Emily M. Douglas Denise A. Hines and Murray A. Straus
The World Health Organization (WHO) (WHO, 2010) draws on work by Heise and Garcia-Moreno (2002) to define PV as acts within an intimate relationship that cause physical, sexual, or psychological harm; this could include physical aggression, sexual coercion, psychological abuse, and controlling behavior. The US Centers for Disease Control and Prevention (n.d.) takes their definition of PV from the work of Saltzman, Holden, and Holahan (2005) and describes four types of PV: (i) physical violence, the intentional use of force that could cause injury or harm; (ii) sexual violence, which involves using physical force to engage in sexual acts against a person’s will, attempting/completing sexual acts without the victims comprehension or consent to participation, or abusive sexual acts; (iii) threats of physical or sexual violence; and (iv) emotional/psychological violence, which includes humiliating, embarrassing, or controlling the victim.
The most widely used tool to measure and assess PV is the Conflict Tactics Scale (CTS), which was developed in the 1970s and is a behavioral checklist that asks research participants to indicate the methods or tactics that they have used to resolve a difference with an intimate partner – dating, married, or otherwise (Straus, 1979). A revised version of the CTS was developed in the 1990s (Straus et al., 1996), which continues to be used today. The CTS has been used in hundreds of studies, translated into dozens of languages (Straus, 2005), and administered on diverse populations (Straus, 2004, 2008a; Anderson and Leigh, 2010); at one point, it was estimated that the CTS was featured in six new family violence publications each month (Straus, 2005). The CTS that is used today measures five different tactics: (i) negotiation, (ii) psychological aggression (minor and severe), (iii) physical assault (minor and severe), (iv) physical injury (minor and severe), and (v) sexual coercion. Participants rate the extent to which they perpetrated or sustained any of the behaviors measured in the CTS within a specified time period, usually 12 months or lifetime experiences. Examples of behaviors that are measured include (i) negotiation – listening to a partner’s side of the story; (ii) psychological aggression – threatening to hurt a loved one or calling a partner names; (iii) physical assault – hitting/slapping a partner or beating up a partner; (iv) physical injury – leaving a small bruise or cut on a partner or needing medical attention; and (v) sexual coercion – using force or threats to make partner have sex (Straus et al., 1996).
The United States leads the world on the study of PV; however, well over 200 studies have been conducted on PV in countries outside the United States. The majority of these studies assess only female rather than male victimization (Esquivel-Santovena, Lambert, and Hamel, 2013). We focus on the prevalence rates established from two relatively recent, major, worldwide studies of PV: (i) WHO’s Multi-Country Study on Women’s Health and Domestic Violence Against Women, a 10-country study on PV against women (WHO, 2005a), and (ii) the International Dating Violence Study (IDVS), a 32-nation study of primarily heterosexual college student dating relationships, the only large-scale, cross-national study that assessed both men and women as potential perpetrators and victims of PV (Straus, 2008a, b). In addition, we discuss a major literature review on PV worldwide (Esquivel-Santovena, Lambert, and Hamel, 2013) to highlight key issues in assessing prevalence.
WHO collected data from 24,000 women in 10 countries with varied geographical, cultural, and regional setting (Bangladesh, Brazil, Ethiopia, Japan, Peru, Namibia, Samoa, Serbia and Montenegro, Thailand, and the United Republic of Tanzania). The underlying theoretical perspective guiding this study was that “violence against women is both a consequence and a cause of gender inequality” (WHO, 2005a, p. viii). Thus, only victimization of women was assessed. They found that the prevalence of lifetime physical PV ranged from a low of 13% in Japan to a high of 61% in Peru, with most cites ranging from 23% to 49%. For severe PV victimization (e.g., hit with a fist, kicked, dragged, choked, burned, having a weapon used against her), the prevalence ranged from 4% in Japan to 49% in Peru. For sexual PV, lifetime prevalence ranged from 6% in Japan and Serbia to 59% in Ethiopia, with most cites falling between 10% and 50%. Finally, for emotionally abusive or controlling acts (e.g., being insulted/made to feel bad about oneself, humiliated, intimidated, threatened, restricting access to family or friends), between 20% and 75% of women experienced one or more of these in the previous year (WHO, 2005a).
