, gradually acquired the power to decide disputes in its own right, a change completed by the law of 24 May 1872.

Dicey attacked the system and his attack is where most examination answers go wrong. In An Introduction to the Study of the Law of the Constitution (1885) he treated droit administratif as the antithesis of the rule of law, on the footing that it gave officials special privileges and exempted them from the jurisdiction of the ordinary courts, so that the citizen was left worse off. His second meaning of the rule of law, equality before the ordinary law administered by the ordinary courts, was framed precisely to exclude it. He was wrong on the facts, and admitted as much in later editions. The Conseil d'Etat is not an arm of the administration but a genuinely independent court, its procedure is cheap, quick and inquisitorial, its members are specialists in the working of government, and its record in controlling the administration is at least as good as that of the English courts. Dicey's misdescription delayed the recognition in England that a distinct administrative law was needed at all, and the Committee on Ministers' Powers (the Donoughmore Committee) reported in 1932 in a climate still shaped by it.

The two notes compared

For full marks, cover: what the quoted sentence actually asserts, which is a limit on review as much as a justification for it; the three propositions packed into it, namely validity, discernible reasons and the absence of ulterior purpose; the constitutional foundation of each in Articles 14 and 21 and in the rule of law; the duty to give reasons as the operative test of "discernible"; the ulterior purpose head, which is malice in law; the limits of review that the words "only to the extent" impose, namely that the merits belong to the administration; and a critical assessment, because "elucidate" invites you to test the statement, not merely to agree with it.

In Tata Cellular itself the facts matter. The Government invited tenders for cellular mobile telephone services in the four metropolitan cities. The Telecom Commission's evaluation, and the participation of the Director General of Telecommunications in the tender evaluation committee when his own department was interested, were challenged. The Supreme Court laid down that the government must have freedom of contract and that the court does not sit as a court of appeal over the tender decision, but that the decision making process is reviewable for illegality, irrationality and procedural impropriety, and that fair play in the joints is what the administration is entitled to expect. It struck the balance the quoted sentence expresses.

Discernible reasons: the duty to give reasons

"Discernible reasons" is the operative phrase, because a reason that cannot be discerned is, for the purposes of review, no reason at all. The requirement is not merely that reasons should exist in the mind of the officer but that they should be recorded and communicated, and Indian law has moved steadily towards making that a general obligation.

S.N. Mukherjee v. Union of India, (1990) 4 SCC 594, is the Constitution Bench authority. An army officer's conviction by court martial and its confirmation were challenged for want of reasons. The Court held that except where expressly excluded, the recording of reasons is a necessary concomitant of every administrative decision affecting rights, for three reasons: it introduces clarity and minimises arbitrariness by disciplining the decision maker; it assures the party affected that the decision was reached after due consideration; and it makes effective the power of judicial review, since a court cannot examine a decision whose grounds it cannot see. On the facts the Court held that the confirming authority need not give separate reasons where it agrees with the findings, but the general principle was settled.

Not whimsically: arbitrariness and Article 14

"Whimsically" points at arbitrariness, and after E.P. Royappa v. State of Tamil Nadu, (1974) 4 SCC 3, arbitrariness is a distinct ground under Article 14. Bhagwati J. held that equality and arbitrariness are sworn enemies, and that where an act is arbitrary it is implicit in it that it is unequal both according to political logic and constitutional law, and therefore violative of Article 14. Maneka Gandhi v. Union of India, (1978) 1 SCC 248, completed the reasoning by holding that a procedure under Article 21 must be right, just and fair and not fanciful, oppressive or arbitrary, and by reading Articles 14, 19 and 21 together as a single guarantee.

The Article 14 route matters practically because it lowers the threshold. A litigant in England must show that the decision is one no reasonable authority could have reached, a very high bar; an Indian litigant may show that it is arbitrary, which is a lower one. The consequence, sometimes criticised, is that Indian courts intervene in administrative decisions more readily than English courts do.

