What’s the point of being a discipline? Four disciplinary strategies and the future of International Relations

While disciplinary identities are among the most fraught subjects in academia, much less attention has been given to what disciplinarity actually entails and what risks different disciplinary strategies involve. This article sets out a theory of disciplinarity that recognises not only their coercive but also their redeeming features, particularly in view of the coexistence of multiple competing disciplines and powerful transdisciplinary movements (such as rationalism). On this basis it identifies four disciplinary strategies and each is assessed in relation to the future of IR: (1) remaining a subdiscipline of Political Science (‘stay put’), (2) becoming an interdisciplinary field (‘reach out’), (3) dissolving into transdisciplinarity or abolishing IR (‘burn down’), or (4) establishing IR as a discipline in its own right (‘break out’). Rejecting the false choice of disciplinary constraint versus epistemic freedom, this framework allows IR and other subfields to more consciously consider a range of disciplinary strategies and to entertain the risks and affordances they each offer. The article concludes that a future independent discipline focused on the implications of ‘the international’ not just for politics but all fields – including disciplinarity – would make for a broader, more diverse IR, ultimately also better able to engage other disciplines.


Introduction
Whether International Relations (IR) is or should be a separate academic discipline is the subject of long-standing debate (Buzan and Little, 2001;Dyer and Mangasarian, 1989;Grenier et al., 2015;Kaplan, 1961;Kennedy-Pipe, 2007;Kristensen, 2012). The discipline-question resurfaced most recently in response to Justin Rosenberg's (2016) startling claim that IR has so far failed to clearly identify its own unique subject matter, defining itself only negatively as a subfield of Political Science concerned with politics beyond settled state confines. Without a notion of 'the international' as a problem in its own right, IR has remained conflicted about its subject matter and narrowly focused on a certain sphere of politics, making it relatively ineffectual in contributing to other disciplines, runs the argument (Rosenberg, 2017, see also Brown, 2013;Halliday, 1994). To remedy this, a positive conception of the core subject matter of IR is offered: '[N]o matter how much we twist and turn it in our hands, the word "international" always ends up presupposing the same basic circumstance, namely, that human existence is not unitary but multiple. It is distributed across numerous interacting societies' (Rosenberg, 2016:135). Since no other discipline has the consequences of societal multiplicity as its object, IR can with this identify its own unique vantage point, and become relevant to the whole of the social world (not just politics), freeing itself from the 'prison of Political Science' (Rosenberg, 2016, see also Albert and Buzan, 2017) and thereby become a fully fledged discipline in its own right.
But whereas it used to be a source of worry that IR fell short as a discipline, there is now widespread anxiety about disciplinarity succeeding. Some fear a new disciplinary prison of colonial modernity (Blaney and Tickner, 2017) or unwanted disciplining at the hands of positivist scientism (Jackson, 2017). Catarina Kinnvall (2019) captures the sentiments of many of those aiming to expand the remit of IR and who are sceptical of disciplines: at a time when we are concerned with decentring IR, recognising that there may be many IRs rather than one [. . .] and when we are increasingly asked to investigate the white mythology of IR [. . .] the call for disciplinary cores and thus boundaries, seems to be problematic. (pp. 153-154).
Others see a 'set of warning lights' going off and 'red flags' waving at the idea of defining a core subject matter risks 'catapulting' innovative IR work out of the discipline (Drieschová, 2019: 156-157). Patrick Jackson (2017) welcomes Rosenberg's argument but declares himself 'simply not interested in defining anyone or anything out of IR ' (p. 83). So what Rosenberg takes as the great selling point is for others the precise reason why his reformulation of IR's subject matter must be rejected.
In this article I argue that, although answers differ as to what the core problem of IR is, the actual crux of this debate is pre-judged: should IR even aspire to become a discipline on a par with history, geography or economics with its own unique angle on the social world? Or should it avoid identifying its subject in positive terms as a discipline in its own right in a bid to avoid further harmful disciplining of knowledge and scholarship about its subject (whatever that is deemed to be)?
In this debate, 'disciplinarity' has been cast as code for a restrictive state-centric IR subject matter and exclusiveness in terms of approaches. It is painfully true that disciplines regulate and restrict knowledge cultivation, and there are very good reasons to be critical of the specific history and content of much of existing IR. In a world of multiple disciplines, however, the alternatives to being an independent discipline might also not be so rosy, nor lead to more diverse and inclusive scholarship. Important strategic choices should not be made without pausing to ask, 'what is disciplinarity' and 'what are the disciplinary options actually on offer?'.
After assembling a theory of disciplinarity I consider four possible disciplinary strategies for IR. Each has advantages and risks, but I argue that it is hard to see a route to broader themed and more diverse and critical scholarship about the international without an independent discipline of IR with a general purchase on the social world. The first section sets out the continuing confusion around whether a core subject matter of IR can be identified and whether IR even has disciplinary potential. Second, key characteristics of academic disciplines are explored building critically on theorists and histories of disciplinarity, arguing that the multiplicity of disciplines makes a rejectionist strategy very risky. Third, the four disciplinary strategies available -subdisciplinarity ('stay put'), inter-disciplinarity ('reach out'), transdisciplinarity or abolition ('burn down'), and independent disciplinarity ('break out') -are developed in turn, noting, in particular, IR's early and ongoing epistemic vulnerability to other disciplines and epistemologies, not least positivism and, more recently, rationalism. Fourth, Rosenberg's suggested 'deep ontology' for IR in the shape of 'societal multiplicity' is reconsidered, arguing disciplinary subjects are constructed but in the shadow of the other disciplines and the international itself, making jettisoning IR as a discipline both hard and risky.

