US Supreme Court Finds Execution
of the Mentally Retarded “Cruel and Unusual”;
You Have to Pass a Test
Before You Can be put to Death?
By Brett Trowbridge, Ph.D., J.D.
This article supported by The Trowbridge Foundation
Presented at the American College of Forensic
Psychology 19th Annual Symposium,
Palm Springs, April 2003
To be published in the American Journal of Forensic Psychology 2003
INTRODUCTION:
This summer the US Supreme Court
reversed its own [i]
1989 decision in deciding that it is now unconstitutional to execute
the mentally retarded. In Atkins v. Virginia [ii]
, the court held that the meaning of “cruel and unusual punishment”
can change over time depending upon shifting public sentiment. The court
pointed out that only a very few retarded persons had been executed since
the 1989 decision, and also pointed out that more states had outlawed execution
of the mentally retarded [iii]
or had prohibited capital punishment completely in the intervening
period, while none had passed laws allowing for the execution of the mentally
retarded. Justice Stevens authored the opinion for the 6-3 majority; the
opinion left to the states to decide the scientific standard for retardation,
the timing of the determination, and whether a judge or jury makes it. [iv]
The Supreme Court stated that
retarded persons are less likely to be deterred by the prospect of capital
punishment, and that principles of retribution do not apply to the retarded,
as they are less culpable, although the court offered no scientific authority
for these propositions. The court did not address the issue of whether IQ
testing measures the ability to be deterred, or whether IQ tests measure culpability.
The Supreme Court also stated
that mentally retarded persons are more likely to make false confessions,
and have less ability to assist their lawyers, and thus as a group face a
higher chance of false execution. [v]
Defense attorneys will be eager
to use this new ruling as a method to challenge executions of those on “death
row”, and will also argue that the prosecution should not ask for the death
penalty in cases with colorably retarded defendants.
Retardation is diagnosed in cases
in which the subject has a low IQ score, and has also shown a pattern of functional
disabilities in his adaptive patterns of living; school and employment records
are obviously important here. However, many criminal defendants will have
histories that arguably might qualify, since many criminal defendants did
poorly in school and have spotty employment records. Furthermore, after being
arrested for aggravated murder, defendants will obviously not be motivated
to perform well on IQ tests.
DEFINITIONS OF RETARDATION:
There is little disagreement as
to definitions of mental retardation, as both the American Association on
Mental Retardation (AAMR) and the Diagnostic and Statistical Manual of Mental
Disorders, 4th edition - Text Revisions (DSM-IV-TR) use basically
the same three criteria. [vi]
DSM-IV-TR states that before a diagnosis of “mental retardation”
can be made the individual must have received a score of approximately 70
or below on an individually administered IQ Test. Onset must be before the
age 18, and there must also be functional disabilities or limitations in the
adaptive pattern of living commensurate with the low IQ number [vii]
. Thus, clearly persons who have college degrees, have successfully
operated businesses, etc. would not be too “retarded” to execute even if they
scored low on IQ tests. However, the vast majority of persons convicted of
serious crimes have not graduated from high school, and have spotty work histories;
many have usually lived with their parents, and thus have never really lived
independently. An argument can easily be made that they have had a pattern
of living that has been “retarded”. For these individuals IQ testing would
probably be the most authoritative “evidence” of retardation, now that the
US Supreme Court has ruled that mentally retarded people cannot be executed.
IQ TESTING
Intelligence testing is a fairly
recent development first introduced by psychologists during World War One
to test ability among recruits. Although originally the concept of Intelligence
Quotient meant the quotient between the person’s “mental age” and his chronological
age, the concept of “mental age” is not now generally utilized or accepted.
To arrive at an “IQ” the psychologist administers a series of short “sub-tests”,
and then compares the subject’s scores with the “norms” for the sub-test.
The psychologist can then state for each sub-test whether the subject’s score
was above or below the average score for his age-group, and how far it was
above or below the average or “mean” score. A statistical term, “standard
deviation”, is used to measure how far above or below average the subject’s
score was. Only about 2% to 3% of the population will score more than 2 standard
deviations below the mean, and these persons are by definition “retarded”
as measured by that sub-test. The scores on the various sub-tests are than
averaged to arrive at a “Full-Scale” IQ number. More sophisticated IQ tests
will also produce “Verbal” and “Performance” IQ scores, which are supposed
to tap verbal and non-verbal forms of intelligence respectively. By convention
the average, or “mean” score on any intelligence test is 100, with a standard
deviation of 15, so by definition on any IQ test a score of 70 or less is
by definition “retarded”.
