There are basically four types of batterer intervention programs: same-sex group therapy (for example, a group of males sent to therapy by the criminal justice system), couples therapy (which focuses on the interaction between the two members of the couple), ‘‘psychoeducational’’ groups (again, mandated by the criminal justice system but having a different focus than treatment groups), and intimate abuse circles (an innovative form of restorative justice that involves public apology to a small group and/or apology to the victim) (Mills 2003; Strang and Braithwaite 2002). Psychoeducational groups do not view intimate partner violence as having psychological causes but as being a case of male power and control (Pense and Paymar 1993) that requires attitude adjustment. For this reason, this approach does not refer to ‘‘treatment’’ but rather to intervention and is designed with male perpetrators in heterosexual relationships in mind. It is legally required in many states (Tolman 2001), although many researchers have complained that the approach is not informed by research on perpetrators, that is, it does not have a complex picture of the subtypes of perpetrators, even within a heterosexual male group (see Maiuro et al. 2001; Dutton and Sonkin 2003; Hamel 2005). For example, the notion that ‘‘attitudes’’ drive violence is naive. Large survey studies (e.g., Simon et al. 2003) find that only 2.1 percent of males in the United States agree with the statement ‘‘A man is justified in using violence to keep his mate in line.’’ Even studies on male perpetrators obtain mixed results on whether attitudes predict use of intimate partner violence (see Dutton, in press). It may well be that attitudes are changed by violent men to be more consistent with their behavior (Bem 1972).
The Duluth Domestic Abuse Intervention Project (DAIP) designed an intervention program to be applied to men who had assaulted their female partners but who were not going to receive jail time. The objective of the program was to ensure the safety of the women victims (i.e., protection from recidivist violence) by ‘‘holding the offenders accountable’’ and by placing the onus of intervention on the community to ensure the woman’s safety. The curriculum of the Duluth model was developed by a ‘‘small group of activists in the battered woman’s movement’’ (p. xiii) and was designed to be used by paraprofessionals in court-mandated groups. It is now one of the most commonly used court-sanctioned interventions for men convicted and having mandatory treatment conditions placed on their probations. This is true in many U.S. states and Canadian provinces. The curriculum of the model stresses that violence is used as a form of ‘‘power and control,’’ and a ‘‘Power and Control Wheel’’ has become a famous insignia of the program. This wheel depicts various forms of abuse (physical, financial, sexual, emotional) as emanating from a need in the abuser to have power over the abuse victim. Also, the need for power and control is seen as being an exclusively male problem. As the authors put it, ‘‘[Men] are socialized to be dominant and women to be subordinate’’ (p. 5). Hence, the ‘‘educational’’ aspect of the program deals with male privilege that exists in patriarchal structures such as those in place in North American countries. The DAIP view of female violence is that it is always self defensive. ‘‘Women often kick, scratch and bite the men who beat them, but that does not constitute mutual battering’’ (p. 5). Male battering stems from beliefs which are themselves the product of socialization. These include the beliefs that the man should be the boss in the family; that anger causes violence; that women are manipulative; that women think of men as paychecks; that if a man is hurt, it is natural for him to hurt back; that smashing things isn’t abusive; that ‘‘women’s libbers’’ hate men; that women want to be dominated by men; that men batter because they are insecure; that a man has the right to choose his partner’s friends and associates; and that a man can’t change if the woman won’t (pp. 7–13). According to the manual, the basis for these beliefs came from a sample of five battered women and four men who had completed the Duluth program.
Outcome studies, which measure recidivism (the success or failure of a treatment after completion, usually assessed either by new police reports or by interviews with the previous victim), have been carried out on the Duluth model. Four separate studies (Davis, Taylor, and Maxwell 1998; Feder and Forde 1999; Levesque 1998; and Shepard 1992) essentially found that men completing Duluth treatment were just as likely as untreated control subjects to reoffend. When treatment ends, violence returns. It seems that clients privately reject the Duluth model’s proposals while publicly realizing that they have to comply with the system. Furthermore, dropout rates are very high for such programs, ranging from 40 to 80 percent (McCloskey, Sitaker, Grigsby, and Malloy 2003). One reason may be that such programs do not form a ‘‘therapeutic bond’’ with the client and can be highly judgmental, promoting a philosophy that the client does not see as fitting his situation.
The single most predictive factor for successful therapeutic outcome (realizing that the Duluth model is not therapy but required of many mental health practitioners) is the therapeutic relationship (see, for example, Luborsky 1984; Schore 2003). However, it becomes extremely difficult to form a positive relationship when the therapist is required to disbelieve clients’ reports of acts of violence by the partner; indeed, therapists can lose their certification with probation if they don’t confront their clients enough and tell them that they have a ‘‘power entitlement’’ when the clients feel powerless in the world, and are considered enabling or manipulated when they advocate for their clients’ continued treatment.
