OPINION OF THE COURT

BECKER, Circuit Judge.

This bizarre case requires us to determine what, if any, evidence of a criminal defendant's mental abnormality is admissible to prove the defendant's lack of specific intent to commit an offense, following the passage of the Insanity Defense Reform Act of 1984, Pub.L. No. 98-473, Title II, § 402(a), 98 Stat. 2057, § 20, recodified at 18 U.S.C. § 17.

The government contends that such evidence is never admissible to negate specific intent because the Act restricts all such evidence to the jury's consideration of a defendant's legal sanity or insanity. We disagree with this broad contention. We conclude that although Congress intended § 17(a) to prohibit the defenses of diminished responsibility and diminished capacity, Congress distinguished those defenses from the use of evidence of mental abnormality to negate specific intent or any other mens rea, which are elements of the offense. While the contours of the doctrines of diminished responsibility and diminished capacity are unclear, the defenses that Congress intended to preclude usually permit exoneration or mitigation of an offense because of a defendant's supposed psychiatric compulsion or inability or failure to engage in normal reflection; however, these matters do not strictly negate mens rea.

Despite our disagreement with the government's broad contention, we agree that the Congressional prohibition of diminished responsibility defenses requires courts to carefully scrutinize psychiatric defense theories bearing on mens rea. Psychiatrists are capable of supplying elastic descriptions of mental states that appear to but do not truly negate the legal requirements of mens rea. Presenting defense theories or psychiatric testimony to juries that do not truly negate mens rea may cause confusion about what the law requires.

In this case, we believe that the evidence provided by appellant Stephen Pohlot and effectively excluded by the district court from the jury's consideration of mens rea could not, even if believed, demonstrate that Pohlot lacked the specific intent to contract for the killing of his wife. By his own admission, Pohlot engaged in considerable planning and activity, and he finalized an agreement to have his wife murdered. While Pohlot may not psychologically have understood the full consequences of this activity — and in one sense may not have wanted his wife to die — the purpose of his activity was the hiring of someone to kill his wife. Pohlot's request for the jury to consider evidence of mental abnormality other than in the context of insanity therefore amounted to a request for a diminished responsibility defense that Congress has abolished.

We therefore affirm the judgment of conviction.

I. Factual and Procedural History

The following evidence was adduced at trial. Until the summer of 1985, Stephen Pohlot was a successful pharmacist and private investor, living with his wife, Elizabeth, and three of their children in Katonah, New York. According to Pohlot, however, beyond this facade lay a strange set of relationships, dominated by his wife's abuse.

Pohlot testified, for example, that his wife had broken his thumb by crashing a coffee pot down on it; deeply gouged his face with her nails; threatened him with a hunting knife; shot him in the stomach; and often locked him out of their house and bedroom. Pohlot also blamed his wife for the psychiatric illnesses of two of his four children, who were seriously anorexic. Illustrating her behavior, Pohlot said that she had insisted on keeping an enormous number of pets in or about the house: sixty rabbits, six goats, tanks full of fish, tanks full of snakes, a pony, six indoor cats and nine outdoor cats, numerous ducks and dogs, and a variety of birds.

In the summer of 1985, Elizabeth obtained a court order removing Pohlot from their home. In July 1985, she filed for divorce, freezing Pohlot's assets. These events, according to the government, triggered the murder plot, which the government presented chiefly through the testimony of George Neustadt, a friend and business associate of Pohlot and Michael Selkow, the man Pohlot contacted to arrange the murder.

Shortly after Elizabeth filed for divorce, Pohlot told Neustadt that he was thinking of killing his wife. Pohlot said that he had hired a contract killer in Atlantic City, but that the killer had committed suicide. Pohlot asked if Neustadt knew of anyone else in Atlantic City who could arrange the murder. Neustadt thought he might and agreed to try to contact him.

Neustadt then called Selkow, whom he had met at a casino and whom he suspected of having connections to organized crime. Unknown to Neustadt, Selkow had become a government informer. Neustadt communicated the general purpose of his call, and they arranged to meet in Atlantic City. In Atlantic City, in a conversation tape-recorded by the F.B.I., Neustadt explained what Pohlot needed, and Selkow told Neustadt that he would bring a killer in from Italy. Neustadt relayed this representation to Pohlot.

Several days later, Pohlot phoned Selkow, and the two arranged to meet in Philadelphia. In the course of the meeting, which was also tape-recorded, Selkow asked Pohlot why he wanted to have his wife killed. Of relevance to the government's trial theory of financial motivation, Pohlot referred several times to the fact that Elizabeth was aware of money that he had stashed away (apparently in safe deposit boxes). He stated, for example:

You know there's just, there's just no way . . . you know I uh implore her to, let's, let's talk about this. You know, don't, don't be a pig, don't expose any of them, them papers that are drawn, there's the x number of dollars in this box and x number of dollars in that box. And the day that come, that, that it comes, that they open these . . . boxes and they see this kind of money you know it's uh, [unintelligible] gonna be looking over here and, saying, let me have it.

Pohlot also made references to his wife "cheating," to her fabrication of charges of assault to gain the order barring him from their home, and to the problems of his children. Pohlot related the story of having talked to an earlier close friend about killing his wife before the friend committed suicide.

Pohlot and Selkow agreed that Selkow would arrange the murder of Pohlot's wife for $25,000. Pohlot paid $8,000 immediately and then gave Selkow a description of his wife, her cars, and her routine. Pohlot described particularly his wife's appointments with her psychiatrist and the route she travelled to get there. He stressed that his wife's murder had to "appear to be an accident." Selkow explained the plan:

[I]t'll be an accident, and it'll be en route to the shrink. Now what they're, what he's gonna do, this guy's an expert, comes in from Italy. . . . [H]e will be in after Thursday. (Pause) After that it'll take him about, less than a week.

Pohlot stated that "I would like the least amount of exposure," but stressed that he wanted "this thing" done by September 18, 1985 because that was the scheduled date of a court hearing on his wife's abuse allegations. Selkow briefed Pohlot on how to react to the investigations that would follow the killing. They arranged for Pohlot to pay Selkow after the murder by giving an envelope with cash to one of Neustadt's drivers.

