The Standard of Proof of Intellectual Disability in Georgia

Abstract

Despite being the first state to abolish the capital punishment of defendants with intellectual disability (ID), Georgia is currently the only state to uphold what is considered to be the most stringent standard of proof of ID in the United States: beyond a reasonable doubt. Other states have implemented less stringent standards of proof (i.e., a preponderance of the evidence and clear and convincing evidence). Although the U.S. Supreme Court ruled the execution of persons with ID unconstitutional in the 2002 Atkins decision, states are at this juncture considered to have a great deal of discretion in defining what constitutes ID. In addition to raising concerns about the Eighth Amendment’s protection against cruel and unusual punishment, variation between state-determined definitions of ID raises important questions of equal protection of the law. On January 27, 2015, Mr. Warren Lee Hill was executed by method of lethal injection in the state of Georgia after the U.S. Supreme Court denied Mr. Hill’s claim to have met the state’s legal definition of ID beyond a reasonable doubt. This article provides a historical and legislative background for the case of Warren Lee Hill, while examining the definitions of ID (including adaptive functioning) in the legal and clinical arenas. Last, this article will take a critical stance with regard to the current diagnostic criteria being used in the state of Georgia.

Keywords

intellectual disability adaptive functioning beyond a reasonable doubt a preponderance of the evidence Warren Lee Hill

In 2014, Blume, Johnson, Marcus, and Paavola reported that, from 2002 to 2013, approximately 7.7% (n = 371) of death row inmates and capital defendants raised claims of intellectual disability (ID), and an estimated 55% of such claims were substantiated by the courts. Furthermore, they found that in states that significantly deviated from accepted clinical methods for determining ID (formerly termed Mental Retardation [MR]), death row inmates and capital defendants raising claims of ID had the lowest success rates. Wide variance within success rates was shown even within the same region of the United States: North Carolina = 82%, Mississippi = 57%, Georgia = 11%, and Florida = 0% (Blume, Johnson, Marcus, & Paavola, 2014). Table 1, modeled on Blume et al.’s findings, showcases the more extreme variations between states in terms of the number and percentage of successful claims of ID raised by capital defendants as of 2013. The issue of between-state variance raises troubling implications about the role of arbitrary features such as geographic location in the determination of justice—and for Warren Lee Hill of Georgia, in the determination of his life and death.

Table 1.

Variation Between States in Successful ID Claims as of 2013.

State	Total number of ID claims as of 2013	Number of successful claims of ID as of 2013	Number of unsuccessful claims of ID as of 2013	% successful claims of ID as of 2013
Alabama	34	5	29	14.7%
Georgia	9	1	8	11.1%
Kentucky	9	1	8	11.1%
North Carolina	34	28	6	82.4%
South Carolina	6	5	1	83.3%
Tennessee	8	0	8	0%
Texas	45	8	37	17.8%
Virginia	7	0	7	0%

Note. ID = intellectual disability.

Despite being the only state with the strict beyond a reasonable doubt standard of proof of ID, Georgia was the first state to ban the executions of individuals with ID in the 1980s (Lohr, 2013). The standard of proof, beyond a reasonable doubt, is contrasted with the intermediate standard, clear and convincing evidence, and the lowest standard, a preponderance of the evidence. Although the majority of states with death sentencing require proof to a preponderance of the evidence, several states still require clear and convincing evidence—and one state, Georgia, requires proof beyond a reasonable doubt (Saviello, 2014). Richard Dieter, Executive Director of the Death Penalty Information Center, stated that Georgia’s standard of proof is the most rigorous in the United States, and the question at hand is whether Georgia has defined its law too narrowly to comply with the Atkins v. Virginia (2002) decision (Lohr, 2013).

In the Atkins v. Virginia (2002) decision, the U.S. Supreme Court simultaneously barred the execution of persons with ID while also allowing for state-level discretion in the definition of ID. Because the Court in Atkins v. Virginia gave license to establish procedures at the state level, standards of proof of ID vary greatly between states (DeMatteo, Marczyk, & Pich, 2007; Saviello, 2014). One such option available to states is the level of the standard of proof a capital defendant raising a claim of ID must meet (Saviello, 2014). Thus, in the state of Georgia, the lawful execution of Warren Lee Hill hinged not upon his criminal offense but upon his point of legal defense (i.e., whether or not Mr. Hill could demonstrate beyond a reasonable doubt that he had an ID according to the state’s narrow legal definition).

Purpose

This article provides a summative historical and legislative context for the case of Warren Lee Hill, and critically examines the constitutionality of the beyond a reasonable doubt standard of proof of ID. The aims of this article are threefold. The first will be to raise relevant questions concerning the differences between states’ legal definitions of ID versus clinical criteria put forward by the American Psychiatric Association (APA) in the Diagnostic and Statistical Manual of Mental Disorders (DSM). ID is at this time the preferred nomenclature and is used in place of the former term, mental retardation. The second aim will be to question whether a single definition of ID can or should be used across legal and clinical arenas. The final aim will be to critically examine the constitutionality of Georgia’s beyond a reasonable doubt standard of proof.

To address the above-mentioned aims, we first provided a background of Warren Lee Hill’s case. Next, in the second section of the article, we examined the constitutionality of the beyond a reasonable doubt standard of ID. The third section explores the evolving and disparate definitions of ID (e.g., Georgia’s legal definition of ID, as well as those put forward by the APA). The fourth section raises issues of uncertainty in measurement as related to the finding of ID. The fifth section consists of a general discussion, and the final section includes recommendations so that capital defendants in Georgia can successfully raise claims of mild to moderate ID.