One must balance confrontation with support, belief, and caring in order to develop a solid therapeutic alliance. Building a therapeutic alliance without colluding with dangerous acting-out behaviors is one of the greatest challenges facing treatment providers working with domestic violence perpetrators. Because so many of these individuals experienced abuse by authority figures, the process of building a trusting relationship is particularly difficult.
According to Lester Luborsky of the Penn Psychotherapy Project, the therapeutic alliance may be defined as ‘‘that point in the therapeutic relationship when the client on one hand elevates the therapist to a position of authority, but on the other hand believes that this power and authority is shared between them, that there is a deep sense of collaboration and participation in the process. In this way a positive attachment develops between the client and the therapist’’ (Luborsky 1984).
Cognitive-behavioral treatment (CBT) of intimate partner violence is based on the assumption that beliefs or cognitions about violence and its causes sustain a habit of intimate violence and that by challenging and changing these beliefs, abusive behavior can also be stopped. CBT is implemented in a same-sex group with one or two therapists. Treatment typically lasts for sixteen to fifty-two weeks on a once-a-week basis. The treatment is far broader in its targeting of abusive beliefs than is psychoeducational intervention. Topics covered in CBT groups include the following:
Focus on the unacceptability of abuse:
- Confrontation of beliefs and behavioral choices
- Emphasis on attitudes and choices
Generation of client agreement with the unacceptability of abuse:
- Generation of a ‘‘Violence Contract’’ (getting the clients to write out their personal violence policy—the conditions under which they believe the use of violence is acceptable)
- Generation of commitment to therapy
- Emotional labeling
- Anger management (including keeping anger diaries)
- Self-soothing (stress reduction skills)
- Redirecting power needs
- Assertiveness awareness
Focus on specific ‘‘problem’’ emotions:
- Dealing with anger, jealousy, anxiety, depression
- Attitudes toward the use of violence
- Attitudes toward women
- Violence potential awareness
Managing contact with partner:
- Crisis strategies
Connection of learned patterns in family of origin to present dysfunctional action patterns
Therapists typically attempt to confront abusive behaviors while forming and maintaining a therapeutic bond (sportive milieu) with the client.
Dutton (in press) has developed a ‘‘blended CBT’’ model that expands the targets of CBT to include identity disturbances (called borderline personality organization), trauma, substance abuse, and insecure attachment. Research had shown that all four are risk factors for abusiveness, what Dutton (2003) called the associated features of abuse. As of this writing, no evaluation has been done of programs utilizing this expanded focus.
Evaluations of earlier CBT models showed the following. Babcock, Green, and Robie (2004) conducted a meta-analytic study of twenty-two studies of treatment outcome. (Meta-analysis combines several or all known studies to arrive at an overall evaluation.) For Duluth treatment, the effect size (differences in success of treatment groups and control groups in remaining violence-free after treatment) was .19. (An effect size of .20 is considered small, of .40 moderate, and of .75 large.) Comparisons between CBT and Duluth were not significant. However, ‘‘pure’’ Duluth models were hard to find; as the authors of this study state, ‘‘modern batterer groups tend to mix different theoretical approaches to treatment, combining feminist theory of power and control as well as specific interventions that deal with anger control, stress management and improved communication skill’’ (p. 1045).
The effect size of .34 for most therapeutic outcomes, also reported by Babcock, Green, and Robie (2004), is less than optimal. The average effect size in psychotherapy studies is .85, but it is substantially lower for court-mandated treatment. By standards of court-mandated client populations, however, this is an average result. By expanding the focus of treatment in a blended model, this outcome may improve.
Several studies have found couples therapy effective with violent couples (Brannen 1994; Heyman and Schlee 2003; Klein 1991; O’Leary, Heyman, and Neidig 1999). Obviously the form of treatment is dictated by an assessment of violence levels and danger but to rule it out a priori, as the Duluth model does, operates against treatment efficacy.
The decision regarding whether an individual or a couples approach to therapy is best may depend on the client. A partner who has a history of violence in several relationships may be a conflict generator capable of creating the system pattern in the current relationship, as observed by the systems therapist. Certainly an ‘‘abusive personality’’ requires extensive therapeutic work at an individual level before couples treatment seems viable.
