Women’s discursive agency in transitional justice policy-making: A feminist institutionalist approach

Abstract Scholars have studied how women’s domestic and transnational civil society activism addresses the gendered nature of transitional justice. In contrast, they have paid scant attention to women’s impact on transitional justice policy-making in institutions. We leverage the feminist institutionalist perspective that makes visible gendered norms, rules, and discourses in institutions. Homing in on women’s influence in parliaments where women are outnumbered by men and marginalised by adversarial discourse, we develop a conceptualisation of women’s discursive agency. Foregrounding discourse in women’s ability to drive change, women’s agency is enacted through their linguistic communication style and substantive normative positions that constitute micro- and macro-level structures of domination. Quantitative and qualitative discourse analysis is applied to a corpus of parliamentary questions about transitional justice in the Croatian parliament from 2004 to 2020. Our results show that women adopt the adversarial style of questioning, which they use to broaden the scope of entitlements and press for reparations for female and male victims. They overcome constraints posed by partisanship and ideology, while constraints of nationalism are less easily broken. The article advances feminist transitional justice by demonstrating how women’s language contributes to dismantling men’s policy domination in institutions, with implications for mixed-sex interactions in non-institutional domains. Video Abstract


Introduction
Feminist analysis of transitional justice has highlighted the importance of women's agency in shaping processes and outcomes of transitional justice. 1 These insights have been gained by studying women's agency in the context of civil society activism within national and transnational networks, 2 women's participation in transitional justice instruments, 3 and women's contributions to transitional justice in the context of post-conflict peace-making. 4The exercise of women's agency has been critical in addressing the gendered nature of transitional justice, whose masculinist conception results from transitional justice 'negotiations that are predominantly male affairs' . 5Having recognised the problem, the United Nations' Women, Peace and Security (WPS) agenda has provided a normative and policy framework for promotion of women's equal representation and integration of their needs in peace and justice processes. 6However, women's increased participation has not ended marginalisation of their input in policy-making. 7Cahn and Ní Aoláin have highlighted that women's formal gains, such as their increased representation in legislatures during peace-building, can be 'nullified or significantly constrained' by the realities of a given context that accommodates male domination. 8As Hamber points out, we need to analyse how structures, for example parliaments, that are essential for post-conflict peace-building reinforce masculine power. 9It has become evident that this analysis requires close examination of internal dynamics of institutions and processes to account for women's inability to exercise their agency. 10his article turns to women's agency in shaping transitional justice policy-making in national parliaments.The extant scholarship has neglected this domain of transitional justice, despite parliaments' critical role in localising global norms of transitional justice. 11Scholars have paid attention to outcomes of parliamentary activity, such as the adoption of discriminatory laws that recognise male victims and marginalise female victims. 12But, scant attention has been paid to various aspects of parliamentary activity, such as parliamentary discourse and the speaking behaviour of members of parliament (MPs), that lead to those gendered outcomes.We address this gap by presenting a conceptualisation of women's discursive agency, which reveals how women challenge male domination in institutions, as exemplified by women legislators' contribution to transitional justice policy-making.
This study is informed by feminist institutionalism that makes visible the gendered nature of institutions, which is constructed and reproduced through male domination. 13The feminist institutionalist agenda is preoccupied with transformative change.To bring about change, we need to better understand how actors can exercise their agency in contexts structured by unequal gender relations. 14Interested in the role of language in gendering politics, 15 we approach parliaments as one type of gendered 'communities of practice' (CofP).These communities construct ways of talking that, in turn, have material impact on power relations between men and women. 16We examine women's discursive agency in parliaments as a community of practice where adversarial discourse is a dominant norm of communication that disadvantages women policy-makers. 17Discourse is thus implicated in perpetuating structures of masculine domination after a conflict. 18omen's discursive agency foregrounds discourse when conceiving of women's ability to enact transformation in line with their identities, attitudes, values, and needs, whereas the possibility of transformation captures the essence of their agential power. 19It is distinct from women's narrative agency, which highlights women's ability to express their needs, for example through story-telling. 20ur discourse-centred interrogation of women's participation in parliaments focuses on how 'agency is spoken into existence at any moment' , 21 while considering micro-and macro-level structures of domination, manifested respectively in the communication style and normative content of public discourses.Accordingly, we propose that women's discursive agency, which captures their power as political actors, is enacted through a combination of micro-level interactional characteristics of women's language with its linguistic features and macro-level normative dimensions of discourse, such as partisanship, ideology, and nationalism.These dimensions are significant because party politics, a liberal or conservative outlook on rights and entitlements, and a view of one's nation as civic or ethnic 22 represent macro-level discursive structures that constrain women's ability to act on behalf of women in post-conflict contexts, mediating their influence on policy-making.
We conduct a multi-modal discourse analysis of an original corpus of parliamentary questions about post-conflict transitional justice in the Croatian parliament between 2004 and 2020.The Croatian case represents a typical case, whose generalisability is contingently limited to post-conflict societies where transitional justice as a norm and practice is gendered. 23We find that female legislators' parliamentary questions about transitional justice are as adversarial as men's, when we measure linguistic features of parliamentary questions that capture direct pressure put on the answerer. 24This result challenges arguments that women's legislative speech is more polite, less adversarial, and, therefore, less influential on policy-making. 25Arguably, the alignment of female legislators' communication style with what is considered to be a masculine norm of adversarialness in parliaments at the micro-level of discourse enables women to challenge other macro-level discursive structures of domination and broaden transitional justice policy deliberation to address women's interests as well as a range of other concerns.Women legislators contest the masculine framing of transitional justice through robust questioning, despite constraints imposed by partisanship, ideology, and nationalism, although constraints imposed by nationalism, related to dominant normative interpretations of conflict, are the most difficult for women legislators to overcome.
Our contribution advances feminist perspectives on transitional justice, which have centred gender in the analysis of the traditionally de-gendered concerns of this field. 26These perspectives have focused on the need for inclusion of harms against women in transitional justice, participation of women in transitional justice, and addressing structural gender inequalities. 27We engage with the question of women's participation, while heeding Davies and True's observation that 'the fixation on the quantifiable nature of the number of women with a seat at the peace table must not become an end in itself, detracting from the substantive post-conflict gender-justice outcomes' . 28Mobilising feminist institutionalism concerned with how the gendered character of institutional norms, rules, and relations produces their gendered effects, 29 this article reveals how gender relations are reordered through women's participation in discourse in parliaments.Our findings challenge binarised and essentialist views on women's language in institutions and reveal women's ability to take on the enduring power of 'hegemonic masculinities' 30 after a conflict and transform masculinist and, to an extent, nationalist conceptions of transitional justice.
We first discuss the role of language in establishing and maintaining masculine domination in parliaments by drawing on the feminist institutionalist analysis in relation to the communities of practice approach.The article then addresses and defines women's discursive agency and its enactment at the micro-level and macro-level of discourse.The section on research design introduces the case study of Croatia and describes the textual data and the coding of parliamentary questions that captures directness as a feature of adversarial discourse.It also presents the methods used to analyse micro-and macro-dimensions of discourse.Next, we analyse and discuss the results of quantitative and qualitative analysis of women's and men's parliamentary questions.The conclusion elaborates the article's contribution to feminist transitional justice.

