Does it make sense to theorise international relations (or global politics) in terms of ‘practices’ and, if so, what does it mean to speak of ‘practices’, i.e. how would one want to conceptualise ‘practices’ and approach studying them? These questions lie at the core of Silviya Lechner’s and Mervyn Frost’s book Practice Theory and International Relations (henceforth referred to as PT). It is a most welcome contribution to an intensifying and broadening debate about ‘praxis’, ‘practice’ and ‘practices’ in the social sciences more generally and in IR specifically.^{Footnote 3} It is welcome from this reader’s angle in particular because Lechner and Frost actually engage key authors of praxis, such as Ludwig Wittgenstein, who, in IR, have often only been referenced in passing even though their central status in any serious discussion of praxis and social practices has been acknowledged early on (and often, if equally briefly, ever since).^{Footnote 4} Practice Theory and International Relations is an important contribution to IR debates because it more systematically develops what the authors call an ‘internalist’ perspective on studying practices via a ‘descriptivist’ approach.

While I am very sympathetic vis-à-vis (and was initially even enthusiastic about) the emphasis on Wittgenstein and a form of social science which highlights ‘internalism’ (in contrast to ‘externalism’, PT 27–8) and ‘descriptivism’ (in contrast to prescriptivism, PT 10), I have been increasingly struggling with the actual interpretation (or usage) of all three in the course of the authors’ development of their argument. ‘Struggling’ here means that I found it increasingly difficult to understand and accept the vocabulary Lechner and Frost are using (or introducing) in their book in order to theorise practices anew. Two dimensions of this increasing struggle are worth highlighting.

First, my own reading of Wittgenstein in general, and of his take at praxis ^{Footnote 5} in particular, leads me to think quite differently of praxis and ‘practices’ – and I would also claim that my Wittgensteinian alternative conceptualisation is closer to ordinary uses of the terms in both ordinary language and established scholarly language games about practice and practices in IR. To be sure: ‘ordinary’ uses are not necessarily superior uses – and scholarship is often about teaching/training and enticing us to acquire a new vocabulary and speak differently in order to see (and act in) the world differently (Rorty Reference Rorty1989: 5–7). However, with Wittgenstein I would claim that there might be good reasons why particular habits of speaking about ‘practice’ and ‘practices’ have sunk in which we should, at a minimum, acknowledge and respect, and which lead us not to use them synonymously with the concept of ‘institution(s)’, as the authors do.^{Footnote 6} One reason not to change established language games hastily is that new ways of speaking about ‘practice’ and ‘practices’ should connect with old ways, either in order to entice speakers to adopt the new way of speaking via Kuhnian ‘language learning’ or to at least encourage dialogue.^{Footnote 7} Moreover, to change a language game implies that changes in one concept reverberate with a supporting net of additional concepts. This is what Donald Davidson calls the ‘holism of the mental’ – meaning that to have a concept (say, of ‘practice(s)’) implies that one has to have many other related concepts (Davidson Reference Davidson2001: 123–7 and Davidson Reference Davidson2004: 3–18 and 135–50). Therefore, in order to be meaningful (i.e. resonate with potential interpreters of our way of speaking) a concept ought to fit within established ways of speaking or be refitted with novel ways of actually acquiring a ‘new vocabulary’.^{Footnote 8}

Lechner and Frost are quite aware of these requirements since their whole book can be read as an invitation to switch to such a ‘new vocabulary’. However, I am not convinced (yet), whether potential gains in switching to the new vocabulary (or ‘theory’) on offer (i.e. the focus on macro-practices as institutions) will actually compensate for major losses which I see mainly in a fairly rigid and unnecessary denigration of other ways of speaking about (or theorising) ‘praxis’, ‘practice’ or ‘practices’ which link more easily with ordinary and established scholarly use.^{Footnote 9} More specifically I cannot yet see the advantage of eliminating the close link between (of course: a much fuller, i.e. pragmatist, notion of) ‘action’ on the one hand and practice(s) on the other because this severs (or, at least, significantly weakens) the critical link between agency and social responsibility.

In response to such a criticism of their position Lechner and Frost might refer to sentences like the following:

From my Wittgensteinian angle it sounds quite odd (or extra-ordinary), however, to conceptualise ‘the society of states’ in ‘agency’ terms and associate ‘responsibilities’ with it, especially when ‘deliberation’ and ‘intentionality’ (but not necessarily purposiveness) in the sense of pursuing a plan (‘do something’) are explicitly said to be associated with all ‘common practices’ (PT 202). In my ears this sounds like a ‘nonsensical combination of words’ – to use a Wittgensteinian formula (Wittgenstein Reference Wittgenstein, Anscombe, Hacker and Schulte2009 (1953) PI § 512) – because the practice(s) vocabularies I am familiar with^{Footnote 10} would not combine the equivalent of an ‘institution’ (such as the ‘society of states’) with deliberateness or intentionality as suggested by Lechner and Frost. Now, the full quote from Wittgenstein just referred to actually allows for ‘nonsensical combinations of words’ (in contrast to ‘nonsensical imagining’)^{Footnote 11} in the same way as the second introductory Wittgenstein quote above rejects the view that ‘certain concepts are absolutely the correct ones’. In other words, Lechner and Frost certainly ought to be listened to carefully when they suggest an unfamiliar new way of theorising ‘practices’. However, it ought to become clear (or at least clearer as it is for me at this point) what internationally relevant practices we may grasp better if we start to theorise what we have thus far called ‘institutions’ as ‘practices’ henceforth, especially when the introduction of this new vocabulary is accompanied with an almost wholesale rejection of alternative ways of conceptualising ‘practice(s)’.

