Johnson did not answer …; but talking for victory and determined to be master of the field, he had recourse to the device which Goldsmith imputed to him in the witty words of one of Cibber’s comedies. “There is no arguing with Johnson; for when his pistol misses fire, he knocks you down with the butt end of it.”
— J. Boswell, Life of Samuel Johnson
Dr. Johnson had an unwieldy frame, a face scarred by scrofula, the laughter of a hippopotamus, always left home with a solid stick to defend himself from hoodlums and always argued for victory. After having shot blanks, the anti-realist with something like Johnson’s tenacity will go after the realist with the butt end of his pistol. Even if he concedes that realism about universals is very hard to evade as I tried to argue in the previous post, maybe he can find an objection powerful enough to force the realist into a stalemate.
In this fourth installment, I will consider some of the objections put forward against realism. The first two are somewhat technical; they are here because they illustrate the ways in which anti-realists find universals abhorrent, and also to pave the track to the last three, which form the spiritual core of the post.
A. Identity conditions
Some philosophers (Quine in particular, pressed this objection) argue that compared to classes, the identity conditions for universals are obscure. The identity conditions for classes are clear: two classes are identical iff they have exactly the same members. But no such condition exists for universals, and if there are no identity conditions for the posited entities, then how can there be such entities in the first place? This argument can be interpreted in two ways, an ontological and an epistemological one.
The former does not pose a special problem as the realist can always use a variation of Leibniz’s law of the indiscernibility of identicals now applied to universals. If the latter, then the objection seems to be a variation of methodism applied to the identity conditions of universals, that one must know how one knows before one can know, and if one cannot answer the skeptical question of how one knows, then the skeptic becomes the master of the field.
A first possible response is to deny the need for specifying identity conditions for universals. One reason for doing this is that providing non-circular and informative identity conditions for material objects is equally fraught with problems. For example, certain facts of quantum mechanics make the identity conditions for elementary particles highly problematic. Presumably, this does not bother anyone so why should we be bothered by the absence of explicit identity conditions for universals? Following this, we could adopt some form of epistemological particularism, which seems unavoidable anyway on pain of infinite regress. To see how it can be done, consult J. P. Moreland, Universals, chapter 5, pg. 118 ff.
B. Vicious regresses
Some vicious regress arguments have been aimed at realism. The Third Man argument is a particularly notorious one directed against the Platonist version of realism. Since I favor the Aristotelian-Thomist stripe, I will not bother with it and instead focus my attention on an argument devised by F. H. Bradley. It goes like this. The core of the realist doctrine is the instantiation relation, call it R, between concrete particulars, say x, y, etc. and universals X, Y, etc. Now Bradley says the following: for the relation x R X to obtain, and given the realist ontological commitments, then both x and X must enter into a relation with R and so on ad infinitum, and the realist has a vicious regress in his hands.
But the realist can respond that just as one does not need superglue to connect two objects to normal glue in order to tie them together with normal glue, relations do not need to enter in relations with their relata for they to relate those relata to each other. In the same way, the fact that x instantiates X does not need putative higher-order relations R’, R” such that x R’ R and X R” R’ if R is to relate x and X. As a primitive metaphysical fact, the instantiation relation, or nexus or tie if we want to avoid any confusion, connects particulars and universals and does not need any mediators for the relation to obtain.
The two previous objections are admittedly of a technical kind. The next three on the other hand are probably what hover on most anti-realist minds. They usually come tied together by some prior metaphysical commitment, for example, to naturalism. This is spelled out in the fifth and last objection. To be fair though, the problem, from a Thomist point of view, lies in assumptions going back to Descartes and co., naturalism just being the logical outcome of rejecting the Thomist essentialist picture. It is not, as many ignoramuses think, that the Thomist will want to hold on to his metaphysics with its “ghostly” entities as the last bulwark against the onslaught of materialist science that explains, and explains away, everything as congeries of particles in motion. For the Thomist, the modern metaphysical conception of matter is just as wrongheaded, if not more, as the conceptions of the soul or the mind. And just in case it is not clear, let me repeat that the bone of contention is metaphysical, not scientific.
C. Ockham’s razor
One common complaint is that the realists’ ontology is bloated, dragging in together with the common objects of our experience an extravagant abundance of abstract objects like universals, relations, properties, etc. The anti-realist might even say that universals, for example, are invoked to explain features of language, and that a wielding of Ockham’s razor should instead cut them off. But what are we to make of this suggestion? Given the arguments to the effect that an appeal to universals is necessary to not only explain linguistic features such as predication and abstract reference but that it is also the ground for a robust correspondence theory of the truth, Ockham’s razor only applies if the anti-realist can offer a better and simpler account of such phenomena. However one judges realism, it is clear that such an account has not emerged so the argument has little force.
