Summary
affirming on narrower grounds; trial court held pathological gambling, as matter of law, could not form basis of ALI insanity defense
Summary of this case from United States v. DavisOpinion
No. 82-5340.
Argued November 3, 1983.
Decided August 9, 1984.
Jodie A. English, Asst. Federal Public Defender, Baltimore, Md. (Fred Warren Bennett, Federal Public Defender, Baltimore, Md., on brief), for appellant.
J. Frederick Motz, U.S. Atty., Baltimore, Md. (Larry A. Ceppos, Asst. U.S. Atty., Baltimore, Md., on brief), for appellee.
Appeal from the United States District Court for the District of Maryland.
Before WINTER, Chief Judge, and PHILLIPS and ERVIN, Circuit Judges.
The dispositive issue on this appeal is whether a pathological gambling disorder can serve as the basis for an insanity defense to charges of unlawful entry of a bank with intent to commit robbery or larceny. The district court held that it could not, and on this basis excluded evidence of the condition from jury consideration. On defendant's appeal from his ensuing conviction, we affirm, though on a somewhat different basis than that relied upon by the district court.
I
Jay Dennis Gould was indicted on two counts of unlawful entry of a bank with intent to commit robbery and unlawful entry of a bank with intent to commit larceny, in violation of 18 U.S.C. §§ 2113(a) and (f). Gould pleaded not guilty to both counts, and in advance of trial, duly noted his intention to assert a defense based on insanity.
Following the completion of several pretrial mental examinations and evaluations of Gould, the government filed a motion in limine seeking to exclude from jury consideration any evidence of Gould's pathological gambling disorder as the basis for an insanity defense. The court deferred ruling on the motion, indicating that the proffered evidence would be received subject to a ruling in the form of an appropriate jury instruction. Faithful to this indicated course, the court allowed extensive expert opinion related to Gould's pathological gambling disorder, but at the conclusion of all the evidence excluded it from jury consideration in the course of its instructions. Those instructions, in pertinent part, were as follows:
The defendant in this case asserts the defense of insanity. So, in addition to proving beyond a reasonable doubt the evidence of the offenses charged, the government must also prove beyond a reasonable doubt the sanity of the defendant, that is, that the defendant, at the time of the defense was [sic] charged did not have a mental disease or defect as defined in the American Law Institute test or that despite the mental disease or defect from which he suffered, he had substantial capacity both to appreciate the wrongfulness of his conduct and to conform his conduct to the requirements of the law.
As to pathological gambling as an insanity defense, the court further instructed the jury.
that as a matter of law pathological gambling disorder is not a disease or defect within the meaning of the American Law Institute test which would constitute a basis for the insanity defense. In other words, the mere fact that one may suffer from a pathological gambling disorder and may wish to obtain money with which to gamble does not in and of itself make that person incapable of conforming his conduct to the requirements of the law; and you should disregard any evidence in this case to the contrary, as a matter of law.
The jury returned a verdict of guilty on both counts of the indictment. The defendant was sentenced to concurrent terms of 10 years on each count. This appeal followed.
II
The substantive "insanity defense" rule in this circuit is the ALI test as adopted in United States v. Chandler, 393 F.2d 920 (4th Cir. 1968) (en banc) (Haynsworth, J.):
(1) A person is not responsible for criminal conduct if at the time of such conduct as a result of mental disease or defect he lacks substantial capacity either to appreciate the criminality of his conduct or to conform his conduct to the requirements of law.
(2) The term "mental disease or defect" do[es] not include an abnormality manifested only by repeated criminal or otherwise anti-social conduct.
Id. at 926.
Procedurally, the "defense" of insanity, of non-responsibility under the ALI test, is only injected into a case by the defendant's production of sufficient evidence to raise it; "slight" evidence suffices for this purpose; and once the issue is raised, the burden is cast upon the government to prove beyond a reasonable doubt that the defendant is "sane," i.e., responsible under the ALI test. Hall v. United States, 295 F.2d 26, 28 (4th Cir. 1961) (test of sufficiency adopted pre-ALI rule); see also United States v. McGirr, 434 F.2d 844, 849 (4th Cir. 1970) (test applied post-ALI rule).
That these are the relevant rules for decision here is not challenged. The dispositive issue is only whether the district court properly excluded consideration of the evidence proffered here under these substantive and procedural-evidentiary tests.
