Engendering Violence: The Acceptability of Domestic Violence as a Determinant of Intrastate Conflict

Abstract: 

Recent studies indicate that women’s status may be a factor in the likelihood of conflict. Prior work illustrates that women and men typically hold different attitudes and behaviors regarding war and the use of violence. Much of the literature linking women’s equality to violent conflict, however, has focused on women’s experiences or roles in society rather than the attitudes towards violence against women. This project contributes to the literature by analyzing the influence of attitudes regarding domestic violence upon the likelihood of civil conflict, ultimately finding support for the hypotheses that women’s attitudes towards domestic violence exhibit a positive relationship with the occurrence of civil conflict and its frequency. This study illustrates that a meaningful connection exists between attitudes towards household violence against women and the likelihood of civil conflict.

Table of Contents: 

    Introduction

    In the Western tradition, a long-held connection between men and war and, concomitantly, of women and peace, is evident. Indeed, the Ancient Greek god of war was personified as the masculine Mars, while the goddess of peace was embodied as a woman, Irene. The legacy of such a gender division can be identified today through attitudinal surveys illustrating male support for and female aversion to violent conflict (Smith, 1984; Wilcox, Conover & Shapiro, 1993; Togeby, 1994; Hewitt & Allsop, 1996). Men, however, are consistently overrepresented in the decision-making process of both war and peace, while women’s presence has been strikingly absent (Tickner, 1992; Hudson, 2009). This gender disparity has repercussions, not only because women are also affected by the presence of war, but because recent studies indicate that women’s status may be a factor in the likelihood of war and the stability of peace (Caprioli, 2000, 2003, 2005; Caprioli & Boyer, 2001; Regan & Paskeviciute, 2003; Melander, 2005a; Hudson, Caprioli, Ballif-Spanvill, McDermott, & Emmett, 2009; Hudson, Caprioli, Ballif-Spanvill, & Emmett, 2012).

    To understand women’s connection with conflict and peace, we must understand that war is often conceptualized as a male domain: a byproduct of male testosterone and aggression. Men predominantly initiate conflict and promote escalation to violence (Caprioli & Boyer, 2001), in addition to making up the vast majority of armed combatants and peacekeepers (Hudson, 2009). Due to this disproportionate male involvement, women’s contributions and concerns are often discouraged or blatantly ignored (Hudson, 2009). Yet, studies have indicated a relationship between higher levels of domestic gender equality or inclusion and reduced levels of militarized conflict, indicating that ignoring the voices of women can have devastating results (Caprioli, 2000, 2003, 2005; Caprioli & Boyer, 2001; Regan & Paskeviciute, 2003; Melander, 2005a; Hudson et al., 2009, 2012). The presence of female peacemakers has also been found to be associated with a higher likelihood of sustainable peace (Hudson, 2009).

    The inquiry into women’s roles in conflict and peace has often been neglected, leaving an important aspect of the field understudied. One feminist peace researcher maintains that sexism and war are “twin manifestations of the same underlying cause” (Reardon, 1985, p. 2). If this assertion is indeed true, a large gap remains in the conflict literature. Thus, a gendered perspective could do much to enhance our understanding of the field. This project draws on the constructivist argument, maintaining that gender is socially constructed through a process of socialization. This perspective helps us to understand gender violence, specifically the likelihood of violent conflict occurring within a society. Assuming that acceptance of violence against women within the household drives and reinforces the broader use of violence in war and conflict (Hudson et al., 2012), we expect to find that the more a society holds attitudes accepting domestic violence, the higher the likelihood that these nations will experience civil conflict. This paper will first identify common explanations and causal factors influencing conflict, and then discuss recent feminist researchers’ attempts to incorporate women’s perspectives into peace and conflict studies.

    Literature Review

    The study of how civil wars begin is not a recent inquiry. Studies have identified contributing factors of civil war in concepts such as high income inequality and the deprivation of political rights (Collier & Hoeffler, 2004). Boix (2008) further develops the theory that the emergence of civil conflict can be attributed to economic inequality. There is a strong case that diverse ethnic identities and religious associations, which can create divisions in society, can cause civil conflict (Fearon & Laitin, 2000; Horowitz, 2000; Collier & Hoeffler, 2004). Other researchers have identified links between geographical characteristics, such as adjacency of international borders (Buhaug & Gates, 2002), natural resources (Buhaug & Gates, 2002), and, in particular, oil (Fearon, 2005). Yet, these studies have traditionally ignored the voices of women (Gizelis, 2011; Sjoberg, 2013), and when acknowledged, viewed women merely as passive victims without recognition of their agency (Hudson, 2009). The gender perspective on the study of peace and conflict is fairly recent and serves as the theoretical foundation for this project.

