International Courts and Tribunals and Their Linguistic Practices: A Communities of Practice Approach

This paper argues that the framework of Community of Practice is beneficial for an understanding of the linguistic practices that international courts and tribunals employ in their interpretative approaches. Other than the frameworks of the social network, the speech community, and the epistemic community, the framework of Community of Practice can be said to allow for a more critical assessment of the social context in which international courts and tribunals function. Such an assessment is crucial in that it is in that social context that interpretative approaches can be said to take form and in return shape the social institutions that international courts and tribunals comprise. That is, the framework of Community of Practice entails the notion that any form of meaning and its subsequent reifications, including, actual language use, are continually negotiated communally as a result of social interaction and, hence, shape the social constellations in which such interaction takes place accordingly.


Introduction
How do international courts and tribunals (ICs) do things with words? More concretely, what linguistic practices characterise their language use? In order to make sense of these practices, it is important to take into account that (1) we are dealing with international courts and tribunals, and (2) that these in general can be said to have a clear, circumscribed task, namely that of adjudicating legal cases. That is, international courts and tribunals are defined here in line with Romano's updated ''taxonomy of international rule of law bodies'' [45, p. 243]. The main point is that the individuals on international courts and tribunals are judges or legal experts and that they decide cases on the basis of international law and legal procedure [45, pp. 254 and 261]. 1 Decision-making processes can be said to lie at the heart of this task of settling legal disputes. Making sense of ICs' linguistic practices therefore implies an understanding of their engagement in legal decisionmaking, or more precisely, judicial decision-making, as that term more accurately captures the fact that we are dealing with bodies of judges/legal experts.
Thus, in order to understand how ICs do thing with words, we have to have an understanding of the activity that ICs engage in and the actors that are involved. Although we can distinguish between the two, we cannot separate them in the sense that it requires an actor's engagement in a certain activity for the activity to materialise. With respect to the linguistic practices of ICs, this then means that these practices can be said to emerge against the backdrop of ICs' involvement in judicial decision-making. However, this does not suggest that linguistic practices 'come after the fact'; they are very much part of this involvement because of the linguistic nature of this decision-making. It is through language that ICs decide cases and it is that language that displays these linguistic practices. That such language use might be characteristic of certain practices brings up the notion of legal discourse. ICs' language use can be said to be a type of legal discourse in the sense that it is determined by ICs' involvement in judicial decision-making. Consequently, any linguistic practices depend on ICs' involvement in judicial decision-making as well; their makeup is determined by what Goodrich aptly describes as ''the roles, purposes and ideologies of its participants or subjects'' [24, p. 79], so that linguistic practices ''are accessible to study through their expression in the lexicon, syntax and semantics of [a] text'' [24, p. 79].
A recognition of linguistic practices as discursive entails a recognition of an interplay between, on the one hand, the social-institutional context of the way in which ICs do things with words, comprising an international court or tribunal as actor with its judges as actual agents and its engagement in judicial decisionmaking, and, on the other hand, the linguistic realisation of the way in which ICs do things with words as characterised by certain linguistic practices. This notion that the social and the linguistic are somehow intertwined has been well recognised in the fields of sociolinguistics and linguistic anthropology. 2 It has also been observed 1 As Romano nonetheless points out, the World Trade Organization Appellate Body forms an exception, as the individuals on the bench are generally not referred to as judges but as members [45, p. 254]. 2 For sociolinguistics, see e.g., [30,31]. For linguistic anthropology, see e.g., [14,19]. in legal theory. 3 Furthermore, this interplay between context and linguistic realisation can be depicted as follows-see Fig. 1. Figure 1 may be seen as a first attempt at how we might make sense of ICs in relation to their use of language in interpreting and applying the law in cases that are brought before them. However, to fully grasp this relationship between context and linguistic practices, such a preliminary step might not suffice. Recognising such interplay is not enough, as actors can be said to engage in a certain activity in a specific way. Conceptualising ICs as types of social constellations thus means that we see them as engaging in judicial decision-making in a particular way, yielding particular forms of language use. In order to render possible an understanding of linguistic practices, we might therefore make use of certain frameworks that enable a further conceptualisation of this interplay.