D. The epistemological objection
A fourth objection is the so-called epistemological challenge. If universals are abstract objects as the traditional realist contends, then they do not exist in space-time. If they do not exist in space-time then they are causally inert. And if they are causally inert how can we ever perceive them and come to know them? In the philosophy of mathematics, the epistemological challenge was launched in the influential article of P. Benacerraf, “Mathematical Truth”. If you are like me and favor an Aristotelian-Thomist metaphysics, reading the article followed by a reading of modern responses to it such as B. Hale and C. Wright’s “Benacerraf’s Dilemma Revisited”, for all the undeniable virtues of these articles, it makes for something of an infuriating exercise, for you will note just how much they are suffused with metaphysical assumptions that a Thomist will outright reject.
What do people mean when they say that abstract objects are causally inert? Usually, it is meant in the sense that abstract objects have no efficient causal powers, but efficient causality is just one of the four modes of causality, and its modern usage is a narrow construal of the classical one at that. But why assume that to know the abstract universals greenness or triangularity, there must be a physical chain of efficient causation involved?
For Aquinas, when we sense-perceive that Fido is green, followed by the intellectual act of abstracting the universal greenness, the same universal that exists in re in Fido comes to exist post rem in the intellect, in a different mode of being. For Aquinas, an intellect just is the kind of thing that can grasp or get a hold of universals without instantiating them, that is, that can grasp greenness without itself becoming green. This is not only what guarantees the objectivity of knowledge but its truth, the correspondence between the universals abstracted by the mind in act and the universal as instantiated in the concrete particular.
But even leaving Aquinas’ epistemological account aside, it is simply not clear why abstract objects cannot be the objects of thought. For we can think about non-existent objects, such as unicorns, a thing which even the anti-realist must concede. The anti-realist may retort that such fictional objects like unicorns, if they existed would be material objects and thus capable of being perceived and it is precisely because we do not perceive them that we have grounds to say that they do not exist. But this is muddleheaded; for given that abstract objects do not exist in space-time, our failure to perceive them tells us nothing whatsoever. Moreover, given that we can think about them, even pose the question of whether they do exist or not as extra-mental entities, it seems it is at least metaphysically possible that they do exist. And given that we can reason about them, as we surely can, it seems at least possible that we can reason to their existence, which is precisely what the realist will contend he
E. The naturalist objection
An elaboration of objections C and D is the naturalist charge and contain the metaphysical commitment underlying not only most of the objections but the very rejection of realism. Naturalism is notoriously hard to define; here I will take it to be the view that the spatio-temporal physical universe of entities studied by science, especially the hard empirical sciences, is all there is. Everything that exists is located in space-time and is part of the efficient causal system known as the universe. If naturalism is true, then it is clear that the traditional realist account must be false.
But what arguments are there for naturalism? As far as I can glean from what the Prophets of the Sect affirm, it all boils down to two arguments: first, non-naturalists have no evidence for the existence of non-physical entities and second, the inductive successes of the modern empirical sciences give strong evidence that concrete particulars are the only thing that exist.
Starting with the former, the claim is patently false. This series of posts is evidence for the existence of one type of non-physical entities, universals. And this is just one, rather paltry example. But even if we granted the truth of the claim (which to repeat myself, I do not, not even for a second), the only thing you could squeeze from it is that all the purported arguments for the existence of non-physical entities fail. This does not by itself give warrant to believe in the claim P = “non-physical entities do not exist”. For if it did, it would mean that you would be warranted in believing in P without the least shred of evidence in favor of it which is absurd. So by itself, it is useless.
And we come to the second argument. This argument only has force if we judge the modern, hard empirical sciences to either exhaust the field of knowledge, in the sense that everything can be ultimately reduced to scientific explanations, or that the empirical sciences are the ultimate epistemic arbiter of warranted knowledge. The first is patently false; mathematics cannot be reduced to the empirical sciences. It is false, if the arguments in this series are correct. There are other arguments that purport to show this, but here, the only thing I need to notice is that this is nothing more than a promissory note, something like “We have not pegged it yet, but just you wait”, so the appropriate response is to call on the bluff.
The second take is either circular or self-refuting. To put it in a different way: what is the object of proper study of the hard empirical sciences? Material bodies in motion or change. Is it a great wonder that the hard empirical sciences have not found anything besides their object of proper study? This is like a man trying to convince us that there are no stars by saying to look through the microscope and point out that there are none to be found upon looking. This is one of the reasons why many science-fetishists when pressed against the wall to justify their metaphysical foundations, will retreat more often than not to a bizarre concoction of relativism and pragmatism, which itself is ultimately circular or self-refuting. The ultimate irony being that this stance is what most undermines the science to which they pay lip service.
