Brecht v. Abrahamson – Oral Argument – December 01, 1992

Media for Brecht v. Abrahamson

Audio Transcription for Opinion Announcement – April 21, 1993 in Brecht v. Abrahamson

del

William H. Rehnquist:

We’ll hear argument now in 91-7358, Todd A. Brecht v. Gordon A. Abrahamson.

Mr. Shoenberger.

Allen E. Shoenberger:

Mr. Chief Justice and may it please the Court:

Mr. Brecht was convicted after a jury trial after the special prosecutor in this case first on cross-examination of Mr. Brecht then on re-cross initial closing argument and final closing argument, breached a promise that had been made to Mr. Brecht that if he remained silent at a certain point in time that silence would not be used against him.

The case involves a core due process violation.

It does not involve the Miranda case itself, nor is it a case that involves a prophylactic rule or a prophylactic right.

Let me explain.

If this Court… pardon?

Byron R. White:

So it isn’t a Fifth Amendment case.

Allen E. Shoenberger:

Not primarily.

This is a due process case.

The promise that was made–

Byron R. White:

Which is Doyle… which is Doyle.

Allen E. Shoenberger:

–The promise that was made to him could have been any promise.

The promise was made, and that’s important, but the promise is analogous to the promise that was made in Johnson v. United States, United States v. Hale, Raley v. Ohio, Santobello v. New York, which is the plea bargain line of cases, as well as what was stated in the Cupps v. Louisiana case, where the officers told certain civil rights demonstrators that they had the right to go and demonstrate at a certain place, and then those same demonstrators were prosecuted for demonstrating near the courthouse.

In all of these cases, the State told somebody they had a right to do something, then attempted in one way or another to penalize them, and this Court in all of those cases said that the use of that penalty against that person violated due process.

It’s also not… this case is also not a prophylactic case.

It does not involve a prophylactic rule, because all that Mr. Brecht is seeking is the specific performance of the promise that was given to him that if he remained silent, that silence would not be used against him.

On a scale of 1 to 10, where 10 is prophylactic, this case is a zero.

He’s seeking exactly the due process performance, the specific performance of the promise that was made that at the trial which was going to be subsequent, or was subsequent to the promise that was made, that the State not be permitted to use its breach of that promise to attempt to secure his conviction.

He seeks that to be excluded.

The constitutional violation in Mr. Brecht’s case would occur at trial.

It’s not like the questions of illegal searches and seizures where this Court may have to deal with cases after the constitutional violation has occurred, and the question of remedying that violation that’s already occurred is part of what this Court has to consider.

In addition, the case involves, and Doyle involves, a question of the legitimacy of inferences.

This case does go to what kind of inference can be drawn from the silence.

At a certain point, we would contend, an inference can become so tenuous that it violates due process to draw the inference.

I would suggest that Justice O’Connor’s opinion in Estelle v. McGuire suggests the possibility of such matters, but in this case there’s a problem of unreliability, that the inferences that may be drawn from his silence and the use of that silence after the Miranda warning is certainly unreliable.

Antonin Scalia:

Mr. Shoenberger, he wasn’t explicitly told that if he remained silent that silence would not be used against him, was he?

Allen E. Shoenberger:

No.

That explicit statement–

Antonin Scalia:

There was no explicit statement in this case.

You’re just relying upon the general inference from a Miranda warning.

Allen E. Shoenberger:

–Yes, and as this Court has interpreted in Doyle and subsequent cases.

Actually, it’s not exactly clear what statement was made to him.

The record only reflects that at the first appearance in court he was given his Miranda rights, but the record that I have seen does not actually contain a text of the rights that were given to him, so we’re assuming that the Miranda rights were the typical set of Miranda rights.

David H. Souter:

Well, it’s true, isn’t it, every court that’s dealt with this so far has found a Doyle violation, so that’s the… the premise for your argument is clear, isn’t it?

Allen E. Shoenberger:

Every court except the first court, the trial court… the most important one.

It didn’t find a Doyle violation, and it let the prosecutor go ahead and use the arguments and the questions that he was using despite the fact that they were objected to repeatedly by counsel.

David H. Souter:

That’s right, it was found to be error and harmless error in the State appellate court.

Allen E. Shoenberger:

The appellate court… the State appellate court found it was not harmless error, the Wisconsin Supreme Court found it was harmless error, and that’s the only court that has found it to have been harmless error.

Byron R. White:

Well, that was the standard they applied… harmless error.

Allen E. Shoenberger:

That’s correct.

Byron R. White:

The Chapman standard.

Allen E. Shoenberger:

That’s what they purported to apply, yes.

William H. Rehnquist:

So we are the sixth court to have considered the question in this case.

Allen E. Shoenberger:

Yes, Your Honor, you’re the sixth.

The fact that each court has reversed the decision, in effect, of the court before suggests the closeness of the issue in terms of this particular case, and since this is a case about the allocation of risks, of who loses in close cases, we think it’s a particularly appropriate case that the current Chapman standard remain the standard test.

Byron R. White:

So what happened in the district court on Federal habeas?

Allen E. Shoenberger:

The district court… the district court properly applied the Chapman standard and found that the error was not harmless beyond a reasonable doubt, that the State had failed to meet its burden, and found specifically that the use of the silence, the impermissible use of the silence, may have been critical to the actual–

Byron R. White:

So it disagreed with the State supreme court.

Allen E. Shoenberger:

–Yes, it did.

Byron R. White:

And the court of appeals reversed the district court–

Allen E. Shoenberger:

The court of–

Byron R. White:

–And applied a lower standard.

Allen E. Shoenberger:

–The court of appeals applied a different standard, reversing the district court, yes, Your Honor.

Sandra Day O’Connor:

Do you think that a Federal court on habeas owes some deference to State court findings as in this case that the error was harmless?

Allen E. Shoenberger:

Not to the findings that the error was harmless.

At least, there’s no absolute deference that is supposed to be paid.

This Court in Brown v. Allen many years ago suggested that these kinds of issues are ones in which you should certainly pay attention to the State’s–

Sandra Day O’Connor:

Yes, the State court is presumably in a better position to make that determination than the Federal court.

Sandra Day O’Connor:

I wonder why deference to those findings isn’t appropriate?