Given the very explicit recognition of an intertwinement between the so-called 'social' and the 'linguistic', considering the usefulness of the social network 4 and the speech community 5 seems self-evident, as both have featured prominently in sociolinguistic inquiries, although the latter has also been of use in linguistic anthropology. 6 Another framework that cannot readily be ignored is the epistemic community. 7 It has gained ground in international relations theory in conceptualising non-state actors in the international community. Although it does not directly reflect on the relation between linguistic practices and their social-institutional context, it does frame language use in terms of persuasion and seems to acknowledge an existence of concomitant linguistic practices (see Sect.

below).
A framework that is not borne out of linguistics either but nevertheless has proved important for the study of legal interpretation is the interpretive community as described by Fish [20]. Mainly written in light of the interpretation of literary texts, an interpretive community denotes a group of people that ''share interpretive strategies […] for writing texts, for constituting their properties and assigning their intentions'' [20, p. 483]. Readers are part of an interpretive community to the extent that they share and use the strategies of a particular group of writers. Differences in interpretation do not come about because of differences in texts but because of intersubjective differences: ''the only 'proof' of membership is fellowship, the nod of recognition from someone in the same community'' [20, p. 485].
Despite its seeming relevance, I will not consider the interpretive community as a possible framework for conceptualising international courts and tribunals. The reason for doing so is that it only meaningfully allows for a commentary of the interpretive approaches that ICs employ while I am interested in the more concrete level of language use of ICs. To Fish, the notion of language use can be said to be subsidiary to a primarily substantive rendition of judicial decision-making with 3 See e.g., [52], in particular Chapter 2, 'The Language of Concepts: A Case Study', pp. 22-45; and [5], who points to a unity of form and content. 4 See e.g., [6,39]. 5 See e.g., [28, 29, p. 105, 30]. 6 See e.g., [41]. 7 See e.g., [11, 26, 27]. interpretative strategies lying at the heart of such rendition, an objective which I am not concerned with here.
In the same vein, we could, of course, also think of other ways of conceptualising international courts and tribunals, such as institutions that are characterised by pathdependence, meaning ''that how litigation and judicial rule-making proceeds […] is fundamentally conditioned by how earlier legal disputes in that area of law have been sequenced and resolved'' [46, p. 113], 8 or as groups of professionals working within the legal field, 9 or perhaps even as 'actor-networks'. 10 Given the focus of this paper on the question what ICs do with words and, thus, on an understanding of ICs as 'language users', I will not consider these other frameworks in more detail but instead focus on the three mentioned above. This is not to say that these three frameworks mentioned here cannot be useful; rather, an assessment of their usefulness merely falls outside the scope of the present paper given the legallinguistic focus of the paper.
A framework that I will consider and has been adopted in sociolinguistics and in international relations theory is that of the Community of Practice (CofP). Originally proposed by Lave and Wenger [34], and further explored by Wenger [51] and in sociolinguistic studies, 11 central to the concept of Community of Practice is the notion that any form of meaning, including actual language use, is continually negotiated communally as actors within that community participate with one another, and shapes the constellations in which such negotiation takes place accordingly [51, p. 52].
The present paper purports the view that an understanding of ICs in terms of the framework of CofP will be especially useful for an understanding of the linguistic practices of ICs. That is, it is argued that the Community of Practice is better able at conceptualising ICs in the act of judicial decision-making than the social network, the speech community or the epistemic community, so as to render possible an understanding of ICs' linguistic practices. To further explore this notion, the present An internaƟonal court or tribunal ENGAGEMENT IN Judicial (with judges as actual decision-making agents) LINGUISTIC REALISATION e.g., actual language use characterised by linguisƟc pracƟces Fig. 1 A depiction of the way in which we might make sense of an IC, its engagement in judicial decision-making, and actual language use as a realisation of that engagement paper proceeds as follows. Section 2 describes the main characteristics of the social network (2.1), the speech community (2.2) and the epistemic community (2.3) as well as the CofP (2.4). Section 3 compares the Community of Practice with these frameworks in their ability to grasp the relation between ICs as actors engaging in judicial decision-making, and their concomitant linguistic practices. Section 4 provides some concluding remarks.
