United States v. William Bennett

460 F.2d 872, 148 U.S. App. D.C. 364, 1972 U.S. App. LEXIS 11771

• Court: Court of Appeals for the D.C. Circuit
• Decided: January 19, 1972
• Docket: 24387
• Status: Published

Opinions

BAZELON, Chief Judge:

Convicted of a sexual assault on a thirteen-year-old boy, appellant raises important questions concerning the insanity defense he asserted at his trial. While our test of criminal responsibility is currently under reconsideration en banc,1 we can resolve appellant’s contentions without awaiting a decision on the substantive definition of the test. His arguments are not directed at the definition of responsibility, but rather at the procedures through which the defense — whatever its precise definition— must be channeled.² We hold that appellant was denied a full and fair hearing on his insanity claim and therefore that he should have a second opportunity to persuade a jury that they should not hold him criminally responsible for his action.

Appellant was charged on July 10, 1968, with taking indecent liberties with a minor child in violation of D.C.Code § 22-3501 (a), and with committing sodomy on a male person under the age of sixteen in violation of D.C.Code § 22-3502. On August 23, 1968, he was committed to St. Elizabeths Hospital for a sixty-day examination into his competency to stand trial and his mental condition at the time of the alleged offense. On the basis of the hospital’s report, the trial court found Bennett competent and he was brought to trial before a jury on March 4, 1969. At trial, Bennett denied that he had committed the acts charged, but relied primarily on a responsibility defense. The jury rejected that defense and returned a verdict of guilty on the sodomy count. Bennett received a sentence of four to fifteen years, and he brought this appeal. He argues that the conviction must be reversed because (1) government psychiatrists failed to disclose information which might have supported his insanity defense, (2) the trial court erred in denying him a bifurcated trial, and (3) the trial court erred in admitting over objection his confidential communications with a government psychiatrist.

I.

Two psychiatrists testified for the defense that Bennett was suffering from a mental disease at the time of the alleged offense.³ In rebuttal, the government offered the testimony of two psychiatrists who had examined appellant at St. Elizabeths and who maintained that Bennett was a malingerer without any mental disease or defect. The government psychiatrists pointed out that they had observed appellant continually for a substantial period of time, and they disparaged the testimony of defense psychiatrists by alluding to the defense psychiatrists’ limited opportunity for observation of appellant.⁴ What the government witnesses failed to disclose, however, was that at St. Elizabeths appellant was receiving daily doses of major tranquilizers which could be expected to abate many of the symptoms of acute mental disturbance, and appellant insists that the failure to disclose this fact was reversible error. He points out that the medication could have had a significant impact on his behavior, and hence on the psychiatrists’ description of his behavior, and that the jury should therefore have been advised of the impact the drugs may have had.

Records at St. Elizabeths Hospital indicate that at the time of his commitment for observation in connection with the trial that led to this appeal, appellant was administered substantial doses of Thorazine and Stelazine, both anti-psychotic agents, for more than a month.⁵ The drug treatment was discontinued on October 15, 1968, just one day before the staff conference at which Bennett was found to be without mental disease or defect and competent to stand trial.

Without referring to the drug therapy, Dr. Nicola Kunev of St. Elizabeths defended his conclusion that Bennett lacked mental disease by describing the absence of any symptoms of mental illness. He explained to the jury that at the hospital Bennett was “in very good contact with us, he [was] orientated as to why we were with him, what the problems are, what the difficulty that he is facing. He has good recollection of his latest day in St. Elizabeths Hospital. What happened in the hospital for the two months that he was there * * * * ”⁶ In the same vein, the other government psychiatrist, Dr. Mauris Platkin, told the jury that Bennett was without mental disease or defect because he lacked “any symptomology which in my opinion would indicate he was suffering from a mental illness.” ⁷ Both psychiatrists contrasted Bennett’s behavior at the time of their examination with the behavior of a person they would consider mentally ill.

