"[T]he legislative, executive, and judicial powers, of every well constructed government," said john marshall in osborn v. bank of the united states (1824), "are co-extensive with each other; … [t]he executive department may constitutionally execute every law which the Legislature may constitutionally make, and the judicial department may receive from the legislature the power of construing every such law." The articles of confederation fell far short of this model. Not only was there no federal executive with authority to enforce congressional measures against individuals, but, apart from a cumbersome procedure for resolving interstate disputes, Congress was authorized to establish courts only for the trial of crimes committed at sea and for the determination of "appeals in all cases of captures." The remedy for these shortcomings was one of the major accomplishments of the Constitution adopted in 1789. As Article II gave the country a President with the obligation to "take care that the Laws be faithfully executed," Article III provided for a system of federal courts that more than satisfied Marshall's conditions for a "well constructed government."
Article III consists of three brief sections. The first describes the tribunals that are to exercise federal judicial power and prescribes the tenure and compensation of their judges. The second lists the types of disputes that may be entrusted to federal courts, specifies which of these matters are to be determined by the supreme court in the first instance, and guarantees trial by jury in criminal cases. The third defines and limits the crime of treason.
"The judicial Power of the United States," Article III declares, "shall be vested in one Supreme Court, and in such inferior Courts as the Congress may from time to time ordain and establish." The text itself indicates that the Supreme Court was the only tribunal the Constitution required to be established, and the debates of the constitutional convention demonstrate that the latter words embodied a deliberate compromise.
In fact, however, Congress created additional courts at the very beginning, in the judiciary act of 1789. Since 1911 the basic system has consisted of the united states district courts—at least one in every state—in which most cases are first tried; a number of regional appellate courts now called the united states courts of appeals; and the Supreme Court itself, which functions largely as a court of last resort. From time to time, moreover, Congress has created specialized courts with jurisdiction to determine controversies involving relatively limited subjects. All this lies well within Congress's broad discretion under Article III to determine what lower courts to create and how to allocate judicial business among them. Specialization at the highest level, however, seems precluded; Congress can no more divide the powers of "one Supreme Court" among two or more bodies than abolish it altogether.
"The Judges, both of the supreme and inferior Courts," section 1 continues, "shall hold their Offices during good behaviour and shall, at stated Times, receive for their Services, a Compensation, which shall not be diminished during their Continuance in Office." Under the second section of Article II the judges have always been appointed by the President subject to Senate confirmation; under the fourth section of that article they may be removed from office on impeachment and conviction of "Treason, Bribery, or other high Crimes and Misdemeanors." The central purpose of the tenure and salary provisions, as alexander hamilton explained in the federalist #78, was to assure judicial independence.
The Supreme Court has repeatedly enforced the tenure and salary provisions. In ex parte milligan (1867), for example, the Court held even the Civil War no excuse for submitting civilians to military trials in states where the civil courts were open, and in O'Donoghue v. United States (1933), it held that the Great Depression did not justify reducing judicial salaries.
On a number of occasions, however, the Court has permitted matters within the judicial power to be determined by legislative courts whose judges do not possess tenure and salary guarantees. State courts may decide Article III cases, as the Framers of the Constitution clearly contemplated; the tenure and salary provisions do not apply to the territories or to the district of columbia, where there is no separation of powers requirement; Article III did not abolish the traditional courtmartial for military offenses; federal magistrates may make initial decisions in Article III cases provided they are subject to unlimited reexamination by tenured judges.
Early in the twentieth century the Supreme Court appeared to give judicial blessing to the numerous quasi-judicial bodies that have grown up since the creation of the Interstate Commerce Commission in 1887, although scholars have debated heatedly whether there is any satisfactory way to distinguish them from the nontenured trial courts plainly forbidden by Article III. That these developments did not mean the effective end of the tenure and salary requirements, however, was made clear in 1982, when the Court in northern pipe line construction co. v. marathon pipe line co. invalidated a statute empowering judges with temporary commissions to exercise virtually the entire jurisdiction of the district courts in bankruptcy cases. Where to draw this line promises to be a continuing problem.
