Measurement = generative: bridging IR’s epistemological divides with quantum mechanics’ measurement apparatus

Abstract

This article reflects on possibilities for bridging International Relations’ (IR) epistemological divides, through recourse to concepts from quantum mechanics. The article draws on quantum mechanics’ measurement problem to address polarisation between IR’s realist/positivist and interpretivist/critical ‘camps’. I argue this polarisation poses an obstacle to unified conversation between IR’s disciplinary traditions. I note epistemological divergences which underlie the polarisation, namely, different attitudes to the relationship between ‘measurement’ and ‘reality’, across realist/positivist and interpretivist/critical traditions. And I consider possibilities for overcoming this divergence, through attention to quantum mechanics’ foundational measurement problem. Engaging with writings from quantum mechanics’ founding parents, Niels Bohr and Werner Heisenberg, I position the measurement problem, and its central vehicle the ‘apparatus’, as a basis for more unified exchange in IR. This exchange would start from an understanding of measurement and reality, not as binary or hierarchical (as in IR’s present epistemological divide), but rather as phenomenally intertwined. That is, as inseparable features of a single phenomenon, taking shape from within the particular configurations of IR’s generative apparatus.

Introduction

It is nearly two decades since Christine Sylvester (2007) described International Relations’ (IR) dispersion into diffuse ‘camps’. Writing at the ostensible ‘end of IR’, Sylvester deployed the campfires metaphor to frame the discipline as having ‘lost the common ground that would enable all its practitioners to participate in a shared conversation’ (Kurki and Rosenberg, 2020: 397). Lamenting the dissipation of IR’s paradigmatic debates, Sylvester (2007) found ‘we [in IR] agree on very little’ (pp. 562–563) – with a previously consolidated discipline having been ‘split into a growing number of separate campfires, each hosting its own introspected conversation, and no longer making up a single discussion’ (Kurki and Rosenberg, 2020: 397).

This article invokes Sylvester’s metaphor of ‘camps’ as a departure point to explore, and assess possibilities for reunifying, one of IR’s lost conversations. This is the conversation on IR epistemology, which I argue is the archetype of Sylvester’s campfire model. Sylvester made her intervention on IR ‘camps’ with the purpose of demonstrating the dissipation of binary paradigmatic debates. Sylvester (2007) felt these paradigmatic debates had exercised a positive centripetal force upon the discipline, drawing scholars into direct conversation around ‘a couple of positions [or] attempts at camp syntheses’ (p. 563). Such conversations included what Sylvester (2007) called IR’s ‘third debate’: that which emerged ‘in the 1980s and 1990s pitt[ing] interpretive or reflectivist epistemologies against positivism’ (p. 553). Sylvester argued that, by the mid-2000s, this third debate had disintegrated into a cosmology of different positions, with no unifying centre to drive transdisciplinary engagement.

My take on the third debate is different to Sylvester’s, although I think her metaphor of ‘camps’ remains useful for framing the problem. For I argue the divide Sylvester identified between positivists and interpretivists remains largely operational in our field, albeit less as a source of engaged debate than as a void of silence between increasingly distinct epistemic communities. Other reviews in the years since Sylvester’s article have argued the same thing. Contributors to a 2013 special issue of this journal on ‘the end of IR theory’, for example, concluded IR was still ‘shaped’, in important ways (Dunne et al., 2013: 406), by a ‘real and emerging divide. . . between positivists and post-positivists’ (Lake, 2013: 577–578); while more recent sociological network analyses of the field by Peter Marcus Kristensen (2016, 2018) and Stephen Aris (2021) found scholars in Sylvester’s post-paradigmatic period ‘continue to distinguish themselves from one another on the basis of distinctive macro-epistemological commitments’ (p. 178). Twenty years on from Sylvester, IR remains divided along epistemological lines. Where Sylvester’s campfires metaphor continues to resonate, as a description of today’s discipline, is in its image of a scholarly community suffering limited interaction between groups. IR’s polarisation between positivist and interpretivist epistemologies has become a true case of isolated campfires: no raging debate on mutual interests, but rather an acceptance of each side’s essential differences, and a retreat into sub-disciplinary conversations – with neither camp showing strong interest in the other’s discussions.

Such is the problem to which the present article is addressed. Drawing on sociological reviews of the field, the article reflects on persistent (and growing) distance between IR’s realist/positivist and interpretivist/critical traditions. This distance has seen shrinking levels of transdisciplinary interaction. Evidenced by minimal returns for cross-disciplinary citation practice, the shrink exemplifies Sylvester’s fears of a retreat from a consolidated conversation: involving the crystallisation of camps along the third debate’s binary epistemological lines, rather than that debate’s disappearance – and the dispersal of knowledge exchange into starkly separate spheres. The article considers underlying reasons for IR’s epistemological divide. Above all, these include foundational differences between realists/positivists and interpretivists/critical scholars regarding the relationship between ‘measurement’ and ‘reality’. Being epistemological in nature (having to do with first principles of research), these differences make it difficult to curate a shared conversation in the discipline – with scholars from realist/positivist and interpretivist/critical traditions flying at fundamentally different altitudes in researching the international, thus limiting the scope for cross-pollination of knowledges.

The article responds to this problem by exploring possibilities for bringing IR into more unified conversation through recourse to concepts from quantum mechanics. Specifically, the article considers lessons on the relationship between measurement and reality, as contained in quantum mechanics’ measurement problem. These lessons include measurement’s generative relation to reality: bringing reality into being according to particular material and conceptual configurations, as defined by the parameters of an applied measurement apparatus. Such lessons can open doors to a more unified conversation in IR, grounded in a shared understanding of what it means to research (or ‘measure’/‘generate’) the international. The article draws this understanding through a ‘diffractive’ (Barad, 2007: 66–67) reading of texts from quantum mechanics’ founding parents, Niels Bohr and Werner Heisenberg. The article explores Bohr and Heisenberg’s efforts to theorise the ‘apparatus’, as the central vehicle for thinking about measurement’s generative effects for reality. Though rarely treated in IR (including quantum IR), I believe this ‘apparatus’ can serve as a basis for a more unified conversation between IR’s epistemological camps. Rather than understanding measurement and reality as distinct or hierarchical (as in IR’s present epistemological divide), this conversation would start by approaching each as phenomenally intertwined. That is, as indissoluble parts of a single phenomenon, which takes its shape through interaction with the particular configurations of IR’s material-conceptual apparatus.

What are the stakes of this intervention? Positioning the apparatus as a basis for a more unified conversation across IR has important implications. First, it can help move IR’s epistemological camps into closer communication – no longer separated at birth through divergent epistemic commitments, but rather connected through a shared conceptualisation of the relationship between measurement and reality (this conceptualisation being itself grounded in ‘revolutionary’ insights of 20th century physical science, Barad, 2007: 106). Second, drawing quantum’s measurement problem into IR can get our discipline’s diffuse epistemological camps into asking questions of renewed mutual interest. For example: if measurement is generative of conditions for reality, as defined by the particular configurations of the applied apparatus and as revealed in physical quantum theory – then what is the nature of the apparatus which IR applies to measure world politics? And what kinds of realities might this apparatus’ application generate? These are questions in whose exploration IR’s realist/positivist and interpretivist/critical scholarships have equal stakes. The necessary start-point for asking them, however, is to lay for these divided scholarships some common epistemological ground. I find that common ground in my readings on quantum mechanics’ measurement apparatus.

