Concept
Structural Limits and Klein
Congressional power to define and limit
Structural Limits and Klein
The preceding sections asked a definitional question about the Exceptions Clause: how large an exception may Congress make? Article III gives the Supreme Court appellate jurisdiction over the listed cases outside its original categories “with such Exceptions, and under such Regulations as the Congress shall make.”1 The plenary and mandatory readings of that text disagree about whether Congress may repeal one statutory route to the Court, withdraw a whole subject, or leave a constitutional claim without any federal forum. Both readings, though, ask only whether a statute falls within the power the clause confers. This section asks what comes after that answer. A statute can be an exception in the clause’s sense and still be invalid because of what Congress uses it to do.
Any exercise of constitutional power faces three limits. It must come within the defined scope of the power, which is why the power to make exceptions to appellate jurisdiction cannot reach the Court’s original jurisdiction. It must not violate limits inherent in the Constitution’s structure, particularly the separation of powers. And it must not transgress external guarantees such as those in the Bill of Rights.2 This section concerns the second limit; the third comes later in the chapter. The structural limit has one clear application: Congress cannot use its exceptions power to vacate a Supreme Court decision, because doing so would usurp a judicial prerogative.3
Jurisdiction is where the structural limit operates because the Constitution shares the judicial power. The President and the Senate select federal judges, Congress may remove them by impeachment, and Congress may narrow the jurisdiction of the lower federal courts and the Supreme Court’s appellate jurisdiction.4 Congress therefore has a legitimate role in deciding which courts hear which cases. It has no role in deciding the cases themselves, and the difficulty lies in telling the two apart.
Ex parte McCardle shows the legitimate side. After the Court had acknowledged jurisdiction over McCardle’s appeal, and after the case had been argued, Congress withdrew the statutory right of appeal, seeking to avoid a Supreme Court determination that the Reconstruction legislation was unconstitutional. The Court complied and dismissed for want of jurisdiction.5 United States v. Klein shows the other side. Three questions organize what follows. What did Klein hold, and why? Where does the line run between changing the law for pending cases and directing their outcome? And does Klein protect only individual cases, or also the Court’s wider role as a check on the other branches?
The Klein Principle
In United States v. Klein, decided in 1871, the Court held that Congress had “inadvertently passed the limit which separates the legislative from the judicial power.”6 The statute had the form of an exception to appellate jurisdiction, so it cleared the definitional test, and the Court objected to what the exception accomplished. That is the general rule at work: Congress may not use its exceptions power in a manner that violates constitutional limitations on government power.7
The dispute began with an 1863 statute allowing recovery of property captured or abandoned during the Civil War by a claimant who could prove he had not aided the rebellion. Klein sued in the Court of Claims and relied on an earlier Supreme Court decision treating a presidential pardon as conclusive proof of that loyalty. He won, and the government appealed. While the appeal was pending, Congress, perhaps emboldened by its victory in McCardle, passed a statute to reverse the result.8
The new proviso worked in three steps. It declared that the Supreme Court would not consider pardons on appeal. It made a pardon that recited participation in the rebellion conclusive evidence of that participation, if the claimant accepted it without disclaimer. And it directed that whenever a Court of Claims judgment rested on a pardon without other proof of loyalty, the Supreme Court would lose jurisdiction and dismiss the case.9 Taken together, the steps stripped the Court of power to decide exactly the cases in which its own settled law required a ruling against the government.10
The Court read the proviso as a rule of decision wearing a jurisdictional label. The denial of jurisdiction was, it said, “founded solely on the application of a rule of decision, in causes pending, prescribed by Congress.” The Court had jurisdiction up to a point, but once it found a pardon in the record, its jurisdiction was to cease and it was to dismiss.11 Jurisdiction therefore turned on the merits. The Court would lose power only when a pardon, which its own decisions treated as equivalent to proof of loyalty, would have carried the judgment for the claimant, and it was directed to give the pardon “an effect precisely contrary” to the one its own judgment assigned. The Court asked what the proviso was “but to prescribe a rule for the decision of a cause in a particular way.”12 It then asked whether Congress could require the Court to deny itself jurisdiction “because and only because its decision, in accordance with settled law, must be adverse to the government and favorable to the suitor,” and answered that the question “seems to us to answer itself.”13
The limit follows from what the exceptions power is for. It is power to decide which cases a court may hear, and Congress must exercise it consistently with the independence of the judiciary. A jurisdictional limit therefore must be neutral: Congress may not decide the merits of a case under the guise of limiting jurisdiction.14 The government was a party in Klein, and a rule that switched off the Court’s power exactly when the government would lose let one party decide the controversy in its own favor, the very possibility the Court raised when it asked whether it could comply “without allowing one party to the controversy to decide it in its own favor.”15 Neutrality is also what protected the repeal in McCardle. The denial of appeal applied to the Government and the private party alike, so a Government loss below would have left it, like McCardle, without an appeal, and that symmetry helps assure that a withdrawal is not an attempt to alter the results of specific cases.16 The test looks to what the statute does, meaning whom it helps and what it requires the court to find.