The limit: "only to the extent"

The sentence is also a statement of restraint and an answer that ignores that half misreads it. Judicial review is directed at the decision making process, not the decision, as Chief Constable of the North Wales Police v. Evans, [1982] 1 WLR 1155, put it, warning that review is not an appeal and that the court is concerned with the manner in which the decision was reached. Indian courts have applied that restraint most consistently in three fields: contract and tender, where Tata Cellular and Michigan Rubber (India) Ltd. v. State of Karnataka, (2012) 8 SCC 216, hold that the terms of a tender are for the authority and that the court interferes only if the process is arbitrary or mala fide; economic policy, where Balco Employees Union v. Union of India, (2002) 2 SCC 333, refused to examine the wisdom of disinvestment, holding that in matters of economic policy the court is not the forum and that judicial review is not concerned with whether a better decision was available; and academic and expert evaluation, where the court defers to the assessment of experts.

The second difficulty is that the standard is applied unevenly. Om Kumar v. Union of India, (2001) 2 SCC 386, keeps Wednesbury ([1948] 1 KB 223) for service penalties and proportionality for fundamental rights, so the intensity of review depends on the classification of the case and not on the quality of the reasons. In 2025 the Court has continued to press accountability as a general principle, observing in BSES Rajdhani Power Ltd. v. Union of India, 2025 INSC 937, decided on 6 August 2025, that accountability is itself an essential principle of administrative law, but the level of scrutiny still varies with subject matter, and candour about that is what distinguishes a good answer.

Conclusion. The sentence states the modern justification for judicial review and the modern limit on it in one breath. State action is amenable to review because public power is held on trust and must be exercised for the purpose for which it was conferred, which is why an ulterior purpose is fatal whether or not it is dishonest, as Padfield and Barium Chemicals show. It is amenable only to the extent that the court examines validity, reasons and purpose, because the choice between lawful courses belongs to the administration, which is what Tata Cellular and Balco protect. The load bearing element is the middle one, discernible reasons: after Mohan Lal Capoor, S.N. Mukherjee and Kranti Associates, the recording of intelligible reasons is the ordinary duty of every authority deciding anything that affects rights, and it is the mechanism by which the other two requirements are made testable. An administration that records its reasons makes review possible and usually survives it; one that does not converts every challenge into an inquiry into its motives, which is the very whimsicality the sentence condemns.

What is being controlled, and why it exists

Delegated or subordinate legislation is law made by an authority other than the legislature, under a power conferred by the legislature. Rules, regulations, byelaws, orders, notifications and schemes are its forms. It exists because a modern legislature cannot do the whole job: parliamentary time is scarce, the subject matter is technical, conditions change faster than statutes can be amended, emergencies require immediate action, and experimentation is safer in a rule that can be withdrawn than in a statute that cannot. The Committee on Ministers' Powers (Donoughmore, 1932) in England and the fourteenth report of the Law Commission of India both accepted its necessity while warning against its abuse.

The constitutional problem is that Article 245 vests legislative power in Parliament and the State legislatures, and a body that has been given power by the legislature has not been given power by the Constitution. All the grounds below are, at bottom, ways of asking whether the rule making authority has stayed within the boundary the legislature was itself entitled to draw.

In re Delhi Laws Act, 1912, AIR 1951 SC 332, is the foundation. A Presidential reference asked whether three provisions were valid: a power to extend to Delhi any law in force in any other part of India with modifications; a similar power for Part C States, including a power to repeal or amend any corresponding law already in force there. The seven judges wrote seven opinions, but a majority position emerged. Delegation of the power to extend an existing law, with restrictions and modifications not touching its essential features, is valid; the power to repeal or amend an existing law is an essential legislative function and cannot be delegated. The reasoning is that the legislature must retain in its own hands the determination of policy, because that is what it was elected to do.

Devi Das Gopal Krishnan v. State of Punjab, AIR 1967 SC 1895, applies the doctrine to taxation, holding that a provision empowering the Government to fix the rate of sales tax without any maximum was excessive delegation, the Court adding that the legislature must fix the maximum or lay down the guidelines. And in the modern era the same idea has been applied to the structure of the State itself: in Madras Bar Association v. Union of India, 2025 INSC 1330, decided on 19 November 2025, the Supreme Court struck down key provisions of the Tribunals Reforms Act, 2021, because they gave the executive control over tribunal appointments, tenure and service conditions and repeated almost verbatim provisions that had already been declared invalid, which the Court held to be an impermissible legislative override, and it directed the Union to constitute a National Tribunals Commission within four months.