Maybe the naturalist will retort that I am mischaracterizing the argument. What he is saying is that if such non-physical entities existed, then we should expect to see physical effects of their existence. Facepalm. No, we do not expect to see such a thing, because the traditional realist will concede that abstract objects are causally inert, causality understood in the sense of efficient causality operative in the empirical sciences. So what is the naturalist asking? Maybe the naturalist intends the request only against God and such “personal”, “supernatural” entities. The argument is still worthless, but as far this post is concerned, discussion is over.
Oh what the heck, I still want to know what is this particular brand of science-beholden naturalist asking? Let us plumb these depths of irrationality. A personal intervention from God in his life? That is putting the evidential bar a little too high, methinks. I would advise such a person to read the epistle of James 1:5-8. And even if he did witness a miracle what is to stop him from saying that aliens with ultra-advanced technology are just goofing around and thus give a completely naturalist explanation?
Maybe the naturalist is asking for something more modest, a state of affairs that is not explainable by natural science. But what would this be other than a gap argument? Gap arguments are fallacious. There are now three options. One, the naturalist is asking for a gap argument not knowing that it is fallacious, in which case he is clueless. Two, he does know it, in which case it is a mere rhetorical ploy and he is intellectually dishonest. The third and last option is that he knows that gap arguments are fallacious, but given his (self-refuting) epistemic strictures he then must say that there can be no evidence for God. This I presume, is what in some quarters passes for being open-minded and committed to evidence and reason…
Since naturalists tend to be science-fetishists, here is a final argument in favor of realism. Consider any given scientific law, say Newton’s law of gravitation, and for the sake of simplicity assume it obtains exactly in our universe. If the anti-realist is right and only concrete particulars exist then this law is a law about pairs of concrete particulars, concrete material bodies with specific mass and at a specific distance from each other. But if the law is merely a law about concrete particulars it could be a mere “cosmic accident” and there would be no reason why it applies to any future contingent, existent pair of concrete particulars.
In other words, it fails to account for the truth of the corresponding counterfactual conditionals. But what becomes of the predictive power of science if it cannot account for counterfactual conditionals? Vanished, poof, gone with the pigs. To avoid this and other difficulties, we must accept that a law is a relationship between universals and that it applies to all concrete particulars instantiating the relevant universals. For example, in case of Newton’s law the universals involved are mass, force and distance. And insofar as the aim of science is to uncover objective, mind-independent truths, the subjects of those truths, universals, must be objective and mind-independent. Thus, it seems acceptance of science commits us to the existence of universals.
As the saying goes, one man’s modus ponens is another man’s modus tollens. Realism is true so naturalism is false. Good riddance.
 There is an obvious joke lurking here. But serious, sober philosophical discourse is above such petty vulgarities. (He He He He He).
 Sometimes one hears that abstracts objects exist outside of space-time, but this is confused and confusing as “outside” implies a spacial relationship which contradicts the fact that abstracts do not exist in space-time.
 Benacerraf’s article appeared in Journal of Philosophy, 70 (1972) 661-680 and is reprinted in Philosophy of Mathematics: Selected Readings, ed. by P. Benacerraf, H. Putnam. The article of B. Hale and C. Wright can be found here.
 I am glossing over many details, such as the difference between sense and perception, the role of phantasms, etc. The crucial point is that Aquinas is adamant on a distinction in kind between the animal powers of sensation and perception, memory and imagination on the one hand and the intellectual activity proper on the other, to which the powers of abstraction, concept formation and the various modes of reasoning belong. This distinction, lost to many modern philosophers, is at the heart of many (misguided) objections against realism.
 Aquinas’ epistemology follows his metaphysics, as one should expect in a realist. A good starting point to know more about it is S. MacDonald, “Theory of Knowledge” in The Cambridge Companion to Aquinas, ed. by N. Kretzmann and E. Stump.
 A modal argument for the necessary existence of abstract objects, numbers more precisely, parallel to Plantinga’s modal ontological argument can be found in S5, God and Numbers by Chad McIntosh. One obvious difference between the two arguments is that the crucial premise in the case of numbers is much harder to dispose of, and is in fact practically all but granted by many (most?) philosophers.
 And thus, by shedding a sound, realist metaphysics and philosophy of nature, another typical modern “problem” is born, the problem of induction.
 For an elaboration of the argument from science in favor of realism, see D. M Armstrong, What is a law of nature?