Women, armed conflict and language – Gender, violence and discourse

Abstract

Facilitating critical reflection on the words and concepts used to write policy enables practitioners to avoid unconsciously reproducing the different forms of oppression and exclusion that their policies seek to overcome. In this article, the author provides an analysis of Chapter 5.10 of the United Nations Integrated Disarmament, Demobilization and Reintegration Standards, arguing that policy makers, scholars, students and practitioners cannot avoid making and/or changing meaning through their well-meaning interventions, but that this need not lead to political or practical inertia.

The politics of language

When I was a graduate student, I attended a conference and posed a question to a panel of senior academics, explaining my doctoral research (an analysis of United Nations Security Council Resolution 1325 on women, peace and security)³ as a precursor to asking how I might make this research useful to practitioners and, ultimately, to those whose lives were affected by the violence and insecurities of which I wrote. I was told that such engagement was unlikely, that I wasn't doing ‘that kind of research’. The response I received was premised on the assumption that theoretically driven work – work that engages with French philosophers rather than statistical analysis software – is of little use to policy makers and stakeholders. This is a widely held and frequently repeated opinion which will be critiqued in this article. Against the suggestion that paying close analytical attention to language necessitates either using overly complex terminology to explain relatively simple ideas or produces research that is of no practical relevance (or both), it is proposed instead that language is the medium of politics, as well as its vehicle, its shield and its disguise. We must know language, theoretically speaking, before we can comment on its effects (and the relevance – or otherwise – of studying it); the formulation of a politics of language is one of the most directly practical efforts I have ever made. Language matters. We know that language matters. The UK government knows that language matters. Article 29B(1) of the 2006 Racial and Religious Hatred Act states that ‘A person who uses threatening words or behaviour, or displays any written material which is threatening, is guilty of an offence’.⁴ Why else would the government legislate against the use of specific types of language other than to illustrate that language affects and effects our being in the world?

In our personal lives, we know that language matters, that words are constitutive of reality. There are words that have been excised from our vocabularies, deemed too damaging to use. There are forbidden words that children whisper with guilty glee. There are words we use daily that would be meaningless to our grandparents. Moreover, the cadence and content of our communications vary by context; words that are suitable for the boardroom may not be appropriate for the bedroom or the bar. In our personal lives, we admit that words have power, and in formal politics we do the same. It is not such a stretch to admit the same in our professional lives. I am not claiming that all analysis must be discourse-theoretical – must take language seriously – to be policy-relevant, for that would clearly be nonsense. I am, however, claiming that poststructural theories of language have much to offer policy makers and practitioners, and arguing that in order to understand how best to implement policy we first need to understand ‘how’ a policy means, not just what it means. That is, we must understand a policy before we can implement it. This article argues that we need to engage critically with how that understanding is mediated through and facilitated by our ideas about the world we live in. If we are to avoid unconsciously reproducing the different forms of oppression and exclusion that different forms of policy seek to overcome, we need to take seriously Jacques Derrida's suggestion that ‘il n'y a pas de hors-texte’.⁵

[T]here is nothing ‘beyond the text.’ […] That's why deconstructive readings and writings are concerned not only with library books, with discourses, with conceptual and semantic contents. They are not simply analyses of discourse […]. They are also effective or active […] interventions that transform contexts.⁶

Writing from a discipline (International Relations) that has a very clear idea of what constitutes an ‘effective or active’ intervention (and writing about language is not it), I have a strong desire to have the politics I espouse recognized as a legitimate and useful form of intervention. It ‘is not politics as a means to truth but as the activity of contesting truths’.⁷ Discursive practices maintain, construct and constitute, legitimize, resist and suspend truth as they (re)produce meaning, and it is these practices that we can interrogate. The simple formulation of an essay title is a discursive practice: there is a difference between writing on ‘women, armed conflict and language’ and ‘gender, violence and discourse’. The former fits comfortably within an empiricist framework – we can see, and therefore know, language, women and armed conflict – whereas the latter does not. The latter demands conceptual, rather than experiential, engagement. I include both in the title of this essay not only to draw attention to the ways in which two sets of three words purport to mean the same thing while having very different connotations, but also to suggest that we should (that tricky exhortation) include both in our studies and practices of politics and policy.

