Abstract
We begin with a brief overview of the Not Guilty by Reason of Insanity (NGRI) and Guilty but Mentally Ill (GBMI) verdicts in the United States and then report on a study of qualified jurors (n = 96) in which we examined jurors’ understanding and attitudes about mental illness verdicts and the disposition of mentally ill defendants. Results indicate that although the jury pool was highly educated, only 4.2 percent of jurors could correctly identify both the definitions and dispositions of defendants found NGRI and GBMI. Jurors with lower educational levels were less likely to identify the dispositional outcome of a GBMI verdict (p < .05). Eighty-four percent of respondents believed that juries should be informed of dispositional outcome before deciding a verdict. Also, 68.4 percent of jurors erroneously believed that a defendant found GBMI could not receive the death penalty. Among jurors who correctly identified the definition of GBMI, those with lower educational levels were more punitive in their attitudes toward disposition of the GBMI defendants, believing they should eventually be sent to prison (p < .05).
The disposition of mentally ill offenders continues to be a controversial issue as lawmakers try to balance the public’s demand for retribution and safety with the offender’s civil rights and need for treatment. A large majority of states have enacted legislation that establishes an insanity defense. These statutes are based on the common law principle that the accused should not be held criminally responsible if he or she was “insane” at the time of the offense. This principle was summarized by Judge David Bazelon: “Our collective conscience does not allow punishment where it cannot impose blame” (Ref. 1, p 876). The criteria used to determine eligibility for a verdict of Not Guilty by Reason of Insanity (NGRI) vary across jurisdictions. Most states use some version of the M’Naghten rule, which requires that defendants be mentally ill to the extent that they do not know the nature and quality of their criminal acts or that their criminal acts are wrong.2 Most of the remaining states use the American Law Institute’s (ALI) standard, which holds that defendants are not criminally responsible if, as a result of mental disease or defect, they lacked substantial capacity either to appreciate the criminality of their conduct or to conform their conduct to the requirements of the law.3 In most jurisdictions, defendants found NGRI are hospitalized for treatment in a psychiatric hospital and released when they are no longer mentally ill or dangerous.
In addition to these insanity statutes, several states have enacted legislation that includes a separate Guilty but Mentally Ill (GBMI) verdict. The first GBMI statute was enacted in Michigan in 1975 in reaction to the Michigan Supreme Court’s decision in People v. McQuillan.4 Before this decision, insanity acquittees were automatically committed to the Michigan Department of Mental Health for an indeterminate time. This practice was declared unconstitutional because it violates due process and denies the insanity acquittees equal protection by failing to provide them with release procedures available to individuals committed under other civil commitment statutes. As part of the McQuillan decision, the court ordered that all previously committed insanity acquittees be evaluated and released if they no longer met criteria for civil commitment. A total of 270 patients were evaluated and 214 were released. Within a year of their release, two of the acquittees committed heinous and highly publicized crimes.5 The Michigan GBMI statute6 was enacted a year later in response to public outcry. The statute stated that a defendant could be found GBMI if the trier of fact found that the defendant was guilty of the offense, was mentally ill at the time of the offense, and did not lack the substantial capacity either to appreciate the nature and quality or wrongfulness of his conduct or to conform his conduct to the requirements of the law. It further stated that the same criminal sanctions should be imposed on a person found GBMI as on a person found guilty.
At least eight additional states created GBMI statutes in response to the insanity acquittal of John Hinkley, Jr., in 1982, after his attempted assassination of President Reagan. Currently, 12 states utilize some form of GBMI verdict: Alaska, Delaware, Georgia, Illinois, Indiana, Kentucky, Michigan, New Mexico, Pennsylvania, South Carolina, South Dakota, and Utah.7
Mental Illness Verdicts in South Carolina
South Carolina defines insanity (NGRI) as an affirmative defense that a defendant, as a result of mental disease or defect, “lacked the capacity to distinguish moral or legal right from moral or legal wrong or to recognize the particular act charged as morally or legally wrong.”8
The GBMI statute was enacted in South Carolina in April of 1984. The statute states that a person is GBMI9: …if at the time of the commission of the act constituting the offense, he had the capacity to distinguish right from wrong or to recognize his act as being wrong…but because of a mental disease or defect he lacked sufficient capacity to conform his conduct to the requirements of the law.