Second, I am struggling with the interpretation of Wittgenstein, internalism and descriptivism offered because it combines very different conceptual vocabularies. This relates in particular to Wittgenstein, Oakeshott and Hegel which I am not equally familiar with. However, I find it difficult to combine these conceptual vocabularies as suggested in PT based on my (partly limited) understanding of the work of all three referenced authors. For instance, in my reading of Wittgenstein, especially as far as his insistence is concerned that we should not waste too much time on definitions or try to legislate particular ways of speaking when we observe ‘praxis’ and practices, certain ways of writing about ‘misconceived’ (or ‘proper’) notions of practice in PT^{Footnote 12} strike me as utterly un-Wittgensteinian – besides being not particularly constructive in advancing dialogue about possible different and equally meaningful (or at least sensible) ways of theorising practice(s).

In particular, I find it very hard to agree with Lechner’s and Frost’s claimed ‘agree(ment) with Wittgenstein and Oakeshott that the proper task of the theorist qua theorist is to elucidate the practices under investigation by providing coherent and accurate descriptions’ (PT 30, italics added). This claim strikes me in many ways as the opposite of what Wittgenstein was arguing for. Whereas Lechner and Frost seem to be largely in agreement with Wittgenstein’s scathing criticism of a type of ‘craving for generality’ (Wittgenstein Reference Wittgenstein1958: 16–20) he associated with what we have come to call the ‘social sciences’, their understanding of ‘theorising’ (or the synonymously used word ‘philosophising’) as ‘a search for conceptual synthesis over and above a taxonomy of practices’ is at odds with the later Wittgenstein’s criticism that strange ‘philosophical problems’ only arise ‘when language goes on holiday’.^{Footnote 13} In the reading of most Wittgenstein scholars of his late work^{Footnote 14} this view marks a clear distancing from the Tractatus phase’s emphasis on ‘logic’ and ‘definitions’, i.e. the early Wittgenstein Lechner and Frost rely on when they approvingly quote his take at philosophy and ‘theory’ as aiming ‘at the logical clarification of thoughts’ (PT 4). For good reasons the late Wittgenstein was quite critical of his early views and most likely would have characterised Lechner’s and Frost’s depiction of Oakeshott´s understanding of theorising as being ‘directed at the understanding of a not-yet-understood identity’ (PT 4) as a sentence ‘produc(ing) in us mental cramp’ because a ‘substantive’ (identity) makes us look for a ‘substance’ or ‘a thing that corresponds to it’.^{Footnote 15}

Moreover, Lechner’s and Frost’s claim that Wittgenstein urges us to provide ‘coherent and accurate descriptions’ insinuates a correspondence theory of truth which (at least) the late Wittgenstein was fundamentally opposed to. He repeatedly criticised (and even mocked) the fixation of much of philosophy on ‘clarity’ or ‘exactness’ (even ‘ideal exactness’ (Wittgenstein Reference Wittgenstein, Anscombe, Hacker and Schulte2009 (1953) PI § 88).^{Footnote 16} In this sense he probably would also have been highly critical of the huge effort Lechner and Frost invest in developing their ‘theory’ with extended definitional elaborations and a critique of alternative ways of theorising or conceptualising practice(s) as ‘defective’, ‘too crude’ or ‘misconstrued’ (e.g. PT 63, 82, 91). Rather than legislating new vocabularies the later Wittgenstein’s ‘descriptivism’ boils down to a strong plea for the description of ‘language games’ (or ‘forms of life’) as ‘proto-phenomena’.^{Footnote 17} This is particularly so when we think about what it may mean to ‘describe’ (or ‘theorise’) praxis and practice(s). Here Wittgenstein’s ‘anthropological’ observations (as manifested in his critique of Frazer’s ‘The Golden Bough’ (Wittgenstein and Rhees 1967)) – are relevant because the analogy between observing ‘strange’ (foreign) cultures can productively be compared in analogous fashion with the observation and description of ‘international practices’.^{Footnote 18} In other words, his descriptivism is a plea for the description of descriptions ^{Footnote 19} which has no place for notions of ‘true’ or ‘truth’ in combination with ‘description’ (e.g. PT 24, 43–4) and which contrasts starkly with Lechner’s and Frost’s argument in favour of ‘normative descriptivism’ (PT 184–5).

