Gender Inequality and Internal Conflict
Abstract and Keywords
Prior research has found robust support for a relationship between gender inequality and civil war. These results all point in the same direction; countries that display lower levels of gender equality are more likely to become involved in civil conflict, and violence is likely to be even more severe, than in countries where women have a higher status. But what does gender inequality mean in this area of research? And how does research explain why we see this effect on civil war? To explore this, we start with reviewing existing definitions and measurements of gender inequality, noting that the concept has several dimensions. We then proceed to outline several clusters of explanations of how gender inequality could be related to civil war while more equal societies are better able to prevent violent conflict, as described in previous research. It is clear that existing misconceptions that gender inequality primarily involves the role of women are clouding the fact that it clearly speaks to much broader societal developments which play central roles in civil war. We conclude by identifying some remaining lacunas and directions for future research.
Research has found robust support for a relationship between gender inequality and internal armed conflict.1 These results all point in the same direction: countries that display higher levels of gender inequality are more likely to become involved in civil conflict, the violence is likely to be more severe, and postconflict peace appears to be more fragile compared to countries where women have a higher status (Caprioli, 2005; Demeritt, Nichols, & Kelly, 2014; Gizelis, 2009, 2011; Melander, 2005a). These associations hold in global statistical analyses and when controlling for numerous alternative explanations. In fact, gender inequality may potentially even trump level of democracy and economic development in terms of explanatory power in civil war research (Bjarnegård et al., 2015; Hudson, Caprioli, Ballif-Spanvill, McDermott, & Emmett, 2008–2009). As recently pointed out by Bjarnegård et al. (2015), there is no systematic empirical study that has refuted the relationship between gender inequality and war.2 But what does gender inequality mean in this area of research? And how does research explain why we see this effect on civil war?
As noted by Melander (2005a), if we succeed in “disentangling” the complex content of the gender inequality concept, and thereby move the theoretical explanation forward, the contribution to the understanding of civil war could be substantial. In International Relations (IR) and security studies generally, and in the study of civil war specifically, gender inequality has long been an important research agenda for feminist scholars with constructivist and/or postmodern approaches.3 In contrast, positivist scholars have typically not put gender inequality at the forefront (Reiter, 2015). Rather, the determinants of civil wars have mainly been sought in, for example, levels of poverty, regime characteristics, treatment of ethnic minorities, population size, and physical terrain. As such, the most frequently cited scholarly works that statistically investigate the correlates of civil war (e.g., Collier & Hoeffler, 2004; Fearon & Laitin, 2003; Hegre & Sambanis, 2006), as well as articles that review extant findings (e.g., Blattman & Miguel, 2010; Dixon, 2009; Florea, 2012; Lacina, 2008), make no mention of gender inequality. This is a puzzling omission, especially in light of the striking role gender appears to play in the mobilization and conduct of political violence of all kinds.
To contribute to progress, the aim of this article is to explore the concept of gender inequality in greater depth and to iron out the existing theoretical explanations of how gender inequality could be related to civil war.4 As such, the contribution is primarily made to systematic empirical, that is, positivist, scholarship on civil war. We begin by disaggregating the notion of gender inequality and its various dimensions both conceptually and through the manner in which it has been captured by different indicators in previous research on gender inequality and armed conflict. We then proceed to outline several clusters of explanations of how gender inequality could be related to civil war as described in this research. We conclude by discussing a few central potential paths for future research and the challenges these entail.
Conceptualizing and Capturing Gender Inequality
In previous research there are two underlying understandings of gender inequality. One is to consider it as a specific relation among groups where the level of inequality refers to an almost mathematical description of the distribution of certain resources between these groups. In the case of gender inequality, it concerns the resource distribution between men and women. The form of the resource varies. The more prominent forms focus on power, and material and immaterial resources. In that sense, research treats gender inequality as speaking to the level of capacity in a society (see Caprioli, 2000; Gizelis, 2011; Regan & Paskeviciute, 2003), as women constitute about 50% of the adult population. There are interesting variations of this understanding of capacity, encapsulating ideas related to human capital (investments in the individual) as well as social capital. Regan and Paskeviciute (2003), for example, see the limitations in women’s access to the labor market as decreasing the actual capacity of society, as fewer of its members will have skills and resources to affect developments. As an example of gender inequality as social capital, Gizelis conceives of gender inequality as the level of “resources embedded in social structures” in the form of “relation interaction.” It describes the situation on the social level, although it basically depends on human capital, measured in the form of access to education (Gizelis, 2011, p. 524). In its essence, research where gender inequality is understood as social capacity uses this to capture the distribution of skills, power, and resources in a society.
The other underlying understanding of gender inequality is normative. An equal, or balanced, distribution of resources is considered fair and desirable and an unequal resource distribution is assumed to indicate lesser value being assigned to a specific group (see, for instance, Caprioli, 2000; Kuper & Kuper, 1996; Olsson, 2009). In this understanding, gender inequality is considered in terms of one kind of normative intolerance more broadly. For example, in her study of the impact of gender inequality on civil war, Caprioli makes this link by comparing discrimination against women and ethnic groups as two manifestations of a domestic environment characterized by inequality. She correctly observes that gender inequality and discrimination of women have received much less scholarly attention than other types of inequalities: “Women, however, also constitute a minority group in terms of power, yet are largely excluded from systematic studies on intrastate conflict” (Caprioli, 2005, p. 166).