    The Feminist Perspective on the Emergence of War

    Recent feminist literature has been directed toward answering questions of war and peace and explaining these issues from a gendered perspective (Tickner, 1992, 2001). The feminist perspective focuses on how war, conflict, and peace emerge from patriarchy. Feminist peace scholars maintain that patriarchy is entrenched in the most basic of human relations and prescribes male and female roles, as well as violence within the household (Reardon, 1985; Hudson et al., 2009). While the emergence of war cannot be attributed to gender relations alone, it has been shown to be an influence (Sjoberg, 2013). Some researchers have posited that the unequal relationship between men and women creates a hierarchical structure at this fundamental level of human relationships, thus predisposing a significant aspect of a nation’s behavior leading into war (Hudson et al., 2009).

    As will be discussed below, men and women hold different attitudes due in large part to their social experiences. In fact, these experiences are mandated in some case by law. For example, men of a certain age may be drafted, while women may be barred from certain positions in the military. This scenario thus prevents women from attaining the experience necessary for expertise in military affairs and decision-making. In addition to having different experiences with regard to war, studies show that women and men hold divergent attitudes on the acceptability of the use of force. Women’s attitudes, in general, tend to be more reserved about accepting the use of violence to deal with international conflict while men typically show more support for the use of force (Smith, 1984; Conover et al., 1993; Togeby, 1994; Wilcox et al., 1996). While some researchers stress that gender is an important difference, it is not the sole influence on attitudes towards violence. Factors such as political ideology can influence support for the use of violence in conflictual situations as well (Conover et al., 1993). Nonetheless, gender has been shown to be a factor in influencing one’s support of the use of violence to resolve political conflict.

    This ‘gender gap’ in attitudes has been explained in various ways. Some suggest it may be attributed to women’s ignorance of foreign affairs (and thus an ignorance of when force is necessary) or the fact that in many places, women tend to be poorer economically (a situation that would lead them to support the funding of social services as opposed to costly wars) (Wilcox et al., 1996). Other studies (Smith, 1984) find a similar pattern of male and female attitudinal differences when respondents are asked their views towards the use of violence in general, non-policy related questions. This would seem to support the argument that men and women maintain different values relating to violence in general. These attitudes are then consistent with responses regarding the use of force militarily. To explain this discrepancy, it is assumed that men and women are either inherently different or are socialized in distinct ways which result in such attitudes (Melander, 2005a). While much of the literature confirms a relationship between female equality and decreased state militarism, this disparity is often attributed to one of two competing explanations akin to those of “nature” or “nurture” debate – that of the “essentialist” or “constructivist” perspectives.

    Essentialist and Constructivist Approaches

    The essentialist explanation argues that the biological and innate characteristics of women make them predisposed to cooperation rather than competition (Melander, 2005a). The traditional hunter-gatherer perspective identifies women as caregivers who remain in the home, providing emotional support to their husband and children. Under this paradigm, men are considered hunters characterized by strength, aggression, and domination (Goffman, 1977). Male and female behavior in war, then, will follow a similar pattern where women are viewed as nurturing and cooperative while men are considered aggressive and warlike. Indeed, many women’s peace groups have drawn on their unique positions in society, particularly as mothers, to advocate for nonviolence (Segal, 2008). According to this framework, behavioral and attitudinal variance between males and females simply reflects a natural difference between the sexes at a biological level.

    Other researchers reject this formulation and find essentialist explanations of male and female variance to be too universalistic to explain the complex phenomena of gender and war (Hudson, 2009). They argue that while the attitudinal gender gap may be instinctual, it is not innate (Hudson et al., 2009). According to this explanation, men’s and women’s divergent attitudes are ascribed by socialization through society’s values of femininity and masculinity. For instance, beginning at a young age, boys are exposed to war through toy soldiers and weaponry and are expected to participate in rough play. This exposure conditions not only aggression in boys, but also normalizes it throughout society (Goffman, 1977). Similarly, girls are expected to be passive and submissive, fulfilling a caregiving role for their families (Goffman, 1977). It is in these societies that female subordination occurs due to the inferior position of women and women’s needs within this male-female hierarchy (Hudson et al., 2009). Some researchers go so far as to assert that essentialist formulations of women as inherently peaceful and averse to conflict  has served to “disempower women and keep them in their place” (Tickner, 1999, p. 8). This association is normalized within broader society to such a degree that women will essentially be prohibited from the decision-making process of war, as they are considered unfit to handle such topics.