The article is structured as follows. I begin by clarifying IR’s divide between realist/positivist and interpretivist/critical epistemologies. I note sources of this divide (a foundational divergence on first principles of IR scholarship) and its effects (inhibiting consolidated conversation across the discipline). I précis the recent emergence of interest in ‘quantised IR’ (Der Derian and Wendt, 2022). And I flag possibilities for more unified cross-disciplinary conversation which engagement with quantum concepts might afford (most centrally, concepts of measurement and the apparatus which have yet to make headway in IR). I engage in a reading of texts from quantum mechanics’ Copenhagen interpretation to explore quantum’s measurement problem and position its apparatus as a basis for bridging IR’s epistemological divide. I end with thoughts on the types of questions IR might be empowered to ask, were it to absorb lessons of quantum mechanics’ measurement problem more fully.

Before I begin, two caveats. The first concerns the sources from which I develop my argument. My argument is grounded in a close reading of texts by quantum mechanics’ ‘primary authors’, Bohr and Heisenberg (Barad, 1997: 167). My conclusions are limited entirely to what these authors wrote about quantum mechanics, the measurement problem or the apparatus. There are strong grounds for accepting these conclusions, given their roots in texts authored by the progenitors of a theory now widely accepted as the basis for understanding universal physics. Nonetheless, these texts represent only one interpretation of quantum mechanics’ experimental findings. Such findings have precipitated a ‘plethora’ of interpretations (French, 2023: 2), engendering as many quantum ‘schools and cosmologies as there are religious sects’ (Der Derian and Wendt, 2020: 4). Among this plethora, I draw my argument from literatures of the Copenhagen interpretation, specifically. This is a logical choice, given Copenhagen’s status as the ‘dominant’ source of inspiration for quantum IR (Der Derian, 2022: 921). But it remains important to note that other interpretations might have led me in different directions – for example, the many-worlds interpretation advanced by Hugh Everett (1957), whose argument on the universal wavefunction carries altogether different implications for quantum’s measurement problem. That being the case, my second caveat is to frame my interventions on measurement and the apparatus, throughout, as a question, rather than an answer. My reflections on quantum are to the effect of asking could we think this way? about measurement problem and apparatus in IR, rather than prescribing that we necessarily must. There are grounds for, and benefits to, adopting the conceptualisations advocated here. But there are also, and always, grounds to think ‘otherwise’ (Khatibi, 2019).

The present situation in IR

Sylvester deployed the campfires metaphor to describe the dissipation of IR’s centripetal debates into increasingly diffuse camps, including the binary ‘third debate’ between realist/positivist and interpretivist/critical epistemologies. Yet, sociological reviews of the field find IR remains a discipline divided along these binary epistemological lines. Where Sylvester’s campfires metaphor resonates most in today’s IR is less in its prediction of a proliferating cosmology of epistemological positions – and more in its depiction of a discipline enjoying minimal interaction between essentially separate sub-disciplinary conversations. That is, a discipline lacking cross-pollination of findings, concepts, and knowledges between traditions founded upon fundamentally divergent epistemological premises.

IR’s epistemological divide

This division of IR into separate epistemological camps is affirmed by recent network and citation analyses of the field. I want to cite two main studies, in this respect: those conducted by Kristensen (2016, in this journal, and 2018) and Aris (2021, also in this journal). Kristensen’s (2018) study deployed citation data from 106 IR journals indexed by the Web of Science to ‘dissect the structure of the discipline’ through network analysis (p. 246). On IR’s theoretical ‘isms’ (realism, liberalism, constructivism), Kristensen (2018) found these still manifested as separate communities in IR, but that they were also ‘more closely intertwined and cross-contaminated than [Sylvester’s “end of IR”] fragmentation thesis suggests’ (p. 245). Where Kristensen (2018) did find a disconnect within IR, was in the lack of cross-citation between two distinct epistemological camps – camps which Kristensen described as either ‘reflectivist’ (including critical and post-structuralist work) or ‘positivist’ (including ‘quantitative modelling [and] hypothesis testing’ approaches) (pp. 254–255). Kristensen (2018) found these epistemological camps exhibited minimal interaction, as defined by negligible co-citation of each other’s work, concluding that

a horizontal divide stretch[es] across the entire network. . . Authors at each end of the network certainly represent very different positions. . . but they are rarely used explicitly for position-taking. . . the relationship. . . is more a divide than a debate; it is a long way from text to test, social to formal theory, Foucault to Fearon. (p. 256)

In short, Kristensen’s analysis framed IR scholarship as split along binary epistemological lines, with scholars on each side of the divide showing minimal interest in (or perhaps even knowledge of) each other’s work. IR’s ‘third debate’ had indeed crystallised into distant campfire conversations.

Aris’ (2021) study affirmed this binary picture. Like Kristensen’s, the study employed bibliometric data to assess IR’s citation networks – this time, focussing on co-citation between scholars of IR and those in cognate disciplines, including economics, geography, history, law, and political science. Aris’ analysis deepened Kristensen’s picture of binary epistemological division within IR, by noting equally binary position-taking in IR’s interdisciplinary alliances. Where IR scholars drew inspiration from other disciplines, Aris (2021) found this was ‘largely in terms of interdisciplinary epistemological commitments along the positivist/interpretivist binary’ (pp. 190–191). When scholars from IR’s realist/positivist camp cited frameworks and findings from outside IR, for instance, these came predominantly from political science (whose ‘neopositivist and quantitative orthodoxy’ matched realists/positivists’ own epistemological positions). Meanwhile, when interpretivist/critical scholars engaged non-IR sources, these came from disciplines of sociology and history (which are more ‘strongly identified with a “critical” reflectivist research agenda’) (Aris, 2021: 191–195). In short, for Aris as for Kristensen, network analysis exposed a cleavage between two essentially distinct communities (realist/positivist and interpretivist/critical) – with each community drawing interdisciplinary (as much as intradisciplinary) inspiration from interaction with work that confirmed their own epistemological principles. While theoretical isms were now ‘a less relevant differentiator [between IR camps], different groups of IR scholarship continue[d] to distinguish themselves from one another on the basis of their distinctive macro-epistemological commitments’ (Aris, 2021: 178) – realist/positivist versus interpretivist/critical.

Sociological reviews confirm IR’s division into distinct epistemological camps. These camps exhibit limited interaction with each other – exemplifying the campfires metaphor of which Sylvester warned. They also tend to draw their sources of interdisciplinary inspiration along epistemological lines, structuring their engagement with cognate literatures through the prism of shared epistemological commitments. Of course, this division is not totalising. There remain sub-disciplines and authors in IR that straddle this epistemic divide. But, at an abstract level, these network analyses reveal IR to be an increasingly polarised field. The discipline suffers a divide between distinct epistemological conversations. As Ole Wæver put it in 2009, this divide, a leftover from IR’s unresolved ‘third debate’, may no longer be ‘lively and actively conducted’ (as in Sylvester’s prediction), but it remains ‘still the most relevant general map of the field. A polarised rationalism/reflectivism debate [has] mutated into an axis with more and more people located towards the middle, but still defining themselves in relation to this axis’ (Wæver, 2009: 217). Or, as Milja Kurki and Colin Wight (2021) summarised more recently, ‘rationalism and reflectivism have produced a dichotomous logic that has fashioned two wings of the discipline: a “proscience” viewpoint versus an “antiscience” position’ (p. 23).

