The Impact of Roy Bhaskar and Critical Realism on International Relations

Roy Bhaskar, who died in November 2014, is considered the originator of the critical realist approach, a variant of scientific realism. This short summary considers the potential impact of Bhaskar’s philosophy on international relations.

The previous sentence may appear somewhat odd, considering Bhaskar has already had an impact on IR. His will be a familiar name to most IR scholars, due to the work of Alexander Wendt (1987), who first introduced Bhaskar’s ideas to an IR audience nearly thirty years ago. Subsequent works by Dessler (1989), Patomaki (2002), Wight (2006), and Kurki (2008) are all well-known. Even sceptics like Brown (2007) feel entitled to wonder whether critical realism might become the next big thing in IR theory.

But actually, when we consider whether young scholars are working with critical realism today, it is not easy to think of many examples. Indeed, critical realism, although well-known, is no longer considered fashionable. Furthermore, it has been attacked by recent ‘turns’ to such things as practice theory, pragmatism, Actor Network Theory, and new materialism, even though some of the current trends like new materialism and complexity theory owe much to Bhaskar’s concept of emergence as interpreted by DeLanda (2006).

Perhaps critical realism has not been more impactful because it does not set out to provide a theory of IR and, in contrast to constructivism and poststructuralism, does not make strong philosophical claims about what the social world is like (distinguishing between scientific and philosophical ontologies). Bhaskar’s early work saw the role of critical realism as akin to an underlabourer to the sciences, not a substitute for substantive research (Bhaskar 1989b: vii). In any case, metatheory is not popular in the current climate of practice-turn thinking (for critical scholars) and methods-driven research (for less critical ones). Even if given a more humble underlabouring role, its ontological and epistemological arguments are either dismissed or taken for granted as correct, but not worth dwelling on.

So perhaps it is time to state explicitly just what Critical Realism has offered and can continue to offer to IR.

1) There is a real world out there! The realist position on the existence of a world independent of the knowledge we have of it has been subjected to critiques from constructivism and poststructuralism. It is now subject to renewed critique from pragmatists, complexity theorists, practice theorists, and actor network theorists. What these approaches all share in common is their tendency to conflate the world itself with the knowledge we have of it. By contrast, positivism does assert the independent existence of the world, but reduces our knowledge of it to simplified models or identifications of regularities – a kind of empirical realism. In fact, what the ‘new’ approaches share with positivism is a form of naive empiricism that denies the need for conceptual or philosophical abstraction, regardless of how complex the world is considered to be. By contrast, critical realism regards knowledge as meaningful precisely because of its relation to something out there in the world. Indeed, the different positions mentioned above can only properly be understood once it is recognised that they are disputing the nature of the world itself, not just the knowledge we have of it.

2) This leads to the second major point in favour of critical realism, its advocacy of the primacy of ontology. Critical realism tells us that if knowledge is meaningful and, indeed, if disputes are to have any significance, then they must be about something other than just understanding, and that it is the world itself, rather than epistemological framework, that is ultimately the basis of meaningful knowledge. The critical realist approach asks, what must it be about the world itself that makes knowledge possible? This transcendental realism breaks us from the vicious circularity of epistemological approaches or those views that proclaim ‘practices all the way down’. The latter would in fact be seen as making a significant (and wrong) claim about the nature of the world. The realist position thus frees us from the epistemological fallacy shared by constructivist and positivist approaches of reducing the real world to the knowledge we have of it.

3) Using causal analysis is essential to how we understand the world. However, this is not the sort of causal approach as understood by the IR mainstream. The critical realist position is again best explained in terms of its opposition to constructivist and positivist views alike. In the case of constructivism, causal analysis tends to be rejected in favour of ‘constitutive’ analysis without recognising that the way that things are constituted is itself a form of causality. In this sense constructivists are in negative agreement with the positivist understanding of causality as the constant conjunction of events whereby empirical event A is said to cause or correspond to empirical event B. Constructivism rejects such a view and, in so doing, also rejects the idea of causation. Critical realists like Kurki (2008) argue that there are varieties of causality and that ultimately we should look beyond empirical outcomes such as ‘if A, then B’, and instead seek to identify the underlying generative mechanisms. These mechanism reside in the properties of things themselves, as well as in tendencies which may or may not be exercised under various different conditions (Bhaskar 1997: 14). Positivists reduce such causal mechanisms to their exercise, whereas critical realists view causality as complexly overdetermined and irreducible to either the actualised or the experienced.