The Case of Warren Lee Hill

As will later be discussed in greater detail, 10 years following their initial examinations of Mr. Warren Lee Hill, state medical experts recanted their original testimony, determining that Mr. Hill did indeed demonstrate cognitive disorder with adaptive skill deficits since early childhood, with additional evidence that he consistently tested in the 2–3 percentile in childhood achievement and intelligence testing (Sachy Aff. 17). State experts also found evidence of a history of trauma (Sachy Aff. 10) along with significant deficits in functioning (e.g., self-care, functional academics, interpersonal skills, and home living prior to age 18; Sachy Aff. 17). Following a period of psychological decompensation while serving in the U.S. Navy, Mr. Hill was sent home to live with relatives and in fact never lived alone (Sachy Aff. 15). It was after this period, at age 25, that Warren Lee Hill was convicted of the murder of his then 18-year-old girlfriend, Myra Wright, whom he shot 11 times in 1986 (Blanco, n.d.; Connor, 2015; Lohr, 2013; West, 2013). Mr. Hill was subsequently sentenced to life imprisonment for the murder of Ms. Wright. While serving his life sentence, Warren Lee Hill was, in 1991, charged with beating to death a fellow Lee County Correctional Institute inmate, Joseph Handspike, in 1990 with a two by six nail-embedded wooden board as Mr. Handspike slept (Renaud, 2003). In the wake of his 1991 sentencing, Mr. Hill’s defense attorneys did not ask the jury for a verdict of not guilty but instead for the finding that Mr. Hill indeed met state legal criteria for having an ID (Renaud, 2003). This would have effectively rendered his execution unconstitutional (Brumback, 2013; Connor, 2015; Lohr, 2013; Renaud, 2003; Saviello, 2014).

Overview of Judicial Procedures

Throughout this process, Mr. Hill, under advisement from his legal defense, filed four counseled state habeas cases and three counseled federal habeas cases (Hill v. U.S. Court of Appeals, 2015). Habeas corpus is a protection against unconstitutional confinement, such as holding a person without charges, or when due process obviously has been denied (Habeas Corpus, n.d.). It is a court order that directs the law enforcement officials (i.e., prison administrators, police, or sheriff) who have custody of a prisoner to appear in court with the prisoner to help the judge determine whether the prisoner is lawfully in prison or jail (Habeas Corpus, n.d.). Such a writ is obtained by petition to a judge in the county or district where the prisoner is incarcerated, and the judge sets a hearing on whether there is a legal basis for holding the prisoner (Habeas Corpus, n.d.). In total, Mr. Hill’s execution was stayed 3 times (Connor, 2015): (a) July 2012, (b) February 2013, and (c) July 2013 (Davis, 2013). Only one stay was related to ID (Brumback, 2013): The 11th U.S. Circuit Court of Appeals halted the February 2013 execution less than 1 hr before the scheduled time—after prison officials had already administered one of the injectable drugs to Mr. Hill (Lohr, 2013). The Court granted a last-minute stay to weigh the claim of ID and to determine whether Mr. Hill could file a second petition for a writ of habeas corpus (Connor, 2015). In April 2013, however, the 11th Circuit, in a 2-1 decision (West, 2013), denied Mr. Hill’s application to file a second writ and vacated the stay of execution (Davis, 2013). On January 27, 2015 (the fourth and final scheduled execution), Mr. Hill was put to death by method of lethal injection and pronounced dead at 7:55 p.m. in the Jackson, Georgia State Prison (Connor, 2015). He declined to issue a final statement and to request a last meal (Connor, 2015). No bill was introduced to the Georgia General Assembly during the 2015 legislative session that would alter the standard of proof of ID in the state (Jones, 2015).

Beyond a Reasonable Doubt and the U.S. Constitution

Defining the Standards of Proof of ID

Beyond a reasonable doubt is the highest of three legal standards of proof (Reasonable Doubt, n.d.), requiring sufficient doubt on the part of the jurors for acquittal of a defendant based on a lack of evidence (Reasonable Doubt, n.d.). It stands in contrast to the intermediate standard of proof, clear and convincing evidence, and the lowest standard of proof, a preponderance of the evidence. Beyond a reasonable doubt is said to be met when jury members cannot say with moral certainty that the defendant is guilty (Burden of Proof, 2002). This high standard of proof is used in criminal trials because such trials can result in the deprivation of a defendant’s liberty or in the defendant’s death (Reasonable Doubt, n.d.). Beyond a reasonable doubt is protected under the due process clause of the Fifth Amendment (Reasonable Doubt, n.d.), as per the 1970 ruling of the U.S. Supreme Court in the case of In Re Winship, 397 U.S. 358 (Burden of Proof, 2002). According to Justice John M. Harlan, the reasonable doubt rule is intended to decrease the likelihood of wrongfully convicting an innocent defendant and signifies: “A fundamental value determination of our society that it is far worse to convict an innocent man than to let a guilty man go free” (Winship, 1970). It would seem, then, that the beyond a reasonable doubt requirement is intended to weigh the scales of justice in favor of the defendant, not against. Georgia’s application of this standard in capital cases involving defendants with ID seems incongruent with protections afforded under the Eighth Amendment of the U.S. Constitution and accordingly the Atkins v. Virginia (2002) decision. Table 2 differentiates between the three standards of proof according to level of burden to the defendant, definition of the standard and traditional applications.

Table 2.

Standards of Proof of Intellectual Disability.