Allen E. Shoenberger:

–The deference that is due, is what this Court said in Brown v. Allen, is what those findings are entitled to, but no more than that.

This Court, or Federal courts under habeas have a different command, a command that’s narrowed to the guardianship of particular Federal rights, normally Federal constitutional rights, and in its special charge given to it by Congress, both to it and to the Federal district courts, it has an obligation… or the Federal courts have an obligation to see that fundamental rights remain to be protected.

Sandra Day O’Connor:

Why do you suppose we don’t apply a rule that in determining whether there is a due process violation in the first instance you have to examine whether there’s been any harm done?

Why shouldn’t the harmless inquiry occur at the level of determining whether there was a violation of due process in the first place?

Allen E. Shoenberger:

This Court can certainly establish due process rules in that manner, and I think in some cases it may have.

Sandra Day O’Connor:

I mean, that would be typical.

In some cases we’ve done exactly that, have we not?

Allen E. Shoenberger:

But in the Doyle case line this Court has not started off that way.

It started off by saying that this is a violation itself, creating a bright line test which at least eliminates the difficult questions that often occur of whether or not a constitutional violation has occurred.

The fact of the violation is not really contested in this Court.

Hale isn’t contested, I think, in even the Federal district court in this case.

The question’s always been the things that can be drawn from that violation.

For most of this Court’s history, habeas corpus review, once habeas corpus review reaches the merits, has applied the same standard of review for habeas corpus as this Court has applied on direct appeal.

Now, there’s a big caveat.

Today, as the Court knows, there are many hurdles to getting into a Federal habeas corpus court and reaching the merits.

They include questions of exhaustion, the abuse of the writs, waiver, and various other procedural hurdles.

This case doesn’t raise any of those questions.

Mr. Brecht has properly preserved all of the issues that are presented today to this Court.

There’s no allegation that none has not been.

There’s no abuse.

This is his first attempt to seek Federal habeas corpus relief.

So he’s one of those few cases that have managed to make it through the entire minefield of procedural barriers to seek to reach a determination about the merits of the core issue that’s presented.

Moreover, it involves no Teague issue.

He’s not seeking any kind of different right that would otherwise have not been available to him on direct appeal.

We see the Teague rule as simply stating that you get no more when you go up on habeas corpus than you would have gotten had you gone up directly on certiorari on direct appeal and submitted to the court under the rules of law applicable at that point in time.

Anthony M. Kennedy:

What was your explanation for the phenomenon that the rule on 2255 in Federal habeas, the standard of review, seems to differ from that of Chapman–

Allen E. Shoenberger:

It does–

Anthony M. Kennedy:

–And why should this be, if there is this difference?

Allen E. Shoenberger:

–Well, the cases that I’ve seen… I’m assuming I’m thinking about the same sort of case line… would suggest that in Federal… Federal habeas corpus in 2255 relief, it’s quite analogous to, or the Court treats it as analogous to, in some instances such as out of the District of Columbia convictions, that are effectively the same as in State court.

Allen E. Shoenberger:

The more recent decision… the case name slips my mind right now… is really an application of Teague without announcing at the time, because it predates Teague but Teague principles are involved, suggesting that the same rule, the same substantive rule, ought to apply on 2255 at the time of the case, had it gone up on direct appeal.

Anthony M. Kennedy:

What about the standard for determining whether or not there’s reversible error?

Should the standard be the same in both proceedings, State habeas and 2255, or Federal habeas to a State proceeding on the one hand and 2255?

Allen E. Shoenberger:

I think different considerations may sometimes apply, particularly when you’re coming from Federal jurisdictions and Federal convictions.

Anthony M. Kennedy:

Why?

Allen E. Shoenberger:

This Court’s supervisory power may also be implicated.

Anthony M. Kennedy:

Well, that would indicate–

Allen E. Shoenberger:

Part of the–

Anthony M. Kennedy:

–That would indicate that we would have a broader standard of review over a Federal conviction.

Allen E. Shoenberger:

–I don’t think it speaks to the standard of review, or the appropriateness of the standard of review.

I think it speaks to the question of this Court’s power.

The interplay between the Federal Rules of Criminal Procedure and the rules for purposes of habeas corpus relief that this Court has announced, or… and Congress has enacted, that imports a whole set of different considerations that are not applicable in 2254 cases, and so it may be that certain–

Anthony M. Kennedy:

Well, I’m asking what those different considerations are, because it seems to me that at root is the proposition or the premise that State courts are not as adept or as capable as enforcing Federal rights of determining them as the Federal courts are.

Allen E. Shoenberger:

–I don’t think that’s part of the proposition.

The constitutional plan indicates that Federal courts, at least this Court in particularly, ultimately is to decide Federal constitutional questions.

Anthony M. Kennedy:

So you’re willing to submit the argument on the theory that State courts are as adept and as capable and as willing to enforce Federal rights as Federal courts are.

Allen E. Shoenberger:

I don’t think that’s at issue.

Congress has said that Federal–

Anthony M. Kennedy:

Well, if it’s not at issue, are you willing to put the argument on that basis?

Are you willing to accept that as a premise?

Allen E. Shoenberger:

–No, I’m not willing to accept that, and I think this Court has several times suggested that that argument is inappropriate, that there is an obligation that Federal courts have to ensure that State courts tow the constitutional line, or tow the constitutional mark when reviewing the Federal constitutional questions which they sometimes must review, and that in the ideal world there would be no lesser standard and no lesser guarantee of Federal constitutional rights in State courts.

Anthony M. Kennedy:

But in applying those rights to determine whether or not a conviction should be reversed under a Chapman standard or some other standard, aren’t the State courts in an equal or better position than the Federal courts to make that determination?

Allen E. Shoenberger:

No, they’re not, and they’re not because they have so many other considerations to worry about in cases that come up on direct appeal in the State courts.

Federal constitutional rights are certainly something they’re concerned with.

It’s only one of many things they are concerned with.

The Federal habeas corpus remedy, particularly 2254, focuses in and narrows in on one set of peculiar rights, Federal constitutional rights… nothing else.

State error, State procedural error, State law interpretations, all those kinds of issues, are stripped out of the case.

They’re not available to this Court or to Federal district courts on Federal habeas corpus, so what the Federal court has is a very different kind of question.