    Conover and Sapiro (1993) illustrate the need to understand the gender gap in terms of the constructivist conception. The researchers found that women were not averse to promoting violence under certain conditions; instead, their threshold for advocating force was higher than for men. Women were simply more likely to advocate non-violent, cooperative solutions before resorting to violence. The researchers concluded, “The point is not that women learn early in life never to engage in conflict nor use violence, but rather that they learn to put off the use of violence until later in the course of a conflict than do men, to escalate its use more slowly, and to be more emotionally upset by it” (Conover & Sapiro, 1993, p. 1096). This concurs with Segal’s (2008) caution against broad generalizations of the gender gap in conflict, as women have often been just as complicit in their support of wars as have men.

    Gendered norms and hierarchies are taught to young children and reinforced through gendered experiences. Adult men and women have vastly different experiences in war. Women, for instance, are less likely to be injured as combatants but more likely to constitute civilian refugees (Enloe, 1990) and victims of wartime sexual violence (Seifert, 1996). These divergent experiences may make women more opposed to war and more invested in facilitating a durable peace. Second, women’s frequent position as noncombatants may appear less threatening and this in turn may strengthen trust with the enemy, a necessary component of building peace (Demeritt, Kelly & Nichols, 2014). The effectiveness of emerging women’s civil society and peace groups lends further support to this relationship (Segal, 2008; Gizelis, 2011), though many of these are nongovernmental, grassroots organizations (Regan et al., 2003). This finding lends further support to the idea that women are still often marginalized from the formal political processes of war and peace.

    According to the constructivist viewpoint, the presence of war and peace can be ascribed, to a certain extent, to the value of gender equality within a nation. According to the essentialist explanation, the occurrence of war and peace can be accounted for by the presence of (or lack thereof) women and their innate cooperative qualities. Through either the essentialist or constructivist explanation, researchers have identified a link between gendered instincts and/or social constructivism and attitudes towards war, thus addressing the question of why the inclusion of women in conflict decision-making processes has a pacifying effect (Regan et al., 2003).

    However, the essentialist explanation may be inadequate in recognizing the varied experiences amongst women participants in war. Indeed, women have been involved in wars as combatants and aggressors (Segal, 2008; Hudson, 2009), illustrating that women are not always averse to the use of violence. While the link between women and preferences for peace may be strong, it does not tell the whole story. Thus, this project aims to emphasize the role of constructivism in accounting for the complex relationship between gender and conflict. A main assumption of this theory is that violence is not a manifestation of some inherently male characteristic; rather, it is the presence of a gendered hierarchy that has established one sex over the other which promotes violence. An attitudinal survey covering four Middle Eastern nations supports this claim regarding gender and violence. The study discovered that men and women were more likely to disagree with the use of force equally when both groups favored gender equality (Tessler & Warriner, 1997). This survey indicates that it is not women’s innate preferences that makes peace more likely, but instead societal attitudes towards equality. The equality between men and women serves as a foundation for accepting equality among other groups of people in the national and international sphere. Without holding notions of equality, it is argued, one is likely to be more willing to use violence against their opponent.

    Addressing the Literature Gap through Feminist Theory

    Despite the substantial amount of extant literature on the topic of the gender gap regarding both behavior and attitudes, much of the prior scholarly work considered only men’s and women’s distinct behavior and its influence on conflict. An empirical study of internalized attitudes is yet to be done, and this paper will identify the ways in which women’s acceptance of inequality between the sexes can influence civil conflict. In the process of operationalizing the terms “gender equality” or “women’s rights,” previous studies have utilized various conceptualizations, including women’s political representation (Caprioli, 2000; Caprioli & Boyer, 2001; Regan & Paskeviciute, 2003; Melander, 2005a,b), prevalence of female heads of state (Caprioli, 2003, Melander, 2005a,b), duration of female suffrage (Caprioli, 2000), women’s access to education (Melander, 2005a), women in the labor force (Regan et al 2003), women’s physical security (Hudson et al., 2009), and women’s fertility rate (Caprioli, 2000; Regan & Paskeviciute, 2003). Statistical analyses were then conducted to determine their relationship with the likelihood of intrastate armed conflict (Melander, 2005a), state militarism (Caprioli, 2000; Regan et al., 2003), severity of violence (Caprioli et al., 2001), state human rights abuse (Melander, 2005b), the first use of force (Caprioli, 2003), and overall peacefulness (Hudson et al., 2009). While research has documented the negative relationship between female empowerment and conflict, to date no study has examined the link between attitudes towards gender equality (and attitudes towards domestic violence against women) in relation to such conflict.

