The royal road to sociological truth? Is there a need for critical realism?

A post by Tim Rutzou.

In a recent review of the 2015 books Reconstructing Sociology: The Critical Realist Approach by Douglas V. Porpora and The Relational Subject by Pierpaolo Donati and Margaret S. Archer, Neil Gross “rips critical realism a new one”. Gross takes aim at critical realism as expressed in the work of Porpora, Donati, and Archer, lands a few blows and raises a few good questions, above all, why the need for critical realism? Similar assertions have been made by Kieran Healy who suggests “eat something else”, fill your theoretical “diet” with more accessible engagements, and avoid indigestion by not touching “the low-quality, confusing, misleading” and ultimately dissatisfying “cream cheese” of critical realism. As fabiorojas wants to suggest “Friends don’t let friends do critical realism”.

While I think a lot of their claims and analyses are overstated and misleading (see [endnote 1]), I also think both Gross and Healy are right on a number of fronts. I think it is fair to say that a number of critical realists have historically been prone to overreaction in the name of defending certain theoretical and political stakes. Critical realism has been prone to some gross generalizations. This has included overly strong reactions to forms of social constructionism, relativism, and post-modernism. Critical realism has sustained itself on a rich diet of dense philosophy, sometimes to its detriment. It has lacked nuance. It has been obscure. It has not been critical enough. It has often been too excited by its own claims, too dogmatic and maybe even imperialistic in its tendencies – over-promising and under-delivering. This is to say nothing of Bhaskar’s own descent into a metaphysical system that attempted to encompass the nature of the universe within an arcane framework that has all the gilded grandeur of a truly metaphysical metanarrative, complete with an emancipatory heartbeat of the universe. Given all of that, why would you even touch someone who has even looked at the philosophy of critical realism? It sounds positively awful!

#%@? that!

Well, like all things, some of it is awful, and some of it isn’t. Far from being a unified whole there is not one unitary framework, set of beliefs, methodology, or dogma that unites critical realists into a whole. It would not be wrong to say that “there is no critical realism”. Instead, critical realists are ‘united’ in what more properly resembles a series of family resemblances – and like all families, there is no shortage of disagreement or contention. Very few people are drawn to Bhaskar’s later works and see them as a seriously problematic. Most critical realists are promiscuous, combining insights from critical realism with insights from Bourdieu, Latour, feminist theory, analytical sociology, post-structuralism, pragmatism, and/or critical theory [endnote 2] …. There are a few reasons why engaging with critical realism might not be the nausea-inducing diet of cream cheese it is made out to be by some … but neither is it the royal road to sociological truth as is sometimes implied by others.

Here, I simply want to explore one reason why sociologists might want to engage with critical realism – sociologists, and not just the elite few, should want to be more engaged with philosophy because, ultimately, it allows you to do better and more sophisticated empirical work. Critical realism opens up a space to do this, by emphasizing reflexivity, providing some orienting stances, as well as offering some useful theoretical and ontological concepts to help structure this engagement.

As Gross mentions critical realism positions itself as a meta-theory. But its meta-theory is not simply about providing a critique and alternative to the covering law approaches to explanation. Meta-theory is a much more general reflexive stance concerned with wanting to provide a philosophically informed account of science and social science, which can in turn inform our empirical investigations. In other words, meta-theory is a second order reflection – an attempt to systematize knowledge, and feedback lessons learned in the course of our empirical engagements into a framework that makes more sense of the world.

It is a series of historical and contingent philosophical reflections built upon the systematization of forms of practice and knowledge. It is a fallible attempt to reconstruct and systematize at an abstract level what we know of the world, and things in the world, in light of knowledge gained through experience and practice, including such practices as the natural and social sciences. It is concerned with uncovering the conditions of possibility for particular experiences or practices and trying to clarify key concepts in light of this. This actually links critical realism very closely to early (rather than neo-) pragmatism (particularly, Pierce, Dewey, Hook, and Mead).

