Quick viewing(Text Mode)

Empirical Research on the Insanity Defense and Attempted Reforms: Evidence Toward Informed Policy

Empirical Research on the Insanity Defense and Attempted Reforms: Evidence Toward Informed Policy

Law and Human Behavior, Vol. 23, No. 3, 1999

Errata This article entitled 'Empirical Research on the and At- tempted Reforms: Toward Informed Policy' by Randy Borum and Solo- mon M. Fulero originally appeared in Volume 23, Number 1 (February 1999). This article was printed with numerous references mistakenly omitted. Reprinted below is how the article should have appeared.

Empirical Research on the Insanity Defense and Attempted Reforms: Evidence Toward Informed Policy

Randy Borum1 and Solomon M. Fulero2

The paper addresses some common questions about the insanity defense and issues raised by commonly proposed "reforms." The first section begins with a brief descrip- tion of the insanity defense and the reasons for its existence in the . It then examines some of the popular myths and public misperceptions surrounding the insanity de- fense. The next three sections discuss proposed "reforms" and the empirical research that addresses their effect. These reforms, including various procedural changes in definitions, , and expert testimony, the institution of a guilty but mentally ill verdict, and the abolition of the insanity defense itself, are reviewed, along with relevant research findings and policy issues. Finally, the development of sound conditional release programs for criminal defendants found not guilty by reason of insanity is proposed as a reform option which could serve the objectives of enhancing public safety and access to appropriate treatment while continuing to meet the objec- tives of the insanity defense within criminal jurisprudence.

Virtually all of us can recount the details of a well-publicized insanity defense case in our home state. For example, on January 26, 1995, a 26-year-old law student named Wendell Williamson armed himself with a rifle he took from his father's house. He then walked through the streets of a college town in North Carolina shooting at people in his path. By the end of this incident, two people were dead

1Department of and Behavioral Sciences, Duke University Medical Center, Durham, North Carolina 27710; e-mail: [email protected]. 2Department of Psychology, Sinclair College and Wright State University School of Medicine, Dayton, 45402; e-mail: [email protected].

375 0147-7307/99/0600-0375$16.00/1 C 1999 American Psychology-Law Society/Division 41 of the American Psychology Association 376 Borum and Fulero

(a university student and a local resident), and several others, including one police officer, were wounded. It quickly came to light in the media that the suspect had been treated for paranoid schizophrenia, but in recent months he had allegedly stopped taking his psychotropic medication. At trial, he put forth a defense of "not guilty by reason of insanity" (NGRI) and was found NGRI by a jury in the county where the shootings took place. During and after the trial there was strong public reaction to the incident, both to the jury's decision and to the insanity defense in general. Two state senators responded by proposing legislation to repeal the defense of not guilty by reason of insanity and replace it with a "guilty but mentally ill" (GBMI) provision. This controversial case, the public reaction to it, and the subsequent to "reform" the insanity defense in North Carolina all mirror what has happened in other states such as Ohio and Arizona (Borum, 1997; Fulero, 1997; O'Connor & Bayer, 1997). This paper addresses some common questions about the insanity defense and issues raised by commonly proposed "reforms." The next section begins with a brief description of the insanity defense and the reasons for its existence in the law. It then examines some of the popular myths and public misperceptions sur- rounding the insanity defense, since they are commonly part of the impetus for, and discourse about, its reform. The subsequent three sections discuss proposed reforms and the empirical research that addresses their effect. These reforms, ranging from various procedural changes to the institution of a GBMI verdict and the abolition of the insanity defense itself, are reviewed, along with relevant research findings and policy issues. Finally, the development of sound conditional release programs for criminal defendants found NGRI is proposed as a reform option which could serve the objectives of enhancing public safety and access to appropriate treatment while continuing to meet the objectives of the insanity defense within criminal jurisprudence.

THE INSANITY DEFENSE

Doctrine of the Insanity Defense To establish that an individual is guilty of a , the state must prove at least two components: first, that the defendant engaged in proscribed conduct or committed the illegal act in question (referred to as ); and second, that the defendant committed this act with criminal intent (referred to as )3. Different offenses may require different levels of intent. For example, in North Carolina, an indictment for a charge of requires a declaration that the defendant killed "willfully and with aforethought," whereas that element for man slaughter only requires that the killing be done "willfully." (North Carolina, 1998, §15-144). Similarly, in Ohio, an aggravated murder charge requires that the

3Block's law dictionary defines mens rea as "As an element of criminal responsibility: a guilty mind; a guilty or wrongful purpose; a criminal intent; guilty knowledge and willfulness" (p. 680). Insanity Defense and Attempted Reforms 377 defendant killed "purposely, and with prior calculation and design," while murder requires only that the killing be done "purposely" (Ohio, 1998, §§2903.01,2903.02). The assumes that persons act with "free will" and that they should be held responsible for their own behavior (Perlin, 1994). The insanity defense doctrine, however, excepts from criminal responsibility and culpability a narrow class of persons who have some form of severe mental illness or disability which impairs either their cognitive abilities (e.g., the ability accurately to perceive reality and to make rational and reasonable inferences based upon them) or their volitional abilities (e.g., the ability to control their own behavior) in such a way and to such an extent that their will is compromised. Most states use one of two insanity defense standards: the M'Naghten or the American Law Institute (ALI) definitions.4 The ALI standard provides that "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 either to appreciate the criminality [wrongfulness] of his conduct or to conform his conduct to the reqirements of the law" (American Law Institute [ALI], 1962, §4.01). The M'Naghten standard requires proof that at the time of the offense, as a result of a mental disease or defect, the defendant was unable to know the nature or quality of his act or was unable to know that the act was wrong. [See, e.g., Ohio, 1998, §§2901.01(A)(14), 2945.391]. The M'Naghten test is generally considered to be the more narrow and conservative one because it includes only a cognitive component (knowing that one's actions are wrong), whereas the ALI test adds a volitional component (see Wettstein, Mulvey, & Rogers, 1991). Under the insanity defense, persons meeting this narrow definitional exception are not held criminally responsible for behavior that meets these conditions. Because they are not criminally culpable, they are not "guilty" of a crime in the eyes of the law, even if they actually committed the act in question. Many who misunderstand the essence of the insanity defense ask how someone can be called "not guilty" when it is clear that they did the act, since the term "guilty" is generally thought to apply when a person has engaged in some proscribed conduct.5 However, in assigning "guilt," the law is not only concerned with an individual's actions (the actus reus element), but also with an individual's state of mind (mens red) at the time he or she commits an act. For example, the law generally prohibits the inten- tional killing of another person. However, if an individual kills someone because he/she reasonably believes that his/her life is in imminent danger, that individual might not be held criminally responsible for the other person's death. In legal terms, then, that person might be found "not guilty" even though he/she intentionally killed another person. Because those found NGRI are "not guilty" in the legal sense, they are not

4Not all jurisdictions use these definitions; law in the Virgin Islands, in the United States Third Circuit, states that: "All persons are capable of committing or offenses except ... (4) Persons who are mentally ill and who committed the act charged against them in consequence of such mental illness. ... Virgin Islands Code, Title 14, Chapter 1, Section 14. 5In recognition of this issue, Oregon has revised the name of the insanity defense to "guilty except for insanity" (Oregon Revised Statutes 161.295) This idea is one which, arguably, should be adopted in other states as well. 378 Borum and Fulero subject to criminal sanctions for their behavior. However, they are likely to be civilly committed to a psychiatric hospital (see below). This alternative disposition is rooted, in part, in a belief that punishment would not serve as a deterrent, or that one should not be punished for an act that was not performed with criminal intent, and also that treatment is a more appropriate disposition for such persons than imprisonment.