The reluctant discipline
The subject of IR is taught in various ways: through 'great debates' or mythical moments (De Carvalho et al., 2011), via paradigms, isms and 'schools' (Waever, 2004), or via 'research traditions' (Sil and Katzenstein, 2010). Such narratives organise IR into positions and categories that facilitate (or regiment) teaching, scholarship and theorising in IR, (re)producing 'IR' as they go. But curiously, these almost always take as given what makes them specifically 'IR'. IR theories are designated in a circular fashion as theories of 'international events' (Dunne et al., 2013) and bibliometric studies of the field of IR determine almost by methodological fiat that IR is simply whatever is published in journals classified as 'IR journals' (e.g. Aris, 2020;Kristensen, 2012Kristensen, , 2018. Practice theorists make this circularity explicit by declaring that '(d)efining what does and what does not count as an international practice becomes an empirical question best left to the practitioners themselves in their actual performance of world politics' specifying only that international practices are the ones that 'pertain to world politics, broadly construed' (Adler and Pouliot, 2011: 6). IR, it seems, is what happens while you're busy making other plans.
As a result, ambiguity reigns about IR's subject matter. Diverging labels for it appear alongside and on top of each other, including 'the field of international studies' (Lake, 2011(Lake, , 2013, 'world affairs' or 'global politics' (Lee-Koo, 2015: 381;Owens and Smith, 2020: 5:5). The IR textbook Global Politics teaches the apparently synonymous objects of 'international relations/global politics' specifying, that 'global politics' concerns 'politics at, and, crucially, across all levels -the worldwide, the regional, the national and subnational' (Heyward, 2014: 1). The specificity of the 'international' as a subset of globality, students are told, concerns 'states as key actors on the world stage' (Heyward, 2014: 1). This then frames 'the international' narrowly as state interactions, but identifies global politics 'across all levels' as the object of IR! When the question of IR's subject matter is addressed explicitly, statements to the effect that IR simply has no unique subject matter are not uncommon. For Christine Sylvester (2007), the days of agreement about the focus of IR are over and as such 'IR is at an end' (p. 551). For others, it never really was a thing: 'the subject matter that IR is concerned with is not distinct to IR' and 'the more IR tries to be a discipline, the greater its failure' (Baron, 2015: 260-261). As George Lawson and Robbie Shilliam (2010) note, 'IR appears as a kind of disciplinary Polo mint -an enterprise without a centre' (p. 70).
Puzzlingly, the deep hesitancy about defining the subject matter of IR coexists with a confident insistence on diversity in something nonetheless still referred to as 'IR'. Dunne et al. (2013) justify their call for theoretical pluralism by the idea that '(t)he sheer diversity and complexity of what is studied ensures that there will be multiple perspectives on what the most important factors are and how inquiry should proceed' (p. 405). Others aim for even the Polo-mint condition to be radicalised via 'fragmenting the international' in order to further disrupt state-centric views of politics in IR (Huysmans and Nogueira, 2016: 299).
The need to widen the subject area beyond state interaction is part of the case against considering IR 'a discipline': 'a lot of what now counts for IR is not about [interaction between states] but about ideas, peoples, norms, aboriginal rights, culture, multinational corporations and the environment' (Baron, 2015: 261). Caroline Kennedy-Pipe's suggestion 'for the future of the discipline is that, first, we forget about it as a discipline. It is not' (2007). Self-identifying as a discipline, it is held, only 'legitimizes the reproduction of particularly uniformizing and (epistemically) intolerant practices in the academic field' (Grenier, 2015: 250). Critiques of Rosenberg's idea that IR should embrace the idea of 'societal multiplicity' as the essence of the international echo such concerns: it risks providing a new standard with which to discipline IR scholars, assumed to favour statist, elitist interpretations of IR (e.g. Jackson, 2017;Shepherd, 2017).
For those reluctant to be a discipline, what should IR then be? A field of study, perhaps, oriented towards multiple objects and drawing on multiple other subjects -'a crossroads for other disciplines' (Kennedy-Pipe, 2007: 325). Jackson argues that -yes, IR was born in the prison of Political Science (in the United States at least), but no, it need not be 'a discipline' of its own with a settled point of focus or object of study -and absolutely not one particular epistemology. Why not instead remain 'a meeting place for a variety of academic disciplines around a common concern with international affairs' he asks (Jackson, 2017: 83).
Let us examine the options more carefully and deliberately.

What is a discipline?
A common assumption behind all this is something along the following lines: since disciplines police knowledge and scholars, the less 'disciplinary' IR is, the freer the IR scholar or student of world politics will be to include all that which has been excluded and to broaden knowledge and perspectives. Without disciplinary boundaries and cores, IR may better escape its current narrow and prejudiced incarnation, and make way for a more wide-ranging and plural object of scholarship focused on world politics/global politics/IR/international affairs/world politics and so on (the label immediately becomes problematic but also cannot be dropped completely). While IR undoubtedly must change (and is changing), what does disciplinarity entail? At one extreme, disciplinarity is simply considered a sociological phenomenon, consisting of institutions, journals and professional practices. In this view, IR's disciplinarity is more akin to a 'simulation' (Baron, 2015: 261): it need not involve a settled or unique subject matter, agreed upon theories or methods. Demanding that would in any case disqualify most academic fields from being disciplines, and moreover, 'it is almost impossible for any discipline to be truly autonomous from other realms of academic pursuit' (Turton, 2015: 4). Yet focusing on the institutionalisation of IR hardly does away entirely with the question of how it should conceive of itself, as Lene Hansen (2015) points out, 'as it would be difficult to sustain self-identification and resource allocation without a substantial narrative that identifies what one does as at least distinctive if not better than other fields or disciplines' (p. 267).