It immediately becomes apparent
that the real question is not only what the IQ score is, but also what the
tasks are that make up the test. Most people will perform relatively well
on some tasks but will be below average on others, depending, for example,
on whether the tests involve timing the subjects in solving mazes, correct
definitions of vocabulary, or remembering a series of numbers backwards.
The more sophisticated intelligence tests use more sub-tests, so that more
forms of “intelligence” are measured. An individually administered intelligence
test with numerous sub-tests may take a psychologist up to two hours to administer
and score. Less sophisticated tests utilizing only one or two “sub-tests”
may be administered and scored in as few as fifteen minutes, but do not tap
as many different domains of intelligence, and thus may not be as accurate
in estimating a person’s “true” level of intelligence. There are at least
one hundred readily commercially available individually administered intelligence
tests that have published norms for adults, so psychologists have many tools
available, but measurement of intelligence is certainly not an exact science,
in the sense that one individual given a number of different IQ tests would
usually show a considerable distribution of different scores, since each IQ
test taps different domains of intelligence depending upon the tasks involved.
Most psychologists would agree
the best-accepted intelligence test for adults is the Wechsler Adult Intelligence
Scale III (WAIS III) [viii]
, published in 1997, the latest iteration of a test first devised
in 1939. The WAIS-III was eleven sub-tests, so there are eleven “scale scores”
which produce a resultant Verbal IQ (average over 6 verbal sub-tests), Performance
IQ (average over 5 performance sub-tests), and Full Scale IQ (average over
all eleven sub-tests).
Very frequently an individual
will score in the “retarded” range on one or more sub-tests, but because he
also has relatively higher scores on other sub-tests, his overall, or “Full-Scale”
IQ is not in the “retarded” range on the WAIS-III. This points to the issue
of whether an “overall” or “Full-Scale” IQ is a more meaningful concept, or
whether the more meaningful discussion is what the subject’s strengths and
weaknesses are as measured by the scale scores. This is a boiling controversy
in psychology now, with prominent psychologists in both camps. Muriel Lezak,
a leading neuro-psychologist, believes that Full-Scale IQ scores are not meaningful.
In her article entitled “IQ: Rest in Peace” [ix]
, she states, “When the many and various neuropsychological observations
elicited by so-called “intelligence” tests are lumped together into a single
IQ score or even three the product of this unholy conversion is a number that
in referring to everything represent nothing.” Dr. Lezak comes out a background
in neuropsychology, a profession which uses a large number of different psychological
tests administered to one subject in an attempt to evaluate the true reason
for the subject’s difficulties in functioning. Various patterns of test results
are indicative of various types of brain-dysfunction. After all, individuals
can suffer from mental retardation for numerous reasons including heredity,
specific chromosomal changes, problems in utereo, (e.g. Fetal alcohol syndrome),
or because of diseases or toxic conditions encountered soon after birth. [x]
“Retarded” scores (in the statistical sense of being in the bottom
2% to 3%) can also result from brain disease or brain –injury occurring during
adulthood, such as chronic alcoholism, Huntington’s disease, Altzheimer’s,
etc., although such persons are not technically mentally retarded since the
onset was after age 18. Thus, “mental retardation has many different etiologies,
and may be seen as a final common pathway of various pathological processes
that affect the functioning of the central nervous system” [xi]
.
The issue seems to boil down to
whether there is really meaning to the concept of ”overall” intelligence,
or whether these are a number of different “intelligences” that are important
in understanding a person’s functioning, and a number of different types of
“retardation” with various causes and various patterns of deficits. Taken
to its extreme the proponents of “overall” IQ could administer numerous different
IQ tests, each with numerous sub-tests, and average them all in an attempt
to come up with a more accurate estimate of the “true” IQ. By the same token,
taken to its extreme those adhering to Lezak’s position might also administer
numerous IQ tests, each with numerous sub-tests, but instead of averaging,
would focus on the patterns of the lowest scores in an attempt to make a more
definitive diagnosis of the subject’s exact deficits. Proponents of the “overall”
IQ approach would criticize this approach, suggesting that if you give somebody
enough tests you will almost always be able to find one or two on which that
person will score poorly.
