Transcript

IN THE SUPREME COURT OF THE UNITED STATES

CONNECTICUT, Petitioner v. LINDSAY B. JOHNSON

No. 81-927

October 13, 1982

The above-entitled matter came on for oral argument before the Supreme Court of the United States at 2:00 o'clock p.m.

APPEARANCES:

LINDA K. LAGER, ESQ., Special Assistant State's Attorney, New Haven, Connecticut; on behalf of the Petitioner.

JERROLD H. BARNETT, ESQ., New Haven, Connecticut; on behalf of the Respondent.

PROCEEDINGS

CHIEF JUSTICE BURGER: We will hear arguments next in Connecticut against Johnson.

Ms. Lager, I think you may proceed whenever you are ready.

ORAL ARGUMENT OF LINDA K. LAGER, ESQ., ON BEHALF OF THE PETITIONER

MS. LAGER: Mr. Chief Justice, and may it please the Court, the principal issue in this case is whether as a matter of federal constitutional law the state prosecution is entitled to show that a constitutional error in an unobjected to portion of a lengthy, detailed, and elaborate jury instruction is harmless beyond a reasonable doubt.

In 1981, the Connecticut Supreme Court foreclosed the prosecution that opportunity by automatically reversing the Respondent's 1976 convictions on attempted murder and robbery after finding Sandstrom error. That finding of Sandstrom error was premised solely on the court's use of the word "conclusively" twice, and the balance of the instructions were in fact described by the Connecticut Supreme Court as "precise, elaborate, and cast in highly permissive language," and the Connecticut Supreme Court also found that if the word "conclusively" had not been used in this jury instruction, it would not have found Sandstrom error.

The state maintains that a finding of an arguably conclusive presumption did not warrant automatic reversal in this case. Furthermore, the state of Connecticut maintains that a reviewing court can determine the impact of an unconstitutional jury instruction on the verdict, and that in this case, because the instruction could have no impact on the outcome of the case, the instruction was harmless beyond a reasonable doubt.

The Respondent in this case was charged with four crimes that took place on Saturday night and early Sunday morning of December 20th and 21st, 1975, the weekend before Christmas. It is not necessary to repeat to this Court the sordid details of the rape and the kidnapping aspects of this case, because they are not before this Court.

QUESTION: Ms. Lager, that is seven years ago, isn't it?

MS. LAGER: That is correct, Your Honor.

QUESTION: Why did it take four years in the Connecticut court to get decided?

MS. LAGER: The case was -- excuse me. The Connecticut Supreme Court decided the case in 1981 principally because there were problems in obtaining the complete transcript. In fact, most of the trial transcript had been obtained in 1977 and 1978. Then it was discovered that there were portions missing. Those portions involved the hearing on the motion to suppress. In this case, there were initially five defendants, and some of the hearings on the motions to suppress were conducted jointly with other defendants.

The last portion of transcript involving the motion to suppress was received by counsel on November 26th, 1980. The case was then rapidly briefed and argued in June of 1981 under a special briefing schedule that the Connecticut Supreme Court had set up.

What was involved in this case --

QUESTION: Does this happen frequently in the state of Connecticut?

MS. LAGER: This was a somewhat unusual circumstance. There have been problems in the state of Connecticut --

QUESTION: Just somewhat unusual?

MS. LAGER: -- with transcription. I think the length of the delay in this case was somewhat -- extremely unusual.

In this case, the victim clearly had ample opportunity in over five and a half hours of being with the Respondent and his five companions to view them. In fact, in the course of her testimony before the trial court, she was able to give clear identification of the Respondent and his co-assailants by details of clothing and physical description. It was also undisputed that the Respondent was the leader of the group, that he initiated the incident, that he took the most active role, and that he was the most vicious, that he gave directions, that he restrained the victim, that he threatened her, and that he was the first one to rape her.

The crucial facts, I believe, before the Court are those that are surrounding the robbery and the attempted murder, and I will concentrate this afternoon on the facts surrounding the attempted murder because in my opinion, and the Respondent virtually concedes that there was overwhelming evidence of intent necessary for robbery by not even discussing those facts in his brief.

QUESTION: May I ask you a question? You are going to argue that the error was harmless, I think, now.

MS. LAGER: That is correct.

QUESTION: Before you get to that, I am not quite clear on your position with respect to when the harmless error argument should be made in Connecticut. In the state against Truppi, the court had an opinion in which it discussed that it would sometimes entertain a harmless error challenge and sometimes it would not. Now, if as a matter of Connecticut practice the Connecticut Supreme Court should decide, well, Sandstrom errors are going to give rise to automatic reversal, would we have power to tell it it could not do that?

MS. LAGER: I believe that what the Connecticut Supreme Court did is decide as a matter of federal constitutional law that it would not review the error in Truppi as being harmless. Its citations are solely to federal cases.

QUESTION: Well, they cite State against Zeco, which is a Connecticut case, State against Sorbo, a Connecticut case, Elman against State, a Connecticut case, State against Briggs, a Connecticut case.

MS. LAGER: Those citations are to cases in which the court, the Connecticut Supreme Court was willing to apply a harmless error rule --

QUESTION: Right.

