United States v. Hasting

PETITIONER:United States
LOCATION:Dr. Simopoulos’ Clinic

DOCKET NO.: 81-1463
DECIDED BY: Burger Court (1981-1986)
LOWER COURT: United States Court of Appeals for the Seventh Circuit

CITATION: 461 US 499 (1983)
ARGUED: Dec 07, 1982
DECIDED: May 23, 1983

John F. DePue – on behalf of the Petitioner
Paul V. Esposito – on behalf of the Respondent

Facts of the case


Audio Transcription for Oral Argument – December 07, 1982 in United States v. Hasting

Warren E. Burger:

We will hear arguments first this morning in the United States against Hasting.

John F. DePue:

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

Warren E. Burger:

Mr. DePue.

John F. DePue:

This case presents the question whether a federal appellate court may summarily reverse a conviction because of the prosecutor’s improper comment on the defendant’s failure to present evidence, without first making an assessment for prejudice under the constitutional harmless error standard.

Briefly summarized, the government’s evidence showed that during the early morning of October 11th, 1979, three teenaged girls and a male companion were driving in the vicinity of East St. Louis, Illinois, when their car was forced off the road by five men driving a turquoise Cadillac.

After both cars had come to a halt, the girls were pulled from their car.

One of them was raped.

The men then forced them into the Cadillac, drove them across the interstate boundary into St. Louis, Missouri, where they repeated their sexual assaults at three different locations.

Following the girls’ release, one of them led the police to the apartment where she had been taken by her abductors.

There, the officers arrested respondent Stewart and found items of clothing belonging to the victims.

They also located the turquoise Cadillac, learned that it was registered to respondent Williams, and found Williams’ latent prints on the car from which the girls had been taken.

These developments also resulted in the apprehension of the remaining respondents, each of whom was identified by at least one of the victims during a lineup the following morning.

At their ensuing trial for rape and a Mann Act violation, each of the victims again identified each respondent as a participant in the abductions.

The defense did not put on the defendants themselves as witnesses.

Instead, their defense was calculated to discredit the girls, their credibility and the accuracy of their identifications by reference to the girls’ activities just before and just after the abductions took place.

For example, they elicited testimony from the girls during cross examination that they had been out drinking, in violation of parentally-imposed curfews and that they’d gotten lost on the way home from a tavern.

And that just after the abductions, their initial identifications and descriptions of respondents were not very accurate and were confused.

William H. Rehnquist:

Well, that’s a fairly standard tactic, isn’t it?

How does that bear on the issue here?

John F. DePue:

It bears, Justice Rehnquist, on the argument that we will see that the prosecutor presented subsequently, because we maintain it was a direct reference to what the defendants were actually doing.

In addition, respondents presented witnesses of their own who testified that just before and just after the alleged abductions, the physical appearances of respondents were not at all correspondent with those as described by the victims.

The prosecutor in his summation first discussed the posture of the government’s case.

He then turned to respondents’ case and he made the following statement, which appears at page 21 of the Joint Appendix.

And I’m going to quote it verbatim from the record because of its centrality to these proceedings.

He said,

“Let’s look at the evidence the defendant put on here for you so that we can put that in perspective. “

“I’m going to tell you what the defendant did not do. “

“Defendants on cross examination and– “

When respondents objected on the ground that they didn’t have to testify at all, the prosecutor continued,

“That’s right, they don’t. “

John F. DePue:

“They don’t have to. “

“But if they do put on a case, the government can comment on it. “

“The defendants at no time ever challenged any of the rapes, whether or not that occurred, any of the sodomies. “

“They didn’t challenge the kidnapping, the fact that the girls were in East St. Louis and were taken across to St. Louis. “

“They never challenged the transportation of the victims from East St. Louis, Illinois to St. Louis, Missouri, and they never challenged the location or whereabouts of the defendants at all relevant times. “

“They want you to focus your attention on all the events that were before all of the crucial events of that evening. “

“They want to pull your focus away from the beginning of the incident in East St. Louis after they were bumped, and then the proceeding events. “

“They want to focus to the events prior to that. “

Following this, the defense moved for a mistrial on Fifth Amendment grounds.

The trial judge denied the motion, commenting that to his view, this was simply a commentary of the posture of the cases and didn’t implicate Fifth Amendment issues at all.

The Seventh Circuit, however, disagreed, characterizing the remarks as a pointed allusion to respondents’ failure to take the stand, and held that they constituted a clear violation of this Court’s decision in Griffin versus California.

Although the court acknowledged that the crimes were heinous and that the evidence of guilt was clear, it refused to assess for prejudice under the constitutional harmless error standard.

Instead, after alluding to the fact that prosecutors in the Seventh Circuit had continuously disregarded admonitions not to make arguments that implicated Fifth Amendment rights, they stated that to their view, application of the constitutional harmless error standard would impermissibly compromise the defendants’ Fifth Amendment rights.

It’s the government’s position that this reasoning is in direct violation of this Court’s decision in Chapman versus California, violates the federal harmless error statute and contravenes broader rules regarding the formulation of remedies in criminal cases.

William H. Rehnquist:

Mr. DePue, are you in a position to state why the government didn’t raise the question here of whether or not this comment was, in fact, a violation of Griffin?

