Você está na página 1de 7

Professional Psychology: Research and Practice Copyright 2006 by the American Psychological Association

2006, Vol. 37, No. 1, 59 – 65 0735-7028/06/$12.00 DOI: 10.1037/0735-7028.37.1.59

Unresolved Controversies Concerning Psychopathy: Implications for

Clinical and Forensic Decision Making
John F. Edens
Southern Methodist University

At some point in their careers, clinicians who work or consult in forensic and correctional settings will
almost certainly encounter individuals who exhibit psychopathic personality features. Because of the
widespread use of this disorder to inform legal and clinical decision making, psychologists should be
exceedingly familiar with the relevant research literature on this topic before venturing into these settings.
This article reviews the empirical bases of several clinically relevant claims and assertions regarding
psychopathy and concludes that many areas of research are decidedly more equivocal in their findings
than is commonly perceived. Although there is much to be gained by assessing psychopathy in various
contexts, clinicians need to be cautious about drawing overzealous and empirically questionable con-
clusions about an important disorder that also has great potential for abuse.

Keywords: psychopathy, PCL-R, forensic, controversies, decision making

Few disorders would seem to elicit as visceral a reaction among assessment instruments, and professional guidelines. As such, fa-
mental health professionals as psychopathic personality (psychop- miliarity with the ever-burgeoning psychopathy literature would
athy). Almost by definition, highly psychopathic individuals are seem to be a prerequisite to engaging in many types of forensic
those who use deception, subterfuge, coercion, and perhaps even assessments or providing therapeutic services in correctional
outright violence to accomplish their goals, with a callous disre- settings.
gard for the potentially devastated lives they leave in their wake Because of the ever-increasing use of psychopathy assessments
(Hare, 1993, 1996). Informal discussions with clinicians and an- to inform life-altering (e.g., indeterminate confinement) or even
ecdotal accounts in the published literature suggest that mental life-ending (e.g., capital punishment) legal decisions around the
health professionals are by no means immune to such manipulation world (Cunningham & Reidy, 2002; Edens & Petrila, 2006), one
in that they frequently are duped or conned in both their clinical could argue that the bar is higher in terms of the need for clinicians
and forensic work (Cleckley, 1976; Hare, 1993, 1998). Because to avoid erroneous or at least highly controversial conclusions
many clinicians appear to be increasingly augmenting their prac- about those identified as “psychopathic.” In addition, because of
tices by providing services in forensic or correctional settings the highly prejudicial associations that the label psychopath evokes
(Otto & Heilbrun, 2002), it appears likely that greater involvement among the lay public (e.g., Ted Bundy, Charles Manson; see
with the criminal justice system will lead to increased contact with Helfgott, 1997; see also Edens, Colwell, Desforges, & Fernandez,
highly psychopathic clients or examinees, who typically account 2005), practitioners should be clear about what such a label does
for 20% to 30% of prison populations (Hare, 2003). Because good and does not imply about adults and children who display person-
clinical practice is founded on appropriate knowledge, skills, and ality features that appear to be consistent with historical concep-
experience, it seems ethically mandated that those who work in tions of this disorder.
such settings should be familiar with relevant empirical literature, Given the ramifications of the psychopathy label, the purpose of
the current review is to address several contentious issues regard-
ing this disorder that are directly relevant to its use in legal and
JOHN F. EDENS received his PhD in clinical psychology from Texas clinical decision making. The particular topics reviewed reflect
A&M University in 1996. He is currently an associate professor in areas in which I have observed clinicians and researchers drawing
the Department of Psychology at Southern Methodist University. overly forceful, categorical, or sweeping conclusions about psy-
His clinical and research interests focus on forensic assessment chopathy in the courtroom, in formal and informal talks, and/or in
and the use of psychological data to inform legal decision making. print.1 I believe that close inspection of the available research
He frequently serves as a consultant on criminal cases involving evidence indicates that considerable restraint is warranted in rela-
mental health evidence. In 2001, he was the corecipient of the tion to these topics. In the sections below, I describe several
Saleem Shah Award for Early Career Contributions to Law and
Psychology, awarded by the American Psychology-Law Society 1
Which is not to suggest that the present review constitutes an exhaus-
and the American Academy of Forensic Psychology. tive list of controversies by any means. For example, the use of the
CORRESPONDENCE CONCERNING THIS ARTICLE should be addressed to Rorschach to assess psychopathy has come under extensive criticism (e.g.,
John F. Edens, Department of Psychology, Southern Methodist Wood, Lilienfeld, Garb, & Nezworski, 2000) and the construct validity and
University, P.O. Box 750442, Dallas, TX 75275-0442. E-mail: utility of self-report measures of psychopathy continues to be a widely
jedens@smu.edu debated topic (e.g., Lilienfeld & Fowler, 2006).