MS. LAGER: -- and the Connecticut Supreme Court has indicated in a decision that was decided on January 5th of 1982, Turcio versus Manson, which was cited at Page 3 of the Respondent's brief, in a footnote that they would not decide in the Turcio case nor did they decide in the Johnson case, this case, that there can be no language or other circumstances which could operate to cure an intent instruction containing the phrase "conclusively presumed."

Subsequently, in a decision of March 30th, 1982, State against Kervin, the Connecticut Supreme Court appeared to be moving towards this Court's total record review analysis in determining whether a jury charge could be harmful. In that case, they specifically emphasized that the charge should not be examined in a vacuum, that it should be viewed in the context of the factual issues raised at the trial. The challenge in that case was premised on Sandstrom and based on a claim that an elemental portion of one of the offenses was improperly given. In fact, there was a dissent that would have held that charge in violation of Sandstrom.

QUESTION: I am not sure I understand what your explanation of what the state of Connecticut's position is on harmless error on this issue. Does the court sometimes review it and sometimes not review it?

MS. LAGER: I think that the state of Connecticut has a very ambivalent position with the application of the harmless error rule, and I think it has adopted a case by case review. I think the reason that the Court declined to apply the harmless error doctrine in State against Truppi, although they considered the issue when it was raised by the prosecution, was because of the nature of the instruction that was given in that case, and in this particular case --

QUESTION: Well, does that mean -- I just want to be sure I follow what you are explaining to me. Does that mean that the Connecticut Supreme Court as a matter of Connecticut practice sometimes entertains the challenge to instructions and sometimes does not?

MS. LAGER: It appears that the Connecticut Supreme Court as a matter of federal constitutional law sometimes entertains the instruction -- the challenge that the error was harmless and sometimes it does not. The only stated position that the Connecticut Supreme Court has given in this regard is the one in State against Truppi, is that we sometimes apply the harmless error rule.

It appears that if the --

QUESTION: Well, is it your position that they must apply it in every case?

MS. LAGER: No, Your Honor, my position is not that they must apply the harmless error rule in every case, but if the prosecution asks the Connecticut Supreme Court to review an error for its harmlessness as a matter of federal constitutional law, that it must at least use the harmless error test.

QUESTION: Let me just state it to be sure I understand it correctly. As a matter of federal constitutional law, a state supreme court must entertain a harmless -- must make a harmless error examination every time the prosecutor asks it to.

MS. LAGER: In the context of an unconstitutional jury instruction, yes.

QUESTION: Ms. Lager, before you go on, would you mind telling me where "conclusively presumed" was used in the instructions at 23A and 25A? Was it used in respect of only one of these offenses, or all of them? And if less than all, which?

MS. LAGER: At 23A, Your Honor, the court was giving a general instruction with respect to the question of a finding of intent. The second time that the word --

QUESTION: This would cover all of the offenses charged.

MS. LAGER: That is correct. The Connecticut Supreme Court found, however, that the language of the instructions on the kidnapping offense was sufficiently permissible and phrased solely in terms of inference therefore that any error in the general instruction was cured by that specific one.

QUESTION: The court drew back from that instruction in the latter part, did he not, where he said, "If you believe the victim's version as to the defendant's conduct at the bridge, you may presume he intended what would be the natural and necessary consequences of his actions."

MS. LAGER: That is correct, Your Honor, and in fact the argument was made to the Connecticut Supreme Court that what was operating in this case was not merely a conclusive presumption at all, that the court's language was ambiguous. The Respondent in fact makes quite a to-do about the fact that the court instructed on a rule of law, but the rule of law that the court instructed on is as follows. "The intent is a question of fact for the jury to find." That was clearly told to the jurors. And then the court said, "However, you should be aware of the rule of law that will be helpful to you, and that is that a person's intention may be inferred from his conduct, and every person is conclusively presumed to intend."

QUESTION: In the court's opinion, that discussion relates only to the first count of attempted murder. The Connecticut Supreme Court did not discuss that, as I recall it, in the counts on kidnapping and robbery.

MS. LAGER: What the Connecticut Supreme Court found is that this general instruction that appears at the bottom of Pages 22A and 23A of the appendix was unconstitutional despite the other language that was used in the qualifying language in the charge. It then proceeded to examine each of the four separate offenses to determine whether there was any spill-over from that general instruction.

QUESTION: The first being the charge of attempted murder?

MS. LAGER: The first being the charge of attempted murder.

QUESTION: That is where the "conclusively presumed" was repeated at Page 25A?

MS. LAGER: That is correct. In that portion, after the judge explained to the jury that they must find that the Respondent had the specific intent to cause the death of the victim, and they must ask themselves what intention did the defendant have in mind, did he alone or participating with others intend to commit the crime of murder, and that no one could look into a person's mind and see what the intention is. The court went on to say again in an ambiguous way, "The only way to decide that question is to infer from the accused's conduct in the light of the surrounding circumstances, but as previously stated, every person is conclusively presumed." We have --

QUESTION: Ms. Lager, am I right in thinking this about the state's position, that the questions as to whether the instructions considered as a whole might have improperly influenced the jury on the constitutional issue of burden of proof go into deciding whether or not there was so-called Sandstrom error, and then the question of harmless error is whether on that, on the issue that the instructions were addressed to, the evidence was so overwhelming that probably no damage was done?

