Introduction

Despite efforts to increase public understanding of intimate partner violence (IPV), staggering statistics indicate that intimate partner violence (IPV) accounts for 21% of violent crimes against women (Bureau of Justice Statistics 1998) and 30% of all female murder victims (Federal Bureau of Investigation 1996). Research has shown that male-perpetrated IPV can result in grave consequences for victims, including physical injury (Bureau of Justice Statistics 2000; Foa et al. 2000), psychological distress (Busch and Valentine 2000; Fine and Weis 2000; Foa et al. 2000), and spiritual upheaval (Giesbrecht and Sevcik 2000). Additionally, treatments for adult IPV victims are estimated to have cost Americans $67 billion from 1987 to 1990 (National Institute of Justice 1996). Given the costs and consequences associated with IPV, an understanding of its many causes and correlates is important. This qualitative study explores the ways that perpetrators of IPV experience their religion as interacting with IPV to enhance this understanding.

Although women have been shown to assault their partners at similar rates, they remain the primary victims due to the greater physical, financial, and emotional injuries experienced at the hand of male partners (Archer 2000; Straus 1997). In addition, IPV seems to influence female victims’ sense of personal control but not male victim’s sense (Umberson et al. 1998). In addition, national statistics indicate that 92% of reported IPV cases are perpetrated by men against women (Bureau of Justice Statistics 1994), and suggest that an understanding of male-violence also may have broader relevance to the problem of understanding and treating those IPV perpetrators who enter the legal system.

This paper is part of a tripartite program of research that began with qualitative studies of the intersection of faith and IPV beliefs within victims’ (Knickmeyer et al. 2004) and religious leaders’ perspectives (Levitt and Ware 2006a, b; Ware et al. 2003), and that presently develops an understanding of the perspective of male perpetrators of violence upon female victims. These studies have focused on these experiences within a Southern US context (within the Memphis region) where religion plays an important role in the “bible belt” culture. In addition to spiritual guidance and community, religious institutions in this area often offer services like healthcare, daycare, and social services and can have a strong influence on civic life.

Many different theories have been put forth to explain male-perpetrated IPV. More recent theories have tended to describe reasons why perpetrators might commit violence as framed in relation to either sociological and cultural factors, or individuals factors. Although a complete review of all these theories is not possible in this article, other works review these theories in great detail (e.g., Barnett et al. 2004; Jasinski 2001). Instead, we summarize theories pertaining to the intersection of religion and IPV perpetration, and emphasize the psychological literature rather than the theological literature on this topic (see Fortune 1987; McClure and Ramsay 1999).

Patriarchal Religious Belief and IPV Perpetration

The construct of masculinity in our culture has long been studied in relation to IPV and religion. These studies tend to focus upon patriarchial cultural beliefs that create a climate in which men expect to dominate their partners and control the family’s resources (e.g., Dobash and Dobash 1979; Jasinski 2001; Steinmetz and Lucca 1988). Religious beliefs that exemplify this stance include those that place the husband as the head of the household with the primary decisional power, or that cast wives as the primary caretakers of children, limit their vocational or economic prospects, or charge them with submission to their husbands (see Fortune 1987; McClure and Ramsay 1999). Feminist theorists (e.g., Walker 1988) have cited these teachings as a contributing factor to domestic violence, believing that the assignation of disproportionate power within the relationship can lead to the abuse of that power.

It appears that the acceptance of spousal violence in the US is high overall. In a national survey of family violence, Dibble and Strauss (1980) documented that 27.6% of respondents thought that slapping a spouse could be necessary, normal, or good. The incidents of IPV might be expected to be higher, however, within communities where violence is highly visible or where more patriarchal relationships are normative due to social learning processes (Dutton 1995; Saunders 1988; Viano 1992).

Because of this interest in patriarchal beliefs, conservative Christians are the group that has been studied most frequently, however, the research findings appear to be mixed. Investigators have found that, overall, conservative Christian men were not more abusive than other men (Brinkerhoff et al. 1992; Ellison et al. 1999). There is some evidence, however, that discrepancies in partners’ religious beliefs or congregations may lead to increased risk of violence (Gelles 1974), particularly when men who held more conservative beliefs about the inerrancy than their wives and authority of the Bible were more likely to be abusive (Ellison et al. 1999). As well, Shupe et al. (1987) found that most perpetrators rationalized their violence as biblically acceptable and thought it helped the family run more smoothly. IPV victims also have reported that their perpetrators used patriarchal religious beliefs to justify their acts of abuse (e.g., Knickmeyer et al. 2004).

Although patriarchal religious messages might support IPV, other religious messages might be preventative. For instance, as perpetrators of IPV are thought to be over-dependent on their wives (Hale et al. 1988), and to abuse as a way to counter a sense of powerlessness (see Taubman 1986), an involvement within a congregation might help diffuse this focused dependency and isolation. While religious groups might offer interpersonal support which is associated with decreased IPV rates (Brinkerhoff and Lupri 1988), relief from some forms of stress (e.g., Turner and Avison 2003), and social deterrents to IPV (e.g., Gibbs 1977; Ware et al. 2003), it is unclear whether perpetrators can make use of these resources due to their denial of abuse (Edleson and Brygger 1986; Wetzel and Ross 1983), poor problem solving skills (e.g., Morrison et al. 1987), or difficulty with moral reasoning (Buttell 1999; Kane et al. 2000).

