INTERNATIONAL CRIMINAL COURTS AND TRIBUNALS The powers of silence: Making sense of the non-definition of gender in international criminal law

Silence has often been studied in international law as a mechanism tied to passivity and oppression. In this study, I propose an exploration of other ontologies of silence by unravelling its possibilities as an active mechanism, namely: (i) a tool for resistance; and (ii) a linguistic device for managing disagreement. For this, I use as an exploratory ground the construction of a non-definition of gender for the crime of persecution in international criminal law (ICL). Analysing the Rome Statute negotiations, I examine how gender-conservative actors successfully opposed the proposal for a non-definition of gender, arguing that such a solution would harm the clarity required by the principle of legality in ICL. By establishing that legal rules must be clear, specific, and cohesive, I argue that the legality principle imposes a burden of speech upon non-state voices in ICL, one that encircles them within a subalternity scheme where speech is demanded but can only be performed or mediated by states. Exploring the negotiations of the Convention on Crimes Against Humanity draft, I examine how the non-definition of gender allowed feminist and queer activists to resist such a burden of speech for the conceptualization of gender. Simultaneously, silence also provided an opportunity for International Law Commission members to propose a draft that avoids cacophony around a contentious term. By reflecting on the active roles of silence, this study contributes to new modes of analysing resistance to dominant modes of legal discourse, as well as exploring dynamics of order(ing) in international law-making.


Introduction
As an argumentative practice and discipline, texts and discourse have taken centre-stage in international law (IL). While considerably side-lined, the usesand the politicsof silence in IL have also garnered some attention. Whilst more traditional approaches have often tied their explorations of silence in terms of tacit state consent, critical takes on the topic have frequently examined silence as a symbolic representation of sites of exclusion, oppression, *For comments, enlightening conversations, and supportive insights on this article, I am grateful to Nico Krisch, Elisabeth Prügl, and Tanja Aalberts. For their engaged discussions and suggestions on earlier versions of this article, I am also thankful to Victor Tozetto da Veiga, Karem Cárdenas Ynfanzón, the participants of the 13 th Melbourne/UNSW Doctoral Forum on Legal Theory, and the feminist writing group of students from the International Relations/Political Science department of the IHEID. For kind mentoring and advice in the submission of this article, I am thankful to Tamsin P. Paige and Justina Uriburu. The research for this article was supported by the Swiss Federal Commission for Scholarships for Foreign Students (scholarship grant 2019.0113). and (forced) submission. 1 More specifically, silence is explored as both evidence of and a mechanism for (re-)enforcing the subalternity of women and third-world peoples: that is, their embeddedness in a material, political, and discursive context in which they are denied the power to speak or to have their speech heard as meaningful. 2 Further, sociolinguistic approaches to silence in international law have considered it as a speech-dependent linguistic mechanism, whereby it gains meaning when it frustrates the expectation of a spoken response. In this sense, whereas particularly different in their scopes and assumptions, different strands of international legal scholarship have understood and circumscribed silence to a specific ontology of oppression and passivity.
However, across a myriad of studies on political theory, cultural sociology, and sociolinguistics, silence has featured many different meanings and characteristics. 3 Instead of solely being considered as a tool for deliberate or imposed acquiescence, a dependent linguistic mechanism, or a theft of voice, silence has also been studied in its active and resisting possibilities. Within this framework, silence has been rethought of as a self-contained mechanism with specific sociolinguistic roles, as well as a useful tool for resistance and subversion. My aim with this study is to build from these perspectives and provide a different approach to silence in international law. By looking at it as a potentially active mechanism, I intend to explore how relevant actors utilize, as well as grapple with, silence in their practices of negotiating and constructing legal concepts and norms in the international scenario.
For such an analysis, I will use as an exploratory ground the construction of a non-definition of gender as a viable legal concept for the crime of persecution in ICL. This exploration begins at the Rome Statute negotiations, when the meaning of gender was a highly contentious issue. At the time, feminist NGOs and allied countries defended that term related to the socially constructed roles expected from men and women. Conservative states and NGOs, on the other hand, defended that the concept meant the 'biological' differences contingent on the male and female sex. To curb this apparent unsurmountable disagreement, a non-definition for gender was proposed during the negotiations. This solution was based on the previous agreements achieved for the Beijing Declaration and Platform of Action of 1995, which established that the term should be left undefined in the text of the document and 'interpreted and understood as it was in ordinary, generally accepted usage'. 5 However, the proposal for the non-definition was rejected on the basis that it would potentially harm the clarity required for criminal rules, as dictated by the principle of legality in ICL. 6 The result of this impasse was the oddly constructed definition of Article 7(3), which combines both views into one and defines the term as 'refer[ring] to the two sexes, male and female, within the context of society '. 7 More than 20 years later, the feminist proposal for a non-definition of gender has been fairly successful as a constructive and practical solution for the Convention on Crimes Against Humanity draft (CAH draft). This was the result of a broad campaign coalition between NGOs, states, and several UN experts, in which the campaigners urged the International Law Commission (ILC) to remove the re-employment of the Rome Statute gender definition from the draft convention. 8 In its fourth report regarding the document, the ILC Special Rapporteur on the topic noted and accepted the arguments propelled by the campaigners, ultimately deciding to remove the Rome Statute definition and leave the term undefined altogether. 9 Here I am interested in the different layers, roles, and meanings of silence in relation to this non-definition. First, it is possible to inquire about the contextual signification of the nondefinition in the Rome negotiations, when it was considered as a threat to legality. This leads us to a second question, that is: how did actors re-signify the non-definition itself in order to accommodate silence as a legal solution now in the CAH draft? To answer these questions, I conduct a case study of both 'gender negotiations', relying mostly on a triangulation between primary and secondary sources, namely official statements, records of meetings, as well as theoretical and empirical academic literature.
This article is organized as follows. Section 2 articulates an analysis of the current studies on silence in international legal scholarship, noting their shortcomings when they look at silence merely as an instrument for social domination or as a linguistic mechanism dependent on speech. Section 3 draws from sociolinguistic studies, political theory, and cultural sociology to substantiate the argument that silence can present more active functions, namely as a discursive tool for both resistance and order(ing). Within this framework, I explain how silence can serve as a resistance mechanism against a subalternity scheme that imposes upon the subaltern a burden of speech. Further, I also explore how silence can be an active linguistic mechanism by co-ordinating and pacifying a multitude of dissentient speeches. There I draw from feminist epistemologies to explain how this ordering/pacification function carries a gendered dimension in its imposition of order and cohesiveness. Section 4 turns to the Rome Statute negotiations to examine the legality principle in ICL as a gendered, (neo)colonial demand for legal clarity, which encodes non-state voices in ICL within a subalternity scheme where legal norms are demanded to be clear while such precision can only be defined or mediated by public sovereigns (states). I then go back to feminist theory to point out how the legality principle also embodies a gendered epistemology of order(ing), one that imposes unity and cohesiveness even when legal concepts are inherently 5 Fourth World Conference on Women, 'Report of the Fourth World Conference on Women, Annex IV ("Reservations and Interpretative Statements on the Beijing Declaration and Platform for Action")', UN Doc. A/CONF.177/20/Rev.1 (1996). contested and heterogeneously interpreted. Section 5 examines the efforts of the campaign led by NGOs, academics, and UN independent experts in making the non-definition of gender a viable legal path. It explores the non-definition as proposed by the campaign coalition as resisting the subalternity scheme of legality in ICL. Section 6 analyses the debate regarding the non-definition of gender by states and ILC members, looking at its role as a mechanism to prevent cacophony and pacify eventual differences. Section 7 concludes and suggests topics for future research.