The WHO study examined only female victimization; the IDVS, however, assessed perpetration and victimization of both men and women. This is important because in both the United States (Langhinrichsen-Rohling et al., 2012) and other nations (Esquivel-Santovena, Lambert, and Hamel, 2013), bidirectional PV, when both partners are violence with each other, is the dominant pattern. Population-based studies show that PV is predominantly bidirectional in 15 countries (China, Hong Kong, Philippines, Thailand, Botswana, Namibia, Swaziland, Zimbabwe, Barbados, Brazil, Jamaica, Trinidad and Tobago, Portugal, the Ukraine, and the United States); in 11 countries, female victimization/male perpetration is the dominant pattern (Esquivel-Santovena, Lambert, and Hamel, 2013).
The IDVS represents over 136,000 university students in 32 nations; rates of any physical PV perpetration for men ranged from a low of 10% in Singapore to a high of 95.5% in Iran; for women, perpetration rates ranged from 15.8% in Malta to 71.4% in Iran. For severe PV perpetration, rates for men ranged from 0% in Malta and Singapore to 42.9% in South Africa; for women, perpetration rates ranged from 2% in Sweden to 25.7% in the Netherlands (Straus, 2008a). Additionally, at least 50% or more of the violent couples in 31 of the 32 nations examined showed some level of bidirectionality. Moreover, if just severe physical PV is assessed, data from 22 nations show that at least half of the severely violent couples were bidirectionally violent.
There are several major theoretical perspectives that guide the field’s understanding of PV. We review the most dominant of these theories here, but this discussion is not exhaustive of all theoretical perspectives.
Patriarchy theory is the dominant perspective on PV, both within Western nations (Dutton and Corvo, 2006) and around the world (WHO, 2005b). Patriarchal theorists hold that the cause of PV is the gendered structure of world societies. In all nations of the world, men have economic, political, social, and occupational power over women, and this power structure is reflected in heterosexual intimate relationships. This theory purports that men strategically use violence to maintain their dominant status over women (Dobash and Dobash, 1979). WHO (2005b) argues that “the abuse of women and girls, regardless of where it occurs, should be considered as ‘gender-based violence,’ as it largely stems from women’s subordinate status in society with regard to men” (p. 11).
There is limited empirical support for this theoretical perspective. A meta-analysis of studies addressing PV against women and the male partner’s patriarchal ideology (Sugarman and Frankel, 1996) provided little support for patriarchal theory. A more recent meta-analysis (Stith et al., 2004) found that traditional sex role ideology and PV against women were moderately associated. A cross-national literature review by WHO (2010) showed that “traditional gender norms and social norms supportive of violence” was not a consistent or strong predictor of PV against women. Moreover, there is an abundance of evidence that contradicts patriarchy theory: the consistent findings that women physically assault their partners at rates comparable to men (Archer, 2000) – including the large majority of nations assessed in the IDVS (Straus, 2008a); and, rates of PV in gay and lesbian couples are at least as high as in heterosexual couples (Hines, Malley-Morrison, and Dutton, 2013).
The IDVS assessed both men and women as potential PV perpetrators and victims; thus, it provides a more nuanced understanding of how dominance, subordination, and attitudes toward women and men are related to PV on a worldwide level. For example, Straus (2008a) found that the predominant pattern of physical PV in all nations was bidirectional; the next most common form of PV was female-only perpetrated. According to the participants’ self-reports, in 24 of 32 nations, women were more dominant on average than men within their romantic relationships, with little gender difference overall in dominance. Furthermore, dominance by either partner was associated with an increased probability of bidirectional PV, female-only PV, and male-only PV. Also using the IDVS, Hines (2007a) found that rates of sexual PV by intimate partners within nations were influenced by both the status of women in that nation (a composite measure of women’s political, educational, and workforce participation) and general levels of hostility toward members of the opposite sex, providing little evidence to support the patriarchy theory.
