The great writ and the right to silence.

In Withrow v. Williams (113 S. Ct 1745 (1993)), the Supreme Court for the fourth time refused to extend the rule of Stone v. Powell (428 U.S. 465 (1976)). In Stone, the Supreme Court held that state prisoners may not obtain review of state court denials of Fourth Amendment suppression motions by filing a federal habeas corpus petition. The stone majority reasoned that federal review would do little to deter police misconduct. Not only is the prospect of ultimately "losing" a conviction in post-conviction proceedings too remote to affect police behavior, but so long as the state courts respect federal search-and-seizure law, police must expect to lose the right to present illegally obtained evidence in state court.

Stone combined two trends in Burger Court doctrine. One was hostility to the exclusionary rule, which the Burger Court limited in a wide variety of contexts. (See, e.g., United States v. Payner, 447 U.S. 727 (1980); United States v. Janis, 428 U.S. 433 (1976); United States v. Calandra, 414 U.S. 338 (1974).)

The other trend was hostility to federal habeas jurisdiction over claims by state prisoner. Justice Lewis Powell, the author of the Stone opinion, believed in confining habeas claims to allegations of errors implicating the reliability of the state court's determination of the petitioner's guilt. (See Schneckloth v. Bustamonte, 412 U.S. 218, 250 (1973) (Powell, J., concurring).)

Three post-Stone cases strongly imply that the dominant factor in the Stone decision was hostility to the exclusionary rule, not to the habeas remedy. (Kimmelman v. Morrison, 477 U.S. 365 (1986) (claim that counsel's ineffectiveness led to loss of suppression motion cognizable on federal habeas, Stone notwithstanding); Jackson v. Virginia, 443 U. S. 307 (1979) (federal court may review sufficiency of evidence on habeas); Rose v. Mitchell, 443 U.S. 545 (1979) (claim of discrimination in selection of grand jurors cognizable on federal habeas).)

Jackson claims, of course, go to guilt and innocence. Nonetheless, almost every losing litigant can raise an insufficiency-of-the-evidence claim. Jackson suggests that considerations of judicial economy alone are not enough to bar the federal habeas remedy.

In both Kimmelman and Mitchell, the petitioner's claim did not implicate the state court's finding of guilt. But in these cases, there were reasons to be skeptical about the adequacy of state judicial processes. Without effective counsel, no court - state or federal - can be expected to vindicate the defendant's Fourth Amendment rights. Nor can state courts be trusted to fully and fairly hear claims of discrimination by the state court system itself.

Miranda Claims

Given the Jackson, Kimmelman, and Mitchell cases, the doctrine announced in Stone appeared to have only limited applicability. The most likely subject to be withdrawn from the federal habeas jurisdiction seemed to be claims based on Miranda v. Arizona. (384 U.S. 436 (1966).)

Just as in Stone, claims of Miranda, violations do not undermine the reliability of the fact-finding process. There is as much reason to trust state court decisions about police interrogation practices as there is to trust those decisions in the search-and-seizure area. When the courts review the work of the police, the problem of self-regulation, present in Mitchell, is absent.

Relying on these points, Justice Sandra Day O'Connor, joined by Justice Antonin Scalia, made a plea for extending Stone to Miranda claims. (See Duckworth v. Eagan, 492 U.S. 195, 208 (1989) (O'Connor, J., concurring).) The similarities between the Miranda context and Stone led many to conclude that "an extension of Stone to Miranda violations seems likely." (Wayne R. LaFave & Jerold H. Israel, Criminal Procedure 1202 (2d ed. 1992).)

The Supreme Court, however, delights in surprises. In Williams, by a vote of five to four, the Court held that Miranda claims rejected in state court may be relitigated on federal habeas review. (113 S. Ct. 1745.)

After making inculpatory statements to police officers, Williams was convicted of two murders. The officers claimed that Williams had voluntarily accompanied them to the station for questioning; Williams claimed he was under arrest and therefore in custody and entitled to Miranda warnings.

The Michigan courts rejected his claim. (See Michigan v. Williams, 429 N.W.2d 649 (Mich. Ct. App. 1988), appeal denied, 440 N.W.2d 416 (Mich. 1989).) On federal habeas, both the district court and the court of appeals agreed that Williams's rights under Miranda had been violated. (See Williams v. Withrow, 944 F.2d 284 (6th Cir. 1991).)

