"The defense of insanity is probably the most controversial issue in all criminal law," according to Melton et al. (1997, p. 187). In the earlier edition of Psychological Evaluations for the Courts, Melton et al. (1987) wrote "… insanity cases often provide instances for society's expression of more general frustrations or disagreements about law, personal responsibility, and justice." (p. 157)

Public myths about the insanity defense abound. Pasewark and Seidenzahl (1979) found that college students believed that 37% of felony cases involve a NGRI plea and that 44% of these attempts are successful. In contrast, Steadman, McGeevy, Morrissey, Callahan, Robbins, & Cirincione (1993) reported that the insanity defense is raised in about 1% of felony cases and has a success rate of about 26%. Perlin (1996) detailed numerous misconceptions about the insanity defense, including perceptions that it is (a) overused, (b) is limited to murder cases, (c) entails no risk to the defendant, (d) if successful, results in release, (e) is frequently faked, (f) results in frequent battle of experts, (g) is misused by defense attorneys, and (h) that mental health experts cannot accurately assess MSO. In fact, the great majority of NGRI cases are settled by agreement between prosecution and defense, indicating that both sides see the defendant as insane at the time of the offense (Perlin, 1996).

Insanity Standards
Assessment of legal sanity must be distinguished from a diagnosis of mental illness: The presence of mental disorder, even psychosis at the time of the offense, does not equate to a finding of NGRI. The Texas insanity standard is set out in the Section 8.01 of the Penal Code, is that "… at the time of the conduct charged, the actor, as a result of severe mental disease or defect, did not know his conduct was wrong." Several points must be emphasized. A qualifying condition must be a (a) severe (b) mental disease or defect (c) that the defendant from knowing (d) at the time of the alleged offense (e) that the illegal behavior was wrong.

To qualify for an insanity defense, the mental illness in question must be "severe." Definitions of "severe" and "mental disease or defect" are not provided, at least in the Texas statute. Rogers and Shuman (2000) point out that it is not clear whether "severe" is meant to limit the defense to specific disorders (i.e., schizophrenia), or to restrict consideration only to severe forms of many disorders.

The mental condition must prevent the person from knowing the quality of his or her actions. The word "knowing" can be interpreted narrowly (i.e., the person intellectually knew the act was wrong) or broadly (i.e., understood and appreciated the implications and moral context). The defect in knowledge or appreciation must be at the same time as the offense, and it must interfere with the knowledge or appreciation that the action was "wrong." However, it is also clear that juries do not always hold to such a strict construction.

The Texas standard has its roots in the M'Naghten standard, which had its origin in England following the acquittal of John M'Naghten in 1843. Recognized as outmoded at the time it was introduced, the M'Naghten test held force for over a hundred years, both in Great Britain and the U.S., until Durham v. United States in 1954. The Court proclaimed a major revision of the traditional standard, holding that "an accused is not criminally responsible if his unlawful act was the product of mental disease or defect." In other words, if the psychiatric condition caused the criminal act, an insanity defense was warranted. Spurred by concerns that the legal standard trailed advances in psychiatric knowledge, the court sought to allow psychiatric experts to readily translate their terminology into relevant legal observations and conclusions (Melton, et al., 1997; Rogers & Shuman, 2000). However, the legal community quickly became disenchanted with the new rule, finding that experts often gave conclusive statements on the ultimate question (NGRI) without offering reasoning to support their conclusions. In 1962, the Court (MacDonald v. United States) added guidance in defining the requisite mental disorder: "The jury should be told that mental disease or defect includes any abnormal condition of the mind which substantially affects mental or emotional processes and substantially impairs behavior controls." (p. 851)

Continued reflection on the NGRI defense resulted in a position statement by the American Law Institute (ALI). Based on years of study and discussion between legal and mental health scholars, the ALI standard found a person not criminally responsible 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 conform his conduct to the requirements of the law.

The ALI standard was significant for several substantial reasons. It specifically included "irresistible impulse" as an excusing condition, which later was to prove controversial. Secondly, it used the word "appreciate" rather than "know," and required only "substantial" impairment of appreciation. As such, it was a liberal rule by today's standards. It was formally adopted into law in United States v. Brawner (1972).

The ALI test was adopted by a majority of jurisdictions, but criticism gradually mounted. Some felt that psychiatric testimony was displacing the trier-of-fact in NGRI determinations, and that terms such as "substantial" and "appreciate" were too amorphous. But the primary problem was the so-called volitional prong. Both the American Psychiatric Association and the American Bar Association deemed the distinction between an irresistible impulse and one that is not resisted as ephemeral and not subject to scientific scrutiny. Both organizations called for the removal of the irresistible impulse criterion (Melton et al., 1997). However, the revamped standard may fail to protect people suffering from bipolar disorder (manic depression), who are prone to acting impulsively and who obviously suffer from a serious mental illness.

After John Hinckley attempted to assassinate Ronald Reagan and was acquitted on an insanity defense, the public was outraged. Congress responded by reversing the thirty-year trend, begun with Durham, of liberalizing the NGRI defense. The new standard, the Insanity Defense Reform Act, is binding in federal courts and was adopted by many other jurisdictions. Under it, a defendant pleading NGRI must prove that "as a result of severe mental disease or defect, he was unable to appreciate the nature and quality or wrongfulness of the act." In contrast to ALI, the IDRA explicitly places the burden of proof on the defendant, specifies that the mental condition must be "severe," and the volitional prong is eliminated. In the aftermath of Hinckley, all but eight states revised their insanity statutes (Rogers & Shuman, 2000), presumably with the intent to limit NGRI acquittals. Utah, Montana, and Idaho went so far as to eliminate their insanity defense.

