IV. Court-Created Limitations on Collateral Review.

The purpose of this paper is to explore possible limitations on habeas review by Congress. A discussion of the limitations already created by the Supreme Court is useful both as the legal background against which a statute may be enacted and to illuminate where Congress ought to follow the Court and where it ought to depart from the case law.

A. Pre-Teague limitations.

Each of the limitations on habeas corpus that the Supreme Court has adopted has accommodated the enforcement and correction of injustice purposes and rejected the federal forum and perfection rationales.

In Stone v. Powell,⁽¹²⁵⁾ the Supreme Court removed from habeas corpus claims based on the exclusionary rule of the Fourth Amendment. Stone recognizes that the exclusionary rule has no relevance to actual innocence and hence habeas review of such claims does not further the injustice correction purpose.⁽¹²⁶⁾ Stone emphatically rejects the contention that a federal forum is required merely because a federal constitutional question is to be litigated.⁽¹²⁷⁾ It accepts the possibility of imperfection at trial. The requirement is full and fair litigation, not absolute correctness.⁽¹²⁸⁾ The enforcement purpose is not explicitly discussed, but presumably a state court decision which defies outright a controlling Supreme Court precedent would not constitute full and fair litigation.⁽¹²⁹⁾

In Wainwright v. Sykes,⁽¹³⁰⁾ the Supreme Court clarified that habeas corpus relief would not be granted when the defendant failed to timely raise his claim as required by the procedural rules of the state, unless he had good cause for the failure and made a showing of prejudice. This rule implicitly recognizes that when the trial judge's failure to rule on a question is due to defendant's own failure to raise it, there is no need for the enforcement function.⁽¹³¹⁾ In addition, the Court carved out an additional exception for habeas petitioners who are actually innocent in Murray v. Carrier,⁽¹³²⁾ thus recognizing the importance of the correction of injustice. Under Sykes, imperfections not objected to are usually tolerated, and they are generally not litigated on the merits in any forum, state or federal.

The procedural default doctrine serves as one method of separating fundamental from nonfundamental claims. Failure to object at trial to a fundamentally unfair procedure will not insulate that error from habeas review, since such a failure is necessarily ineffective assistance.⁽¹³³⁾ Conversely, the defects in trial to which a competent lawyer does not object are necessarily less than fundamental.

B. Partially adopted proposals.

Three proposed limitations on federal habeas are worth mentioning for the light they shed on the broader problem. Professor Bator's extensive article, Finality in Criminal Law and Federal Habeas Corpus for State Prisoners,⁽¹³⁴⁾ is devoted primarily to demonstrating that relitigation de novo on federal habeas is unjustified. He concludes with a recommendation that federal habeas courts consider actual guilt or innocence as an important, if not determinative, factor in deciding whether habeas relief ought to be granted.⁽¹³⁵⁾

Judge Henry Friendly spoke more directly to the same point. He proposed that "with certain exceptions, an applicant for habeas corpus must make a colorable showing of innocence . . . " before habeas relief could be granted.⁽¹³⁶⁾

Neither proposal has been adopted for the first federal review. For successive petitions, however, Judge Friendly's view was endorsed by a four-Justice plurality of the Supreme Court in Kuhlmann v. Wilson,⁽¹³⁷⁾ with Justice Stevens indicating that guilt or innocence should be considered in the judge's discretion,⁽¹³⁸⁾ along the lines of Professor Bator's proposal. Kuhlmann has since been superseded by McCleskey v. Zant, discussed below.

The partial acceptance and partial rejection of these innocence-driven proposals indicate that innocence is relevant but that there was thought to be a continuing need for the enforcement function as well. A mechanism is needed to correct state court disobedience of controlling precedent, even in the case of clearly guilty defendants. It is most unlikely, however, that such disobedience would go uncorrected on the first federal petition.

A different but equally substantial reform was proposed by Justice Stevens in his dissent in Rose v. Lundy.⁽¹³⁹⁾ Justice Stevens divided constitutional claims into four categories: meritless claims, harmless errors, reversible but nonfundamental errors, and fundamental errors.⁽¹⁴⁰⁾ The last category is described primarily by examples.⁽¹⁴¹⁾ The third category includes all claims under rules which had been held nonretroactive.⁽¹⁴²⁾ These claims, said Justice Stevens, should not be grounds for habeas relief at all.

Like the Friendly and Bator proposals, this one is aimed at preserving habeas for the innocent, but in a less direct manner. It excludes by category those claims of error which are unlikely to result in the conviction of an innocent person, rather than by examining guilt or innocence in the individual case.

The separation of constitutional claims into the categories suggested by Justice Stevens would reject the implicit holding of Brown v. Allen that constitutional claims are fundamentally different as a category from other claims. Instead, constitutional claims would be subject to inquiry as to whether the claimed violation is fundamental, similar to the inquiry presently made for nonconstitutional claims.⁽¹⁴³⁾ Those not meeting the test would not be grounds for collateral relief.

C. Teague and Butler.

In this broader context, Teague can be seen as part of a larger effort to preserve the enforcement and injustice-correcting functions of habeas corpus while applying some limits to the relitigation of the reasonable decisions of the appellate courts.