Two weeks later, at the FBI's direction, Selkow notified Pohlot that the hitman had arrived. He told Pohlot to stay away from his wife. FBI agents then informed Selkow that they planned to arrest Pohlot and Neustadt and that they would relocate Selkow for his own protection. Selkow, however, was unwilling to "disappear in the middle of the night" because of unresolved personal matters. In an attempt to "buy time" by slowing down the progress of the Pohlot case, he contacted Neustadt and Pohlot and arranged to meet Pohlot at a rest station on the New Jersey Turnpike.

At the turnpike station, Selkow told Pohlot of doubts about the Italian hitman. He stated that he thought his telephone might be bugged. He instructed Pohlot to telephone him that night to call the murder off and thereby throw the authorities off the track. In response to a question from Pohlot, however, he assured Pohlot that the murder would still occur in a couple of weeks once Selkow was sure that they had "the right people."

That night, Pohlot did call Selkow and express his desire to cancel the murder. Selkow played the tape of this phone call to the FBI agents. But the FBI agents did not believe this call. They suspected that Selkow had planned it. They therefore arrested Pohlot and Neustadt the next day and eventually arrested Selkow on charges of obstruction of justice. Pohlot was indicted on five counts of using interstate commerce facilities, i.e., the telephone, in the commission of murder for hire and one count of conspiring to commit that offense. At trial, the government theorized that Pohlot intended to have his wife murdered because of concern that she knew where he had hidden assets from the IRS.

Pohlot disputed few of the government's factual assertions. He did, however, dispute the following. He disagreed that he had proposed the murder initially to Neustadt, contending instead that the idea had "evolved" out of numerous conversations. He also claimed that his statements about discussing his wife's murder earlier with a potential killer were lies. Most importantly, he claimed that Selkow never told him, in their meeting on the New Jersey Turnpike, to telephone and cancel the murder. Instead, Pohlot claimed that he finally "came to his senses," understood that a crime was going to be committed and decided to take action to prevent it.

These limited disagreements with the government alone do not negate the thrust of the government's allegations. The district court instructed the jury, and Pohlot did not object, that even if Pohlot had truly decided on his own to abandon the murder after the New Jersey meeting, he would still be guilty if he had conspired and made telephone calls for the purpose of committing murder for hire. Instead, Pohlot's defense, based primarily on the affirmative defense of insanity, concentrated on claims that his entire experience had been a fantasy, representing an unrealistic attempt to overcome an inability to deal with his wife's abuse.

In the pursuit of this defense, Pohlot brought forth the many examples of abuse by his wife mentioned above. In the face of this abuse, Pohlot testified that he had shown a psychological inability to respond. For example, after his wife shot him, and after his resulting operation, Pohlot stated that he had spoken to his attorney in the hospital as follows:

"I said, `What are my options?' He said, `You can prosecute or you can waive prosecution.' I said, `I have to waive prosecution. I am sure that was an aberration. That was an inadvertent incident and I am very much in love with my wife and I want to waive prosecution and I want her to stay home'"

As a result, Pohlot claimed:

. . . I no longer had a say — I am sure I had a lot of say but it didn't carry any weight in what went on in the house. My function then deteriorated to provide for her way of life and the children's way of life and I had a place to sleep.

This psychological inability to respond, Pohlot claimed, led to the murder plot as "a weak attempt to fight back." But Pohlot hatched the plot to fulfill psychological motivations, not truly to kill his wife. "It was a fantasy," he said.

Buttressing Pohlot's own statements, the defense presented the testimony of a psychiatrist, Dr. Gary Glass. On the basis of a series of meetings with Pohlot, but relying only on Pohlot's version of events, Dr. Glass diagnosed Pohlot as a "compulsive personality, passive dependent personality and passive aggressive personality." Pointing to various events in Pohlot's life history, Glass explained that Pohlot had characteristically shown an inability to assert himself and to gain control of his life. Unable to take decisive action or even to acknowledge problems to himself, "his only way of development so as to have control over his life . . . was to hurt himself." For example, to protest being sent to a different high school, "What he did was essentially to hand in blank papers in his third year until the school was forced into the position of calling his mother and throwing Stephen out of school. . . ."

According to Dr. Glass, this pattern of behavior continued into Pohlot's married life and explained Pohlot's inability to respond to his wife's abusive behavior. As his relationship with his wife became increasingly one of verbal and physical struggles, and as his children became sick, Pohlot became more and more disturbed. Then, the same year as the murder plot, the death of Pohlot's father inflicted an enormous shock. Finally, when his wife served him with divorce papers, Pohlot felt it "was the end of my life." Citing these facts, Dr. Glass presented the following understanding of Pohlot's mental state:

He becomes depressed. He takes to drinking quite heavily. He is unable to sleep. . . . And things go on from there in which he meets with a friend, a business associate and the idea of, "Helping you with your dilemma" begins to surface.

Originally, there was some confusion as to whether or not Stephen really knew what that meant, at first, and then, I suspect there is knowing and there is knowing much like he did not know his wife was having several affairs. He did not really know until it was discussed in the context of psychiatric treatment. Here again he knew it, but he didn't know. Things progressed until the behaviors I am sure you heard of where he did meet a gentleman at a motel in Pennsylvania and transact[ed] a cash transaction for the purpose of, `killing his wife.' Now, I say, `killing his wife' — I will come back to that in a moment. Following that cash transaction, he feels a great relief, much as he did when he left school as a teenager and finally, took control, albeit in a somewhat pathological way of his life. . . .

He also felt uncomfortable and the exhilaration was very short lived, but that was the event. I don't believe that there was anything — the issue there was taking control. . . . [A]ll of these terrible feelings had been building up for years and years. . . .

Finally, he takes a step to do something about it and it's like the sky opens up and the sun shines until he realized, of course, that this is a problem and he tries his best and is fortunate to be able to undo it, but that is the climactic event for him, the transaction, not a murder. He felt in many ways — this may sound a little unusual, but I believe there is truth to it. He felt as if he would hire somebody to kill Betsy and after that happened, they would go home and live together and be happier. . . . [H]e thinks when all of this nonsense is over, they're going to get back together and be happy. . . .

Another analogy I might use to this cash transaction, is similar to an adolescent who is jilted by their boy friend or girl friend and trying all sorts of behaviors to make it right and unable to do so and finally resorts to a carefully orchestrated suicide attempt where they perhaps take a dull knife or an overdose of what is presumed to be a safe medication like aspirin.

