Looking Back: In defence of inclusive realism in psychology

David Pilgrim offers an alternative to radical constructionism and naive realism

Fact-claiming’ has been central to the rhetoric of justification of modern psychology in Britain.

For example, on the very page of the very first editorial of the British Journal of Psychology in 1904 we find this statement of intent: Psychology, which till recently was known among us chiefly as Mental Philosophy and was mainly concerned with problems of a more or less speculative and transcendental character, has now at length attained the position of a positive science; one of special interest to the philosopher, no doubt, but still independent of his control, possessing its own methods, its own specific problems and a distinct standpoint altogether its own. ‘Ideas’ in the philosophical sense do not fall within its scope; its inquiries are restricted entirely to ‘facts.’ (emphasis added)

In this important statement, issued in the early days of the British Psychological Society, psychology is intended to be: a positive science (asserting its positivist credentials); independent from philosophy in the academy; not idealist (in the philosophical, not vernacular, sense); and, accordingly, now only interested in facts. By 1939 only six Chairs existed in Britain in psychology, and this was largely because of efforts on the part of philosophy to restrict its development (Hearnshaw, 1964). And yet, the editorial statement also reflected the practical implications of a particular form of philosophy – British empiricism.  

The ‘postmodern turn’ of the past 30 years or so can, in turn, be seen as part of a historically derived tension between idealist and empiricist aspirations about human understanding. The radical constructionism associated with the French post-structuralism of Foucault (1973), Derrida (1982) and Lyotard (1984) then found its way into an understanding of psychological phenomena during the 1980s in Britain (e.g. Parker et al., 1995; Rose, 1989) and can be seen as a return to philosophy. But, by this time, the orthodoxy of British psychology had, by and large, followed the course set for it by the founders of the BPS and articulated in the stall-setting exercise of that early editorial.
This ‘French’ attack was only joining other idealist movements that had already challenged disciplinary orthodoxy. For example, Kelly’s personal construct theory and Rogers’ phenomenology can be traced to North American humanism and, to an extent, the pragmatism of Dewey (Butt, 2005). Methodological arguments (about nomothetic versus idiographic assessments) were a symptom not the deeper fracture. The latter was philosophical and about the privileging
of idealism or empiricism.   

British empiricism created a particular cultural context of legitimacy for the psychological orthodoxy being encouraged. This became evident when it had to deal with another ‘alien’ approach from Continental Europe. Psychoanalysis was contained on the margins of metropolitan psychology, until it was virtually expunged by British empiricist orthodoxy: hunted out and skewered as a form of pseudo-science in the academy and clinic (Eysenck, 1953). An irony and unintended consequence of British empiricism being the main philosophical basis for modern psychology in the UK is that it diminished a proper philosophical sensibility in the new discipline and its development. (How many psychology graduates today could properly articulate the philosophical basis of their discipline?) If the philosophers were retarding the growth of psychology as a discipline, then once independence was achieved, philosophy was no longer required and it could be rejected, with the serious risk of its content and value then being lost. And so, when philosophy returned in the ‘postmodern turn’ within human science, a minority of psychologists could embrace and champion a new and critical heterodoxy. However, the majority could simply ignore it from the comfort of a traditional orthodoxy, with their own philosophical legacy simply being presumed legitimate and with no need of a defence.

The fate of naive realism and the rise of social constructionism  
In this disciplinary context after the 1980s, Stam (2001), the editor of Theory and Psychology, noted that a ‘disconnect’ was appearing between the extensive theoretical arguments, which were heavily reliant on philosophical justifications for versions of social constructionism, and the dominant practices and practical rationales in academic and applied psychology. The battle for one version of realism had been won by the historical founders of fact-claiming psychology, in the early 20th century; the course of disciplinary legitimacy was both set and correct. Philosophical justification could now be overlooked or considered only as a historical curio. Maybe philosophy had simply outlived its utility for most academic and applied psychologists and only the social constructionists now delighted in the legitimacy of philosophy and the dense writings that laboured beneath. But, eventually, in response to Stam’s musing, or lament, came the retort of the situational realists (Mackay & Petocz, 2011).