Yet at no point did the government psychiatrists make clear to the jury that major tranquilizers administered to Bennett could have obliterated the “symptomology of mental illness,” and that they were administered for precisely that purpose. Hostility and uncooperativeness are thought to be especially amenable to drug therapy.⁸ Within two weeks major tranquilizers may end hallucinations and delusional ideas,⁹ and under medication “[t]he sullen, sarcastic and antagonistic patient is less irritable and frequently becomes quiet, cooperative, and accessible.”¹⁰ There is thus a striking similarity between Bennett’s behavior as described by the government psychiatrists and the behavior which the drugs could be expected to induce.¹¹ Accordingly, the conviction must be reversed because the government’s experts withheld information that calls into question the validity of their entire testimony, and that might— if disclosed — have led the jury to disregard their stated conclusions.¹² Since their testimony was the only evidence offered by the government in opposition to Bennett’s insanity defense, it can hardly be disputed that the jury might have reached a different result if the crucial information had been properly presented.

In reversing appellant’s conviction, we recognize that trial counsel made no effort to explore the issue below even though he arguably knew of the medication.¹³ But we find it impossible to ascribe fault to appellant or trial counsel, who may have been unaware of the significance of the major tranquilizer medication. Lawyers are not expected to be informed about the effects of anti-psychotic drugs, but we can hardly assume that the government psychiatrists failed to understand their potential impact. On remand, the government’s experts must disclose all of the factual information which underlies their conclusions.

The defect in the expert testimony at issue here is especially disquieting not only because of the importance of the information withheld, but also because of the special nature of the insanity defense. It should be obvious that a jury cannot make the “intertwining moral, legal, and medical judgments”¹⁴ on which the resolution of an insanity defense depends if it has been deprived of a piece of information that might substantially alter its evaluation of expert testimony. Without this information, which may be crucial to an understanding of appellant’s behavior and behavioral controls, the jury can do little more than choose between the competing labels proffered by the prosecution and the defense. Yet we have repeatedly emphasized that “neither the court nor the jury is bound by ad hoc definitions or conclusions as to what the experts state is a disease or defect.”¹⁵ Where psychiatrists offer testimony in eonclusory terms and fail to reveal all of the crucial underlying facts, the jury’s dependence on the experts is substantially increased and the underlying purposes of the insanity defense are correspondingly jeopardized.¹⁶ For the almost inevitable result will be a hopeless jumbling of the uniquely psychiatric elements of the determination of responsibility with the legal and moral elements of that determination, leading the jury to the mistaken conclusion that a defendant’s responsibility turns on whether or not the experts have given his condition a name and the status of disease. We have repeatedly emphasized, however, that

there is more to a determination of responsibility than psychiatric conclusions, unanimous or not. The gravity of an impairment and its relevance to the acts charged are both questions of degree, which can only be resolved with reference to the community’s sense of when it is just to hold a man responsible for his act.¹⁷

If the evidence presents an incomplete or misleading representation of a defendant’s behavior, the jury will almost invariably abdicate its responsibility and acquiesce in the conclusions of the most persuasive experts. And the process suffers no less than the defendant when crucial information is withheld and the jury’s function is disrupted. Accordingly, we reverse and remand for a new trial so that all of the available information relevant to the responsibility defense can be presented to the jury.

II.

If appellant raised no further claim of error, it would not be necessary on remand to redetermine whether or not he committed the acts in question. Only his insanity defense would be at issue. But appellant does argue that his entire defense was prejudiced by the intermingling at trial of his insanity defense and his defense on the merits. The prejudice, he asserts, was of two types. First, he argues that the introduction of evidence on the merits prejudiced his insanity defense because the jury was exposed to the “unpleasant details” of the “shocking sexual assault on a 13-year-old boy” with which appellant was charged.¹⁸ Second, his defense on the merits was allegedly prejudiced by the introduction of evidence, ostensibly relevant only to the issue of sanity, which was tantamount to a confession of guilt by appellant. The evidence in question was the testimony of Dr. Kunev to the effect that Bennett’s sanity was evidenced by his “very good recollection of the events of the alleged events. He recalls minutely what happened prior to the offense, of the alleged offense, and following it. He expressed his own version of the story, his feeling about it * * *."¹⁹ Appellant maintains that both types of prejudice could have been cured if the trial court had granted him a bifurcated trial.

Bifurcation lies in the first instance within the “sound discretion” of the trial court.²⁰ And we have repeatedly held that the trial court does not abuse its discretion in refusing bifurcation where the defendant does not present a substantial defense both on the merits and on the issue of responsibility.²¹ Given that standard, Bennett’s bifurcation claim would ordinarily fail, for he offered nothing more on the merits than a bare denial that he had committed the offense.