A.
Two other circuits, the Eighth and the Second, recently have held in closely analogous cases that, under the ALI test, evidence of a compulsive gambling disorder could not be considered as the basis for an insanity defense to charges of theft-type federal offenses. See United States v. Torniero, 735 F.2d 725 (2d Cir. 1984) (interstate transportation of stolen property); United States v. Lewellyn, 723 F.2d 615 (8th Cir. 1983) (embezzlement; false statement; mail fraud). In both cases, the courts reserved decision upon the general question whether the compulsive gambling disorder might under any circumstance and in respect of any offenses constitute a "mental disease or defect" within the ALI substantive test. Each focused more narrowly upon whether the evidence sufficed under the appropriate procedural-evidentiary test to establish the requisite causal connection between the "disorder" and the specific offenses in issue, a causal connection that is required by the "as a result of" language in the ALI test. Torniero, at 731-32; Lewellyn, at 617-618. Both courts also concentrated only upon the "volitional" prong of the ALI test since in both cases, as in the instant case, the defense was proffered only in that respect. Torniero, at 730; Lewellyn, at 616.
Though a thoroughgoing philosophical determinism would presumably hold cognition and volition to be equally "determined" and human responsibility equally absent as to each, street learning is clearly more ready to find the will than the cognition determined by "uncontrollable" forces. Hence, presumably, the more frequent reliance by criminal defendants on the volitional prong. Hence also, presumably, the basis for recent judicial and congressional uneasiness specifically with the volitional prong. The Fifth Circuit, sitting en banc, has recently abandoned the volitional prong of its ALI test, retaining the cognitive. See United States v. Lyons, 731 F.2d 243 (5th Cir. 1984) (en banc). And two bills currently pending in Congress would, inter alia, eliminate the volitional prong as applied in federal courts. See S. 1762, 98th Cong., 2d Sess. Title IV (1984); H.R. 3336, 98th Cong., 2d Sess. (1984).
The gist of the Eighth Circuit's rationale was that, applying the test of Frye v. United States, 293 F. 1013, 1014 (D.C. Cir. 1923), the expert testimony offered to establish the requisite causal link between the gambling disorder and the defendant's volition in respect of the "collateral" criminal offenses there charged did not sufficiently show that the underlying scientific principle had gained "general acceptance" in the relevant discipline. Lewellyn, at 618.
The Second Circuit, analyzing the problem as one of foundational relevance of the proffered evidence, upheld a district court's exclusion of comparable expert testimony on the basis that it did not show that the "hypothesis" of a causal link between this disorder and commission of the type offenses there charged commanded "substantial acceptance" in the relevant discipline. Torniero, at 731.
While there are undoubtedly shadings of conceptual difference between us, we basically agree with the approaches which led both of these courts to reject compulsive gambling as the basis for an insanity defense to criminal offenses collateral to gambling. In particular, we agree with the Second Circuit that the question whether a novel insanity defense of this type may properly be considered is properly first addressed as one of relevance under Fed.R.Evid. 401, applying a special test appropriate to the peculiar nature of this wholly exculpating defense. Our reasons are as follows.
B
In Chandler, Judge Haynsworth made two wise and prescient observations about the insanity defense in general. First, he noted that in adopting this undoubtedly improved substantive standard we "ought not to fall into the egregious error of the last century"; that "[a]pproval of any standard . . . need not, and should not, freeze the language or solidify thought." Id. at 926-27. This was simply to say that so long as the general principle of exculpation by reason of insanity exists in the criminal law, courts must be prepared to adapt the principle and its application to emerging scientific knowledge about the mysterious workings of the human mind in its cognitive and volitional elements. Cf. Torniero, at 734-35.
His other observation emphasized, however, that at some point the process of adaptation and expansion of the insanity defense might well have to yield notion scientific grounds but on the basis of overriding policy considerations of society's interests in protection of its members. A "qualified freedom of will" on the part of persons charged with crime must, for purely prudential reasons, continue to be assumed; not every "psychiatric explanation" for "deviant conduct" that might be forthcoming over time must, when first advanced, be accepted by the law. 393 F.2d at 926.
