The Adjudication of Criminal Responsibility:
A Review of Theory and Research
Stephen L. Golding, Ph.D.
Professor and Director of
Clinical Training
Department of Psychology
Adjunct Professor of Law
University of Utah
Salt Lake City, UT
84112
In D. Kakehiro and W. Laufer (Eds.), Handbook of Psychology and Law, Springer-Verlag: Berlin, 1992, Pp. 230-250
Table of Contents
Criminal Responsibility in Historical and Current Context 1
Historical Perspectives 1
Current psychological and jurisprudential perspectives 4
Conceptualizations of Criminal Responsibility 5
Lay attitudes and the ascription of criminal responsibility 5
Methodologies for the assessment of criminal responsibility 9
Characteristics of NGRI Verdicts and Dispositions 12
NGRI verdicts 12
Characteristics of NGRI populations 13
Correspondence of professional and judicial decision making 16
Length of NGRI dispositions 17
Criminal and psychopathological "recidivism" 19
The alternative verdict: "Guilty but mentally ill" 21
History of the GBMI verdict 21
Reduction in "abuse" of the NGRI verdict 23
GBMI as a dispositional alternative 24
Research directions and social policy 27
References 30
The attribution of criminal responsibility has been a central problem of social policy and moral philosophy that antedates the formal mechanism of the "insanity defense." After placing the attribution of criminal responsibility in historical perspective, this chapter develops the modern versions of the insanity defense and then identifies a set of principle psychological perspectives for study of the insanity defense. Utilizing those perspectives, the chapter then reviews and integrates the current status of theory and research in the following areas: a) lay attitudes toward the ascription of criminal responsibility; b) professional conceptions of criminal responsibility as manifested in methodologies for its assessment; c) relevant characteristics of NGRI (not guilty by reason of insanity) verdicts, the persons to whom they are applied, and the outcomes associated with the verdicts; and d) characteristics of alternatives to the insanity defense, with special reference to the "guilty but mentally ill" verdict option. Finally, future directions for psychological research and social policy in this area are discussed.
Criminal Responsibility in Historical and Current Context
Historical Perspectives. The historical roots of the relevance of mental state at the time of offense to the attribution of criminal responsibility have been the subject of several well documented scholarly reviews (Golding & Roesch, 1987; Gray, 1972; Hermann, 1983; Keilitz, 1987; Platt & Diamond, 1965, 1966; Pollack & Maitland, 1952; Sayre, 1932; Shah, 1986; Smith, 1981; Walker, 1968). These reviews demonstrate that an historic characteristic of Judeo-Christian cultures has been a fundamental concept that certain mental states are a necessary condition for the just ascription of moral responsibility. Moral responsibility, in turn, has been tied to the idea of having certain mental capacities and having acted voluntarily under the influence of certain intentions. This fundamental logic was imported into the criminal law at a very early stage, and became the central assumption in criminal law by the end of the Middle Ages (Sayre, 1932). Blackstone's Commentaries present the classical statement of the principle:
An involuntary act, as it has no claim to merit, so neither can it induce any guilt; the concurrence of the will, when it has its choice either to do or to avoid the fact in question, being the only thing that renders human actions either praiseworthy or culpable. Indeed, to make a complete crime cognizable by human laws, there must be both a will and an act. [Cited in State v. Strasburg (1910, p. 1021)].
Hence, "An act is not legally cognizable as evil, and hence criminally punishable, unless it is committed by a person who has the capacity to cognize the act as evil and then freely chooses to do it" (Golding & Roesch, 1987, p. 395). Insofar as the issues of mental capacity, cognition and choice are intimately involved in the legal ascription of criminal responsibility, mental health professionals have been involved in the process, and the formulation of the legal standard, for several hundred years.
The first formulations of legal standards for insanity were crude but in synchrony with the understanding of mental disorder at the time, as can be seen in the judge's summation in Arnold's Case (1724), "If a man be deprived of his reason, and consequently of his intention, he cannot be guilty. ...it must be a man that is totally deprived of his understanding and memory, and doth not know what he is doing, no more than an infant, than a brute, or a wild beast" (cited in Walker, 1968, p.56). By the time of Daniel M'Naghten's trial in 1843, over a century of debate about intentionality and the influence of mental disorder, led to this interpretation of legal insanity,
to establish a defense on the ground of insanity, it must be clearly proved that, at the time of the committing of the act, the party accused was labouring under such a defect of reason, from disease of the mind, as not to know the nature and quality of the act he was doing; or, if he did know it, that he did not know he was doing what was wrong (M'Naghten's Case, 1843, p. 722)
Almost a century and a half later, the intense debate as to the legal demarcation between "culpable" and "delusional" intentionality, between "narrow" and "broad" definitions of cognitive capacity (i.e, "knowing" vs. "appreciating") and between "knowledge of wrongness" and "irresistible impulses" continues. Various attempts to "reform" the insanity standard, to restrict its scope, or to abolish the plea altogether occur at regular historical intervals (Golding & Roesch, 1987; Keilitz, 1987). In the mid-1920's, state and Federal jurisdictions were divided in terms of following "broad" versus "narrow" constructions of M'Naghten, supplementing it with irresistible impulse rules, or an even broader "product rule" (see Ballantine, 1919; Crotty, 1924; Keedy, 1917). In 1954, Judge David Bazelon initiated another period of experimentation with insanity rules in Durham v. United States (1954), in which the District of Columbia adopted the "product test" out of frustration with the strictures and ambiguities of M'Naghten. Ultimately, the District of Columbia scrapped Durham and adopted the American Law Institute's (1962) Model Penal Code definition of legal insanity,
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 (wrongfulness) of his conduct or to conform his conduct to the requirements of the law" (p. 74).
The "ALI" rule was adopted in many jurisdictions and was the "majority view" until a series of sensational state cases and the attempted assassination of President Reagan. At the Federal level, the United States Congress adopted the Insanity Defense Reform Act (1984), eliminating the "volitional prong" of the ALI rule ( "to conform his conduct to the requirements of law") and changing Rule 704 of the Federal Rules of Evidence to bar expert witnesses from conclusory testimony in the form of an opinion or inference as to a defendant's mental state or condition constituting an element of the crime charged. Many states also responded by making changes in the legal standard for insanity, by shifting the burden of proof to the defendant, by altering procedural aspects of the adjudication, by adopting "alternative verdicts" (such as "guilty but mentally ill," discussed below), or by a combination of changes (Callahan, Mayer & Steadman, 1987).
As might be expected, there is considerable and intense debate in the legal and empirical literature as to whether the "insanity standard" adopted produces differences in the rate or type of criminal defendant who is successful with the insanity defense (Keilitz, 1987). In part, the empirical difficulty is due to the fact that changes in the "insanity standard" tend to covary with other changes in the criminal justice-mental health "mega-system" (Golding & Roesch, 1987; Keilitz, 1987), and cross-jurisdiction studies with a longitudinal data base are notoriously difficult to carry out (Golding, Eaves & Kowaz, 1989; Steadman & Morrissey, 1986). Keilitz (1987) believes that the weight of evidence supports the notion that the "ALI rule" produces a broader and larger class of acquittees than the "M'Naghten test." While an extensive analysis of the reasons for continued dissatisfaction with "insanity standards," and the impact which these standards may have on juror decision-making is beyond the scope of this chapter (see Finkel, 1989; Golding & Roesch, 1987; Hermann, 1983; Keilitz, 1987), it is clear, from both psychological and jurisprudential perspectives, that certain scholarly debates and empirical issues repeatedly arise in the consideration of the attribution of criminal responsibility, and it is to these that we now turn.
Current psychological and jurisprudential perspectives. Of the many issues which arise with respect to the insanity defense, a subset have the most interest from a psychological perspective. First and foremost, what sort of an expert data base exists that concerns the relationship between psychological characteristics of defendants and their capacities to intend, reason, cognize, appreciate and control their behavior and its consequences? While critics continue to assault the existence of any reasonable data base (Faust & Ziskin, 1988), most of the empirical data and scholarly commentary show this position to be wildly exaggerated (Fukunaga et al., 1981; Golding & Roesch, 1987; Melton et al., 1987; Morse, 1986; Petrila, 1982; Resnick, 1986; Rogers et al., 1984; Steadman et al., 1983). Much of this chapter concerns a review of sub-issues that pertain directly to this question. However, the most "sensitive" question, concerning the ability to conceptualize the exact mechanisms by which certain kinds of mental disturbance influence an individual's ability to cognize and intend, continues to need empirical and scholarly evaluation. While the evidence reviewed indicates that, by and large, the individuals we do identify are rationally classified, it is less clear that we understand the underlying psychological mechanisms of the influence of mental state on volitional capacities.
Second, research on the insanity defense is a natural laboratory for the study of psychological processes underlying attribution in an applied social cognitive context (Golding & Roesch, 1987; Roberts & Golding, 1990). Whether or not it occurs in a "naive" context (global attitude) or in an "informed" context (during a trial), the interaction of prior attitudes and beliefs, personality factors, and "objective" evidence is a useful laboratory for the study of attributional processes. Thus, research on the insanity defense is also reviewed from this perspective.