On the uses of silence in international law
As already alluded to in the introduction, silence is not a novel object of analysis in IL. Nevertheless, current conceptualizations have been somewhat generally circumscribed to an idea of silence as an enactment of passivity both in political and linguistic terms, with legal speech representing its opposing, active pole. By engaging with Acheson's invitation to look at silence in dialectical terms ('both/and', instead of 'either/or'), 10 I do not reject that silence can represent, signify, or enforce passivity and subjugation. However, there is still the need for a better look at the active roles of silence as an effective counter-act to speech, in order to give a fuller account of its uses in international law.
In more doctrinal terms, silence has been considered as an important tool for and evidence of acquiescence and tacit consent in international lawboth for subsequent practices concerning the interpretation of treaties or for building new practices in customary international law (CIL). 11 In the voluntarist scheme of classical ILwhere states' consent is the cornerstone of legal obligations, and legal duties must be explicit 12silence is read as a non-objection of states towards the practices of other states, 13 especially when a response is called for in the circumstances. It is then taken as tacit acceptance or toleration, something that allows the non-objected practice to solidify into custom if re-enacted with time and consistency. Within this reading, silence is, therefore, crucial for building as well as for evidencing the elusive opinio juris, i.e., states' subjective understanding that a certain practice is legal and thus should be followed as such. 14 However, not every silence is considered to have this value for transforming a practice into international custom, or yet to (re)confirm its legal relevance for states. Discussions on the relationship between silence and the construction of international custom have often been peppered with concerns about differentiating between legally relevant silences and merely political ones. Within this context, for example, the ILC has made a case for caution when interpreting silences, arguing that inaction should generally be interpreted 'in relative terms, account taken of the specific (sequence of) facts and the relationship between the States involved'. 15 This is to highlight the importance of differentiating silence as meaningfully enacteda deliberate choiceor as a result of an inability to speak or lack of direct interest. 16 Within this framework, silence is regarded as a relevant mechanism to accord and uphold the legality of an emerging practice in international lawan enabler mechanism, so to speak.
While such an exploration could be understood as taking silence in its active functions for representing volition, the way silence is encoded within a dichotomous relationship with the persistent objector rule constructs a different image. The idea of silence as an enabler mechanisma 10 Acheson, supra note 3.  (2015), supra note 1, at 11-12, para. 22. 16 Ibid., at 13, para. 24. deliberate acceptance that makes way for a new practice to solidify into international customgrounds the power of silence in CIL as being a legally meaningful passivity. Within this context, silent acceptance is taken as the 'normal' state of affairs, one that does not elicit any action to effect meaning. Only speech is able to break with this 'normalcy': the vocal protest of the persistent objector actively pierces through this stasis, thus standing in powerful contrast with the placidity of tacit acquiescence.
Aside from its relevance to CIL, silence has also been analysed through more critical lenses. In an attempt to unearth the structural biases of IL, critical legal scholars have frequently looked at what is said as well as what is unsaid in international legal practices and in the discipline, exploring the power dynamics and asymmetries therein. In feminist approaches to international law, for instance, men's voices are seen as overpowering and dominant, whereby womenand women's concernsare constantly silent and silenced. In this sense, feminist scholars in IL have often pointed out that gender issues are ghettoized or even completely left out of international legal instruments and institutions. 17 Moreover, the consistent underrepresentation of women in international legal bodies has been considered as a marker of oppressive deafness of the international legal system towards women's voicessomething that is read as contributing to their exclusion or tokenization in the field. 18 Third World Approaches to International Law (TWAIL) have likewise underscored the politics of silence in international law. In such analyses, international legal silences are read against broader continuities of colonial inequality and the imposition of consensus, 19 thus being considered as perpetuating frameworks of colonial domination and subalternity. In this regard, for instance, Chimni has explored the colonial legacies and reinstatements of CIL, explaining how third-world states have frequently been silenced in the production of international custom. 20 He notes that in the accession to independence, third-world nations were not able to object to previously developed customary rulesan imposed silence to pay the entry cost into the international community of sovereign and independent states. 21 Moreover, the neglect of third-world state practice in the consideration and studies of CIL is also symptomatic of deeper inequalities. In this sense, the absence of systematic organization and publication of the practices of the third world are often the result of a 'lack of human and financial resources to gather and disseminate legally relevant practice'. 22 It also stems from asymmetries in international legal knowledge production, where the role of scholars in clarifying CIL is frequently overrepresented by western academics. 23 For these reasons, silence in international law for both feminist and TWAIL approaches seems to be understood both as evidentiary and oppressive: silence exists as proof of an inherently (masculine or western) biased international law, as well as it is imposed upon women and third-world peoples. Such understandings connect with a myriad of other critical works on silence across the social sciences, which have linked silence to silencing and the theft of voice. 24 In such perspectives, silence is commonly seen as an imposition of dominant groups in order to out-shout marginalized peoples, thus being closely linked to disciplinary technologies, lack of agency and power, and structural  20 Chimni, supra note 1. 21 Ibid., at 24-5. 22 Ibid., at 21; Byers, supra note 1, at 153. 23 Chimni, supra note 1, at 25-6. 24 One reflection that neatly encapsulates this perspective is Anzaldúa's notion of 'linguistic terrorism' which imposes a tradition of silence upon the marginalized. See G. Anzaldúa, Borderlands: La Frontera: The New Mestiza (1987), at 58-9.
inequalities that impede the expression and assertion of the marginalized. 25 Unsurprisingly, many critical scholars have proposed political strategies of resisting silence, either by exposing it, taking control of language, re-appropriating words, and, ultimately, creating spaces and possibilities for languages that are detached from patriarchal and western schemes of subjugation. 26 In this perspective, resistance is placed in opposition to silence: it needs to be speech-based and vocal to effect a liberating praxis.
Another analysis of silence in international law is the one given by sociolinguistic approaches to the topic. Building on the works of sociolinguists such as Schröter and Kurzon, 27 Schweiger digs deeper into silence's communicative functions and explains that, for silence to effect a meaning, it needs to be linked to a question (prompt) that is expected to be answered but is frustrated when there is inaction. 28 For example, when a controversial action is performedsuch as a unilateral use of force by a state, for instancethe legal and political weight of this action triggers an expectation of response. 29 When this expectation is frustrated, silence gains a communicative function by becoming an unsaid counterweight to the expectation of explicit condemnation or support.
In this framework, communicative silence is therefore contextual: it gains this function because of a context that orients actors to understand a silent response as inappropriate. What Schweiger interestingly adds to her analysis is that, despite the contextual relationship between silence and international law practices, not all silences are equally noted or accounted for in this discussion. She argues that silences, as any other mode of expression, can be interpreted and construedsomething that does not occur in a political and power vacuum. More importantly, she highlights that the interpretation and construction of silences can steer the discussion of legal practices, especially when some silences are picked as legally relevant over others. 30 Although this sociolinguistic analysis brings the notion of silence closer to an idea of a message in itselfsomething that blurs the understanding between silence and speechit still does not regard it as a self-contained mechanism. 31 Silence is only properly understood if prompted by a spoken legal question and only gains meaning if it frustrates the expectation of a verbalized response. Albeit in more nuanced terms, silence here is still understood as dependent and passive: a frustration, a mechanism bound to inertia that can only be put into motion by the verbal prompt. 32 Not only that, but silence is also dependent on speech to be interpreted: without the frustration of a spoken response, it is taken as meaningless and non-communicative.