Anthony M. Kennedy:

So is the difference one of competence between the two types of tribunals?

Allen E. Shoenberger:

No.

Allen E. Shoenberger:

The difference is one of the charge that Congress made to Federal district courts, to courts of appeals, and this Court, that they are supposed to enforce Federal constitutional rights under the Federal habeas corpus provisions in 2254.

To go back for a second to 2255, in 2255 cases, there has already been a Federal court, or possibly several Federal courts, that in theory have already addressed the issue, so the interests of ensuring that courts tow the constitutional mark applies in a different relationship to 2255 cases than it does in terms of 2254.

There’s already a Federal tribunal that has addressed the case at some point.

That’s not true on 2254, and it may be an additional reason for different treatment in these cases.

I would not suggest that it’s still appropriate that the Chapman standard should be changed for 2255, although that’s not at issue in this case.

The harmless error standard, we would suggest, is the appropriate standard.

It’s the appropriate standard for a number of reasons.

For one reason, it’s stare decisis certainly on direct appeal since Chapman.

I don’t think anybody would contest that.

It’s also stare decisis on collateral attack certainly since Rose v. Clark in 1986, and I would suggest as well, since six justices expressed that opinion in Dutton v. Evans in 1970, and all of the cases that are cited in the reply brief, justice after justice after justice in opinion after opinion has suggested that the harmless error standard is appropriate in 2254 cases.

It’s a standard that works.

It’s worked for many years.

It involves the application of de novo review, certainly of questions of law and mixed questions of law and fact, once the case reaches appropriately a Federal district court.

I would also suggest that it serves interest in efficiency as well as interests in comity.

William H. Rehnquist:

Has this Court ever decided, Mr. Shoenberger, whether harmless error is a question of law or fact, or a mixed question of law and fact, such that whether or not deference might be due to a State for determination?

Allen E. Shoenberger:

I’m not aware of this Court ever deciding that particular proposition.

Byron R. White:

Well, the district court specifically accepted the facts, the historical facts of the State court, but then said that whether a constitutional error is harmless is a question of Federal law, it is not subject to the presumption of correctness given to the State court’s findings of fact, and it cited… it cited a court of appeals case in the 7th Circuit which in turn cited Harrington v. California.

That was a direct appeal.

Allen E. Shoenberger:

That’s… Harrington is direct appeal.

Byron R. White:

Yes.

Allen E. Shoenberger:

And we’d certainly–

Byron R. White:

Well, yes, but it’s a question of whether it’s a question of law or fact.

Allen E. Shoenberger:

–We believe it’s a question of law–

Byron R. White:

Yes.

Allen E. Shoenberger:

–As a practical matter, and we would suggest that what’s at issue here is certainly an issue for de novo review by the Federal courts.

As a practical matter, there really weren’t any facts at dispute in this case except for one, the fact that disputes with the intent or alleged intent or purpose of Mr. Brecht in killing.

John Paul Stevens:

Well, there is another… at least an unresolved matter of fact, and that’s whether or not he received Miranda warnings before the Wisconsin Supreme Court mentioned when he was at the first… apparently after he had been arrested.

Allen E. Shoenberger:

That certainly… if there’s a question of fact about that, that fact question’s not been–

John Paul Stevens:

So at his initial appearance is the only thing in the record about getting Miranda warnings.

Allen E. Shoenberger:

–That’s correct.

Allen E. Shoenberger:

That’s correct.

John Paul Stevens:

And isn’t there some uncertainty… incidentally, in your brief you quote the erroneous statements by the prosecutor.

Is that a complete… are those all the erroneous statements that you rely on, those at pages 3 and 4 of your brief?

Allen E. Shoenberger:

No.

Well, those are the ones that have been found by one of the courts already to rule in the matter to have violated the post-Miranda period Doyle rule.

John Paul Stevens:

You see, because it seemed to me that one could read at least several of those as not necessarily referring to the time after his initial appearance, because they’re talking about contacts with the police, and I take it those contacts might have been before his initial appearance and therefore before, as far as the record shows, he got Miranda warnings.

There’s some ambiguity, is all I’m saying.

Allen E. Shoenberger:

There’s ambiguity about at least one of those particular references, but the fact that it’s ambiguous is something that goes directly to the jury’s fact-finding power.

It’s the jury that’s supposed to resolve ambiguity in terms of these kinds of factual issues, not the court, and–

John Paul Stevens:

Well, but they didn’t have a–

Allen E. Shoenberger:

–There is dispute–

John Paul Stevens:

–The jury didn’t have a question put to it that raised the Doyle issue, did it?

Allen E. Shoenberger:

–That’s right.

John Paul Stevens:

They weren’t asked when were Miranda warnings given, or was he silent before or after he got Miranda warnings.

They weren’t asked that.

Allen E. Shoenberger:

That’s correct.

They were not told at all about the Doyle issue.

There was no instruction given to them.

John Paul Stevens:

And see, some of these statements about the… he didn’t say anything until he bumped into the… at the time he talked to the police.

I don’t have the exact language, but I can’t tell from the papers whether that’s before or after he got Miranda warnings.

Allen E. Shoenberger:

The question is not the statements.

The statements that are quoted are all quoted from the trial.

That’s long after–

Anthony M. Kennedy:

Right.

Allen E. Shoenberger:

–The Miranda warnings.

The question is, what the referent of the statement is–

John Paul Stevens:

That’s exactly what I mean.

Allen E. Shoenberger:

–And sometimes it’s a little bit unclear.

On certain of them it’s clear that it does include the period of Miranda–

John Paul Stevens:

But, see, to the extent that it’s unclear, it’s not quite as clear as we assumed at the outset that any promise was broken.

John Paul Stevens:

You see, if they’re pre-Miranda warning statements, your argument about breaking a promise is not really valid.

Allen E. Shoenberger:

–Out of what I count up as five particular statements that the courts that have reviewed below have found to have breached the Doyle rule, only one, I believe, legitimately fits into a dispute category in terms of its referent, and that, the ambiguity in terms of that referent under Doyle and under Chapman ought to count in favor of Mr. Brecht in terms of trying to figure out whether there was a possible impact under Chapman to the conviction.

Byron R. White:

Well, the district court in granting habeas certainly identified the post-Miranda statements that he relied on as saying that it might have made a difference.