Finally, research on the insanity defense uncovers certain societal assumptions that influence our policies with respect to the mentally ill, in both civil and criminal contexts. This perspective generates a useful link between psychological data and public policy. We now turn to a selective review of the psychological research on the insanity defense from these perspectives.
Conceptualizations of Criminal Responsibility
Lay attitudes and the ascription of criminal responsibility. Within Judeo-Christian cultures, the ascription of guilt and criminal responsibility has always recognized the importance of disordered mental state, even though surrounded by social debate and moral conflict. Thus, while such cultures have recognized the existence of insanity and its relevance to guilt and moral culpability, tension has always surrounded the personal and public processes involved in its ascription. Ancient texts (e.g., the Babylonian Talmud, pre- and post-Christian Greek and Roman texts) overtly address the problems of idiocy and lunacy, the distinctions between negligent and intentional mental states, and the problem of their determination (Platt & Diamond, 1965, 1966; Sayre, 1932; Stroud, 1914). Equally present are trace records of conflict over the role of mental state, motives for feigning insanity, and the role of expert witnesses.
Modern surveys of public attitude and experimental measurement of individual attitudes and beliefs (Ellsworth et al., 1984; Hans, 1986; Hans & Slater, 1983; Pasewark, 1986; Roberts, Golding & Fincham, 1987) find considerable support for the abstract logic of the insanity verdict, but also undercover widely held negative attitudes. The primary problems identified concern protection of the public, inadequate release and supervision of acquittees, perception of misuse and malingering, and ease of achieving the verdict. These bases of negative attitude are virtually identical to those found in historical records of public attitude as manifested in the media (Golding & Roesch, 1987). Interestingly, the empirical data on these points continues to accumulate, as it has since Guy's (1869) brilliant study over a century ago, that the public's actual knowledge of the defense is considerably inaccurate, their negative attitudes notwithstanding. Thus, the lay public seriously overestimates the use of the insanity plea and its success, underestimates the degree of agreement among professionals and the courts, and misunderstands the consequences of insanity verdicts and its associated dispositions (Hans, 1986; Pasewark, 1986; Roberts, Golding & Fincham, 1987; Rogers, Bloom & Manson, 1984; Utah v. Shickles, 1988).
General attitudes toward the insanity defense and their role in decision-making have been investigated only recently. Different studies, using either college populations or public opinion surveys, have uncovered similar, but not identical, attitudinal components using different item pools (Hans, 1986; Roberts & Golding, 1990; Roberts, Golding & Fincham, 1987). The components reported, labels aside, tap attitudes and beliefs about the underlying logic of the insanity defense (strict liability versus mens rea), concerns about the misuse of the plea and verdict (misuse, loophole, influence of experts), dispositional and safety concerns (length and type of supervision and release, future dangerousness), general attitudes toward the criminal justice system and punishment (consequences of guilt, treatment, death penalty), and a series of specific factors having to do with verdict options (confusion over "guilty but mentally ill," "insanity," "diminished capacity," etc.). Further development of the measurement of this attitudinal domain is needed. Specifically, the sufficiency and articulation of the item domain, its stability, and the overlap with related attitudinal domains (attitudes toward law and order, control, moral and religious issues) need to be examined. While the measurement of the emerging attitudinal components needs to be viewed as preliminary, it is clear that such attitudes exert considerable influence on both ultimate decision making and construal of information prior to decision making. Ellsworth, Bukaty, Cowan and Thompson (1984), Homant and Kennedy (1987; 1986), Roberts and Golding (1989), and Roberts, Golding and Fincham (1987) have all uncovered strong associative relationships between attitudes toward criminal responsibility and tendency to view a fact pattern as supporting a finding of not guilty by reason of insanity.
In an attempt to partially untangle the influence of attitudes, information, construal of information, and verdict options, Roberts and Golding (1990) experimentally manipulated the presentation of information, and measured the influence of attitudes and information construal on individual pre-deliberation judgments of insanity and criminal responsibility. Undergraduate subjects were randomly assigned to conditions of a between-subjects design that manipulated the degree of planfulness, characteristics of the defendant's thought disorder and verdict options. Subjects' construal of the information, their attitudes and pre-deliberation verdicts were measured. In a separate study, attitudes were measured before exposure to case information. The essential results demonstrate the complex interactions between objective characteristics of the judgment task (information and verdict form) and subject's attitudes and construal of case information. When both objective design variables (characteristics of the defendant, judicial instruction and verdict form) and subjective case-construal variables and attitudes span the predictor space in a discriminant analysis of predicted verdicts as the criterion, 66% of the criterion variance is predictable from a linear combination of attitudinal components (particularly, degree of endorsement of a strict liability orientation), case construal variables (particularly, whether the defendant knew he was killing a person, whether the defendant was perceived to be capable of acting differently, and whether the defendant was perceived as able to appreciate the wrongfulness of his actions) and verdict form (as a design variable). (The impact of verdict form is discussed in a subsequent section on the GBMI verdict option). Thus, subjects who reached different pre-deliberation verdicts construed the cases in systematically different terms. Regardless of design variables, when defendants were perceived as being more disordered, as having less knowledge that their victim was a person, less appreciation of the wrongfulness of their conduct, and less capacity for acting differently, reasoning logically in their situation, and understanding reasonable alternative conduct, then subjects strongly "voted" in favor of an insanity verdict. Their attitudes toward the underlying premise of the insanity defense also played a significant role in determining verdict choice independent of its correlation with case construal variables. Using a rather different methodology with respect to the presentation of legal evidence and standards, Finkel and Handel (1989) also focussed upon the construals of decision makers in the sanity context, and found that mock jurors engage in complex and discriminating construals of evidence in line with their verdict choices: guilt versus not guilty by reason of insanity. Similarly, in a different type of fact situation, Pennington and Hastie (1986) found that decision makers actively construed the fact patterns into "schema" or "stories" that are consistent with their judgments.
Results of this type are of interest both theoretically and practically. At the theoretical level, they underscore the importance of complex interactions between attitudes, construal and social attribution, and perception in social contexts (Wyer & Srull, 1986). Subjects' attitudes toward the insanity defense, when measured prior to information about a particular case, were inconsistently related to judgments and construals. As (Abelson, 1988) has argued, it is understandable that generalized attitudes, measured out of applied or "committed" contexts, will only weakly relate to other characteristics of psychological information processing and judgment. While Roberts and Golding's (1990) use of a between-subjects design precludes analysis of change of attitude within the person, comparison of mean levels suggests that subjects changed their attitudes as a function of exposure to contextualized information. In addition, their attitudes-in-context were significantly related to both their judgments and their processing of information. Such findings reinforce the conceptual importance of examining the complex interaction between juror attitudes and the manner of presentation of evidence as they influence individual and group decision making.
These data should not be construed as implying that individual decision makers "ignore" evidence (Visher, 1987) or are confused or "bamboozled" by mental health information or experts at insanity trials. Rather, the scant empirical evidence (Finkel & Handel, 1989; Golding & Roesch, 1987; Roberts & Golding, 1990; Roberts, Golding & Fincham, 1987) seems to imply that individuals pay very close attention to the case information, and make informed, critical and conservative evaluations of the evidence presented to them. The empirical data (Janofsky et al., 1989; Pasewark, 1986; Pasewark & McGinley, 1985) clearly indicate that the plea is seldom successful and that triers of fact, whether judge or jury, rarely make such findings. Decisional variance, in either simulation studies or statistics reflecting court outcomes, can be markedly influenced by choosing case descriptions at the extremes. Thus, the classic delusional defendant, with a long mental health history, who kills the object of his torment in a spontaneous manner, has a high probability of being adjudged NGRI, whereas the "wise con" who attempts to feign insanity and has no history of disorder, or the generally personality disordered defendant with a long history of rule violation has a much lower (but non-zero) probability of such an outcome. Within most cases, however, and particularly with "close cases" that raise extremely problematic moral and forensic issues, attitudinal and construal factors play an extremely important role. Thus, the clinical evidence on the degree of planfulness present in non-criminal patients' behavior and cognitions will sometimes be used to argue that the planfulness evident in a criminal defendant's actions suggest, that schizophrenic history aside, this act was the product of choice. Jurors or judges will then have to weigh implicit moral, psychological and attributional theories more carefully, and their cognitive and affective processes will influence naturally their information processing and decision making.
Methodologies for the assessment of criminal responsibility. Rogers and his colleagues have attempted to devise an rating scale methodology to quantify the professional mental health examiner's decision-making process. The Rogers Criminal Responsibility Assessment Scales (RCRAS) (Rogers, Wasyliw & Cavanaugh, 1984; Rogers and Cavanaugh, 1981) were originally designed to map the underlying conceptual terms of the "ALI rule" into reliable and quantifiable variables. Based upon comprehensive interviews and testing of a client, the forensic examiner rates a series of scales grouped into five sub-areas: a) reliability of report; b) organicity; c) psychopathology; d) cognitive control; and e) behavioral control; in addition, a series of more global ratings on final judgments of insanity and impairment are typically made. Thus, the RCRAS is an instrument that reflects relative importance assigned by examiners to the "first-order" elements of an "insanity-decision." It should be noted that these "elements" are fairly abstract psychological and legal terms (e.g., "delusions at the time of alleged crime") and do not necessarily represent the cues that are actually utilized by professional examiners in making their decisions.
