Prevention & Treatment, Volume 3, Article 21, posted June 2, 2000
Copyright 2000 by the American Psychological Association

IN MEMORY OF NEIL S. JACOBSON

Jacobson's Domestic Violence Research: Empirical, Social, Political Contributions

Gayla Margolin
University of Southern California

Correspondence concerning this article should be addressed to Gayla Margolin, Department of Psychology, SGM 930, University of Southern California, Los Angeles, California 90089-1061.
E-mail: margolin@rcf.usc.edu

Neil liked nothing more than to broaden and intensify discussion of important issues in clinical psychology, which he most definitely accomplished with his writings and presentations on domestic violence. Over the past two decades, the topic of domestic violence has been addressed by two diverse camps—the advocacy community and clinical researchers. Neil was unique in the extent to which he embraced and integrated both of these perspectives. By his own admission, his focus on domestic violence sprung "not from couple therapy or research but, rather, from an interest in political issues as they relate to marriage (Jacobson, 1989) and, more specifically, in interpretations of marital dynamics that were informed by feminist perspectives" (Jacobson, 1994b, p. 81).

The article by Jacobson and colleagues presented here (Jacobson et al., 1994) is a good representation of the varied perspectives and methodologies that Neil brought to his work on domestic violence. This article is noteworthy in its blending of feminist and contextual values, which often are viewed as contradictory (Jacobson, 1994b). Feminist values are evidenced in the assumptions that domestic violence is not simply physical aggression toward the spouse but is a form of men's control over women, and that men's and women's violence in relationships serve very different functions. The contextual perspective is reflected in the assumption that it is necessary to study environmental and relationship factors to identify variables associated with increases or decreases in men's battering.

Methodologically, Jacobson et al. (1994) used multiple forms of data collection to examine wide-ranging phenomena surrounding violence in couples. The task of having participants provide descriptions of violent and nonviolent arguments allowed for act-by-act sequential coding of affect and content behaviors. These data were used to examine the function and sequential patterns of behavior in violent and nonviolent arguments. The laboratory-based observations of nonviolent arguments were coded for specific types of affective responses and were also used to measure cardiovascular arousal, thereby examining both interpersonal and intrapersonal variables associated with battering. Although other investigators have reported observational data in violent versus nonviolent relationships (e.g., Margolin, John, & Gleberman, 1988), the careful delineation of types and seriousness of anger is a unique feature of this study and illustrates differences between low-level anger and more provocative anger, exhibited through belligerence and contempt. This delineation of provocative anger also proves important at the two-year follow-up, predicting which violent couples are no longer together and which men maintain versus reduce their level of serious violence (Jacobson, Gottman, Gortner, Berns, & Shortt, 1996).

In Jacobson et al. (1994), the questions and hypotheses are informed by a feminist perspective and attend to the realities of gender and the differences between victim and perpetrator. The rigorous scientific underpinnings provide controlled, systematic comparisons to either confirm or disconfirm anticipated gender differences and to determine whether clinical observations about domestic violence can indeed be attributed to violence. For the most part, the data reported by Jacobson et al. draw sharp distinctions between male and female violence. Approximately half of the female participants in this study engaged in physical aggression that would qualify them for the domestic violence group, and most all reported at least low-level violence. Accordingly, these investigators were able to examine empirically whether husbands' and wives' violence serve different functions and are elicited under different circumstances. Whereas husbands' violence produced fear in wives, wives' violence had no such effect on husbands. Whereas husbands' violence occurred in response to a variety of wives' behaviors, wives' violence was predictably a response to husbands' emotional and physical abuse. Moreover, once husbands' violence began, wives could do nothing to stop their husbands' attack. These data provide important empirical support for the hypothesis that husbands' violence is an effective form of control over their wives.

However, using adequate controls also led to the disconfirmation of certain assumptions about batterers and victims. Jacobson et al. (1994) showed that coding belligerence and contempt in violent versus nonviolent men could lead to the conclusion that those characteristics are a defining feature of male batterers. Yet, contrary to the hypothesis, wives in domestically violent relations are as belligerent, contemptuous, and angry as are their husbands. Jacobson et al. included a nonviolent group that was matched with the violent group on marital satisfaction in order to attribute group differences to violence and not marital dissatisfaction. Unexpectedly, the comparison between wives in violent versus nonviolent relationships failed to find significant differences on any coded behaviors other than anger.

One of Neil's most important contributions to the domestic violence literature was clarifying the role of interactional data vis-à-vis issues of causation and responsibility for battering (Jacobson, 1994a, 1994b; Jacobson, Gottman, & Shortt, 1995). In defending a contextual perspective, Jacobson (1994b) wrote,

Most important behaviors are best understood by examining their covariation with factors in the environment; and whether or not I understand a phenomenon is determined by whether I can predict its occurrence/nonoccurrence, and influence its frequency, intensity, or duration. (p. 82)

The study presented here introduces physiological data that were elaborated on in later studies. The Jacobson et al. (1994) comparisons show that wives of the violent husbands—compared to control wives—exhibited more cardiovascular arousal during conflict discussions, whereas no differences were found for husbands. Gottman, Jacobson, Rushe, and colleagues (1995) provided a departure from other proposed typologies of batterers (e.g., Holtzworth-Munroe & Stuart, 1994) when they indicated that heart rate reactivity can serve as a basis for distinguishing among types of batterers. They differentiated Type 1 batterers (approximately 20% of their sample), who showed decelerating heart rates during the course of conflict discussions, from Type 2 batterers (80%), who showed heart rate acceleration. Type 1 men, additionally, were characterized by violence toward others, as well as antisocial and aggressive–sadistic personality disorders. During nonviolent conflictual discussions, Type 1 men, as compared to Type 2 men, displayed more belligerence and contempt, whereas their wives displayed more sadness and less anger.

The research findings of Jacobson et al. (1994) and related articles are summarized for the public in the book that Neil and John Gottman (1998) coauthored, When Men Batter Women: New Insights Into Ending Abusive Relationships. This book speaks directly to battered women about the dynamics of battering, the choices they face, the resilience and heroism of women in battering relationships, and the availability and importance of community support. Although Neil did not conduct intervention studies on domestic violence, he took every available opportunity to inform battered women about how to seek help in their community. He also worked hard to educate marital therapists to be aware of violence and not to view violence as just another marital problem. Neil's comments about batterers, particularly in the Type 1 group, generally reflected pessimism about their responsiveness to psychotherapy.

Neil's research on domestic violence brings to light the multidimensional, complex nature of this major societal problem. He and his colleagues applied highly sophisticated observational and psychophysiological technologies that they "might find out some things that would provide hope to battered women, and even some clues as to how to reduce battering" (Jacobson & Gottman, 1998, p. 32). Jacobson's synthesis of scientific, humanitarian, and feminist values has provided a unique understanding of this problem and important information for generating solutions.

References