Skip to main content
Find a Lawyer

Did Justice Stevens Pull a Fast One? The Hidden Logic of a Recent Retroactivity Case in the Supreme Court


Monday, Feb. 25, 2008

Last week, in Danforth v. Minnesota, the U.S. Supreme Court ruled that a state court was free to give greater protection to defendants' rights than the Supreme Court itself requires. Stated that way, the decision is hardly news. In our system of federalism, federal constitutional law is not a ceiling, but a floor. It sets out the minimum protections to which people are entitled. If states--through their constitutions or otherwise--choose to add protection, that is their prerogative.

Yet Danforth was no ordinary application of the floor-but-not-a-ceiling principle, because the question in the case was not whether Minnesota could interpret its own state law more broadly than federal law. Everyone accepts that it (like every other state) can. The question in Danforth was whether Minnesota could over-protect federal law. Perhaps surprisingly, the Supreme Court said yes.

Although the Danforth case involved highly technical and somewhat convoluted doctrine, it nonetheless warrants unpacking, for it may reveal an unexpected and important shift in the Justices' thinking about the relationship of state law to federal law.

The Danforth Case and the Retroactivity Question

At Stephen Danforth's 1996 trial for sexual conduct with a minor, the prosecution introduced a videotaped interview of his six-year-old victim. Danforth objected to this evidence, but the state courts rejected the objection under the standard the Supreme Court had set forth in the 1980 case of Ohio v. Roberts: whether the evidence was sufficiently reliable to satisfy the Sixth Amendment right of the accused to confront his accuser. Danforth was convicted.

After Danforth had exhausted his direct appeals, in 2004, the Supreme Court overruled the Roberts decision. In Crawford v. Washington, the Justices held that the admission into evidence of videotaped testimony of a witness who is otherwise available for trial, violates the Sixth Amendment (which applies to the states via the Fourteenth Amendment). After the Crawford decision, Danforth filed a habeas corpus petition in Minnesota state court, seeking a new trial.

If Danforth had sought habeas relief in federal court, he would have lost. Under the Supreme Court's landmark 1989 decision in Teague v. Lane, federal courts do not grant habeas relief to state prisoners based on "new rules" of constitutional law, except in two narrow circumstances. The rationale for the Teague rule is straightforward: States have a strong interest in the finality of criminal convictions; if a defendant had a trial that conformed to the constitutional standards that were understood to apply at the time of that trial, then the state generally should not have to re-try the defendant simply because it failed to anticipate a novel decision by the Supreme Court.

A contrary rule would put too great a burden on the state and could also create a disincentive for the Justices to recognize constitutional rights: If recognizing a new constitutional right required new trials for defendants who had originally been tried decades earlier, then the Justices would be very reluctant ever to recognize new rights.

So, did Crawford announce a new rule? Of course it did. Further, the Court said in the 2007 case of Whorton v. Bockting, Crawford's new rule requiring live testimony--overruling Roberts--did not fit into either of the exceptions to the Teague non-retroactivity principle. Thus, Danforth could not get federal habeas relief based on Crawford.

Moreover, all the Justices agreed that the Minnesota courts were not required to apply Crawford in Danforth's case. After all, the same fear of opening old cases that leads to the application of the Teague non-retroactivity rule in federal court, applies to state habeas cases also.

The Minnesota Supreme Court went a step further, however: It also said that the state courts were not permitted to apply Crawford retroactively in Danforth's case.

Is Retroactivity a Choice-of-Law Question or a Remedial Question?

In the U.S. Supreme Court, two Justices--Chief Justice Roberts and Justice Kennedy--agreed with the Minnesota Supreme Court. Crawford was an interpretation of federal constitutional law, they said, and therefore the question whether to apply it retroactively is also a question of federal law. And citing precedents going back to the early Eighteenth Century, the Chief Justice said that the Supreme Court, not state courts, decides federal questions.

The 7-2 majority, in an opinion by Justice Stevens, did not exactly disagree with the Chief Justice's general point. If retroactivity were a pure question of federal law, Stevens acknowledged, then the Minnesota courts would not be free to fashion a view of retroactivity that differed from the Supreme Court's view. But the majority did not view retroactivity as a purely federal question.

Whether to apply Crawford or any other federal decision retroactively is a question of remedy, the majority said, and "the remedy a state court chooses to provide its citizens for violations of the Federal Constitution is primarily a question of state law." Federal law, in this now-familiar view, creates a floor but not a ceiling: States must give retroactive remedial effect to decisions the Supreme Court says apply retroactively, but states are also free to choose whether or not to give retroactive remedial effect to decisions the Supreme Court says they are not required to apply retroactively.