Gender, language, and domination in parliaments
National parliaments are public domains in which differences between women's and men's language reflect and perpetuate male domination. 31Masculine parliamentary discourse accompanies an enduring historic pattern of men's proportional domination in parliaments, despite the more recent global trend of women's increased representation. 32The impact of different patterns of men's and women's speech behaviour on policy-making is the subject of a growing body of scholarship. 33his literature has shown how female legislators can contribute to the recognition of women's needs and concerns as well as to enriching discussion of issues beyond those conventionally understood as 'female' . 34ngaging with arguments about men's and women's distinct communicative styles both in informal and formal contexts, 35 we operationalise feminist institutionalist analysis by leveraging the 'communities of practice' (CofP) approach put forward by Eckert and McConnell-Ginet. 36The premise of the CofP approach is that understanding the relationship between gender and language requires that 'we look locally, closely observing linguistic and gender practices in the context of a particular community's social practices' . 37Integrating the CofP language-focused approach with the feminist institutionalist analysis allows us to elucidate how gender differences are constituted by interactional strategies and how these, in turn, construct gender relations out of 'a repertoire of similarities and differences and ideas about [men and women]' . 38Notably, by observing the differences in men's and women's language in parliaments, we can infer women's agency enacted through their participation in policy formulation.

Conceptualising women's discursive agency
This study's conceptualisation of women's discursive agency draws on feminist theories of gender and language, which aim at identifying and explaining differences in how men and women speak. 39ven within the same community of practice, men and women may be positioned differently, a difference which, as Cameron argues, will be manifested in their respective speaking behaviour. 40t the same time, scholars have questioned generalisations about the characteristics of women's speech, as evidenced in their speaking style and in their content, which often portray women as 'more conservative, more polite, more cooperative, and more egalitarian than men' . 41These binaries between feminine and masculine speech need to be interrogated when studying women's agency, which can be observed both at a micro-and a macro-level of women's discourse in any given community of practice.We first elaborate the micro-level interactional feature of women's discursive agency.

Directness: The micro-level interactional dimension of legislators' discourse
Adversarial discourse is a prominent aspect of parliaments.This masculine norm of communication is constitutive of asymmetrical gender relations in parliaments, characterised by men's domination and women's subordination. 42Gender-based differences in the mode of interaction can have policy repercussions. 43The way women engage with dominant discursive norms in any given community of practice can undermine or enhance their influence on policy formulation.The aggressive nature of parliamentary discourse can be observed from the content of MPs' speeches.However, linguistic and syntactic features of language are also a potent but less understood aspect of gendered discourse in parliaments.Therefore, the adversarialness of parliamentary questions can find linguistic expression in directness, which refers to 'the level of coercion encoded in the question' . 44Directness is one of the most debated examples of manifestation of difference between women's and men's language.
A considerable body of sociolinguistic research has found that women tend to be more indirect than men in their speech. 45Indirectness is an expression of politeness.Politeness as a feature of women's speech has been observed in a variety of cultural contexts and conversational domains. 46s a linguistic strategy, politeness works well in private domains.It may also be effective in public settings, such as meetings.Politeness can lead to 'communicative satisfaction' 47 and can also encourage collaborative discussion and problem-solving.
But, politeness can be inappropriate or unhelpful in certain public contexts where, as Holmes has argued, it 'can give the impression that the speaker is weak and ineffectual' . 48This is particularly the case in the political domain of parliaments and during parliamentary questions, which are intended to probe government policy.Given the role of discourse in entrenching male domination in parliaments, the exercise of women's discursive agency can be gleaned from whether female legislators enact or reject the interactional communicative norm of adversarialness at the microlevel of parliamentary interactions.This aspect of women's discursive agency is inseparable from the content of their contributions.