The present study used a qualitative interviewing method to gain more insight on how low-income IPV perpetrators experience their abusive behavior, both generally and in relation to their religious beliefs. Abusers in conditions of poverty might be expected to require the supports offered by religions more than other abusers. They also may face additional challenges in terms of experiencing helplessness and reduced access to resources.

Method

Participants

Participants were male perpetrators of IPV from the Memphis area who were recruited in person at court-appointed anger management groups and at the Exchange Club—the only court-mandated assessment center in Memphis that conducts psychological evaluations of individuals arrested with charges of IPV and refers them to treatment programs. Individual interviews were conducted with 12 perpetrators who admitted to having committed physical abuse of a female intimate partner. Diversity was sought in terms of the men’s religion, race, age, attendance at services, marital status and SES in order to develop results that are as rich and encompassing as possible. This method of “maximal variation sampling” (Patton 1990) is aimed at capturing central themes that cut across differences in participant experience and is a strength in qualitative research.

Although some diversity was achieved along the other characteristics considered to be of theoretical importance, it was difficult to find diversity in participants’ income and the sample was entirely one of low socioeconomic status—perhaps as higher income abusers are better equipped to avoid being mandated to assessment or can afford private treatment. It may be for similar reasons that there was limited racial diversity. Ten participants were African American and two were Caucasian. Eight of the men reported making under $10,000 annually, three indicated that their income was between $10,000–20,000, and only one participant reported an annual income that ranged between $20,000–30,000. Three participants were unemployed, two had sales or clerical jobs and the rest were employed in unskilled labor.

The mean age of participants was 31.25 (range of 20–48). Of the 12 men, nine described themselves as Christian, one as Jewish, one as Islamic and one as Jehova’s Witness. Eight of the participants reported attending religious services for major holidays and six of these men attended services at least monthly. The other four attended rarely or not at all. One interviewee was divorced, four were single and never married, three were co-habitating, three were separated, and one was married.

Instruments

For descriptive purposes, a demographic questionnaire was distributed to the participants and, because of literacy problems, reviewed orally. It asked for their age, race, faith affiliation, marital status, and employment status. It also included a version of The Conflict Tactic Scale (Straus 2004) to screen potential participants. To participate, all participants indicated, either by marking on this scale or verbally endorsing an item on the scale, that they had engaged in the physical abuse of a romantic partner.

Procedure

Interviewing

The interviewers and primary analysts in this study (the first two authors) were Jewish and Christian in their backgrounds respectively and did not enter the study expecting to hear particular responses from the perpetrators. Memos or notes were kept throughout the research process to help the researchers become aware of and limit the affects of assumptions and biases, to keep a record of procedural decisions, and to record theoretical ideas that arose during the process of analysis. The primary author has expertise with qualitative research methods (teaching qualitative methods courses and having published extensively using grounded theory methods). She co-conducted the initial five interviews with the second author to provide supervision and training in qualitative interviewing. After having become a skilled interviewer, the second author conducted seven interviews alone.

The questions in this study were designed to be open-ended and to avoid leading the participant to respond in a particular way. The main question inquired about the perpetrators’ experiences of IPV as it interacted with his faith beliefs. For instance: “Has your faith affected the course of IPV in your relationship? If so, how?” Sub-questions explored how religious beliefs about gender roles influenced their experiences of IPV, and the ways in which social expectations and supports from religious and social communities influenced their experiences of IPV (e.g., “What do you think a woman should do for a man in a healthy relationship?”). The interview was geared to shed light on whether religion acted as either a support to avoid IPV or as force that may promote IPV. Prompts were used to encourage the participant to elaborate further about or to clarify responses (e.g., “Tell me more about that.”). As well, participants were asked how to best prevent IPV and asked if their experience with IPV has changed their lives (and if so, how).

Interviews were conducted at the site of anger management groups, or at the Exchange Club. Interviews were approximately 1 h in duration and were audio-taped with the consent of the participants. Participants were given a thirty-dollar honorarium to recompense them for their time.

Analysis

The interviews were analyzed using a grounded theory approach. This inductive approach (Glaser and Strauss 1967) was developed to generate empirically based theories. In accordance with Rennie’s (2000) argument that the classic version of grounded theory is most coherent within its philosophical framework, this form of analysis was utilized. In this analysis, transcripts were divided into meaning units. Meaning units are segments of texts that each contain one main idea (Giorgi 1970). In the initial stages of the analysis, the meaning units were labeled in a manner that remains very close to the language used by the participants. The meaning unit labels were compared to one another and then organized according to their similarities, creating descriptive categories. Then, these categories were compared to one another and, again based upon similarities, higher order categories were formed. This process was continued until a hierarchical model was developed. Ultimately, a central theme was represented in one core category that topped the hierarchy. In this study, the last three interviews did not add additional categories to the hierarchy and so data collection was halted after 12 interviews. At this point, the hierarchy was considered ‘saturated’ which occurs when new information does not seem to be forthcoming and the hierarchy appears to be comprehensive.

Credibility Checks

Three credibility checks were conducted to assess how well the analysis represented the participants’ experiences. First, participants were asked for feedback after each interview on its thoroughness (e.g., “Is there anything relevant we haven’t discussed?”). Second, after the data analysis was complete, five participants who agreed to be contacted for this purpose were telephoned and the results were reviewed with them orally. All the participants contacted endorsed the results, some remarking on how they shed light on their own experiences and understandings. Third, as consensus between investigators increases credibility (see Elliott et al. 1999), the two primary analysts conducted the analysis together, reviewing transcripts and discussing and creating categories that were congruent to their shared experiences. In the case of conflict, the second researcher’s interpretation was privileged, in keeping with the authors’ hermeneutic approach to qualitative analysis, as she had the lived experience of being present at all the interviews of participants in this study.