If you’re not, feel free to say you’re not.

John F. DePue:

Justice Rehnquist, we didn’t seek to address… we didn’t feel we needed to address that in our brief, although we do not concede that it was, in fact, a violation of Griffin.

We didn’t feel that that was a cert worthy issue, and therefore, we didn’t raise it.

Sandra Day O’Connor:

Well, we have to assume that it was a Griffin type error for purposes of resolving this case.

John F. DePue:

Yes, Justice O’Connor, you certainly do.


Well, we don’t have to.

John F. DePue:

–No, Your Honor.


The way you phrased the question presented, you assume it is.

You say whether the closing argument that constitutes improper comment upon the defendant’s failure to testify justifies reversal.

So you are assuming–

John F. DePue:

Yes, Justice Stevens, we assume that it is.

We, however, do not concede that it is.

–Well, are there other ways than testifying that they could have challenged this?

In other words, by establishing that they were in New York City attending the World’s Fair at the time?

John F. DePue:

Oh, certainly, Chief Justice Burger.

And as a matter of fact, they presented witnesses of their own who said that they saw these people shortly after midnight and approximately at 6:00 o’clock in the morning.

And in addition to this, one of respondents’ mothers, respondent Stevens’ mother actually saw Stevens and the girls and testified that to her observation, everything that was going on was consensual.

So in fact, these were challenged to a certain extent, and there were people who could have testified to the contrary.

Now, of course, in Chapman versus California, a state prosecutor, relying on a provision of the state constitution, directly and repeatedly commented on the defendant’s failure to testify and the inferences of guilt that could be extracted from that.

The trial judge followed suit in his instructions to the jury.

Following that trial, however, this Court in Griffin versus California held that the provision of the California Constitution upon which he had relied was unconstitutional.

When Chapman’s own case came before this Court, Chapman held, argued that all errors of a constitutional magnitude, and Griffin errors in particular, should be deemed reversal per se, and thus not subject to review under the harmless error standard.

This Court, however, squarely rejected that proposition.

Mr. DePue, if this Court had accepted the argument in Chapman against California, I assume they pretty much would have emptied the California jails, wouldn’t they, because that had been the standard practice in California before Griffin was decided.

John F. DePue:

Yes, Justice Stevens.


John F. DePue:

Now, when the case came before this Court, the Court rejected that argument, and held that Griffin violations, just like any other non-constitutional violations, could be deemed harmless and thus not a basis for reversal.

It reversed Chapman’s conviction, but only after first satisfying itself that given the less than overwhelming posture of the evidence, it could not be satisfied beyond a reasonable doubt that the prosecutor’s pointed and repeated allusions to the defendant’s failure to testify was not harmless.

In the wake of Chapman, and prior to the decision of the court below, this Court and the courts of appeals have virtually without exception held that the constitutional harmless error standard applies to violations of Griffin.

They have therefore reversed convictions only when after canvassing the evidence in its totality, looking at the argument in its proper context, they couldn’t be satisfied that it didn’t contribute to the conviction.

This approach, we maintain, is not only mandated by the Chapman decision itself, but by broader principles of revenue recently articulated by this Court.

In United States versus Morrison, the Third Circuit dismissed an indictment because a government agent had made improper remarks to the defense counsel about the competency of her own lawyer, to the defendant about the competency of her own lawyer.

This Court, however, reversed, finding that dismissal was unwarranted absent any evidence that the alleged Sixth Amendment violation could have prejudiced the defendant and the proceedings in any way.

In doing so, the Court admonished the courts of appeals that after identifying a constitutional error, the proper approach was to assess the damage, and to tailor a remedy that was commensurate to the injury received as a result of that violation, and then a broader approach would impermissibly infringe upon other legitimate interests.

In the context of a Chapman violation… of a Griffin violation, this means that once the court has determined that the violation occurred, it has got to look at the amount of prejudice that resulted in determining whether other events at trial offset it.

It cannot summarily reverse.

Of course, harmless error principles are not simply a matter of judicial self-restraint.

The federal harmless error statute and Rule 52(a) of the federal rules of criminal procedure both expressly prohibit a court of appeals from reversing a conviction on the basis of errors or defects that do not substantially affect the nature of the proceedings.

Respondent argues, however, that the harmless error statute and Rule 52(a) are inapplicable to this case because the error involved is of a constitutional magnitude.

This Court has, however, squarely rejected this reasoning in Chapman itself.

There, the Court accepted the policy embodied by the federal harmless error statute which was to require an assessment of prejudice in all cases without regard to whether the error was of a constitutional or non-constitutional magnitude.

Respondent’s principal contention, however, is that the Court in the exercise of its supervisory power was free to disregard the Chapman standard and fashion a remedy of its own as a prophylactic measure because of the alleged repeated violations of the Griffin standard within the circuit.

In the first place, nothing in the decision of that court even suggests that it was in fact relying on its supervisory power.

John F. DePue:

The court’s language that it felt the application of a harmless error standard would impermissibly compromise the defendant’s Fifth Amendment rights suggests that it simply didn’t believe that the harmless error standard was sufficient, a sufficient mechanism to protect the defendants against Fifth Amendment violations.

But be that as it may, there are three reasons why the Seventh Circuit could not adopt a reversal per se rule in the exercise of its supervisory power.