Partisanship, ideology, and nationalism: The macro-level dimensions of legislators' discourse
Sociolinguists have warned that the understanding of parliamentary discourse as gendered -that is, adversarial and masculine as opposed to cooperative and feminine -risks entrenching misplaced binary conceptions between impolite men and polite women. 49Not only do these assumptions need to be tested empirically, but the binaries between 'dominant' and 'adversarial' men and 'dominated' and 'polite' women can also be unsettled when other components of identity, such as ethnicity, class, and (sub)culture are taken in account. 50These intersections manifest themselves in discourse and need to be addressed when attempting to capture women's discursive agency.
Partisanship, ideology, and nationalism are macro-level normative structures that mediate policy positions on the legacy of human rights violations in post-conflict contexts.Partisanship as a form of lasting associative relation between the partisan and their party is defined by a distinct set of obligations and in-group loyalty. 51Scholars of legislatures and political representation have shown that for 'women legislators, representing women is an inherently partisan endeavour' . 52Concerns of partisanship appear paramount; female legislators tend to give precedence to party-sponsored non-female issues rather than step out of the party line and advocate in favour of women's interests. 53Whether women legislators follow or challenge their party line on transitional justice corresponds to one macro-level dimension of their discursive agency.The next dimension concerns their ideological preferences.Scholars have observed that women tend to be more liberal than men even within the same party. 54In the context of transitional justice, women's agency can be observed in whether and how they challenge communitarian approaches to human rights and entitlements.Lastly, women's discursive agency in post-conflict transitional justice needs to consider the normative impact of 'post-conflict nationalism' , 55 which constrains the scope of political positions considered legitimate.Nationalism demands ultimate loyalty to the nation, whereby national identity overrides other identities, including gender.It is rooted in the ethno-centric interpretations of violent conflict that construct victim hierarchies. 56These hierarchies afford recognition only to the victims from one's own national group and deny recognition to victims from adversary group(s).Yet recognition as a victim, regardless of their national identity, precedes 'the possibility of redress' 57 and, therefore, is necessary for broadening the scope of a transitional justice policy in post-conflict contexts.
Consequently, women's discursive agency in parliaments can be observed in both how women engage with adversarial discourse and how they navigate normative discursive structures imposed by partisanship, ideology, and nationalism.These two dimensions of discourse that reflect the communication style and the content of their parliamentary speech can be captured by two distinct empirical approaches: Conversation Analysis (CA) and Critical Discourse Analysis (CDA).Before turning to the methods and data used in this study, we present our research design in the following section.

Transitional justice in Croatia: A case study
We apply a case-study research design.Croatia's transitional justice is a typical case of gendered transitional justice in a post-conflict country, representative of 'a broader population' of cases in other post-conflict contexts. 58Specifically, women are outnumbered by men in the Croatian parliament, a pattern which is typical of other Western and non-Western parliaments.Further, the masculine character of Croatia's transitional justice reflects common, gendered, and ethnic framing of transitional justice in other post-conflict contexts. 59For example, a law on wartime rape established and legally operationalised an ethnically-centred definition of wartime rape. 60Such an intersectional nature of transitional justice in the Croatian case study delimits its generalisability to that universe of cases of gendered post-conflict transitional justice where conflict was fought along the lines of ethnic identity. 61

The background
Transitional justice policy in Croatia is embedded in the dominant nationalist understanding of the 1991-5 war, called the 'Homeland War' ('Domovinski rat' in Croatian), fought between Croats and Serbs from Croatia, who were supported by neighbouring Serbia.The associated narrative, supported by the majority of ethnic Croats, is that Serbia was an aggressor attempting to destroy both the Croatian state and the Croatian nation. 62This view of the war was made official when the Croatian parliament ('Sabor' in Croatian) adopted the 'Declaration on the Homeland War' in 2000. 63The declaration introduced a single, hegemonic ethno-centric narrative about war.
At the same time, party politics, involving the left-of-centre parties spearheaded by the Social Democratic Party (SDP -from the Croatian 'Socijaldemokratska partija Hrvatske') and the Croatian Democratic Union (HDZ -from the Croatian 'Hrvatska demokratska zajednica'), the wartime party, impacted transitional justice.The SDP was prepared to support Croatia's collaboration with the International Criminal Tribunal for the former Yugoslavia (ICTY), the main instrument of transitional justice in the early post-war years, 64 arguing that Croats who committed war crimes should be held accountable like war criminals from other ethnic groups.Unlike the SDP, the HDZ professed and practised cooperation with the ICTY while maintaining that suspected Croatian war criminals were in fact blameless defenders of the Croatian nation.At the same time, the HDZ has nurtured a strong bond with war veterans, passing legislation supporting them and their families. 65In addition, supported by the right-wing intelligentsia and conservative segments of Croatian society, 66 the HDZ has accused SDP members and supporters of being unpatriotic or 'improper' Croats. 67The HDZ has also attempted to discredit the left for failing to stand by war veterans during and after the war. 68These sensitive charges have narrowed the space for the centre-left's political manoeuvring, 69 revealing its vulnerability on national issues.
It is within this national and partisan context that transitional justice policy in Croatia has emerged as a gendered policy field.Nationalist framing of transitional justice has been a subject of the growing body of scholarship on transitional justice in Croatia. 70 on the persistence of public discourse about the war that elevates the sacrifice made by veterans. 71However, little is known about how this powerful masculine discourse, which defines the political parameters for the formulation of transitional justice policy in the Croatian parliament, is contested, including the role of women legislators therein.