Rogers has reported that inter-rater reliabilities average 0.58 at the item level, kappa coefficients range from 0.49 to 1.00 at the sub-area level, and are quite high, 0.93, at the level of final decision (Rogers, Seeman & Clark, 1986; Rogers, Wasyliw & Cavanaugh, 1984). Interestingly, the lowest reported reliability (0.49) is for the "product question," (i.e. was the loss of control attributable to underlying psychopathological disturbance), one of the most frequent sources of disagreement in contested trials.
Rogers has also reported an high level of overall agreement (88%) between examiners and triers of fact (96% with respect to sanity with lower levels of agreement on insanity [70%]) (Rogers, Cavanaugh, Seeman and Harris, 1984). These findings are in general accord with the levels of agreement between clinicians and courts found in other studies of final judgment that use no formalized interviews or rating scales (Rogers, Bloom & Manson, 1984; Fukunaga et al., 1981; Stock & Poythress, 1979). Unfortunately, the validity of the RCRAS in these situations is difficult to assess because the published validity data use criterion-contaminated groups, i.e. either a) groups defined as sane or insane on the basis of the RCRAS itself, thus measuring in reality only item-total correlations, or b) groups based on court decisions where the same examiner's conclusions heavily influence court outcome. Thus, Rogers et al. (1986) show that the RCRAS scales do a good job of discriminating between groups judged as "insane" under ALI, M'Naghten and GBMI standards, but the criterion groups are formed by clinician judgment, not by independent judges.
Slobogin, Melton & Showalter (1984) have attempted to make mental state at the time of the offense (MSO) evaluations more legally and forensically relevant by developing an investigative technique that is influenced strongly by legal criteria and psychopathology in legal context (see, also, Golding & Roesch, [1987] for a similar analysis of an idealized view of methods for conducting such investigations). The MSO (for greater detail, see Melton et al., 1987) is a semi-structured interview technique that assumes considerable sophistication in legal and psychopathological issues. Unlike the RCRAS, the interview material is not reduced to scores, and little data have appeared on the psychometric characteristics of the technique, although Slobogin et al.(1984) have analyzed the ability of the MSO to function as a screening device. It achieves its highest level of inter-rater agreement on "screen out" decisions (97%), i.e. decisions that a defendant is an inappropriate candidate for the insanity defense.
The critical area in which devices like the RCRAS and MSO can make their most important contributions is in clarifying for the trier of fact the underlying bases for professional judgment (Golding, 1990; Morse, 1985; Rogers, 1987), hence potentially highlighting the areas of disagreement so that expert testimony can be of more assistance to judge or juror. Because the central issue in the admissibility and usefulness of expert testimony on the insanity issue remains the association between organic or psychopathological disturbance and control/moral judgment capacities (Golding, 1990; Golding & Roesch, 1987), these devices are most useful when they serve the heuristic value of a) highlighting the aspects of the defendant's psychological state that are relevant, b) describing a purported relationship to control and judgment capacities, and c) organizing know data about the empirical relationships between disorder and psychological capacities in various states and situations. Thus, devices such as RCRAS are somewhat useful in describing the normative processes of professional judgment, but the critical need, both theoretically and empirically, is to articulate and test theories of criminal responsibility and control, and to compare the critical elements of such theories to juror and judge decisional strategies and to normative moral models.
Characteristics of NGRI Verdicts and Dispositions
While a number of studies have described the NGRI population in traditional demographic and diagnostic terms (summarized in Pasewark, 1986), fewer studies have concentrated on variables that are of inherent theoretical interest. From the perspectives on the insanity defense identified at the beginning of this chapter, three aspects of the NGRI population are of most interest: a) do the characteristics defendants who are successful or unsuccessful indicate that juror decision-making is "rational," and do the data indicate that the jurors are weighing factors according to the "proper schemata?"; b) do the characteristics of those "successfully" acquitted differ from those convicted or withdrawing the plea in ways that are jurisprudentially sensible?; and c) do the empirical characteristics of NGRI cohorts suggest that the social policies surrounding the process are sound?
NGRI verdicts. A reasonably coherent set of data emerge about the NGRI verdict. It is seldom raised and not often successful (Janofsky et al., 1989; Pasewark, 1986). Contrary to the public statements of advocates of the "guilty but mentally ill" (GBMI) option, there is no support for the contention that the insanity plea is either abused or a loophole. In 1978, for example, there were approximately 2.3 million felony arrests in the United States of which 1,625 (0.07%) were adjudicated as not guilty by reason of insanity (Steadman & Braff, 1983). Similarly, Janofsky, Vandewalle and Rappeport (1989) found that only 1.2% of all indicted defendants entered insanity pleas, and, of those, only 10% were "successful." Furthermore, most studies of these "insanity acquittals" find that between 80% to 90% are by pretrial agreement between the defense and prosecution (e.g. Rogers, Bloom & Manson, 1984). In successful cases, there is considerable agreement between the mental health professionals, and between their recommendations and court outcome (see section of professional agreement below). Successful insanity defenses rarely occur in front of jurors (Boehnert, 1989) and most often involve plea negotiations or bench trials. The public's view (Hans, 1986) notwithstanding, there is little indication that the role of the mental health professional improperly influences jurors' insanity verdicts (Cohen, 1988; Golding & Roesch, 1987; Rogers & Ewing, 1989), although the mechanisms through which they do communicate with judges and juries alike is an important area of future study. Successful insanity acquittees typically have long histories of serious mental disorder and involvement in the justice system, and they are most often so disturbed at the time of arrest that they are adjudicated as incompetent to proceed even before they ultimately are found NGRI (Packer, 1987; Roesch & Golding, 1987; Golding, Eaves & Kowaz, 1989). Contrary to the thrust of some reform movements which are based upon the notion that many antisocial individuals misuse the insanity defense, there never has been any empirical support for this assertion (Golding & Roesch, 1987; Rogers, 1988; Rogers & Zimbarg, 1987; United States v. Lyons, 1984a, 1984b). Finally, post-acquittal release procedures are quite strict in most states and the most of the available evidence indicates that NGRI acquittees spend as much or more time confined as do those convicted of similar crimes (Braff, Arvanities & Steadman, 1983; Cooke & Sikorski, 1974; Pantle, Pasewark & Steadman, 1980; Pogrebin, Regoli & Perry, 1986).
Characteristics of NGRI populations. Jurisprudentially, there are obvious and fundamental differences between a civil commitment population and an NGRI population (Parry, 1987). Empirically, however, it is not clear whether NGRI acquittees form a meaningfully different class with respect to civil committees, or whether they represent a sub-set defined by longitudinal events. As might be expected, both populations are characterized by prior criminological and hospitalization experiences. Insofar as stable data are available, they tend to indicate that prior arrest records are more frequent in the NGRI population. Rates generally fall into the 30 - 45% range (Bogenberger et al., 1987; Pasewark, 1986; Steadman & Braff, 1983) for NGRI acquittees and somewhat less in civil committee populations (Heilbrun, Heilbrun & Griffin, 1988; Hiday, 1988). There is considerable variation among reported studies reflecting important jurisdictional differences in the way individuals are processed (Keilitz, 1987; Steadman, 1987). It is important to note, however, that when reported, it seems clear that most prior charges in both cohorts are for relatively non-serious crimes (Golding, Eaves & Kowaz, 1989; Hiday, 1988; Pasewark, 1986).
Most studies of NGRI cohorts have reported high rates of prior hospitalizations and mental health contacts. Typically, more than two-thirds of samples have prior hospitalizations, with rates ranging from 20 to 80% (Bogenberger et al., 1987; Cooke & Sikorski, 1974; Criss & Racine, 1980; Golding, Eaves & Kowaz, 1989; Pasewark, 1986). Again, the data are similar to the civil committee population (Hiday, 1988) with the NGRI cohort tending to have somewhat higher rates. Very few studies report a breakdown by types of prior hospitalizations (civil vs. criminal vs. voluntary), outcomes of prior hospitalizations, or other mental health contacts. Pasewark et al. (1979b) found that 44% of their male NGRI subjects had prior hospitalizations; the average individual had two prior civil and 0.6 prior criminal hospitalizations. Phillips and Pasewark (1980) report that 61% of an NGRI sample had prior hospitalizations, and that 60% of these were "criminal" hospitalizations (competency, lateral transfer, etc.). A detailed examination of a large NGRI cohort by Golding, Eaves and Kowaz (1989) found that 78.7% had been previously hospitalized and an additional 5.3% had received outpatient treatment only. Of those who had prior admissions, the mean was 4.11 hospitalizations. Of the aggregate of prior admissions, 43.4% were for forensic reasons involving remand, fitness or civil commitment. Thus, almost 80% of the sample had significant mental health histories involving frequent hospitalizations, and approximately three-quarters of the sample having prior contacts were hospitalized at least once for reasons that tied their mental disorder to either criminal charges or concerns over their dangerousness to self or others.
