OUR JUDICIAL FEDERALISM*

Justice Sandra Day O'Connor**

IN PREPARING for today's talk I knew you might like to hear how I set about deciding cases, just what really goes on inside the Court, and how I feel personally about my colleagues on the bench, and so on. But one of the qualities desirable for a Justice is to be judicious and, therefore, those subjects are best avoided. It is not even advisable to speak about most of the interesting issues of the day, since, as Alexis de Tocqueville noted, almost every issue in American life is likely, sooner or later, to end up before the courts.1 When a Justice expresses an opinion on war or peace, on religion or politics, or even on science or literature, she always risks having her words embarrassingly quoted back at her in a brief or oral argument. As a result, we usually follow the advice of Calvin Coolidge:

"If you don't say anything, you won't be called on to repeat it." 2

It is clear, however, that it is appropriate for me to say some thing today, and that what I say should somehow be suitable for the Canary Lecture. The Sumner Canary Lectureship honors a man who served as a leader of the Bar, as a United States Attorney, and a judge of the Ohio Court of Appeals.3 The Canary Lecture series is affiliated with the Case Western Reserve Law School, an institution which, as you know, originated as a department of Western Reserve and ultimately flourished as an integral part of the federated Case Western Reserve University.4

* The text of this Article, with minor changes, was presented as the Third Sumner Canary Lecture, Case Western Reserve University School of Law, Cleveland, Ohio, November 13, 1984.

** Associate Justice, United States Supreme Court.

1. A. DE TOCQUEVILLE, DEMOCRACY IN AMERICA 248 (G. Lawrence trans. 1966).

2. J. BARTLETT, FAMILIAR QUOTATIONS 736 (15th ed. 1980).

3. See A. Austin, Sumner Canary Lectureship (1983) (available at the Case Western Reserve Law Review).

I can think of no topic more fit to honor this man and this institution than federalism. Like me, Judge Canary served on a state appellate court. I suspect that his experience in that office inspired a concern for the interaction between state and nation in the courts. As a judge on the Arizona Court of Appeals, I found that the tensions inherent in our "indestructible union of indestructible states"5 were a subject of frequent concern.

And surely federalism is an appropriate topic at a university whose very name is a witness to its federal origins. A few short decades ago, Case and Western Reserve were separate universities, and Western Reserve was itself segmented into the separate faculties of Mather, Adelbert, and Cleveland Colleges.6 The federation of this university, like the federation of our nation, took years of planning. It required compromises to assure that the strengths and diversity of its constituent parts would not be submerged in a monolithic, homogenous whole. And like this nation, the federation of this university had its skeptics and opponents. One can hear the echo of the antifederalists of 1787 in the cheers of the students of the Case Institute at their football game against Western Reserve in 1965: "Two, four, six, eight, we don't want to federate!"7 Despite early opposition, both the university and the nation have achieved the delicate balance that is the hallmark of a federal system.

I believe any administrator at Case Western could tell us that the balancing inherent in a federal system is never a static one. The relations of the states with the nation, or of the schools with the university, require constant and flexible accommodation of often conflicting interests. Justice Hugo Black, in a memorable opinion for the Court, once described what he saw as the essence of federalism, and he gave that essence a name.8 He called it "our federalism."

The concept [of our federalism, Justice Black said,] does not mean blind deference to "States Rights" anymore than it means centralization of control over every important issue in our National Government and its courts. The Framers rejected both these courses. What the concept does represent is a system in which there is sensitivity to the legitimate interests of both State and National Governments, and in which the National Government, anxious though it may be to vindicate and protect federal rights and federal interests, always endeavors to do so in ways that will not unduly interfere with the legitimate activities of the States.9

4. See c. CRAMER, THE LAW SCHOOL AT CASE WESTERN RESERVE UNIVERSITY(1977).

5. E.g., United States v. Bekins, 304 U.S. 27, 53 (1937).

6. See c. CRAMER, CASE WESTERN RESERVE (1976).

7. Id. at 282.

8. Younger v. Harris, 401 U.S. 37 (1971).

It would take far too long to explore all the aspects of what Justice Black called our federalism. Instead, I will confine myself to what Justice Black might have called "our judicial federalism." It has been my privilege to witness the accommodation of state and nation from both sides in the interaction of the state and the federal courts. This afternoon, I wish to highlight a few of the currently prominent features of the ever-evolving law that defines the federal state court relationship. I hope to describe not only what our judicial federalism looks like today, but also the directions in which it may be or should be moving.