One of the most consistent findings in the research is that PV passes through the generations; children who are exposed to aggression in their families of origin are more likely to perpetrate or be victimized by aggression in their relationships as adults than children who are never exposed to familial aggression (Stith et al., 2000; Ehrensaft et al., 2003). This has been supported by a meta-analysis of 39 studies, which found a significant association between both the witnessing of interparental aggression and the experience of child abuse, with PV in adulthood among both genders in Western nations (Stith et al., 2000). WHO (2005a) reports on studies from other world regions, stating that exposure to violence during childhood increases men’s perpetration of PV by – three to four times (Gil-González et al., 2008); exposure to childhood sexual abuse triples men’s perpetration of sexual violence toward women (Jespersen, Lalumiere, and Seto, 2009); and childhood exposure to violence increases women’s victimization from both PV and sexual violence (Söchting, Fairbrother, and Koch, 2004; Martin, Taft, and Resick, 2007; Vung and Krantz, 2009). Further, using the IDVS data, researchers found that violence socialization in childhood was related to both victimization and perpetration of aggression in adulthood for both men and women (Douglas and Straus, 2006; Hines, 2007b; Gamez-Guadix, Straus, and Hershberger, in press).
The most often cited theory to explain the intergenerational transmission of PV is social learning theory (Bandura, 1979), which posits that when children witness individuals in their family rewarded for using aggression, they learn to resolve frustrations and conflicts with family members through the use of their own aggression (Eron, 1997). There is some evidence to support this theory, but not necessarily as the sole cause of the intergenerational transmission of IPV (Straus and Yodanis, 1996). Instead, attachment theorists claim that aggression witnessed in childhood may create an avoidant–ambivalent bonding style that persists into adulthood romantic relationships and leads to overly demanding and angry behavior with their partners (Dutton, 2007). Bowlby (1988) states that the attachment bond formed in infancy becomes the internal working model for all later significant relationships. Empirical studies show that individuals who are most at risk for expressing anger are those who are anxiously attached to their partners; these individuals usually experienced violence in their families of origin (Dutton, 2007). Attachment theory has been well established in Western nations, but there is debate about the applicability of this theory in other cultures (van Ijzendoorn and Sagi-Schwartz, 2008; Jin et al., 2012). Nevertheless, multinational data from the IDVS shows that higher rates of neglect during childhood are related to physical PV perpetration against dating partners (Straus and Savage, 2005).
A consistent predictor of PV is alcohol abuse. This link has been primarily demonstrated among samples in the United States, including male batterers (Fals-Stewart, 2003), community samples of men (Leonard, 1993), and population-based samples (Caetano, Schafer, and Cunradi, 2001). When male alcoholics remit from using alcohol through the administration of empirically based alcohol treatment programs, their rates of PV perpetration decrease significantly (O’Farrell et al., 2003). This finding demonstrates the importance of screening for and treating alcoholism in male batterers.
Alcohol abuse has been shown to predict PV perpetration on a multinational level as well. WHO (2010) reports that alcohol use is strongly associated with the perpetration of PV, including low- and middle-income countries. Overall, these studies show that the abuse of alcohol is associated with a 1.6–4.8 times greater likelihood of men perpetrating PV. The IDVS showed that across nations, there are no gender differences in the prediction of PV perpetration from alcohol abuse (Hines and Straus, 2007), showing that alcohol abuse equally predicts PV perpetration for both men and women.
There is also empirical support for the relationship between PV perpetration and personality disorders. For example, Dutton et al. (1994) found that several personality problems – borderline personality, antisocial personality, aggressive-sadistic personality, and passive-aggressive personality – are related to PV perpetration by men. It is argued that men with such mental health concerns have a fearful-angry attachment style, which results in them lashing out violently toward their female partners during confrontations and perceived separations. The relationship between attachment, personality dysfunction, and PV has been confirmed by others as well (Holtzworth-Munroe and Stuart, 1994).