In the first place, as Justice Frankfurter stated in Sherman versus United States long ago, the supervisory power entails the formulation of standards for the administration of criminal justice when Congress has not specifically legislated to that end.

Where, however, Congress in the exercise of its constitutional power to regulate the courts has expressly articulated a standard that is applicable to the issue, here, the federal harmless error statute and Rule 52(a) of the federal rules of criminal procedure, the courts of appeals are not free to disregard that standard and fashion a remedy of their own in the exercise of their supervisory power.

What about this Court?

John F. DePue:

Pardon, Justice–

What about this Court?

John F. DePue:

–This Court can’t do so either, Justice Brennan, because the federal harmless–

Any decision of this Court that has said that?

John F. DePue:

–No, Your Honor, but the federal harmless error statute says that.

How do you reconcile that with the Bruno case, which indicated that Section 2111 was concerned basically with etiquette of trials and formalities?

John F. DePue:

Yes, Justice O’Connor.

I have two responses to that.

First, and most importantly, the version of the federal harmless error statute upon which Justice Frankfurter relied in Bruno was amended in 1948.

The language upon which Justice Frankfurter relied was deleted.

That is the word “technical”, the phrase “technical errors”.

As the Ninth Circuit said in Broulette, and as I believe is one of the principal cannons of construction, that where a court deletes critical language, the presumption is, it intends to change the standard.

Second, in Chapman versus California, of course, this Court expressly held that the federal harmless error statute applies to errors of a constitutional magnitude.

Therefore, Bruno has no–

I wonder if that is correct, Mr. DePue.

Was the federal harmless error statute at issue in the California trial that gave rise to the appeal in Chapman against California?

John F. DePue:

–No, Your Honor, it wasn’t.

That was a state case.

So it wasn’t… the federal harmless error statute was not involved at all in that case?

John F. DePue:

No, it wasn’t, Justice–

And it’s a different standard that the court applied than the harmless… the statutory harmless error standard.

You don’t suggest that the Griffin standard is the same standard that was described in Justice Rutledge’s opinion in Kotteakos, do you?

John F. DePue:

–No, I certainly don’t, Justice Stevens.

Which was an interpretation of the federal statute.

John F. DePue:

This Court has made it quite clear that the degree of certitude between the constitutional harmless error standard and the standard for non-constitutional errors is different.

The statute says that you’ve got to be sure that the error does not affect the substantial rights of the parties.

Now, do you suggest that commenting on the failure of the witness to get on the witness stand, failure of the defendant to get on the witness stand does not affect a substantial right?

John F. DePue:

I suggest, Justice Stevens, that there may be some instances, indeed many instances, where indirect allusions to the defendant’s failure to take the stand may be deemed harmless, yes.

They may be deemed harmless, but within the meaning of the statute, would you say that a comment on the defendant’s failure to take the witness stand does not affect his constitutional right not to testify?

John F. DePue:

No, it wouldn’t.

Then the statute simply does not apply.

John F. DePue:

Well, I believe, Justice Stevens, as I said before, that it depends upon the context of the… that it depends on the context of the situation.

That you simply… you simply cannot say that–

You are saying, in some situations the comment on the failure to testify affects the right, constitutional right; in others it does not affect the right?

John F. DePue:

–It always affects the constitutional right.

All right, but then it may well be harmless, but that doesn’t mean the statute doesn’t apply.

John F. DePue:

Of course, the statute… the statute applies, but the statute simply says, you’ve still got to assess the presence.

The standard is simply different.

Second, even where–

Is it your view that, for example, if television covers an event in which a crime, a murder takes place, that the evidence then corroborated by other eye witnesses is so conclusive that no comment by the prosecutor would constitute reversible error?

John F. DePue:

–There may be some cases of that nature, yes, Mr. Chief Justice.

We don’t think this is one of them, but there certainly may, yes.

In other words, the error, conceding that there is an error, would be harmless in view of the overwhelming nature of the evidence.

Is that your view?

John F. DePue:

Yes, Mr. Chief Justice, harmless beyond a reasonable doubt under the constitutional harmless error statute of Fahy versus Connecticut.

Now, even–

Under your view, if the defendant didn’t have a lawyer, and the evidence was just as clear as you hypothesize, could that error ever be harmless?

John F. DePue:

–Absolutely not, because that would pervade the entire nature of the proceedings.

That sort of thing is not subject to any kind of an accurate assessment, but this, we believe, is, because the Court can look at that… look at that remark, look at how pervasive and how direct it is, look at the evidence, look at the judge’s instruction, and all sorts of other things, and determine whether there was adequate damage control done to deem that it was harmless beyond a reasonable doubt.

Now, even where it is otherwise permissible to employ the supervisory power for didactic or prophylactic purposes over agents of the executive branch, this Court has made it clear that such authority should be exercised sparingly, and only in the face of intentional transgressions of legal norms.

In this regard, we believe that this particular case is a most inappropriate vehicle for exercising such authority.

Any didactic benefits that arguably might result from a reversal here will surely be outweighed by the cost to the victims who must again relive their harrowing ordeal at respondent’s hands.

But be that as it may, the universally accepted test for determining whether a Griffin violation has occurred embraces both intentional and unintended reference to the defendant’s failure to testify.