At the close of trial, the court permitted Pohlot's counsel to argue to the jury that Pohlot's mental abnormality indicated a lack of mens rea. The district court then proceeded to instruct the jury, without objection, on the elements that the government must prove for the conspiracy count and for the counts of using interstate commerce for the commission of murder for hire.

Describing the necessary mental state, the court stated that for Pohlot to be guilty of conspiracy, he must have "willfully participated" in the conspiracy "with the intent to advance [sic] or some other object or purpose of the conspiracy." "To participate willfully means to act and participate voluntarily intentionally with specific intent to do something the law forbids. . . ." For Pohlot to be guilty of using interstate commerce in the commission of murder for hire, the court stated that Pohlot must have "intended that the murder of his wife be committed."

In considering whether Pohlot possessed the requisite mens rea, the court told the jury, "You may consider any statements made or done or admitted by the Defendant or all the other facts and circumstances and evidence which indicates his state of mind." The court rejected, however, Pohlot's proposed jury instruction that in determining whether the government had met this burden, i.e., of proving specific intent, the jury "may consider the testimony as to the defendant's abnormal mental condition." The parties dispute whether other portions of the court's instruction effectively precluded the jury from considering evidence of mental abnormality on the specific intent issue.

The jury found Pohlot guilty of five counts of using interstate commerce facilities in the commission of a crime of violence, and one count of conspiring to do the same. Pohlot then moved for a judgment of acquittal or for a new trial on the grounds, inter alia, that the court had failed to instruct the jury that it could consider evidence of Pohlot's mental condition in deciding whether Pohlot possessed the requisite mens rea. The court rejected this motion on the grounds that Pohlot was "asserting a diminished capacity defense" that Congress had abolished in the Insanity Defense Reform Act. United States v. Pohlot, Crim. No. 85-00354-01, slip op. at 9-13 (E.D.Pa. March 27, 1986). Pohlot now appeals, presenting solely the mens rea issue.

II. Did the District Court Actually Preclude Evidence on The Issue of Mens Rea?

At the threshold, the government contends that we need not decide whether evidence of mental abnormality is admissible to prove lack of mens rea because the trial court did not actually preclude such a use. The government notes that the district court permitted Pohlot's counsel to argue to the jury that Dr. Glass's testimony indicated a lack of mens rea. And the government points to the court's general instruction that the jury could consider "all the . . . facts and circumstances and evidence which indicates [Pohlot's] state of mind." At one point, the district court even acknowledged that Pohlot's "fantasy defense" was different from the insanity defense and appeared to instruct the jury to consider it:

I think Mr. Pohlot's defense was that [driving away from the Trenton meeting] was the first time he came to realize what he was doing and withdrew as a result of becoming rational, I suppose, as opposed to being legally insane. However, that is a matter that you should consider.

Even assuming that evidence of abnormality was admissible to prove lack of mens rea, the government concludes, Pohlot received the jury instruction he was due, and he had no right to have the evidence of mental abnormality heightened with a particular instruction. Cf. United States v. Castro, 776 F.2d 1118 (3d Cir. 1985), cert. denied, ___ U.S. ___, 106 S.Ct. 1233, 89 L.Ed.2d 342 (1986) (failure to instruct jury to consider psychologist's testimony on defendant's susceptibility to entrapment does not constitute error where court did not instruct jury to ignore evidence).

Pohlot responds that the district court's jury instruction did effectively exclude the psychiatric evidence from the jury's consideration of mens rea. Pohlot also argues that even if the district court did not exclude the evidence, the court committed reversible error by failing to instruct the jury specifically to consider the psychiatric evidence in that context. Pohlot considers this case analogous to cases in which we have reversed trial courts for failing to instruct juries specifically that they may acquit on the basis of character or alibi evidence. See United States v. Logan, 717 F.2d 84 (3d Cir. 1984) (character evidence); United States v. Frischling, 160 F.2d 370, 370-71 (3d Cir. 1947) (same); United States v. Booz, 451 F.2d 719 (3d Cir. 1971), cert. denied, 414 U.S. 820, 94 S.Ct. 45, 38 L.Ed.2d 52 (1973) (alibi evidence).

We need not consider Pohlot's demand for a specific supporting instruction because we agree that the district court's instructions probably led the jury not to consider the psychiatric evidence on the issue of mens rea. We acknowledge the statements mentioned above suggesting the contrary. After describing the insanity defense, however, the court stated:

Mental disease or defect does not otherwise constitute a defense. . . . A mental disease or defect, other than one which renders a Defendant unable to appreciate the nature or quality or wrongfulness of his acts, as I said, does not in itself constitute a defense. (834a)

If the court intended this statement only to mean that mental disease did not constitute any affirmative defense other than insanity, the instruction would not necessarily preclude the use of evidence of mental defect from the jury's consideration of the elements of the offense. See the discussion infra at 897-98. The court followed this statement, however, with a discussion of the psychiatric evidence. Without any instruction to the contrary, therefore, this language suggests that evidence of mental abnormality was relevant only to the issue of insanity.

Later instructions were more specific. After the court had concluded its original instructions, counsel for both parties requested additional instructions on mens rea. Pohlot's counsel reiterated his request for an instruction that psychiatric evidence could prove lack of mens rea. The court rejected Pohlot's suggestion but provided additional instruction on mens rea:

I believe if this was a mistake by the Defendant, and if you believe that somehow it was a mistake by the Defendant, that he did not understand, appreciate what was going on, [then] he didn't have the mens rea. If on the other hand, you think he did understand what was going on as defined, when I say understand what was going on, I mean within the framework of his legal sanity or insanity, then the mens rea has been met.

This instruction conflated the issues of mens rea and the issues of the insanity defense and suggested that the evidence of mental abnormality could prove only insanity.

Although the issue is a close one, we conclude that the instructions as a whole had the probable effect of excluding evidence of Pohlot's mental abnormality from the jury's consideration of mens rea. If this evidence was admissible to prove a lack of mens rea, the court's instruction "was capable of confusing and thereby misleading the jury." United States v. Fischbach and Moore, Inc., 750 F.2d 1183, 1195 (3d Cir. 1984), cert. denied, 470 U.S. 1029, 105 S.Ct. 1397, 84 L.Ed.2d 785 (1985). We therefore proceed to a consideration of whether that exclusion was proper.

