So, you have pledged allegiance to critical realism – what next?

Critical realism, initiated by Roy Bhaskar (1944–2014), is a popular package of meta-theories and principles to help guide how we conduct research. It is often framed as an alternative to positivism and postmodernism. Margaret Archer and eight fellow critical realists (2016) composed a helpful summary of four key critical realist principles:

  • Ontological realism
  • Epistemic relativity
  • Judgemental rationality
  • Ethical naturalism

Here are some thoughts on what these may mean for the everyday work of conducting social research and evaluations.

1. Ontological realism

What is it? There is a social and material world existing independently of people’s speech acts. “Reality is real.” One way to think about this slogan in relation to social kinds like laws and identities is they have a causal impact on our lives (Dembroff, 2018). Saying that reality is real does not mean that reality is fixed. For example, we can eat chocolate (which changes it and us) and change laws.

What to do? Throw radical social constructionism in the bin. Start with a theory that applies to your particular topic and provides ideas for entities and activities to use and possibly challenge in your own theorising.

Those “entities” (what a cold word) may be people with desires, beliefs, and opportunities (or lack thereof) who do things in the world like going for walks, shopping, cleaning, working, and talking to each other (Hedström, 2005). The entities may be psychological “constructs” like kinds of memory and cognitive control and activities like updating and inhibiting prepotent responses. The entities might be laws and activities carried out by the criminal justice system and campaigners. However you decide to theorise reality, you need something.

How an intervention may influence someone’s actions by influencing their desires, beliefs, and/or opportunities (Hedström, 2005, p. 44)

2. Epistemic relativity

What is it? The underdetermination of theories means that two theorists can make a compelling case for two different accounts of the same evidence. Their (e.g., political, moral) standpoint and various biases will influence what they can theorise. Quantitative researchers are appealing to epistemic relativity when they cite George Box’s “All models are wrong” and note the variety of models that can be fit to a dataset.

What to do? Throw radical positivism in the bin – even if you are running RCTs. Ensure that you foreground your values whether through statements of conflicts of interest or more reflexive articulations of likely bias and prejudice. Preregistering study plans also seems relevant here.

There may be limits to the extent to which an individual researcher can articulate their biases, so help out your colleagues and competitors.

3. Judgemental/judgmental rationality

What is it? Even though theories are underdetermined by evidence, there often are reasons to prefer one theory over another.

What to do? If predictive accuracy does not help choose a theory, you could also compare them in terms of how consistent they are with themselves and other relevant theories; how broad in scope they are; whether they actually bring some semblance of order to the phenomena being theorised; and whether they make novel predictions beyond current observations (Kuhn, 1977).

You might consider the aims of critical theory which proposes judging theories in terms of how well they help eliminate injustice in the world (Fraser, 1985). But you would have to take a political stance.

4. Ethical naturalism

What is it? Although is does not imply ought, prior ought plus is does imply posterior ought.

What to do? Back to articulating your values. In medical research the following argument form is common (if often implicit): We should prevent people from dying; a systematic review has shown that this treatment prevents people from dying; therefore we should roll out this treatment. We could say something similar for social research that is anti-racist, feminist, LGBTQI+, intersections thereof, and other research. But if your research makes a recommendation for political change, it must also foreground the prior values that enabled that recommendation to inferred.

In summary

The four key critical realist principles provide a handy but Big metaphysical and moral framework for getting out of bed in the morning and continuing to do social research. Now we are presented with further challenges that depend on grappling with substantive theory and specific political and moral values. I wish you the best of luck on your endeavour!

References

Archer, M., Decoteau, C., Gorski, P. S., Little, D., Porpora, D., Rutzou, T., Smith, C., Steinmetz, G., & Vandenberghe, F. (2016). What is Critical Realism? Perspectives: Newsletter of the American Sociological Association Theory Section, 38(2), 4–9.

Dembroff, R. (2018). Real talk on the metaphysics of gender. Philosophical Topics, 46(2), 21–50.

Fraser, N. (1985). What’s critical about critical theory? The case of Habermas and gender. New German Critique35, 97-131.

Kuhn, T. S. (1977). Objectivity, Value Judgment, and Theory Choice. In The Essential Tension: Selected Studies in Scientific Tradition and Change (pp. 320–339). The University of Chicago Press.

Hedström, P. (2005). Dissecting the social: on the principles of analytic sociology. Cambridge University Press.

