Another round with Michael Bérubé

Famed ice-hockey scholar and literary critic Michael Bérubé has written in several places about the notorious Science Wars, but not always to my satisfaction, especially as we both march under the banner of post-Rortyan pragmatism. We've gone a few rounds in the past, and I haven't yet been able to make my objections clear to him; but his recent article (see also here, for an invigorating comment thread) gives me a chance to try to do better in this space.

One of Michael's concerns is to defend “theory” and “science studies” in a broad sense from its attackers like Alan Sokal (of Sokal Hoax fame). He admits that things got a little out of hand in the 80s, what with the pony-tailed left-academic brigade making the humanities look bad with (in Michael's sublimely witty rendition) “their queering this and their Piss Christ that and their deconstructing the Other”. The Hoax seemed to many to burst that Theoretic bubble and restore sanity to the academic realm, or at least provide a clear criterion for same (which, alas, not everyone meets, even now). But what is its real significance when science seems now to be threatened from another front? Rhetpage

As Michael relates, “[i]n my academic-left circles, Sokal’s name was mud, his hoax an example of extraordinary bad faith” while “everywhere else […] Sokal was a hero, the guy who finally exposed the naked emperor.” Michael's verdict, and mine, is more mixed. In our view, Sokal got them good, no question: anyone who knows what the axiom of choice is (or the axiom of identity, or even non-linear dynamics), would catch the joke immediately. And they didn't. This sorry result corroborates Sokal's charge that, as Michael puts it, the Social Text crowd “were overstepping their disciplinary bounds and doing 'science studies' without any substantial knowledge of science.” This is a problem, because if this is right, then they can't be familiar enough with the practices of science to say anything useful about it theoretically, as they claim to do.

On the other hand, Sokal and his fans seem to think that the hoax proved a graver charge than mere ignorance and Dunning-Kruger style hubris: that is, that among the “howlers” inserted by Sokal but missed as such by Social Text were blatantly nonsensical claims, self-refuting in the familiar way, by goofy French types like Derrida and Lyotard to the effect that objectivity is a phallogocentric myth, that there's no real world, and so on. This failure supposedly established that science studies types are soft on, or even sold on, the sort of anything-goes relativism (again supposedly) found in English departments and across the Channel.

Michael wants to preserve a role for Theory's constructive claims, so he provides a corrective designed to acknowledge the former of Sokal's charges and deflect the latter. If successful, this will allow the academic left to overcome its tradition of self-laceration long enough to confront its common enemy: right-wing irrationalism and its politicized attacks on evolutionary theory and climate science. In a way, this means that he is trying to do well what Sokal did poorly, which is to show that it is not the very idea of science and rationality, but instead adolescent rebellion against same, which — especially now — serves anti-progressive aims. This is better, again, in Michael's view, because it leaves room for the real contributions socially-minded theory can provide, rather than discarding them as pernicious nonsense and ceding the entire task to the sciences.

I will focus here on one promising but elusive slogan in Michael's corrective; but in true hermeneutic fashion, I will insist that, well, it depends on what he means.

I must refer you to Michael's article for the details; here let me focus on Michael's charge that Sokal's use of “the phrase 'objective reality (both natural and social)' […] makes the terrible mistake of conflating two different things, and of suggesting that the analysis of social reality should proceed like the analysis of physical reality–as if the pursuit of social justice is a matter of discovering the physical properties of the universe.” Michael's response resists this conflation, preserving “social reality” as a separate domain from the physical, one which thus responds to different methods, and so resists the scientistic overreaching which so worries humanists. At the same time, it deflects the suggestion that interpretive types lack any sense of the objectivity of physical reality — that they think that oxygen, Neptune, and X-rays (Michael's examples) are “merely socially constructed” — and thus not “real” or “objective” phenomena.