    Understanding attitudes is a critical step because reports of violence alone do not tell us about the attitudes towards or the acceptability of such acts. If the constructivist explanation is correct in asserting that gendered hierarchies create the violence necessary for war, then we would expect to see that states with greater acceptance of inequality will have a stronger hierarchy, thus increasing the likelihood of violent civil conflict. Because attitudes and values are often deeply ingrained, the chances of a society rejecting such aggressive behavior are unlikely. Thus, violent tendencies may be difficult to overcome during efforts to promote stable peace, leading us to posit that states where violence against women is seen as acceptable by a larger proportion of the population would have a higher frequency of such conflict. Examining gender violence and gender inequality through attitudinal survey data allows for an understanding of the extent to which societies find such violence acceptable, or to what degree gendered hierarchies have been established. Assuming Hudson et al.’s (2012) view that violence within the household strengthens the use of force in interstate conflict, we would expect to find that the more a society holds attitudes accepting violence against women, the more likely these nations will be to accept violence on a broader, national scale. For the purposes of this study, we will focus on the outbreak of civil conflict and its frequency.

    We seek to add to the existing literature in several important ways. While domestic violence can be an accurate gauge of the prevalence of a gendered hierarchy, domestic violence is only one of its manifestations. Not all prejudice against women can be captured through domestic violence, so it is important to examine attitudes directly. Developing a better understanding of the acceptability of such violence and inequality will allow for great insight into the degree to which a gendered hierarchy has been established in societies. It is this hierarchy which is responsible for causing the necessary inequity for both violence within the household and violence within broader society. Through an examination of women’s personal attitudes towards violence against women, we can better understand to what degree women have accepted their inferior position, exposing the degree to which sexism and hierarchical violence pervade the society. Through the use of attitudinal survey data, we can understand gendered hierarchy directly, without speculation or having to use indirect measures such as domestic violence.

    Theory

    This study draws on constructivist theory to understand war and peace through a gender perspective. As noted previously, this explanation holds that if there are more women in positions of power, society is relatively more accepting of gender equality. In other words, such a society accepts an absence of a gendered division of labor regarding political power and decision-making. A single female leader may be an exception (she may be a proxy for a man, such as a widow or daughter), but a larger proportion of women in positions of political power suggests there is something more fundamental about the equality of the role of women and men. It would follow that states emphasizing gender equality tend to reduce reliance on traditional gender norms and incorporate women into the decision-making process (Melander, 2005a). In addition, it is societies that value equality and tolerance between the sexes that would extend such equality and tolerance to others when involved in conflict. This reduces the likelihood of violence occurring (Hudson et al., 2009). According to this explanation, a state in which women are unequal with men and less visible as decision-makers in politics would presumably experience a higher frequency of civil conflict not because of the lack of women’s contribution per se, but due to the lack of values of equality and tolerance among those facilitating peace (values which typically allow women to attain positions of power). This is not to negate the importance of women in ensuring peace, but to underscore the importance of broader, more fundamental female empowerment within society.

    One underlying cause of conflict, then, is misogyny and other similar ideologies that prioritize one group of people over another (Reardon, 1985; Hudson et al., 2009). Tickner (1992) argues that sexual hierarchies are not the only form of hierarchy, but one of the most fundamental and widespread. We would argue that other forms of inequality and oppression (ethnic, religious, etc.) should lead to similar levels of violence, but for the purposes of this study we focus solely on gender inequality.

    This gendered hierarchy has two implications. First, this account explains why women, who typically hold more pacifistic views, are largely ignored in foreign policy decisions (Caprioli, 2003). Second, the type of inequality that situates women at a lower position on the gendered hierarchy will lead to replication of such inequality on national and international levels. In this sense, the goals of feminism are intertwined with the goals of peace research; sexist hierarchy is considered the root cause of violence as it normalizes the use of force at the most fundamental of human relationships, within the household. This violence is then transposed to states’ relations in the national and international sphere (Caprioli, 2003; Hudson et al., 2009, 2012). According to this perspective, war is simply the apex of violence that has been fostered within domestic relations due to inequity and intolerance, as it is gendered hierarchies that create the violence necessary to cause war (Reardon, 1985; Caprioli, 2003).

    Studying attitudes in particular is important to this analysis because the acceptance of gender inequality, from a constructivist perspective, forms a hierarchy that normalizes violence within a society. Attitudes towards gender equality, therefore, are a significant element of the constructivist theory that has been largely unstudied. This gendered hierarchy is indicative of an unequal society, which then manifests itself in both domestic violence and civil violence. The acceptance of inequality between human beings makes it easier to perceive enemies as less valuable people, thus justifying violent attacks on them. This is contrasted with a cooperative approach that recognizes the equality of all parties involved, an approach that is generally considered feminine.