MS. LAGER: That is an aspect of the state's position. The state is not before this Court to relitigate the Connecticut Supreme Court's finding that the jury instruction violated -- the "conclusively presumed" jury instructions violated Sandstrom. The state's argument is that the other instructions in the charge and the totality of the entire record in this case, including the overwhelming evidence of guilt, would allow a reviewing court to assess the impact of this unconstitutional instruction on the outcome of the case.

QUESTION: But I think your response assumes that the Connecticut Supreme Court at any rate simply took one instruction in isolation, and said, look, this violates the Sandstrom case, and then didn't treat the other portions of the instructions that might be assumed to counteract that effect. I read the Supreme Court of Connecticut's opinion somewhat differently, that they assimilated all the balancing of the various parts of the instructions into the final question, did it improperly influence the jury in an unconstitutional way, into the question of was there Sandstrom error, and therefore the question of the harmlessness of that error would have to be taken somehow out of the -- outside of the instructions, which have already been fully canvassed in the determination of whether there was constitutional error.

MS. LAGER: I would agree that if the test that was adopted would simply require a canvass of the instructions to determine whether the Sandstrom error was harmless, as the Respondent has suggested in his brief, you would have the equivalent of an automatic reversal rule given what the Connecticut Supreme Court did in this case. The state's argument before this Court is that one must look at the instructions as part of the entire trial record, and determine --

QUESTION: Well, would that extend to a situation where the instruction on intent is clearly wrong under Sandstrom, and there is no redeeming instruction whatever, so that if you are simply canvassing the instructions, you have to reach the conclusion that the jury was improperly instructed in violation of Sandstrom, but the testimony at the trial is that the defendant took the stand on the question of intent, perhaps was asked the question, did you intend to do this, and the defendant said, sure. They were defending maybe on self-defense or something like that, so that you would say the error was harmless not because of anything contained in the instructions, but because the evidence on that issue was so overwhelming the jury would have reached that conclusion no matter what burden of proof it had been assigned.

MS. LAGER: Yes, I would agree with Your Honor's statement.

QUESTION: Could I put it another way, that the only way to be successful is that the appellant has to first show, has the burden of showing that there is a violation of Sandstrom, and secondly, he must show that there is not harmless error.

MS. LAGER: No, Your Honor. The state's position would be that the appellant has the initial burden of establishing -- the defendant below would have the initial burden of establishing that there was Sandstrom error, and then according to well established principles of this Court, the burden would shift to the prosecution to establish that error to be harmless beyond a reasonable doubt, and in shifting that burden to the prosecution, the appellate court would be allowed to examine how the instruction affected the outcome of the case.

If the appellate court could conclude beyond a reasonable doubt that there is no reasonable possibility that a rationale trier of fact could have entertained a reasonable doubt about intent, so that the instruction resulted in a conviction where there otherwise would have been an acquittal, then, despite the finding of Sandstrom error --

QUESTION: You mean, you assume that you put yourself in the position of a juror who is told about this conclusive presumption. And it has no effect on him at all.

MS. LAGER: That's not precisely the analysis that --

QUESTION: It's close, though, isn't it?

MS. LAGER: Not exactly, Your Honor, because --

QUESTION: Well, how do you get at the word "conclusive?"

MS. LAGER: What an appellate court --

QUESTION: What do you think "conclusive" means to the average person?

MS. LAGER: I think that the word "conclusive" had to be viewed particularly in this case in context. I think what the jury in this case would have heard are the following things. In -- questions of facts --

QUESTION: My question was very simple. What does the word by itself, "conclusive", mean to the average person?

MS. LAGER: To reach a conclusion, Your Honor.

QUESTION: Reasonable what?

MS. LAGER: To reach a conclusion.

QUESTION: Well, how can you reach a conclusion as to a presumption that is -- oh, yes, you reach a position he is guilty.

MS. LAGER: Well, that is not exactly what the jury in this case was told. The jury in this case was told that intent was a question of fact, that they were the fact finders, that the prosecution bore the burden of proving intent among all of the elements of the crime beyond a reasonable doubt, that the jury was to examine and assess the credibility of the victim, who was the only one to supply direct evidence of the acts of the defendant at the time of the offense, that the jury was then to assess all of the objective facts concerning the circumstances, such as the time of day -- it was 4:00 in the morning -- such as the temperature -- it was ten degrees below zero -- such as the fact that the victim unequivocally identified the defendant, such as the fact that the jury never came back in this case and asked for reinstruction on either the attempted murder or the robbery charges, such as the fact that the defense counsel never objected to the instruction that was given, that the issue of the defendant's intent was not closely disputed or at issue in this case, and that the theory of the defense did not really in any way involve either the attempted murder or the robbery charges.

QUESTION: Well, after all of that, why did you need the presumption instruction?

MS. LAGER: Well, at the time that that instruction was given, and I don't believe that any court since the Sandstrom decision would give that instruction, it had been a recognized statement of what I would like to describe as the standardized inference of intent, proving intent from an accused conduct. It is pretty clear that in most cases, we don't have the hypothetical situation where the accused comes in and says, I did it, except, for example, in a case -- and that I intended to do it. You may have a case like Sandstrom itself where the accused admitted his act but hotly contested the fact that he could form the specific intent.

