Concept
The Ban on Advisory Opinions
Case or Controversy and Advisory Opinions
The Ban on Advisory Opinions
Federal courts do not answer legal questions that arise outside a lawsuit. The Supreme Court has called this refusal "the oldest and most consistent thread in the federal law of justiciability."1 An advisory opinion is a ruling that expounds a legal issue in general, without a live dispute behind it. It almost always lacks opposing parties, because the person who asks, whether a private individual or a governmental actor, wants guidance before acting and usually hopes to avoid later lawsuits with the force of the opinion.2
The ban also does work beyond its own terms. Courts often defend their justiciability rulings as necessary to avoid the de facto issuance of advisory opinions, so the rule functions as a fixed-star reference point for the other doctrines that limit federal judicial power.3
Other courts handle the question differently, which shows that the ban is not inevitable. English courts regularly gave advisory opinions when the Constitution was written. The Canadian Supreme Court still issues them from time to time, as do ten state high courts, and they remain an integral part of the legal tradition in continental Europe and in international tribunals such as the International Court of Justice.4 The federal courts have consistently refused, and the reason lies in what Article III makes the judicial power for. Article III extends that power to "cases" and "controversies," and the Court reads those words as the Framers' expectation that federal judges would resolve live, real-world conflicts. On this view, courts state principles of law only while umpiring actual and concrete disputes between parties.5
Marbury v. Madison ties the two ideas together. Marshall justified judicial review by the constitutional responsibility to decide cases or controversies, so the Court's proper role is to give a judicial remedy for individual harm; its power in the abstract is not the point.6 A court that answers a question for someone who has no opposing litigant, and whose rights do not turn on the answer, therefore speaks without the footing that justifies speaking at all.
Four episodes turn that idea into working rules, and the rules are among the case-or-controversy doctrines that state fundamental limits on federal judicial power.7 The 1793 correspondence supplies the reasons for the ban. The ban then yields a test for recognizing an advisory request. Hayburn's Case adds a requirement of finality, and Muskrat shows that Congress cannot create a case by enacting one.
The 1793 Correspondence
The refusal to answer abstract questions has a documented beginning. In Flast v. Cohen the Court explained that the rule against advisory opinions implements the separation of powers and confines federal courts to the role Article III assigns them, and it cited the correspondence between Secretary of State Jefferson and Chief Justice Jay.8 A footnote in the same opinion dates the rule to 1793 and reports that the Court has adhered to it without deviation.9 The Justices' own explanation in that exchange remains the best statement of why the ban exists.
The correspondence grew out of the Washington administration's policy of neutrality between the warring powers of Europe. France demanded American help and cited the treaty of 1778, and the demand raised a host of legal questions about the scope of French rights and American obligations.10 Jefferson submitted twenty-nine questions to the Court on international law, neutrality, and the construction of the French and British treaties.11 The executive branch itself noted that the first question would be whether the Justices could, with propriety, offer advice on abstract questions.12
Jefferson's pitch was practical. The President would be much relieved, he wrote, if he were free to refer such questions to the judges, whose knowledge of the subject would secure the nation against errors dangerous to its peace and whose authority would insure the respect of all parties.13
The Justices declined. The departments of government are in certain respects checks upon each other, they wrote, and the Justices sit as judges of a court of last resort. Those considerations "afford strong arguments against the propriety of our extra-judicially deciding the questions."14 They added that the President's constitutional power to call on the heads of departments for opinions seemed to have been "purposely as well as expressly united to the executive departments."15 The reply bundles three arguments, and each carries its own weight.
The first argument concerns checks. The departments restrain one another, and a Justice who has told the President in advance that a course of action is lawful has joined in planning an act the Court may later have to judge. Advice also leaves the courts exposed when another branch ignores it, because the courts are then in an awkward position if the issue later comes before them in litigation.16
The second argument, that the Court is a court of last resort, has a practical edge. No higher tribunal can correct a mistake, and requests for advice tend to arrive in the heat of the moment, which deprives the Court of the detached perspective that an unfolding lawsuit often affords. Requests also have an abstract or hypothetical quality, so they may not focus the issues as concretely as the adverse presentations of disputing parties do.17
The third argument is textual. The Constitution gives the President power to call on the heads of the executive departments for opinions, and the Justices read that provision as placing the advice-giving function in the executive branch.18 The President was not left without legal counsel; the Constitution had supplied it elsewhere.