Policy documents are, among other things, discursive practices, and can be read in this way, with a view to asking what is in some ways the most directly political – and policy-relevant – question: how is it that the reality we take for granted, which includes disparities of power and multiple forms of (sometimes violent) oppression, comes to be accepted as such? This special issue of the Review draws attention to many ways in which gendered logics produce (in)equality and order social life; through the analysis of policy governing women and war, the present essay seeks to contribute a discourse-theoretical perspective in keeping with this theme. Crucially, the distinctively poststructural form of policy analysis I outline here highlights the ambiguities and tensions inherent in any policy document and offers usable strategies for negotiating these, mediating the implementation of policy in a productive and potentially transformative way. The essay is divided into three substantive sections. In the first section, I map out a poststructural approach to discourse that, I argue, facilitates particular kinds of analysis of policy documents and other relevant political materials. The second section then provides an illustrative account of the theory presented, through the analysis of Chapter 5.10 of the United Nations Integrated Disarmament, Demobilization and Reintegration Standards,⁸ which is entitled ‘Women, gender and DDR’. In the final section I offer some concluding remarks and suggest some potentially fruitful avenues for future research.

‘How does a policy mean?’⁹

Elsewhere, I have argued that ‘understanding the ways in which the conceptual organisation of policy documents pre-/proscribe effective policy practice is a significant, and under-theorised, challenge’.¹⁰ I came to this conclusion specifically through the detailed analysis of United Nations Security Council Resolution 1325, which is a crucially important policy document pertaining to gender issues in the management of conflict and post-conflict reconstruction. However, most of the literature on UNSCR 1325 investigates the challenges of its implementation rather than the politics of its formulation and I sought to explore how the resolution came to be written in the way that it did, suggesting that the challenges of its implementation were in a very real sense produced by its formulation. When I began working on UNSCR 1325, my hunch was that the ideas and ideals about gender, violence and security that were represented in the resolution could be tracked back to ideas and ideals held in the institutions involved in the crafting of the document – what I term the ‘discursive terrain’ of the institutions, constituted through time- and location-specific legal systems, cultural and socio-political traditions, geopolitical positioning and histories and so on. If this was shown to be the case (and ultimately I believe I demonstrated that it was), then the implications for policy-making would be profound: the frequently unreflective and unconscious ideas that people have are being written into policy documents and are ordering and organizing those documents in very specific ways.

This theoretical agenda starts from the premise that no ‘thing’ has a material reality prior to language. There is no universal and unproblematic definitional lexicon to which we as scholars or practitioners can refer. All concepts come to have meaning through the context of their articulation. This may seem counter-intuitive. Surely a woman is a woman is a woman, regardless of her ‘context’? This is not in fact the case, for as we can learn through engagement with poststructural gender theory,¹¹ we can never ‘fix’ the identity of ‘woman’ independent of context. It may be strategically useful to speak of women, or directly necessary to speak with women. In some cases it might even be politically justifiable to speak for women, but we can never assume that we know who we are including and excluding in the category of ‘women’. Further, we cannot assume that those to whom we speak have the same understanding of women as we do, that their boundaries of inclusion and exclusion map on to our own. Finally, even if we were to agree with all concerned that we know what the category of ‘women’ is – that it includes, for example, post-operative male-to-female transsexuals and self-identified butches and bois¹² but excludes, for example, drag queens, female-to-male transsexuals and self-identified sissies¹³ – we could not, as the examples given demonstrate, say with any certainty that we know what ‘woman’ means.

