After Feminist Analyses of Bosnian Violence

Darius Rejali is the author of Torture and Modernity; Self, Society and State in Modern Iran. He teaches political philosophy and comparative politics at Reed College. April 24, 1996

Copyright (c) 1996 by Darius Rejali, all rights reserved. This text may be used and shared in accordance with the fair-use provisions of U.S. copyright law, and it may be archived and redistributed in electronic form, provided that the editors are notified and no fee is charged for access. Archiving, redistribution, or republication of this text on other terms, in any medium, requires the consent of the author and the notification of Peace Review at

As events in Bosnia unfolded, American feminists found themselves grappling with how their theories of rape were relevant beyond North Atlantic shores. I build on these analyses because they provide excellent ways of thinking about gender and rape in war. And yet they are limited as well by what comparative political scientists would recognize as an impoverished concept of ethnicity. To think about rape in war comparatively requires relating gender to ethnicity in more complicated ways than these theories allow. And the place to see this most clearly is in Bosnia.

Analysts distinguish three kinds of rape in Bosnia: rapes that occurred when Serbs first occupied a village; rapes committed by prison guards in detention camps; and rape camps, or temporarily commandeered houses where Serbs kept women expressly for that purpose. Reports also emphasized that rapes often took place in public or in front of other witnesses; that rapes included acts designed to degrade the victim; and that often the victims knew the aggressors.

In assessing this information, American feminists agreed that rape in a war context couldn't be reduced to the psychological attributes of the aggressors or the mere aggregate of them in war. Rape must be understood in terms of social structures and practices. Mass rape also cannot be understood by emphasizing its unique or exceptional wartime character; rather, it can only be clarified and rendered intelligible in terms of every day forms of violence that are considered legitimate. Finally, an account of rape must be able to identify the interrelationship between ethnicity and gender. Rape in a national context often touches on, if not centers on, issues of race and ethnicity (e.g., in the American context, the notion of the black rapist or the idea that black women are less worthy rape victims). In a war context, racial and ethnic distinctions take on a particular salience because they are being renegotiated violently, and the relationship of ethnicity and gender becomes much more complicated. So rape in Bosnia was systematic, but in what way was rape systematic, that is, in relation to what policies or socio-cultural forms? Can one specify the processes that make a rapist? How was the seemingly pathological violence related to everyday life in the former Yugoslavia? Here key feminists gave different answers.

Catherine MacKinnon analyzes the Bosnian rapes by focusing on the place that pornography occupied in the former Yugoslavia. MacKinnon notes that in Yugoslavia, pornographic products were more common than any other socialist country. "When pornography is this normal a whole population of men is primed to dehumanize women and to enjoy inflicting assault sexually...Pornography is the perfect preparation - motivator and instruction manual in one- for the sexual atrocities ordered in this genocide." MacKinnon notes that rapes in detention camps always involved performing for a male audience, just as in pornographic shows. Some of the rapes were not only performed publicly but filmed using video technology. These films were sold as pornographic products, since they were indistinguishable from the real thing. Furthermore women were ordered to copy poses from pornographic magazines stuck on the walls during the rapes, thus reproducing pornography as reality. Finally, these pornographic films were turned out for mass consumption much like news and entertainment, except that in this case they were designed to whip up popular frenzy for the Serbian war effort. Analyzing the films, MacKinnon notes details that suggest sophisticated staging by Serbians directing the films, including providing props and dubbing dialogue to implicate Croatian soldiers. All of this points to extensive government involvement. "The world has never seen sex used this consciously, this cynically, this elaborately, this openly, this systematically with this degree of technology and psychological sophistication, as a means of destroying a whole people." Rape pornography in turn encourages more men to enlist who in turn rape and produce more pornographic products which in turn leads to more enlistees.