Social Sciences under Attack in the UK (1981-1983)

Interesting paper by Michael Posner, who was chair of the UK Social Science Research Council (SSRC) when it was under attack by the Conservative Thatcher government in the early 1980s.

Secretary of State Sir Keith Joseph considered dismantling SSRC and asked for an independent review into its utility by an established biologist.

SSRC survived, though one notable change was made…

“Joseph opted for a public, but very light punishment: a change of name. I told him that I could persuade scores of academics to accept a name change if he would promise, on the record, the continuing independence of the SSRC. He agreed, and the SSRC was duly renamed the « Economic and Social Research Council » (ESRC). The significance of this change was the omission of the word « science », which Joseph had insisted upon and which many of us at the council and in academia found it difficult to accept.”

Read more over here.

A more daring approach to writing theory

“What if we took a more daring, modernist, defamiliarizing approach to writing theory? What if we asked of theory as a genre that it be as interesting, as strange, as poetically or narratively rich as we ask our other kinds of literature to be? What if we treated it not as high theory, with pretentions to legislate or interpret other genres, but as low theory, as something vulgar, common, even a bit rude—having no greater or lesser claim to speak of the world than any other? It might be more fun to read. It might tell us something strange about the world. It might, just might, enable us to act in the world otherwise. A world in which the old faith in History is no more, but where there are histories that still might be made—in a pinch.”

– McKenzie Wark (2019). Capital is dead.

Metaphysical isms and theorising gender

I had tried to avoid engaging in grand metaphysical “ism” talk, but it seems that resistance is futile! So here are brief thoughts, in the context of theorising gender.

We can safely assume that there is a reality to people’s gender-relevant experiences and biochemistry which exists independently of our understandings. Taking this (to me obvious) stance is known as ontological realism. Theorising, about gender or otherwise, is done by people who have imperfect and indirect access to reality and theories evolve over time. Our vantage point—beliefs, biases, values, experience, privilege and oppression—has an impact on our theories, so two gender theorists doing the best they can with the available evidence can produce very different explanations (epistemic relativism). This is true of any science where multiple theories are consistent with evidence; in other words, the theories are underdetermined by evidence. It is also true when we theorise about ourselves and try to work out our own gender.

Even with this relativist mess, manifesting as bickering in scientific journals and conferences, consensus can arise and one theory can be declared better than another (judgemental rationality). However, there are often many different ways to classify biological, social, and other phenomena, even with impossibly perfect access to reality (this has a great name: promiscuous realism).

The underdetermination of theories means that something beyond evidence is needed to decide how and what to theorise. Scholars in the critical theory tradition are required to pick a side in a social movement, for instance feminism, anti-racism, trans rights, or an intersectional composition thereof. It is not enough for a critical theory to be empirically adequate; it also has to help chosen social struggles make progress towards achieving their aims. Two theories may be empirically indistinguishable but one transphobic; from a trans rights perspective, the transphobic theory should be discarded.

(For more on epistemic relativity, ontological realism, and judgemental rationality, see Archer et al. (2016).)

Now we can make sense of what it means to be assigned female or male at birth. What is assigned is a sex category. This is not arbitrary, but based on socially agreed and – for cisgender people – reliable biological criteria. However, those criteria could have been otherwise, for instance using a broader range of biological features and more than two categories. Also the supposedly biological male/female sex category quickly takes on a social role that is independent of genitals and operates even when they are hidden.

Unsticking social research through lived experience and citizen control

Having lived experience and knowing people with lived experience are really effective way of researching social conditions – unavoidably, whether you want to or not – and lead to rich theory.

In activist groups with chat groups and regular meetings, the “data collection” isn’t actually data collection, but inseparable from day-to-day conversations, support, and campaigning. The “research” is the shared understanding of what’s going on that is maintained by continuous contact.

Compare this to a model of social research where a professional-run central institute runs surveys and writes reports. This approach can lead to flat, superficial theorising if researchers don’t have lived experience of the problems being researched.

But traditional reports can still be important to get media and government attention: “What’s a Nice Girl Like You Doing in a Job Like This?”, written by the English Collective of Prostitutes, a network of sex workers, is a great example of research drawing on lived experience to set priorities and ask the right questions combined with “traditional” research skills.

To unstick social research requires benefiting from methodological advances whilst radically opening up research to citizen control. Sometimes getting a good estimate of the population prevalence and causes of oppression are important to highlight how bad things are and suggest ways to improve things. Advances in techniques and software for qualitative analysis can be useful too and ensure best use is made of testimony.