Ground 3: the rule is ultra vires the parent Act

This is the commonest ground in practice, and it has two limbs. Substantive ultra vires means the rule goes beyond what the Act authorises, contradicts the Act, or covers a field the Act does not reach.

Ground 4: violation of fundamental rights

A rule is void to the extent that it takes away or abridges a fundamental right, and Article 13(3)(a) puts the matter beyond argument by defining "law" to include rules, regulations, notifications, orders and byelaws. So a rule imposing an unreasonable restriction on a trade offends Article 19(1)(g), and one drawing an irrational classification offends Article 14, whether or not the parent Act is itself attacked. This ground is independent: a rule perfectly within the parent Act is still void if it violates Part III.

Ground 6: mala fides, retrospectivity and exclusion of natural justice

Bad faith in the making of a rule is a ground, though courts are slow to find it against a rule making body, since a rule addressed to the world at large rarely reveals a personal motive. Retrospectivity is the more useful head: a rule cannot operate retrospectively unless the parent Act expressly or by necessary implication authorises it, and a rule that takes away accrued rights from a past date without such authority is void, as Hukam Chand v. Union of India, AIR 1972 SC 2427, held, since the power to make rules with retrospective effect must be conferred and cannot be assumed. A rule that excludes natural justice where the Act does not require exclusion is likewise open to attack, since the principles are read into a statutory power unless the statute displaces them.

The other controls, briefly

Judicial control is not the only control and an answer that mentions the others is stronger. Parliamentary control operates through the requirement of laying, which may be simple, subject to negative resolution, or subject to affirmative resolution, and through the Committee on Subordinate Legislation of each House, which examines whether rules are within the powers conferred, whether they impose taxation or oust the jurisdiction of courts, and whether they have been laid in time. Procedural control operates through requirements of previous publication, consultation with affected interests, and, in the statutes that provide for it, the pre publication of drafts for objections.

Conclusion. Delegated legislation is void in India on grounds that can be arranged by the object of the attack. The parent Act may be unconstitutional; the delegation may be excessive because the legislature has parted with an essential legislative function, which is the Delhi Laws Act and Hamdard Dawakhana line; the rule may be substantively or procedurally ultra vires, which is Kunj Behari Lal Butail and Raza Buland Sugar; it may violate Part III, Article 13(3)(a) making the point unarguable; it may be manifestly arbitrary or unreasonable in the Indian Express sense; it may be retrospective without authority, mala fide, or may exclude a hearing the statute never displaced; and it may fail for want of publication, which is Harla and M.H. George. What holds these together is a single proposition: the delegate exercises borrowed power, and borrowed power is measured against the terms of the loan. The most significant recent application of that proposition is Madras Bar Association (2025), where the Court refused to allow the executive to obtain, by re-enacting struck down provisions, what it had already been denied, which shows that the doctrine still has force where the stake is the independence of adjudication itself.

The text and what it guarantees

Article 32 has four clauses and each carries weight. Clause (1) guarantees the right to move the Supreme Court by appropriate proceedings for the enforcement of the rights conferred by Part III. Clause (2) empowers the Court to issue directions or orders or writs, including writs in the nature of habeas corpus, mandamus, prohibition, quo warranto and certiorari, whichever may be appropriate. Clause (3) permits Parliament to empower any other court to exercise the same jurisdiction. Clause (4) provides that the right guaranteed by the Article shall not be suspended except as otherwise provided for by this Constitution.

The five writs, in outline

Habeas corpus issues to produce a detained person and test the legality of the detention; the burden lies on the detainer, and the writ has been extended to conditions within custody, as Sunil Batra (II) v. Delhi Administration, (1980) 3 SCC 488, shows. Mandamus commands the performance of a public duty; it lies where there is a legal duty of a public nature and a corresponding right in the applicant, and it does not lie against a private person, nor to enforce a contractual obligation simply, nor to direct the legislature to legislate. Prohibition issues to an inferior court or tribunal to stop it exceeding its jurisdiction, while proceedings are pending. Certiorari issues to quash an order already made for want of jurisdiction, violation of natural justice, or error of law apparent on the face of the record. Quo warranto calls upon a person to show by what authority he holds a public office of a substantive character created by statute or the Constitution, and it may be sought by any person, standing being irrelevant.