While these conceptualizations have their own merits and are applicable to certain objects of analysis, their fixity on silence as passive or dependent limits further explorations. On this, even the very own assumption of silence as powerless and speech as powerful should not be taken at face-value. As already suggested by works on linguistics and communicative theory, western cultures are often biased in favour of speech, 33 with silence often taken as a lack, an illness, a 25 On a discussion regarding these perspectives see Butler, supra note 3, at 137-9. This understanding is similar to the notions defended both by the ICJ and the ILC that for a silence to be legally meaningful, it needs to arise from circumstances when the state must and can act (si loqui debuisset ac potuisset). See supra note 11 and accompanying text. 30 Schweiger, supra note 1, at 407-10. 31 On silence as self-contained see generally R. P. Clair, Organizing Silence: A World of Possibilities (1998). 32 Acheson, supra note 3, at 536-7. malfunction, or a defect. 34 Unsurprisingly, it is frequently coded as a sign of femininitysomething that represents the expected role of women to remain submissive and, therefore, silent. 35 Speech, on the other hand, is regarded as the powerful pole: a way to assert opinion, culture, and identity.
Against this background, there is still the need to take the manifold nature of silence more seriously in international law. Aside from a few notable exceptions, the question of how silence can be a powerful and active mechanism both in terms of political resistance and linguistic functions remains underexplored in the field. 36 In the subsequent section, I draw on political theory, cultural sociology, and sociolinguistic studies that have focused on more active and productive roles of silence, namely: (i) resistance against a subalternity scheme grounded on a burden of speech for the subaltern; and (ii) a tool for ordering dissentient speeches and contestation.
3. Expanding the roles of silence: Resistance and order(ing) As explained in the previous section, studies on silence in international law have analysed it as a passive (or passivity-enforcing) mechanism and signifier. However, there is much more to its functions and possibilities. Although the discussion on the active roles of silence is broad and has been covered by a wide range of disciplines, 37 here I focus on two of these roles: resistance and order(ing).

Resistance: Silence as a counteract to subalternity and the burden of speech
Oppression and subalternity have long been connected to silence. As already explained in the previous section, feminist and TWAIL scholars have picked up on such conceptualizations, unearthing how the unsaids of the international legal discipline and practice point to structural biases and matrixes of power/dominance. However, subalternity can be much more complex than the imposition of sheer silence. Instead, it can also be effected through demands of speech that curtail an autonomous voice from those in the margins.
To understand this, there is first the need to expand on the social functions of silence. Whereas social groups have often been analysed as bound by (often spoken) norms and modes of communication, 38 silence can also constitute the social fabric of a group and determine the boundaries between insiders and outsiders. 39 In some Indigenous North American cultures, for instance, silence is a marker of group belonging and intercultural communication: being silent is seen For an edited volume that goes against this trend and introduces interesting ways to look at the political and active functions of silence (but mostly focusing on international relations and political science) see T. N. Cooke and S. Dingli (eds.), Political Silence: Meanings, Functions and Ambiguity (2018). In this volume see, especially, the contribution of Guillaume and Schweiger, supra note 1. 37 See a general review on the topic in Acheson, supra note 3. This is particularly salient on the Bourdieusian notion of 'background knowledge' that demarcates social fields and guides the behaviour of actors. As commented by Pouliot, Bourdieu's practice theory has two modes of intersubjective knowledge that orients practices: representational knowledge and background knowledge. The first refers to conscious, verbalized, and intentional knowledge acquired through formal schemes. It is explicit, based on reasoning and prone to justification. Background knowledge, on the other hand, is rather implicit, inarticulate, and unreflexive. It is hardly salient in the level of the discursive, and for Bourdieu it is even considered as the knowledge learned and deployed at the level of the body. In this regard, a silent knowledge is here depicted as a virtue, and those who know how to appropriately perform and understand it are then recognized as one of the group. 40 Silence can thus be actively practised as a community, 41 therefore helping actors to secure the community's social boundaries and to hold the group's unity and identity. 42 Within this context, knowing its ways around silence can be an important marker of belonging and appropriateness within a social group. In this sense, those within a lesser or unwelcome position can be subject to the imposition of explaining and affirming themselves to be let in and taken into consideration, whereas those in power may use silence as a contrasting performance to demonstrate authority and a stance of prestige. 43 As explained by Glenn, 'silence is expected (and therefore goes unnoticed) from people who wield power, be it religious, therapeutic, bureaucratic or legal: it is the silent listener who judges, and who thereby exerts power over the one who speaks'. 44 In such settings, speaking becomes a burden of the subordinatea signifier of their outside and/or subjugated position.
Against this background, performing silence can yield more complex meanings than simply denoting oppression and subservience. Four of these performances and their generative meanings merit particular attention.
First, silence can be performed in order to mimic authoritative silence. As explained above, those with a more powerful and prestigious stance within a social group may have the privilege to remain silent or be expected to do so. In these situations, individuals with less authority in a given situation may attempt to imitate this privileged position by performing silence, in an effort to simulateor even establisha more powerful position for themselves. 45 Second, it can represent a direct resistance and refusal to sustain a burden of speech. 46 On this, whenever an underlying subalternity scheme feeds off from having the subaltern's speech 'mediated through interpretation and replication mechanisms that foreclose her exercise of power through speech', silence can represent a powerful possibility of resistance. 47 Simply put, silence can break with subalternity when the subaltern is not exactly silenced but enclosed within a position where their speech is demanded but will be invariably interpellated, appropriated, mediated, or limited. 48 This way, silence can be used as a 'means of power, a species of sadism, a virtually Glenn, supra note 33, at 10. 45 Acheson, supra note 3, at 29. 46 Take the Women in Black international peace movement, for instance. For decades now, they have used silent demonstrations as an invitational rhetoric to encourage passers-by think about the issues they are protesting. See Glenn and Ratcliffe, supra note 3. inviolable position of strength : : : [in contrast to] the burden of talking' 49a barrier for speech impositions that enables an autonomous exercise of discursive (and political) power. 50 Third, performing silence can be an important tool to break with the limitations of pre-established modes of discourse whenever verbal participation is imposed. Remaining silent can mean a defiant refusal to speak within forced discursive regimes that have been established without the (autonomous) participation of the subject; likewise, it can represent a possible site of resistance when a disciplining authority orders an individual or group to give an account of their story, to justify their actions, or to prove their innocence. 51 Fourth and finally, the vacuity and ambiguity of silence can provide an interesting tool for breaking with unitary linguistic formations that do not provide an adequate tool for full selfexpression. Building on Foucault's refusal to provide absolute truths and frames of reference in his archaeology of knowledge, silence can be a way to concretize such refusal by 'deploy [ing] a dispersion that can never be reduced to a single system of differences, a scattering that is not related to absolute axes of reference'. 52 By not reducing oneself to precise signifiers, silence can thus create an important subversive space that challenges the limitations of a dominant discursive formation.