conclusions about psychopathy that appear difficult to defend should not be ignored or dismissed. Neither should they be as-
empirically, I briefly summarize evidence bearing on these issues, serted to portend an ominous downward spiral that will inevitably
and I highlight some of the practice implications of what I argue to lead to adult psychopathy, particularly given the developmental
be more qualified but defensible interpretations of the current fluidity of childhood and adolescent characteristics such as gran-
literature. By highlighting some of the current limitations of what diosity, impulsivity, and irresponsibility (Edens, Skeem, et al.,
is known about psychopathy, I hope that practitioners can avoid 2001; Frick et al., 2003). In settings in which the presumed
possible abuses of what Hare (1998) described as “the single most stability of a “bad character” may have profound weight in terms
important clinical construct in the criminal justice system” (p. 99). of subsequent treatment and rehabilitation efforts (e.g., waiving
adolescent offenders out of the juvenile and into the adult criminal
“Once a Psychopath, Always a Psychopath” justice system), mental health examiners should be forthcoming
The Controversy about how little is actually known about the stability of these
characteristics among children and adolescents. In relation to
The position that psychopathy is remarkably stable over time adults, examiners would do well to avoid longitudinal claims about
underlies much that has been written about this disorder in that it the constancy of a particular individual’s psychopathic traits when
is argued essentially to reflect a dispositional, immutable person- these claims are based on cross-sectional studies of PCL-R scores
ality constellation (e.g., Gacono, 2000). One version of the “im- across highly select samples of prison inmates and forensic pa-
mutable personality” assertion applies to the developmental tra- tients (see the “Psychopaths Cannot Be Treated” section below as
jectory of children. Unlike Athena, highly psychopathic adults do well).
not spring forth fully formed from the head of Zeus—this is clear.
Existing evidence suggests that adults who obtain high scores on
“Where the Psychopath Goes, Violence Will Surely
the most widely studied measure of psychopathy, the Psychopathy
Checklist–Revised (PCL-R; Hare, 2003), start showing some signs
of the characteristic traits and behaviors at a relatively early age. In The Controversy
fact, the scoring of some PCL-R items is based specifically on such
conduct. It cannot be stated emphatically enough that these find- One could argue that the extensive attention given to psychop-
ings do not lead to the inevitable conclusion that the presence of athy over the last few decades has been largely attributable to its
psychopathic-looking “traits” among children and adolescents will association with violence. In fact, some commentators have even
ultimately result in adults who continue to demonstrate these traits described the PCL-R as the “unparalleled” measure of violence
and behaviors (Edens, Skeem, Cruise, & Cauffman, 2001; Hart, risk (Salekin, Rogers, & Sewell, 1996; cf. Gendreau, Goggin, &
Watt, & Vincent, 2002).2 The field lacks any significant longitu- Smith, 2002; Walters, 2003)—a position that also has been fre-
dinal research regarding the stability of these traits, although the quently endorsed in court cases (e.g., R. v. K.S., 2004; R. v. LaRue,
available evidence suggests that more extreme scores identified 2003). In many contexts, clinicians should be encouraged to con-
among young children in particular may regress toward the mean sider the presence of psychopathic traits when attempting to assess
in many instances (e.g., Frick, Kimonis, Dandreaux, & Farell, the likelihood that someone may act out aggressively, particularly
2003; see also Cauffman & Skeem, 2004). when one considers the recurrent finding that clinicians routinely
The second version of the immutability argument relates to focus on violence-risk factors that have little or no empirical
adults. Cross-sectional studies clearly indicate that there are older support (e.g., Elbogen, Mercado, Scalora, & Tomkins, 2002).
individuals in prison settings who exhibit psychopathic traits to a Although it would be defensible, in the aggregate, to conclude
considerable degree, although as a group these older individuals that elevated PCL-R scores identify individuals who may be more
display somewhat lower scores on the factor of the PCL-R (Factor violence-prone, the problem with such assertions in individual
2) more strongly wedded to impulsivity, social deviance, and cases is that examinees are never in aggregate contexts. Although
violence (Hare, 2003). The available research by definition comes several meta-analyses (e.g., Gendreau et al., 2002; Walters, 2003)
from offenders who are in prison settings—perhaps specifically and other large-scale studies have demonstrated that psychopathy
because they have committed serious criminal acts relatively late typically is associated with higher rates of violent and other forms
in life. There are no well-controlled longitudinal studies that have of criminal conduct in the community, such probabilistic findings
followed young, highly psychopathic adults to determine whether obviously do not support absolutistic assertions made by some
they in fact are very likely to end up as highly psychopathic adults examiners that individuals who are psychopathic will necessarily
much later in life. The closest approximation to such a study was engage in violent conduct in the future. More important, the
conducted by Rutherford, Cacciola, Alterman, McKay, and Cooke “average” association between psychopathy and violence in these
(1999) with 200 male and 25 female methaodone patients over a meta-analyses belies the fact that across studies the strength of this
2-year follow-up. Intraclass correlations for dichotomous classifi- relationship is remarkably heterogeneous. The implication here is
cations and continuous PCL-R scores ranged from .48 to .67. that one should not discuss the “global” relationship between
Although generally supportive of a relative degree of stability at psychopathy and violence but, instead, should consider other fac-
the group level, such results offer little support for arguments tors (e.g., context) that might help practitioners and researchers
regarding the immutability of psychopathy.
Practice Implications When confronted with this logical fallacy (affirming the consequent),
a useful strategy is to ask whether because all adult psychopathic offenders
Certainly the early onset of severe callousness, egocentricity, ate baby food as infants, eating baby food would be expected to be a good
and other characteristics or behaviors similar to adult psychopathy predictor of adult psychopathy.