Results

The data derived from the transcripts of interviews conducted for this study consisted of 1,068 meaning units. An eight-level hierarchy was derived from the sorting of these meaning units. In describing this hierarchy, the following lexicon will be used: Seven “clusters” were developed; each encompassing a group of “categories” which, in turn, subsumed “subcategories.” Following the presentation of the clusters is the description of the “core category,” which is at the apex of the hierarchy and reflects the central finding in this study. Table 1 presents the number of participants whose interview contributed meaning units to each cluster, and category.

Table 1 Clusters and categories.
Full size table

Cluster One: The Arrest Incident is a Hurdle or a Test From God That I Alone Have to Deal With, Although the Responsibility for the Abuse Was Not All My Own

The perpetrators (9 of 12) stressed that the fear of sentencing and the inconveniences and fees of the court, assessment and treatment systems motivated them to avoid future IPV incidents (see Table 1). One explained, “I ain’t got no job and can’t find no job. It’s really frustrating. I have to come to these classes every Wednesday and go see my probation officer. It’s running me crazy. I want me a day job, but I can’t work no day job. I have to do all this” (P-04). Although six participants believed that their experience with IPV was a test from god, none of the participants reported that their IPV and subsequent arrest incident was caused by god. In contrast, god was thought to want the perpetrator to become a better person and tended to be described as all-knowing, powerful, kind, and just. In response, perpetrators tended to try to pacify god via prayers but their distress did not move them into a process of religious self-examination or critique of their morality.

Instead, most of the perpetrators (8 of 12) thought that both partners shared the responsibility for the abuse or it was due to the influence of drugs or alcohol (2 of 12).

I think that [drinking] used to bring on a lot of my domestic violence... because, you know, when you get high you [think you] got more power than what you got.... Now when you take that strength out of you, well, you ain’t drinking today. You might not have the conflict (P-07).

At least one of the perpetrators had participated in substance abuse support groups in response to having committed IPV.

When they blamed their partner for their violence, nagging was a common cause of abuse (6 of 12). “Nag, nag, keep on bothering me, especially when I done left alone and you still talking about the same subject.... Constantly talking. Talking, talking and talking. That’s how they push my buttons” (P-03). They experienced their partners as willfully and skillfully attempting to upset them rather than trying to find ways to have their own needs met.

It became clear that at least some (5 of 12) of the participants heard their partner’s complaints as an insult—suggesting that they were inferior. One man described this experience: “It is very easy to make me angry because I feel like someone is playing with my intelligence.... Like a white hot flame shoots on. And then I have to prove to you that you’re not running the game on nobody stupid” (P-05). It seemed to some interviewees (5 of 12) that displays of anger became the only way to regain respect and avoid threats to masculinity. It was striking that the interviewees tended not to see discrepancies between the impact of their partners’ verbal provocation and their physically abusive reactions.

Cluster Two: Passive Avoidance and Withdrawal from Conflict is the Best Way to Prevent Aggression and to Please God

Solutions for violence reportedly were offered in both IPV-groups and in religious sermons (see Table 1). The solutions the men remembered were often passive, however, such as the belief that by just attending IPV groups they would be cured (6 of 12). “I wish the church would kind of have a class like this... an open discussion about anything [on] domestic violence. Be nice for a young person like me, you know, who knows now that any type of verbal or physical assault, you’ll get charged for it” (P-12). Although some (4 of 12) wished they had these classes before, they acknowledged that they attended now because the court had mandated treatment.

Almost all the men (10 of 12) thought god did not approve of IPV, but they could not recall learning any strategies that would help them resolve conflict during their attendance in religious institutions. Although they wanted advice, they thought that IPV was not important to religious leaders, and that leaders would not understand them or would blame them rather than help them learn relationship strategies.

They [leaders are] all talk, but they’re not in the situation you’re in.... To you, it can be, “My gosh, these guys are beating on women” and to me, it’s “Damn, this girl just threw a pot of hot water on me so I kicked the shit out of her,” so I mean, its different...They need to sit down and listen to people instead of preaching so much...I would want them to hear the cries of... the inner city (P-05).

While some interviewees doubted that religious sermons could overcome the negative influences of media violence and poverty, most (9 of 12) thought that hearing sermons on IPV might drive perpetrators away from religion: “[IPV-related sermons] make me feel bad, and I’m in that category [“perpetrator”], because when they talk about it, they make you feel bad... hearing what [religious leaders think] almost makes you feel worse (P-04).” One interviewee expressed his disdain towards religious leaders’ simplistic anecdotes:

The only thing they can say to you is “Pray and stay strong and keep working real hard, and then that might work.” And sometimes that might work. But then you get to a point where you think, “Damn, I’m working hard and I ain’t making no money.” It’s bad. I’ve got a friend who’s 33-years-old and he’s working at McDonalds making $5.25 an hour. How the hell is he supposed to support his family? I mean, seriously (P-05).

Two of the men described that their frustration increased with the only religious guidance they received, which was for conflict resolution to occur in a relatively short time-frame (e.g., “before the sun does down”).