Perceptions and Misperceptions of the Insanity Defense The insanity defense is one of the most controversial areas in the criminal law, and is plagued by many myths and public misperceptions. Often, the impetus for insanity defense reform, including calls for the abolition of the insanity defense, arises from concerns that (a) the insanity defense may be overused, (b) it is used almost exclusively in heinous crimes, (c) people who are acquitted as NGRI simply "go free" or are quickly released, and (d) NGRI acquittees, perhaps because of their , are particularly dangerous. Current research findings, however, contradict many of these commonly held beliefs (Perlin, 1994; Melton, Petrila, Poythress, & Slobogin, 1997). Prevalence of Insanity Pleas and Acquittals Although there is some variability across jurisdictions, on average the insanity defense is raised in less than 1% of all cases, and is only successful about 15-25% of the time. It is the publicity of these cases that makes it appear so much more common.6 In a survey of seven states, the rate of insanity pleas ranged from .29 to 1.73 with an average of .85 (less than 1%) per 100 felony indictments. The aggregated success rate for insanity pleas in that survey was 28.1%. (Cirincione, Steadman, & McGreevy, 1995). Similar findings have emerged from at least two other studies. A four-state study reported that the rate of success for insanity pleas across all states was 22.71% (Steadman, McGreevy, Morrissey, Callahan, Robbins, & Cirincione, 1993, p. 28), and a national survey reported that the median success rate for insanity pleas was one acquittal for every 6.5 pleas (approximately 15%) (Pasewark & McGinley, 1986). Use of the Insanity Defense in Murder Cases Murder cases account for only about one third or less of insanity defenses (Rodriguez, LeWinn, & Perlin, 1983), and the success rate is no better for these than for non-murder defendants (Steadman, Keitner, Braff, & Aravanites, 1983). For example, results from an eight-state study showed that only 14.3% of defendants pleading insanity were charged with murder (Silver et al., 1994). Still, a survey of attorneys found that 80% believed that the insanity defense was used consistently

6In a content analysis of stories from United Press International, violent crime was the focus in 86% of the articles involving a former mental patient (Silver, Cirincione, & Steadman, 1994, p. 64). This type of selective reporting is bound to influence the public's perceptions of a link between "insanity" and crime. Additional factors contributing to this perception are (1) the excessive publicity given to high- profile insanity pleas such as those of , John Hinckley, and Lorena Bobbin, and (2) the increased likelihood that insanity defense cases (as opposed to non-insanity-defense cases) will be reported upon. Insanity Defense and Attempted Reforms 379 for certain types of offenses, the most frequent of which was perceived to be murder (Burton & Steadman, 1983).

Disposition of NGRI Acquittees Most states have a statutory provision for the commitment of NGRI acquittees to a 24-hr psychiatric facility at least for a period of evaluation (Steadman et al., 1993; North Carolina, 1998, §15-144). Although there is a popular perception that insanity acquittees often go free, in many jurisdictions and for most types of offenses, defendants who are found NGRI are actually confined as long as or sometimes longer than defendants who are convicted on equivalent charges.7 A survey of seven states comparing relative lengths of confinement for people acquitted as NGRI versus those found guilty of similar offenses yielded the following findings: in two states (Georgia and Ohio), NGRI acquittees spent less time confined than those convicted; in three states (New Jersey, Washington, and Wisconsin), they spent approximately the same length of time; and in the remaining (and largest) two states (California and ), the NGRI acquittees were confined for longer periods than those found guilty (Silver, 1995, p. 375). Another study in Arizona found comparable lengths of confinement; however, over research in Connecticut, Colorado, and the District of Columbia found longer periods of confinement for NGRI acquittees than for convicted prisoners (Miller, 1994).

Dangerousness of NGRI Acquittees Despite persistent public perception to the contrary, the recidivism rate for NGRI acquittees is actually no greater than that of felons.8 A study in New York compared a matched group of NGRI acquittees and convicted felons, and found that during the follow-up period, 24% of the released NGRI acquittees were rearrested, compared to 27% of released felons (Pasewark, Pantle, & Steadman, 1982). Simi- larly, Steadman and Braff (1983) report comparative rearrest rates of 35% for NGRI acquittees and 39% for convicted felons in their sample of offenders. Public misperceptions about these and other related issues often drive the notion that the insanity defense is "broken," inappropriate or misplaced. As noted by Finkel and Fulero (1992), they are also regularly part of the rhetoric in calls for insanity defense reform. However, the development of sound policy or reform can only be enhanced by reliance on empirical evidence. A number of different types of so-called reforms have been proposed and attempted. These range from procedural changes in definition, burden of proof, and expert testimony, to more drastic reforms such as the implementation of a GBMI verdict and the abolition of the insanity defense. The next three sections examine each of these in turn.

'Although early research on the relative lengths of confinement among NGRI acquittees was somewhat inconsistent (Silver et al., 1994, p. 65), more recent, multisite data tend to suggest that many NGRI acquittees are confined as long as or longer than their convicted counterparts. See also Rodriguez et al. (1983): NGRI acquittees spend almost twice as long in confinement as people convicted of similar charges. 8See Pasewark, Pantle, and Steadman (1979): 19% of 107 NGRI acquittees were rearrested after release from the hospital, most for nonviolent offenses. 380 Borum and Fulero

"TINKERING" WITH THE INSANITY DEFENSE: PROCEDURAL REFORMS

Various procedural reforms have been proposed over the years by those un- happy with the insanity defense. However, when examined empirically, most appear to have had little impact on the actual use or success of the defense. A summary of the research on the impact of these sorts of reforms is presented below.