WHO (2010) also identified antisocial personality as a contributor of male perpetration of both sexual violence and PV. It cites studies (two from high and one from middle-income countries) that provide support for this association. The IDVS provides evidence of associations between both antisocial personality (Hines and Straus, 2007) and borderline personality (Hines, 2008) with PV perpetration across all sites with similar strength of correlations for men and women.
Increasingly, researchers are using an ecological model to understand PV (Dutton, 1985; Gelles, 1998; Hines and Malley-Morrison, 2005). In this perspective, PV researchers tend to focus on four levels of influences, all of which have their roots in Bronfenbrenner’s (1979) work: (i) the individual’s unique bio-psycho-social characteristics that exist even before birth and that individuals bring to every interaction, (ii) the microsystem, which consists of the individual’s immediate settings (e.g., the home), (iii) the exosystem, which includes the larger neighborhood, mass media, and state agencies, and (iv) the macrosystem, which consists of broad cultural factors.
An ecological perspective conveys the notion that in order to understand how someone may commit PV, we need to understand the genetic endowments of that individual, the microsystem in which he/she was raised, the microsystem in which he/she is currently embedded, characteristics of the neighborhood within which he/she functions (including the availability of social services and the criminal justice system), and the larger society that either condones or condemns PV (Hines, Malley-Morrison, and Dutton, 2013). WHO (2010) also promotes an ecological model; however, it maintains its focus on female victims of male-perpetrated violence, and therefore, only focuses on elements of the patriarchal perspective at each level of the model (WHO, 2005b).
Since research on family violence was developed in the early 1970s (Steinmetz and Straus, 1974; Straus, 1976; Carroll, 1977; Steinmetz, 1977), researchers have documented a range of negative factors that are correlates of, or the potential consequence of, PV – including mental and physical health problems of victims, the well-being of children who witness PV, as well as the costs to society.
Community studies of female victims of PV show that they have poorer mental health than nonvictims (Coker et al., 2000; Weinbaum et al., 2001), and are at an increased risk for panic, depression, anxiety, sleep problems, posttraumatic stress disorder, and suicidal ideation (Leserman et al., 1998; Golding, 1999; Hathaway et al., 2000). They are also more likely to abuse alcohol and illegal and prescription drugs (Golding, 1999; Black and Breiding, 2008). Most of this research has been conducted on populations in Western nations, but WHO (2010) cites similar problems for women throughout the world; the research also references a lack of fertility control and the increased risk of HIV/AIDS to women who experience sexual assault at the hands of their male partners.
A small, but growing, body of research concerning males’ experiences with PV victimization and mental health concerns; results are similar to what has been found for females. Males from community samples who have experienced PV are more at risk for posttraumatic stress symptoms/disorder, alcohol abuse, substance abuse, depression, and anxiety (WHO, 2005a; Hines, 2007a; Kaura and Lohman, 2007; Romito and Grassi, 2007; Sabina and Straus, 2008; Hines and Douglas, 2011b; Hines and Douglas, 2012). Results from the IDVS found that male victims across nations experience suicidal ideation, and self-harm (Chan et al., 2008).
Some research has also compared the outcomes of men and women who have experienced PV victimization. US-based research has found that both women and men who experience physical PV report higher levels of psychological distress (Fortin et al., 2012), depressive symptoms, chronic mental illness, and illegal and prescription drug abuse (Coker et al., 2002; Carbone-Lopez, Kruttschnitt, and MacMillan, 2006). Further, research has shown that after controlling for lifetime exposure to aggression, there were no gender differences in the influence of PV on these mental health outcomes (Pimlott-Kubiak and Cortina, 2003). Similar results were found among a cohort study in New Zealand (Fergusson, Horwood, and Ridder, 2005). The IDVS has found that across the globe, men and women do not always have parallel correlates of victimization and mental health concerns, but that in general, victimization raises the likelihood of mental health concerns for all men and women (Chan et al., 2007; Sabina and Straus, 2008).