Remarks, although not intended as comments, are later subject to a construction, a reasonable construction that the jury could reasonably infer that they are related to the defendant’s failure to testify.

It is a rare case indeed that a prosecutor in the wake of Griffin and Chapman sets out comment on the defendant’s failure to testify.

John F. DePue:

Most reversals for Chapman and Griffin violations occur where an appellate court with the benefit of hindsight and after looking at the totality of the evidence, the posture of the defense case, the defense theory, the totality of the prosecutor’s remarks, concludes that the prosecutor overstepped an indistinct line between perfectly permissible commentary upon the posture of the defense case and impermissible reference to the defendant’s personal failure to rebut damaging evidence.

It is not surprising, given the highly subjective nature of this assessment, that appellate courts will frequently reach different conclusions on very similar factual situations and arguments.

I believe that this case is a perfectly good example of that.

The court of appeals found a clear Griffin violation here.

The linchpin of respondent’s argument is that this was intentional misconduct by the prosecutor, yet the trial judge, who was able to listen to this evidence, who saw it come in, to listened to the argument, found that there wasn’t a Griffin violation at all, that what the prosecutor was doing was simply commenting upon the strengths and the weaknesses of the case.

Indeed, while it is true that the prosecutor did use the word “defendant”, no objective person could have viewed his argument as a commentary upon the defendant’s personal failure to testify.

Rather, it is perfectly obvious from the context of the prosecutor’s remarks here that they were intended as a commentary upon what the defendant’s lawyers were doing.

He used the word “defendant” just as I am using the word “respondents” today.

You haven’t made a distinction, or if you have I have missed it, between a case where the defendant testifies and one in which he does not take the stand at all.

Is that a significant difference?

John F. DePue:

I don’t think that you would have this type of violation where the defendant testified, Mr. Chief Justice.

This type of argument arises in the context where you have a non-testifying defendant, the defense puts on a case, and the prosecutor during his closing argument says, points out to the weaknesses in the defense case, which is, of course, one of the mainstays of proper argument.

The question arises when he does that, pointing out to the weaknesses of the defense case, whether he transgresses what I call a very indistinct line between talking about weaknesses and talking about the defendants themselves, and this is precisely why I maintain that most of these so-called Griffin violations are totally unintentional, because most arise in the context of talking about the posture of the evidence, and this is precisely what the prosecutor was doing in this particular case.

He was talking about the scope of cross examination, that respondents during cross examination peripherally examined the victims, talked about their misconduct before, talked about the lack of certitude of their identifications afterward, but left totally unscathed their rendition of the critical events.

Mr. DePue, do you know when 28 USC 2111, the present harmless error statute, was amended to have its present form?

John F. DePue:

Yes, Your Honor, 1948, I believe.

Was that the time of the codification?

John F. DePue:

Yes, Mr. Justice Rehnquist.

The precursor to that was rescinded in 1947, and that precursor had the word “technical” in it.

The key language which–

That was the one that the Court construed in Bruno.

John F. DePue:

–Yes, Justice Rehnquist.

Then it was amended in 1948.

The word “technical” was deleted.

Finally, in United States versus Payner, this Court just 18 months ago held that courts of appeals cannot exercise a supervisory power in a manner that is inconsistent with the controlling decisions of this Court.

Payner, of course, also involved an intentional transgression by a government agent, and by contrast with this case.

The decision of the court below is, however, directly contrary to the controlling decision of this Court in Chapman versus California.

It therefore cannot stand.

Thank you.

May I ask one other question?

Is it… it is not the government’s position that this Court should make the determination as to whether there is harmless error, but rather we should direct the court of appeals to?

John F. DePue:

We maintain, Justice Stevens, that you certainly could make that decision, because the evidence is so clear.

However, we acknowledge that this case will have to be remanded for the assessment of other errors.

The court of appeals left undecided a number of errors in an unpublished decision that it issued contemporaneously with this one.

So, the case is going to have to be remanded, and we would certainly not have any strenuous objections if the Court were to remand this issue as well.

And your view is that… what you are asking us to say is that they may not make a harmless error determination… I mean, they must make a harmless error determination; they may never exercise supervisory power on a question like this, even if… even if they… even if we thought they had here, and I understand your position is, they didn’t.

John F. DePue:

Precisely, Justice Stevens.

And this because of the harmless error statute.

John F. DePue:

This because… for three reasons.


But primarily–

John F. DePue:

–The first is the harmless error statute.

–The harmless error statute you argue forecloses the exercise by the courts of appeals of supervisory power.

John F. DePue:

Yes, Justice Brennan, and my authorities for that again is one case cited in our brief, which is United States versus Palermo, and another case that recently came to my attention that was not cited in our brief but of which I have apprised respondents, United States versus National City Lines, which is cited at 334 US 573, and the pertinent language is at Page 589.

Thank you, Mr. Chief Justice.

Warren E. Burger:

Very well.

Mr. Esposito?

Paul V. Esposito:

Mr. Chief Justice, and may it please the Court, this case involves the remedial powers of the federal courts of appeals.

Like Chapman, it involves comments on the failure to testify, and like Chapman, for purposes of review, the error was conceded, but there are significant differences.