When analyzed chronologically, the data of Golding, Eaves and Kowaz (1989) reveal another important aspect of NGRI cohorts. Over half of the subjects with prior admissions were discharged within one year of their index offense; in fact, 44.6% committed their index offense within six months of their last discharge. A quantitative and qualitative analysis of the nature of these "just prior hospitalizations" is quite revealing. Only 22.5% were voluntary hospitalizations. The majority were involuntary commitments for dangerous acts towards self or others (42.3%). The remaining reasons for admission all include legally significant aspects of the individual's mental disorder (16.2% as mentally disordered offender transferred from jail or prison, 8.5% as unfit to stand trial, and 2.8% for fitness evaluations). Over three-quarters of the subjects had been previously hospitalized one or more times prior to this "just prior" hospitalization. Qualitatively, an analysis of the existing records indicated that the individual's predominantly psychotic symptomatology (76% were psychotic) was substantially related to their legally relevant behavior.
The logical implication of these data is that the vast majority of individuals who were ultimately found NGRI for their index offenses were identified as suffering from legally relevant mental disorders prior to their index offenses. A large group of the NGRI acquittees were within the forensic system and/or the mental health system prior to their offense. It is thus critical for future research to examine subject characteristics that may have identified them as in need of sustained treatment, which treatment may have prevented the index offense. An alternative problem concerns whether or not this sample differs from the civil commitment population, or at least that portion of it that repetitively cycles through hospitalizations. Pursuing this question should be a major research agenda in the future. Can one find ways to predict for an individual in the system that this person is a likely candidate for further re-hospitalizations or legally relevant mental health incidents (civil commitment, remand, unfit, NGRI, dangerousness, etc.). The study would have to sample from an unconstrained sample of individuals having hospitalizations of various types and look for predictors of subsequent forensic contacts.
Thus, while NGRI acquittees have many characteristics in common with the civil committee population, they may represent a certain subpopulation (Bloom et al., 1983) that penetrates the larger criminal justice-mental health "megasystem" in certain, perhaps predictable, ways. The complex "megasystem" interdependencies in forensic populations is illustrated in Menzies' (1987) study of pre-trial assessment unit (see also Menzies & Webster, 1987). Only 14 percent of a cohort of 571 patients experienced no subsequent criminal or mental health contacts over a two year period. During follow-up, half of the sample were re-hospitalized at least twice for an average of two months. Roughly two-thirds of the cohort had multiple subsequent police contacts, arrests, convictions, and incarcerations. One quarter of the cohort had at least one subsequent remand during the follow-up period. For the whole cohort, roughly 25% of their follow-up period was spent in one form of an institution or another. These data on "trans-institutionalization" tend to be quite consistent (Steadman et al., 1989; Morrissey & Goldman, 1986). As elaborated in the concluding comments, they indicate that the primary problems associated with the insanity defense are not its misuse, but rather a confused social-mental health policy for individuals who are quite seriously disturbed, whose disturbance repeatedly has social-behavioral consequences, and who need to be served by a more coherent social and legislative policy.
Correspondence of professional and judicial decision making. A small but consistent body of literature has examined either judgmental agreement between mental health professionals, or agreement between mental health professionals' recommendations and final judicial outcome. Very high levels of inter-examiner agreement and examiner-judicial outcome agreement are reported (Criss & Racine, 1980; Fukunaga et al., 1981; Howard and Clark, 1985; Jeffrey et al., 1988; Pasewark, 1986; Petrila, 1982; Rogers, Bloom & Manson, 1984; Rogers, Wasyliw & Cavanaugh, 1984; Steadman et al., 1983; Stock & Poythress, 1979). Within these studies, a more fine-grained analysis of factors associated with disagreement produces a relatively coherent picture. Individuals with convincing psychopathological histories, seriously incapacitating mental disorders and reliably determinable disturbances of thought, perception and affect, tend to be agreed upon as "insane and not guilty on that account." Individuals whose actions are seen as rational, motivated by understandable impulses, and with antisocial personality and/or behavioral histories are the subject of most disagreement (see also Beckham, Annis & Gustafson, 1989). Rogers, Bloom and Manson's study (1984) typifies results in this area: 86% of their sample were uncontested and agreed upon dispositions, usually involving severely impaired and psychotic individuals. Thus, the long standing public perception of the degree of disagreement with respect to the insanity defense among experts (see Golding & Roesch, 1987, for a comprehensive review), does not correspond to the empirical data, and must be based upon other attitudinal factors.
Length of NGRI dispositions. In Jones v. United States (1983), the Supreme Court of the United States affirmed that an insanity acquittee is not to be treated as a criminal convictee: the issues pertaining to commitment and release must focus on the individual's mental illness and future dangerousness, not on punishment for the "crime" they committed (Parry, 1987). However, because their dangerous acts have been found to occur "beyond a reasonable doubt," the Court would allow the determination of their continuing mental illness and dangerousness to be decided according to a "preponderance of the evidence" standard rather than the "clear and convincing" evidence standard which holds for ordinary civil commitment [Addington v. Texas (1979)].
While the legal standard for the release of NGRI acquittees is thus current mental disorder and its relationship to future dangerousness, the available empirical evidence strongly supports a different view. Length of initial periods of hospitalization for NGRI acquittees has been found, in general, to be primarily determined by the nature of the individual's criminal offense (Braff, Arvanities & Steadman, 1983; Cooke & Sikorski, 1974; Golding, Eaves & Kowaz, 1989; Heilbrun, Heilbrun & Griffin, 1988; Pasewark, 1986; Steadman & Morrissey, 1986). In addition, while there is some variability in findings, NGRI acquittees tend not to be institutionalized for periods of time that significantly shorter than either unsuccessful NGRI pleaders or non-NGRI convictees.
Mulvey, Blumstein & Cohen (1986) provide a detailed and scholarly review of the methodological problems of analyzing institutionalization statistics or the criminal behavior of released mental patients. Like Golding, Eaves and Kowaz (1989), and Steadman and Morrissey (1986), they have argued that research designs and models for understanding the factors associated psychopathological or criminal behavior will need to include complex conditional probabilities. Thus, the "outcome" or "disposition" questions cannot be addressed independent of the different multiple paths representing type of treatment, type of supervision, nature of patient history, and so forth. A few studies of insanity acquittees have taken steps in this direction. Because length of institutionalization data, by themselves, have little clinical or social policy significance without knowing what treatments were provided, and with what outcome.
Golding, Eaves and Kowaz (1989) sought to estimate the degree of clinical response obtained by NGRI acquittees at various points in their institutional careers. By intensively analyzing the careers of 188 NGRI acquittees, they found that the average subject was held 49.9 months in their initial hospitalization, with individuals charged with murder or attempted murder spending a significantly longer 71.9 months prior to first supervised discharge to the community. However, there was no evidence that individuals hospitalized longer were significantly more disturbed than the individuals in other groups, nor did they differ on other demographic, diagnostic or criminological variables (age, sex, marital status, ethnic group, prior charges (type or number), juvenile record, number of prior serious offenses, and diagnostic category). Most importantly, in terms of the Jones standard, the groups do not differ significantly with respect to the clinical profiles obtained at either admission or discharge. However, detailed analyses with respect to legally significant time-points reveals that most significant and substantial psychopathological change occurs between fitness admission and fitness discharge in almost all areas of measurement from both sources. Thus, almost all individuals whose psychopathology responds to the psychotropic medications offered have made most or all of their significant change before they ever enter the hospital for their first post-adjudication hospitalization which lasts, on average, 49.9 months.
Criminal and psychopathological "recidivism". Not surprisingly, the data on subsequent adjustment, re-arrest, and reasons for failure of community tenure of NGRI acquittees is confusing at best. Most studies find that a substantial proportion of the samples followed are re-arrested (Bieber et al., 1988; Morrow & Peterson, 1966; Pantle, Pasewark & Steadman, 1980; Pasewark, Pantle & Steadman, 1982; Phillips & Pasewark, 1980; Spodak, Silver & Wright, 1984; but see Bloom, Rogers & Manson, 1982). Considering the restriction of range within this population, it is not surprising that prediction of post-release recidivism is difficult. Bieber et al.'s (1988) study of 132 insanity acquittees found only modest predictability (80% classification accuracy compared to base-rate predictions of 71%). Past arrests, severity of arrests, psychoticism , homicide as the NGRI offense, and escape attempts were the most significant predictors. Interestingly, length of time hospitalized was not. Similarly, Bloom et al. (1986) found that discharge status (maximum time versus clinical judgment) did not discriminate between post-discharge arrest rates, averaging 51 percent.
The majority of studies examining the issue have also found high re-admission or utilization of psychiatric services by discharged NGRI acquittees (Bogenberger, Pasewark, Gudeman & Bieber, 1987; Pasewark, Pantle & Steadman, 1979a). An additional complicating factor in research of this type is that re-admission for reasons of deterioration of mental status depend upon the quality and intensity of the follow-up procedures, and sometimes mask alternative dispositions for behavior that might have been illegal (Golding, Eaves & Kowaz, 1989; Bloom et al., 1986; Steadman & Morrissey, 1986). Most research also supports the notion that individuals who have had difficulties in the criminal justice and mental health systems prior to their NGRI status, particularly the chronically psychotic and aggressive sub-population, are the primary subjects who re-offend or who are re-admitted.