Academics without lived experience running convenience sample qualitative studies with small numbers of people and pretentious methodology (pages of reflection on whether reality is real and complaining about positivism) are fundamentally limited in what they can discover. But the same sample informed by lived experience, focussing on issues that matter, is very different.

There are many professional researchers with lived experience. Max Weber (1864-1920) was one, with experience of psychiatric inpatient stay:

“Max Weber had made contact with psychiatric thinking as a result of personal experiences. After he had been made professor of economics in Heidelberg and following the death of his father in 1897, he suffered from a severe depressive episode lasting until 1902. After recovering, he became increasingly interested in contemporary philosophical and psychological literature and began to study act theory as a basis of economics.” (Frommer et al., 2000, pp. 346-347)

But higher education is a hostile environment – you couldn’t design a better system to reward junk research and cause burnout if you tried. Such environments are far from ideal for supporting people who have the lived experience needed to help improve the quality of research.

Your various identities, privileges and oppression (racism, sexism, transphobia, homophobia, poverty, etc., and intersections thereof) fundamentally constrain who will answer your calls for research participants, what social phenomena you can understand, who will listen to what you discover. They literally change what you see, hear, and feel, and what you can research.

Some researchers break free of these constraints thanks to contradictory locations; for instance, being articulate and well-connected can be used to resist a position of oppression. Though then you can end up being attacked for having privilege, even by “your own side”.

Academics with more secure positions can help, for instance:

  1. Support PhD students and colleagues who are discriminated against in various ways: grants, decent pay, and mentoring are helpful.
  2. Instead of “giving voice” to people through interview excerpts, give a platform.
  3. Cite blog posts and reports from activists with lived experience, not only peer reviewed journal articles.

Mr Justice Mostyn vs. vague, rhetorical applications of theory

A court case (GM v Carmarthenshire County Council [2018] EWFC 36) has ruled that a social worker’s “generalised statements, or tropes” based on attachment theory are not admissible evidence.

The full judgement by Mr Justice Mostyn has interesting thoughts on the valid application of theory and balance between theory and observation.

“… the local authority’s evidence in opposition to the mother’s application was contained in an extremely long, 44-page, witness statement made by the social worker […]. This witness statement was very long on rhetoric and generalised criticism but very short indeed on any concrete examples of where and how the mother’s parenting had been deficient. Indeed, it was very hard to pin down within the swathes of text what exactly was being said against the mother. […] [The social worker] was asked to identify her best example of the mother failing to meet L’s emotional needs. Her response was that until prompted by the local authority mother had not spent sufficient one-to-one time with L and had failed on one occasion to take him out for an ice cream. […] A further criticism in this vein was that the mother had failed to arrange for L’s hair to be cut in the way that he liked.”

There is also a detailed section on attachment theory:

“… the theory is only a theory. It might be regarded as a statement of the obvious, namely that primate infants develop attachments to familiar caregivers as a result of evolutionary pressures, since attachment behaviour would facilitate the infant’s survival in the face of dangers such as predation or exposure to the elements. Certainly, this was the view of John Bowlby, the psychologist, psychiatrist, and psychoanalyst and originator of the theory in the 1960s. It might be thought to be obvious that the better the quality of the care given by the primary caregiver the better the chance of the recipient of that care forming stable relationships later in life. However, it must also be recognised that some people who have received highly abusive care in childhood have developed into completely well-adjusted adults. Further, the central premise of the theory – that quality attachments depend on quality care from a primary caregiver – begins to fall down when you consider that plenty of children are brought up collectively (whether in a boarding school, a kibbutz or a village in Africa) and yet develop into perfectly normal and well-adjusted adults.”

Much to discuss!

The social model of disability as a case study of social ontology

Picture of a staircase
Photo by Alessia Cocconi on Unsplash

Social ontology is a branch of philosophy that tries to understand the building blocks of the social world. Debates in social ontology can be abstract and seem pointless. Even defining social ontology, and how it differs from, say, sociology, is a challenge (see Epstein, 2021). There has been a case to “rid social sciences of ontology altogether – of all philosophized metaphysics of how the social world is” (Kivinen and Piiroinen, 2007, p.99). This brief post tries to show why social ontology is important, using the social model of disability as an example.