What’s the matter? A discipline flying at different altitudes

Per language from the excerpts above, this epistemological divide has been described, variously, in terms of a binary between ‘positivist’, ‘rationalist’, ‘quantitative’, or ‘orthodox’ scholars (on one hand), and ‘interpretivist’, ‘reflectivist’, ‘constructivist’, or ‘critical’ peers (on the other). In this article, I describe the divide in terms of realist/positivist and interpretivist/critical traditions (this categorisation being based on the kinds of authors found on either side of Kristensen’s network map). I argue that the distance between camps is motivated by a fundamental divergence in their attitudes to the relationship between reality (the actual stuff ‘out there’ that IR aspires to analyse) and measurement (the methods, interpretive frameworks, and concepts by which this stuff comes to be known). Being a matter of first principles informing research, this divergence is hard to resolve. It leaves scholars from each tradition flying at fundamentally different altitudes, in their discussions of world politics. Such difference is a problem for IR because it constrains possibilities for the two conversations to meet. Essentially, the traditions find themselves talking across purposes, and thus enjoy minimal scope for a unified conversation on methods, empirical findings, or conceptual frameworks.

In IR, as Samuel Barkin (2003) characterised,

opinions tend to coalesce into one of two epistemologies: one argues that an identifiable reality exists out there and can be accessed through empirical research [i.e., measurement]; the other contends that we can never know for sure if what we observe really exists independently of our observation of it and, therefore, no true reality exists for empirical research to find. (pp. 326–327)

Realists/positivists fall into the first of these categories. ¹ Scholars of the realist/positivist camp acknowledge that ‘knowledge is never perfect’. Nevertheless, they still assume that ‘through the observance of agreed-upon research criteria, we can aim to make some justifiable judgements between competing knowledge claims’; and they continue to ‘strive for objective knowledge by attempting to define methods and criteria for knowledge production that minimise the influence of value-biased judgements’ (Kurki and Wight, 2021: 28). There are two things to say about these assumptions and commitments. On one hand, they maintain a distinction between reality and measurement – the former existing irrespective of the latter; and the latter being applied independently of the former, to generate knowledge of its qualities. On the other hand, they also maintain equipollence between each category – with ‘measurement’ being in no way subordinate to ‘reality’, since it can (at least in theory) be applied to obtain full knowledge of it.

Interpretivist/critical scholars operationalise a different attitude to the relationship between reality and measurement. Again, at its most basic level, this attitude treats reality and measurement as essentially distinct. But, unlike realism/positivism, it rejects the possibility of any meeting between the two categories – since the very act of applying measurement distorts reality, creating knowledges which tell us more about the situatedness of the observer than about that which was observed. Interpretivist/critical research ‘rejects the positivist approach that a researcher can be an independent observer of the social world’, arguing instead ‘that ideas, and even the particular identity, of a researcher influence what they observe and therefore have an impact upon what they conclude’ (McGlinchey et al., 2022: 87). Since ‘knowledge is by definition always “situated knowledge”; knowledge claims can never be formulated outside the influence of social and political context [meaning] claims about “reality” are always constructions of particular discursive and social systems’ (Kurki and Wight, 2021: 28). For interpretivist/critical epistemologies, in David Campbell’s (1998) words, the world may well exist independently of its measurement. But ‘we can never know that (beyond the fact of its assertion), because the existence of the world is literally inconceivable outside of language and our traditions of interpretation’ (p. 6). Reality and measurement are still separate categories here. But the situatedness of measurement prevents any really accurate knowledge of external realities.

In its most essential form, the difference between IR’s realist/positivist and interpretivist/critical camps boils down to a difference in their attitudes to the relationship between reality and measurement. Both camps envisage reality and measurement as distinct categories. They share a commitment to René Descartes’ (2008) mind-body dualism (pp. 51–64), both maintaining ‘an inherent/Cartesian distinction between the “object” [i.e. “external reality”] and the “agencies of observation” [i.e. “measurement”]’ (Barad, 1997: 170). But they diverge in their understandings of implications of this dualism for knowledge creation. Realists/positivists hold the mind can come to a perfect representation of the body, through scientific measurement. Conversely, interpretivist/critical traditions maintain a continuing distance between mind and body, with measurement’s entrapment in a situated mind preventing it from entering any unmediated knowledge of reality. In fact, as we’ll see, advances in our understandings of ‘reality’ associated with quantum mechanics reveal both positions as flawed. Quantum experimentation has shown the Cartesian split between mind and body/measurement and reality to be ‘illusory’ (Barad, 1997: 188). I’ll come on to this later, in discussing quantum mechanics’ measurement problem. For now, I’ll close this section by underlining the problem to which this article is addressed.

I’ve noted the continuing distance between IR’s realist/positivist and interpretivist/critical camps. This distance is a problem for IR. Above all, it constrains possibilities for transdisciplinary exchange (as borne out by diminishing returns for co-citation between camps in sociological studies quoted above). The result of such constraint is IR’s fragmentation into a conversation that is either disjointed (lacking a foundation in shared principles, a conversation about fundamentally different things), or non-existent (as in the third debate’s dispersal into separate camps showing no urge for shared discussion). Of course, some might embrace this fragmentation as a good thing for IR, empowering a greater diversity of positions and voices. There’s truth in this. But for the purposes of this article, I argue the absence of conversation risks making research agendas invisible to each other. Such mutual invisibility is a problem because it inhibits learning between sub-disciplines (this peer-based learning being the purpose of publishing research in the first place). How many times have we heard critical scholars lament, in this respect, that orthodox colleagues have simply not taken their interventions on board (Austin et al., 2019) – continuing to operationalise concepts like ‘security’ or ‘terrorism’ as stable entities, for example, without acknowledging 20 years of critical research identifying these concepts’ inherently discursive/contested/political natures (Livesey, 2024: 907)? This lack of recognition weakens research on either side of the divide. And it is a product of the distance between epistemological camps.

That the divide is grounded in epistemology, rather than in conceptual differences or divergence on empirical findings, is also significant. As Steve Smith (2021) has noted, IR’s ‘central differences between rationalist and reflectivist accounts are epistemological and methodological [in nature], and only secondarily about what the world is like’ (p. 5). This makes the dislocation harder to overcome. First, because, unlike empirical findings or conceptual frameworks, epistemological positions are not subject to mutual comparison or falsification. Second (and more importantly), because, being a matter of first principles, these positions leave IR’s realist/positivist and interpretivist/critical scholars flying at fundamentally different altitudes, in researching and discussing the international. With the two conversations proceeding from essentially different notions of what it means to ‘do’ IR, there is limited scope for them to meet and exchange perspectives once the IR is ‘done’. That’s precisely why Sylvester’s ‘third debate’ remains operational in the period she christened ‘IR2’. This debate could not be brought to resolution or synthesis, because its divergence was a matter of principle – leaving scholars with little choice but to retreat into sub-disciplinary cliques that did at least speak the same epistemological language. To obtain mutual visibility between camps, therefore, it would be necessary to forge a new epistemological ground, which realist/positivist and interpretivist/critical traditions might share – as a first step towards a more unified conversation across IR.