better understand why such variability exists across different pop- or she may be violent 30 years later is clearly guesswork that
ulations and settings. presently lacks any scientific foundation.
As an example of the importance of context in risk assessment, When using the PCL-R, it would be helpful if examiners pro-
it is worth noting that highly psychopathic individuals often end up vided descriptive information to the consumers of their evaluations
in environments specifically designed to inhibit violence (e.g., regarding exactly how useful particular correlational results are in
maximum-security prisons), which of course may reduce the num- relation to the specific contexts and samples under consideration.
ber of opportunities to act out. Even when acting out does occur, For example, in a large sample of British prison inmates, Hare,
however, there is growing evidence that psychopathy may not be Clark, Grann, and Thornton (2000) reported that 42% of those with
a particularly useful risk factor in U.S. prison systems. For exam- elevated PCL-R scores committed an assault in prison, whereas
ple, Guy, Edens, Anthony, and Douglas (2005) recently reported only 16.4% of those with lower scores engaged in such institu-
mean meta-analytic correlations between the PCL measures and tional misconduct. Similarly, Edens, Buffington-Vollum, Keilen,
prison violence that were considerably weaker in studies of U.S. Roskamp, and Anthony (2005) recently noted that across three
prisoners (rw ⫽ .10) than in studies of non-U.S. prisoners (rw ⫽ U.S. prison samples in which the base rate of violent infractions
.24).3 was 22.2% for the entire group (N ⫽ 293), only 25% of the
Another issue of contention in relation to the “psychopathy psychopathic offenders committed such an infraction. Descriptive
equals violence” argument relates to assertions that psychopathic values such as these arguably are much more informative for
individuals are highly prone to serious aggression even beyond the laypersons than are the more complex summary statistics (e.g.,
age when most criminals tend to “burn out.” This is an issue that receiver operating characteristic curves, biserial correlations) typ-
arises in death penalty cases in which an offender might be ically reported in most research studies. Unfortunately, descriptive
released after serving a 30-year sentence in some jurisdictions statistics often are not provided in these studies.
(Cunningham & Reidy, 2002; Edens, Petrila, & Buffington- A final point to note is that even if an examiner deems psy-
Vollum, 2001). Support for this position perhaps stems most chopathy relevant to include in a risk assessment, it seems unten-
directly from an early study by Hare, Forth, and Strachan (1992), able that someone would ever rely solely on the PCL-R, because
in which they concluded that “the psychopath’s propensity for there are certainly instances in which low scores cannot be equated
violence and aggression may be relatively persistent across much with low risk (see Hare, 1998, for examples) and other circum-
of the life span” (p. 295). One other small study (Harris, Rice, & stances where high scorers can be managed successfully without
Cormier, 1991) reported that psychopathic inmates over the age of major incident (Cooke, 1997; Hare et al., 2000). Practitioners may
40 were more likely to reoffend violently than were comparably well need to consider the relevance of psychopathy when asked to
aged nonpsychopathic offenders. Despite these early findings, one make inferences about risk for violence and criminality, but they
large-scale subsequent study has failed to replicate these results should never base their opinions exclusively on PCL-R measures
(Porter, Birt, & Boer, 2001 [see Figure 1, p. 654; also see Figure 2, when conducting such work (Hemphill & Hare, 2004). For a useful
p. 655]), and a reanalysis of the Hare et al. (1992) results also example of how the PCL-R may be incorporated with other risk
strongly contradicted the “persistent-violence” assertion (Edens, factors, interested readers should examine the HCR–20 (Webster,
Petrila, & Buffington-Vollum, 2001).4 Douglas, Eaves, & Hart, 1997).