Any realistic picture of judicial federalism must acknowledge the primary role of the state courts in our federal system of government. The vast bulk of all civil and criminal litigation in this country is handled in the state courts. In 1982, more than thirteen million civil suits and twelve million criminal actions were filed in the fifty state court systems and the District of Columbia.10 By comparison, only 234,139 civil and 38,449 criminal actions were filed in the federal courts that same year.11

Equally important to a fair portrait of the federal system is an acknowledgment of the role played in each judicial system by law from the other system. State law, for instance, plays a significant role in federal court actions-largely through diversity jurisdiction and through the doctrine of pendent jurisdiction. But in my re marks today, I wish to focus on the pervasiveness of federal law both statutory and constitutional-in state courts. To take what is probably the most important example, state courts day in and day out apply federal constitutional law-most notably, in the multitude of state criminal prosecutions. State trial and appellate judges must be and are fully conversant with case law interpreting such provisions as the fifth amendment privilege against self-incrimination, the sixth amendment right to counsel, and the double jeopardy clause. More to the point, state judges must often wade into the fine

9. Id. at 44.

10. U.S. BUREAU OF THE CENSUS, STATISTICAL ABSTRACT OF THE UNITED STATES:1982, at 191 (114th ed. 1983) (excluding juvenile and traffic charges).

11. Id. at 188-89; REPORT OF THE PROCEEDINGS OF THE JUDICIAL CONFERENCE OF THE UNITED STATES: ANNUAL REPORT OF THE DIRECTOR OF THE ADMINISTRATIVE OF FICE OF THE UNITED STATES COURTS 190, 196, 211, 272 (1982).

points-some would say the excruciatingly fine points-of our fourth amendment search and seizure jurisprudence. Thus, federal law is in fact developed and interpreted by all fifty state court systems as well as by the federal court system.

These basic facts about our judicial federalism indicate the need for some means to assure a consistent and uniform body of federal law among the state and federal courts. The goal of national uniformity rests on a fundamental principle: that a single sovereign's laws should be applied equally to all-a principle expressed by the phrase, "Equal Justice Under Law," inscribed over the great doors to the United States Supreme Court. Justice Holmes recognized that uniformity of federal law also lies at the heart of what binds us together as a nation. Speaking of the power of judicial review of the Supreme Court of the United States, he said: "I do not think the United States would come to an end if we lost our power to declare an act of Congress void. I do think the Union would be imperiled if we could not make that declaration as to the laws of the several states." 12

In our dual system of courts, review of state court decisions on federal law by the Supreme Court of the United States is the principal means we have of encouraging the needed uniformity. The Supreme Court recognized this as early as 1816, when it stated, in Martin v. Hunter's Lessee,13 that its review of state court decisions is demanded by the "necessity of uniformity of decisions throughout the whole United States upon all subjects within the purview of the constitution."14

Of course, the sheer volume of state court decisions on federal questions permits the Supreme Court to review only a relative handful of cases from state courts. That fact guarantees state courts a large measure of autonomy in the application of federal law. At the same time, the inherent limits on Supreme Court resources make it especially important, first, that the Supreme Court give understandable guidance on constitutional questions, and second, that state courts conscientiously follow the constructions of federal law adopted by the Supreme Court. In this way our several courts are dependent on each other for the successful functioning of our judicial federalism. Our founding fathers joined our state and federal court systems in a marriage for better or for worse, a marriage requiring each partner to have appropriate respect and regard for the other.

12. O. W. Holmes, Law and the Court, in COLLECTED LEGAL PAPERS 295-96 (1921).

13. 14 U.S. 304 (1816).