In Chapman, at the time the comment was made, the courts had not yet found that that type of comment, at least in the state court setting, was constitutional error.

However, in this case, long before the comment had been made, the courts of appeals had found that this type of comment was in fact constitutional error.

In your view, is there any circumstance in which comment, direct comment by the prosecutor on the failure to testify, on the defendant’s failure to testify could be harmless error?

Paul V. Esposito:


We are not trying to… we are not here today, Your Honor, asking this Court to overrule Chapman in any regard.

We find… we feel that this case presents a significant… this case is significantly different from Chapman.

Chapman involved the question, of course, of whether a Fifth Amendment comment could ever be harmless error, and the Court said it did.

This case questions… this case questions the power of the federal courts of appeals, whether they have the power to refrain from applying Chapman in an appropriate case in order to deter misconduct which stands in violation of repeated decisions by the courts of appeals.

I think it is very important to recognize that last point.

The federal court of appeals in the Seventh Circuit has repeatedly dealt with this problem, the exact problem, indirect comments that a defendant didn’t contradict or deny or refute evidence, and it has repeatedly held the same thing.

Well, is this discussion not to follow Chapman, is pursuant to what, supervisory power?

Paul V. Esposito:

Respondents contend that it would be pursuant–

That isn’t what the court of appeals said.

I can’t tell from its opinion anything about supervisory–

Paul V. Esposito:

–They did not discuss supervisory power, Your Honor.

Respondents would contend that that power… the existence of that power would support their judgment, and therefore it may be a ground for affirmance of the decision, even though the court didn’t discuss it.

What happened here, Your Honor–

–Well, I might… I guess you have to address the government’s position, then, that the harmless error statute forecloses the court of appeals from doing what it did here.

Paul V. Esposito:

–That’s exactly correct, Your Honor, and we believe that it does not foreclose it.

First of all, in Bruno, the court indicated that the harmless error statute was not intended to apply to commands of Congress and respondents contend that it very logically and easily flows that it would not respond to… would not apply to constitutional violations either.

Let me make a point about the legislative history, since the Court directed some questions to it.

The statute was repealed in 1947, I believe.

It was repealed because of the enactment of the Federal Rules of Civil Procedure.

The following year, in ’48, it was realized that the Federal Rules did not apply to the courts of appeals in the Supreme Court, so it was in fact re-enacted.

Well, the Federal Rules of Civil Procedure were promulgated, as I recall, in 1934.

Why did Congress not get around to repealing this thing that they felt was inconsistent with it until 1947?

Paul V. Esposito:

I am not sure of that, Your Honor.

I know when it was repealed.

I am not quite clear why it–

Was it in conflict with their Rules of Criminal Procedure or–

Paul V. Esposito:

–There was a… federal… civil and criminal procedure, Your Honor.

I should be saying the criminal rules here.

So it was re-enacted, and re-enacted to bring back the provision, albeit with the language change, but it was re-enacted to bring back the provision of 391, which was basically discussed in the Bruno case, back into the federal law.

So, there really is, from respondent’s view, and what we see of the legislative history, no change in the legislative intent.

–Well, don’t you think the dropping of the word “technical” after Justice Frankfurter’s opinion in Bruno had referred to it as kind of a matter of etiquette and that sort of thing, meant that Congress intended a little more bite to the thing than it had before?

Paul V. Esposito:

Your Honor, we believe that if the Court… excuse me, if the Congress had intended to change it, there would have been much more discussion.

Why need there be any discussion if Congress chooses to enact something which in fact changes it?

Paul V. Esposito:

Because the commands of this Court or the reading of this Court was so… was so very clear.

In the language of the legislative report… in the bill that was enacted, in the bill in which it was re-enacted, all that bill was doing was to make typographical change, delete sections that should have been deleted before, add sections that shouldn’t have been deleted.

It was mainly a… how do I say it?

a bill just to make the minor changes.

Paul V. Esposito:

There wasn’t any substantive law change that was intended by the bill.

But how can you be so sure of that when in fact the words come out differently than were in the previous thing, and when it in fact had been repealed for a year?

I mean, it is not as if you are just codifying something that has always been in effect.

Paul V. Esposito:

Well, the closest I can find, Your Honor, is the House report on the bill, which just very briefly talked about it, and again, I would contend that that indicates that Congress was not intending a substantive change that may change the prior readings, prior readings of the court, and there is some language, very briefly, by Senator O’Connor the Senator from Maryland, I believe, who also indicated that this bill doesn’t really mean to… wasn’t meant to make a great substantive impact in the law.

So, we contend that there is still… there is still… the statute is inapplicable, and furthermore, nothing in the legislative history, in fact, nothing in Chapman itself would seem to indicate that the harmless error rule itself was meant to take away the supervisory powers of the court to deter repeated misconduct.

That is a point I think that has to be brought out in this case.

The Seventh Circuit has addressed this problem in the past.

They have told the prosecutors that these types of comments are wrong.

They have threatened summary reversals, and in fact, to make their point perfectly clear, they even went so far as to circulate an unpublished opinion to all the United States attorneys within the circuit with instructions to either give a copy of that opinion to the assistant attorneys or to issue an explanatory memorandum.