They built on the work of the Scottish-Australian philosopher, John Anderson (1927), and his ‘situational realism’. Unlike the positivists, Anderson did not eschew metaphysics, because early Greek, unlike Cartesian, philosophy focused on being rather than on knowledge (Hibberd, 2009). He relied on pre-Socratic philosophy, and especially that of Heraclitus. The latter emphasised that the world is highly complex and in constant flux, hence his most well-known aphorism: ‘No one can step into the same river twice’.

For Anderson, causality is about ‘processes continuing into one another’, a view also held by William James. All psychological phenomena are viewed as located in space and time, and can only be studied scientifically with this specifiable condition in mind. Dynamic flux and the particularities of situations then should be the focus of psychological inquiry (with those particularities assumed to reflect general processes in context). Following Heraclitus, the process of change, not general laws derived from the identification of permanent or replicable events, should be its priority for investigation.  

This guidance from Anderson had different implications for the legitimate scope of psychology inquiry, compared with those articulated in the 1904 editorial. Fact-claiming remains feasible and desirable but ontological claims are now always situated and methodological pluralism is implied. How we might study situated complexity depends on the question we are pursuing about it.

Thus both situational realists and postmodern psychologists agree fundamentally on the matter of context. But they part company emphatically over the matter of ontology. All realists assume that what Anderson called external ‘states of affairs’ exist separately from the knower. Situations precede, proceed, and are independent of, knowing subjects. By contrast, constructionists emphasise that reality, though present in principle, is unknowable in practice and so ‘reality’ becomes a socially constructed epiphenomenon of human understanding (representations and forms of knowledge-in-action, or what Foucault called ‘discursive practices’).

Two points then flow from this constructionist position. First, nothing can be said about reality in and of itself (contra the fundamental position of all realists). Second, the focus of psychology should not be on fact-claiming but on understanding the situated socio-linguistic conventions attending human relationships, because the ‘transcendent domain of the real’ is unknowable (Gergen, 2001).

Let us consider those socio-linguistic conventions. For constructionists such as Gergen, statements are simply there as satisfactory performances, which suit the speaker in their particular relational context with others. On behalf of all realists, Hibberd (2002, p.288) disagrees profoundly with this limited and restrictive conclusion, when noting that statements may also legitimately, as they say, ‘tell it like it is’. That is, statements can truly advance our understanding of reality itself in particular situations. Ontology is thereby rescued and not scorned nihilistically as an unknowable mystery.

However, this does not mean that fact-claims are simple and non-problematic. Critical realism, following Bhaskar (1989, p.79), notes that human science is embedded in its host society and so has to proceed with extreme scepticism. It must recognise that social structures, unlike natural structures, do not exist independently of the activity they govern, or of the agents’ conceptions of what they are doing in their activity; and the tendencies they ground may not be universal in the sense of a space-time invariant.

For realists in psychology this summary of their own context has three major implications if they are to work with a sense of reflective integrity. First, by trying to understand ‘structures’ (relatively stable aspects of reality) psychologists will fail unless they are constantly aware of context, including their own. This is the point of consensus here with postmodern critics: all human activity is part of an open system and cannot be understood by deploying the assumptions of closed systems inquiry (Bateson, 1972), a point emphasised by both situational realism and critical realism (see Hibberd, 2010, and Hartwig, 2007, for distinctions though between these versions of realism.)  

Second, and following this point about context, we have to recognise
that our field of inquiry is governed by interests and values. We pose certain research questions and answer them in accordance with what sociologists call ‘interest work’ in our extant context (i.e. we are part of groups, or work for them, that pursue particular cognitive, ideological and financial interests).

We ask questions at the behest of our paymasters, and our seemingly individual imagination and curiosity are actually determined by social and economic processes. For this reason, whilst reality exists independent of the knower, the latter’s motivation for, and way of understanding, their tasks (as research or applied psychologists) are context-derived. Why do we pose some research questions and not others? When we gather data why do we interpret in this rather than another way? Why do we attend to some implications of our findings and not others and want to pursue further research in this, rather than that, way? It is only after critical reflection on these questions has been pursued honestly and thoroughly that we can start to be confident in our knowledge claims about reality.