The power to be vested in federal courts is the "judicial power," and the various categories of matters that fall within this power are all described as cases or controversies—"Cases," for example, "arising under this Constitution," and "Controversies to which the United States shall be a Party." From the beginning the Supreme Court has taken this language as a limitation: federal courts may not resolve anything but "cases" and "controversies," and those terms embrace only judicial functions.
Thus, for example, when President george washington asked the Justices for legal advice respecting the United States' neutrality during hostilities between England and France, they declined to act "extra-judicially"; and when Congress directed them to advise the war secretary concerning veterans' pensions, five Justices sitting on circuit refused, saying the authority conferred was "not of a judicial nature" (hayburn ' scase, 1792). Washington's request for advice did not begin to resemble the ordinary lawsuit, but later decisions have invoked the "case" or "controversy" limitation to exclude federal court consideration of matters far less remote from the normal judicial function. The essential requirement, the Court has emphasized, is a live and actual dispute between adversary parties with a real stake in the outcome.
One dimension of this principle is the doctrine of ripeness or prematurity: the courts are not to give advice on the mere possibility that it might be of use in the future. Occasionally the Court has appeared to require a person to violate a law in order to test its constitutionality—causing one commentator to remark that "the only way to determine whether the subject is a mushroom or a toadstool, is to eat it." The declaratory judgment act, passed to mitigate this hardship, has generally been applied to allow preenforcement challenges when the intentions of the parties are sufficiently firm, and it has been held consistent with the "Case" or "Controversy" requirement.
At the opposite end of the spectrum is the mootness doctrine, which ordinarily forbids litigation after death or other changed circumstances deprive the issue of any further impact on the parties. A series of debatable decisions essentially dating from Moore v. Ogilvie (1969), however, has relaxed the mootness doctrine especially in class actions, so as to permit persons with no remaining interest to continue litigating issues deemed "capable of repetition, yet evading review."
The "case or controversy" requirement has also been held to forbid the decision of collusive suits, and to preclude the courts from exercising the discretion of an administrator, as by reviewing de novo the decision to grant a broadcasting license. The most important remaining element of that requirement, however, is the constitutional dimension of the doctrine of standing to sue.
While standing has been aptly characterized as one of the most confused areas of federal law, its constitutional component was simply stated in Warth v. Seldin (1975): "[t]he Article III power exists only to redress or otherwise to protect against injury to the complaining party." Injury in this context is hardly self-defining, but it plainly requires something more than intellectual or emotional "interest in a problem." This principle puts under a serious cloud the periodic congressional attempts to authorize "any person" to obtain judicial relief against violations of environmental or other laws. On the other hand, other aspects of the standing doctrine are not of constitutional dimension and thus do not preclude Congress from conferring standing on anyone injured by governmental action.
One of the principal points of contention of the law of standing has been the right of federal taxpayers to challenge the constitutionality of federal spending programs. When a taxpayer attacked expenditures for maternal health on the ground that they exceeded the powers granted Congress by Article I, the Court in frothingham v. mellon (1923) found no standing: "the taxpayer's interest in the moneys of the treasury … is shared with millions of others, is comparatively minute and indeterminable, and the effect upon future taxation, of any payment out of the funds, so remote, fluctuating, and uncertain, that no basis is afforded for an appeal to the preventive powers of a court of equity."
Although the apparent reference to equitable discretion made it uncertain that the Court was saying taxpayer suits were not "cases or controversies" within Article III, the remainder of the passage suggests that the taxpayer could not show the constitutionally required injury because it was uncertain that a victory would mean reduced taxes. Nevertheless, in flast v. cohen (1968) the Court allowed a federal taxpayer to challenge expenditures for church-related education as an establishment of religion in violation of the first amendment. Unlike the taxpayer in Frothingham, who "was attempting to assert the States' interest in their legislative prerogatives," the plaintiff in Flast asserted "a federal taxpayer's interest in being free of taxing and spending in contravention of specific constitutional limitations," for one purpose of the establishment clause was to prevent taxation for religious ends. Whether the distinction was of constitutional scope the Court did not say; interestingly, the taxpayer opinions have tended to avoid entirely the traditional constitutional inquiry into the existence of an injury that will be redressed if the plaintiff's claim prevails.