In 1975, the Union of the Physically Impaired Against Segregation, a group of disability activists, published a series of fundamental principles which challenged the ontology of disability:

“In our view, it is society which disables physically impaired people. Disability is something imposed on top of our impairments, by the way we are unnecessarily isolated and excluded from full participation in society. Disabled people are therefore an oppressed group in society. […] For us as disabled people it is absolutely vital that we get this question of the cause of disability quite straight, because on the answer depends the crucial matter of where we direct our main energies in the struggle for change. We shall clearly get nowhere if our efforts are chiefly directed not at the cause of our oppression, but instead at one of the symptoms.”

Here a distinction is made between impairment and disability. From this perspective, it makes no sense to say that someone “has a disability”; individual people can have impairments, but it is society that determines whether someone is disabled. A vivid example of this is how common it still is for buildings not to be wheelchair accessible – or only partly so, e.g., wheelchair users can enter a building but not use its toilets. Note how the conceptualisation is used to unite people behind a social struggle. It has a practical purpose rather than only adding to our knowledge.

A related example is illustrated by the difference between deaf and Deaf identity:

“To be ‘deaf’ (small d) is to fit into the medical definition of deafness as something to be cured and eradicated. Being deaf means you have a hearing loss, but you choose or don’t feel able to function within the Deaf Community. […] Deaf – with a capital “D” (and occasionally with capital E, A and F too) – is used to refer to people who are culturally Deaf. These people actively use British Sign Language; they see themselves as being culturally Deaf and part of the Deaf community. […] I consider myself to be culturally Deaf; this is my Deaf Identity. […] I don’t see it as a disability – there is nothing I feel I cannot do – rather, I see it as an important aspect of my character that makes and shapes me.”

These conceptualisations of impairment and disability, social barriers, adjustments, aids, community, and Deaf identity, concern social ontology. Debates on these topics occur naturally in social struggles and discussions of social policy and identity, whether or not explicitly articulated as being about ontology. They also have clear implication for how social science is carried out and how research findings are used.

More posts on social ontology

References

Epstein, B. (2021). Social Ontology. The Stanford Encyclopedia of Philosophy (Winter 2021 Edition), Edward N. Zalta (ed.).

Kivinen, O., & Piiroinen, T. (2007). Sociologizing Metaphysics and Mind: A Pragmatist Point of View on the Methodology of the Social Science. Human Studies, 30, 97–114.

Brands

Photo by Polina Tankilevitch

 

CBT versus counselling. Mental models versus mental rules. Bayesianism versus everything. There are lots of brand wars in psychology which stifle the development of theory and lead to tedious arguments. This does get recognised in print now and again. Here are some examples.

The first comes from a discussion of the ACT-R theory (Anderson, 2007, p. 19):

“… this book is not about ACT-R; rather I am using ACT-R as a tool to describe the mind. […] We may be proud of our ACT-R models and think they are better than others […] but we try not to lose track of the fact that they are just a way of describing what is really of interest.”

Later (p. 239) Anderson cites Herbert Simon, who complained that

“… ‘brand names’ tend to make difficult the analysis and comparison of […] mechanisms or the exchange of knowledge between research groups. […] Physicists did not divide quantum mechanics into the Heisenberg Brand, the Schrodinger Brand, and the Dirac brand.”

Another example, from sociology: a book on social class (Wright, 1997, p. 34):

“Readers who are highly skeptical of the Marxist tradition for whatever reasons might feel that there is no point in struggling through the numbers and graphs in the rest of this book. If the conceptual justifications for the categories are unredeemably flawed, it might be thought, the empirical results generated with those categories will be worthless. This would be, I think, a mistake. The empirical categories themselves can be interpreted in a Weberian or hybrid manner. […] As is usually the case in sociology, the empirical categories are underdetermined by the theoretical frameworks within which they are generated or interpreted.”

References

Anderson, J. R. (2007). How can the human mind occur in the physical universe? Oxford University Press

Wright, E. O. (1997). Class counts: student edition. Cambridge University Press.

John Fox on SEM

From an Appendix to An R and S-PLUS Companion to Applied Regression:

“A cynical view of SEMs is that their popularity in the social sciences reflects the legitimacy that the models appear to lend to causal interpretation of observational data, when in fact such interpretation is no less problematic than for other kinds of regression models applied to observational data. A more charitable interpretation is that SEMs are close to the kind of informal thinking about causal relationships that is common in social-science theorizing, and that, therefore, these models facilitate translating such theories into data analysis.”