For Michel Foucault (1970Foucault ( [1966), probably the most famous theorist of discipline (though less known for his thoughts on disciplinarity), a field of knowledge production becomes distinguishable from others when it reaches the 'threshold of positivity ' (p. 196). A discipline has a defined shape and is unique and identifiable as itself, but also, crucially, can reproduce itself and produce new statements according to its own apparatuses and truth-practices. He contrasts this with truth previously being attached to certain individuals or other social spheres (e.g. law). Alluding to today's multiple separate disciplines, Foucault (1980Foucault ( [1977) points to the emergence, emerging in biology and physics, of what he terms the '"specific" intellectual' (savant) who 'intervenes in contemporary political struggles in the name of a "local" scientific truth' (pp. 128-129) rather than the older 'universal intellectual' who posed as a bearer of general truths and values.
In this view, a discipline is defined by a domain of objects, a set of methods, a corpus of propositions considered to be true, a play of rules and definitions, of techniques and instruments: all this constitutes a sort of anonymous system at the disposal of anyone who wants to or is able to use it, without their meaning or validity being linked to the one who happened to be their inventor. [. . .] what is supposed at the outset is [. . .] the requisites for the construction of new statements'. (Foucault, 1981(Foucault, [1971: 59) Thus, while '(t)he discipline is a principle of control over the production of discourse' (Foucault, 1970: 61), and hence restrictive, it also provides 'the possibility of formulating new propositions ad infinitum' (Foucault, 1970: 59) independently of specific inventors or authorities. A discipline is an 'anonymous system at the disposal of anyone who wants to or is able to use it' (Foucault, 1970). Those new propositions must of course acknowledge a set of rules and procedures for their generation in order to be 'in the truth', but within those strictures (and not 'in a wild exteriority', as Foucault calls it) an endless number of statements can be produced.
Disciplines are thus productive as well as restrictive, and outsiders can -in theoryenter and make use of them (after being schooled appropriately of course). Although Foucault is known as the great critic of disciplinary rule, 'it is wholly misleading to focus on the negative functions [of disciplines] alone' (Osborne, 2015: 26): they are generative too. Foucault t (1980Foucault t ( [1977) admits that in Madness and Civilisation, he had focused on how psychiatry repressed 'a sort of living, voluble, and anxious madness', but later realises that 'the notion of repression is quite inadequate for capturing the productive aspects of power' which rather 'transverses and produces things, induces pleasure, forms of knowledge, produces discourse' (p. 118-119).
Disciplines can be enabling in another way: shielding or providing scholarly toeholds in relation to transdisciplinary epistemes, for example, positivism. Foucault (2007Foucault ( [2004), it is true, initially discards disciplines as less significant compared to discursive formations. For example, the discourse of psychiatry is not reducible to the medical discipline of Psychiatry. The latter 'can only be understood as an institution on the basis of something external and general, that is, the psychiatric order' (p. 117); but later he recognises that disciplines allow some insulation from discursive epistemes like structuralism or the 'alternative danger of a complete dissolution of disciplines into positivism' (Osborne, 2015: 25). While they can provide shelter from powerful epistemes then, disciplines also change over time, of course, making them to some degree malleable, though this also makes them subject to political and resource pressures within and beyond the university.
What marks this out from the institutional 'simulation' view of disciplines is that it focuses on how modern social sciences are each crafted, not just via institutions like departments and journals, but through 'a series of abstractive moves, each producing a reified object of analysis, each with its own history' (Bell, 2019: 21). In particular, in order to emerge and be sustained as a discipline requires 'the delineation of a specific domain which its members can claim as their own, demarcating it from other disciplines and providing a focal point for research and debate' (Bell, 2019: 21). This specific domain emerges, often messily, and allows even discipline-sceptics to assume there is a subject that IR is attached to, even while they point out it has been conceived too narrowly. In a thinner version of a disciplinary scholarly 'domain', Ole Waever defends the much maligned 'great debates' and other structuring devices precisely for providing a focal point and means for IR to operate and to critique and develop itself, short of agreement on the object itself. For Waever (2021), 'we can see the discipline [of IR] as real and reproducing -even in the absence of a clear and given object (international relations) and a shared agreement (IR)'. Almost no discipline has complete agreement about its object, but all have a recognisable debate about 'its' remit, presupposing a discipline nonetheless.
Beyond the university, disciplines are in fact only one type of institution in a wider set of 'knowledge-complexes' (e.g. think-tanks, government agencies etc.) (Bell, 2019: 27). Even purely within the academy, academic subcultures and other 'transversal phenomena, criss-crossing and helping to (re)constitute various disciplines and fields' (Bell, 2019: 27) put a question mark over where IR would be without disciplinary structures to support it in relation to knowledge-complexes. Foucault (2019) identifies threats from structuralism and positivism, but Bell updates this with: modernisation theory, neoconservatism, neoliberalism and war. All of these (re)configured IR as a discipline via knowledge practices within and beyond the university.