Underlying the constitutional "case or controversy" limitation are a variety of policy concerns. The first group relates to reducing the risk of erroneous decisions. Concrete facts enable judges to understand the practical impact of their holdings; adverse parties help to assure that arguments on both sides will be considered; as argued by felix frankfurter, "the advisory opinion deprives constitutional interpretation of the judgment of the legislature upon facts." A second group of reasons focuses upon strengthening the Court's institutional position. Lawmaking by appointed judges is least difficult to reconcile with democratic principles when it is the inevitable by-product of the stock business of judging; the courts should not squander their power of moral suasion or multiply conflicts with other branches by deciding unnecessary legal questions. Third, and of considerable importance, is a concern for the separation of powers. The courts are not to exercise a general superintendence over the activities of the other branches.
The costs of the "case or controversy" limitation include the delay, uncertainty, and disruption incident to determining the constitutionality of legislation only in the course of subsequent litigation, and the danger that some legislative and executive actions may escape judicial review entirely. Whether the latter is cause for concern has much to do with one's perception of the function and importance of judicial review itself; it seems reasonable to expect that perception to influence the definition of a "case" or "controversy."
In addition to restricting federal courts to the decision of "cases" and "controversies" of a judicial nature, section 2 of Article III enumerates those categories of "cases" and "controversies" to which the "judicial Power shall extend." As the former limitation serves the interests of separating federal powers, the latter serves those of federalism. In accord with the spirit of the tenth amendment the Supreme Court has held that Congress may not give the federal courts jurisdiction over disputes of types not listed in Article III. John Marshall set the tone in cutting down to constitutional size a statute providing for jurisdiction over cases involving aliens in hodgson v. bowerbank in 1809: "Turn to the article of the constitution of the United States, for the statutes cannot extend the jurisdiction beyond the limits of the constitution."
Article III's provision that federal judicial power "shall extend to" certain classes of cases and controversies has generally been taken to mean that it shall embrace nothing else. From the text alone one might think it even more plain that federal courts must be given jurisdiction over all the matters listed, for section 1 commands that the federal judicial power "shall be vested" in federal courts. Indeed, Justice joseph story suggested just such an interpretation in martin v. hunter ' s lessee in 1816. This conclusion, however, was unnecessary to the decision, contrary to the understanding of the First Congress, and inconsistent with both earlier and later decisions of the Supreme Court.
Article III, in other words, has been read to mean only that Congress may confer jurisdiction over the enumerated cases, not that it must do so. This arguably unnatural construction has been defended by reference to the limited list of controversies over which the Supreme Court has original jurisdiction, the explicit congressional power to make exceptions to the Supreme Court's appellate authority, and the compromise at the Constitutional Convention permitting Congress not to establish inferior courts at all.
This is not to say, however, that Congress has unfettered authority to deny the courts jurisdiction, for all powers of Congress are subject to limitations found elsewhere in the Constitution. A statute depriving the courts of authority to determine cases filed by members of a particular racial group, for instance, would be of highly doubtful vitality under the modern interpretation of the Fifth Amendment due process clause, and one part of Marshall's reasoning in marbury v. madison (1803) supports an argument that closing all federal and state courts to free-speech claims would defeat the substantive right itself. Proposals to remove entire categories of constitutional litigation from the ken of one or more federal courts often follow controversial judicial decisions. Out of respect for the tradition of checks and balances, however, such bills are seldom enacted; we have so far been spared the constitutional trauma of determining the extent to which they may validly be adopted.
The cases and controversies within federal judicial power fall into two categories: those in which jurisdiction is based upon the nature of the dispute and those in which it is based upon the identity of the parties. In the first category are three kinds of disputes: those "arising under this Constitution, the Laws of the United States, and Treaties made, or which shall be made, under their Authority"; those "of admiralty and maritime Jurisdiction"; and those involving competing land claims "under Grants of different States." The provision last quoted is of minor importance; the second formed the staple business of the district courts throughout their early history; the first fulfills Marshall's condition for a "well constructed government" and is by any measure the most critical ingredient of federal jurisdiction today.