    When an individual supports gender equality, it would be expected that their level of support for violence would decrease, regardless of their gender (Tessler & Warriner, 1997). When a society does not value gender equality, this hierarchy will be more prevalent, and thus we would expect both men and women to be more accepting of violence. Of particular interest is the degree to which a high level of gender inequality in society affects women’s attitudes towards violence. Ideally, this study would incorporate an analysis of both men’s and women’s attitudes towards domestic violence, but due to data availability, this project will analyze only women’s responses.

    What if this inequality is so pervasive that women are accepting of high levels of violence, even when such violence is perpetrated against them and other women in the household? If women are indeed inferior on the gendered hierarchy, their negative attitudes towards gender equality and violence against women should be indicative of a strongly established hierarchy within society. By extension of this argument, we would expect that in societies in which women are accepting of high levels of domestic violence, the use of violence is then normalized in society, creating a more belligerent society willing to use violence in civil conflict. Thus, women’s acceptance of gender violence in the home should be associated with higher levels of violent conflict and a higher frequency of civil war. 

    This underscores the necessity of analyzing women’s attitudes in order to understand gender inequality in relation to violence. This includes not only violence between men and women, but also the use of violence to address social conflict and societal disputes that may eventually lead to war. When women accept violence, even when perpetrated against them, we would expect that the gendered hierarchy in a society is so ingrained that inequality and violence have been normalized. These observations lead us to the following two hypotheses:

    H1: The higher the level of women’s acceptance of violence within the household, the higher likelihood of the onset of civil war.

    H2: The higher the level of women’s acceptance of violence within the household, the higher the frequency of civil war (per year).

    Research Design

    Sample

    To assess the relationship between women’s attitudes toward domestic violence and the onset and frequency of civil conflict, this study draws on two main sources of data: the Uppsala Conflict Data Program’s Onset of Intrastate Armed Conflict dataset and USAID’s Demographic and Health Surveys (DHS) for data on the onset of civil conflict and survey data on women’s attitudes towards violence, respectively. Using a country year unit of analysis, the temporal domain of this study is 2000 to 2013. The scope of the study was limited to post-2000 due to the availability of DHS data and to 2013 due to the range of UCDP data. This quantitative analysis included cross-sectional data from 52 countries, consisting of a total of 146 observations spanning Eastern Europe, Central and South America, the Caribbean, Africa, the Middle East, and Asia. The spatial domain was selected due to the availability of DHS survey data and provides an optimal range of cases, as gender violence and violent conflict are not limited to a specific region.

    Dependent Variables

    This project incorporates two dependent variables, the onset of civil conflict and the frequency of civil conflict. The onset of civil conflict serves as the dependent variable for H1 and is operationalized using data from the UCDP Onset of Intrastate Armed Conflict dataset. UCDP terms armed conflict as “a contested incompatibility that concerns government and/or territory where the use of armed force between two parties, of which at least one is the government of a state, results in at least 25 battle-related deaths in one calendar year” (Themnér & Wallensteen, 2012). While other datasets may establish the threshold for conflict at a higher number of deaths, (1,000, for instance), this project will accept the UCDP’s definition, because it captures variance among conflict, as it encompasses a larger, more inclusive range of data than alternative definitions. The onset of civil conflict will be coded as a binary variable; each country that experiences a civil conflict within a given year will be coded as 1, and those countries where civil conflict has not occurred will be coded as 0. This project’s second dependent variable, the frequency of civil conflict, will be used to measure H2. Frequency of civil conflict similarly utilizes the UCDP Onset of Intrastate Armed Conflict dataset, but uses the ordinal variable of incidents of civil conflict per year for measurement.

    Independent Variables

    Variables analyzing women’s attitudes towards domestic violence serve as the primary independent variables of interest for this study. These variables were derived from DHS survey data. The DHS is one of the paramount sources for demographic data and includes several survey questions relating to women’s attitudes towards domestic violence.

    Although it would be optimal to analyze the discrepancy between men’s and women’s attitudes towards such violence, analyses of this nature are not currently feasible due to the lack of availability of men’s responses. Nevertheless, women’s attitudes towards domestic violence (perpetrated by men against women) are paramount to understanding the establishment of sexist hierarchy in society. Consistent with our theory, if women are at the bottom of the gendered hierarchy, their acceptance of violence in general and their specific acceptance of violence against women should be indicative of a strongly established hierarchy. In order to measure women’s attitudes towards domestic violence, five DHS survey items were utilized to capture opinions on the acceptability of domestic violence in certain situations. The survey asks women if it is justifiable for a husband to beat or hit his wife if she: 1) burns the food, 2) argues with him, 3) goes out without telling him, 4) neglects the children, and 5) refuses to have sex with him. Participants responded to these questions with either yes or no, depending on whether or not they agreed with the statement. These numbers were then aggregated per country, thus a 39.9 for Uganda in 2006 indicates that in that year, 39.9 percent of women surveyed agreed with a given statement. These numbers measure women’s acceptance of domestic violence on a state level, as it aggregates survey responses per country to allow researchers to understand the percentage of women in society who agree with the statement. Since the DHS contains aggregated survey data per country, this necessitates a country-year unit of analysis. DHS data that corresponded with the country-year data from UCDP was incorporated to conduct the quantitative analyses.