2 Conceptualising International Judiciaries: The Social Network, the Speech Community, the Epistemic Community and the Community of Practice

The Social Network
It is generally held that the analysis of social networks (henceforth: social network analysis or SNA) emerged in the 1930s [21, p. 31]. It was adopted in different ways in social psychology, social anthropology and sociology [44, pp. 19-20]. As a form of doing research, it has developed into a fully-fledged framework with its own ''conceptual, methodological and analytical tool-kit'' [44, p. 19]. Central to SNA is the detailed mapping of the social ties and relationships that exist among a network's members. As Freeman has pointed out, ''social network analysis is motivated by a structural intuition based on ties linking social actors'' [21, p. 3] and hence, is informed by a strong empirical methodology, including ''graphic imagery'' [21, p. 3] and all sorts of ''mathematical and/or computational models'' [21, p. 3]. An example of a graphic image of a social network is given in Fig. 2. 12 There is no reason to exclude social network analysis as a method of inquiry into international judiciaries a priori. An example of social network analysis in relation to an international judiciary comes from Lupu and Voeten [36] in their study of precedent in case citations by the European Court of Human Rights. Their network included judgments of the Court, the different nodes in their graph indicating case law [36, p. 425], 13 which is probably why they referred to their method as network analysis solely. However, it is clear that they turned to SNA methods in order to gauge precedent in the Court's jurisprudence. Although not resorting to SNA, Terris, Romano and Swigart recognised the importance of social network conceptualisations for an understanding of the dynamics inside international courts and tribunals: ''Judges necessarily become part of personal and professional networks within their institutions, which must, in turn shape the way they perform their work'' [48, p. 49].
This brings us to the question of what the added value might be of mapping and detailing the relations underlying a group of social actors, more concretely, a group of judges acting collectively on an international court or tribunal. Although there may not be a reason for precluding SNA to international judiciaries, an obvious difficulty lies in the fact that judiciaries are expected to act in unison where it concerns judicial decision-making and the handing down of judgments. Individual ties might therefore not be readily clear and may prove difficult to trace. Networks are surely bound to exist as Terris et al. [48, p. 49] aptly observed, but whether these networks truly influence the work of judges cannot readily be foretold. 14 Therefore, SNA may not be ideal for unravelling the social dynamics underlying an international judiciary since individual ties and relationships, as focal point in SNA inquiries, are subordinate to a judiciary acting as a collective which it does as a consequence of its role as dispute settler.
Nevertheless, SNA has been widely adopted in sociolinguistic inquiries, both contemporary and historical, 15 for which reason it is important to consider the social network framework in relation to the linguistic behaviour of international judges. The social network has found sway in sociolinguistics precisely because of its focus on the social ties linking social actors together, which has allowed inquiries to focus less on social variables, such as class or status, which might prove unsatisfactory in their explanatory force when employed in isolation [38, p. 46]. The social network's capacity to ''illuminate patterns of specifically linguistic behaviour'' [38, p. 46] 16 lies in its ability to explain linguistic behaviour on the basis of a combination of underlying social ties and relationships, on the one hand, and social variables, on the other. We do not look at a group and distinguish it solely in terms of class, for  14 Indeed, see [35], who focuses on the so-called ''invisible aspects of judicial internationalization (to the extent that, VW) judges conduct dialogue with their foreign peers'', p. 116. 15 See e.g., [4,37,40,49]. 16 Emphasis omitted. example; we link this social variable to the ties and relationships that exist in that group. Linguistic practices are therefore understood in terms of this combined perspective.
The question ultimately is to what extent the social network framework might be of use in gauging the linguistic behaviour of international judiciaries. Language use can be assessed through an inquiry of judgments and separate opinions, and, if desired, through ethnographic survey. An obvious difficulty nonetheless lies in explicating how linguistic findings should be understood given the direct social ties that may very well exist among international judges but prove difficult to ascertain, thereby obfuscating the possibility of explaining any linguistic features in terms of some social variables that are evidently needed in the social network framework.
Although the social network has potential, its focus on individuality compounded with social variables, such as age, sex or class, in ''illuminating patterns of […] linguistic behaviour'' [38, p. 46] 17 may not directly take into account the collective nature of judiciaries' practice of decision-making, and the linguistic manifestations that arise within that interaction. A conceptualisation of ICs as social networks recognises a relation between actor and the linguistic level. However, it is not clear how judicial decision-making informs this level. There is no explicit recognition of ICs' engagement in judicial decision-making and concomitant linguistic practices. For all that, the social network may prove valuable in comparative studies of the extent to which international judges interact outside their courts with other judges. 18 We could summarise our findings by means of the template presented in Fig. 1 above-see Fig. 3.