Michael's rhetorical strategy is well thought out and consequently very effective, especially in mending fences between science 'n' rationality types and literary types on the left, enabling them to resist both major sorts of right-wing attacks on academia, to wit:

1) those academics are all postmodern skeptics who reject the very idea of objectivity and thus objective morality, like that homosex is eeevil;

2) those academics are all materialist dogmatists who reject transcendent values in favor of their (doomed) faith in modern empirical science and utterly un-Aristotelian Enlightenment “rationality”

where the rhetorical strategy of the latter has recently borrowed a page (drawn “aid and comfort,” one might say, borrowing the language of treason) from postmodernism itself, i.e. moving from the original charge that

3) transcendent values are real [= platonism] so atheistic materialism is false (etc.)


4) materialism is just one perspective on reality, and others are just as good even if they cannot be proven (and only scientistic dogmatism says otherwise).

This is the threat Michael is concerned to meet. He sees that achieving a united front among what it is now fashionable in political contexts to call the “reality-based community” requires that some of that community overcome, and be seen to overcome, their apparent allergy to, well, reality. The promising slogan Michael unveils for this purpose, which I would like to examine here, is this:

Lifepage “[T]he world really is divvied up into “brute fact” and “social fact,” just as philosopher John Searle says it is, but the distinction between brute fact and social fact is itself a social fact, not a brute fact, which is why the history of science is so interesting. Moreover, there are many things–like Down syndrome, as my second son has taught me–that reside squarely at the intersection between brute fact and social fact, such that new social facts (like policies of inclusion and early intervention) can help determine the brute facts of people’s lives (like their health and well-being).”

The phrase I have bolded is a good line, all right. But let's see if it can really do what we want. Taken in the context of his description of what science-studiers can get wrong — a failure to acknowledge the independence of reality as captured in the idea of “brute fact” — this dictum makes it sound like that distinction between brute and social fact maps onto that between the natural and social sciences, as a characterization of each's proper domain. This is what I will dispute.

Not only is that not necessarily right, but it leaves mysterious why, even if it is right, this should be the case, as it appeals to the very idea of “socially constructed fact” whose application in this context is obscure. Of course as a pragmatist Michael has a ready reply for both worries. For some pragmatists at least, truth is “what works.” Now this itself can sound like the goofy French; but all it really needs to mean is that once we have determined what to say — and as we have seen, saying this seems to further our goals as outlined above — then, as Rorty would insist, one should ask no further questions as to whether it is “really true,” which is what I seem to be doing.

But does it really work? This will depend on what we are interested in and what exactly would count as “working” in that context. Complicating this judgment, however, is the inconvenient fact that Michael and I do not share exactly the same interests. While I deplore some of the same deplorables as does Michael (creationist sophistry as well as science 'n' rationality fanboys like Sokal and Gross & Levitt), my main concern is not with “science studies,” let alone the academic left, but instead with pragmatism's ongoing fight against its traditional philosophical opponents: platonists and Cartesians, whether creationist or materialist. From this perspective, even Michael's clever gloss on Searle's distinction can backfire if we're not careful. If this is true, then one might wonder about the longer-term practicality of this move.

The short version of why this is is that this mended fence is unstable. The notion of objectivity cannot simply be construed as the object of its own — properly rigorous and detached — form of inquiry. As a global constraint on inquiry, it must be reconstrued in a way which — diplomacy be damned — causes metaphysical realists like Searle and Sokal some real theoretical pain, or our alliance will fall apart when, if this ever happens, the battle is joined and push comes to theoretical shove. For if modernism is not rid of its poisonous Cartesian heritage, anti-modernists (both post- and pre-) will have perfectly good points to use against us. (I say “us” provisionally here; for myself, I see little functional difference between a post-Kantian project of “modernism criticizing itself” and something worth calling “postmodernism.” Unfortunately, as Rorty also recognized, the latter term has become toxic.)

I don't have much space left for a longer version, but I'll give it a go. As I've already said, however it is construed — even as itself a “social fact,” as per Michael's corrective — Searle's distinction between “brute fact” and “social fact” is most naturally taken to map onto that between natural and social science, as a characterization of each's proper domain. In other words, the idea is (as determined, on Michael's version, by the latter) that natural science is inquiry (into brute fact) and social science is interpretation (of social fact).