Name of standard	Burden of standard	Definition of standard	Traditional application of standard
Beyond a reasonable doubt	Highest	Requires sufficient doubt on the part of the jurors for acquittal of a defendant based on a lack of evidence	Criminal trials
Clear and convincing evidence	Intermediate	Substantially more likely than not to be true	Civil and criminal trials
A preponderance of the evidence	Lowest	A position must have more evidence in its favor than not, even if by the narrowest margin	Civil trials

Cruel and Unusual Punishment

Atkins and the Eighth Amendment

The Eighth Amendment is “most often mentioned in the context of the death penalty” and prohibits cruel and unusual punishments (Eighth Amendment, n.d.). In the case of Atkins v. Virginia (2002), the U.S. Supreme Court banned the execution of individuals with ID, citing the prohibition of cruel and unusual punishment under the Eighth Amendment (Atkins v. Virginia, 2002; Saviello, 2014). In 2003, Georgia State Supreme Court Justice Carley overturned Muscogee County Superior Court Judge Allen’s 2002 finding that the Atkins v. Virginia prohibition on executing defendants with ID could not be enforced properly under Georgia’s stringent standard of proof (Renaud, 2003). Justice Carley ruled instead that nothing in Atkins v. Virginia dictates that states must apply any particular standard of proof (Head v. Hill, 2003). Justice Carley concluded that any risk of wrongful execution is “sufficiently counterbalanced by the joint safeguards of Georgia’s procedure for demonstrating incompetency to stand trial under the preponderance of the evidence standard and mental retardation [sic] under the beyond a reasonable doubt standard” (Head v. Hill, 2003). Three Georgia Supreme Court Justices, however, disagreed with the 2003 ruling upholding the standard of beyond a reasonable doubt, and warned that the majority’s decision to require the high standard of proof is likely to result in the execution of defendants with ID in the state of Georgia (Brumback, 2013). Chief Justice Norman S. Fletcher and Justice Robert Benham allied with presiding Justice Leah W. Sears’s dissenting opinion.

Justice Sears contended that the case of Warren Lee Hill represented more than just what procedures states should use to prove their particular standard of proof of ID (Renaud, 2003). Instead, she found to be at stake the constitutional guarantee of procedural due process and the right of defendants with ID to be exempt from execution (Renaud, 2003). She wrote that the U.S. Constitution could not

simultaneously limit the state to the preponderance of the evidence standard when seeking to prosecute a capital defendant who claims incompetence, yet allow the state to impose the beyond a reasonable doubt standard when that same defendant, after being convicted and sentenced to death, claims mental retardation [sic]. (Head v. Hill, 2003)

According to Justice Sears, individuals with ID are presumed to have diminished capacities that would render those individuals vulnerable to a significant risk of being wrongfully executed (Renaud, 2003). She further noted that persons with ID are more prone to make false confessions, less able to make showings of mitigating factors, are often unable to help their attorneys, and are frequently poor witnesses (Renaud, 2003)—findings that are supported in the scholarly literature (Blume et al., 2014). Similarly, in the Atkins v. Virginia (2002) decision, the U.S. Supreme Court reasoned that offenders with ID had significant impairments in their abilities to (a) process information, (b) logically reason, (c) control their impulses, and (d) learn from experience (DeMatteo et al., 2007). These factors render individuals with ID less morally culpable and more at risk for wrongful conviction (DeMatteo et al., 2007). In Georgia, the only state to uphold the strictest standard of proof (i.e., beyond a reasonable doubt), this also effectively renders capital defendants with ID more at risk for wrongful execution.

States’ Rights Versus Individual Civil Rights

Rather than adopting a uniform definition of ID, the U.S. Supreme Court charged each state with defining ID in a manner congruent with the constitutional ban (DeMatteo et al., 2007). And, according to the findings of DeMatteo et al. (2007) and Blume et al. (2014), definitions of ID vary greatly by state and may not be consistent with accepted clinical standards. As such, it is not clear whether the majority of states are effectuating the intent of Atkins.

The Fourteenth Amendment

Precipitating the second stay of execution in February 2013, Mr. Hill and his attorneys applied for a second writ of habeas corpus arguing that the execution would violate the Eighth and Fourteenth Amendments to the U.S. Constitution (Hill v. U.S. Court of Appeals, 2013). Section 1 of the Fourteenth Amendment formally defines U.S. citizenship and protects various civil rights from being abridged or denied by any state. The Fourteenth Amendment exemplifies tension between states’ rights and uniform protection of civil rights of the individual (i.e., at what level the federal jurisdiction should intervene in state issues of civil injustice). This tension was palpable during the Civil Rights Cases of 1883, in which the U.S. Supreme Court declared the Civil Rights Act of 1875 unconstitutional, finding instead that Congress was unauthorized to bar racial discrimination by private individuals or organizations (Civil Rights Cases, 2008). This led to the enactment of state laws that effectively codified racial segregation in the South (Civil Rights Cases, 2008).

Antiterrorism and Effective Death Penalty Act (AEDPA) of 1996

According to the Cornell University Legal Information Institute, the AEDPA of 1996 was a major mid-1990s reform of habeas corpus as used to challenge criminal convictions limiting both the procedural and substantive scope of the writ. Procedurally, it bans successive petitions by the same person, requiring defendants to put all of their claims into one appeal; substantively, it narrows the grounds on which successful habeas claims can be made, allowing claims only to succeed when the convictions were contrary to “clearly established federal law” or an “unreasonable determination of the facts in light of the evidence” (AEDPA of 1996). The Act (AEDPA of 1996) essentially restricts the power of federal judges to grant habeas corpus relief at the state level. Advocating that Mr. Hill’s was a case that warranted federal intervention, Judge Martin of the 11th Circuit Court wrote, “I cannot agree that we have no choice but to execute [Hill] anyway because his claim does not fit neatly into the narrow procedural confines delimited by AEDPA” (Hill v. U.S. Court of Appeals, 2015, p. 20).

