Administrative Law Notes | LL.B. (3 years) Semester 3 | Mumbai University | munotes

In one line

Administrative law exists because the modern State does far more than the old State did, and somebody has to keep the officials who do it inside the law.

In the words a student can write in an exam: administrative law is the body of law that grew out of the transformation of the State from a laissez faire or police State, whose functions were limited to defence, order and the collection of revenue, into a welfare State which regulates, licenses, taxes, subsidises, employs and provides, and which therefore had to be given vast powers over the individual; administrative law is the law that controls the exercise of those powers.

Two words to get out of the way first

Laissez faire is French and means, literally, "let do", or in plainer English "leave it alone". As a theory of government it says the State should keep out of economic and social life: it should defend the country, keep the peace, enforce contracts, punish crime, and otherwise leave people to arrange their own affairs. A State organised on that view is sometimes called a police State, which does not mean a State full of policemen. It means a State whose whole business is police in the old sense of keeping order.

Welfare State means the opposite: a State that accepts responsibility for the material well being of its people. It runs schools and hospitals, fixes minimum wages, distributes food, builds housing, insures workers against accident and old age, controls prices, and regulates the industries that supply water, power and transport.

Neither expression is a term of art with a fixed legal meaning. They are labels for two ends of a range, and every real State sits somewhere between them. But the movement from one end towards the other is the single fact that produced this subject, so the labels are worth knowing precisely.

Why the law had to change

Picture the government of a district in India in 1850. It collects land revenue, keeps a police force, runs the courts, and maintains a few roads. Almost every dispute a citizen can have with it is about money or about criminal process, and the ordinary civil and criminal courts can handle both. There is very little for a separate body of law to do.

Now picture the same district today. The government decides who gets a ration card, who gets a licence to run a chemist's shop, whether a factory may discharge effluent into the river, what a landlord may charge, which contractor gets the road tender, whether a teacher is dismissed, how much compensation is paid for land acquired for a bypass, and whether a stall is removed from a footpath. Each of those is a decision by an official, taken under a statute, affecting somebody's livelihood.

In one line

Administrative law is the law that governs the administration: who the administrative authorities are, what powers they have, how those powers must be exercised, and what remedies a person has when they are exercised wrongly.

In the words a student can write in an exam: administrative law is that branch of public law which determines the organisation, powers and duties of administrative authorities, prescribes the procedure they must follow, and provides the remedies available to a person aggrieved by their action or inaction.

Why the definition is contested

Every textbook gives a different definition, and a student who has met five of them is entitled to ask why nobody can agree. The reason is that writers disagree about what the subject is for, and a definition is a compressed answer to that question.

Put crudely, there are two instincts. One says the subject is about the administration: describe the machinery, the departments, the corporations, the tribunals, what they do and how they are organised. The other says the subject is about control of the administration: the machinery is the business of political science, and law comes in only when we ask what limits the administration and what a citizen can do about it.

That disagreement is the same one that appears in [Red Light and Green Light Theories] as a debate about the purpose of judicial review. Definitions and theories are two faces of the same question, and saying so in an answer is worth marks.

The definitions, and what each one gets right

Sir Ivor Jennings defined administrative law as the law relating to the administration, determining the organisation, powers and duties of administrative authorities. This is the widest and the most quoted definition, and it is the natural starting point.

Its strength is that it is complete: nothing about the administration falls outside it. Its weaknesses are two, and an examiner expects both. First, it is so wide that it swallows constitutional law, since the Constitution also determines the organisation and powers of authorities. Second, it says nothing about procedure or about remedies, which is a strange omission, because the rules on hearing, bias and reasons, and the writs, are what practising administrative lawyers spend their time on.

K.C. Davis, writing in the United States, defined it as the law concerning the powers and procedures of administrative agencies, including especially the law governing judicial review of administrative action. This corrects Jennings on both counts: it brings in procedure expressly, and it puts judicial review at the centre.

Its weakness is the opposite one. By concentrating on agencies and on review, it leaves out the rule making that the administration does, and it says little about the internal controls, such as departmental appeals, that dispose of the overwhelming majority of grievances before any court is involved.

In one line

A source of administrative law is any place a rule of administrative law comes from, and in India there are six: the Constitution, statutes, delegated legislation, judicial decisions, administrative directions, and reports and custom.

In the words a student can write in an exam: the sources of Indian administrative law are the Constitution, which supplies both the powers of the administration and the writ jurisdiction that controls it; statutes conferring administrative powers; rules, regulations and bye laws made under those statutes; judicial decisions, which are the principal source, since the doctrines of the subject are judge made; administrative circulars and directions, which bind the administration internally; and the reports of commissions and committees, together with settled administrative practice.

Why this is not a list to be memorised

A student who writes the six words in an exam and stops has given half an answer. The examiner wants to see that you know what each source contributes and which source wins when two conflict, because that is the part that decides real cases.

The order below is a hierarchy. Anything lower down that contradicts something higher up is void to that extent. That single sentence organises the whole chapter.

1. The Constitution

The Constitution is the primary and the highest source, and it plays two opposite roles at once, which is the point worth making.

It creates administrative power. Article 53 vests the executive power of the Union in the President and Article 154 vests that of a State in the Governor. Articles 73 and 162 fix how far that executive power extends. Part XIV, Articles 308 to 323, provides for the services and the Public Service Commissions, which is the subject of [Civil Services in India: Nature, Organisation, Powers and Functions]. Articles 323A and 323B authorise Parliament and the State legislatures to set up administrative tribunals, which is where [Administrative Tribunals: Nature and Characteristics] begins.

It controls administrative power. Article 13 makes any law inconsistent with the fundamental rights void, and the definition of "law" there includes rules, regulations, notifications and orders, so delegated legislation is caught. Article 14 forbids arbitrary State action and is the constitutional root of the grounds of review. Article 19 limits restrictions on the freedoms. Article 21 requires a procedure established by law before life or personal liberty is taken, and is the constitutional home of natural justice. Article 300 makes the Union and the States suable, which is the foundation of [Tortious Liability of the State], and Article 299 governs their contracts.

Above all, Articles 32 and 226 give the Supreme Court and the High Courts power to issue the writs, and Article 227 gives the High Courts superintendence over all courts and tribunals in their territory. Without those three Articles there would be no effective administrative law in India at all, because there would be no remedy. They are treated in [Judicial Remedies against Administrative Arbitrariness: the Writs].

In one line

The red light theory says the job of administrative law is to stop the State doing wrong; the green light theory says its job is to help the State do right.

In the words a student can write in an exam: red light theory and green light theory are two competing views of the function of administrative law, associated with Carol Harlow and Richard Rawlings, who used the traffic light image to describe them. The red light view treats administrative power as inherently dangerous and administrative law as a system of external control, chiefly through the courts, whose purpose is to keep the executive within legal limits and protect individual liberty. The green light view treats administrative power as the means by which social goals are achieved and administrative law as a way of improving the quality of administration from within, chiefly through good procedures and internal controls, with litigation as a last resort. The amber light view is the middle position now generally accepted.

Why a chapter on theories at all

Because the whole subject makes more sense once you see that the cases are pulling in two directions, and because MU prints these two theories as an express head of topic 1.1. They are also the part of the syllabus that competing notes reduce to four lines, which makes them a good place for a student to gain marks.

A theory here is not a rule you can cite. It is an answer to the question what is administrative law for, and different answers produce different decisions on the same facts. When a court says that it will not sit in appeal over policy, it is thinking in green light terms. When it says that unchecked discretion is the enemy of liberty, it is thinking in red light terms.

The red light theory

The image is a traffic light showing red: the law's function is to stop.

The theory grows out of nineteenth century liberal thought and out of the fear that concentrated executive power threatens individual freedom. Its intellectual ancestor is Dicey, whose account of the rule of law is treated in [Rule of Law], and its clearest modern exponent is Wade, who treated the control of governmental power as the primary purpose of the subject.

Its propositions are these.

1. The State is a necessary evil and its powers are inherently dangerous. Every grant of power to an official is a subtraction from the liberty of the subject.
2. The primary purpose of administrative law is therefore to control the administration and to protect the individual against it.
3. The proper controller is the court, because it is independent of the administration, and the ordinary courts are preferred to specialised tribunals.
4. The preferred technique is external and after the event: review the decision once taken, and quash it if it is unlawful.
5. Administrative discretion is a problem to be minimised, confined and structured, because discretion is where arbitrariness enters.

In one line

Administrative law grew in four different ways in four countries: England denied it existed and then built it out of the writs; the United States built it out of statutes and regulatory commissions; France built a separate system of courts for it two centuries before anybody else; and India inherited the English machinery and then gave it a constitutional foundation the English system does not have.

Why the comparison is on the syllabus

Because each system solved the same problem differently, and the differences expose what is really at stake. The central question in every country was the same: should disputes between a citizen and the administration be decided by the ordinary courts, or by special courts of their own? England said the ordinary courts, France said special courts, the United States said ordinary courts supervising special agencies, and India said the ordinary courts with a constitutional guarantee.

England

The long denial. Until well into the twentieth century English lawyers denied that England had any administrative law at all. The denial is Dicey's, and it rests on his second meaning of the rule of law, treated in [Rule of Law]: that in England every man is subject to the ordinary law administered by the ordinary courts, and there is no separate body of law and no separate set of courts for officials. Dicey contrasted this with the French system, which he misunderstood, and treated the absence of administrative law as a mark of English liberty.

The denial was never accurate. The Crown and its officers had always been subject to control through the prerogative writs, which the Court of King's Bench issued to keep inferior bodies within their jurisdiction. Certiorari, prohibition, mandamus, habeas corpus and quo warranto are all medieval in origin. What England lacked was not administrative law but the name.

The growth. From the middle of the nineteenth century the modern State arrived: the Poor Law, factory legislation, public health, compulsory education, and later national insurance and the National Health Service. Each statute created officials with powers and disputes to be decided, and Parliament created tribunals to decide them rather than sending the work to the courts.

Three landmarks are worth naming.

• The Committee on Ministers' Powers, which reported in 1932, was appointed because of alarm at delegated legislation and administrative adjudication. It found both to be necessary but recommended safeguards, and it is discussed in [Delegated Legislation and Why It Grew].
• The Franks Committee on tribunals and inquiries, which reported in 1957, laid down that tribunal procedure should be marked by openness, fairness and impartiality, and led to the Tribunals and Inquiries Act.
• The revival of judicial review from the 1960s onwards, which restored natural justice to administrative decisions and developed the modern grounds of review.

In one line

Constitutional law describes the structure of the State at rest; administrative law describes the State in motion.

In the words a student can write in an exam: constitutional law deals with the structure, composition and powers of the principal organs of the State and with their relations to one another and to the citizen, while administrative law deals with the exercise of power by administrative authorities within that structure, the procedures they must follow and the remedies available against them. Both are branches of public law, both draw on the same constitutional provisions, and in India the distinction between them is one of emphasis rather than of subject matter.

The classical distinction

The traditional way of putting it, associated with English writers such as Holland, is that constitutional law describes the State at rest and administrative law the State in motion.

The image is exact enough to be worth keeping. Constitutional law tells you that there is a President in whom the executive power of the Union is vested, that there is a Council of Ministers, that there are Public Service Commissions, and that the High Courts may issue writs. It is a description of the machine. Administrative law tells you what happens when the machine is switched on and a particular officer refuses a particular licence: whether he had the power, whether he heard the applicant, whether his reasons were relevant, and what the applicant can do.

A second traditional formulation is that constitutional law is concerned with the organs of government and administrative law with the functioning of those organs. A third is that constitutional law deals with structure and powers, administrative law with procedure and control.

Where they overlap, and why the overlap is unavoidable

Every distinction offered above breaks down at the edges, and an answer that gives the distinction without the overlap is a weak answer.

Consider Article 311, which protects a civil servant against dismissal without an inquiry in which he has been informed of the charges and given a reasonable opportunity of being heard. Is that constitutional law or administrative law? It is in the Constitution, so it is constitutional law. It prescribes the procedure an administrative authority must follow before taking a decision about an individual, so it is administrative law. The honest answer is that it is both, and it is treated in this book in [Civil Services in India: Nature, Organisation, Powers and Functions].

The same is true of Article 14, which supplies the standard of non-arbitrariness that most of Module III applies; of Article 21, which is the constitutional foundation of natural justice; of Article 12, which decides whether a public corporation can be proceeded against by writ at all, as in [Liability of and Control over Public Corporations]; and of Articles 32 and 226, which supply every remedy the subject has.

In one line

The rule of law means that everybody, including the government, is under the law, and that nobody is punished except for a breach of law proved before an ordinary court.

In the words a student can write in an exam: the rule of law is the principle that the State and its officers possess no power except that which the law gives them, that all persons are equal before the law and subject to the ordinary law administered by the ordinary courts, and that the law itself must be certain, general and applied by an independent judiciary. In India it is not merely a political ideal: it is given legal effect by Articles 13, 14, 20, 21, 32 and 226, and it has been held to form part of the basic structure of the Constitution.

The origin of the phrase

The idea is ancient. The medieval statement of it is Bracton's, that the King is under no man but under God and the law, and Sir Edward Coke used it against the Stuart kings. Its classical modern statement is by Professor A.V. Dicey, in his Introduction to the Study of the Law of the Constitution, published in 1885. The Latin phrase Dicey's idea translates is la principe de legalite, the principle of legality: that government must act according to law and not arbitrarily.

Dicey did not invent the rule of law; he formulated it in three propositions and made those propositions the standard exam question. It is his three meanings that are examined, and each has to be stated and criticised, because an answer that recites all three and criticises none gets half the marks.

Dicey's three meanings

First meaning: supremacy of law, as against arbitrary power

No person may be punished except for a distinct breach of law established in the ordinary manner before the ordinary courts. Government must show legal authority for whatever it does. Wide discretionary power in the hands of officials is, on this view, the opposite of the rule of law, because where there is discretion there is room for arbitrariness.

Criticism. This meaning is too absolute for any modern State. As [The Rise of the Administrative State] shows, the welfare State cannot function without conferring discretion: the licensing authority must be allowed to judge whether premises are suitable, and no statute can specify every case in advance. The modern answer is not to abolish discretion but to structure and control it, which is what [Administrative Discretion and Why It Is Conferred] and the review grounds that follow it are about. Discretion is compatible with the rule of law if it is conferred by law, exercised for the purpose for which it was given, and subject to review.

In one line

Separation of powers is the principle that the three functions of government, making law, executing law and deciding disputes, should be in three different hands, so that no one body has the whole power of the State.

In the words a student can write in an exam: the doctrine of separation of powers, in its classical form stated by Montesquieu, requires that the legislative, executive and judicial functions of government be vested in three separate organs, that no organ exercise the functions of another, and that no person be a member of more than one organ. It is applied strictly in the United States, hardly at all in its formal sense in England, and in India in a modified form under which the functions of the organs are demarcated but not rigidly separated, the doctrine having been held to form part of the basic structure of the Constitution.

Where the doctrine comes from

The idea is Aristotle's in origin and Locke's in modern form, but the classical statement is by Baron de Montesquieu, in The Spirit of the Laws, published in 1748. Montesquieu, writing about England as he understood it, argued that liberty is destroyed when the same body makes the law, executes it and judges breaches of it, because such a body can make a tyrannical law and enforce it tyrannically without any check.

His proposition has three limbs, and an examiner expects all three because most students give only the first.

1. Function: the same person or body should not exercise more than one of the three functions.
2. Personnel: the same person should not be a member of more than one of the three organs.
3. Non-interference: one organ should not interfere with or control another in the discharge of its functions.

Montesquieu's account of England was inaccurate even in 1748: the Lord Chancellor sat in all three branches, and the House of Lords was both a legislative chamber and the highest court. The doctrine was a description of what he thought he saw, and it became prescriptive afterwards.

Why the doctrine matters for administrative law

Because administrative law exists in the space the doctrine says should not exist.

An administrative authority that makes rules is performing a legislative function. One that decides whether a licence should be revoked is performing a judicial function. It is doing both while being part of the executive. On a strict application of Montesquieu, delegated legislation and administrative adjudication are both unconstitutional, and the whole of Modules II and III would collapse.