Persons adjudicated GBMI must be taken to a facility designated by the Department of Corrections for treatment and retained there until, in the opinion of the staff at the facility, such inmates can be safely moved to the general population of the Department of Corrections to serve the remainder of their sentences.
Criticisms of the GBMI Verdict
The American Psychiatric Association,10 American Psychological Association,11 and American Bar Association12 have all opposed the GBMI verdict. GBMI statutes have been widely criticized for several reasons.
One criticism is that persons found GBMI are subject to the same criminal sanctions as offenders who are found guilty, including incarceration and death. Some have argued that the statute may as well be called “guilty but ignorant” or “guilty but remorseful,” since GBMI does not indicate mitigated or diminished capacity.13 In fact, studies have shown that defendants found GBMI receive longer sentences. For example, Callahan et al.14 compared the sentences of those offenders who pled insanity and were subsequently found guilty with those who were found GBMI. Their results indicated that defendants found GBMI often received longer sentences than those found guilty with no mental disorder. Specifically, they found that those receiving a GBMI verdict were sent to prison 80 percent of the time (rather than receiving probation) as opposed to 62.8 percent of those found guilty without a mental illness. Also, those found GBMI received a life sentence 14 percent of the time as opposed to 5.5 percent of the time for those found guilty without mental disorder. This difference was even more significant when the charge was murder, as those found GBMI received life sentences 70.6 percent of the time, as opposed to 49.2 percent of the time for non-mentally ill defendants found guilty.
Another criticism of the GBMI verdict is that it does not always guarantee mental health treatment for the inmate. Some states make no provision for mandatory treatment. For example, the statute in Georgia specifically states that prisoners found GBMI receive treatment only as financial resources permit.15 In addition, in People v. Marshall16 the Illinois Court of Appeals ruled that failure to assure treatment for individuals found GBMI does not make the statute unconstitutional.
Another criticism is that jurors may not understand the difference between the NGRI and GBMI verdicts and may see a GBMI finding as a compromise verdict between guilty and NGRI.5 The GBMI verdict gives jurors two verdicts (guilty and guilty but mentally ill) by which to find a defendant guilty and only one verdict (Not Guilty by Reason of Insanity) by which to find the defendant not guilty when the insanity defense is raised. In one study, prosecutors agreed that the GBMI statute makes it easier to prosecute a case when a defendant pleads insanity.17
The Debate About Juror Instructions
A common misperception held by the public is that defendants found NGRI are released into the community just as any other acquitted individual would be.18 This has led some to argue that juries should always be instructed that NGRI acquittees will be committed for treatment and will not be released until they are judged not to be a threat to society. In response, numerous courts have ruled on whether a defendant has a right to have the jury instructed as to the consequences of an NGRI verdict. A small number of states have approved the use of instructions to a jury as to the outcome of an NGRI verdict (California, Colorado, Florida, Louisiana, Massachusetts, Maryland, Pennsylvania, and Utah).19 However, most courts have ruled that such instructions would distract a jury from its sole function as the trier of fact and that consideration of the disposition of these defendants would influence jury decision-making. These courts have held that jurors should only consider the statutory definitions of these verdicts during their deliberation.
This was the case in South Carolina, as decided in State v. Gary Allen Rimert.20 In this case, the defendant was found GBMI on a charge of murder and appealed the decision, in part, on whether the jury should have been instructed as to the dispositional consequences of the GBMI and NGRI verdicts. The Supreme Court of South Carolina upheld the trial judge’s decision not to instruct the jury regarding the dispositional consequences of these verdicts.
Study Hypotheses
This study was designed to test several hypotheses. Mental illness verdicts are not easily understood and are not part of common, everyday knowledge. Therefore, we hypothesized that most prospective jurors could not correctly identify the definitions and dispositional outcomes of the NGRI and GBMI verdicts. We also hypothesized that jurors would consider the consequences (i.e., perceived dispositional outcomes) of their verdicts and would let those beliefs influence their verdicts, even if instructed not to do so by the trial judge. This study was designed to measure jurors’ knowledge about two subjects: the meaning of these verdicts in South Carolina and the disposition of individuals found NGRI or GBMI. In addition, we assessed juror attitudes as to what they thought the disposition should be of defendants found NGRI or GBMI.