Practice Implications “Psychopaths Cannot Be Treated”

The fact that the magnitude of the psychopathy–violence asso- The Controversy
ciation has varied widely from study to study would seem to
This position has been espoused with varying degrees of cer-
compel examiners to be very specific about exactly what types of
tainty by mental health experts and judges in criminal proceedings
settings and circumstances are being considered when making
(e.g., Barefoot v. Estelle, 1983; U.S. v. Stitt, 1998) and in published
statements about risk for future violence. For example, in relation
research reviews (e.g., Gacono, Nieberding, Owen, Rubel, & Bod-
to U.S. prisons, it would seem hard to defend using the PCL-R in
an attempt to identify inmates who are likely to be violent given
the modest relationship noted in the literature. Even if statistically 3
These meta-analytic findings stand in sharp contrast to some of the
significant because of large sample sizes, a correlation of .10 (Guy opinions offered by prosecution experts in capital trials. Although not
et al., 2005) indicates that a large number of false positives will based on the PCL-R per se, testimony in the now infamous Barefoot v.
occur if the PCL-R is used to sort individuals into higher versus Estelle (1983) case represents perhaps the most extreme form of overreach-
lower risk groups. ing that has been documented. In response to a description of a hypothetical
Additionally, at present there is no evidence to suggest that offender with a criminal history similar to the defendant, James Grigson
practitioners should be using psychopathy scores to make predic- asserted that such a person would be a sociopath of the “most severe
category” (p. 919) and that there was a “one hundred percent and absolute”
tions regarding the likelihood of violent behavior literally decades
(p. 919) chance that he would commit future acts of violence— even if in
later. Even if a highly psychopathic 50-year-old prison inmate
prison. As empirically unsupportable as this conclusion was (Cunningham
were at increased risk for violence if released into the community, & Reidy, 2002), the U.S. Supreme Court’s affirmative decision gave
that does not imply that a highly psychopathic 20-year-old inmate experts considerable latitude to opine with great certainty about violence
will continue to be at increased risk for violence if released at the risk posed by criminal defendants, a precedent that still holds today.
age of 50. The relevance of including the PCL-R when assessing 4
Edens, Petrila, and Buffington-Vollum (2001) reported that in the
recidivism risk for an older offender seems questionable at best at combined 41–50-year-old age range, the rates of violent reoffending were
present, and assessing risk in a 20-year-old to ascertain whether he 11.6% for the psychopathic and 7.4% for the nonpsychopathic groups (ns).

holdt, 1997). Gacono et al. (1997), in fact, argued that there was settings where individual PCL-R ratings are being considered.
“nothing the behavioral sciences can offer for treating those with Results summarized in the most recent PCL-R manual (Hare,
psychopathy” (p. 119). This conclusion seems to have taken hold 2003) clearly suggest that well-trained examiners can score the
in some arenas in that high PCL-R scores increasingly serve as the PCL-R and its derivatives reliably, particularly when there is
rationale for denying treatment services to some offenders considerable file information available to inform these ratings. The
(D’Silva, Duggan, & McCarthy, 2004). The “cannot-be-treated” standard error of measurement (SEM) for the PCL-R reported in
argument is driven mostly by a consideration of research suggest- the new manual ranges from 2.7 to 3.4, depending on the sample
ing that adults and adolescents with elevated scores on the PCL-R and the scoring method used.
measures are less likely to respond to the types of interventions Despite these encouraging findings, such results do not lead to
that have been investigated in the past (e.g., Hare et al., 2000). the inevitable conclusion that a particular PCL-R score is a close
Although it would be hard to argue that psychopathy does not approximation of any given offender’s “true” level of psychopa-
raise serious concerns about treatment responsiveness, the certi- thy— especially in highly adversarial settings. In fact, examples of
tude of negative outcomes has been challenged by recent reviews rather significant disagreements in the scoring of the PCL-R are
(D’Silva et al., 2004) and research findings (Cooke, 1997; Skeem, evident in several legal cases (see, e.g., R. v. K.S., 2004, in which
Monahan, & Mulvey, 2002). Most of the earlier outcome studies one examiner indicated that the defendant’s PCL-R score fell at the
with psychopathic offenders were methodologically weak, and 32nd percentile, whereas the other argued it was at the 91st
none appear to have been based on rehabilitation principles that percentile). Although there are several potential explanations for
have shown some degree of efficacy among criminals (see Dow- such discrepancies (e.g., poorly trained examiners), one possibility
den & Andrews, 2000, for a meta-analytic review of the offender- is that in some cases there simply may be inadequate information
treatment literature). Additionally, one widely cited program in- upon which to make reliable ratings. Related to this latter issue, an
volved “novel” interventions that would be considered unethical interesting and apparently unexplored area of clinical practice is at
by today’s standards (see Harris, Rice, & Cormier, 1994, for an what point examiners believe they have sufficient information to
extensive review of a program that involved, among other “treat- score psychopathy scales (Edens & Petrila, 2006). The PCL-R
ments,” marathon nude encounter groups). manual (Hare, 2003) is clear that it should not be used absent
sufficient file information, but it is not entirely clear on what
Practice Implications constitutes “sufficient” in any given case.