III. Effect of the Insanity Defense Reform Act

A.

(a) Affirmative defense. — It is an affirmative defense to a prosecution under any Federal statute that, at the time of the commission of the acts constituting the offense, the defendant, as a result of a severe mental disease or defect, was unable to appreciate the nature and quality or the wrongfulness of his acts. Mental disease or defect does not otherwise constitute a defense.

(b) Burden of proof. — The defendant has the burden of proving the defense of insanity by clear and convincing evidence.

18 U.S.C. § 17. Through this section, Congress deleted the "volitional prong" of the commonly accepted Model Penal Code approach which had permitted acquittal if the defendant "as a result of mental disease or defect . . . lacks substantial capacity . . . to conform his conduct to the requirements of law." Model Penal Code § 4.01 (1962). Congress also shifted the burden of proof, requiring the defendant to prove insanity by clear and convincing evidence.

The government claims that, in addition to these changes, the Act also bars a defendant from using evidence of mental abnormality to negate mens rea. The government relies on the language stating that apart from insanity "mental disease or defect does not otherwise constitute a defense." The Report of the Senate Judiciary Committee explained that this language was:

intended to insure that the insanity defense is not improperly resurrected in the guise of showing some other affirmative defense, such as that the defendant had a `diminished responsibility' or some similarly asserted state of mind which would serve to excuse the offense and open the door, once again, to needlessly confusing psychiatric testimony.

S.Rep. No. 98-225, 98th Cong., 2d Sess. 229 (1984), reprinted in 1984 U.S.Code Cong Ad.News 3182, 3411 (hereinafter Senate Report) (emphasis supplied).

In the government's view, a claim that a defendant "lacked specific intent due to his mental abnormalities is tantamount to a `diminished capacity' defense," which is in turn equivalent to a "diminished responsibility" defense. Government's Brief at 19. The government draws support from Muench v. Israel, 715 F.2d 1124, 1143 (7th Cir. 1983), cert. denied, 467 U.S. 1228, 104 S.Ct. 2682, 81 L.Ed.2d 878 (1984), in which the court stated:

[T]he courts have used the labels diminished responsibility, diminished capacity, and other nomenclature merely as a shorthand for the proposition that expert evidence of mental abnormalities is admissible on the question of whether the defendant in fact possessed a particular mental state which is an element of the charged offense. . . . When a court rejects the doctrine of diminished capacity, it is saying that psychiatric evidence is inadmissible on the mens rea issue.

Other courts have similarly referred to the use of expert psychiatric evidence to negate mens rea as a "diminished capacity" or "diminished responsibility" defense. The government therefore reasons that § 17(a)'s prohibition of diminished responsibility defenses precludes Pohlot from using evidence of mental abnormality for the purpose he requested.

Pohlot disagrees. He contends that the use of psychiatric evidence to negate mens rea is not a "diminished capacity" or "diminished responsibility" defense. As he views it, the Insanity Defense Reform Act only aims to preclude "affirmative defenses" based on mental disease, not the use of mental disease to negate an element of the offense. He therefore claims that the Act did not justify the district court's restriction of psychiatric evidence to the insanity defense.

B.

We first consider the government's claim that evidence of mental abnormality is never admissible to negate mens rea. We disagree. Both the wording of the statute and the legislative history leave no doubt that Congress intended, as the Senate Report stated, to bar only alternative "affirmative defenses" that "excuse" misconduct not evidence that disproves an element of the crime itself. See United States v. Gold, 661 F. Supp. 1127 (D.D.C. 1987) (insanity act does not bar psychiatric evidence on issue of mens rea); United States v. Frisbee, 623 F. Supp. 1217 (N.D.Cal. 1985) (same).

Pohlot essentially contends that "mental disease or defect is admissible whenever it is relevant to prove that the defendant did or did not have a state of mind that is an element of the offense." Model Penal Code, § 4.02(1) (1962). Although this principle has sometimes been phrased as a version of the "diminished capacity defense," it does not provide any grounds for acquittal not provided in the definition of the offense. Properly understood, it is therefore not a defense at all but merely a rule of evidence. See Arenella, The Diminished Capacity and Diminished Responsibility Defenses: Two Children of a Doomed Marriage, 77 Colum.L.Rev. 827, 833 (1977); Morse, Undiminished Confusion in Diminished Capacity, 75 J.Crim.L. Criminology 1, 5-7 (1984). As several United States Courts of Appeals have therefore stressed, "[t]he use of expert testimony for this purpose is entirely distinct from the use of such testimony to relieve a defendant of criminal responsibility based on the insanity defense or one of its variants, such as diminished capacity." United States v. Demma, 523 F.2d 981, 986 n. 14 (9th Cir. 1975); see also United States v. Staggs, 553 F.2d 1073, 1075 (7th Cir. 1977); United States v. Bennett, 539 F.2d 45, 53 (10th Cir.) cert. denied, 429 U.S. 925, 97 S.Ct. 327, 50 L.Ed.2d 293 (1976); United States v. Brawner, 471 F.2d 969, 998-1002 (D.C. Cir. 1972); Rhodes v. United States, 282 F.2d 59, 60-61 (4th Cir. 1960), cert. denied, 364 U.S. 912, 81 S.Ct. 275, 5 L.Ed.2d 226 (1960).

Staggs, 553 F.2d 1073, provides an example of how psychiatric evidence may help negate mens rea without raising an additional ground of defense. In that case, the defendant, Staggs, faced charges of aggravated assault on the claim of a policeman that the defendant had threatened to shoot if the policeman approached him. Staggs contradicted that testimony and sought to introduce the testimony of a psychiatrist that Staggs' mental state made him unlikely to threaten the policemen. Reversing the district court, the court of appeals ordered that the jury hear the proffered expert evidence.

Because admitting psychiatric evidence to negate mens does not constitute a defense but only negates an element of the offense, § 17(a) by its terms does not bar it. Section 17(a) states only that "mental disease . . . does not otherwise constitute a defense;" it does not purport to establish a rule of evidence.

C.