Bennett’s claim does present special difficulty, however, because his objection to the admission of Dr. Kunev’s testimony rests on a constitutional and statutory foundation. He maintains that the testimony should have been excluded either under the fifth amendment bar against self-incrimination, or under 18 U.S.C. § 4244 (1970), which provides that:

[n]o statement made by the accused in the course of any examination into his sanity or mental competency provided for by this section, whether the examination shall be with or without the consent of the accused, shall be admitted in evidence against the accused on the issue of guilt in any criminal proceeding.²²

We have no difficulty concluding that Dr. Kunev’s testimony amounted to the admission into evidence of a “statement made by the accused in the course of” an examination. The troublesome point, of course, is whether the statement was admitted “on the issue of guilt”²³ or only on the issue of sanity. Dr. Kunev undoubtedly intended to address his remarks to the question of sanity, but the jury could easily find his testimony pertinent to their determination of the merits. The statute does not clearly determine the admissibility of statements made by the accused where the statements bear on both the merits and sanity. And to our knowledge, no court has ever squarely resolved the precise question.²⁴

In light of the statutory prohibition, and the fifth amendment bar which underlies it,²⁵ it seems clear that the defense was, in fact, prejudiced by the admission of a confession which could obviously be expected to have an impact on the jury’s determination of the merits. Under these circumstances, the trial court might ordinarily have declared a mistrial or, at the very least, have offered a limiting instruction that the jury could not consider the statements in deciding whether or not Bennett had committed the. acts in question.²⁶ But the trial judge took neither approach, apparently because he believed that appellant had himself conceded the case on the merits. We cannot accept appellant’s vague and essentially incoherent remarks on the witness stand²⁷ as a confession of guilt or a waiver of his statutory and constitutional privileges,²⁸ and we must therefore decide how the situation should have been handled.

Obviously, a limiting instruction cannot be expected to remove all traces of prejudice where the jury is told, in effect, that the defendant confessed to his crime.²⁹ Even the most conscientious juror will have enormous difficulty forgetting a confession when turning from the insanity issue to the case on the merits, and a limiting instruction is thus an unsatisfactory solution. But a mistrial also has great and obvious disadvantages, and it seems especially wasteful where the entire problem could have been fairly and efficiently avoided. If Bennett’s trial had been bifurcated, there would have been no prejudice. Dr. Kunev’s testimony on the insanity issue could have been presented to the jury in a separate insanity trial after the case on the merits had already been decided. Since the government concedes that his testimony had no relevance whatsoever to the merits, it can hardly object if the merits are resolved before his testimony is introduced. Bifurcation may, to be sure, be somewhat inefficient. But where inefficiency is the sole objection,³⁰ and the alternatives — a dubiously effective limiting instruction or a grossly inefficient mistrial- — are plainly more troublesome, bifurcation emerges as the best possible solution to this otherwise difficult problem.³¹

Of course, the trial court could not have predicted at the outset of Bennett’s first trial that permitting Dr. Kunev to testify on the issue of insanity before the merits had been determined would prejudice the defense on the merits. We emphasize, therefore, that the failure to order a bifurcated trial was not reversible error. But with the aid of hindsight it is clear that bifurcation would have avoided significant prejudice to the defense, and we see no reason why Bennett should be denied a bifurcated trial on remand.³² In subsequent cases, trial judges may decline to grant bifurcation only to encounter the same difficulty that arose here. Since a limiting instruction will not effectively dispel the prejudice and since a mistrial is a patently unattractive result, they may find it advisable, when making an initial ruling on the bifurcation issue, to forestall the entire problem by determining in advance whether the government psychiatrists intend to disclose statements of the accused which could prejudice the defense on the merits. Counsel for defendant could attempt to obtain the same information through pre-trial discovery in order to support a motion for a bifurcated trial.³³

In some eases this process of inquiry may persuade the trial court to bifurcate the trial even though the defense on the merits is minimal and bifurcation would not normally have been granted. But that result is by no means undesirable, since bifurcation can eliminate any question of prejudice and since it is not likely under these circumstances to be especially inefficient. After all, if the defense on the merits is not substantial, the merits portion of the trial should be capable of swift resolution.