Third, and following from this type of healthy sceptical reflection, the ‘agent’s conceptions’, including that of psychologists, are prone to two fundamental errors: the ontic fallacy and the epistemic fallacy. The former refers to the tendency of us to naively accept what we see and hear, without being aware of the acculturated biases and errors accruing when processing sense-data. The epistemic fallacy is when we confuse reality with what we choose (in our current culture) to call reality. The latter fallacy is rife in human science, when it creates categories and seeks generalisations about them, rather than focusing on processes and situations.

For example, ‘depression’ or ‘schizophrenia’ have no a priori ontological status, but psychologists may naively believe that they do and, accordingly, study the supposed embodied expressions of these putative ontological states in ‘depressed’ or ‘schizophrenic’ subjects (Bentall et al., 1988; Pilgrim & Bentall, 1999). This is a fundamental scientific error in the eyes of critical realists. The map is not the territory. The territory does exist; it is just very challenging to investigate (Korzybski, 1931).

The reality of distress is that many people are profoundly miserable at points in their lives. But they do not ‘have depression’. The challenge is to work out with a person and their significant others, in their particular circumstances, why they are feeling so distressed by attending both to the meanings they ascribe to their presenting problem and the actual events in their lives, including current pressures, past losses or insults and future threats that might be causally relevant to their current misery. Realism, unlike constructionism, permits the second part of this logic of inquiry, not just the first. Similarly, people do not ‘have schizophrenia’ but some people act and speak unintelligibly and this creates fear in others and the disruption of their expectations about how life should be proceed in particular contexts of the home, the street and the workplace. The disruption that emergent madness creates is real. And it has antecedent trends (especially in early abuse and family communication systems but probably,as well, in biological vulnerability). 

Misery and madness are constituted by real events in everyday life. They are not ‘socially constructed’ by-products of psychiatric discursive practices. They are real enough in their situated contexts, when people recognise distress or unintelligible speech and conduct. But a good realist psychologist would also recognise that dubious categories like ‘depression’ or ‘schizophrenia’ reflect the epistemic fallacy and so would avoid falling into a version of poor science. Inclusiveness as an implication of a return to realism
Social constructionism and naive realism have led us into cul-de-sacs of differing sorts. In the first case, reality is deemed to be unknowable, leading some of its critics to describe it as a form of ‘psychosis’ (Craib, 1997). This refusal of reality, as reality, has profound implications both for science and justice about the ‘states
of affairs’ we bear witness to in our lives (Pilgrim, 2000).

In this regard, are we not ethically obliged to develop methods of inquiry which lead us to fact-claims that are reasonably confident? For example, if those methods are pre-empted by a rejection of ontology, then how would we know whether 8000 Muslim boys and men were really murdered in Srebrenica? How would we ever judge whether or not Ratko Mladic was probably ultimately responsible for that genocide? And in the routines of psychology departments, if we took social constructionism too literally and too seriously, then we would deny the relevance of neuroscience to our understanding of experience and behaviour and have no confidence in any empirical methodology.

These judicial and scientific consequences of strong constructionism reflect anti-rationalism and nihilism, and it is difficult to then appreciate precisely in what way human progress is being advanced. In reaction to this risk, some have sought to rescue the project of the Enlightenment, from the ravages of postmodern nihilism, in order to respect genuine scientific progress (e.g. Habermas, 1987). Social constructionism is appraised by its advocates as a source of radicalism (e.g. Fox & Prilleltensky, 1997). But its opponents argue that it is ultimately conservative in its consequences, despite its honest critical intentions (e.g. Callinicos, 1989).

Notwithstanding the protests about nihilism and unintended conservatism, there is a paradox here that needs to be noted in partial defence of postmodern human science. The naive realism I am about to address was diverted into a frame of reference that was allegedly value-free, but this cannot be so: psychology is embedded in a set of relationships in society, which are value-derived and value-saturated. In no small part, some criticisms from social constructionists began to move the discipline of psychology ‘beyond the myth of detached neutrality to discover virtue and to recognise politics as forces which determine ethical behaviour’ (Pettifor, 1996, p.1). (The problem was that postmodernists then threw the baby out with the bathwater by rejecting ontology in psychological inquiry.)