Revising the Substantive Test of Insanity and Jury Instructions Although there has been a fair amount of both naturalistic and analogue research on this issue, clear conclusions are still not possible. This may be due in part to a disheartening series of studies showing that juror comprehension of jury instructions is rather poor. For example, using jury instructions for the M'Naghten standard, Elwork and his colleagues (Elwork & Sales, 1985; Elwork, Alfini, & Sales, 1987) have found the rate of juror comprehension to be about 30%, and even when rewritten for maximum comprehensibility, the rate rose only to 51%. James (1959; see Simon, 1967) found that jurors accurately recalled only 58% of the insanity instructions they were given, and that this information was the least well understood of all information presented to them. Arens, Granfield, and Susman (1965) found juror comprehension of insanity defense standards to range between 31% and 40% accuracy. Clearly, if jurors cannot understand instructions on insanity, it would be hard for them to apply them accurately to a case (Ogloff, 1991). Nonetheless, the analogue or mock jury studies tend to show jurors do not produce significantly different verdicts when they are judging a defendant according to different substan- tive insanity criteria given to them in the form of jury instructions (Finkel, Shaw, Bercaw, & Koch, 1985; Finkel, 1989,1991; Ogloff, 1991). While the mock jury studies are important, they are perhaps somewhat less salient for policy decisions than studies of actual impact in legal cases, in part because they tend only to look at acquittal rate rather than at the broader use of the plea, and also because few insanity cases proceed to a jury trial. Several states have examined the actual impact of changing from a M'Naghten standard to an ALI standard, and generally have found an increase in the number of insanity acquittals.9 Pasewark and colleagues, however, had different findings based on their study of changes to the insanity defense in Wyoming,10 although that state's

'See Sales and Hafemeister (1984). When Maryland switched tests from M'Naghten to ALI, the NGRI acquittal rate among evaluated male offenders went from 8% in 1966 to 19% in 1973. Reynolds, (1984) studied the impact of a shift from the M'Naghten to the ALI standard in Oregon; in the 6 years prior to the change, there were 44 NGRI acquittals compared to 734 in the 10 years following the change. When California switched from M'Naghten to ALI, there were 187 NGRI acquittals in the year preceding the change and 270 in the year following (Slobogin, 1985). For a possible explanation, see Wettstein et al. (1991), who conducted a survey of asking them to rate which of the varying criteria from the ALI, APA, and M'Naghten standards were met by the defendants they evaluated. Although there was a strong correlation among the cognitive criteria, almost 25% of those who were assessed as being NGRI met only the ALI volitional criterion, suggesting that use of this criterion might identify a broader group for acquittal. 10In Wyoming trends were examined over a 6 year period in which three different standards, includng ALI and M'Naghten, were used. They also changed to bifurcated trials, then switched back. There were no major differences in the plea rate, success rate, or characteristics of defendants using the defense. Insanity Defense and Attempted Reforms 381

M'Naghten test also had an "irresistible impluse" clause which made it structurally similar to the ALI standard to begin with. Likewise, in 1982, California changed from the ALI standard to the more restrictive M'Naghten rule in an to reduce insanity acquittals; however, there appeared to be no significant changes in rates of insanity pleas, acquittals, or NGRI verdicts. There was a steady decrease in plea rate throughout the study, which was probably attributable to legislation designating a maximum term of confinement for NGRI acquittees (McGreevy, Steadman & Callahan 1991; Steadman et al., 1993, Chapter 4).

Changing Professionals' Expert Testimony In 1984, as part of the federal Insanity Defense Reform Act, the Federal Rules of Evidence were amended to include new Evidence Rule 704(B), which proscribed "" expert testimony by or psychiatrists on the question of insanity (that is, testimony that explicitly concludes that a criminal defendant is or is not insane). In a mock jury study, Fulero and Finkel (1991) explicitly manipu- lated the level of testimony given by the expert, and found no effect for ultimate issue testimony on the number of NGRI verdicts.

Changing the Burden of Persuasion and/or the Standard of Proof In 1982, at the time of John Hinckley's trial for shooting President Reagan and the subsequent NGRI verdict, the federal courts and most state courts required the state to bear the burden of persuasion in insanity cases, with the standard of proof being "beyond a ."11 In the ensuing 8 years, however, 17 states made changes in their requirements regarding burden and standard of proof, so that now the federal courts and virtually all of the state courts place the burden on the defendant to prove by clear and convincing evidence or a preponderance of the evidence that he or she was insane at the time of the offense.12 Simon and Aaronson (1988) conducted a survey of opinions about who should rightly bear the burden of persuasion in insanity cases. Perhaps not surprisingly, public defenders and private attorneys believed that it should be borne by the State, while and judges said it should be the defense. Mental health directors and forensic mental health professionals also said the defense should bear the burden. Ogloff (1991), in his comprehensive analogue study, found that mock jurors did not assign significantly different verdicts given different burdens and standards of proof. In a naturalistic study in Hawaii, Pasewark, Parnell, and Rock (1994) found that shifting the burden of persuasion from the prosecution to the defendant

11Other legal standards of proof are less restrictive and include "clear and convincing evidence" and "preponderance of the evidence," which is the least restrictive. 12See Leland v. Oregon (1952). Seeing the insanity defense as an has upheld the rights of the states to place the burden of persuasion on the defendant. No state currently requires the defendant to prove his or her sanity beyond a reasonable doubt (BYRD), although, according to Leland, this apparently would be constitutional under the Fourteenth Amendment. In jurisdictions where the burden of proof rests with the state, the prosecution is still required to prove the defendant's sanity BYRD. 382 Borum and Fulero did not reduce the frequency or success rate of the insanity plea. However, Steadman et al. (1993, Chapter 5) conducted a similar study in Georgia and New York, which shifted the burden of persuasion to the defendant and the standard of proof to "preponderance of the evidence." Both states showed a significant decrease in the rate of insanity pleas. There was no change in the success rate in Georgia and an increase in the success rate in New York. This latter finding was explained by the fact that more of those who pled and were acquitted as NGRI under the new provisions had a major mental illness. They also concluded that, after the reform, there were fewer "questionable cases" pleading insanity, leading to the overall reduction in plea rates.

MORE DRASTIC PROPOSALS: THE "GUILTY BUT MENTALLY ILL" PLEA

Michigan enacted the first guilty but mentally ill (GBMI) provision in 1975. Since that time, approximately 12 other states have adopted similar provisions.13 By the mid-1980s the momentum for adopting GBMI provisions had stalled. Still, in 1985, Weiner predicted that "it is likely to be revived in those states where a crime occurs which enrages the public when the defendant raises and/or succeeds with the insanity defense" (Weiner, 1985, p. 714). This prediction appears to have come to pass. Despite its initial popularity, the GBMI verdict has sparked considerable debate and criticism. The original intent of GBMI proponents generally was to reduce the number of insanity acquittals and to assure treatment for such individuals within a correctional setting (Robey, 1978). However, research suggests that in many instances these objectives have not been achieved. In addition, the GBMI verdict has been severely criticized on both legal and conceptual grounds. Over time, those who have studied it, including the American Bar Association's Criminal Justice Mental Health Standards, the American Psychiatric Association's Statement on the Insanity Defense, the National Mental Health Association's Commission on the Insanity Defense, the American Psychological Association, and the National Alli- ance for the Mentally Ill, have all opposed or recommended against the adoption of GBMI. This section defines GBMI, reviews research on the impact of GBMI in accomplishing its stated objectives, and outlines some of the legal and concep- tual criticisms.

