Advising the president


back to issue

ON 25 August 1944, Justice Zafarullah Khan, dissenting with the majority, refused to answer a question referred to the Federal Court by the then Governor General of India, the Viscount Wavell. Declining, he said, ‘I do not think any useful purpose would be served by my attempting to frame answers to the questions referred… it is bound to raise ghosts far more troublesome than any it might serve to lay.’1 The question before the court, which Justice Spens answered for the majority, related to whether the federal legislature had the power to make a law to levy estate duty on property or whether such power rightly fell within the domain of provincial legislatures.

Viscount Wavell, exercising powers under Section 213 of the Government of India Act, 1935, referred the question to the Federal Court for an opinion. Justice Zafarullah Khan’s objection was that, ‘A reference relating to a legislative proposal must in a large number of cases be enveloped in a thick fog of hypotheses and uncertainties and an opinion delivered thereon could only rest upon a forest of assumptions which must rob it of all value.’ Justice Spens, who did answer the questions, for the majority, dismissed these objections to the exercise of reference jurisdiction by the Federal Court, acerbically noting that, ‘When Parliament has thought fit to enact Section 213, Constitution Act, it is not in our judgment for the court to insist on the inexpediency (according to a certain school of thought) of the advisory jurisdiction.’

This exchange, in 1944, was the last time that there was a serious dispute in Indian jurisprudence on the merits of the power granted to a federal court to answer questions referred to them by the executive. Like Section 213, Article 143 of the Constitution now permits the President to refer factual or legal questions to the Supreme Court for an opinion, when they arise or are likely to arise, if the President feels it expedient to do so. Since independence, presidents have found it expedient to refer matters to the Supreme Court for an opinion on thirteen occasions. Presumably, on these thirteen occasions, it was not deemed sufficient for the executive to restrain itself to seeking the advice of the Attorney General, whose constitutional duty it is to advise the executive on legal matters.

The Supreme Court has then gone on to exercise its powers to issue an opinion, under Article 143 of the Constitution, in twelve of thirteen references. In all or most of these twelve instances, the Supreme Court has been urged by various private parties and states to decline the reference on various grounds. However, it has refused on only one occasion – on a question of pure fact in 1993, of whether a Hindu temple or any Hindu religious structure existed prior to the construction of the Babri Masjid (a question that the Allahabad High Court was subsequently willing to address at great length).


The readiness of the Supreme Court, therefore, to answer questions that are at best hypothetical, without serious discussion of the merits and demerits of exercising such power, warrants consideration. The Supreme Court is not bound to issue an opinion – it has absolute discretion to decline the same, provided it does not relate to a dispute between states or between the Centre and states, under Article 143(2). When the court does in fact issue an advisory opinion in response to a reference, such opinion is not binding on either legislatures or the executive. There is some debate on whether it amounts to a binding precedent on the judiciary itself, but we may take it that an advisory opinion is just that – an opinion.

What judicial function, therefore, does the Supreme Court exercise in a democratic state, when it acts as advisor to the executive? If ‘expediency’ in the eyes of the President is the only consideration for an overburdened court to take on another responsibility at the cost of separation of powers between the three branches of government, then Article 143 merits reconsideration.

Advisory opinions at the Indian Supreme Court have been issued in respect of executive, legislative and judicial acts. There have been references on questions relating to the constitutionality of proposed bills (such as the Special Courts Bill), and on existing acts (such as the Delhi Laws Act of 1912). These have been acts and bills, both of the state and the central legislature (for instance, the Gujarat Gas Act, 2001 was referred in 2004,2 or the Kerala Education Bill in 19583 ). The Supreme Court has been asked to issue advice on executive acts, such as the implementation of an agreement to resolve territorial disputes between India and Pakistan,4 or on the powers of the Election Commission to reconstitute state legislative assemblies.5 Recently, the Supreme Court was asked to consider the ‘effect’ of a previous order by a bench of the Supreme Court, in what is known as the 2G reference.6 


A common challenge to the power of courts to issue advisory opinions in respect of legislative acts has been that it operates as a form of anticipatory judicial review. The purpose of obtaining such opinions would appear to be to ensure that unconstitutional laws are not enacted. This is especially relevant in India as the President must grant his assent to any bill, under Article 111 of the Constitution, before it becomes law. For instance, the Supreme Court was requested in 1979 to provide an opinion on whether the Special Courts Bill, if enacted, would be constitutionally invalid.7 References such as these are challenged on the grounds that they infringe upon the powers of the legislature to enact laws.