Once this Searlean dichotomy is conceded to the defender of “brute fact,” though, it is cold comfort to retain its actual manifestation as an instance of a “social fact.” What we should say instead, on my view, is ironically similar to what Sokal said in “conflating” natural and social reality: i.e., that everything — both natural science and social science — is both interpretation and inquiry: whenever we say — or hear, or read — anything, we are concerned both with how things are and with what words mean (as well as a few other things, like our or our interlocutor's purposes in saying that rather than something else).

The way to understand the difference between Michael's and my hermeneuticism, in other words, is that Michael, like Rorty, is concerned with the constructive nature of the first-person plural perspective: things are a certain way for us, such that any one of us could be wrong about them, but only because we have agreed among ourselves that they are that way. Michael's corrective assures natural scientists that this cannot be a universal fact in the self-undermining postmodern sense. My view, on the other hand, follows Donald Davidson in concentrating on the first-person singular agent in the process of interpretation — that is, of language use. As I see it, Davidson shows that just as the interpretation of meaning includes an ineliminable aspect of doxastic commitment to how things are in reality (or in other words, belief), inquiry into how things are includes an ineliminable aspect of interpretation — that is, of sensitivity to the interpretive aspect of language use. In Davidson-speak, we must affirm the holism of belief and meaning, and thereby reject the Cartesian scheme/content dualism.

Like Michael's version, this 1) leaves a distinction in place between a determination of how things are and one of what to say for more subjective, e.g., instrumental, reasons; while 2) recognizing the ultimately hermeneutic constitution of that distinction. But seeing that distinction as hermeneutic in my sense means that the former does not determine a “brute fact” as opposed to a “social fact.” It's just a fact like any other: in its relation to the world as it is, it provides a doxastic constraint on the continuing process of interpretation/inquiry. Again, Michael's version makes it sound as if that hermeneutically determined distinction is this: that science does inquiry and the humanities do interpretation. This is not a hermeneutic point at all for me. Everyone does both, because every utterance in whatever context manifests a commitment to both belief and meaning.

So what is that distinction then, on a Davidsonian view? It's this: when we take ourselves to understand each other — that we speak “the same language” — we tend to take meaning as fixed and concentrate on fixing belief. Similarly, when we are concerned to determine the meaning of an alien utterance, we can only do so by seeing the two of us as sharing an objective world, tying the circumstances of the utterance to our shared surroundings, and thus to what it is about, by our own lights as manifested in our beliefs about it. This is the real import of Davidson's much-misunderstood “principle of charity,” which in fact applies just as much to myself as to others. Reality, for me, is necessarily reality as I believe it to be; yet I may still acknowledge that my beliefs, qua beliefs, may need to be revised — or, in other words, that a better-situated interlocutor may properly interpret them as (saying something) false.

But why does this entail inflicting theoretical pain on realists, given that the notion of objectivity is retained rather than (as Rorty urged) given up? Metaphysical realists like Searle believe we need to hang onto the Cartesian notion of an “objective world” lest we fall into a relativism or idealism which dispenses with the commonsense idea that “saying doesn't make it so” — or, in other words, that our beliefs can be false. But we don't need this conception of objectivity (or “brute facts”) to preserve this idea. For Davidson, the very point of the concept of belief is to mark the potential difference between what we believe and how things are. It simply does so without tying the notion of objectivity in the realist manner to a thing so designated, rather than simply denoting an external rational constraint on what we say. (As John McDowell points out, Davidson himself gets into trouble in misconstruing the nature of “rational constraint” in this sense.) Recognizing this constraint — an interpretive rather than metaphysical one — is all it takes to cause us, or even require us by our own lights, to modify our beliefs in order to get things right in inquiry, and thus turn our backs on relativism and skepticism. In fact it is the realist who runs into skeptical difficulties, often relying simply on a practical determination that skeptical worries cause not too little, but too much trouble for us to take them seriously (that is, they render science impossible). Once this is understood, there's nothing for the notion of “brute fact” to do.

Like what you're reading? Don't keep it to yourself!
Share on Facebook
Tweet about this on Twitter
Share on Reddit
Share on LinkedIn
Email this to someone