The provision for jurisdiction in cases arising under the Constitution and other federal laws has two essential purposes: to promote uniformity in the interpretation of federal law, and to assure the vindication of federal rights. The First Congress sought to accomplish the second of these goals by providing, in section 25 of the 1789 Judiciary Act, for Supreme Court review of state-court decisions denying federal rights; the additional uniformity attendant upon review of state decisions upholding federal claims was not provided until 1914. In sustaining section 25, the opinion in Martin v. Hunter's Lessee demonstrated the difficulty of achieving Article III's purpose without Supreme Court review of state courts: while plaintiffs might be authorized to file federal claims directly in federal courts and defendants to remove state court actions to federal courts on the basis of federal defenses, it was not easy to see how a state court opposing removal "could … be compelled to relinquish the jurisdiction" without some federal court reviewing the state court decision.
Conversely, although Congress failed to give federal trial courts general jurisdiction of federal question cases until 1875, Marshall made clear as early as 1824, in Osborn v. Bank of the United States, that it had power to do so. Supreme Court review alone was no more an adequate protection for federal rights, Marshall argued, than was exclusive reliance on litigation beginning in federal trial courts. As the latter would leave claimants without remedy against a recalcitrant state court, the former would give a state tribunal the critical power to shape the factual record beyond assurance of federal appellate correction.
The Osborn opinion also settled that jurisdiction of a federal trial court over a case arising under federal law was not defeated by the presence of additional issues dependent upon state law. In a companion case, indeed, the Court upheld jurisdiction over a suit by the national bank on notes whose validity and interpretation were understood to depend in substantial part upon nonfederal law: it was enough that the plaintiff derived its existence and its right to contract from the act of Congress incorporating it. The courts have not followed this broad approach, however, in determining whether federal question jurisdiction lies under general statutory provisions; when the federal ingredient of a claim is remote from the actual controversy, as in a dispute over ownership of land whose title is remotely derived from a federal land grant, the district courts lack statutory jurisdiction.
In the contract dispute discussed in Osborn, federal and state law were bound together in the resolution of a single claim; in such a case, as henry hart and Herbert Wechsler said, "a federal trial court would … be unable to function as a court at all" if its jurisdiction did not extend to state as well as federal matters. In the interest of "judicial economy," however, as the Supreme Court put it in United Mine Workers v. Gibbs (1966), jurisdiction over a case arising under federal law embraces not only a plain-tiff's federal claim but also any claims under state law based on the same facts. This so-called pendent jurisdiction doctrine, however, is inapplicable when the Supreme Court reviews a state court decision. With one exception, in such a case the Court may review only federal and not state questions, as the Court held in Murdock v. Memphis (1875); for to reverse a state court in the interpretation of its own law would be a major incursion into state prerogatives not required by the purposes for which Supreme Court review was provided.
A corollary of the Murdock principle is that a state court decision respecting state law often precludes the Supreme Court from reviewing even federal questions in the same case. If a state court concludes, for example, that a state law offends both federal and state constitutions, the Supreme Court cannot reverse the state law holding; thus, however it may decide the federal issue, it cannot alter the outcome of the case. This independent and adequate state ground for the state court decision means there is no longer a live case or controversy between the parties over the federal question. In light of this relation between state and federal issues, Martin itself announced the sole exception to the Murdock rule: when the state court has interpreted state law in such a way as to frustrate the federal right itself—as by holding that a contract allegedly impaired in violation of the contract clause never existed—a complete absence of power to review the state question would mean the Court's authority to protect federal rights "may be evaded at pleasure."
"The most bigoted idolizers of state authority," wrote Alexander Hamilton in the federalist #80, "have not thus far shown a disposition to deny the National Judiciary the cognizance of maritime causes"; for such cases "so generally depend upon the law of nations, and so commonly affect the rights of foreigners, that they fall within the considerations which are relative to the public peace." Jurisdiction over what Article III refers to as "Cases of admiralty, and maritime Jurisdiction" has been vested by statute in the district courts since 1789. Today federal admiralty jurisdiction extends, as the Court stated in another context in The Daniel Ball (1871), to all waters forming part of "a continued highway over which commerce is or may be carried on with other states or foreign countries."