There remains, however, Bennett’s second claim of prejudice. He asserted, as we noted above, that his insanity defense was prejudiced by the presentation to the jury of the “shocking details” of the crime with which appellant was charged. We recognize that a jury may be reluctant to find a defendant not guilty by reason of insanity if the crime at issue is especially heinous. In some cases, therefore, the government’s presentation of its case on the merits is indeed likely to prejudice an insanity defense. But bifurcation alone cannot prevent that result unless the two parts of the trial are presented to different juries. It would surely be unreasonable to expect a jury to ignore the lurid details of the crime when turning to the insanity defense even if the defense on the merits had already been resolved. Thus, a defendant can avoid this problem only by stipulating all of the facts in the government’s ease on the merits, or by persuading the trial court to convene separate juries for the two parts of the bifurcated trial.

We have previously held that trial courts do have discretion to impanel two separate juries in a bifurcated trial.³⁴ But it seems entirely clear under our case-law that Bennett is not entitled as of right to a trial before two juries.³⁵ The inefficiency of impanelling two separate juries is plainly maximized where the defense on the merits is slight and its resolution is likely to be rapid.

Thus, we hold that Bennett’s conviction must be reversed and his case remanded for a bifurcated trial on his insanity defense and his defense on the merits. The trial court may, in its discretion, impanel two juries, but it would not abuse its discretion if it refused to enter such an order. It should be noted, however, that Bennett does have the option of waiving his defense on the merits so that his trial on remand would concern only his responsibility defense.³⁶ By exercising that option, appellant could insure that the jury will be able to consider his responsibility defense without being exposed to the allegedly prejudicial testimony concerning the “shocking sexual assault” which the government submitted at his first trial. In addition, the government has the option of cutting back Dr. Kunev’s testimony so as to eliminate any reference to appellant’s pretrial statements. In that case, Dr. Kunev’s testimony would no longer prejudice the defense, and bifurcation would thus no longer be required to avoid that prejudice.

Reversed and remanded.

MeGOWAN, Circuit Judge, concurs in the result only of Part I, and concurs in the Court’s opinion in Part II.

1 . United States v. Brawner, No. 22,714.

2 . Cf, United States v. Trantham, 145 U.S.App.D.C. 113, 118, 448 F.2d 1036, 1041 (1971) (statement of Bazelon, C. J. as to why he would grant rehearing en hone). For the argument that problems in the administration of the insanity defense are far more important than the language of the test itself, see Brief for Prof. David Chambers as Amicus Curiae, United States v. Brawner (lodged June 28, 1971).

3 . Dr. Goldberg, who had examined appellant on several occasions prior to the time of the alleged offense and who had twice signed papers to have Bennett hospitalized for mental illness, testified that Bennett’s symptoms led to a diagnosis of acute, undifferentiated schizophrenia. He described those symptoms in detail to the jury. Although Dr. Goldberg’s last examination of appellant took place more than a year before the time of the alleged offense, he maintained that it was highly unlikely that appellant’s illness would have disappeared during the course of the intervening year. The second defense psychiatrist, Dr. Kirby, examined Bennett once after the time of the alleged offense. He interviewed Bennett in jail, and on the basis of Bennett’s symptoms — which he described to the jury — he concluded that Bennett suffered from a mental illness and should be transferred to St. Elizabeths.

4 . See, e. g., Transcript at 300, 319-323. Cross-examining the defense psychiatrists, counsel for the government attempted to drive home the same point. See United States v. Schappel, 144 U.S.App.D.C. 240, 445 F.2d 716 (1971); Rollerson v. United States, 119 U.S.App.D.C. 400, 343 F.2d (1964); cf. United States v. Leazer, No. 24,799, 148 U.S.App.D.C. - at -, 460 F.2d 864 at 870, (1972) (Bazelon, C. J., concurring).

5 . Letter from Dr. Elizabeth Strawinsky, Acting Associate Director for Forensic Programs, St. Elizabeths Hospital, to J. G. Dail, counsel for appellant, reproduced at Brief for Appellant at Appendix A:

* * * Thorazine 100 mgms b.i.d. and Stelazine 5 mgms b.i.d. were started on September 9. On September 17, Thorazine was increased to 200 mgms b.i.d. and Stelazine to 10 mgms b.i.d. and both were discontinued on October 15.