Quantum IR

In a 2013 special issue of this journal on ‘the end of IR theory?’, David Lake (2013) lamented he was ‘loathe to see international relations split into two separate disciplines sharing the same subject matter and intellectual history but divided by epistemology’ (p. 579). Per the review above, this division is where IR presently finds itself. Such division is a problem for IR, limiting possibilities for transdisciplinary (Kristensen, 2018) as well as interdisciplinary (Aris, 2021) cross-pollination. This article asks: what are possibilities for bridging the divide, in the interests of a more unified conversation? My answer is to locate those possibilities through recourse to concepts from quantum mechanics, and especially through reflection on quantum theory’s central ‘measurement problem’. Attention to the measurement problem, and its vehicle, the ‘apparatus’, can help move IR beyond its present epistemological binary: first, reorienting the relationship between measurement and reality as envisaged by both realist/positivist and interpretivist/critical IR traditions; second, positioning the apparatus as a basis for exploring questions in which both camps have mutual stakes.

I write this article at a good moment for thinking about the use of quantum concepts in IR. For, over the last decade, we have seen a proliferation of interest in ‘quantised IR’: with publication of several major texts ² advocating a ‘“turn” to quantum social theory’ (Murphy, 2021a: 562) and a general ‘vogue’ for quantum terminologies across the wider field (‘entanglement’, ‘uncertainty’, ‘multiplicity’, etc., Torrent, 2023: 2), confirming that the ‘idea of “quantum international relations”’ is ‘definitely in the ether’ (Der Derian and Wendt, 2020: 400).

Quantum IR’s advocates propose a ‘leap’ between ‘Newtonian’ and ‘quantum’ imaginaries ‘will be nothing short of revolutionary’ for IR (Murphy, 2021b: 62). As it stands, this leap is largely framed as landing in favour of interpretivist/critical approaches to IR. Michael Murphy’s (2021a) book Quantum Social Theory for Critical International Relations Theorists is clearest in this respect, arguing the assimilation of quantum concepts in IR offers a fillip to ‘the practice of critical scholarship’ (Murphy, 2021b: 62); namely, by elevating ‘uncertainty and the open-ended potentialities of world politics’ (Katzenstein, 2022b: 8–25) over ‘assumptions of separability, law-like predictability, and objectivity that. . . dominate positivistic projects of social research and IR’ (Heffernan and Murphy, 2023: 3). Quantum IR’s leaning towards interpretivist/critical agendas is understandable. On one hand, many of those involved in quantum IR are also leading figures in interpretivist/critical thought (e.g. James Der Derian, Karin Fierke, Mark Salter, Alexander Wendt, etc.). On the other, many concepts of quantum mechanics tend towards complicating realist/positivist models – as with Wendt’s (2015) use of wave function collapse to frame states as waves, ‘continuously popping in and out of existence with the practices through which they are instantiated’ (p. 264), in opposition to realist/positivist approaches’ ‘simple billiard ball model of international relations’ (Katzenstein, 2022b: 25). It is precisely this synergy between quantum IR and interpretivist/critical traditions that underlies existing interpretivist/critical sub-disciplines’ scepticism (or ‘ambivalence’, Sjoberg, 2020: 127) towards quantum IR (quantum IR’s argument already being, in Kimberly Hutchings’ (2022) words, a ‘staple to. . . existing [interpretivist/critical] understanding[s] of social scientific inquiry’ (p. 188)).

For the present article, however, I argue that quantum theory’s real promise for IR lies less in adjudicating between realist/positivist and interpretivist/critical traditions – and more in building bridges between each side. I draw this argument on bridge-building through engagement with a series of concepts that have not yet become central to quantum IR’s intervention. These are concepts on quantum’s measurement problem and especially the ‘apparatus’ – whose centrality to quantum theory has been overlooked in quantum IR, via a focus on alternative concepts like wave function collapse, entanglement, and superposition (one notable exception to this trend being Fierke, 2022: 10). Yet, the measurement problem and apparatus are absolutely pivotal parts of quantum theory. The apparatus, in particular, features centrally in writings of quantum mechanics’ founding parents, Niels Bohr and Werner Heisenberg: as a way of thinking about the ensemble of technical instruments, material conditions, and conceptual imaginaries whose collective functioning conditions quantum’s measurement problem. It is in this apparatus that I see scope for a more unified conversation between IR’s epistemological camps.

Measurement = generative: theorising the ‘apparatus’

Quantum’s measurement problem complicates existing attitudes to the relationship between measurement and reality in IR. Instead of positioning measurement and reality as separate, quantum’s measurement problem reveals them to be phenomenal – indissoluble parts of a single phenomenon. More particularly, quantum mechanics shows us that measurement generates conditions for reality to come into being according to particular material or conceptual configurations. And it is the apparatus, as the ensemble of technical instruments, material conditions, and conceptual imaginaries through whose application measurement proceeds, that defines these configurations. Absorbing this concept of the apparatus as a principle for understanding measurement’s generative relation to reality can lead IR’s two epistemic camps towards a more unified conversation: centring on questions of mutual interest, like the question of what instruments, material conditions, and conceptual imaginaries comprise IR’s measurement apparatus in the first place; or the question of what kinds of realities this apparatus’ application might actually produce?

The measurement problem

Quantum mechanics encompasses an array of (oftentimes contradictory) frameworks and findings. Hence, Feynman’s (1985) famous claim that ‘nobody understands quantum mechanics’ (p. 129). Yet, there remains one constant across quantum theory, ‘woven through [its] contested, layered histories’, and affording ‘a convenient “hook” on which to hang the various interpretations’ (French, 2023: 3). To John Wheeler (1978), this constant represented ‘the quantum principle’ (p. 29, emphasis added) – the fundamental premise distinguishing quantum from classical physics. This is the premise of quantum’s measurement problem, including measurement’s generative role in shaping conditions for phenomena to appear.

Whereas classical physics grounded itself in an ‘idealisation in which we speak about the world as entirely separated from ourselves’ (Heisenberg, 2000: 63), quantum mechanics maintains ‘the impossibility of any sharp separation between the behaviour of atomic objects and the interaction with the measuring instruments which serve to define the conditions under which phenomena appear’ (Bohr, 1987a: 40, emphasis in original). For Bohr (1928), this was the ‘quantum postulate’: implying ‘an independent reality in the ordinary physical sense can neither be ascribed to the phenomena nor to the agencies of observation’ (p. 580) – with this reality owing its particular qualities to the very parameters of the apparatus by which it was measured (making the apparatus as much a part of a phenomenon’s reality as the phenomenon itself).

One example of early experimentation in quantum mechanics elucidates this phenomenal relationship between ‘measurement’ and ‘reality’. This is the example of quantum mechanics’ famous ‘two-slit’ experiment. Designed by Thomas Young to resolve the puzzle of light’s ‘wave’ or ‘particle’ materiality, the two-slit experiment actually ended up revealing the strange phenomenon of wave function collapse – and the startling role of measurement in animating this collapse. Figure 1 shows Bohr’s (1987b) diagram of Young’s two-slit apparatus (p. 45). As Young explained, the arrangement is ‘very simple really’:

It just has three parts: a device that is the source of the entity being tested [on the left of Bohr’s diagram], a barrier with two holes in it [in the middle], and a screen placed some distance further back [on the right]. Now, if you want to know if an entity is a wave or particle you simply fire a bunch of them at the barrier with the two open slits. One of two patterns will appear on the screen. If most of the entities hitting the screen collect directly across from the slits the entity in question is a particle. On the other hand, if a distinctive pattern with alternating bands of intensity appears on the screen, the entity in question is a wave. (In Barad, 2010: 255)

Figure 1.

Bohr’s two-slit diagram.