The legislative history supports this plain reading. Congress heard and relied on the testimony of leading academic commentators who related the distinctions between a strict mens rea approach and defenses of diminished responsibility or partially diminished capacity. Citing one of these commentators, the House Judiciary Committee stated in a report to the House accompanying a bill that was substantially the same as the Senate bill which eventually became law:

The use of mental disorder to negate mental state elements of crimes should not be confused with the `diminished capacity' defense as developed by the California courts during the 1960's and 1970's. Under that doctrine, a defendant could escape responsibility for a crime by demonstrating not that he or she lacked a required specific intent, but rather that his or her capability of entertaining that intent was not, because of mental disorder, commensurate with that of nondisordered persons.

H.R.Rep. No. 98-577, 98th Cong. 1st Sess. 15 n. 24 (1983) (hereinafter House Report) ( citing Arenella, supra). Describing that provision of the bill prohibiting the creation of affirmative defenses, other than insanity, based on mental abnormality, the Report explained:

Since the current insanity defense is derived from case law, the Committee is concerned that additional defenses based on mental disorders could be developed by the courts in order to circumvent the tighter requirements developed by Congress. Thus, the bill provides that the Committee's test constitutes the only affirmative defense based on mental disorder that will be applicable in Federal courts. Mental disorders will remain relevant, of course, to the issue of the existence of a mental state required for the offense, such as the specific intent required for certain crimes. This accords with current practice.

House Report at 14 (emphasis supplied; footnote omitted).

The Senate similarly distinguished between diminished responsibility or diminished capacity defenses, particularly as recognized by California cases, and the use of evidence to negate mens rea. As Senator Orrin Hatch stated on the Senate floor:

Some critics of the mens rea approach, confusing the issue of incapacity to harbor the requisite mens rea with the doctrine of "diminished responsibility" recognized in a line of California court decisions, contend that its adoption would result in a significant expansion of the scope of psychiatric testimony. Actually, the California doctrine bears little if any relevance to the issue of capacity to form criminal intent that may be relevant under the mens rea approach.

130 Cong.Rec. S418-19 (daily ed. Jan. 30, 1984).

In light of this awareness, the language of the Senate Report is instructive. The report states that § 17(a) was intended to insure only that evidence of mental disease will not resurrect the insanity defense "in the guise of showing some other affirmative defense." Senate Report at 229 (emphasis supplied). It also refers to "` diminished responsibility' or some similarly asserted state of mind which would serve to excuse the offense." Id. (emphasis supplied). These word choices indicate that the Senate Judiciary Committee wished to abolish only diminished responsibility and capacity defenses not to abolish the use of psychiatric evidence to disprove mens rea.

A later portion of the Senate Report confirms this view of the Committee's intent. In addition to changing the insanity defense, the Insanity Defense Reform Act altered Federal Rule of Evidence 704 to prevent psychiatric experts from providing opinions on ultimate issues. The Report explained: "The Committee has fashioned its Rule 704 provision to reach all such `ultimate' issues, e.g., premeditation in a homicide case, or lack of predisposition in entrapment." Id. at 231 (emphasis supplied). Premeditation is, of course, an element of mens rea. If the Judiciary Committee had intended § 17(a) to prevent psychiatrists from testifying about mens rea at all, its report would almost certainly not have indicated its approval of psychiatric evidence on mens rea.

We need not rely only on the language of § 17(a) and on the House and Senate reports. The entire structure of the Congressional debate suggests that Congress did not intend to bar evidence of mental abnormality to prove a lack of mens rea. After a jury acquitted John S. Hinckley of attempting to kill President Reagan, there was widespread support in Congress, in the Administration and in medical and legal professional organizations for some reform of the insanity defense. Many members of Congress wished to preserve some affirmative insanity defense but to delete the "volitional prong" of the Model Penal Code approach, see supra at 896, and to shift the burden of proof. Other members of Congress, backed by the Justice Department, wished to abolish the insanity defense altogether. Even those favoring abolition, however, wished to preserve the defendant's right to use psychiatric evidence to prove lack of mens rea, and their bills explicitly did so.

For example, Congressman Bill McCollum, who favored abolition, stated on the House floor:

There is a longstanding principle in criminal law that no one can be convicted of any crime without having the requisite and necessary mental state of mind. If that person does not have the mental state of mind because he or she is suffering from a mental disease or defect, then there is no way the prosecution can gain a conviction. Mr. Speaker, it is true that it is rarely argued or used today in criminal law. That is so because some years ago we gave the accused a second bite at the apple; that is the opportunity to present a so-called insanity defense as an affirmative defense, regardless of and in addition to the mental state of mind question that the prosecutor has to prove. My proposal . . . would be to abolish the second bite at the apple, do away with the separate insanity defense and allow some expert testimony on the question of the accused having a mental disease or defect as it bears on the state of mind issue. . . .

130 Cong.Rec. H9674 (daily ed. Sept. 18, 1984).

Eventually, the abolition approach did not triumph. Congress preserved a limited affirmative insanity defense. It did so, however, not because of doubts about the use of evidence to negate mens rea, but rather because it felt that concerns about the dangers of an insanity defense were overstated and because abolition "would alter that fundamental basis of Anglo-American criminal law: the existence of moral culpability as a prerequisite for punishment." House Report at 7-8. In light of this legislative history, it would be ironic to view the acceptance of an affirmative insanity defense as a rejection of psychiatric evidence on mens rea.

D.

The government argues that even if Congress did not explicitly intend § 17(a) to bar evidence of mental abnormality on the issue of mens rea, we must do so to preserve an important element of the reform. The Insanity Defense Reform Act not only altered the definition of insanity, but it shifted the burden of proof to the defendant to prove that affirmative defense by clear and convincing evidence. Congress shifted this burden because of testimony that proving a defendant sane beyond a reasonable doubt was virtually impossible. The government contends that admitting evidence to prove a lack of mens rea effectively places the burden of proving a defendant's sanity back on the government.

Only in the rare case, however, will even a legally insane defendant actually lack the requisite mens rea purely because of mental defect. As the House Report stated: "Mental illness rarely, if ever, renders a person incapable of understanding what he or she is doing. Mental illness does not, for example, alter the perception of shooting a person to that of shooting a tree." House Report at 15 n. 23. Similarly, a man who commits murder because he feels compelled by demons still possesses the mens rea required for murder. The government's burden of proving mens rea is therefore considerably less onerous than its previous burden of proving sanity.