DSM-IV-TR deals with this controversial
issue by stating “When there is significant scatter in the sub-test scores,
the profile of strengths and weaknesses, rather than the mathematically derived
full-scale IQ, will more accurately reflect the person’s learning abilities.
When there is a marked discrepancy across verbal and performance scores, averaging
to obtain a full-scale IQ score can be misleading.” [xii]
The problem is, “significant scatter”, is not defined, and psychologists
do not agree on when scores are far enough apart to be “significant”. It
seems to me very likely that in many death penalty cases an argument will
be made that because of wide differences in scores, only the lower scores
should be discussed when determining “mental retardation”. Thus, there will
often not be an agreement even as to which scores on an IQ test should be
used in measuring intelligence.
In any case, there is criticism
of current IQ tests including the Wechsler tests that they are outmoded ways
of measuring intelligence as they do not tap many forms of intelligence now
known or postulated to exist by theorists. [xiii]
IQ TESTS ARE DESIGNED TO “PREDICT”
ACADEMIC PERFORMANCE
It is commonly understood that
IQ tests purport to measure a person’s ability to learn, not what the person
has actually learned. However, it is difficult to tease out “ability” from
“achievement”, and to a certain extent IQ tests do measure what a person “knows”
or “has learned”, particularly those using sub-tests that test for facts,
such as, “Name three men who have been presidents of the United States since
1950.” Indeed, the way that IQ tests are “validated” is that they are shown
to correlate with school grades, or performance on achievement tests. Indeed,
most accept that an IQ test is best used for attempting to estimate (or “predict”)
academic performance, because validation research shows that IQ tests are
somewhat useful at estimating academic performance. Thus, when a test publisher
asserts that his IQ test is “valid”, he is usually saying research shows scores
on that test correlate with academic and/or work performance. Some publishers
simply provide proof that their tests are correlated with other already accepted
IQ tests, such as the Wechsler tests. Correlations between scores of the
same person on different IQ tests are often fairly low. For example, Full
Scale IQ scores on the WAIS III correlate with scores on the Standard Progressive
Matrices [xiv]
at a correlation co-efficient of +.64 [xv]
(accounts for about 41% of the variance), and correlate with scores
on the Stanford Binet [xvi]
at a better correlation co-efficient of +.88 [xvii]
, (accounts for about 74% of the variance). Thus, it is entirely
possible for the same person to be “retarded” on one IQ test, but not on another.
ARE IQ TESTS “RELIABLE”
So far we have seen that IQ tests
are a somewhat “valid” but still fairly inexact way of measuring a person’s
potential to do well at work or on a job [xviii]
. Although there is no evidence that IQ tests really measure an
ability to be deterred by capital punishment, or measure culpability in some
sense, society apparently assumes a test which is supposed to correlate with
academic and work skills will somehow measure those domains as well.
Why is it that IQ tests are inexact
measurements of the construct of “intelligence”? We have already seen that
there is not general agreement as to what ”intelligence” even is, or on what
tests best measure it. One problem is that IQ tests are not even very “reliable”.
In psychology the “reliability” of a test has to do with how consistent the
scores are if you measure them more than once, and it is axiomatic that a
test cannot be any more “valid” than it is “reliable”. One would hope that
if you gave the same person the same IQ test twice the scores would usually
be near the same. However, research shows (for example) that upon re-test
with the WAIS-III the average score in the 16-29 age group will rise by 5.7
points, probably because the subject has better learned how to take the test
through more practice (so called “practice effects”) [xix]
. These test re-test differences become even more pronounced among
the neurologically and psychiatrically disturbed, disorders many low-functioning
prisoners would have. [xx]
For example, in one study among this group using IQ scores on
the WAIS-R (an earlier version of the WAIS-III), test – retest reliabilities
were even somewhat lower than published in the standardization sample. Within
that small group there was one person who showed a drop of 13 IQ points upon
retesting, while another saw a gain of 18 points of overall “IQ” [xxi]
. Thus, IQ scores may be especially unreliable in the lower ranges.