These considerations are not merely ‘academic’, in the pejorative sense of the term, as they have significant implications for policy formulation and implementation. In the remainder of this section, I discuss three ways in which the concepts we use to think/write policy matters and relate these ideas to poststructural theories of discourse. First, there is the issue of common sense. Poststructural theorists frequently endeavour to ‘show that things are not as self-evident as one believed, to see what is accepted as self-evident will no longer be accepted as such’.¹⁴ Representations of political activity in media coverage (who is seen as authoritative, who is included at the negotiating table, and whose presence – or absence – is accepted as ‘normal’?), representations in policy documents and representations of global politics more broadly, in popular culture and arts, are all constitutive of how we understand global political processes and our place within them. Representations ‘are never merely descriptive, but always normative, and, as such, exclusionary’,¹⁵ as they prescribe – and proscribe – our acceptance of the way things are, thus delimiting the boundaries of common sense, which Stuart Hall called ‘the moment of extreme ideological closure’.¹⁶

Second, we must take seriously the question of inclusion and exclusion alluded to above. In her work on gender theory and development, Susie Jolly reminds us that ‘norms are all-pervasive, and not only determine the sexual aspect of our lives, but also shape our access to economic resources, and our ability to participate in social and political activities’.¹⁷ The marginalization of queer sexualities in security, economic/development and state-building policy, whether by omission or by design, both affects and demonstrates whose participation is considered legitimate, whose interests are represented and, ultimately, whose modes of being in the world are deemed to be of value. Relationships that have conventionally been considered ‘private’ are increasingly being publicly addressed. Even if relationships between individuals are assumed to be heterosexual and are most frequently heterosexual in ‘fact’ (i.e. in the specific case in which the relevant policy is enacted), the beliefs about marriage, monogamy and power that are intrinsic to the model of heterosexuality propagated by influential institutions may still have negative consequences.¹⁸ If intimate relationships are not conceptually bounded within a heteronormative model, practitioners need to ensure that due consideration is given to how sexual behaviours are thought of in the specific social and political context. Ultimately, the approach I espouse seeks to challenge the mostly silent norms that pervade policy-making. This challenge is posed in an effort to ensure that the types of social and political spaces that are produced through development practices and built through state-building processes are inclusive rather than exclusive, and that no mode of being in the world is marginalized or devalued because of particularly powerful notions of ‘common sense’.

Third, and finally, this approach draws attention to the process of implementing policy. It is not the overall aim of this approach to juxtapose different readings of various policies with a view to dismissing one or another of the readings as ‘untrue’, nor to suggest that one set or another of the meanings read in the documents is somehow ‘better’. All words carry meaning and have value; the process of writing value into policy documents is therefore inescapable, but it has profound implications for the interpretation and therefore implementation of that document. Meaning cannot be fixed; the sense we make of a policy document or strategic plan is conditioned by our own discursive context and the productive context of the document or plan in question. The challenges in implementing UNSCR 1325 – or, more recently, United Nations Security Council Resolution 1820 that seeks to eliminate all forms of sexual violence as weapons of war – will vary from place to place, and over time. This variation does not inhere in the inaccurate representation of specific cultural and historical contexts; rather, it is a function of language itself, according to poststructural philosophy. Therefore, looking for the origin or root of meaning, the reality to which a representation purports to relate, is an irrelevance. ‘Truth is a thing of this world. […] Each society has its regime of truth […], the type of discourses which it accepts and makes function as true’.¹⁹ This has profound implications for political research, in that a search for the ‘truth’ of the matter/‘reality’ becomes in this mode of investigation a search for the ‘systems of power which produce and sustain it [truth], and the effects of power which it induces and which extend it’.²⁰ It is hoped that this article will encourage critical interpretations of and reflections on the policy documents that order the lives of individuals everywhere, employing as they do concepts that, like all concepts, are inherently value-laden. As Dvora Yanow points out, ‘[i]nterpretations […] are more powerful than “facts”. That makes the policy process, in all its phases, a struggle for the determination of meanings’.²¹

Representations of gender and violence in the International Disarmament, Demobilization and Reintegration Standards