The Speech Community
Roughly speaking, the speech community describes a group of people as a social aggregate with a certain linguistic makeup. The speech community is well established in the fields of sociolinguistics and linguistic anthropology, and has been defined in many different ways, each definition emphasising some other aspects, although all seem to stem from a belief in a relation and interaction between what could loosely be described as the social and the linguistic. Hymes, for example, has defined a speech community as ''a social, rather than linguistic, entity. One starts with a social group and considers the entire organization of linguistic means within it'' [28, p. 47]. To Hymes, then, a speech community is not necessarily characterised by recognised, standardised form of language use. Rather, it is more likely that members speak some language variety [28, p. 47]. Labov has stressed the importance of norms in that a speech community is ''united by a common set of evaluative norms, though divergent in the application of these norms'' [30, p. 355]. Here, Labov seemed to suggest that evaluative norms comprise both subconscious and conscious standards of language use which are informed by social factors and as such are applied in different ways giving rise to variation in linguistic performance.
What is absolutely essential is that there has to be some commonality linking the members of a speech community together. What this may be is not readily clear; yet, Morgan denotes that speech communities may well be ''part of a dominant culture and [that] others may result from an event, activity, etc.'' [42, p. 16]. Although the concept of speech is central to the definition of a speech community, speech communities do not only exist where actual speech can be heard, but rather, where people interact and participate with each other [42, p. 1]. Labov seem to have hinted at a speech community's participatory nature when he defined a speech community ''by participation in a set of shared norms'' [31, pp. 120-121]. To Gumperz [25], then, face-to-face interaction underlines the entire concept of a speech community.
Morgan has identified three analytical concepts that can help determine whether a social aggregate can be considered a speech community: (1) there has to be some degree of communicative competence, that is, ''skill in applying the norms, values, and rules of language use'', (2) there has to be some degree of socialisation, that is, the knowledge necessary in order to participate, ''using language and discourse norms and standards'', and (3) there have to be some contextualisation cues, such as ''the level of immersion, learning, and trial and error necessary to function as a communicatively competent person in a speech community'' [42, pp. 36-41]. Thus, what is central to the framework of the speech community is the idea that some shared social and linguistic norms exist as a consequence of social interaction.
There is no direct reason to assume that international judiciaries cannot be conceptualised as speech communities. Their commonality seems to reside in the fact that they comprise bodies of judges/legal experts so that communicative interaction is a given. A first glance at the three analytical concepts teaches us that some degree of communicative competence, of socialisation and of contextualisation are surely there, but to what extent requires some detailed ethnographic study, which, given the scope of this paper, cannot be elaborated upon here.
Although a speech community might result from a certain event or activity, to what extent that event or activity actively plays a role in constituting and constraining linguistic practices is not readily clear, while the contextualisation of social interaction necessarily entails the norms underlying a certain activity. It seems that once a speech community has been established, it is very much a given,

An IC
Judicial decision-making (WITH JUDGES AS ACTUAL AGENTS) ? (NO SPECIFIC ACKNOWLEDGMENT) Social Ɵes and LinguisƟc pracƟces relaƟonships (AS LINGUISTIC REALISATION) + Social variables Fig. 3 A summary of the findings concerning the application of the social network framework to ICs by means of the template presented in Fig. 1 V. Willems providing a site where speech events can emerge and be identified, and thereby appearing as a somewhat static, isolated foundation upon which linguistic behaviour is built provided there is social interaction. Although a relation between the social and the linguistic realms is well recognised, once constituted, a speech community seems to provide a mere background against which linguistic behaviour can be studied, overlooking a possible interplay between linguistic and social norms which may be continuous and contained in linguistic practices. In this sense, the findings for the speech community are similar to those for the social network. ICs and their linguistic practices can be linked to one another, but it is not clear how judicial decision-making informs this dynamics. International courts and tribunals are not just being competent in a language variety; they do something, which does not seem to be taken into account explicitly.
Just as with the social network, we can summarise our findings by using the template in Fig. 1-see Fig. 4.  [26, p. 3] who are steered towards achieving a particular goal by relying on the same set of principles and beliefs. Its means of achieving that goal is through persuading others of its motivations and knowledge [13, p. 11], since an important characteristic of an epistemic community is that it usually is not the one in control of actual decisionmaking [26, p. 4]. Epistemic communities are called upon in order to clarify and untangle the intricacies and difficulties involved in decision-making processes so that actual decisions can be made.