Justice David Souter's majority opinion affirming the lower federal courts distinguishes Stone from Williams on two major grounds - one ideological and unconvincing, the other pragmatic and compelling. The ideological ground characterizes Miranda, and the Fifth Amendment privilege against self-incrimination more generally, as part and parcel of a fair trial. (See Williams, 113 S. Ct. 1745, 1752-53).) Justice Souter here echoes Wigmore's arguments that systemic reliance on confession ultimately leads to abusive questioning, and that coercion induces statements that arc of doubtful reliability.

Courts, however, routinely, receive confessions. Courts compel, through the contempt power, unwilling testimony that does not incriminate the witness, and sometimes they, compel incriminating testimony by way of immunity. Often a confession is corroborated by other evidence, and in every case its credibility is a matter for the trier of fact to decide.

Souter's description of Miranda as a component of a fair trial, as distinct from a safeguard against abusive police questioning whether or not a trial ever occurs, simply does not ring true. Souter's pragmatic point has more force. Before Miranda, the Supreme Court had held that the Due Process Clause of the Fourteenth Amendment prohibits introducing a coerced confession at a state trial. (See, e.g., Brown v. Mississippi, 297 U.S. 278 (1936).) This voluntariness requirement was distinct from the privilege against self-incrimination and remains so, as shown in cases where a statement that is admissible under Miranda is nonetheless excluded as involuntary. (See Mincey v. Arizona, 437 U.S. 385 (1978).)

Voluntariness Test

Under the voluntariness test, failure to comply with Miranda is one factor suggesting coercion. (See Procunier v. Atchley, 400 U.S. 446, 454 (1971).) Compliance with Miranda does not, however, bar a coercion claim on a per se basis. In consequence, denying habeas review of Miranda claims would do little to reduce the workload of the federal courts. Most petitions dismissed on Miranda grounds could be recast as coercion claims and promptly refiled. (See Williams, 113 S. Ct. 1745, 1754.)

In dissent, Justice O'Connor, joined by Chief Justice William Rehnquist, argued that Miranda sweeps more broadly than the voluntariness test or the Fifth Amendment privilege itself. In the view of these two, any "true Fifth Amendment violations" would also violate the voluntariness test; indeed the voluntariness test all along has been the means by which "the courts enforced [the Fifth Amendment's] prohibition" on compelled self-incrimination. (Id. at 1761.)

This characterization of the scope of a suspect's rights under the Fifth Amendment privilege, Miranda, and the due process voluntariness test is flat-out wrong. In the first voluntariness case, Brown v. Mississippi, the Supreme Court said that "the question of the right of the state to withdraw the privilege against self-incrimination is not here involved. . . ." (297 U.S. 278, 285.)

Until the 1960s, the only references to the Fifth Amendment privilege in Supreme Court confessions cases came in dissenting opinions. (See Walter V. Schaefer, The Suspect and Society 15 (1967).) Indeed, not until 1964 did the privilege become binding on the states. (Malloy v. Hogan, 378 U.S. 1 (1964).)

Testimony, can be "compelled" by, pressures far slighter than those that amount to "coercion." Prosecutorial comment on the failure to testify violates the privilege, but it hardly smacks of coercion. (See Griffin v. California, 380 U.S. 609 (1965).) One cannot equate the Fifth Amendment privilege, the Miranda rules governing pre-trial interrogation, and the venerable due process test. Doing so ignores both legal history and the distinctive policy concerns that support each of these quite different doctrines.

Justice Scalia's dissent is more cogent but more radical. His opinion, joined by Justice Clarence Thomas, states that re-litigation of a federal question on habeas corpus review is discretionary with the federal courts, and petitioners seeking review must demonstrate unusual equities weighing in favor of review. (See 113 S. Ct. 1745, 1765-70).) Whatever its merits, this position calls less for extending Stone than for overruling Brown v. Allen (344 U.S. 443 (1953)), which held that even nonjurisdictional constitutional violations might be challenged on habeas.

Difference Among Justices

Thus, only two Justices - O'Connor and Rehnquist - were willing to extend the logic of Stone in the context of Miranda violations. Given the seemingly strong parallels between Fourth Amendment claims and Miranda claims, the decision in Williams might suggest that Stone is confined to Fourth Amendment claims.

But two Justices, Scalia and Thomas, stand ready to withdraw habeas scrutiny from constitutional claims absent special equities favoring review, and the majority in Williams was heavily influenced by a pragmatic point in that case. Similarly, in Jackson, Kimmelman, and Mitchell, the facts of each of these cases seemed to have priority over more general ideology logical considerations.

So it is possible-although it seems unlikely-that the Court might decide that, say, Sixth Amendment claims under Massiah v. United States might not be heard on federal habeas. (377 U.S. 201 (1964).) To put it another way, this is one area of the law where nothing is set in Stone.