Not everything occurring on navigable waters, however, is a proper subject of admiralty jurisdiction; in denying jurisdiction of claims arising out of an airplane crash in Lake Erie, the Supreme Court made clear that the case must "bear a significant relationship to traditional maritime activity … involving navigation and commerce on navigable waters." Conversely, the relation of an activity to maritime concerns may bring it within admiralty cognizance even if it occurs on land. Marine insurance contracts, for example, are within the jurisdiction although both made and to be performed on land. Similarly, the Court has acquiesced in Congress's provision for jurisdiction over land damage caused by vessels on navigable waters.
Because an additional purpose of federal judicial power over maritime cases is understood to have been to provide a uniform law to govern the shipping industry, the Supreme Court also held in Southern Pacific Company v. Jensen (1917) that Article III empowers the federal courts to develop a "general maritime law" binding even on state courts, and that Congress may supplement this law with statutes under its authority to adopt laws "necessary and proper" to the powers of the courts. Indeed the Court has held that this aspect of the judicial power, like the legislative authority conferred by the commerce clause of Article I, has an implicit limiting effect upon state law. Not only does state law that contradicts federal law yield under the supremacy clause, but, as the Court said in rejecting the application of a state workers' compensation law to longshoremen in the case last cited, no state law is valid if it "interferes with the proper harmony and uniformity" of the general maritime law "in its international and interstate relations."
The remaining authorization of federal court jurisdiction protects parties whose fortunes the Framers were for various reasons unwilling to leave wholly at the mercy of state courts. Many of these categories involve government litigation: "Controversies to which the United States shall be a Party; … between two or more States; between a State and Citizens of another State, … and between a State, or the Citizens thereof, and foreign States, Citizens or Subjects." A federal forum for the national government itself protects against possible state hostility; federal jurisdiction over interstate conflicts provides not only a neutral forum but also a safeguard against what Hamilton in the federalist #80 called "dissentions and private wars"; that "the union will undoubtedly be answerable to foreign powers, for the conduct of its members," was an additional reason for jurisdiction over disputes involving foreign countries as well as the related jurisdiction over "Cases affecting Ambassadors, other public Ministers and Consuls."
The most interesting issue concerning these provisions has been that of sovereign immunity. In chisholm v. georgia (1793), ignoring the assurances of prominent Framers like James Madison and Alexander Hamilton as well as the common law tradition that the king could not be sued without his consent, the Supreme Court relied largely on the text of Article III to hold that the power over "Controversies … between a State and Citizens of another State" included those in which the state was an unwilling defendant. Obviously, as the Justices pointed out, this was true of the parallel authority over "Controversies … between two or more States," and Justice james wilson added his understanding that the English tradition was a mere formality, since consent to sue was given as a matter of course.
Whether this decision was right or wrong as an original matter, within five years it was repudiated by adoption of the eleventh amendment, which provides that "[t]he Judicial power of the United States shall not be construed to extend to any suit in law or equity, commenced or prosecuted against one of the United States by Citizens or Subjects of any Foreign State." Notably, the amendment does not mention admiralty cases, suits by foreign countries, suits against a state by its own citizens under federal law, or suits against the United States. Nevertheless the Supreme Court, taking the amendment as casting doubt on the reasoning underlying Chisholm, has denied jurisdiction in all of these instances. The best explanation has been that, although not excepted by the amendment, they are outside the power conferred by Article III itself. One state may still sue another, however, and the United States may sue a state. The Court has found such jurisdiction "essential to the peace of the Union" and "inherent in the constitutional plan." Why this is not equally true of a suit by a state against the United States has never been satisfactorily explained.
At least since the 1824 decision in Osborn v. Bank of the United States, however, both the Eleventh Amendment and its related immunities have been construed to allow certain actions against state or federal officers even though the effect of the litigation is the same as if the government itself had been named defendant. The theoretical explanation that the officer cannot be acting for the state when he does what the Constitution forbids is inconsistent with the substantive conclusion, often reached in the same cases, that his action is attributable to the state for purposes of the fourteenth amendment. A more principled explanation is that suits against officers are necessary if the Constitution is to be enforced at all; the response is that those who wrote the amendment could not have intended to allow it to be reduced to a hollow shell.