The proviso had a second flaw, narrower and less portable. The Court said the rule was open to objection as “impairing the effect of a pardon,” and so infringing the constitutional power of the Executive.17 Klein thus rests on two separation-of-powers grounds: the statute attempted both to dictate the result in pending cases and to limit the effect of a pardon.18 Only the first travels well, because few jurisdictional limits touch the pardon power. An answer about a jurisdiction-stripping bill should develop the first ground and give the second a sentence.
McCardle supplies the contrast. Klein’s proviso reached the Court of Claims as well as the Supreme Court, leaving a pardoned claimant no avenue of relief at all, and it operated only against claimants whose pardons would otherwise have won.19 The McCardle repeal removed one statutory route to the Court, and shortly afterward the Court held that it could still issue original writs of habeas corpus and certiorari and thus review a case like McCardle’s.20 Whether the availability of that alternate route was essential to McCardle’s holding remains unclear, and one should not read too much into the decision.21 The distinction between the two statutes does not depend on how that question is answered. Both were enacted while a case was pending, and both reflected a wish to avoid an unwelcome result. The difference lies in operation: the repeal told the Court nothing about how to decide any case, while Klein’s proviso ended the Court’s authority precisely where settled law pointed toward the claimant.
The safe statement of the principle is the narrow one. Klein does not bar every jurisdictional withdrawal prompted by a wish to avoid an unwelcome ruling; Congress withdrew the appeal in McCardle for exactly that purpose, and the Court gave the repeal effect.23 Nor does Klein stop Congress from changing the law that governs pending cases. The Court distinguished an earlier decision in which the court was “left to apply its ordinary rules to the new circumstances created by the act,” because in Klein “no new circumstances have been created by legislation.”24 The next subsection develops that line. The limit also follows Congress into the lower courts: Congress may not manipulate inferior-court jurisdiction in a manner designed to achieve a particular result in a pending case.25
Step two decides most fact patterns. A statute that triggers only when one side would win reveals itself there, and a statute that applies to every case of a kind passes. Steps three and four supply further evidence of a rule of decision and a second ground of invalidity. A genuinely neutral withdrawal still faces the external limits taken up later in this chapter.
Changing Law Versus Directing Outcomes
A careless reading of Klein would bar Congress from legislating anything that touches a case already in court. The opinion rules that reading out. The Court contrasted an earlier decision, Pennsylvania v. Wheeling Bridge Co., in which Congress legalized a bridge the Court had held to be a nuisance. There the Court was “left to apply its ordinary rules to the new circumstances created by the act,” while in Klein “no new circumstances have been created by legislation.”28 Two lines follow. Congress may change the law that applies to pending cases, but it may not tell a court what to find or conclude under existing law, and it may not reopen a judgment that is already final.
The reason lies in what courts do when they decide. On appeal, a court ordinarily applies the law that exists at the time of appeal, even if that law did not exist at trial, a rule the Supreme Court has applied since the time of Chief Justice Marshall.29 A litigant whose claim rests on a statute holds a cause of action that lasts only while the statute stays in force, not a vested property right.30 A legislature that amends the statute changes what the court applies; it does not tell the court how to apply it. The judicial power sits elsewhere. The federal judiciary has the power “not merely to rule on cases, but to decide them,”31 and a legislature that dictated the finding, or reopened a finished judgment, would be taking the deciding for itself.