Therefore, I don't think the judge could be faulted for giving this instruction, given the state of the law, and that there is really a very limited salutary effect in an automatic reversal rule in this type of a case. Indeed, it would probably be -- the more likely effect would be probably be for the public confidence in the administration of justice to be undermined when the general public hears that under the facts and circumstances of a case such as this one, a new trial has been ordered on the attempted murder and the robbery counts.

Even if we assume that the jury followed the presumption, the question is also what in fact would this jury have done that would have been any different --

QUESTION: I think that you and I are in the same position. We don't know because we were never on a jury.

MS. LAGER: Well, I think that --

QUESTION: Or am I wrong?

MS. LAGER: I think that in fact an appellate court --

QUESTION: Am I wrong?

MS. LAGER: I think that --

QUESTION: I know I've never been on a jury.

MS. LAGER: Well, I've never been on a jury either, Your Honor.

QUESTION: I've never been in a jury room.

MS. LAGER: But I think that appellate courts accepting and in fact sanctioning a standardized inference of intent understand what jurors do when they attempt to find intent. What they do is, they look at the evidence to see what acts and conduct have been proven beyond a reasonable doubt, and they assess credibility and perform the other functions that jurors do. They then look to the evidence to see the natural and necessary consequences of those acts under the attendant circumstances that are established by the evidence.

They then ask themselves, has any countervailing evidence been introduced in this case which raises a reasonable doubt, that is, a doubt based on reason, not a speculative doubt, about the defendant's intent, and then they reason from all of the above to find that intent has been established beyond a reasonable doubt, and in fact the reasoning process that this jury engaged in, assuming they followed the charge in this case, had to be that process, because that's what they were told to do specifically in the attempted murder instructions. They were told to examine all of the circumstances, to assess credibility, to draw inferences.

Furthermore, they were told, properly instructed on the kidnapping charge on how to draw an inference of intent from conduct, and the only reinstruction that the jury heard on the issue of intent was a reinstruction on the kidnapping charge.

QUESTION: Incidentally, the Respondent is now incarcerated on what conviction?

MS. LAGER: The Respondent is incarcerated on the kidnapping and sexual assault convictions.

QUESTION: And what are his sentences?

MS. LAGER: He is serving now as a result of the reversal an 18 to 36 year sentence.

QUESTION: And if the consequence of the Supreme Court of Connecticut's decision is what?

MS. LAGER: The consequence of the Supreme Court of Connecticut's decision is that the state is now obliged to conduct a second trial of the attempted murder and the robbery convictions, bring this victim back in after seven years of the incident to repeat the details of the story, establish the same case.

QUESTION: Well, what sentences did he get on those two?

MS. LAGER: He got a -- the consequence of the reversal was that the minimum sentence was reduced by ten years, and the maximum sentence was reduced by 20 years.

QUESTION: So he had --

MS. LAGER: He got a 28 to 56 year sentence originally.

QUESTION: Twenty-eight to 56, and he is now serving a --

MS. LAGER: An 18 to 36 year sentence.

QUESTION: -- 18 to 36.

QUESTION: Why must the state reprosecute?

MS. LAGER: Why must the state reprosecute?

QUESTION: That is what you just said.

MS. LAGER: The state's position in a case such as this one is that the gravity of the offense is so serious that it would be necessary to reprosecute the defendant.

QUESTION: Well, that's a matter of discretion on the state's part.

MS. LAGER: It is a matter of prosecutorial discretion.

QUESTION: So it doesn't have to.

MS. LAGER: It does not have to, but as the case stands now, it is in a posture of having a retrial on the two counts that were ordered to be retried.

I would like to reserve --

QUESTION: Let me have those figures again. You said 18 to 36 under the adjusted sentence.

MS. LAGER: Twenty-eight to 56 under the original sentence.

QUESTION: Well, it is down to 18 to 36, you said.

MS. LAGER: That is correct.

QUESTION: Now, on the assault charge, what was the sentence on that? And were they consecutive or --

MS. LAGER: If I may have a moment.

(Pause.)

QUESTION: Well, if it takes time, don't use the time now. I can check that.

MS. LAGER: If you look at Pages 51A and 52A of the appendix --

QUESTION: Thank you. Thank you.

MS. LAGER: -- that would give the sentence.

I would like to reserve the rest of my time for rebuttal.

CHIEF JUSTICE BURGER: Mr. Barnett?

ORAL ARGUMENT OF JERROLD H. BARNETT, ESQ., ON BEHALF OF THE RESPONDENT

MR. BARNETT: Mr. Chief Justice, and may it please the Court, this case -- this case involves the question of whether an admittedly unconstitutional instruction that posited the element of intent as a question of law, contrary to decisions of this Court, and which remained uncured in a viewing of the charge as a whole, can ever be considered harmless error by virtue of the strength of the state's evidence. That is the position that the Petitioner has taken here today, and it is the position taken in the Petitioner's brief.

There is really no question that the error was not corrected by the remainder of the charge. There has been some mention that the inference charge on intent, concededly permissible in the specific instructions on the kidnapping charge, constituted some sort of a cure, but the Connecticut Supreme Court ruled otherwise when it refused to import the kidnapping instructions into the instructions on robbery which was the third count, for the reason that the structure of the charge did not permit such a construction.