In particular, critical realism has been concerned with trying to congeal our knowledge of such concepts as structure, emergence, and causation, unpacking the logics of explanation (notably, retroduction and abduction), and trying to root out from philosophy and the philosophy of science, useful theoretical gems; without falling back into forms of positivism that want to reproduce the natural sciences in the social sciences.

Why is this important? Well, as Gaston Bachelard once observed, “all philosophy, explicitly or tacitly, honestly or surreptitiously … deposits, projects or presupposes a reality” (Bachelard, cited in Bhaskar 2008:13). This is also true of sociology. As we investigate anything in the social world, objects, phenomena, or a domain, we operate with a tacit understanding of the nature of those entities, phenomena, or domains. We cannot avoid having existential commitments, certain expectations, and certain abstract ideas about the social world: certain things we presuppose and operate with – particular ontological models of what the social world is, what it looks like, and how it works. A classic historical case of an ontological model is of course ontological naturalism, born from a mistaken attempt to mirror the natural sciences; a doomed endeavor that resulted in an ontology of the social world which was populated by laws. This naturalism seeped into many accounts of causation and structures, as well as regularities, objectivity, and methods. It shaped the sociological imagination… and maybe it still does.

Against this account, most sociologists would want to affirm that the social world is open, complex, and chaotic. It is not defined by anything resembling laws. Instead, events and phenomena in the social world are contingent. Discourses, social structures, cultures, bodies, intermingle in various ways to create events and situations that are overdetermined by the interaction between these different and sometimes competing elements – often resulting in context specific and unique events, complete with inconsistencies and contradictions. While there are pockets of order the social world is dynamic and heterogeneous, even schizophrenic at times, and can only ever be understood ‘in the wild’. We need thick accounts of structures and causation that are able to do justice to the different phenomenon sociologists encounter, without reducing structure and causation to simple ideas, or elevating one aspect of the social world into a silver bullet. *&%# one-sidedness. *&%# simplicity.  Causation is not just about how event a is related to event b, and social structures are not just in the head of individuals .These are not just epistemological or methodological pragmatics – they are ontological claims about the nature of social reality, and we need ways of thinking through these issues and discussing and adjudicating between rival or incommensurable accounts. We need orienting frameworks and thick and rich concepts to do this.The world is rich in different things that behave in different ways according to different ‘logics’ and we need to capture that, as difficult as it may be, without resorting to one-sided resolutions and the tedious pendulum swings and oscillations this inevitably generates. This is not just an “empirical question” – it is a philosophical one.

So, how about an example? Let’s talk big data. There is often an unquestioned assumption that emphasizing large amounts of data will lead to more accurate conclusions… and sometimes it does. It reveals patterns we would otherwise not see. But how are we to understand these patterns, and the implications of big data for description and explanation of social phenomena? What is the nature of the relationships we discover through big data – what is their ontological status? What distinguishes an accidental relationship or pattern from a real one? What is generating these patterns? Are these the product of single dynamics? A few ‘mechanisms’? Is it possible that different dynamics, practices, social structures, etc. are producing similar patterns and effects? Or that the same dynamics, etc. are producing different effects? These are certainly empirical questions, but they also raise philosophical ones.  When it comes to big data, it would not be unfair to suggest that empiricist and positivist ideals are creeping back into sociology. Or let’s take the example of racism, which, in spite of work by sociologists and works outside of sociology like The New Jim Crow, is still often viewed or investigated in cultural terms – namely, through the analysis of attitudes and behaviors, or alternatively through demographics and statistics which leave us with variables, often without any real explanatory power. Indeed, population studies and demography are generally valued more than the experience of agents. Experimental methods and regressions are thought to give stronger evidence of causation than ethnographies or interviews. There is a strict imperative to quantify across the board.

So does philosophy actually solve any of these problems? Yes. No. Maybe. Maybe not … but it helps.