Another issue is whether scores
are reliable across different examiners. Virtually all test publishers require
an individual wishing to purchase an IQ test to certify he or she is a doctoral
level psychologist with training in test construction and test administration.
Thus, usually only licensed psychologists (or those working for them or being
trained by them) are considered qualified to administer, score, and interpret
IQ tests, or to make the diagnosis of “mental retardation”. Research shows
that even qualified psychologists make a surprisingly high number of administrative
and scoring errors. [xxii]
Another issue is that probably
because of better schools, better medicine and nutrition, and better dissemination
of information, people are gradually getting “smarter” on average, at least
in the sense that they are scoring higher on average on IQ tests. [xxiii]
This is one reason why some IQ tests are regularly updated and
“re-standardized” every several years with a modern sample of “normal” people.
This effect is so pronounced that it’s size has even been estimated – it is
estimated to be on average about 3/10’s of an IQ point a year. [xxiv]
This means that an IQ score derived from an older IQ test which
has not recently been “re-standardized” will probably overestimate a person’s
IQ, on average about 3/10’s of an IQ point times the number of years ago the
test has standardized. For example, on the WAIS-R (standardized in 1981)
people will score on average about 4.8 points higher than on the WAIS-III
standardized in 1997, 16 years later (16 x 0.3 = 4.8). [xxv]
For that reason, it is not really correct to compare an individual’s
IQ as measured at some time in the past on an older version of an IQ test
with his current scores on the modern version of the test.
Another frequently over-looked
issue is that some subjects are not appropriate for traditional IQ testing
because of language and/or cultural factors. Those whose native language
is not English or those who did not grow up in the U.S. or Canada should not be compared to standardization samples of people whose native
language is English and who come from our culture. [xxvi]
Although some efforts have been made to develop a “culture-free”
IQ test [xxvii]
, by their very nature these tests may be even more inexact than
standard IQ tests.
For all at the reasons stated
above, as well as some others, IQ scores are relatively inexact. Indeed,
many feel expressing intelligence in a numerical score suggests more precision
than actually exists, and instead suggest expressing IQ’s in a range. On
the Wechsler scales the 67% confidence interval is about plus or minus 5 points.
In other words, if a person receives a full-scale IQ score of 75 on the WAIS-III,
67% of the time his “true” IQ will be between 70 and 80. Although psychologists
conventionally use a 67% confidence interval [xxviii]
, in order for 95% confidence to be achieved the range would have
to be plus or minus 13 points. In other words, if a person receives a Full
Scale IQ of 75 on the WAIS-III, there is a 95% chance his “true” IQ would
fall between 62 (75-13) and 88 (75+13). Because of the fact that measurements
of IQ are so imprecise, most commentators (including the DSM IV-TR) allow
a diagnosis of mental retardation even if the IQ score is as high as 75 when
the person’s life functioning suggests significant adaptive deficits [xxix]
. Just above the mildly mentally retarded range is the range of
borderline intellectual functioning (IQ 71-84). Differentiating mildly mentally
retarded from borderline intellectual functioning requires careful consideration
of all available information.
MENTAL RETARDATION AND THE DEATH PENALTY
It must be remembered that prisoners incarcerated in prison have all been
found competent to stand trial (or competent to enter a guilty plea), and
competent to be sentenced. Thus, presumably some of the very low functioning
individuals have been excluded from the prison population because of legal
incompetencies. Nonetheless, on average those incarcerated in prison have
lower IQ’s that “normals”, and thus the mentally retarded are over-represented
in a prison sample. Whereas by definition in the general population 2% to
3% of individuals are “retarded”, estimates for mental retardation among those
in prison range as high as 10%. [xxx]
However, there is evidence that the mentally retarded commit
disproportionately fewer serious crimes than minor crimes. [xxxi]
Even before this summer’s US Supreme
Court ruling courts were often or even usually factoring in IQ data when deciding
whether to impose the death penalty, and presumably a number of persons were
not sentenced to death because of their retardation. It appears that what
is different now is that courts will begin using low IQ scores as a dispositive
factor instead of merely a contributing factor. Defense attorneys with low
functioning clients who are facing a possible death sentence will certainly
begin focusing even more on IQ testing than was the case before the ruling.