So every legal system has had to decide how much separation it will insist on. The three answers below are the syllabus.

In one line

Administrative authorities do three different kinds of thing, make rules, decide disputes and simply act, and which kind a particular act belongs to decides what legal controls attach to it.

In the words a student can write in an exam: the functions of the administration are classified as legislative, where the authority lays down a general rule for the future; judicial or quasi-judicial, where it determines a question affecting the rights of a particular person after considering the matter; and administrative or executive, where it takes action to give effect to policy. To these is usually added the ministerial function, where the authority has no choice at all. The classification matters because different consequences follow from it: the duty to observe natural justice, the duty to give reasons, the availability of certiorari and the standard of review all depend on it.

Why bother classifying at all

This is the question a student should ask, because a classification with no consequences is a waste of a chapter. There are five consequences and they are the whole reason the topic is on the syllabus.

Note the second row and the third. Before 1969, everything turned on getting a decision into the quasi-judicial box, because natural justice attached only there. After A.K. Kraipak v. Union of India, AIR 1970 SC 150, the difference between the two narrowed a great deal, and the practical importance of the classification fell with it. That change is the story of this chapter and the next.

The tripartite classification

The syllabus calls these the tripartite functions of administration, and the three are these.

1. The legislative function, or rule making

The authority lays down a general rule of conduct, for the future, applying to a class of persons or situations, under power delegated by a statute. The rule binds people who were never heard and are not identified. Examples: rules made under an Act, regulations of a statutory corporation, municipal bye laws, a notification fixing the price of a drug, a scheme framed under a welfare statute.

The marks of a legislative act are: generality of application, prospectivity, and the absence of any particular person whose case is being determined.

In one line

A legislative act makes a rule for the future for a class; a judicial act decides a dispute between parties on evidence; a quasi-judicial act decides a question affecting one person's rights where the law requires a judicial approach; and an executive act carries policy into effect.

In the words a student can write in an exam: the three distinctions the syllabus requires are those between legislative and judicial functions, between legislative and executive functions, and between judicial and quasi-judicial functions. The tests are, respectively: whether the act lays down a general rule for the future or determines an existing dispute; whether it creates a new rule of general application or applies an existing rule to a particular case; and whether the body is a court bound to decide a lis between parties according to law or a body outside the judicial hierarchy which is nevertheless under a duty to act judicially.

Why three distinctions and not one

Because MU prints three, and because each is drawn for a different reason.

The legislative against judicial distinction decides whether natural justice attaches, and whether the act is challenged as ultra vires the parent Act or quashed by certiorari. The legislative against executive distinction decides whether the act had to be published, whether it binds a court as law, and whether a person adversely affected by it can complain of not being heard. The judicial against quasi-judicial distinction decides whether the body is a court, which in turn affects appeal, contempt, and the scope of Article 227.

Distinction 1: legislative and judicial functions

The test

A legislative act creates a rule; a judicial act applies an existing rule to found facts and determines rights.

Four practical indicators follow, and they are what a student should apply to a fact pattern.

The classic illustration

If a Government fixes the maximum price of a drug at forty rupees for all manufacturers, that is legislative: prospective, general, and nobody's individual case is decided. If a Government determines that a particular manufacturer has overcharged and must refund a sum, that is judicial or quasi-judicial: a named person, past facts, an existing rule applied.

The point is not the size of the group affected but how the group is defined. A rule that applies to "all manufacturers of a drug" is legislative even if there is only one such manufacturer, because the rule is expressed by description. A determination about "Sharma Pharmaceuticals Private Limited" is not legislative even though it names only one, because it is expressed by identity.

In one line

Delegated legislation is law made by somebody other than the legislature, under a power the legislature has given them.

In the words a student can write in an exam: delegated legislation, also called subordinate or subsidiary legislation, means the rules, regulations, bye laws, orders, notifications and schemes made by the executive or by another authority under a power conferred by an Act of the legislature; it has the force of law and binds citizens and courts in the same way as the parent Act, provided it is within the limits of the power conferred and does not offend the Constitution.

Two senses of the expression

The phrase is used in two ways and a good answer separates them at the start.

As a power: the authority given by the legislature to somebody else to make rules. In this sense we say that an Act "contains a delegation".

As a product: the body of rules actually made under that authority. In this sense we say that a rule "is delegated legislation".

Both senses are correct and examiners use both.

The vocabulary, defined

Students meet six words for the same kind of thing and are rarely told the difference. There is no rigid legal distinction, and the label used depends on the parent Act, but the ordinary usage is this.

Parent Act is the Act which confers the power. It is also called the enabling Act or the delegating statute. Ultra vires is Latin for "beyond the powers" and means that the maker has gone outside what the parent Act allowed. Intra vires is its opposite.

Delegated legislation is law

This is the proposition that makes the topic serious, and it has three consequences.

1. It binds citizens exactly as an Act does, and a breach can be punished if the parent Act provides a penalty.
2. It binds the courts, which must apply it unless it is shown to be invalid.
3. It is "law" within Article 13(3)(a) of the Constitution, which defines law as including any ordinance, order, bye law, rule, regulation, notification, custom or usage having the force of law. So a rule inconsistent with a fundamental right is void, just as an Act would be.

In one line

The question is whether a legislature which has been given the power to make law may hand part of that job to somebody else, and the Indian answer is yes, but not the essential part.

In the words a student can write in an exam: the Constitution vests legislative power in Parliament and the State legislatures by Articles 245 and 246, and the question of the constitutional validity of delegated legislation is whether a body so entrusted may delegate that power to the executive. The Supreme Court answered the question in the Delhi Laws Act reference: delegation is permissible, because a modern legislature cannot function without it, but the legislature must itself determine the legislative policy and formulate it as a rule of conduct, and it cannot delegate that essential legislative function or abdicate its own position.

Why there is a problem at all

Two maxims and one constitutional argument produce the difficulty, and the answer has to meet all three.

The maxim delegatus non potest delegare means that a person to whom power has been delegated cannot delegate it further. If the people, through the Constitution, delegated legislative power to Parliament, Parliament cannot pass it on.

The separation of powers argument is that legislative power is vested in the legislature, so the executive exercising it is exercising a function that belongs to another organ: [Separation of Powers].

The democratic argument is that law made by an unelected department has not been consented to by the governed.

Against these stand the practical reasons in [Delegated Legislation and Why It Grew]. Every legal system has had to find a compromise, and the American compromise is the requirement of an intelligible principle, while the English position is that Parliament being sovereign may delegate as widely as it wishes. India's compromise came out of a single reference in 1951.

The Indian position: the Delhi Laws Act reference

This is the foundation case for the whole of Module II. It is long, there are seven separate opinions, and the standard student mistake is to state a single ratio it does not have. What can be stated is the proposition the later cases have taken from it.

Facts. In re The Delhi Laws Act, 1912, AIR 1951 SC 332, was a reference by the President under Article 143(1), which allows the President to refer a question of law or fact of public importance to the Supreme Court for its opinion. Doubts had arisen after a decision of the Federal Court about laws delegating legislative power to the executive, and three provisions were referred.

• Section 7 of the Delhi Laws Act 1912 allowed the Provincial Government, by notification, to extend to Delhi, with such restrictions and modifications as it thought fit, any enactment in force in any part of British India.
• Section 2 of the Ajmer-Merwara (Extension of Laws) Act 1947 did the same for Ajmer-Merwara.
• Section 2 of the Part C States (Laws) Act 1950 allowed the Central Government to extend to any Part C State, with such restrictions and modifications as it thought fit, any enactment in force in a Part A State, and further provided that an enactment so extended might provide for the repeal or amendment of any corresponding law then applicable to that Part C State.

In one line

Delegation becomes excessive, and therefore unconstitutional, when the legislature leaves the delegate to decide the policy instead of deciding it itself.

In the words a student can write in an exam: the doctrine of excessive delegation holds that although a legislature may delegate the power to make rules, it may not delegate its essential legislative function, which is the determination of the legislative policy and its formulation as a rule of conduct; a provision which confers power without laying down any policy, standard or guideline for its exercise is void as an excessive delegation, and the test is whether the Act, read as a whole, supplies guidance by which the delegate's power is canalised.

The question the doctrine answers

The previous chapter established that delegation is permitted in India. This chapter asks the next question: how much?

The answer cannot be a quantity. There is no rule that a legislature may delegate seventy per cent of a subject and no more. What the courts ask instead is a question about kind: has the legislature done the legislating, or has it passed the legislating on?

That question is answered by looking for policy. If the Act contains the policy, the delegate is administering it, and the delegation is permissible however wide the rule making power looks. If the Act contains no policy, the delegate must invent one, and inventing policy is legislating.

The two formulations of the test

Both appear in the cases and they mean the same thing.

The essential legislative function test. From In re The Delhi Laws Act, 1912, AIR 1951 SC 332: the essentials of a legislative function are the determination of the legislative policy and its formulation as a rule of conduct, and those essentials are the characteristics of a legislature itself.

The policy and guidelines test. Does the Act lay down a policy, and does it give the delegate a standard or guideline by which the power is to be exercised? A power which is, in the language of the cases, uncanalised, meaning without banks to keep it in a channel, is bad.

Permissible delegation

The following may be delegated, and a student should be able to list them.

1. Commencement. Power to appoint the day on which an Act, or a provision of it, comes into force.
2. Extension and application. Power to extend the operation of an Act to an area or to a class, where the Act supplies the criteria.
3. Inclusion and exclusion. Power to add to or subtract from a schedule, provided the Act states the basis on which the addition is to be made.
4. Suspension. Power to suspend the operation of a provision in stated circumstances.
5. Prescribing detail. Forms, fees, registers, returns, procedure, qualifications and the machinery of administration.
6. Modification. Power to apply an existing law with modifications, so long as the modifications do not touch the policy of the Act.
7. Removal of difficulties. Power to make orders removing difficulties in giving effect to the Act, if narrowly framed and time limited.

In one line

Delegated legislation is classified in three different ways, by the title of the instrument, by the width of the discretion given, and by the purpose the power serves, and the classification that matters legally is the second.

In the words a student can write in an exam: delegated legislation may be classified by title, into rules, regulations, bye laws, orders, notifications and schemes; by the nature and extent of the power conferred, into normal delegation and exceptional delegation, the latter including Henry VIII clauses, sub-delegation and power to modify or repeal a statute; and by the purpose for which the power is given, into enabling, extension, inclusion, exclusion, suspending, dispensing, taxing, modifying and framing legislation. Conditional legislation is treated separately, because in it no legislative power is delegated at all.

Classification 1: by title

This is the classification students meet first and it is the least important legally, because the label a parent Act chooses does not decide the instrument's legal character. It is nevertheless examinable, and it is set out in [Delegated Legislation and Why It Grew] with a table of the six titles: rule, regulation, bye law, order, notification and scheme.

The single point worth adding here is that the label does not determine the effect. An "order" may be legislative if it is general and prospective, or administrative if it decides an individual case. What decides its character is what it does, not what it is called: [Distinguishing Legislative, Judicial, Quasi-judicial and Administrative Functions].

Classification 2: by the width of the power, which is the one that matters

Normal delegation

The parent Act states the policy and confers a power that is limited and clearly defined. The delegate fills in forms, fees, procedure, technical standards and machinery. The great bulk of delegated legislation is of this kind, and it raises no constitutional question.

Normal delegation is sometimes subdivided into:

• Positive delegation, where the Act states what the delegate may do.
• Negative delegation, where the Act states what the delegate may not do, for example by providing that rules shall not impose a fee exceeding a stated amount.

Exceptional delegation

Delegation is called exceptional where the power goes beyond filling in detail. Four kinds are recognised, and each carries its own risk of being held to be excessive delegation.

1. Power to legislate on matters of principle. Where the Act allows the delegate to decide what the policy shall be. This is the vice in Hamdard Dawakhana (Wakf) Lal Kuan v. Union of India, AIR 1960 SC 554.

2. Power to impose a tax. Taxation is a legislative function in the fullest sense, since Article 265 provides that no tax shall be levied or collected except by authority of law. A power to fix a rate may be delegated if the Act sets the limits and the principle.

In one line

A court can strike down a rule on two grounds: that the Act which authorised it was bad, or that the rule goes beyond what the Act authorised.

In the words a student can write in an exam: judicial control over delegated legislation operates through the doctrine of ultra vires. A rule is void if the parent Act itself is unconstitutional, whether for want of legislative competence, for violation of a fundamental right or for excessive delegation; and it is void if the rule, though made under a valid Act, is substantively ultra vires because it exceeds the power conferred, is inconsistent with the parent Act or another statute, violates a fundamental right, is unreasonable, is made in bad faith or is unconstitutionally retrospective, or is procedurally ultra vires because a mandatory procedural requirement such as publication or consultation has not been complied with.

The structure of the whole topic

Everything in this chapter fits under one Latin phrase. Ultra vires means "beyond the powers". A rule is valid only if the person who made it had the power to make it and stayed within that power.

That gives the two branches, and every ground below belongs to one of them.

Keep these apart. The commonest error in this topic is to answer a question about a rule by discussing excessive delegation, which is an attack on the Act. See [Excessive Delegation: Permissible and Impermissible Delegation].

Branch 1: the parent Act is invalid

Three grounds, each of which destroys the rules made under the Act.

1. Want of legislative competence. The legislature had no power to legislate on the subject under the Seventh Schedule, or legislated beyond its territory contrary to Article 245.

2. Violation of a fundamental right. The Act itself offends Part III and is void under Article 13(2).

3. Excessive delegation. The Act delegated the essential legislative function or amounted to abdication: In re The Delhi Laws Act, 1912, AIR 1951 SC 332, and Hamdard Dawakhana (Wakf) Lal Kuan v. Union of India, AIR 1960 SC 554, treated in the previous two chapters.

Branch 2: the rule is invalid although the Act is good

This is where most litigation happens, and it divides into substantive and procedural ultra vires.

Substantive ultra vires

1. The rule goes beyond the power conferred. The enabling section allows rules on A, and the rule deals with B. This is the primary ground and it is a pure question of construction: read the enabling section, read the rule, and ask whether the rule falls within the words.

In one line

The legislature keeps control over the rules its delegates make in three ways: by how it drafts the enabling section, by requiring the rules to be laid before it, and by a committee that scrutinises them.

In the words a student can write in an exam: legislative control over delegated legislation is exercised at three stages. At the stage of delegation, by confining the power in the enabling section, stating the policy and prescribing procedural conditions. After the rules are made, by the laying procedure, under which rules must be placed before the House in one of three forms, simple laying, laying subject to negative resolution, or laying subject to affirmative resolution. And continuously, by the Committee on Subordinate Legislation of each House, which examines whether the rules made are within the power conferred and reports to the House.

Why legislative control matters at all

Because the legislature is the body that gave the power away, and it is the only body that can take it back. Judicial control is powerful but it is occasional: it works only when somebody with standing and money brings a case, and it can only ask whether the rule is lawful, never whether it is wise. Legislative control is continuous and can ask both questions.

This is the green light half of the topic. Judicial control in the previous chapter is the red light half: [Red Light and Green Light Theories].

Stage 1: control at the moment of delegation

The most effective control is exercised before any rule exists, by the way the enabling section is drafted. A legislature which drafts carefully has less to police afterwards.

The devices are these.

1. State the policy in the Act, which is also what saves it from excessive delegation.
2. Confine the subject matter by listing the matters on which rules may be made, rather than conferring a power to make rules generally.
3. Set outer limits, for example by providing that a fee prescribed by rules shall not exceed a stated sum, or that a penalty shall not exceed a stated amount.
4. Prescribe the procedure, requiring previous publication of a draft, consultation with a named body, or the sanction of a superior authority.
5. Require laying before the House.
6. Limit exceptional powers in time, as with a removal of difficulties clause exercisable for two years only.

The Committee on Ministers' Powers made recommendations to this effect in 1932, and they remain the standard: define the powers precisely, use exceptional powers only where essential and for a limited period, and provide for scrutiny.

Stage 2: the laying procedure

Laying means placing a copy of the rule before the House. The parent Act says whether laying is required and in what form. There are three forms and the difference between them is examinable.