The letter did not, in explicit terms, set out an absolute prohibition, yet the Court has long understood it as doing so.19 The Court's account of the ban's source also reaches beyond the letter. Because English courts could give advisory opinions when the Constitution was drafted, history alone cannot explain the American rule, and the Court rests the ban on policies implicit in Article III.20 Two policies are in play. The rule implements the separation of powers, and it avoids questions that reach the Court without "that clear concreteness provided when a question emerges precisely framed and necessary for decision from a clash of adversary argument exploring every aspect of a multifaced situation embracing conflicting and demanding interests."21
The Framers' choices point the same way. The Constitutional Convention four times rejected the Virginia Plan's proposal to associate Justices in a council of revision that would advise on pending legislation, and it likewise rejected suggestions that the President or either House be able to request advisory opinions from the Supreme Court, as an opinion collecting the Convention records recounts.22 The Framers never accepted a proposal authorizing the President and the House and Senate to obtain advisory opinions from the Supreme Court.23
The Justices' own early experience shows what the abstract setting costs. In 1790 Chief Justice Jay and some of the other Justices wrote to President Washington that they thought the requirement of riding circuit unconstitutional. Washington was apparently unmoved, and the practice continued. When it was challenged in litigation over a decade later, the Court upheld circuit riding on historical precedent and did not mention the earlier correspondence.24 Once the Court faced a precise case in concrete and real litigation, it decided contrary to its earlier informal opinion.25
What changes if a treaty question like Jefferson's reaches the courts through an ordinary lawsuit?
Yes. The legal question has not moved, but its setting has. The merchant has a concrete injury, the government is an adverse party defending its own act, and the judgment will decide who holds the vessel. The court answers the treaty question as a by-product of resolving that dispute, which is the way Article III contemplates. Neither the requester's rank nor the gravity of the subject supplied what was missing in 1793. The requester was the President acting through his Secretary of State, and Jefferson tied the questions to the peace of the United States, yet the Justices declined.26
Requests Without Adverse Parties
The Justices' reasoning in 1793 never depended on who was asking, so it can be restated as a two-part test. The Court's standard description of a controversy, taken from Aetna Life Insurance Co. v. Haworth, is "a real and substantial controversy admitting of specific relief through a decree of a conclusive character, as distinguished from an opinion advising what the law would be upon a hypothetical state of facts."28 The description has two halves. The first asks whether the parties are adverse and dispute their legal relations. The second asks whether the court's answer would bind them or would amount to a nonbinding recital of the court's views.29 A request that fails both halves, like the questions put to the Justices in 1793, is advisory in the plainest sense. Both halves belong to the case-or-controversy doctrines that state fundamental limits on federal judicial power.30
The test examines the structure of a request, so the requester's rank and the weight of the question drop out. Federal courts may not reach out to resolve even vital constitutional questions at the behest of Congress and the President.31 Congress may not ask the Supreme Court to comment on the constitutionality of proposed legislation, and the President may not ask it for an opinion on the legality of pending executive action. Neither request involves adverse parties, and neither promises more than a nonbinding recital of the Court's views.32 A private person fares no differently. Someone who wants guidance about a quarrel that has not yet happened must turn to a lawyer or another adviser, because courts exist only to resolve actual disputes.33
The judge can do nothing with it. No adverse party is before the court, and no decree could issue. The landowner has asked for exactly what the Justices declined to give the President.
The two halves of the test protect different things. Adversity supplies concreteness: litigants with real stakes frame the question precisely and press every side of a multifaced situation, while a request that arises from no clash leaves the issue abstract or hypothetical.34 The Court has said that a real stake "preserves the vitality of the adversarial process by assuring both that the parties before the court have an actual, as opposed to professed, stake in the outcome, and that the legal questions presented ... will be resolved, not in the rarified atmosphere of a debating society, but in a concrete factual context conducive to a realistic appreciation of the consequences of judicial action."35 The second half addresses a separate defect. An opinion that settles nothing turns the court into a consultant.
A realistic request shows both halves failing and then succeeding once the posture changes.
The Court may not answer the President's request. It arises from no dispute between adverse parties, and it seeks no binding or conclusive relief, so any reply would be an advisory opinion.36 The same question reaches a court by a different road once the order issues. The state and the executive branch now dispute their legal relations, and the relief ordered will bind them, so the court's ruling is no longer merely advisory.37 The statutes and the President's doubts are what they were before. What has changed is the posture: an injured party, an adverse officer, and a judgment that decides who must do what.
The pattern can also run the other way. If a state refuses to comply with the order, a federal officer may sue it for a declaration that the order is lawful. A declaration issued in an actual controversy is not advisory, provided the relief applies to the parties before the court.38 A later section takes up declaratory judgments in detail.
Statutory authorization does not displace this check. In Buckley v. Valeo, Congress had intended its provision to give judicial review to the extent Article III permits. The Court nevertheless began by asking whether the case presented a case or controversy, because Congress may not require the Court to render opinions in matters that are not cases or controversies. It then found that at least some of the appellants had a sufficient "personal stake" in the validity of each challenged provision.39 The authorization was the starting point, and the Court still looked for a litigant with a real stake.
