Critical Realism and Health

By | October 20, 2016

This is an adaption of a piece I wrote for Bill Cockerham’s Encyclopaedia of Medical Sociology a while ago. It stands alone but I hope it also complements other blogs I have written on critical realism’s range and merits. Since I first wrote it: (a) sadly (he was a lovely as well as talented man) Roy Bhaskar has passed on; and (b) perhaps the most accessible of his works has been published posthumously: the title is Enlightened Common Sense and I highly recommend it.

Critical realism is a philosophical approach that has come to exercise a growing influence on medical sociology. It emanates from the philosophy of science and social science propounded by Roy Bhaskar (1978, 1989) in the UK. A cornerstone of this philosophy is the exposure of the ‘epistemic fallacy’, namely the reduction of what exists to what we can know of what exists. A concentration on epistemology, or the theory of knowledge, has eclipsed interest in ontology, or the theory of being. Critical realism seeks to recover interest in the ontological dimension of our natural and social worlds.

These worlds are ‘stratified’. Events are made accessible to us via the empirical or experiential; but it is at the level of the real however that the mechanisms that give causal shape to the events we experience are active. The argument at this point is Kantian or transcendental, using retroduction rather than the more conventional mix of deduction and induction. Retroduction takes the following form: certain ‘generative mechanisms’ must exist at the level of the real for us to experience the events we do. These mechanisms are ‘intransitive’ (that is, they exist whether or not they are detected); ‘transfactual’ (that is, they are enduring not transitory); and they govern events. On the face of it this argument seems more plausible in relation to the natural than to the social world. After all Newtonian gravity, or something like it, must exist for apples to fall so predictably from trees when ripe or otherwise disturbed. But Bhaskar makes a parallel case for generative mechanisms in the social world.

Critical realism and sociology

Unlike structures in the natural world like gravity or magnestism, social structures depend for their existence on the actions of individuals. Social structures, for Bhaskar, comprise enduring relations between individuals and groups. To simplify, physicians are only physicians by virtue of their relations with patients and other healthcare workers. Society ultimately consists of a complex lattice-work of relations, and it is the structuring of these relations that affords sociology its subject matter. Marx writes: ‘society does not consist of individuals, but expresses the sum of interrelations, the relations within which these individuals stand’. Marx too gives an apt, proto-critical realist summary of the relationship between structure and agency: ‘men make their own history, but … not under circumstances they themselves have chosen but under the given and inherited circumstances with which they are directly confronted. The tradition of the dead generations weighs like a nightmare on the minds of the living’ (cited in McDonnell et al, 2009: 121-2, 124). It is not for nothing that Bhaskar sees himself as deepening and extending the Marxian project.

In the social world in particular, but not exclusively so, events tend to be ‘unsynchronized with the mechanisms that govern them’, and ‘conjointly determined by various, perhaps countervailing influences so that governing causes, though necessarily ‘appearing’ through, or in, events can rarely be ‘read straight off’’ (Lawson, 1997: 22). The governing causes, or generative mechanisms, can rarely be ‘read straight off’ because they only manifest themselves in open systems (that is, in circumstances where numerous mechanisms are simultaneously active and there is therefore limited potential for experimental closures). They cannot be identified independently of their effects. Sociology differs from, say, biology, in that the objects of its enquiries not only cannot be identified independently of their effects but do not exist independently of their effects. Moreover, sociology has to accept an absence of spontaneously occurring, and the impossibility of creating, experimental-type closures. This denies sociology, ‘in principle’, decisive test situations for their theories. Since the criterion for the rational corroboration of theories in sociology cannot – after the positivist imperative – be ‘predictive’, it must be ‘exclusively explanatory’ (Bhaskar, 1989). Thus explanation displaces prediction; and to explain a phenomenon is to provide an account of what might be termed its ‘causal history’.

Creaven (2000) has argued that a strong explanatory account of human nature, and of the ‘non-social subject’, is indispensable in providing ‘micro-foundations’ for the theory of social structures, cultures and human agency. He develops this transcendentally, contending that human nature and the non-social subject denote ‘an ensemble of species powers, capacities, dispositions and psycho-organic needs and interests’ that ‘logically must be held to exist in order to account for the existence of human society’ (Creaven, 2000: 139). At the same time as humanity’s species-being and its attendant powers and capacities are transmitted upstream into socio-cultural relations and interaction, constraining, enabling and ‘providing a certain kind of universal impetus towards the universal articulation of particular kinds of cultural norms or principles’, social structural, cultural and agential inputs are transmitted downstream to human persons, investing them with specific social interests, shaping their make-up, and furnishing them with cultural resources for identity-formation. Thus are human beings constituted (Creaven, 2000: 1.40-141).

Critical realism and emancipation

But critical realism is not just concerned with identifying the structured relations that are causally efficacious for events. Echoing Marx, critical realists maintain that the point is to change the world not merely to interpret it. In as far as critical realist sociology ‘exposes’ structured relations, it affords an opportunity also to address them and move on. Structures that enhance human freedom and dignity might be reinforced, those that inhibit them contested or minimized (Bhaskar, 1989). As Bhaskar readily acknowledges, however, this is a far from straightforward task:

‘to be free is (i) to know, (ii) to possess the opportunity and (iii) to be disposed to act in (or towards) one’s real interests … Emancipation … is both causally presaged and logically entailed by explanatory theory, but … can only be effected in practice’ (quoted in McDonnell et al, 2009: 126).