Order(ing): Managing and pacifying cacophony
As explored in the previous section, sociolinguistic analyses of silence in international law have mostly understood it as a mechanism dependent on speech. Its communicative functions are comprehended as hinging on a spoken or written prompt, and the meaning of silent responses are contingent on a worded context. The linguistic and communicative possibilities of silence in acting onto speech are therefore largely left out of the picture.
Perhaps one of the most interesting effects of silence on speech is its function of managing and pacifying dissentient voices. As already explored, performing silence can hold an important marker of authority and belonging in social groupsbut, more than that, it can also serve as an important element in securing cohesiveness within the group. Whereas social fields have been largely explored as containing shared 'rules of the game' that orient what actors do within that field, 53 heterogeneity is inevitable. Each member of that particular community may be a complex entity in itself, with distinct situationalities that condition how they understand their reality and shared knowledges. 54 Uniformity within a social field is then much more tenuous than stable: discontinuities, disagreement, and change are the norm rather than the exception. 55 Within this context, silence can be an interesting tool to co-ordinate and pacify disagreement. By not speaking of differences, it can maintain both real and imagined commonalities of a social group. 56 It can also be regarded as an element of wisdom and construction of co-operative meaning: it disarms opposition as its ambiguity allows for the co-creation of meaning without the limiting interpretive structures of verbalized discourse. 57 Silence here is not seen as a lack of dialogue or miscommunication but as a speech co-ordinator in the face of eventual discontinuities. 58 It limits the cacophony that can be generated through discussions, co-ordinating and pacifying a multitude of dissentient speeches by not selecting a precise response to the issue in contestation.
While this function of managing cacophony and differences can be understood in terms of sociability of actors, it also harbours a potentially gendered dimension. As explained by feminist theorist Elizabeth Grosz, notions of (supposed) homogeneity, unity, order, and wholeness were crucial in the western construction of the 'knowing subject' as a shorthand for a white, western man. 59 By imposing western and masculinized knowledges as the exclusive paths to access and comprehend the realities around us, a gendered form of power seeps into how we construct our social epistemologies: one that enforces those gendered ways of knowing as universal and unitary. 60 In this sense, aside from a social, historical, and material matrix of dominance, gender power also represents an epistemological ordering force in the sense it imposes a gendered homogenizing and orderly pull upon ways of understanding our inherently diverse selves and social worlds. 61 Within this context, silence can provide an interesting path to concretize this gendered ordering and homogenizing drive. As explained above, when the difference is too salient and apparently unsurmountable, silence can become a mechanism to pacify and accommodate all differences. This has mixed blessings. While the creation of an ambiguous space where all can speak may generate an aura of all-encompassing inclusivity, it may also muffle dissension. Within this context, the creation of a silent, ambiguous space adheres to an ordering logic where no view is allowed to stand in marginal defiance, thus forcing the incorporation of every dissentient voice into a dominant core. In such incorporation, compromises and re-adaptations can be imposed, something that may undermine the power of dissent previously held by counter-dominant voices. 62 However, the strength of this gendered ordering drive is not omnipotent. In fact, its own absoluteness and rigidness often go against itself, as complete order and cohesiveness with one dominant core are never reachable nor stable in reality. 63 Because of this, such rigidity often creates 56 Ferguson, supra note 3, at 50. 57 Acheson, supra note 3, at 23. See also Guillaume and Schweiger, supra note 1, at 161-9. 58 Acheson, ibid., at 23; Guillaume and Schweiger, ibid., at 165-6. 59 E. Grosz, 'Bodies and Knowledges: Feminism and the Crisis of Reason', in L. Alcoff and E. Potter (eds.), Feminist Epistemologies (1992), 187, at 205. Here she explains that conventional epistemologies and production of knowledge as inherently tied to ideals of masculinity: '[m]any features of contemporary knowledges-knowledges based on the presumption of a singular reality, pre-existent representational categories, and an unambiguous terminology able to be produced and utilized by a singular, rational, and unified knowing subject who is unhampered by "personal" concerns-can be linked to man's disembodiment, his detachment from his manliness in producing knowledge or truth' (ibid.) (emphasis in original). cleavages and, most importantly, points where change can take place. 64 For instance, to pacify disagreement and avoid cacophony, agents can acknowledge difference as an essential characteristic of a concept, thus leaving no other option but to have it as an open term. Silence here can represent a viable path to enable openness and consequent pacification; however, it also provides an important 'democratizing' potential for a term by enabling it to remain flexible and mobile for its different users. 65 With this in mind, while gendered ordering epistemologies can demonstrate how gender operates by trying to 'make coherent' all of that which is essentially messy, 66 it should also be regarded as something that produces the conditions of its own subversion.
With all these perspectives in consideration, I now turn to a case study of the negotiations on the gender definition for the crime of persecution in ICL. First, I analyse the Rome Statute negotiations and how the issue of legality posed a hefty challenge to the proposal of a non-definition of the term back in 1998. I then provide a feminist analysis of legality in international criminal law, examining it as an enabler of a subalternity scheme that grounds itself on a burden of speech for the subaltern. After this analysis, I then study the discussions of the ILC regarding the nondefinition of gender and how it has been employed as a tool for both resistance and order(ing) in international criminal law-making.

The (gender) politics of legality in Rome: Subalternity and homogeneity in criminal law-making
Although the Rome Statute was not the first time gender issues were being considered in international criminal law, the course and results of its negotiations brought hard-fought advancements to feminist interventions in the field. The Nuremberg Charter did not include gender as a protected category for the crime of persecution, nor did the Statutes for the International Criminal Tribunals for the former Yugoslavia and Rwanda. However, the jurisprudence arising from the cases before these international tribunals, coupled with the engagement of feminist activism in international conferences in the 1990ssuch as the World Conference on Human Rights of 1993 and the Fourth World Conference on Women of 1995 (Beijing Conference)opened the doors for gender issues to be integrated into the Rome Statute negotiations. 67 The preliminary text drafted in 1994 by the Working Group on the Establishment of an International Criminal Court (ICC) had no reference to gender. 68 The term was first included as a protected category under crimes against humanity in February 1997 by PrepCom's Working Group on definition of crimes. 69 During the Diplomatic Conference of Plenipotentiaries on the Establishment of an ICC, gender was fiercely debated. Gender-conservative states strongly opposed the inclusion of the term in the Statute, especially as part of the no-adverse distinction clause. 70 The no-adverse clause related to the paragraph regarding the application and interpretation 64 of the law by the Courtwhich would later become Article 21(3) of the Rome Statutewhere gender was one of the enumerated grounds for non-discrimination.