Hayburn's Case and Finality
A court exercises judicial power only when its decision is the last word of the judicial branch, and the founding-era source of that limit is Hayburn's Case. A 1792 statute let disabled Revolutionary War veterans apply for pensions in the circuit courts. Those courts were to decide whether the applicant was a disabled veteran and certify a pension amount to the Secretary of War, who was to place the applicant on the pension list unless he had cause to suspect imposition or mistake.41 The claim then went on to Congress, which decided whether the pension would be paid.42 The courts' finding was only the first step in a process that executive officers and the legislature could end differently.
Five Justices of the Supreme Court, sitting on circuit, refused to act under the statute.43 The circuit courts uniformly gave two grounds: the business was not properly judicial in nature, and the Secretary of War's power to suspend the decision deprived the proceeding of the finality that a court's judgment requires.44 The grounds are connected. Work that another branch may undo is a recommendation, and a recommendation is not the business a court exists to transact.
The rule makes sense once you ask what a judgment is. A judgment declares the parties' legal rights and ends their dispute unless a higher court intervenes. When an executive officer may disregard what the court found, the officer, and not the court, decides whether the veteran receives a pension, and the court's finding becomes one input into another branch's decision. The constitutional guarantees that protect federal judges may bear on the point, since they may insulate the judges' decision-making from personal retaliation, though whether they forbid further action after the judicial process has ended is debatable. Whatever their scope, the federal courts scrupulously protect the principle of Hayburn's Case.45 Finality also completes the logic of the previous subsection. A decision that another branch may revise settles nothing, so it resembles the advice the Justices declined to give in 1793, and finality is one of the fundamental limits on federal judicial power that the case-or-controversy doctrines state.46
The principle remains in force. Judicial finality forbids review by the political branches, even though superior courts in the Article III hierarchy may review a federal court's decision.47 Plaut v. Spendthrift Farm, Inc. applied the same logic to legislative interference. The Court struck down a statute that directed federal courts to reopen final judgments, because a final judgment is the last word of the judicial department in that case and Congress may not rescind it.48
Two clarifications keep the rule from overreaching. First, review by a superior Article III court never offends it. The courts of appeals and the Supreme Court belong to the same judicial department as the trial courts, so a trial court's decision is not that department's last word until the appeals have run their course.49 An appellate court must therefore apply Congress's latest enactment, even a clearly retroactive one, to a judgment still on appeal, and it may overturn the lower court's result in doing so.50 The bar on reopening applies only after a judgment has become final. Congress may change the law that governs a pending case, but it may not command that a decided case be tried again.
Second, finality differs from efficacy. Finality asks whether another branch may revise what the court decided. Efficacy asks whether the decision gets carried out, and that depends on the other branches. Congress can frustrate a money judgment against the United States by refusing to appropriate payment, and any other federal judgment is ineffective if the executive fails to enforce it.51 In those situations the court's decision stands unrevised, which may be why one well-known commentator called finality the least important element of the case-or-controversy requirement. The Secretary of War in Hayburn's Case held a different power, the power to disagree with the court's finding about who deserved a pension.
A problem built on this rule typically dresses the Hayburn statute in new subject matter. The question to put to it is whether anyone outside the judiciary can revise the court's decision.
The judge may decline for the reason the circuit judges gave in 1792. The Secretary's veto makes each award a recommendation, so no court's decision would be the last word on a farmer's entitlement.52 If a right of appeal to the court of appeals replaced the veto, the objection would vanish, because review would then stay inside the Article III hierarchy, and the court of appeals, not an officer of the executive branch, would have the last word.53
A related limit concerns which duties Congress may assign to judges. As a general rule, executive or administrative duties of a nonjudicial nature may not be imposed on judges holding office under Article III.54 The rule has exceptions. In Morrison v. Olson the Court found no Article III objection to a special court's power to appoint an independent counsel, because that power comes from the Appointments Clause, a source of authority for judicial action independent of Article III.55 In Mistretta v. United States the Court likewise recognized significant exceptions and approved some nonadjudicatory activities by the Judicial Branch, describing a "twilight area" in which the activities of the separate Branches merge.56
Muskrat and Congressional Test Cases
Congress cannot create a case or controversy simply by enacting one, and Muskrat v. United States establishes the point.58 Flast v. Cohen cited Muskrat when it explained that, where the judicial power is invoked to pass on actions of the Legislative and Executive Branches, the rule against advisory opinions implements the separation of powers.59 Mistretta cited it for the point that the judicial power is limited to cases and controversies, and it added that refusing advisory opinions protects the Judiciary's independence by avoiding entanglements with the political branches and by keeping courts out of areas reserved to them.60 The 1793 episode involved a request from the President. Muskrat involved one from Congress, written into a statute.
