Pigliucci on metaphysics

At Scientia Salon, philosopher Massimo Pigliucci admits to “always having had a troubled relationship with metaphysics.”  He summarizes the reasons that have, over the course of his career, made it difficult for him to take the subject seriously.  Surprisingly -- given that Pigliucci is, his eschewal of metaphysics notwithstanding, a professional philosopher -- none of these reasons is any good.  Or rather, this is not surprising at all, since there simply are no good reasons for dismissing metaphysics -- and could not be, given that all purported reasons for doing so themselves invariably embody unexamined metaphysical assumptions.  Thus, as Gilson famously observed, does metaphysics always bury its undertakers.

Pigliucci’s misgivings began, he tells us, when he first encountered the medieval Scholastics while in high school in Italy.  Though he admits that “medieval logicians actually did excellent work,” he says that “as a teenager prone to (intellectual) rebelliousness… I couldn’t but reject the Scholastics.”  He adds that “the Scholastics still have a bad reputation in philosophical circles.”  Now of course, neither adolescent rebelliousness nor appeal to contemporary intellectual fashion constitutes a serious argument.  So, does Pigliucci actually have any substantive grounds for rejecting Scholastic metaphysics, specifically?  He doesn’t tell us.  Does Pigliucci even understand Scholastic metaphysics?  For example, does he understand how it differs (profoundly!) from the kind of metaphysics one finds in rationalist philosophers like Leibniz and in the work of most contemporary metaphysicians?  From other things he says in his post, it seems not.

We’ll come back to that.  First, consider the other factors which, Pigliucci tells us, deepened his suspicion of metaphysics.  While in college, he says, he was impressed by the logical positivists’ famous verification principle, and their application of it to a critique of metaphysics.  The basic idea, as is well known, is that any meaningful statement must (the verification principle claims) be either analytically true (like “All bachelors are unmarried”) or empirically verifiable. Yet metaphysical statements are (the argument continues) neither.  Therefore they are strictly meaningless, not even rising to the level of falsehood.

There are various problems with the verification principle, the most notorious being that it is self-refuting, insofar as the principle itself is neither analytically true nor empirically verifiable.  It is thus no less “meaningless” and indeed “metaphysical” (as verificationists conceived of metaphysics) as the claims it was deployed against.  Alternative formulations of the principle have been attempted, but the trouble is that there is no way to formulate the principle in such a way that it both avoids self-refutation and still has the anti-metaphysical bite the positivists thought it had.  These are such well-known points that it is unlikely that Pigliucci still regards verificationism as a serious challenge to metaphysics.  So, even if it impressed Pigliucci as a student, what does that have to do with why he is still suspicious of metaphysics now, as a professional philosopher?

The third influence on his suspicions, he says, was “Hume’s Fork” -- David Hume’s famous doctrine that any proposition that concerns neither “relations of ideas” nor “matters of fact” can contain only “sophistry and illusion” and might as well be “commit[ed] to the flames.”  Naturally, the suspect propositions included, in Hume’s view, those of traditional metaphysics, and Pigliucci tells us that on first encountering it he found Hume’s position “a neat and no nonsense kind of view.”  The trouble, though, is that Hume’s Fork is an anticipation of the positivists’ verification principle, and has similar problems.  In particular, it appears to be no less self-refuting, for Hume’s Fork is not itself either true by virtue of the relations of the ideas that enter into its formulation, or true by virtue of empirically discernible matters of fact.  Hence it is no less “metaphysical” than the propositions it was used to criticize.  And as with the verification principle, while one can attempt to reformulate Hume’s Fork in such a way as to keep it from being self-undermining, doing so also strips it of its anti-metaphysical bite.  And again, Pigliucci presumably realizes this, since it is well-known.