Allen E. Shoenberger:

That’s right.

That’s right, and that’s why we’re here today, because we believe it did make a difference, that there are enough post-Miranda rights, post-Doyle violations in this case that it might have made a difference to this particular jury.

What we don’t know is how this jury weighed any of those particular arguments, but we do know… and the joint appendix includes a large part of the summation by the prosecutor.

We do know how central the argument about silence, about the thrust of truthfulness of Mr. Brecht’s testimony, was to the argument of the case below to the jury, and so it’s not something we can say well, this is a side issue.

This went right to the core of the case, right to the core of whether or not there was a constitution… whether or not there was a finding, an appropriate finding of his purpose or intent to kill which was required under the law of Wisconsin before a conviction could come down validly.

We would suggest that maintaining the Doyle rule, or maintaining Chapman, is quite appropriate in terms of comity, because it does allow Federal courts to use the analysis of the State courts when they analyze on direct appeal under Chapman.

That analysis can fully inform the reviewing Federal court of how the State courts saw the particular interrelations of fact and law in the particular case.

If you apply any other standard on Federal habeas corpus review, what you’re doing is you’re deprecating that particular analysis and telling the Federal court, well, you have to do something else.

And there’s argument about that may or may not be something that is more burdensome or less burdensome in the Federal court, but it certainly would be something different, and the particular analysis that the State supreme court or appellate court went through would be something that would not be directly relevant to that issue.

William H. Rehnquist:

Unless we were to say that that is the sort of determination by the State court that is entitled to some deference in Federal habeas proceedings.

Allen E. Shoenberger:

But unless this Court said that complete deference of the type that Stone v. Powell is involved, would be appropriate, the Federal courts would still have to decide something, and in fact we suggest in our brief that the standard on Kotteakos developed for direct appeal requiring review of the entire record means that it’s at least as burdensome, if not more burdensome.

William H. Rehnquist:

Well, but I’m not talking about introducing the Kotteakos standard.

I’m suggesting a possible way that this case could be decided is that a finding of harmless error, a determination to that effect by the State court, would be entitled to deference under 2254 in the Federal habeas proceedings.

It wouldn’t be a different standard, but some deference would be required to the fact that the State court had reached that conclusion.

Allen E. Shoenberger:

Well, we believe there already is a deference inherent in the Brown v. Allen decision, but the deference is not a deference to say that we’ll back off and a reasonable application of Federal law by State courts is what the defendant in this particular case was entitled to.

We believe that the Federal court should take, for what it’s worth, the State court determination, but the Federal court has, under the habeas corpus statute, a duty and an obligation to review the Federal constitutional rights that are involved and to make the right decision at that point.

William H. Rehnquist:

Well, you say take it for what it’s worth, which of course is a somewhat ambiguous phrase.

You can say, it means no more than if we agree with it, we’ll accept it, if we don’t we’ll reject it.

What I’m suggesting is that maybe a Federal court is obliged, even if it would not have come out that way itself, to say that the State court finding of harmless error shouldn’t be set aside.

Allen E. Shoenberger:

That would radically change the law of Federal habeas corpus from the way it has been for the last, I would suggest, 200 years, that once you reach the merits… once a Federal district court or a Federal court is supposed to reach the merits on Federal habeas corpus, that it’s supposed to apply the same rules that it would apply on direct appeal.

William H. Rehnquist:

Well, that’s concededly true on questions of law, but I don’t think it’s necessarily true on other questions.

Allen E. Shoenberger:

Well, in one sense there are very few questions of law, for example, now, that are cognizable in habeas corpus at least in terms of new cases since the Teague rule establishes that we have to simply take the case of law, applying existing legal rules, so in effect you’d be closing the doors to virtually all Federal habeas corpus over State courts, were you to adopt a rule that does what you’re suggesting.

William H. Rehnquist:

Oh, I don’t see that… I don’t see that at all, because all we’re talking about is a harmless error determination.

Allen E. Shoenberger:

But harmless error is… the harmless error determination speaks to virtually all constitutional violations except for those few categories where it’s an automatic reversal rule, so it doesn’t just speak to Doyle violations, it speaks to Miranda violations, to Sixth Amendment… other kinds of Sixth Amendment or Seventh Amendment violations.

There’s no way of saying that Doyle is in any way unique in terms of these kinds of things.

William H. Rehnquist:

Oh, I agree with you on that.

Allen E. Shoenberger:

Okay.

Allen E. Shoenberger:

I’d like to save some time for rebuttal.

William H. Rehnquist:

Very well, Mr. Shoenberger.

Mr. Wellman, or Ms. Wellman… pardon.

Sally L. Wellman:

Thank you, Mr. Chief Justice, and may it please the Court:

If I may try to clarify a factual point first, at pages 7 through 9 of our brief with appropriate cites to the transcript and the joint appendix, we do quote each of the remarks that were found by the Wisconsin Supreme Court, the district court, and the Seventh Circuit, to be Doyle violations.

There was some ambiguity initially as to whether the petitioner had been given his Miranda rights at any earlier point in time, the ambiguity earlier than the initial appearance.

The ambiguity came about only because everyone involved in his arrest and immediate proceedings thereafter did not testify, but there never was any testimony that he did receive his Miranda warnings before his initial appearance.

The Wisconsin Supreme Court therefore proceeded on the assumption that he did not receive his Miranda rights until his initial appearance.

That is a factual finding binding on the Federal courts which has not been challenged by petitioner.

John Paul Stevens:

May I ask, was the questioning by Officer Papke before or after the initial appearance?

Sally L. Wellman:

Before.

John Paul Stevens:

Well, the district court thought that questioning was post-Miranda warnings.

Judge Crabb thought that was–

Sally L. Wellman:

Yes, we think she is wrong on that one.

John Paul Stevens:

–I see.

Sally L. Wellman:

The others, the only violation, however… and I would like to make this clear… is the prosecutor at no time singled out the post-Miranda silence.

The only thing he did wrong was speak a little too broadly, so that in asking questions or commenting, he used words like, isn’t this the first time you ever, and isn’t it true that you never, and those words got him into trouble because they encompassed both the pre-Miranda and the post-Miranda, and in this case, of course, there was a ton of pre-Miranda silence and conduct that was very, very probative of guilt, such as him fleeing the scene and stealing the victim’s car, and heading for the State line, instead of calling for help, which one would expect if it had been an accidental shooting.