    Control Variables                                                                                                                             

    Factors commonly linked to civil conflict were controlled for, such as poverty, level of democracy, male gender skew, and degree of women’s political representation. Child marriage was also included as a control, as early ages of female marriage is an indicator of gender inequality and evidence of a gendered hierarchy within society. Thus we expected this variable to have a negative relationship with the dependent variables. Poverty data was derived from United Nations poverty data on the percentage of the population below national poverty lines. Consistent with prior work, we expected this variable to positively correlate with both the onset of civil conflict and its frequency. Democracy was controlled for using the Polity IV Project’s dataset, which scores countries on a scale of -10 to 10 depending on their level of democracy, with higher scores indicating stronger levels of democracy and lower levels representing lower scores. Democracy was assumed to have a negative relationship to conflict. Male gender skew was captured using World Bank data; this figure indicates the percentage of a country’s population that is female. Women’s political representation was also drawn from World Bank data and represented the percent of females in national legislatures. Both World Bank variables were theorized to have a negative association with the dependent variables. Our final control, child marriage, used measures from the WomanStats dataset, which ranked countries on a scale of 0-4, with higher scores indicating higher levels of child marriage and lower ages at female marriage. This control was similarly expected to have a negative relationship with the dependent variables.

    Methodology

    Logistic regression was used to test H1 (the higher the level of women’s acceptance of violence within the household, the higher likelihood of the onset of civil war). This method was chosen as the distribution of the dependent variable (onset of civil conflict) ranges from 0 to 1, thus necessitating a logistic model. H2 (the higher the level of women’s acceptance of violence within the household, the higher frequency of civil war (per year) was tested with an ordered logistic model due to the distribution of the frequency of civil conflict data, which ranges from 0-7 in incidents of civil conflict. Due to the limited number of observations, missing data was accounted for through interpolation, including computing averages for each country to make up for missing DHS survey year data; the values for one country were averaged to carry across many years. Although this method did incorporate assumptions about the variability of such data, we ultimately found that it was necessary in order to obtain an appropriate sample size necessary to conduct analyses. Further, this interpolation is justified in that we expect that these attitudes should not vary greatly across time.

    Results and Analysis

    This study’s hypotheses were partially supported with mixed results across the two models presented in Table 1 and Table 2. Model 1 consisted of a logistic regression to test H1. After accounting for the five control variables, we found that there is a statistically significant and positive relationship between women’s attitudes towards domestic violence and the onset of civil conflict.

    Model 2 was generated using an ordered logistic regression in order to test H2. This finding was similarly significant, upholding the hypothesis that a higher acceptance of domestic violence against women influences the number of intrastate conflicts a country experiences per year. In both models, three of the five variables performed as hypothesized, providing general support for the theory put forth in this study. However, two variables did not perform as theorized; those variables measuring women’s responses to domestic violence based on arguing with their husband or going out without telling their husband were found to have a negative relationship with both the incidence of civil conflict and its frequency per year. The inclusion of both models serves as a robustness check on the findings of this study.

    Although there was statistically significant data on three of the five independent variables, providing support for this project’s theory, there were two remaining variables that did not perform as hypothesized. One way of accounting for this discrepancy is that these two groups of variables are measuring different concepts. Indeed, this project conducted analyses with the five explanatory variables operationalized as individual components, as they were measuring different aspects of acceptance of domestic violence against women. This was due to the fact that respondents had often given vastly different responses across the five constituent questions. While the questions about burning food, neglecting children, and refusing to have sex with one’s husband may relate to one aspect of women’s expectations in society, the other variables, arguing with one’s husband and going out without telling him, may relate to another. One explanation is that the former three variables, which were found to have a positive association with conflict, pertain to women’s role as caregivers, often understood to be women’s roles under the paradigm of traditional gender norms. When women accept these roles, consistent with the gendered hierarchy theory, they will be more accepting of violence if they fail to fulfill these caregiving tasks.