In general, the data reported on various follow-up studies of NGRI acquittees point to the fact that the characteristics of their outcome are heavily dependent upon the conditions of their supervision and treatment, and their prior histories. For example, Cavanaugh and Wasyliw (1985), reporting on a sample of NGRI acquittees treated on an outpatient basis, report that, under close supervision, only 1/44 patients were either arrested or convicted, and that significant change in psychopathology and level of adjustment were noted. Similarly, Cohen, Spodak, Silver & Williams (1988) provide one of the few studies that finds sufficient variability in hospital adjustment and discharge data to find significant relationships between psychological status on discharge and future adjustment in the community. As discussed in the policy section, these findings are tentative because of the dearth of studies utilizing a long-term individualized follow-up design. Steadman (1987) makes the compelling argument that research in the area of mental health and the law needs to address a set of interrelated design issues: a) it needs to be cross-jurisdictional, so as not to sample the idiosyncracies of particular jurisdictions, decision-makers, and the like; b) it needs to systematically and empirically address the impact of various legal rules, decisions and options on the actual systems-wide outcome; c) it needs to take the longitudinal and multi-decision point nature of the criminal justice-mental health interface into account. Avison and Speechly (1987) in their review of the community outcome data for discharged psychiatric patients in general also found a chaotic pattern of weak predictors, and a lack of attention to detailed analysis of reasons for the success or failure of community tenure. Thus, while these data are only tentative, they point to the need to think of this population as being a high-risk group that will need continual monitoring (Lamb, Weinberger & Gross, 1988; Bloom et al., 1986). The task then becomes one of designing such a system that is cost-efficient, constitutional, and reflective of the values of our society.
The alternative verdict: "Guilty but mentally ill"
History of the GBMI verdict. The "guilty but mentally ill" verdict form has an interesting, if confused, history which presents current challenges with respect to the psychology of the attribution of criminal responsibility, and the jurisprudential meaning of guilt and mental illness. As has been documented in this chapter and elsewhere (Golding & Roesch, 1987; Hermann, 1983), the moral and psychological attribution problems surrounding the insanity defense have always found expression in calls for reform in the defense and the processes surrounding it. Forensic professionals had already been calling for an empirical evaluation of the validity of underlying premises, and an examination of the processes and dispositions for insanity acquittees by the time that Guy (1869), in the Nineteenth Century, published an analysis of the rates of insanity acquittals, before and after controversial insanity cases. Guy described the situation as an attempt to
discover some facts...which may go far to set these questions at rest; and perchance the truth, if we can discover it will be found to set us free from the exaggerations which hang about...as well as from the panics which are apt to seize on those who make the safety of the State, and the protection of the innocent against violence and fraud, their supreme law (p.159).
In a time-series from 1836 to 1867, of total acquittal rates (as a function of type of crime, adjusted to a constant population), Guy shows that not only did celebrated trials not result in an increase in insanity pleas and acquittals, but, to the contrary, the entire series is reasonably flat, that the fluctuations are as observable in minor crimes as in high profile murderous assaults, and in convictions as well as in insanity acquittals.
The reasons given for the need to adopt a reform in the verdict, from "not guilty by reason of insanity " to "guilty but mentally ill" fall into certain overlapping classes. Jurisprudentially, the argument has been that fundamental mens rea assumption is an unworkable construct in modern criminal law (see Golding & Roesch, 1987; Hermann, 1983; Morse, 1988, 1986, 1985, 1982 for reviews), hence such mitigating psychological factors ought to be part of the disposition after a finding of guilt. In terms of legislative or media-based opinion, the concern has been over protection of the public (insanity acquittees at large) or the duping of the judicial process by the feigning of insanity or the "bamboozling" of the jury by expert witnesses who are "hired guns" (Golding, 1990; Hans, 1986). Some professionals have also advocated the alternative plea on grounds of providing mental health services to a larger segment of the criminal population. Finally, some advocates of the alternative verdict have been quite direct in their view that it represents a way of "curbing or eliminating" the insanity verdict (see Keilitz, 1987; McGraw, Farthing-Capowich & Keilitz, 1985; and Roberts, Golding & Fincham, 1987 for reviews).
Twelve states have now enacted GBMI legislation in one form or another (see McGraw, Farthing-Capowich & Keilitz, 1985, for detailed reviews of these statutes). The original GBMI legislation in this century was introduced in Michigan in 1975 following People v. McQuillan (1974), where 150 patients were released after a finding that Michigan's automatic commitment of NGRI acquittees was unconstitutional. Subsequent to McQuillan, two acquittees re-offended most spectacularly. When John Hinckley was acquitted as not guilty by reason of insanity for the attempted murder of President Reagan, the stage was set for another round of attempted reform of the insanity defense. The Congress passed the Insanity Defense Reform Act (1984) and other states passed GBMI legislation or otherwise continued to alter their criminal responsibility codes (Golding & Roesch, 1987; Hermann and Sor, 1983; Keilitz, 1987) in response to the perceived abuses of the insanity plea. Logically and empirically, all of the premises and reasons for the adoption of the GBMI alternative can be shown to be either false or unrealized, with unanticipated consequences.
Reduction in "abuse" of the NGRI verdict. A primary aim of the GBMI legislation was to make it more difficult to win "acquittal" by reason of insanity. While it is difficult to interpret time-series data using the point of reform as the interrupt (because other changes in legislation have tended to accompany the implementation of GBMI legislation), these data and some laboratory simulation studies are the only existing sources of information as to the effect of the GBMI verdict option. However, careful analysis of the existing data point to reasonably secure conclusions.
In the first place, there wasn't much abuse in the traditional insanity plea to be reformed, public and media-opinion to the contrary. As detailed above (see section, "Attitudes and the ascription of criminal responsibility," and references therein), the formal sanity issue is infrequently raised as a defense, it is difficult to "win," and most cases (averaging 90%) where the issue is raised are resolved without dispute, media coverage, or "battles of the experts." Data on the number of NGRI verdicts reached in various jurisdictions are notoriously unstable, but survey data in most jurisdictions and hard data in a few, led Keilitz (1987) to conclude that "if the GBMI alternative has limited the insanity defense, it has done very little indeed" (p. 320). In fact, in most jurisdictions the number of NGRI verdicts stays constant or rises, and additional individuals are found GBMI.
Since both NGRI and GBMI outcomes are modally the result of accepted pleas or plea bargains, however, these data do not address directly the subtler issue of whether the GBMI verdict option has its other intended effect, namely, an attempt to influence juror decision making, with the hope that it would allow them a "middle ground" between guilty and NGRI. Several simulation studies, using somewhat different methodologies, have shown significant effects in decision making by individuals (but not constituted as "mock juries"). Roberts, Golding & Fincham (1987) found large within-subject displacement effects in experimental conditions at either end of the traditional criminal responsibility continuum. When the GBMI verdict option was added to the traditional not guilty, guilty, and NGRI provisions (described according to the ALI rule), the percentage of subjects judging a "NGRI prototype" vignette (a clearly delusional schizophrenic defendant whose delusion was related to the victim and whose aggressiveness showed a relative lack of planfulness) as NGRI dropped from 95% to 18%. At the other end of the continuum, when the GBMI option is added, subjects judging a prototype of the antisocial personality disorder changed their "guilty" verdicts from 91% to 35%. Roberts and Golding (1990) replicated this effect in a between-subjects design in several samples. Also using vignette methodologies, Finkel has found effects from the adoption of the GBMI verdict option (see studies reviewed in Finkel and Handel, 1989) as have Savitsky and Lindblom (1986).
Thus, the introduction of the GBMI verdict, in simulation studies, appears to create a "collapsing of the extreme," precisely by creating a verdict that is inherently contradictory, mirroring the moral tension in societal policies and attitudes. Individual decision makers find the face logic of the GBMI option (Roberts, Golding & Fincham, 1987) extremely attractive -- of course the evidence supports that he did it (guilty), but he clearly has serious mental problems. Jurors are likely to also be swayed by such arguments, although no attempts to uncover actual decision making by juries or judges (which is rare with the GBMI option) have yet been made. Thus, the ultimate generalization question is to the decision making behavior of judges or jurors under "real" conditions. The attitudinal and individual simulation data point clearly in the direction of such displacement effects, and this is consistent with the informal reports of attorneys, prosecutors and judges (Hermann & Sor, 1983; Klofas & Weisheit, 1986; Utah v. Shickles, 1988).
GBMI as a dispositional alternative. GBMI legislation was defended and advocated on other grounds that served to add a gloss of rehabilitation. The new verdict would explicitly recognize that some of those sent to prison were in need of mental health treatment. Unfortunately, this assumption bears no relationship to reality. There is no evidence that individuals found GBMI are any more likely to receive treatment for their mental disorder (Keilitz, 1987). Treatment, when provided, is usually under the auspices of the Department of Corrections. There is no evidence of new funds that have been appropriated to the prison systems to provide more treatment (Beasley, 1983). Additionally, provisions already existed, in every state that passed GBMI, to laterally transfer a disturbed prisoner into mental hospital settings for treatment if that was necessary (Golding & Roesch, 1987). Hartstone, Steadman and Monahan (1982), in a survey of treatment facilities for mentally disordered offenders within the correctional systems of six state jurisdictions found that mental health problems among inmates were both "over-identified" and "underidentified." That is, approximately half of the staff of both correctional and secure mental health facilities felt that prisoners in need of mental health treatment were not being appropriately transferred to treatment settings, whether within the department of corrections or within the department of mental health. Approximately one-third of the staff also felt that significant mental health problems amongst inmates were not even being identified, much less dealt with appropriately. At the same time, these same staff felt that a significant group of inmates were being transferred "inappropriately," that is for management of behavioral problems rather than for treatment of mental disorders. Thus, the provisions exist, but tend not to be properly utilized, with or without the GBMI verdict option.