Thus naive realism (the ‘positive science’ view from 1904) has taken us in the erroneous direction of scientism. This has meant that values have to be added after the event (for example in ethical codes), rather than being recognised as being ubiquitous, at all stages of psychological inquiry.

As well as this mistaken ‘detached neutrality’, naive realism cannot deliver on the predictive generalisations it aspires to in its work. It has been wrong to pursue the fantasy of generalisability because this is logically and empirically pre-empted by the real character of open systems. Open systems are ultimately unpredictable but they can be studied and provisional statements can be made about what is happening in them and probability statements made about particular outcomes (von Bertalanffy, 1969; Weiss, 1977). The latter are not random, but predictions can only be improved by adding more and more data under shifting objective conditions, and ultimately uncertainty must still be tolerated.

Adding to the dilemma of context-specific predictions is the fact that in psychology we are studying reflective and meaning-seeking subjects: language itself is a real input in the system. We are like scientists who have test tubes that talk back at us. (The constructionist Bannister, 1966, made a fair comment in this regard, in an early critique of psychological orthodoxy.) But the social constructionists, with their linguistic preoccupations, then inevitably over-value language. This itself can develop into a form of reductionism. Language is an important determinant in open human systems, but it is not the only determinant. Other real determinants need to be considered as well.

Psychology could now pull back from the aspiration about being a strong predictive science and proceed cautiously about multiple causal or generative mechanisms in dynamic flux that might be discerned reasonably within particular human contexts (Pawson & Tilley, 1998). Moreover, the preoccupation with methodological purity and a putative hierarchy of knowledge, which privileges quantification and the ‘controlling out’ of some variables, have deemed some forms of inquiry scientifically dubious or unworthy. This restrictive rhetorical starting point has ruled out of court what our host society and fellow citizens might reasonably expect from us.

Starting with the intriguing insights of both ethnomethodology and existentialism we can note that ordinary people with no formal psychological knowledge succeed quite well in life, when understanding their particular situations. When they need to, they describe, anticipate and estimate (for example in dealing with an approaching real car and its real speed when crossing a real street). When they need to, they interpret events and one another (for example in their anxious sense-making about the sudden change of habit of a colleague or relative). When they need to, they explore their feelings and seek meaning in their lives. When they need to take action in relation to all of the above aspects of doing life, they may do so with consummate ease or extreme difficulty, but always without any formal psychological knowledge.

They know the difference immediately between loving eggs on toast and loving their grandchildren (Butt, 2012). Equally they know that a feather dropping on their head will not cause them brain damage but a falling rock may well do so. Both causes and meanings are part of life and so why should psychology want to skew its activity more to one than the other? To skew in either direction is disrespectful to our fellow human beings.

In the language of philosophy, left to us by Dilthey, there is a place for both erklaren (the explanation of causes) and verstehen (the attention to interpreted meanings) in psychology. Explanations are possible but they must be offered very cautiously. At the same time, description and understanding in particular contexts is always required. Context is all important for both, and prediction becomes highly risky in the pull between them. The weighting we might give to either depends on the question we are addressing. The method should follow the question. Given the vast range of ordinary human inquiry, decision making and interpretation, it is clear that psychology could legitimately extend its range to it all, but should do so in a systematic way that respects both causes and meanings.

If ordinary people deal with the complicated intersection in their lives of the symbolic, the imaginary and the real, then surely the discipline of psychology has an obligation to mirror and improve on those efforts.

It is quite possible to have a realistic view (in the philosophical sense) about psychological inquiry, while embracing methodological pluralism. Indeed realism should require the latter. Neuroscience should have its place at the table, but so should psychoanalysis and existentialism. But all should proceed with the critical reflection I have suggested, and equal attention could be given to both ontology and epistemology. Philosophy (the old enemy?) still has much to offer us when negotiating this middle way between naive realism and social constructionism.