Although most of the men (8 of 12) who were interviewed had been taught that ‘calming down’ or ‘walking away’ could help them de-escalate, most of the men (10 of 12) expressed uncertainty about how to resolve conflict beyond that point. Such a strategy in which conflict was perpetually ignored did not seem to work well. “Me and my wife, it was mostly mental or emotional. If she did something to me I wouldn’t talk to her. I would just go off in a corner or just be by myself or just leave the house. She couldn’t take that” (P-11). They had observed few role models of conflict resolution and were aware of few strategies.

Accordingly, when conflicts accumulated that could not be easily resolved, many of the men (5 of 12) thought that ending the relationship was the only way to stop violence from recurring.

I don’t deal with anybody so I don’t have any conflicts.... It makes me feel bad because I be lonely sometime, but at the same time, it’s the best thing going for me right now. I’m trying to rebuild me. I’m trying to put me on a foundation to where I can be a total leader. Like I teach my sons, “Be leaders instead of followers”(P-05).

Although this interviewee’s choice to isolate himself was a strategy to avoid relational dependency and conflict, it left him without interpersonal support and it could be difficult for him to model healthy relationships for his children, despite his intentions.

Half of the participants described the need to separate from families in order to end the pattern of abuse. “It was scary [moving away from his partner], because I’m not a traveling person, so they [my children] would probably be coming to see me before I went to see them. So I’d probably be sending some money for them to come see me” (P-06). In contrast, another participant responded with an apparent nonchalance at not being able to see his son, saying, “I’ll see him when he gets older” (P-03). For some, engagement with their children was difficult to retain in the face of parental conflict.

Although withdrawal from family through mandatory or willful separation was described as a defense against further violence or “doing something really bad” (P-07), some participants reported missing their families (3 of 12). One man described his prayers in jail:

Lord, Jesus, I didn’t mean to do this. Please help me get out of this situation. Please help me get by. You know, you miss your family. You miss your family so much and you can’t cry no more because you know you’re gonna be there [jail].... It changes everything because you’re not used to being locked down. You’re not used to a cellblock (P-08).

The termination of relationships might appear necessary to avoid IPV, as the methods of conflict resolution known seemed unrealistic or abusive, but it was not always desirable.

Cluster Three: Ambivalence about Marital Roles was Created by Conflicting Messages between Religion and Society

Whereas in the second cluster the participants’ described ending relationships with partners, these beliefs about marriage were challenged by their religious teachings (9 of 12; see Table 1). This quote exemplified this stance on marital relations:

I believe in the marriage for better or worse.... I always felt like if you were married you should stay with that person till ya’ll pass away. I feel like if you get a divorce and get married again.... you’re gonna share your life with too many women—you ought to be sharing it with one (P-07).

At the same time though, some interviewees (4 of 12) seemed to think that the problem in marriage was one of finding the right person, with whom there would be no conflict.

You can’t find the right woman or the right man (inaudible), but if ya’ll meet and got a lot in common and ya’ll understand a lot of things, then ya’ll agree and disagree, it won’t lead to no big argument. If you find someone like that and ain’t got problems, then you do everything in your power to hold onto them (P-09).

This belief allowed them to sever ties with partners, despite hopes for lasting monogamy.

Some participants (3 of 12) experienced frustration with the ongoing changes in a society advocating an equality of the genders in marriage. Half of them (6 of 12) described and endorsed conservative religious teachings that men should be the leader, provider and decision maker in relationships. This belief was at odds, with messages they received from their culture, though, in which women expected to be equals.

Whatever the man says is supposed to go. That’s what I feel like. But it don’t always work that way, though. Nowadays...the woman wants to run everything. She wants to be the boss of everything...I think god’s saying the man ought to be the leader.... I think a woman should have some household duties.... The man should run everything else (P-04).

Many participants (7 of 12) thought women were responsible for emotional caretaking within the relationship: “He would be the one who brings home the bread. She would be there to comfort him.... [Saying] like ‘How was your day?’.... If it wasn’t good, have him sit down and have a talk... A woman would be there to talk him through it” (P-02). Another interviewee offered the following advice to women:

Stop trying to be so much in control. Be the woman... know what being a woman is.... But you have to give your man space to be a man.... You can’t be right all the time. Some time, let him make a mistake. I got an uncle and he’s got a wife who tells him to shut up when he’s in the middle of a conversation. He folds up to her. And that type of—that’s a time bomb waiting to explode. I can see him kicking the shit out of her in another month (P-05).

As the majority of the men interviewed had an income of under 10,000 per year, their ability to be traditional family leaders or breadwinners was seriously compromised. They were in a situation where they had neither the resources nor the partners who wished them to be the heads of the family, yet they were embittered about their lack of power rather than wishing to share power. They struggled with the question of how they could retain their identity and pride in their masculinity under these conditions.

Other interviewees argued that women and men should work together within a relationship. These men (8 of 12—sometimes the same men who wished for dominance) offered a variety of reasons why equality might be better: if women did not rely upon men financially so they could escape in cases of abuse and not be financial burdens on their partners; if power within relationships was situational, partners of either sex could have decisional power when they had more expertise or resources in a certain area; if men and women both were imperfect, they needed to work together; and there was a concern that submission could be harmful for women’s self esteem. Although these men espoused beliefs in equality, often some indecision existed about how this ideal should be practiced. For instance, this participant began explaining how the church should teach marital equality.

Dismiss our stereotypes.... If they got rid of the stereotypes and just concentrate on everybody being equal and, I mean, the Bible I think might say that.... I used to always think that the Church advocated that men and women are not equal.... But then again, I don’t think the Bible says that a man and woman are equal. I think the Bible says the man is superior to women. But I could be wrong (P-11).