In one line

Besides the courts and the legislature, delegated legislation is controlled by the requirements that it be published before it binds anybody, that affected interests be consulted before it is made, and by the ordinary pressures of publicity and public opinion.

In the words a student can write in an exam: the other controls over delegated legislation are procedural and political. The procedural controls are publication, without which a rule does not become operative, and antecedent publicity and consultation, under which a draft is published or affected interests or an expert body are consulted before the rule is made. The political controls are the ordinary accountability of a minister to the legislature, the work of the press and of affected trade and professional bodies, and, in modern practice, pre-legislative consultation policy and the disclosure obligations of the Right to Information Act 2005.

Why a third chapter of controls

Because the first two are both incomplete. Judicial control needs a litigant and can ask only about legality. Legislative control comes after the rule is already in force and is defeated by volume. The controls in this chapter operate before the rule exists, which is where a green light theorist would say control belongs: [Red Light and Green Light Theories].

Publication

The rule

A law that nobody can find does not bind. This is not a technicality but a principle of natural justice, and the leading Indian case says so in terms.

Facts. Harla v. State of Rajasthan, AIR 1951 SC 467, concerned the Jaipur Opium Act. The Council of Ministers appointed by the Crown Representative for the government of the Jaipur State passed a Resolution in 1923 purporting to enact that Act, but it was neither promulgated nor published in the Gazette nor otherwise made known to the public. The Jaipur Laws Act 1923, in force from 1 November 1924, provided that the law to be administered by the courts should include the regulations then in force and enactments thereafter passed by the State and published in the Official Gazette. In 1938 the Opium Act was amended by adding a clause that it should come into force from 1 September 1924. The appellant was convicted under section 7 of it.

Held. Natural justice requires that before a law can become operative it must be promulgated or published. It must be broadcast in some recognisable way so that all men may know what it is, or at least there must be some special rule or regulation or customary channel by or through which such knowledge can be acquired with the exercise of due and reasonable diligence. The mere passing of the Resolution of the Council, without further publication or promulgation, was not sufficient to make the law operative, so the Jaipur Opium Act was not a valid law. Nor was it saved by the Jaipur Laws Act 1923, since it was not a valid law in force on 1 November 1924, and the addition in 1938 of a clause that it should come into force in 1924 was of no use.

In one line

Natural justice means that before a decision is taken against you, the person deciding must be impartial and must let you put your side.

In the words a student can write in an exam: the principles of natural justice are the minimum standards of fair procedure which the law requires of any authority deciding a matter affecting the rights or interests of a person. They rest on two maxims, nemo judex in causa sua, that no one shall be a judge in his own cause, and audi alteram partem, that both sides shall be heard, to which Indian law has added a third requirement, that a decision be supported by reasons. They are not embodied rules, they are not a fixed code, and what they require in a given case depends on the facts, the framework of the statute and the constitution of the body deciding.

The two maxims, translated

Nemo judex in causa sua means "no one should be a judge in his own cause". Sometimes written nemo judex in re sua, "no one a judge in his own matter". It is the rule against bias, and it is treated in [The Rule Against Bias].

Audi alteram partem means "hear the other side". It is the right to notice and a hearing, and it is treated in [Audi Alteram Partem].

Natural justice itself is a phrase of some age and no precision. It does not mean the law of nature, and it has nothing to do with natural law in the philosophical sense. In modern English and Indian usage it means simply fair procedure, and courts increasingly use the phrase "the duty to act fairly" to say the same thing without the misleading adjective.

Where the principles come from

Three sources, and an answer should name all three because the Indian position uses all three.

1. The common law. English courts developed the rules to control inferior tribunals and, later, administrative bodies. The old cases put it picturesquely, saying that even God heard Adam before expelling him from Eden.

2. The Constitution. In India the principles have a constitutional home, which they do not have in England, and this is the difference to emphasise.

• Article 14 forbids arbitrary State action. A decision taken without hearing the person affected is arbitrary, so unfairness of procedure is a breach of Article 14.
• Article 21 requires that a person be deprived of life or personal liberty only by a procedure established by law, and that procedure must be fair.
• Article 311 writes a specific hearing requirement into the Constitution for civil servants, requiring an inquiry in which the officer is informed of the charges and given a reasonable opportunity of being heard: [Civil Services in India: Nature, Organisation, Powers and Functions].

In one line

The person who decides must not have an interest in the outcome, and must not appear to have one.

In the words a student can write in an exam: the rule against bias, expressed in the maxim nemo judex in causa sua, requires that no person shall be a judge in his own cause, and that a decision be taken by an authority free from any interest, personal, pecuniary or otherwise, in the subject matter or in the parties. It is not necessary to prove that the decision was in fact influenced; it is enough that there was a real likelihood, or in some formulations a reasonable suspicion, of bias, because justice must not only be done but must manifestly be seen to be done.

Why appearance is enough

This is the part students find odd, so it is worth meeting head on. If bias had to be proved to have actually operated, the rule would be almost useless: what happened in the mind of a decision maker cannot be shown, and he will always say he was not influenced.

The rule therefore looks at the position from outside. If a reasonable person, knowing the facts, would think that the decision maker might not bring an impartial mind to bear, the decision goes, whether or not he was in fact impartial. The classical English formulation is that justice should not only be done but should manifestly and undoubtedly be seen to be done.

The consequence, which is worth stating in an answer, is that a finding of bias is not a finding of dishonesty. A judge who owns a few shares in a company before him is not accused of corruption. He is simply disqualified.

The kinds of bias

1. Pecuniary bias

Any financial interest in the outcome, however small, disqualifies. There is no de minimis rule, meaning there is no minimum below which the interest is ignored. A shareholding in a company that is a party, a fee that depends on the outcome, or a share in a fine imposed will each disqualify.

This is the strictest of the categories, and the reason is practical: money is measurable, so no inquiry into degree is needed.

2. Personal bias

Arising from a relationship between the decision maker and a party: kinship, friendship, hostility, a business relationship, a professional connection, or the fact that the decision maker is himself a rival for what is being decided.

Personal bias does not disqualify automatically. The question is whether the relationship is such that there is a real likelihood of bias, and that is a question of degree on the facts.

Facts. A.K. Kraipak v. Union of India, AIR 1970 SC 150, concerned selection to the Indian Forest Service from officers of the Jammu and Kashmir forest department. A Special Selection Board was constituted under the Indian Forest Service (Initial Recruitment) Regulations 1966. One member of the Board was the Acting Chief Conservator of Forests, who held that post by superseding another Conservator whose appeal against the supersession was pending before the State Government, and who was himself a candidate for selection. He withdrew when his own name was considered, but sat and participated when the claims of his rivals were considered and when the list of selected candidates was prepared in order of preference. His name came top of the list; three Conservators who were his rivals, including the officer he had superseded, were omitted. The list went to the Home Ministry and then to the Union Public Service Commission, which examined the records afresh and made its recommendations, and the Government of India notified it. The officers left out petitioned under Article 32.

In one line

Before an authority decides against you, it must tell you what is alleged and let you answer.

In the words a student can write in an exam: audi alteram partem means "hear the other side", and it requires that a person whose rights or interests are to be affected by a decision be given notice of the case against him, disclosure of the material on which it is based, and a fair opportunity to answer it before the decision is taken. Its content is not fixed: it varies with the nature of the decision, the gravity of what is at stake and the framework of the statute, and it may extend to an oral hearing, legal representation and cross-examination where those are necessary for a fair decision.

The two limbs

The rule has two parts, and an answer that gives only the second is incomplete.

1. Notice. The person must be told, before the decision, what is alleged against him and what is proposed. 2. Hearing. The person must be given a real opportunity to answer.

Notice without a hearing is useless, and a hearing without notice is a trap: a person who does not know what is alleged cannot answer it.

Limb 1: notice

Notice must satisfy four requirements.

It must be given before the decision. Notice after the event is not notice at all, subject only to what is said in [Post-Decisional Hearing].

It must be adequate and specific. A notice saying that the recipient's conduct "has been found unsatisfactory" tells him nothing. It must state the specific charge, the grounds relied on and the action proposed, so that he knows both what to answer and what he stands to lose.

It must give reasonable time. A notice requiring an answer within twenty-four hours to a charge covering three years of accounts is not a real opportunity.

It must disclose the material. A person cannot answer evidence he has not seen. Where the authority proposes to rely on an inspection report, an audit note or a complaint, the substance of it must be disclosed. This is the point on which administrative decisions most often fail: the authority hears the person but decides on a file he has never seen.

Limb 2: the hearing, and what it contains

The content varies, and the variation is the whole difficulty. The elements that may be required are these.

In one line

An authority deciding a matter affecting a person must say why it decided as it did.

In the words a student can write in an exam: the requirement of a reasoned decision, also called a speaking order, is now treated as the third principle of natural justice in India. An administrative authority exercising judicial or quasi-judicial functions is required to record the reasons for its decision, except where the requirement has been dispensed with expressly or by necessary implication; the reasons need not be as elaborate as those of a court, but must be clear and explicit enough to show that the authority has given due consideration to the points in controversy.

What a speaking order is

An order speaks when it discloses on its face the reasons that led to it. Its opposite is an order that states only the conclusion: "the application is rejected", "the licence is cancelled", "the representation is not accepted".

The distinction matters because a conclusion without reasons cannot be tested. Nobody, including a court, can tell whether the authority considered the right things, applied the right test, or considered anything at all.

Why reasons are required: four arguments

Each is worth a sentence in an answer, because an examiner is testing whether you understand the purpose rather than the rule.

1. They discipline the decision maker. Somebody who must write down why is forced to work out why. The obligation to give reasons improves the decision before anybody challenges it, which is the green light argument: [Red Light and Green Light Theories].

2. They make review possible. Without reasons, a court asked whether the authority took irrelevant considerations into account has nothing to look at. Reasons are what convert a right of appeal or review from a formality into a reality.

3. They satisfy the person affected. A decision explained is a decision that can be accepted, or answered. A bare rejection invites the suspicion that no reason existed.

4. They exclude arbitrariness. An order without reasons cannot be distinguished from a whim, and Article 14 strikes at arbitrariness in State action: Maneka Gandhi v. Union of India, AIR 1978 SC 597.

The leading Indian case

Facts. S.N. Mukherjee v. Union of India, AIR 1990 SC 1984, was decided by a Constitution Bench. The appellant, an army officer officiating as a Major, took over as Officer Commanding a supply company attached to a Military Hospital. In his absence another officer submitted a contingent bill for Rs.16,280 for the winter liveries of depot civilian chowkidars and sweepers, which was returned by the Controller of Defence Accounts with objections; the appellant then submitted a fresh contingent bill for Rs.7,029.57. In view of the wide difference between the two bills the Controller reported the matter for investigation, a Court of Enquiry blamed the appellant for certain lapses, and disciplinary proceedings followed. His post-confirmation petition to the Central Government under section 164 of the Army Act 1950 was rejected without reasons, and the question was whether the Central Government was bound to give reasons.

In one line

A post-decisional hearing is a hearing given after the order has already been made, and it is a second best that the law allows only where a hearing beforehand was genuinely impossible.

In the words a student can write in an exam: a post-decisional hearing is an opportunity to be heard afforded after an order affecting a person has been passed, instead of before. The normal rule is that natural justice requires a pre-decisional hearing; a post-decisional hearing is accepted only where prompt action was necessary in the public interest, where the statute so provides, or where the pre-decisional hearing would defeat the purpose of the order, and even then it must be a full, fair and effective hearing by an authority genuinely open to changing its mind.

Why the idea exists at all

Because some administrative powers are useless if exercised slowly. If a food inspector must give a week's notice before seizing adulterated milk, the milk will be sold. If the Reserve Bank must hear a bank before suspending it, the deposits will be gone. If a passport can only be impounded after a hearing, the holder will leave the country.

So the law faces a conflict: fairness says hear first, effectiveness says act first. The post-decisional hearing is the compromise. It lets the administration act at once and gives the person affected the hearing afterwards.

The problem with the compromise

An honest treatment says plainly that the compromise is unsatisfactory, and the criticism is examinable.

1. Minds close once a decision is taken. An authority which has already acted, has committed itself publicly and may have to admit error is not in the same position as one deciding for the first time.
2. The damage may be done. A licence cancelled, a business closed or a passport impounded causes losses which a later hearing cannot undo, even if the order is revoked.
3. The burden shifts. In a pre-decisional hearing the authority must be satisfied before acting; in a post-decisional hearing the person affected is arguing to have something reversed.
4. It can become a formality. If every power could be exercised first and explained later, the pre-decisional hearing would disappear.

That is why courts treat it as an exception requiring justification, not as an alternative the administration may choose.

The Indian position

Facts. Maneka Gandhi v. Union of India, AIR 1978 SC 597, concerned a passport issued to the petitioner on 1 June 1976 under the Passports Act 1967. On 4 July 1977 she received a letter dated 2 July 1977 from the Regional Passport Officer telling her that the Government had decided to impound her passport under section 10(3)(c) "in public interest" and requiring her to surrender it within seven days. She was given no hearing before the order, and when she asked for the reasons the Government declined to furnish them on the ground that it was not in the interest of the general public to do so.

In one line

Natural justice can be displaced, but only where a statute clearly says so or where a hearing would be pointless, impossible or dangerous.

In the words a student can write in an exam: the principles of natural justice are not absolute and may be excluded in recognised situations: by express statutory provision or by inevitable implication from the statutory scheme; by the Constitution itself, as in the second proviso to Article 311(2); in cases of emergency or urgency where immediate action is required; where a hearing would serve no useful purpose, under the useful formality theory; where the decision is legislative rather than adjudicatory; in matters of confidentiality and national security; in academic evaluation; where the facts are admitted; and under the doctrine of necessity, where the person alleged to be disqualified is the only authority competent to act.

Why exceptions exist, and the limit on them

Natural justice supplements the law; it does not override it. So a valid statute can displace it, and situations exist in which a hearing is impossible or useless.

But the exceptions are dangerous, because an administration that can name an exception can avoid a hearing whenever it likes. The law therefore hedges them in two ways.

Exclusion must be clear. Swadeshi Cotton Mills v. Union of India, AIR 1981 SC 818, holds that the rules of natural justice operate only in areas not covered by any validly made law, and that a statutory provision may exclude them specifically or by inevitable implication. Silence is not exclusion, which is why the audi alteram partem rule was read into section 18AA of the Industries (Development and Regulation) Act 1951 in that very case.

An express exclusion is not immune from challenge. After Maneka Gandhi v. Union of India, AIR 1978 SC 597, a statutory procedure must be right, just and fair to satisfy Articles 14 and 21. So a provision denying a hearing may be valid or may itself be unconstitutional, and stating that in an answer separates a good one from an average one.

The exceptions, one by one

1. Statutory exclusion

Where the Act expressly says no notice or hearing is required, or where the scheme of the Act makes it inevitable that none was intended. The test of inevitable implication is a strict one: it is not satisfied merely because the Act is silent, or because a hearing would be inconvenient.

An examiner will often set a section that is merely silent and expect you to say that silence is not exclusion.

2. Constitutional exclusion

The Constitution itself excludes a hearing in one well known case. The second proviso to Article 311(2) dispenses with the inquiry otherwise required before a civil servant is dismissed, removed or reduced in rank, in three situations: where he is dismissed on the ground of conduct which has led to his conviction on a criminal charge; where the authority is satisfied, for reasons to be recorded in writing, that it is not reasonably practicable to hold the inquiry; and where the President or Governor is satisfied that in the interest of the security of the State it is not expedient to hold it. This is treated in [Civil Services in India: Nature, Organisation, Powers and Functions].

In one line

An order made in breach of natural justice is bad, but whether it is a nullity from the start or merely liable to be set aside, and what the court actually does about it, are three different questions.

In the words a student can write in an exam: an order passed in violation of the principles of natural justice is void, in the sense that it is made without jurisdiction and can be challenged by a writ under Article 226 or Article 32; but the Indian courts have not treated the consequence as automatic, and the practical position is that such an order is voidable at the instance of the person aggrieved, who may waive the breach, and that the remedy is discretionary, so that relief may be refused where no prejudice was caused, where the petitioner has been guilty of delay or has himself acted improperly, or where quashing would serve no purpose, and the usual order is to quash the decision and remit the matter for a fresh hearing rather than to decide the merits.