Let me give one example. In the philosophy of science there has been a movement towards a powers or dispositionalist account of causation, that emphasizes complexity, conjunctions, constellations, singularities, context sensitivity, uniqueness, accounts of causation which allow for causal factors to have different effects when placed in different contexts, different causes to generate similar effects, causes that may be “present” but not observable or known only through their effects, or effects that take place over time rather than in punctuated moments  (see here). This move to a dispositionalist account of causation is concerned with rethinking causation in terms of powers, emergence, demergence, capacities, dispositions, tendencies, liabilities, and affordances. Such accounts attempt to move away from thinking about relations between events and statistical correlations, and instead boot straps contingent and historical regularities (or demi-regularities) as insights into thinking about things (taken in the broadest maximal sense), stratified structures, capacities, powers, dispositions, and tendencies. Critical realism has been on this bandwagon for a while now but many of these concepts remain foreign or under-utilized within sociology, but hold a great deal of potential.

For reasons like this, critical realism tries to make room for and concerns itself with philosophy and philosophical questions in order to have clarity about what we are doing when we do empirical work and explanation. This is not first philosophy, but it is built upon reflections from empirical engagements and the theoretical problem these reflections raise, looking wide . This is often captured by the metaphor of “underlaboring”. The term is taken from the British empiricist John Locke who contrasts the underlaborer to the master-builder. The master-builders are those whose “mighty designs, in advancing the sciences, will leave lasting monuments to the admiration of posterity” while for the underlaborer it is ambition enough to clear the ground a little, and removing some of the rubbish that lies in the way to knowledge [endnote 3]. To this end, critical realism has focused upon mining the philosophy of science for concepts which may be useful to the sociologist – terms like open systems, emergence, structure, stratification, and above all powers, for the purpose of awareness, as well as to correct and prevent certain biases from emerging.

Of course, critical realism is not alone in this. We can see similar similar gestures in sociologists like Bourdieu, Latour, Foucault, Sewell, Tilly, C. Wright Mills. Bauman, pragmatism, and analytical sociology, who have focused upon clarifying our concepts of structures and mechanisms, sometimes reaching similar conclusions, sometimes very different ones. Theorists, such as Sandra Harding, Elizabeth Grosz, Victoria Pitts-Taylor, have been engaged in questions of materiality and ontology from a feminist perspective. Critical Race Theory has centrally concerned itself with structural and institutional accounts of white supremacy and racism, rather than simply cultural grounds, and has developed sophisticated accounts about how structures operate. Anthropology has undertaken an ontological turn, with anthropologists such as Webb Keane and David Graeber, arguing for a realist stance. Continental philosophy has taken a realist turn in the form of speculative realisms and object-oriented ontologies, championed by people such as Manuel DeLanda, Graham Harman, and Timothy Morton, all of whom are theorizing the social world. Analytic philosophers and realists, some such as Nancy Cartwright who are more known, and others such Sally Haslanger, Stephen Mumford, Rani Lill Anjum, Ruth Groff, and Dan Little, who are unfortunately largely unknown to sociologists, and are doing work in the philosophy of science and social science that should be of immense interest to sociology. There are gems to be mined – gems that are likely to be generative. Philosophy has a way of providing concepts, orienting concepts, sensitizing concepts, and background theoretical beliefs that provide clarity for the researcher when they are in the field, and explanatory devices when they are writing up their field notes.

So, do practicing sociologists need to worry about philosophy at all? Does sociology need philosophy in order to survive? Maybe. Maybe not. But at the very least, why shouldn’t sociology be interested in philosophy!? If we want to accept that philosophy is about meta-theoretical reflection on the world, on experience, on practice, and the concepts we use …  it seems like something sociologists should be interested in. Gleaning insights from philosophy, at the very least, seems like a fruitful idea. In fact, historically, this has been the case. The early founders of sociology Marx, Weber, Durkheim, and Du Bois all either studied philosophy, or engaged heavily with philosophy. It provided them not only with orienting concepts, but insights that were generative and often profound – profound enough to shape a discipline.

But surely most of what I have outlined is just good sociology? Well, yes! I’m not disputing that you can point to a number of good sociological empirical and theoretical works that are trying to think along the same lines and think through the same problematics. In fact, I want to go a step further and suggest this shouldn’t be surprising at all! In fact, whether we recognize it or not, and whether you like it or not, a lot of good work in sociology is implicitly making the kind of distinctions and positions that critical realism is trying to make, reconstruct, and systematize. All sociologists are already doing philosophy – whether explicitly or implicitly. All sociologists are already doing ontology – whether explicitly or implicitly. We are always making claims about and against the world. Critical realism just argues that we need to be more reflexive about this, and provides tools to do just that.