The issue in this case as we see it is what standard should be applied to use to determine when a Federal habeas court should release a State prisoner on habeas?

Obviously, Federal habeas review of final State court convictions is a statutory remedy, so we look first to the statute itself to see if it tells us when relief should be granted.

It does not.

The statute 2254 does not say anything about relief per se, or harmless error, or error per se.

The only guidance we find is the terminology in 224 that a Federal court may not entertain a petition unless there is an allegation that the person is in custody in violation of the Constitution, and the language in 2243 that the habeas court shall dispose of the matter as law and justice require.

Because the statute has not defined when relief is required, this Court must do so, and it is not legislating for this Court to do so.

This Court has an obligation… when there is a broad grant of jurisdiction and a broad legislative grant that does not fill in the details and the definitions this court has an obligation to do so, and it does so, of course, by looking at the purpose underlying the statute.

It’s very clear, of course, that in habeas the purpose is twofold.

There are two constitutional principles at issue.

The first, of course, is Federalism, and the second is the duty to protect Federal constitutional rights.

This Court has repeatedly recognized in Stone v. Powell and a number of other cases two primary points about Federalism.

The first is, of course, that the States are sovereigns.

They have the duty and responsibility and power to define and punish crime, and they may do so without Federal interference unless, of course, by doing so they violate Federal constitutional rights.

Sally L. Wellman:

The Court has also repeatedly and properly recognized that it is the State courts that are the primary enforces and protectors of Federal constitutional rights, and they are every bit as able, capable, and competent as the Federal courts to do so, so it’s clear–

Harry A. Blackmun:

You find no exception to that.

Sally L. Wellman:

–That’s right, no exception.

Harry A. Blackmun:

Anywhere.

Sally L. Wellman:

Not that I am aware of.

It’s clear to us, then, that what habeas is is an equitable, special, extraordinary remedy to step in in those few cases where in spite of their duty and their best efforts to do so, the State courts have failed to adequately protect Federal constitutional rights.

We do not believe that the Chapman harmless error standard is a proper measure of when that extraordinary step should be taken.

Chapman is perfectly appropriate on direct review to say that the State shall not profit from its error unless it can show there was no reasonable possibility that the error contributed to the verdict, but that is a very high standard.

It’s just short of automatic reversal, and on direct review it serves many functions.

It serves an educative function, a deterrent function.

You set a very high risk of violating the Constitution, and that encourages State prosecutors and police and judges to take care in imposing and following the constitutional rules, but that same high level is not necessary on the collateral review of habeas.

Anthony M. Kennedy:

You give no effect, then, to 28 U.S.C. 2111 with reference to direct review in the Federal courts.

Sally L. Wellman:

No, we do not.

We have not seen that as implicated in this case.

Anthony M. Kennedy:

So the Chapman decision is correct, in your view, and the Federal statute under… that I’ve cited is irrelevant to that determination and to that rule.

Sally L. Wellman:

I have not thought about that as a separate matter.

We have not seen it as having any implication as to Federal review of a State conviction, no, on either direct review to this Court or on Federal habeas, but as I say, I’ve not considered that so I may be missing something.

David H. Souter:

I’m not sure that I understand your argument that there isn’t much of an implication for the deterrent function of habeas or for the need for such a high standard, given the deterrent function of habeas.

The fact is, most State convictions are not reviewed on direct review at all.

We don’t grant many cert petitions out of State supreme courts, so the fact is, whatever effect Federal jurisdiction generally has on State practice is an effect that is a function of habeas.

Why, therefore, shouldn’t the standard be just as high as it would be on direct review, otherwise we’re going to have a much lesser deterrent function.

Sally L. Wellman:

Because this Court has mandated through Chapman that the State courts themselves shall apply the harmless error beyond a reasonable doubt standard.

David H. Souter:

Well, and the Constitution of the United States mandates that the State court shall apply the Constitution of the United States, but we don’t therefore necessarily adopt a lesser standard on collateral review in determining what’s a constitutional violation.

Sally L. Wellman:

I think that you look at the costs and balances between whether that additional benefit of applying the highest… or nearly the highest standard of review, gives any added benefit on collateral review as opposed to, of course, the obvious significant… more costs on collateral review, and what we’re saying is we don’t think it’s realistic to think that State prosecutors or judges or police are going to be willing to take a risk that might overturn a conviction when Chapman is applied in their State courts just because they know that on Federal habeas a lesser standard will be… a less stringent standard will be applied.

So we think that it does provide adequate deterrence to enforce the rules in the first place on the State courts by mandating that they apply the beyond-a-reasonable-doubt standard themselves on review, and then to have habeas review as a back-up, but you don’t need such a high standard of… of review.

David H. Souter:

And the fact that our direct review is necessarily as limited as it is really, in your judgment, is nonetheless a sufficient interrorum deterrence mechanism.

Sally L. Wellman:

I think so, in combination with the other factors, which of course is primarily that Chapman is applied on direct review.

Byron R. White:

Well, I suppose… we denied cert in this case, did we?

Sally L. Wellman:

He did not seek cert in this case.

Byron R. White:

I see.

Sally L. Wellman:

We believe that habeas should be limited to those situations in which it can be said that the petitioner has been deprived of fundamental fairness, that somehow the error is not just error that we cannot prove harmless beyond a reasonable doubt, but that it is error that so infected the entire trial that the conviction itself is a denial of due process.

One way to state that is the way that Justice Stevens stated in Miller v. Greer and Rose v. Lundy, which is that errors that are so fundamental that they infect the validity of the underlying judgment itself should be granted relief.

We think that that is an appropriate view, but it does not give a measuring stick, it doesn’t tell lower courts how to figure out when we have reached that level, and what we are suggesting is that any of the three tests that we have proposed, either Kotteakos, which the 7th Circuit used, or our Jackson v. Virginia formulation, or our Strickland formulation, would be an adequate test.

It would be sufficient to identify those cases where the conviction and the resulting custody have truly deprived the defendant of fundamental fairness as opposed to those constitutional trial errors that we simply cannot prove are harmless beyond a reasonable doubt.