The United Nations International Disarmament, Demobilization and Reintegration Standards (IDDRS) are a pivotal policy platform on which to found efforts at lasting, equitable post-conflict reconstruction. The Standards were published in 2006 and, according to the accompanying website, are:

a comprehensive set of policies, guidelines and procedures covering 24 areas of DDR. The IDDRS consolidate policy guidance on DDR, providing a United Nations integrated approach on the planning, management and implementation of DDR processes. They are also the most complete repository and best practices drawn from the experience of all United Nations departments, agencies, funds and programmes involved in DDR.²²

This document is a testament to the development of an institutional agenda within the United Nations that takes gender seriously, as one entire chapter is devoted to addressing gender issues pertinent to post-conflict reconstruction. However, my analysis of the Standards proceeds from the assumption that even materials that are not overtly or explicitly ‘about’ gender have something interesting and important to say about gender. That is, the entire document, rather than just the chapter devoted to gender issues in disarmament, demobilization and reintegration, is produced through and is productive of gendered identities that are assumed to be both social and sexual.

This is perhaps a slightly contentious claim, but if we consider that post-conflict reconstruction frequently involves the organization of individuals into productive domestic units (for example, the household or community), we can begin to see how sexuality may figure in DDR programmes. Moreover, a post-structural approach does not see gender as the social construction of identity in opposition to sex as a biologically fixed variable. Rather, following Judith Butler,²³ such an approach sees the concept of biological sex as a regulative fiction: our understanding of gender claims, and therefore constitutes, a biological sexed identity but both gender and sex are performative, in that they come to be seen as ‘real’ through the iterated performance of behaviours that are widely understood to characterize gender/sex identities. This approach sees the human body materialize as gendered as a product of our theories of gender, rather than deriving our theories of gender from the materiality of the body. That is, we have a dimorphic theory of gender and label/amend bodies to fit that theory, rather than vice versa, so the bodies are literally constituted by this discourse. Furthermore, discourses of gender require specific performances of sexuality to retain intelligibility, the dominant mode of which Butler refers to as compulsory heterosexuality.²⁴ Therefore, in seeking to determine what kinds of subjects are being produced through the IDDRS, the performances of gender and sexuality that the Standards recognize as intelligible may be interrogated. These have serious implications for the implementation of the policy in post-conflict situations. However, the analysis for current purposes is restricted to Chapter 5.10 of the IDDRS and, specifically, to the representations of gender and violence in that chapter.

Gender

The IDDRS were drafted by an Inter-Agency Working Group on Disarmament, Demobilization and Reintegration, which was formed in March 2005. The Working Group has fifteen member organizations²⁵ and represents an incredibly broad spectrum of issue-priorities and interests within the United Nations system. This diversity would inevitably have led to controversy and contestation in the discursive terrain of the Working Group, with institutional variations in the conceptualization of core ideas needing uniform clarification. It is by no means clear, for example, that what ‘gender’ means to the International Labour Organization is the same as what it means to the Development Fund for Women. ‘These “commonsense” notions are constituted by particular configurations of discourse that, as a whole, has specific horizons of possibility, including certain modes of operation and excluding others’.²⁶

There are therefore inescapable conceptual ambiguities and tensions in the IDDRS, evidenced even by the title of Chapter 5.10: ‘Women, gender and DDR’. I would speculate that the inclusion of both ‘women’ and ‘gender’ was a result of the negotiating processes within the Working Group, perhaps the same processes that led to the inclusion of ‘gender’ but not ‘women’ in Annex A, ‘Terms, definitions and abbreviations’, which accompanies the chapter. So why include women in the title, when gender is defined in the chapter as the ‘social attributes and opportunities associated with being male and female and the relationships between women, men, girls and boys’?²⁷ According to this definition, women are included by virtue of the focus on gender. I suggest that it is likely that certain members of the Working Group fought to retain the framing of Chapter 5.10 as referring specifically to women, such as UNIFEM, for instance.²⁸ This would be in keeping with academic and policy debates over the utility and deployment of gender as a depoliticized alternative to women. While it is refreshing to see a policy document that explicitly does not treat gender as ‘loosely synonymous with “sex” and lazily synonymous with “women”’,²⁹ the separation of the two terms and inclusion of both (as well as ‘gender-aware interventions’ alongside ‘female-specific actions’, which are discussed below) renders the very title of the chapter ambiguous and confusing, which does not bode well for the implementation of the policies contained therein.