Does such a critique really matter? Why should IR scholars care about very different readings of a dead philosopher when one wants to theorise (international) praxis and practices anew? My argument is that (the late) Wittgenstein is indeed an extremely helpful source to theorise (international) praxis and practices anew because he sensitises us for observing praxis carefully and self-critically and for rejecting the ‘pneumatic conception of thinking’ which somehow associates something more ‘concrete’ or ‘substantive’ ‘behind’ our linguistic signs. This is also why Wittgenstein insisted ‘that our considerations must not be scientific ones’ and that

Wittgenstein finds ‘theories’ (as they are practised in the natural sciences) absolutely misplaced in understanding praxis and practices because the very equation of social action and natural facts fails to grasp that language as our key tool for relating to and making sense of ‘the world’ is not ‘picturing’, but constituting ‘reality’.^{Footnote 21} Because this is so we are, in understanding what motivates social action, indeed dependent on practising a form of ‘internalism’ with regard to what (or whom) we conceptualise as actor or agent (in contrast to ‘externalism’). Practising social science has to take into account, and reflect the fundamental difference between, praxis which indeed makes up ‘a meaningful whole’ (PT 15) to the agent here and now on the one hand, and an observer, on the other hand, who tries to make sense of such a meaningful whole via some form of description or explanation which, by necessity, has to reconstruct this meaningful whole ‘from the outside’. This does not mean that there is no difference between what Lechner and Frost, correctly in my view, call the significant difference between externalist and internalist accounts of praxis. Yet, in my reading of Wittgenstein the type of ‘internalist’ account he values would be one which looks at phenomena as ‘proto-phenomena’ – i.e. that he would precisely not claim that ‘we are searching not just for any set of descriptions but for appropriate ones’ (PT 24, emphasis added) because in his view we lack any measure which might enable us to establish ‘appropriateness’. After all, in encountering fellow human beings we ought to approach them with ‘an ethnological way of looking’ rather than assimilating an observed praxis of others to practices we are used to.^{Footnote 22}

In this sense a ‘non-pneumatic’ conception of thinking lets us look at our uses of language as very individual ways of coping in the world. Language (or, for that matter, the vocabularies or theories we use) don’t represent anything ‘out there’ in the world (cf. Rorty Reference Rorty1992 (1967): 373 and Rorty Reference Rorty1991: 1 on ‘anti-representationalism’). Rather it enables us in our daily praxis to cope with the world. This world does not ‘speak’ to us or suggest a vocabulary it wants to be described in (e.g. a ‘society of states’ or ‘global rights’, cf. Rorty Reference Rorty1989). Rather, we as humans are inventing such (necessarily and essentially contested) descriptions about this world and the praxis and practice(s) we observe. Whether these descriptions are ‘true’ or not does not matter, because ‘truth’ is a nonsensical concept combined with ‘description’. What matters is whether these descriptions resonate and/ or are acknowledged by others in terms of their willingness to act (pragmatism’s maxim that ‘beliefs are rules for action’). In this sense ‘knowledge’ is, in the end, always ‘based on acknowledgement’ (Wittgenstein Reference Wittgenstein, Anscombe, Wright, Paul and Anscombe1969: § 378). This also means that rule-following in a (late) Wittgensteinian understanding is praxis, not ‘practice’ (‘Praktik’ or ‘Handlunsweise’)^{Footnote 23} in the insinuated sense of patterned or habitual action. It is precisely because there is indeed an ‘endless variety of social practices’ (PT 23) why any ‘theory’ that seems to claim (almost) all-encompassing validity should be very careful to limit the notion of ‘practice(s)’ to a specific set of practices as institutions.

Describing international praxis, then, would focus on both agency/action and interaction – i.e. practices such as waging war; conducting foreign policy; negotiating, signing and ratifying treaties; or deterring with nuclear weapons. However, it would obviously also focus on ‘single cases’ – i.e. the praxis of the current Syrian war and Iranian, Israeli or Russian strategising; the conduct of German foreign policy at a time when the EU is under increasing stress; the negotiation, signing and ratification of BREXIT; or the revision of NATO’s deterrent posture in the light of the collapse of the INF treaty – just to name a few contemporary problems of international praxis we ought to be concerned with. In other words, rather than inflating the ‘generalising’ aspect of practice theory, Wittgensteinian descriptivism would be open to the observation (and/or linguistic invention) of ‘patterned’ practice(s) as much as it would appreciate praxis ‘proto-phenomena’. Such an understanding of international praxis theory would be built around what we ordinarily conceptualise as ‘agency’ based on ‘intentionality’ and ‘purposiveness’. However, given that international praxis is social action (different from the ‘simple actions’ or ‘doings’ such as walking etc. which Lechner and Frost are distinguishing) this understanding would also appreciate the dimensions of situatedness and interaction of praxis. Nothing more is needed in order to understand international praxis – and all of that can also be realised in terms of sense making via social inquiry.