Byron R. White:

There are going to be very, very few convictions, State convictions that would be upset on Federal habeas under the Jackson standard.

Sally L. Wellman:

That’s true.

That is–

Byron R. White:

We know that from experience.

Sally L. Wellman:

–That is the least protective.

We think it is still consistent with habeas because what we’re saying is, if you can take the infected or bad material out and you still have sufficient proof beyond a reasonable doubt, you cannot say that the State really depended on that… that they could not have a conviction without that evidence, but we agree that either Kotteakos or Strickland would serve the same purpose and yet be more protective or more… give more credence to the risk that even though yes, the State could prove its case without that evidence, that evidence might have influenced the verdict.

So all of these really are sort of on a continuum of looking at what would have happened if this error hadn’t occurred, and how much risk of error are we willing to take?

The automatic reversal rule–

Byron R. White:

So what… on direct review, suppose we reverse a case and find, contrary to the State court, that this was not harmless error, and so there’s going to be a new trial.

What is our judgment based upon, that there’s been a violation of the Constitution?

Sally L. Wellman:

–And that the State has not been able to prove that error harmless beyond a reasonable doubt.

Byron R. White:

Well, I know, but that’s a… it’s a constitutional rule, then.

Sally L. Wellman:

Well, it’s a constitutional rule in the sense that… I don’t think that in Chapman this Court would have had to adopt that test as the test for harmless error.

It has also a deterrent purpose and an educative purpose.

I mean, I think courts have chosen a lesser test in Chapman as well, or have chosen no test and said no, as Justice Harlan suggested, we will just look to see whether whatever harmless error test the State is imposing is consistent with due process.

John Paul Stevens:

May I just ask you to comment on this thought–

Sally L. Wellman:

Sure.

John Paul Stevens:

–The question of the difference between the harmless… the Chapman standard and the Kotteakos standard.

The intermediate court in Wisconsin apparently applied… has found prejudice, which would have found a violation of Kotteakos, unanimously.

The Wisconsin Supreme Court unanimously found harmless error under the Chapman standard, and the district… I wonder if the standard made any difference to any court.

Sally L. Wellman:

Well, I think it does, because I think you have to remember that the Wisconsin Court of Appeals also found all the pre-Miranda references to be error.

John Paul Stevens:

That’s right, but they did not apply the Chapman standard.

They said there was prejudice, as I… if I read the opinion correctly.

Sally L. Wellman:

But I believe in doing that they meant to include the… I mean, they thought they were applying the Chapman standard.

John Paul Stevens:

Well, I don’t know how you can tell that from the opinion.

They don’t cite it.

John Paul Stevens:

They don’t cite any Federal cases on that point.

Well, what did the… did the… what did the 7th Circuit… it used the Kotteakos standard.

Sally L. Wellman:

That’s right, the 7th Circuit used Kotteakos.

Byron R. White:

Did it apply it, then?

Sally L. Wellman:

Yes, it did.

Byron R. White:

Well, it made a difference to them.

Sally L. Wellman:

Yes, it–

John Paul Stevens:

Well, that’s what Judge Easterbrook said, but then he also said he didn’t want to trudge through the record, so I’m not sure he thought he should trudge through the record under either standard.

Well, you have to do it under both.

Sally L. Wellman:

–Right.

You have to–

John Paul Stevens:

But he seems to think under one you trudge through the record and under the other you don’t.

Sally L. Wellman:

–Well–

John Paul Stevens:

Do you think that’s right?

Sally L. Wellman:

–No, I don’t think that’s right.

I–

John Paul Stevens:

Do you think… they did have the duty to look at the entire record.

Sally L. Wellman:

–Yes, and I think he did so.

John Paul Stevens:

Even under Kotteakos.

Sally L. Wellman:

Yes, definitely.

Byron R. White:

And at least… and when it got to the district court, the district court had to act contrary to what it did originally.

It dismissed the petition for cert… or, the petition for habeas, Federal habeas.

Sally L. Wellman:

I’m sorry, I–

Byron R. White:

I mean, consistent with the court of appeals ruling–

Sally L. Wellman:

–Oh.

Byron R. White:

–It dismissed the petition for… denied Federal habeas.

Sally L. Wellman:

Right.

I think that the district court erred in how it applied Chapman as well, because it seemed to think that the only way you could meet Chapman is to find overwhelming evidence, and we don’t think that is correct, and I think that’s what accounts for the difference between the Wisconsin Supreme Court and the district court, so I don’t think that it necessarily means because we do have this history of different courts, each overturning the one before it, that it means that this was that close of a case.

I think that would be not necessarily the inference to draw.

I think what we are searching for in this case is a standard that will permit us to limit habeas relief to those cases where law and justice truly require that it be granted, and the way to measure that is to look at a broad sense at causality.

Sally L. Wellman:

Did the error, did the constitutional violation cause this person to be in custody, and at what level or degree or likelihood is a fair measure of that on collateral relief, where the petitioner has had full review, appellate review in the States below, and the petition can be brought at any time, therefore, of course, putting the State to great disadvantage in terms of being able to reprove the case if that should become necessary or to prove underlying errors that may be raised.

We think that the Constitution and the criminal justice system of course protect other values besides guilt or innocence, and we’re not suggesting that any of these three tests would protect only guilt or innocence.

The other values of the Constitution, such as Doyle itself, it protects something that is not just a question of guilt or innocence, but a question of fairness.

We think the Government has to treat the petitioner fairly.

You can’t tell him one thing and then double-cross him at trial, but that value is fully served by applying the Doyle rule in the first place and secondly requiring the State courts to apply the Chapman beyond-a-reasonable-doubt test.

Byron R. White:

Do you think that silence after Miranda rulings has the same significance as before the warnings?

Is it as probative of guilt?

Sally L. Wellman:

I think the probativeness is the same.

I think it’s really the fairness that is at issue.

To the extent that a person… it’s less probative because the person might be relying on the Miranda rights as opposed to making up this cockamamie story.

Byron R. White:

So it is less probative.

Sally L. Wellman:

No, I think two inferences can be drawn.

Byron R. White:

It only might… it only might be less probative.

Sally L. Wellman:

It might be, just like any other evidence that is ambiguous, or from which two inferences might be seen as less probative, but that isn’t the reason that we don’t let it in, I don’t think.