Who may move the Court

The traditional rule was that only the person whose right was infringed could apply. That rule has been substantially relaxed. In S.P. Gupta v. Union of India, 1981 Supp SCC 87, the Court held that where a legal wrong is caused to a person or class of persons who by reason of poverty, helplessness, disability or socially or economically disadvantaged position are unable to approach the court, any member of the public acting bona fide may move the court. In People's Union for Democratic Rights v. Union of India, (1982) 3 SCC 235, the Asiad workers case, a letter complaining of violations of labour laws at the construction sites for the 1982 Asian Games was treated as a petition, and the Court held that non payment of minimum wage to a workman amounts to forced labour under Article 23, so that the complaint raised a fundamental right and Article 32 was available.

Against whom: the reach of Article 12

A writ under Article 32 lies to enforce a fundamental right, and most fundamental rights are enforceable only against the State, so the scope of Article 32 is limited by the meaning of "State" in Article 12. The expansion of Article 12 has therefore expanded Article 32. In R.D. Shetty v. International Airport Authority of India, (1979) 3 SCC 489, the Court laid down the tests for an instrumentality or agency of the State, including the financial resources of the State being the chief funding source, deep and pervasive State control, a monopoly status conferred or protected by the State, and functions of public importance closely related to governmental functions. Ajay Hasia v. Khalid Mujib Sehravardi, (1981) 1 SCC 722, generalised those tests to a society registered under the Societies Registration Act, and Pradeep Kumar Biswas v. Indian Institute of Chemical Biology, (2002) 5 SCC 111, restated them, holding that the question is whether the body is financially, functionally and administratively dominated by or under the control of the Government, and that the control must be particular to the body and pervasive.

The limits

Five limits should be stated. First, a fundamental right must be shown: Article 32 is not available for the enforcement of an ordinary legal right, however clear, and this is the principal difference from Article 226. Second, the power is discretionary in its exercise even though the right to move is guaranteed, so the Court may decline relief where the petitioner has suppressed material facts or has come with unclean hands. Third, res judicata applies: in Daryao v. State of U.P., AIR 1961 SC 1457, the Court held that a petition under Article 32 is barred where the same matter was decided on the merits by a High Court under Article 226 and no appeal was taken, since finality of judicial decisions is a rule of public policy, though a dismissal in limine or on the ground of laches does not bar. Fourth, laches: unexplained delay may defeat the petition, since a stale claim may have upset settled rights, though there is no fixed period. Fifth, alternative remedy is not a bar in the way it is under Article 226, but the Court routinely relegates a petitioner to the High Court, and in Kanubhai Brahmbhatt v. State of Gujarat, AIR 1987 SC 1159, it said that a petitioner should ordinarily approach the High Court first, so that the Supreme Court's time is preserved for matters that require it.

Compensation and the modern reach

Rudul Sah v. State of Bihar, (1983) 4 SCC 141, added a remedy the Article does not mention. Rudul Sah was acquitted of murder in 1968 and kept in jail for fourteen years afterwards. He was released before the habeas corpus petition was heard, so on the classical view there was nothing left to order. The Court awarded Rs. 30,000 as compensation in the writ jurisdiction, reasoning that Article 21 would be denuded of significance if the Court's power were limited to ordering release and the victim were left to a civil suit. Nilabati Behera v. State of Orissa, (1993) 2 SCC 746, put the doctrine on a firm footing, holding that the award of compensation in a proceeding under Article 32 or 226 is a remedy in public law, based on strict liability for contravention of fundamental rights, to which the defence of sovereign immunity does not apply, and it is distinct from a private law remedy in tort.

