A Secret Service agent recently asked if I was familiar with a 1992 FBI report that almost half of the killers of law enforcement officers met the criteria for antisocial personality. I replied that I had not seen the report but that the finding did not seem surprising or noteworthy to me. My comment was based on the assumption that the report had used antisocial personality as a synonym for antisocial personality disorder (ASPD), a category listed in the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) and applicable to the majority of criminals.
However, the agent explained that the description of the killers in question indicated to him that they matched the profile of the psychopath defined by the Psychopathy Checklist-Revised (Hare 1991). When I later saw a copy of the FBI report, I realized that he was correct in his assessment and that the report’s findings were indeed noteworthy and chilling, particularly for law enforcement officers.
The killers’ characteristics referred to as antisocial personality in the FBI report were as follows: sense of entitlement, unremorseful, apathetic to others, unconscionable, blameful of others, manipulative and conning, affectively cold, disparate understanding of behavior and socially acceptable behavior, disregardful of social obligations, nonconforming to social norms, irresponsible. These killers were not simply persistently antisocial individuals who met DSM-IV criteria for ASPD; they were psychopaths- remorseless predators who use charm, intimidation and, if necessary, impulsive and cold-blooded violence to attain their ends.
The distinction between psychopathy and ASPD is of considerable significance to the mental health and criminal justice systems. Unfortunately, it is a distinction that is often blurred, not only in the minds of many clinicians but in the latest edition of DSM-IV.
Source of the Problem
Traditionally, affective and interpersonal traits such as egocentricity, deceit, shallow affect, manipulativeness, selfishness, and lack of empathy, guilt or remorse, have played a central role in the conceptualization and diagnosis of psychopathy (Cleckley; Hare 1993; in press); Widiger and Corbitt). In 1980 this tradition was broken with the publication of DSM-III. Psychopathy-renamed antisocial personality disorder- was now defined by persistent violations of social norms, including lying, stealing, truancy, inconsistent work behavior and traffic arrests.
Among the reasons given for this dramatic shift away from the use of clinical inferences were that personality traits are difficult to measure reliably, and that it is easier to agree on the behaviors that typify a disorder than on the reasons why they occur. The result was a diagnostic category with good reliability but dubious validity, a category that lacked congruence with other, well-established conceptions of psychopathy. This “construct drift” was not intentional but rather the unforeseen result of reliance on a fixed set of behavioral indicators that simply did not provide adequate coverage of the construct they were designed to measure.
The problems with DSM-III and its 1987 revision (DSM-III-R) were widely discussed in the clinical and research literature (Widiger and Corbitt). Much of the debate concerned the absence of personality traits in the diagnosis of ASPD, an omission that allowed antisocial individuals with completely different personalities, attitudes and motivations to share the same diagnosis. At the same time, there was mounting evidence that the criteria for ASPD defined a disorder that was more artifactual than “real” (Livesley and Schroeder).
Coincident with the publication of DSM-III in 1980, I presented some preliminary findings on efforts to provide researchers with an operational definition of psychopathy in offender populations (Hare 1980). During the next decade those early efforts evolved into the Hare Psychopathy Checklist-Revised (PCL-R) (Hare 1991), a 20-item construct rating scale that uses a semi-structured interview, case-history information and specific diagnostic criteria for each item to provide a reliable and valid estimate of the degree to which an offender or forensic psychiatric patient matches the traditional (prototypical) conception of the psychopath (Fulero; Stone). Each item is scored on a 3-point scale (0, 1, 2) according to the extent to which it applies to the individual. The total score can range from 0 to 40, with between 15 percent and 20 percent of offenders receiving a score of at least 30, the cutoff for a research diagnosis of psychopathy. To put this into context, the mean scores for offenders in general and for noncriminals typically are around 22 and 5, respectively.
A 12-item version of the PCL-R was developed for use in the MacArthur Foundation study on the prediction of violence in the mentally disordered (Hart and others 1994). Published in 1995 as the Hare Psychopathy Checklist: Screening Version (PCL:SV) by Hart and colleagues, it is highly correlated with the PCL-R and is used both to screen for psychopathy in forensic populations and as a stand-alone instrument for the assessment of psychopathy in noncriminal populations. The PCL:SV formed the basis for the psychopathic personality disorder items used in the DSM-IV field trial for ASPD.
The items fall into two clusters: One cluster, referred to as Factor 1, reflects core interpersonal and affective characteristics; the other cluster, Factor 2, consists of items that reflect a socially deviant and nomadic lifestyle. The similarity between these factors and the behaviors and characteristics described above in the FBI report are obvious.
Most psychopaths (with the exception of those who somehow manage to plow their way through life without coming into formal or prolonged contact with the criminal justice system) meet the criteria for ASPD, but most individuals with ASPD are not psychopaths. Further, ASPD is very common in criminal populations, and those with the disorder are heterogeneous with respect to personality, attitudes and motivations for engaging in criminal behavior.
As a result, a diagnosis of ASPD has limited utility for making differential predictions of institutional adjustment, response to treatment, and behavior following release from prison. In contrast, a high PCL-R score depends as much on inferred personality traits as on antisocial behaviors, andwhen used alone or in conjunction with other variableshas considerable predictive validity with respect to treatment outcome, institutional adjustment, recidivism and violence (Hare 1991; Harris and others; Hart and Hare, in press).
For example, several studies have found that psychopathic offenders or forensic psychiatric patients (as defined by the PCL-R) are as much as three or four times more likely to violently reoffend following release from custody than are nonpsychopathic offenders or patients. ASPD, on the other hand, has relatively little predictive power, at least with forensic populations (Hart and Hare, in press).
It might be argued that a diagnosis of ASPD is useful in civil psychiatric settings, particularly as a general risk factor for substance abuse (Leal and others). Even here, however, psychopathy may be more important than ASPD in understanding substance abuse (Alterman and colleagues; Cacciola and others).