Hall v. Florida

In Hall v. Florida (2014), the Supreme Court struck down Florida’s strict IQ cutoff of 70 as violating the Eighth Amendment’s prohibition on cruel and unusual punishment on the ground that the rule “misuse[d] IQ score on its own terms” in a way that risked the execution of those with ID (Hill v. U.S. Court of Appeals, 2015, p. 10). Specifically, the Supreme Court in Hall v. Florida noted evidence of accepted medical practice that (a) an IQ test result taken as sole evidence of ID was “not conclusive evidence of intellectual capacity” and (b) “each IQ test had a standard error of measurement” (Hill v. U.S. Court of Appeals, 2015, p. 10). Leading up to this decision, the Florida Supreme Court had interpreted a Florida statute to mean that a prisoner sentenced to death was required to show an IQ test score of 70 or below before presenting any additional evidence of ID. According to Taylor and Krauss (2014), Florida’s procedure to limit Atkins v. Virginia’s (2002) evaluations to a bright line 70-point IQ threshold appears contrary to standard clinical practice and norms.

Evolving and Disparate Definitions

Definitions and classifications that result in categorizing people for purposes of accessing services or participating in government are important for efficient and fair operation of a society or country. The precision with which these definitions and classifications are designed sometimes fails to validly afford the guidance they were designed to provide. Furthermore, ambiguity around the margins and lack of agreement among professionals in fields that study characteristics of those groups results in suboptimal definitions. This can result in multiple definitions put forth by groups with competing interests, raising the question as to whether different agencies of the state should rely on different definitions (e.g., one definition to access educational or social services and another to guide legal proceedings).

Georgia’s Legal Definition

The state habeas court employed the definition of ID in O.C.G.A. § 17-7-131(a)(3), which provides that “‘mentally retarded’” [sic] (now intellectual disability) means (a) having “significantly sub-average general intellectual functioning,” (b) “resulting in or associated with impairments in adaptive behavior,” (c) “which manifested during the developmental period” (Proceedings upon plea of insanity or mental incompetency at time of crime, 2014). Qualified examiners include court-appointed licensed psychologists or psychiatrists, or physicians or licensed clinical psychologists chosen and paid for by the defendant (Proceedings upon plea of insanity or mental incompetency at time of crime, 2014).

The APA

In 2000, at the time Mr. Warren Lee Hill was clinically assessed by three state doctors (i.e., Thomas Sachy, MD, MSc; Gary Carter, MD; Donald Harris, PhD in clinical psychology), the current edition of the DSM was the revised fourth edition (Diagnostic and Statistical Manual of Mental Disorders [4th ed.; DSM-IV]; APA, 1994; Hill v. U.S. Court of Appeals, 2013). At this time, each of the three doctors determined that Mr. Hill did not have an ID according to the clinical standards set forward by the DSM (Lohr, 2013). These criteria include measures of intellectual functioning (i.e., IQ), as well as adaptive functioning. Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.; DSM-IV-TR; APA, 2000) defines adaptive functioning as the following:

How effectively individuals cope with common life demands and how well they meet the standards of personal independence expected of someone in their particular age group, sociocultural background, and community setting . . . (p. 42)

It is possible to diagnose Mental Retardation [sic] in individuals with IQs between 70 and 75 who exhibit significant deficits in adaptive behavior. Conversely, mental retardation [sic] would not be diagnosed in an individual with an IQ lower than 70 if there are no significant deficits or impairments in adaptive functioning. (pp. 41–42)

As discussed later, before recanting their original testimony, the assessing clinicians initially determined that Mr. Hill did not meet criteria for ID based on his involvement with the U.S. Military. They reasoned that Mr. Hill’s work with the military precluded any significant deficits in adaptive functioning that would be required to substantiate his claim of ID. In this vein, Blume et al. (2014) have suggested that many courts rely on stereotypes when determining the presence of substantially impaired adaptive functioning. Frequently, if individuals demonstrate any of the following, they are considered not to meet criteria of the adaptive functioning prong: the ability to read, write, perform basic math, maintain social relationships, drive, or perform daily personal care (Blume et al., 2014). We will later discuss the construct, adaptive functioning, in greater detail.

The Uncertainty of Measurement

The information used in the diagnostic process is inherently uncertain and diagnosis involves value-laden, subjective analysis and interpretation (Saviello, 2014). The U.S. Supreme Court recognized the impact of the imprecision of psychiatric diagnosis on the standard of proof for competency to stand trial in Cooper v. Oklahoma (1996), finding only a standard of preponderance of the evidence to satisfactorily offset that imprecision (Saviello, 2014). In Cooper v. Oklahoma, the Supreme Court ruled that the preponderance of the evidence standard was the most appropriate standard for proving a defendant’s incompetence to stand trial, and the only standard that adequately safeguarded against an individual with mental illness going to trial while experiencing an acute mental health episode (Mobbs & West, 2013). In a unanimous decision, the Court ruled that the higher standard of proof, clear and convincing evidence, stood in violation of due process requirements by placing the burden of proof on the defendant (Ellis, 2003).

The Uncertainty of Medical Expert Testimony

In the early 2000s, Thomas Sachy, MD, MSc; Gary Carter, MD; and Donald Harris, PhD, were hired by the state of Georgia to examine Warren Lee Hill (Doctors Who Evaluated Death Row Prisoner for Georgia, 2013). All concluded that he did not meet criteria for ID (Lohr, 2013). However, after reviewing the evidence again in 2013, all three doctors have recanted their original decisions (Lohr, 2013). According to Lohr (2013), the three doctors claim in affidavits that their initial evaluations were extremely and unusually rushed. One doctor further claimed that what would have been a close case back then would be viewed differently today in light of changing opinions about the abilities of persons with ID—and that he now believes Mr. Hill met criteria for having had a mild ID and should have been eligible to have his sentence commuted to life in prison (Lohr, 2013; West, 2013). With regard to the expert medical testimony, Judge J. Martin of the 11th Circuit Court of Appeals wrote,:

No one disputes that every mental health expert who has ever evaluated Mr. Hill, now including the state’s three experts, has concluded he is intellectually disabled [sic]. The fact that there is now unanimity among these experts makes it all the more striking to recall that even before this unanimity existed, the state habeas court made a finding that Mr. Hill had proved, beyond a reasonable doubt, that his IQ showed he had significantly subaverage intellectual functioning. (Hill v. U.S. Court of Appeals, 2015, pp. 32–33)

Dr. Sachy was one of the three doctors hired by the state of Georgia to examine Warren Lee Hill in 2000 based on Mr. Hill’s prior claim to have an ID (Lohr, 2013). At that time, Dr. Sachy concluded that Mr. Hill did not meet criteria for ID (Lohr, 2013). However, after reviewing the evidence a second time, in 2013, Dr. Sachy filed a legal affidavit (i.e., a written statement of facts voluntarily made by an affiant under an oath; (Affidavit, 2005). In his affidavit, Dr. Sachy recanted his previous decision, stating instead that he now believes Mr. Hill did meet criteria for a mild ID and should have been eligible to have his sentence commuted to life in prison (Lohr, 2013; West, 2013). Dr. Sachy (2013) made the following statements throughout his affidavit:

I did not have experience evaluating patients for mental retardation [sic]. I had almost no experience testifying in a forensic context . . . This was my first experience working on a capital case (Sachy Aff. 3). The whole process, including my evaluation of Mr. Hill, was rushed . . . (Sachy Aff. 4). I felt that my previous conclusions about Mr. Hill’s mental health status were unreliable because of my lack of experience at the time . . . (Sachy Aff. 5). In the psychiatric community, we now know that reliance on the DSM criteria has resulted in an extremely high rate of false findings of malingering . . . I do not believe now that Mr. Hill was deliberately feigning a cognitive disorder in 2000, and I believe that his responses to my questions were consistent with mild mental retardation [sic] (Sachy Aff. 9). Finally, the totality of evidence shows that far from “malingering a cognitive disorder,” Mr. Hill has had a cognitive disorder with adaptive skill deficits since early childhood. He consistently tested in the 2-3 percentile in childhood achievement and intelligence testing, consistent with mild mental retardation [sic] . . . There is also evidence of significant deficits in such areas of his functioning as self-care, functional academics, interpersonal skills, and home living since prior to age 18. (Sachy Aff. 17)

The Uncertainty of Intelligence Testing

Measures of intelligence raise important logical, empirical, and ethical issues related to the determination of ID. When a capital defendant makes a claim of ID in a state with death sentencing (as did Warren Lee Hill), the decision of life and death may hinge not only on the validity of the intelligence test scores themselves but as well on the underpinning philosophical justifications for the use of that test in the first place.

Charles Murray coined the term, “Flynn effect” in The Bell Curve: Intelligence and Class Structure in American Life (Herrnstein & Murray, 1996) to describe James Flynn’s empirical finding that populations improve over time in performance on intelligence tests, suggesting the need for “the periodic re-standardization” of such assessment tools (Silverman et al., 2010, p. 242). Despite this well-documented effect—and other logical and philosophical issues—the validity of IQ tests (when properly administered) has been and continues to be broadly accepted (Silverman et al., 2010). Historically and currently, the two instruments most widely used in the United States to assess intelligence have been the Stanford–Binet Scale and Wechsler Adult Intelligence Scale (WAIS; Silverman et al., 2010).

In practice, both scales are often used to assess intelligence in children and young adolescents (Silverman et al., 2010). However, because earlier versions of the Stanford–Binet were not normed for older ages, the various editions of the WAIS became the dominant IQ assessment for adults (Silverman et al., 2010). In one study, Silverman et al. (2010) compared Stanford–Binet and WAIS scores for a group of 74 adults with ID:

In every case, WAIS Full Scale IQ was higher than the Stanford-Binet Composite IQ, with a mean difference of 16.7 points . . . Additional comparisons with other measures suggested that the WAIS might systematically underestimate severity of intellectual impairment . . . The present data for individuals with relatively higher IQs, though sparse, indicate that differences between the Stanford-Binet and WAIS IQ tests can no longer be summarily dismissed as merely reflecting the scales’ different floors. When test results are informing judgments of literal life and death, any suspected uncertainty regarding the validity of outcomes must be addressed aggressively. (Silverman et al., 2010, pp. 242, 247)

The concern for test score validity has been raised for adults with ID throughout the past 65 years: Bensberg and Sloan (1950) found that WAIS IQs of a group of adults with ID were higher than their Stanford–Binet IQs by between 7 and 20 points, with the magnitude of difference increasing with age at test administration; Brengelmann and Kenny (1961) reported an advantage of 8 points; Spitz (1989) reported an advantage of 12.4 points; and Nelson and Dacey (1999) an advantage of 14.6 points (Silverman et al., 2010). This seems to be recognized by even the Stanford–Binet test officials, as the Fourth Edition Technical Manual describes a 9.3-point advantage for the WAIS for adults with ID (Silverman et al., 2010). Although professionals recognize that test imprecision can allow people with “true” IQs just under the 70-point threshold to score higher by chance, it is clear, according to Silverman et al. (2010), that any IQ result in the 60s and above imposes an increased burden of proof on the individual in question, especially for scores between 70 and 75 points. Silverman et al. (2010) concluded with the precaution, “if two gold-standard procedures consistently produce divergent results, then the validity of test-based inferences should be questioned” (p. 243).

Stanford–Binet

In 1904, the minister of public education commissioned French psychologist, Alfred Binet to develop techniques for identifying children whose lack of success in normal classrooms suggested the need for some form of specialized education (Gould, 1996). Binet “rejected the hereditarian interpretation of intelligence,” wanting only to use his test as a device to identify children who were in need of intervention, and expressed the concern that his test could be used to apply ineffaceable labels rather than as a practical assessment tool (Gould, 1996, p. 40). However, American psychologists later reified Binet’s scores, and “took them as measures of an entity called intelligence” (Gould, 1996, p. 187).