So far, then, if Pigliucci intends to give us serious rational grounds for being suspicious of metaphysics, he’s 0 for 3.  But he cites a fourth influence on his skepticism: James Ladyman and Don Ross’s book Every Thing Must Go, which, while it advocates a “scientific” or “naturalized” metaphysics, is hostile to traditional metaphysics.  On what grounds?  In Ladyman and Ross’s view, the trouble with any metaphysics that isn’t essentially just the book-keeping department for empirical science is that it is going to amount to mere “conceptual analysis.”  And “conceptual analysis” is grounded in ordinary language, commonsense intuitions, and “folk” notions -- all of which often conflict with the picture of the world science gives us.  The concepts the metaphysician analyzes and the intuitions to which he appeals thus may well float free of objective reality.  Hence any metaphysics that isn’t essentially just the systematization of what the various sciences have to tell us lacks (so the argument goes) any solid foundation.

This might seem to be a more formidable challenge to metaphysics than either the Humean or the verificationist challenge.  After all, Ladyman and Ross do not eschew metaphysics entirely, since they allow that metaphysics is respectable if suitably “naturalized” or made “scientific.”  And many contemporary metaphysicians do indeed ground their arguments in “conceptual analysis,” “intuitions,” and the like.  Hence, Ladyman and Ross might seem more sober than the likes of Hume, A. J. Ayer, and Co., neither directing their attacks at a straw man nor advocating an unreasonably extreme alternative position.

In fact, though, the Ladyman/Ross position is not only not a better argument than the Humean and verificationist arguments, it is on closer inspection really just the same argument superficially repackaged.  For Hume’s “matters of fact” and the positivists’ “empirically verifiable propositions,” read “naturalized (or scientific) metaphysics.”  And for Hume’s “relations of ideas” and the positivists’ “analytic statements,” read “conceptual analysis.”  Hence the Ladyman/Ross thesis that if a proposition isn’t a claim of natural science/”naturalized” metaphysics, then the only other thing for it to be is “conceptual analysis,” is essentially just a riff on Hume’s Fork.  And it has the same problem.  For the Ladyman/Ross thesis is not itself either a claim of natural science/”naturalized” metaphysics, or knowable via “conceptual analysis.”

Of course, some “naturalized metaphysicians” might suggest that neuroscience or cognitive science supports the Ladyman/Ross thesis, but if so they are deluding themselves.  For the actual empirical results of neuroscience or cognitive science would support the thesis only if interpreted in light of a naturalistic metaphysics, but not if interpreted in light of (say) an Aristotelian metaphysics, or an idealist metaphysics, or a panpsychist metaphysics, or a Cartesian metaphysics, or a Whiteheadian process metaphysics, etc.  Hence any attempt to appeal to the results of neuroscience or cognitive science naturalistically interpreted, in order to support the Ladyman/Ross thesis, would be question-begging.

So, the fourth influence on Pigliucci’s skepticism about metaphysics really gives him no better a reason for his skepticism than the first three do.  Nor is the self-refutation problem the only problem with the critiques of traditional metaphysics in question.  Another problem is that the verificationist, Humean, and Ladyman/Ross objections all presuppose too narrow and parochial a conception of metaphysics.  In particular, they tend unreflectively to frame the issues within a rationalist/empiricist/Kantian dialectic inherited from the early moderns.  But the Aristotelian-Scholastic tradition -- against which these early modern positions reacted and defined themselves -- rejects the basic assumptions underlying them.   

Like the rationalists, Aristotelian-Scholastic philosophers hold that there are metaphysically necessary truths which can be known with certainty, but they reject the rationalist view that such truths are innate or that metaphysics is an essentially a priori discipline.  Like the empiricists, Aristotelian-Scholastic philosophers hold that our concepts and knowledge derive from experience, but they also reject both the empiricists’ desiccated conception of “experience” and the empiricist tendency to conflate the intellect and the imagination.  They regard the intellect as capable of “pulling out” from experience far more than either the rationalist or the empiricist supposes.  Hence they reject the assumption that if a proposition isn’t empirical in the thin empiricist (as opposed to thick Aristotelian) sense of “empirical,” then it must be a matter of “conceptual analysis,” with the only remaining question being whether “conceptual analysis” is to be understood in rationalist, Humean, Kantian, Wittgensteinian, Strawsonian, or Frank Jackson-style terms. 