Conclusion. The scope of Article 32 is fixed by three propositions. It is a guaranteed remedy, so the Supreme Court cannot decline to entertain a petition that discloses infringement of a fundamental right, which Romesh Thappar established in the Court's first year. It is a flexible remedy, since the Court may issue any direction, order or writ that is appropriate, which allowed it in Bandhua Mukti Morcha to receive a letter, appoint commissioners and devise a procedure, and in Rudul Sah to award compensation where release was no longer possible. And it is a limited remedy, confined to fundamental rights, subject to Daryao res judicata, laches and the Court's own practice of relegating petitioners to the High Courts, and dependent for its reach on the meaning of "State" in Article 12. The Article's history is the best summary of its extent: reduced almost to nothing in ADM Jabalpur, restored by the Forty Fourth Amendment, expanded through public interest litigation, protected against exclusion in L. Chandra Kumar, and now guarded by the Court itself against frivolous invocation under Balwant Singh Chauffal. Ambedkar's description of it as the heart and soul of the Constitution is accurate, but the record shows that a remedy is only as strong as the willingness of the court that holds it.

(a) Injunction and declaration as judicial remedies

These are the two ordinary civil remedies available against the administration, and they matter because the writ jurisdiction is not the only route. A writ lies under Articles 32 and 226 against the State and its instrumentalities; a suit for injunction or declaration lies in an ordinary civil court, may be brought against any defendant, allows evidence to be led and facts to be tried, and results in a decree that is executable. Where the dispute turns on disputed questions of fact, which a writ court will not ordinarily investigate, the suit is the appropriate remedy.

An injunction is an order restraining a party from doing an act, or requiring the doing of an act. It is prohibitory when it forbids, mandatory when it commands under Section 39 of the Specific Relief Act, 1963, temporary when granted during the pendency of a suit under Order 39 Rules 1 and 2 CPC, and perpetual when granted by the decree itself under Section 38. Against the administration it is used to restrain an authority from acting under an invalid rule, from enforcing an order made without jurisdiction, from dispossessing a person otherwise than by due process, or from acting in breach of a statutory duty.

A declaration is a judicial statement of a legal right or status which binds the parties but grants no consequential relief of itself. Section 34 of the Specific Relief Act allows any person entitled to any legal character, or to any right as to any property, to sue for a declaration, with the important proviso that no declaration shall be made where the plaintiff, being able to seek further relief than a mere declaration, omits to do so. So a dismissed public servant who wants reinstatement and arrears must ask for them; a bare declaration that the dismissal is void will be refused. Section 35 makes the declaration binding only on the parties and those claiming through them.

(b) Conclusive evidence clause

A conclusive evidence clause provides that a specified document, certificate or finding shall be conclusive evidence of a stated fact, so that no evidence may be led to contradict it. Its purpose is administrative certainty: it saves the authority from having to prove satisfaction of preliminary conditions in every proceeding. Its danger is that it can convert a jurisdictional fact into an unexaminable one, and thereby exclude review by the back door.

It must be distinguished from its two neighbours. A finality clause says that the decision of the authority "shall be final", which excludes appeal but, as the courts have consistently held, does not exclude certiorari for jurisdictional error. An exclusionary or ouster clause says that the decision "shall not be called in question in any court", and attacks the jurisdiction itself. A conclusive evidence clause does neither expressly: it operates on the law of evidence, and its effect is to make a fact unprovable to the contrary, which may leave the court with jurisdiction but with nothing to examine.

The English authority that shaped the Indian approach is Anisminic Ltd. v. Foreign Compensation Commission, [1969] 2 AC 147. Section 4(4) of the Foreign Compensation Act, 1950, provided that a determination by the Commission of any application made to it shall not be called in question in any court of law. The Commission rejected the claim of Anisminic, whose Egyptian property had been sequestrated in 1956, on the ground that its successor in title was not a British national, a requirement which on the true construction of the Order did not apply to that class of claim. The House of Lords held that a determination made on a misconstruction of the authority's powers is a nullity, and that a clause protecting a "determination" does not protect something that is not in law a determination at all. It is the most important decision on ouster clauses in the common law world, and the same reasoning applies to a conclusive evidence clause, since a certificate cannot be conclusive of a state of affairs the authority had no power to certify.

(c) Res judicata and judicial review

Res judicata means that a matter finally decided by a competent court between the same parties cannot be reopened. Section 11 CPC states the rule for suits, requiring the same matter directly and substantially in issue, the same parties or those claiming under them, a court competent to try the subsequent suit, and a decision on the merits after hearing. Explanation IV contains constructive res judicata: any matter which might and ought to have been made a ground of attack or defence in the former proceeding is deemed to have been directly and substantially in issue.