14. Id. at 347-48.

It is no wonder, then, that one of the Supreme Court's most important functions-and perhaps the most important function-is to oversee the systemwide elaboration of federal law, with an eye toward creating and preserving uniformity of interpretation. It is precisely because of the importance of this unifying function that the jurisdiction of the Supreme Court of the United States has been made ever more discretionary over the years. Today, this function is uppermost in the minds of the Justices in exercising the discretion to take cases for review. I breach no confidence in saying that the most commonly enunciated reason for granting review in a case is the need to resolve conflicts among other courts over the interpretation of federal law.

The chief problem encountered by the Supreme Court in exercising its power to review state court judgments is the problem of deciding when a federal question is presented for review. When a state court has decided a case on both federal and state law grounds, the Supreme Court's jurisdiction is limited to reviewing only the federal law grounds.15 A state court's view on issues of state law is, of course, binding on the federal courts.16 When the outcome of a state court case could not be changed, even if the federal law issue were resolved differently by the United States Supreme Court, the Supreme Court must decline review. The state court judgment in such a case rests on an adequate and independent state ground.17 The roots of this important jurisdictional limitation are found in the nation's longstanding recognition of the importance of preserving the vitality of both components of our dual federal-state judicial system.

I will give several illustrations of how the strength of both the federal and state courts is promoted by the doctrine forbidding Supreme Court review of state court judgments supported by an adequate and independent state ground. The point I wish to stress is that, especially in the constitutional context, state courts have substantial power to grant or withhold jurisdiction to the Supreme Court by the choice and articulation of the grounds for the state court decisions.

Generally speaking, if a state court decides that a particular

15. See, e.g., Leathe v. Thomas, 207 U.S. 93, 98 (1907).

16. See, e.g., American Ry. Express v. Kentucky, 273 U.S. 269, 272 (1927). 17. See, e.g., Murdock v. Memphis, 87 U.S. 590, 635 (1874).

state action violates both federal and state law, the final state court judgment is not reviewable by the Supreme Court.18 For example, if a state court finds that a challenged state action or law violates a state constitution's due process clause and, independently, the fourteenth amendment's due process clause, the United States Supreme Court cannot review the decision. This is true, of course, because even if the state court is wrong about the federal question, the Supreme Court cannot change the state law holding, and the final state court judgment would stand regardless of what the Supreme Court were to decide on the federal question. A decision by the Supreme Court on the federal issue would be merely advisory.

If, in contrast, a state court upholds a particular state action on the ground that it offends neither federal nor state law, then the Supreme Court has power to review the state court judgment though on the federal question only. Thus, a state court's rejection of both the federal and state due process challenges may be re viewed by the Supreme Court. 19 The decision on the federal grounds in these circumstances could change the result in the case, since the Supreme Court's disagreement with the state court on the federal question would render the challenged state action unlawful.

These first two situations present few difficulties. A third situation is also straightforward. If a state court holds that a particular state action violates state law because it violates a parallel provision of federal law, then the Supreme Court has power to review the case.20 This would be the case, for example, if a state court were to decide that the due process clause in its state constitution automatically means whatever the fourteenth amendment's due process clause is held to mean. Simply put, the state law decision in such a case is not independent of federal law.

So much for the easy questions. Most of the difficult problems concerning the adequate-and-independent-state-ground doctrine arise when a state court's opinion in a case does not make clear whether the decision is based on federal or state law. Over the years, the Supreme Court has adopted various approaches to the problem of determining whether a decision rests on an adequate and independent state ground. None of the methods has been entirely satisfactory either to the state courts or to the Supreme Court itself.

18. See, e.g., Henry v. Mississippi, 379 U.S. 443 (1965).

19. Ohio v. Johnson, 104 S. Ct. 2536 (1984).

20. Zacchini v. Scripps Howard Broadcasting Co., 433 U.S. 562 (1977); United Air Lines v. Mahin, 410 U.S. 623 (1973).

The Court has recently revisited this problem and adopted a new approach.