However, the no-adverse clause was not the only point of contestation regarding the incorporation of gender in the Statute. In the plenary meetings of 9 July 1998, the representative of Azerbaijan strongly reacted against the use of the term as a protected category under persecution, asking whether such provision would 'imply that a national court for homosexual acts might be regarded as persecution and thus fall within the jurisdiction of the court'. 71 Pakistan expressed the need to define the term in relation to males and females, a statement echoed by Qatar. 72 The pressure to define the term increased on the part of conservative NGOs and states, promoting a heavily polarized debate on how to elaborate a working definition for it. 73 The impasse was centred on how, for feminist NGOs and allied countries, the term related to the social construction of gender, i.e., the socially constructed roles expected from men and women, whereas the oppositional side insisted that gender meant the 'biological' differences contingent on the male and female sex. 74 During such debates, the Chair of the Working Group on Applicable Law suggested employing the strategy that had been successful for the Beijing Platform, i.e., not defining the term in the text of the negotiated document. This, however, was rejected by more conservative states, who argued that a non-definition for a 'vague' term such as gender would harm the legality principle of ICL. 75 This argumentative move around the legality principle proved crucial for gender-conservative actors. Since, at that point of the negotiation process, many delegations were both supporting the retention of gender in the Statute and the non-definition proposal for it, the legality argument allowed gender-conservative actors to change the course of discussions and push for a more 'precise' definition of the termone where they could include their own specific visions on the concept. 76 From that point onwards, the discussions on gender at the Rome Conference shifted from retaining or not the term in the Statute to the elaboration of a working definition that could accommodate both conservative and feminist perspectives on the topic. The result of such an impasse was the oddly constructed definition of Article 7(3), which clumsily combines these opposing views.
The principle of legality, considered one of the corollaries of domestic and international criminal law systems, is multifaceted. It encapsulates the maxims nulla poena sine lege and nullum crimen sine lege, 77 but the exact extent and meaning of these maxims are contested. Among its various meanings, legality in criminal law can be interpreted as a prohibition against retroactive law-making in criminal law; a requirement for criminal rules to be enacted by legal authorities and institutions; a demand for criminal rules to be clear and specific, or yet a principle that requires criminal offences to be construed leniently. 78 It has then a close link to a strict construction of criminal rules and the maintenance of their certainty and stability. 79 Oosterveld, supra note 4, at 64-5. 75 Steains, supra note 6, at 372-3; Oosterveld, ibid., at 63-4; Oosterveld, supra note 73, at 566-7. 76 Ibid. In the jurisprudence of the International Criminal Tribunal for the former Yugoslavia, for example, the legality principle is understood as establishing a narrow scope for judicial interpretation and innovation in ICL, whereby criminal courts are not allowed to create offences by analogy or 'by giving a definition to a crime which had none so far'. 80 Law-making in ICL is therefore taken to be strictly limited (in relation to judicial interpretation of crimes), or dependent upon either a robust track record of customary international law or lengthy processes of treaty negotiations and amendments (concerning the creation of new crimes by states). 81 For that reason, the legality principle in ICL grounds the field firmly upon an expectation of constancy, certainty and stability, something that mounts hefty opposition to normative change without a strong base in already established legal provisions.
All these dimensions of the legality principle in ICL can be considered as rules about rules, that is: rules that dictate how legal rules are to be produced, adopted, and interpreted in international criminal law. In Foucauldian terms, the legality principle could thus be considered as a technology of law-making governmentality in ICLa way to 'conduct conduct' in international criminal legal production. 82 More than that, I argue that legality is part of a (neo)colonial, gendered scheme of subalternity in international criminal law, one that elects public sovereignsstatesas the only agents able to speak in the law. By dictating that criminal provisions need to be clear, specific, and based on the traditional sources of ILi.e., treaties or custom 83the legality principle is particularly tied to a traditional paradigm of law-making based on the sovereignty of states.
Sovereignty in the international scenario has, as one of its core prerogatives, the ability to enter and, therefore, create international obligations. 84 Despite the burgeoning legal pluralism in the global system or the idea that international organizations and judges are law-makers in their own right, the formal paradigm that this capacity is held only by statesor derived from their delegationis still strong. 85 The monopoly held by states in speaking the law is then a marker of sovereignty and a foundational rule that demarcates who is inside or outside the field of public sovereigns. This law-making capacity is also the only avenue through which sovereignty is traditionally conferred upon, regulated, and modifiedsovereignty is, after all, a legal construct. 86 By enacting statutes, charters, declarations, treaties, and custom, international sovereigns can welcome new members to the 'sovereignty club', create duties and obligations, as well as modify and extinguish forms of political organization. 87 Not only that, all these acts are dependent upon the consent of the audience of sovereigns in order to be recognized and therefore taken as legally valid. 88 80 Prosecutor v. Mitar Vasiljevic, Judgment, Case No. IT-98-32-T, T. Ch. II, 29 November 2002, at 75, para. 196. See also Art. 22(2) of the Rome Statute. 81 UN Security Council, 'Report of the Secretary General Pursuant to Paragraph 2 of Security Council Resolution 808', UN Doc. S/25704 (1993), at 7, paras. 20-2. 82 Here I am particularly inspired by Jauhola's (supra note 64) work on looking at gender mainstreaming practices through a governmentality lens. Werner and De Wilde, supra note 84, at 288-96; Aalberts, supra note 86, at 92-124. 88 Ibid.
The circularity of the dynamic between sovereignty and law-making capacity demonstrates not merely the co-constitution of the scheme but also its arbitrariness. While states are taken as the natural sovereigns of the international legal order, they are also the gatekeepers of the fieldand this gatekeeping is far from unbiased. As already explained by feminist critiques to international law, the concept of sovereignty is both sustained by and supportive of gendered notions of authority. The public realm of the sovereign state is historically tied to a gendered division between public/private, which is homologous to a men/women division of labour that assigns to women the sphere of the hearth and the home and to men the sphere of the polis and the public. 89 This dichotomous division between public/private plays out to the oppression and exclusion of women in social life and institutions, including in the law. 90 Furthermore, the concept of sovereignty in international law is historically grounded on imperialist notions of 'civilization', which have long been used to exclude third-world peoples from the field. 91 The very own constitution of sovereigntybased on the notion of a 'civilized family of nations'has been continuously built through differentiation practices vis-à-vis the Others of international law, 92 be they third-world peoples, 'rogue', or 'failed' states. 93 Sovereignty is, therefore, a key concept in IL that excludes all of those who do not conform to the masculine, western tenets of state authority. More importantly, it precludes any meaningful possibility of speech in the law that does not come from states or is mediated by them. Without sovereignty, the Others of international law are embedded within a scheme of subalternity where they cannot ever speak autonomously in the field.
This scheme of subalternity is strengthened by the principle of legality in ICL. When the principle demands clarity and specificity, it is, ultimately, asking for worded precisiona working definition of concepts that can cast away ambiguity and arbitrariness, as seen in the case of the Rome Statute negotiations. However, since states are the gatekeepers of law-making through their sovereign status, this imposes a burden of speech that roughly excludes any autonomous voice that is not uttered, authorized, or mediated by them.
Furthermore, here, it is also possible to see legality as embodying a gendered ordering pull for law-making in ICL. It acts as an imposition of conformity with a singular, specific legal language consolidated by prior practices recognized as legally relevantwhich, again, are conventionally protagonized, selected or mediated by states. 94 Hence, whereas already established terms can have the privilege of being non-definedor at least defined by the long history of precedents constructed over timenovel concepts hold the burden of speech and linguistic conformity in order to be considered clear, specific and, ultimately, legal. 89 In this sense, the legality principle is not merely a corollary of a normative system but a part of a gendered and (neo)colonial boundary work practice within the juridical field. 95 By imposing rules and constraints on legal normative production, it seeks to draw clear guidelines on what belongs to the legal field and what should be factored out as extraneous to law. However, what is crucial is that the existence of such legality guidelines does not mean they are understood and shared across the board. By discussing the legality of a normthat is, which requirements it needs to attain to be considered as legaldifferences and heterogeneity will be inevitable: actors will perceive and interpret what is the law and what is non-law differently, be it because of their positionalities, interests or identities. 96 Indeed, whereas the legality principle was argued to be important for the clarity of the gender definition in the Rome Statute negotiations, the boundaries of this clarity were visibly contested. For feminist NGOs and allied states, it was clear enough that gender was a social construct; for gender-conservative states and NGOs, it was clear that the term was bound to 'biological sex'. While the legality principle drove this unsurmountable heterogeneity to be pacified by forcing the amalgam of two opposing discourses, the non-definition proposal would be reappropriated with new strengths two decades later.