This ambivalence made it hard for these interviewees to articulate their marital role clearly and develop a construct of being a husband that was both coherent with their religious beliefs and their socioeconomic and social reality.

Cluster Four: An Instrumental Relationship with God: Religious Leaders are Needed to Teach us What God will Punish or Reward

The fourth cluster in this hierarchy reflected participants’ experiences with religious leaders and their relationships with god (see Table 1). All participants affirmed a belief in a higher power, and 11 of the 12 participants believed that having a relationship with god was important in their lives. “He’s the center of my life. Everything revolves around Him (P-09). A strong religious belief is common in this Southern city.

God could offer a safe mode of communication: “God doesn’t judge. He’s looking over you and knowing exactly why you made a decision. It wasn’t the right decision, you know, but He knows what was going on in your life, what caused it” (P-02). At the same time, God also was thought to set trials for people that could lead to growth or to obstruction. It was important to be in God’s favor as result. Three of the participants reported feeling closer to god after the IPV as it gave them the time or impetus for reflection.

Although interviewees did pray for forgiveness and guidance once they had a problem, they tended not to pray to consider how to change their own ways of relating. In contrast, prayer was most often a request for god to deliver goods or to solve problems.

I have an example. I violated probation because... I was supposed to take a counseling session [but didn’t].... I think it was because of god they gave me another chance.... And when like I caught the charge, they tried to give me aggravated assault because a knife was involved, but you know... they dropped the aggravated part and made it simple assault (P-11).

When we go to the Lord in prayer, on bended knee, and we talk to Him with our heart and our mind, you know, with all sincerity, then He’ll give it to you. Whatever you ask for, He’ll give it to you (P-09).

Four of the men described a somewhat conflicted relationship with god, as they had experienced pressure to be religious from others that made them wary of religious leaders. Some of the men (6 of 12) implied that biblical texts contained straightforward instructions for all problems and were suspicious that preachers were deliberately misleading them: “Don’t put personal opinions in other people’s minds... especially if it’s a religious leader, people who hold authority in the church, they should get all their information from the Bible (P-02). They wanted religious leaders to more clearly indicate to them the answers to their problems as they are written in biblical texts. As a result their relationship with leaders was dependent and instrumental when it was not suspicious outright.

Cluster Five: Isolation is Scary to Bridge and so Relational Skill Development is Limited, and God can be a Source of Solace

Most of the men (9 of 12) had impoverished social support networks and reported trusting no one or few others (see Table 1). A number conveyed that our 1-h interview was the deepest exchange they had had with anyone in years. This lack of trust also extended to religious leaders, who often were seen as out of touch with the poverty that these men lived within. “I think the reason why the preachers try to get the congregation to work is ‘cause when they pass around the money tray and everything. I think that’s the whole purpose... that’s like a big racket” (P-11). Because religious leaders did not appear to suffer from poverty, their sincerity was questioned.

The domestic violence groups offered the perpetrators an opportunity to bridge their isolation and the men (9 of 12) spoke positively of their gains from this experience. Talking to others who shared the same experiences helped them to cope with feelings of guilt and to develop greater self-awareness.

I don’t trust nobody...It leaves me in a situation where I feel lacking when it comes to help and support.... I was court-ordered to... take a domestic violence program.... [But] actually coming here and realizing these people are going through the same thing I am... I listen to their stories about things they have done a little bit differently that worked better than I handled it. It makes me feel better. I trust the person, and that’s something I can work on. I trust one person (P-02).

Most of the participants (10 of 12), however, said that they chose to keep their problems to themselves, because they either did not feel safe being vulnerable with others or did not want to burden others with their difficulties. “I have a problem trusting people...it’s because I’ve been betrayed before, as far as letting people in. Like you tell somebody something and it gets back. You didn’t want it to get out, but it gets out, and you hear about it. No, that’s not going to happen again” (P-11). This powerful fear of gossip made it difficult to confide in friends, seek guidance for problems, or find support for stressors within relationships.

Within an intimate relationship though, the men (6 of 12) reported valuing communication and mutual respect, and realized that problems in communication often led to other difficulties. “You should work things out with your wife. But how can you work things out with your wife when she don’t want to talk to you? She don’t want to have anything to do with you” (P-04). The implementation of healthy interactional styles appeared difficult in the face of their relational realities.

For most of the men interviewed (7 of 12), intimate partners were their sole source of emotional support and so relational conflict might be expected to be particularly threatening. Five of the men, describing trust and closeness in partnerships, thought it could best be developed through the doctrine of mutual submission, rather than the unilateral submission of the wife to the husband. Financial pressures, however, were not always conducive to the development of an intimate relationship: “It kind of be the best if—we were balanced. [If] she had a job in the morning, I had a job at night. So someone always watched the kids, we saved money on child care” (P-12). Although they hoped that equality would lead to increased intimacy, their strategies might lead to isolation and distrust instead.