Dimensions of Gender Inequality
The two understandings of gender inequality—as social capacity and as a norm—underlie much of the debate but to analyze how they function empirically they must be disaggregated into something observable. Gender inequality, like other inequalities in civil war research, is measured through many different forms or “dimensions.” At its core are often questions of access to power to decide (over everything from the more “private” matters of one’s own body to influencing the larger “public” structures) and access to varying forms of material and immaterial resources (such as land, money, education, etc.). Different dimensions bring up a number of potentially relevant processes and dynamics for understanding how gender inequality can be related to civil war. Let us look closer at four central dimensions in previous research—political, economic, social, and physical (in)security—and outline a few concerns and considerations about content, interrelations, and measures.
The first dimension, often employed in capturing gender inequality, relates to access to political power. In fact, many would say that power is at the basis of the inequality concept and that the political dimension could be argued to most closely capture “real” power to exert influence in society. In research on conflict and inequality, Caprioli conceptualizes power as “a divisible, infinite resource and/or as the ability to reach goals” (Caprioli, 2000, p. 55). Thus, if women are not allowed to participate in the public sphere equally to men, then that entails a lower access to power (Caprioli, 2000; see also Olsson, 2009). Two often-used measurements to capture political power are representation in political institutions and decision-making capacity. This indicates also the conundrum of being able to detect the difference between mere representation and actual influence.
Several studies suggest that high representation of women in political institutions, primarily the parliament, indicates, on average, a more gender-equal society (Caprioli, 2000, 2005; Caprioli & Boyer, 2001; Koch & Fulton, 2011; Melander, 2005a, 2005b; Regan & Paskeviciute, 2003). At its most basic, the argument is that parliaments represent the people. The more even the balance in parliament, the more even the power distribution among men and women in the population is assumed to be. As noted by several scholars, however, a high proportion of women in parliament is not always a reflection of gender equality in society. For instance, many highly authoritarian states do have parliaments, but these lack real influence. Hence, a large proportion of women in such rubber-stamp legislatures is not a valid indication of women having equal say in politics (Bjarnegård & Melander, 2011; Melander, 2005a). Importantly, even with functioning parliaments, research notes that there can be a substantial difference between women in the political elite and the “average” woman’s ability to participate in politics. This underlines the question of what we expect to measure in this dimension; is it the distribution of resources or the norm? Other scholars instead measure the political dimension of gender inequality by examining the duration of female suffrage, suggesting that if women have had the right to vote for a long time, it is more likely that a more gender-equal norm has made its way more broadly into society and that women practice their political power (Bussmann, 2007; Caprioli, 2000; Caprioli & Boyer, 2001).
Particularly in early research, the underlying question concerning power is thus whether it is women’s capacity to influence that in itself is central, or if it is the role of the gender equality norms in politics that is important and of which the level of women in politics is a reflection. Concerning the former, research notes that it is possible to use direct access to the highest decision-making level as an indicator. Scholars have here tried to use the existence of women as state leaders to measure political gender equality (Caprioli & Boyer, 2001; Koch & Fulton, 2011; Melander, 2005a, 2005b). What they have found, however, is that this measure may be a less adequate manifestation of gender equality. First, the reason for expecting women to behave differently in politics remains questionable. Rather, research finds more and more support that it is the gender equality norm that is having an effect on both men and women’s political behavior (Asal, Legault, Szekely, & Wilkenfeld, 2013; Bjarnegård, 2013; Bjarnegård et al., 2015; Melander, 2016). Second, many female leaders have come to power for dynastic reasons rather than as a force of female empowerment. Third, female executives may be forced to act more hawkish and “masculine,” being the only female in an all-male environment (Melander, 2005a, 2005b).5
The second dimension is the economic. At its basis, this centers on access to material resources through which individuals in a group can affect their own lives.6 In addition to contributing to the understanding of the broader distribution of financial resources between men and women in a society, studies which focus on the economic dimension often use employment rates of women to capture the growth of individual capacity. If women are holding jobs outside of the household, that is assumed to foster a sense of political participation and capacity to influence. In both senses, high levels of employment are considered indicative of a more gender-equal society (Caprioli, 2000, 2005; Caprioli & Boyer, 2001; Regan & Paskeviciute, 2003). Hence, this indicator, similar to that of suffrage, is argued to capture the breadth, or spread, of the gender equality norm in a society. In addition, this indicator can be relevant for understanding the more general distribution of physical resources, as women constitute about 50% of the adult population. In effect, if women do not work to generate income, a natural consequence is that control of financial material resources may be quite concentrated.7
It should be noted that using female participation in the labor market as a measure of economic gender inequality can be quite crude. For example, Caprioli (2000) underlines that a caveat should be placed on equalizing women holding a job with women controlling the use of their paycheck. In addition, she notes that it does not include type of employment or household responsibilities, in particular so considering that some countries have a large informal work sector. It also does not capture the size of the gender gap in wages, a problem that has become more in focus and which may be related to women’s status and capacity in many respects (Caprioli, 2000, p. 56). Finally, the economic dimensions underline the importance of gender equality as relative; ratios between men and women are perhaps more central than the actual number. For example, women’s participation in the labor force must be placed in relation to the level of men participating in the labor force, as the general level of official employment in some societies can be quite low.8
A third dimension often emphasized by scholars concerns the social dimension of gender inequality. This is discussed in research both in terms of the value given to individuals depending on their sex and also in terms of that attached to the more structural perceptions of women/femininity compared to men/masculinity in a society, that is, gender. As noted by Caprioli, it is a dimension which is quite difficult to capture. For example, it touches on perceptions of gender roles, guiding, for example, distributions of labor in a society, or affecting women’s political participation (Caprioli, 2000, p. 56). In that sense, research often considers this as fundamental to the other dimensions in that it can encapsulate both basic demographic aspects and normative values and understandings. It is here that empirical research on civil war and gender inequality in many respects comes closest to the feminist theory which builds its understanding of gendered discrimination based on patriarchal power structures and the social constructions of gender (see Olsson, 2009; Regan & Paskeviciute, 2003).