As a cautionary note, we remind readers that the research which has been conducted on health and mental health problems associated with PV victimization has been correlational in nature and has not established causal links between PV victimization and health concerns. It is possible that individuals who have poorer mental health are more likely to become engaged in violent relationships.
The study of children’s exposure to PV is a developing field. Most research focuses on child witnesses of male-to-female PV (Levendosky et al., 2003, 2006; Stover, Horn, and Lieberman, 2006; Heugten and Wilson, 2008; Spilsbury et al., 2008) using small program or community-based samples in Western nations (Lang and Stover, 2008; Owen et al., 2009), although a small body of research exists on non-Western nations/cultures (Speizer, 2010; Vameghi et al., 2010; Chan, 2011). Exposure to PV between one’s parents can include a wide range of behaviors, such as hearing PV, seeing it, calling for emergency services, being caught in the middle, seeing the destruction/injuries resulting from PV, or becoming a victim of PV (Fantuzzo, Mohr, and Noone, 2000; Martiny, 2006). Research has shown that being exposed to PV between parents can have a direct effect on children’s mental health (Sternberg et al., 2006), including externalizing problems such as disobedience, defiance, and aggressiveness (Kalil, 2003), and, internalizing problems such as depression, anxiety, (Clements, Oxtoby, and Ogle, 2008; Heugten and Wilson, 2008), and posttraumatic stress symptoms. (Kilpatrick and Williams, 1998; Lang and Stover, 2008). Other research has documented physical health problems in child witnesses, including maladaptive brain development (Gunnar and Barr, 1998), and increased heart rates and cortisol levels, which place children at risk for memory and attention problems (Saltzman, Holden, and Holahan, 2005). Differences in children’s responses to witnessing PV based on parental gender are inconclusive (Stover, Horn, and Lieberman, 2006; Clements, Oxtoby, and Ogle, 2008), but suggest that child witnesses of female-to-male PV may be at an increased risk for emotional and behavioral problems (Leisring, Dowd, and Rosenbaum, 2002) and be more aggressive toward romantic partners and peers in adolescence (Moretti et al., 2006).
Since the 1970s, a number of different types of services have been developed to help victims of PV (Higgins, 1978). Today, there are a number of ways in which victims seek help, including domestic violence shelters (sometimes called battered women’s shelters) (Bowker and Maurer, 1985) and hotlines/helplines specializing in domestic violence issues and concerns; these are common in most countries and can be one of the first low-cost steps taken to address PV (Mršević and Hughes, 1997; Bennett et al., 2004; Sun-Hee Park, 2005; Hines, Brown, and Dunning, 2007). The passage of the 1994 Violence Against Women Act in the United States helped to formally criminalize domestic violence (Crais, 2005), which means that victims in the United States commonly seek help from law enforcement officers (Buzawa and Austin, 1993; Shannon et al., 2006; Johnson, 2007), as they do in other countries (Stanko, 2000). Victims also often contact attorneys for advice on how to leave a violent relationship, how to document the abuse, and how to protect children in the process (Bowker, 1983; Erez and King, 2000; Stanko, 2000). PV victims seek help from health professionals, such as in emergency rooms, mental health professionals (Leone, Johnson, and Cohan, 2007; McNamara, Tamanini, and Pelletier-Walker, 2008), and members of the clergy (El-Khoury et al., 2004). Victims also use more informal sources of support, such as on the Internet (Web sites for information about PV, forums, LISTSERV, e-mail groups, etc.) (Douglas and Hines, 2011), and seek assistance from family and friends (Leone, Johnson, and Cohan, 2007; Douglas and Hines, 2011). WHO (2005a) states that many women tell family and friends about their experiences, but they rarely seek help from agencies, police, or other providers.