According to the IDDRS, it is ‘impossible to tackle’ various problems women face ‘without paying attention to how men's experiences and expectations also shape gender relations’.³⁰ The inclusion of the modifier ‘also’, however, suggests that the problems women face (manifesting as ‘political, social and economic marginalization’ and ‘high levels of violence against women’³¹) shape gender relations: to put it another way, the marginalization of, and violence against, women shape – influence, affect, form – gender relations. Thus gender relations are implicitly defined from the outset as reliant on the subordination and abuse of women. However, the IDDRS articulates a non-essentialist understanding of gender (where ‘roles and relations are by definition constructed…’³²) that separates gender from sex (which is defined as ‘[t]he biological differences between men and women, which are universal and determined at birth’³³). While defining sex as universal biological dimorphism is problematic (especially given that studies such as those undertaken by Melanie Blackless et al. suggest that 1% of all live births do not fit the ‘standard’ definitions of ‘M’ and ‘F’),³⁴ drawing attention to the social, cultural and political dimensions of gender constructions is potentially a progressive conceptual move. However, despite an overt commitment to the recognition of gender as constructed, there are many instances in the chapter where it is either implicitly or explicitly assumed that one can derive political will from physical form, which points to an underlying essentialism in the document. For example, the assumption that ‘female representatives’ will ensure the inclusion of ‘women's interests’ during the negotiation phase of DDR³⁵ would seem to imply that biological sex determines political interests, which is in tension with much feminist scholarship on representation.³⁶

Further, the IDDRS follow conventional logic of political participation when they suggest that facilitators and envoys should be ‘made aware of the internationally agreed minimum standard of 30 per cent female participation in any democratic decision-making forum’.³⁷ This is in spite of the increasingly well-evidenced feminist claim that ‘there is neither a single nor a universal relationship between the percentage of women elected to political office and the passage of legislation beneficial to women as a group’.³⁸ Despite explicit efforts to challenge stereotypes that associate women with peace,³⁹ it is assumed that the presence of women in formal decision-making will stabilize post-conflict societies. Interestingly, leadership remains a masculine privilege, according to Chapter 5.10, while women feature as ‘female representatives’ and in ‘women's groups’, as ‘the main caregivers’ and heading ‘up to 75 per cent of all households’ in post-conflict societies.⁴⁰ The IDDRS seem to expect rather a lot of women, in return for rather a limited amount of formal political power. For example, in order to ensure that gender issues are understood during the DDR negotiation phase, the chapter suggests that ‘facilitators of meetings and gender advisers should organize gender workshops for women participants before the start of formal negotiation’.⁴¹ Why only for women? And who has access to these workshops? The same questions arise as have been asked in relation to UNSCR 1325: ‘which women are included and are we expecting more from women (super heroines) than we expect of men?’⁴²

The question of what we – or the IDDRS – expect of men is an interesting one. As mentioned above, there is an assumption of masculine privilege with regard to formal political participation. However, masculinity is also represented in the document as pathological, manifesting in behaviour that is maladapted and unhealthy. Men need to ‘learn to resolve interpersonal conflicts without using violence’,⁴³ they ‘use their start-up cash irresponsibly, rather than to the benefit of family and community’,⁴⁴ they benefit from the ‘subordination’ and ‘domination’ of women.⁴⁵ By far the most egregious discursive violence perpetrated against the masculine subject, however, is the rendering of the subject invisible in the title of the chapter and its recommended actions. There are ‘gender-aware interventions’ (many of which are about women) and ‘female-specific interventions’, this despite the acknowledgement of ‘gender equality as a core principle of UN-supported DDR programmes’.⁴⁶