Method
This study was approved by the Institutional Review Board of the University of South Carolina and the Chief Administrative Judge for the Fifth Judicial Circuit (Richland and Kershaw Counties). A written, multiple-choice instrument was given to volunteers from a qualified circuit court jury pool in Richland County, South Carolina, by the Clerk of Court as they were discharged from jury duty. Residents of Richland County who are called for jury duty serve for an entire week. At the end of that week, our questionnaires were handed out to discharged jurors, whether they had participated in a jury trial or not. In addition to the questionnaire, jurors were provided a stamped, self-addressed envelope to return the survey. Richland County contains the state’s capital (Columbia) and has a population of approximately 325,000 persons. The high school graduation rate is 85 percent and the median income is $40,000.
The survey questioned jurors about the following demographic variables: sex, age, educational level, occupation, and the juror’s relationship to the law enforcement and legal communities (juror, friend, family member, or none). Jurors were also asked if they and/or anyone they knew “suffered from a severe mental illness such as Schizophrenia, Bipolar Disorder (Manic Depression), or Major Depression,” with the same choice of response (juror, friend, family member, or none).
The prospective jurors were asked to choose the correct definition of the NGRI and GBMI verdicts using the following choices: (1) the defendant has a severe mental illness and cannot stand trial; (2) the defendant was mentally ill at the time of the crime; (3) the defendant had a mental illness that prevented him or her from understanding that what he or she did was wrong (correct response for NGRI); and (4) the defendant had a mental illness that prevented him or her from controlling his or her actions according to the law (correct response for GBMI). The prospective jurors were asked to identify the dispositional outcome of the NGRI and GBMI verdicts and what the outcomes should be, using the following choices: (1) the defendant goes home; (2) the defendant goes to prison; (3) the defendant goes to a psychiatric hospital for treatment and is transferred to prison when stable enough to complete the sentence (correct response for GBMI); and (4) the defendant goes to a psychiatric hospital and is then released to go home when he or she is no longer a danger to him- or herself or others (correct response for NGRI). The prospective jurors were asked whether a defendant found guilty but mentally Ill could receive the death penalty.
Finally, the prospective jurors were asked if they thought that jurors should be instructed as to the dispositional outcomes of the NGRI and GBMI verdicts before deliberation. They were also asked if this knowledge would influence their verdict, even if the judge instructed them not to consider dispositional outcomes in reaching a verdict.
Results
Of the 200 surveys distributed to the qualified jury pool, 101 (50.5%) were returned. Five responders gave multiple answers to a single-answer question or failed to respond to one or more questions. Therefore, 96 of the 101 questionnaires were included in the analysis. The demographic variables are summarized in Table 1. The responding jurors were highly educated, with 97.9 percent having at least a high school education, and 54.2 percent having completed college and/or post-graduate education. Six percent of jurors admitted to having a mental illness, and 34 percent indicated that they knew someone with a major mental illness. The occupational profile of the responder group was diverse: 47 percent were nonprofessionals, 35 percent were professionals, 16 percent were unemployed, and 2 percent did not identify their occupations.
Juror knowledge is summarized in Table 2. Only 4.2 percent of all respondents identified the correct meanings and outcomes of both the NGRI and GBMI verdicts. When questioned as to the legal definition of the NGRI verdict, 55.3 percent of the prospective jurors answered correctly, while 24.5 percent chose the definition of the GBMI verdict. When questioned as to the legal definition of the GBMI verdict, 37.2 percent answered correctly, while 27.7 percent chose the definition of the NGRI verdict.
Regarding the disposition of an NGRI verdict, 62.5 percent of responders identified the correct disposition; however, 27.1 percent chose the GBMI disposition (hospital, then prison), and 10.4 percent believed the defendant would go home. Among jurors who chose the correct legal definition for NGRI, 71.2 percent also chose the correct disposition, while 15.4 percent chose the disposition for GBMI, and 13.5 percent answered that they believed the defendant would go home (Table 3).
When asked about the dispositional outcome of the GBMI verdict, 78.1 percent answered correctly, and 17.7 percent chose the disposition of those found NGRI. Among jurors who chose the correct legal definition of a GBMI verdict, 88.6 percent chose the correct outcome, while 11.4 percent chose the outcome for an NGRI verdict. The majority of respondents (68.4%) answered incorrectly, believing that a defendant found GBMI could not receive a death sentence (Table 4).