    Although a wife arguing with her husband may often be seen as a direct challenge or threat, and thus an instigation of violence, arguing could in some cases serve a woman’s traditional role. For instance, it may be considered a wife’s duty to argue with her husband about certain things such as excessive drinking or other behavior that may threaten the family. A woman’s going out without telling her husband may similarly be defensible on the grounds of traditional feminine roles; perhaps a woman may leave the house to collect water or take the children somewhere. Tasks such as these are integral to a woman’s traditional role. Although arguing and going out may not generally be considered acceptable female behavior, they can be defended on the basis of a woman’s wifely or motherly duties, whereas failing to cook food (through burning it), neglecting children, or refusing sex with their husband may be considered direct challenges to women’s roles and the gendered hierarchy that makes women and their needs subservient to those of men. These variables appear to be more associated with being a good wife or mother, whereas arguing and going out can be understood or at least defended on the basis of serving one’s feminine role. This could account for the inconsistent findings concerning the five DHS variables, as the three positive variables can be understood as more of a direct challenge to the lower position of women on the gendered hierarchy. Thus, their acceptance of violence for these offenses should indicate a broader acceptance of violence, which is consistent with this project’s theory and findings.

    Conclusion

    The role of women and conflict as a research agenda in peace and conflict studies is a fairly new development, and an inquiry into their treatment as a determinant of conflict has become an area of inquiry even more recently. This project addresses a gap in the literature by examining women’s attitudes towards domestic violence against women. Through the theory of gendered hierarchy with an attitudinal focus, this project draws upon constructivist theory to explain the relationship of attitudes towards violence within the household and the occurrence of conflict.

    This study lends further support to the developing body of literature linking the domestic status of women to peace and conflict. We found a statistically significant and positive relationship between women’s attitudes towards domestic violence and both the onset and frequency of civil conflict. The caregiving role of women, often understood as women’s traditional role in society, can be understood to be a driving force of these attitudes. According to our theory, women will accept violence, and moreover violence against women themselves, when there is a strong gendered hierarchy established in society. These societies then have a higher manifestation of violence at a state level. This project contributes to the literature on gender and conflict by incorporating the use of survey data and provides further evidence of the salience of gender variables in the study of civil conflict.

    Future research into this topic could involve the investigation of the relationship between DHS attitudinal survey data and the duration of peace, the effect of prior conflict on women’s attitudes towards domestic violence, and an investigation of the impact of attitudes on the intentional targeting of civilians during conflict. Further, should applicable data become available, an examination of the influence of male attitudes towards domestic violence would be informative to the topic. The study of attitudinal data in relation to gender and conflict has much to contribute to the civil war literature, and this study strengthens the constructivist theory of male and female behavior by acknowledging the importance of socially constructed attitudes towards violence. In order to challenge societal norms of violence, we will need to identify and address the attitudes that justify them.