The initial error was made common to all crimes involved in the information.

QUESTION: Was that the one at Page 23A?

MR. BARNETT: That's the one that appears on Page 23A, Justice Brennan.

QUESTION: With the word "conclusive."

MR. BARNETT: The word "conclusive," where the jury was instructed on a rule of law that every person is conclusively presumed to intend the natural and necessary consequences of his act, and of course in Sandstrom versus Montana, this Court noted that "presume" had a common meaning of acceptance of a proposition as true without proof, and "conclusively," of course, has common meanings of irrefutable, final, and decisive.

QUESTION: I am not sure I understood the state's position just as you have stated it. I thought the state's position was that taking the record as a whole, that is, the instruction as a whole, that there was sufficient ambiguity to afford a basis for a harmless error holding.

MR. BARNETT: I do not understand the state's position to be such, Your Honor. I understand the state's position to be that a cure may have been effected by the specific instructions on kidnapping. However, the Connecticut Supreme Court refused to follow such a suggestion because of the structure of the charge, and further, it could only mean that if the state's position were made and considered, that the jury would have been submitted these crimes on two theories, one concededly unconstitutional, and one claimed to be constitutional, and I believe that settled law is that if the theories are to be viewed as alternatives, that the unconstitutionality, the admitted unconstitutionality would require the convictions to be set aside.

Yes, Justice Rehnquist.

QUESTION: Mr. Barnett, supposing that an appellate court reviewing a claim of Sandstrom error in the charge to the jury saw there was one sentence in a very long set of instructions that might be interpreted as suggesting a conclusive presumption, but came across seven or eight restatements of the proposition, as you often get in a series of complicated instructions, which simply used the -- clearly spoke in terms of an inference that was permissible under the Sandstrom decision. It concludes on the basis of all that that the jury simply wasn't misled.

Now, is the proper result for that appellate court to say there was no Sandstrom error, or that there was Sandstrom error but it was harmless, without examining into the evidence at all?

MR. BARNETT: My position, if Your Honor please, is that it would be proper for an appellate court to rule on the possibility of misunderstanding solely from the language of the charge, that if there was language which imported a conclusive presumption that the error could only be corrected in the charge.

QUESTION: Well, but supposing the appellate court concludes that it was corrected in the charge, that there is something that taken all by itself might be Sandstrom error, but other parts of the charge really cured it, so that the impact on the jury was de minimis. Now, is the answer that the appellate court reaches in that situation that there was no violation of Sandstrom or that there was a violation but it was harmless error?

MR. BARNETT: I think that the appellate court would have to rule or would rule in my opinion that there was no violation of Sandstrom because the charge has to be considered as a whole.

QUESTION: Mr. Barnett, I gather, or do I, that you think that an examination of the evidence is irrelevant to the determination of the issue we have to decide?

MR. BARNETT: I so think, Your Honor, and I base that on what I am going to posit to the Court is a concept of what is a jury's verdict, and I refer to language used by Justice Rutledge in Kotteokos versus the United States, where he said that a verdict is more than a judgment on the facts. A verdict is a judgment of law rendered by a court of laymen, and I think that that language explains why this Court has treated incidents of instructional error differently than it has treated incidents of evidentiary error.

QUESTION: Now, do you suggest that we have never treated for harmless error purposes instructional error?

MR. BARNETT: No, I do not make that suggestion. My suggestion is that although Petitioner has not placed much emphasis on the word "conclusively," that when the Connecticut Supreme Court said that the convictions had to be overturned because the word "conclusively" was used, "conclusively" became more than a word, and I think the decision of the State Supreme Court shows that. "Conclusively" became a concept.

The Connecticut Supreme Court overturned these convictions because in its opinion this instruction withdrew the element of intent from the jury's factfinding function insofar as attempted murder, where it was specifically stated for the second time, and insofar as the crime of robbery was concerned, where nothing was stated as to how the element of intent should be ascertained, and it became a sort of guessing game between the invalid presumption and the subsequent mention of inference in the kidnapping instructions, and by taking it from the factfinding function of the jury, by effectively eliminating intent as an element of the crime, what has been done is that the factual elements of the crime have been taken from the decision of the only body which is constitutionally able to make that decision, namely, the jury.

And in brief, my position here today is that the invalid presumption amounted to a directed verdict. It was the functional equivalent of a directed verdict.

QUESTION: Well, Mr. Barnett, isn't your position basically an attack on the whole notion of harmless error? Certainly in Harrington against California, the court seemed to have been faced with a similar argument, the argument that you can't put yourself in the position -- an appellate court can't put itself in the position of jurors, but as I understand it, the court rejected that argument. It said, we don't know, of course, what jurors sat, but our judgment has to be based on our reading of the record and what seems to be the probable impact of two confessions on the minds of an average jury.

Certainly, in that case, they thought they could decide what a reasonable jury would have done without that particular unconstitutional Bruton violation.

MR. BARNETT: That is true, Justice Rehnquist, and I stated that the Court has treated instances of instructional error differently from instances of evidentiary error, such as existed in Harrington.

QUESTION: What cases are you relying on for treatment of instances of instructional error?

MR. BARNETT: I am relying on Wyler versus the United States, the United Brotherhood of Carpenters and Joiners versus the United States, Bollenbach versus the United States.