Breaking the burden of speech: Silence as resistance and authority in the CAH draft
Differently from the Rome Statute draft, the preliminary text elaborated by the ILC for the CAH considered gender as a protected category under persecution from the start. 97 For this purpose, the ILC replicated the Rome Statute definition, in light of the broad consensus achieved for Article 7 on crimes against humanity. 98 The ILC preliminary text was open to comments from states and non-state actors, with proposals for reform gaining momentum in late 2018. Among these submissions, the campaign coalition between MADRE, Outright, and the City University of New York (CUNY) Law School requested the definition to be either removed or replaced by the definition laid out by the Office of the Prosecutor (OTP) of the ICC in its 2014 Policy Paper on Sexual and Gender-Based Crimes. 99 The letter brought three main arguments for these propositions.
First, the coalition put forward the claim that the gender definition in the Rome Statute was not only outdated and opaque, but that it has also created obstacles to the prosecution of sexual and gender-based crimes. 100 They argued that given its faulty composition, no other mechanism had 95 adopted a similar conceptualization of gender. 101 Second, because of the unduly limitations of the definition, the campaigners also explained it has not been able to keep up with the progressive evolution of human rights in protecting gender minorities against discrimination. For that, they provided a detailed compilation of jurisprudence, treaty-law, general recommendations, and reports of UN special procedures on the issue. In this compilation, they underscored how the term gender had been generally understood as a social construct across multiple legally relevant documents and instruments. Indeed, in the letter, the campaigners are careful to assert that the notion of gender as a social construction is the overall accepted usage of the term, therefore not comporting other interpretations. 102 Third, they added that no other protected category in the crime of persecution was defined. In this regard, they pointed out that having 'a definition may imply that persecution on the basis of gender is secondary or qualified, and not equivalent to other persecutory categories'. 103 The submission of interACT: Advocates for Intersex Youth and Intersex Human Rights Australia followed similar argumentative lines. 104 They recommended the amendment of the provision or simply its deletion, in order to allow the term to be 'unrestricted[ly] interpret[ed]'. 105 Their position seemed more inclined to the non-definition of the term, arguing that ' : : : "gender", standing alone, can be interpreted to encompass all aspects of sex and gender : : : in addition to traditional notions of "the two sexes"'. 106 Matching the submission spearheaded by MADRE, they also argued that striking the definition would put the category 'on the same footing as other protected classes insofar as it would no longer be modified by what some would perceive as a limiting construction'. 107 The submission by UN Independent experts brought similar considerations as well. Whereas their letter affirmed that international human rights law (IHRL) would recognize 'gender as the social attributes associated with being male and female', it stressed that it is 'an evolving social and ideological construct that justifies inequality and provides a means to categorize, order and symbolize power relations'. 108 Their letter also proposed either the modification of the definition of gender to fit within the notion of a social construct or the removal of a definition altogetheragain, based on the argument that no other protected category of persecution is defined. 109 The interlocking of all these arguments provides an interesting example of when the silence of a non-definition can serve as a strategic path for resistance. For that, the consolidation of gender as a social construct in IHRL is crucial for two reasons.
First, advancing the need to update the gender definition in the CAH draft because of the developments in IHRL gave the campaigners a roomier framework than the one provided by ICL. As it is possible to see in the campaign, while non-definition is one of the proposed solutions, the campaigners did not claim that gender is undefinable or undefined in international law. Instead, they relied heavily on the understanding of the term within IHRL, arguing that the progressive 101 Davis and Bradley,supra note 97,at 9. 102 In their submission, the coalition campaigners affirm that, by replicating the Rome Statute definition: 'the convention in its current draft could be misread as promoting the misconception that the term "gender" has multiple understandings codified under international human rights law and as such denies its consistently recognized definition as a social construct' (Davis et al.,supra note 8,at 4.) development of the concept in ICLwith a few exceptions, of course 110had been significantly impaired by the opaque definition provided by the Rome Statute. Whereas not explicitly argued in the campaign, this can be read as an illustration of how the principle of legality in ICLwhich, as explained in the previous section, creates a more rigid construction and interpretation of the law based on existing legal provisionsrendered the Rome Statute definition a limiting provision that curtailed more progressive understandings of gender. IHRL, on the other hand, developed conceptualizations for the term that were not dragged down by such a limitation, thus encapsulating a more open framework where some feminist and queer concerns in relation to gender discrimination have gained more traction.
However, the campaigners could not rely only on IHRL for their proposals. While the CAH draft is intended to develop into a treaty that is couched within IHRL, it alsoand one might say, predominantlysits within the legal framework of international criminal law. And this is where the non-definition of gender, instead of a mere update to reflect the usage of the term in IHRL, comes as a powerful strategy. This is because electing a worded definition drawing mostly from IHRL could be read by other actors (especially the ILC) as incongruous with the broader legal complexity and design of the CAH draft. 111 The non-definition provided the campaigners with an avenue to break with the outdated and limiting moulds of the existing definition of gender in international criminal law, all the while allowing them to avoid having their concerns and proposals being dismissed as mere 'human rights lingo'. It is then an insightful example of using silence as a way to deploy a dispersion that is not related to absolute axes of reference. 112 Second, the argument that no other protected category has been defined brings an interesting possibility to mimic an authoritative silence for gender. Although the concept was not the only new addition brought by the ICC Statute to the crime of persecution, it is indeed the only category that is followed by a definition given the record of its contested negotiations in the Rome Conference. Aside from the arguments explored in the previous section, gender-conservative states also argued, back in 1998, that gender persecution and gender violence were not wellestablished concepts and thus merited further clarification. 113 Gender was therefore placed as a category that needed to be explained due to its vagueness and novelty, whereas other categories were taken for granted and understood as sufficiently precise via the developments of prior jurisprudence and statutes of international criminal tribunals. 114 As an act of precaution against similar contentions, there was a visible effort from the campaigners to compile an extensive track record of jurisprudence, general recommendations and reports of UN independent experts to attest to the legal pedigree of gender in IL. By showcasing a consistent evolution of international precedents and quasi-legal documents where actors have engaged with the term, the campaigners were able to ground gender as a usual concept in international legal practice. This approximates it to a category that, given its prior legal usage, speaks for itself. The attempt to mimic an authoritative silence for gender is therefore not completely wordless: it is backed up by a legal-technical language and expertise, something that helps lend 110 Davis et al.,supra note 8,[2][3] I thank one of the anonymous reviewers for indicating this point in earlier versions of this article. 112 See supra note 52, and accompanying text. 113 For example, the representative of the Syrian Arab Republic argued that he 'knew of no speciality called "gender violence"'; while the representative of Azerbaijan asked for clarification if the term 'gender persecution' would not raise issues with interpretation and translation (UN Doc. A/CONF.183/13 (1998), supra note 71, at 228, para. 46;272, para. 61). 114 The categories of political, racial, and religious persecution were established since the Nuremberg Charter and the statutes of the International Criminal Tribunal for the Former Yugoslavia, and the International Criminal Tribunal for Rwanda mirrored this wording.
an authoritative claim to the campaigners' proposal by adhering to the dominant epistemologies of the international legal field. 115 However, while legality pushes the campaigners to use traditional sources of law such as treaty law and judicial decisions as much as possible, the campaigners also make use of the murky legal authority of IHRL general recommendations and reports from UN independent experts. This inbetween legal characteristic of these documents is the result of ongoing fragmentations and evolutions on the concept of law-making authorities in IL, 116 but more importantly, they provide a fertile ground for the campaigners and their proposals. As demonstrated in their campaign, feminist, and queer concerns in relation to persecution have been particularly picked up by IHRL and expert-based normative production. This makes such frameworks more spacious sites where those perspectives have flourished, all the while still being clothed within (quasi-)legal language.