Finally, Foucault also recognised that disciplines, rather than being purely a negation of freedom, are important for technologies of the scholarly self (Bell, 2019: 37) because the latter 'require truth obligations', providing the self at least some tools with which to work (Foucault cited in Osborne, 2015: 27). For Foucault (2019Foucault ( [1994), 'truth isn't the reward of free spirits, the child of protracted solitude, nor the privilege of those who have succeeded in liberating themselves. Truth is a thing of this world: it is produced only by virtue of the multiple forms of constraint' (p. 131). The most powerful of scholarly selves in the post war era are 'the "rational" social scientist -the individual who comes to view the world in a certain way, as capable of explanation (even prediction) through the application of the arts of "scientific" reasoning' (Bell, 2019: 21). But other scholarly selves such as 'radical historicist' also exist (Bell, 2019: 38). Such identities work through disciplines, making the latter a potential vector for them, as well as a potential foothold for those struggling against or between them.
Where discipline-sceptics see boundaries as purely coercive 1 , the above view situates disciplines in the context of other power-knowledge vectors. But for that reason, it is important not to underplay the power dynamics and influences involved in the emergence of disciplines. Foucault here over-plays the degree to which they in practice are anonymous and 'at the disposal of anyone who wants to or is able to use' them, as the rest of his writings on discipline attest. Colonisation and later industrialisation and the growth of modern (European) state apparatus all underpinned how knowledge production came to be organised in distinct disciplines in the first place. In that process, many contingent knowledge distinctions emerged or were constructed around which modern disciplines were formed and continue to be organised. For instance, European colonialism solidified and spread the 'Cartesian' nature-human divide as an organising epistemological and normative distinction, in the process placing much of the human and non-human world on the 'nature' side of the equation, subject to scientific objectification and militaryadministrative domination (Patel and Moore, 2017: 51).
The institutionalisation of the multiple modern disciplines is relatively recent. Within the academy, natural and social sciences only began to diverge formally by the 19th century after natural philosophy had previously split into natural history and the other physical sciences of chemistry and physics and biology (Heilbron, 2003: 42). Sociology, in some countries more clearly than others, emerged to facilitate the study of social life, carving itself out from the 'general theory of life', namely biology (Renwick, 2012: 2) or for August Comte, out from physics (hence his 'social physics'). 'Social science', first thought of as a meta-discipline but later an umbrella term for several disciplines distinct from the physical sciences and humanities, appeared only in the late-19th century (Porter and Ross, 2008: 2). Social sciences divided variously including into Sociology, Economics, Anthropology Psychology and Political Science, where IR currently mostly resides (departments of just IR exist but are rare).
Institutionally, only in the last decades of the nineteenth Century [. . .] do we see disciplines with three key characteristics: deciding what is taught in the departmental structure, deciding what is good research through dedicated journals, and deciding who gets hired and promoted. (Repko et al., 2020: 30-32) The international mattered throughout all this, and the now dominant 'departmental' anchoring of disciplines evolved first in the US context, only spreading outside the United States 'well into the postwar period' (Abbott, 2001: 123). This rested on US hegemony after the Second World War, and its need to organise knowledge production to serve the purpose of first fighting the Cold War and managing decolonisation and then from the 1980s organising a 'global' world order with the United States at the helm (Kamola, 2019: 44). Funded by huge flows of philanthropic and federal money, the previously more diverse pre-war international studies scholarship began to be shaped into 'a discipline capable of providing the American security apparatus the conceptual tools needed to imagine -and therefore to manage -the world as an international state system' (Kamola, 2019: 44). While disciplines have become ubiquitous and provide means of articulating and building specialist knowledge, they help reproduce and are imbricated in wider processes and inequities. So what can be done?

IR's four disciplinary options
Given how disciplines function and emerged historically, and the near-ubiquity of them worldwide today, scholars who wish to depart from or adjust given categories or objects of knowledge face a set of strategic dilemmas. The following identifies four ideal-typical strategies available to IR today: 'stay put', 'reach out', 'burn down' and 'break out'. Some of these could plausibly be pursued simultaneously, although some trade-offs may also be inevitable. Here I explicate and assess them and suggest that all of them ultimately rely in some way on IR becoming a distinctive discipline, ideally with a broader purview.

'Stay put': IR as subdiscipline
First, mainstream IR currently de facto accepts a subdiscipline strategy in the sense of typically (though not exclusively) being institutionally located in or with Political Science department and virtually always having for itself 'an ontology of political power (operating in the absence of central authority) rather than an ontology of the international per se' (Rosenberg, 2016: 5). This limits IR to a subset of politics, sometimes narrowly, sometimes broadly conceived.
To be sure, Political Science was not the original home (or prison) for international scholarship everywhere and a rich and deep history of ideas about multiple societies preexisted the Euro-centric version of IR found widely today (Acharya and Buzan, 2019;Buzan and Acharya, 2021). But in Western canonical texts, IR's object was defined negatively as politics outside or beyond sovereign authority: 'the way [E.H. Carr] proceeds is not to identify what premises of its own the international might uniquely contain. It is rather to extend the premises of Politics into the international sphere' (Rosenberg, 2016: 131). Hans Morgenthau (1959) decided that 'for theoretical purposes international relations is identical to international politics' and furthermore: 'a theory of international politics is but a special instance of a general theory of politics' (p. 15). Morten Kaplan (1961) knew of 'no convincing discussion that a specifically international relations discipline in its own right exists' (p. 465) but considered the field of international politics (just about) viable as a subdiscipline. A politics-centric and statist designation of IR's subject matter arguably remains the dominant stance today among scholars and teachers of the international (Maliniak et al., 2012: 4).
Besides a restricted subject matter, the main problem of being a subdiscipline lies in negotiating the relationship to the dominant host discipline. Crucially, if a subfield is essentially a corner of another discipline, does it really need its own theories and methods? Helen Milner (1998) argued explicitly that international politics is now so empirically intertwined with domestic politics that there is hardly any reason to have separate theories, never mind distinct disciplines: the rationalist paradigm's common 'concern with the mechanisms of collective choice in situations of strategic interaction' (p. 760) obviates the need for a intellectual endeavour distinct from Political Science.