    References

    • Boix, C. (2008). Economic roots of civil wars and revolutions in the contemporary world. World Politics, 60(3), 390-437. Retrieved from http://www.jstor.org/stable/40060203
    • Buhaug, H. & Gates, G. (2002). The geography of civil war. Journal of Peace Research, 39(4), 417-433. Retrieved from http://www.jstor.org/stable/1555275
    • Caprioli, M. (2000). Gendered conflict. Journal of Peace Research37(1), 51-68. Retrieved from http://www.jstor.org/stable/425725
    • Caprioli, M. (2003). Gender equality and state aggression: The impact of domestic gender equality on state first use of force. International Interactions, 29(3), 195-214.
    • Caprioli, M. (2005). Primed for violence: The role of gender inequality in predicting internal conflict. International Studies Quarterly, 49(2), 161-178. Retrieved from http://www.jstor.org/stable/3693510
    • Caprioli, M. & Boyer, M.A.. (2001). Gender, violence, and international crisis. Journal of Conflict Resolution, 45(4), 503-518. Retrieved from http://www.jstor.org/stable/3176309
    • Collier, P. & Hoeffler, A. (2004). Greed and grievance in civil war. Oxford Economic Papers, 56(4), 563-595. doi:10.1093/oep/gpf0
    • Conover, P. J. & Sapiro, V. (1993). Gender, feminist consciousness, and war. American Journal of Political Science, 37(4).1079-1099. Retrieved from http://www.jstor.org/stable/2111544
    • Demerrit, J., Kelly, E., & Nichols, A. (2014). Female participation and civil war relapse. Civil Wars. Retrieved from http://jdemeritt.weebly.com/uploads/2/2/7/7/22771764/demnickel_civilwars.pdf
    • Enloe, C. H. (2000). Bananas, beaches and bases: Making feminist sense of international politics. Oakland: University of California Press.
    • Fearon, J. D. (2005). Primary commodity exports and civil war. Journal of Conflict Resolution, 49(4), 483-507. Retrieved from http://search.proquest.com/docview/224557393?accountid=7113
    • Fearon, J. D. &Laitin, D. D.. (2000). Violence and the social construction of ethnic identity. International Organization, 54(4), 845-877. doi: http://dx.doi.org/10.1162/002081800551398
    • Gleditsch, N. P., Wallensteen, P. Eriksson, M., Sollenberg, M.,& Strand, H. (2002). Armed conflict 1946-2001: A new dataset. Journal of Peace Research, 39(5). Retrieved from http://www.jstor.org/stabl,e/1555346
    • Gizelis, T. I. (2011). A country of their own: Women and peacebuilding. Conflict management and peace science28(5), 522-542. doi: 10.1177/0738894211418412
    • Goffman, E. (1977). The arrangement between the sexes. Theory and Society, 4(3), 301-331. Retrieved from http://www.jstor.org/stable/656722
    • Horowitz, D. L. (1985). Ethnic groups in conflict. Oakland: University of California Press.
    • Hudson, H. (2009). Peacebuilding through a gender lens and the challenges of implementation in Rwanda and Côte d'Ivoire. Security Studies18(2), 287-318. doi: 10.1080/09636410902899982
    • Hudson, V. M., Caprioli, M., Ballif-Spanvill, B., McDermott, R., & Emmett, C. F.. (2009). The heart of the matter: The security of women and the security of states. International Security, 33(3), 7-45. Retrieved from Project Muse.
    • Hudson, V. M., M. Caprioli, B. Ballif-Spanvill, & C.F. Emmett. (2012). Sex and world peace. New York City: Columbia University Press.
    • Melander, E. (2005a). Gender equality and intrastate armed conflict. International Studies Quarterly49(4), 695-714. Retrieved from http://www.jstor.org/stable/3693506
    • Melander, E. (2005b). Political gender equality and state human rights abuse. Journal of Peace Research, 42(2), 149-166. Retrieved from http://www.jstor.org/stable/30042271
    • Reardon, B. A. (1996). Sexism and the war system. Syracuse: Syracuse University Press.
    • Regan, P. M. & Paskeviciute, A. (2003). Women's access to politics and peaceful states. Journal of Peace Research40(3), 287-302. Retrieved from http://www.jstor.org/stable/3648332
    • Segal, L. (2008). Gender, war and militarism: Making and questioning the links. Feminist Review, 88(1), 21-35. Retrieved from http://www.jstor.org/stable/30140873
    • Seifert, R. (1996). The second front: The logic of sexual violence in wars. Women's Studies International Forum, 19(1), 35-43. doi: 10.1016/0277-5395(95)00078-X
    • Sjoberg, L. (2013). Gendering global conflict: Toward a feminist theory of war. New York City: Columbia University Press.
    • Smith, T. W. (1984). The polls: Gender and attitudes toward violence. Public Opinion Quarterly, 48(1), 384-396. Retrieved from http://www.jstor.org/stable/2748632
    • Tessler, M., & Warriner, I. (1997). Gender, feminism, and attitudes toward international conflict: Exploring relationships with survey data from the Middle East. World Politics, 49, 250-281. Retrieved from http://www.jstor.org/stable/25053999
    • Themnér, L. & Wallensteen, P. (2012). Armed Conflicts, 1946-2011. Journal of Peace Research, 49(4). doi: 10.1177/0022343312452421
    • Tickner, J. A. (1992). Gender in international relations: Feminist perspectives on achieving global security. New York City: Columbia University Press.
    • Tickner, J. A. (2001). Gendering world politics: Issues and approaches in the post-Cold War era. New York City: Columbia University Press.
    • Tickner, J. (1999). Why women can't run the world: International politics according to Francis Fukuyama. International Studies Review, 1(3), 3-11. doi: 10.1111/1521-9488.00162
    • Togeby, L. (1994). The gender gap in foreign policy attitudes. Journal of Peace Research, 31(4), 375-392. doi: http://www.jstor.org/stable/424593
    • Wilcox, C., Hewitt, L. & Allsop, D. (1996). The gender gap in attitudes toward the Gulf War: A cross-national perspective. Journal of Peace Research, 33(1), 67-82. Retrieved from http://www.jstor.org/stable/425134

    Table 1: Model 1 Dependent Variable: Onset of Intrastate Conflict

    Number of observations: 146
    *significance level at p<0.10  **significance level at p<0.05   ***significance level at p<0.01

    Table 2: Model 2 Dependent Variable: Frequency of Intrastate Conflict (Per Year)

    Number of observations: 146
    *significance level at p<0.10  **significance level at p<0.05   ***significance level at p<0.01