In the reference on the Special Courts Bill, A.K. Sen, arguing for the state of Karnataka, recommended that the Supreme Court dismiss the reference, as the bill itself might or might not become law, and might undergo further changes in Parliament. The question, therefore, he argued, was wholly hypothetical and served no purpose. The counsel for the Leader of the Opposition argued that it was rightly the power and the privilege of the Parliament, and not the Supreme Court, to decide whether the bill should become an act and whether its provisions were unconstitutional. He was supported by the Advocate General for Karnataka, who argued that the Supreme Court was effectively being treated as a Joint Select Committee of Parliament.

These objections were dismissed by the majority. Justice Y.V. Chandrachud considered these arguments seriously, as he went on to say, ‘We were, at one stage of the arguments, so much exercised over the undefined breadth of the reference that we were considering seriously whether in the circumstances it was not advisable to return the reference unanswered.’ The reference was, in fact, answered, but with the stricture that general roving inquiries into constitutionality were not easily entertained and that references should ideally be specific. Constitutionality of a bill, just as an act, wrote Justice Chandrachud, fell within the legitimate domain of courts to decide, given the entrenched powers of judicial review. ‘The ultimate decision on the validity of a law’, he wrote, ‘has to be of the court not of the Parliament.’ The fact that this was, in fact, a bill, and not a law, was not addressed in this context.

The conflict that advisory opinions engender, between courts and the legislature, was more apparent in the reference concerning Keshav Singh, a journalist who published a pamphlet levying charges of corruption against a Member of the Legislative Assembly of Uttar Pradesh. In 1964, Uttar Pradesh was subject to a series of increasingly odd events, as the legislative assembly first charged Keshav Singh with contempt of the House and sentenced him to prison for seven days. Keshav Singh was granted bail by two judges of the Allahabad High Court, and the legislative assembly, taking further offence, charged these two judges and Keshav Singh’s lawyer with contempt as well. A full bench of twenty-eight judges of the Allahabad High Court heard petitions from the two judges claiming that the legislative assembly through its Speaker had committed contempt of court, and the matter was ultimately referred by the President to the Supreme Court for an opinion.


As in a previous reference, this was challenged on the grounds that the President had nothing to do with the matter, that it fell outside the legislative or executive powers of the Union, and consequently, the reference was itself invalid. This contention was rejected on two grounds. The Supreme Court held, first, that the President was empowered to make reference on any question of fact and law which was of public importance, and this included matters outside the Union’s executive or legislative powers. Second, it pointed to the exact utility envisioned for such opinions rendered by the court: ‘ may be also be open to him [the President] to formulate for the advisory opinion of this court questions of constitutional importance like the present, and it may be that the President may, on receiving our answers consider whether the Union government or the state government should be requested to take any suitable or appropriate action, either legislative or executive in accordance with the opinion expressed by this court.’


In addition to reviewing acts (and proposed acts) of legislatures, presidential references have been utilized to review acts of the executive. In 1960, the court was asked to render an opinion on how an agreement between India and Pakistan on resolution of border disputes should be best implemented. Since the resolution of the dispute required cessation of some territory, the question was whether such act by the executive required ratification by the legislature and whether such ratification had to be in the form of a constitutional amendment. The court found that the executive would have to ratify this act by constitutional amendment.

There have been attempts to use presidential references to reconsider judicial pronouncements as well. In the reference concerning the Cauvery River Water Disputes Tribunal, the Supreme Court was asked to clarify the scope of its own judgment on whether the water disputes tribunal could grant interim relief. As the answer in the court’s judgment was specific on the facts, but did not answer a general question on interim relief, the matter was referred for opinion. The Supreme Court in reference, found the answer to the reference ‘implicit’ in its previous decision.