Cluster Six: Anger is a Way to Manage Interpersonal Insecurity and Maintain Masculinity through Asserting Control, but God Doesn’t Support its Expression in Violence

Within the interviews, participants (10 of 12) were able to identify that there was a fear underlying their aggression, the fear of being seen as weak and of being rejected (see Table 1). One man described his experience at the point of engaging in abuse, “Everything that was important to me was put on the back burner.... She was very verbally abusive. And I tried to ignore it, and then after a while it got to where I felt like she was another man trying to make me smaller and we used to fight” (P-05). In this quote, the participant imaginatively altered his partner’s gender to permit his assertion of masculinity through physical aggression. As suggested in cluster one, a perceived threat to their masculinity was an impetus to abuse across many of the interviews. Although this interviewee could identify his fear to seem week in the interview, the following was his response when asked if he could convey this fear to his partner directly at that moment:

No. You’d probably say “I’m really pissed off this is happening.” I don’t think you would say, “I’m scared....” Probably because it was ego.... Make him look weak to the woman... (Interviewer: What is scary about that?) Well, I guess it just ruins your—every man wants to be masculine. He wants to be an Alpha man, and if he shows weakness to his woman, she might not look at him as being an Alpha man no more.... His fear of that [is] she wouldn’t respect him any more (P-05).

These men (5 of 12) asserted that as women did not respect men who expressed fear, it was safer for them to express anger.

It was hard for them to be certain that they would not become violent again if their masculinity was threatened in the future. “I never know what I might do. I might just been so mad and angry I might just click [lose control] again, I don’t know what might be going through my mind... I just try to hold my composure” (P-04). It was frightening for them to wonder, “Will I do it again? The next interaction with my wife, will it end up as a hostile situation” (P-02).

Although most of the interviewees thought that violence was not condoned by their religion (10 of 12), eight thought that expressing anger was acceptable. “If you got to go in your back yard or get a bag or take it out on a tree, that’s good to relieve your anger” (P-07). Violence though was expected to be punishable by god, “If I got to hit her to make her love me and want to be with me, I don’t want her. Cause what? What is that? That’s being controlling and abusive, and where is that going to leave me? In the place down there [hell]” (P-09). Although their rage response might have been empowering in the moment, it led to eventual regret, albeit a self-focused regret.

Cluster Seven: Cultural Expectations and the Witnessing of Domestic Violence Teaches that it is Unavoidable and can be Socially Beneficial, Despite Religious Beliefs

Many participants (5 of 12) described the occurrence of violence and the absence of conflict resolution as common in their youths (see Table 1). Several participants (4 of 12) endorsed the fact that they had experienced, witnessed, or been exposed to domestic violence repeatedly as children.

It wasn’t no problem. I grew up around it all my life. My mama and my stepdaddy used to get to fighting all the time. I used to see it all the time. Well, it doesn’t affect me a whole lot. It surrounds me, but there’s not a lot I can do about it. I can’t get away from it. I’m just in it. Like that man my mama was with, he died. Now she got her another man. Now she’s beating him (P-04).

The early and repeated witnessing and experiencing of violence left many men with difficulties trusting others. Abusive parental figures often modeled paranoia as well, teaching them explicitly at times to avoid trusting others.

After seeing women family members remain in violent relationships they (2 of 12) endorsed the idea that women prefer to stay with their abusive partners. This participant described learning that he could abuse women:

I wanted to help my mama [when she was beaten], but I couldn’t because I was too young. It was a hurting feeling... They’d be drinking all night and he’d start clicking. [I learned] I could be abusive to them [women], I could do anything I wanted to them...I asked my mama about that and she said she loved her man.... Some women love for a man to beat them. They feel like a man love them if they treat them like that.... Ain’t no sense in me helping them (P-04).

It appeared to them that women favored abusers and gave them preferential treatment.

I’ve blocked so much of that out. It’s hard, you know. I know she [mother] used to always take care of her men better than she took care of us, and that was one thing that pissed me off, because I’m like, “You do everything in the world for this guy that’s kicking the shit out of you every other week” (P-05).

I asked my sisters “Why do you like for a man to beat [you]....” And they do.... Like I want to protect them, keep them away from this person...but women like to be abused because it makes them feel like the man loves them...I see it all the time. They get black eyes real bad and they still be with that same man...Maybe they learned from their mothers to always be with someone (P-06).

Seven of the men reported never seeing successful conflict resolution in their homes. When conflict did not lead to violence, it was discussed behind closed doors. “If they had a conflict it never was in front of us” (P-07). As there was little awareness of the verbal negotiation of conflict, it was not surprising that violence was considered an inevitable result of conflict: “If I stay angry I might gonna do something to her.... Somebody is gonna disagree and then somebody is gonna click.... Somebody is gonna get hurt”(P-04).

Other participants (3 of 12) reported learning from parents by seeking counsel from them, or by using them as positive or negative models. One participant described realizing he was like his father and subsequently trying to change:

They were never together but my mom has always been in violent relationships, and my dad has just never cared, so even when I was reaching out to him because I didn’t know who else to go to, he pretty much was “I don’t care....” That was his attitude.... Raising me didn’t benefit him, so he didn’t want no part in doing it... Uh, I used to be same way.... I used to fight dudes, women, especially if you hit me.... So now, I pride myself... I stay away from people. If I can’t get along with you on a conversation level, I just leave you alone. (P-05)

In the face of violent family histories, it appeared that some of the participants learned to focus on getting their own needs met. One relayed the following reaction to abuse:

I wanted to help my mama, but I couldn’t because I was too young. It was a hurting feeling. What really messed me up we was sitting on the porch one day and he just came by and shot my mama. He could have shot me cause I was sitting on the porch, too. It really messed my mind up... It could have been me (P-04).

This lack of empathy was in keeping with his belief that abuse was desired by women.

If their peers knew a woman was putting them down, an act of violence might be expected so they could save face.