The social dimension is perhaps the one where the question of what the different measures capture is the most pertinent. To begin with two dominant demographic entry points, fertility rates and sex ratios, there have been several studies which use these indicators to capture the basic state of gender inequality. For example, Caprioli (2000, 2005) and Regan and Paskeviciute (2003) suggest that high fertility rates indicate persistent gender inequality in society. The reasons are several. The consequences of high fertility rates is that women spend a large proportion of their lives either pregnant or breastfeeding, which leads women in such societies to have less time to educate themselves, seek employment, and become involved in politics. As a consequence, women will have less influence and say in many important matters, and their potential to constrain government behavior will be decreased (Regan & Paskeviciute, 2003), again supporting the proposition that social inequality may be fundamental for other dimensions of inequality. High fertility rates by themselves can also strongly indicate if women have the right to decide over their own body. In addition, high fertility rates have serious health consequences which can affect the level of participation in society (see, for instance, Blumberg & Clark, 1989; Dasgupta, 1995). In that sense, the indicator of fertility rates can capture both direct and indirect forms of power in gender relations.
The other demographic indicator used to measure the social dimension of gender inequality is distorted sex ratios, resulting in a male surplus (den Boer & Hudson, 2004; Forsberg & Olsson, 2013; Hudson & den Boer, 2002, 2005). As argued by Hudson and den Boer (2002), if a common practice is to allow one child to live and not another (due to sex-selective abortion, or active or passive infanticide) or when one child is consistently prioritized in terms of nutrition and health care on the basis of gender, it is a clear indication of exaggerated gender inequality. Since such practices are almost universally the result of a son preference, it indicates that the value and status of females are substantially lower than that of males.9 While distorted sex ratios among children can be an indicator of very grave versions of gender inequality, there are other forms of demographic imbalance that instead can be aspects of gender distributions of labor and gendered patterns of migration. This can, for example, be overrepresentation of young men in urban areas (Urdal, 2008), which, in effect, can speak to gender inequality but of a more indirect and less grave form than that of distorted sex ratios among children.
Apart from the demographic indicators, the social dimension of gender inequality can focus on the value-based distribution of investment in individuals based on gender considerations. Hence, access to education is a core aspect of more immaterial resource distributions (see Melander, 2005a, p. 698). As education is central for an individual’s development and for society in general, basic education for girls has increased globally over the last decades. This is partly due to the international norm development, formalized in, for example, the former Millennium Development Goals,10 which have directed development aid to improving girls’ education. Basic education has, therefore, become a less relevant measurement for understanding the level of gender inequality in a given society on the macro-level (see Caprioli, 2000, p. 57).11 As noted by Gizelis (2011), however, it can still be a useful measurement on the micro-level, as access to education still can vary greatly within a state. To come to terms with the measurement problems on the macro-level, many studies nuance the measurement of access to education by studying female-to-male higher education attainment and deviations in literacy rates. Being costly, the former would capture the amounts of resources that are invested in women. It would also, probably, be related to other economic aspects of gendered distribution of labor, such as on what level women could be employed in business and state administration. Hence, it could speak to the elite aspects of developments of gender inequality. Deviations in literacy rates among the population at large, on the other hand, would capture the broader spread of a gender inequality norm. Such deviations would capture the level of social capital of a society in general (Bussmann, 2007; Forsberg & Olsson, 2013; Melander, 2005a).
Last, the dimension of physical (in)security of women has been used as a measure of gender inequality. There are some measures that capture the security of women indirectly, for example, life expectancy and maternal mortality, or de jure rather than de facto security, such as legislation against domestic violence. However, Hudson et al. (2008/2009) suggest a measure that intends to be more immediate—direct violence against women. This is fruitful, they argue, as it to a high degree speaks to the situation in the most basic of institutions in a society, the family. They therefore focus on rape and violence against women (see Caprioli, 2000, p. 55; Melander, 2005a for a discussion). However, it is not an easy measurement to use. Many forms of violence against women come with a stigma or a blaming of the victim. In addition, in some societies, domestic violence may not be considered illegal. Therefore, many women do not report abuse. This results in a lack of data on violence against women, which is sketchy at best in most states. To address this, Olsson (2009) uses the concept of security equality as a proxy for physical insecurity of women. This concept focuses on the resource distribution to protective measures set in place for different forms of violence, as violence tends to follow gender-specific patterns. It is therefore possible to see if, for example, domestic violence and rape are given a substantive amount of efforts and resources in comparison to violence which affects primarily men.