The three questions, kept apart

Students lose marks here by running three separate questions together. Keep them apart.

1. Is the order void or voidable? A question of legal theory.
2. Can the breach be waived or cured? A question about the person affected.
3. What will the court actually do? A question about remedy and discretion.

Question 1: void or voidable

The theoretical position

The orthodox view is that a breach of natural justice goes to jurisdiction. An authority is empowered to decide only after hearing; if it decides without hearing it has done something the law never authorised, and its order is a nullity, void from the beginning, of no effect for any purpose and incapable of being validated.

The English revival of this view came in Ridge v. Baldwin, where a chief constable was dismissed without notice or a hearing and the House of Lords held the dismissal void. That case is the reason natural justice was restored to administrative decisions in England, and it is named in most Indian discussions of the topic.

Why "void" is an awkward word here

The difficulty is practical. If the order were truly void, then anybody could ignore it, it would need no court to set it aside, and it could be attacked collaterally in any proceeding at any time. That is not how the law works in practice, for three reasons.

1. Somebody must decide whether there was a breach. Until a court says so, the order looks valid and is acted on.
2. The person affected may not object. If he takes no steps, the order stands, which is not what one expects of a nullity.
3. He may waive the breach, and a nullity cannot be waived into existence.

In one line

An administrative tribunal is a body outside the ordinary court system which Parliament has set up to decide a particular class of dispute, quickly, cheaply and by people who understand the subject.

In the words a student can write in an exam: an administrative tribunal is an authority, other than a court, vested by statute with the power to adjudicate disputes of a specified kind between a citizen and the administration, or between citizens, and required to act judicially; it is characterised by its statutory origin, its adjudicatory function, its freedom from the strict rules of procedure and evidence, the presence of technical or administrative expertise alongside legal qualification, and its subjection to the writ jurisdiction of the High Courts under Articles 226 and 227.

Where tribunals came from

The reasons are the reasons in [The Rise of the Administrative State], applied to adjudication. Once the State began to regulate wages, rents, tariffs, service conditions, taxes and licences, disputes arose in numbers and of a kind the ordinary civil courts could not absorb.

Four features of the ordinary courts made them unsuitable.

1. Cost. A workman claiming compensation cannot fund a civil suit.
2. Delay. A tenant facing eviction or an employee dismissed cannot wait years.
3. Procedure. The Code of Civil Procedure and the law of evidence are designed for a contested trial between private parties and are heavy machinery for a small administrative dispute.
4. Expertise. A civil judge has no special knowledge of tariff structures, industrial relations or drug pricing.

Tribunals were the answer: statutory bodies with narrow jurisdiction, simple procedure and members who know the subject.

The constitutional foundation in India

Before 1976 tribunals existed under ordinary statutes. The Constitution (Forty-second Amendment) Act 1976 inserted Part XIVA, containing two Articles.

Article 323A empowers Parliament to provide by law for the adjudication or trial by administrative tribunals of disputes and complaints with respect to recruitment and conditions of service of persons appointed to public services and posts under the Union, a State, or a local or other authority or a government company. Only Parliament may legislate under it, and it covers service matters only.

Article 323B empowers the appropriate Legislature, meaning Parliament or a State legislature, to provide for tribunals for other specified matters: taxation, foreign exchange and import and export, industrial and labour disputes, land reforms, ceiling on urban property, elections to Parliament and the State legislatures, production and supply of essential goods, and rent and tenancy matters.

In one line

A court is part of the ordinary judicial system and can try any dispute of its kind; a tribunal is a statutory body created for one class of dispute and can do only what its Act allows.

In the words a student can write in an exam: a court is a permanent organ of the State forming part of the ordinary judicial hierarchy, presided over by a judicially trained person, possessing inherent jurisdiction to try all disputes of a civil nature unless barred, bound by the rules of procedure and evidence, and deciding solely according to law. A tribunal is a body created by statute for the adjudication of a specified class of dispute, whose jurisdiction is confined to what the statute confers, which may include technical or administrative members, which is not bound by the Code of Civil Procedure or the Indian Evidence Act, which may take policy into account, and which is subject to the supervisory jurisdiction of the High Court under Articles 226 and 227.

Why the distinction is asked

Three practical consequences turn on it, and an answer that gives the table without the consequences is thin.

1. Which writ or remedy is available, and whether Article 227 superintendence applies.
2. Whether the body has contempt power, since a court has it inherently and a tribunal only if the statute confers it.
3. How its decisions are treated: whether they bind as precedent, and what appeal lies.

The distinctions

In one line

The Administrative Tribunals Act 1985 took every service dispute of a government servant out of the civil courts and the High Courts and gave it to a specialised tribunal, and the courts then spent twelve years deciding how much of that was constitutional.

In the words a student can write in an exam: the Administrative Tribunals Act 1985 was enacted under Article 323A to provide for the adjudication of disputes relating to recruitment and conditions of service of persons appointed to public services under the Union and the States. It establishes the Central Administrative Tribunal and State Administrative Tribunals, prescribes their composition, qualifications and terms of office, confers on them the jurisdiction previously exercised by the courts in service matters, lays down a simplified procedure, and by section 28 purported to exclude the jurisdiction of all courts except the Supreme Court under Article 136; that exclusion was held unconstitutional, so far as it touched Articles 226, 227 and 32, in the decision of seven judges worked below.

The scheme of the Act, section by section

The Act has 37 sections and they fall into five groups. Reading them in groups is the way to remember them.

Group 1: establishment and constitution, sections 1 to 13

• Section 1 short title, extent and commencement. Section 2 provides that the Act does not apply to certain persons, principally members of the armed forces, officers and servants of the Supreme Court and the High Courts, and persons in the secretarial staff of Parliament and the State legislatures. Section 3 contains the definitions.
• Section 4 establishes the Central Administrative Tribunal for Union services, and empowers the Central Government, on receipt of a request from a State, to establish a State Administrative Tribunal, and by agreement between two or more States a Joint Administrative Tribunal.
• Section 5 deals with composition and Benches: a Chairman, and such number of Vice-Chairmen and Judicial and Administrative Members as the appropriate Government may deem fit. Ordinarily a Bench consists of one Judicial and one Administrative Member. The proviso to section 5(6) requires that where a question involving the interpretation of a statutory provision or rule in relation to the Constitution arises before a single Member, it be referred to a Bench of at least two Members.
• Section 6 prescribes qualifications for the Chairman, Vice-Chairman and Members. Section 7 allows a Vice-Chairman to act as Chairman. Section 8 fixes the term of office and section 9 provides for resignation and removal. Section 10 deals with salaries, allowances and other conditions of service; section 10A saves the terms and conditions of service of a Vice-Chairman; and section 10B governs the qualifications, terms and conditions of service of the Chairman and Members, which is the provision the Tribunals Reforms Act 2021 amended by its section 15. Section 11 restricts the holding of offices after ceasing to be Chairman or Member. Section 12 confers financial and administrative powers on the Chairman, and section 13 provides for the staff.

In one line

Tribunalisation means the steady transfer of judicial work from the courts to tribunals staffed and controlled by the executive, and the objection to it is that the body which is the usual litigant also appoints, pays and removes the people who decide against it.

In the words a student can write in an exam: tribunalisation is the process by which adjudicatory functions formerly exercised by the ordinary courts have been transferred to specialised tribunals created by statute. It threatens the independence of the judiciary because the members of those tribunals are appointed, remunerated and removed by the executive, which is the standing respondent before them, because their tenure is short and re-appointment is possible, and because administrative members without judicial training may outnumber judicial members. The Supreme Court has responded through a line of decisions requiring that any tribunal substituted for a court be no less efficacious than the court it replaces, culminating in the striking down of provisions of the Tribunals Reforms (Rationalisation and Conditions of Service) Ordinance 2021, after which Parliament enacted the Tribunals Reforms Act 2021 containing substantially the same provisions.

What is objectionable, precisely

The objection is not that tribunals exist. It is structural, and it has five parts. Naming them separately is what distinguishes a good answer.

1. The appointing authority is the usual litigant. In a service tribunal, a tax tribunal or a regulatory appellate tribunal, the Union is a party in nearly every case, and the Union appoints the members.

2. Short tenure with re-appointment. A member appointed for four years who hopes to be re-appointed depends on the goodwill of the executive. Security of tenure is the oldest guarantee of judicial independence, and Articles 124 and 217 secure it for judges precisely because it cannot be left to the appointing power.

3. Administrative control of the tribunal. Where the tribunal's staff, budget and premises come from the ministry whose decisions it reviews, its independence is compromised even if its members are upright.

4. Composition. A bench in which administrative members can outnumber judicial members transfers judicial work to persons without judicial training.

5. The parent ministry. A tribunal attached to the ministry whose orders it hears appeals from is not seen to be independent, whatever the reality.

Behind all five stands Article 50, the Directive Principle that the State shall take steps to separate the judiciary from the executive in the public services: [Separation of Powers].

The line of cases

The Supreme Court has decided this question repeatedly. The 2021 decision describes itself as the seventh in a line beginning with the National Company Law Tribunal case. Three decisions carry the argument, and two of them are worked in the earlier chapters.

In one line

Discretion is the freedom a statute gives an official to choose, and administrative law exists largely to make sure that the choice is a real one, made for the right reasons.

In the words a student can write in an exam: administrative discretion is the power conferred by a statute on an administrative authority to choose between alternative courses of action, or to decide whether to act at all, according to its own judgment of what is appropriate in the circumstances. It is indispensable in a welfare State because no legislature can foresee every case, but it is also the point at which arbitrariness enters, and it is therefore confined by the statute conferring it, structured by published criteria and reasons, and checked by judicial review on the grounds of abuse of discretion and failure to exercise discretion.

How to recognise discretion in a statute

Discretion is conferred by particular verbs and phrases, and a student should be able to spot them.

The commonest examination trap is a section using both: "the authority shall grant a licence to an applicant who satisfies the prescribed conditions, but may refuse where it considers the grant contrary to the public interest". That confers discretion on the second limb only.

Why discretion is necessary

The reasons follow from [The Rise of the Administrative State] but they are worth stating for themselves, because the syllabus asks for the "need for conferring discretion" as its own head.

1. The legislature cannot foresee every case. A statute regulating the grant of a licence cannot list every configuration of premises, applicant and locality.

2. Circumstances change faster than statutes. A rule fixed in 2010 may be senseless in 2026, and discretion allows the administration to respond without an amending Act.

3. Individual justice. A rigid rule applied to everyone produces hardship in the exceptional case. Discretion lets the administration make an exception where the rule would do injustice, which is precisely the green light argument in [Red Light and Green Light Theories].

4. Technical judgment. Whether a bridge is safe, a drug effective or a tariff reasonable are questions of judgment that a statute cannot settle in advance.

5. Prevention and management. Modern administration must plan, allocate and prioritise. Allocating hospital beds, deciding where to build a road, or choosing among tenderers all require a choice.

In one line

An authority abuses its discretion when it uses a real power for the wrong reason.

In the words a student can write in an exam: abuse of discretion covers the cases in which an authority has exercised the power conferred on it, but has exercised it improperly. The recognised grounds are that the power was used for an improper or ulterior purpose; that irrelevant considerations were taken into account or relevant considerations ignored; that the authority acted in bad faith; that it exercised the power in a colourable manner; that its decision was so unreasonable that no reasonable authority could have reached it; that the decision was based on no evidence or on a mistake of fact; and, in the appropriate class of case, that the interference was disproportionate.

The two families of ground

Module III's grounds divide into two, and MU prints them as separate heads, so the division must be observed.

The English classification, which Indian courts also use, groups all of it under three headings: illegality, irrationality and procedural impropriety, with proportionality added later. Illegality and irrationality are the subject of this chapter; procedural impropriety is Module II's natural justice.

Ground 1: improper purpose

A power conferred for one purpose may be used only for that purpose. If the Act allows land to be acquired for a public purpose, it may not be acquired to spite the owner or to benefit a private developer.

How the purpose is found. From the Act itself: its preamble, its long title, the scheme, and the section conferring the power. The question is never what the authority thought the purpose was, but what the statute says it is.

Mixed purposes. Where an authority acts for two purposes, one proper and one not, the usual test is whether the improper purpose was the dominant one, or whether the authority would have reached the same decision on the proper purpose alone.

Ground 2: irrelevant considerations, and ignoring relevant ones

The authority must take into account what the statute requires it to consider, and must exclude what the statute makes irrelevant. Both halves matter and students usually give only the first.

What is relevant is determined by the statute, expressly or by implication from its purpose. A licensing statute concerned with public safety makes the applicant's caste, politics and religion irrelevant, and makes the condition of the premises relevant.

In one line

An authority fails to exercise its discretion when it goes through the motions of deciding but does not actually decide: it obeys somebody else, applies a rule it has bound itself to, hands the job to another, or does not apply its mind at all.

In the words a student can write in an exam: failure to exercise discretion is the second family of grounds of judicial review. It arises where the authority in whom the discretion was vested has not brought its own independent judgment to bear on the particular case, whether because it acted under the dictation of another, because it fettered its discretion by a rigid rule or a self imposed policy applied mechanically, because it unlawfully sub-delegated the power, because it surrendered its judgment by acting on an agreement or a direction, or because it did not apply its mind to the matter at all.

Why this is a separate family

Because the complaint is different in kind. In [Grounds of Judicial Review: Abuse of Discretion] the authority made a choice, but made it for the wrong reason. Here it made no real choice at all.

The principle underlying the whole family is one sentence: the discretion is to be exercised by the authority on whom the statute conferred it, on the facts of each case. Everything below is a way of failing to do that.

MU prints the two families as separate items in topic 3.2, so an answer must not merge them.

Ground 1: acting under dictation

The authority takes the decision because somebody else told it to. It has the power, it goes through the process, and then it does what a minister, a superior officer or a committee directed.

It is no answer that the person giving the direction was senior. The statute chose the deciding authority, and a superior cannot exercise a power vested in a subordinate unless the statute allows it.

How to tell it from lawful guidance. An authority may seek advice, consult, and consider a departmental policy. It fails to exercise discretion only when it treats the direction as binding and decides because of it rather than on its own view. The question is whether the authority applied its own mind or merely relayed another's decision.

Ground 2: self imposed fetters

The authority adopts a rule or policy and then applies it mechanically, refusing to consider the particular case.

The law's balance is worth stating carefully, because both halves are examinable.

1. An authority may adopt a general policy. Consistency is a virtue, and a policy structures discretion and protects against arbitrariness: [Administrative Discretion and Why It Is Conferred].
2. But it must keep its mind ajar. It must be willing to hear that this case is exceptional and to depart from the policy if the case warrants it.

In one line

The writs are the five orders a High Court or the Supreme Court can make to keep the administration inside the law: release the person, do your duty, quash the order, stop the proceeding, and prove your title to the office.

In the words a student can write in an exam: the judicial remedies against administrative arbitrariness in India are the writs of habeas corpus, mandamus, certiorari, prohibition and quo warranto, issued by the Supreme Court under Article 32 and by the High Courts under Article 226, together with the supervisory jurisdiction of the High Courts over courts and tribunals under Article 227 and the discretionary appellate jurisdiction of the Supreme Court under Article 136; the ordinary remedies of a declaration, an injunction and a suit for damages remain available in addition.

The two Articles, and why the difference matters

The words "for any other purpose" in Article 226 are the most important five words in Indian administrative law. They mean a High Court can issue a writ for the breach of an ordinary legal right, not only a fundamental right, which is why almost all administrative law litigation in India is under Article 226 and not Article 32.

Article 32 is itself a fundamental right, which no other constitution guarantees in the same way, and it is part of the basic structure, as is Article 226: L. Chandra Kumar v. Union of India, AIR 1997 SC 1125, holds that the jurisdiction conferred on the High Courts under Articles 226 and 227 and on the Supreme Court under Article 32 is part of the inviolable basic structure and cannot be excluded.

Against whom does a writ lie

Against the State as defined in Article 12, and against any person or authority discharging a public function.