While this letter is plainly not part of the record before us, it should be noted that the government raises no objection to what it terms an “evasion of the record.” On the contrary, the government points out that inclusion of the letter was “unnecessary because, contrary to the way appellant now chooses to interpret it, the record is replete with testimony indicating that appellant received medication while at St. Elizabeths (see Tr. 263, 313, 351). In fact appellant himself testified to this effect (Tr. 188).” Brief for Appellee at 15 n. 25. Since our holding could be grounded either on the somewhat opaque record references to which the government alludes or on the detailed information — not contested by the government — in Dr. Strawinsky’s letter, we need not decide how the case would be resolved if our holding rested only on extra-record information. Compare Suggs v. United States, 132 U.S.App.D.C. 337, 341 n. 5, 407 F.2d 1272, 1276 n. 5 (1969) (per curiam); 132 U.S.App.D.C. at 343 n. 6, 407 F.2d at 1278 n. 6 (Bazelon, C. X, dissenting). It is clear, however, that none of the references to medication in the record came close to making the jury understand that defendant had been administered major tranquilizers at the time of his psychiatric examination, and they certainly could not have indicated to the jury the drugs’ potential impact on his behavior.

6 . Tr. 237-38.

7 . Tr. 296.

8 . Cole, Goldberg & Davis, Drugs in the Treatment of Psychosis: Controlled Studies 157, in Psychiatric Drugs (P. Solomon ed.) (1966).

9 . Kolb, Noyes’ Modern Clinical Psychiatry 393 (7th ed. 1968).

10 . Id. See generally D. Klein & J. Davis, Diagnosis & Drug Treatment of Psychiatric Disorders (1969).

11 . For example, Dr. Kunev testified that Bennett was a “perfect patient” who did not want to return to jail and who could turn inappropriate behavior off and on in his effort to convince the staff that he was mentally ill. It was only when trying to appear ill, Dr. Kunev stated, that Bennett was uncooperative. Tr. at 233. Dr. Plat-kin indicated in his testimony that Bennett was “well aware of what was going on around him and was capable of making intelligent and rational responses if he wanted to.” Tr. at 302.

12 . See Rollerson v. United States, 119 U.S.App.D.C. 400, 343 F.2d 269 (1964); Carter v. United States, 102 U.S.App.D.C. 227, 252 F.2d 608 (1957); cf. United States v. Amburgey, 189 F.Supp. 687, 692-693 (D.D.C.1960); J. Maguire, J. Weinstein, J. Chadbourn, J. Mansfield, Cases & Materials on Evidence 262-264 (5th ed, 1965); Rheingold, The Basis of Medical Testimony, 15 Vand.L.Rev. 473 (1962).

13 . Appellant argues that reversal is required by Brady v. Maryland, 373 U.S. 83, 83 S.Ct. 1194, 10 L.Ed.2d 215 (1963) and Levin v. Clark, 133 U.S.App.D.C. 6, 408 F.2d 1209 (1967), on the grounds that the government unlawfully withheld information favorable to the defense. The government, on the other hand, contends that “it cannot be seriously argued that the Government somehow suppressed this information,” since “both appellant and his trial counsel knew that he received the medication.” Brief for Appellee at 15 n. 25. See note 5 supra. In our view, the conviction must be reversed irrespective of Brady and Levin, and it is therefore unnecessary to resolve this dispute.

14 . United States v. Eichberg, 142 U.S.App.D.C. 110, 111, 439 F.2d 620, 621 (1971); Adams v. United States, 134 U.S.App.D.C. 137, 142, 413 F.2d 411, 416 (1969); King v. United States, 125 U.S.App.D.C. 318, 324, 372 F.2d 383, 389 (1967).