Women in Western nations have relatively positive experiences when seeking help for PV victimization. A study of women who sought services from a domestic violence shelter found that 89% of the clients believed that they were helped by the services that they received and a high proportion reported that they felt better because of these services (McNamara, Tamanini, and Pelletier-Walker, 2008). These findings are similar to a study which examined women’s impressions of a hospital-based DV support group (Norton and Schauer, 1997). These findings are consistent with other literature which states that women are often very satisfied with the services that they received for PV (Bowker and Maurer, 1985; Molina et al., 2009). Battered women also report being satisfied with the assistance that they receive from the police. One study of female PV victims indicated that the female victims found the police to be very helpful and the majority would contact the police again for assistance (Apsler, Cummins, and Carl, 2003). WHO (2010) reports that many similar efforts have or are being tried in less developed countries, but that women do not seek help very often. There are multiple barriers to seeking help for PV victimization, including cultural attitudes, not knowing where to obtain help, economic constraints, personal and family shame, and being revictimized when seeking help (Mason, 2009; Shen, 2011; Ogunsiji et al., 2012).
Men in Western nations report more difficulty in obtaining services. One study of men’s help-seeking experiences found that family and friends were overwhelmingly the most helpful resource to them in coping with their PV victimization (Douglas and Hines, 2011). Mental health and medical professionals were rated as being the most helpful formal resources; they took the male victims seriously and inquired about the origin of the men’s injuries. The resources providing the least support to men were those at the core of the DV service system: DV agencies, DV hotlines, and the police. Many men reported being turned away. This research is consistent with Cook (2009) who found that some DV hotline workers say that they only help women, infer/state that the men must be the actual instigators of the violence, ridicule them, or refer them to batterers’ programs. Some men report that when they call the police they sometimes fail to respond, ridicule them, or the men are incorrectly arrested as the primary aggressor. Other research has found that male victims do not feel that the police take their concerns seriously and are significantly less satisfied with the police response than female victims of PV (Buzawa and Austin, 1993). Other populations in Western nations which cite difficulty in obtaining assistance include gay men, lesbians, and older women; these populations report that services are not available for them or are not tailored to meet their specific needs (Renzetti, 1989; Donnelly, Cook, and Wilson, 1999; McClennen, Summers, and Vaughan, 2002; Beaulaurier et al., 2007). This research is consistent with the analyses of US DV agencies, which report being less able to support older men and women, adolescents, and gays and lesbians (Hines and Douglas, 2011a).
A small body of research has estimated the costs to society (other than children) associated with PV, such as within the health care, social service, and criminal justice systems (Max et al., 2004; Rivara et al., 2007; Chan and Cho, 2010; Fishman et al., 2010; Logan, Walker, and Hoyt, 2012). Several studies have demonstrated the increased costs within the health-care system to individuals, health insurance companies, government-sponsored insurance, and health-care facilities. For example, the utilization and cost of US women who experience PV victimization is 1.6–2.3 times higher than women who do not experience PV victimization (Ulrich et al., 2003). One study estimated these costs to be $1700 higher over a 3-year period of time, per individual female victim (Jones et al., 2006); another study estimated the total cost of PV against all female victims to be between 2.3 and 7.0 billion dollars – depending on how it is calculated (Brown, Finkelstein, and Mercy, 2008). Similar results have been found in other countries as well. A Danish study found higher rates of health-care use and costs for both men and women who had experienced PV victimization (Helweg-Larsen et al., 2011).
PV has additional costs to society. Women who experience PV are more likely to miss days at work and to be distracted when they are at work, than women who have not experienced PV victimization (Reeves and O’Leary-Kelly, 2007). There are also significant tolls on the criminal justice system, as they process and enforce restraining orders, investigate and prosecute cases of assault, stalking, rape, and murder. These costs have estimated to be around $8.3 billion a year in the United States (Max et al., 2004). Another US study estimated the combined costs of PV to women and society in a 6-month period of time to be $17,497. This calculation included health and mental health services, victim services, lost time from work, family, etc., lost property and transportation, health-related quality of life costs, and legal/criminal justice system costs (Logan, Walker, and Hoyt, 2012). A British study concerning the combined costs of PV to society, including productivity, the criminal justice system, health care, social services, housing, and legal fees, amounted to £23 billion (~38 billion US dollars) in 2008 (Walby, 2009). A similar study of costs to the Australian society estimated the total costs of health, productivity, consumption, administration, and costs to children to be about 15.6 million Australian dollars (~14 million US dollars) a year (Commonwealth of Australia, 2009). Most of the research concerning the costs to society have focused on female victimization in Western nations; WHO (2004) comes to similar conclusions, although it does not cite research from non-Western nations.