Also, as some therapists have shown (Richter 1974), an individual is capable of generating entire interaction patterns within a family on the basis of his or her own pathology. Richter describes how a paranoid personality who holds power in a family can generate a shared paranoia in the entire family system. Men with abusive personalities, one may suspect, are conflict generators in all their intimate relationships, regardless of the personality or style of their female partner. Of course, such men may also pick women with their own backgrounds of abuse victimization and personality disorders. Therefore, obtaining detailed social histories of clients and their partners is recommended prior to embarking on a systems approach. If a male batterer has a history of violence with women that predates his current relationship, or strong indicators of an abusive personality, couples treatment may not be advisable. Where the female feels threatened by the man’s violence potential or where violence is still recent, couples therapy might be delayed until the man has successfully completed an anger management program and has been violence free for a lengthy period. In general, where the violence and conflict seem specific to the present relationship, couples treatment may be more useful after the man’s anger treatment.
Neidig and Friedman (1984) begin their description of their couples treatment program by stating that ‘‘abusive behavior is a relationship issue but it is ultimately the responsibility of the male to control physical violence.’’ Their view is that approaches which attribute total responsibility to either party lead to blaming, which compounds the problem. According to these authors, it does so by beginning a chain of retributional strategies by the victim and the aggressor whereby each tries to ‘‘get even’’ for the other’s most recent transgression. A systems approach avoids blaming by getting couples to think of the causes of violence from a circular feedback perspective rather than a linear one. This leads to ‘‘constructive interventions in the escalating process’’ which permit each partner to accept a portion of the responsibility. Having said that, however, Neidig and Friedman assign ‘‘ultimate responsibility to the male for controlling violence’’ [emphasis added], as a recognition that both parties are not equal in physical strength. If a man is responsible for his violence, then why is he not to blame if he acts violently? One answer may be that his violence occurred in a state of high arousal when he perceived no alternatives to the actions he took. Therapeutically, a couples approach and an individual approach have a fundamental disagreement: The couples approach tries to reduce blame, and the individual approach tries to increase responsibility.
Cascardi and Vivian (1995) found that in the majority of couples clients seeking marital therapy, both partners engaged in aggressive acts, though the woman got the worst of it. Vivian and Langhinrichsen-Rohling (1994) classified couples seeking therapy as (1) mild bidirectional, in which about 50 percent report low-level aggression (pushes, slaps, grabs) committed by both husband and wife, (2) moderate, and (3) severe wife victimization, in which 30–40 percent report high levels of wife victimization and much lower levels of husband victimization. This leads to the same question posed by Stets and Straus (1992): What happens to violent couples in which the female is the predominant aggressor? These couples do not appear to seek marital therapy. Interestingly, only a small percentage (6–14 percent) of women seeking marital therapy report physical violence as a problem, despite reports from the Conflict Tactics Scale (CTS) revealing higher levels of physical aggression in the marriages.
Heyman and Schlee (2003) assessed for levels of aggression prior to their treatment program and found that very few couples reported severe levels of aggression (p. 145). When someone was injured or fearful or when the husband was in denial, the couple was screened out. They did not comment on wives in denial. Post-treatment assessment revealed significant drops in aggression and increases in reported marital adjustment by both parties. The reduction in aggression was still significantly lower than its pre-treatment level one year after cessation of treatment. Complete cessation was found among 26 percent of the couples one year later. Additionally, reductions occurred in a substantial subgroup.
Klein (1991) did a follow-up on a ten-week conjoint (couples) therapy group. The results were mixed: 80 percent of the couples were violence free at a two-month follow-up, but 80 percent were continuing to be verbally abusive. However, the sample was small and the follow-up period too short.
Stith, Rosen, McCollum, and Thomsen (2004) also found significant reductions in male violence recidivism six months after couples treatment cessation in a study of forty-two couples (only 25 percent recidivated). This was in a couples group-therapy format. In an individual-couples format, 43 percent recidivated. In a nontreated control, 66 percent recidivated. By comparison, in a treatment outcome study done on the Duluth model, Shepard (1987, 1992) found a 40 percent recidivism rate in a six-month follow-up of Duluth clients, higher than most control recidivism levels, and Dutton (1987) found a recidivism rate of 16 percent (or 84 percent complete cessation) based on wives’ reports for a CBT court-mandated group for men. Rosenfeld (1992) found that wives’ reports of husbands’ violence revealed four times as much violence as police reports. (This does not mean that four times as many men were being violent, but that the violent men were more violent than the police realized.) Stith, Rosen, and McCollum (2003) reviewed six outcome studies of couples treatment and concluded that they were at least as effective as so-called traditional treatment.