QUESTION: All of those were trials in federal court, weren't they?

MR. BARNETT: Yes, they were.

QUESTION: And wasn't the court in those cases interpreting a statute of Congress providing that technical errors shouldn't be used as ground for reversal?

MR. BARNETT: That is correct, but I believe that this Court in Chapman versus California mentioned that the rule adopted in that case, very much like the federal harmless error statute that preceded it, was not designed to relate to substantial rights, so I think those cases do become very relevant.

QUESTION: But there is no reason why one should govern the other, is there?

MR. BARNETT: Well, I think that the idea of directed verdicts and their functional equivalents have been discussed by this Court lately. I refer to Justice Brennan's concurrence in Chiarella versus United States, which is cited in both briefs, that where it -- where the error is evidentiary, apparently from the nature of the violation, the material is admitted improperly for the jury's factfinding determination. Therefore, on review, a court can eliminate that tainted item and make a determination of what impact it had upon the jury, and furthermore, in evidentiary matter, the jury, of course, is told that it is the supreme judge of facts, it may accept or reject any evidence.

When it comes to instructional error, the jury has no choice. The jury is always told that it has a duty to abide by the instructions given to it by the court, and I believe that this Court has said that the crucial assumption behind the constitutional scheme of trial by jury is that the jury will listen, follow, and obey the instructions.

So therefore in the case of instructional error, you must start with the proposition that the error had an impact on the jury in the sense that they understood and followed.

QUESTION: And that it is irreparable other than by better instructions?

MR. BARNETT: It cannot be cured except by other language in the charge.

QUESTION: Well, now, in this case, Mr. Barnett, I appreciate that argument as to the attempted murder, where twice "conclusive" was used, but how do you carry this as instructional error into the other convictions --

MR. BARNETT: Into the robbery --

QUESTION: Yes.

MR. BARNETT: -- conviction. Justice Brennan, I believe it is Footnote 4 of my brief where I mention that the robbery -- the effect of this court's decision on the robbery conviction will not afford Mr. Johnson one less day in the prison. Footnote 3 sets forth the sentences, which are consecutive and which are concurrent. And Footnote 4 mentions that Mr. Johnson petitioned this Court, and his petition was denied, and as a result of concurrent sentences on robbery and kidnapping, it really is not factually going to make much difference.

QUESTION: Even on parole?

MR. BARNETT: Even on parole. It will not make much difference.

QUESTION: Well, you say it won't make much difference. Will it make any difference?

MR. BARNETT: That, on parole, I cannot tell you, Your Honor, whether it will definitely make any difference or not. We have had some recent statutory changes, and I am not quite certain how much credits would be allocated to him.

QUESTION: Would it make a difference if he were ever paroled and convicted of a new crime for recidivism purposes?

MR. BARNETT: It would increase his past record. He is a recidivist at the present time, Justice O'Connor.

QUESTION: Is there another one? That's the robbery.

MR. BARNETT: It's the robbery. What the Connecticut Supreme Court did in its construction of the charge is that it took the general instructions and analyzed them and found uncured error. Then it made an analysis between the general instructions and the specific instructions on each crime. For attempted murder, the error was compounded. For kidnapping in the second degree, the second charge in the information, the court felt that the error was cured by the permissive inference language that attended the description on how intent should be ascertained for that crime.

On the third charge, robbery in the first degree, the court went to great lengths in describing the nature of the intent, but it never described how the intent should be ascertained, so the jury was left with two specific instructions which could be construed as at variance with each other, the attempted murder, admittedly invalid, and the kidnapping, admittedly valid, plus the general instruction which had been made common to all crimes, and in the Truppi case, which is referred to in a footnote in my brief, this procedure is set forth. The Connecticut Supreme Court in the Truppi case held that by giving a general instruction on intent, the jury would interpret that general instruction every time the court mentioned intent.

So therefore, when it came to robbery, the court refused and properly so to speculate on which prior instruction the jury had adopted, and on that basis overturned the conviction also. And on that basis overturn the robbery conviction also.

QUESTION: Do you think it is always improper for an appellate court to say what the jury might have done under other circumstances?

MR. BARNETT: Well, Justice White, I think that much of it depends upon the instruction. Where the instruction has been given as a conclusive presumption and is removed from the jury, I believe that this Court recognized in the United Brotherhood case that -- and in the Wyler case -- that to ask the Court to decide a factual element that was not properly submitted to the jury was, in effect, asking the court --

QUESTION: What about an evidentiary error?

MR. BARNETT: In an evidentiary --

QUESTION: -- in which the jury is told, you consider all the evidence; and the jury considers all the evidence. And then some evidence is found to have been improperly admitted, and the Court finds harmless error on the grounds the jury would have come out that way anyway.

MR. BARNETT: Yes. In the evidentiary situations it's somewhat different because of the nature of the violation and the fact that the inquiry proceeds at a lower level, it does not proceed on the level --

QUESTION: Well, not unless the appellate court is coming to a conclusion as to what the jury would have done absent the evidence or with the evidence.

MR. BARNETT: Oh, absent the evidence?

QUESTION: Yes. So it's performing in a way a jury function.

MR. BARNETT: It is, but it can perform it because it can say with some certitude that this tainted item of evidence had no impact on the jury.