6. Organizing cacophony: Silence as speech management in the CAH draft Before their submission to the ILC, the campaigners were met with discouraging reactions to their proposals of either deleting or updating the gender definition from the CAH draft. 117 Since the Rome Statute provided a somewhat consolidated text for crimes against humanity, it was regarded by treaty supporters as a format that could enable quick negotiations without dissensionthus mounting a particular challenge to any proposal of change. 118 Following the call for comments and submissions to the ILC draft, the campaigners started a worldwide effort to rally other actors (such as supportive states, UN agencies, and civil society) to elaborate submissions to the Commission on the issue, as well as in organizing workshops and briefings with activists. After such concerted efforts, the campaigners were able to gather broader momentum for their proposals. 119 This is visible, for instance, in the comments and observations provided by states to the treaty draft. On that occasion, several delegations supported either the update or the complete deletion of the gender definition. 120 In multiple instances, the main arguments of the campaigners were rehearsed, and even the same language used in the campaign was employed by state 115 representatives. 121 However, states' responses seemed generally more inclined to update the definition, with only a few countries specifically asking to drop the definition of gender completely. 122 In the discussion among ILC members, the deletion was more broadly accepted, following the lead of the Special Rapporteur on crimes against humanity, Mr. Sean Murphy. In his fourth report, the Special Rapporteur revised the draft text and decided to delete the definition of the term altogether. 123 For that deletion, he asked 'whether it is necessary or appropriate to impose the same definition on all States for the purpose of their national laws regarding crimes against humanity'. 124 By citing a report from the independent expert on protection against violence and discrimination based on sexual orientation and gender identity, Mr. Murphy highlighted the vast differences in the concepts of gender identities across the world, which not only go beyond men and women, but are also a result of the diversity of cultural traditions around the globe. 125 When discussing the Fourth Report, ILC members widely agreed to delete the Rome Statute gender definitionbut often with caveats. During the 3453 rd meeting of the Commission, Mr. Murase, for instance, affirmed that the ICC definition did not reflect current developments in IHRL. He further noted that, although amendment would be more desirable than deletion, 'the issue [of gender] was an evolving one'. 126 This way, he considered that leaving the term undefined should not be regarded as a 'negative response from the Commission, but as a positive action aimed at removing obstacles to the healthy future development of international law on the issue of gender '. 127 In the 3454 th meeting, Mr. Park supported the deletion of the ICC definition, arguing that it no longer reflected the development of the concept and that it could improperly limit the scope of the protection against persecution. 128 Nevertheless, he still indicated the need for the omission to be adequately explained by the Special Rapporteur should the Commission agree to it. 129 In his intervention, Mr. Wood affirmed that 'unless there was a pressing need to make changes, the Commission should not depart from the language used in earlier, widely accepted conventions', since this could 'lead to uncertainty by appearing to call into question the accepted understanding of existing provisions'. 130 However, he found that there was a strong case for the deletion of the gender definition and accepted it. 131 Likewise, Mr. Rajput demonstrated an uneasiness with innovations in the CAH draft. He found, however, that there was enough clarity in the deletion of the gender definition, even if in contrast with the already established Rome Statute provision. 132 During the 3455 th meeting, Ms. Lehto considered it relevant to expand the grounds protected from persecution, including gender identity and sex characteristics. 133 She also supported the 121 deletion of the Rome Statute definition, arguing it had already been contested at the time it was adopted, was 'blatantly outdated', and did not comport to the current understandings of the term. 134 She noted, however, that while providing a more updated version for the definition was ideal, the deletion was agreeable as a 'practical' solution. 135 Mr. Nguyen also supported the deletion of a gender definition from the CAH draft, arguing that the ICC definition had not kept up with the recent developments in IHRL. 136 Mr. Nolte also supported the deletion, noting, however, that the commentary should clarify that the interpretation of the term should be made in accordance with IL, especially IHRL. 137 In the 3456 th meeting, Ms. Galvão Teles aligned herself with the position brought by Ms. Lehto, agreeing that deletion of the definition was a 'pragmatic solution'. 138 Mr. Hmoud also agreed with the deletion of the ICC definition, arguing that mirroring its limitations would be 'both inoperable and unjust'. 139 Mr. Reinisch explained that he was 'generally in favour of adhering as closely as possible to the wording of the [Rome Statute]', but that the reasons presented by the Special Rapporteur had made a convincing case for deleting the gender definition. 140 As to the 3457 th meeting, Mr. Saboia argued that the deletion of the definition altogether was a pragmaticbut not idealsolution. He also advanced that if the term were to be left undefined, a proper explanation in the commentary should highlight the 'concerns that existed in most societies regarding the situation of lesbian, gay, bisexual and transgender persons'. 141 Ms. Oral accepted the deletion and supported that the draft articles should reflect current realities and be flexible to enable the evolutive process of international law. 142 Ms. Escobar Hernández unequivocally supported the deletion of the definition, arguing it was 'totally baseless and out of step with current social reality and the legislation of a good number of States that had a different understanding of the concept'. 143 She further argued that defining the term would be extremely difficult and pointless, especially given the ongoing addition of new categories to be protected under gender concerns. 144 In the 3458 th meeting, Mr. Jalloh noted that while a more modern definition of gender was preferable, he could fully endorse the Special Rapporteur's deletion proposal. 145 The Chair, Mr. Šturma, also agreed with the deletion, noting that the definition had been the result of compromise during the Rome Conference and agreed it should not be re-employed in the draft. 146 What these excerpts demonstrate is that, very much like the campaigners, state representatives and ILC members regarded the deletion as a viablebut not necessarily the bestsolution.
Here the positions are much more cautious and preoccupied with making the multiple social meanings of gender more precise or at least more detailed. While states seemed to generally favour an update for the term, ILC members seemed more open to the deletion. However, this deletion for the ILC does not necessarily represent a move to resist limiting definitions or to assert gender as a category that speaks for itself. Instead, ILC members entertained the deletion as a pragmatic or practical solution either due to the term's historical evolution and social contingency, its contentiousness back in the Rome Conference, or yet because it would be impractical to impose a single gender definition for a treaty designed to be applied and implemented by national jurisdictions.