Defining GBMI Distinguishing between the concepts of not guilty by reason of insanity (NGRI) and GBMI can be difficult. In brief, NGRI is an affirmative defense to a crime.

13Alaska (October 1, 1982), Delaware (July 2, 1982), Georgia (July 1, 1982), Illinois (September 17, 1981), Indiana (September 1, 1980), Kentucky (March 26, 1982), (August 6, 1975), Nevada (October 1, 1995), New Mexico (May 19, 1982), Pennsylvania (December 15, 1982), South Carolina (May 16,1984), South Dakota (March 19,1983), (March 31,1983-guilty and mentally ill). Note that all states except Utah and Nevada added GBMI to the existing insanity defense, rather than replacing it. Insanity Defense and Attempted Reforms 383

That is, if a criminal defendant successfully argues that he or she meets the insanity defense standard, he or she is determined to be "not guilty" (or "not responsible") in the eyes of the law and is then subject to civil proceedings for their confinement, but not to criminal incarceration or punishment (see above). In contrast, GBMI is not a defense, but is rather a verdict that implies that one is "guilty" or criminally culpable, and is subject to criminal sanctions including incarceration and possibly even death (Harris v. State, 1986; People v. Crews, 1988). The addition of the term "but mentally ill" only denotes a finding that the defendant had a mental disorder at the time of the offense and/or sentencing, but it does not lessen his or her legal guilt or criminal culpability. All states except Utah and Nevada with GBMI have added it to their existing insanity defense as a fourth option, rather than using it to replace the insanity defense. The definitions and provisions for GBMI do vary across different jurisdictions (Keilitz, Farthing-Capowich, McGraw, & Adams, 1984; Perlin, 1993). In most states, a verdict of GBMI would be indicated if it is proven beyond a reasonable doubt that the defendant (a) committed the act and (b) was aware of the nature, quality, and wrongfulness of the act, but (c) had a mental disorder which substantially impaired his or her ability to conform his or her conduct to the requirements of the law.14 If found GBMI, the defendant may be subject to any appropriate sentence, including the death penalty (Harris v. State, 1986; People v. Crews, 1988).

Impact Research on GBMI Does GBMI Reduce Insanity Acquittals? The current literature does not show a clear trend for reductions in the rate of insanity acquittals following the adoption of GBMI.15 In general, data from states in which outcomes have been studied (Michigan, South Carolina, Georgia, and Illinois) suggest that the implementation of GBMI did not significantly reduce the overall rate of insanity acquittals. In fact, in Michigan, following enactment of a GBMI statute, the number of NGRI findings actually increased. The exceptions in the existing research have come from studies in Pennsylvania and Georgia, although in light of other preexisting trends and cooccurring changes in the law, the specific effect of GBMI in these states is difficult to ascertain. Because these studies had discrepant findings, they deserve some additional explanation. Pennsylvania did report a reduction in NGRI acquittals after GBMI. However, there were an unusually large number of NGRI acquittals in the year preceding the implementation of GBMI, and the change also happened at the same time that the burden of proof in insanity cases was shifted from the state to the defendant. Similarly, Georgia showed a decline in NGRI success rates for certain kinds of offenses, but there was no change in the overall success rate. Callahan, McGreevy,

14Note that persons found to meet this definition of GBMI would presumably be found NGRI in states that have adopted the American Law Institute standard, because they would meet the "volitional" prong of the test. 15These data come from states that have retained the insanity defense and added GBMI, rather than those that have eliminated the insanity defense. 384 Borum and Fulero

Ciricone, and Steadman (1992) also reported a decrease in the acquittal rate in Georgia for certain types of offenses in the year following implementation of GBMI. Their data show a slight but statistically significant downward trend between mid- 1982 and 1983. However, it appears that the offenses that showed the greatest decline showed another increase and upward trend in the following interval. Furthermore, subsequent analyses from these same data reported "no statistically significant effect associated with the introduction of the GBMI verdict" (Steadman et al., 1993, p. 108). In addition, these subsequent analyses suggested that the decline in NGRI acquittal rates occurred before the GBMI reform and that the observed decline was also affected by a non-GBMI-related major policy shift stemming from a court case which altered judicial control over the commitment of NGRI acquittees. Thus, it is difficult to separate the effects of these reforms from any independent effects of GBMI. Does GBMI Increase Access to Treatment? It has been demonstrated in several states that a designation of guilty but mentally ill generally does not ensure opportunities for treatment beyond those ordinarily provided to other offenders, and typically adds nothing to existing provi- sions for the hospitalization of prisoners requiring inpatient care. Thus, for example, a Georgia study found that there were 150 defendants designated as GBMI, only 3 of whom were being treated in hospitals. Pennsylvania reported a higher rate of treatment with just over one-fourth (27.5%) of GBMI inmates never receiving inpatient psychiatric treatment,16 but in Illinois, of the first 44 defendants found GBMI, none received any hospital treatment (Steadman et al., 1993). Research has shown, however, that prosecutors and defense attorneys both overestimate the likelihood that GBMI inmates will receive treatment.17 Thus, the GBMI verdict may create "false treatment expectations" which would be misleading to defendants, juries, judges and the general public.

Legal and Conceptual Criticisms of GBMI No Enhanced Public Protection Contrary to the intent of some of its proponents, it appears that abolition of the insanity defense and implementation of GBMI would probably not enhance public protection. Once GBMI inmates serve the minimum time necessary for parole, their eligibility for release is dependent on their perceived level of dangerous- ness, the same factor that determines the release of NGRI acquittees. And, as noted above, defendants who are acquitted as NGRI are often confined as long as or sometimes longer than defendants who are convicted on equivalent charges. More- over, in absolute terms, GBMI displaces very few insanity acquittees and it is even unclear as to whether the few whom it does replace are the ones who would reoffend.