The Epistemic Community
In international relations theory the epistemic community has been taken up to describe various non-state actors in international politics, in particular groups of

An IC
Judicial decision-making (WITH JUDGES AS ACTUAL AGENTS) (NO SPECIFIC ACKNOWLEDGMENT) ? Competence in a parƟcular language LinguisƟc pracƟces variety through socialisaƟon (AS LINGUISTIC REALISATION) Fig. 4 A summary of the findings concerning the application of the framework of the speech community to ICs by means of the template presented in Fig. 1 International Courts and Tribunals and Their Linguistic… scientists. However, Davis Cross [11, p. 155] is quick to note that ''[d]iplomats, judges, defence experts, high-ranking military officials, bankers, and international lawyers, among others, all have just as much of a claim to authoritative knowledge as scientists'' so that epistemic communities do not always and only consist of scientists. Davis Cross clearly believes judges to be able to constitute epistemic communities as well. Yet, given that they are in control of actual decision-making processes, it is not readily clear to what extent they merely persuade others. Nevertheless, their decisions might steer the perceptions of policymakers, especially when they are delivered in final instance without further remedies. For this reason, in their study of precedent in the jurisprudence of the European Court of Human Rights, Lupu and Voeten [36, p. 417] assumed among other things that judges of the ECtHR are aware of the effects their decisions might have on policy. 19 In his book on the relation between judicial behaviour and judges' audiences, Baum [3, p. 6] points to strategic models of judicial behaviour which embrace the notion that judges are much aware of the influence they can exert so that they often decide cases strategically. Baum [3, p. 4] takes this one step further by arguing that judges are not only concerned with influencing their audiences ''as a means to other ends'' but also with the way audiences perceive them as such; ''judges engage in self-presentation to audiences whose esteem is important to them'' [3, p. 4]. Whatever the extent to which judges are concerned with audiences besides the direct litigants involved in a legal case brought before them, it is very likely the case that judges influence policymakers, whether domestic or international, and in that sense could be gauged in terms of an epistemic community.
Central to the framework of epistemic community is its ability to create a common ground through shared beliefs, a shared enterprise and a common goal. Its success therefore depends on what Haas and Adler [27, p. 368] have referred to as ''the creation of collective meaning'' within the community and its ability to act as a coherent entity to the outer world. Cohesion is thus crucial, even to the extent that when consensus proves difficult to maintain, an epistemic community is likely to collapse [27, p. 384]. Differences of opinion might threaten a community's viability in such a way that they severely jeopardise its persuasive nature.
Although international judiciaries act collectively, there certainly is room for difference of opinion. Not in the least can this be seen in the opportunity of separate opinions (i.e. concurrences and dissents), such as is possible with respect to the European Court of Human Rights, the International Court of Justice, or the International Criminal Court. 20 Nevertheless, the very existence of separate opinions has in the course of time prompted debate about the extent to which they create unity or disharmony, and, more broadly, to what extent they contribute to the authority of international courts and tribunals. 21 19 On the relation between law and politics surrounding the European Court of Human Rights more broadly, see [7]. 20 21 See further e.g., [2,12].
In order for an epistemic community to function well, it needs a firm consensual basis, both as a constitutive means and as a means to exert influence. Persuasion is the central modus through which this is achieved. Epistemic communities try to persuade policymakers of their policy knowledge concerning a particular problem. They do not merely present information or advise on how to proceed but actively seek to further and spread their solution to a problem. Thus, epistemic communities have an interest in the act of persuading others. Undoubtedly, their interest will be reflected in the way in which they seek to persuade policymakers. Despite this evident persuasive nature, not much is known about the way in which persuasion is manifested linguistically. Haas does point to possible places of inquiry, including ''speeches, biographical accounts, and interviews'' [26, p. 35], and of course, the materials directed at policymakers, but he remains silent on any linguistic aspects.