On the other hand, evidence of mental abnormality may help indicate a lack of mens rea even when a defendant is legally sane. In Staggs, discussed supra, the defendant wished to use psychiatric evidence to bolster his claim that he did not threaten the policeman and therefore did not commit assault. Staggs did not claim to be insane and probably could not have done so successfully. Limiting evidence of mental abnormality only to the issue of insanity might therefore not merely shift a burden of proof but also completely preclude evidence that is relevant to guilt or innocence.

Finally, the government's request would require us to raise a serious constitutional question that Congress explicitly determined to avoid. Under In re Winship, 397 U.S. 358, 363-64, 90 S.Ct. 1068, 1072, 25 L.Ed.2d 368 (1970), due process requires that the government prove every element of a criminal offense beyond a reasonable doubt. The defendant's right to present a defense to one of those elements generally includes the right to the admission of competent, reliable, exculpatory evidence, and the Supreme Court has struck down "arbitrary rules that prevent whole categories of defense witnesses from testifying." Washington v. Texas, 388 U.S. 14, 22, 87 S.Ct. 1920, 1925, 18 L.Ed.2d 1019 (1967) (striking down Texas statute prohibiting testimony of defendant's alleged accomplice); see also Crane v. Kentucky, 476 U.S. 683, 106 S.Ct. 2142, 90 L.Ed.2d 636 (1986) (reversing Kentucky conviction for excluding evidence of circumstances surrounding confession); Chambers v. Mississippi, 410 U.S. 284, 299-302, 93 S.Ct. 1038, 1047-49, 35 L.Ed.2d 297 (ordering new Mississippi trial because, inter alia, trial court excluded adverse witness' out of court confession). Evidentiary rules that would bar the testimony of the defendant himself, as would a rule barring all evidence of mental abnormality on the issue of mens rea, need particular justification. See Rock v. Arkansas, ___ U.S. ___, 107 S.Ct. 2704, 97 L.Ed.2d 37 (1987) (striking down state rule preventing defendant from testifying on issues previously the subject of his hypnosis). In light of these cases, a rule barring evidence on the issue of mens rea may be unconstitutional so long as we determine criminal liability in part through subjective states of mind.

Furthermore, the mere fact that a defendant has the right to introduce psychiatric evidence in support of the affirmative defense of insanity does not justify barring the evidence from negating the government's case in chief. The Supreme Court has indicated that although a state may constitutionally shift the burden of proving insanity to the defendant, Leland v. Oregon, 343 U.S. 790, 72 S.Ct. 1002, 96 L.Ed. 1302 (1952), it did not sanction, and probably would not sanction, a jury charge that prevented a jury from considering evidence of mental abnormality in determining whether the state had proven premeditation and deliberation beyond a reasonable doubt. In upholding the verdict, the Court stated: "It is apparent that the jury might have found appellant to have been mentally incapable of the premeditation and deliberation required to support a first degree murder verdict . . . and yet not have found him to have been legally insane." Id. at 794, 72 S.Ct. at 1005.

More recently, in an analogous case, the Supreme Court has held that a state may place the burden on a defendant of proving self-defense. See Martin v. Ohio, ___ U.S. ___, 107 S.Ct. 1098, 1101-02, 94 L.Ed.2d 267 (1987). The Court indicated, however, that a state's right to shift the burden on self-defense does not include the right to prevent a defendant from showing self-defense in an effort to prove that she did not act with the mens rea of "prior calculation and design." Id. at 1102. The exclusion of such evidence "would relieve the state of its burden and plainly run afoul of Winship's mandate" that the prosecution prove each element of the offense beyond a reasonable doubt. Id. at 1102 (citing In re Winship, 397 U.S. at 364, 90 S.Ct. at 1072).

Rather than challenging these Constitutional arguments, Congress, supported by the Justice Department, appears explicitly to have embraced them. In the words of Attorney General William French Smith, the abolition of the insanity defense coupled with the continued admission of psychiatric evidence to negate mens rea:

would abolish the insanity defense to the maximum extent permitted under the Constitution. . . . Under any approach, the Government will always be required to prove every element of the statutory offense that is charged. This includes any specific intent or knowledge required by the statute. In the rare case, therefore, in which a defendant is so deranged that, for example, he did not know that he was shooting a human being, one of the elements of the offense could not be proved — the mental element of mens rea — and he could not be convicted under current law or under any constitutionally supportable change in the law.

Hearings on Bills to Amend Title 18 to Limit the Insanity Defense, Senate Judiciary Committee, 97th Cong., 2d Sess. 30 (1982) (Serial No. J-97-126); see also 1982 House Hearings, supra note 5, at 25 (almost identical testimony by Associate Attorney General Rudolph Giulliani). Accepting this view, the House Judiciary Committee stated:

By distinguishing the affirmative defense of insanity from the narrow mens rea/mental state requirements, the Committee's approach meets the constitutional requirement that the prosecutor prove all elements beyond a reasonable doubt while placing on the defendant the burden of demonstrating a reason for exculpation that presumes the existence of these elements.

House Report at 14.

We do not decide the constitutionality of any Congressional attempt to bar evidence of mental abnormality from the issue of mens rea. The constitutional issues are sufficiently substantial, however, that we are unwilling to create a rule of evidence that would raise them in the absence of explicit Congressional direction. See NLRB v. Catholic Bishop of Chicago, 440 U.S. 490, 501, 99 S.Ct. 1313, 1319, 59 L.Ed.2d 533 (1979) (in absence of clear Congressional direction, court should construe statute to avoid serious constitutional question). For all the foregoing reasons, we therefore reject the government's contention that the Insanity Defense Reform Act either explicitly or implicitly bars a defendant from introducing evidence of mental abnormality on the issue of mens rea.

IV. Differentiating Mens Rea from Diminished Responsibility

Although we reject the government's broader argument, the Senate Report makes clear that § 17(a) does preclude defenses akin to partially diminished capacity or diminished responsibility. Senate Report at 229 (§ 17(a) abolishes "diminished responsibility" defense); see also 130 Cong. Rec. S 425 (daily ed. Jan. 30, 1984) (comments of Senator Laxalt) (§ 17(a) abolishes "diminished capacity" defense). The Senate Report indicates disapproval in this context not just of the creation of actual technical defenses but also of presenting the jury with "needlessly confusing psychiatric testimony." As the House Report and floor statements of Senator Hatch indicate, see supra typescript at 22-23, Congress focused its disapproval on the version of diminished responsibility adopted by the California courts in the 1960's and 1970's.

