The study of serial murder has, unfortunately and unnecessarily, become intertwined with criminal profiling (hereafter, CP). As serial murder has spread across the world, CP has followed closely behind. Often, when there is a conference or a special publication devoted to serial murder, authors submit papers devoted solely to CP (see, e.g. Petee & Jarvis, 2000). Since the U.S. provides more models (examples) of serial murder than any other nation and the FBI is the main force spreading profiling, both here and abroad, American researchers, if not the FBI, need to take some responsibility for evaluating the effectiveness of CP in curtailing serial murder, and in decreasing the number of unsolved homicides. This three-part article reviews sobering efforts in this direction that have not been well publicized.
Only in the past few years have researchers begun to seriously assess the efficacy of CP. Thus the spread of profiling (through both fictional depictions on television, e.g. "Criminal Minds" and through pseudo-scientific and true crime journals and books) has occurred in the absence of any effort to verify its effectiveness - a fact which does not recommend those advancing the technique.
However, empirical psychologists have now looked into the scientific basis for profiling (Snook, Cullen, Bennell, Taylor & Gendreau, 2008). They examined the FBI's form of profiling, which is based on an either/or typology (or classification) into which, the FBI believes, all serial offenses and offenders can be fit. In this binary typology, offenses can be divided into those that are organized (i.e. planned) or disorganized (unplanned) which is determined through analysis of (intangible) evidence at the crime scene. Offenders are either organized (sane, competent), or disorganized (performing poorly or irrationally) based on background characteristics, and lastly, "(c) there is a correspondence between offenses and offenders (i.e., organized offenders commit organized crimes and disorganized offenders commit disorganized crimes)" (pg. 1259).
The belief that there is a correspondence between the type of offense and the background of offenders, such that the latter can be predicted from the former, is present in most, if not all, forms of criminal profiling, along with the use of typologies. (See, for example, Salfati, 1998; Salfati & Canter, 1999; Keppel & Walter, 1999.)
Snook, Cullen et al. (2008) and Snook, Eastwood, Gendreau, Goggin, & Cullen (2007) report that, (1) most studies attempting to fit serial murderers into the FBI's typologies failed to find empirical support for those efforts, (2) typologies failed to provide an adequate theoretical framework, and were based on outmoded theories of personality dispositions that psychologists rejected some 4 decades ago, and (3) there is minimal evidence to show that professional profilers are better able to predict personality or background characteristics from crime scene behavior than are e.g. college students, psychologists or police. In other words, in those areas of human behavior that criminal profilers claim expertise, no such expertise was found. Snook, Cullin et al. (2008) conclude by arguing that CP should not be used as an investigative tool until adequate scientific support has been demonstrated. They further argue that it is up to the FBI and other profilers to take away the secrecy that has thus far shrouded their studies and prove profiling efforts work.
In case the reader still believes that this will eventually happen, the next section continues this evaluation and considers why such a result is highly unlikely.