I go tell one of my partners, he might tell me something completely stupid, like “Whip her ass! Make her stay at home!”... [If I] go to a girlfriend of mine, she might make it even worse: “Well, you know she only going out so she can see ‘so and so’ at the club.” So, I mean, then they added fuel to a fire that’s already festering...[So], I try to keep all my problems to myself (P-05).

Isolation from others preserved their masculinity in the public sphere and made violence less likely, despite the loneliness and difficulty seeking advice that it created.

Core Category: How to Develop Trust in Conflict Resolution? The Struggle to Maintain a Relationship in Conditions that Exacerbate The Fear of Losing Masculinity and the Need for Isolation

The core category was identified from the analysis of the commonalities across all the clusters. The construct of masculinity as powerful and self-reliant was key in several of the clusters and can be understood as implicit in others. As a result of the belief that being a man was to be entirely independent and that dependency invited humiliation and rejection, perpetrators’ support systems were severely curtailed. This notion of masculinity left them suspicious of religious resources or supports that necessitated ties to a community. This isolation made it harder for them to develop alternate perspectives on conflict situations or obtain suggestions for better coping. Their masculinity was compromised by their poverty that left them unable to enact the male role of supporter that many saw as a religious obligation. Also, changing societal gender roles left female partners unwilling to adopt the religious role of the submissive wife. Interpersonal conflict then was experienced as their masculinity being further questioned by the, often only, person with whom they shared intimacy. As they did not have confidence or experience in their abilities to resolve intimate partner conflict verbally, a sense of vulnerability could surface quickly. This vulnerability could not be expressed as it would further undermine their masculinity, so the perpetrators would seek to gain control by expressing rage instead. They described that this expression bolstered their masculinity, was normative among their peers, and was thought to be acceptable or desirable by women. In this way, the inability to trust in the possibility of conflict resolution worked with their construct of masculinity to undermine their abilities to seek supports, make use of resources, or develop successful conflict resolution strategies.

Discussion

This study was designed to shed light on how perpetrators of violence experience their abuse and understand the place of religion in connection with IPV, but the analysis went further. As the questions were designed to be open ended, in the interviews, participants raised relevant issues relating to masculinity, poverty, and emotional regulation that became key themes in the analysis. The sample selected in this study was from one southern city and was a sample largely living in poverty, so readers should use caution when generalizing these findings to other groups of IPV perpetrators. Because of the need for small numbers of participants in qualitative research, it often is difficult to obtain diversity across all desired characteristics. The little diversity in race and socioeconomic status reflected the pool of participants in the treatment centers and groups that provided assessment and treatment for perpetrators. It may be that perpetrators in higher income brackets avoid court-ordered assessments and mandates to treatment in publicly-funded anger-management groups.

Also, as this study gathered data only from male perpetrators, it does not comment on the experience of female perpetrators, victims of IPV, or the perspectives of religious leaders on IPV (for research within the same region on the latter two topics see Ake and Horne 2003; Knickmeyer et al. 2004; Levitt and Ware 2006a,b). At the same time, the study of male perpetrators’ experience is helpful in developing an understanding of IPV from a perspective rarely explored. While the purpose of this study is not to excuse abusive behavior, the authors do believe that such an understanding can lead to the development of treatment and prevention options that are better attuned to the needs of perpetrators. Also, the descriptions of participants’ experience, in their own words, can have clinical utility themselves as they can be discussed within group treatments for IPV and may promote useful reflection upon themes that otherwise might be difficult to discuss.

Violence as an Assertion of Masculinity in a Context of Poverty

While patriarchal religious beliefs might allow men to justify abuse (Knickmeyer et al. 2004), this study sheds light on how their beliefs interact with other relational dynamics. Poverty has been tied to severity of abuse (Sutherland et al. 2001) and these participants had incomes that were less than the local living wage (Economic Policy Institute 2007). Although their conservative religious beliefs charged men with the role of provider and leader, these roles were not viable possibilities. Many did not question the marital roles they endorsed, but talked instead as though they should fulfill these roles, alongside of talking about the impossibility of financial security. This contradiction appeared to leave the participants in positions where they might feel shame and emasculation due to their socioeconomic class (see Brown 2004 for a discussion of perpetrators’ shame). This work supports prior research that has suggested that high gender-role conflict with respect to success, power, and competition is positively related to abusive behavior (Schwartz et al. 2005) and work by Anderson and Umberson (2001) that suggests that violence can help to construct a sense of masculinity where one is weak.

This understanding suggests the need to develop and promote constructs of masculinity in which dependency or interdependency is reconcilable. While struggling to maintain an image of masculinity, it was difficult for the participants to empathize with victims as they reported feeling so powerless and injured themselves. In contrast to previous research that has found that sexist views are related to attitudes that legitimize the abuse of women (e.g., Glick et al. 2002; Luddy and Thompson 1997), this research draws to light a contradictory understanding of relationship roles in which men’s authority is valued but equality was espoused as well. This contradiction might be discussed usefully in men’s groups and fathering groups where explorations of masculinity might help foster healthier constructs of masculinity within communities. In addition, men in poverty may be educated about anti-poverty and living wage organizations to help them recognize the additional stressors they face.

