Résumé
Keppel and Walter’s (1999) classification system for serial sexual homicide is commonly used as the basis for generating offender profiles of unknown offenders despite the fact that it has yet to be empirically validated. Keppel and Walter’s model assumes that serial sexual homicides and serial sexual homicide offenders can be classified into four categories - power-assertive, power-reassurance, anger-retaliation, and anger-excitation -according to the degree of anger and power exhibited in the criminal and non-criminal context. As a first step in assessing the validity of this model, the crimes of 53 serial sexual murderers were examined to determine if the categories proposed by Keppel and Walter could be identified. Although no support was found for this classification system, support was found for an alternative model - the FBI’s organized/disorganized dichotomy. The second phase of the study tested the organized/disorganized classification system to determine if evidence could be found for the three basic assumptions underlying the profiling process, namely, the temporal stability of criminal behaviour across crimes, the cross-situational consistency of behaviours between an offender’s criminal and non-criminal life, and the homology of background characteristics for offenders who display similar behaviours (compared to one another) in their criminal life. Consistent with previous literature, moderate support was found for temporal stability, whereas lower levels of support were found for cross-situational consistency and homology. These results are discussed in terms of their theoretical and practical implications.