Similarly, judges will have to worry more about being reversed if they sentence
an individual to death who had received test scores of 70 or below. Experts,
whether hired by the prosecution or the defense, and whether proponents of
“global” IQ or adherents of Lezak’s analysis of low scores, will almost certainly
be doing more IQ testing and using more different tests in each case. Psychologists
will be the experts used, as professionals such as psychiatrists, counselors
and social workers have no expertise in IQ testing. In jurisdictions where
the jury decides whether to impose the death penalty (such as Washington) some sort of IQ evidence will have to be presented to the jury, and some
sort of determination as to whether the defendant is retarded will have to
be made. The most likely scenario is that there will usually be a “battle
of the psychologist experts”.
One obvious issue that would almost
certainly be brought up by the prosecution in every case would be the issue
of possible malingering. The defendant certainly has a motivation to do poorly
on IQ tests, especially if he understands his “informed consent” form, which
would explain to him, presumably in language that he could understand, exactly
what was going to be done with the results of the testing and evaluation,
namely that the results would be used to help decide whether or not he would
be put to death! One wonders what the result would be if the defendant told
the examiner he did not wish to sign the informed consent form because he
did not feel he understood it, or because he did not wish to consent to the IQ testing [xxxii]
.
Assuming that he does consent
to an IQ test administered either by a defense psychologist or a state psychologist,
a defendant facing possible execution might likely attempt to depress his
score by missing items he could actually solve. Most IQ tests are administered
sub-test by sub-test, instead of mixing up the items from the various sub-tests
with each other. Furthermore, the very easiest item is listed first, and
the next easiest, and so forth. This is done to facilitate the work of the
examiner, who can establish a “floor” and “ceiling” for each sub-test, thus
avoiding even administering items way too easy or way too hard for the subject.
Scoring is also facilitated by progressing through each sub-test from the
easier to the harder items.
Unfortunately, however, these
same procedures for making the examiner’s job simpler also help facilitate
successful malingering. A malingerer knows that he must pass at least the
very easiest items on each group of questions, and then should start missing
items consistently after a certain point. The fact that the questions increase
gradually in difficulty makes it easier for the malingerer because he can
simply miss all of the questions in each group past the point where he thinks
a “retarded” person would score. In other words, on most IQ tests the malingerer
has the advantage of knowing which questions are easier and which are harder
simply because their order of presentation progresses from easier to harder.
Psychologists have developed tests
to help determine whether individuals are malingering symptoms of mental retardation
or cognitive dysfunction. One strategy to detect malingering is to mix up
the difficulty of the questions, so that it is not immediately apparent which
questions are easier and which are harder. [xxxiii]
Persons who answer too many of the easy items wrong while still
getting some of the harder ones right are thought to be malingering. Another
strategy used to detect malingering is to ask true-false questions. Malingerers
have a tendency to deliberately miss items, but if they get a statistically
improbably high number of the items wrong, their scores are worse than would
be achieved by chance alone, and thus constitute proof they knew at least
some of the items but deliberately marked them wrong. For example, out of
50 true-false items, a person who knew none of the answers but who guessed
on each item would get about 25 right. A person who gets less than 17 right
has an improbably high number wrong, so scores of 17 or less are thought to
be indicative of malingering. Unfortunately, some malingerers may actually
be retarded, so it is not possible to definitively conclude that all those
testing as malingerers are not retarded.
Given the stakes and the potential
for malingering it would appear that every expert evaluation by a psychologist
of a defendant facing the death penalty should include not only IQ testing
but also assessment of the possibility of malingering using specialized testing
to detect malingering.
Although few prisoners have been put to death in many states recently (including
the State of Washington), and although even fewer colorably retarded individuals
have recently been put to death in most states, the question of retardation
still applies to a fairly broad range of cases, i.e., to all cases in which
the possibility exists that the state could ask for the death penalty. In
many such cases now the prosecutor’s decision may be made based on what IQ
testimony the defense has submitted that it intends to use should the prosecution
ask for capital punishment.
Obviously IQ testing has now become much more important in death penalty
cases than it was before the US Supreme Court’s ruling. Psychologists hired
to do such IQ testing should use tests with up-to-date norms. Ideally a number
of tests or sub-tests should be employed, some of which are designed to assess
malingering. Because of practice effects it may not be advisable to administer
the same IQ test repeatedly to the same individual. IQ tests administered
years ago and thus available through school or medical records will be based
on older norms, and may not be directly comparable to newer tests given after
incarceration based on contemporary norms. However, old records, such as
school testing, testing for entry to the military, SSI records, etc., have
the advantage that the individual was probably not motivated to malinger at
the time of taking of those tests.