As the conflicting cases cited above indicate, the terms "diminished responsibility" and "diminished capacity" do not have a clearly accepted meaning in the courts. To the extent that American courts have adopted cognate doctrines, they generally have done so sub silentio. By reference to the careful work of many commentators heard and cited by Congress, however, we believe that we can identify the "diminished responsibility" defenses that Congress intended § 17(a) to prohibit. See Arenella, supra (cited in House Report at 15 n. 24); Morse, supra (cited in House Report on other issues at 11, 16 n. 28); see also Dressler, Reaffirming the Moral Legitimacy of the Doctrine of Diminished Capacity: A Brief Reply to Professor Morse, 75 J.Crim.L. Criminology 953 (1984); Note, Diminished Capacity and Diminished Responsibility: Irreconcilable Doctrines Confused in State v. Wilcox, 14 Toledo L.Rev. 1399 (1983).

The first variant of what courts have called "diminished capacity" defenses inappropriately is the evidentiary doctrine which we have already noted is not a defense at all but merely a rule of evidence; specifically, the admission of evidence of mental abnormality to negate mens rea.

A second strain of diminished capacity permits a defendant to show not only that he lacked the mens rea in the particular case but also that he lacked the capacity to form the mens rea. Whether a defendant has the capacity to form mens rea is, of course, logically connected to whether the defendant possessed the requisite mens rea. Commentators have agreed, however, that only in the most extraordinary circumstances could a defendant actually lack the capacity to form mens rea as it is normally understood in American law. See Arenella, supra, at 834. Even the most psychiatrically ill have the capacity to form intentions, and the existence of intent usually satisfies any mens rea requirement. Commentators have therefore argued that permitting evidence and arguments about a defendant's capacity to form mens rea distracts and confuses the jury from focusing on the actual presence or absence of mens rea. See Morse, supra, at 44-45 and Arenella, supra, at 863.

Finally, commentators have identified defenses generally categorized as "diminished responsibility" or "partially diminished capacity." Pure "diminished responsibility" exists in many European countries as a formal defense. It "permits the jury to mitigate the punishment of a mentally disabled but sane offender in any case where the jury believes that the defendant is less culpable than his normal counterpart who commits the same criminal act." Arenella, supra, at 829; see also Morse, supra, at 21-22. Although such a formal defense is largely unknown in the United States, many legal doctrines may work similarly. At least one controversial decision of the California Supreme Court, discussed below, effectively adopted such a defense.

In addition to explicit doctrines of diminished responsibility, commentators have contended that courts may permit juries to excuse a defendant's criminal conduct because of mental abnormality and thereby covertly create a "partially diminished capacity" defense when they admit evidence of mental abnormality that does not truly negate mens rea. See Arenella, supra, at 844-49; Morse, supra, at 42-43. In some cases, defendants have attempted to present such theories of defense as the claim that a defendant lacked the mens rea to distribute cocaine because of the psychological domination of her mother. See United States v. White, 766 F.2d 22 (1st Cir. 1985). In others, psychiatrists have proved willing to testify that a defendant who planned and executed a murder did not possess the requisite mens rea because of a mental disease such as chronic schizophrenia. See, e.g., Commonwealth v. Tempest, 496 Pa. 436, 437 A.2d 952 (1981). Mens rea is generally satisfied, however, by any showing of purposeful activity, regardless of its psychological origins. Such testimony may therefore mislead the jury about the mens rea requirements of a crime.

The progression of cases in the California Supreme Court illustrates the distinctions between these various defenses.

In People v. Wells, 33 Cal.2d 330, 202 P.2d 53 (1949), the California Supreme Court considered the conviction of a prisoner for assaulting a prison guard with "malice aforethought," a more serious offense than simple assault. The defense had sought through expert testimony to show that the prisoner misinterpreted and overreacted to external stimuli to such an extent that he mistakenly believed he was defending himself. The trial court rejected this testimony, but the California Supreme Court, following a strict mens rea approach, held that this evidence should have been admitted. Obviously, if the defendant thought he was acting in self-defense, he could not have planned the crime in advance and therefore could not have acted with the mens rea of "malice aforethought." The court carefully distinguished between evidence showing that the defendant did not possess the required mental state and evidence showing the defendant incapable of entertaining the mental state.

In People v. Gorshen, 51 Cal.2d 716, 336 P.2d 492 (1959), however, the Court abandoned the distinction between evidence of intent and capacity to form intent and held admissible evidence of whether the defendant "could not and therefore did not deliberate" about committing a murder. 51 Cal.2d at 731, 336 P.2d at 502. The case also interpreted the mens rea requirement of premeditation and deliberation for first degree murder to require considered reflection. As a result, the Court ordered a retrial at which the defendant could use psychiatric evidence of his paranoid schizophrenia to show that he could not have reflected sufficiently to be guilty of first degree murder.

Finally, in People v. Wolff, 61 Cal.2d 795, 821, 394 P.2d 959, 975, 40 Cal.Rptr. 271, 287 (1964), the California Supreme Court effectively recognized a diminished responsibility defense. The defendant in that case had carefully planned and executed a series of murders and rapes. Despite unanimous psychiatric testimony to the contrary, the jury found the defendant sane. On appeal, the California Supreme Court considered whether the evidence had shown the defendant to be guilty of first or second degree murder. The distinction, the court held, turned on the "quantum of personal turpitude of the offenders" reflected in the extent to which the offender could meaningfully reflect upon the gravity of their acts. 40 Cal.Rptr. at 288, 394 P.2d at 976. Because the defendant's mental abnormality prevented him from realizing "the enormity of the evil," the court found that he was not guilty of first degree murder. In this way, the court made mental illness that did not directly correlate to any particular element of mens rea a ground for reducing the severity of an offense.

