Among the arguments for PSR I put forward in Scholastic Metaphysics are a retorsion argument to the effect that if PSR were false, we could have no reason to trust the deliverances of our cognitive faculties, including any grounds we might have for doubting or denying PSR; and an argument to the effect that a critic of PSR cannot coherently accept even the scientific explanations he does accept, unless he acknowledges that there are no brute facts and thus that PSR is true. Della Rocca’s argument bears a family resemblance to this second line of argument.
Della Rocca notes, first, that even among philosophers who reject PSR, philosophical theses are often defended by recourse to what he calls “explicability arguments.” An explicability argument (I’ll use the abbreviation EA from here on out) is an argument to the effect that we have grounds for denying that a certain state of affairs obtains if it would be inexplicable or a “brute fact.” Della Rocca offers a number of examples of this strategy. When physicalist philosophers of mind defend some reductionist account of consciousness on the grounds that consciousness would (they say) otherwise be inexplicable, they are deploying an EA. When early modern advocates of the “mechanical philosophy” rejected (their caricature of) the Aristotelian notion of substantial forms, they did so on the grounds that the notion was insufficiently explanatory. When philosophers employ inductive reasoning they are essentially rejecting the claim that the future will not be relevantly like the past nor the unobserved like the observed, on the grounds that this would make future and otherwise unobserved phenomena inexplicable. And so forth. (Della Rocca cites several other specific examples from contemporary philosophy -- in discussions about the metaphysics of dispositions, personal identity, causation, and modality -- wherein EAs are deployed.)
In responding to these different examples of EAs, one could, says Della Rocca, take one of three options:
(1) Hold that some EAs are legitimate kinds of argument, while others -- in particular, any EA for some claim about why things exist at all -- are not legitimate.
(2) Hold that no EA for any conclusion is legitimate.
(3) Hold that all EAs, including any EA for a claim about the sheer existence of things, are legitimate kinds of argument.
Now, the critic of PSR cannot take option (3), because that would, in effect, be to accept PSR. Nor could any critic of PSR who applies EAs in defense of other claims -- and the EA approach is, as Della Rocca notes, a standard move in contemporary philosophy (and indeed, in science) -- take option (2).
So that leaves (1). The trouble, though, is that there doesn’t seem to be any non-question-begging way of defending option (1). For why should we believe that EAs are legitimate in other cases, but not when giving some account of the sheer existence of things? It seems arbitrary to allow the one sort of EA but not the other sort. The critic of PSR cannot respond by saying that it is just a brute fact that some kinds of EAs are legitimate and others are not, because this would beg the question against PSR, which denies that there are any brute facts. Nor would it do for the critic to say that it is just intuitively plausible to hold that EAs are illegitimate in the case of explaining the sheer existence of things, since Della Rocca’s point is that the critic’s acceptance of EAs in other domains casts doubt on the reliability of this particular intuition. Hence an appeal to intuition would also beg the question.
So, Della Rocca’s argument is that there seems no cogent way to accept EAs at all without accepting PSR. The implication seems to be that we can have no good reason to think anything is explicable unless we also admit that everything is.
Naturally, I agree with this. Indeed, I think Della Rocca, if anything, concedes too much to the critic of PSR. In particular, he allows that while it would be “extremely problematic” for someone to bite the bullet and take option (2), it may not be “logically incoherent” to do so. But this doesn’t seem correct to me. Even if the critic of PSR decides to reject the various specific examples of EAs cited by Della Rocca -- EAs concerning various claims about consciousness, modality, personal identity, etc. -- the critic will still make use of various patterns of reasoning he considers formally valid or inductively strong, will reject patterns of reasoning he considers fallacious, etc. And he will do so precisely because these principles of logic embody standards of intelligibility or explanatory adequacy.
To be sure, it is a commonplace in logic that not all explanations are arguments, and it is also sometimes claimed (less plausibly, I think) that not all arguments are explanations. However, certainly many arguments are explanations. What Aristotelians call “explanatory demonstrations” (e.g. a syllogism like All rational animals are capable of language, all men are rational animals, so all men are capable of language) are explanations. Arguments to the best explanation are explanations, and as Della Rocca notes, inductive reasoning in general seems to presuppose that things have explanations.
So, to give up EAs of any sort (option (2)) would seem to be to give up the very practice of argumentation itself, or at least much of it. Needless to say, it is hard to see how that could fail to be logically incoherent, at least if one tries to defend one’s rejection of PSR with arguments. Hence, to accept the general practice of giving arguments while nevertheless rejecting EAs of the specific sorts Della Rocca gives as examples would really be to take Della Rocca’s option (1) rather than option (2).
Della Rocca also considers some common objections to PSR. In response to the claim that PSR is incompatible with quantum mechanics, Della Rocca refers the reader to Alex Pruss’s response to such objections in his book The Principle of Sufficient Reason, but also makes the point that appealing to QM by itself simply does nothing to rebut his own argument for PSR. For even if a critic of PSR thinks it incompatible with QM, he still owes us an answer to the question of where we are supposed to draw the line between legitimate EA arguments and illegitimate ones, and why we should draw it precisely where the critic says we should. (For my own response to QM-based objections, see pp. 122-27 and 142 of Scholastic Metaphysics.)
Della Rocca also considers an objection raised by philosophers like Peter van Inwagen and Jonathan Bennett to the effect that PSR entails necessitarianism, the bizarre claim that all truths, including apparently contingent ones, are really necessary truths. Della Rocca thinks van Inwagen and Bennett are probably right, but suggests that the defender of PSR could simply bite the bullet and accept necessitarianism, as Spinoza notoriously did. And in that case, to reject Della Rocca’s argument for PSR on the grounds that necessitarianism is false would just be to beg the question.
Here again I think Della Rocca concedes too much. As I argue in Scholastic Metaphysics (pp. 140-41), objections like the one raised by van Inwagen and Bennett presuppose that propositions are among the things PSR says require an explanation, and that for an explanans to be a sufficient reason for an explanandum involves its logically entailing the explanandum. But while rationalist versions of PSR might endorse these assumptions, the Thomist understanding of PSR does not.
Della Rocca also remarks:
I suspect that many of you simply will not see the force of the challenge that I am issuing to the non-rationalist. (I speak here from long experience, experience that prompted me to call my endeavor here quixotic.) Philosophers tend to be pretty cavalier in their use of explicability arguments -- using them when doing so suits their purposes, refusing to use them otherwise, and more generally, failing to investigate how their various attitudes toward explicability arguments hang together, if they hang together at all. We philosophers -- in our slouching fashion! -- are comfortable with a certain degree of unexamined arbitrariness in our use of explicability arguments. But my point is that a broader perspective on our practices with regard to explicability arguments reveals that there is a genuine tension in the prevalent willingness to use some explicability arguments and to reject others.
Amen to that. As with the urban legend about First Cause arguments resting on the premise that “everything has a cause,” the notion that the PSR is a relic, long ago refuted, is a mere prejudice that a certain kind of academic philosopher stubbornly refuses to examine. It doesn’t matter how strong is an argument you give for PSR; he will remain unmoved. He “already knows” there must be something wrong with it, because, after all, don’t most members of “the profession” think so?
Why, it’s almost as if such philosophers don’t want the PSR to be true, and thus would rather not have their prejudice against it disturbed. Can’t imagine why that might be, can you?
Some related posts:
Marmodoro on PSR and PC
An exchange with Keith Parsons, Part IV [on “brute facts”]
Nagel and his critics, Part VI [on rationalism, PSR, and the principle of causality]