Initial runs of the two-slit experiment seemed to confirm light’s ‘wave’ character, with light beams fired through the screens registering an interference pattern on the end plate – as in Bohr’s diagram, and as expected of an entity with wave-like characteristics. However, subsequent investigation found that firing particles (e.g. electrons) through the apparatus also produced an interference pattern on the end plate, implying that ‘particles’ might also behave like ‘waves’: passing through both slits and interfering with each other to produce the diffraction pattern, in direct contradiction of distinctions between wave and particle materiality. To clear up the confusion, Bohr proposed an adjustment to the two-slit apparatus: adding a detection device to each slit, to verify whether the ‘particles’ were indeed travelling through both slits as ‘waves’, as the experiment seemed to suggest. The results of this amendment were ‘astonishing’ (Barad, 2007: 102). Namely, that once a detector was added to the apparatus, the diffraction pattern on the end plate disappeared. Instead, what appeared on the plate were spots directly opposite each slit – with the wave-particles now behaving entirely as particles: passing through one slit and hitting the screen opposite in distinct loci, without interference.

The only change between the two versions of the apparatus was Bohr’s addition of a detector in the central screen. That this addition was enough to engender ‘collapse’ from wave to particle states was ‘nothing short of revolutionary’ (Barad, 2007: 106). It affirmed the generative role observation plays, in determining conditions for physical reality. As Wendt (2015) notes, ‘the correct conclusion’ from the experiment was

that as long as the electron is not being observed it behaves as if it is a wave and as soon as it is observed it behaves as if it is a particle. . . Depending on the experiment, we always see only one aspect or the other: put detectors on the slits and we get particles, take them off and we get waves. (pp. 46–48)

The distinction crystallises measurement’s phenomenal relationship with matter. For ‘particles only manifest their particle nature when explicitly “asked” to do so within an experimental setup. When not, that is, when not being watched too closely, they morph into an “imaginary probability-wave”’ (Schnabel, 2022: 89). In an empirical sense, we find the forms comprising our material reality are watching us, and that they will adjust their qualities, depending on whether we are watching them too – ‘reality varies’, per Heisenberg (2000), ‘depending on whether we observe it or not’ (p. 20) (see Figure 2).

Figure 2.

Two-slit meme.

Measurement = generative

This is the essence of quantum mechanics’ measurement problem. The two-slit experiment (alongside others, for example, the delayed choice/quantum eraser experiments, Barad, 2007: 315; Wheeler, 1978: 14) indicated measurement was in fact a ‘creative process’ (Heisenberg, 1971: vi). Rather than being something applied by the observer to a distinct external object, as in Descartes’ mind-body dualism, measurement had somehow insinuated itself into the conditions by which phenomena assumed material form. Measurement = generative. Bohr and Heisenberg described this generative relation between measurement and reality in terms of ‘indeterminacy’ or ‘uncertainty’. For Bohr (1987a), ‘any attempt to acquire a knowledge of [physical] processes involve[d] a fundamentally uncontrollable interference with their course’ (p. 100), rendering ‘reality’ (in the material sense by which disciplinary physics uses the term) as much a product of the parameters of an applied apparatus as of phenomena themselves. Likewise, for Heisenberg, ‘the atom actually has no specific attributes until it interacts with something else that has’ (Gilder, 2009: 307) – making the dimensions of that ‘something’ with which the atom interacted (i.e. the apparatus) an integral part of its real form. In either operationalisation, there could be no strict distinction between measurement and reality, since both were shown to be ‘an inseparable part of the phenomenon’ (in Rovelli, 2022: 119).

John Wheeler made a useful comparison, in this respect, between measurement and a game of 20 questions. Wheeler proposed quantum’s measurement problem as an inverted version of the game, in which the answerer does not choose a word for the questioner to guess in advance – but rather allows that word to emerge through the process of questioning. The answerer

could respond ‘yes’ or ‘no’ as he pleased to whatever question was put to him. But however he answered, he had to have a word in mind compatible with his own reply – and with all the replies that went before.

To Wheeler (1978), this inversion of the game’s conventional logic paralleled quantum’s inversion of classical physics:

First, the puzzled participant thought the word already existed ‘out there’ as physics once thought that the position and momentum of the electron existed ‘out there’, independent of any act of observation. Second, the information was brought into being step by step by the experiments that the observer chooses to make. Third, if the participant had chosen to ask different questions he would have ended up with a different word, as the experimenter would have ended up with a different story for the electron if he had done different measurements or the same measurements in a different sequence. (pp. 28–29)

In sum, measurement and reality are not independent but phenomenal. Measurement involves the ‘creation-discovery’ (Aerts, 1998: 223) of reality: gradually bringing the latter into being, in an entirely physical sense and according to the particular configurations of an applied apparatus, just as the word in Wheeler’s version of 20 questions is gradually brought into being by the questions players choose to ask.

Measurement and IR

This ‘creation-discovery’ view of reality carries implications for IR’s epistemological camps. On one hand, it opens doors to a different set of epistemological premises than that from which both camps begin. Earlier, I depicted these camps as united in accepting the Cartesian dualism of mind and body, or measurement and reality. Realists/positivists accepted this dualism by operationalising reality as a distinct something – which existed outside the mind of the observer, but which could be fully known inside the mind through accurate scientific method. Interpretivists/critical scholars also accepted this dualism, even in proposing a different relationship between measurement and reality – by posing the particular situatedness of knowledge as an obstacle to fully internalising external realities (‘the world exists independently of language, but we can never know that (beyond the fact of its assertion), because the existence of the world is literally inconceivable outside of. . . our traditions of interpretation’, Campbell, 1998: 6).

The first lesson IR can learn from the measurement problem concerns the illusory nature of this distinction between measurement and reality. Quantum-mechanical experimentation and theory show us that measurement and reality are not distinct. They are phenomenal. Hence, quantum’s ontological ‘holism’ (Fierke, 2022: 14). Quantum does not establish separation between measurement and reality, but amalgamation: it being impossible to operationalise ‘reality’ without ‘measurement’, or ‘measurement’ without ‘reality’, even to the extent of rendering the phrase ‘relationship between’ measurement and reality effectively meaningless. What is a relationship without relata, after all? Northern Irish physicist John Bell (1990) was emphatic on this principle, reflecting of quantum’s measurement problem that ‘when it is said that something is “measured” it is difficult not to think of the result as referring to some pre-existing property of the object’; whereas, in actual fact, the measurement problem suggested the utter inseparability between act of measurement and object being measured – in the constitution of whose reality ‘the apparatus as well as the system is essentially involved’ (p. 35).

The second lesson IR can draw from the measurement problem concerns the character of this ‘apparatus’, which is so essentially involved in generating conditions for reality to emerge. I argue the apparatus can be a fruitful concept for IR: as the vehicle for conceptualising measurement’s phenomenal relation to reality; and as the basis for a new, unified conversation between IR’s two epistemic camps.

The apparatus

In Bohr and Heisenberg’s writings on quantum mechanics, the apparatus appears centrally: as the principal mechanism within which measurement’s generative relation to reality unfolds. In a 1946 paper ‘On the Problem of Measurement’, for instance, Bohr (1998) argued,

While until now it has been assumed that all phenomena. . . can be traced back to the behaviour of the objects independent of observation, in the study of the individual quantum effects we are by the very nature of the situation prevented from differentiating sharply between an independent behaviour of atomic objects and their interaction with the means of observation. (E.g., the apparatus, p. 125)

Conversely, Heisenberg (2000) reflected ‘it does not matter whether the observer is an apparatus or a human being; but the registration, that is, the transition from the “possible” to the “actual”, is absolutely necessary and cannot be omitted from the interpretation of quantum theory’ (p. 89). For both authors, the apparatus in which quantum phenomena were manifested was essential to their unambiguous description. Quantum’s measurement problem hinged upon the apparatus. But what did Bohr and Heisenberg mean by this concept?

