THE UTILITY OF DETERRENCE-BASED SANCTIONS IN THE PREVENTION OF VIOLENCE AGAINST ABORTION PROVIDERS: TESTING A BLENDED MODEL OF DETERRENCE AND BACKLASH

The American abortion debate has existed since the early 19th century; however, the previous four decades have born witness to fundamental changes within the abortion opposition movement. Beginning in the 1970s, activists started to focus more of their attention on abortion providers. Soon thereafter, the traditionally peaceful protest activities of the activist movement began to share space with acts of harassment, arson, bombings, assaults, and even assassination. Today, abortion provider-related crime has become an unwelcome staple within the broader pro-life movement. In an attempt to prevent future attacks, state and federal legislatures have enacted a series of protection laws designed to raise the penalties associated with crimes against abortion providers. Despite the recent proliferation of these laws, their impact on abortion provider-related crime has seldom been the subject of rigorous empirical research. In this dissertation, I aim to address this shortcoming by using zero-inflated negative binomial regression modeling to present the first longitudinal test of the relationship between protection laws and abortion provider-related crime using incident-level data from 1975 to 2008, collected during a year-long research project at The National Consortium for the Study of Terrorism and Responses to Terrorism (START). The results of this study offer considerable support for the backlash hypothesis and the notion that traditional deterrence-based policy is often demonstrably unsuccessful in the prevention of this particular type of crime. Additionally, the findings suggest that not all protection laws are created equal with respect to their impact on crime. While state laws prohibiting minor forms of anti-abortion crime are shown to produce a backlash effect for crimes of harassment and vandalism, other types of state protection laws were shown to have no effect on crime whatsoever. Furthermore, the presence of the highly visible FACE Act is shown to generate similar increases for both major and minor crime types