Robertson v. Seattle Audubon Society shows a legitimate change. An environmental group had sued over timber harvesting in federally managed forests, and while the suit was pending Congress changed the harvesting standards and provided that compliance with the new standards would satisfy the applicable requirements. The Court sustained the change.32 The analysis turns on how the statute is characterized. A statute that instructs the trial court to make specific findings of fact or conclusions of law in a pending case invades the province of the judiciary. A statute that merely changes the legal standards under which the case is judged is an exercise of lawmaking authority, and it stays one even when the new standards lead to a different result.33 The timber statute ordered no finding. It replaced the yardstick, so the court still had a legal question to answer.
Bank Markazi v. Peterson pushed the idea further. Victims of terrorist acts sponsored by Iran had won judgments against that country. In 2012 Congress enacted a provision specifying that assets held by a New York bank on behalf of Iran’s central bank, Bank Markazi, could be used for post-judgment execution to satisfy those judgments. The Court, per Justice Ginsburg, held 6–2 that Congress may enact new legislation that changes the outcome of a pending civil case without violating the separation of powers.34 Together, Robertson and Bank Markazi show that a statute does not become a Klein violation just because it affects pending litigation or changes who wins. The question is whether Congress supplied new law or dictated a conclusion under old law.
Klein sits on the other side of that line because the proviso supplied no new standard. It forbade the Court to give a pardon the effect its own judgment assigned and directed an effect “precisely contrary.”35 Wheeling Bridge, Robertson, and Bank Markazi each left the court a legal question to resolve under a changed rule. Klein left the Court nothing to resolve once a pardon appeared in the record.
The second limit concerns finality. Plaut v. Spendthrift Farm, Inc. involved a statute that changed a limitations period and required courts to reinstate suits they had already dismissed as untimely under the earlier rule. The Court held that Congress may alter a limitations period for all pending and future cases but may not require a federal court to reopen a final judgment, because the judicial power includes the authority to issue dispositive judgments in pending cases. “By retroactively commanding the federal courts to reopen final judgments, Congress has violated this fundamental principle.”36 Plaut thus separates two moves Congress might make with a limitations period. Applying a changed period to cases still pending is new law of the kind Robertson upheld, and commanding courts to reopen final judgments is invalid.
Finality matters because a judgment that Congress could reopen would be only a recommendation. A losing party could seek relief from the legislature instead of appealing, and Congress would sit as a court of review. The federal judiciary will not have its decision frustrated by allowing the executive or legislative branches to revise, suspend, modify, or otherwise review it.37 Federal decisions may be reviewed by superior courts within the Article III hierarchy, but judicial finality forbids review by the political branches.38 McCardle fits the same line. Congress may repeal appellate jurisdiction after a case has been briefed and argued but before it has been decided, but it may not reopen and redetermine cases the courts have fully and finally resolved.39
One qualification matters. Finality applies with reduced force to injunctions, which remain under the court’s continuing supervision and may be modified when circumstances change, including a change in the underlying law, so a decree yields to the extent it conflicts with the new law.40 Wheeling Bridge is the old example, and the Court relied on the same supervisory role in upholding parts of the Prison Litigation Reform Act in Miller v. French.41 A final money judgment is the contrast. Once final, it is not affected by a later federal law.42
The second step does the work. The common error is to stop at the observation that the statute affects a pending case or will change its outcome. Neither fact decides the question, because Robertson and Bank Markazi both involved legislation that did exactly that and survived. What decides it is what the statute left for the court to decide.