On one hand, we can think of the apparatus as a specific configuration of technical instruments used to measure (read: generate) phenomena. Bohr’s (1987a) descriptions of the apparatus dwelt repetitively, in this regard, on ‘the measuring rods and clocks used for observation’ (p. 98) – ‘the measuring instruments which serve to define the conditions under which the phenomena appear’ (Bohr, 1987b: 40, emphasis in original). It is in this sense of the apparatus that we can understand experiments like the two-slit experiment. In these experiments, there was a mechanical ‘disturbance’ (Dirac, 1947: 3) of the system under observation, which forced it to adopt a particular kind of material reality, in conformity with the apparatus’ instrumental configurations. This instrumental disturbance informed Heisenberg’s uncertainty principle – the observation that defining a physical system’s properties (for example, a particle’s position/momentum) required application of a measurement apparatus designed for the purpose. But any apparatus designed to recognise one property must also preclude its manifestation of other properties (the more a particle’s position was defined through application of an apparatus with fixed parts, for instance, the less its momentum could be known – given the latter requires an apparatus with moving parts). Thus, the act of mechanical measurement enacted ‘a complete rupture’ in the system’s physical qualities (Bohr, 1928: 584), taking place ‘at the cost of the connection between the past and future course of phenomena’ (Bohr, 1987a: 11).

On the other hand, we can also think of the apparatus in a more expansive sense – to include material conditions underlying measurement practice, as well as conceptual imaginaries by which it is pursued. This more expansive operationalisation of the apparatus is of greater relevance to IR, reminding us that our concepts, theories, languages, and discourses on the international are as much involved in measuring-generating its realities, as were the physical dimensions of any technical apparatus.

Bohr himself understood the apparatus in this more expansive sense. Bohr’s (1987c) work was unequivocal that the ‘unambiguous account of proper quantum phenomena must, in principle, include a description of all relevant features of the experimental arrangement’ (p. 4), that in any ‘specification of the conditions for any well-defined application of the formalism, it [was] essential that the whole experimental arrangement be taken into account’ (Bohr, 1987b: 50, emphasis in original). The question here is where this ‘whole experimental arrangement’ ends? Was the person operating an experimental setup part of the apparatus (as Heisenberg assumed, above)? What of the concepts informing their operations, and languages by which they were communicated? Karen Barad’s (2003) reading suggests Bohr’s ‘whole experimental arrangement’ did indeed encompass these categories: capturing the entirety of measurement practice, the fullness of the conditions (material or conceptual) under which measurement was possible (pp. 819–820). Barad makes an analogy in this respect, with the voice of Stephen Hawking. At public lectures, Hawking’s voice reaches his audience through a chain encompassing the public address (PA) system, speech-generating device, handheld switch, neural pathways connecting Hawking’s hand to his brain, and the brain itself where Hawking formulates his words. Of this chain, Barad (2007) asks ‘where is Hawking?’; or, more to the point, where is Hawking’s voice? Is it in the PA system sound, the speech-generating device, Hawking’s brain, or even the brains of Hawking’s listeners? Arguably, ‘Hawking/Hawking’s voice’ includes all of these: ‘in an important sense, Hawking doesn’t stop being Hawking at the edge of his body. . . a serious part of Hawking extends into the box in his lap. No box, no discourse’ (p. 159). Likewise, the experimental apparatus does not stop at the edge of its technical arrangement. For a serious part of the apparatus extends into the person operating its instruments, and into material/conceptual conditions by which they operate. No person/conditions/concepts, no findings.

Bohr’s most powerful recourse to this expansive definition of the apparatus came through his response to Erwin Schrödinger’s findings on quantum entanglement. In quantum mechanics, entanglement describes a situation whereby two particles that have been united in a singlet system at one time will subsequently exhibit entangled natures: with measurement of one precipitating instantaneous, opposite change in the other – even when the two are separated by great magnitudes of space, and thus enjoy no mechanical connection. When it comes to the apparatus, entanglement’s importance is in its expansion of quantum’s measurement problem – to include disturbances that are entirely un-physical in nature in determining physical reality (i.e. the upwards/downwards spin of entangled particles), alongside mechanical disturbances like those seen in the two-slits. What entanglement tells us is that measuring one particle will simultaneously affect the physics of another – even when the particles are separated by many kilometres, which, ‘given the enormous distance’ this separation represents ‘relative to the sub-atomic scale’, implies entangled particles’ mutually correlated polarisations applies ‘no matter how far apart they were, even across the universe’ (Wendt, 2015: 53–54, emphasis in original). Given this distance between the particles, it seems there can be no question of the disturbance being mechanical in nature – the startling upshot of which is that measurement’s generative relationship to reality must apply as much to abstracted mathematical conceptualisations (e.g. the quantum formalism that ties particles into their entanglement) as to the material dimensions of an experimental apparatus.

It was this un-physical dimension of the measurement problem that aggravated Albert Einstein’s famous objections to entanglement (and thereby quantum formalism). In their 1935 paper critiquing entanglement, Einstein, Podolsky, and Rosen grounded their scepticism in a disbelief that a particle could change its spin state purely as a result of mathematical prediction/conceptualisation – that is, without mechanical disturbance. Their investigations found that ‘by measuring either [particle] A or B we are in a position to predict with certainty, and without in any way disturbing the second system, either the value of the quantity P. . . or the value of the quantity Q’ (P and Q being properties of the unmeasured particle, Einstein et al., 1935: 780). Crucially, Einstein/Podolsky/Rosen (EPR) concluded,

this makes the reality of P and Q depend upon the process of measurement carried out on the first system, which does not disturb the second system in any way. No reasonable definition of reality could be expected to permit this. (Einstein et al., 1935: 780)

Similarly, Schrödinger held that

when two systems interact, their Ψ-functions. . . immediately cease to exist and a single one, for the combined system takes their place. . . Thus one disposes provisionally (until the entanglement is resolved by an actual observation) of only a common description of the two. (Schrödinger and Trimmer, 1980: 338)

In this case, quantum’s measurement problem required the application of no physical apparatus for its realisation – but rather a purely conceptual one. As Bohr reflected in a 1935 interview on EPR, with entanglement the problem of measurement was purely mathematical. But this did not mean it should not be considered part of the measurement problem. On the contrary, it meant the measurement problem had to be expanded to acknowledge that even conceptualising a system mathematically enacted a change in possibilities for its reality. With entanglement, there was ‘no question of a mechanical disturbance of the system under investigation’, since the entangled particle’s properties appear based on their calculation/prediction alone. Yet, for Bohr, this very act of prediction was enough to sustain measurement’s generative effects because, ‘even at this stage. . . there is essentially the question of – this should be underlined, too – an influence on the very conditions which define the possible types of predictions regarding the future behaviour of the system’ (in Gilder, 2009: 166, emphasis in original). This is to say, the act of abstractly predicting (i.e. imagining) the entangled particle’s properties was enough to enact the adjusted calculus of possibilities for its physical reality comprising quantum’s measurement problem. That’s exactly why Einstein chose to dismiss entanglement as ‘spooky actions at a distance’ (Einstein and Born, 1971: 158). Simply, it seemed entanglement had taken quantum’s measurement apparatus into the realm of the ‘spooky’ supernatural.