Whether a body is State under Article 12 is therefore the gateway question, and it is answered by the instrumentality test.

Facts. Ajay Hasia v. Khalid Mujib Sehravardi, AIR 1981 SC 487, concerned admissions to the Jammu and Kashmir Regional Engineering College, Srinagar, which was run by a society registered under the Jammu and Kashmir Registration of Societies Act 1898. The preliminary question was whether that society was a State within Article 12.

In one line

Where a public authority has led somebody to expect a benefit or a procedure, it cannot simply disappoint that expectation without at least hearing them and giving a reason.

In the words a student can write in an exam: the doctrine of legitimate expectation holds that where an administrative authority, by an express promise or by a settled and consistent past practice, has led a person to expect either a substantive benefit or a particular procedure before a decision is taken, that person acquires an interest which the law will protect, so that the authority may not depart from the expectation without giving him an opportunity of being heard and without a reason that is rational and in the public interest; the expectation does not create an enforceable right to the benefit itself, and it yields to a change of policy made in good faith and in the public interest.

The gap the doctrine fills

Consider Sadhana Kale, who has held a licence renewed every year for fifteen years, always without objection. This year the authority refuses renewal without notice and without explanation.

• She has no legal right to renewal, because the statute gives a discretion.
• She has no contract with the authority.
• She cannot rely on estoppel, because no representation was made to her personally on which she altered her position in the technical sense: [Estoppel and Waiver against the Government].
• She may not fall within natural justice on the older view, because nothing of hers is being taken away; a renewal is a fresh grant.

And yet it is obviously unfair. The doctrine of legitimate expectation exists to close exactly that gap: she expected renewal because of fifteen years of consistent practice, and the authority must at least hear her before disappointing the expectation.

How an expectation arises

Two routes, and both must be given.

1. An express promise or representation. The authority publishes a scheme, announces a policy, gives an assurance, or issues a circular stating how applications will be dealt with. A person who comes within its terms may expect it to be applied to him.

2. A settled and consistent past practice. The authority has, over a period, dealt with a class of case in a particular way, and the person could reasonably expect the same treatment.

The expectation must be legitimate, not merely a hope. It is not legitimate if:

• it is contrary to a statute, because no expectation can be founded on an illegality;
• the practice was irregular or unauthorised;
• the person knew the policy was under reconsideration;
• the representation was made by an officer without authority to make it; or
• it is founded on a mere wish rather than on any promise or practice.

In one line

Proportionality asks whether the administration used a sledgehammer to crack a nut.

In the words a student can write in an exam: the doctrine of proportionality requires that administrative action which restricts a right be no more restrictive than is necessary to achieve its legitimate object. It asks whether the measure pursues a legitimate aim, whether it is rationally connected to that aim, whether a less restrictive measure would have achieved the aim equally well, and whether the benefit to the public outweighs the harm to the individual. In India it applies with full force where a fundamental right is restricted, and in the review of a punishment where the penalty is shockingly disproportionate to the misconduct, while ordinary administrative decisions not touching a fundamental right are reviewed on the narrower standard of unreasonableness.

The contrast that defines the topic

Proportionality is best understood against the older ground it competes with.

Unreasonableness, in the Wednesbury sense, asks whether the decision is so unreasonable that no reasonable authority could have reached it. The threshold is deliberately extreme, and the court says nothing about whether the decision was correct or sensible. It is a test of outrageousness.

Proportionality asks whether the means chosen were necessary to the end. It is a test of fit. The court examines whether a lesser measure would have done, which requires it to look much more closely at the merits.

The single sentence that answers most exam questions: proportionality is a more intense standard of review than unreasonableness, and Indian courts apply it where a fundamental right is at stake and the Wednesbury standard where it is not.

The four part test

The modern test, as used in constitutional adjudication, has four steps, and they must be given in order.

1. Legitimate aim. The measure must pursue an object that the law permits. If the aim itself is impermissible, the enquiry stops.

2. Rational connection, or suitability. The means chosen must be capable of achieving that aim. A rule which cannot advance the object fails here.

3. Necessity, or minimal impairment. There must be no less restrictive measure that would achieve the aim equally well. This is the heart of the test and the step that does the work.

4. Balancing, or proportionality in the strict sense. The benefit to the public must outweigh the harm to the right. Even a necessary measure fails if the cost to the individual is out of all proportion to the gain.

In one line

Public accountability means that everybody who exercises public power must be answerable for how they used it, to somebody, by some process, with some consequence.

In the words a student can write in an exam: public accountability is the principle that the holders of public office and public power are answerable to the public for the exercise of that power, and it is secured in India through a set of overlapping mechanisms: political accountability to the legislature through ministerial responsibility, questions and parliamentary committees; financial accountability through the Comptroller and Auditor General and the Public Accounts Committee; legal accountability through judicial review, the writs, and the liability of the State in tort and contract; institutional accountability through the Lokpal, the Lokayuktas, the Central Vigilance Commission and the Prevention of Corruption Act 1988; and public accountability in the direct sense through the Right to Information Act 2005 and a free press.

The three questions accountability asks

Any account of accountability must answer three questions, and structuring an answer around them is what turns a list into an argument.

1. Accountable to whom? The legislature, the courts, a specialised body, or the public directly.
2. Accountable for what? Legality, financial regularity, efficiency, propriety, or policy.
3. With what consequence? Nothing at all, a report, quashing of the decision, compensation, dismissal, or prosecution.

A mechanism that answers the first two but not the third is weak, and most of the criticism of Indian accountability machinery is at that point.

The mechanisms

1. Political accountability

Ministerial responsibility. By Articles 75(3) and 164(2) the Council of Ministers is collectively responsible to the House of the People and to the State legislative assembly respectively. Individually, a minister answers for the department. The instruments are questions, calling attention motions, adjournment motions, debates and the ultimate sanction of a vote of no confidence.

Its weakness should be stated plainly. Collective responsibility with a disciplined majority means the sanction is rarely applied, and a minister answers for the department's policy rather than for an individual officer's decision affecting an individual citizen.

Parliamentary committees are the working part of this machinery and are treated in [Congressional and Parliamentary Committees].

2. Financial accountability

The Comptroller and Auditor General, appointed under Article 148, audits the accounts of the Union and the States, and by Article 151 his reports are laid before Parliament and the State legislatures. Those reports go to the Public Accounts Committee, which examines them and reports to the House.

This is the strongest continuous accountability mechanism in India for money, and its weakness is that it operates after the expenditure and its reports depend on the House and the Government acting on them.

3. Legal accountability

The whole of Modules II and III. Judicial review of rules and of discretion; natural justice; the writs in [Judicial Remedies against Administrative Arbitrariness: the Writs]; and the liability of the State in tort and contract in the chapters that follow.

In one line

The State can be sued for the wrongs its servants commit, and Article 300 is the provision that says so, but it says so by pointing back at the law as it stood in 1950.

In the words a student can write in an exam: a tort is a civil wrong, other than a breach of contract, for which the remedy is an action for unliquidated damages. The tortious liability of the State in India rests on Article 300(1) of the Constitution, which provides that the Government of India may sue and be sued by the name of the Union of India and the Government of a State by the name of the State, in the like cases as the Dominion of India and the corresponding Provinces might have sued or been sued if the Constitution had not been enacted. The Article therefore fixes no substantive rule of its own but carries forward the position under the earlier Government of India Acts, which in turn carried forward the liability of the East India Company, under which the State was liable for acts done in the exercise of non-sovereign functions but not for those done in the exercise of sovereign powers.

The vocabulary first

Tort: a civil wrong independent of contract, for which the remedy is an action for damages. Negligence, trespass, nuisance, false imprisonment, conversion and malicious prosecution are torts.

Vicarious liability: the liability of one person for the tort of another, most commonly of an employer for the acts of an employee committed in the course of employment. The State's liability is vicarious in this sense: it is sued for what its officers did.

In the course of employment means while doing the work he was employed to do, or doing it in an unauthorised way. It does not cover a frolic of the officer's own.

The text of Article 300(1)

The Article provides that the Government of India may sue or be sued by the name of the Union of India, and the Government of a State by the name of the State, and may, subject to any provisions which may be made by Act of Parliament or of the legislature of such State, sue or be sued in relation to their respective affairs in the like cases as the Dominion of India and the corresponding Provinces or the corresponding Indian States might have sued or been sued if this Constitution had not been enacted.

Three things follow, and each is examinable.

1. It is a procedural provision that carries forward substantive law. It tells you who to sue and by what name, and then sends you back to the earlier law for whether the claim lies at all.

In one line

The State used to escape liability by saying that what its officers were doing could only be done by a State, and the courts have spent sixty years cutting that defence down to almost nothing.

In the words a student can write in an exam: the distinction between sovereign and non-sovereign functions determines whether the State is liable in tort for the acts of its servants. Functions which by their nature can be performed only by a sovereign, such as defence, the maintenance of armed forces, the making of peace or war, foreign affairs and the acquisition and retention of territory, attract immunity; functions which a private individual could equally perform, such as running a transport service, a factory or a hospital, do not. The distinction originated in the liability of the East India Company for its trading as distinct from its sovereign acts, was applied at its widest to statutory powers of arrest and seizure, and has since been confined to the primary and inalienable functions of the State, the Supreme Court itself observing that watertight compartmentalisation into sovereign and non-sovereign is unsound.

Where the distinction came from

The East India Company both traded and governed. When it was sued, the courts had to ask which capacity the act belonged to, and the rule that emerged was that the Company was liable for acts done in its commercial or trading capacity and not for acts done in the exercise of its sovereign powers.

That rule was carried into the Government of India Act 1858 and, by the chain traced in [Tortious Liability of the State], into Article 300(1) of the Constitution. So a distinction devised for a trading company that also governed is the one by which the liability of a modern welfare State is still measured.

The old test is usually put in the form: could a private individual have done this act? If yes, the function is non-sovereign and the State is liable; if only a sovereign could have done it, immunity applies.

The three cases, in order

The whole topic is the story of three decisions, and the answer to almost any question on it is these three in sequence.

Stage 1: the ordinary law applies

Facts. State of Rajasthan v. Mst. Vidhyawati, AIR 1962 SC 933. On 11 February 1952 the husband of the first respondent and father of the second was knocked down by a Government jeep car driven rashly and negligently by an employee of the State of Rajasthan, while the vehicle was being taken from the repair shop to the Collector's residence, and he died in hospital. The trial court decreed the suit ex parte against the driver but dismissed it against the State, holding that because the car was maintained for the use of the Collector in the discharge of his official duties, that alone absolved the State of vicarious liability, even though it was not being used for any State purpose at the time. The High Court decreed against the State as well.

In one line

The Government can make contracts, but only in the form Article 299 prescribes, and a contract that does not meet that form is void, however honest the dealing.

In the words a student can write in an exam: Article 298 empowers the Union and the States to carry on trade and to make contracts for any purpose, and Article 299(1) prescribes the form: every contract made in the exercise of the executive power of the Union or a State must be expressed to be made by the President or the Governor, and must be executed on behalf of the President or the Governor by such persons and in such manner as he may direct or authorise. A contract not satisfying those requirements is void and unenforceable against the Government; but Article 299(2) protects the President, the Governor and the executing officer from personal liability, and a party who has performed under a void contract may recover under section 70 of the Indian Contract Act 1872.

The power to contract

Article 298 provides that the executive power of the Union and of each State extends to the carrying on of any trade or business, and to the acquisition, holding and disposal of property, and the making of contracts for any purpose.

Two points follow. First, the Government does not need a statute authorising each contract; the executive power suffices, which is consistent with Rai Sahib Ram Jawaya Kapur v. State of Punjab, AIR 1955 SC 549, holding that the executive may act without prior legislation so long as it does not infringe rights or contravene a law. Second, because the power is executive, it is subject to the constitutional limits on executive power in Articles 73 and 162.

The form: Article 299(1)

Article 299(1) provides that all contracts made in the exercise of the executive power of the Union or of a State shall be expressed to be made by the President, or by the Governor of the State, as the case may be, and that all such contracts and all assurances of property made in the exercise of that power shall be executed on behalf of the President or the Governor by such persons and in such manner as he may direct or authorise.

That yields three requirements, and they are the heart of the topic.

In one line

The Government may refuse to produce a document in court if disclosing it would harm the public interest, but it is the court and not the Government that decides whether that is true.

In the words a student can write in an exam: public interest privilege, also called Crown privilege or State privilege, is the rule that evidence derived from unpublished official records relating to the affairs of State may not be given without the permission of the head of the department concerned, and that a public officer may not be compelled to disclose communications made to him in official confidence where he considers that the public interest would suffer. The rule is now contained in sections 129 and 130 of the Bharatiya Sakshya Adhiniyam 2023, which replaced sections 123 and 124 of the Indian Evidence Act 1872 with effect from 1 July 2024. The privilege belongs to the public and not to the Government, the claim must be made properly and on affidavit, and the final decision whether disclosure would injure the public interest rests with the court, which may inspect the document.

The section numbers have changed

This is the single most important practical point in the chapter, and every textbook printed before 2024 has the old numbers.

The Indian Evidence Act 1872 was replaced by the Bharatiya Sakshya Adhiniyam 2023 with effect from 1 July 2024. Give both numbers in an answer. The older case law is all decided on sections 123 and 124, and a student who cites only the new numbers cannot connect the cases, while one who cites only the old numbers is stating repealed law.

The two provisions

Section 129: evidence as to affairs of State

No one shall be permitted to give any evidence derived from unpublished official records relating to any affairs of State, except with the permission of the officer at the head of the department concerned, who shall give or withhold such permission as he thinks fit.

The elements are: the record must be unpublished; it must be an official record; and it must relate to affairs of State. If any one is missing the section does not apply at all.

Section 130: official communications

No public officer shall be compelled to disclose communications made to him in official confidence, when he considers that the public interests would suffer by the disclosure.

The elements are: the person must be a public officer; the communication must have been made to him in official confidence; and he must consider that the public interest would suffer.

In one line

If the Government promises something and a citizen acts on the promise, the Government may be held to it, unless holding it there would require the Government to break the law.

In the words a student can write in an exam: promissory estoppel is the rule that where a party makes a clear and unequivocal promise intended to create a legal relationship, and the promisee acts on it and alters his position, the promisor is not permitted to go back on the promise, even without consideration. It applies to the Government in its administrative and executive dealings, so that a citizen who has acted on a government assurance may hold the Government to it; but it cannot be invoked to compel the Government to act contrary to a statute or the Constitution, it does not operate against the exercise of legislative power, and it yields where the public interest genuinely requires a departure, though the burden of showing that lies on the Government.

The vocabulary

Estoppel is a rule of evidence that prevents a person from denying what he has previously asserted, where another has acted on the assertion. The ordinary form, in section 115 of the law of evidence, concerns a representation of existing fact.

Promissory estoppel, also called equitable estoppel, concerns a promise as to future conduct. It is the form that matters here, because a government assurance is almost always about what will be done.

Waiver is the intentional relinquishment of a known right. It differs from estoppel: waiver looks at the conduct of the person who gives up a right, estoppel at the conduct of the person who made a representation and the reliance of the other.

The elements of promissory estoppel

Four, and each must be stated.

1. A clear and unequivocal promise or representation, by words or conduct, intended to create legal relations or to affect a legal relationship.
2. Made with the knowledge or intention that it would be acted upon.
3. The promisee in fact acted on it and altered his position. Reliance is essential and this is what distinguishes the doctrine from [The Doctrine of Legitimate Expectation], which does not require it.
4. It would be inequitable to allow the promisor to go back on it.

Consideration is not required. That is the point of the doctrine: it makes a bare promise binding because it was acted on.

Why the Government is treated specially

Because two competing considerations meet.

In favour of applying estoppel to the Government: a State governed by the rule of law must keep its word. If a citizen who relies on an official assurance can be defeated by the answer that the Government is not bound, then no assurance is worth anything, and the administration gains an advantage no private person has. Article 14 requires fairness in State action, and going back on a promise arbitrarily is unfair.

In one line

An act of State is an act of sovereign power done against a foreigner which no court will examine, and statutory immunity is the protection a particular Act gives to officers acting in good faith under it.