QUESTION: Well, what about the certitude based on the evidence that the jury couldn't possibly have found anything but that the defendant intended the act?

MR. BARNETT: First, the appellate court must assume that the jury followed the instruction. So if the jury --

QUESTION: Well, you have just told us we should assume the jury's followed instructions.

MR. BARNETT: Yes. That the jury followed the instruction, so if in the decisional process the jury used the invalid presumption as a way to find guilt, it does not affect a cure merely because the appellate court can say the jury would have convicted on the evidence anyway, because it would show that the jury did not act properly in considering intent as a factual element to be adjudged in terms of all the evidence.

QUESTION: But if the jury, you say the appellate court is foreclosed from saying, if the jury had been properly instructed, it would have found him guilty anyway because it couldn't have done anything but found that he intended what he did.

MR. BARNETT: No, Your Honor.

QUESTION: The Court can't do that?

MR. BARNETT: No, Your Honor, I am saying that if the jury used an improper decisional process fostered by the instruction which it must be presumed that they followed, then that is an error that affected the verdict. And that is the test under Chapman v. California.

QUESTION: Well, it affected the verdict.

MR. BARNETT: And the test is not whether the same result would have been achieved had the error not have occurred. That is the test.

QUESTION: What about the hypothesis that I posed to your opponent? Supposing that the jury is erroneously instructed along the conclusively presume lines on the issue of intent; the evidence at trial shows the defendant took the stand, admitted intent, simply advanced a self-defense defense by way of avoiding a guilty verdict. And during closing argument the counsel for the defendant says, we concede intent, what we are really arguing about here is self-defense; we think you should find him not guilty for that reason.

It goes to the appellate court. You say they could not say in the light of this situation that whatever burden of proof was put on the State with respect to the element of intent, a reasonable jury would have found it met here?

MR. BARNETT: I must answer the question this way, Justice Rehnquist --

QUESTION: You cannot answer it yes or no?

MR. BARNETT: Yes, I can answer it yes or no.

QUESTION: But you are going to take some time doing it. Okay.

MR. BARNETT: But may I explain my answer a little. My answer is that, no, an appellate court could not in the case of an instruction worded as this one, because the wording of this instruction means that it did not reach the jury in the manner in which it is constitutionally required to. And for that reason the confession and the admissions of counsel would not act to cure. And that, I think, is one of the difficulties that courts have had with the Sandstrom situation where the instruction was capable of two interpretations and where the claims may before this Court, somewhat belatedly, were that the error was harmless because assume it was burden shifting, the burden didn't matter because of Sandstrom's confession.

QUESTION: But on this kind of an instruction, is it your position that no amount of evidence, no matter how overwhelming, could permit a reviewing court to say that it was harmless error?

MR. BARNETT: That is my position, Mr. Chief Justice, that if the instruction posited as conclusive, and I draw that to the analogy to the directed verdict, that no court has the power to direct a verdict in a criminal case no matter how conclusive the evidence may be.

QUESTION: Even if a reviewing court would say that no rational mind could have been misled by the ambiguity in the instruction in the light of the overwhelming evidence? You would maintain that position?

MR. BARNETT: I maintain that position.

QUESTION: It doesn't leave much left to the harmless error rule, does it?

MR. BARNETT: Well, it depends, if Your Honor please, I submit, on what type of error has been committed. I maintain that position because of decisions from this Court that there are institutions that have certain functions and that if the function of one institution is taken away from it, then the error cannot be cured because of the constitutional requirement of trial by jury.

QUESTION: May I ask, Mr. Barnett, do you concede that the instruction in this case was ambiguous?

MR. BARNETT: No, I most certainly do not. I concede it was the most unambiguous instruction there could be.

QUESTION: Mr. Barnett, may I just ask you what your understanding of the holding of the Connecticut Supreme Court was? I don't read in their opinion the same theory of defending this judgment that you advance in argument today. They don't say anything like that.

MR. BARNETT: Justice Stevens, perhaps I should start with answering something that you asked of Ms. Lager. The harmless error rule in Connecticut for criminal cases generally revolves about a case called the State v. Evans, which is cited in the respondent's brief in the section which explains how the error was considered below.

State v. Evans rests on two premises: the Connecticut Supreme Court will consider on a plain error basis claims where error was not claimed at the trial level when a newly articulated constitutional position is announced by this Court and where the record shows that the petitioner may have been deprived of a fair trial because of denial of constitutional right.

And in this case the decision notes that they have accepted claims made on the basis that this Court's decision in Sandstrom v. Montana. But they have never ruled on a case where the claim of error involved an instruction stated conclusively as this one.

Subsequently, in Turcio v. Manson in a habeas corpus case, convictions were reversed. The instruction was either the same or similar to this.

QUESTION: Well, I am still not sure I understand. Does that mean in your view that they have taken a position that as a matter of Connecticut practice they will always consider a Sandstrom error to be not subject to harmless error?

MR. BARNETT: No, if Your Honor please, it means that they have taken the position that failure to object to the instruction at trial will not bar you from appellate review in the sense of Wainwright v. Sykes.