4) This, in turn, is linked to Bhaskar’s advocacy of a stratified ontology. Reality is understood as comprised of complex overlapping layers. These have their own distinctive properties and characteristics, but are part of an interacting whole. There are deeper, underlying layers that produce causal effects at higher levels. Bhaskar captures this through his distinction between the empirical, actual, and real (Bhaskar 1997: 56). While the empirical refers to that which we can observe, and the actual to the occurrence of events, the real points to the real powers and liabilities of things that may or may not be manifested as actual occurrences or empirical events. This stratified approach can also be applied to emergent levels so that the social world is emergent out of the natural world, the biological out of the physical.

5) This relates to critical realism’s approach to structural or emergent powers. Rather than a ‘levels of analysis’ approach, as might be found in mainstream IR, critical realism would advocate an emergent approach that sees international phenomena as having distinctive qualities, but also as emergent out of particular social conditions. The ‘lower levels’ from which things are emergent provide the necessary conditions of possibility, such as physical or cultural resources, but the ‘higher levels’ cannot be reduced to these conditions. The natural world has long been understood in such a way – clearly H2O is dependent upon hydrogen and oxygen for its existence, yet its properties are clearly distinctive from and irreducible to those of its component parts. Likewise, political processes might be dependent upon certain economic relations that help (causally) constitute them, but they cannot be reduced to these and, contrary to reductionist explanations, have their own specific dynamics.

6) That critical realism highlights the stratified nature of reality points to a commitment to view the world in a certain way that considers it intelligible and open to investigation. While seeing the world as independent of the knowledge we have of it, this view, rather than taking the post structuralist route of seeing only knowledge as meaningful and the world itself as meaningless, takes the approach that the meaningfulness of knowledge depends upon the meaningfulness of that which the knowledge is about. The meaningfulness of knowledge comes not from other forms of knowledge, but from the structure of the world itself. In contrast to constructivism, knowledge is not simply the product of human practices, but the way these practices are situated in relation to deeper structures. This presupposes a belief in enduring relations that can be investigated in an intelligible way. Starting with the idea that the practices of scientists are intelligible, we can take note of how natural science works through the presumption of enduring natural processes and causal mechanisms. The social world should, in turn, be regarded as relatively enduring, but with important qualifications.

7) Critical realism believes in the enduring character of social relations. However, unlike natural structures, social structures do not exist independently of the activities they govern. Nor are social structures independent of agent’s conception of their activities, although this intentional activity may have unconscious or unintentional consequences (of structural reproduction / elaboration). Social structures, in contrast to natural ones, are only relatively enduring. Their existence is ‘spatio-temporally moored’ and ‘geo-historically reproduced, distantiated and transformed’ (1989a: 175).

8) Bhaskar’s notable contribution to the structure-agency debate comes via his transformational model of social activity (TMSA). According to this model, society is both the ever present condition and the continually reproduced outcome of human activity (1989a: 34-5). Ironically, this claim has strongly influenced constructivists like Wendt (1987). However Bhaskar’s claim that people’s conscious activity for the most part unconsciously reproduces the structures governing their activity presupposes an ‘ontological hiatus’, rather than ‘mutual constitution’ of structure and agent. Society is regarded as an ensemble of structures, practices, and conventions which do not exist independently of human activity (reification), but are not merely the product of such activity (voluntarism) (1989a: 36). In contrast to constructivist arguments for ‘practices all the way down’, critical realism puts practices in their appropriate place as the mediating point linking action to structure. Agents acquire certain duties, roles, and responsibilities by virtue of the occupation of certain positions within practices (1989a: 40-1).

9) Critical realism takes a pluralistic approach to research methods. While the majority of scholars in IR probably do support a mixed methods approach, the justification for this is often not explicitly made (see Helen Louise Turton’s post ‘International Relations is not an American Discipline (Well, Maybe it is a Little)’ on The Disorder of Things). Rationalists in IR commit the epistemic fallacy of reducing the real world to the knowledge we have of it as gained though empirical research. The current trend is not to challenge the primacy of methods, but to challenge the positivist focus on quantitative data. A critical realist approach challenges the primacy of methods in IR and suggests instead that methods are part of the toolkit by which scientists gain knowledge of the world, but that the unobservable character of social structures and generative mechanisms means that theoretical abstraction and conceptual analysis are also essential parts of the scientific process. Different methods can help to reveal different aspects of the social world. Methods should therefore be chosen on the basis of their appropriateness, and fit with conceptual arguments about the nature of the processes they help to investigate.