In conclusion, these four key dimensions bring up numerous aspects of gender inequality. As we can see, they all relate in different manners to both forms of understanding of gender inequality, that is, as social capacity or as a norm. As a consequence, regardless of understanding, most research dealing with the relationship between gender inequality and civil war has used a combination of different indicators, such as women in parliament, fertility rates, access to education, women’s participation in the labor force, and so forth. As noted, previous research also considers these dimensions to be interrelated, that is, affecting each other (see Regan & Paskeviciute, 2003 for a discussion). To complicate matters, the indicators of the dimensions do not co-vary in a way which makes it easy to spot a pattern. For example, in Georgia, the general views on women’s social roles are very conservative, with women being seen as primary caretakers and responsible for the household. The same survey found, however, very positive views toward women getting higher education and holding a well-paid job outside of the household (Naskidashvili, 2011). In Italy, economic equality is rising, resulting in very low fertility rates as the social dimensions of gender equality lag behind—thus, individual women may find it very difficult to reconcile the social role with the economic and thereby choose to have few or no children.12 In addition, there can be substantial differences between urban and rural areas in a country (Regan & Paskeviciute, 2003, p. 294). Thus, while extremely high fertility rates may be an indication of gender inequality, this cannot be assumed to hold to a linear function. Given social welfare systems that subsidize child care and pay for extended parental leave, with the explicit purpose of providing more equal rights for women and men, families may choose to have more children than in a context where such support systems are absent. Hence, fertility rates may be a consequence of the level of social gender inequality but may, in turn, also affect the economic and political dimensions of the phenomenon. In sum, there is a need for further research to develop the measurement of gender inequality, as the social processes are very complex.
Why Gender Inequality and Internal Armed Conflict?
Why is gender inequality related to internal armed conflict? The analyses found in previous research revolve around two strands of explanations; why gender inequality may be associated with an increased risk for intrastate armed conflict or why gender equality instead may contribute to resolving a conflict without violence. Interestingly, these two strands of explanations are not simply two sides of the same coin. Theoretically, it is possible that severe inequality can affect the risk of violence while a positive presence of high gender equality may be completely unrelated to peace, and vice versa. Within these two strands, there are subthemes related to aspects of norms and social capacity which relate to different dimensions of gender inequality. Is gender inequality picking up on the very unequal resource distribution—in the sense of very unequal investments in members of the society and, hence, lower social capacity and weaker networks—thereby resulting in a society’s incapacity to handle existing conflicts without resorting to violence and oppression? Or is it the norms of gender inequality that is affecting the risk of conflict violence?
Gender Inequality and Risk for Conflict
There are two explanations regarding why gender inequality may contribute, directly or indirectly, to increasing the risk for violent conflict. One focuses on norms that may enhance violence, in particular a masculinized political culture lowering the threshold for violence. The other concerns the direct provision of capacity to mobilize for conflict, in particular recruitment of young men.
Norms enhancing violence
All societies consist of groups with conflicting interests and a competition for power and other scarce resources. How these grievances are addressed affect the risk of war. As gender is a dominant organizing principle in all societies and the level of inequality varies greatly in this relationship, it has been argued that the treatment of women in highly unequal societies may serve as a blueprint for how grievances by other “out-groups” are perceived (see, for example, Caprioli, 2000, 2005; Caprioli & Boyer, 2001; den Boer & Hudson, 2004; Gizelis, 2009, 2011; Goldstein, 2001; Hudson, 2012; Hudson et al., 2008–2009; Hudson & den Boer, 2002; Melander, 2005a, 2005b; Peterson, 1992; Reardon, 1996; Tickner, 1992). As noted by Caprioli (2005, p. 163), “norms of intolerance and inequality should have an incendiary impact on domestic and international behavior by legitimizing violence as a tool of conflict resolution.” Using the psychological concept of “othering,” Hudson and colleagues (Hudson et al., 2008–2009; Hudson & den Boer, 2012) suggest that in highly patriarchal societies, where children grow up seeing women being dominated and controlled by men, they are provided with a template where violence and domination is considered normal. Moreover, highly patriarchal societies often espouse traditional gender roles based on a culture where manhood is linked to toughness and more “warlike” attitudes and behavior, and prescribing that men are superior to women. For example, Enloe states that “militarization of ethnic nationalism often depends on persuading individual men that their own manhood will be fully validated only if they perform as soldiers, either in the state’s military or in insurgent autonomous or quasi-autonomous forces” (Enloe, 1989, p. 55). As a consequence, these cultures are associated with more violence (e.g., Goldstein, 2001; Tickner, 1992).
As such, norms of gender inequality may be seen as one form of intolerance, where some groups consider it legitimate to oppress and dominate over other groups, be it women, sexual minorities, ethnic minorities, or political opposition groups. Evidence indeed suggests that such norms of intolerance correlate; for example, attitudinal survey-based studies show a correlation between sexist and racist attitudes (see, for instance, Henley & Pincus, 1978).13 Such societies are more likely to see the “superior” group dominate over other “inferior” groups. It is postulated that societies with a very high level of male dominance in politics tend to be dominated by hypermasculine political cultures. This norm prescribes violence as a means to resolve conflict also on the highest decision-making levels. For example, this could be related to the risk of losing face or appearing weak if opting, for example, for negotiations (Melander, 2005a). Hence, the level of gender inequality can be considered as capturing how a society deals with existing grievances, that is, how elites deal with horizontal differences between groups.