I think one of the reasons why the opinion is so short is because the court was just so terribly frustrated with what was going on.

It had said enough, when you look back at the decisions in–

But you have to go look back at them, don’t you?

Paul V. Esposito:

–Yes, you do, Your Honor.

Your suggestion that this is why they did it they certainly didn’t reveal in what they wrote in this case.

Paul V. Esposito:

That’s correct.

I think… as I say, I think they just felt enough was enough, and they just decided to put the foot down right here and now, and it was this case that they did.

Mr. Esposito, you speak of this series of Seventh Circuit cases.

How do the facts of those cases compare with this one?

Do you have any idea?

Paul V. Esposito:

Closely, Your Honor, in that… in the type of comment that was involved.

In the comments… in the cases that are referred to in our brief, the comments were either that the defendant had not denied evidence, had not refuted evidence, had not contradicted evidence.

In this case, the contention was that the defendants had not challenged evidence, a different word, but the same intent.

What about the underlying facts of the charges and the like?

Paul V. Esposito:

Different cases.

Some involved robberies.

I think there was a mail forgery, one case.

The basic facts, the substantive crimes were different, but the language was the same, and the standard was the same.

There was–

Why did they pick this case particularly when they had let a dozen or so go by?

Paul V. Esposito:

–Because they were tired of giving warnings, Your Honor, and they felt, we believe, from the history of this problem, that warnings were no longer going to do the job.

Paul V. Esposito:

It was becoming increasingly clear to the court of appeals that if they had again applied Chapman, it wouldn’t do any good.

What they were trying to do is eliminate the problem.

After speaking to the problem a number of times, after warning prosecutors, the problem wasn’t being eliminated, and that is what those–

So these defendants as contrasted with the others in perhaps less offensive crimes were particularly fortunate.

Paul V. Esposito:

–As opposed to the others, their conviction is reversed.

Yes, Your Honor.

I think the Court was trying to clear up a problem that won’t… or hopefully will not reoccur in the future.

They… in a sense, they do get a benefit, but so does the system, and I think that is what the… that’s what the court was interested in, protecting… protecting the system of justice so that these types of comments would not occur in the future.

Mr. Esposito, if you are correct that the court does continue to have some supervisory power to set aside convictions despite the provisions of Section 2111, should not the court be required to at least articulate the fact that it is exercising the supervisory power, and to set forth the factors that informed its decision, and to consider the public interest in maintaining a conviction in a crime of this nature?

Aren’t those things all appropriate for the court to have to consider and to articulate?

Paul V. Esposito:

Your Honor, it may very well have been appropriate for the court to state the basis on which it was acting for the benefits of this Court and for people who have to read the decisions.

In fact, if I might ask this question, isn’t it particularly anomalous for the court of appeals to set an example, which it seemed to have been doing, in an order which is an unpublished order which the parties are forbidden to cite in other proceedings?

Paul V. Esposito:

I think–

This is not a published opinion, as I understand it.

Paul V. Esposito:

–There are two parts to this opinion, Your Honor.

The first part, the part that is up for review today, is a published opinion.

Oh, it is.

Paul V. Esposito:


Oh, I misunderstood.

I thought–

Paul V. Esposito:

The second… the second part is an unpublished order.

–I misunderstood.

Paul V. Esposito:

The unpublished sections are not before the Court.

So this part they did publish, the part we are–

Paul V. Esposito:

This part is published, Your Honor.

–Has the… excuse me.

Has the court of appeals ever undertaken to exercise any form of supervisory power by disciplinary action against a lawyer who breaks some of the ground-rules?

Paul V. Esposito:

Your Honor, in the last decision, United States versus Rodriguez, the decision that immediately preceded this case, the Seventh Circuit said that this matter may be something that should be brought to the attention of the Attorney General himself, the matter of his repeated comments.

The court did not bring this matter, as far as I determined from the decision, the court did not bring the matter up to the Attorney General, and I can only conclude that they had reassessed the matter and said that it would be best to handle this through a reversal.

I think there is one… well, there are two problems possibly in terms of discipline.

Paul V. Esposito:

One is is a possible separation of powers problem.

What type of… In terms of the types of discipline that the court may be able to invoke against a prosecutor.

Well, does not the district court have plenary jurisdiction on admitting lawyers to practice in the district court and in turn in the court of appeals?

Paul V. Esposito:

They do, Your Honor.

In your view, would the court of appeals exceed its supervisory powers or its powers generally if they issued an order to show cause why a lawyer who violated rules in their view should be disciplined, or the order to show cause would be to show cause why he should not be disciplined?

Paul V. Esposito:

Disciplined by the district court, Your Honor?

Or by the court of appeals?

Either, or both.

Paul V. Esposito:


Well, I gather the district court wouldn’t have been disciplining him, because they thought it wasn’t even a violation of the Fifth Amendment.

Paul V. Esposito:

They thought the comment was proper in the first place.

What I am speaking of is the supervisory power of the court of appeals over the district court.

Paul V. Esposito:

The court may have been able to issue an order, a rule to show cause, as you suggest.

The problem–

Well, I don’t want to divert you.

I just wondered whether–

Paul V. Esposito:

–No, I think–

–if the court had ever undertaken to exercise such a power.