One approach the Court has followed when the basis for a state decision is ambiguous is simply to refuse to review it. Thus, the court has on occasion simply dismissed an appeal or declined to accept a petition for certiorari.21 Such a course of action is obviously deferential to state courts, though it requires some sacrifice of uniformity in federal law.

A second approach the Court has taken is to make its own in dependent determination whether the state court judgment rests on an adequate and independent state ground.22 Thus, the Supreme Court has at times attempted to decide itself both whether the state court actually relied on state law and what the state law is. The disadvantages to this approach are manifold. The state decision may be so obtuse, and state law may be so unclear, that no sure answer is possible. This approach is even less attractive because it requires the Supreme Court to interpret state law. When you imagine nine Supreme Court Justices trying to decide a fine point of Hawaiian land law, or Louisiana civil procedure, I think you can understand why this is a task the Court is reluctant to undertake.

A third approach taken by the Court when faced with an ambiguous state court judgment is to ask the state court to resolve the confusion. Thus, the Court has several times returned a case to state court for clarification of the ground of decision.23 This procedure avoids the problems of having a federal court interpret matters of state law. But it is perceived by some to be coercive and intimidating to state courts. It is the judicial equivalent of a law review's sending an article back to its author for a second draft. And state judges may understandably react in the same way an author does, quite unfavorably.24

None of the three options I have mentioned has proved to be wholly satisfactory. Recently, therefore, in the case of Michigan v. Long, 25

21. See, e.g., Lynch v. New York, 293 U.S. 52 (1934).

22. See Texas v. Brown, 103 S. Ct. 1535 (1983); cf. South Dakota v. Neville, 103 S. Ct. 916 (1983) (state supreme court's finding that coerced blood alcohol tests violated right against self-incrimination reversed despite state constitutional provision similar to the fifth amendment).

23. See, e.g., California v. Krivda, 409 U.S. 33 (1972); Minnesota v. National Tea Co., 309 U.S. 551 (1940) (cases vacated and remanded); Herb v. Pitcairn, 324 U.S. 117 (1945) (case continued).

24. See, e.g., State v. Jackson, 672 P.2d 255, 260, 261 (Sheehy & Shea, JJ., dissenting), on remand from Montana v. Jackson, 103 S. Ct. 1418 (1983).

the Supreme Court adopted a new approach to resolving ambiguity about the existence of an adequate and independent state ground. In that case, we held that when a state court decision fairly appears to rest primarily on federal law or on grounds interwoven with federal law, and the adequacy and independence of the possible state law ground is not clear from the face of the opinion, the Supreme Court will assume that the decision was based on a federal ground.26 We reasoned that, since the existence of an adequate and independent state ground is, by hypothesis, not clear, the Supreme Court's review in such cases is not advisory.

I believe that the Michigan v. Long rule both preserves state court autonomy and ensures Supreme Court oversight in the interests of uniformity. The rule promotes uniformity by enabling the Supreme Court to review state court decisions that may be read as based on federal law. Because in such situations state and federal grounds are not clearly distinguished, such decisions may have precedential force on the federal law issue and, at the same time, may inhibit the appropriate state legislative bodies from freely considering whether to change the state law rule that may have been adopted by the state court.

The Michigan v. Long rule also promotes state court autonomy, by virtue of the simple fact that state courts retain complete control over whether the rule will be applied and whether the case can be reviewed. The assumption of a federal ground of decision can easily be avoided by a state court's clear and express articulation of its separate reliance on bona fide adequate and independent state grounds. In these ways, I believe, the Michigan v. Long approach responds to both the uniformity and state-autonomy policies that inform so much of the law governing the relation of federal and state courts.

Notwithstanding the advantages of the rule, I must add a word of caution about Michigan v. Long. The Court did not adopt this rule unanimously.27 It will take time to see whether the principle Michigan v. Long adopts will serve our judicial federalism well. It is my prediction and my hope that it will.

The effort of the Supreme Court to protect our judicial federalism

25. 103 S. Ct. 3469 (1983).