Capacity to mobilize for conflict
While Fukuyama (1998) provocatively argued that male aggression was directly related to war, the current consensus in civil war research is that armed conflict involves organized groups rather than individuals confronting each other at random. That said, perhaps the level of gender inequality can make it easier to access men for mobilization? A key resource for conducting a violent conflict is access to soldiers. The primary target group to mobilize is young men. As noted under the discussion on the social dimension of equality, high levels of gender inequality can affect the demographic balance. For example, a society characterized by gender inequality in the form of a persistent and strong son preference may, as a consequence, end up with a large male surplus. This is particularly evident in some Asian countries. In fact, the large number of missing women in India and China alone skews the world average so that the female-to-male ratio is smaller than 1, although it is expected to be larger than 1 (as women, on average, live longer). In regions with large male surpluses, it is likely that there will be a higher-than-average “supply” of men who have low opportunity costs and hence are easy to mobilize for political violence. For example, Urdal (2008) finds that youth bulges are related to increased risk of conflict. Hudson et al. (2008–2009) expand this argument by looking closer at bulges resulting from very uneven sex ratios which, in turn, stem from serious forms of gender inequality. The authors claim that such a male surplus can result in a large number of dissatisfied, that is, aggressive, men. According to this explanation, mobilization for war is made easier for two reasons: a) the society in which these men live has a hypermasculinity-based culture that normatively encourage violence as a means of resolving conflict; b) these men, being excluded from society as many of them cannot find a spouse or a job, have a higher likelihood of gathering and organizing into groups, or “gangs” (Hudson et al., 2008–2009). Thus, gender inequality creates a larger number of men with low opportunity costs (being unmarried and unemployed); these men are more susceptible to gender-based language in recruitment, and they are already connected in groups through which they can be more easily mobilized.
Gender Equality and Ability for Peaceful Resolution
If we look at the flipside, perhaps it is instead higher levels of gender equality which provide both the norms and the resources to handle conflicts in more nonviolent manners?
Norms promoting nonviolence
Highly unequal societies were postulated to be dominated by a political culture based on negative masculine roles which overemphasize violent means to resolve conflict. However, although some conflict over power and resources is inevitable, it is possible that it instead is the case that more equal states are dominated by norms more prone to nonviolent forms of handling grievances. Gender equality as a norm prescribes respect and resolution of conflict without violence (i.e., decreases the role of hypermasculinity). Hence, equality norms may both prevent grievances from escalating to violent conflict in the first place and decrease the risk that a postconflict country relapses into violence (Demeritt et al., 2014). As noted by Melander (2005a, 2005b), societies characterized by gender equality are ingrained with norms that prescribe that men and women treat each other with respect. These norms then transfer to other societal relations, such as between ethnic groups and political parties, and can explain the relative peacefulness of such states.
Hence, where socially constructed gender roles are more equal, it is expected that respect for others in the private sphere will carry over into society at large. The cost for using violence would increase substantially and other methods of addressing grievances may be more institutionalized. Thus, societies with higher levels of equality are likely to have elites that are better at handing grievances by different groups (Caprioli, 2000, 2005; Melander, 2005a, 2005b). This suggestion is not only supported by studying the link between gender equality and violent conflict across different countries. Melander (2016) notes that the relationship between gender inequality and internal conflict is found also at levels of analysis other than the country level. As found by Asal et al. (2013), groups that proscribe a gender-inclusive ideology are less likely to pursue their objectives using violent means. At the individual level it has been found that gender-equal attitudes are correlated with advocating peaceful conflict resolution; in fact, it appears that such attitudes carry more explanatory power than biological sex (Conover & Sapiro, 1993; Tessler & Warriner, 1997).
Capacity to prevent violence
Let us now turn to the explanation which focuses on the capacity of a society to resist violence and handle conflict peacefully. Examining the risk for conflict relapse, Gizelis (2009) finds that higher levels of investment in women and their having access to more resources result in higher capacity to influence and to create and uphold networks—both vertical and horizontal—among a larger part of society. These capacities can be used to forward peace and create other means to resolve conflict. Rather than excluding competent individuals by upholding patriarchal structures, more gender equality, that is, higher status of women, hence indicates that there is more competence to draw from in a society. Thus, in the aftermath of conflict, peacebuilding efforts can gain more ground if finding effective local partners includes both women and men. The result is that the higher the status of women, the lower the risk of armed conflict. A continued development of this argument is that more benign gender roles—and the mobilization of women based on these roles in particular—are considered a central aspect of the argument for why these actors can play a mediating role to decrease violence. The same dynamic could come into play in the prevention of violence. In a similar vein, Regan and Paskeviciute argue that in a society with higher levels of gender equality, power is going to be more diffused. This can play out on the grassroots level (Regan & Paskeviciute, 2003, pp. 289–290) where grassroots networks of women can constitute a positive resource for peace (Gizelis, 2011). Liberia is such an example; there, women organized across warring lines in order to influence and pressure the warring parties (Gizelis, 2011). Similarly, Kutz-Flamenbaum (2012) notes that women’s organizations can instrumentally play on gender roles in order to forward norms related to conflict resolution in situations where the security risks otherwise would be too high for, for example, male-dominated peace groups to operate.
While we have focused mainly on the potential direct link between gender inequality and internal armed conflict, it should be noted that the link may be, in part, indirect or conditioned upon specific circumstances. A related explanation to that of capacity is that the connection can be indirect if gender equality operates by strengthening societal institutions. Bussmann (2007) notes that a higher engagement of women in various societal sectors including politics, due to higher education levels and activity in the labor force, will work to constrain government behavior and strengthen state capacity and governance. Strengthened state capacity and governance are, as supported by several studies, in turn associated with peace. Similarly, several researchers find interaction effects with democracy and economic development. For instance, a recent study demonstrates that democracies are more peaceful than nondemocracies only when interacted with gender equality (Melander, 2005a).