Congress had enacted statutes that extended restrictions on the alienation, encumbrance, and leasing of Cherokee allotments and enlarged the group entitled to share in the final distribution of lands and funds. A special act then authorized the plaintiffs to sue in the Court of Claims to determine the validity of those statutes, with a right of appeal to the Supreme Court. The United States was the named defendant, and if the plaintiffs won, the statute authorized the government to pay their attorneys' fees.61 A defendant that pays the winning plaintiffs' fees has little reason to resist them.
The Supreme Court reversed the Court of Claims and directed it to dismiss for want of jurisdiction.62 The government purported to be the defendant but had no interest in the outcome, and the plaintiffs did not assert a property right against it or demand compensation for anything it had done. The special act was merely an attempt to obtain a Supreme Court test of an act of Congress without regard to the case-or-controversy requirement, and the attempt was unconstitutional.63 The Court relied on two points: that the United States had no interest adverse to the plaintiffs, which one commentator calls debatable, and that a judgment which could not legally be executed amounted to no more than an expression of opinion on the validity of the statutes.64
The special act misunderstood why courts ever hold statutes invalid. Marshall's justification for judicial review rested on the responsibility to decide cases or controversies; the Court exists to give a judicial remedy for individual harm from constitutional violations, and its power in the abstract is not the point.65 Judicial review is therefore a by-product of deciding a dispute. A statute that directs the Court to rule only on another statute's validity reverses that order and makes the by-product the product. The Framers never accepted a proposal authorizing the President and Congress to obtain advisory opinions from the Supreme Court, and the special act sought the same result by another route.66
Each defect in the special act matters for its own reason. A nominal defendant that funds the plaintiffs cannot press the other side, and a court understands a constitutional question best when someone with a stake does. A decree that fixes no one's rights is an opinion, which is what the Justices declined to give in 1793. The act also confined the Court to the constitutional question alone. In a real dispute the Court might never reach that question, since it could decide that the plaintiffs were not really Cherokee Indians or rest on another, narrower ground. An act that makes the constitutional issue the whole business of the case removes the exits that ordinarily tie a constitutional ruling to the facts.67
Congress's endorsement therefore carries no weight on the Article III question, because the case-or-controversy doctrines state fundamental limits on federal judicial power.68 Congress may open the federal courthouse as widely as Article III permits, as the statute in Buckley intended, but the Court still asks whether a litigant with a real stake faces an adverse party.69 Congress cannot make the United States a defendant in federal court unless the United States truly has an interest adverse to the plaintiff.70 An answer that treats a statute naming the United States as defendant and promising Supreme Court review as the end of the analysis has skipped the real question, which is whether anyone on the other side has something to lose.
The special act in Muskrat drew a negative answer at every step. The plaintiffs claimed no right against the government and sought no compensation for its acts, the government was a defendant in name that would pay the plaintiffs' fees if they won, and the judgment could not be executed in any legal sense.71 A statute that fails any one of the questions asks for an opinion, however carefully Congress words the grant of jurisdiction.
The same constitutional question could have reached the Court through genuine disputes, and one was already underway. The enrollment provisions were not self-executing, because federal officers had to add names to the list. Indians had already sued in a lower federal court to enjoin the Secretaries of the Interior and Treasury from enlarging the enrollments, a suit to which Muskrat itself referred. The next year that case reached the Supreme Court, which affirmed the lower court's decision upholding the acts.72 Jurisdiction lay in that case because it was an actual controversy: parties injured by the action of government officials sought to enjoin those officials from carrying out statutorily mandated duties on the ground that the statute was unconstitutional.
A private suit could have raised the restriction on alienation just as well.
It does. The allottee and the buyer are adverse over title to the land, and the judgment will bind them. Deciding who owns the land requires the court to rule on the restriction's validity, so the constitutional question that Muskrat could not reach arrives as part of a genuine case.
The lesson of Muskrat is narrow. The defect lay in the statute's design, and a more carefully written statute could have created the appropriate test case. What Congress cannot do is grant jurisdiction to give what is really an advisory opinion, even though the Court might reach the same constitutional issue if an appropriate case presents it.73
Two side points keep the holding in proportion. Coercive relief is no longer essential to a case or controversy: if an actual controversy exists, a court may issue only declaratory relief, and the constitutionality of declaratory judgment statutes is now accepted. A federal court may even issue prospective rulings, so long as the relief applies at least to the parties before it, so that the opinion is not merely advisory.74 The defect in Muskrat was therefore the missing controversy, not the absence of an injunction. The Court of Claims was also an Article I court at the time of Muskrat, which Article III does not bind, so it could give advisory opinions and, until it became an Article III court, routinely did so for Congress under its reference jurisdiction.75 The Supreme Court, hearing the case on appeal, was an Article III court and could not. A later section returns to courts outside Article III.