The corpus
We analyse an original textual corpus comprised of the transcripts of 390 parliamentary questions asked in the Croatian parliament about issues concerning transitional justice, mainly referring to the 1991-5 Croat-Serb conflict.The questions were extracted from the official Croatian parliament webpage by applying multiple search strategies that encompass comprehensively terms related to transitional justice in the Croatian context. 72They do not include questions on other policy areas (or topics), although certain aspects of questions about transitional justice also relate to other policy areas, for example social policy or education.The questions were asked orally by MPs during the designated question time in the Croatian parliament from 2004 to 2020.

Discourse analysis methods: Conversation Analysis and Critical Discourse Analysis
Our examination of the discourse of female legislators uses both Conversation Analysis (CA), focused on the sequential organisation of talk, and Critical Discourse Analysis (CDA), which reveals the role of discourse in constructing and refracting the social world. 73CA and CDA are analytically different, and their analytical pay-offs differ as well. 74onversation Analysis, which can be both qualitative and quantitative, focuses on the microlevel of 'talk in action' . 75What distinguishes CA from other types of analysis of discourse is its attention to the 'organisation of conduct within interactions' . 76We use CA to analyse the interactional dimension of parliamentary questions that emerges from the bottom-up treatment of textual data. 77The application of CA, which focuses on the orderly nature of talk in conversational sequences and turn-taking, is combined with CDA, which captures disorderly interactions and reveals obscured meanings that reflect, reinforce, or challenge the lifeworld beyond speakers' interactions, while taking into account how speakers are positioned within the macro-level of discourse.The constructionist assumption in CDA defines discourse as a set of practices implicated in the social production of reality, including subjects, objects, and relations between them. 78e draw on multi-method analysis involving CA and CDA to capture the 'dilemmas of voice' , since the investigation of talk in action can highlight the substantive content of speech and reveal hidden social implications of interactions. 79In this study of women's discursive agency in transitional justice policy-making, we combine CA and CDA to analyse how female legislators orient themselves towards the adversarial nature of parliamentary discourse to evaluate their use of language at the micro-level of discourse in the community of practice represented by a national parliament, and to capture how their discourse is fluid, contested, and embedded in normative power structures at the macro-level of discourse, 80 as defined by partisanship, ideology, and nationalism.
To measure directness as an interactional feature of questions, which we analyse quantitatively, we created a dataset by compiling and applying a coding scheme, anchored in CA.Directness denotes the extent to which a question is blunt 81 and coerces the answerer to respond.The level of directness is expressed in self-and other-referencing phrases with corresponding syntactic forms.Indirect questions tend to contain self-referencing phrases, such as: 'I wonder whether…' , 'I would/want to ask you…' , etc., or other-referencing phrases, such as: 'Can you/Could you tell us …' , or 'Will you/Would you tell us …' . 82By contrast, direct questions lack these polite frames.As such, they reflect the absence of deference and represent adversarial engagement. 83The entire corpus was coded independently by two coders to ensure that coding is reproducible. 84The inter-coder reliability test demonstrates the acceptable level of agreement. 85A is combined with a qualitative approach -rooted in CDA -to questions asked by male and female MPs that were coded as direct, i.e.only to adversarial questions.We hold this feature of questions constant so that we can rigorously identify differences, if any, in the content of men's and women's parliamentary questions.Comparing men's and women's adversarial questions allows us to establish how women exercise their agency and to what extent it contributes to dismantling masculine and other discursive hegemonies in transitional justice. 86