If we were to conceptualise international courts and tribunals as epistemic communities, any linguistic practices should nonetheless be understood in terms of this persuasive mode of operation. That international judiciaries resort to specific language use as a means of persuading litigants in a legal dispute is not all too unlikely. That they may have an interest in reaching the general public seems plausible as well, especially where it concerns cases that are of social importance and/or have caused a stir among groups in society. 22 But in general, the idea that international judges are set out to achieve some greater goal by attempting to influence policymakers does not sit well with the concept of sovereignty of states and the notion that judges' main turf is in principle restricted to matters of legal import. 23 This is not to say that international judiciaries always confine themselves to legal matters; reality is far less circumscribed, so that the boundary between law and politics is not always clear. Nevertheless, as was pointed out above, Baum has proposed that judges may very well be concerned with the way in which they are viewed by their audiences for the sake of regard, so that their interest cannot merely be subsumed under the notion of serving justice.
Again, we can summarise our findings by using the template in Fig. 1-see Fig. 5.

The Community of Practice
It is generally held that the phrase Community of Practice was first coined by Lave and Wenger [34]. As Lave and Wenger [34, p. 42] themselves indicated, the concept was ''left largely as an intuitive notion, which serves a purpose […] but which requires a more rigorous treatment''. Wenger later seized the opportunity to develop it into a fully-fledged framework [51]. Thus, it is Wenger's conceptualisation that forms the foundation for the present discussion. 22 See e.g., [1] where the European Court of Human Rights explicitly stated that Article 8 of the European Convention on Human Rights does not confer a right of abortion, so that women cannot claim to have that right under European law when national legislation prohibits abortion, para. 214. 23 See in this light, [43], who argue that judges are independent so long as ''they remove themselves from the formation of public policy. Those are the terms on which the judiciary wins its 'independence''', p. 58. I am not concerned here with the way in which judges are recruited and appointed.
Wenger defines practice as follows: The concept of practice connotes doing, but not just doing in and of itself. It is a doing in a historical and social context that gives structure and meaning to what we do. In this sense, practice is always social practice. Such a concept of practice includes both the explicit and the tacit. … [Thus,] the social production of meaning is the relevant level of analysis for talking about practice … [so that] practice [is] the source of coherence of a community [51, pp. 47; 49].
Central to the concept of practice is the production of meaning: ''[p]ractice is about meaning as an experience of everyday life'' [51, p. 52]. 24 More importantly, it is about the negotiation of meaning, that is, meaning is never a given, but instead has to be formed and continuously restated-it is a constant process. To Wenger then, ''we produce meanings that extend, redirect, dismiss, reinterpret, modify or confirm-in a word, negotiate anew-the histories of meanings of which they are part'' [51, pp. 52-53]. In the case an event arises that has not been foreseen by the law, yet asks for a legal interpretation, the application of a legislative act, for example, may very well involve the process of reinterpretation in order to provide that event with some legal grounding.
To Wenger, the process of negotiation ''involves the interaction of two constituent processes … call[ed] participation and reification'' [51, p. 52]. These processes form a duality in that they constantly interact with and complement each other. Wenger defines participation as: a process of taking part and … the relations with others that reflect that process. It suggests both action and connection. … [It describes] the social experience of living in the world in terms of membership in social communities and active involvement in social enterprises [51, p. 55].
Wenger's conceptualisation is very close to a common understanding of the term participation which can be defined as people taking part in something. But Wenger's conceptualisation implies more than social partaking and connectedness.

An IC judicial decision-making (WITH JUDGES AS ACTUAL AGENTS)
? (UNDERSTOOD AS PERSUASION?) clear in Shared principles, beliefs, and a common LinguisƟc pracƟces goal, i.e. the (AS LINGUISTIC REALISATION) advancement of a parƟcular worldview Fig. 5 A summary of the findings concerning the application of the framework of the epistemic community to ICs by means of the template presented in Fig. 1 For all that, Wenger [51, p. 56] asserts that participation does not only include peaceful, harmonious relations-conflictual ones can also characterise participation in a social community. Participation is not the mere ''engagement in practice'' [51, p. 57]; participation is constitutive of the identities of the members of a social community and as such shapes the community as a whole. Participation encompasses the entire involvement of each member in the process of the negotiation of meaning.
Wenger's use of reification is rather distinct from usages that have been salient in legal theory. 25 In brief, it is about ''giving form to our experience by producing objects that congeal this experience into 'thingness''' [51, p. 58]. This 'giving form' should be understood in a broad sense and surely does not always involve a linguistic encapsulation or verbalisation of an experience; it involves ''making, designing, representing, naming, encoding, and describing, as well as perceiving, interpreting, using, reusing, decoding, and recasting'' [51, p. 59]. Hence, reification comes in a variety of forms: a monument dedicated to remember those that fell during World War II is just as much a reification as a document spelling out the rules of a board game.