Thus, when Ladyman and Ross -- with, it seems, Pigliucci’s approbation -- describe contemporary “conceptual analysis” and “intuition”-based metaphysics as “neo-Scholastic,” they demonstrate thereby only their own utter ignorance of (or, worse, perhaps indifference to) what Scholastics themselves actually believe.  For from an Aristotelian-Scholastic point of view, contemporary “conceptual analysis” and “intuition”-based metaphysics is essentially an anemic successor to early modern rationalist metaphysics -- a metaphysics which Scholastics would reject, and which defined itself in opposition to the Aristotelian-Scholastic tradition.

As an example of the sort of thing he regards with suspicion, Pigliucci cites the contemporary metaphysician’s appeal to “conceivability,” as in arguments to the effect that “if it is conceivable, say, that there could be a being that is made exactly like me, atom per atom, and who however doesn’t experience any phenomenal consciousness, then this is sufficient to show a lacuna in physicalism.”  Writes Pigliucci: “I reject the very idea that conceivability is a reliable guide to metaphysics at all.”

The example is ironic in two respects.  First, Aristotelian-Scholastic metaphysicians would agree that conceivability doesn’t have the significance for metaphysical inquiry that many contemporary analytic metaphysicians suppose it to have.  But second, it is quite comical for someone who thinks Hume a paradigm of “no nonsense” anti-metaphysical thinking to cite the appeal to conceivability, of all things, as an Exhibit A piece of metaphysical sleight of hand.  For the principle that “whatever we conceive is possible, at least in a metaphysical sense,” is, as is extremely well known, a key component of Hume’s own method.  (The quote is from the Abstract of Hume’s Treatise of Human Nature.)  For example, this conceivability principle is central to Hume’s critique of the principle of causality, a key thesis of traditional metaphysics.  To reject the conceivability principle is thus to reject precisely one of Hume’s key weapons against Scholastics and rationalists alike.

But it’s worse even than that.  Hume conflates intellect and imagination, so that to “conceive” something is, for him, essentially to form a mental image of it.  This “imagist” account of concepts has been widely regarded as a philosophical howler at least since Wittgenstein (though any Scholastic or rationalist could have told you what is wrong with it).  The Humean thesis that we can read off sweeping metaphysical conclusions from the mental images we form is a thesis far more preposterous than any of those held up by Pigliucci for ridicule.

A further irony: Pigliucci (no surprise) makes some dismissive remarks about theology, a subject about which he seems to know as much as he knows about Scholastic metaphysics, viz. not much at all.  In particular, he evidently knows nothing about the crucial role played historically by the theological voluntarism of Ockham and Nicholas of Autrecourt, the occasionalism of Malebranche, and the Cartesian and Newtonian replacement of substantial forms and causal powers with “laws of nature” understood as divine decrees, in setting the stage for the Humean conception of natural objects and events as “loose and separate.”  Understood in light of its historical background, Hume’s philosophy can be seen to owe largely to bad theology.

In fact, when Hume’s various philosophical errors are exposed -- the assumptions inherited from bad theology, the conflation of intellect and imagination, the self-undermining character of Hume’s Fork, and so forth -- little is left in the way of actual argumentation to support the anti-metaphysical and anti-theological conclusions for which he is famous.  His bloated reputation notwithstanding, Hume is exactly what Anscombe said he is: a “mere -- brilliant -- sophist.”

Why that reputation is as bloated as it is, everyone knows:  Skeptics simply like Hume’s conclusions, and don’t care to investigate too carefully how plausible, at the end of the day, are the arguments by which he arrived at them.  F. H. Bradley, though a metaphysician himself, famously characterized metaphysics as “the finding of bad reasons for what we believe on instinct.”  Never was it more obvious than in the case of Hume and his fans how true this can be of opponents of metaphysics.

As it needn’t be said, a lot more could be said.  Since I say a lot more in Scholastic Metaphysics -- about the difference between Scholastic metaphysics and what passes for metaphysics in much contemporary philosophy, about scientism, about Hume’s foibles and intellectual forebears, about laws of nature and much else -- I direct the interested reader to that.