Therapeutic nihilism, although perhaps understandable given

the affective and interpersonal traits associated with psychopathy, Practice Implications
would seem to be somewhat premature in relation to the treatment
and management of psychopathic individuals. Moreover, promis- Although there is no mandated level of training that individual
ing alternative intervention strategies (e.g., Wong & Hare, in examiners should have prior to using the PCL-R or its derivatives
press) have yet to be investigated in any systematic fashion. It may in applied settings (Hare, 2003), clearly it is not an instrument that
very well be that psychopathic offenders ultimately will need some can simply be “bought and used.” The PCL-R manual provides
type of intervention that is fundamentally different from what has excellent guidance in terms of what sorts of background and
been demonstrated to be effective with inmates in general. At training experiences are needed prior to using this instrument, but
present, however, it is not even known whether effective rehabil- it is apparent that examiners have misused the PCL-R and made
itation methods will have some impact with this subgroup of ratings that were quite suspect in some cases (Edens, 2001; Hare,
offenders, because such programs have yet to be studied. 1998).
Regarding clinical opinions about treatment responsiveness, in One potentially useful modification of current practices would
relation to adult offenders it would seem defensible for examiners be to report PCL-R scores with a confidence interval rather than as
to assert that there is no known intervention that has been shown simply a number (Edens & Petrila, 2006). Consistent with Hare’s
to be particularly effective with individuals high in psychopathic (2003, pp. 66 – 67) description of the expected range in which an
traits, although there appears to be some evidence that psychiatric examinee’s “true” score is likely to fall, this procedure would help
patients high in psychopathic traits who can be engaged in treat- the psychometrically unsophisticated understand that such assess-
ment may respond in a positive fashion (Skeem et al., 2002). ment results are subject to a certain degree of imprecision or error,
Practitioners also should be aware, however, that therapist impres- even under the best of circumstances. For example, on the basis of
sions of “positive treatment outcome” should not be assumed to SEMs reported in the manual the 95% confidence interval for a
reflect actual positive treatment outcome—at least one study of sex hypothetical PCL-R score of 25 for a male offender would be
offenders found that psychopathic individuals who were presumed approximately 19 –31.6 Aside from perhaps fostering a more real-
to have benefited from treatment were at heightened risk for istic appraisal of the stability of psychopathy scores among lay-
criminal recidivism (Seto & Barbaree, 1999).5 persons who review such information (e.g., jurors), such score
ranges also might be helpful for countering criticisms when cross-
“Clinical Evaluations of Psychopathy Are Highly
Reliable” 5
Despite this finding, which has been widely cited in the literature,
The Controversy Barbaree (2005) recently published an extended follow-up study on this
sample that found that treatment behavior was unrelated to recidivism.
This position, which is a quite reasonable summary of published 6
In a blatant distortion of the stability of a psychopathy score, I recently
research studies, is a potential source of confusion in applied observed an expert testify to a jury that his PCL-R rating of a defendant

examination suggests that a particular item score for an examinee ation’s (2001) recommendations regarding the use of nonstigma-
is questionable. tizing language (see also Forth, Kosson, & Hare, 2003).