Susan Brownmiller has written her own, quite different analysis of the events in Bosnia. Rape in Bosnia, Brownmiller asserts, is not simply produced by a crisis in ethnic identity, but a crisis in male ethnic identity. Balkan men have found an ethnic cause to fight and die for in the issue of "protecting their women" from their enemies. Whereas MacKinnon locates rape in male/female dynamics, Brownmiller locates rape in the conflict between males. "Sexual trespass on the enemy's women is one of the satisfactions of conquest, like a boot in the face, for once he is handed a rifle and told to kill, the soldier becomes an adrenaline-rushed young man with permission to kick in the door to grab, to steal to give vent to his submerged rage against all women who belong to other men." Women's bodies here are the battlefield in which men communicate their rage to other men.

Brownmiller is keenly aware that the scale of the Bosnian rapes is not historically unprecedented. Whereas MacKinnon here sees technological sophistication and systematicity unique in modern history, Brownmiller sees the recurrence of the same. She recognizes that rape dehumanizes women, but she notes that there is more than one force in war time that leads towards dehumanization. "Rape of a doubly dehumanized object - as woman, as enemy, carries its own terrible logic. In one act of aggressiveness, the collective spirit of women and the nation is broken, leaving a reminder long after the troops depart." Pornography is not the only social form that dehumanizes women: war itself introduces a process whereby the enemy is dehumanized. The possibility of rape in a war context is overdetermined for Brownmiller.

There is an odd symmetry to these analyses. MacKinnon maintains that the motor for rape in a war context is the social dynamic between men and women. Yet her focus on the eroticized body leaves almost no space for discusing how ethnicized bodies are constituted. That ethnic bodies exist primordially is assumed. Brownmiller's analysis, on the other hand, focuses on the ethnic dynamic between men and men, which leads her to diminish the eroticization of torture and marginalize the women in her analysis. So singleminded a focus can be blinding. If Brownmiller is right about the intra-male dynamics of rape, heterosexual rape is often a substitute for homosexual rape, and yet Brownmiller does not look for such evidence to vindicate her thesis and MacKinnon's focus on male/female dynamics rules this inquiry out by fiat.

To move beyond these problems, it's helpful to look at American feminist analyses of rape in Latin America. Throughout the 1980s, rape in a war context by government soldiers was common throughout Central and Latin America (notably in Peru and El Salvador). For Cynthia Enloe and Julie Phillips, who have been concerned with these issues, the feminist debate on Bosnia seemed completely disproportionate to the gravity of the issue. "As the highly visible rapes in Bosnia reshape feminist thinking on women and war," Phillips remarked bitterly, "the conflict in Peru quietly goes into its 13th year." And Enloe emphasizes that the limited knowledge we possess about this phenomena. "We still know surprisingly little about why the government's male soldiers in Guatemala or the Contra's male insurgents in Nicaragua engaged in sexual assault on women so insistently." Were these soldiers in or out of control? Was this a conscious government strategy to intimidate or simply one more product of a mostly militarized society? What would count as evidence for one or the other?

Unlike many kinds of violence, rape is shot through with symbolic significance and must be contextually analyzed. Enloe criticizes analysts who list rape as one of "an assortment of repressive acts, as if rape were not qualitatively different in both its motivations and repercussions." Rape's consequences may be either shame as in Bosnia or heroic martyrdom. Similarly, rapists may be seeking to intimidate the woman being raped, the men whose property she represents, or even other rapists. For example, Enloe maintains that in Central America, rape socialized the new military recruit and separated him permanently from his civilian compatriots; many soldiers were forced recruits who had to be isolated from their communities through rape before they could undertake repressive operations.