In any event, the Osborn exception has not been held to embrace all suits against government officers. At one time it was said generally that an officer could be prevented from acting but could not be ordered to take affirmative action such as paying off a government obligation, for if he was not acting for the state he had no authority to reach into its treasury. The simplicity of this distinction was shattered, however, by opinions acknowledging the availability of a writ of mandamus to compel an officer to perform a nondiscretionary duty. The more recent formulation in edelman v. jordan (1974), which essentially distinguishes between prospective and retrospective relief, seems difficult to reconcile with the language of the Constitution, with its apparent purposes, or with the fiction created to support the Osborn rule.
Even when the government is itself a party, it may consent to be sued, and the books are filled with a confusing and incomplete array of statutes allowing suits against the United States. Some judges and scholars have argued that suits against consenting states are inconsistent with the language of the amendment, which declares them outside the judicial power; the Court's persuasive explanation has been that, like venue and personal jurisdiction, immunity is a privilege waivable by the party it protects (Clark v. Barnard, 1883). More debatable was the Court's decision in Parden v. Terminal Railway (1964) that a state had "waived" its immunity by operating a railroad after passage of a federal statute making "every" interstate railway liable for injuries to its employees; in Edelman v. Jordan, retreating from this conclusion, the Court emphasized that "[c]onstructive consent is not a doctrine commonly associated with the surrender of constitutional rights." Still later, however, in fitzpatrick v. bitzer (1976) the Court held that Congress had power to override a state's immunity in legislating to enforce the Fourteenth Amendment, although it has never suggested that that amendment allowed Congress to ignore other constitutional limitations, such as the bill of rights.
The two remaining categories of disputes within federal judicial power are "controversies … between Citizens of different States" and between state citizens and "Citizens or Subjects" of "foreign States." Once again the reasons for federal jurisdiction are generally said to be the avoidance of state-court bias and of interstate or international friction. In contrast not only to the admiralty cases but also to those between states, federal jurisdiction based solely on the diverse citizenship of the parties does not carry with it authority to make substantive law. Absent a federal statute, the Court held in erie railroad v. tompkins (1938), "the law to be applied … is the law of the State." Later cases such as Textile Workers Union v. Lincoln Mills (1957) have qualified the effect though not the principle of this decision by finding in silent statutes implicit authorization to the federal courts to make law. An occasional decision has upheld federal common law, without the pretense of statutory authority, on matters mysteriously found to be "intrinsically federal"; an example was the Court's refusal in Banco Nacional de Cuba v. Sabbatino (1964) to look behind official acts of foreign governments. (See act of state doctrine.)
In early decisions the Supreme Court took a narrow view of what constituted a controversy between citizens of different states for purposes of the statute implementing this provision of Article III. More recently, however, the Court has generously interpreted the power of Congress to confer diversity jurisdiction on the federal courts. And as early as the mid-nineteenth century, recognizing that corporations can be the beneficiaries or victims of state court prejudice without regard to the citizenship of those who compose them, the Court effectively began to treat corporations as citizens by employing the transparent fiction of conclusively presuming that the individuals whose citizenship was determinative were citizens of the state of incorporation.
The best known decision involving the diversity jurisdiction was dred scott v. sandford (1857), in which three Justices took the position that a black American descended from slaves could never be a state citizen for diversity purposes because he could not be a citizen of the United States. Questionable enough at the time, this conclusion was repudiated by the Fourteenth Amendment's provision that all persons born in this country are citizens of the United States "and of the state wherein they reside." Nevertheless the courts have held that only American citizens are "Citizens of … States" within Article III, and conversely that only foreign nationals are "Citizens or Subjects" of "foreign States."
"In all Cases involving Ambassadors, other public Ministers and Consuls, and those in which a state shall be Party," Article III, section 2 provides, "the supreme Court shall have original jurisdiction. In all the other Cases before mentioned, the supreme Court shall have appellate jurisdiction, both as to Law and Fact, with such Exceptions, and under such Regulations as the Congress shall make."