Wechsler

The WAIS manual interprets scores in terms of confidence intervals around point scores to account for the standard error of measurement (SEM) of approximately 2.5 points, or 5 points for the suggested 95% confidence interval (Taylor & Krauss, 2014). In the range around 70, the SEM may be substantially larger scores due to the smaller number of people in that range of the standardization sample (Taylor & Krauss, 2014). One study reported an SEM of 7.65 in low-IQ scoring persons, suggesting that a 95% confidence interval around a true score of 70 may test as 85 (Whitaker, 2010), calling into question the accuracy of ID/non-ID threshold.

The Uncertainty of Adaptive Functioning

Based on the evidence presented by Mr. Hill and the State, the state habeas court found that Mr. Hill did not have impairments in the adaptive functioning areas of “communication, self-care, home living, social/interpersonal skills, use of community resources, self direction, functional academic skills, work, leisure, health, and safety” (Hill v. U.S. Court of Appeals, 2015, p. 4)—Criterion B of the DSM-IV-TR (APA, 2000, p. 41). Specifically, court documents cited Mr. Hill’s military record, which allegedly included five promotions, his successful courses in military education, leadership management and instruction training, and his purportedly excellent oral and written English language skills (Hill v. U.S. Court of Appeals, 2015; Lohr, 2013):

The court noted Hill’s (a), extensive work history and apparent ability to function well in such employment; (b), disciplined savings plans pursued to purchase cars and motorcycles; (c), military service; (d), active social life; (e) writing skills; and (f) ability to care for himself. (Hill v. U.S. Court of Appeals, 2015, p.4)

Accordingly, it was found that Mr. Hill did not meet the legal criteria for having an ID (Connor, 2015).

Thomas Sachy affidavit

Conversely, Thomas Sachy, neuropsychiatrist and 2000 state medical expert found that Warren Lee Hill did demonstrate “significant deficits in adaptive functioning” upon his later and further review of the evidence (Sachy Aff. 18). He wrote in his 2013 affidavit:

I simply found it difficult to believe that someone with mental retardation [sic] could function even minimally in the United Sates Navy as a petty office, although I have never served in the military (Sachy Aff. 7) . . . What the [2000] testimony of [Navy psychologist Dr. Jerry Brittain, Army psychiatrist Dr. Donald Stonefeld, and original trial attorneys William Erwin and Al Grieshaber, also a former military service member] helps me to understand now is that a mildly mentally retarded [sic] individual like Mr. Hill could have functioned adequately at least for a time in the structured setting of the Navy at the rank of E5 or petty officer . . . As Dr. Brittain describes, Mr. Hill began to decompensate once he transferred from the Naval Air Station in Boston to NAS Atlanta, where he was given additional responsibilities which he did not have the coping skills to manage . . . There came a time when Mr. Hill decompensated and had to be brought home to live with relatives. Mr. Hill never lived alone (Sachy Aff. 15) . . . Although Mr. Hill was recommended for advancement to the rank of E6, he was passed over for promotion. Thus, there appears to be recognition, consistent with the affidavits of Mr. Hill’s Navy associates, that Mr. Hill was not able to maintain a level of functioning past a certain point. (Sachy Aff. 16)

In addition, in his affidavit, Dr. Sachy references “past documented trauma in Mr. Hill’s background” (Sachy Aff. 10) and that, contrary to the purportedly excellent language skills of Mr. Hill, “a memo by a defense social worker . . . indicates that even with defense team members prior to his trial, Mr. Hill showed ‘poor verbal skills and [found] it difficult, almost impossible to express thoughts and feelings’” (Sachy Aff. 12).

Discussion

Georgia appears poised to maintain the beyond a reasonable doubt standard of proof, and despite the controversy that surrounded Warren Lee Hill’s execution, no bill was introduced to the Georgia General Assembly during the 2015 legislative session moderating the standard of proof (Jones, 2015). Between-state variations in adjudication procedures and standards of proof invite charges of capricious application of capital punishment. Furthermore, when capital defendants raise the claim of ID, those in the legal system begin to practice outside their area of expertise, ruling on the validity of a clinical diagnosis (the providence of a licensed psychologist) rather than on issues of law. This is tantamount to a psychologist providing legal advice to a client. Simply, if a standard of proof for diagnosis requires something other than what a qualified expert in that field can provide, Georgia is asking for something that is not only impossible but outside the bounds of what lawyers and judges are qualified to do. A change in Georgia’s standard of proof would bring Georgia more in line with other states and would help the judiciary to safeguard against the uncertainty and ambiguity involved in reaching a clinical diagnosis of ID. In issues where qualified professionals dispute diagnosis, a more moderate standard would permit the court to err on the side of not executing someone that at least some qualified professionals deem to have an ID.

Conclusions Regarding Judicial Interpretation and Procedure

Misapplication of beyond a reasonable doubt

The U.S. Supreme Court recognized the impact of the inherent imprecision of psychiatric diagnosis on the standard of proof for competency to stand trial in Cooper v. Oklahoma (1996), finding that only a standard of a preponderance of the evidence to satisfactorily offset that imprecision (Saviello, 2014). As Timothy Saviello (2014) wrote, the Cooper v. Oklahoma decision suggests that, to weigh the appropriate standard of proof for capital defendants raising claims of ID, the U.S. Supreme Court should take the same approach as it did in Cooper v. Oklahoma and reach the same conclusion that a preponderance of the evidence is the only standard which complies with Atkins v. Virginia (2002).