That’s why this Court was able to say that pre-Miranda silence can come in, that it’s not just the probative value but it’s the fairness element.

Antonin Scalia:

I guess if we told him you have a right to make a confession, and then his making of a confession, if we used the confession against him, that would be unfair too, right?

Sally L. Wellman:

I think so.

I think so, so I don’t think that it hinges on Miranda per se.

I mean, I think in a situation where, for example, Miranda was not required because the person was not in custody, if nonetheless he was told he had the right to remain silent and then that silence was used against him, I think that could be held to be unfair, so in that sense we do depart from Judge Easterbrook.

We agree with him that a different standard should be used on habeas because of all the differences between direct review and collateral review, but we don’t focus it on it being because Doyle is prophylactic, because there are… it is prophylactic to some extent, but it is not the same kind of prophylactic as the Fourth and Fifth Amendment exclusionary rules are.

We believe that either the Kotteakos substantial and injurious effect on the verdict test or our Jackson formulation or a Strickland type of formulation that would require a showing of a reasonable probability that but for the error the result would have been different are all adequate to protect the defendant’s Federal constitutional rights and to pick up those cases where the State courts, even though they had a duty to protect those rights, and we can assume strove to do so, nonetheless failed.

We also believe that any of these three tests would be sufficient to protect the interest of Federalism and the interest of the State as a sovereign and its interest in its final State judgments.

Therefore, we would ask this Court to hold that… thank you.

William H. Rehnquist:

Thank you, Ms. Wellman.

General Barr, we’ll hear now from you.

William P. Barr:

Mr. Chief Justice, and may it please the Court:

The position of the United States is that the harmless error standard that should apply on collateral review of constitutional errors is the traditional Kotteakos standard.

The Kotteakos rule is the traditional formulation of the harmless error doctrine derived from the common law approach which later found its way into the Federal statutes of 1919 and 1949 and now the Federal rules.

Chapman is the exception to that traditional doctrine, and in Chapman this Court formed a stricter, a ratcheted up harmless error standard, and in that context, on direct review of constitutional errors, that ratcheted-up standard serves a deterrent function promoting rigorous adherence to constitutional norms at the trial phase.

Now, we submit that this should not be carried over to collateral review, the Chapman standard, and this Court has never directly addressed this issue.

William P. Barr:

We think there are three reasons for this.

First, Kotteakos fully effectuates the purposes of habeas corpus.

Habeas is not supposed to be a full replay of direct review, and there should be no presumption that the rules and the standards that apply on direct review necessarily carry over to habeas corpus review, and for that reason this Court has repeatedly recognized that the rules that apply on habeas should be tailored to the purposes of habeas, and may be different than the rules and the standards that apply in direct review.

Now, the core purpose of habeas is to protect against the kind of serious constitutional defect that gives rise to a substantial risk that an innocent person has been convicted, and Kotteakos is fully sufficient to meet that standard.

After all, it is the rule that we use on direct review of nonconstitutional errors which can be more prejudiced than many constitutional errors such as 404(b) violation, and in that context we rely upon the Kotteakos standard, whether there’s been a substantial influence in determining the jury’s verdict, to assure us–

John Paul Stevens:

General Barr, would you go so far as to say that’s true in those cases that have been automatic reversals such as total denial of counsel, or something like that?

William P. Barr:

–No, Justice Stevens.

What I would say is, there are essentially three boxes.

The first issue is, is a claim reviewable at all under habeas corpus, or should it receive Stone v. Powell treatment, but if it’s a cognizable claim, the next issue is, is it automatically reversible, and there may be some debates about that test, but I don’t think Doyle meets that test, and that would put, then, Doyle in the subject-to-harmless-error analysis box.

So if we can rely on Kotteakos on the direct review of nonconstitutional errors to sort out the reliability of the verdict, I think we can rely upon it in habeas corpus context.

The second reason, I think, that we shouldn’t just carry over the Chapman standard, is that the benefits of Chapman… and this goes to your comments, Justice Souter… the benefits of Chapman are largely achieved on direct review.

It’s the presence–

David H. Souter:

How, as a matter of fact, do we know that?

You say that.

How do we know that?

William P. Barr:

–That… the… if it doesn’t perform a deterrent function on direct review, then what’s the purpose–

David H. Souter:

Well, I’m sure it performs some, but again, given the fact that to the extent that a Federal… that the guarantee of the Federal system is an element at some point in enforcing the Chapman standard, and given the fact that direct review by a Federal court is extraordinarily limited, how do we know, as matter of fact, that a substantial deterrent function is being performed by direct review alone, and it will not be sacrificed if we adopt a lesser standard on collateral?

William P. Barr:

–That question boils down to this: will a prosecutor facing Chapman on direct review through the State system and potentially on cert in the Federal system really play fast and loose with the Constitution because he may be facing down the road on collateral review Kotteakos rather than Chapman?

David H. Souter:

Well, the… you can focus it on the prosecutor only if you, I suppose, limit your argument in this case to the application of the Kotteakos standard in a Doyle case and not to habeas generally.

But even aside from that, it isn’t merely the prosecutor whom we are supposedly influencing, it’s the State courts, and I think one of the questions about deterrence that has to be asked is whether the State courts will simply as a fact feel as great a need to be on their toes if there is a lesser standard for the vast majority of State cases that will be reviewed later by a Federal court, and I don’t see anything self-evident about your statement that they will be.

William P. Barr:

Well, the Kotteakos standard has worked for nonconstitutional errors, which can be just as prejudicial as a constitutional error, and it’s also essentially the standard that we use… the Kotteakos standard, whether there’s a substantial probability of prejudice… is the standard we use on Donnelly v. DeChristoforo potential due process violations in the State system, it’s the standard we use for ineffective assistance of counsel in the Strickland case, although the burden is shifted by Strickland and DeChristoforo.

So it hasn’t… there’s no evidence that has led to a breakdown of the State criminal justice system or ignoring or flouting constitutional standards.

While the benefit, we maintain, of carrying over the Chapman rule to collateral review is slight, if any, from a deterrence standpoint, we believe the costs are high.

Overturning final criminal convictions for errors that did not substantially influence the outcome of the case derogate from the truth-seeking function of the criminal justice system, undermine the Federalism interest and the interest in finality, and breeds disrespect for the criminal justice system as a whole.

