Assessing Message Affect in Crisis Negotiations An Exploratory Study



    1. Randall Rogan (Ph.D., Michigan State University, 1990) is an assistant professor in the Department of Speech Communication at Wake Forest University and incident adviser to the Winston-Salem Police Department hostage negotiation unit.
    Search for more papers by this author

    1. Mitchell Hammer (Ph.D., University of Minnesota, 1982) is an associate professor in the School of International Service at the American University in Washington, DC.
    Search for more papers by this author

  • The authors thank the Special Operations and Research Unit of the FBI Academy for providing the tapes of the incidents used in this investigation. The assistance of the FBI has been invaluable to the completion of this project.


In this study, patterns of perpetrator and negotiator message affect behavior in three actual crisis negotiation incidents are examined. The unit of analysis is the uninterrupted talking turn. Overall findings offer reliable support for the coding schemes employed in the study to measure message affect. Using the Michigan State Police four-stage (Introduction and Establishing Contact, Relationship Building, Problem Negotiation, and Resolution) strategic model of negotiation as a template, the study points to some preliminary results concerning patterns of emotional arousal during different phases of crisis negotiation. In all three incidents, perpetrators experienced an increase in message affect during the Establishing Contact stage of negotiation, followed by a general decrease in affect as the interaction moved to Relationship Building. However, when negotiators attempted to shift into the Problem-Solving and Resolution stages, perpetrator message affect varied by incident type and outcome. In the one resolved incident (Mental/Emotional Instability), perpetrator message affect increased in positiveness through the Resolution stage. For the Suicide incident, perpetrator affect became increasingly more negative up through the fatal end of the interaction. The Domestic incident was marked by notable variance in perpetrator affect up to the point where the perpetrator reneged on the plan to surrender. These findings offer initial insight into message affect patterns in successfully and unsuccessfully resolved crisis incidents.