15 . McDonald v. United States, 114 U.S.App.D.C. 120, 312 F.2d 847 (1962) (en banc). In that case we defined mental illness, for purposes of the insanity defense, as any abnormal condition of the mind which substantially affects mental or emotional processes and substantially impairs behavior controls, and we pointed out that testimony relevant to the jury’s determination would concern the “development, adaptation and functioning of these processes and controls.” 114 U.S.App.D.C. at 124, 312 F.2d at 851. See also Washington v. United States, 129 U.S.App.D.C. 29, 390 F.2d 444 (1967); Rollerson v. United States, 119 U.S.App.D.C. 400, 402, 406, 343 F.2d 269, 271 275 (1964); Carter v. United States, 102 U.S.App.D.C. 227, 236, 252 F.2d 608, 617 (1957). Compare A. Goldstein, The Insanity Defense 104 (1967).

16 . In Washington v. United States, 129 U.S.App.D.C. 29, 390 F.2d 444 (1967), we concluded that eonclusory expert testimony on one aspect of the insanity defense so substantially impairs the jury’s ability to carry out its function that eonclusory testimony on that issue should henceforth be barred. Nevertheless, we affirmed the conviction notwithstanding our dismay over the use of eonclusory testimony, pointing out that the jury “obtained enough concrete information to preclude us from disturbing the verdict. The defense psychiatrist and, on cross-examination, the Government psychiatrists gave some meaningful descriptions of defendant’s mental and emotional processes * * * *” 129 U.S.App.D.C. at 35, 390 F.2d at 450. That rationale cannot be carried over to save Bennett’s conviction. For here, as opposed to Washington, the government psychiatrists presented information without crucial explanation and qualification, and the failure to explain and qualify may well have misled the jury in its resolution of Bennett’s insanity defense. It is surely no answer that defense psychiatrists presented information which supported Bennett’s claim. Their testimony was discredited by the government on the grounds that they had had limited opportunity to observe appellant, see p. 874, supra, and the jury may, as a result, have placed very great emphasis on the data provided by government psychiatrists.

17 . United States v. Eichberg, 142 U.S.App.D.C. 110, 113, 439 F.2d 620, 623 (1971) (Bazelon, C. J., concurring); see United States v. Leazer, No. 24,799, 148 U.S.App.D.C. -, at -, 460 F.2d 864, at 869 (1972) (Bazelon, C. J., concurring).

18 . Brief for appellant at 20.

19 . Tr. at 224-225.

20 . Contee v. United States, 133 U.S.App.D.C. 261, 262, 410 F.2d 249, 250 (1969), citing Holmes v. United States, 124 U.S.App.D.C. 152, 154, 363 F.2d 281, 283 (1966) .

21 . See, e. g., Harried v. United States, 128 U.S.App.D.C. 330, 389 F.2d 281 (1967). But cf. United States v. Grimes, 137 U.S.App.D.C. 184, 186-188, 421 F.2d 1119, 1121-1123 (1969).

22 . Bennett’s examination at St. Elizabeths was not carried out pursuant to § 4244, but under a similar provision in the District of Columbia Code, D.C.Code § 24r-301. While that section does not contain the same exclusionary rule embodied in § 4244, we have held that the § 4244 prohibition applies even where the commitment is carried out under § 24-301. See Edmonds v. United States, 104 U.S.App.D.C. 144, 147, 260 F.2d 474, 477 (1958) (en banc). See also Battle v. Cameron, 260 F.Supp. 804 (D.D.C.1966).

23 . It could be argued, of course, that a defendant who lacks responsibility is not guilty, and therefore the “issue of guilt,” as that phrase is used in the statute, should be taken to embrace not only the merits but also the issue of sanity. The precise relationship between guilt and responsibility lias long been the subject of debate, however, and it is not clear that the issue of sanity does fall within the issue of guilt. See generally Louisell and Hazard, Insanity as a Defense: The Bifurcated Trial, 49 U.Calif.L.Rev. 805, 805-806 (1961). In any case, the evident purpose of § 4244 was to permit statements of the accused to be admitted into evidence only on the issue of sanity, and not on the issue of whether or not the defendant committed the acts charged. See Judicial Conference of the District of Columbia, Report of the Committee on Problems Connected With Mental Examination of the Accused in Criminal Cases, Before Trial 110 n. 1 (1966).