Two qualifications are important. First, habeas corpus stands apart to a degree: a successive petition on the same grounds is not entertained, but a fresh petition on fresh grounds is maintainable, because the liberty of the subject is a continuing matter and the legality of a detention is examined as at the date of the return. In Lallubhai Jogibhai Patel v. Union of India, (1981) 2 SCC 427, the Court held that the doctrine of constructive res judicata does not apply to a habeas corpus petition, so a ground not taken earlier may be taken later, though the Court will be astute to prevent abuse.

Second, constructive res judicata applies to writ proceedings and is the trap for the litigant. In Direct Recruit Class II Engineering Officers' Association v. State of Maharashtra, (1990) 2 SCC 715, the Court applied the principle to service litigation, holding that a party who has litigated a claim and lost cannot come back with a ground that was available and was not taken. The reason is the same public policy: a litigant is expected to bring his whole case at once, and a rule permitting instalments would make finality impossible.

Forward Construction Co. v. Prabhat Mandal (Regd.), Andheri, (1986) 1 SCC 100, confirms the position for both Articles 32 and 226, holding that a matter which might and ought to have been raised in the earlier proceeding is barred in the later one, and that the law discourages a litigant who withholds a ground for future use. Between them these two decisions, and not Direct Recruit alone, are what an examiner expects on constructive res judicata in this subject.

For full marks, cover: for legitimate expectation, its definition and source, the distinction between procedural and substantive expectation, how it arises, the three Indian decisions that fix its scope, its relation to promissory estoppel, the grounds on which it may be defeated, and the 2025 limitation; for fair hearing, the content of audi alteram partem set out as a list of components rather than a single rule, Ridge v. Baldwin and Maneka Gandhi as the two turning points, the exceptions, and the modern doctrine of useless formality.

Union of India v. Hindustan Development Corporation, (1993) 3 SCC 499, is the leading Indian exposition. The Railway Board changed its policy on the procurement of cast steel bogies, allocating a share of orders to smaller manufacturers at a rate different from that offered to the established suppliers, who complained that they had a legitimate expectation of the previous allocation. The Court refused relief but stated the doctrine at length. It held that a legitimate expectation may arise from an express promise or from the existence of a regular practice which the claimant can reasonably expect to continue; that it gives locus standi to seek judicial review and is a ground on which a decision may be attacked as arbitrary; but that it is not a right in the conventional sense, that a mere disappointment does not found a claim, and that the doctrine yields to an overriding public interest and to a bona fide change of policy. The Court also warned that the doctrine must not be stretched to the point where it fetters the administration's power to change its policy in the public interest.

Food Corporation of India v. M/s Kamdhenu Cattle Feed Industries, (1993) 1 SCC 71, supplies the constitutional link. The Corporation invited tenders for stocks of damaged foodgrains, and after receiving the highest tender invited all tenderers to a negotiation, in which a higher offer was obtained. The highest original tenderer complained. The Court held that every holder of public office must act fairly and non arbitrarily, that this is the requirement of Article 14, that a legitimate expectation is a relevant factor which the authority must take into account, but that where the expectation is outweighed by a larger public interest, here obtaining the best price for public property, the claim must yield. The case is useful because it states both halves: the expectation must be considered, and considering it may be enough.

The limits are as important as the doctrine. An expectation is defeated by a change of policy made bona fide in the public interest; by the fact that the expectation is contrary to law, since no expectation can be legitimate if giving effect to it would require an illegality; by the absence of any promise or practice, a pious hope not being enough; and by the claimant having no dealings with the authority at all. The most recent statement is K. Purushottam Reddy v. Union of India, 2025 INSC 894, where petitioners sought, under Article 32, an increase in the Assembly seats of Andhra Pradesh from 175 to 225 and of Telangana from 119 to 153, as had been done for the Union Territory of Jammu and Kashmir, relying on Article 14 and on the voters' legitimate expectation founded on Section 26 of the Andhra Pradesh Reorganisation Act, 2014. The Court dismissed the petitions, holding that Section 26 must yield to Article 170(3), which bars readjustment until the first census after 2026, and that the doctrine of legitimate expectation cannot be invoked to claim an enforceable right contrary to the constitutional arrangement. That is the sharpest modern limit: the doctrine operates within the constitutional and statutory scheme, never against it.