Now, there is no constitutional or statutory command to select either the Chapman standard or the Kotteakos standard under habeas corpus.

We see no constitutional reason for doing it, and there’s certainly nothing in the statute dictating the selection of one or the other of those standards, and therefore it’s a prudential matter as to how much respect is going to be given to the Federalism interest, the interest in finality, and so it is a prudential decision, and we maintain that the best balance is struck by selection of the Kotteakos standard.

Now, to answer a question that Justice O’Connor raised at the outset, as you know, we argued in Wright v. West that there should be deference given to mixed questions of law and fact.

We believe that this kind of judgment, the determination of harmlessness, is a mixed question of law and fact akin to a sufficiency determination, and as you know, we argued there that that should be given deference.

The Court did not reach that issue, but if deference were given here, then we belief the judgment below should be affirmed.

Byron R. White:

Well, the next question, I suppose, is that there are historical facts to be considered and then there’s a question of law to be drawn from those facts–

William P. Barr:

Correct, Justice White.

Byron R. White:

–And the district court purported to do just that.

It gave complete deference to the historical fact findings of the State court–

William P. Barr:

That’s correct.

Byron R. White:

–And then said that the ultimate judgment of harmlessness is a question of law.

Do you think it is?

William P. Barr:

No.

I think it’s a mixed question of law and fact.

Byron R. White:

Well… mixed question.

What do you mean by a mixed question?

Isn’t it just the inferences that are to be drawn from the facts?

William P. Barr:

It’s a determination–

Byron R. White:

The judgment of harmlessness, isn’t it a question of law?

It’s the inferences to be drawn from the facts, I suppose.

There’s no reason to think that’s a question of law, I would think.

You can have something that’s genuinely a mixture of legal issues and factual issues, and those are what are called mixed questions of law and fact.

Well, is it any difference in deciding whether a confession is voluntary or involuntary?

William P. Barr:

–Well, here, Justice White, the record is before the Court, and the Court has to make a judgment–

Byron R. White:

Right.

William P. Barr:

–as to whether or not that jury was influenced by the error in the case and whether it had some kind of determining influence on the verdict, and it seems to me that that’s a mixed question.

John Paul Stevens:

May I ask, General Barr, in your… in the notion of deference, do you think it is part of the duty of the Federal court to read the entire record under either standard, Kotteakos or Chapman?

William P. Barr:

Yes, I do, Justice Stevens.

I think the Kotteakos case discusses that extensively, the obligations of the court.

The… so it’s our position that if a claim is cognizable under habeas, and if it’s subject to harmless error review, then one standard should apply, and that is the Kotteakos standard.

Now, in this case, as I indicated, there doesn’t seem to be a plausible argument that this is the kind of fundamental or structural defect that puts it into the automatic reversal category.

David H. Souter:

May I ask you just one final question?

Is there any reason to, if we accept your view, to confine the rule you want to Doyle cases?

William P. Barr:

There is an argument for, obviously, limiting it to the Doyle case or just deciding the Doyle case here, based on what the purpose of the Doyle rule is.

Is it a truth-seeking rule, or is it not a truth-seeking rule?

Does the violation of the rule, therefore, undermine our confidence directly in the reliability of the verdict?

William P. Barr:

And if those considerations are taken into account, then a more limited holding could be issued.

David H. Souter:

But there’s nothing essential in your argument that confines it to Doyle, really.

William P. Barr:

No.

I don’t think that the truth-seeking or the nontruth-seeking distinction for harmless error analysis should generate two different harmless error rules.

I think that distinction should be taken into account in determining whether or not the rule is automatically… the error is automatically reversible, or whether it’s subject to harmless error at all.

William H. Rehnquist:

Thank you, General Barr.

Mr. Shoenberger, you have 3 minutes remaining.

Allen E. Shoenberger:

Thank you, Chief Justice.

Let me state first that the rule in Chapman is not based upon any particular statute.

It was an attempt by this Court to devise what it described in the decision itself as a Federal right or a Federal rule, and it appears it may be a constitutional rule because it’s so closely allied with the enforcement of constitutional rights, but it certainly… Chapman certainly was not interpreting the Federal habeas corpus statutes itself, because the Federal habeas corpus statutes weren’t involved.

That came later, although a case was argued 4 days before Chapman was announced, the first case in which Justice Black indicated that Chapman ought to apply in Federal habeas corpus, but it’s at least until 1970 that the first, if you will, majority of the Court in collective different decisions indicated that Chapman applied on Federal habeas corpus, as far as I can find.

The standards in Jackson are completely inappropriate for this kind of a case.

Jackson assumes that the jury fact-finding is untainted by any error and then looks at the result of that fact-finding to see whether that fact-finding is so out of the realm of permissibility that it has to be overturned.

We’re talking here about a case where constitutional error clearly occurred, and we submit it had a direct truth-finding relationship to the jury fact-finding, so you start off with a tainted jury determination.

How much that taint… I almost heard, and I don’t think I heard correctly, but I almost heard that the question was a question of percentage.

Did it have a 5 percent or 20 percent difference in terms of the verdict?

We can’t tell.

The jury is in the jury room.

Only it knows.

Indeed, only an individual jury member can know how much a particular factual inference or fact had in relationship to that jury member’s determination of what the vote should be.

Byron R. White:

Do you think it would be error if the district judge on Federal habeas said well, this is an awfully close case, I probably would have come out differently in deciding harmlessness than the State Supreme Court did, but it’s a close case and I’ll just go along with the State Supreme Court?

Allen E. Shoenberger:

I think that’s error.

I think the duty on the–

Byron R. White:

Even though the State courts may know an awful lot more about what might affect the jury in their State than some Federal panel, or some district judge?

Allen E. Shoenberger:

–They may know more, or they may know less in a particular case.

May I finish?

William H. Rehnquist:

You can finish answering.

Allen E. Shoenberger:

The particular application is something that we can’t tell about, and that’s the essence of it.

The duty is the Federal court’s duty to make what it… is the right decision under the existing Federal law, which includes the Chapman standard in this case.

William H. Rehnquist:

Thank you, Mr. Shoenberger.

William H. Rehnquist:

The case is submitted.

The honorable court is now adjourned until tomorrow at ten o’clock.