16Pennsylvania'a treatment rates may be somewhat higher because (1) they require a presentencing hearing on the issue of whether inpatient hospitalization is required and (2) their GBMI population is largely comprised of persons with severe mental disabilities. 17See McGraw, Farthing-Capowich, and Keilitz (1985) and Finkel (1991). Mock jurors selected the GBMI verdict because they believed that the defendant would have better access to mental health treatment. Insanity Defense and Attempted Reforms 385

Jury Confusion There are several bases for concerns about jury or fact-finder confusion in GBMI cases. The concept and criteria for the insanity defense alone can be difficult to understand. As noted above, research tends to show that jurors comprehend only about 30% of jury instructions on the insanity defense (maybe up to 50% under ideal circumstances with careful wording) (Elwork et al, 1987). The addition of GBMI further complicates the issue, making discrimination between verdicts even more difficult and potentially confusing. Indeed, "mental illness" is a component of both verdicts, and some have suggested that the phrase "but mentally ill" connotes a type of diminished capacity that is not truly part of the verdict. It has also been argued that the addition of GBMI can lead to improper verdicts by allowing fact finders an "easy way out" so that they will not grapple with the central issue of culpability and by its being applied to people who actually meet the insanity defense criteria to insure that they are confined (Slobogin, 1985). Constitutional Concerns Constitutional challenges to and concerns about GBMI statutes have arisen in a number of jurisdictions. McGraw et al. (1985) have identified several bases for challenge. Equal-protection challenges have claimed that GBMI statutes create "irrational classifications" which discriminate against people found GBMI. challenges relate to assertions of a right to a pretrial hearing on the issue of defendant's present mental condition, to allegedly "broad and vague" definitions of mental illness and insanity, and to whether or not a "strict scrutiny" analysis should be applied to disputes and claims about GBMI. It has also been alleged that GBMI statutes violate Eighth Amendment protections against cruel and unusual punishment by not providing appropriate treatment to people with this special designation of mental illness. Despite these criticisms, GBMI has so far withstood constitutional challenge (Perlin, 1993; People v. Sorna, 1979).

ABOLITION OF THE INSANITY DEFENSE

Five states have eliminated the affirmative defense of NGRI: Montana (1979), (1982), Utah (1983), Nevada (1995), and (1996). Even in these states, however, mental disorder can still be introduced into evidence, but it is only admissi- ble to negate or disprove the requisite intent or mens rea element of the offense charged. The wording of Utah's statute is illustrative: "It is a defense to a prosecution under any statute or ordinance that the defendant, as a result of mental illness, lacked the mental state required as an element of the offense charged. Mental illness is not otherwise a defense" (Utah, 1998). This minimal provision for the admissibility of evidence on mental illness is apparently constitutionally required. In the early 1900s, three states [Louisiana (State v. Lange, 1929), Mississippi (Sinclair v. State, 1931), and Washington (State v. Strasbury, 1910)] also sought to abolish the defense of mental nonresponsibility, but the statutes were found to be unconstitutional, presumably because they did 386 Borum and Fulero not allow introduction of any evidence concerning one's mental disorder, not even to negate mem rea. This mens rea approach, however, is different from the NGRI approach. Unlike the insanity defense, the mens rea exception is not an affirmative defense and it is not concerned with whether or not an individual knew what he or she was doing was wrong. The mens rea approach is only concerned with whether or not an individual acted with the simple mental state that the law defines as an element of the charged crime, such as intent to engage in a particular kind of action, intent to cause a particular result, or knowledge of a particular kind of circumstance. To illustrate this distinction, consider the following example: Suppose a man with schizophrenia has a delusional belief that a group of men in red hats are stalking him in order to kill him so that they can take over the planet. Along the street, he encounters a man in a red hat walking quickly toward him. Fearing his own death, he shoots the red-hatted man to death. Under the mens rea approach, the individual's fear of immediate harm from the red-hatted man and his belief that he was acting to defend himself are not relevant to determining his culpability. Because he intended to kill the man, regardless of his reason, he would be found guilty of murder and subject to criminal sanctions. Under the NGRI approach, however, the judge or jury would determine whether the man's delusions caused him not to know his actions were wrong. If they determined that he did not know that it was wrong, he would be found NGRI and involuntarily committed to a state facility to receive treatment rather than sentenced to prison for his actions.

Legal and Policy Issues A question which is still unsettled in the law, however, is whether states are constitutionally required to provide some form of affirmative insanity defense. Thus far, the statutes in Idaho (State v. Searcy, 1990), Montana (State v. Korrell, 1984; State v. Byers, 1993; State v. Cowan, 1993/1994), and Utah (State v. Herrera, 1995) have all survived constitutional challenge before their respective state supreme courts. The Utah Supreme Court attempted to avoid the issue of whether or not it was constitutional to "abolish" the insanity defense by claiming that their mens rea exception was simply an alternate approach to dealing with "the insane crimi- nal defendant." The U.S. Supreme Court has never directly addressed the issue of whether any particular insanity test is minimally required by the Constitution, or even whether an affirmative defense of insanity is required at all. In Leland v. Oregon (1952), the U.S. Supreme Court declined to specify that any particular insanity test was required under federal due process. Some have interpreted that to mean that states may be free to choose among the various substantive tests for insanity, but that they may not altogether eliminate this affirmative defense. Others, however, point to dicta in previous cases which seem to suggest that at least some of the current justices on the U.S. Supreme Court would not find the insanity defense to be constitutionally required. For example, in Foucha v. Louisiana (1992), Justice Kennedy in his dissent declared that "States are free to recognize and define the insanity defense as they Insanity Defense and Attempted Reforms 387 see fit" (p. 96), and Justice O'Connor indicated that Foucha did not restrict "States' freedom to determine whether and to what extent mental illness should criminal behavior. The Court does not indicate that States must make the insanity defense available" (pp. 88-89). Similarly, Justice Rehnquist in his dissent in Ake v. Oklahoma (1985) opined that "it is highly doubtful that due process requires a state to make available an insanity defense to a criminal defendant" (p. 91). Of course, such dicta carry no binding authority or precedential weight. In 1994, the U.S. Supreme Court had an opportunity to rule on the constitution- ality of these "abolition statutes" in Montana v. Cowan (1993/1994). However, the Court declined to review the case (i.e., it denied ). Although some media commentators and policy makers have apparently interpreted this as "opening the door" for states to eliminate their insanity defenses, it is a well-established principle of constitutional law that a denial of certiorari has no legal authority or precedential value, nor is it necessarily an index of how the Court would decide the issue if it chooses to review it at a later date (Perlin, 1995). Thus, this nonaction provides no additional insight concerning the Court's view of the constitutionality of "abolition statues." Nevertheless, the case probably served as part of the impetus for the subsequent abolition statutes in Nevada and Kansas. From a policy perspective, one of the key issues facing those states that abolish the insanity defense is the question of what to do with "insane offenders." This might include two groups: (a) those who were or are mentally ill and in need of treatment, but who are not found to meet the narrow mens rea exception and (b) those who are found not guilty because their mental illness negated mens rea. The convicted mentally ill offender is often either treated/managed in the correctional facility or temporarily transferred to a state psychiatric facility under statutory provisions. For the acquitted mentally ill offender, "abolition" states have generally sought to civilly commit such individuals if they are determined to be dangerous. Over time, states have approached this issue with increasing specificity and sophistication. Kansas, the most recent state to abolish its insanity defense, has a detailed set of statutory procedures for handling defenses based on "lack of mental state." These include pretrial notification and dispositional provisions that closely parallel those which existed for NGRI acquittees. For example, if a defendant is found not guilty solely because he or she had a mental disease or defect which rendered him or her incapable of possessing the required criminal intent, the defen- dent is "committed to the state security hospital for safekeeping and treatment" (Kansas, 1998, §22-3428. (1)(a)). Within 90 days of this commitment, the chief medical officer of the facility will send a written evaluation report to the court, after which there must be a hearing (within 30 days) "to determine whether or not the defendant is currently a mentally ill person" (§22-3428. (1)(b)). If the court finds by clear and convincing evidence that the defendant is not a mentally ill person, "the court shall dismiss the criminal proceedings and discharge the defen- dant, otherwise the court may commit the defendant to the state security hospital for treatment or may place the defendant on conditional release" (§22-3428. (1)(d)). The constitutionality of these provisions, as distinguished from ordinary civil com- mitment proceedings, has not yet been before the courts. 388 Borum and Fulero