Other California cases took a similar approach. See, e.g., People v. Poddar, 10 Cal.3d 750, 111 Cal.Rptr. 910, 916, 518 P.2d 342, 348 (1974); People v. Conley, 64 Cal.2d 310, 322, 49 Cal.Rptr. 815, 822, 411 P.2d 911, 918 (1966). These decisions eventually led a broad alliance of political forces through referendum to abolish the defense of diminished capacity. The new law abolished the use of evidence to show that a defendant lacks the capacity to form mens rea. The law, however, explicitly permitted the use of evidence of mental disease to show that a defendant actually lacked mens rea. See Cal.Penal Code § 28 (as amended) (The amendment restricted use of this evidence only to specific intent crimes.).

As the California experience points out, the strict use of psychiatric evidence to negate mens rea may easily slide into wider usage that opens up the jury to theories of defense more akin to justification. In rejecting the "diminished responsibility" defenses as accepted by California courts, Congress barred these theories of justification from consideration by juries. Only the first of our typology of "defenses" is permissible, namely the use of evidence to prove that a defendant actually lacked mens rea.

In light of the strong danger of misuse, we join other circuits that have directed district courts to examine proffered psychiatric testimony carefully "to determine whether the proof offered is grounded in sufficient scientific support to warrant use in the courtroom, and whether it would aid the jury in deciding the ultimate issues." Bennett, 539 F.2d at 53; see also Demma, 523 F.2d at 986; Brawner, 471 F.2d at 1002. As this case shows, however, expert psychiatric evidence is not the only evidence of mental abnormality bearing on mens rea, for a defendant or other witnesses may testify about mental abnormality. Courts should also be careful in deciding whether to issue jury instructions or to permit defense arguments directing the jury to consider whether any evidence of mental abnormality negates mens rea. Notions of intent, purpose and premeditation are malleable and at their margins imprecise. But the limits of these concepts are questions of law. District courts should admit evidence of mental abnormality on the issue of mens rea only when, if believed, it would support a legally acceptable theory of lack of mens rea. In deciding such a question, courts should evaluate the testimony outside the presence of the jury. Brawner, 471 F.2d at 1002.

We note one particular area of caution relevant to this case. As Professor Morse theorizes, psychiatric testimony suggesting that a defendant lacks mens rea may often focus not on the defendant's intent but on the defendant's awareness of intent:

At nearly all times, human beings are conscious of themselves, they perceive and are aware of what they are doing as they do it. . . . This self-reflective split in consciousness that allows self-monitoring is an important regulator of behavior, for it provides constant feedback that allows us to correct maladaptive behaviors. . . . [Is the state of lacking self-awareness a state] in which mens rea is lacking? On the one hand, the defendant knows at some level what he is doing and intends to do it; on the other hand, he is not fully conscious of his actions in the usual sense. I believe that this situation is better handled as a matter of affirmative defense. Mens rea is present but the usual control structures are compromised.

Morse, supra, at 46-47 (footnotes omitted).

Professor Morse's point is that a lack of self-reflection does not mean a lack of intent and does not negate mens rea. We agree.

Criminal responsibility must be judged at the level of the conscious. If a person thinks, plans and executes the plan at that level, the criminality of his act cannot be denied, wholly or partially, because, although he did not realize it, his conscious was influenced to think, to plan and to execute the plan by unconscious influences which were the product of his genes and his lifelong environment.

State v. Sikora, 44 N.J. 453, 210 A.2d 193, 202 (1965).

V. Application of the Standard to This Case

As we construe the Insanity Defense Reform Act, Congress did not intend to bar all evidence of mental abnormality from the jury's consideration of mens rea, but it did require the exclusion of evidence that does not support a legally acceptable theory of a lack of mens rea. Deciding whether Pohlot's evidence was admissible under that standard, while superficially difficult, is ultimately straightforward. Pohlot testified that all his efforts to hire Selkow to murder his wife were "a fantasy." Dr. Glass testified that Pohlot "did not really know" what hiring Selkow meant until it was discussed in the context of psychiatric treatment. Dr. Glass testified, in essence, that Pohlot's real hope was not to kill his wife but to live with her again.

All of these statements, taken out of context, could suggest that Pohlot did not know the consequences of his plotting and did not intend to achieve the death of his wife. Because the district court instructed the jury that Pohlot could be found guilty only if he "intended that the murder of his wife be committed," these statements at first appear to be relevant and admissible.

Rather than focusing on Pohlot's conscious mind, however, the testimony focused on what Pohlot "really" knew. Pohlot acknowledged that he was attempting "to fight back." Dr. Glass agreed that Pohlot transacted a cash transaction "for the purpose of killing his wife." Similarly, Dr. Glass testified that Pohlot knew in some senses what he was doing, "but he didn't know." Thus, according even to the defense, whatever Pohlot "really" knew and intended, at least a substantial part of his mind knew and intended exactly what were the natural consequences of his plotting.

Placed in the context of the remaining evidence, it is clear that Pohlot acted with considerable awareness of what he was doing and with considerable purpose. Pohlot engaged in careful activity, over a lengthy period of time, sending first a deputy and then engaging in several phone calls and personal meetings all directed at one purpose: hiring someone to murder his wife. Pohlot came to a firm agreement, made payment, and expressed concern and took action to assure that the crime was not ultimately traced to him.

In the context of the facts, both Pohlot's own testimony and that of Dr. Glass relate clearly not to Pohlot's intent in a legal sense but to Pohlot's meaningful understanding of his actions and their consequences. We often act intending to accomplish the immediate goal of our activity, while not fully appreciating the consequences of our acts. But purposeful activity is all the law requires. When one spouse intentionally kills the other in the heat of a dispute, he or she will rarely at that moment fully appreciate the consequences of the murder. The spouse is guilty of homicide nonetheless.

Taken in context, suggestions that Pohlot did not intend Selkow to commit murder focuses not on Pohlot's conscious mind but on his unconscious. To accept this theory as a defense to mens rea requires manipulation of the concept of intent beyond what the "intent" element of 18 U.S.C. § 1952A requires. Pohlot therefore offered his evidence of mental abnormality in support of a legally unacceptable theory of lack of mens rea that amounts covertly to a variation of the partially diminished capacity defense precluded by § 17(a). Whether the district court applied this correct analysis or accepted the incorrect broader view advocated by the government, it was correct in instructing the jury under the circumstances not to consider this evidence in deciding whether Pohlot possessed the requisite mens rea.

The judgment of the district court will be affirmed.