24 . See Holmes v. United States, 124 U.S.App.D.C. 152, 154 n. 7, 363 F.2d 281, 283 n. 7 (1966). Statements by the accused relevant to the merits, as well as to sanity, have been admitted where the defendant clearly conceded that he had committed the acts in question and the merits were thus no longer in issue. See, e. g., Ashton v. United States, 116 U.S.App.D.C. 367, 324 F.2d 399 (1963); Bailey v. United States, 101 U.S.App.D.C. 236, 248 F.2d 558 (1957), cert. denied, 355 U.S. 919, 78 S.Ct. 351, 2 L.Ed.2d 279 (1958). Statements by the accused relevant to the merits have been held inadmissible where the statements had no conceivable relevance to sanity. See Edmonds v. United States, 104 U.S.App.D.C. 144, 260 F.2d 474 (1958). While it appears that no court has resolved the question, it has been argued that statements simultaneously relevant to the merits and sanity should not be held admissible:

We assume that a colloquy such as this would not pass muster under § 4244 even if accompanied by the clearest kind of instruction that it could be considered only on the issue of the defendant’s sanity: “Doctor, what makes you believe the defendant is sane?” “The fact that he described how he killed his wife in just the nervous guilty way any normal person would.”

Judicial Conference Report, supra note 31, at 111 n. 1.

25 . See generally Thornton v. Corcoran, 132 U.S.App.D.C. 232, 236-238, 407 F.2d 695, 699-671 (1969); United States v. Driscoll, 399 F.2d 135, 139 (2d Cir. 1968); State v. Whitlow, 45 N.J. 3, 210 A.2d 763 (1965).

26 . See United States v. McClain, 142 U.S.App.D.C. 213, 440 F.2d 241 (1971); Jones v. United States, 128 U.S.App.D.C. 36, 385 F.2d 296 (1967). It has always been assumed in this situation that a defendant would be entitled, at a minimum, to a limiting instruction. See, e. g., Model Penal Code § 4.09, Comment (Proposed Final Draft No. 1, 1961).

27 . Asked if he had any statement to make to the judge or jury, Bennett said:

Well, other than I am not denying the fact that something actually happened, but as far as I am concerned, that it was democratic and no matter how it happened, I was high on jive. Like I told the doctor, Dr. Kunev, he got on a green coat and a sweater and balding.

Tr. at 191.

28 . Compare, e. g., Hall v. United States, 410 F.2d 653, 661 (4th Cir.), cert. denied, 396 U.S. 970, 90 S.Ct. 455, 24 L.Ed.2d 436 (1969); Bailey v. United States, 101 U.S.App.D.C. 236, 248 F.2d 558 (1957), cert. denied, 355 U.S. 919, 78 S.Ct. 351, 2 L.Ed.2d 279 (1958).

29 . Cf. Bruton v. United States, 391 U.S. 123, 88 S.Ct. 1620, 20 L.Ed.2d 476 (1968); Sims v. United States, 132 U.S.App.D.C. 111, 405 F.2d 1381 (1968); Nash v. United States, 54 F.2d 1006, 1007 (2d Cir. 1932) (limiting instruction is a “recommendation to the jury of a mental gymnastic which is beyond, not only their powers, but anybody’s else.”) H. Kalven & I-I. Zeisel, The American Jury 127-128, 177-180 (1966).

The efficacy of a limiting instruction in this context has generally been viewed with skepticism. See, e. g., Model Penal Code § 4.09, Comment (Proposed Final Draft No. 1, 1961) citing Mass.Gen. Laws ch. 233, § 23B; Judicial Conference Report, supra note 23, at 111 n. 1. Cf. United States v. Albright, 388 F.2d 719, 726 n. 9 (4th Cir. 1968). But see State v. Whitlow, 45 N.J. 3, 210 A.2d 763, 773 (1965).

In response to this precise problem the Model Penal Code was amended to add an optional provision, borrowed from the Massachusetts statute cited above, barring the use of a statement by the accused even as to the issue of defendant’s mental condition, where “such statement constitutes a confession of guilt of the crime charged.” Plainly, under that standard the introduction of Dr. Kunev’s testimony would have led to a mistrial and not merely a limiting instruction.