On a worldwide level, the theory that has the strongest influence on policy, prevention, and offender treatment is patriarchal theory. Both the United Nations and WHO have issued recommendations for policy review and development, primary prevention programming, and treatment protocols that are consistent with this theory. They state that violence against women is rooted in gender biases, stereotypes, and socializations, and the United Nations calls upon all nations to adopt national action plans to prevent violence against women (WHO, 2005b; United Nations Women, 2012).
An initial problem with the framing of PV as a form of violence against women is that PV is perpetrated by both men and women toward both men and women; therefore, it is not an accurate or encompassing definition of the problem of PV. Second, lumping all forms of violence that can be perpetrated against women into the catchall phrase, violence against women, is misguided because it assumes that PV has similar etiologies and can be remediated with similar tactics; yet, there is no evidence for this assumption (Dutton, 2012).
Both the United Nations and WHO have had considerable influence on the development of policies throughout the world. The United Nations Women (2012) has called upon all countries to review and revise their existing legislation to conform to international law. At least 115 governments have revised or adopted laws to address violence against women or PV specifically (Ortiz-Barreda, Vives-Cases, and Gil-González, 2011). These include the Violence Against Women Act in the United States, the Law on the Protection of Women’s Rights and Interests in China (Zhang, 2009), and the Protection of Women from Domestic Violence Act in India (Germain, 2007). Moreover, the African Charter on Human and People’s Rights on the Rights of Women, the Council of Europe, the European Parliament of the European Union, and the Inter-American Convention on the Prevention, Punishment, and Eradication of Violence Against Women have urged all member states to take necessary action to protect women and eradicate violence against women (United Nations Women, 2012).
Although the revision and development of policy to prevent violence is good practice, the legislation has been developed under the misleading premise that PV only happens to women in heterosexual relationships. There is a danger to forming policy that protects only women from PV. Such policy “may implicitly encourage a lack of understanding and relaxed response to other types of physical assault towards intimate partners, such as same sex, female to male, or reciprocal PV” (Dixon and Graham-Kevan, 2011, p. 1152). In addition, without acknowledging that same-sex, female-to-male, and reciprocal violence occurs, a nation is not protecting a substantial portion of PV victims.
As part of national action plans, the United Nations Women (2012) urges that all nations develop prevention strategies. They explicitly state that the prevention of and response to violence against women “will necessarily be distinct to other forms of violence” (p. 12), despite overwhelming evidence that PV is linked to other forms of violence (Malone, Tyree, and O’Leary, 1989; Giordano et al., 1999; Felson, 2002; Ehrensaft, Moffitt, and Caspi, 2004; WHO, 2005b; Medeiros and Straus, 2006). Both the United Nations and WHO call for prevention plans to address the root causes of violence against women, which they say are “gender inequality, gendered social constructions, and inadequacies in education” (United Nations Women, 2012, p. 32). Thus, a crucial element in the prevention of PV is increasing the status of women in employment, education, political participation, and legal rights. All such prevention programming should take place both universally and in targeted groups, such as primary and secondary educational institutions, workplaces, the military, faith-based and cultural institutions, and all professionals who may have contact with women victims (WHO, 2005b; United Nations Women, 2012).
