Previous research on subtypes of batterers has revealed at least two distinct types of batterers. One group (Type 1) demonstrates suppressed physiological responding during conflicts with their wives, tends to use violence in nonintimate relationships and manifests Millon Clinical Multiaxial Inventory (MCMI-II) scale elevations on the Antisocial and Aggressive-Sadistic scales. The second group (Type 2) manifests violence in the intimate relationship only and reports dysphoria. The current study extends our knowledge of these two groups by using a cluster analysis to assess personality disorder and relating the results to each group’s attachment style, anger, trauma scores, and scores on a self-report of Borderline Personality Organization (BPO). An instrumental group (Type 1) showed an Antisocial-Narcissistic-Aggressive profile on the MCMI-II and reported more severe physical violence. An impulsive group (Type 2) showed a mixed profile on the MCMI-II with Passive-Aggressive, Borderline, and Avoidant elevations, high scores on a self-report of BPO, higher chronic anger, and Fearful attachment. Both types of abusive men reported a Preoccupied attachment style, but only the Impulsive men reported an accompanying Fearful attachment style.

A review of the experimental literature on wife assault causation indicates that differing “profiles” of wife assaulters have been developed by different research strategies. Profiles based on interviews with victims suggest a tyrannical, personality-disordered type of wife assaulter. Clinical assessments reveal several profiles, only one of which is consistent with this view. Other types of wife assaulters are dependent and unassertive. By comparing the emerging data on wife assaulters with the initial clinical descriptions, an assessment is made of the contribution of empirical studies to date. While the potential for a major contribution to our understanding of wife assaulters exists through use of systematic empirical methods, it is concluded that this potential has not yet been fulfilled. Some suggestions are made for future research strategies to improve empirical capabilities in furthering this understanding, including a more thorough assessment of early trauma as a major causative factor affecting a large subcategory of wife assaulters and greater attention to the self-selection of wife assault samples.

Previous evaluations of wife assault treatment outcome have focused generically on whether groups “succeed” or not without a clear criterion of what constituted success. The present study examines the question for whom groups generate the greatest reduction in post-treatment abuse and for whom they work least well. It was found that certain types of personality disordered men had the worst post-treatment prognosis. Specifically, men with high scores on borderline personality, antisocial personality, and avoidant personality fared least well after treatment. However, taken as a generic group, men in treatment had significantly reduced post-treatment abusiveness whether reported by themselves or their wives.

Shame-proneness has been found to be related to anger arousal and a tendency to externalize attributions for one’s own behavior, both common features of men who assault their wives. The present study examined a potential origin of a shame-prone style by analysing reports of shaming experiences by ones’ parents as reported by a population of assaultive males. Significant relationships were found for recollections of shaming actions by parents on adult anger, abusiveness (as reported by the men’s wives), and a constellation of personality variables related to abusiveness in prior research. These associations maintained even after corrections were made for response sets such as social desirability. These shaming actions were largely comprised of recollections of parental punishment that were public, random, or global. The role of shame experiences in disturbances of self-identity and rage is discussed.

This study represents an attempt to assess the effectiveness of court-mandated treatment for wife assault. A quasi-experimental design examined post-conviction recidivism rates for men convicted of wife assault. Fifty men who completed a 16-week treatment program had a 4% recidivism rate for a posttreatment period of up to 3 years. A comparable group who were not treated had a 40% recidivism rate in the same period. Hence, the “success” rate of treatment was 36% according to police records (Rosenthal, 1983). Straus Conflict Tactics Scale scores reported both by the treated men and their wives demonstrated significant posttreatment decreases from pretreatment levels. Treated husbands’ average annual use of severe violence dropped from 10.6 to 1.7 times per year (p <0.01). Eighty-four percent of wives reported no posttreatment violence. Rates of verbal aggression also dropped significantly from pretreatment levels. Comparison with CTS scores of a group of men who were arrested but not treated for wife assault (Jaffe, Wolfe, Telford, & Austin 1986) revealed significant decreases in the use of Physical Aggression subscale tactics (as reported by their wives) as a result of treatment. Interpretative difficulties with the quasi-experimental design used in this study are discussed and a randomized design with appropriate psychological assessment of subjects is recommended.

An empirical test of traumatic bonding theory, the notion that strong emotional attachments are formed by intermittent abuse, is reported. In-depth assessments (interviews plus questionnaires) were conducted on 75 women who had recently left abusive relationships (50 where physical violence had occurred). The study found support for the effect of relationship dynamic factors such as extremity of intermittent maltreatment and power differentials on long-term felt attachment for a former partner, experienced trauma symptoms, and self-esteem, immediately after separation from an abusive partner and again after a six month interim. All three of these measures were significantly intercorrelated within each time period. Each measure at Time 1 correlated significantly with each corresponding measure at Time 2. After six months attachment had decreased by about 27%. Relationship variables (total abuse, intermittency of abuse and power differentials) accounted for 55% of the variance in the attachment measure at Time 2 indicating prolonged effects of abuse suffered in the relationship.

Wife assaulters attending a treatment group and women who had just exited an abusive relationship were asked to report on the extent of physical violence and emotional abuse in their relationship. Measures of socially desirable responding (SDR) were administered to both groups. Wife assaulters' self-reports of physical abuse correlated negatively with one SDR measure (the Balanced Inventory of Desirable Responding) but not another (the Marlowe-Crowne Social Desirability Scale); emotional abuse correlated negatively with both measures. Although physical abuse was primarily related to impression management, psychological abuse was affected by both impression management and self-deception aspects of SDR. Wife assaulters' reports of their own anger also correlated negatively with SDR. Both self-deception and impression management appear to contribute to underreporting of anger. Finally, abuse victims' reports of both physical and emotional abuse were unrelated to SDR.

Research has consistently found that partner violence, defined as physical abuse between married, cohabitating, or dating partners, is not the only type of abuse with long-term deleterious effects on victims. Male and female victims alike report that emotional abuse, along with controlling behaviors, are often as or more traumatic. Existing instruments used to measure emotional abuse and control have either been limited to male-perpetrated behaviors, as conceived in the well-known Duluth “Power and Control” wheel, or field tested on dating or general population samples. This study discusses the genesis and evolution of a gender-inclusive instrument, the Controlling and Abusive Tactics (CAT) Questionnaire, which was field tested on males and females with both a clinical and general population sample. For perpetration, a preliminary comparison across gender found no significant differences across gender for the great majority of items, with women reporting significantly higher rates on 9 items, and men reporting significantly higher rates on 6 items. Women reported higher rates of received abuse than men on 28 of 30 items in which gender differences were found to be significant, but both males and females reported higher victimization than perpetration rates on all items. Exploratory and confirmatory factor analyses resulted in the CAT-2, a valid and reliable instrument appropriate for clinical use by treatment providers as well as for research purposes.