26. Id. at 3476.

27. Chief Justice Burger, Justices White, Powell, and Rehnquist agreed with Justice O'Connor's articulation of the rule. Id. Justice Blackmun concurred in the result but dissented in regard to the rule. Id. at 3483. Justices Brennan and Marshall dissented, id., as did Justice Stevens, id. at 3489.

has proceeded during the past decade on several other fronts as well. I wish to focus briefly on two of the more prominent issues. Both concern not Supreme Court review of state court decisions, but rather the other principal means by which state court proceedings are reviewed in federal court-through suits in the lower federal courts. The two areas-abstention doctrine and federal habeas corpus-are both grounded in respect by the federal courts for state court proceedings.

Respect for the integrity of state court proceedings was the explicit basis for Justice Black's important opinion in Younger v. Harris in 1971.28 In that case, the Supreme Court recognized the "longstanding public policy against federal court interference with state court proceedings,"29 and held that a federal court may not enjoin a state court criminal proceeding. Accordingly, a federal court must "abstain" from adjudicating any case brought to enjoin a state prosecution. This abstention principle has been recognized in recent years as extending beyond criminal proceedings to other contexts, such as bar disciplinary proceedings, 30 that implicate important state interests. The Younger abstention doctrine thus generally "counsel[s] federal courts to abstain from jurisdiction whenever federal claims have been or could be presented in ongoing state judicial proceedings that concern important state interests."31

I need hardly point out that this doctrine accords broad protection to pending state proceedings. Under the Younger abstention doctrine, as under the Michigan v. Long rule, state courts exercise substantial control over the extent of federal court review. Just as a state court can avoid Supreme Court review by clearly articulating a state ground of decision, state courts can generally ensure federal court abstention under the Younger doctrine by providing an opportunity for the parties before them to raise their federal claims. The Younger doctrine thus illustrates a pervasive theme in our judicial federalism: state court respect for federal law is inextricably linked to federal court respect for state court proceedings.

The same theme is evident in the law of federal habeas corpus review of state criminal convictions. Review by federal courts of state criminal cases is a point of particularly sensitive federal-state interaction. The federal habeas law reflects the strong presumptions

28. 401 U.S. 37 (1971).

29. Id. at 43.

30. See, e.g., Middlesex County Ethics Comm. v. Garden State Bar Ass'n, 457 U.S. 432 (1982); Anonymous v. Association of the Bar, 515 F.2d 427 (2d Cir. 1975).

31. Hawaii Housing Auth. v. Midkipp, 104 S. Ct. 2321, 2329 (1984).

that state court criminal judgments are final and that state court proceedings are fully adequate to resolve federal claims. No better statement of the Court's policy in this regard can be found than in our 1976 decision in Stone v. Powell:32 "[W]e are unwilling, [the Court said] to assume that there now exists a general lack of appropriate sensitivity to constitutional rights in the trial and appellate courts of the several states. State courts, like federal courts, have a constitutional obligation to safeguard personal liberties and to up hold federal law." 33

There are at least four important ways in which federal habeas corpus law pays respect to state courts. First, federal courts by statute must give deference to state court findings of fact relevant to federal claims. Congress has provided that in federal habeas corpus proceedings, a state court's findings of fact are presumed to be correct-unless, broadly speaking, the state court proceedings that resulted in those findings were in some way procedurally inadequate for a fair decision on the federal claim.34 This deference to state findings of fact means that most habeas issues will be decided by the state rather than the federal courts, for, as Justice Jackson once ob served in another context, "most contentions of law are won or lost on the facts." 35 Second, deference to state court determinations on habeas issues is embodied in the statutory requirement that a state prisoner exhaust his state remedies before making application to a federal court for a writ of habeas corpus.36 This exhaustion requirement rests on the congressional judgment that it is appropriate that state courts have the first opportunity to address challenges to their own proceedings.37

Third, the waiver rules in federal habeas law also rest on federal respect for state court proceedings. If a state prisoner has commit ted a procedural default with respect to a particular federal claim that is, if he has failed to comply with a state procedural rule requiring him to raise the claim at a specified time, on penalty of forfeiting the claim for state court purposes-he is generally precluded from raising the claim in federal court. This is because the state procedural

32. 428 U.S. 465 (1976).