This debate seems perennial. At the Rockefeller-funded 1954 conference on IR, often considered a founding moment, the fortunes of IR as a discipline were thought to depend on 'a substantive' or 'general theory' of international politics. This was crucial, participants realised, if 'doubts about the disciplinary nature of IR as well as about the analytical distinctiveness of its subject matter' (Schmidt, 2011: 86) were to be put to rest. David Singer opined that 'without theory we have only the barest shadow of a discipline' (cited in Schmidt, 2011: 87). Morgenthau warned in that context against 'eclecticism which is the exact opposite of theory' (cited in Guilhot, 2011: 257) and grounded the need for theory in the threat from social scientific impetuses bearing down on IR as a part of Political Science (yet he also decided IR theory would be 'not in essence different from that of general political theory ' (1962: 77). In contrast, today eclecticism is called on by some to save the discipline from tiresome theory (Lake, 2011(Lake, , 2013Sil and Katzenstein, 2010) while others insist that 'one of IR's comparative advantages over other disciplines might just be its strong sense of being a theory-led and theory-concerned field' (Dunne et al., 2013; see also Mearsheimer and Walt, 2013;Reus-Smit, 2012). Yet without an object distinct from politics, why should such theories differ fundamentally, and not just regionally, from those in Political Science? And why should methods prevalent there not suffice here?
Strapped to politics, IR seems unable to find its focus or feet.

'Reach out': IR as interdisciplinary field
Interdisciplinarity is often a productive and positive strategy for bringing multiple lenses to bear on cross-cutting societal problems. Yet contact and cross-fertilisation between disciplines is not preordained to be equal or to end well. Power differences exist between, as well as within, disciplines (Rosenberg et al., 2022). Disciplines that import or borrow theories from others, especially if they do so without being clear on their own disciplinary contribution, risk becoming what David Long (2011) has dubbed invaded disciplines providing 'little more than data sets, issue areas, or context -the empirical domain, the field, on which the [invading] theoretical paradigm operates' (p. 43). The discipline reaching out can also do so in exploitative ways. IR has arguably often treated History in this way, plucking examples to illustrate supposed logics of the international (e.g. Waltz, 1979). But, more often than not, IR has itself been invaded by other disciplines, not least by Economic methods and rationalist axioms, often posing as objective policy advice (Rathbun, 2017).
On the more critical wing of the discipline the multiple 'turns' in IR suggest an almost limitless appetite to draw on other disciplines. One tally registered six such turns in recent IR scholarship (Baele and Bettiza, 2020) importing variously from psychology and neurosciences (affect); sociology and ethnomethodology (practice); while the 'material turn' has imported ideas and concepts also from and Science and Technology Studies (STS) and social theory. These have all generated valuable insights when applied to subject matter normally identified as 'international', but often without making clear what is specifically international about the practices, matter, affect and so on, in question. If the imported ontology (matter or affect, for example) is centred in analysis, the specifically international dimension can disappear from view. Theories and concepts routinely get imported and deployed without being internationalised, as if they fit seamlessly into IR (Corry 2022). At the same time an equivalent 'international turn' in Psychology or neuroscience, STS or even Anthropology has not yet happened (although perhaps it has in historiography, see Armitage, 2004;Bell, 2002). Thus, internalism -the 'tendency, at a deep theoretical level, to conceptualize society in the singular' (Rosenberg, 2016: 140) -abounds, albeit largely unwittingly, in neighbouring disciplines and even in IR itself as it grafts concepts and theory from sociology, anthropology and so on -disciplines not dedicated to the significance of the international -onto its subject matter.
Interdisciplinarity has not been all bad of course. The extraordinarily narrow conception of politics that developed in mainstream (US) IR by roughly the end of the 1980s (rationalist and focused largely to states in cooperation or conflict) has been vigorously contested by a broad field of critical IR that typically draws on sociological and anthropological theories that see the social as 'constituted by practices, as relational, processual, assembled' (Guzzini, 2017: 369). This sets up a much wider notion of politics than the mainstream rationalist one pertaining also to 'the way the fixtures of social reality are constituted' (Guzzini, 2017: 339; see also Bigo and Walker, 2007). But without a distinct and positive notion of the international, this tends to fall back on the mainstream delimitation of it, in an attempt to anchor the specificity of the object it expands upon or goes 'beyond'. In an editorial, the journal International Political Sociology celebrates that it no longer has to justify 'why our work is legitimate in the context of IR' focusing on diverse 'formations of the international', but the editors call for 'submissions that do not shy away from the difficult work of explicitly addressing why a particular topic / question / puzzle matters to the terrain of the global, broadly conceived' (Lisle et al., 2017: 2). This 'terrain of the global' turns out to refer to many quite standard IR issue areas: 'key areas and processes that remain central to the understanding of transformations of the international, including war, militarization, migration, weapons, security, and colonial legacies' (Huysmans and Nogueira, 2016: 300) , or 'border sites' and 'objects that evade sovereign borders' (Lisle 2014: 69). What makes those factors 'international' is considered obvious, while Mark Salter (2016) also pin-points IR's focus as the politics of sovereignty: '[w]hat the field of IR can contribute to ANT and assemblage thinking is a concern with the constitutive nature of sovereignty and the politics that it limits' (p. xiii).
Thus, in multiple ways the problem of the object 'the international' reappears, evenor especially -when interdisciplinarity is pursued. Interdisciplinarity is paradoxical in that it presupposes what it seeks to go beyond -namely, a discipline.