Results and analysis
This analysis of women's discursive agency in transitional justice policy-making locates female MPs' discourse in national parliaments as a type of 'community of practice' . 87A community of practice conditions language use through the habitual interaction of its members, which in the case of parliaments is characterised by male domination.As Table 1 shows, most questions (82.56%) in our corpus were asked by male MPs, as compared to 17.44% asked by female MPs.The share of questions asked by female MPs about transitional justice is somewhat below the share of women in the Croatian parliament, who have accounted for about 27% of legislators on average therein since 2000. 88This indicates men's marginal domination of parliamentary questions about transitional justice and begs the question whether there is masculine control of transitional justice policy formulation.We first investigate the communication style in the Croatian parliament.The analysis of directness -the linguistic feature of parliamentary questions -allows us to establish whether men's numerical domination in parliament and in parliamentary questions coexists with adversarialness as a masculine discursive norm of parliamentary speech. 89Table 2 shows that adversarialness is the dominant feature of parliamentary questions about transitional justice in the Croatian parliament.We found that of 390 questions asked in the period between 2004 and 2020, 82.31% were direct questions that put pressure on the answerer, while 13.33% were indirect questions.
Considering that patriarchy is entrenched in parliaments through dominant adversarial discourse, 90 we further check how men and women participate in this type of discourse.We find that the distribution of direct and indirect questions is similar for men and women legislators.As seen in Table 3, the vast majority of questions by both men and women about transitional justice were direct; 86.97% of questions asked by men were direct, while 81.82% of questions asked by women were direct. 91he use of a particular linguistic form or conversational strategy depends in part on previous interactive experiences; however, as Tannen points out, it also represents the individual's choice and preference for some strategies over others. 92Our findings, based on a linguistic measure of adversarialism, indicate that women legislators adopt the adversarial norm of discourse in parliaments when they ask parliamentary questions about transitional justice.Challenging the stereotype of female legislators' style as less adversarial and more cooperative than that of male legislators, 93 our results contribute to growing evidence pointing to many features of women's speech in parliaments and in politics, more generally, that do not differ from men's.For example, women's speaking style has become more adversarial over time in the UK parliament. 94Similarly, when testing arguments about 'gendered deliberation' , 95 scholars have shown that there is no difference between women's and men's quality of deliberation in parliaments and civil society fora. 96The analysis now turns to macro-level dimensions of female legislators' discourse in transitional justice policy-making and examines whether it is different from men's, when the women's style of questioning is adversarial like men's.
Seven parliamentary questions analysed below represent different types of interactions in the corpus of 321 direct questions about transitional justice asked by male and female legislators.After reviewing and classifying these questions, we selected the questions that are representative 89 Walsh, Gender and Discourse. 90Mackay, Kenny, and Chappell, 'New institutionalism through a gender lens'; Holmes, Women, Men and Politeness. 91Proportions are calculated excluding the questions coded as neither.An example of this type of 'neither' question is a rhetorical question.The inclusion of these questions does not change substantially the proportions of direct and indirect questions by gender. 92  of different combinations of askers' and answerers' attributes (e.g. the government and opposition) and specific topics within the transitional justice portfolio.The qualitative analysis probes how legislators navigate party politics (partisanship), invoke the concepts of human rights and equality (ideology), and frame the nation (nationalism).A comparative analysis of the content of men's and women's parliamentary questions reveals whether they are characterised by a feminist or masculinist engagement with transitional justice.The first question was asked by a male opposition MP from the centre-right nationalist party HDZ.It was addressed to the prime minister, who was from the centre-left SDP.The question was asked during the 555-day-long protest by war veterans in front of the Ministry of Defence (2014-16).The protest was triggered by Deputy Defence Minister Bojan Glava š evi ć 's statement that all civilian victims of war, including ethnic Serb victims, should be treated as war victims. 97is statement challenged the accepted norm that only ethnic Croat war veterans should be entitled to state support by virtue of their participation in the conflict.The MP asked: Why are you doing it?We all ask ourselves this question.And, they [war veterans] asked me to ask you how you can fall asleep while being aware that the most severely disabled have been on the road for 90 days; and, you continue to question their honesty and attempt to diminish their loyalty to the homeland as well as their honourable role in the Homeland War by revealing their earnings while they are nailed to a wheelchair for the rest of their lives.Don't you dare insult them more, belittle them!Do you have a conscience at all, or are you so in love with yourself and your ministerial chair?
The opening line of this question is confrontational, signalling ideological confrontation.The MP's subject position is embedded in an understanding of the primacy of the ethnic nation.The question reflects the nationalist understanding of the 'Homeland War' as a rebirth of a nationstate, whose primary duty is to safeguard entitlements of war veterans.The MP identified with the veterans, as illustrated by his erasure of the difference between himself and the veterans in the ambiguous use of the pronoun 'we' . 98The MP acted simultaneously as the veterans' spokesman and as an opposition politician, openly disagreeing with the policies of the centre-left government.His partisanship is evident in the way he discredits the proposed amendments to the 2013 Draft Law on the Rights of Croatian Veterans and Members of their Families, which aimed to revoke their special treatment and entitlements.Although this MP originally supported the proposed amendments, 99 97 Koska and Matan, 'Croatian citizenship regime' , p. 143. 98Norman Fairclough, New Labour, New Language (London: Routledge, 2000), pp.35-7. 99Croatian Sabor, War Veterans' Committee, Session on 29 November 2013, available at:{https://www.sabor.hr/radna-tijela/odbori-i-povjerenstva/izvjesce-odbora-za-ratne-veterane-o-prijedlogu-zakona-o-12}.about a year after the veterans' protests began, the MP characterised the SDP's proposed law as disloyal to the legacy of the 'Homeland War' because it stated that all war victims should be treated as war victims. 100His alignment with the ethno-centric interpretation of the conflict is consistent with his support for privileging the veterans to whom the nation purportedly owes its freedom. 101urthermore, the MP asserted that the veterans' 'loyalty and honesty' should not be questioned, tacitly indicating that veterans must not be accused of any wrongdoing (whether during the war, such as war crimes and human rights violations, or after the war, such as financial malpractice related to their benefits).
His question reflected the dominant nationalist narrative of the war, which limited the range of contributions to transitional justice policy-making that are considered legitimate.Additionally, the MP shamed the prime minister for being 'in love with [his] ministerial chair' .This attack across party lines demonstrates how partisanship intersects with nationalism: the prime minister was criticised for daring to question entitlements of war veterans.Lastly, the silence in this question reaffirmed nationalist framing of transitional justice as collective and explicitly gendered.The MP glossed over the fact that the protests for reparations were not supported by all veterans, let alone all Croatian citizens; 102 neither did he mention the involvement of women and minorities in the war on the Croatian side.His question shows how partisanship and ideology are used to construct transitional justice as a gendered and ethnicised policy field that prioritises the recognition of men who are members of the dominant ethnic group. 103he next question illustrates the reverse position of Croatia's main parties: an MP from the opposition, centre-left SDP addressed a question to the centre-right government run by the HDZ.The question was prompted by a strike by teachers, who were dissatisfied with low salaries and the government's offer of 'only' a 2% pay increase. 104His question was: Prime Minister, have you considered stopping giving many billions to the Catholic Church?Further, [have you considered] finally putting an end to the ever-increasing number of veterans and their privileges?Also, preventing the President of the Republic from distributing the privileged pension to whomever she wants and how she wants, and finally [have you considered] ending money transfers to Herzegovina [in Bosnia and Herzegovina]?You might then have enough [money] to meet the demands of those who teach children.