In the words a student can write in an exam: an act of State is an exercise of sovereign power against a person who is not a subject of the State and who owes it no allegiance, done or subsequently adopted by the State, and it is not cognisable by the municipal courts, which cannot examine its legality. Statutory immunity is different: it is a protection expressly conferred by an Act on the Government or its officers, typically barring suits or prosecutions for anything done or intended to be done in good faith under the Act, and it is confined by the terms of the section conferring it, in particular by the requirement of good faith. To these must be added the constitutional immunities and privileges in Articles 361, 105 and 194.

Act of State

The elements

Four, and all must be present.

1. An exercise of sovereign power, not an ordinary administrative or commercial act.
2. Against a person who is not a subject of the State and owes it no allegiance. This is the decisive element.
3. Done by the State, or done by an officer and subsequently ratified or adopted by it.
4. Not justiciable, in the sense that the municipal courts will not enquire into its legality.

The second element is where almost every examination answer goes wrong. An act of State cannot be pleaded against a citizen. The plea is available in relation to foreigners and, historically, in relation to the annexation of territory and the treatment of the rights of the inhabitants of a newly acquired territory before the new sovereign recognises them.

The historical setting

The doctrine reached India through the acquisition of territory. When a State conquers or annexes territory, the rights that existed under the previous sovereign do not automatically bind the new one; they bind it only when it recognises them, expressly or by conduct. Until then, refusal to recognise a right is an act of State and no court may question it.

Once the new sovereign has recognised the rights of the inhabitants, or once the persons concerned have become its subjects, the doctrine ceases to apply and ordinary rights and remedies revive.

Its place today

Small, and an honest answer says so.

• It cannot be pleaded against a citizen, so it plays no part in ordinary administrative law.
• Its historical function, disposing of claims arising out of annexation, has largely been exhausted.
• The Constitution's guarantees operate against the State in favour of persons within India, and Article 21 is available to non-citizens as well, so the space in which a plea of act of State could defeat a claim is narrow.

In one line

A public corporation is a body created by a statute to carry on a public purpose, with a legal personality of its own, so that it can act with the freedom of a business while remaining answerable as an arm of the State.

In the words a student can write in an exam: a public corporation is a body corporate created by a special statute of the legislature, which defines its powers, duties, functions and pattern of management, having perpetual succession and a common seal, capable of suing and being sued in its own name, holding property and entering into contracts in its own right, financed wholly or largely from public funds, and managed by a board appointed by the Government; it is a device by which the State performs commercial, industrial, financial or developmental functions through an autonomous body rather than through a government department.

Why the device exists

The problem it solves is worth stating, because it explains every one of the characteristics.

A government department is answerable to a minister and through him to the legislature, is bound by government financial rules, must observe the service rules and the procedure of government, and cannot easily take a commercial risk or move quickly. Those features are virtues in the administration of justice or the collection of revenue. They are handicaps in running a bank, an airline, an insurance company or a power station.

A private company, on the other hand, moves quickly and is commercially efficient, but it answers to its shareholders and pursues profit, which is not what a public utility is for.

The public corporation is an attempt to have both: the flexibility of a company with the accountability of a public body. It is sometimes described as combining the initiative of private enterprise with public ownership and public purpose.

The characteristics

Seven, and an answer should explain rather than list them.

1. Created by a statute. Each public corporation owes its existence to a specific Act which defines its objects, powers, capital, management and duties. It has no existence apart from that Act, and its powers are only what the Act confers, so an act beyond them is ultra vires.

2. A body corporate with separate legal personality. It has perpetual succession and a common seal, may sue and be sued in its own name, may hold property and enter into contracts in its own right. Its employees are its employees and not government servants, and its funds are its funds.

3. Financed from public funds. Its capital is provided wholly or largely by the Government, and it is not financed by public issue of shares in the way a company is.

In one line

Public corporations are grouped by what they do, and the grouping matters because the more commercial the function, the less the law treats the body as an arm of the State.

In the words a student can write in an exam: public corporations may be classified by function into commercial or industrial corporations, developmental or promotional corporations, financial corporations, social service corporations and regulatory corporations; by the manner of their creation into statutory corporations, government companies and registered societies; and by the level of government to which they belong into central, State and joint corporations. The classification is not merely descriptive: whether a body is an authority within Article 12, and therefore subject to the fundamental rights and to the writ jurisdiction, depends on the degree of government control over it rather than on the label it carries.

Classification by function

This is the classification the syllabus asks for.

1. Commercial or industrial corporations

They produce goods or provide services for sale, and are expected to run at a profit or at least to cover their costs. Transport undertakings, power generation companies, steel and mining undertakings and manufacturing enterprises fall here.

Legal features. They contract in the ordinary way, employ under contracts of service, and are liable in tort like a private employer, since running a business is not a sovereign function: [Sovereign and Non-Sovereign Functions].

2. Developmental or promotional corporations

Their object is to promote an activity rather than to profit from it: industrial development corporations, tourism development corporations, small industries development bodies and agricultural marketing boards.

Legal features. They give assistance, subsidy, land or finance, and their decisions on whom to assist are administrative decisions subject to Article 14 and to the review grounds in [Grounds of Judicial Review: Abuse of Discretion].

3. Financial corporations

Banks, insurance corporations, term lending institutions and State finance corporations. They handle public money and are usually the most heavily regulated.

Legal features. They contract commercially, but their lending and recovery decisions affect citizens directly and attract fairness obligations.

4. Social service corporations

Their object is welfare rather than commerce: housing boards, water supply and sewerage boards, health and education bodies, and welfare corporations for particular groups.

Legal features. They are the closest to the State in substance, they allocate scarce benefits, and their allotment decisions are the classic subject of administrative law litigation about arbitrariness and legitimate expectation.

5. Regulatory corporations and commissions

They regulate an industry rather than participate in it: electricity regulatory commissions, telecom and securities regulators, and similar bodies.

Legal features. They exercise all three functions discussed in [Classification of Administrative Functions]: they make regulations, which is legislative; they adjudicate disputes and grant or revoke licences, which is quasi-judicial; and they administer. They must therefore observe natural justice in the adjudicatory part and the delegated legislation rules in the rule making part.

In one line

A public corporation is liable in contract and in tort like any other legal person, and it is controlled by Parliament, by the Government, by the auditor and by the courts, the last of which depends on whether it counts as the State.

In the words a student can write in an exam: a public corporation, having a separate legal personality, is liable in contract, in tort and in crime in its own name, and Article 299 does not apply to its contracts because they are not made in the exercise of the executive power of the Union or a State. It is controlled in four ways: by Parliament or the State legislature through its annual report and accounts, questions and the Committee on Public Undertakings; by the Government through appointment of the board and directions on policy; by audit, ordinarily by the Comptroller and Auditor General; and by the courts, which exercise writ jurisdiction over it if it is an authority within Article 12, and otherwise entertain ordinary civil proceedings.

Liability

Contractual liability

The corporation contracts in its own name, and its contracts are governed by the ordinary law of contract and by its own statute and regulations.

Article 299 does not apply. That Article governs contracts made in the exercise of the executive power of the Union or of a State. A statutory corporation is a distinct legal person and its contracts are its own, not the Government's, so the three formal requirements discussed in [Contractual Liability of the State] have no application. This is a favourite examination point, because students apply Article 299 to everything.

The corporation is nevertheless bound by its Act: a contract beyond its objects or powers is ultra vires and void, since the corporation exists only to the extent its statute permits.

Where the corporation is the State within Article 12, its contracting is also subject to Article 14, so a tender may not be awarded arbitrarily.

Tortious liability

The corporation is liable for the torts of its servants committed in the course of employment, on ordinary principles of vicarious liability. Article 300 does not apply, because it concerns suits by and against the Government; the corporation is sued in its own name under its own Act.

The sovereign function defence is not available to it in the ordinary case, since running a transport service, a bank or a factory is precisely the kind of activity that N. Nagendra Rao and Co. v. State of Andhra Pradesh, AIR 1994 SC 2663, places outside the primary and inalienable functions of the State: [Sovereign and Non-Sovereign Functions].

Criminal liability

A corporation, being a legal person, can be prosecuted for offences that do not require imprisonment as the only punishment, and its officers may be liable personally where the statute so provides.

In one line

Most grievances against the administration are settled without any adjudication at all, by somebody helping the parties reach an agreement, and that is what conciliation and mediation are.

In the words a student can write in an exam: conciliation and mediation are informal methods of settling disputes in which a neutral third person assists the parties to reach a settlement, without imposing a decision on them. They differ from adjudication, whether by a court or by a tribunal, in that the outcome is an agreement of the parties rather than a determination binding on them, and they differ from arbitration in that the neutral has no power to decide. In the administrative field they are important because the overwhelming majority of grievances against public authorities never reach a court, and because they belong to the green light approach of preventing and resolving disputes rather than reviewing decisions after the event.

Where this fits in the subject

This is the green light half of the syllabus. Everything in Modules II and III is about what a court does after an administrative decision has gone wrong. This chapter and the next are about the machinery that resolves the grievance without any court at all: [Red Light and Green Light Theories].

The scale is worth stating. For every citizen who files a writ petition about a licence, a pension, a ration card or a water connection, there are thousands who complain to the department, to a grievance cell, to a Lokayukta or to a consumer forum, or who settle. A subject that describes only the writ jurisdiction has described the exception.

The vocabulary

The line between conciliation and mediation is not sharp, and Indian statutes have used the words interchangeably at times. The traditional distinction is that a conciliator may put forward proposals for settlement while a mediator confines himself to helping the parties find their own; but the Mediation Act 2023 defines mediation broadly enough to include processes described as conciliation, so an answer should state the classical distinction and note that the statute has largely merged them.

The Indian statutory framework

Section 89 of the Code of Civil Procedure 1908 requires a court, where it appears that there exist elements of a settlement acceptable to the parties, to formulate the terms and refer the dispute to arbitration, conciliation, judicial settlement including settlement through a Lok Adalat, or mediation. It is the provision that brought court referred mediation into ordinary civil litigation.

In one line

Administration is influenced not only by law but by publicity, by organised persuasion and by letting the people affected have their say before a decision is taken.

In the words a student can write in an exam: besides the legal controls on administrative action there are political and social influences on it, and the syllabus names three. The media informs the public about administrative action and creates the pressure that makes formal accountability mechanisms operate, its freedom deriving from Article 19(1)(a) and being subject to Article 19(2). Lobbying is the organised representation of an interest to those exercising public power, which is legitimate where it is transparent and objectionable where it is concealed or purchased. Public participation is the involvement of affected persons and the general public in administrative decision making before the decision is taken, through consultation, publication of drafts, public hearings and access to information.

Why these belong in a law syllabus

Because administrative law is not only about courts, and the syllabus is right to include them. The formal mechanisms in [Public Accountability] all depend on somebody knowing that something has gone wrong and being willing to raise it. That is what the media does. And the controls in Module II which operate before a rule is made, publication and consultation, are the legal form of public participation: [Other Controls: Publication, Consultation and Procedural Safeguards].

The media

Its constitutional basis

There is no separate guarantee of freedom of the press in the Indian Constitution. The freedom of the press is part of the freedom of speech and expression under Article 19(1)(a), and it is subject to the reasonable restrictions in Article 19(2): the sovereignty and integrity of India, the security of the State, friendly relations with foreign States, public order, decency or morality, contempt of court, defamation and incitement to an offence.

Its functions in administrative accountability

1. Disclosure. Reporting what an authority has done, which is often the first anybody outside the department knows of it.
2. Scrutiny. Examining a decision against the reasons given for it.
3. Agenda setting. Making an issue important enough that the legislature, an auditor or a court takes it up.
4. Sanction. Publicity is itself a consequence, and often the only immediate one, as noted in [Public Accountability].
5. Supplying the raw material of litigation, since a report often prompts the petition.

Its limits

Trial by media, in which an accusation is treated as a finding, prejudices both the person accused and any fair inquiry. Commercial and ownership pressures, which affect what is reported. Access, since the media can only report what it can find out, which is why the Right to Information Act 2005 matters more to journalism than any press law. Article 19(2) restrictions, and the law of contempt and defamation.

In one line

When something goes badly wrong in public life, the Government appoints a commission to find out what happened, and the Commissions of Inquiry Act 1952 is the statute under which it does so.

In the words a student can write in an exam: the Commissions of Inquiry Act 1952 empowers the appropriate Government to appoint a Commission of Inquiry to inquire into any definite matter of public importance, either on its own motion or when a resolution of the legislature so requires. The Commission has the powers of a civil court for specified purposes, may require the production of documents and the attendance of witnesses, must give a person likely to be prejudicially affected an opportunity of being heard and a right of cross-examination and legal representation, and reports its findings to the Government. Its report is not a judgment: it binds nobody, creates no liability and is not enforceable, and its value lies in establishing facts and fixing responsibility for the information of the Government and the public.

The scheme of the Act, section by section

The Act is short, twelve numbered sections with several lettered insertions, and the whole of it is manageable.

Appointment: section 3

The appropriate Government may, if it is of opinion that it is necessary so to do, and shall, if a resolution in that behalf is passed by the House of the People or the Legislative Assembly of the State, by notification in the Official Gazette, appoint a Commission of Inquiry for the purpose of making an inquiry into any definite matter of public importance, and performing such functions and within such time as may be specified in the notification.

Three points, and all are examinable.

In one line

An Ombudsman is an independent officer who investigates a citizen's complaint of maladministration against the government, free of cost and without the formality of a court.

In the words a student can write in an exam: the Ombudsman is an institution of Scandinavian origin consisting of an officer appointed by and answerable to the legislature, independent of the executive, who investigates complaints of maladministration by public authorities, has access to official files, conducts his inquiry informally and in private, and reports his findings with recommendations rather than making binding orders. The Indian adaptations are the Lokpal at the Union level and the Lokayukta in the States, recommended by the Administrative Reforms Commission in 1966 and given statutory form by the Lokpal and Lokayuktas Act 2013, though the Indian bodies are concerned principally with corruption rather than with maladministration generally.

The word and the original

Ombudsman is a Swedish word meaning a representative or agent, and it is the office of the Justitieombudsman established in Sweden in 1809 that is the ancestor of all the others. The office spread to Finland, Denmark and Norway in the twentieth century, and from there to the common law world: New Zealand in 1962 was the first common law country to adopt it, and the United Kingdom followed with the Parliamentary Commissioner for Administration in 1967.

What an Ombudsman is for

The gap it fills should be stated first, because it is the whole justification for the office.

A citizen who has been treated badly by a department has, in theory, several remedies. In practice each has a difficulty.

Maladministration is the gap. It covers delay, incompetence, ineptitude, neglect, arbitrariness, rudeness, failure to follow procedure, failure to reply, and bias falling short of anything a court would call illegality. None of that is unlawful in a sense a writ court can address, and all of it is what citizens actually complain of.

The classical characteristics

Seven, and the Indian institutions should be measured against them.

1. Independence of the executive, secured by appointment through a process the executive does not control, a fixed tenure, a salary charged on the Consolidated Fund and removal only by a difficult process.
2. Answerable to the legislature, to which the Ombudsman reports.
3. Jurisdiction over maladministration, not merely illegality.
4. Access on complaint by a citizen, ordinarily free of cost and without a lawyer.
5. Wide powers of investigation, including access to official files.
6. Informal, inquisitorial and private procedure, rather than an adversarial hearing.
7. Recommendation, not adjudication. The Ombudsman reports and recommends; the force of the recommendation lies in publicity and in the report to the legislature.

In one line

The Lokpal is a statutory body which inquires into allegations of corruption against public servants of the Union, including the Prime Minister, and can order investigation, prosecution and the attachment of assets.

In the words a student can write in an exam: the Lokpal and Lokayuktas Act 2013 establishes the Lokpal for the Union, consisting of a Chairperson and not more than eight Members of whom not less than half are Judicial Members, appointed by the President on the recommendation of a Selection Committee. It has jurisdiction over allegations of corruption, under the Prevention of Corruption Act 1988, against a person who is or has been the Prime Minister, a Union Minister, a member of either House of Parliament, or an officer or employee of the Union or of a body established by an Act of Parliament or financed or controlled by the Central Government. It has an Inquiry Wing and a Prosecution Wing, the powers of a civil court, powers of search and seizure, of provisional attachment and confiscation of assets, and power to grant sanction for prosecution, and section 63 requires every State to establish a Lokayukta.