For all of the reasons discussed
in this article, in many cases the determination of whether an individual
is “retarded” may not be cut and dried. In many if not most cases there will
be a considerable “gray area”, with room left for argument and disagreement.
IQ tests are fairly imprecise measuring instruments. They were designed
to “predict” (i.e. correlate with) school and job achievement, but they do
better at predicting school performance then they do at predicting job performance,
because they are fundamentally academic tests. [xxxiv]
IQ tests are imperfect yardsticks for estimating or “predicting”
school achievement. Examiners using different tests to measure the IQ of
an individual would usually come up with a relatively wide-band of scores.
And even if “intelligence” as measured by “intelligence testing” is a meaningful
(although inherently “squishy”) concept, no scientific evidence exists that
IQ testing actually measures the ability to be deterred by capital punishment,
or “culpability”. Just as with society in general, the US Supreme Court
seems to have an over-inflated belief in the meaning and the powerfulness
of IQ testing.
COMPETENCY TO BE EXECUTED
The term “competency to be executed”
has its origins in the common law [xxxv]
, and is not the same as whether a condemned person may not be
executed due to mental retardation. In 1986 the United States Supreme Court,
recognizing the long common law origins of the concept, declared it is unconstitutional
to execute those deemed incompetent. [xxxvi]
Although the decision itself did not define the standard for
competency to be executed, Justice Powell, in a concurring opinion, enunciated
the standard by holding that the 8th Amendment ”forbids the execution
only of those who are unaware of the punishment they are about to suffer and
why they are to suffer it.” Another formulation is that the court must assess
the “condemned prisoner’s capacity to understand the nature and purpose of
the death penalty in relation to his or her homicidal conduct” [xxxvii]
. A standard forensic psychology textbook definition of when a
person is competent to be executed is when he knows “what the physical finality
of death means, and the causal link between the act of murder and the penalty”. [xxxviii]
How is the “competency to be executed”
standard different from the “too mentally retarded to execute” standard?
Besides the obvious differences in definition, we see immediately that there
are many similarities. First, the rationales cited by courts for the prohibitions
are remarkably similar. We saw above that the U.S. Supreme Court feels normal
principles of retribution and deterrence do not apply to the retarded, and
that the court is concerned retarded persons might not always get a fair shake
from the criminal justice system. The same reasons are generally cited by
the courts as rationales for the prohibition against executing the “incompetent”. [xxxix]
Second, an expert opinion as to
either ”competency to be executed” or “retardation” can only be reached by
a highly trained and qualified mental health professional after a thorough
examination. Either a psychologist or a psychiatrist could address “competency
to be executed” [xl]
, but only a psychologist is qualified to administer IQ tests,
which are required for a diagnosis of mental retardation. In either case,
the professional must do a face-to-face interview, review records, take a
detailed history, and do cognitive testing in order to make a diagnosis.
Malingering is an obvious issue
with competency to be executed evaluations, just as it is with assessing mental
retardation in the death penalty context. As suggested above, psychological
tests specifically aimed at assessing malingering of cognitive symptoms are
probably essential. Assessment of “premorbid functioning”, or the individual’s
life-style and adaptive functioning in the period before the crime, becomes
important. Thus, school records, military records, SSI records, etc., may
play a large role given the ever present suspicion of malingering in death
penalty cases. Truly retarded persons will usually have had a pattern of
very poor academic performances in school (including placement in special
education classes), and usually would not have been able to pass the entrance
exams required for military service.
We see that the database the professional
must draw on to do either assessment is virtually the same. For that reason
many courts may choose to allow professionals to testify to both issues based
on the same evaluation. Procedures for courts to use in determining whether
an individual is mentally retarded were not set forth in Atkins v. Virginia,
but procedural requirements are likely to be similar to those found by the
U.S. Supreme Court to be minimally required in the competency to be executed
context: there must be an opportunity for the prisoner and/or his lawyer to
be heard, challenges to the findings of state-employed mental health professionals
must be allowed, and the final decision must be up to the judicial, not the
executive branch.