The resulting problem for courts is that highlighted in this case: how to strike a proper balance between overly rigid assumptions about the scope of criminal defendants' qualified freedom of cognition and volition and too ready an acceptance of new scientific hypotheses suggesting hitherto unrealized further qualifications upon that freedom. We believe that this difficult policy determination is best made within the framework of a threshold relevance assessment pursuant to Fed.R.Evid. 401, and generally in line with the Second Circuit's analysis in Torniero.
On this basis we hold that the proper test of foundational relevance is whether the general scientific hypothesis of a putative causal relation between specific disorder and specific conduct has substantial acceptance in the relevant discipline. Only if that test of evidentiary relevance is met need the inquiry proceed to the further question of the sufficiency of the particular evidence to meet the defendant's burden of producing sufficient evidence of his own "insanity" or lack of responsibility under the ALI test. Put otherwise, whenever an insanity defense is sought to be raised by the proffer of evidence of a newly-identified mental "disease or defect," the proffered evidence is relevant for the purpose only if there is shown to be substantial acceptance within the relevant discipline of the general hypothesis that the disorder may deprive some persons of the substantial capacity either to appreciate the wrongfulness of the particular conduct in issue or to conform their conduct to the particular requirements of law in issue. Cf. Lewellyn, at 617 (aspect of "minimum showing" required to raise insanity defense).
This is a less stringent test than the "general acceptance" test first announced in Frye v. United States, 293 F. 1013 (D.C. Cir. 1923), and since widely adopted as a foundational test of relevance for all kinds of scientific evidence. While less stringent, it nevertheless remains a test of relevance dependent upon some degree of proven acceptance within the appropriate discipline of the "generalized proposition that constitutes the major premise of the relevance syllogism," 22 Wright Graham, Federal Practice and Procedure: Evidence § 5165 at 58, and to this extent it retains the essential feature of the Frye test.
We are aware that there is a respectable body of scholarly and judicial opinion that the Frye test, presumably in any form requiring any measure of proven acceptance of the "generalized proposition" underlying a proffer of "scientific evidence," has been wholly undercut by the extremely liberal "any tendency" test of relevance in Fed.R.Evid. 401. See, e.g., id. § 5168; but see Note, Expert Testimony Based on Novel Scientific Techniques: Admissibility Under the Federal Rules of Evidence, 48 Geo.Wash.L.Rev. 774 (1980) ( Frye test still permissible under Rules construed in pari materia.)
With all respect for those opinions, we are persuaded that, for reasons expressed in the body of this opinion, it remains necessary to subject proffers of evidence of "new" insanity defenses to a threshold test of relevance that requires substantial acceptance of the underlying hypothesis of causal connection between "disease or defect" and conduct. The alternative is to leave to fact-finders case-by-case the resolution of what will ordinarily be at least substantial questions about the validity of new hypotheses. This may be acceptable practice where acceptance of a new scientific hypothesis involves no more than the consideration of an additional item of evidence in a chain of proof. But we think it is not acceptable where acceptance of the general hypothesis is tantamount to exculpation from otherwise proven guilt of a criminal offense and, in the process, expansion, even on a one-case basis, of the critical insanity defense.
Furthermore, we believe that to apply such a test comports with the essential spirit of Fed. RR. Evid. 401 and 402. The determination whether proffered scientific evidence has the requisite "tendency" to prove the existence or nonexistence of the consequential fact in issue must certainly contemplate a determination, for example, that the generalized proposition upon which it is based is, as a matter of judicial notice, manifestly false, or so manifestly untrustworthy, given its proffered source (e.g., a phrenologist), that the requisite "tendency" is lacking. From this, it is but an obvious step to determining that the "generalized proposition" has simply not been shown by the proponent of the evidence to have the requisite validity to stand as the major premise of the relevance syllogism. See Wright and Graham, supra § 5168 at 89, 90.
If relevance in this sense is found, there yet remains a further question of the sufficiency of the proffered evidence to raise the insanity issue in respect of the conduct charged to the particular defendant. And it is to the latter question — not the former one — that our "slight evidence" test of sufficiency applies. See Hall v. United States, 295 F.2d at 28; cf. Torniero, at 731 n. 9. Once the general hypothesis has been found sufficiently accepted to meet the test of foundational relevance, only slight evidence that the general hypothesis might have application to the particular defendant's case is required to raise for him the insanity defense. Cf. id. (distinguishing threshold relevance test from "helpful-to-trier of fact" standard for admissibility of expert testimony under Fed.R.Evid. 702).