One could argue that such widespread calls for this type of prevention programming are premature, as there is little evidence that prevention programs that take these strategies work to reduce PV, particularly all forms of PV. For example, evaluations of programs that take a patriarchal perspective, such as Skills for Violence-Free Relationships (Levy, 1984), show that they do not work in changing attitudes or knowledge regarding PV, either in the short- or long term (Avery-Leaf and Cascardi, 2002). One large-scale review of PV prevention studies between the years 1993 and 2012 found that there were 19 experimental or quasi-experimental studies, only 9 of which were methodologically sound (Whitaker et al., 2013). They found that the Safe Dates program in schools was one of the only prevention programs that reduced PV perpetration and victimization over 4 years. It was also equally effective for boys and girls, across races/ethnicities, and for adolescents who had previously experienced PV and those who had not. The Safe Dates program is a gender-neutral, mixed-gender program that includes a theater production, a 10-session curriculum, and a poster contest (Foshee and Langwick, 2004).
Another consideration with the United Nations and WHO mandates is that they overlook a major portion of PV that is experienced by people worldwide. Prevention programming should accurately reflect what the majority of any population experiences; “otherwise the message may not be internalized by the majority as something that applies to them” (Dixon and Graham-Kevan, 2011, p. 1151). When prevention campaigns contain messages that PV is severe, unidirectional violence that men perpetrate toward women, then victims who experience other forms of PV might think that their experiences do not fall under the rubric of PV (Dixon and Graham-Kevan, 2011). Thus, we argue for public education which states that all forms of violence are unacceptable, so that we reduce the rates, severity, and frequency of all forms of PV.
The United Nations Women (2012) says that intervention programs for male batterers should be part of every nation’s national action plan. Such programs should have minimum standards and be reviewed by women’s NGOs and female survivors. The minimum standards include “a programme commitment to work within a gendered structural analysis of violence against women” (p. 59), despite the evidence that programs with such an orientation have been shown to be ineffective (Davis, Taylor, and Maxwell, 1998; Babcock, Green, and Robie, 2004).
Nevertheless, Batterer Intervention Programs (BIPs) that are built around this gender paradigm are the main treatment programs mandated by law in many nations. In the United States, BIPs are the primary diversionary programs for batterers and are mandated by most state laws. In Great Britain, PV is considered a special case, unrelated to other forms of violence, and thus, general violence programs are unavailable to or explicitly exclude PV perpetrators; perpetrators are routed to BIPs there as well (Dixon, Archer, and Graham-Kevan, 2012). Because of the strict adherence to programs that do not work, some researchers have argued that the primary goal of these programs is not to change perpetrator’s behavior or keep victims safe, but rather to deconstruct male privilege in an effort to reeducate the male participants (Corvo, Dutton, and Chen, 2009). BIPs assume that PV is the result of underlying gender biases, stereotypes, and socialization. This assumption may be one of the reasons why BIPs are ineffective. We argue that PV perpetrators – both men and women – need to be assessed for the wide range of potential risk factors we have already discussed, including alcohol abuse, impulsivity, personality disorders, history of trauma and abuse, poor emotional regulation, coping skills, hostility, etc. (Dixon, Archer, and Graham-Kevan, 2012), in order to design/select treatment programs that will be most suitable to perpetrators. We also support assessing the behavior of both partners, which is consistent with systems theory (Dixon, Archer, and Graham-Kevan, 2012). Given that there is little evidence of any substantial sex differences in the risk factors or motivations for PV (Moffitt et al., 2001), it is likely that programs can be developed for both male and female perpetrators. Current evidence suggests that the best programs focus on constructs such as readiness to change and motivational enhancement (Eckhardt et al., 2013).
As noted throughout this chapter, we disagree with and do not think that there is evidence to support the dominant story of PV, that it is something which only men do to women. Thus, we maintain the need for a broader and more comprehensive understanding of PV, and one that is based in evidence. In this regard, we encourage researchers to examine victimization and perpetration among both men and women; for public education campaigns to acknowledge the bidirectionality of PV as a way for victims and perpetrators to accurately name and identify their experiences; for human service providers to reach out and to serve all individuals who seek assistance; and for policy on all levels – from agency policy to national policy – to be gender-inclusive and to be based in research and evidence, so that we as a global society maximize our potential to reduce PV regardless of gender, class, sexual orientation, race/ethnicity, or nationality.