In commenting on the GBMI verdict from an academic perspective, Bonnie (1983) observed bluntly, "(The guilty but mentally ill verdict) should be rejected as nothing more than moral sleight of hand" (p.194). A person found or pleading GBMI may be sentenced to any term that is statutorily specified for the crime charged, including the death penalty (McGraw, Farthing-Capowich & Keilitz, 1985); GBMI convictees are also, in many states, subject to more stringent conditions of probation and parole than other convicts. Existing research also indicates that GBMI convictees are likely to serve longer sentences than those in general population. Finally, they are no more likely to receive mental health treatment than they would as ordinary convicts in departments of correction (Keilitz, 1987; Morgan et al., 1988) despite some claims to the contrary (McKay & Kopelman, 1988; Mickenberg, 1987).
Unlike the NGRI verdict, the GBMI verdict is simply a guilty verdict, and all meaning after that is superfluous. "Guilty but mentally ill" means guilty but confers no special rights with respect to treatment or other considerations due to mental illness. No state which has adopted such statutes has provided either additional treatment resources to the Departments of Correction, nor have state courts held that a finding of GBMI confers a right to treatment for such convicts (People v. Marshall, 1983). At most, defendants found GBMI have the right to periodic review of their mental condition ("periodic inquiry and examination ... concerning the nature, extent, continuance, and treatment of the defendant's mental illness" [Ill. Rev. Statutes, 1983, Ch. 38, 1005-2-6{b}], but there is no statutory obligation to actually provide mental health treatment beyond minimally adequate medical care, and few states do so within their correctional systems.
The GBMI verdict, in its generalized form (Keilitz, 1987), requires the trier of fact to find that the defendant committed the acts charged, but that was mentally ill, though not legally insane. Michigan, for examples, "defines" this distinction [mental illness (not rising to insanity)] as "substantial disorder of thought or mood which significantly impairs judgment, behavior, capacity to recognize reality, or ability to cope with the ordinary demands of life" (Michigan Revised Statutes, 1982). While many have argued that the distinction is inherently confusing and unreliable (Golding & Roesch, 1987; Hermann & Sor, 1983), courts have uniformly rejected appeals on that ground. The GBMI verdict option appears to be here to stay, but there is little evidence that it produces any greater degree of protection for the public or treatment for defendants, nor does it render the underlying jurisprudential and societal conflict over the attribution of criminal responsibility more coherent. Paradoxically, although GBMI was designed as an alternative verdict, presumably to correct perceived deficiencies in NGRI verdict, and allow jurors to find a middle ground between verdicts of guilty and not guilty by reason of insanity, the empirical evidence available consistently points to GBMI being primarily a plea used by defendants (Keilitz, 1987; Klofas & Weisheit, 1986; McKay & Kopelman, 1988; Mickenberg, 1987; Morgan et al., 1988; Smith & Hall, 1982), with more than two-thirds of the verdicts being a result of pleas rather than verdicts by judge or jury. In terms of attitudes toward the GBMI verdict option, it is surprising that attorneys who are involved in work with the option tend to have serious misconceptions of the effect of being found GBMI. Thus, in Klofas and Weisheit's (1986) study, the majority of defense attorneys believed, like jurors, that an advantage to a defendant being found GBMI was that they would receive treatment not otherwise available.
Research directions and social policy
Research and scholarship on the insanity defense has concentrated on a number of issues that have both jurisprudential and psychological importance. Clearly, most lay opinions about the verdict bear little resemblance to the actual practice surrounding insanity defenses and acquittals. While this has obvious implications for eduction and social policy, it also uncovers more interesting aspects of the attributional processing that surrounds the determination of lack of criminal responsibility. Several studies have uncovered reliable and substantial effects that prior "naive" attitudes have upon the social construction of reality ("evidence"), operating primarily through an individual's construal of the evidence presented. As seems to be the case with respect to the construal of social interactions (Golding, Valone & Foster, 1982), information about the "objective" world of attribution of responsibility (Roberts & Golding, 1990; Roberts, Golding & Fincham, 1987) is apprehended through the prism of attitudes, beliefs, and personality. The research reported here is the logical starting place for a more comprehensive analysis of the process. By sampling an individual's construal of the evidence presented, Roberts and Golding (1990) were able to uncover differential profiles of construal of evidence, as a function of both objective reality, subject's schemata, and attitudes. Thus subjects made subtle discriminations in the attribution of capacity to control between delusional defendants whose psychotic beliefs involve the threat of imminent danger from the delusional object and equally psychotic defendants whose delusions were un-related to their acts. We need to explore the dimensions of construal of acts with respect to the issue of responsibility and the implicit theories of psychopathology (Golding & Roesch, 1987) held by persons of different cultural, personological and religious backgrounds. Simultaneously, the structure of the domain of attitudes in this area needs to be explored and related to patterns of construal.
The research reviewed here also provides directions for future research on the longitudinal development of psychopathology and response to various classes of intervention. Data reported by Golding, Eaves & Kowaz (1989) suggests, but certainly does not prove, that even with proper regimes of psychotropic medication and supervision, certain types of chronic mental patients re-experience legally, socially and personally problematic re-institutionalization. Is this pattern significantly different for involuntarily committed patients who are dangerous to self only, or to other types of chronic patients? More importantly, what psychopathological mechanisms of disturbance remain untouched by the current, widely used, psychotropic treatments?
The longitudinal study of individuals whose capacity to control their thoughts, emotions or volitional cognitions is impaired is also a prime goal for future research. Lay subjects make differential "implicit psychopathological assumptions" based upon the characteristics of the defendant's delusions and cognitions and behaviors that reflect intentionality, but we have a scant data base that explores these relationships empirically. While Justice Rehnquist may believe that considerations of mental disorder would "import into this area of constitutional law notions of `free will' that have no place here" (Colorado v. Connelly, 1986, p.523), it clearly does belong there, but we need to understand the mechanisms of influence more precisely. This sort of inquiry needs to focus on methods for the empirical study of volition and its conceptualization in severe mental disorders.
Research on the insanity defense also has direct and significant implications for social and legislative policy. In one sense, the Jones court was perfectly correct: the issue is response to treatment and future dangerousness. However, the transinstitutional system for monitoring and adjudicating the chronically mentally ill whose psychotic behavior is legally meaningful, is based upon some erroneous assumptions and is quite inefficient by any standard. Golding, Eaves and Kowaz (1989) estimated that the per individual cost for an insanity acquittee, during commitment and supervision periods alone, came close to one-half of a million dollars. The data described here provide a cogent basis for the financial, legal and social reform of the adjudication of these defendants. This sub-population of the chronic mental patient domain has strong mental health treatment needs, which should be addressed in the larger mentally ill population prior to a sub-population of this group engaging in high levels of aggressive behavior that are a function of their psychopathology more than their criminality. The safety of society is not served by cosmetic change that exacerbates the problem by adding even more individuals (GBMI's who would have been found guilty) to the disordered offender population, but adding no additional resources.
References
Abelson, R. P. (1988). Conviction. American Psychologist, 43, 267-275.
Addington v. State of Texas. 99 S. Ct. 1804 (1979).
American Law Institute. (1962). Model penal code. Philadelphia: Author.
Avison, W. R., & Speechley, K. N. (1987). The discharged psychiatric patient: A review of social, social-psychological, and psychiatric correlates of outcome. American Journal of Psychiatry, 144, 10-18.
Ballantine, H. W. (1919). Criminal responsibility of the insane and feeble-minded. Journal of the American Institute of Criminal Law and Criminology, 9, 485-499.
Beasley, W. R. (1983). An overview of Michigan's guilty but mentally ill verdict. Michigan Bar Journal, 62, 204-205;215-217.
Beckham, J. C., Annis, L. V., & Gustafson D. J. (1989). Decision making and examiner bias in forensic expert recommendations for not guilty by reason of insanity. Law and Human Behavior, 13, 79-87.
Bieber, S. L., Pasewark, R. A., Bosten, K., & Steadman H. J. (1988). Predicting criminal recidivism of insanity acquittees. International Journal of Law and Psychiatry, 11, 105-112.
Bloom, J. D., Faulkner, L., Shore, J. H., & Rogers J. L. (1983). The young adult chronic patients and the legal system: A systems analysis. In D. L. Cutler (Ed.), Effective aftercare for the 1980's: New directions for mental health services. San Francisco: Jossey-Bass.
Bloom, J. D., Rogers, J. L., & Manson S. (1982). After Oregon's insanity defense: A comparison of conditional release and hospitalization. International Journal of Law and Psychiatry, 5, 391-402.
Bloom, J. D., Rogers, J. L., Manson, S. M., & Williams M. H. (1986). Lifetime police contacts of discharged Psychiatric Security Review Board clients. International Journal of Law and Psychiatry, 8, 189-202.
Boehnert, C. E. (1989). Characteristics of successful and unsuccessful insanity pleas. Law and Human Behavior, 13, 31-39.