When the “world is in flames” theory and philosophizing seems not only inadequate but pathological. It has all the feel of fiddling while Rome burns [endnote 4]. As Gross rightly mentions “we need good, clear, accurate, and powerful explanations” so that we can work out how we might move forward as a discipline and as a society which strives for democracy. We need concrete on the ground work uncovering the various dynamics at work in our current conditions – structures of racial expression, misogyny, inequality, and perhaps even the conditions under which human beings are able to live a good life [endnote 5]. But we need to do this with a degree of reflexivity and sophistication about what we are doing and how we are going about our explanations. The current situation was in part caused by a lack of sociological reflexivity and a widespread (mis-)placed belief in the power and accuracy of statistical methods and big data. We need to be more reflexive about what we are doing. More philosophically reflexive about the assumptions our methods are makings. More ontologically reflexive about how we are thinking about the social world. We need to stop and examine the foundations. Clean the sociological carburetor of theoretical blockages. Build up the sociological imagination. Rethink what we are taking for granted.

At the very least I want to suggest critical realism opens a space in sociology for these discussions to take place. It tries to reflect upon the best practices of sociology and systematize those insights. It identifies certain problematics, and explores the traction certain philosophical concepts might have for sociology. It wants to explore the relationship between philosophy and sociology, and how one can inform the other. It creates a space for theoretical reflections, gives a useful orientation for how to do philosophy in sociology, and it provides access to a few good tools for thinking through certain problematics. Critical realism has been doing this for a while, and brings different but often overlapping and complementary perspectives and concepts than other theoretical positions. In short, critical realists tries to make space for different forms of reflexivity in sociology by engaging with certain traditions of philosophy. And in summation, frankly, friends should let friends do philosophy … particularly since they are already doing it (whether they want to or not).

[Confession time. Not only am I deeply imbedded in the church of critical realism – complete with a membership card and a brand – and not only am I helping to master-mind the looming theoretical coup d’etat of American sociology, I have a Ph.D… in philosophy…. and I am a philosopher employed in a sociology department. Defending the place of philosophy in sociology is very much in my best interests. Of course a philosopher was going to defend philosophy! Regardless of this, I also firmly believe that the division between philosophy and sociology is a false and artificial one, and that the “philosophy of social science” should arise from reflections upon concrete work and concrete problems in the social world. Sociologists should be philosophers and philosophy should not simply be left to the philosophers … myself included]

[*** keep an eye out in the near future for a post looking at empirical studies drawing on critical realism]
End Notes

[1] Unsurprisingly, I think both Gross and Healy are also a little reductive, misleading, and give an unnecessarily bad reading to both critical realism and the texts they engage.

Given that something like this will be expected, here are a few points where I think this is the case. Rather than defending the claims of Porpora, Donati, or Archer, I will focus upon certain formalities. This is not to say I agree or disagree with their positions.