Quantum’s material-conceptual apparatus

So, the apparatus encompasses more than just its technical arrangement. To Bohr and Heisenberg, the apparatus had also to include material conditions by which measurement proceeded – and conceptual imaginaries according to which it unfolded. Of the former, the story of the Stern-Gerlach experiment is a good example. This experiment produced important findings vis-à-vis angular-momentum quantisation in atomic-scale systems. However, these findings would have been impossible without human-scale material relations’ insertion into the experimental setup. Otto Stern recalled, in this regard, the ‘strange experience’ by which he returned his experimental findings following multiple inconclusive runs:

Gerlach removed the detector flange. But he could see no trace of the silver atom beam and handed the flange to me. With Gerlach looking over my shoulder as I peered closely at the plate, we were surprised to see gradually emerge two distinct traces of the beam. Several times we repeated the experiment, with the same mysterious results. Finally we realised what it was. I smoked cheap cigars. These had a lot of sulphur in them, so my breath on the plate turned the silver into silver sulphide, which is jet black so easily visible. It was like developing a photographic film. (Herschbach, 2021: 9)

Stern’s cigar smoking – or, more specifically, his status as an underpaid academic smoking cheap cigars – was as much a part of the apparatus as the magnetic field through which he beamed his silver atoms. Per Murphy (2022),

what if [Stern/Gerlach] had not been males of a social elite class whose norms included cigar smoking?. . . What if the economic status of 1920s Germany had permitted greater remuneration such that junior researchers [could] afford high-quality cigars that did not contain enough sulphur to produce silver sulphide? (p. 124)

That the answer to these questions is ‘no cigar, no experiment’ leads us to the ineluctable conclusion that ‘class, nationalism, gender. . . among other variables, are all part of this apparatus’ (Barad, 2007: 165). Far from being a finite technical device, the apparatus encompasses the full web of connections, practices, and relations informing knowledge production: exercising a centripetal force that brings a plethora of contingencies to bear upon ‘the larger material configuration of the world’ (Barad, 2007: 171).

Among this plethora, concepts play as significant a role as agential/material considerations. From the start, quantum mechanics’ progenitors understood its findings to indicate the centrality of concepts/language within measurement. For concepts, as crystallised in language, were a fundamental part of the conditions under which measurement took place. According to Bohr (1987b), ‘the aim of every physical experiment – to gain knowledge under reproducible and communicable conditions – leaves us no choice but to use everyday concepts. . . not only in all accounts of the construction and manipulation of the measuring instruments but also in the description of the actual experimental results’ (p. 26); ‘language is indeed our main tool for the expression and communication of knowledge’ (Bohr, 1998: 189). Equally, Heisenberg (2000) felt ‘any kind of understanding, scientific or not, depends on language, on the communication of ideas. Every description of phenomena, of experiments and their results, rests upon language as the only means of communication’ (p. 94). Both Bohr and Heisenberg were sensitive, then, to the ‘sticky problem of humanity’s own captivity within language’ (Barad, 2003: 812). And by this captivity, they meant three things. Concepts determined conditions for:

• 1. Most basically, the communicability of measurement.

Bohr and Heisenberg appreciated that concepts/language define possibilities for exchanging knowledge about quantum phenomena, including phenomena emerging through experimentation. To Bohr, early efforts to understand atomic phenomena fell short because, while

we intend to say something about the structure of the atom[, we] lack a language in which we can make ourselves understood. We are in much the same position as a sailor, marooned on a remote island where conditions differ radically from anything he has ever known. . . He simply must make himself understood, but has no means of doing so. (Heisenberg, 1971: 40)

Likewise, in desperate efforts to convince Einstein of quantum theory in 1926, Heisenberg (1971) explained his difficulties in having ‘no idea in what language we must speak about processes inside the atom [such that] we cannot hope to speak of the path of the electron in the cloud chamber without inner contradictions’ (p. 66). This understanding of quantum mechanics as hinging upon a series of ‘communication problems’ (Petersen, 1963: 10) was not the trivial one of simply failing to find the right words to describe independent realities. For, as noted earlier, quantum’s whole gambit was that there is no independent reality, aside from practices by which it is measured/conceptualised/brought into being. Quantum’s early founders thus considered the problem of communicability to be one of entanglement within a conceptual apparatus for creating-discovering reality.

• 2. More intrinsically, the construction of technical apparatuses.

This non-trivial captivity becomes more apparent when we think of conceptual a priori underlying any measurement practice. On this aspect of the apparatus, it was Heisenberg who took the lead. Bohr (1987c) acknowledged ‘the decisive point is to recognise that the description of the experimental arrangement must be given in plain language, suitably refined by its usual physical terminology’ (p. 3) – meaning the apparatus’ physical arrangement was limited to what it was possible to think, and to construct in relation to that thought. But Heisenberg went further: posing the problem of an impassable ‘cut’ between a technical apparatus built on classical concepts, and the quantum phenomena that apparatus was designed to unveil. As Heisenberg (2000) noted, ‘the Copenhagen interpretation starts from a paradox. Any experiment in physics, whether it refers to the phenomena of daily life or to atomic events, is to be described in terms of classical physics’, meaning ‘the concepts of classical physics form the language by which we describe [and thus assemble] the arrangement of our experiments’ (p. 14). In this sense (again, in Heisenberg’s words), ‘in an experiment on a quantum object two worlds are meeting – a quantum world and a classical world’: both of which ‘behave in totally different ways. Something crucial is lost in that meeting. . . Both on the quantum-mechanical side, with the object, and on the classical side, with the apparatus’ (in Gilder, 2009: 152). The possibility of meeting quantum phenomena required absorption of concepts in the experimental apparatus’ design (‘fixed parts. . . to make the notion of “position” intelligible. . . moveable parts. . . for “momentum” to be intelligible’, Barad, 2010: 253), placing concepts at the heart of measurement. Yet this necessity put quantum in the bind of using classical concepts to measure quantum phenomena, whose very properties exposed the former as untenable.

• 3. Most subtly, the reception of sense (i.e. experimental) data.