Original jurisdiction is the power to determine a dispute in the first instance; appellate jurisdiction, the power to review a decision already made. Marbury v. Madison (1803) held that Congress had no power to give the Supreme Court original jurisdiction of a case to which neither a diplomat nor a state was a party; a contrary result, Chief Justice Marshall argued, would make the constitutional distribution between original and appellate jurisdiction "mere surplusage." This reasoning is not especially convincing, and the converse is not true; in cohens v. virginia in 1821 Marshall himself conceded that Congress could give the Court appellate jurisdiction over cases for which Article III provided original jurisdiction. Cohens also held that the Supreme Court had original authority not over all Article III cases in which a state happened to be a party but only over those "in which jurisdiction is given, because a state is a party," and thus not over a federal question case between a state and one of its own citizens. Inconsistently, however, the Court allowed the United States to sue a State in the Supreme Court in United States v. Texas (1892).
Marbury 's implicit conclusion that the exceptions clause quoted above does not allow Congress to tamper with the original jurisdiction strongly suggests that the enumeration of original cases is a minimum as well as a maximum, and the Court has described as "extremely doubtful" the proposition that Congress may deprive it of original power over state or diplomat cases; yet the Court has concluded that it has discretion not to entertain cases within its original jurisdiction.
Unlike the original jurisdiction provision, that giving the Court appellate authority in "all the other" Article III cases contains an explicit escape valve: "with such Exceptions … as the Congress shall make." In the federalist #81, Hamilton explained that this clause permitted Congress to limit review of facts decided by juries, but he did not say this was its sole objective. From the beginning Congress has denied the Court jurisdiction over entire classes of controversies within the constitutional reach of appellate power—such as federal criminal cases, most of which were excluded from appellate cognizance for many years even if constitutional issues were presented. The Court itself accepted this particular limitation as early as United States v. More (1805), without questioning its constitutionality. Moreover, when Congress repealed a statute under which a pending case attacking the Reconstruction Act had been filed, the Court in ex parte mccardle (1869) meekly dismissed the case, observing that "the power to make exceptions to the appellate jurisdiction of this court is given by express words."
As the McCardle opinion noted, however, other avenues remained available for taking similar cases to the Supreme Court, and three years later the Court made clear in United States v. Klein (1872) that Congress could not under the guise of limiting jurisdiction effectively dictate the result of a case by directing dismissal if the Court should find for the plaintiff. Respected commentators have contended that the Supreme Court must retain appellate authority over certain constitutional questions, arguing that the exceptions clause cannot have been intended, in Henry Hart's words, to "destroy the essential role of the Supreme Court in the constitutional plan." The persuasiveness of this position depends on one's perceptions of the function of judicial review. (See judicial system.)
In order for the Court in Marbury v. Madison to dismiss an action that it found Congress had authorized, it had first to conclude that it had the right to refuse to obey an unconstitutional act of Congress. Marshall's argument that this power was "essentially attached to a written constitution" is contradicted by much European experience; and his assertion that choosing between the Constitution and a statute was an inescapable aspect of deciding cases begged the question, for the Constitution might have required the courts to accept Congress's determination that a statute was valid. For the same reason one may object to his reliance on Article VI's requirement that judges swear to support the Constitution: one does not offend that oath by enforcing an unconstitutional statute if that is what the Constitution requires.
The supremacy clause of Article VI is no better support; the contrasting reference to "Treaties made, or which shall be made" in the same clause strongly suggests that the phrase "laws … which shall be made in Pursuance of" the Constitution, also invoked by Marshall, was meant to deny supremacy to acts adopted under the Articles of Confederation, not to those that were invalid. Most promising of the provisions brought forward in Marbury was Article III's extension of judicial power to "Cases … arising under this Constitution"; as Marshall said, it could scarcely have been "the intention of those who gave this power, to say that in using it the constitution should not be looked into." Yet even here the case is not airtight. For while Article III provides for jurisdiction in constitutional cases, it is Article VI that prescribes the force to be given the Constitution; and while the latter article plainly gives the Constitution precedence over conflicting state laws, it appears to place federal statutes on a par with the Constitution itself.
