Wed Sep 4, 2013

A Word on Critical Realism

Note: The original version of this post, with lots of comments including several follow-ups, clarifications, and further argument from me can be found at OrgTheory. A very useful contribution by Omar Lizardo can also be found there.

Seeing as Fabio has promoted some off-the-cuff remarks I made on Twitter about Critical Realism, I suppose I should say something a little more about it. All the moreso seeing as some anonymous commenters have been getting quite huffy at the very idea that anyone who called themselves an academic could make a dismissive comment without, presumably, devoting themselves full-time to “thoughtful debate and analysis” on the work in question. I have a general and a specific response to that. Speaking generally, online commentary should not be a kind of Markov process where every single contribution must start from scratch with no memory of anything that has gone before. The demand that any particular comment or post provide a full and complete accounting of everything on the topic (before it can count as “thoughtful debate and analysis”) is a hallmark of annoying Internet discussion. My specific response is that some time ago I did in fact devote myself full-time to thoughtful debate and analysis about Critical Realism, for a period of about eighteen months. I read pretty much everything on the topic that had come out until that time, which was a real barrel of monkeys let me tell you. I wrote and published an article on a current debate in the field, and then I moved on to other projects.

My conclusion, then as now, was that Critical Realism is a low-quality, confused, and misleading body of work. It is a justly peripheral branch of 1970s philosophy of science. The philosophical demands it satisfies amongst sociologists could be met elsewhere at much higher quality and far lower cost. In practice it does literally nothing substantive for the work of the sociologists who have taken it up, and I am dismayed to see it gain a foothold in the United States.

Why has CR gotten traction in sociology? In the UK, it was initially attractive mostly to a certain sort of humanist Marxist, especially after it became clear that very strong versions of structuralist Marxism had some unpleasant qualities. The backlash against structure (whether in its Marxist or Parsonian forms) led to various middle-way programs, most notably Giddens’s structuration theory. To many observers this was thin gruel indeed. (They were right about this.) In response, some began to cast around for a way to say social structures were “real” in some strong sense. They sought some philosophically respectable means of defending that idea. Bhaskar’s work, beginning with A Realist Theory of Science, presented itself as an attractive answer. A little later, Margaret Archer retrofitted her already in-progress critique of Giddens with Bhaskar’s vocabulary. This was one of the key pieces of CR’s consolidation in British sociology, as Archer’s theoretical work was a clear and effective attack on the whole structuration program.

In the United States, meanwhile, there seem to have been two channels of diffusion. The first, via Michigan, was through historical sociologists like George Steinmetz and, latterly, Philip Gorski. The second was through religiously-minded sociologists like Chris Smith. I admire the work of these scholars a great deal, by the way. The same goes for Archer, whose work got me reading CR in the first place in 1996. Between them, they jump-started American interest in CR, which up until then had been confined to papers written by a small number of scholars mostly publishing in the Journal for the Theory of Social Behavior.

The diffusion of CR was slightly hampered by the transformation of Bhaskar from fringe philosopher of science to full-blown guru. Having recruited followers in sociology on the basis of his realism, he began to pull the rug out from under them in the late 1990s, first with the merely absurd Plato, Etc and then with the frankly embarrassing From East to West: Odyssey of a Soul (which closed with a final chapter titled “The Dance of Shiva in the Age of Aquarius”). This work, in retrospect, seems like the culmination of the unpleasant cult of personality that grew up around Bhaskar in CR circles in the 1980s and which he seems to have done little to discourage.

CR presents itself in a way that some social scientists—with next to no real background in philosophy—feel gives them just what they need to shore up their empirical research and metaphysical intuitions. You want to be realist in your philosophy of social science? Sure! You want your preferred level of analysis to also be an ontologically emergent level of reality? No problem! You want to talk about social structures as irreducible in some serious-but-not-really-analyzed fashion? You got it. You want your theory to be critical? I mean, who doesn’t, right? Just call yourself a Critical Realist and cite some Bhaskar. After all, he has repeatedly asserted that his work is a “Copernican Revolution” in the philosophy of science—the near-total indifference of everyone else in that field notwithstanding, but never mind that for now.

It’s understandable that sociologists have been susceptible to this sort of thing. The field has long been attracted to strongly expressed emergentist claims. At least since Durkheim, many sociologists have felt that a scientific discipline needs a self-subsistent and “ontologically real” subject matter all of its own. Otherwise, the fear is that the field would become “reducible to” psychology, or worse. Sociological journals are filled with a repeated insistence on the reality of “society” or “the social”, and–conversely—a rejection of “methodological individualism” or “reductionist” approaches to explanation associated with the less social social sciences. No wonder that we’re attracted to a philosophy that seems to “ground” this idea, that guarantees that society or social structure or some supra-individual entities are really real, really. Conveniently, CR also links the allegedly fundamental layered structure of reality with the social organization of academic disciplines round about 1975, with Physics and the bottom, then Chemistry on top of them, then Biology, then Psychology, and then Sociology. (Above that is the Divine bonus layer which you can take or leave as you like.) Each layer has its own emergent properties and causal powers.