Finally, concepts/languages entered into measurement’s generative quality through their (unavoidable) use in receiving experimental results. Bohr acknowledged this problem during a skiing holiday with Heisenberg (1971), in which he posed the paradox of using classical concepts to interpret quantum investigations: ‘we know that the concepts of this language are imprecise, that they have a limited area of application, but we have no other language, and, after all, it does help us to grasp the phenomenon at least indirectly’ (p. 130). Later, he accepted

that is the whole paradox of quantum theory. On the one hand, we establish laws that differ from those of classical physics; on the other, we apply the concepts of classical physics quite unreservedly whenever we make observations, or take measurements. (Heisenberg, 1971: 129)

Concepts were thus part of Bohr’s apparatus, as the lens through which quanta were seen. He developed this notion for Heisenberg on a separate holiday, walking from Copenhagen to Helsingør. Arriving at Kronborg castle, Bohr asked,

isn’t it strange how this castle changes as soon as one imagines that Hamlet lived here? As scientists we believe that a castle consists only of stones, and admire the way the architect put them together. . . None of this should be changed by the fact that Hamlet lived here, and yet it is changed completely. Suddenly the walls and the ramparts speak a different language. The courtyard becomes an entire world, a dark corner reminds us of the darkness in the human soul, we hear Hamlet’s ‘to be or not to be’. . . Kronborg becomes quite a different castle for us. (Heisenberg, 1971: 51)

Bohr indulged this analogy between quantum and everyday experience to demonstrate conceptual a priori’s role in determining conditions for measurement. For Bohr, concepts were part of the apparatus by which sense data were received, interpreted, given meaning – in such a way as to change the nature of the thing being observed. Heisenberg (1971) was gripped by this final entanglement between concepts and apparatus: writing later

we cannot grasp things directly, we must first transform them into ideas, and then shape these into concepts. . . Only through this transformation, through this fitting together of individual impressions into a ‘comprehensible’ whole, can we claim to have perceived anything. . . For ideas are obviously prior to experience; indeed, they are the prerequisite of all experience! (p. 4)

Bohr and Heisenberg made ‘one of the strangest moves in the history of physics’ in these respects: directing our attentions ‘to the question of. . . language!’ within the configuration measurement = generative (Barad, 2010: 252–253, emphasis in original). Of concepts’ role within the apparatus, Bohr concluded ‘we are suspended in language in such a way that we cannot say what is up and what is down’ (in Petersen, 1963: 11). Language, as the expression of concepts, had insinuated itself into the apparatus according to which reality acquired its particular characteristics. If quantum mechanics’ measurement problem made it impossible to speak of reality in isolation from the apparatus with which it was measured, then these further insights made it impossible to conceive of a reality in whose generation concepts had no principal hand. Put more simply, for quantum, ‘our discourse on reality is itself part of that reality’ (Rovelli, 2022: 166). As Heisenberg (1971) concluded towards the end of his life, anyone who had ‘thought about Plato’s philosophy will know that the world is shaped by images’ (p. 246).

Herein lies the apparatus’ relevance for IR. Quantum mechanics’ measurement problem suggests measurement exists in a generative, phenomenal relation to reality. But quantum’s founding parents rejected any effort to restrict this relation to narrow instrumental terms. Instead, they found their apparatus to be a conceptual as much as a material device – reaching beyond the experimental arrangement itself to include concepts by which it was designed, a priori through which its sense data were received, and languages in which its results were communicated. For IR, this means recognising our theories, observations, concepts, languages, institutional arrangements, material relations, and discourses as all part of an apparatus that brings the phenomena we measure into being. And I don’t mean this in the traditional interpretivist/critical sense, whereby research is seen as embedded in a sociology of knowledge precipitating change in that which is researched. Nor do I mean it in the Cartesian or relativistic sense that research constructs a world within the mind of an observer, constituting its own reality with equal status to ‘external’ equivalents. Rather, I mean it in the much ‘weirder’ (Ball, 2018) quantum-mechanical sense: of the theories, observations, concepts, languages, institutional arrangements, material relations, and discourses that constitute our apparatus being part of a process that brings world politics into being according to a particular physical-phenomenal-observed nature.

If this sounds hard to understand or envisage, that’s a good thing. After all, anyone who thinks they’ve understood quantum mechanics hasn’t (Feynman, 1985: 129). The point, as I’ll come to shortly, is not so much to understand the implications of the measurement problem and its apparatus – as it is to use them as a basis for conversation between IR camps: regarding what it might mean to bring non-physical phenomena into being according to the configurations of IR’s particular material-conceptual apparatus.

Conclusion

This article has used concepts from quantum mechanics to address a binary epistemological divide in IR. The article considered the resilient distance between IR’s realist/positivist and interpretivist/critical traditions. And it raised problems engendered by this distancing, including especially the problem of absent cross-pollination between mutually invisible campfires. Engaging a reading of texts by quantum mechanics’ founding parents, the article explored possibilities for bridging the divide between camps through assimilation of quantum theory’s measurement problem – including its overarching principle, that measurement = generative of conditions for reality; and its central conceptual vehicle, the material-conceptual ‘apparatus’, as theorised by Bohr and Heisenberg. At its heart, this article aspires to position the apparatus as a basis for more unified conversation between IR’s epistemic camps. The lesson quantum mechanics affords for these epistemic camps is that measurement and reality are not distinct but phenomenal. For quantum, measurement is ‘constitutive of reality’ such that reality cannot be understood as ‘composed of things-in-themselves’, but rather as ‘things-in-phenomena’ (Barad, 1997: 176), with ‘phenomena’ here referring to the holistic apparatus of observation that ‘is creative of the possibility for meaning’ (Murphy, 2021b: 21).

Absorbing this lesson would open doors to asking questions in whose pursuit both camps share a mutual interest. These questions start with the apparatus from within whose confines and according to whose parameters measurement’s generative relation to reality unfolds. As I’ve noted, this apparatus’ relevance for IR derives from its inclusiveness of concepts, theories, languages, discourses (i.e. the materia of social scientific investigation) as much as experimental instruments (the materia of early quantum physics). It was precisely in this expansive sense of the apparatus/the measurement problem that quantum mechanics’ founders understood its importance. Bohr (1987c) himself felt quantum mechanics’ ‘significance. . . does not merely lie in the steady increase of our experience of inanimate matter, but above all in the opportunity of testing the foundation and scope of some of our most elementary concepts’ (p. 1). His various written/spoken engagements were directed above all to developing ‘the general epistemological lesson’ of quantum’s measurement problem, which he felt would ‘illuminat[e] our position in many other fields of human interest’ (Bohr, 1987b: 12). Rather than being reducible to pure formalism, this general lesson shed ‘new light upon the old [global] philosophical problem of the objective existence of phenomena independent of our observations’ (Bohr, 1987a: 115). There’s no reason to believe this light shouldn’t also carry implications for IR, as I’ve tried to elucidate here.

Measurement generates conditions for reality. This generative relationship takes place from within the bounds of an applied apparatus, which encompasses material contingencies and conceptual imaginaries underlying measurement, as much as physical instruments by which it proceeds. How would absorbing these lessons help IR? First, it would enable us to settle our epistemological differences (grounded, as they are, in an outdated model of the relationship between measurement and reality) – and begin a more unified conversation on world politics. To what questions might this conversation be directed? As a start point, I think it would be worth asking: if our apparatus is indeed generative of the reality of political phenomena, what is its exact nature? According to what material relations has it been forged? And through what inherited images does it think? Of course, many of these questions have already been explored, or are currently in the process of being explored. This is a good thing; it means we already have tools readily available to answer them. However, doing so through the prism of the apparatus would mean something slightly different – precisely because of what this apparatus stands for (namely, a vehicle for thinking about how IR’s material contingencies or conceptual imaginaries contribute to generating world politics’ physical realities). Considering the qualities of IR’s apparatus, then, would also invite us to ask: what kinds of realities is our apparatus really generative of? What is the nature of the reality IR’s apparatus brings into being? And, indeed, what realities can it not generate? ³ Finally, what does it really mean for IR’s apparatus to be generative of political phenomena – in the sense by which quantum mechanics means this generative-ness, specifically?

The present article isn’t disposed to answer these questions. But it is disposed to ask them. It has drawn them through engagement with the writings of those responsible for the ‘revolution’ (Bohr, 1998: 1) in our understandings of reality comprising quantum mechanics. And it advocates them here, as a door to different conversations in IR, between our discipline’s presently divided campfires.