It stands to reason that if the intermediate standard of proof, clear and convincing evidence, violates due process protections, then the more stringent standard of proof, beyond a reasonable doubt, is guilty of the same offense and to a greater degree. As Justice John M. Harlan interpreted the U.S. Constitution, the standard of beyond a reasonable doubt is intended to offset the chances of wrongfully convicting an innocent defendant, and signifies an essential value determination that it is ethically far worse to convict an innocent person than it is to allow a guilty person to go free (Winship, 1970). It is an interpretation that favors the protection of legal innocence over the exaction of retribution. If the beyond a reasonable doubt requirement is intended to weigh the scales of justice in favor of a criminal defendant—and not against—then it seems that (a) the beyond a reasonable doubt standard is being applied in a manner antithetical with the intended purpose, and (b) Georgia’s application of the standard in capital cases involving defendants with ID is misaligned with protections afforded under the Eighth Amendment of the U.S. Constitution and accordingly the Atkins v. Virginia (2002) decision.

AEDPA and the Fourteenth Amendment

The AEDPA of 1996 and the Civil Rights Cases of 1883 exemplify the historic caution employed by the federal jurisdiction when determining the appropriateness of state-level intervention. This raises issues concerning “equal protection of the law” afforded under Section 1 of the Fourteenth Amendment to the U.S. Constitution. The point of contention rests upon the AEDPA (1996) stipulations that habeas claims may go forward only if they are contrary to “clearly established federal law” or an “unreasonable determination of the facts in light of the evidence.” Because, as Justice Sears found, the U.S. Supreme Court simultaneously limits the state to the preponderance of the evidence standard when seeking to prosecute a capital defendant who claims incompetence, yet allows the state to impose the beyond a reasonable doubt standard when that same defendant, after being convicted and sentenced to death, claims ID (Head v. Hill, 2003), there is an incongruence that undermines “clearly established federal law” (AEDPA, 1996). Considering that Mr. Hill’s execution was stayed 3 times, it seems that the U.S. Supreme Court would have been justified in state-level intervention.

Conclusions Regarding Georgia’s Standard of Proof

Qualified witnesses

Under Georgia law, qualified examiners include court-appointed licensed psychologists or psychiatrists, or physicians or licensed clinical psychologists chosen and paid for by the defendant (Proceedings upon plea of insanity or mental incompetency at time of crime, 2014). This point of consideration merits further attention, as it is questionable as to how “qualified” examiners are determined. Dr. Thomas Sachy readily admits in his 2013 affidavit that he lacked experience evaluating individuals for ID, had “almost no experience testifying in a forensic context,” and that Warren Lee Hill was his first capital case (Sachy Aff. 3). Yet, at least nominally, Dr. Sachy met the state’s criteria.

Conclusions Regarding Widely Held Clinical Definitions

Adaptive functioning

The legal finding of ID rests upon three prongs: (a) significantly subaverage intellectual functioning, (b) deficits in adaptive functioning, and (c) onset during the developmental period (Proceedings upon plea of insanity or mental incompetency at time of crime, 2014). This combination of criteria is consistent with that used by the APA in the Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-5; APA, 2013). As was earlier argued, intelligence measures (e.g., the Stanford–Binet and WAIS IQ tests) are inherently uncertain, imbued with individual and cultural biases, and not intended to be used as a stand-alone criterion. Instead, it is the second prong of ID, adaptive functioning, that is meant to corroborate IQ scores. What we hope to raise here is the concern that adaptive functioning is a paradoxical benchmark that has resulted in unintended and adverse consequences.

Recommendations

To address the concern that between-state variations in adjudication procedures and substantive definitions of ID invite charges of capricious application of capital punishment, we offer the following recommendations: (a) at a minimum, the federal government should establish greater substantive and procedural regularity across all states (i.e., a national model of standards); (b) as a part of this regularity, the standard of proof should be set at a preponderance of the evidence, state legal definitions should adhere to those accepted in the clinical community, expert witnesses should be clinically credentialed and experienced specifically with ID—and the interpretation of standardized test scores—as well as with capital cases, and the age of onset prong seems tangential to the question of functionality and should be removed as a legal criterion; and (c) the assessment of adaptive functioning should be met with the same skepticism as intellectual functioning. However, if imprecision in diagnosis precludes in any way the constitutional protection of persons with ID against capital punishment as per the ruling of the U.S. Supreme Court in the 2002 Atkins decision, this may be grounds for a national moratorium.

Footnotes

Acknowledgements

I wish to disclaim at this time that this article is being completed as part of a doctoral training program and as such, I would like to request permission to include it as part of my dissertation.

Authors’ Note

The article represents the original work of the two authors, Lauren A. Ricciardelli and Kevin M. Ayres.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: All work was funded by the authors.

References

Affidavit. (2005). West’s Encyclopedia of American Law. Retrieved from http://www.encyclopedia.com/doc/1G2-3437700183.html

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author.

American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., text rev.). Washington, DC: Author.

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Arlington, VA: American Psychiatric Publishing.

Antiterrorism and Effective Death Penalty Act, 28 U.S.C. §2254 (1996).

Antiterrorism and Effective Death Penalty Act. (n.d.). Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA). Retrieved from https://www.law.cornell.edu/wex/antiterrorism_and_effective_death_penalty_act_of_1996_aedpa

Atkins v. Virginia, 536 U.S. 304 (2002).

Bensberg

G. J.

Sloan

(1950). A study of Wechsler’s concept of “normal deterioration” in older mental defectives. Journal of Clinical Psychology, 6, 359–362.

Blanco

(n.d.). In Hill

W L.

Blanco

J.I.

(Eds.) Murderpedia. Retrieved from http://www.murderpedia.org/male.H/h/hill-warren-lee.htm

10.

Blume

J. H.

Johnson

S. L.

Marcus

Paavola

(2014). A tale of two (and possibly three) Atkins: Intellectual disability and capital punishment twelve years after the Supreme Court’s creation of a categorical bar. William & Mary Bill of Rights Journal. Retrieved from http://papers.ssrn.com/sol3/Papers.cfm?abstract_id=2549737

11.