There has been considerable research on the effect of different insanity standards on acquittal rates and mock juror decisions. Generally, differences between insanity standards do not translate into action: Jurors tend to acquit about the same number of people under "right from wrong," "appreciates the wrongfulness," and "irresistible impulse" standards. This may be because jurors follow their own moral standards and ask, in short, "How crazy was he?" In fact, studies of mock jurors suggest that this issue is the most important consideration for those assessing NGRI, followed by the defendant's culpability in causing (or failing to treat) his or her own mental illness (Finkel & Slobogin, 1995).

Lastly, the need for jury instruction on the insanity issue has been questioned. Both researchers and legal scholars have argued that juries can make subtle and sophisticated judgments on individual cases. Unlike any formal insanity standard, jurors consider the defendants' role in bringing about their insanity. In other cases, limiting the judgment to knowledge of right from wrong, or appreciation, may fail to capture the whole sanity issue. Compare the M'Naghten, Durham, or ALI standards with the one proposed by Judge David Bazelon of the DC Circuit Court of Appeals-"if at the time of his unlawful conduct his mental or emotional processes or behavioral controls were impaired to such an extent that he cannot justly be held responsible." (italics added)

Referral for Insanity Evaluation
In local county criminal and district courts, MHMRA is contracted to provide competency and sanity evaluations. Evaluations can also be sought through private psychologists and psychiatrists. In Ake v. Oklahoma, the Supreme Court ruled that a defendant seeking the insanity defense is entitled to a partial expert to help develop an insanity defense. Regardless of who performs the evaluation, the attorney should seek to request medical records and facilitate the examiner's contact with family members and others who can provide information about the defendant's psychiatric problems. This is especially important if the defendant has legitimate mental illness and tends to minimize his condition. The attorney should also consider the risk of information being disclosed that is potentially harmful to the defendant-what is said during the evaluation is not confidential. Some examiners will include the defendant's account of the offense in the report, which may amount to a confession or be used to impeach other statements the defendant has made.

The Evaluation
Examiners at MHMRA typically conduct two competency and insanity evaluations per day. This is a brisk pace and means that not every information lead can be explored. Several issues will probably orient the examiner to the probability of a valid insanity defense: 1) Is there a documented history of mental illness? 2) Does the crime have a rational motive? 3) Is the act of out character? 4) Is there indication of bizarre behavior or intoxication in the Offense Report? If an attorney believes that the defendant has a valid insanity defense that does not fit this mold, it is especially desirable to contact the examiner and facilitate the recovery of information that may otherwise be overlooked.

Several pieces of information may be crucial, including the defendant's account, medical chart notes at the jail, previous psychiatric, medical or school records, and statements of family members, arresting officers and witnesses. Generally, sources are more valuable to the extent they can attest to the defendant's thinking and awareness of right and wrong at the time of the offense. Therefore, medical notes immediately before and after the alleged offense are usually more valuable than more thorough records from five years earlier.

Many insanity referrals are relatively straightforward and the defendant may have little to say about his or her mental state. Often, guilt and a rational motive are admitted, or any involvement in the offense is denied. Sometimes the nature of the crime (selling crack to an undercover officer, burglary) makes an insanity defense unlikely, as mental illness is unlikely to lead to the behavior or prevent interfere with basic knowledge of legal consequences. However, for those cases in which the person was legitimately mentally ill at the time, and is willing to openly discuss his or her thoughts and feelings, a great deal of relevant information can be gathered. Although one could legitimately examine virtually every aspect of the person's history, it is usually most useful to focus on the defendant's life situation and concerns on during the week leading up to the offense and on the offense itself (Rogers & Shuman, 2000). The examiner should concentrate on the defendant's actions, perceptions (what he or she was aware of), emotions, and especially the thoughts and thinking processes before, during, and after the offense.

Traditional psychological tests are of little help in assessment of NGRI, unless the person was tested shortly before or after the offense, or if the person's mental status is very stable. This is because tests gauge primarily the person's current mental state and are not focused on issues most relevant to the insanity issue (the defendant's thinking, symptoms, and judgment regarding the offense). There is only one published instrument that is designed to assist in assessment of the insanity issue. The Rogers Criminal Responsibility Assessment Scales (R-CRAS) is a summary rating form that the examiner completes at the end of the assessment. The examiner provides ratings on thirty variables that are relevant to an insanity determination, such as the severity of a mental disorder present, the degree of planning, and the degree of disorganization in the defendant's speech and behavior. Each rating is accompanied by six short descriptions that define the severity of the feature present. The R-CRAS helps ensure that relevant topics are covered and it forces a specific judgment on each issue. Research by Richard Rogers and colleagues show it can be rated reliably and that the results are highly predictive of NGRI verdicts. The R-CRAS has not gained wide acceptance. More importantly, it does not provide much information for making the inferential leap from psychiatric symptoms to moral responsibility.

The finding of NGRI is a moral decision, not just a medical or psychological judgment. In federal court, mental health examiners are forbidden to give their opinion on the "ultimate issue" in insanity cases by FRE 704b. The issue has been controversial among forensic evaluators as well. As a practical matter, state statute in Texas, as well as the standard court orders, direct examiners to provide an opinion on the defendant's legal sanity.

Forensic examiners can help address the moral dimension more rigorously by defining the issue better and creating standardized instruments. For example, to what degree did the defendant plan the crime? Did he think about what would happen if he were caught? What the chance of this was? How this would affect his family? How it would affect the victim? Of course, many criminals do not ask themselves these questions either, or at least skirt them. But if a normally responsible person does not do so because of severe mental illness, one must question if he really understands the nature and quality of his actions.

Page 2