'Burn down': transdisciplinary ends and anti-disciplinary means
In contrast to interdisciplinarity, 'transdisciplinarity rejects the parallelism of established disciplinary methods, theories and paradigms, and calls instead for an alternative or novel approach at odds with one or more of the established disciplines' (Long, 2011: 38). Transdisciplinarity is typically motivated by the complexity and multifaceted nature of 'real life' problems. It is often invoked to 'overcome the mismatch between knowledge production in academia, on the one hand, and knowledge requests for solving societal problems, on the other' (Hoffmann-Riem et al., 2008: 4).
This idea animates much of the problem-solving and critical opposition to IR as a discipline today, but the clearest exposition of transdiscipinarity is provided by Immanuel Wallerstein (1996) and his somewhat older idea for the abolition of social science disciplines and the creation of an open and historically informed uni-disciplinarity. The report 'Open the Social Sciences' (Wallerstein, 1996) was defined in opposition to modern disciplinary boundaries, which Wallerstein (2000) deemed to be organisationally strong but to have 'lost most of their historic intellectual justification' (p. 33).
However, worryingly for those aiming for more pluralist knowledge production, transdisciplinarity also comes with risks. As David Long (2011) warns, 'while the hope is for new theory, in most instances transdisciplinary exchange involves the application of theory from one discipline onto the field of another, or the borrowing of concepts from one by another' (p. 42). Does opening the social sciences inadvertently roll out 'the Trojan horse for the dissolution of particular disciplines by bringing them into a hierarchical relation with more powerful disciplines' (Burawoy, 2013: 7)? And which are more powerful? Arguably, Economics or systems approaches emanating from natural sciences look much more likely contenders for the role of super-discipline than Wallerstein's World Systems approach. Wallerstein himself recognised that disciplines reflected deep world-spanning inequalities and that changing them would take more than innovative scholarship (see also Kamola, 2019): Today, for obvious economic reasons, the bulk of social science is done in a small corner, the rich corner, of the globe. This distorts our analysis, and the distortion is structural not individual. No amount of virtuous self-discipline on the part of individual scholars will correct this situation. (Wallerstein, 2000: 34) Despite this risk, might an abolitionist strategy concerned with 'clearing the decks'precisely due to entrenced biases -still be a necessary risk to be run? If IR is based on inherently flawed foundations, it may need to be not just adjusted but taken out -uprooted at its conceptual foundations before it can be re-formed. This goes for the mainstream canon of IR which, for example, through epistemologies of ignorance 'exclude or exceptionalise the central role of racialised dispossession, violence, and discrimination' in IR (Sabaratnam, 2020: 12). But it could also apply even to critical IR, the argument runs. Critical scholars may be busy unveiling such epistemologies of ignorance and exclusions, but what if the tools they use to do so are infected by the same foundations, privileging another coterie of white Eurocentric thinkers like Derrida and Foucault (Grosfoguel, 2007: 211). Discussing Critical Security Studies (CSS), Chandler and Chipato (2021) state that '[a]ny project of reparative work for generative ethico-political openings would have to be undertaken after the abolition or dismantling of critical security studies, not as a substitute for this' (p. 64). The only strategy left is then 'the refusal of the settled order of academia, the flight from the institutional demands of disciplinarity, the embrace of dissonance instead of clarity' (Chandler and Chipato, 2021: 65). They end with a call to arms: 'The only thing we can do is tear this shit down completely and build something new' (p. 66).
Highlighting the abolitionist's dilemma, however, what 'build something new' might refer to is left unattended and the authors self-avowedly 'build on the important existing critiques of race in security studies' (Chandler and Chipato, 2021: 66) to stake out their abolitionist position. Although abolitionists are not obliged to provide an answer, the first questions in a post-abolition space would be 'what is now our object of study?', 'how do we wish to study it?' and 'for what purpose?'. After which an institutional strategy would be required concerning which outlets, professional structures and curricula to aim forand how to relate to any other disciplines that had not yet been abolished. Either new host disciplines would be required or a transdisciplinary meta-disciplinarity á la Wallerstein would be on the cards.
An abolitionist strategy may thus be useful for avoiding superficial makeovers that conceal underlying problems born of the history of disciplines and their entanglement with worldly inequities. But it may also backfire by ceding ground to other stronger disciplines while ultimately relying on disciplinary rebuilding. The question reappears: should IR be an independent discipline?

'Break out': augmenting IR's distinctive disciplinarity
The remaining position is the one that attempts to reground or augment IR's credentials as a distinct discipline with a unique subject matter requiring bespoke theories and research programmes. Unlike the other disciplinary strategies, this considers IR's current disciplinarity 'stunted' (Long, 2011: 36), that is, limited, but potentially valuable and salvageable.
Despite some ambivalence, an example of this on the critical side is the ongoing project of 'Global IR' (Acharya, 2014;Rojas, 2016;Tickner and Waever, 2009) aiming to rethink or reworld IR (Bilgin, 2014). Global IR pushes for a broader and less Eurocentric, less rationalist and Western discipline, but from within 'IR', explicitly or implicitly recognising it as a distinct discipline. Roland Bleiker's (1997) call to 'forget IR theory' emphasised also that 'forgetting' involves not an abandonment of IR but rather a supplementary strategy designed 'to open up more inclusive ways of perceiving and practicing IR' (p. 76). Non-Western thinkers and 'historical patterns of interstate relations in the non-Western world' should on this account be 'viewed as sources of IR theorizing' (Acharya, 2014: 652; see also Shilliam, 2011). Critical of IR, Amitav Acharya nevertheless aims for 'reimagining IR as a global discipline' (Acharya, 2014: 649, emphasis added). Those who find that the global IR project retains too singular a global imaginary also make their case with the express aim of improving the discipline of IR (Anderl and Witt, 2020: 26;Smith, 2020; see also Blaney and Tickner, 2017). Others seek to uncover a 'pluriverse' of perspectives, but the purpose is to expand the scope of 'the international', contesting its colonial roots (Rojas, 2016;Vitalis, 2015) or its current Judeo-Christian cosmology (Shani and Behera, 2021).