33. Id. at 494 n.35.

34. 28 U.S.C. § 2254(D) (1982).

35. Jackson, Advocacy Before the Supreme Court: Suggestions for Effective Case Presentations, 37 A.B.A. J. 801, 803 (1951).

36. 28 U.S.C. § 2254(B) (1982).

37. See Freiser v. Rodriguez, 411 U.S. 475 (1973) (federal habeas corpus is rooted in considerations of federal-state comity that require that the convicting state be given the first opportunity to correct its errors).1984)

rule constitutes an adequate and independent state ground for the state courts' rejection of the claim. The prisoner can overcome the bar to federal habeas consideration of his claim only by demonstrating cause for and prejudice from his waiver. The basis for these restrictive waiver rules, once again, is federal respect for "the state's interest in the integrity of its rules and proceedings and the finality of its judgments." 38

The last aspect of federal habeas law I wish to mention is the holding of Stone v. Powell.39 The Supreme Court held there that lower federal courts cannot entertain a state prisoner's habeas petition that alleges a fourth amendment violation if the state court criminal proceeding provided a full and fair opportunity to litigate the issue.40 If there has been such an opportunity, only discretionary review by the United States Supreme Court remains available to a convicted state defendant alleging that evidence obtained in violation of the fourth amendment was unlawfully admitted. This principle has recently been extended to preclude a state criminal defendant from bringing a civil damages action against the police who seized evidence if the legality of the search and seizure has previously been resolved against the defendant in the criminal case.41 Because of the rarity of Supreme Court certiorari review, these rules, of course, accord substantial finality to state court resolutions of fourth amendment claims.

Each of the doctrines I have discussed-the Michigan v. Long rule, the Younger abstention principle, and the rules of federal habeas-are designed to preserve the vitality and autonomy of the state court component of our judicial federalism. I think it is clear that the Supreme Court of the United States has been increasingly sensitive to the role of state courts within the federal system. This recognition of the role of state courts, in my view, necessarily places a reciprocal burden and responsibility on state court judges to deal with federal issues in a thorough and receptive manner. Hearings on federal issues in criminal cases must be conducted with great care and with knowledge of the applicable principles. Adequate findings must be made and clearly articulated. This kind of careful attention by the state courts to their role in deciding questions of

38. Reid v. Ross, 104 S. Ct. 2901, 2907 0984) (state's interests are undermined if federal courts can freely ignore the procedural forfeitures of a state's action).

39. 428 U.S. 465 0976).

40. Id. at 482.

41. See Allen v. McCurry, 449 U.S. 90, IOI 0980).

federal law is precisely what enables state courts to exercise the substantial degree of control they have over our dual judicial system.

There are today as in the past skeptics and critics of various aspects of federalism as we experience it in this country. When in 1815 the liberator of much of Latin America, Simon Bolivar, was choosing a system of government for the nations he had helped to create, he was skeptical of federalism: "Among the popular and representative systems of government [Bolivar said,] I do not approve of the federal system: it is too perfect; and it requires virtues and political talents much superior to our own."42

In my remarks today I do not embrace Bolivar's notions that federalism is perfect, or that it requires virtues and talents beyond human capacity. Our judicial federalism is, by its very nature, a flexible and dynamic accommodation of the sometimes conflicting interests of the state and federal courts. Judicial federalism can never be perfect as long as our country continues as a sovereign union of equally sovereign and vital states. But despite the inevitable imperfections and conflicts in federal and state court relation ships in a dynamic federal union, I sincerely believe that we have the ability and virtues necessary to make federalism work. I take heart in this from the vitality of this law school within its federated university. Our judicial federalism can and will work. But the marriage between our state and federal courts, like any other marriage, requires each partner to respect the other, to make a special effort to get along together, and to recognize the proper sphere of the other partner.

42. 1 S. BOLIVAR, SELECTED WRITINGS OF BOLIVAR 118 (H. Bierck ed. 1951).