“Psychopaths Are Qualitatively Different From Other Concluding Remarks

The PCL-R is a widely used and valuable instrument, but like
The Controversy any assessment procedure, it can be misused or oversold in various
ways to support numerous objectives. In hindsight, the areas of
It has been argued by some commentators that psychopaths
concern noted above would seem to reflect two general themes
“comprise a discrete natural class of individuals” (Harris, Skilling,
related to potential overreaching on the part of mental health
& Rice, 2001, p. 197). This conclusion is driven in large part by a
experts that are not unique to psychopathy per se. The first is the
widely cited article by Harris, Rice, and Quinsey (1994) in which
longstanding nomothetic-versus-idiographic problem of what
they claimed to identify a latent taxon on the PCL-R in a large
types of conclusions or predictions are appropriate to make about
Canadian sample of forensic patients using complex statistical
individuals on the basis of group-level data. This problem is
procedures first developed by Paul Meehl and his colleagues.
evident in relation to making statements about the stability or
Although perhaps an intuitively appealing position that helps to
treatability of psychopathy as well as regarding the likelihood of
psychologically distance the psychopathic offender from the rest
future violence. Categorical predictions (e.g., “This individual
of society, blanket assertions that psychopaths are a “discrete
will/will not be violent in the future”) based on psychopathy
natural class” are much more questionable than would be gleaned
ratings go well beyond what reasonably can be concluded on the
from the Harris et al. (2001) review. More recent taxometric
basis of the probabilistic nature of this relationship—and the
studies using both the PCL-R and other psychopathy measures
probabilistic relationship itself seems to vary widely from one
have been much more consistent with the argument that psycho-
study to the next.
paths differ from other offenders in degree rather than in kind
The second general theme is the effect of working in an adver-
(Edens, Marcus, Lilienfeld, & Poythress, in press; Guay, Ruscio,
sarial criminal context, where there is a press to move from
Hare, & Knight, 2004; Marcus, John, & Edens, 2004; cf. Skilling,
objective “expert” neutrality toward asserting positions or opinions
Quinsey, & Craig, 2001).
that are most favorable to the side that has retained the clinician,
or at least asserting them in the most favorable way possible
Practice Implications (Shuman & Greenberg, 2003). As such, it is perhaps not entirely
surprising that examiners might be somewhat prone to draw con-
At first glance, taxometric debates may seem to be something of clusions or make assertions that overstate or perhaps misconstrue
an ivory tower topic that is far removed from clinical practice— what is currently known about this particularly troubling person-
particularly given that identifying a particular class of individuals ality disorder. There is ample evidence that the objectivity of
on the basis of even an arbitrarily defined score may be useful for mental health experts already is viewed with considerable skepti-
applied decision-making purposes (e.g., identifying an optimal cism in the legal system (e.g., Otto & Heilbrun, 2002), however,
PCL-R score to differentiate more vs. less violence-prone offend- and drawing definitive conclusions about topics that can easily be
ers). Despite this, I believe there are clear real-world implications demonstrated to be at least very controversial will only undermine
regarding this controversy. For example, referring to an examinee experts’ credibility and give further ammunition to those who
or client as “a psychopath” implies a dichotomous distinction question whether mental health professionals should be involved
between those who “are” and those who “are not” psychopaths, in the legal system (e.g., Hagen, 1997).
despite the markedly inconsistent findings in the available scien- As a final point, I would suggest that some of what drives what
tific studies on this topic. Moreover, many legal decision makers I perceive to be the overzealousness about psychopathy is the
already seem to have either implicitly or explicitly adopted this desire of some professionals to bolster the relevance of clinical
categorical model, reflected in recurrent debates in court cases psychology in forensic and correctional decision making, where its
regarding whether a particular defendant is “diagnosable” as a contributions over the past few decades have been (somewhat
psychopath (i.e., is 30 or above on the PCL-R). As noted earlier, speciously, I would argue) devalued (see, generally, Andrews,
if examiners were to focus attention on confidence intervals rather Bonta, & Hoge, 1990). A PCL-R score is not a simple actuarial
than on raw scores or categorical placements, perhaps decision checklist that can be completed by a correctional technician with
makers would be less preoccupied with the difference between a minimal training. It reflects a complex psychological construct that
score of 29 and a score of 30 on this instrument (Edens & Petrila, is not easily dismissed or explained by most sociological and
2006). criminological models of crime and violence, which typically
Finally, even if psychopathy were ultimately found to be tax- minimize the role of individual differences. But the PCL-R is only
onic, from a semantic perspective, referring to individuals by their one tool in the armamentarium of the forensically proficient cli-
disorder or diagnosis (e.g., “the schizophrenic,” “the retardate”) nician (Otto & Heilbrun, 2002), who should be well versed in its
only encourages others to define such persons by their pathology strengths and limitations. Ten years ago, Hare (1996) asserted that
rather than seeing them as an individual with such a condition. psychopathy was “a clinical construct whose time has come” (p.
Although referring to offenders with an elevated score on the
PCL-R as “highly psychopathic” or “high in psychopathic traits”
may seem less attractive to some, terminology such as “the psy- was highly reliable and that, if 100 other examiners readministered it, 90 of
chopath” is inconsistent with the American Psychological Associ- them would give him the exact same score.