• First, I will admit to being a little disappointed with Gross’ laconic responses to the “myths”. These terse responses and general lack of engagement with the positions, resulted in an anlaysis that was a little misleading about some of the claims of the book. It should be noted that the word myth on Porpora’s own definition is used to designate both 1. A belief filled with larger significance for some group that tells a group who they are, where they come from, and where they are going, and 2. “false beliefs” generated from “faulty philosophy” that are “somewhat arbitrarily identified” but nevertheless mark out certain terrains. On Porpora’s own account, this was done deliberately provocatively to engender debate – it seems to have succeeded. So let me quickly look at a few example “myths” Porpora provides and the responses they elicit from Gross and :
• Myth 1 “Ethnography and historical narrative too are only exploratory or descriptive. They are not explanatory” fabiorojas seizes on this myth in particular and points to a special edition in the AJS devoted to inference in ethnography, and refutes the myth with the delightful line “Bro, do you even J-stor?” Firstly, it is important to note that my critical realist “bro” Porpora does actually cites Tavory and Timmerman’s article in that AJS edition to show that people are trying to think through the issue of ethnography and explanation, and that this is in fact a live debate. So, apparently he does J-stor. Bro, do you even read citations and bibliographies? 😉 Secondly, I will leave it to the reader as to whether the perceived need for a special edition in AJS on inference, ethnography and explanation, lends validity to Porpora’s claim or not.
• Myth 3 “There is no truth. Everything is constructed” Gross, somewhat ironically if we believe Porpora, responds with “Yes, this is indeed the guiding belief behind editorial decision making at our leading journals”. Social construction is norotiously difficult to define and means many different things [5] but Porpora is making a few claims here that are worth unpacking and not so easily dismissed: 1 we need to distinguish between beliefs and reality – not all reality is belief or concept-dependent. Processes and structures in the social world exist apart from our conceptualization of them – racism and misogyny are cases in point. In other words, there are mind-independent aspects of social reality and an overly-wrought focus on beliefs, attitudes, and behaviors (as opposed to social structures and material distributions – such as structural racism) is problematic to say the least. Porpora claims this insight is needlessly lost in the cultural turn and the practice turn. 2. We need a way of adjudicating between different claims and criteria of objectivity – a concepts of realism, truth and objectivity – particularly if we want a critical sociology that is able to identify structures of domination and oppression. We need to be able to discuss and assess competing positions and competing claims. For Porpora both claims fundamentally amount to the same thing – the need for ontology.
• Finally, in his critique of Archer and Donati, I think that Gross’ claim that Donati “refuses to engage the impressive scholarship produced by economic sociologists, economists, anthropologists of finance” is a little misleading … and frankly a slightly curious critique to make of the book. At no point does Donati explicitly refuse an engagement, rather, the book from start to finish is explicitly a theoretical and ethical meta-analysis of social relationa; and the chapter on the financial crisis is seeking to show how the theoretical and ethical framework they develop might be used and applied to a case like the financial crisis – at an explanatory and moral level. But this raises a question…. is this still sociology? I think this touches on a fundamental difference in orientation between American sociology on the one hand, and British and European sociology on the other. British and European sociology (and anthropology) is, in general, more open to philosophical and speculative programs and even grand theories; while American sociology is more empirically-driven, heavily favoring theories of the middle-range (Archer and Donati might even reference Porpora’s myth 4 – “The most important scientific questions are empirical”). These are very different traditions and could be said to represent the troubled and contested fault-line between “science” and “humanities”. While Gross may be justified in his frustration, or not, there are important disciplinary dynamics at play here, as well as different concepts about what theory is and what it should do. There is an important discussion to be had here.

[2] As I and others have argued elsewhere critical realism, “Defining critical realism is not an easy task. While there is a pool of scholars that critical realists often draw upon … there is not one unitary framework, set of beliefs, methodology, or dogma that unites critical realists as a whole. Instead, critical realism is much more like a series of family resemblances in which there are various commonalities that exist between the members of a family, but these commonalities overlap and crisscross in different ways. There is not one common feature that defines a family, instead, it is a heterogeneous assemblage of elements drawn from a relatively common “genetic” pool. Critical realism is a philosophical well from which Marxists, Bourdieusians, Habermasians, Latourians, and even poststructuralists have drawn”. For broader contextualization of this quote, and a broader introduction to critical realism, see What is Critical Realism

[3] notably, in his later work, Roy Bhaskar makes a disastrous transition from underlaborer to the master builder.

[4] I do feel the need to point out, and I am sure Gross would agree with me, that the world has been “burning” for some time, and not just the last few years or since November 2016.

[5] As Philip Gorski has argued elsewhere, the fact-value distinction is leaky and values can be objective, and not just subjective. Accordingly, the social sciences should endeavor to be both critical AND constructive.

[6] See, for example, Ian Hacking’s famous work, The Social Construction of What? (2000) or Sally Haslanger’s increasingly well-known Resisting Reality (2012) [both philosophers who should be of immense interest to sociologists]. For a critical realist take see Dave Elder-Vass The Reality of Social Construction 2012