Like the previous question, this question is confrontational, openly questioning the priorities of the ruling party.It was motivated by the unequal financial treatment of different social groups.The male MP singled out for his criticism the inflated privileges enjoyed by war veterans at the expense of teachers and, by implication, children as well.The question and the choice of words illustrate the challenge of criticising the government, whose legitimacy is largely built on the sacrosanctity of the 'Homeland War' and whose critics are often portrayed as disloyal and unpatriotic.Referring to 'children' let down by nationalists in power gave this MP the leverage to criticise the government and its nationalist policy.Ideologically, he stood for universal benefits as opposed to privileging certain groups, such as war veterans.He portrayed the government as irresponsible, oriented towards the past and its nationalist mythology. 105 discursive strategy for challenging and discrediting the 'deserving' few: the Catholic Church in Croatia, the ethnic kin, i.e.Croats in Bosnia and Herzegovina, and war veterans. 106These groups have been symbolic pillars of Croatia's conservative ethno-nationalism, which defines the Croatian state in primarily religious and ethnic terms. 107This MP challenged indirectly the concept of an ethnic nation in an attempt to demystify the 'Homeland War' . 108Demystification was also achieved by his criticism of the veterans' morality and their abuse of privileges.The MP referred to an 'increasing' number of veterans.This is a reference to the abuse of veteran status in Croatia, whereby people resorted to fraudulently claiming veteran status because it brought privileges and financial benefits. 109The question marks an attempt to shift away from the nationalist construction of privilege and redefine the basis of the transitional justice policy, despite the risks of being portrayed as disloyal to the nation.By advocating care for all children, he affirmed civic understanding of the nation, as supported by the SDP, and demonstrated a critical approach towards the war.
The last of the three illustrative questions by men was asked by an independent MP who addressed a member of the SDP government.The question concerned redress for gender-based sexual violence during the Croat-Serb conflict.He asked: What will be done to find and extradite the Croatian police officer convicted of the crime of rape in the Homeland War [and] who is on the run outside Croatia?Until when will he keep receiving the salary from the Ministry of Interior Affairs and will he be made to return the pay he has received since he was indicted?
The independent MP, critical of all parties, whose discursive agency is not constrained by partisanship, raised the question referring to one of the two police officers of Serb ethnicity who raped a Croatian woman on the eve of the Croat-Serb conflict in 1991. 110She subsequently recognised the offenders and reported them to the police in 2000. 111At that point in time, both worked for the Croatian police force in Vukovar.They were suspended from duty in 2002 after the indictment was filed but continued to receive 50% of their salary.One policeman escaped to Serbia in 2002. 112The other was tried in Croatia in 2006 and found guilty in 2012.Just before he was scheduled for sentencing, he escaped to neighbouring Bosnia and Herzegovina to evade justice.Croatia's Supreme Court reduced his five-and-a-half-year sentence to two years in 2013. 113He died in Bosnia and Herzegovina in the same year. 114he MP was entrusted by women in his constituency (as we learn from the preface of his question) to ask the question about sexual violence during the 'Homeland War' .He delivered the question as a demand for extradition and the return of the salaries paid to the offenders.The victim's gender and ethnicity were not mentioned.In other words, a human rights issue was framed as a case of financial crime, marginalising the victim's gendered experience of war.This independent MP was not constrained by partisanship and could have foregrounded the victim's right to justice and opened the debate about justice for women victims of sexual and gender-based violence, which also included Serb victims of wartime rape.This question demonstrates how women and their need for recognition are marginalised by certain framings of transitional justice issues, even when the issue of redress for conflict-related sexual violence that they suffered is raised in the parliament.
Our analysis now turns to the questions about transitional justice asked by female MPs, whose linguistic style is adversarial, to find out whether women exercise their discursive agency to promote feminist transitional justice.
The first question, asked by an MP of the centre-left SDP party was addressed to a government minister from the same party.The question concerned the legislation regarding wartime sexual and gender-based violence, which had been in preparation for a long time.She asked: I am asking you a question relating to the problems involving women victims of wartime rape.More than a year ago, the preparations started for the systemic resolution of this issue with legislation that would finally address, on the one hand, their status as victims of rape, and, on the other, the ways to compensate them.All of us cannot make up for the twenty years of the state institutions doing nothing about this issue; we cannot compensate for lost documents and statements which the women have given several times, and for humiliation they have suffered.But, we can offer them some satisfaction with this law, which is in the making.Mr Minister, please tell us at what stage is the preparation of this Law [on Sexual Violence during the Homeland War] and how do you plan to resolve the status of these women?
Like the previous question asked by a male MP, this question was about the issue of rape during the 'Homeland War' .It was one of few parliamentary questions about the recognition of victims of sexual and gender-based violence, rather than reparations for war veterans.Here, transitional justice policy was framed as an issue of justice owed to civilian victims, especially women, who deserve treatment equal to that given to predominantly male military victims.According to the United Nations Development Programme (UNDP) estimates, there were between some 1,500 and 2,200 (or 2,400 depending on the methodology) cases of serious sexual violence in the Croat-Serb war, most of them women. 115his confrontational question put pressure on the government led by the MP's own party to account for the lack of action on the issue of wartime sexual and gender-based violence.The MP acted as a representative of all women victims. 116She not only challenged the government to acknowledge women victims symbolically and financially but also expressed sympathy with their sense of mistreatment over having to testify multiple times about the crimes they suffered.Without mincing words, she stated plainly that women victims of violence had been 'humiliated' .Identification with survivors of rape based on this MP's gender identity empowered her to override the constraints of partisanship and criticise successive governments, including those led by her party.It took some twenty years from the end of the war to acknowledge women victims of 115 UNDP, Procjena broja ž rtava seksualnog nasilja tijekom Domovinskog rata na podru č ju Republike Hrvatske i optimalni oblici obe š te ć enja i podr š ke ž rtvama (The estimate of the number of victims of sexual violence during the Homeland War on the territory of the Republic of Croatia and optimal forms of compensation and support for victims) (Zagreb: UNDP, 2013), (pp.33-38).