Such efforts to change IR are typically in opposition to the dominance of realist statecentrism and rationalist approaches to IR, but in their grammar they echo how realists themselves earlier thought about disciplinarity. As we saw, the latter regarded theory as the key to warding off an unwanted positivist episteme. The protagonists at the Realism conference considered themselves to be representatives of IR and viewed behaviouralists outside the discipline as the main threat (Waever, 2011: 111). Participants noted with concern that the journal World Politics had been taken over by behaviouralists and considered theory the bulwark for the independence of IR as a discipline. This was considered crucial as IR faced transdisciplinary positivism and 'social science' paradigms: 'unable to contrast the positivist trends transforming the discipline [political science] these scholars settled for controlling a smaller but independent disciplinary territory . . . making it immune to the cues of behaviouralism' (Guilhot, 2008: 282).
This largely failed, at least in the United States. Waltz's post-positivist position set out in Chapter 1 of Theory of International Politics was subsumed by the sheer force of the positivist-behaviouralist ascendancy in Political Science (Waever, 2009), and IR never left Political Science in the United States. With that, 'disciplinarity' itself came to be associated with 'science' rather than theory (Hoffmann, 1977). Michael Brecher (1999: 213) in his ISA presidential address argued 'the field' had not yet 'crystallized into a mature social science discipline' due to a 'retreat from science in international studies'. But if theorising could be overcome and scientific empiricism entrenched, 'a genuine social science discipline' beckoned.
Today, some critics of current IR content also seem to take the mainstream notion of 'disciplinarity as scientism' at face value (Jackson, 2017). In this vein, Roland Bleiker (1997) pointed to King, Keohane and Verba's positivist methods manifesto as an example of why disciplinary power needs to be subverted in IR (p. 163). But in doing so, he perhaps missed that such positivism emanates from a cross-disciplinary social science movement and Bleiker (1997) himself leans on IR theory from the likes of Carol Cohn, Richard Ashley (p. 67) and Cynthia Weber (p. 78) for counterpoints -that is, resistance to positivism again comes from (theorists) within the discipline against positivists outside it.

Societal multiplicity reassessed
Rosenberg's attempt to reground IR on more independent terms in terms of 'societal multiplicity' is unambiguously a disciplinary move (Rosenberg and Tallis, this issue). It falls clearly in our 'break out' category (even more than 'Global IR' which has been content to stay politics-focused) and amounts to something of a classic disciplinary 'jailbreak', theorising a subject and problem area distinct from a 'host' discipline. Societal multiplicity, Rosenberg argues, provides a specific angle on human life with general applicability, since all social phenomena are affected by the coexistence of multiple and uneven societies. Just as all social situations have a history, geography, economics and so on, so do they all have an international dimension. IR is (or should be) the discipline that theorises and studies that dimension, which -crucially -is not limited to just politics, and not only found beyond the state.
The international in this sense is significantly broader than state relations or even global politics: 'where societal multiplicity obtains, its significance is not restricted to politics and relations of power. It extends into the social, economic, cultural and developmental dimensions too' (Rosenberg, 2016: 136). This means that the international becomes identifiable in much more diverse domains and registers than just politics or security -something potentially appealing to many critics (Guillaume and Huysmans, 2018;Shepherd, 2017). In relation to decolonising IR, multiplicity has even been argued to counteract what Sabaratnam calls out as 'epistemological immanence' in IR, by taking as foundational 'an interactive and pluriversal conception of totality that critically incorporates the problematic of historical difference central to Post-/Decolonial Theories' (Matin, 2022: 320).
Those keen to import from Sociology, Anthropology or STS (and so on) into IR fear that 'the consequences of societal multiplicity' as a basis for disciplinarity would limit this by requiring agreement about what those consequences are. Yet this buys into the idea of disciplines being based on substantive agreement, which seems not to apply to any disciplines, and even suggests that IR must somehow stop engaging in interdisciplinary work if it does become a discipline proper. Since interdisciplinarity relies on each discipline being able to identify its unique contribution to collaborative work, and relies on ingesting concepts and ideas from other disciplines without losing sight of this contribution, this is strange. Clarifying the unique contribution of IR and doing so in a way that gives it potential import in any setting could hugely strengthen interdisciplinarity. Realism limits itself to the state-related security consequences of societal multiplicity and is as such only a very partial theory of societal multiplicity. Viewing realism this way relegates it from 'leading' IR theory to a provincial and partial voice in a wider discipline dedicated to exploring the many consequences of societal multiplicity. Rosenberg thus arguably offers a key part of what many in the heterodox camp want: a less restrictive object of study as well as greater interdisciplinary potential.