The dissenters strenuously objected to the characterization of retroactivity as a question of remedy. Teague and subsequent cases, the dissenters said, treat retroactivity as a question of what law applies to a given case--old law or new law. And that question, they added, can, itself, only be a matter of federal law.

The Puzzling Split in the Danforth Case

For non-specialists--indeed, even for specialists--it is difficult to see why the result in the Danforth case should have turned on the semantic question of whether we characterize retroactivity as a matter of remedy or a matter of choice of law (old or new). Surely, one thinks, there must be something else driving the decision. But what?

Ideology is not a very good candidate. Although the Court's most liberal Justices--Stevens, Souter, Ginsburg, and Breyer--were all in the majority, the case split the conservatives: Justices Scalia, Thomas and Alito joined the majority, while Chief Justice Roberts and Justice Kennedy dissented. Nor is that result entirely surprising as an ideological matter, because the case pitted two conservative principles against each other: On one hand, allowing the state courts to give greater retroactive effect to Supreme Court decisions than the Justices themselves requires, vindicates states' rights. On the other hand, denying state courts that right vindicates the competing conservative principle of finality in criminal law.

Although general principles of conservatism cannot explain the split in Danforth, perhaps other "psychological" analyses of the Justices can shed some light on the ruling. In his majority opinion in Danforth, Justice Stevens repeatedly invoked--and quoted at length from--earlier decisions written by Justice Scalia. Moreover, the underlying principle at stake came from the Crawford opinion, which Justice Scalia authored for the Court. Nonetheless, it is hard to believe that Justice Scalia would have voted with Justice Stevens simply out of vanity or gratitude for the respect granted his prior work.

Did Justice Stevens Hoodwink Six of His Colleagues?

Moreover, Danforth is an especially puzzling decision in light of another case--one not mentioned by either the majority or dissent. In 1983, in Michigan v. Long, the Court confronted the question of whether it had jurisdiction to entertain a petition by a state alleging that its own state high court had over-protected federal constitutional rights. In that case, the Michigan Supreme Court's opinion was unclear as to whether the ruling was based on federal law alone or a combination of state and federal law. The Court said that jurisdiction was proper in the case, but more importantly for our purposes, Justice Stevens wrote a remarkable solo dissent.

In Long, Justice Stevens said that the Supreme Court should never reverse a decision by a state court over-protecting federal rights. Over-protection of American citizens' federal constitutional rights by Michigan, he said, is no different from over-protection of American citizens' rights by a foreign nation such as Finland.

Not surprisingly, no other Justice agreed with this characterization of Michigan and Finland as standing in the same relationship relative to the United States, for the obvious reason that Michigan is part of the United States, while Finland is not. As the majority opinion in the Long case noted in response to Justice Stevens, state courts issue thousands of final rulings of federal law each year. Accordingly, the view expressed by Justice Stevens would threaten the uniformity of federal law.

The same point was raised in vain by Chief Justice Roberts last week in his Danforth dissent. The retroactive effect of a federal constitutional ruling, he said, is a question of federal law, and so giving states latitude to give such a ruling extra retroactive effect results in the disuniformity of federal law.

Justice Stevens had a response in Danforth that did not, as a technical matter, depend on his view that Michigan (or Minnesota) is no different from Finland: Retroactivity, he pointed out, is primarily a matter of state law. And everyone agrees that state courts can fashion their own rules of state law.

Nonetheless, it is difficult to shake the impression that Justice Stevens pulled a fast one in Danforth, mostly because he did not clearly explain why the retroactivity of a decision of federal law should be a matter of state law. Article I, Section 6 of the Minnesota Constitution contains a Confrontation Clause that is virtually identical to the Sixth Amendment of the U.S. Constitution. No one doubts the authority of the Minnesota Supreme Court to interpret its state constitutional provision to embody the Crawford rule, and to apply that interpretation retroactively. However, the Danforth decision confers on the state courts a very different power.

If, God knows why, Finland were to decide to apply Crawford retroactively, that would be no concern of the U.S. Supreme Court. But Minnesota is not Finland.

Michael C. Dorf is the Isidor & Seville Sulzbacher Professor of Law at Columbia University. He is the author of No Litmus Test: Law and Politics in the Twenty-First Century and he blogs at

Was this helpful?

Copied to clipboard