Nevertheless the Marbury decision should be regarded as neither a surprise nor a usurpation. Though Marshall did not say so, judicial review had a substantial history before Marbury, and despite occasional scholarly denials it seems clear that most of the Framers expected that the courts would refuse to enforce unconstitutional acts of Congress. Moreover, there is force to Marshall's argument that a denial of this power would effectively undermine the express written limitations on congressional power; the natural reluctance to assume that the Framers meant to leave the fox in charge of the chickens lends credence to the conclusion that judicial review is implicit in the power to decide constitutional cases or in the substantive constitutional limitations themselves.
In fact the Marbury opinion espouses two distinct theories of judicial review that have opposite implications for a number of related issues, some of which have been discussed above. If, as Marshall at one point seemed to suggest, judicial review is only an incidental by-product of the need to resolve pending cases, it is no cause for constitutional concern if Congress eliminates the Supreme Court's jurisdiction over First Amendment cases, or if no one has standing to attack a federal spending program. If, on the other hand, as argued elsewhere in Marbury, judicial review is essential to a plan of constitutional checks and balances, one may take a more restrictive view of Congress's power to make exceptions to the appellate jurisdiction, and perhaps a broader view of what constitutes a case or controversy as well.
Dissenting from the assertion of judicial authority over legislative reapportionment cases in baker v. carr (1962), Justice Felix Frankfurter argued for a broad exception to judicial review of both federal and state actions: even unconstitutional acts could not be set aside if they presented political questions. Some have attempted to trace this notion to Marbury itself, where the Court did say that "[q]uestions in their nature political" were beyond judicial ken. The context suggests, however, that Marshall meant only that the Court would respect actions taken by other branches of government within their legitimate authority, and Louis Henkin has shown that most later decisions using "political question" language can be so explained.
The Court itself, however, spoke in Baker of a general "political question" doctrine preventing decision of the merits when, among other things, there was "a lack of judicially discoverable and manageable standards for resolving" a "political" issue. A number of lower courts relied on such a doctrine in refusing to decide the legality of the vietnam war. While the doctrine as so conceived appears at cross-purposes with the checks-and-balances aspect of Marbury, nothing in that decision bars a finding that a particular constitutional provision either gives absolute discretion to a nonjudicial branch (such as the power to recognize foreign governments) or makes an exception to Article III's grant of the judicial power itself (as, arguably, in the case of impeachment).
In most respects, then, Article III amply satisfies Marshall's conditions for a "well constructed government." Though the governmental immunities associated with the Eleventh Amendment may seem anachronistic today, unsympathetic judicial interpretation has blunted their interference with the enforcement of federal law. Decisions since the 1950s have generally rejected Justice Frankfurter's broad conception of the political question. Thus with rare exceptions the federal judiciary, as Marshall insisted, may be given authority to construe every federal law; and the extension of judicial power to controversies between citizens of different states means that the federal courts may often be given power to apply state law as well. Though increased mobility has led to serious efforts to repeal the statutory basis for the diversity jurisdiction, it served an important function in the past and conceivably may become more important in the future. Moreover, the Framers were farsighted enough to assure federal judges the independence necessary to do their appointed job. The weakest point in the system is the arguable authority of Congress to take away all or a substantial part of the Supreme Court's appellate power in constitutional cases; for such an authority undermines other elements of the system of checks and balances that the Framers so carefully constructed.
David P. Currie
Bickel, Alexander 1962 The Least Dangerous Branch. Pages 111–199. Indianapolis: Bobbs-Merrill.
Borchard, Edwin 1928 Hearings on H.R. 5623 before the Subcommittee of the Senate Committee on the Judiciary. 70th Cong., 1st Sess., pp. 75–76.
Hart, Henry and Wechsler, Herbert 1973 The Federal Courts and the Federal System. Pages 309–418, 833–1103. Mineola, N.Y.: Foundation Press.
Henkin, Louis 1976 Is There a "Political Question" Doctrine? Yale Law Journal 85:597–625.