Conclusions and Implications for Future Research
This article has sought to iron out and explore the explanations for the correlation between gender inequality and intrastate armed conflict found in systematic empirical contributions in the field of civil war research. In that way, the article contributes to the development of the positivist scholarship on civil war. To do this, the concept of gender inequality was disaggregated and four key dimensions of the concept in previous research were identified: political, economic, and social (in)equality, and physical (in)security of women. In addition, the measurement of these dimensions was reviewed, and it was noted that while previous research considers them to be interrelated, the indicators of the dimensions do not co-vary. This raises interesting questions about the gender inequality concept. Based on the discussions on understandings and dimensions, several explanatory clusters which related to a society’s risk of armed conflict or ability to find peaceful resolutions were then identified. These clusters can then be used in further empirical studies. It is here interesting to speculate on whether the different clusters of explanations can be relevant for understanding the conflict dynamics in different phases of a conflict. That said, what do these observations then say about how to best proceed? And what should be considered to be particular challenges? We will end by considering some opportunities and challenges for future research, focusing on important areas that remain to be addressed.
The first area that needs to be addressed in order to develop the theoretical ideas around the connection between gender inequality and civil war concerns identifying the most relevant indicators and finding valid and reliable data for an appropriate unit of analysis. Previous positivist research suffers from a paucity of valid comparable data that is available longitudinally and cross-nationally. As a consequence, most researchers have used the same set of indicators (women in parliament, fertility rates, access to education, women’s participation in the labor force) to assess what they clearly believe to be different types of mechanisms. As a result, one and the same indicator has been invoked in support of multiple competing hypotheses. Due to such data scarcity, existing studies predominantly use proxies and overly aggregated country-level data. On the positive side, Hudson et al. (2008–2009, p. 32), quoting Nussbaum (2000), claim that some variables that are available across cases and over time, “such as highly abnormal sex ratios in favor of males, or denying girls the legal right or access to education, can be applied cross-nationally to determine gender status beliefs and the security and status of women.” Hence, a second area to consider concerns the level of analysis. While cross-national data and measures may be relevant to assess the effect of political inequality (such as women in parliament), their applicability is less straightforward when pursuing a socioeconomic analysis of gender inequality. We know that gender inequality levels can vary between groups living in the same state. For instance, as shown by Gizelis (2011), gendered deviations in education and literacy rates may vary substantially within a country, where some regions appear to be relatively equal while others clearly invest more in men. To assess the relevance of such explanations, it may be more appropriate to use subnational data, as the country average is less informative.14 Third, Melander (2005a) raises interesting questions in terms of if we need to see certain levels of gender equality (or gender inequality) before we see an effect on conflict. Here, it is conceivable that this question is more relevant for some dimensions than for others. For example, it is possible that demographic indicators such as sex ratios require a certain level to have an effect, while previous research has not been able to observe such effects concerning, for example, women’s participation in parliament (most notably, the argument of critical mass has no support; see Bjarnegård et al., 2015; Melander, 2005a).
These are all possible to address. In so doing, researchers must consider the current two underlying understandings of gender inequality—as a norm and or as social capacity—and what these diverging understandings mean for the theory on civil war. It is apparent that gender is relevant for understanding both attitudes to violence and the access to resources to mobilize for peace or conflict. These two understandings could thereby play central roles in developing current “nongendered” theoretical explanations. As eloquently noted by Melander (2005a), in disentangling the concept and moving the theoretical explanation forward, the contribution to the understanding of civil war could be substantial. However, the two understandings of gender inequality are themselves in need of development and clarification. Notably, for example, the understanding of gender inequality as social capacity tended to draw on ideas of both social capital and human capital. Hence, this is an interesting conceptual area to develop in future research. Lastly, it is apparent that an existing misconception in current debates, namely, that gender primarily involves the role of women (a misunderstanding which exists in both research and policy), is clouding the fact that gender inequality clearly speaks to much broader developments in a society. These developments directly involve and affect both men and women, along with where different forms of masculinity and femininity appear to play central roles in civil war.
This research was supported by a grant from the Swedish Research Council. The authors thank Elin Bjarnegård, Mats Hammarström, and Theodora-Ismene Gizelis for excellent suggestions on how to improve this essay. Authors are listed alphabetically; equal authorship is implied.