I am not seeking, as others have done,⁴⁷ to redress a perceived imbalance by drawing attention to the absence of positive representations of masculinity in the IDDRS. Rather, I suggest that we need to take seriously the implications of the absence. Given the overwhelmingly negative representation of masculine subjects in the IDDRS, it would be almost understandable if male stakeholders, facilitators and DDR experts professed little or no sympathy for the concerns raised in Chapter 5.10. Furthermore, the failure to recognize that ex-combatant men, just like ex-combatant women, may have ‘specific health and psychosocial needs’,⁴⁸ that men – and particularly boys – may have been ‘sexually abused during the war’,⁴⁹ that men and boys may be dependants of ex-combatants,⁵⁰ these failures ultimately perpetuate a conceptualization of gender that does not adequately separate gender relations from biological sex. Men and women are still constrained, almost defined, by their bodies: (male) fitness/aggression/power in opposition to (female) weakness/passivity/lack. Instead of valorizing the male subject, however, the chapter perpetuates a crude reversal of masculine privilege such that the needs (and bodies) of women are prioritized. This need not be the case. A truly gender-sensitive policy agenda would not contribute to the conceptualization of gender as a zero-sum game, where either men or women benefit only at the expense of the other. It would pay attention instead to the differences among men and among women (not reducible to bodily signifiers) rather than exaggerating those between them.

Much poststructural feminist work draws on queer theory to challenge the ideas that we have about bodies, about the status of biology and about the things to do with sex and sexuality that are generally considered to be ‘natural’.

Our experiences have implications for the appearance, condition, and performance of our bodies. For example, women may have hysterectomies, bear children or not, remove or grow body or facial hair. Both men and women may or may not exercise until they are muscular, or suffer from war or sport injuries. […] Similarities between bodies of one ‘biological’ sex are exaggerated, and [similarities] between bodies of different sexes are played down. […] Thus, although the categories of sex appear natural and absolute, they are ‘cleaned up’ by human intervention.⁵¹

A poststructural approach submits, therefore, that representations of gender, sex and sexuality in policy documents do not simply describe bodily realities, but rather are constitutive: of domestic units, sexual relations, parental care and a host of other social arrangements as well as of the physicality of the gendered body, predicated on a norm of heterosexuality. Gender is therefore profoundly implicated in our figuring of human life, literally defining the parameters of the question: ‘Who counts as human?’ In policy terms, this may be rephrased as: ‘To whom does this policy apply?’ The inclusion and exclusion of groups of people on the basis of their adherence to a culturally determined set of ‘natural’ ideals (sex binaries mapping to physical dimorphism, sexual desire following an oppositional logic) does a violence to those whose lives are already likely to be marked by violence, by virtue of their perceived transgressions of gender norms.

Violence

The first mention of violence, in the opening paragraph of Chapter 5.10, is in association with the predicate ‘sexualized’, represented as a threat to women.⁵² In the paragraph that follows, the chapter highlights the problem of ‘violence against women’.⁵³ As has already been noted, this same summary section suggests that men must ‘learn to resolve interpersonal conflicts without using violence’.⁵⁴ Violence, then, is something that happens to women, something that is perpetrated by men. In precluding the notion that men can be victims of gendered violence, this construction is patently false. It also further circumscribes female agency, already limited by the essentialist conceptualization of gender that informs the IDDRS, as violence is a ‘potential [cause] of insecurity for […] women’⁵⁵ and therefore women are in need of protection. ‘Gender-based’ or ‘sexualized’ violence is prioritized in the chapter, which serves to suggest that gender relations are somehow violent by necessity, indeed, by nature. ‘Female military personnel should […] play an important role in receiving and transmitting information on gender-based violence and/or sexual exploitation’,⁵⁶ presumably because it is assumed that the victims themselves will be female. Men, on the other hand, ‘may express their frustration in increased violence’.⁵⁷ The most obvious implication of this representation is the limited discursive and material opportunities available to male victims of gender-based violence to report violence and seek redress. A secondary implication is that, through the association of gendered violence with femininity, those men that do experience such violence may well feel feminized, making them less likely to speak out even where such facilities exist.