Most importantly, reification is never absolute in that forms of reification do not exist in isolation with a clearly identifiable meaning. Thus, the monument that is founded to remember the victims of World War II might become a symbol for the atrocities that war in general engenders, allowing the remembrance of victims of other wars as well. Undoubtedly, its meaning today is not the same as the meaning of those who founded it. Similarly, the document laying down the rules of a board game is continually reinterpreted, as some rules may be considered as outdated, too strict or simply obnoxious. Thus, participation and reification should be understood as being part of a cyclic motion of interplay-as a corollary, meaning, which is negotiated in that interplay, ''is at once both historical and dynamic, contextual and unique'' [51, p. 54].
Wenger's conceptualisation of community implies the concept of practice, so that a community can only be understood as having a certain practice 26 : To associate practice and community, I will describe three dimensions of the relation by which practice is the source of coherence of a community …: 1. mutual engagement 2. a joint enterprise 3. a shared repertoire [51, pp. 72-73].
With respect to the first dimension, i.e. mutual engagement, Wenger [51, p. 73] notes that ''[p]ractice resides in a community of people and the relations of mutual engagement by which they can do whatever they do''. This engagement comes 25 The term reification has found its way in legal theory, most notably in the realm of Critical Legal Studies. In this light, see e.g., [22,23]. However, the Critical Legal Studies movement has been quite sceptical about the concept. In the present article, it is used as Wenger deploys it, that is, as being conducive to a Community of Practice without questioning its applicability to the realm of law as such. 26 This is not to say that Wenger [51, p. 72] only acknowledges the existence of communities that have a clearly identifiable practice. On the contrary, there may be communities that lack such practice, but these then are not Communities of Practice. about through the inclusion of members into a community and intensive work that Wenger [51, p. 74] describes as ''community maintenance''. Mutual engagement does not imply that all members in a community are or become alike. On the contrary, heterogeneity is very much part and parcel of a community in that members retain their identities although these may converge. Thus, heterogeneity as well as homogeneity can be observed in a CofP. Mutual engagement also means that relationships are established. These may be close or somewhat distant, friendly or, as was seen above, even hostile.
Intensive involvement in a community may be superfluous if there is not a joint enterprise driving that involvement. However, Wenger [51, p. 78] is quick to note that a joint enterprise does not merely equate ''a stated goal, but creates among participants relations of mutual accountability that become an integral part of the practice''. That is, ''[i]t is the result of a collective process of negotiation that reflects the full complexity of mutual engagement [and] … is defined by the participants in the very process of pursuing it'' [51, p. 77]. Joint enterprises may not be verbalised, yet only tacitly understood. The do's and don'ts in law, for example, which are sometimes implicitly understood, can be under continuous influence from outside, i.e. from outside the community, but are nonetheless negotiated within the confinements of the community [51, p. 80]. Members within the community understand what drives their practice and what it is driven by.
Pursuing such a joint enterprise over time yields a shared repertoire of negotiated meaning. According to Wenger: the repertoire of a community of practice includes routines, words, tools, ways of doing things, stories, gestures, symbols, genres, actions, or concepts that the community has produced or adopted in the course of its existence, and which have become part of its practice [51, p. 83].
Interestingly, to Wenger the shared repertoire includes both reificative and participative aspects. At first glance, however, the shared repertoire may be perceived of as mainly reificative; yet, keeping in mind the interplay between reification and participation, it should also be understood as incorporating elements that yield participation, which continues the process of the negotiation of meaning as being central to a CofP. In this way, a shared repertoire is constantly renewed. That is, some instances of the repertoire might lose their potential to express a sense of sharedness and will be dropped from the repertoire accordingly, while other meaning formations come to be included as new instances. By having judges on a IC participate in the practice of judicial decision-making, leading to instances of reifications, such as, most notably, a body of case law, meaning can come about. For instance, with respect to the European Court of Human Rights, the notion that the European Convention on Human Rights should be understood as a living instrument was established in case law. 27 In terms of the CofP, the living instrument doctrine as a form of reification constrains ensuing participation in the practice of negotiating meaning, leading to other forms of reifications, which well include linguistic manifestations. The CofP allows for a combined understanding of ICs and their engagement in judicial decision-making. Judicial decision-making as a form of negotiating meaning is in a constant cyclical motion because of continuous participation among judges and the concomitant establishment of forms of reification. Understanding linguistic practices as forms of reification incorporates the notion that they are manifestations of a continuous process that involves the participation of judges on an IC in judicial decisionmaking. Hence, the CofP framework takes into account this interplay depicted in Fig. 1 more than the social network, the speech community and the epistemic community appear to do so.