10) This also fits with critical realism’s advocacy of judgmental rationality, insofar as we can assess theories and methods on the basis of their explanatory adequacy. Bhaskar argues for a combination of ontological realism, epistemic relativism, and judgmental rationalism. Ontological realism supports the view that the real world is independent of our knowledge. Epistemic relativism recognises that this means there is no guarantee that our knowledge corresponds to the way things actually are. However, in contrast to poststructuralist relativism, critical realism argues that there are still rational grounds for preferring some explanations over others. This is rooted not in the nature of knowledge, but in the nature of the world that this knowledge is about. Judgmental rationalism implies that there are good grounds for preferring some explanations over others based in their attempts to explain what goes on in the world.

11) Finally, although critical realism is opposed to the ontological stance taken by poststructuralist positions, there are some strong parallels with certain epistemological arguments. Bhaskar’s later critical realism (1993) develops a strong critique of ‘presentism’ or purely positive conceptions of reality that ignore its underlying uncertainty and contradictions. This can clearly be directed at rationalist accounts of international relations and their fixation with the actual rather than the possible. This approach is compatible with Adorno’s critique of identity thinking and is therefore complimentary with critical theory’s critique of ‘problem solving’ approaches, which support the status quo by taking things ‘as they are’. This is compatible not only with Bhaskar’s earlier critique of actualism, but also with critical theory, and indeed Derrida’s deconstruction and its critique of presence (Derrida 1974: 49). Indeed, Bhaksar’s approach can provide a critique of the ‘metaphysics’ of positivism and is compatible with many of the insights of deconstruction without buying into poststructualism’s (anti) ‘metaphysics’ or ontological evasions.

Bhaskar’s later philosophy is controversial, first for making more and more ambitious claims about the power of absences and significance of freedom, then for taking this in a spiritual direction. However, in highlighting the power of absences as real and determinate causes, Bhaskar is perhaps best summing up the status of his own ideas in IR at this present moment. At a time when new approaches are either knowingly or unknowingly touching on critical realist themes such as complexity and emergence, the thought of Roy Bhaskar stands in the background, perhaps best summed up as a hugely significant absence or, maybe more positively, as an under-actualised possibility.


Bhaskar, Roy (1989a) [1979] The Possibility of Naturalism, Hemel Hempstead: Harvester Wheatsheaf.

Bhaskar, Roy (1989b) Reclaiming Reality: A Critical Introduction to Contemporary Philosophy, London: Verso.

Bhaskar, Roy (1993) Dialectic: The Pulse of Freedom, London: Verso.

Bhaskar, Roy (1997) [1974] A Realist Theory of Science, London: Verso.

Brown, Chris (2007) ‘Situating Critical Realism’, Millennium – Journal of International Studies 35: 409-416

DeLanda, Manuel (2006) A New Philosophy of Society: Assemblage Theory and Social Complexity, London and New York: Continuum.

Derrida, Jacques (1974) Of Grammatology, Baltimore: The Johns Hopkins University Press.

Dessler, David (1989) ‘What’s at stake in the Agent-Structure Debate?’, International Organization, 43, 3: 441-473.

Kurki, Milja (2008) Causation in International Relations: Reclaiming Causal Analysis. Cambridge: Cambridge University Press.

Patomäki, Heikki (2002) After International Relations. Critical Realism and the (Re)Construction of World Politics, Routledge: London and New York.

Wendt, Alexander (1987) ‘The Agent-Structure Problem in International Relations Theory’, International Organization, 41, 3: 335-370

Wight, Colin (2006). Agents, Structures and International Relations: Politics as Ontology. Cambridge: Cambridge University Press.

Further Reading on E-International Relations

Tags: , , ,

Please Consider Donating

Before you download your free e-book, please consider donating to support open access publishing.

E-IR is an independent non-profit publisher run by an all volunteer team. Your donations allow us to invest in new open access titles and pay our bandwidth bills to ensure we keep our existing titles free to view. Any amount, in any currency, is appreciated. Many thanks!

Donations are voluntary and not required to download the e-book - your link to download is below.