Marx might be discerned here too, reflecting Bhaskar’s early intent to provide a more satisfactory philosophical basis for a Marxist theory of society than existed hitherto; although it is a moot point whether or not critical realism ‘must’ be read this way.

Critical realism and chronic illness

Sociologists drawing on critical realism to explore the domains of health, illness, medicine and healthcare have increased in number and ambition over the last decade or so. A pioneering and at the time controversial attempt to explore options was made by Simon Williams (1999). After the manner of Creaven, he reminds sociologists inclined to forgetfulness that the body, ‘diseased or otherwise’, is a real entity, whatever we call it and however we observe it. The body possesses its own mind-independent structures and mechanisms: ‘it has an ontological depth independent of epistemological claims, right or wrong, as to its existence’ (Williams, 1999: 806). Disease labels/diagnostic categories are descriptive, not constitutive of disease itself.

Scambler et al (2010) apply a critical realist perspective to studies of the impact of epilepsy on quality of life. They argue that epilepsy-related quality of life (ERQOL) involves: (1) a mix of the biological, psychological and social, causality travelling both upstream and downstream; (2) the ‘interaction’ of biological, psychological and social mechanisms; and (3) the intrusion of contingency as well as of agency. It is generally accepted that biological, psychological and social mechanisms can each affect ERQOL, whether this is understood objectively or subjectively. But the causal history of ERQOL can be complex. For example, deleterious effect of epilepsy can be present in the absence of any salient biological structures or mechanisms. This is the rule rather than the exception when epilepsy is misdiagnosed: the mere application of the diagnostic label can trigger psychological and/or social mechanisms to negative effect. If epilepsy is but one symptom of a particularly severe underlying pathology on the other hand, biological mechanisms can cancel out or override the causal relevance of psychological and social mechanisms for ERQOL. Or genetic predisposition and brain insult can be mediated by psychological mechanisms that over time ‘decide’ ERQOL.

In other words, biological, psychological and social structures and mechanisms can vary in their causal efficacy for ERQOL from individual to individual as well as by context. Moreover they interact: one genus, acting upstream or downstream, can ‘cancel out’ or ameliorate the impact of others. Danemark et al (2002: 55) discriminate helpfully between ‘objects’, ‘structures’, ‘powers’, ‘mechanisms’ and ‘tendencies’: ‘the objects have the powers they have by virtue of their structures, and mechanisms exist and are what they are because of this structure; this is the nature of the object. There is an internal and necessary relation between the nature of an object and its causal powers and tendencies. This can be expressed as follows (Collier, 1994: 43): ‘things have the powers they do because of their structures … Structures cause powers to be exercised, given some input, some ‘efficient cause’, eg the match lights when you strike it’. This in turn is an example of a mechanism having generated an event. A mechanism is that which can cause something in the world to happen, and in this respect mechanisms can be of many different kinds’.

So a generative mechanism operates when it is being triggered. Unlike the internal and necessary relation between objects and their causal powers, however, the relation between causal powers or mechanisms and their effects is external and contingent. This is because there are many biological, psychological and social mechanisms that are concurrently active. Thus ERQOL is a complex effect of influences emanating from an array of multi-level mechanisms, with some mechanisms reinforcing and some frustrating others. Danemark et al (2002: 56) again:

‘Taken together this – that objects have powers whether exercised or not, mechanisms exist whether triggered or not and the effects of the mechanisms are contingent – means we can say that a certain object tends to act or behave in a certain way’.

Scambler et al (2010) conclude their discussion of ERQOL with three arguments in favour of a critical realist approach. First, it recognizes and allows for an adequate ontology of objects, powers/mechanisms and tendencies in open systems, and does so without falling foul of the naturalistic fallacy (the collapsing of society into biology or human nature). Second, it demands a move beyond the positivistic pursuit of statistical associations between variables, be they biological, psychological, social or a mix. And third, it requires methodological rigour while challenging the positivistic assumption that phenomena can be predicted, and thereby explained, given empirical study of the ‘actual, that is, without resort to Bhaskar’s ‘real’.

In their commentary on disability research, Bhaskar and Danemark (2006) open with a critique of the reductionism implicit in the historically dominant models. They maintain that the phenomenon of disability has the character of a ‘necessarily laminated system’. This alludes to the fact that ‘physical’, ‘biological’, ‘psychological’, ‘psycho-social’, ‘socio-economic’, ‘cultural’ and ‘normative’ mechanisms, contexts and characteristic effects are all essential to the understanding of a phenomenon like disability. This involves recognition of multiple levels of reality (their list is more comprehensive than Scambler et al’s ‘shorthand’ of the biological, psychological and social). Moreover they go on to discern considerable complexity at the level of the social, suggesting that there are four dimensions pertinent to every social event: material transactions with nature; social interactions between agents; ‘social structure proper’; and the stratification of embodied personalities of agents. ‘To say that all social events involve each of these dimensions is to say that social events and social systems are ‘laminated’; and we could call explanations involving mechanisms at several or all of these levels ‘laminated explanations’’ (Bhaskar & Danemark, 2006: 289). So the level of the social also has its levels. They then identify a number of aspects of ‘agency and collectivity’ that disability theorists and medical sociologists alike should be cognisant of:

  • the sub-individual or psychological, as invoked in Goffman;
  • the individual or biographical;
  • micro- and small-group analysis, ventured by ethnomethodologists and others;
  • meso-analysis focused on the relations between functional roles such as disabled worker and capitalist;
  • macro-analysis oriented to relationships, ‘whole societies’ like contemporary financial capitalist society;
  • ‘mega-analysis’ aimed at traditions and civilisations and encompassing notions like normality;
  • ‘planetary analysis’, for example of the impact of globalisation on disability.

Critical realism and health inequalities

Critical realism has also been pressed into service to underpin the sociology of health inequalities. While positivistic research has sought and been satisfied with statistical associations between assorted socio-economic classifications (SECs) and health and longevity, critical realism encourages retroductive inference to the existence of ‘real’ objects, structures/mechanisms and tendencies. In this vein Scambler (2001, 2007) argues that there must exist a new dynamic involving structures or mechanisms of class and command if we are to explain positivistic ‘demi-regs’ linking SECs with health and longevity. Since the early to mid-1970s capitalism has entered a novel phase. One core feature of this new phase of ‘financial capitalism’ has been a resurgence of (rapidly globalizing) class relations relative to the (more slowly globalizing) command relations of nation-states. Class as a generative mechanism has gained in causal salience for health and longevity in the ‘post-welfare statist’ era of financial capitalism (although paradoxically class has become less salient for identity-formation over this same period). It is argued that the core of the capitalist-executive now exercises greater influence over the power elite of the state, hence the expansion of inequality in all its guises. The effect of this is to stem the flow of assets known to be protective for health – biological, psychological, social, cultural, spatial, symbolic and material – amongst the more vulnerable and disadvantaged. It is the growing asymmetry between class and state, it is suggested, that is the key for a sociological explanation of the recent rapid growth in health inequalities manifested in SEC and health/longevity correlations. It has been argued, further, that this thesis on health inequalities in Western nation-states might be extended to emphasize the significance of (1) the ‘contradictions of capitalism’ and of (2) ‘class conflict’ for health and longevity on the global stage (Scambler,2012).

Concluding comments

Sociologists of health, illness and medicine attracted to critical realism emphasize its philosophical infrastructure. What Bhaskar refers to as the ‘holy trinity’ is appealing to many: ontological realism (or the acknowledgement of an independent, multi-level and complex ‘reality’ irreducible to human attempts to capture it in once-and-for-all discourses); epistemological relativism (or the acceptance that what we know is bounded by time and place); and judgemental rationality (or the realisation that ‘epistemological relativism’ does not preclude deciding ‘rationally’ between rival theories about what is happening and why). For others, however, the jury is out. They remain to be convinced of the credibility or potency of retroductive versus inductive and deductive inference. For many the balance between ‘making a theoretical case’ and ‘making an empirical care’ is pivotal.


Bhaskar,R (1978) A Realist Theory of Science. Hemel Hempstead; Harvester Wheatsheaf.

Bhaskar,R (1989) The Possibility of Naturalism: A Philosophical Critique of the Contemporary Human Sciences. Hemel Hempstead; Harvester Wheatsheaf.

Bhaskar,R (2016) Enlightened Common Sense: The Philosophy of Critical Realism. London; Routledge.

Bhaskar,R & Danemark,B (2006) Metatheory, interdisciplinarity and disability research: a critical realist perspective. Scandinavian Journal of Disability Research 8 278-297.

Creaven,S (2000) Marxism and Realism: A Materialist Application of Realism in the Social Sciences. London; Routledge.

Danermark,B, Ekstrom,M, Jakobsen,L & Karlsson,J )2002) Explaining Society: Critical Realism in the Social Sciences. London; Routledge.

McDonnell,O, Lohan,M, Hyde,A & Porter,S (2009) Social Theory, Health and Healthcare. London; Palgrave Macmillan.

Scambler,G (2001) Critical realism, sociology and health inequalities: social class as a generative mechanism and its media of enactment. Journal of Critical Realism 4 35-42.

Scambler,G (2007) Social structures and the production, reproduction and durability of health inequalities. Social Theory and Health 5 297-315.

Scambler,G (2012) Review article: health inequalities. Sociology of Health and Illness 34 130-146.

Scambler,G, Afentouli,P & Selai,C (2010) Discerning biological, psychological and social mechanisms in the impact of epilepsy on the individual: a framework and exploration. In Eds Scambler,G & Scambler,S: New Directions in the Sociology of Chronic and Disabling Conditions: Assaults on the Lifeworld. London; Palgrave Macmillan.

Williams,S (1999) Is anybody there? Critical realism, chronic illness and the disability debate. Sociology of Health and Illness 21 797-819.

Leave a Reply