116 Jane Mansbridge, 'Should blacks represent blacks and women represent women?A contingent "Yes"' , The Journal of war,117 unlike war veterans, whose benefits received legislative approval even before the war was over (in 1994). 118t the same time, this MP avoided talking explicitly about war veterans and their entitlements.This tactic reflects, once again, the difficulty of criticising publicly the nationalist narrative of the war.Instead, she focused on the need to acknowledge suffering of women victims during the war, since the debates in the Croatian parliament have frequently concerned the suffering of the veterans and their contribution to the war effort and nation-building.Speaking about justice for women victims of sexual and gender-based violence in the parliament, she gave voice to the marginalised.Notably, she did not position herself explicitly against the dominant narrative about the 'Homeland War' , in which the prioritisation of male and ethnic Croat victims of armed violence is embedded.Further, she spoke about 'these women' survivors, without explicitly addressing the fact that the victims were both ethnic Croats and Serbs.This is another indication that constraints of partisanship are easier to overcome than constraints of nationalism.
The next question was asked by a female MP of the centre-right HDZ and was addressed to the deputy prime minister, who was from her own party.Her question was: Will the costs of preparatory courses for enrolment in higher education institutions in the academic year 2009/2010 be financed again?
The background to the question was legislation passed in early 2005.The law gave priority to the children of war veterans in university enrolment if they satisfied certain criteria.The Supreme Court abolished this law when the HDZ government was in power (December 2006).The court deemed it unconstitutional on the grounds that it contravened the principle of equality, the rule of law, and respect for the autonomy of higher education. 119Instead, the HDZ government decided to help the children of war veterans by establishing grants for preparatory courses for secondary schools and universities. 120This decision reflected the HDZ policy of prioritising support for war veterans and their families.
The MP skilfully manoeuvered around the issue of the entitlements of the children of war veterans over other children.She relied on a tacit dissatisfaction within the HDZ about the abolition of the law to ask a question critical of her own party's policy. 121Her question did not refer directly to the children of war veterans.Nonetheless, the intent of her question is clear; she asked whether 'the preparatory courses will be financed again' [our italics].The question indicated that the children of war veterans need and deserve support because their parents had fought in the 'Homeland War' , consistent with the HDZ narrative that the nation is indebted to the veterans and their families. 122Although she drew attention to the rights of children, this MP located these rights within the nationalist understanding of the 'Homeland War' .She was a good partisan and loyal to party ideology.Although this parliamentary question contributed to the diversification of transitional justice policies in invoking the welfare of all children, the framing of transitional justice policy was within the bounds of the nationalist narrative of the war.
The next question, concerning the rights of children, was asked by a woman MP from the centreleft SDP and addressed to a female member of government from the centre-right HDZ.As in the previously examined question, the MP scrutinised the law that gave educational privileges to the children of war veterans between January 2005 and December 2006.She asked: Therefore, encouraged by the questions of veterans who are disabled, I am asking you, what are you planning to do to correct this injustice towards the children of the disabled and the children of war veterans who have been paying their fees until now, and what are you planning to do to grant equal rights to all children, that is the children of all disabled and all war veterans?Unlike the previous example, in which the female MP directed a question to a member of her own party, this question illustrates how partisanship plays into a confrontation over the framing of compensation for war victims.Once again, children were at the forefront of a partisan challenge, but in this case the approach to education-related privileges was radically different.This MP accepted the notion that children of war veterans were entitled to certain rights but also insisted that children of all disabled people were entitled to the same rights. 123She signalled her position that rights are universal and that therefore no child should be discriminated against.Like other MPs, she was aware of the nationalist narrative about the war and of the risks of appearing unpatriotic if she were to criticise it openly.Her question put the HDZ minister in the difficult situation of having to explain the differential treatment the law established in the process of enrolment, by privileging veterans' children over other children with disabled parent(s).Furthermore, the law also privileged children of some war veterans, because not all veterans' children benefited from it.The question points to an unexpected alliance between a centre-left politician and war veterans, who are traditionally aligned with centre-right parties. 124It reflects her political acumen in that she seized a political opportunity presented by the fact that traumatised war veterans were less likely to vote for the nationalist parties. 125She therefore acted on their behalf.However, rather than fall into the nationalist pattern of advocating entitlements for war veterans only, she made the case for the support of all children.The criticism of the government's discriminatory policy in this parliamentary question was embedded in her party's advocacy of the rule of law, universal rights, and equal treatment of all children.In this way, she was also able to criticise the government's policy-making, which reflected the nationalist understanding of the conflict.
The last question selected for analysis was posed by a female MP who belonged to one of Croatia's national minorities.The question was addressed to the minister of culture, from the ruling HDZ.In the preface to the question, the ethnically Serb MP stated that her question related 'to one of 3,000 monuments, dedicated to the anti-fascist struggle, which have been damaged and destroyed' .She asked: My question is, what are you going to do in this case as well as in other similar cases so that we don't have a situation like this again?
The question referred to memorialisation as a form of transitional justice.She had pointed out that this particular monument was dedicated to 816 partisans (who resisted the Nazis in the former Yugoslavia during the Second World War) and 1,836 victims of fascist terror from different ethnic groups, including Serbs (carried out by the Independent State of Croatia, a fascist puppet state, during the Second World War).Instead of being repaired and protected, as had been suggested by the conservation department in Zadar and the Croatian police, 126 the monument was to be pulled down under pressure from the local, right-wing HDZ leadership.
The MP's ideological position reflected her support for peaceful reintegration of Serbs into Croatia.She recognised the need to bring together different histories into a collective memory of the Croatian nation.The monument to which she referred was built in a small town of Gra č ac, populated by both ethnic Serbs and Croats.The Serbs were a majority population in the area before the start of the war in 1991; they became a minority, after many left in August 1995 as a result of the Croatian military operation aimed at capturing the Serb-populated territory of the self-declared Republic of Serb Krajina. 127Since 1991, the monument has been considered a symbol of the suppressed Croatian nation and its independent statehood, especially by right-wing politicians. 128For them, the multi-ethnic communist Yugoslav federation, of which Croatia was a part, was an illegitimate state. 129The MP presented the monument's destruction as an attempt to destroy a multi-ethnic heritage that, according to her, was vital for nurturing a new post-conflict Croatian democracy.
Speaking as a member of a minority community, she projected a commitment to building a civic nation by criticising the nationalist framing of the transitional justice policy (specifically regarding memorialisation).However, her question also indicates the political sensitivities that members of the Serb minority have to navigate when addressing the questions about transitional justice, past and present.She opted not to refer to those ethnic Serbs and Croats who did not value their common historical heritage. 130Ultimately, her critique of the nationalist framing of transitional justice policy drew attention to memorialisation as a form of transitional justice and raised the politically delicate issue of the rights of minority Serbs in Croatia.