Yet again, in the most recent presidential reference, the Supreme Court was asked to clarify whether its decision in the case commonly known as the ‘2G Case’,8 would lead by implication to the cancellation of various licenses. In the 2G reference, the Supreme Court categorically refused to reopen its previous decision, finding that to revisit it on merits would be ‘absolutely impermissible’. It was argued, and rightly so, that the purpose of a reference is to provide an opinion to the President, and not to serve as an indirect form of appeal in violation of prescribed procedures for review and appeal. The court declined to answer questions specifically relating to the implementation of its previous decision, issuing an opinion only on the first five general questions relating to allocation of natural resources.


A reference, therefore, appears to have served three purposes in the past. The first, as in the Special Courts Bill, is to save the exchequer the time and difficulties contingent upon the passage of a law that is likely to be found to be unconstitutional. Thus, this preliminary review acts as advice on whether to proceed with the bill, or whether to amend it, which, as has been pointed out, is broadly the function of parliamentary committees constituted for this reason. The second purpose that the opinions seems to have served are to guide the executive on actions to be taken in moments of acute constitutional crisis; in other words, advice that the Attorney General is legally obliged to render to the executive under the Constitution.

The third purpose is to ask the Supreme Court to clarify its own opinions in decisions, or, as it is occasionally alleged, to seek an indirect review of past decisions. In both of these instances, the Supreme Court may be approached, either under Article 137 to review, or under the Supreme Court Rules, to file an application for clarification and if required, modification of an order. Recently, curative petitions to correct past orders have also been permitted. What unique purpose, then, does a reference to the Supreme Court serve in a democratic state?

Our Constitution envisions a republic where the separation of powers between the three branches of government safeguards us against excesses of either one of them. When we allow the Supreme Court to transgress its function as a neutral arbiter of disputes between parties, and permit it to act as advisor to the executive, what is the rationale behind such power? This power was the subject of very limited debate during the framing of the Constitution – limited only to a clarification from Ambedkar that the Supreme Court was not, in fact, bound to answer all references. The answer accordingly lies, perhaps, not in what may be referred, but what an advisory opinion implies.


Article 141 of the Constitution provides, ‘The law declared by the Supreme Court shall be binding on all courts within the territory of India.’ It would be consistent with the principle of separation of powers if the Supreme Court’s opinions on presidential references constituted authoritative binding law on the entire nation. The implication would then be that the Supreme Court in advising, as in adjudicating, acts neutrally to determine real and existent questions of law. Unfortunately, this is not the case: ‘...Nothing we can say in this opinion can deter Parliament from proceeding with the Bill or dropping it…’ conceded the then Chief Justice Chandrachud in the reference on the Special Courts Bill. He went on to gently express the hope that, ‘Parliament will not fail to take notice of the court’s decision.’ As in the case of laws, in executive action, the government may in fact proceed with acting as it desires, regardless of the advisory opinions that the Supreme Court issues, though the court may subsequently intervene and restrain such acts if they are challenged.

In the case of judicial precedent as well, it appears that the Supreme Court is not always bound by its own opinions in reference, although it has on occasions restated, utilized and endorsed the reasoning in its advisory opinions. In the reference on the Special Courts Bill, Justice Chandrachud suggested that advisory opinions ‘ought’ to be binding, noting, ‘It would be strange that a decision given by this court on a question of law in a dispute between two private parties should be binding on all courts in this country but the advisory opinion should bind no one at all even if as in the instant case, it is given after issuing notice to all interested parties, after hearing every-one concerned who desired to be heard, and after a full consideration of the questions raised in the reference.’