Analyzing the Peruvian rapes in the 1980s, Phillips emphasizes that rape does not merely have one function, but is multifunctional, "a uniquely versatile type of violence." Rape serves to intimidate independent community organizers. "If traditional 'women's work' - tending families, raising children, holds communities together, assaults on women do much to tear the social fabric apart. When you are out to subdue a population, women are the population." Rape may be used to force confessions or destabilize whole areas as a tactic of terror. In the case of female guerillas, rape is a particularly vicious means of reinforcing the gendered division of labor. Rape also imitates and reinforces ethnic and racial divisions: in Peru, mestizo guerillas are raped by mestizo and white soldiers, whereas Indian guerillas are raped by dark skinned as well as mestizo soldiers. Similarly, Enloe emphasizes how rape reinforces a sense of community. "Men in war almost always relate to each other in terms of rank and in a lot of circumstances, rape serves to rebond men across personal differences and hierarchies." This becomes even more likely, as in the case of US soldiers in Vietnam, where divisions in rank mirrored racial differences between blacks and whites. While Enloe agrees with Brownmiller that "we need to look at what dynamics between men lead to rape and what dynamics between men influence how men think about rape," she includes a broader range of possible male relationships than the the simple categories of enemies and friends.

Rape can be analyzed by its effects on military rankings, ethnic formations, forms of political recruitment, state formation, family reorganization, gender divisions, economic differentiation, rural/urban differentiation, forms of religious purity and pollution. The effects may not explain why mass rape occurred, but they may clarify why it continues in an organized fashion. It would be important, then, to work outwards from rape, not from a particular social system to rape. One might examine rapes in terms of their specific constellation of practices and discourses about rape and then ask "what are the effects of this way of organizing rape or this way of talking about rape" in multiple fields.

One clear difficulty with the feminist debate on Bosnia is that while American feminists have a rich understanding of gender formation, their thought on ethnicity is remarkably impoverished. American feminists, like the popular press, think of ethnic identity as a property of individuals, assuming that it's somehow inherited or readily visible. Yet, the notion of ethnicity as a distinct property is very difficult to sustain in light of the ethnic conflicts we have today. In the Balkans, for example, there are no immediate racial, physical or even linguistic traits that would facilitate immediate identification, yet group solidarity and ethnic conflict is quite pronounced.

Donald Horowitz has argued persuasively that ethnicity is not a property, but a relation between groups in specified contexts. Horowitz distinguishes between two ethnic systems, what he has called "ranked" and "unranked" systems. In ranked systems, groups stand in clear relations of super- and sub-ordination to one another, positions which are often linked to particular occupational positions. In unranked systems, each group spans the whole available range of occupations and statuses. Horowitz argues that ranked systems are stabler, but can breakdown due to either changes in technology that deskill the superordinate group or upward mobility by the subordinate group. When ranked systems do collapse, the violence is much more extreme than in unranked systems typically because of desperate efforts of the higher-ranked group to maintain its position. By contrast, in unranked societies, of which Yugoslavia was one, each group is potentially a whole society and the groups relate to each other as if they were in a small international system. Conflict in an unranked system is endemic as it is in the international community, but can escalate markedly if one group progresses much more (or less) rapidly than others, and so threatens to monopolize key positions in the economy.

I would argue that this analysis fits well with some feminist analyses of rape. Feminists emphasize rape often involves the intersection of race and gender, but this does not capture the particular saliency racial or ethnic categories take on in relation to rape in a war context. For in the context of a stable state, rapes move through ethnic categories that are relatively secure. In rape in war situations, however, the racial and ethnic conventions are themselves at stake, and their renegotiation is the context of rape, not merely something through which rape moves. In this context, rape is an ethnomarker. This is a difference in degree, instead of kind, but I think this way of conceiving of rape is helpful analytically.

If ethnic groups identify themselves mainly in relation to each other, if ethnic war is a renegotiation of those relations, if rape is a way of marking the new boundaries of those relations, then we may expect rape to vary with the kind of ethnic system we are dealing with. In conflicts involving ranked ethnic systems, rape serves to mark how groups are subordinated and superordinated. Take the example of Peru. According to Phillips, rape serves to minimize hierarchical relations within a ranked, ethnically divided Peruvian military and at the same time to reinscribe those same relations of hierarchy onto the female enemy through rape. By contrast, when conflict breaks out in unranked systems, rape cannot serve the purpose of reinscribing hierarchical relations between males of different ethnicities because no clear pattern of dominance exists with which to begin. What exists instead is competition for an economic niche and also for a demographic majority in an ethnically divided country. As Horowitz points out, in unranked systems, the national census and women's birthrate are highly politically contested among men in politics. When an unranked system collapses, as in Bosnia, women's bodies become a battlefield in which men communicate their rage to other men because women's bodies had been the political battlefield implicitly all along in this kind of system.