In assessing juror attitude about the disposition of NGRI acquittees, 56.8 percent of jurors who identified the correct legal definition of NGRI believed that defendants should go to a hospital and then home when they were no longer dangerous, while 39.2 percent thought that defendants should go to prison after being hospitalized. Jurors were more unified in their attitudes about the disposition of defendants found GBMI. Eighty percent of jurors who correctly identified the legal definition of GBMI believed that the defendant should go to a hospital and then to prison.
A large majority (84%) of respondents believed that jurors should be informed of the outcome of these verdicts before deliberation and 70.6 percent reported that knowing the outcome would influence their decisions, even if the judge instructed them not to consider the outcome in arriving at a verdict.
There was little statistical correlation among the demographic variables studied and knowledge of mental illness verdicts and their dispositional outcomes. The only statistically significant finding was that among those jurors with a post-high school education, those who did not complete a degree were less likely to identify the outcome of a GBMI verdict correctly (p < .005).
Finally, among jurors who knew the correct legal definition of NGRI, those with at least some college education were more likely to think that insanity acquittees should go to a hospital and then home while jurors with a high school education or less believed that insanity acquittees should go to a hospital and then prison (p < .026). This difference did not exist for the GBMI verdict.
Discussion
Among demographic variables, the educational level of the responders was well above the 85 percent high school graduation rate for this area. Less than 2.1 percent of the responders did not have a high school education, and 55 percent had a college degree. This probably reflects sampling bias. Because the survey involved a written instrument, those jurors who are illiterate may have been excluded. In addition, because this study was presented as a project of the University of South Carolina (located in Richland County), those associated with that institution may have been more likely to complete the study. Finally, because these surveys were distributed to jurors during the summer months, the jury pool may have been more educated, because professors, teachers, and students may defer jury duty to the summer months. Despite the high educational level of this sample, only 4.2 percent of the prospective jurors correctly identified both the legal definitions and dispositional outcomes of the NGRI and GBMI verdicts. This finding was not surprising to us, as these are difficult concepts and are not part of common, everyday knowledge or general education.
Six percent of the responders reported having a major mental illness (defined in the question as Schizophrenia, Bipolar Disorder, or Major Depressive Disorder). Thus, the point prevalence of major mental illness in this sample may be higher than in the general population and may also represent sampling bias, as mentally ill individuals may have been more interested in taking the survey.
The lack of significant correlation between demographic variables and knowledge base was consistent across age, sex, relationship to law enforcement, relationship to someone in the legal community, and relationship to someone with a major mental illness. The finding that, compared with jurors with a college or post-graduate degree, those who attended college but did not complete a degree were less likely to identify the outcome of the GBMI verdict correctly (p < .05) is probably true of jurors with a high school diploma or less. However, this finding was probably obscured in groups in the lower educational levels by the low response rate. Therefore, the difference did not reach statistical significance. Although knowledge about mental illness verdicts is poor across all educational levels, a higher educational level may predict better knowledge about dispositional outcomes of the NGRI and GBMI verdicts. Among jurors who correctly identified the meaning of NGRI, juror attitude about the disposition of mentally ill defendants appears to be influenced by educational level. The less educated jurors were significantly more punitive, believing that those found NGRI should eventually go to prison after hospitalization rather than home when no longer dangerous.
In the wake of the recent U.S. Supreme Court decision barring the execution of the mentally retarded,21 the finding that a majority of jurors did not believe that a defendant found GBMI could receive a death sentence may become an appellate issue. If jurors believe that GBMI individuals may not be executed, they may be more likely to decide a defendant is GBMI rather than NGRI in capital cases if they base their decision on perceived dispositional outcome. Because both the NGRI and GBMI verdicts require a unanimous jury finding, only one juror with such a misperception could be crucial in deciding the ultimate outcome in a capital case. The existence of GBMI inmates on death row may become the next battleground in the debate over capital punishment.
In this study, we found that a large majority (84%) of prospective jurors wanted to know the dispositional outcomes of mental illness verdicts and would consider these outcomes when reaching a verdict, even if instructed not to by the trial judge (70.6%). These results suggest that jurors may decide verdicts partly on what they believe will be the dispositional outcome of the verdict. Because only 4.2 percent of prospective jurors in this study correctly identified the meaning and disposition of both the NGRI and GBMI verdicts, it can be hypothesized that jurors may be making decisions based on erroneous perceptions. An argument can be made that jurors should be provided with accurate information regarding dispositional outcomes prior to jury deliberation.
- American Academy of Psychiatry and the Law