Notice must be given, and it must state the time, place and, above all, the substance of the case to be met, with sufficient particularity to allow an answer; a vague notice is no notice. Disclosure of material follows: the party must be told of the evidence and the report relied on, and a decision founded on material not disclosed is bad, which is the rule established in Dhakeswari Cotton Mills v. CIT, AIR 1955 SC 65, where an assessment based on information the assessee was never shown was set aside. An opportunity to state his case, orally or in writing according to the nature of the proceeding; an oral hearing is not invariably required, but it is required where the matter turns on credibility or where the statute so provides. An opportunity to rebut, including, where the proceeding is quasi judicial and evidence is taken, cross examination, though this is not an inflexible requirement in every administrative inquiry. Legal representation is not a general right and depends on the statute and on the gravity of the matter. The decision must be by the authority that heard, so that he who decides must hear, and it must be supported by reasons, which S.N. Mukherjee v. Union of India, (1990) 4 SCC 594, made a general requirement of decisions affecting rights.

A.K. Kraipak v. Union of India, (1969) 2 SCC 262, brought that change into Indian law. Naqishbund, the acting Chief Conservator of Forests in Jammu and Kashmir, was a member of the selection board for the Indian Forest Service while himself a candidate for selection. He withdrew when his own name was considered but sat when the claims of his rivals were considered, and he was placed at the top of the list. The Supreme Court quashed the selection, holding that the dividing line between quasi judicial and administrative functions is thin and being gradually obliterated, and that the principles of natural justice apply to administrative action as well, since the aim of both is to secure justice or to prevent a miscarriage of justice.

The exceptions must be stated, or the note is one sided. Natural justice is excluded or attenuated in cases of genuine urgency or emergency, as in the preventive seizure of adulterated food; where confidentiality or national security is genuinely engaged; where the statute expressly excludes it and the exclusion is constitutionally valid; in legislative action, including subordinate legislation, as against administrative action; where the person has no right or interest at stake; and under the doctrine of useless formality, where a hearing could have made no difference at all. That last exception is applied narrowly: in M.C. Mehta v. Union of India, (1999) 6 SCC 237, the Court held that the useless formality theory is an exception of limited application, and that a court must be satisfied that on the admitted or undisputed facts only one conclusion is possible before it will refuse relief on that ground.

The Indian authorities the examiner expects

Three Indian decisions carry this branch of the subject, and an answer resting only on the English cases is incomplete. In State of Orissa v. Dr. (Miss) Binapani Dei, AIR 1967 SC 1269, an Assistant Surgeon in the Orissa Medical Service was compulsorily retired after the State unilaterally altered her recorded date of birth on an enquiry at which she had no real opportunity to meet the material against her. The Supreme Court set the order aside and held that even an administrative order which involves civil consequences must be made consistently with the rules of natural justice. That proposition preceded A.K. Kraipak by two years and is the true Indian starting point for everything said above.

Conclusion. The two notes are the substantive and the procedural faces of fairness in the administrative process. Legitimate expectation identifies when a person is entitled to be treated in a particular way although he has no legal right, and the Indian courts have settled its shape: it arises from an express promise or a consistent practice, it is enforced readily in its procedural form as Navjyoti shows, sparingly in its substantive form, it must be weighed against a larger public interest as Kamdhenu holds, it yields to a bona fide change of policy under Hindustan Development Corporation, and after K. Purushottam Reddy in 2025 it can never be used to claim a right contrary to the constitutional arrangement itself. Fair hearing identifies what that treatment must contain, and its Indian history is a single movement of expansion: from a rule confined to bodies with a superadded duty to act judicially, through Ridge v. Baldwin and A.K. Kraipak, to Maneka Gandhi, where fairness of procedure became a constitutional requirement of Article 21 itself. What legitimate expectation most often produces is precisely a fair hearing, which is why the two are set together, and why an answer that treats them as unrelated topics loses the point of the question.