David Pilgrim is Professor of Health and Social Policy in the Department of Sociology  at the University of Liverpool [email protected]


Anderson, J. (1927). Studies in empirical philosophy. Sydney: Angus and Robertson.
Bannister D. (1966). Psychology as an exercise in paradox. Bulletin of the British Psychological Society, 19, 21.
Bateson, G. (1972). Steps to an ecology of mind. Chicago: University Of Chicago Press.
Bentall, R., Jackson, H. & Pilgrim, D. (1988). Abandoning the concept of schizophrenia: Some implications for validity arguments about psychotic phenomena. British Journal of Clinical Psychology, 27, 303–324.
Bhaskar, R. (1989). The possibility of naturalism. London: Routledge.
Butt, T.W. (2005). Personal construct theory, phenomenology and pragmatism. History and Philosophy of Psychology, 7, 23–35.
Butt, T.W. (2012). Commentary. Conference of the History and Philosophy Section of the British Psychological Society. Oxford.
Callinicos, A. (1989). Against postmodernism. London: Routledge.
Craib, I. (1997). Social constructionism as a social psychosis. Sociology 31, 1–15.
Derrida, J. (1982). Margins of philosophy. Brighton: Harvester.
Eysenck, H.J. (1953). Uses and abuses of psychology. Harmondsworth: Penguin.
Foucault, M. (1973). The order of things: An archaeology of the human sciences. New York: Vintage Books.
Fox, D. & Prilleltensky, I. (Eds.) (1997). Critical psychology: An Introduction. London: Sage.
Gergen, K. (2001). Construction in contention: Towards consequential resolutions. Theory and Psychology, 11, 419–432.
Habermas, J. (1987). The philosophical discourse of modernity. Cambridge, MA: MIT Press.
Hartwig, M. (Ed.) (2007). Dictionary of critical realism. London: Routledge.
Hearnshaw, L.S. (1964). A short history of British psychology. London: Methuen.
Hibberd, F.J. (2002). Reply to Gergen. Theory and Psychology 12(5), 685–694.
Hibberd, F.J. (2009). John Anderson’s development of (situational) realism and its bearing on psychology today. History of the Human Sciences, 22(4), 63–92.
Hibberd, F.J. (2010). Situational realism, critical realism, causation and the charge of positivism History of the Human Sciences, 23(4), 47–51.
Korzybski, A. (1931). A non-Aristotelian system and its necessity for rigour in mathematics and physics. Paper presented before the American Mathematical Society at the New Orleans meeting of the American Association for the Advancement of Science, 28 December.
Lyotard, J-F. (1984). The postmodern condition: A report on knowledge. Manchester: Manchester University Press.
Mackay, N. & Petocz, A. (Eds.) (2011). Realism and psychology: Collected essays. Boston, MA: Brill.
Parker, I., Georgaca, E., Harper, D. et al. (1995). Deconstructing psychopathology. London: Sage.
Pawson, N. & Tilley, P. (1998). Realistic evaluation. London: Sage.
Pettifor, J.L. (1996). Ethics: Virtue and politics in the science and practice of psychology. Canadian Psychology, 37, 1–12
Pilgrim, D. (2000). The real problem for postmodernism – invited millennium essay. Journal of Family Therapy, 15(1), 1–10.
Pilgrim, D. & Bentall, R.P. (1999). The medicalisation of misery: A critical realist analysis of the concept of depression. Journal of Mental Health, 8(3), 261–274.
Rose, N. (1989). Governing the soul. London: Routledge.
Stam, H.J. (2001). Introduction: Social constructionism and its critics. Theory and Psychology, 11, 3, 291–296.
von Bertalanffy, L. (1969). General system theory: Foundations, development, applications. New York: Braziller.
Weiss, P.A. (1977). The system of nature and the nature of systems: Empirical holism and practical reductionism harmonized. In K.E. Schaefer, H. Hensel & R. Brady (Eds.) A new image of man in medicine, Vol. I: Towards a man-centered medical science. New York: Futura.

BPS Members can discuss this article

Already a member? Or Create an account

Not a member? Find out about becoming a member or subscriber