Conclusion
Identifying differences in the language between female and male legislators offers a novel way of understanding how transitional justice is constituted as a gendered field.While language can be used to maintain male dominance in institutional settings, research that foregrounds gender and language in specific 'communities of practice' can also yield insights into how this dominance can be dismantled. 131Dismantling male domination depends in part on capturing women's discursive agency, which, as we have shown in this analysis of parliamentary questions in the Croatian parliament, is constituted both by the interactional features of language and by the substantive normative content of their discourse.
Discourse as a form of social practice helps produce and reproduce power relations and encodes particular assumptions as common sense. 132The implications of this dynamic in relation to women's ability to exercise their agency require us to interrogate the kind of discourse that The police warned that repair is needed for safety concerns.Goran Borkovi ć , 'Prvo dinamitom pa HDZ-om' (First with dynamite and then with HDZ), Novosti (News) (2 January 2020), available at: {https://www.portalnovosti.com/prvodinamitom-pa-hdz-om}. 127Sokoli ć , International Courts and Mass Atrocity.is normalised and legitimised in different 'communities of practice' . 133Directing attention to the question about how discursive norms gender power relations in parliaments and impact on policymaking has allowed us to enhance feminist institutional analysis preoccupied with the processes of gendering in institutions through formal and informal rules and procedures. 134omen's engagement with masculinist hegemonic structures in different policy areas in postconflict contexts, including in transitional justice policy-making, has been poorly understood. 135iscovering latent patterns of masculine domination vitally depends on an empirical strategy. 136ven in legislative studies, the style and the substantive content of women's parliamentary speech have been analysed in isolation from each other. 137Moreover, speaking style has commonly been inferred from the content of legislators' speech rather than from its linguistic features. 138vercoming these analytical shortcomings, we show that women can annul male domination by aligning with dominant, masculine norm of adversarial interaction, which, in turn, opens up political space for them to make their mark on transitional justice policy-making.In the Croatian parliament, women's adversarial questions broadened the agenda by addressing the needs of different groups within society as a whole, such as women, children, and national minorities.Their questions reflect women's political acumen and their ability to transcend constraints posed by partisanship.They openly criticise their co-partisans, especially when it comes to substantive representation of interests of women victims of wartime sexual violence.But, considering the political costs of being perceived disloyal to the nation, female legislators largely refrained from openly criticising Croatian nationalism.
This article furthers our understanding of women's agential power in transitional justice.It contributes to a better understanding of structures that explain women's continued marginalisation despite their increased inclusion and participation in transitional justice processes promoted by the WPS agenda. 139We develop discourse-focused feminist institutionalist analysis of men's domination, previously illuminated by studying the representational function of discourse in the poststructuralist tradition. 140Our analysis of men's and women's 'talk in action' 141 reveals the power dynamics at the level of micro-level interactions in mixed-sex institutions.When women challenge domination at the micro-level of discourse, this can open up space for asserting gender-sensitive policy positions and contribute to dismantling masculine gendered hierarchy in institutions.These opportunities are also likely to exist at the micro-level of discourse in the public sphere, more broadly, and in informal settings.This assumption charts new directions for theoretical and empirical explorations of feminist transitional justice and possibilities for its practical realisation.
Lastly, as this study shows, future interrogations of women's agency also need to go beyond the advancement of only gender-specific concerns, although their inclusion in transitional justice is paramount.As Charlesworth warned, women's agency should not be restricted to 'women's work' , with women's influence assessed solely in terms of how they advance women's issues. 142omen's participation 'offers better solutions for all of the dilemmas of transitional justice' . 143Our analysis shows that women's contributions enrich and diversify agenda-setting and policy deliberation on transitional justice, 144 which include, but are not limited to, policy-making on exclusively women-related issues.Ultimately, the task is to find out how women can ensure that their voices are heard and that they matter in transitional justice for women they represent as well as for all others seeking justice.

Table 1 .
Frequency of parliamentary questions by gender.

Table 2 .
Directness in parliamentary questions.

Table 3 .
Directness by gender of MP.