Paul V. Esposito:

–I have never… I have never seen the power exercised, but I would like to address your point, Your Honor.

The problem that respondents see from the mere approach of discipline is that while it very well may deter this prosecutor, and in fact it may deter other prosecutors, it doesn’t really get the point home to the prosecutorial system as to what this problem is.

It doesn’t necessarily get them to change their practices, to change the educational practices, the training practices of prosecutors before they get to the trial.

Well, what business is that of the court of appeals?

The court of appeals isn’t in the business of educating prosecutors.

It is in the business of reviewing on an appellate basis decisions of the district courts, isn’t it?

Paul V. Esposito:

Your Honor, I would suggest that part of the business is the education of prosecutors, and that is part of the basis for the harmless error rule itself.

Certainly the harmless error rule is designed in the first instance to affirm a conviction, to affirm the result when the result did not prejudice the defendant, but the rule goes farther.

It tells a prosecutor, it tells the government itself that conduct that was committed at trial was wrong, it was improper, and hopefully through the force of that decision itself it coerces a prosecutor not to repeat those comments in the future.

That is where the education comes in, through the proper use of the rule.

I think what is happening in the Seventh Circuit, and I believe what the Seventh Circuit is despairing of through a reading of all these opinions is the fact that in a sense that harmless error, the rationale is being broken.

Rather than using the harmless error rule to affirm the result, it is being used to justify the conduct.

Paul V. Esposito:

People are forgetting that there is error and just–

Well, but so long as it is harmless error, why should the court of appeals worry about it?

Paul V. Esposito:

–Because it is still, Your Honor, a violation of constitutional rights, and it still has an effect on the system itself.

Well, what effect does it have on the system if it is harmless error?

Paul V. Esposito:

It affects the system in the way a prosecutor… a trial is conducted.

Our system requires certain norms of conduct within it.

Yes, and the norms of conduct are set up, and then reversals are not to occur if there is only harmless deviation from them, and by hypothesis, what we are talking about is a harmless deviation from them.

Why should that concern anybody?

Paul V. Esposito:

Just because, Your Honor, of the basic manner in which the prosecution is being gathered, in which the system is being gathered.

I guess you can say the same thing, Your Honor, of a coerced confession or any other type of constitutional problem.

It is bigger than both of us?

Paul V. Esposito:

It may be.

Was this widespread throughout the Seventh Circuit?

Paul V. Esposito:

The Seventh Circuit has dealt with this particular issue in eight cases, and they have come throughout different districts in the circuit.

Yes, Your Honor.

But not in other circuits?

Paul V. Esposito:

The court wasn’t… the problem has appeared in other circuits.

As a matter of fact, in the First Circuit, eleven years ago, the First Circuit adopted a rule under its supervisory powers to provide for reversals for comments on the… indirect comments like this comment on the failure to testify where certain preconditions aren’t established.

The court of appeals had threatened to employ that rule in the Seventh Circuit, but didn’t go so far as to do it.

I would like to address the question of standards that can be applied.

I don’t believe that this is something that has to be an unlimited review.

It is something that the Seventh Circuit can use to reach a result whenever it wants.

Supervisory powers, of course, are to be reluctantly exercised.

I think that is clear from this Court’s case law.

But it is a discretionary power, too, and where… and in determining whether to exercise the discretion in cases of this type, in the case of this type is repeated misconduct which the prosecutor’s failure to conform to decisions, the court should look first of all to the type of error involved, dealing with constitutional error, command of Congress.

It should look to whether they have addressed, that particular court has addressed the problem in the past, whether it has set forth a clear legal principle, whether it is a principle that can be applied broadly to many types of cases such as this type of case, and whether that principle has been disseminated in sufficient time to give prosecutors the practical means of incorporating the principle into the methods of prosecution.

Any of these factors are met, and they are certainly met in this case, the number of times that the court has… has to deal with the problem in a sense becomes an aggravating factor, making the exercise even more appropriate.

If the court didn’t abuse its discretion in this regard, the decision should be affirmed.

Mr. Esposito, I gather your answer to the government’s argument is that exercise of supervisory power is not foreclosed by the harmless error statute.


Because the statute, you argue, is inapplicable to this situation.

Paul V. Esposito:

I argue first of all, Your Honor, that the statute is inapplicable.


Why is it inapplicable?

Paul V. Esposito:

–The statute, because it wasn’t, as I read the case law, Bruno, and extend the case law, Your Honor, it wasn’t designed to apply… the case cited wasn’t designed to apply to violations of statutes.

In fact, in Bruno, the statute involved was the present day 304.81, commenting on the failure to testify.

It seems to me it very logically follows that if it doesn’t apply to the statutes it wasn’t meant to apply to the constitutional violation.

Even assuming, Your Honor, assuming arguendo that the statute could be applied, may be applicable, we contend that the harmless error statute was not intended to deprive the court of their traditional supervisory powers, powers that are recognized in McNabb, powers that are recognized in Donnelly versus DeChristoforo, for instance, to deter prosecutorial misconduct.

Those powers, we believe, are still there, and absent something much clearer that would indicate that the Congress meant to take away that power, that supervisory power is there.

In sum, Your Honor, we believe that above all, the supervisory powers are the powers of flexibility.