The case of emergent properties is a nice example both of what makes CR attractive to some sociologists and of how detached CR became from both the philosophy of science and metaphysics in the 1970s. Many of the claims about emergence characteristic of CR rest on repeated (and probably unwitting) conflation of different senses of what is “emergent” about emergent properties, and what it is to be able to “reduce” something to something else. CR writers typically make the case for their emergence and irreducible properties by way of some stock examples. For instance, they will say that new properties emerge at “higher” levels, as when atoms of hydrogen and oxygen—separately, flammable gases—combine to form water, a molecule that has a property—”wetness”—not “deducible” (CR people will say) from the properties of its component parts. This kind of reasoning is extremely attractive to social scientists for obvious reasons, but it is also very slippery.

Reduction can mean two things. It can refer to the business of translating or defining sentences from one theoretical language in the language of another. For example, we might try to translate statements or theories about water in chemistry, or whatever, into the theoretical language of, say, physics. Or, we might want to “reduce” the propositions and theories of psychology into the language of biology. Whether and to what degree this is possible was the topic of an extensive debate in the ‘60s and ‘70s (on the “special sciences”, with Putnam, Fodor, and others, involved). The result of this debate was that you generally can’t reduce theories in this way. (Fodor is especially clear on this.)

Second, and separately, there’s the question of ontological reducibility. This is question of whether, independent of your theoretical language, something like a water molecule just is a matter of hydrogen and oxygen (or more fundamental physical entities) in some particular configuration. There’s a further wrinkle here depending whether you are talking about the reducibility of some particular thing or a natural kind.

This distinction between explanatory and ontological emergence is why it doesn’t make sense to talk about about physical properties being deducible from other properties. That can only be true of the linguistic stuff — axioms, theorems, sentences, predicates. CR advocates repeatedly seem to miss this distinction as they want to use examples of alleged ontological irreducibility to shore up claims for the need for a number of distinct disciplines at particular levels of analysis — physical sciences, biology, psychology, sociology, etc. One minute they’re talking about emergence of physical properties in the world (a question of ontology), the next about relationships between theories from this or that discipline at various levels of analysis (a linguistic matter). Running the two together reliably generates an enormous mess.

This is just one of the many ways CR fails to deliver. I am not going to catalog them all here. Again, this isn’t to say that, for example, substantial parts of Archer’s critique of Giddens weren’t correct, or that sociologists shouldn’t—when it comes to developing or testing explanations—rely on “higher level” concepts in their explanations. (Remember: facts about linguistic reduction don’t entail ontological claims about what fundamentally there is.) But there is an awful lot of philosophy that will give sociologists the frankly minimal machinery they need to go ahead and do this work without fear of incoherence. None of the problems CR presents itself as solving are uniquely explicable by (or even originate with) the CR approach itself. Sociologists interested in emergence or macro-level explanations have no need to run together that interest with the specific CR position or view. (This was the point of my old article on Archer, Mouzelis, and CR.) There are large and immediately accessible literatures on all of these topics in philosophy, leading naturally to more technical or specialist work. Consider the SEP article on Emergent Properties, for instance, and the one on Supervenience, or the one on Scientific Explanation, or Scientific Realism. Go from there to, say, Jonathan Schaffer’s Is there a Fundamental Level? (for the metaphysics) or Michael Strevens’ Depth for one take on the philosophy of science, or—for something in parts similar to Bhaskar but far more creative and central—read Nancy Cartwright) on laws of nature. I feel confident in asserting that different sociologists could ally themselves with quite incompatible positions in these debates and much of our work would go on as before.

You’ll notice, if you work through this material, that Bhaskar and Critical Realism are basically not discussed at all. There are good reasons for this, some of which I have mentioned. You might be tempted to feel offended at my dismissal of CR as peripheral. Maybe you think there is some kind of conspiracy in philosophy. “They laughed at Einstein!” you say. (They didn’t really, but never mind.) Well, of course expert scholarly opinion is not infallible. The center of a field doesn’t have a monopoly on insight or valuable argument. But I don’t see anything wrong with asking and taking seriously the views of experts in other fields, especially when CR itself makes such strong claims about its revolutionary effects on those fields. Hence my line on Twitter about Toffler. I bet you’d be at least bemused if you came across a bunch of philosophers building some quasi-sociological research program around the insights of Sociology’s central Tofflerian school. And you might get a little annoyed if they dismissed your efforts to explain that, although sociological insights might be found here and there his work, Toffler is also a somewhat unreliable guide to both the field and the world—his own grand claims to the contrary notwithstanding.

My own view is that sociology needs very little philosophy in order to thrive. Most of the exciting work in sociology has not come from self-conscious theorists seeking to fit the field to some philosophical system, least of all when that system is not even defensible. Much of it, indeed, arises interstitially and out of cross-cutting research groups and programs. This is an imagery of science very much at odds with the carefully-layered picture provided by CR. Indeed, one of the things philosophers of science began noticing in the 1970s was that when two sciences met at a boundary the result typically was not some process of tectonic subduction, with the more “fundamental” science consuming the other by reduction, but rather the appearance of a new science with its own specialist sphere and field-specific concepts and theories—molecular biology, biochemistry, cognitive anthropology, or what have you. CR claims to ground social science with its picture of the fundamental ontology of things, tightly bound to a particular view of what scientific explanation is. And yet in CR-inspired books and articles, it appears more like a lump of cream cheese sitting on top of a bagel: laid on too thick, added after the baking was done, obscuring the stuff you want to chew on, and probably bad for you in the long run. Eat something else.