QUESTION: Well, all right. So you are over the Wainwright against Sykes problem, you have got the error before the Supreme Court of Connecticut. When, if ever, will they allow the prosecutor to say, well, granting there was Sandstrom error, nevertheless it was harmless because we look at the entire record. Have they passed on that question?

MR. BARNETT: I don't believe that the Connecticut Supreme Court has passed on that question. Error has been found to exist or not to exist according to the language of the instructions.

QUESTION: And what do you understand them to have done in this case as a reason for rejecting the prosecutor's argument that this error was harmless? Why did they reject that?

MR. BARNETT: They rejected it because the error, the instructional error, was uncured by the remainder of the charge and because of its conclusive, expressly stated conclusive nature, removed the element that could have been interpreted by a reasonable jury.

QUESTION: But that doesn't really meet the argument that the prosecutor makes that it could also be cured by the abundant evidence of guilt.

MR. BARNETT: No. They --

QUESTION: And then they don't really deal with that argument, do they?

MR. BARNETT: The decision does not expressly address that. But I think it is implicit in the decision that an error of this type can be cured only in the charge.

QUESTION: Mr. Barnett, as I understand your position, the Sandstrom function is a functional equivalent of a directed verdict. If so, why was the case submitted to the jury on the attempted murder question at all? Why didn't the Judge simply say that as a matter of law he concluded that the evidence was insufficient to go to the jury on that issue?

MR. BARNETT: That certainly would have been error, Justice Powell. I think it certainly would have been error.

QUESTION: Why? You say the instruction was the equivalent --

MR. BARNETT: Yes.

QUESTION: -- of a directed verdict --

MR. BARNETT: Yes.

QUESTION: -- on the issue of intent. Intent is a necessary element of the crime, isn't it?

MR. BARNETT: Yes, it is.

QUESTION: What's the difference in this instance?

MR. BARNETT: May I, Judge?

QUESTION: If the Judge had said --

QUESTION: Can you have a directed verdict in Connecticut in a criminal case?

MR. BARNETT: No, you most certainly cannot.

QUESTION: Never?

MR. BARNETT: Never. And it is my claim that this instruction was the equivalent of a directed verdict.

QUESTION: Can the Judge just dismiss that count?

MR. BARNETT: Well, the question, Your Honor, is why would the Judge dismiss the count? We claim that the count was improperly submitted to the jury because the jury was directed to find on an element of the count of attempted murder.

QUESTION: But it still had to find the other elements, didn't it?

MR. BARNETT: The jury still, as a factual situation, would -- that is the only, the only, element in which the conclusive presumption was used. Everything else was posited as a question of fact.

QUESTION: You haven't said very much about the failure to object here.

MR. BARNETT: No, I haven't, Mr. Chief Justice, because the Connecticut Supreme Court accepted the claim. I have pointed out in my brief in a footnote that the statement in the State v. Ruiz, referred to by Ms. Lager, was not made in the situation of an approved jury instruction. It was made in a comment by the Supreme Court of Connecticut to a claim that the evidence in a certain case was insufficient on which to base a jury's verdict of guilt.

QUESTION: The whole purpose of objections is to give the trial in the courts an opportunity to correct an asserted error, is it not?

MR. BARNETT: Yes, it is, sir. However, this is a case on direct review, and I think the Connecticut Supreme Court was quite proper in treating it under --

QUESTION: Well, as I understand it, Connecticut Supreme Court practice, or at least that's what happened here, was that notwithstanding the failure to object --

MR. BARNETT: That is correct.

QUESTION: -- they were willing to entertain the --

MR. BARNETT: Yes, yes, Justice Brenner.

QUESTION: And did.

MR. BARNETT: It did. And fully considered the issue.

Thank you, Your Honors, for the privilege of addressing you.

CHIEF JUSTICE BURGER: Ms. Lager, do you have anything further?

MS. LAGER: With respect to the claim of the concurrent sentence doctrine that was raised by Mr. Barnett, the sentences, in response to your question earlier, Mr. Chief Justice, were 10 to 20 years on the attempted murder count, 8 to 16 years on the kidnapping count, 10 to 20 years on sexual assault, and 5 to 10 years on the robbery, which was concurrent with the kidnapping offense.

The effect of the Connecticut Supreme Court's reversal is that the the State of Connecticut is obliged to retry the defendant on both the robbery and the attempted murder counts.

QUESTION: Now, you say obliged again.

MS. LAGER: Or can exercise its prosecutorial discretion. As it stands now, the --

QUESTION: But if it does not, the respondent has 15 to 36, is that right?

MS. LAGER: If it does not, the respondent has 18 to 36. But there is certainly no question of mootness with respect to the the State of Connecticut, who is the petitioner today, and the consequences to the the State of Connecticut. And as far as the reversal of the defendant's conviction on those charges, notwithstanding the fact that the sentence was concurrent.

The other point that I would like to make is that harmfulness of an error should not depend on an absolute rule concerning the point in a trial where the error occurred. An evidentiary error may be devastating, and an instructional error may be insubstantial or insignificant. And that is one of the reasons that this Court has announced a harmless-error rule and why it should be appropriately applied in the proper case under the correct circumstances. Thank you.

CHIEF JUSTICE BURGER: Thank you, Counsel. The case is submitted.

(Whereupon, at 3:00 o'clock p.m., the case in the above-entitled matter was submitted.)