Justice Chandrachud’s suggestion was tempered in that he took note that H.M. Seervai, the noted scholar on constitutional law, held the view that advisory opinions ‘served no judicial function’ and should not be treated as binding law. Unfortunately, this carefully modulated suggestion has been characterized in the opinion on the 2G reference as categorical endorsement that advisory opinions are binding. Quoting a part of Justice Chandrachud’s suggestion, Justice D.K. Jain for the majority also finds that in the course of a reference, the Supreme Court may not only issue a binding opinion but also ‘overrule a previous view delivered by it’, opening the gateway to all sorts of mischief by a discontented executive.


An evaluation of the way the Indian Supreme Court narrates advisory opinions demonstrates three things. First, for the most part it is recognized that the advisory opinions have no binding authority, but are in the form of guidance. Second, it is apparent that the court expects, and even assumes that the legislature and the executive will take note of the advisory opinions and act accordingly. Third, great care is consistently taken to ensure that a variety of public and private opinions are heard before advisory opinions are delivered.

The danger with treating presidential references as binding law on the nation lies in excessive reliance on their mimicry of fair procedure. Article 145(4) requires that opinions are pronounced in open court before they are delivered to the President. The Supreme Court has usually issued public notice, or at the least, notice to all states and union territories, inviting views before issuing the opinion. Nevertheless, an advisory opinion is distinct – and markedly so – from decisions in adjudication.


First, in an advisory opinion, the Supreme Court is bound to accept the facts placed before it and may not question them. Accordingly, the opinion that it delivers is on the basis of facts in the reference made by the executive alone, and this provides the first, inevitable bias. Second, a reference is made on an abstract point. It is not a lis between parties, who contest against each other, with rights and interests at stake. Justice Zafarullah in 1944 condemned references because the facts were ‘enveloped in a thick fog of hypotheses and uncertainties’ and any opinion thereon would only rest upon a ‘forest of assumptions, which must rob it of all value.’

If we are to accept the argument that an advisory opinion is binding upon all courts in the country, then each opinion would be rightfully open to challenges on multiple grounds, not least that it anticipates facts, operates on presumptions and determines the rights of parties without consideration of the specific facts and circumstances of their individual instances. Inevitably, this also raises absurd practical implications. Can one file for execution of an advisory opinion? If the government fails to comply with an advisory opinion, does it amount to contempt of court? If not, then is an opinion binding on the judiciary as law, but on no one else?

The last and third reason that renders a presidential reference distinct is that it is neither open to appeal nor review. In consequence, a binding advisory opinion, as Justice D.K. Jain would have it, contravenes the Supreme Court’s own established notions of justice and procedure. ‘Courts cannot declare a limitation or constitutional requirement under the notion of having discovered some ideal norm... The strength of constitutional adjudication’, as he himself wrote in the 2G reference, ‘lies in case to case adjudication.’ Canada, one of the few other nations to permit its federal court to issue advisory opinions, provided earlier for an appeal to Her Majesty in Council, and advisory opinions by provincial courts are still appealable to the Canadian Supreme Court. This is not the case in India.

This leaves us in an unenviable position of having advisory opinions of the Supreme Court occupy two possible spaces. In the first instance, advisory opinions are nothing but opinions – not binding on the executive, legislature or indeed, the judiciary, and consequently, an additional and unnecessary responsibility for an already overburdened court. In the second, advisory opinions are an indirect means for the executive to obtain authoritative binding orders from the Supreme Court, based on facts put forth by the executive alone, against which other persons have no recourse. Neither appears to be wholly acceptable.



1. The powers of the federal legislature to provide for the levy of an estate duty in respect of property, other than agricultural land, passing upon the death of any person, AIR 1944 FC 73.

2. The Gujarat Gas Act, 2001, (2004) 4 SCC 489.

3. The Kerala Education Bill, 1957, (1959) 1 SCR 995.

4. The Berubari Union and Exchange of Enclaves, (1960) 3 SCR 250.

5. Gujarat Assembly, (2002) 8 SCC 237.

6. Special Reference 1 of 2012.

7. The Special Courts Bill, 1978, (1979) 1 SCC 380.

8. Centre for Public Interest Litigation v. Union of India, (2012) 3 SCC 1.