They enable the courts to determine whether a rigid application to a particular legal principle is really going to frustrate rather than promote the interests of justice, and when it does, it allows the courts to bend slightly the system to really meet and to effectuate in a good sense the overall administration of justice.

We believe that is what the court did here.

They weren’t trying to overrule Chapman.

They did not create a per se rule in the Seventh Circuit.

They were just trying to say, in this case, we will put our foot down here.

Well, now, I gather from your colloquy with Justice O’Connor, since there is nothing in this opinion to tell us this was reached on the basis of supervisory power, we can’t conclude that it was without going back to all those cases, the eight of them that you have cited in your brief.

Paul V. Esposito:

Your Honor, I don’t… I think that is correct, Your Honor.

I think in going back to those cases, you will not see the words “supervisory power” invoked except to the extent that the court discusses Flannery, which, in that case the First Circuit invoked their supervisory powers.

So that is correct.

If the Court wants specific findings as to the power basis, it may very well have to remand the case or at least read the decision.

When the Court reaches its decision, you can–

And tell the court of appeals if in fact U.S. did this on supervisory power, tell us so, and your reasons for doing it.

Paul V. Esposito:

–That may be appropriate, if the court does not feel that a decision just on the… does not feel that it cannot just act on the basis of supervisory powers.

It is something that would be used to affirm the judgment of the court.

If there are no further questions, I would ask the Court to affirm.

Warren E. Burger:

Very well.

Mr. DePue?

John F. DePue:

Thank you, Mr. Chief Justice.

I just have–

Before you go on, let me inquire on this point.

In your view, if a prosecutor who was admitted to the court of appeals to practice before the court of appeals had committed a gross, not a marginal or the kind that you say here is incidental, but a gross violation of the matter of commenting on the failure of the defendant to testify, could the court of appeals discipline that lawyer?

Or suspend him?

John F. DePue:

–I think it is questionable.

I think… If he was a member of the court of appeals bar, it certainly could.

Yes, I assume that.

I am not speaking of the district court.

The district court has exclusive jurisdiction there.

But you agree that they could suspend him from practice for some period of time as one kind of discipline.

John F. DePue:

Yes, Mr. Chief Justice.

But your point would be that the conduct here doesn’t rise to any such level.

John F. DePue:


And the two alternatives that we pointed out in our brief was, the district court itself could have disciplined the individual, and finally, the Department of Justice could have and would have disciplined an individual who intentionally violated Chapman.

The point I would like to–

Can you name me any case that it happened?

John F. DePue:

–Yes, Justice Marshall.

I believe I can.

I don’t particularly want to use the individual’s name, but as we pointed out in our brief, within the last eight months a prosecutor–

Did you have any before eight months?

John F. DePue:

–We have never had an incident like this reported.

No, Justice Marshall.

Where this particular type of violation was reported to the Office of Professional Responsibility.

But as I have said, within the last eight months, we had a prosecutor who was reported for making improper comments before a grand jury, and he ended up resigning.

I am talking about before the court, where the court disciplined the U.S. Attorney.

John F. DePue:


Any court.

John F. DePue:

I know of no examples.

I would like to dispel the notion that has been created by respondents and by the court below that this is a pervasive and recurring problem in the Seventh Circuit.

To support this proposition, in both the decision in this case and the preceding Rodriguez decision, they cited a total of eight cases, I believe, that occurred over some 20 years.

In only one of these cases was there an actual comment on the defendant’s failure to testify which the Seventh Circuit found was inadvertent.

In two of the cases, they found that there was no violation of Griffin.

John F. DePue:

In the remaining cases, they found that remarks like “undisputed” or “uncontested” transgressed this indistinct line that I have discussed previously, and therefore constituted what I would term technical violations of Griffin.

You simply cannot use a supervisory power as a prophylaxis, assuming it was otherwise amenable, as a prophylaxis for prohibiting unintended misconduct.

This was the essence of Judge Larson’s dissent in Rodriguez.

Inadvertent remarks aren’t subject to a didactic lesson.

In response to Justice Stevens’ question concerning whether the phrase “substantial rights” would preclude application of the Constitutional harmless error standard in this case, or in any other case involving Constitutional errors, no, the substantial relates not to the nature of the right but to the nature of the injury.

If substantial right were construed to implicate the nature of the right involved, no Constitutional error could ever be deemed harmless, because they are all substantial.

Well, that doesn’t follow at all.

It would mean no Constitutional error would be covered by the harmless error statute.

It could still be deemed harmless under Chapman against California under the reasoning of that opinion, which did not rely on the harmless error statute.

John F. DePue:

No, it didn’t rely on the harmless error statute, but it certainly said, let the statute and the principles embodied in that statute apply to errors of a Constitutional magnitude, which are invariably substantial rights.

It did not say… I just reread it… that the principles of that statute applied.

It said there is a Constitutional harmless error statute which they formulated in that case.

John F. DePue:

And this Court has repeatedly applied, of course, the Constitutional harmless error doctrine to errors of a Constitutional magnitude.

Justice Brennan asked me before whether the court has ever held that it can override a statute under its supervisory power.

In both Palermo and in National City Lines, the court said precisely that, that it could not.

Thank you.

Warren E. Burger:

Very well.

Thank you, gentlemen.

The case is submitted.