Empirical Issues with Regard to Abolition Because abolishing the insanity defense has been done in very few states, there is little empirical information available. Existing research, however, tends to show that abolishing the insanity defense does not reduce the overall volume or rate of mental nonresponsibility pleas, and that essentially the same people continue to be handled by the forensic mental health system, not the correctional system. For example, in Utah (Heinbecker, 1986), there were as many mens rea mental nonresponsibility acquittals (seven) in the 2 years following abolition of the defense as there had been NGRI acquittals in the 9 years prior with the affirmative defense. In reviewing these cases, the author concluded that only one of the mens rea acquittals even appropriately met the narrow criteria, whlie the others were reached through stipulation and negotiation. The state with the longest history of operating under the mens rea model is Montana. Consequently, there is better longitudinal data from that state than from any other. In their study on the effect of abolishing the insanity defense in Montana, Steadman et al. (1993) found that although there was a decline in the success rate of insanity pleas, there was no change in the volume or rate of mental nonresponsibil- ity pleas overall. Despite the apparent decline in success, the same people were essentially being handled in the forensic mental health (as opposed to the correc- tional) system. After abolition, the number of people found incompetent to stand trial (IST) went up in approximately the same proportion that "mental nonresponsi- bility" acquittals went down. It appeared that these "new" ISTs were people who would have been found not guilty by reason of insanity under the prior law. How- ever, under the new law more of them had their charges dismissed or deferred. In addition, after the reform, more defendants were also being released rather than being hospitalized. Thus, it is at least questonable whether or not abolition of the insanity defense results in enhancing public safety or in confining for longer periods of time the class of offenders who meet the insanity defense test.

Unintended Policy Effects of Abolition As a practical matter, the most serious policy implications of the abolitionist position are ones likely to have been unintended by those advocating it. First, it would appear that some individuals who are insane under current insanity might otherwise have been acquitted. Second, the "diminished capacity" defense which the public finds to distasteful would preforce be restored to criminal law as a result of abolition. Formerly "Insane" Defendants Now To Be Acquitted Without an insanity defense, an individual found to be so mentally ill that he or she could not form the specific intent required for the crime would have to be acquitted. For example, Kadish (1968) states: A total inability to know the nature and quality of the act quite plainly precludes convicting a defendant of any crime whose definition requires that he have that knowledge. And any crime which requires intent, or knowledge, or recklessness, surely posits that knowing. If Insanity Defense and Attempted Reforms 389

it were not for the special, preemptive defense of legal insanity, therefore, the defendant would have a complete defense on the merits to any such crime, namely, the lack of mens tea. Other legal scholars (Monahan, 1973; Bonnie, 1983) have acknowledged the same problem with abolishing the insanity defense. We will await with interest the public reaction (and the resulting legislative response) to the first case in which a defendant who is clearly insane is actually acquitted. The Return of the "Diminished Capacity" Defense In the "diminished capacity" defense, a defendant introduces testimony about his or her mental state in order to show that he or she lacked the mental capacity to form the specific mental state required for a particular crime or degree of crime. For example, a defendant charged with "knowingly" attempting to harm someone might be charged with felonious , a second-degree felony. In states that have abolished the diminished capacity defense, such as Ohio (see State v. Wilcox, 1982) that defendant cannot now put on expert testimony on mental state, unrelated to an insanity defense, in order to show that he could only have formed the specific intent of "recklessness," which would reduce the charge to assault, a first-degree . In California, this became known as the " defense" after invoked it successfully in the murder case in (People v. White, 1981). Mr. White could not have argued insanity, as his actions (concealing the weapon, sneaking in, etc.) clearly showed that he understood their wrongfulness. Instead, he argued that his consumption of junk food prior to the shooting had so affected him mentally that he could not form the intent to murder. He was eventually convicted only of , which requires a lesser intent, to a chorus of intense outrage by the public. If the abolitionists are successful in limiting evidence of mental state only to the question of specific intent, they will inadvertently restore the diminished capacity defense. It is simply not possible to have a mens rea element and then not allow the defendant to introduce evidence on it. This will open the floodgates to the use of mental state evidence in many more cases than the ones seen now—with the burden of proof on the State beyond a reasonable doubt, rather than on the defen- dant by a preponderance of the evidence, as it is now with the insanity defense. In the words of Kadish (1968, p. 282), "you can change the name of the game, but you cannot avoid playing it so long as mens rea is required." Abolition, then, is likely to have unintended effects that not only fail to solve perceived existing problems, but actually create worse problems. It is therefore prototypical of policy changes that are influenced by a few cases, by myths and misconceptions, and by political considerations instead of logic, rationality, or empirical data. In a sense, abolition of the insanity defense would not only throw out the baby with the bath- water, but would dismantle the bathtub.

AN EFFECTIVE SOLUTION: CHANGING PROCEDURES FOR RELEASE OF NGRI ACQUITTEES

As noted above, although there is a popular perception that insanity acquittees often "go free," defendants who are acquitted as NGRI are often confined as long 390 Borum and Fulero as or sometimes longer than defendants who are convicted on equivalent charges.18 Thus, when the insanity defense is retained, this class of offenders (persons currently found NGRI) will likely be confined at least as long as or longer than if there were no insanity defense (if they were to be found guilty). One additional way to enhance public safety is to modify the procedures governing the release and postrelease supervision of NGRI acquittees. Conditional release (CR) programs have been used successfully in Oregon, Connecticut, Mary- land, Illinois, Ohio, and Virginia to provide intensive monitoring of insanity ac- quittees (Miller, 1994). These programs operate much like specialized outpatient commitment or "mental health probation" programs. If an individual becomes eligible for release, the case is carefully reviewed, often by a multidisciplinary committee or psychiatric security review board, and any decision to release is usually made contingent on compliance with a set of individualized conditions designed to reduce an individual's risk (e.g., comply with treatment, not possess or use alcohol or drugs, etc.). The individual is then monitored in the community, and any failure to comply with the conditions may lead to the individual's return to the 24-hr facility. Often these provisions are even more stringent and more individually tailored than those used in probation, and because people on CR also tend to be more intensively monitored, this appears to be an extremely promising mechanism to reduce the potential for violence and criminal recidivism among mentally ill offenders. In some jurisdictions, conditional release programs have provided increased and earlier detection of individuals who need rehospitalization when compared to NGRI acquittees whose release is not part of a conditional release program. More- over, they have been shown to reduce recidivism. For example, a study in California compared a group of NGRI acquittees who were unconditionally released after their commitment time expired to a group who were conditionally released under a CR program. During the follow-up period, 27% of those who were unconditionally released were rearrested compared to only 5% of the CR group. Much of the difference appeared to be attributable to increased surveillance in the community leading to rehospitalization before the problem behaviors occurred (Weidrand- ers, 1992).

