Inferring the Characteristics of an Arsonist From Crime Scene Actions: A Case Study in Offender Profiling

• NCJ Number: 200618
• Journal: International Journal of Police Science and Management Volume: 5 Issue: 1 Dated: Spring 2003 Pages: 1-15
• Author(s): Pekka Santtila; Helina Hakkanen; Katarina Fritzon
• Date Published: 2003
• Length: 15 pages
• Publication Format: Article
• Publication Type: Case Study
• Language: English
• Country: United States of America

Annotation

This article uses a framework of offender profiling to infer the characteristics of a serial arsonist.

Abstract

Variations in arson behavior could be distinguished along two facets. The first facet is a distinction between person-oriented arsons and object-oriented arsons. The second facet differentiated between actions relating to the motivational category underlying the act, either expressive or instrumental. The two facets give rise to four distinct themes of arson: (1) an expressive arson directed at an emotionally significant person; (2) an instrumental arson directed at person with the attack directed outwards against another person; (3) an expressive arson directed at an object; and (4) an instrumental arson directed at an object reflecting an attempt to obtain benefits. This framework was applied to a suspected series of arsons that took place in Finland. The first step was to determine the type of arson the offender’s behavior most closely fitted. The next step was to see which of the four themes of arson in the model the identified crime scene actions would seem to fit best. Based on the model, it was suggested that the offender would likely be male, over 30 years old, had previously set fires, was psychologically disturbed, the recipient of welfare benefits, and did not have a wide social network. The results suggest that there is utility in using psychological models of both the relationship between crime scene actions and offender characteristics in arson and of offender spatial behavior. The characteristics predicted by the model were in good accordance with the actual characteristics of the suspect in the case. It was possible to provide good estimates of the home location of the suspect based on the crime locations even if this prediction was damaged by the introduction of faulty input into the profiling process. The psychological models proved useful even if they had been developed in another society. 2 figures, 2 tables, 22 references