The scheme of the Act

Sixty-three sections. Grouping them is the way to hold them.

Establishment and composition, sections 1 to 13

• Section 3: establishment of the Lokpal, consisting of a Chairperson and not more than eight Members, of whom not less than fifty per cent shall be Judicial Members, and not less than fifty per cent shall be from the Scheduled Castes, Scheduled Tribes, other backward classes, minorities and women.
• Section 4: appointment of the Chairperson and Members by the President on the recommendation of a Selection Committee consisting of the Prime Minister as Chairperson, the Speaker of the House of the People, the Leader of Opposition in the House of the People, the Chief Justice of India or a Judge of the Supreme Court nominated by him, and one eminent jurist nominated by the President on the recommendation of the first four.
• Sections 5 to 9: filling of vacancies, term of office, salary and conditions, restriction on employment after ceasing to hold office, and a Member acting as Chairperson.
• Sections 10 to 13: the Secretary and staff, the Inquiry Wing under section 11 for conducting preliminary inquiry, the Prosecution Wing under section 12 for prosecuting public servants, and the charging of the Lokpal's expenses on the Consolidated Fund of India under section 13, which is a guarantee of financial independence.

Jurisdiction, sections 14 to 19

Section 14 is the central provision and should be known in detail. The Lokpal shall inquire, or cause an inquiry to be conducted, into any matter involved in, arising from or connected with any allegation of corruption made in a complaint in respect of:

In one line

Maharashtra's Lokayukta still operates under an Act of 1971, and a new Act passed to replace it has been only partly brought into force, so the old one is still the law.

In the words a student can write in an exam: the Maharashtra Lokayukta and Upa-Lokayuktas Act 1971 established the office of the Lokayukta and Upa-Lokayuktas for Maharashtra to investigate grievances and allegations against Ministers, Secretaries and other public servants, on complaint or of the Lokayukta's own motion, and to report findings and recommendations to the competent authority. A new statute, the Maharashtra Lokayukta Act 2023, enacted as Maharashtra Act XLVI of 2025, has been passed on the lines of the Lokpal and Lokayuktas Act 2013 and will repeal the 1971 Act by its section 66; but only sections 1, 3 and 4 have been brought into force, the remaining sections awaiting notification by the State Government, so the 1971 Act remains the operative law.

Why Maharashtra has one at all

Two reasons, separated by forty years.

The Administrative Reforms Commission's recommendation of 1966, described in [The Ombudsman], proposed a Lokayukta for each State. Maharashtra was among the earliest States to act on it, enacting the 1971 Act while the Union Lokpal Bills were still lapsing in Parliament.

Section 63 of the Lokpal and Lokayuktas Act 2013, which provides that every State shall establish a body to be known as the Lokayukta for the State, if not so established, constituted or appointed, by a law made by the State Legislature, to deal with complaints relating to corruption against certain public functionaries, within a period of one year from the date of commencement of that Act. That section is why States revisited their Lokayukta legislation after 2014, and it is recited in the preamble of the Maharashtra Act of 2023.

The 1971 Act

Twenty-two sections. The scheme is that of a classical Ombudsman rather than of a corruption body, and that is its most important feature.

Appointment and independence, sections 3 to 6

• Section 3: appointment of the Lokayukta and one or more Upa-Lokayuktas by the Governor.
• Section 4: the Lokayukta and Upa-Lokayuktas to hold no other office.
• Section 5: term of office and other conditions of service.
• Section 6: removal, which is made difficult so as to secure independence.

What may be investigated, sections 7 and 8

Section 7 is the central provision. Subject to the Act, the Lokayukta may investigate any action taken by, or with the general or specific approval of, a Minister or a Secretary, or a public servant referred to in section 2(k)(iii), or any other public servant of a class notified by the State Government in consultation with the Lokayukta, in any case where a complaint involving a grievance or an allegation is made in respect of that action, or where the action can be or could have been, in the Lokayukta's opinion, the subject of a grievance or an allegation.

In one line

The Central Vigilance Commission is the apex body that supervises vigilance work in the central government, advises departments on disciplinary cases of corruption, and superintends the anti-corruption work of the Delhi Special Police Establishment.

In the words a student can write in an exam: the Central Vigilance Commission was set up by a government resolution in 1964 on the recommendation of the Santhanam Committee, was given statutory status by the Central Vigilance Commission Act 2003 following the directions of the Supreme Court, and consists of a Central Vigilance Commissioner as Chairperson and not more than two Vigilance Commissioners, appointed by the President on the recommendation of a committee of the Prime Minister, the Minister of Home Affairs and the Leader of the Opposition. It exercises superintendence over the functioning of the Delhi Special Police Establishment in so far as it investigates offences under the Prevention of Corruption Act 1988, inquires or causes inquiries into offences alleged against specified categories of public servants, tenders advice to the Central Government and its authorities on vigilance matters, and functions as the designated agency under the Public Interest Disclosure and Protection of Informers Resolution 2004.

The history

Three steps, and the sequence is examinable.

1. The Santhanam Committee on Prevention of Corruption, which reported in 1964, recommended a central body to advise and guide central government agencies in vigilance matters.

2. The Commission set up by a government resolution in 1964. For over thirty years it existed only as an executive body, without statutory backing, which limited both its powers and its independence.

3. Statutory status by the Central Vigilance Commission Act 2003, which followed the Supreme Court's directions in the case commonly known as the hawala case, in which the Court directed that the Commission be given statutory status and that the Central Bureau of Investigation be insulated from executive interference. That decision is identified in authorities/cases.json and has not been read in a session, so it is named here and its holding is not stated in detail.

The scheme of the Act

Twenty-seven sections. The important ones are these.

Constitution and appointment, sections 3 to 7

• Section 3: constitution of the Central Vigilance Commission, consisting of a Central Vigilance Commissioner as Chairperson and not more than two Vigilance Commissioners as Members.
• Section 4: appointment of the Central Vigilance Commissioner and Vigilance Commissioners by the President by warrant under his hand and seal, on the recommendation of a committee consisting of the Prime Minister as Chairperson, the Minister of Home Affairs and the Leader of the Opposition in the House of the People.

The inclusion of the Leader of the Opposition is the independence safeguard, and it should be mentioned. It follows the pattern used for other watchdog appointments and is comparable to the Selection Committee under section 4 of the Lokpal and Lokayuktas Act 2013: [Lokpal: the Lokpal and Lokayuktas Act 2013].

In one line

A legislature cannot scrutinise the administration on the floor of the House, so it does it in committees, and those committees are the working part of legislative control.

In the words a student can write in an exam: parliamentary committees are bodies of members appointed or elected by the House, or nominated by the Speaker, which examine matters referred to them and report to the House. They are classified into standing committees, which are permanent, and ad hoc committees, appointed for a particular purpose. The financial committees, being the Public Accounts Committee, the Estimates Committee and the Committee on Public Undertakings, together with the departmentally related standing committees and the Committee on Subordinate Legislation, are the principal instruments by which the legislature controls the administration, since they can examine officials, call for documents and report publicly, which the House sitting as a whole cannot do.

Why committees exist

The House as a whole cannot do this work, for four reasons.

1. Time. A House sits for a limited number of days and must legislate, debate and vote supplies.
2. Numbers. Five hundred and more members cannot examine an official.
3. Publicity. Proceedings on the floor are adversarial and political; committee proceedings are in private and are markedly less partisan, which is why they produce agreed reports.
4. Expertise. A small committee returning to the same department year after year develops knowledge that the House cannot.

Classification

Standing committees are permanent and are constituted every year or every term: the financial committees, the departmentally related standing committees, and committees on the business of the House such as those on privileges, petitions, subordinate legislation and government assurances.

Ad hoc committees are appointed for a particular purpose and cease to exist when they report: select and joint committees on a Bill, and committees appointed to inquire into a particular matter.

The power to constitute them comes from the rules of procedure made under Article 118 for Parliament and Article 208 for a State legislature, and the privileges of members and committees are those in Articles 105 and 194.

The three financial committees

These are the heart of the topic.

1. The Public Accounts Committee

What it does. It examines the accounts showing the appropriation of sums granted by the House, the annual finance accounts and appropriation accounts, and above all the reports of the Comptroller and Auditor General. It asks whether the money was spent on the purpose for which it was voted, whether the expenditure was authorised, and whether there was waste, loss, extravagance or irregularity.

Its relationship with the auditor. The Comptroller and Auditor General is appointed under Article 148 and his reports are laid before the House under Article 151. The Committee is where those reports are examined, and the Comptroller and Auditor General assists it. That pairing, an independent auditor and a committee to act on his reports, is the strongest financial accountability mechanism India has: [Public Accountability].

In one line

The civil service is the permanent body of officials that carries on the administration whatever government is in office, and Part XIV of the Constitution both secures its position and limits it.

In the words a student can write in an exam: the civil services are the permanent, professional and politically neutral body of officials who advise ministers on policy and carry it into execution. In India they are organised into All India Services, common to the Union and the States, Central Services and State Services. Part XIV of the Constitution governs them: Article 309 empowers the appropriate legislature to regulate recruitment and conditions of service and the executive to make rules until it does; Article 310 embodies the doctrine of pleasure, under which a person serves during the pleasure of the President or the Governor; Article 311 cuts that doctrine down by requiring that no such person be dismissed or removed by an authority subordinate to that which appointed him, and that none be dismissed, removed or reduced in rank except after an inquiry in which he has been informed of the charges and given a reasonable opportunity of being heard; Article 312 provides for the creation of new All India Services; and Articles 315 to 323 establish the Public Service Commissions.

The nature of a civil service

Five characteristics, and they explain the constitutional provisions that follow.

1. Permanence. Governments change; the service continues. That is what allows administration to be carried on across a change of ministry.
2. Professionalism. Entry by competitive examination, training, and a career structure.
3. Political neutrality. The civil servant serves the government of the day whatever its politics, which is what makes permanence possible.
4. Anonymity. The minister answers publicly; the officer advises privately. This is the counterpart of ministerial responsibility described in [Public Accountability], and it is what makes it possible for the minister to be answerable for the department.
5. Hierarchy and rules. Decisions are taken within a structure and according to rules, which is what makes them predictable and reviewable.

Anonymity is under strain, and it is worth saying so. The Right to Information Act 2005 makes file notings obtainable, so the officer who advised is identifiable: [Obligations of Public Authorities and How a Request Is Made]. The convention was built for a world in which the file was closed.

The organisation

In one line

Accountability asks whether an official can be called to answer for what he did; responsiveness asks whether he does what the citizen actually needs; and the Indian civil service is criticised on both counts for reasons that are structural rather than personal.

In the words a student can write in an exam: accountability is the obligation of an official to answer for the exercise of his authority to a person or body entitled to demand an account, and responsiveness is the readiness of the administration to attend and adapt to the needs and grievances of the citizens it serves. The two are distinct: an administration may be fully accountable in the sense of being auditable and reviewable, and yet be slow, rigid and indifferent. The problems in India are structural, arising from the tension between security of tenure and accountability, from rule bound procedure that displaces the purpose of the rule, from the diffusion of responsibility across a hierarchy and a file, and from the absence of measurement, and the reform perspectives address each of them.

Accountability and responsiveness distinguished

The two can pull against each other, and saying so is the mark of a good answer. An official who fears being held accountable for a wrong decision will avoid deciding, refer the file upwards, ask for another opinion and insist on one more document. That is perfect accountability and complete unresponsiveness.

The accountability mechanisms as they apply to the civil service

Each is treated in full elsewhere; here they are gathered as they bear on the individual officer.

1. Hierarchical. The officer answers to his superior, whose control includes inspection, reporting and the annual assessment.
2. Political. The minister answers to the legislature under Articles 75(3) and 164(2), and the officer answers to the minister. Anonymity means the officer is not publicly named, which is what makes ministerial responsibility work.
3. Financial. The Comptroller and Auditor General under Article 148 and the reports laid under Article 151, examined by the financial committees: [Congressional and Parliamentary Committees].
4. Disciplinary. Departmental proceedings, subject to Article 311's requirement of an inquiry with charges and a reasonable opportunity of being heard: [Civil Services in India: Nature, Organisation, Powers and Functions].
5. Criminal. The Prevention of Corruption Act 1988: [The Prevention of Corruption Act 1988].
6. Vigilance. The Central Vigilance Commission and departmental vigilance officers: [The Central Vigilance Commission].
7. Judicial. Judicial review, subject to the limit that the court examines legality and not merit: Asif Hameed v. State of Jammu and Kashmir, AIR 1989 SC 1899.
8. Public. The Right to Information Act 2005, and the press.

In one line

Administrative deviance is the general name for an official departing from the standard his office requires, and it takes three forms: taking a bribe, favouring his own, and simply administering badly.

In the words a student can write in an exam: administrative deviance means conduct by a public official which departs from the norms of his office. Its principal forms are corruption, the use of public office for private gain; nepotism and favouritism, the conferring of appointments or benefits on relatives or associates rather than on merit; and maladministration, which is bad administration falling short of illegality, comprising delay, incompetence, neglect, inattention, failure to follow procedure and unreasonable, unjust or oppressive action. The three call for different remedies: corruption for the criminal law, nepotism for the constitutional guarantee of equality and the discipline of merit based recruitment, and maladministration for the Ombudsman, the citizens' charter and the right to information, because no court can address it.

The three defined and distinguished

Corruption

The use of public office for private gain. Its forms include demanding or accepting a bribe, extortion under colour of office, misappropriation of public money or property, using inside information for gain, and criminal misconduct in the sense the Prevention of Corruption Act 1988 gives to that phrase.

Its distinguishing feature is a private benefit obtained by the abuse of a public position. It is a criminal offence and is treated in [The Prevention of Corruption Act 1988].

Nepotism and favouritism

Nepotism is the favouring of relatives; the word comes from the Latin nepos, a nephew, and from the practice of popes appointing their nephews. Favouritism is the wider term, covering friends, caste, community, region and political affiliation.

Its distinguishing feature is that the criterion applied is a relationship rather than merit. Note that no money need change hands, which is why it is a distinct category and not a species of corruption.

Its primary remedy is constitutional, not criminal. Article 14 forbids arbitrary State action and Article 16 guarantees equality of opportunity in matters of public employment. An appointment made because of a relationship rather than on the prescribed criteria is an irrelevant consideration and is arbitrary: [Grounds of Judicial Review: Abuse of Discretion]. If the person deciding is himself related to the candidate, it is personal bias: [The Rule Against Bias].

Maladministration

Bad administration that is not illegality: delay, incompetence, ineptitude, neglect, failure to reply, failure to follow the department's own procedure, arbitrariness and rudeness.

The Maharashtra Lokayukta and Upa-Lokayuktas Act 1971 supplies a statutory definition, and quoting it is worth doing: maladministration means action taken or purporting to have been taken in the exercise of administrative functions in any case where the action, or the administrative procedure or practice governing it, is unreasonable, unjust, oppressive or improperly discriminatory, among other limbs: [Lokayukta in Maharashtra: the 1971 Act, and the 2023 Act awaiting commencement].

In one line

The Prevention of Corruption Act 1988 makes it an offence for a public servant to take an undue advantage, for anybody to give one, and for a public servant to enrich himself illicitly, and it was substantially rewritten in 2018.

In the words a student can write in an exam: the Prevention of Corruption Act 1988 consolidates the law relating to the offences of bribery and criminal misconduct by public servants. As amended by the Prevention of Corruption (Amendment) Act 2018, which came into force on 26 July 2018, it makes it an offence under section 7 for a public servant to obtain or accept an undue advantage with the intention to perform a public duty improperly, under section 8 for any person to give one, under section 9 for a commercial organisation to bribe, under section 11 for a public servant to obtain a valuable thing without consideration from a person concerned in a proceeding or business transacted by him, and under section 13 to commit criminal misconduct by misappropriating property or by intentionally enriching himself illicitly. It provides for trial by special judges, requires previous approval under section 17A before an enquiry or investigation into an offence relatable to a recommendation made or decision taken by a public servant in the discharge of official functions, and requires previous sanction under section 19 for prosecution.