SUMMARY AND CONCLUSIONS

Advocates of the various procedural changes to the insanity defense, most of which were proposed as an indirect way of limiting the use of the insanity defense (Finkel, 1989; Fulero & Finkel, 1991), have generally been frustrated by the lack of substantive effect that those changes were supposed to have had. Changing the substantive test or definition of insanity to what would appear to be a more restrictive one has little or no effect on the number of insanity acquittals, though the results are somewhat mixed. Changing the burden of proof and the standard of proof

18Rodriguez, J., LeWinn, L., & Perlin, M., supra note 15, at 403-404. (NGRI acquittees spend almost twice as long in confinement as people convicted of similar charges.) Insanity Defense and Attempted Reforms 391 required has had virtually no effect. Restricting mental health expert testimony appears to have had no effect. Later, the GBMI option appeared on its face to be an attractive alternative for those interested in limiting the insanity defense. However, GBMI does not appear to increase effective treatment options or enhance public safety. In addition, the empirical research reviewed here does not support the position that NGRI acquittals will decrease with the introduction of the GBMI option. Indeed, it appears that those found GBMI come not from the group of defendants that would formerly have been found NGRI, but instead from the population that would formerly have been found guilty. And on a policy or conceptual level, it could be argued that the notion of "guilty but mentally ill" does nothing more than extend or perpetuate the "excuse" or "blame game" that so many commentators have criticized (Dershowitz, 1994) (why not, for example, "guilty but diabetes," or "guilty but abused" as an alternative plea?). Finally, those who have advocated for the most extreme solution-the abolition of the insanity defense-appear to have little empirical support for their position. As set forth above, the arguments that form the underpinnings of the abolitionist position-that the insanity defense is overused and often successful, that it is used mostly in heinous crimes, that successful NGRI acquittees spend little or no time in custody, and that NGRI acquittees are more dangerous than other types of releasees-appear to be myths without empirical foundation. In addition, the unin- tended policy implications of the abolition position, apart from the legal questions raised by abolition, are disturbing. Overall, then, efforts to "fix" the insanity defense by "tinkering" or by more drastic means have generally had little effect. This, we would argue, is largely because there is no real evidence that it is broken. It is a defense that is rarely raised, and usually unsuccessful. Even when defendants raising insanity are successful, they will generally spend in confinement as much time as if or more time than if they had simply been found guilty. In addition, the moral integrity of the criminal law would seem to necessitate at least some provision for lessening the culpability of individuals who, as a result of severe mental illness, do things that they do not know are wrong. Punishing people whose conduct is driven by mental disability, rather than wrongful intent, does not serve any legitimate objective of general deterrence. Certainly, the specific deterrent objective of protecting the public from future harm by a mentally ill and dangerous person is a legitimate one. However, it is clearly one not best achieved by attempts to tinker with, change, or abolish the insanity defense. Instead, sound and carefully implemented conditional release programs appear to have the greatest effect on enhancing public protection, while simultaneously providing necessary treatment. These programs provide greater supervision than is generally afforded to persons on probation or parole; further, the conditions of release are rehabilitative in nature and specifically tailored to the reduction of future problematic behavior that might result from untreated mental illness. It has been said that insanity might be defined as doing the same thing over and over and expecting a different result. Perhaps it is time to apply this maxim 392 Borum and Fulero

to efforts at insanity defense reform. Policy changes that are informed and driven by empirical evidence rather than anecdotes, myths, and political considerations are always more likely to be effective. We urge that changes in this area of law be made in that spirit.

ACKNOWLEDGMENTS

The authors wish to thank Michael Perlin and Lynda Olsen-Fulero for their comments on an earlier draft.

REFERENCES

Ake v. Oklahoma, 470 U.S. 68 (1985). American Law Institute (1962). Model penal code. Proposed official draft. Philadelphia: Author. Arens, R., Granfleld, D., & Susman, J. (1965). Juries, juror charges, and insanity. Catholic University Law Review, 14, 1-29. Black's. (1991). Black's law dictionary. St. Paul, MN: West Publishing Co. Bonnie, R. (1983). The moral basis of the insanity defense. American Bar Association Journal, 69, 194-197. Borum, R. (1997, August). Insanity defense reform in North Carolina. Paper presented at the symposium, "Into the lion's den: Politics, science, and insanity defense reform," S. Fulero (Chair), American Psychological Association Annual Meeting, Chicago. Burton, M., & Steadman, H. (1983). Legal professionals' perceptions of the insanity defense. Journal of Psychiatry and Law, 6, 173-187. Callahan, L. A., McGreevy, M. A., Cirincione, C, & Steadman, H. J. (1992). Measuring the effects of the guilty but mentally ill (GBMI) verdict: Georgia'a 1982 GBMI reform. Law and Human Behavior, 16, 447-462. Cirincione, C., Steadman, H., & McGreevy, M. (1995). Rates of insanity acquittals and the factors associated with successful insanity pleas. Bulletin of the American Academy of Psychiatry and Law, 23, 399-409. Dershowitz, A. (1994). The abuse excuse and other cop-outs, sob stories, and evasions of responsibility. Boston: Little, Brown. Elwork, A., Alflni, J., & Sales, B. (1987). Toward understandable jury instructions. In L. Wrightsman, S. Kassin, & C. Willis (Eds.). In the jury box: Controversies in the courtroom (pp. 161-179. Newbury Park, CA: Sage. Elwork, A., & Sales, B. (1985). Jury instructions. In S. M. Kassin & L. S. Wrightsman (Eds.), The psychology of evidence and trial procedure (pp. 280-297). Beverly Hills, CA: Sage. Finkel, N. (1989). The Insanity Defense Reform Act of 1984: Much ado about nothing. Behavioral Sciences and Law, 7, 403-419. Finkel, N. (1991). The insanity defense: A comparison of verdict schemas. Law and Human Behavior, 15 533-555. Finkel, N., & Fulero, S. (1992). Insanity: Making law in the absense of evidence. International Journal of Medicine and Law, 11, 383-404. Finkel, N., Shaw, R., Bercaw, S., & Koch, J. (1985). Insanity defenses: From the jurors' perspective. Law and Psychology Review, 9, 77-92. Foucha v. Louisiana, 504 U.S. 71(1992). Fulero, S. (1997, August). Insanity defense reform in Ohio. Paper presented at the symposium, "Into the lion's den: Politics, science, and insanity defense reform," S. Fulero (Chair). American Psychological Association Annual Meeting, Chicago. Fulero, S., & Finkel, N. (1991). Barring ultimate issue testimony: An "insane" rule? Law and Human Behavior, 15, 496-507. Harris v. State, 499 N.E. 2d 723 (Ind. 1986). Heinbecker, P. (1986). Two years' experience under Utah's mens rea insanity law. Bulletin of the American Academy of Psychiatry and Law, 14, 185-191. James, R. (1959). Jurors' assessment of criminal responsibility. Social Problems, 7, 58-67. Insanity Defense and Attempted Reforms 393