25). Let us hope that in the next decade it does not become, 4 year stability of psychopathic traits in non-referred youth. Behavioral
because of overly enthusiastic misapplication, a clinical construct Sciences and the Law, 21, 713–736.
in need of restraint. Gacono, C. B. (2000). Preface. In C. Gacono (Ed.), The clinical and
forensic assessment of psychopathy: A practitioner’s guide (pp. xv–
xxii). Mahwah, NJ: Erlbaum.
References Gacono, C. B., Nieberding, R. J., Owen, A., Rubel, J., & Bodholdt, R.
(1997). Treating conduct disorder, antisocial, and psychopathic person-
American Psychological Association. (2001). Publication manual of the alities. In J. Ashford, B. D. Sales, & W. H. Reid (Eds.), Treating adult
American Psychological Association (5th ed.). Washington, DC: Author. and juvenile offenders with special needs (pp. 99 –129). Washington,
Andrews, D. A., Bonta, J., & Hoge, R. D. (1990). Classification for DC: American Psychological Association.
effective rehabilitation: Rediscovering psychology. Criminal Justice and Gendreau, P., Goggin, C., & Smith, P. (2002). Is the PCL-R really the
Behavior, 17, 19 –52. “unparalleled” measure of risk? A lesson in knowledge cumulation.
Barbaree, H. E. (2005). Psychopathy, treatment behavior, and recidivism: Criminal Justice and Behavior, 29, 397– 426.
An extended follow-up of Seto and Barbaree. Journal of Interpersonal Guay, J., Ruscio, J., Hare, R., & Knight, R. A. (2004, October). The latent
Violence, 20, 1115–1131. structure of psychopathy: When more is simply more. Paper presented at
Barefoot v. Estelle, 463 U.S. 880 (1983). the 19th Annual Meeting of the Society for Research in Psychopathol-
Cauffman, E., & Skeem, J. (2004, March). The developmental (in)appro- ogy, St. Louis, MO.
priateness of assessing juvenile psychopathy. In J. Skeem (Chair), Con- Guy, L. G., Edens, J. F., Anthony, C., & Douglas, K. S. (2005). Does
temporary issues in juvenile psychopathy. Symposium conducted at the psychopathy predict institutional misconduct among adults? A meta-
biennial meeting of the American Psychology-Law Society, Scottsdale,
analytic investigation. Journal of Consulting and Clinical Psychology,
73, 1056 –1064.
Cleckley, H. (1976). The mask of sanity (5th ed.). St. Louis, MO: Mosby.
Hagen, M. (1997). Whores of the court: The fraud of psychiatric testimony
Cooke, D. J. (1997). The Barlinnie Special Unit: The rise and fall of a
and the rape of American justice. New York: HarperCollins.
therapeutic experiment. In E. Cullen, L. Jones, & R. Woodward (Eds.),
Hare, R. D. (1993). Without conscience: The disturbing world of the
Therapeutic communities for offenders (pp. 101–120). New York:
psychopaths among us. New York: Guilford Press.
Hare, R. D. (1996). Psychopathy: A clinical construct whose time has
Cunningham, M. D., & Reidy, T. J. (2002). Violence risk assessment at
come. Criminal Justice and Behavior, 23, 25–54.
federal capital sentencing: Individualization, generalization, relevance,
Hare, R. D. (1998). The Hare PCL-R: Some issues concerning its use and
and scientific standards. Criminal Justice and Behavior, 29, 512–537.
misuse. Legal and Criminological Psychology, 3, 99 –119.
D’Silva, K., Duggan, C., & McCarthy, L. (2004). Does treatment really
Hare, R. D. (2003). Hare Psychopathy Checklist—revised manual (2nd
make psychopaths worse? A review of the evidence. Journal of Person-
ed.). Toronto, Ontario, Canada: Multi-Health Systems.
ality Disorders, 18, 163–177.
Hare, R. D., Clark, D., Grann, M., & Thornton, D. (2000). Psychopathy and
Dowden, C., & Andrews, D. (2000). Effective correctional treatment and
the predictive validity of the PCL-R: An international perspective.
violent reoffending: A meta-analysis. Canadian Journal of Criminology,
Behavioral Sciences and the Law, 18, 623– 645.
42, 449 – 467.
Edens, J. F. (2001). Misuses of the Hare Psychopathy Checklist–Revised in Hare, R. D., Forth, A. E., & Strachan, K. E. (1992). Psychopathy and crime
court: Two case examples. Journal of Interpersonal Violence, 16, 1082– across the life span. In R. D. Peters, R. J. McMahon, & V. L. Quinsey
1093. (Eds.), Aggression and violence throughout the life span (pp. 285–300).
Edens, J. F., Buffington-Vollum, J. K., Keilen, A., Roskamp, P., & An- Newbury Park, CA: Sage.
thony, C. (2005). Predictions of future dangerousness in capital murder Harris, G., Rice, M., & Cormier, C. (1991). Psychopathy and violent
trials: Is it time to “disinvent the wheel”? Law and Human Behavior, 29, recidivism. Law and Human Behavior, 15, 625– 637.
55– 86. Harris, G., Rice, M., & Cormier, C. (1994). Psychopaths: Is a therapeutic
Edens, J. F., Colwell, L. H., Desforges, D. M., & Fernandez, K. (2005). community therapeutic? Therapeutic Communities, 15, 283–299.
The impact of mental health evidence on support for capital punishment: Harris, G., Rice, M., & Quinsey, V. (1994). Psychopathy as a taxon:
Are defendants labeled psychopathic considered more deserving of Evidence that psychopaths are a discrete class. Journal of Consulting
death? Behavioral Sciences and the Law, 23, 603– 625. and Clinical Psychology, 62, 387–397.
Edens, J. F., Marcus, D. K., Lilienfeld, S. O., & Poythress, N. G. (in press). Harris, G., Skilling, T., & Rice, M. (2001). The construct of psychopathy.
Psychopathic, not psychopath: Taxometric evidence for the dimensional Crime and Justice, 28, 197–264.
structure of psychopathy. Journal of Abnormal Psychology. Hart, S. D., Watt, K. A., & Vincent, G. M. (2002). Commentary on
Edens, J. F., & Petrila, J. (2006). Legal and ethical issues in the assessment Seagrave and Grisso: Impressions of the state of the art. Law and Human
and treatment of psychopathy. In C. Patrick (Ed.), Handbook of psy- Behavior, 26, 241–245.
chopathy (pp. 573–588). New York: Guilford Press. Helfgott, J. B. (1997, March). The popular conception of the psychopath:
Edens, J. F., Petrila, J., & Buffington-Vollum, J. K. (2001). Psychopathy Implications for criminal justice policy and practice. Paper presented at
and the death penalty: Can the Psychopathy Checklist–Revised identify the 34th Annual Meeting of the Academy of Criminal Justice Sciences,
offenders who represent “a continuing threat to society”? Journal of Louisville, KY.
Psychiatry and Law, 29, 433– 481. Hemphill, J. F., & Hare, R. D. (2004). Some misconceptions about the
Edens, J. F., Skeem, J. L., Cruise, K. R., & Cauffman, E. (2001). Assess- Hare PCL-R and risk assessment. Criminal Justice and Behavior, 31,
ment of “juvenile psychopathy” and its association with violence: A 203–243.
critical review. Behavioral Sciences and the Law, 19, 53– 80. Lilienfeld, S. O., & Fowler, K. A. (2006). The self-report assessment of
Elbogen, E., Mercado, C., Scalora, M., & Tomkins, A. (2002). Perceived psychopathy: Problems, pitfalls, and promises. In C. Patrick (Ed.),
relevance of factors for violence risk assessment: A survey of clinicians. Handbook of psychopathy (pp. 107–132). New York: Guilford Press.
International Journal of Forensic Mental Health, 1, 37– 47. Marcus, D. K., John, S., & Edens, J. F. (2004). A taxometric analysis of
Forth, A., Kosson, D., & Hare, R. (2003). Psychopathy Checklist: Youth psychopathic personality. Journal of Abnormal Psychology, 113, 626 –
version. Toronto, Ontario, Canada: Multi-Health Systems. 635.
Frick, P. J., Kimonis, E. R., Dandreaux, D. M., & Farell, J. M. (2003). The Otto, R. K., & Heilbrun, K. (2002). The practice of forensic psychology: A