By means of such a threshold relevance determination courts may attempt rationally to maintain a proper balance in assessing new hypotheses respecting the insanity defense. The requirement of substantial acceptance within the relevant discipline forces some measure of scientific validation of new hypotheses as a precondition to, though not a guarantee of, their incorporation by judicial decision into this highly sensitive and historically uneasy body of legal doctrine. Because the underlying determination is so largely one of policy, it is properly cast as one of law or judicial discretion rather than as one to be made by various fact-finders from case to case.
Cf. United States v. Lyons, supra, at 246 n. 4 (even the fundamental question of "what definition of `mental disease or defect' is to be employed by courts . . . is . . . a question of legal, moral and policy — not of medical — judgment").
The relevance determination under Fed.R.Evid. 401 is one committed in the first instance to trial court discretion, but that discretion is controlled by standards that make it subject to principled appellate review. So tightly constrained is the discretion in some directions that the determination is essentially one of law subject to relatively free review. See generally 22 Wright Graham, Federal Practice and Procedure: Evidence § 5166. It is a determination, therefore, that has the capacity to emerge as legal rule or principle with precedential force. By this means the acceptance or rejection of the general hypothesis is not left to fact-finders as a matter of assessing credibility or probative weight on a case-by-case basis with potentially inconsistent expansions or contractions of the "insanity defense." See note 2, supra.
III
In applying these principles, we need not decide in this case whether pathological gambling is a "mental disease or defect" under the ALI test; whether in relation to some conduct it might constitute an effective insanity defense; and whether Gould was afflicted with the disorder at the critical times. We need address only the narrow question whether, applying the test of foundational relevance described in Part II, the general hypothesis that pathological gambling may deprive some persons of the capacity to conform their conduct to the requirements of the law prohibiting such conduct as the entry of banks with the intent to rob or steal, was shown to have substantial acceptance in the relevant scientific discipline. We hold that it was not, and that the evidence offered to raise the insanity defense was therefore properly excluded from jury consideration.
By reserving decision on the first of these points, we rest decision on a more narrow basis than that relied upon by the district court which apparently ruled that, as a matter of law, pathological gambling was not a "mental disease or defect" under the ALI test.
We decide it on this different basis under the general principle that we may affirm a judgment on any basis supported in the record if to do so does not unfairly prejudice the appellant in presentation of his case. SEC v. Chenery Corp., 318 U.S. 80, 88, 63 S.Ct. 454, 459, 87 L.Ed. 626 (1943). We find ample support for the alternative basis of affirmance in the record summarized; and we are satisfied that in attempting to establish his entitlement to the specific insanity defense, Gould made his best possible case for the proposition that the general causal hypothesis was one substantially accepted in the discipline.
Gould also assigns as error the court's refusal to admit the proffered testimony of his mother related to his mental condition at the time of the offense charged. A district court's authority to admit a lay opinion is largely discretionary. Fed.R.Evid. 701; United States v. Skeet, 665 F.2d 983, 985-86 (9th Cir. 1982). While in the usual case the lay opinion should be admitted, that discretion was not abused here. The district court justifiably could have concluded that with the particular permutations of the insanity defense on which defendant relied, his mother's lay opinion would not aid the jury in the resolution of the issues. The mother was permitted to testify fully regarding Gould's mental state, short of expressing a conclusory opinion, and Gould was permitted to adduce a significant amount of expert testimony on the same subject.
The evidence from which relevance under this test could have been found was of two kinds: documentary evidence in the form of excerpts from the 1980 third edition of the American Psychiatric Association Diagnostic and Statistical Manual of Mental Disorders (DSM-III), and expert testimonial evidence. We look to these in order.
A.
Pathological gambling was not included in DSM's until the 1980 edition. In that edition it is described as
a chronic and progressive failure to resist impulses to gamble and gambling behavior that compromises, disrupts, or damages personal, family, or vocational pursuits. The gambling preoccupation, urge, and activity increase during periods of stress. Problems that arise as a result of the gambling lead to an intensification of the gambling behavior. Characteristic problems include loss of work due to absences in order to gamble, defaulting on debts and other financial responsibilities, disrupted family relationships, borrowing money from illegal sources, forgery, fraud, embezzlement, and income tax evasion.