This non-definition then comes as a strategic device of lending an aura of inclusivity to the convention's conceptualization of gender, one that allows a more ambiguous space for the term to remain mobile to its different (state) users. By not engaging with a working definition for the term, it avoids the contentious route of electing a fixed definition of a contested concept, thus preventing a possible cacophony in the negotiations or later in the adoption and implementation of the convention. Moreover, by characterizing multiple interpretations as inherent to the term, this non-definition encodes into the core of the concept the capacity to incorporate a wide range of different interpretations and understandings. Silence then acts here as a co-ordinator of contestation and heterogeneity, creating a seemingly 'democratizing' and 'inclusive' space where (legal) dissent and disorder are avoided at all costs.
In the long run, however, this can harbour potential risks for the critical edge of dissentient voices. As already explained, incorporation comes with a price: one that often muffles radical difference so it can be included in a dominant core. Particularly in a discipline and practice where more subversive conceptualizations of gender are constantly dismissed or under attack, 147 this may further subdue the critical potential of progressive gender thinking and theorizings, especially of those that have not been well internalized or accepted into the discursive frames of international law.

Conclusion
This article sought to expand the understandings of silence in IL by using as a case study the negotiations of a non-definition for gender both in the Rome Statute and in the CAH draft. While silence is understood in international law mainly in its non-active roles, this article departed from a curiosity as to how silence can also be employed for more active functions, namely as a device for resistance and as a mechanism for managing disagreement. By exploring such roles for silence, the examinations of this article have led to deeper understandings of legality and lawmaking in international (criminal) law more broadly.
First, by looking at silence as a tool for resistance, this study has examined its subversive legalpolitical potential. By proposing a silence in issues where clarity in the law circumscribes lawmaking activities to the voices, interpellation, or mediation of states, actors can defy the dominant and homogenizing culture of (public sovereign) speech in IL. In this regard, actors can break with a subalternity scheme that does not out-shout them but demands a response through narrow channels of communication that make an autonomous response virtually impossible. To demonstrate this, the case of the campaign led by feminist activists, academics, and UN independent 147 A relevant example of this is the recent partially dissenting opinion issued by Elizabeth Odio Benito in the Vicky Hernandez v. Honduras case at the Inter-American Court of Human Rights. There, she argues against the application of the Belém do Pará Convention to protecting trans women against gendered violence (without, however, foreclosing their protection under other human rights instruments). She does so by drawing a stark division between 'biological women' and 'trans women' on the basis that 'sex' would be irrefutable, further defending that the subject of feminism is the first group. If Judge Odio Benito, considered as one of the major feminist figures in international law, propels forward trans-exclusionary ideas that are so contrary to many theoretical and political developments in queer and feminist critique over the past decades, one is left to wonder how much more averse the international juridical field may be to such theoretical and political thinking. experts for the non-definition of gender was particularly illuminating. It enabled a more profound look into how silence can be employed strategically to break with limiting moulds of discourse. In that regard, by proposing a non-definition of gender to be interpreted in light of relevant developments in IHRL, campaigners were able to include a concept of gender in the CAH draft that spoke to their concerns. At the same time, the ambiguous silence enclosing this non-definition pre-emptively foreclosed a powerful counterargument: one that could dismiss this proposal by claiming it would be incongruous to have a 'human rights definition' in a draft convention that predominantly caters to ICL. Silence thus provided a strategic and ambiguous space for legal innovation and inclusion, even among the limitations particular to international (criminal) law-making.
Second, by looking at silence as a mechanism for managing disagreement, this study has ventured into the possible limitations of the subversive uses of silence. The ambiguity of silence, while defiant to a dominant culture of speech, can also be a pacifying place that co-ordinates dissentient voices by not electing one single response. It allows for the coexistence of different views, construing an image of an all-encompassing whole that embraces all perspectives. The position of states and the ILC in accepting the non-definition of gender in the CAH draft was telling in this respect. By agreeing that heterogeneity and contestation were essential to the legal concept of gender, ILC members coined the non-definition as a practical solution to avoid cacophony in the future steps of the CAH draft. Silence served there as a pacifier of contestation and instability in the international law-making practice, reinstating an image of order within the juridical field.
Regarding suggestions for future research, perhaps the most intriguing path forward would be to accompany how the non-definition of gender will be progressively constructed should the CAH draft be adopted by states as is. The resistant vacuity of the non-definition can be, of course, co-opted for conservative purposes, especially as the CAH is to be applied through national jurisdictions.
Nevertheless, the use of silence as resistance still has a relevant value for critical research in international law. As explained in Section 2, silence is often met with distrust by feminist and TWAIL scholars and activists. As demonstrated in this article, there is much more to its uses and roles in IL. This is because the ambiguity of silence can provide a powerful avenue to defy an adherence to dominant frames of legal discourseframes that do not (and were never devised to) fit the realities of subaltern peoples.
Against this background, perhaps an interesting research avenue to further explore the potential of resistant silence is within the context of transitional and international justice. As already introduced by previous works on these fields, standing in silence in a courtroom or in justice processes that demand verbal participation can be read as a multidimensional act of resistance. 148 It can signify, among others, opposition to the formalism of courts and justice processes (including their discursive frameworks), a practice that encapsulates the unspeakability of traumas, or yet a strategy of survival. 149 This opens up the need to better explore how processes and institutions of global justice fail to address the needs of victims, witnesses, and survivors by foreclosing or misinterpreting their silences. Further, it can illuminate when, how, or whether those institutions can be reformed to allow silence to be recognized as both politically and legally relevant for global justice.
Another fruitful research path to pursue would be to explore further instances where strategic uses of silence may contribute to constructing more spacious rooms for subaltern voices and epistemologies in international law-making. Perhaps a stimulating exploration in this regard would be concerning the definition of 'environment' in the proposed amendment to add the crime of ecocide to the Rome Statute. While recognizing that the term has no single agreed definition in international lawand that the ILC has even left it undefined on a previous occasion 150the Independent Expert Panel for the Legal Definition of Ecocide proposed that the term 'environment' could be defined as 'the earth, its biosphere, cryosphere, lithosphere, hydrosphere and atmosphere, as well as outer space'. 151 While the Panel did acknowledge that environmental destruction has cultural impacts on Indigenous lives, 152 their specific definition of 'environment' draws from technoscientific knowledge for its justification as an appropriate concept. 153 However scientifically sound such a definition may be, it may run the risk of side-lining the meanings of environment for different Indigenous epistemologies, ontologies, and cosmologiesall of which are not fully grasped by technoscientific inquiries and languages. Against this background, there is space for further exploration of what a non-definition of environment could mean in terms of providing a roomier definition that could accommodate the particular meanings of the concept to different Indigenous peoples. Exploring the strategic uses of silence in international law-making can thus open up for more careful analyses of when and how to employ strategies against a dominant culture of speech in international lawstrategies that can destabilize oppressive limitations of what is considered as a valid legal concept and, hopefully, allow for more radical plurality and difference. 150 Stop Ecocide Foundation, 'Independent Expert Panel for the Legal Definition of Ecocide -Commentary and Core Text', at 11, available at static1.squarespace.com/static/5ca2608ab914493c64ef1f6d/t/60d1e6e604fae2201d03407f/1624368879048/ SEFoundationCommentaryandcoretextrev6.pdf (accessed 4 July 2022). 151 Ibid. 152 Ibid., at 8. 153 They explain that their definition 'draws upon scientific recognition of the interactions that make up the "environment"', citing a paper published in Nature. See ibid., at 11, note 12.