Bogenberger, R., Pasewark, R. A., Gudeman, H., & Beiber S. L. (1987). Follow-up of insanity acquittees in Hawaii. International Journal of Law and Psychiatry, 10, 283-295.
Bonnie, R. J. (1983). The moral basis of the insanity defense. American Bar Association Journal, 69, 194-197.
Braff, J., Arvanities, T., & Steadman, H. (1983). Detention patterns of successful and unsuccessful insanity defendants. Criminology, 21, 439-449.
Callahan, L., Mayer, C., & Steadman, H. (1987). Insanity defense reform in the United States -- Post-Hinckley. Mental and Physical Disability Law Reporter, 11, 54-59.
Cavanaugh, J. L., & O. E. Wasyliw. (1985). Treating the not guilty by reason of insanity outpatient: A two year study. Bulletin of the American Academy of Psychiatry and Law, 13, 407-415.
Cohen, D. (1988). Punishing the insane: Restriction of expert psychiatric testimony by Federal Rule of Evidence 704(b). University of Florida Law Review, 40, 541-562.
Cohen, M. I., Spodak, M. K., Silver, S. B., & Williams K. (1988). Predicting outcome of insanity acquittees released to the community. Behavioral Sciences and the Law, 6, 515-530.
Colorado v. Connelly, 479 U. S. 157 (1986).
Cooke, G., & Sikorski, C. R. (1974). Factors affecting length of hospitalization in persons adjudicated not guilty by reason of insanity. Bulletin of the American Academy of Psychiatry and the Law, 2, 251-261.
Criss, M. L., & Racine, D. R. (1980). Impact and change in legal standard for those adjudicated not guilty by reason of insanity 1975-1979. Bulletin of the American Academy of Psychiatry and Law, 8, 261-271.
Crotty, H. D. (1924). The history of insanity as a defence to crime in English criminal law. California Law Review, 12, 105-123.
Ellsworth, P. C., Bukaty, R. M., Cowan, C. L., & Thompson, W. C. (1984). The death-qualified jury and the defense of insanity. Law and Human Behavior, 8, 81-93.
Faust, D., & Ziskin, J. (1988). The expert witness in psychology and psychiatry. Science, 241, 31-35.
Finkel, N. J. (1989). The Insanity Defense Reform Act of 1984: Much ado about nothing. Behavioral Sciences and the Law, 7, 403-419.
Finkel, N. J., & Handel, S. F. (1989). How jurors construe "insanity". Law and Human Behavior, 13, 41-59.
Fukunaga, K., Pasewark, R., Hawkins, M., & Gudeman, H. (1981). Insanity plea: Interexaminer agreement and concordance of psychiatric opinion and court verdict. Law and Human Behavior, 5, 325-328.
Golding, S. L. (1990, in press). Mental health professionals and the courts: The ethics of expertise. International Journal of Law and Psychiatry.
Golding, S. L., & R. Roesch. (1987). The assessment of criminal responsibility : A historical approach to a current controversy. In I. B. Weiner & A. K. Hess (Eds.), Handbook of forensic psychology New York : Wiley.
Golding, S. L., Eaves, D., & Kowaz A. (1989). The assessment, treatment and community outcome of insanity acquittees: Forensic history and response to treatment. International Journal of Law and Psychiatry, 12, 149-179.
Golding, S. L., Valone, K. E., & Foster, S. W. (1982). Interpersonal construal: An individual differences framework. In N. Hirschberg & L. Humphreys (Eds.), Multivariate applications in the social sciences, 163-193. Hillsdale, J.J.: Erlbaum.
Gray, S. (1972). The insanity defense: Historical development and contemporary relevance. American Criminal Law Review, 10, 559-583.
Guy, W. A. (1869). On insanity and crime-and on the plea of insanity in criminal cases. Royal Statistical Society (London Journal Series A), 1869, 159-191.
Hans, V. P. (1986). An analysis of public attitudes toward the insanity defense. Criminology, 24, 393-414.
Hans, V. P., & Slater, D. (1983). John Hinckley, Jr. and the insanity defense: The public's verdict. Public Opinion Quarterly, 47, 202-212.
Hartstone, E., Steadman, H. J., & Monahan J. (1982). Vitek and beyond: The empirical context of prison-to- hospital transfers. Law and Contemporary Problems, 45, 125-136.
Heilbrun, K., Heilbrun, P. G., & Griffin, N. (1988). Comparing females acquitted by reason of insanity, convicted, and civilly committed in Florida: 1977-1984. Law and Human Behavior, 12, 295-311.
Hermann, D. H. J. (1983). The insanity defense: Philosophical, historical, and legal perspectives. Springfield, Ill.: C. Thomas.
Hermann, D. H. J., & Sor, Y. S. (1983). Convicting or confining? Alternative directions in insanity law reform: Guilty but mentally ill versus new rules for release of insanity acquittees. Brigham Young University Law Review, 1983, 499-638.
Hiday, V. A. (1988). Civil commitment:a review of empirical research. Behavioral Sciences & The Law, 6, 15-43.
Homant, R. J., & Kennedy, D. B. (1986). Judgment of legal insanity as a function of attitude toward the insanity defense. International Journal of Law and Psychiatry, 9, 67-81.
Homant, R. J., & Kennedy, D. B. (1987). Subjective factors in clinicians' judgments of insanity: Comparison of a hypothetical case and an actual case. Professional Psychology: Research and Practice, 18, 439-446.
Howard, R. C., & Clark, C. R. (1985). When courts and experts disagree : Discordance between insanity recommendations and adjudications. Law and Human Behavior, 9, 385-395.
Illinois Revised Statutes, Chapter 38, Sections 6-2, 115-1 to 115-4, and 115-6 (1983).
Janofsky, J. S., Vandewalle, M. B., & Rappeport, J. R. (1989) Defendants pleading insanity: An analysis of outcome. Bulletin of the American Academy of Psychiatry & Law, 17, 203-211.
Insanity Defense Reform Act of 1984. (1984). Public Law 98-473, Sections 401-406.
Jeffrey, R. W., Pasewark, R. A., & Bieber S. (1988). Insanity plea: Predicting not guilty by reason of insanity adjudications. Bulletin of the American Academy of Psychiatry and Law, 16, 35-40.
Jones v. United States. 103 S. Ct. 3043 (1983)
Keedy, E. R. (1917). Insanity and criminal responsibility. Harvard Law Review, 30, 535-560, 724-738.
Keilitz, I. (1987). Reforming and researching the insanity defense. Rutgers Law Review, 39, 289-322.
Klofas, J., & Weisheit, R. (1986). Pleading guilty but mentally ill: Adversarial justice and mental health. International Journal of Law and Psychiatry, 9, 491-501.
Lamb, H. R., Weinberger, L. E., & Gross B. H. (1988). Court-mandated community outpatient treatment for persons found not guilty by reason of insanity: A five-year follow-up. American Journal of Psychiatry, 145, 450-456.
M'Naghten's Case, 8 Eng. Rep. 718 (1843).
McGraw, B. D., Farthing-Capowich, D., & Keilitz I. (1985). The "guilty but mentally ill" plea and verdict: Current state of the knowledge. Villanova Law Review, 30, 117-191.
McKay, R. D., & J. Kopelman. (1988). The operation of the "guilty but mentally ill" verdict in Pennsylvania. Journal of Psychiatry and Law, 16(2), 247-268.
Melton, G. B., Petrila, J., Poythress, N. G., & & Slobogin C. (1987). . Psychological evaluations for the courts: A handbook for mental health professionals and lawyers New York : Guilford Press.
Menzies, R. J. (1987). Cycles of control: the trancarceral careers of forensic patients. International Journal of Law and Psychiatry, 10, 233-249.
Menzies, R. J., & Webster, C. D. (1987). Where they go and what they do: the longitudinal careers of forensic patients in the medicolegal complex. Canadian Journal of Criminology, 29, 275-293.
Michigan Compiled Laws Annotated, Section 330.1400a, (1982).
Mickenberg, I. (1987). A pleasant surprise: The guilty but mentally ill verdict has both succeeded in its own right and successfully preserved the traditional role of the insanity defense. University of Cincinnati Law Review, 55, 943-996.
Morgan, D. W., McCullough, T. M., Jenkins, P. L., & White W. M. (1988). Guilty but mentally ill: The South Carolina experience. Bulletin of the American Academy of Psychiatry and Law, 16, 41-48.
Morrissey, J. P., & Goldman, H. H. (1986). Care and treatment of the mentally ill in the United States: Historical developments and reforms. In S. A. Shah (Ed.), The law and mental health : Annals of the American Academy of Political and Social Science, 484, 12-27.
Morrow, W. R., & Peterson, D. B. (1966). Follow-up of discharged psychiatric offenders- 'Not guilty by reason of insanity' and 'criminal sexual psychopaths' Journal of Criminal Law Criminology and Police Science, 57, 31-34.
Morse, S. J. (1982). Failed explanations and criminal responsibility: Experts and the unconscious. Virginia Law Review, 68, 971-1084.
Morse, S. J. (1985). Excusing the crazy: The insanity defense reconsidered. Southern California Law Review, 58, 777-836.