look toward the future in light of the past. American Psychologist, 57, Skilling, T. A., Quinsey, V. L., & Craig, W. M. (2001). Evidence of a taxon
5–18. underlying serious antisocial behavior in boys. Criminal Justice and
Porter, S., Birt, A. R., & Boer, D. (2001). Investigation of the criminal and Behavior, 28, 450 – 470.
conditional release profiles of Canadian federal offenders as a function U.S. v. Stitt, Transcript, Death Penalty Phase, File No. 2:98, November 2,
of psychopathy and age. Law and Human Behavior, 25, 647– 661. 1998.
R. v. K.S., O.J. No. 3826, LEXIS 4498 (2004). Walters, G. D. (2003). Predicting criminal justice outcomes with the
R. v. LaRue, B.C.D. Crim.J. 14785 (2003). Psychopathy Checklist and Lifestyle Criminality Screening Form. Be-
Rutherford, M., Cacciola, J., Alterman, A., McKay, J., & Cooke, T. (1999). havioral Sciences and the Law, 21, 89 –102.
The 2-year test–retest reliability of the PCL-R in methadone patients. Webster, C. D., Douglas, K. S., Eaves, D., & Hart, S. D. (1997). HCR–20:
Assessment, 6, 285–291.
Assessing risk for violence (Version 2). Vancouver, British Columbia,
Salekin, R. T., Rogers, R., & Sewell, K. W. (1996). A review and
Canada: Simon Fraser University.
meta-analysis of the Psychopathy Checklist and Psychopathy Checklist–
Wong, S., & Hare, R. D. (in press). Program guidelines for the institutional
Revised: Predictive validity of dangerousness. Clinical Psychology:
treatment of violent psychopaths. Toronto, Ontario, Canada: Multi-
Science and Practice, 3, 203–215.
Seto, M. C., & Barbaree, H. E. (1999). Psychopathy, treatment behavior, Health Systems.
and sex offender recidivism. Journal of Interpersonal Violence, 14, Wood, J., Lilienfeld, S., Garb, H., & Nezworski, M. (2000). The Rorschach
1235–1248. test in clinical diagnosis: A critical review, with a backward look at
Shuman, D. W., & Greenberg, S. (2003). The expert witness, the adversary Garfield (1947). Journal of Clinical Psychology, 56, 395– 430.
system, and the voice of reason: Reconciling impartiality and advocacy.
Professional Psychology: Research and Practice, 34, 219 –224.
Skeem, J. L., Monahan, J., & Mulvey, E. P. (2002). Psychopathy, treatment Received September 30, 2004
involvement, and subsequent violence among psychiatric patients. Law Revision received June 14, 2005
and Human Behavior, 26, 577– 603. Accepted July 1, 2005 䡲