Treatment Groups for Female Perpetrators
Evidence is beginning to show that women are as violent as men and that the profiles of female abusers are the same as those of male abusers (Archer 2000; Moffitt, Caspi, Rutter, and Silva 2001; Ehrensaft, Moffitt, and Caspi 2004; Babcock and Dutton, in press). Henning, Jones, and Holdford (2003) found that both women and men in court-mandated treatment had adverse childhood experiences and high levels of personality and mood disorder. Borderline personality disorder, which Dutton (2003) noted as being central in male batterers, was even more frequent in female batterers. Hence it seems that blended CBT groups designed for male abusers may prove to be most effective with female abusers.
Intimate Abuse Circles
Linda Mills (2001), in a thoughtful and provocative article in the Harvard Law Review, argues that state intervention itself has become abusive to ‘‘victims’’ who don’t want that intervention. Battered women, she argues, are safest and feel most respected when they willingly partner with state officials to prosecute domestic violence crimes. Mandatory state interventions do not allow clinical healing to occur. The unwanted state intervention replicates ‘‘rejection, degradation, terrorization, social isolation, missocialization, exploitation, emotional unresponsiveness and close confinement that are endemic to the abusive relationship’’ (p. 551). Mills advocates what she calls a ‘‘survivor-centered approach,’’ which focuses on listening to the woman, discussing the options with her, and leaving control of the outcome in her hands.
This approach can involve what are called intimate abuse circles, involving conferences between victims and perpetrators in the midst of a caring community chosen by both the victim and the perpetrator. These ‘‘restorative justice circles’’ have been tried in South Africa, New Zealand, the United States, and Canada. As a group and by consensus, a contract is developed to restore to the victim what has been lost (e.g., dignity, property). The contract must be agreeable to both sides and is prepared only after two events have occurred: First, there has been a full examination of the impact of the violence on those most affected; and second, violent offenders express remorse for their actions. This is referred to as the ‘‘healing process.’’ Conferences can be formed only with the consent of both parties and the participation of the care community. It is a radical alternative to the ‘‘adversarial’’ justice system now in place, in which both sides spin the truth to self-advantage. Braithwaite provides considerable empirical evidence indicating high levels of victim satisfaction; this process leaves victims feeling empowered by their participation in the conference. The offender’s apology offers symbolic reparation and enhanced empathy for the offender (sometimes as a prerequisite for making the apology). Strang and Braithwaite (2002) report that the use of justice circles in Indianapolis had a 40 percent lower recidivism rate than a control group (after six months) and a 25 percent lower recidivism rate after twelve months. A quasi-experimental study in Winnipeg among ‘‘serious adult offenders’’ produced a recidivism rate one-third that of the matched control group. A similar improved reduction in recidivism is reported in a study from New Zealand (Strang and Braithwaite 2002). The only evaluation available as of this writing on the application of justice circles to family violence is a study in Newfoundland by Burford and Pennell (1998), reported in the Strang and Braithwaite book. It found ‘‘marked reduction in both child abuse/neglect and abuse of mothers after the intervention.’’ Thirty-two families who underwent the restorative justice intervention reduced violence by 50 percent in the year after the intervention; by comparison, thirty-one control families saw violence increase.
Caution must be employed here, though; the use of intimate abuse circles in domestic violence treatment probably requires that participants be screened (by criminal justice officials) for psychopathology prior to using this system. Also, the couple has to have access to the ‘‘caring community’’ group that Mills describes. It should be added that victim veto and careful monitoring of community group composition by criminal justice professionals are necessary to ensure that no ‘‘stacking’’ of the community group occurs.
As a summary statement, it must be said that no one treatment modality is so superior that others can be eliminated. What is more important is the fit between participants (the abusive couple) and the treatment or intervention system. Psychoeducational models and/or intimate abuse circles may work better with immigrant groups among whom cultural values upholding patriarchy are at odds with the host culture. Couples therapy may be better suited to mutually violent couples. CBT is probably better for individuals who habitually use violence but should be augmented with attention to the features of abusiveness described above.
- Babcock, J., J. Waltz, N. Jacobson, and J. M. Gottman. ‘‘Power and Violence: The Relation between Communication Patterns, Power Discrepancies and Domestic Violence.’’ Journal of Consulting and Clinical Psychology 61, no. 1 (1993): 40–50.
- Bem, D. J. ‘‘Self-Perception Theory.’’ In Advances in Experimental Social Psychology, edited by L. Berkowitz. New York: Academic Press, 1972.
- Burford, G., and J. Pennell. Family Group Decision Making Project: Outcome Report. St. John’s Newfoundland, Memorial University of Newfoundland, 1998.
- Davis, R. C., B. G. Taylor, and C. D. Maxwell. ‘‘Does Batterer Treatment Reduce Violence? A Randomized Experiment in Brooklyn.’’ Justice Quarterly 18 (1998): 171–201.