Asal, V., Legault, R., Szekely, O., & Wilkenfeld, J. (2013). Gender ideologies and forms of contentious mobilization in the Middle East. Journal of Peace Research, 50, 305–318.Find this resource:
Bjarnegård, E. (2013). Gender, informal institutions and political recruitment: Explaining male dominance in parliamentary representation. New York: Palgrave Macmillan.Find this resource:
Bjarnegård, E., & Melander, E. (2011). Disentangling gender, peace and democratization: The negative effects of militarized masculinity. Journal of Gender Studies, 20, 139–154.Find this resource:
Bjarnegård, E., Melander, E., Bardall, G., Brounéus, K., Forsberg, E., Johansson, K., et al. (2015). Gender, peace and armed conflict. SIPRI Yearbook. Stockholm: Stockholm International Peace Research Institute.Find this resource:
Blattman, C., & Miguel, E. (2010). Civil war. Journal of Economic Literature, 48, 3–57.Find this resource:
Blumberg, R. L., & Clark, M. H. (1989). Making the case for the gender variable: Women and the wealth and well-being of nations. Washington, DC: United States Agency for International Development, Office of Women in Development.Find this resource:
Bussmann, M. (2007). Gender equality, good governance, and peace. Prepared for presentation at the SGIR Sixth Pan-European International Relations Conference, Turin (Italy), September 12–15, 2007.Find this resource:
Caprioli, M. (2000). Gendered conflict. Journal of Peace Research, 37, 53–68.Find this resource:
Caprioli, M. (2005). Primed for violence: The role of gender inequality in predicting internal conflict. International Studies Quarterly, 49, 161–178.Find this resource:
Caprioli, M., & Boyer, M. A. (2001). Gender, violence, and international crisis. Journal of Conflict Resolution, 45, 503–518.Find this resource:
Collier, P., & Hoeffler, A. (2004). Greed and grievance in civil war. Oxford Economic Papers, 56, 563–595.Find this resource:
Conover, P. J., & Sapiro, V. (1993). Gender, feminist consciousness, and war. American Journal of Political Science, 37, 1079–1099.Find this resource:
Dasgupta, P. S. (1995). Population, poverty and the local environment. Stockholm: Beijer International Institute of Ecological Economics.Find this resource:
Demeritt, J. H. R., Nichols, A. D., & Kelly, E. G. (2014). Female participation and civil war relapse. Civil Wars, 16, 346–368.Find this resource:
den Boer, A. M., & Hudson, V. M. (2004). The security threat of Asia’s sex ratios. SAIS Review of International Affairs, 24(2), 27–43.Find this resource:
Dixon, J. (2009). What causes civil wars? Integrating quantitative research findings. International Studies Review, 11, 707–735.Find this resource:
Enloe, C. (1989). Bananas, beaches and bases: Making feminist sense of international politics. Berkeley: University of California Press.Find this resource:
Fearon, J. D., & Laitin, D. D. (2003). Ethnicity, insurgency, and civil war. American Political Science Review, 97, 75–90.Find this resource:
Florea, A. (2012). Where do we go from here? Conceptual, theoretical, and methodological gaps in the large-N civil war research program. International Studies Review, 14, 78–98.Find this resource:
Forsberg, E., & Olsson, L. (2013). Gender inequality and armed conflict: A disaggregated analysis of Northeast India, 1989–2011. Paper presented at the annual convention of the International Studies Association, San Francisco, April 2013, and the joint meeting of CRS and ENCoRe, University of Essex, September 2013.Find this resource:
Fukuyama, F. (1998). Women and the evolution of world politics. Foreign Affairs, 77(5), 24–40.Find this resource:
Gizelis, T.-I. (2009). Gender empowerment and United Nations peacebuilding. Journal of Peace Research, 46, 505–523.Find this resource:
Gizelis, T.-I. (2011). A country of their own: Women and peacebuilding. Conflict Management and Peace Science, 28, 522–542.Find this resource:
Gleditsch, K. S., Wucherpfennig, J., Hug, S., & Reigstad, K. G. (2011). Polygyny or misogyny? Reexamining the “first law of intergroup conflict.” The Journal of Politics, 73, 265–270.Find this resource:
Goldstein, J. S. (2001). War and gender. Cambridge, U.K.: Cambridge University Press.Find this resource:
Hegre, H., & Sambanis, N. (2006). Sensitivity analysis of empirical results on civil war onset. Journal of Conflict Resolution, 50, 508–535.Find this resource:
Henley, N. M., & Pincus, F. (1978). Interrelationship of sexist, racist, and antihomosexual attitudes. Psychological Reports, 42, 83–90.Find this resource:
Hudson, V. M. (2012). Sex and world peace. New York: Columbia University Press.Find this resource:
Hudson, V. M., Caprioli, M., Ballif-Spanvill, B., McDermott, R., & Emmett, C. F. (2008–2009). The heart of the matter: The security of women and the security of states. International Security, 33(3), 7–45.Find this resource:
Hudson, V. M., & den Boer, A. (2002). A surplus of men, a deficit of peace: Security and sex ratios in Asia’s largest states. International Security, 26(4), 5–38.Find this resource:
Hudson, V. M., & den Boer, A. M. (2005). Missing women and bare branches: Gender balance and conflict. ECSP Report, 11, 20–24.Find this resource:
Hudson, V. M., & den Boer, A. M. (2012). A feminist evolutionary analysis of the relationship between violence against and inequitable treatment of women, and conflict within and between human collectives, including nation-states. In T. K. Shackelford & V. A. Weekes-Shackelford (Eds.), The Oxford handbook of evolutionary perspectives on violence, homicide, and war (pp. 301–323). Oxford: Oxford University Press.Find this resource:
Kanazawa, S. (2009). Evolutionary psychological foundations of civil wars. The Journal of Politics, 71, 25–34.Find this resource:
Koch, M. T., & Fulton, S. A. (2011). In the defense of women: Gender, office holding and national security policy in established democracies. The Journal of Politics, 73, 1–16.Find this resource:
Kuper, A., & Kuper, J. (1996). The social science encyclopedia (2d ed.). London: Routledge.Find this resource:
Kutz-Flamenbaum, R. (2012). Mobilizing gender to promote peace: The case of Machsom Watch. Qualitative Sociology, 35, 293–310.Find this resource:
Lacina, B. (2008). Insights from macro studies of the risk of civil war. In M. Öberg & K. Strøm (Eds.), Resources, governance, and civil conflict. London: Routledge.Find this resource:
Melander, E. (2005a). Gender equality and intrastate armed conflict. International Studies Quarterly, 49, 695–714.Find this resource:
Melander, E. (2005b). Political gender equality and state human rights abuse. Journal of Peace Research, 42, 149–166.Find this resource:
Melander, E. (2016). Gender and civil war. In D. T. Mason & S. McLaughlin Mitchell (Eds.), What do we know about civil wars? Lanham, MD: Rowman & Littlefield.Find this resource:
Naskidashvili, M. (2011). How does gender determine roles and behaviors of women in and outside of Georgian families. Unpublished. Yerevan, Armenia: Caucasus Research Resource Center.Find this resource:
Nussbaum, M. C. (2000). Women and human development: The capabilities approach. Cambridge, U.K.: Cambridge University Press.Find this resource:
Olsson, L. (2009). Gender equality and United Nations peace operations in Timor Leste: International peacekeeping. Leiden, The Netherlands: Brill Publishers.Find this resource:
Peterson, V. S. (1992). Gendered states: Feminist (re)visions of international relations theory. Boulder, CO: Lynne Rienner.Find this resource:
Reardon, B. (1996). Sexism and the war system. Syracuse, NY: Syracuse University Press.Find this resource:
Regan, P. M., & Paskeviciute, A. (2003). Women’s access to politics and peaceful states. Journal of Peace Research, 40, 287–302.Find this resource:
Reiter, D. (2015). The positivist study of gender and international relations. Journal of Conflict Resolution. 59, 1301–1326.Find this resource:
Sleebos, J. (2003). Low fertility rates in OECD countries: Facts and policy responses (OECD Social, Employment and Migration Work Papers, No. 15). Paris: OECD Publishing.Find this resource:
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.Find this resource:
Tickner, J. A. (1992). Gender in international relations: Feminist perspectives on achieving global security. New York: Columbia University Press.Find this resource:
United Nations General Assembly, Transforming our world: the 2030 Agenda for Sustainable Development, A/RES/10/1, 21 October 2015. https://sustainabledevelopment.un.org/content/documents/21252030%20Agenda%20for%20Sustainable%20Development%20web.pdf(accessed June 8, 2016).Find this resource:
Urdal, H. (2008). Population, resources, and political violence: A subnational study of India, 1956–2002. Journal of Conflict Resolution, 52, 590–617.Find this resource:
(1.) The terms civil war, civil conflict, internal conflict, and intrastate armed conflict are used interchangeably, referring to a violent conflict over government power and/or territory involving the government of a state and one or multiple armed opposition groups.
(2.) Closely related to studies of gender inequality is the question of whether women are more peaceful than men more generally. This peacefulness is discussed to be either biological or social. The more essentialist version is based on a biological understanding. Women are considered peaceful by nature due to their nurturing mother role. The other is the social, constructivist explanation. In this women’s assumed peacefulness is linked to the social construction of femininity, for which nonviolent resolution is prescribed, but it is not considered to be an innate quality. However, women’s direct access to power and resources can in both of these forms have an impact on the likelihood of peace. While recognizing the potential interlink, this article is focused on gender inequality rather than women per se. This is because gender inequality has been found to be a more relevant explanation than women’s peacefulness (for a discussion, see Caprioli, 2000; Melander, 2005a).
(3.) As noted by Hudson et al. (2008–2009), this literature argues that standard social science norms, including hypothesis testing, are at odds with a feminist perspective. Combined with the striking neglect of including a gender perspective in positivist security studies, the consequence has been little cross-fertilization between the fields. However, there are now a growing number of studies that demonstrate that gender inequality is central for understanding civil war as well as other types of political violence.
(4.) The entire field is not reviewed; instead, the focus is on key publications which are considered as driving the research debate forward. Along with studies that analyze the link between gender inequality and the onset of internal armed conflict, a few references that focus on armed conflict more broadly (including interstate conflict and postconflict peacebuilding) are also included, as the theoretical explanations appear to be similar. For an overview, see Bjarnegård et al. (2015).
(5.) There is also a methodological problem involved in using female state leaders as a measure of political gender equality; the sample is too small to enable statistical inference (Melander, 2005a).
(6.) In addition to this focus on the broader spread of resources in a society, the economic dimension can be used to understand the distribution of financial resources on the elite level, for example, women on the boards and/or in leadership positions of large companies.
(7.) Another aspect of economic gender equality which has not been used in research on civil war so far is the right to own property or the distribution of ownership of material resources. This aspect of equality is now target under the fifth goal of the 2030 Agenda for Sustainable Development, which says that states should “undertake reforms to give women equal rights to economic resources, as well as access to ownership and control over land and other forms of property, financial services, inheritance and natural resources, in accordance with national laws” (p. 20).
(8.) We are grateful to Elin Bjarnegård for underlining this point.
(9.) A variant of the male surplus measure is widespread use of polygyny, which in effect leads to a male surplus. Kanazawa (2009) suggests that polygynous societies are particularly prone to civil war. However, the measure is not discussed in terms of gender inequality. As suggested by Gleditsch et al. (2011), who could not reproduce Kanazawa’s result in a test more aligned with the field of civil war research, misogyny is likely a better predictor of violence than polygyny.
(10.) Now also included in Goal Four of the 2030 Agenda for Sustainable Development (p. 19).
(11.) Given this relatively small gender gap in education, the persistent gaps in political representation and employment strongly indicate gender-based discrimination, as such differences cannot be attributed to differences in educational competence.
(13.) However, within civil war research, we are not aware of any attempt to empirically assess whether gender inequality is connected to other types of inequality and how the interplay of different forms of inequality is related to conflict risks.
(14.) Forsberg and Olsson make this argument in a study of gender inequality and insurgency violence in Northeast India (Forsberg & Olsson, 2013). Among the 85 districts that make up the seven states of Northeast India, there is great variation across various dimensions of gender inequality, such as sex ratios and deviations in literacy rates.