30 . Cf. Bruton v. United States, 391 U.S. 123, 131 & n. 6, 88 S.Ct. 1620, 20 L.Ed.2d 476 (1968). Compare Harris v. New York, 401 U.S. 222, 91 S.Ct. 643, 28 L.Ed.2d 1 (1971), where the government asserted an interest in having certain evidence introduced for an admittedly limited purpose and where no procedural device, such ns bifurcation, was available to permit the government to accomplish its objective without prejudice to the defendant, with Bruton v. United States, supra, where the government’s sole interest in having the confession of one defendant admitted at the trial of a co-defendant was to avoid the inefficiency of bifurcation.

31 . See Holmes v. United States, 124 U.S.App.D.C. 152, 154 n. 7, 363 F.2d 281, 283 n. 7 (1966):

[U]nless the two issues [sanity and commission of the acts charged] are tried separately, the accused might be unwilling to give to the examining psychiatrists information relevant to his mental condition but which supports belief that he committed the offense. And defense counsel may be deterred by the threat of prejudice from eliciting at trial information relevant to sanity from psychiatric witnesses or from the accused. A bifurcated trial would avoid these impediments to full disclosure and presentation.

Cf. Mass.Gen.Laws ch. 233, § 23B, supra note 29. Compare the rule we apply to joint trials of multiple defendants. Bruton v. United States, 391 U.S. 123, 88 S.Ct. 1620, 20 L.Ed.2d 476 (1968); United States v. Gambrill & Hunter, 146 U.S.App.D.C. 72, 83, 449 F.2d 1148, 1159-1163 (1971); Sims v. United States, 132 U.S.App.D.C. 111, 405 F.2d 1381 (1968).

32 . Although trial counsel did not specifically request bifurcation in objecting to Dr. Kunev’s testimony, we need not decide whether the failure to declare a mistrial and grant a bifurcated trial was l>lain error under Fed.R.Crim.P. 52(b). For the reasons stated in part I of this opinion we held that Bennett should have a second trial on the insanity issue. And since the trial court failed to offer even a limiting instruction with regard to Dr. Kunev’s testimony, it is clear that Bennett must also have a second opportunity to present to a jury his defense on the merits. To avoid any further prejudice, the trial on remand should be bifurcated so that the defense on the merits can be heard and resolved before the insanity issue is raised.

33 . Under Fed.R.Crim.P. 16(a) a defendant may be able to inspect his own written or recorded statements or confessions which are in the possession of the government, and also results or reports of mental examinations made in connection with the particular case and which are in the custody of the government. Thus, defendant might be able to obtain the pertinent information under one or both of these two approaches. A defendant’s ability to obtain this information would be enhanced by adoption of the liberalized discovery standards contained in the proposed amendments to rule 16. See generally Preliminary Draft of Proposed Amendments to Rules of Criminal Procedure for the United States District Courts, 48 F.R.D. 547, 587-610 (1970).

It is normally defendant’s responsibility, of course, to raise the issue of bifurcation, Parman v. United States, 130 U.S.App.D.C. 188, 399 F.2d 559, cert. denied, 393 U.S. 858, 89 S.Ct. 109, 21 L.Ed.2d 126 (1968); Harried v. United States, 128 U.S.App.D.C. 330, 389 F.2d 281 (1967), and defendant carries the burden of demonstrating the need for bifurcation. Higgins v. United States, 130 U.S.App.D.C. 331. 401 F.2d 396 (1968).

34 . Parman v. United States, 130 U.S.App.D.C. 188, 190-191, 399 F.2d 559, 561-562, cert. denied, 393 U.S. 858, 89 S.Ct. 109, 21 L.Ed.2d 126 (1968); Holmes v. United States, 124 U.S.App.D.C. 152, 363 F.2d 281 (1966).

35 . See Parman v. United States, 130 U.S.App.D.C. 188, 191-192, 399 F.2d 559, 562-563, cert. denied, 393 U.S. 858, 89 S.Ct. 109, 21 L.Ed.2d 126 (1968). See also United States v. Huff, 409 F.2d 1225, 1228 (5th Cir. 1969); cf. Spencer v. Texas, 385 U.S. 554, 567-568, 87 S.Ct. 648, 17 L.Ed.2d 606 (1967).

36 . A waiver of the defense on the merits does, of course, raise difficult questions, but they must be answered by the trial court when, and if, appellant chooses this approach. See United States v. Brown, 138 U.S.App.D.C. 398, 428 F.2d 1100 (1970).