The Essential Function Reading
The Klein principle described above polices the exceptions power case by case, asking whether Congress is using an exception to prescribe how a pending case must come out. A broader reading of Klein shifts the inquiry from the individual case to the Court’s place in the constitutional system. On that reading, the Supreme Court performs an essential function: it exercises judicial review as a check on the federal political branches and on the states. Congress, the argument runs, may not use its exceptions power to undermine the Court’s ability to perform that function, or to destroy its essential role in the constitutional plan.46 The clause itself, which grants the exceptions power, is the textual starting point for the debate.47
The narrow and broad readings locate the defect in different places. Klein itself involved Congress’s usurpation of one judicial function, the resolution of a pending case, and the narrow reading stops there. The broad reading finds a larger principle behind the case and locates the harm in the loss of the check. That harm can occur even when no statute dictates the outcome of any case, because a jurisdictional statute can leave every result open and still remove the one institution able to enforce the Constitution against the legislature that wrote it.
The broad reading challenges orthodoxy. Orthodox doctrine holds that Congress has complete control over the jurisdiction of the inferior federal courts and essentially unrestricted power over the Supreme Court’s appellate jurisdiction, and a long list of distinguished observers have aligned themselves with that view.48 The Court itself has not recognized the broader restriction in express terms. Its rulings on lower-court jurisdiction seem to recognize a broad congressional power, though the case law is sparse and dated, and McCardle seems at first glance to support an all-but-plenary power over the Court’s appellate jurisdiction.49 The stakes are also disputed. Some predict that state courts, freed from federal oversight, would reject the Court’s precedent and develop their own constitutional doctrine, giving proponents of stripping a change in constitutional law by ordinary legislation. Others reply that the precedents remain on the books and state courts might keep applying them.50
Three arguments support the broad reading. First, judicial review is the mechanism that makes constitutional limits binding. If Congress could withdraw jurisdiction over exactly the claims that test those limits, it would define the scope of its own powers without any check from the federal judiciary.51 Second, the Convention clearly expected the federal judiciary, with the Supreme Court as the focus, to maintain the uniformity and supremacy of federal law. That expectation suggests that Congress can make only minor procedural adjustments, like the jurisdictional-amount requirement.52 Third, jurisdiction stripping leaves constitutional rights to state judges who almost universally serve limited terms and face election, creates tension with Marbury’s endorsement of the federal courts’ power to say what the law is, and invites a cacophony of differing lower-court interpretations of key provisions.53
Supporters frame the limit as internal to Article III, in contrast to the external limits from the Bill of Rights taken up later. Henry Hart argued that the exceptions power may not be exercised in a way that interferes with the Court’s “essential” or “core” functions or destroys its essential role in the constitutional plan.54 A related version ties the limit to the Court’s supremacy over inferior bodies. Congress may restrict the Court’s as-of-right appellate jurisdiction, but it arguably may not foreclose the supervisory review needed to preserve the inferiority of lower courts and tribunals.55
The reading’s force varies with the statute. It is strongest where Congress insulates its own conduct. A statute stripping the Supreme Court and all lower federal courts of jurisdiction over cases involving the commerce power would permit Congress to define the scope of its own powers without any check from the federal judiciary, a serious separation-of-powers problem that directly undermines constitutional structure.56 It is weakest where one route to the Court closes and another stays open. The argument is that the Court’s essential role is not destroyed when, as in 1868, one means of approach is taken away and another is left open, since original habeas writs remained available after McCardle, and that depriving litigants altogether of access to the Court in cases involving the supremacy of federal law would be invalid.57 On that reasoning, a statute that closed the Court to every Establishment Clause challenge to school prayer, with no other route, would draw the objection.
None of this is settled law. The Court has not recognized the restriction in express terms, and the reading has little or no direct authority.58 To the extent a measure only interferes with the Court’s essential function of checking the federal political branches and the states, the outcome is much less certain, and the composition of the Court that considers the question will be of paramount importance.59 The ambiguity also tempers both branches. If it were clear that Congress could limit the Court’s appellate jurisdiction at will, the Court’s authority would be severely undermined; if it were equally clear that the Court could trump every exception, the hubris of unrestrained judicial review might damage the balance with republican democracy. The practical result has been a history of very few conflicts in this sphere.60 An exam answer should present the essential function reading as an argument drawn from the larger principle in Klein, apply it to the facts, and say plainly that it is unresolved.