Teague v. Lane,⁽¹⁴⁴⁾ is the culmination of the Supreme Court's long and difficult struggle with the question of how far back a new decision should reach in undoing the work of judges, lawyers, and juries who had duly obeyed the rules in effect at the time. Whenever a rule is not applied retroactively, the contention is made that the Court has sanctioned a constitutionally flawed conviction or denied someone his constitutional rights.⁽¹⁴⁵⁾ That would be true if one subscribed to the Blackstonian belief that the Court merely discovers pre-existing law and does not make new law.⁽¹⁴⁶⁾ However attractive the Blackstone approach may be in theory, though, it loses any connection with reality when applied to requirements as far removed from the text and history of the Constitution as the Miranda warnings, the exclusionary rule, or the detailed code of capital sentencing procedure which has been promulgated under the name of the Eighth Amendment.

Teague is premised squarely on the dual purpose of habeas corpus review: enforcement of Supreme Court precedents and correction of grave injustices. The Teague rule, as further explicated in Butler v. McKellar⁽¹⁴⁷⁾ and Saffle v. Parks,⁽¹⁴⁸⁾ excludes from habeas review a class of claims where the enforcement value is zero: claims based on rules announced after the date of the state decision. The corollary that new rules cannot be created on habeas corpus inhibits the lower federal courts from doing indirectly what they have no authority to do directly: effectively imposing their own precedents on state courts which have no legal obligation to follow them.

The two exceptions to the Teague rule bring back into habeas two categories of cases where the injustice-correcting function is paramount: cases of people punished for constitutionally protected conduct and people convicted under fundamentally unfair procedures raising grave doubt as to the reliability of the determination of guilt.⁽¹⁴⁹⁾ As applied to the retroactivity question, the latter category is largely theoretical at this point in constitutional history; the fundamental rules have all been made.⁽¹⁵⁰⁾

D. McCleskey v. Zant.

Whatever argument there may be for an enforcement function on the first round of federal habeas corpus vanishes on the second round. State courts are not going to alter their decisions based on the possibility of their judgments being affirmed on the first round but set aside on the second.

Under the rule of McCleskey v. Zant,⁽¹⁵¹⁾ a second round of federal review is precluded unless the defendant shows either "cause and prejudice" or actual innocence. The purpose of the actual innocence exception is clear. It serves the injustice-correcting function of habeas corpus. The cause and prejudice test, which is borrowed from the law of procedural default, is somewhat more opaque.

At one time, a defendant could show "cause" for not having raised the claim earlier by showing that the Supreme Court had unexpectedly created a new rule that his lawyer could not have anticipated.⁽¹⁵²⁾ The Teague rule has virtually abolished this variant of cause. Any rule that new is necessarily nonretroactive, unless it fits within two very rare, narrow exceptions.⁽¹⁵³⁾

The principal form of "cause" today is discovery of new facts that could not have been discovered earlier with reasonable diligence. This exception serves an enforcement function, particularly where agents of the state failed to disclose facts known to them, or even actively suppressed them.⁽¹⁵⁴⁾ The McCleskey test thus indirectly reserves a second round of habeas review for the two classes of cases presenting the most compelling need: actual innocence and government misconduct.

E. The roads not taken.

Some proposals for limiting habeas corpus have not been accepted by the Supreme Court. The reasons for the Court's refusals shed some light on the path of legislative change.

In Rose v. Mitchell,⁽¹⁵⁵⁾ the Supreme Court rejected the proposition that racial discrimination in the selection of grand juries should not be considered on habeas corpus when the defendant has been convicted by a properly selected trial jury, and the grand jury claim has been fully considered by the state courts. The Court noted that trial courts themselves typically select the grand jury, and thus their objectivity in adjudicating a claim of their own wrongdoing is less than complete.⁽¹⁵⁶⁾ The other basis for the decision was that racial discrimination claims have always had a special place in the law, for very good reason.⁽¹⁵⁷⁾

In Wright v. West,⁽¹⁵⁸⁾ the Court discussed, but did not decide, whether the Teague/Butler rule of deference to reasonable state-court resolution of questions of law ought to be extended to mixed questions of law and fact.⁽¹⁵⁹⁾ The opposition to the execution was premised squarely on precedent and the perceived intent of Congress. Justice O'Connor relied heavily on what the Court had said in the past and on Congress' failure to pass proposals to change the rule.⁽¹⁶⁰⁾ Similarly, Justice Kennedy relied on what he perceived to be a "settled principle."⁽¹⁶¹⁾

The West case is also interesting as a stunning counterexample to the notion that federal courts are inherently superior to state courts in applying federal constitutional law. All nine Justices easily concluded that the state courts were right on the merits, and the federal Court of Appeals was clearly and obviously wrong.⁽¹⁶²⁾

Finally, there is Withrow v. Williams,⁽¹⁶³⁾ in which a bare majority of the Court declined to exclude fully litigated Miranda claims from de novo habeas review. The most important reason for this holding was the majority's belief that the excluded Miranda claims would just come back into the habeas court as involuntariness claims, thus producing little benefit to a system where de novo review is the rule and not the exception.⁽¹⁶⁴⁾

Of the three cases, only Mitchell is based primarily on reasons that Congress need consider. The rule that precedents interpreting statutes have special weight in the courts is premised squarely on the belief that Congress can and will act when it disagrees with those precedents.⁽¹⁶⁵⁾ Congress has never, from day one, specified the grounds for use of habeas corpus as a collateral attack, and it is high time it did. The cost-benefit analysis of Withrow v. Williams is based on an assumption which will no longer be valid if Congress does abrogate the rule of Brown v. Allen.