There is an explicit temporal aspect to the representation of violence as conflict, as the chapter frequently differentiates between periods of ‘conflict’ and ‘post-conflict’. This is problematic because it assumes a level of order and regulation that may simply not be present.

[W]ar can surely never be said to start and end at a clearly defined moment. Rather, it seems part of a continuum of conflict, expressed now in armed force, now in economic sanctions or political pressure. A time of supposed peace may come later to be called ‘the pre-war period’. During the fighting of a war, unseen by the foot soldiers under fire, peace processes are often already at work. A time of post-war reconstruction, later, may be re-designated as an inter bellum – a mere pause between wars.⁵⁸

Although DDR processes are to some extent premised on the labelling of a period of time or a society as ‘post-conflict’, it is still necessary to be attentive to the unlikelihood of an immediate end to hostilities, the immediate alleviation of resource scarcity or the immediate dissolution of strongly held beliefs about who supports or sympathizes with one side or another.

The Standards also implicitly differentiate between conflict that is armed and conflict that is not, although the latter is never discussed. Representing violence as conflict severs the signifying links to pain, physicality and specificity of experience that ‘violence’ can sustain. Not only does the language of conflict act to distance the materiality (and therefore reality) of violence assumed by the IDDRS, ‘there is very little discussion in much of the writing on “conflict analysis” or “conflict resolution” on the impact of certain types of social relations on the specific forms of violence’.⁵⁹ Furthermore, ‘the question hardly arises as to how or why this “conflict” situation is different from what is normal’.⁶⁰ As Kurtz and Turpin eloquently argue, ‘[t]he tendency to see violence as the consequence of aberrant behaviour committed by deviant individuals at the margins of society obscures the central roles violence plays at the very foundations of the social order’.⁶¹ That is, inclusion and exclusion, negotiation and coercion, even parenting and caring can be violent processes, according to a poststructural analysis. Such an approach asks which acts of violence are considered worthy of recognition and when these acts occur. Expanding the concept of violence which underpins feminist analysis, as outlined above, allows us to take seriously what Arthur Kleinman refers to as ‘the violences of everyday life’.⁶² Beyond a narrow focus on war and state-based violence lies a plethora of everyday violences we might wish to include in our analysis of DDR and security.

The language of politics

As Roxanne Doty explains, discourse is more than just language. ‘A discourse delineates the terms of intelligibility whereby a particular reality can be “known” and acted upon.’⁶³ Discourses are systems of meaning-production rather than simply statements or language, systems that ‘fix’ meaning, however temporarily, and enable us to make sense of the world. In suggesting that discourses ‘fix’ meaning I do not wish to imply that there is any trans-historical continuity or universality to meaning. This is a politics of iterated practices, these processes are always ongoing and never complete. Rather, the ‘terms of intelligibility’ are multiple, open and fluid and must be continually re-articulated and re-ordered if what was once ‘common sense’ is to remain so. It is the partial and limited nature of fixity that allows critical space for engagement. Above, I have outlined the ways in which the realities of gender and violence are ‘known’ in Chapter 5.10 of the IDDRS, for how these realities are known has direct and material implications for how they are acted upon. In this section, I offer some concluding remarks and suggest some potentially fruitful avenues for future research.