A good test of multiplicty as an aide to interdisciplinarity is to ask how importing a theory of disciplinarity into IR from Sociology of Knowledge (as done in section 'The reluctant discipline') might take proper account of the international. With a conventional politics-focused object, IR might have little to say on the theory of disciplines. For its part, Foucault's theorisation of disciplinarity is quite internalist. It does not explicitly address the role of the international in how disciplines are formed or how they work. Yet it is hardly a coincidence that modern academic disciplines emerged as the modern international system spread unevenly across the globe. National contexts and transnational flows have been significant for different developments in IR (Waever 1998) and in intellectual history more widely (Baring, 2016: 571). In relation to natural sciences too, the international is usually considered irrelevant to 'science', assumed, internalistically, to be universal and objective. However, recognition of 'epistemic geography' (Mahony and Hulme, 2018) or a 'geopolitical approach to transdisciplinary science' is beginning to explore how scientific knowledge originates and circulates via the unevenness and inequalities of the world (Meehan et al., 2018). Another emerging field studies how theories and paradigms 'evolve differently in different contexts -in different disciplines and countries' showing how 'the social and human sciences do not form a unified global social field' (Sapiro et al., 2020).
Thus, there is a double dialectic to grasp: interactions between different uneven societies simultaneously affects how multiple disciplines interact and develop through such interaction. Like all disciplines, IR is itself heavily marked by (and reproduces) the unevenness of the international. Until recently this was ignored despite -or because -'[t]he modern discipline of IR was developed precisely when the West had a whip hand over everyone else', leading to extreme Eurocentrism in IR (Buzan and Acharya, 2021: 4). Histories of the birth of IR are beginning to recognise that IR had multiple births, not just in Aberystwyth in 1919 but in many different national, colonial and post-colonial contexts (Thakur and Smith, 2021). This is progress, although whatever its eventual content 'the ways IR travels through the ordered paths of international hierarchy' (Hamati-Ataya, 2016: 339) may ultimately limit that progress in diversifying and equalising the discipline.
It would therefore be problematic for his own theory if Rosenberg's proposal for a core subject matter for IR were theorised as a transhistorical 'deep ontology' that had been hiding 'in plain sight' (Rosenberg, 2019: 147) -as a brute fact with effects flowing from it that IR simply has to finally pick up and examine. If that were the case, then the international itself would be inconsequential for IR's subject matter which would be a transhistorical given.
In contrast, for historical ontologists, there would be no fact of societal multiplicity without an (international) history of its own. Historical ontology, involves, as Jens Bartelson (2019: 109) summarises, a commitment to dynamic nominalism, a doctrine according to which the ways in which people name and classify themselves and the things around them will interact with the resulting categories across time and space and prompt forever new conceptualisations of things and people.
The subject matters of disciplines are written, practised and reified into being through knowledge-practices, institutionalisation and identity-work, nested within wider knowledge-complexes, which help constitute the societal entities and their inter-relations -which then form the basis of more writing and reification. Responding to the critique that 'societal' implies, in effect, 'states', Rosenberg and Tallis in this issue (section 'Introduction') instead deliberately leave open the specific historical form that 'society' or 'the inter-societal' takes. But to that it must be added that the analytical lenses used to identify and construct such entities are also historically contingent. For this reason, Rosenberg's 'societal multiplicity' would perhaps be better considered a 'shallow ontology' -or perhaps a 'deep epistemology' -rather than a 'deep ontology' for IR.

Conclusion
Disciplines raise instinctive 'red flags' because they inevitably involve policing of scholarship -something many IR scholars rightly critique, especially those pushing for broadly oriented, pluralist and de-colonised IR scholarship. Therefore, 'an obvious temptation arising from complaints about the limits of a discipline is to call for its erasure' (Bigo and Walker, 2007: 730). Some of the strongest objections to IR taking 'the international' as its general but unique dimension of the social world have therefore come from those challenging the underlying assumption that IR should aspire to becoming a more independent discipline.
However, in this article, I have argued that the opposition between disciplinarity and critical pluralist scholarship is false. That would assume that disciplines are purely limiting, or that beyond disciplinarity lies unfettered epistemological sunlit uplands. In contrast, disciplines provide one kind of basis for local truth-claims, tools for interdisciplinarity and communities of knowledge-production or 'scholarly selves' that potentially insulate against transdisciplinary epistemes like positivism. Disciplinary choices should recognise that just as individual societies are marked by their coexistence with multiple other different societies, individual disciplines exist within in a multiplicity of disciplines and are not shaped purely by dynamics internal to each discipline.
Broadly speaking, IR is currently somewhere between a discipline and a subfield with some of the institutional trappings of a discipline but a subject matter derived from another one. Mainstream IR is hemmed in by rationalist axioms and an object located within Political Science (and methods from Economics) while critical IR, pursuing a wider notion of politics but a vague sense of the international, is split between 'stay put' and 'burn down' strategies. This is in many ways the worst of all worlds. A sub-field has by definition a restricted subject matter and is perennially at risk of being swallowed up by the host discipline. Navigating interdisciplinary engagements without an anchor as a subfield is bound to lead to one-sided traffic and ultimately to less disciplinary diversity. IR has imported a succession of ontologies from other disciplines -perhaps because it lacked clarity about its own. Outside disciplines altogether, a free-floating 'field' of IR would be a vulnerable meeting place for other disciplines and epistemes, most of which are congenitally oblivious to 'the international' as a distinct problem or only haphazardly take it into account.
While there are no easy options, as the contours of a different and more global IR are clearly emerging (Krishna, 2021) an opportunity does exist to reground IR as a discipline trained on more than politics without a state. Mainstream IR appears unable to do this, focused on a narrow object and standards of scientific enquiry defined by other disciplines. Critical IR could, but hesitates because it defines itself largely against mainstream IR -a negative identity, reliant on a field that itself relies on a negatively defined object. What's the point in that?

Funding
The author(s) received no financial support for the research, authorship and/or publication of this article.