Brengelmann

J. C.

Kenny

J. T.

(1961). Comparison of Leiter, WAIS, and Stanford-Binet IQ’s in retardates. Journal of Clinical Psychology, 17, 235–238.

12.

Brumback

(2013, April 24). Warren Lee Hill’s stay of execution lifted by federal court. The Huffington Post. Retrieved from http://www.huffingtonpost.com/2013/04/24/warren-hill-execution_n_3149435.html

13.

Burden of proof. (2002). Encyclopedia.com . Retrieved from http://www.encyclopedia.com/topic/Burden_of_proof.aspx

14.

Civil Rights Cases. (2008). West’s Encyclopedia of American Law (2nd ed.). Retrieved from http://legal-dictionary.thefreedictionary.com/Civil+Rights+Cases

15.

Connor

(2015, January 27). Warren Lee Hill executed. NBC News. Retrieved from http://www.11alive.com/story/news/local/2015/01/27/warren-lee-hill-execution/22430827/

16.

Cooper v. Oklahoma, 517 U.S. 348 (1996).

17.

Davis

(2013, July 9). Execution date set for Warren Hill. Jackson Progress-Argus. Retrieved from http://www.jacksonprogress-argus.com/news/2013/jul/09/execution-date-set-warren-hill/

18.

DeMatteo

Marczyk

Pich

(2007). A national survey of state legislation defining mental retardation: Implications for policy and practice after Atkins. Behavioral Sciences & the Law, 25, 781–802.

19.

Doctors who evaluated death row prisoner for Georgia now say that Warren Hill has mental retardation. (2013, February 14). Death Penalty Information Center. Retrieved from http://dev.deathpenaltyinfo.org/documents/HillPRFeb14.pdf

20.

Eighth amendment. (n.d.). Cornell University, Legal Information Institute. Retrieved from https://www.law.cornell.edu/constitution/eighth_amendment

21.

Ellis

J. W.

(2003). Mental retardation and the death penalty: A guide to state legislative issues. Mental & Physical Disability Law Reporter, 27, 11–13.

22.

Gould

S. J.

(1996). The mismeasure of man. New York, NY: W.W. Norton.

23.

Habeas Corpus. (n.d.). Legal dictionary. Retrieved from http://dictionary.law.com/Default.aspx?selected=848

24.

Hall v. Florida. (2014). Oyez. Retrieved from https://www.oyez.org/cases/2013/12-10882

25.

Head v. Hill, 277 Ga.255, 587 S.E.2d 613 (2003).

26.

Herrnstein

R. J.

Murray

(1996). Bell curve: Intelligence and class structure in American Life. New York: The Free Press.

27.

Hill v. U.S. Court of Appeals, No. 13-10702-P (11th Cir. Feb. 19, 2013). Retrieved from http://d35brb9zkkbdsd.cloudfront.net/wp-content/uploads/2013/02/eleventh-circuit-warren-hill.pdf

28.

Hill v. U.S. Court of Appeals, No. 15-10192 (11th Cir. Jan. 26, 2015). Retrieved from http://media.ca11.uscourts.gov/opinions/pub/files/201510192.order.pdf

29.

Intellectual Disability. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Retrieved from http://www.dsm5.org/Documents/Intellectual%20Disability%20Fact%20Sheet.pdf

30.

Jones

(2015, February 11). Death-penalty opponents rally at state capitol. Athens Banner-Herald. Retrieved from http://onlineathens.com/local-news/2015-02-10/death-penalty-opponents-rally-state-capitol#.VNtktqaipks.mailto

31.

Lohr

(2013, February 13). Georgia death penalty under renewed scrutiny after 11th-hour stay. National Public Radio. Retrieved from http://www.npr.org/2013/02/20/172516106/georgia-death-penalty-under-renewed-scrutiny-after-11th-hour-stay

32.

Mobbs

West

(2013). Burden of proof for mental retardation. Journal of the American Academy of Psychiatry and the Law Online, 41, 585–588.

33.

Proceedings upon plea of insanity or mental incompetency at time of crime, O.C.G.A. §§ 17-7-131 (2014). Retrieved from http://www.lexisnexis.com/hottopics/gacode/Default.asp

34.

Reasonable doubt. (n.d.). Legal Information Institute, Cornell University. Retrieved from https://www.law.cornell.edu/wex/reasonable_doubt

35.

Renaud

(2003, October 7). Standard of proof stands in claims of retardation. Fulton County Daily Report. Available from http://www.murderpedia.org

36.

Sachy

(2013). Affidavit of Thomas H. Sachy M.D., MSc. Jones County, GA. Retrieved from http://s3.documentcloud.org/documents/603976/warren-hill-sachy.pdf

37.

Saviello

(2014, August). The appropriate standard of proof for determining intellectual disability in capital cases: How high is too high? ExpressO. Retrieved from http://works.bepress.com/timothy_saviello/1

38.

Silverman

Miezejeski

Ryan

Zigman

Krinsky-McHale

Urv

(2010). Stanford-Binet and WAIS IQ differences and their implications for adults with intellectual disability (aka mental retardation). Intelligence, 38, 242–248.

39.

Taylor

Krauss

(2014). Revisiting intellectual disability and the death penalty. Monitor on Psychology, 45(4). Retrieved from http://www.apa.org/monitor/2014/04/jn.aspx

40.

West

(2013, July 12). Advocacy group wants execution halted. The Albany Herald. Retrieved from http://www.gcdd.org/news-a-media/in-the-news/2503-advocacy-group-wants-execution-halted.html

41.

Whitaker

(2010). Error in the estimation of intellectual ability in the low range using the WISC-IV and WAIS-III. Personality and Individual Differences, 48, 517–521.

42.

Winship, in Re, 397 U.S. 368 (1970).