Kadish, S. (1968). The decline of innocence. Cambridge Law Review, 26, 273-290. Kansas (1998). Kansas Statutes Annolated. Keilitz, I., Farthing-Copowich, D., McGraw, B., & Adams, L. (1984), The guilty but mentally ill verdict: An empirical test. Williarasburg, VA: National Center for State Courts. Leland v. Oregon, 343 U.S. 790 (1952). McGraw, B., Farthing-Capowich, D., & Keilitz, I. (1985). The guilty but mentally ill pleas and verdict: Current state of the knowledge. Villanova Law Review, 30, 117-191. McGreevy, M., Steadman, H. & Callahan, L. (1991). The negligible effects of California's 1982 reform of the insanity defense. American Journal of Psychiatry, 148, 744-750. Melton, G., Petrila, J., Polythress, N., & Slobogin, C. (1997). Psychological evaluations for the courts (2nd ed.). New York: Guilford Press. Miller, R. (1994). Criminal responsibility. In R. Rosner (Ed.). Principles and practice offorsensic psychia- try (pp. 198-215). New York, Chapman and Hall. Monahan, J. (1973). Abolish the insanity defense?-Not yet. Rutgers Law Review, 26, 719-740. North Carolina (1998). North Carolina General Statutes. O'Connor, M., & Bayer, K. (1997, August). Insanity defense reform in Arizonia. Paper presented at the symposium, "Into the lion's den: Politics, science, and insanity defense reform," S. Fulero (Chair), American Psychological Association Annual Meeting, Chicago. Ogloff, J. (1991). A comparison of insanity defense standards on juror decision making. Law and Human Behavior, 15, 509-531. Ohio (1998). Ohio Revised Code. Pasewark, R., & McGinley, H. (1986). Insanity plea: National survey of frequency and success. Journal of Psychiatry and Law, 13, 101-108. Pasewark, R., Pantle, M., & Steadman, H. (1979). The insanity plea in New York State, 1965-1976. New York State Bar Journal, 53, 186-189, 214-225. Pasewark, R., Pantle, M., & Steadman, H. (1982). Detention and rearrest rates of persons found not guilty by reason of insanity and convicted felons. American Journal of Psychiatry, 139, 892-897. Pasewark, R., Parnell, B., and Rock, J. (1994). Insanity defense: Shifting the burden of proof. Journal of Police and Criminal Psychology, 10, 1-4. Pasewark, R., Randolph, R., & Bieber, S. (1984). Insanity plea: Statutory language and trial procedures. Journal of Psychiatry and Law, 1984(Fall), 399-422. People v. Crews, 122 I11.2d 266 (1988). People v. Sorna, 276 N.W.2d 892 (1979). People v. White, 117 Cal. App. 3rd 270, 172 Cal. Rptr. 612 (1981). Perlin, M. (1987; 1993 pocket part). Mental disability law: Civil and criminal (§15.09). Charlottesville, VA: Michie. Perlin, M. (1994). Law and mental disability. Charlottesville, VA: Michie. Perlin, M. (1995). Refusal to hear challenge to insanity abolition statute fuels speculation as to Supreme Court's "hidden agenda." Report on Mental Disability Law, 1(1), 8. Reynolds, S. (1984). Battle of the experts revised: 1983 Oregon legislation on the insanity defense. Willamette Law Review, 20, 303-317. Robey, A. (1978). Guilty but mentally ill. Bulletin of the American Academy of Psychiatry and Law, 6, 374-381. Rodriguez, J., LeWinn, L., & Perlin, M. (1983). The insanity defense under siege: Legislative and legal rejoinders. Rutgers Law Journal, 14, 397-430. Sales, B., & Hafemeister, T. (1984). Empiricism and legal policy on the insanity defense. In L. Teplin (Ed.), Mental health and criminal justice (pp. 253-278). Newbury Park, CA: Sage. Silver, E. (1995). Punishment or treatment: Comparing the lengths of confinement of successful and unsuccessful insanity defendants. Law and Human Behavior, 19, 375-389. Silver, E., Cirincione, C., & Steadman, H. (1994). Demythologizing inaccurate perceptions of the insanity defense. Law and Human Behavior, 18, 63-70. Simon, R., & Aaronson, D. (1988). The insanity defense: A critical assessment of law and policy in the post-Hinckley era. New York: Praeger. Sinclair v. State, 161 Miss. 142 (1931). Slobogin, C. (1985). The guilty but mentally ill verdict: An idea whose time should not have come. George Washington University Law Review, 53(3-4), 494-527. State v. Byers, 261 Mont. 17(1993). State v. Cowan, 260 Mont. 510 (1993), cert, denied, 114 S.Ct. 1371 (1994). State v. Korell, 213 Mont. 316 (1984). State v. Herrera, 895 P. 2d 359 (1995). State v. Lange, 168 La. 958 (1929). 394 Borum and Fulero

State v. Searcy, 118 Idaho 632 (1990). State v. Strasburg, 60 Wash. 106 (1910). State v. Wilcox, 70 Ohio St. 2d 182, 436 N.E. 2d 523 (1982). Steadman, H., & Braff, J. (1983). Defendants not guilty by reason of insanity. In J. Monahan & H. Steadman (Eds.), Mentally disordered offenders: Perspectives from law and social science (pp. 109-129). New York: Plenum. Steadman, H., Keitner, L., Braff, J., & Aravanites, T. (1983). Factors associated with a successful insanity plea. American Journal of Psychiatry, 140, 401-405. Steadman, H., McGreevy, M., Morrissey, J., Callahan, L., Robbins, P., & Cirincione, C. (1993). Before and after Hinckley: Evaluating insanity defense reform. New York: Guilford Press. Utah. Utah Code Ann. §76-2-305(1). Virgin Islands (1998). Virgin Islands Statutes, Title 14, Chapter 1, § 14(4). Weidranders, M. (1992). Recidivism of disordered offenders who were conditionally vs. unconditionally released. Behavioral Sciences and the Law, 10, 141-146. Weiner, B. (1985). Mental disability and criminal law. In S. J. Brakel, J. Parry, & B. Weiner (Eds.), The mentally disabled and the law (3rd ed., pp. 693-801). Chicago, American Bar Association. Wettstein, R., Mulvey, E., & Rogers, R. (1991). A prospective comparison of four insanity defense standards. American Journal of Psychiatry, 148, 21-27.

Top View