I am arguing in other words that we can make better use of the theories of rape I have discussed in this paper if we situate the effects of rape in relation to the kind of ethnic conflicts that exist. Brownmiller's analysis is well-suited for understanding how rape shapes an unranked ethnic conflict but falters when it encounters the complexity of the Peruvian situation. Phillips' analysis of Peru would probably be unsuited for understanding how rape shapes an unranked ethnic conflict since the male solidarity-through-rape presupposes ethnic divisions of a ranked sort. Rape is an ethnomarker in a massive social conflict, but as an ethnomarker its effects differ depending on the ethnic field. Again, this is not to say that rape is an ethnomarker solely in war situations, but that gender and ethnicity gain greater saliency in these situations. Rape in a war context is the means by which differentials of power and identity are defined; it does not turn on other social practices that maintain those distinctions in non-war situations.

So how did rape serve as an ethnomarker in Bosnia? I will answer this by building on the work of the Croatian sociologist Silva Meznaric. Meznaric studied the discourse of rape in the Serbian Albanian conflict in Kosovo in 1990 in order to illustrate how gender is used and abused in ethnic conflict. Like Bosnia, Kosovo was an unranked ethnic society with each group spread across of many occupations. The groups were bound by actors who crossed the boundaries either because of their stable social role (Serbian and Albanian women) or by informal networks within the local community (Albanian men and Serbian women). Phillips argued that in Peru women became targets in the guerilla war because "traditional women's work" was precisely what kept communities together and stable. Meznaric's account, by contrast, how females working in the public sector acted as crucial bridges in a divided communities. It was the daily interactions of Serbian women and Albanian males, the life-work that went on in these interactions, that bound Kosovans together Meznaric argues further that population growth, modernization and migration in the 1980s upset the relationship between Serbs and Albanians: Serbs moved out, but Albanians stayed. While both groups spent enormous amounts of energy and money to emphasize tradition and continuity, there was a serious difficulty. As Meznaric puts it:

"Identification of ethnic markers frequently entails recognition of ethnic identifiers such as language, dress, lifestyle and housingÑthese are customary for every ethnic group. What happens when these identifiers are similar and cannot easily distinguish the groups? In the case of Serbs and Albanians, not only did they demarcate themselves in terms of language, dress, and housing, but they also insisted on separation of schools, instruction, literature, history and so on. Even so, overt signals and signs for defining boundaries were not enough...Classic markers of ethnic difference (language, religion, housing, territory) were insufficient because they could be erased or blurred by the modernization of life"

In this context, rape became an ethnomarker because it defined moral excellence and "by constructing ethnic difference on the basis of one group's cultural proclivity to violence and rape, the boundary between the two groups became fixed." Meznaric covers a media campaign on rape (which presented Albanian men as rapists) that served to sharpen the ethnic border between Serbs and Albanians. Serbian criminal law was amended to include the category of ethnic rape, that is, "if a perpetrator and victim are of different ethnic origin, the criminal law recognizes this fact as aggravating" the crime. As a result, public interaction between working women, who were mostly Serbian, and Albanian men was rendered unpredictable and dangerous. The informal channels connecting Albanians and Serbs in Kosovo were interrupted and "Kosovo ceased to exist as a multiethnic region. Kosovo became a deeply divided society...." There was, in other words, "a total breakdown in communication between ethnic groups."