Commonly these individuals have the attitude that money causes and is also the solution to all their problems. As the gambling increases, the individual is usually forced to lie in order to obtain money and to continue gambling, but hides the extent of the gambling. There is no serious attempt to budget or save money. When borrowing resources are strained, antisocial behavior in order to obtain money for more gambling is likely. Any criminal behavior — e.g., forgery, embezzlement, or fraud — is typically nonviolent. There is a conscious intent to return or repay the money.
DSM-III also provides the criteria for diagnosing pathological gambling. These include:
A. The individual is chronically and progressively unable to resist impulses to gamble.
B. Gambling compromises, disrupts, or damages family, personal, and vocational pursuits, as indicated by at least three of the following:
(1) arrest for forgery, fraud, embezzlement, or income tax evasion due to attempts to obtain money for gambling
* * * * * *
C. The gambling is not due to Antisocial Personality Disorder.
While the professional assessment reflected in this language suggests that some criminal activity may be associated with pathological gambling, nowhere is it posited that the disorder may result in a lack of capacity to refrain from such criminal activity as attempted bank robbery. In fact, it is suggested that resulting criminal activity "is typically nonviolent." Attempted robbery, in the context of the criminal law, is plainly not a nonviolent crime.
Of course, the attempted bank robbery here was as non-violent as a crime of this nature could be. Defendant did no more than hand a demand note to a teller. When stopped and searched, he was found to have a toy gun in his pocket, but he never displayed the gun in the bank. Nonetheless, the crime of bank robbery, even as attempted to be committed by defendant, is a demand for money enforced by a threat against the person. It is thus inherently violent.
We conclude that the scientific judgment reflected in DSM-III does not suggest substantial acceptance within the scientific community of a general hypothesis that pathological gambling may cause some persons to be incapable of conforming their conduct to the requirements of the law that they not engage in such conduct as entering banks with the intent to commit robbery or larceny.
B.
Gould presented the expert testimony of two psychologists, Jules Moravec, Ph.D., and Robert Rovner, Ph.D., both of whom had evaluated Gould under the ALI standard. Dr. Moravec concluded that on the date of the offense, Gould suffered from a pathological gambling disorder, that this is a mental disease or defect, and that as a result of this disorder, Gould did not have the capacity to refrain from the illegal activity. Dr. Moravec acknowledged that crimes typically associated with pathological gambling, as delineated in DSM-III, included forgery, fraud, embezzlement, and tax evasion, but stated that this list was not all inclusive.
Dr. Rovner testified that Gould was a pathological gambler and that pathological gambling was a disease or defect within the meaning of the ALI test. Dr. Rovner concluded that on the date of the offense, Gould lacked the capacity to conform his conduct to the requirements of the law as a result of the pathological gambling disorder.
Dr. Moravec had studied pathological gambling since 1974 and had evaluated over 300 persons to determine whether they suffered from a pathological gambling disorder. He had testified on behalf of approximately 18 persons with respect to the question of legal responsibility. However, he gave no instance in which another pathological gambler had resorted to attempted bank robbery. Neither did Dr. Rovner identify any instance in which any other pathological gambler had engaged in conduct such as that charged to Gould.
Gould also presented the expert testimony of Clark J. Hudak, a licensed clinical social worker with a specialty in pathological gambling disorders. Hudak gave it as his opinion that typical pathological gamblers "will go to extraordinary extremes to obtain money to subsidize their gambling." On the basis of an evaluation of Gould, Hudak gave it as his opinion that Gould did suffer from a pathological gambling disorder at the time of the offense charged.
This is merely a general summary of the testimony offered in Gould's behalf by these expert witnesses. It contains, however, their testimony that bears most directly on the question whether there is within the discipline these witnesses represent substantial acceptance of the causal hypothesis with which we are concerned. Looking only to that testimony and disregarding a volume of conflicting expert testimony by Government witnesses which directly countered many of the basic premises of Gould's experts on the essential nature of the disorder as well as its implications for cognitive and volitional capability, it is impossible to conclude that there is at this time the required degree of acceptance of the causal hypothesis to allow consideration of this proffered defense. The evidence was therefore properly excluded from jury consideration.
While we do not disapprove the district court's reservation of ruling upon the proffered insanity defense until submission of the case to the jury, we observe that, despite the duplication of testimony that might be entailed, a preliminary ruling may be preferable in view of the potential for jury confusion.
AFFIRMED.