Morse, S. J. (1986). Psychology, determinism and legal responsibility. In G. B. Melton (Ed.), Nebraska Symposium on Motivation: Vol. 33. The law as a behavioral instrument (pp. 35-85). Lincoln : University of Nebraska Press.
Morse, S. J. (1988). Treating crazy people less specially. West Virginia Law Review, 90, 353-385.
Mulvey, E. P., Blumstein, A., & Cohen J. (1986). Reframing the research question of mental patient criminality. International Journal of Law and Psychiatry, 9, 57-65.
Packer, I. K. (1987). Homicide and the insanity defense: A comparison of sane and insane murders. Behavioral Sciences and the Law, 5, 25-35.
Pantle, M., Pasewark, R., & Steadman, H. (1980). Comparing institutionalization periods and subsequent arrests of insanity acquittees and convicted felons. Journal of Psychiatry and Law, 8, 305-316.
Parry, J. (1987). The civil-criminal dichotomy in insanity commitment and release proceedings:Hinckley and other matters. Mental and Physical Disability Law Reporter, 11, 218-223.
Pasewark, R. (1986). A review of research on the insanity defense. In S. A. Shah (Ed.), The law and mental health : Annals of the American Academy of Political and Social Science, 484, 100-114.
Pasewark, R. A., & McGinley, H. (1985). Insanity plea: National survey of frequency and success. Journal of Psychiatry and Law, 13(1-2), 101-108.
Pasewark, R. A., Pantle, M. L., & Steadman H. (1979a). The insanity plea in New York State, 1965-1976. New York State Bar Journal, April, 1979, 186-225.
Pasewark, R. A., Pantle, M. L., & Steadman H. (1979b). Characteristics and disposition of persons found Not Guilty by Reason of Insanity in New York State, 1971-1976. American Journal of Psychiatry, 136, 655-660.
Pasewark, R., Pantle, M. L., & Steadman H. J. (1982). Detention and rearrest rates of persons found not guilty by reason of insanity and convicted felons. American Journal of Psychiatry, 139, 892-897.
Pennington, N., & Hastie, R. (1986). Evidence evaluation in complex decision-making. Journal of Personality and Social Psychology, 51, 242-258.
People v. Marshall, 448 N. E. 2d 969 (Ill. App. 3d, 1983).
People v. McQuillan, 221 N.W. 2d 569 (Supreme Court of Michigan, 1974).
Petrila, J. (1982). The insanity defense and other mental health dispositions in Missouri. International Journal of Law and Psychiatry, 5, 81-101.
Phillips, B. L., & Pasewark, R. A. (1980). Insanity plea in Connecticut. Bulletin of the American Academy of Psychiatry and the Law, 8, 335-344.
Platt, A. M., & Diamond, B. L. (1965). The origins and development of the 'wild beast' concept of mental illness and its relation to theories of criminal responsibility. Journal of Historical and Behavioral Science, 1, 355-367.
Platt, A. M., & Diamond, B. L. (1966). The origins of the "right and wrong" test of criminal responsibility and its subsequent development in the United States: An historical survey. California Law Review, 54, 1227-1259.
Pogrebin, M., Regoli, R., & Perry, K. (1986). Not guilty by reason of insanity: A research note. International Journal of Law and Psychiatry, 8, 237-241.
Pollock, F., & Maitland, F. (1952). History of English Law: Vol. I (2nd ed.). Cambridge: Cambridge University Press.
Resnick, P. J. (1986). Perceptions of psychiatric testimony: A historical perspective on the hysterical invective. Bulletin of the American Academy of Psychiatry and Law, 14, 203-219.
Roberts, C., & Golding, S. L. (1990, manuscript in preparation). The role of thought disorder, planfulness, and judicial instructions on the social construction of criminal responsibility and insanity.
Roberts, C., Golding, S. L., & Fincham, F. (1987). Implicit theories of criminal responsibility: Decision making and the insanity defense. Law and Human Behavior, 11, 207-232.
Roesch, R., & Golding, S. L. (1987). Defining and assessing competency to stand trial. In I. B. Weiner & A. K. Hess (Eds.), Handbook of forensic psychology New York : Wiley.
Rogers, R. & Ewing, C. P. (1989). The ultimate opinion proscription: a cosmetic fix and a plea for empiricism. Law and Human Behavior, 13, 357-374.
Rogers, J. L., Bloom, J. D., & Manson, S. M. (1984). Insanity defenses: Contested or conceded? American Journal of Psychiatry, 141, 885-888.
Rogers, R. (1987). Assessment of criminal responsibility: Empirical advances and unanswered questions. Journal of Psychiatry and Law, 15, 73-82.
Rogers, R. (1988). APA's position on the insanity defense:empiricism versus emotionalism. American Psychologist, 42, 840-848.
Rogers, R., Cavanaugh, J. L., Seman, W., & Harris, M. (1984). Legal outcome and clinical findings: A study of insanity evaluations. Bulletin of the American Academy of Psychiatry and the Law, 12, 75-83.
Rogers, R., & Cavanaugh, J. L. (1981). Rogers Criminal Responsibility Scales. Illinois Medical Journal, 160, 164-169.
Rogers, R., & Zimbarg, R. (1987). Antisocial backgrounds of defendants evaluated for insanity: A research note. International Journal of Law and Psychiatry, 10, 75-80.
Rogers, R., Seeman, W., & Clark C. R. (1986). Assessment of criminal responsibility: Initial validation of the R-CRAS with the M'Naghten and GBMI standards. International Journal of Law and Psychiatry, 9, 67-75.
Rogers, R., Seman, W., & Stampley J. (1984). A study of socio-demographic characteristics of individuals evaluated for insanity. International Journal of Offender Therapy & Comparative Criminology, 28, 3-10.
Rogers, R., Wasyliw, O. E., & Cavanaugh, J. L. (1984). Evaluating insanity: A study of construct validity. Law and Human Behavior, 8, 293-304.
Savitsky, J. C., & W. D. Lindblom. (1986). The impact of the guilty but mentally ill verdict on juror decisions: An empirical analysis. Journal of Applied Social Psychology, 16, 686-701.
Sayre, F. B. (1932). Mens rea. Harvard Law Review, 45, 974-1026.
Shah, S. A. (1986). Criminal responsibility. In A. L. McGarry, S. A. Shah & W. J. Curran (Eds.), Forensic psychiatry and psychology: Perspectives and standards for interdisciplinary practice. Philadelphia, PA : Davis.
Slobogin, C., Melton, G. B., & Showalter, C. R. (1984). The feasibility of a brief evaluation of mental state at the time of offense. Law and Human Behavior, 8, 305-321.
Smith, G. A., & Hall, J. A. (1982). Evaluating Michigan's guilty but mentally ill verdict: An empirical study. Michigan Journal of Law Reform, 16, 75-112.
Smith, R. (1981). Trial by medicine: Insanity and responsibility in Victorian trials. Edinburgh: Edinburgh University Press.
Spodak, M. K., Silver, S. B., & Wright C. U. (1984). Criminality of discharged insanity acquittees: Fifteen year experience in Maryland reviewed. Bulletin of the American Academy of Psychiatry and Law, 12, 373-382.
State v. Pike. 49 N. H. 399 (Sup. Ct. N.H., 1869).
State v. Strasburg. 110 P. 1020 (Sup. Ct. Wash.,1910).
Steadman, H. J. (1987). Mental health law and the criminal offender: Research directions for the 1990's. Rutgers Law Review, 39, 323-337.
Steadman, H. J., & Braff, J. J. (1983). Defendants found not guilty by reason of insanity. In J. Monahan & H. Steadman (Eds.), Mentally disordered offenders: Perspectives from law and social science New York : Plenum.
Steadman, H. J., & Morrissey, J. P. (1986). The insanity defense: Problems and prospects for studying the impact of legal reforms. In S. A. Shah (Ed.), The law and mental health : Annals of the American Academy of Political and Social Science, 484, 115-126.
Steadman, H. J., Callahan, L. A., Robbins, P. C., & Morrissey J. P. (1989). Maintenance of an insanity defense under Montana's "abolition" of the insanity defense. American Journal of Psychiatry, 146, 357-360.
Steadman, H. J., Keitner, L., Braff, J., & Arvanities, M. A. (1983). Factors associated with a successful insanity plea. American Journal of Psychiatry, 140, 401-405.
Stock, H. V., & Poythress, N. G. (1979, August). Psychologists' opinions on competency and sanity: How reliable? Paper presented at the American Psychological Association, New York.
Stroud, D. A. (1914). Mens rea or imputability under the laws of England. London: Sweet & Maxwell.
Trial of Lunatics Act, 46 & 47 Vict., c. 38 (1883).
United States v. Lyons, 731 F. 2d 243 (Fifth Cir., 1984a).
United States v. Lyons, 739 F. 2d 994 (Fifth Cir., 1984b).
Utah v. Shickles. 760 F. 2d 291 (Utah Sup. Ct. 1988).
Visher, C. (1987). Juror decision making: The importance of evidence. Law and Human Behavior, 3, 95-106.
Walker, N. (1968). Crime and insanity in England, Volume I: The historical perspective. Edinburgh: University of Edinburgh Press.
Wyer, R. S., & Srull, T. K. (1986). Human cognition in its social context. Psychological Review, 93, 322-359.