What are the implications of the issues discussed here? The theoretical musings may be interesting (to some, at least) but how can the study of women and war be enriched by the arguments presented, by the type of analysis espoused? The first point of critical engagement must be with the language used to write policy. Every policy document at the international level, no matter how short or seemingly insignificant, already goes through a process of drafting and revising in consultation with various advocates, stakeholders and interested parties. Academic interest is rarely focused on these negotiations (a notable exception being the analysis of the ‘holy brackets’ deployed during the UN Fourth World Conference on Women in Beijing, 1995),⁶⁴ but I would suggest that more attention should be paid to how these crucial documents come to be written in the ways that they are. This is not because it is somehow possible to write the perfect policy, one that will neither privilege nor marginalize any group or individual, nor because those who are involved in the policy-making process require the expertise of interested academics in order to say what they really mean. Instead, learning more about which groups are organized – both conceptually and strategically – around which ideas, and how tensions are resolved when two or more groups experience conflict, can only give us a better idea about how commonsense notions are being circulated and (re)produced through the writing of policy itself. I have referred to this kind of analysis as an exercise in mapping, where the terrain in question is discursive.⁶⁵ Again, I emphasize that the aim of such an analysis is not to criticize or exclude from the negotiating table any organization on the basis of its discursive terrain. Rather, given that ideas matter, it simply behoves us to know more about the ideas these key actors hold.

The second type of critical engagement encouraged by the approach to research advocated above is with the translation of policy from document to action plan. A policy document must always be translated, possibly literally (UNSCR 1325, for example, is currently available in one hundred different languages) but certainly metaphorically. The recommendations, suggestions and guidelines for best practice contained within the policy must be rendered coherent within the immediate context of its implementation. Obviously, the contexts vary, and we need to pay attention to the ways in which those variations have an impact on the implementation of policy in order to inform our interjections in decision-making in the future. Even the most groundbreaking policy is not an end point in itself, and those working in this area of policy analysis seek to offer critical insights into how the language used to write policy affects (and effects) its implementation on the ground with a view to these insights feeding forward to the next iteration of the policy process.⁶⁶ The end-users affected by a given policy will undoubtedly have much to say about the efficacy or otherwise of that policy's implementation; this second type of analysis incorporates the consideration of grassroots or indigenous organizing in relation to policy issues. We know that global governance is not necessarily the preserve of the formal, centralized institution, just as we know that it is likely that informal, decentralized networks will write their engagement differently. In order to complement the discursive mapping suggested above, then, spatial mapping might be undertaken in a similar fashion, to trace the functions and effects of translating policy into action, and the ways in which stakeholders engage with the end results.

Finally, I offer a third strand of analysis in the form of temporal mapping. In conjunction with mapping discursive and spatial horizons of a given policy document, it would also be interesting to build up a base of knowledge about that policy and particularly to trace how secondary writings change over time. There is already a shift, for example, in the representation of UNSCR 1325, with its initially positive reception being tempered with critique, and UNSCR 1820 will no doubt attract the same kind of attention. While it would clearly be unmanageable for any one researcher or research team to engage in all three kinds of analysis outlined here across more than one issue area, archiving knowledge in a temporal map would make it far easier to identify and therefore potentially understand the shifts in meaning and emphasis – both in policies themselves, those which are updated, and in secondary writing on an issue – that inevitably occur.

These three forms of mapping, then, are all premised on the assumption that we need a politics of language in order to better comprehend policy formulation and implementation, but they also recognize that the language of politics is impossible to pin down. We can never simply say what we mean, even when we mean what we say. ‘No discursive formation is ever a sutured totality’,⁶⁷ and each formation is contingent, meaning that critical engagement with dominant discourses (for instance, of gender, violence, security) is always possible, and with that possibility comes the potential for change. We – policy makers, scholars, students, practitioners – cannot avoid making meaning through our well-meaning interventions. This does not need to lead to inertia, however. We should not dismiss policy documents nor throw away the baby of strategic gains made with the bath water of value-laden concepts. Rather, we can offer multiple readings of multiple policies with full awareness of the lacunae in our own knowledge and politics. It is hoped that this article will encourage critical readings of and reflections on the policy documents that order the lives of individuals everywhere, employing as they do concepts that, like all concepts, are inherently value-laden.