I want to take a moment to reconstruct what I take to be Mezanaric's implicit argument about Bosnia. In Bosnia, the lines between ethnic groups were blurred; ethnic identification was not easy. Unlike Kosovo, the war in Bosnia was not preceded by a gradual process through which unranked peoples who once knew each other learned to fear each other through rape-talk. In this context, rape served as an ethnomarker particularly in areas where the interaction between groups was densest. Meznaric argues that if one plots the location of camps and information about rapes against a map of the population distribution, "one can notice that massive rapes occur in areas where Serbs are a minority and Muslims are in relative or absolute majority...or else in areas where Serbs are in the majority position, but where there are significant Muslim and/or Croat minorities." Just as in Central America, rape served to socialize the new military recruit and separate him from his civilian compatriots. This is why rapes occurred in front of fellow soldiers and victims who knew the perpetrators. Rape isolated the recruit from his community and prepared him for further military operations. Furthermore, just as in Kosovo, rape was used to underline the moral superiority of the Serbians. This is the best way to understand efforts the videotaping of the rapes which were later used to implicate Croatian forces. Finally, this account generates possible hypotheses about rape in war which one couldn't do with the American feminists' accounts. If it is right, one should predict a decline in Bosnian war rapes over time as the work of isolating the ethnic groups is accomplished.

Fred Halliday writes that "A history of world war as a gendered conflict, ranging from the Japanese 'rape' (in both senses) of Nanking in 1937 through the legitimation of rape by the Red Army as it advanced westwards, remains to be written." Such a history, like the unwritten history of rape and decolonization, would inevitably have to relate gendered conflict to ethnic conflict in specified contexts. The account I have sketched here does so in more specific and contextualized ways than the accounts given by Brownmiller, MacKinnon, Enloe or Phillips; indeed, it incorporates these theories of rape by situating them in relation to variations in ethnic, as well as gender relations.

Selected References

Writings by Enloe, Brownmiller, and MacKinnon are assembled in Stiglmayer, Alexandra, ed. 1994. Mass Rape: The War against Women in Bosnia-Herzogovina, Lincoln: University of Nebraska.

Blatt, Deborah. 1992. "Recognizing Rape as a Method of Torture," Review of Law and Social Change, 19:4, 821-865.

Halliday, Fred. 1994. Rethinking International Relations, Vancouver, Canada: University of British Columbia.

Horowitz, Donald. 1985. Ethnic Groups in Conflict, Berkeley, University of California Press.

Meznaric, Silva. 1994. "Gender as an Ethno-Marker: Rape, War, and Identity Politics in the Former Yugoslavia." In Valentine M. Moghadam, Identity Politics and Women: Cultural Reassertions and Feminisms in International Perspective, Boulder, CO, Westview Press.

Phillips, Julie. 1993. "Crossfire's Targets: Women in Peru Fight Violence from Both Sides," The Village Voice, 13 July, pp. 28-29.

Rape in Mass Social Conflict

Ethnic System Prior to War (from Horowitz)
Analytic Optic Ranked Unranked
Male/Male Interactions
(from Brownmiller)

Rape as a Liminal Act

Peruvian Army 1980s (Phillips),
US Army, 1960-1975, (Enloe)

Rape as an Act of Intimidation and Conquest

Pakistani Army in Bangladesh 1973 (Brownmiller)
Bosnia, 1992-1994 (Brownmiller, Meznaric)

Male/Female Interactions
(from MacKinnon)

Rape as a Reassertion of prior division of labor and status.

Peruvian Army, 1980s (Phillips)

Rape as an Isolating Act

Guatemalan Army and Contra Army, 1980s (Enloe)
Bosnia (Meznaric)

Copyright (c) 2004 by Darius Rejali, all rights reserved. This text may be
viewed, downloaded, redistributed or republished in any medium for the express
and non-commercial purposes of criticism, comment, news reporting, teaching
(including multiple copies for classroom use), scholarship, or research,
provided such use is in accord with Section 107 of the United States Copyright
Act. Archiving, redistribution, or republication of this text on other terms, in
any medium, requires the consent of the publisher and/or author at