- Dutton, D. G. The Abusive Personality: Violence and Control in Intimate Relationships. Revised paperback ed. New York: Guilford Press, 2003.
- ———. The Domestic Assault of Women. 3rd ed. Vancouver: University of British Columbia Press, in press. Dutton, D., and D. Sonkin. Intimate Violence: Contemporary Treatment Innovations. New York: Haworth Press, 2003.
- Ehrensaft, M., P. Cohen, and J. Johnson. ‘‘Development of Personality Disorder Symptoms and the Risk for Partner Violence.’’ Journal of Abnormal Psychology 115, no. 3 (2006): 474–483.
- Ehrensaft, M. K., T. E. Moffitt, and A. Caspi. ‘‘Clinically Abusive Relationships in an Unselected Birth Cohort: Men’s and Women’s Participation and Developmental Antecedents.’’ Journal of Abnormal Psychology 113, no. 2 (2004): 258–271.
- Feder, L., and D. Forde. ‘‘A Test of Efficacy of Court Mandated Counseling for Convicted Misdemeanor Domestic Violence Offenders: Results from the Broward Experiment.’’ Paper presented at the International Family Violence Research Conference, Durham, NH, 1999.
- Hamel, J. Gender Inclusive Treatment of Intimate Partner Abuse: A Comprehensive Approach. New York: Springer, 2005.
- Henning, K., A. Jones, and R. Holdford. ‘‘Treatment Needs of Women Arrested for Domestic Violence: A Comparison with Male Offenders.’’ Journal of Interpersonal Violence 18, no. 8 (2003): 839–856.
- Heyman, R., and K. Schlee. ‘‘Stopping Wife Abuse via Physical Aggression Couples Treatment.’’ In Intimate Violence: Contemporary Treatment Innovations, edited by D. Dutton and D. Sonkin. New York: Haworth Press, 2003.
- Klein, P. ‘‘Efficacy of Conjoint Group Treatment in Therapy for Spouse Abuse.’’ Unpublished doctoral dissertation, Pacific School of Psychology, Palo Alto, CA, 1991.
- Levesque, D. ‘‘Violence Desistance among Battering Men: Existing Intervention and the Application of the Transtheoretical Model for Change.’’ Unpublished doctoral dissertation, University of Rhode Island, 1988.
- Luborsky, L. Principles of Psychoanalytic Therapy: A Manual for Supportive-Expressive Treatment. New York: Basic Books, 1984.
- McCloskey, K. A., M. Sitaker, N. Grigsby, and K. A. Malloy. ‘‘Characteristics of Male Batterers in Treatment: An Example of a Localized Program Evaluation Concerning Attrition.’’ Journal of Aggression, Maltreatment and Trauma 8, no. 4 (2003): 67–95.
- Mills, L. Insult to Injury: Rethinking Our Responses to Intimate Abuse. Princeton, NJ: Princeton University Press, 2001.
- Moffitt, T., A. Caspi, M. Rutter, and P. Silva. Sex Differences in Antisocial Behaviour: Conduct Disorder, Delinquency, and Violence in the Dunedin Longitudinal Study. Cambridge and New York: Cambridge University Press, 2001.
- Pennell, J., and G. Burford. ‘‘Feminist Praxis: Making Family Group Conferencing Work.’’ In Restorative Justice and Family Violence, edited by H. Strang and J. Braithwaite. Cambridge and New York: Cambridge University Press, 2002.
- Pense, Ellen, and Michael Paymar. Education Groups for Men Who Batter: The Duluth Model. New York: Springer Publications, 1993.
- Richter, H. The Family as Patient. New York: Farrar, Straus and Giroux, 1974.
- Schore, A. Affect Regulation and Repair of the Self. New York: Norton, 2003.
- Shepard, M. ‘‘Predicting Batterer Recidivism Five Years after Community Intervention.’’ Journal of Family Violence 7, no. 3 (1992): 167–178.
- Stith, S. M., K. H. Rosen, and E. E. McCollum. ‘‘Effectiveness of Couples Treatment for Domestic Violence.’’ Journal of Marital and Family Therapy 29 (2003): 407–426.
- Stith, S. M., K. H. Rosen, E. E. McCollum, and C. J. Thomsen. ‘‘Treating Intimate Partner Violence Within Intact Couple Relationships: Outcomes of Multi-Couple Versus Individual Couple Therapy.’’ Journal of Marital and Family Therapy 30 (2004): 305–318.
- Strang, H., and J. Braithwaite. Restorative Justice and Family Violence. New York: Cambridge University Press, 2002.