That is, in light of this interplay, the social network and the speech community may not offer the best possible framework for conceptualising ICs. Their social dynamics seem mainly dependent on social interaction and do not seem restricted by a particular social endeavour, which might have a steering effect on concomitant language use and social behaviour. They can nonetheless be used, but a clear weakness lies in their lack of acknowledging explicitly the way in which judicial decision-making might inform actual language use; that is, that decision-making is  Fig. 6 A depiction of the application of the CofP framework to ICs by means of the template presented in Fig. 1 27 It was first noted in [50], where the Court observed: ''The Court must also recall that the Convention is a living instrument which, as the Commission rightly stressed, must be interpreted in the light of presentday conditions. In the case now before it the Court cannot but be influenced by the developments and commonly accepted standards in the penal policy of the member States of the Council of Europe in this field'', para. 31. not merely a social variable, but delineates the boundaries of social interaction and language use within which an international judiciary behaves.
If one wishes to know whether a group of people contributes to policymaking and in what way, the epistemic community might be a good starting point. Nevertheless, it is premised in the notion that a group's enterprise is steered towards persuading others and is therefore characterised by a great degree of cohesion. Disagreement is regarded as a potential threat to the existence of an epistemic community. With respect to international judiciaries, then, it may be too ambitious to consider them as homogeneous bodies of dispute settlement, though some degree of homogeneity may surely be detected. It is perhaps more realistic to view them as mainly heterogeneous bodies that continuously work towards a common-not necessarily shared-outcome as they are instructed with a task that requires collective action.
The CofP framework seems to provide the rigour needed to make sense of the underlying social-institutional dynamics that drive an international judiciary in its endeavour of settling disputes in the international legal order. At the same time, it appreciates the notion that an international judiciary as a group of social actors is likely to present itself with a great degree of diversity and heterogeneity. In this way it enables an understanding of commonalities and idiosyncrasies in the language used to negotiate meaning. When it comes down to the linguistic practices that international courts and tribunals employ, a CofP approach might therefore be better able at taking stock of the shared linguistic norms inherent to a judiciary's shared linguistic practices. These norms will have to be acquired for a successful participation in these shared practices, yielding forms of reification that constitute a judiciary's language use and characterise it in the act of decision-making.
Methodologically, then, any framework poses a different approach and its usefulness is determined by the research question one sets out to answer. In light of this observation, the argument in favour of the Community of Practice in relation to international judiciaries does not necessarily foreclose the possibility of adopting any of the other three frameworks. On the contrary, future research might gain from assessing the applicability and usefulness of these three frameworks in more detail for an understanding of the relation between ICs' social-institutional and linguistic realms.

Conclusion
Given the difficulty involved in conceptualising an aggregate of people, one might simply decide not to conceptualise at all. The danger in refusing to conceptualise, however, lies in the fact that one might not be able to explicate in detail the social dynamics that characterise a particular group of people and which might form the background against which subsequent activities and practices emerge. In the present case, in not choosing to conceptualise international judiciaries, one might run the risk of overlooking potentially relevant social-institutional aspects, such as, the existence of different chambers that are assigned different tasks, as well as the existence of guidelines and protocols that ought to be taken into account, and which can help explain what is taking place on a linguistic level. A clear understanding of the social-institutional context is thus paramount in order to gauge any linguistic practices in more detail, not only for the sake of linguistics, but conversely, for a potentially better understanding of the ''institutional structure'' by which law is ''produced'' and in which law is ''embodied'' [10, p. 251]. That is, empirically, a Communities of Practice approach allows us to grapple with linguistic patterning as well as variation in language use, by taking into account the institutional processes that are at work as well as the individual forces that contribute to the making and remaking of shared repertoires. Where language is pivotal for the law to work as a social-institutional ordering mechanism, some insights from linguistics might thus not prove undeserving. This paper has hopefully made a first step in outlining the sociolinguistic depth inherent to the question of how international courts and tribunals can be said to do things with words.