[xli]
States might choose to address both issues at the same time using
the same procedures.
Findings of incompetency to be
executed are extremely rare. As of 1992 there had only been two cases in
which a death row inmate was found incompetent for execution [xlii]
since the death penalty was reinstated by the U.S. Supreme Court
in 1976. [xliii]
In one case, upon the court’s ruling of incompetence, his death
sentence was reduced to life in prison, and he was removed from death row. [xliv]
However, in the other case the individual was treated at a mental
hospital in order to attempt to restore his competency to be executed, and
then was returned to death row. [xlv]
Thus, it is not clear whether persons judged incompetent to be
executed would remain on death row, or go to the general prison population.
Similarly, nothing in Atkins v. Virginia makes it clear whether those
judicially determined to be mentally retarded would remain on death row.
However, mental retardation is thought to be a permanent condition, and it
is unlikely anyone would attempt to do treatment to overcome or “cure” mental
retardation. Presumably, however, an individual who did not know what death
was and did not know the causal link between the act of murder and the penalty
could conceivably be taught to understand that concept.
Both types of evaluation are made
much more difficult if they must be performed on an individual already sentenced
to death and on death row. [xlvi]
The expert must measure current cognitive functioning in a context
in which malingering is an obvious issue, and must then compare the results
of that testing of cognitive functioning with the individual’s life-style
and adaptive behavior. But what is “normal” adaptive behavior for those on
death row, and what is “retarded” or “incompetent” behavior? In such an impoverished
environment an inmate who does virtually nothing except sleep and eat may
still not be showing “maladaptive” behavior. Again, reliance on how the individual
functioned before his arrest may become more relevant. In both types of evaluation
a problem arises if the individual says he wants to be executed, and thus
does not wish to pursue further legal options, including evaluations for mental
retardation or for competency to be executed. Are those who are possibly
retarded or incompetent able to give “informed consent” and make such a waiver?
How would one make such a determination? [xlvii]
Perhaps the largest difference
between the two standards of “competency to be executed” and “mental retardation”
is that the new prohibition against executing the mentally retarded will apply
to many more individuals. As mentioned above, rarely have courts found individuals
incompetent to be executed, since even most severely disturbed individuals
can know what death is, and can understand they are to be put to death for
committing murder. On the other hand, using a conservative estimate of 5%
of death row inmates being retarded, and there are perhaps 3500 people currently
on death row in the United States [xlviii]
, the ruling might potentially affect as many as 175 current death
row inmates. Looking at it another way, since the death penalty was reinstated
in the United States in 1976, at least 34 mentally retarded persons have been executed in the
United States. [xlix]
A prominent psychologist, Stanley
Brodsky, has questioned whether mental health professionals can be objective
and neutral given the emotional context of the entire death penalty issue. [l]
Some psychologists are not willing to participate in competency
to be executed evaluations because of their objections to the death penalty, [li]
while others decline to treat prisoners in order to help them
regain competency to be executed. [lii]
One Board of the American Psychological
Association has opposed the participation of psychologists in “routine certification
of competency for execution”. In the Board’s view, “Justification for participation
in such evaluations should be based solely on the possibility of bringing
new information which might change the legal verdict and subsequent death
sentence”. [liii]
One wonders how members of such boards would feel about routine
IQ testing of death row inmates to determine if they are mentally retarded.
Perhaps the chance that an inmate might be retarded would be enough to constitute
a “possibility of new information”. Although there are psychologists of all
persuasions concerning the death penalty, a larger percentage of psychologists
are opposed to the death penalty than is true among the general population.
Yet, only psychologists are competent to do IQ testing to determine if an
individual is mentally retarded. Undoubtedly, some psychologists will complain
that using IQ tests to determine whether a person can be executed is an unethical
and improper use of psychological testing.
If the U.S. Supreme Court feels
that mentally retarded persons do not possess the requisite culpability to
justify their execution, could there be other groups who should be given the
same consideration? Are people with major mental illnesses, such as schizophrenia
or bipolar disorder, also deserving of leniency? What about individuals who
commit their crimes as juveniles [liv]
? Would such persons be more likely to be falsely convicted?
The death row population is dropping now in the United States, apparently because juries and judges are now worried about the possibility
of executing innocent people due to the recent high number of reversals due
to DNA evidence. [lv]