The structure of the Act

Thirty-one numbered sections with lettered insertions. Four groups.

Group 1: courts and procedure, sections 3 to 6

• Section 3: power of the Central or State Government to appoint special Judges to try offences under the Act.
• Section 4: cases triable by special Judges, and the direction that the trial be held as expeditiously as possible, on a day to day basis.
• Section 5: procedure and powers of a special Judge.
• Section 6: power to try summarily in specified circumstances.

Group 2: the offences, sections 7 to 16

This is the heart of the Act and the part the 2018 amendment rewrote.

In one line

India passed a law to protect whistleblowers in 2014 and has never brought it into force, so the protection that actually exists is a government resolution of 2004 administered by the Central Vigilance Commission.

In the words a student can write in an exam: the Whistle Blowers Protection Act, being Act 17 of 2014, establishes a mechanism to receive complaints disclosing allegations of corruption or wilful misuse of power or discretion against a public servant, to inquire into such disclosures, and to provide safeguards against the victimisation of the person making the complaint. It has not been brought into force, because section 1(3) makes its commencement dependent on a notification by the Central Government and no such notification has been issued; the Government's stated reason is that the Act requires amendment to safeguard against disclosures affecting the sovereignty and integrity of India and the security of the State. The protection that operates in practice is the Public Interest Disclosure and Protection of Informers Resolution 2004, under which the Central Vigilance Commission is the designated agency.

Why a whistleblower needs protection at all

The logic should be stated first, because it explains everything that follows.

Corruption is transacted privately between two people who both benefit. There is no complainant, no injured party present at the transaction and no document. The only people who know are the participants and those inside the organisation who see the pattern.

So the detection of corruption depends on somebody inside deciding to speak, and that person faces immediate and predictable consequences: transfer, suspension, a departmental inquiry on a manufactured charge, denial of promotion, harassment of family, and in a number of documented Indian cases, physical harm and death.

Hence the design principle: protection must be automatic and identity must be concealed, because a person who must first prove that his transfer was retaliatory has already been transferred.

The Act of 2014

Its identity, which is confusing

Three names for one statute, and an examiner may use any of them.

• MU's syllabus calls it the Whistleblowers' Protection Act, 2014.
• The Act's own short title is the Whistle Blowers Protection Act, 2011, because the short title kept the year of the Bill.
• It is Act 17 of 2014, because the assent came on 9 May 2014 and it was published in the Gazette of India on 12 May 2014.

All three refer to the same statute. It began as the Public Interest Disclosure and Protection to Persons Making the Disclosures Bill 2010, was renamed, was passed by the Lok Sabha on 27 December 2011 and by the Rajya Sabha on 21 February 2014.

Its scheme

Thirty-one sections in seven chapters.

In one line

The right to know is part of the freedom of speech, because a citizen who cannot find out what the government is doing cannot meaningfully speak about it or vote on it.

In the words a student can write in an exam: the right to information is not created by the Right to Information Act 2005; it is a facet of the freedom of speech and expression guaranteed by Article 19(1)(a), and is subject to the reasonable restrictions in Article 19(2). The Supreme Court recognised it before any statute existed, holding that in a government of responsibility where all agents of the public must be responsible for their conduct there can be but few secrets, and that the people have a right to know every public act. The Act of 2005 gave that constitutional right a practical machinery: a designated officer, a time limit, a fee, an appeal and a penalty, without which the right existed but could not be exercised.

Why transparency is a constitutional question and not merely a good idea

Four arguments, and a strong answer gives all four.

1. The argument from free speech. Article 19(1)(a) guarantees the freedom of speech and expression. Speech about public affairs that is uninformed is worth little, and a citizen who cannot obtain the facts cannot form or express an opinion about them. The right to receive and impart information is therefore an element of the freedom itself, not an addition to it.

2. The argument from democracy. In a representative democracy the people are the principal and the government is the agent. An agent who need not account to his principal is not an agent. The right to know follows from the theory of representative government.

3. The argument from accountability. Every mechanism in [Public Accountability] needs evidence. A question in the legislature, an audit objection, a complaint to a Lokayukta, a departmental proceeding and a writ petition all begin with somebody knowing something. Transparency is the precondition of the rest.

4. The argument from the control of discretion. Where an official has wide discretion, the citizen's protection is that the criteria are published, the decision is reasoned and the file can be seen. That connects this chapter to [Administrative Discretion and Why It Is Conferred] and [Reasoned Decisions].

The judicial foundation

The right existed before the statute, and Indian courts built it out of Article 19(1)(a). The line of decisions is identified in authorities/cases.json, and the two that matter most for this chapter are the case on the disclosure of the Blue Book concerning the Prime Minister's security, where the Court said that in a government of responsibility there can be but few secrets and that the people have a right to know every public act, and the case on judges' appointments and transfers, where the Court held that disclosure of information about the functioning of government is the rule and secrecy the exception. Those two decisions have not been read in a session and are therefore named without citations here, in accordance with the rule in authorities/cases.json.

In one line

The Act is short, it applies to every public authority in India except a listed few, and everything in it turns on four defined words: information, record, right to information and public authority.

In the words a student can write in an exam: the Right to Information Act 2005, being Act 22 of 2005, was enacted to set out the practical regime of the right to information for citizens, so as to promote transparency and accountability in the working of every public authority. It extends to the whole of India, applies to all public authorities as defined in section 2(h), which includes bodies established by or under the Constitution, by any law of Parliament or a State legislature, and by notification or order of the appropriate Government, together with bodies owned, controlled or substantially financed by it and non-governmental organisations substantially financed by it. Its salient features are a broadly defined right, a designated officer in every authority, a simple procedure with no requirement to give reasons, a strict time limit, defined exemptions, a two tier appeal to independent Information Commissions, a personal penalty on the defaulting officer, and a duty of proactive publication.

The object

The long title states that it is an Act to provide for setting out the practical regime of right to information for citizens to secure access to information under the control of public authorities, in order to promote transparency and accountability in the working of every public authority, the constitution of a Central Information Commission and State Information Commissions and for matters connected therewith or incidental thereto.

The words "practical regime" are the whole point and are worth quoting. The right existed already; what the Act provides is a regime in which it can be used, which is the argument developed in [Transparency and the Right to Information: a Constitutional Imperative].

The preamble adds the balance: transparency is vital to democracy and to containing corruption; disclosure conflicts with other public interests including efficient operation of governments, optimum use of limited fiscal resources and the preservation of confidentiality of sensitive information; and these are to be harmonised while preserving the paramountcy of the democratic ideal.

Application

Section 1 provides for the short title, extent and commencement. The Act extends to the whole of India.

Two limits on application must be given.

1. It applies to public authorities, as defined, and not to private bodies as such. But information relating to a private body which can be accessed by a public authority under any other law is within the definition of information in section 2(f), so a regulator's file about a private company is obtainable.
2. Section 24 excludes certain organisations, being the intelligence and security organisations specified in the Second Schedule, with a proviso for information about allegations of corruption and human rights violations. That is treated in [Exemptions from Disclosure].

In one line

A public authority must publish a great deal without being asked, must appoint an officer to receive requests, and must answer a request within thirty days or be treated as having refused it.

In the words a student can write in an exam: section 4 imposes on every public authority a duty of proactive disclosure, requiring it to maintain its records catalogued and indexed, to publish seventeen classes of information within one hundred and twenty days of the enactment of the Act and to update them, to publish all relevant facts while formulating important policies, to provide reasons for its administrative and quasi-judicial decisions to affected persons, and to endeavour to provide information suo motu so that the public has minimum resort to the use of the Act. Section 5 requires the designation of Public Information Officers and Assistant Public Information Officers. Section 6 provides for a request in writing or electronically, with no reason required. Section 7 requires disposal within thirty days, or forty-eight hours where life or liberty is concerned, failing which the request is deemed refused.

Section 4: the duty to publish without being asked

This is the most important and least used section of the Act, and it deserves the emphasis. Sections 6 to 19 create a right to ask. Section 4 creates a duty to tell, which reaches every citizen and not only the one who applies.

Section 4(1)(a): records

Every public authority shall maintain all its records duly catalogued and indexed in a manner and form which facilitates the right to information, and shall ensure that records appropriate to be computerised are, within a reasonable time and subject to the availability of resources, computerised and connected through a network so that access is facilitated.

The practical significance is large: an authority that cannot find its records cannot be made to disclose them, so record management is a precondition of the whole Act.

Section 4(1)(b): the seventeen classes

Every public authority shall publish, within one hundred and twenty days from the enactment of the Act, among other things:

In one line

The Act lists ten kinds of information a public authority need not give, one further ground of rejection, and a list of organisations to which it does not apply, and over all of them stands a public interest override.

In the words a student can write in an exam: section 8(1) of the Right to Information Act 2005 sets out ten heads of exemption from the obligation to disclose; section 9 permits rejection where disclosure would infringe a copyright subsisting in a person other than the State; and section 24 excludes the intelligence and security organisations listed in the Second Schedule, save in respect of allegations of corruption and human rights violations. Section 8(2) contains a public interest override, under which a public authority may allow access notwithstanding the Official Secrets Act 1923 and notwithstanding any exemption in section 8(1), if the public interest in disclosure outweighs the harm to the protected interests; section 8(3) makes information more than twenty years old disclosable subject to three clauses; and section 10 requires severance so that the non-exempt part of a record is given.

The ten exemptions in section 8(1)

Section 8(1) opens: notwithstanding anything contained in this Act, there shall be no obligation to give any citizen the following.

The provisos to clause (i)

The first proviso is important and is often missed: the decisions of the Council of Ministers, the reasons for them, and the material on the basis of which the decisions were taken shall be made public after the decision has been taken and the matter is complete or over. So cabinet secrecy protects deliberation while the matter is live, not the decision afterwards. The second proviso adds that matters coming under the exemptions specified in the section shall not be disclosed.

In one line

The Information Commissions are the independent appellate bodies that decide whether information must be given, and the 2019 amendment moved their tenure and salary from the statute into rules made by the Government.

In the words a student can write in an exam: the Central Information Commission is constituted by the Central Government under section 12 and consists of the Chief Information Commissioner and not more than ten Central Information Commissioners, appointed by the President on the recommendation of a committee of the Prime Minister as Chairperson, the Leader of Opposition in the Lok Sabha and a Union Cabinet Minister nominated by the Prime Minister. State Information Commissions are constituted under section 15 on the same pattern, with the Governor appointing on the recommendation of the Chief Minister, the Leader of Opposition in the Legislative Assembly and a State Cabinet Minister nominated by the Chief Minister. The Commissions decide second appeals under section 19 and complaints under section 18, and in deciding them have the powers of a civil court.

The Central Information Commission

Constitution, section 12

Section 12(1): the Central Government shall, by notification in the Official Gazette, constitute a body to be known as the Central Information Commission to exercise the powers conferred on and perform the functions assigned to it under the Act.

Section 12(2): the Commission shall consist of the Chief Information Commissioner and such number of Central Information Commissioners, not exceeding ten, as may be deemed necessary.

Section 12(3): they shall be appointed by the President on the recommendation of a committee consisting of:

1. the Prime Minister, who shall be the Chairperson of the committee;
2. the Leader of Opposition in the Lok Sabha; and
3. a Union Cabinet Minister nominated by the Prime Minister.

An Explanation provides that where the Leader of Opposition in the House of the People has not been recognised as such, the Leader of the single largest group in opposition to the Government in that House shall be deemed to be the Leader of Opposition. That Explanation exists because the office is not always filled, and it prevents the selection committee being reduced to two members of the Government.

Note the composition: two of the three members are the Prime Minister and a Minister he nominates. That is the structural criticism of the appointment process, and it should be stated.

Section 12(4): the general superintendence, direction and management of the affairs of the Commission vest in the Chief Information Commissioner, who is assisted by the Information Commissioners, and who may exercise all such powers and do all such acts as may be exercised or done by the Commission autonomously without being subjected to directions by any other authority under the Act.

In one line

A citizen refused information appeals first inside the department and then to the Information Commission, and the officer who refused without reasonable cause pays two hundred and fifty rupees a day out of his own pocket.

In the words a student can write in an exam: section 19 provides a two tier appeal. A first appeal lies within thirty days to an officer senior in rank to the Public Information Officer in the same public authority, and a second appeal lies within ninety days to the Central or State Information Commission. In any appeal the burden of proving that a denial was justified lies on the officer who denied it. Section 18 provides a separate complaint jurisdiction. Section 20 empowers the Commission to impose on the Public Information Officer a personal penalty of two hundred and fifty rupees for each day of default, subject to a maximum of twenty-five thousand rupees, and to recommend disciplinary action, after giving him a reasonable opportunity of being heard, the burden of proving that he acted reasonably and diligently being on him.

The appeal structure

Section 19(1): any person who does not receive a decision within the time specified in section 7(1) or 7(3)(a), or is aggrieved by a decision of the Public Information Officer, may within thirty days prefer an appeal to an officer senior in rank to that officer in each public authority. A proviso allows the appeal to be admitted after thirty days where the appellate officer is satisfied that the appellant was prevented by sufficient cause from filing in time.

Section 19(2): where an appeal is preferred against an order made under section 11 to disclose third party information, the appeal by the third party shall be made within thirty days from the date of the order.

Section 19(3): a second appeal lies within ninety days to the Commission, with a like proviso permitting condonation of delay for sufficient cause.

Section 19(4): where the decision appealed against relates to information of a third party, the Commission shall give the third party a reasonable opportunity of being heard.

The remaining sub-sections of section 19 provide, in substance:

In one line

The Act is well drafted and the difficulty is everywhere else: vacancies in the Commissions, backlogs, records that were never kept, officers who refuse and are not penalised, applicants who are attacked, and two amendments that have weakened it.

In the words a student can write in an exam: the hurdles in the implementation of the Right to Information Act 2005 are institutional, administrative, legal and social. Institutionally, the Information Commissions function with vacancies and large backlogs, and their independence was reduced by the Right to Information (Amendment) Act 2019, which moved the tenure and salary of the Commissioners from the statute into rules made by the Government. Administratively, section 4 proactive disclosure is poorly complied with and record management is weak, so that information which should never have been asked for is asked for and cannot be found. Legally, the exemptions are invoked expansively and section 8(1)(j) was substituted with effect from 13 November 2025 to exempt personal information without the balancing test it formerly contained. Socially, applicants have faced harassment and violence, and the whistleblower statute meant to protect them has never been brought into force.

1. Institutional hurdles

Vacancies and backlogs

The Commissions are the enforcement machinery of the Act, and an appeal that takes two years to be heard is a right denied. Vacancies in the posts of Chief Information Commissioner and Information Commissioners have repeatedly been left unfilled, and the resulting backlog is the most frequently cited failure of implementation.

The structural point worth making: the Act gives the Government the power to appoint the body that hears appeals against the Government's own refusals, and imposes no deadline for doing so. Section 12(2) says the Commission shall consist of the Chief Information Commissioner and such number of Information Commissioners not exceeding ten as may be deemed necessary, which leaves both the number and the timing to the appointing authority.

Independence weakened by the 2019 amendment

Sections 13 and 16, as substituted by sections 2 and 3 of the Right to Information (Amendment) Act 2019, being Act 24 of 2019, with effect from 24 October 2019, replaced the fixed statutory term of five years with "such term as may be prescribed by the Central Government", and made salaries, allowances and other conditions of service likewise a matter of prescription by that Government, with a proviso only against variation to disadvantage after appointment.

The objection is the one made about tribunals in [Tribunalisation and the Threat to Judicial Independence: the Tribunals Reforms Act 2021]: an adjudicator whose tenure and pay are fixed by the party that appears before him is not structurally independent.

The 2019 amending Act was itself repealed by the Repealing and Amending Act 2023, which is why India Code lists it as "(Rep., Act 37 of 2023)". That is housekeeping and does not restore the earlier wording, because repealing a spent amending Act does not undo the amendment: section 6A of the General Clauses Act 1897.