Identifying and Expressing Fear Rather than Regulating Anger

This work coincides with other research that has shown that batterers have difficulty identifying their actions as abusive and have instead described responding to a threat of being perceived as emasculated (e.g., Anderson and Umberson 2001; Goodrum et al. 2001). The literature on perpetrators seems to suggest that they have a reduced capacity for psychological mindedness (Barnett and Hamberger 1992). Therefore, it was striking to the authors that, within the interview context, participants could admit that their violence was triggered by the threat that they were perceived as weak or unmanly and could label this feeling as fear. At the same time, none of the interviewees would consider confiding this feeling to their partners. They were convinced that disclosing fear would reduce them further in their partners’ eyes. Instead of receiving reassurance for their fears, they abused drugs or alcohol to numb this feeling and at times seemed profoundly detached from both their own emotions and their partners or children. Even when directly questioned, they could not believe that women would prefer a confession of vulnerability to a response of rage. Instead, they felt attacked and felt a need to self-defend. Clinically, these issues may be important to discuss within a group setting in which women perpetrators may be able to convince male perpetrators that violence is not preferable. Indeed, the repression of emotional responses within a situation that is viewed as threatening has been linked with violent behavior (Umberson et al. 2002). These findings emphasize the importance of developing an understanding of masculinity which is not irreconcilable with the expression of fear or support seeking, but in which these behaviors are signs of confidence and maturity (see Horne and Kiselica 1999; Pollack and Levant 1998 for examples of treatments reconstructing masculinity).

It might be helpful in IPV or anger management groups to have men roleplay being angry and, instead of becoming enraged, calming down and then identifying and expressing a specific fear and to receive feedback from women group members that this expression is preferable to violent rage. Through this type of roleplay practice, perpetrators might learn to communicate their own feelings as well as empathize with the pain and fear experienced by their partners (see Wolfus and Bierman 1996 for support for emotion-focused therapy with batterers and Greenberg et al. (1993) for general information on emotion focused therapy). Through emotion-training, participants in conflict-management groups could learn how to resolve a problem peaceably after they walk away from conflict, while the participants in this study had few ideas on how to proceed beyond this point.

Religious Beliefs and IPV: How Religious Communities Might Help

Although religious community involvement might offer different types of supports to the IPV perpetrator, there were several reasons why perpetrators did not appear to be seeking help from religious leaders: (1) They had difficulty trusting others in general; (2) They had difficulty believing that religious leaders were altruistic rather than motivated by financial gain; (3) They thought that leaders would not be able to relate to the stressors they faced in their lives and were not interested in learning about them; (4) Sermons typically did not address martial conflict, but when they did, they either cast the interviewees as villains and increased their sense of guilt, or offered vague platitudes or strategies that were not helpful; and (5) The perpetrators interviewed appear to view the leaders in an instrumental way—as a means to receive advice on how to secure god’s favor—and did not consider them as resources for relational support. These findings were in stark contrast to comparable studies of victims of IPV in this same area who tended to seek out support from religious communities, to engage in much self-reflection in their prayers and to seek emotional support from their relationship with god (see Knickmeyer et al. 2004).

Wilcox (2004) has conducted research that suggests that conservative Protestant men who attend services regularly are less likely to commit IPV than those who do not and explains that the religion is teaching men to be more family-oriented. This research, however, suggests that this association may be due instead to a common factor of interpersonal mistrust, so that men who commit IPV also may be more suspicious of religious leaders and institutions, less invested in the community and less interested in attending regularly. Further research is needed to compare these hypotheses.

Research on religious leaders in this same city (Levitt and Ware 2006a,b; Ware et al. 2003) has indicated that leaders often are perplexed on how best to address the issue of IPV and prevent it within their congregations. As the perpetrators thought that god disapproved of violence, however, they seemed to have some motivation to learning other ways of dealing with conflict. They may be more open to trusting religious leaders who show evidence of altruism and who seem to understand the burden of poverty. Also, they seemed to seek counsel on how to negotiate conflict within their relationships. These findings suggest different avenues for intervention. Psychologists may be invited to provide relationship training on how to resolve conflict in relationship and men’s groups or fathering groups may be organized.

In addition, religious leaders that teach that women should be submissive to men, may wish to be careful about the messages that this teaching gives to men about masculinity. In addition to findings here that suggest that half the perpetrators were endorsing these beliefs, preliminary data from a survey of 2,500 women in this region (Norwood et al. 2004) indicated that 24% of women in current abusive relationships said that their partners used religion to justify their IPV, supporting previous hypotheses of this connection (e.g., Dobash and Dobash 1979). As well, women who have been victims of IPV have reported that the ideal of wifely submission contributed to their remaining in abusive relationships longer (e.g., Knickmeyer et al. 2004; Nason-Clark 1997).

Within evangelical circles, there is a great debate on how scripture on marital structure should be understood. Partially in response to concerns about oppression within marriage, conservative Christian feminists are contesting patriarchal meanings of submission, suggesting that it be reinterpreted as mutual submission between husband and wife (Bartkowski 1997; Nason-Clark 1997). Mutual submission can be enacted using a variety of methods, such as by turn-taking in submission, a process of compromise, or having the spouse with the most competence in an area lead in that area (see Bartkowski 1997 for a more thorough review of these methods).

Based on the analysis of reports of male IPV perpetrators, this paper proposes multi-faceted suggestions to reducing IPV that can be incorporated in anger management or violence prevention groups. If these men learn to trust in the possibility of verbal conflict resolution and to believe that they can practice it without sacrificing their masculinity, they can learn to begin to successfully resolve conflict after deescalating. Discussion of egalitarian religious beliefs about marital relationships, and the ability to trust in others may help these men to forge an interpersonal network that might decrease their isolation and provide support. Also, by working with male IPV perpetrators to consider the ways their beliefs about masculinity influence these aspects, we may help them to develop healthier attitudes about what it means to be a man and to be in relationships.