Robert Pasnau’s paper presents a strong thesis, which it does not manage to substantiate. The thesis in question is that the Aristotelian doctrine of the identity of the knower and the known, as interpreted by St. Thomas, cannot possibly be used to fend off skepticism.

[Before addressing this thesis, however, I have to note that the alleged "tension" between Aristotle’s and Aquinas’s understanding of the doctrine is entirely spurious. In the passage referred to in Aristotle’s De Anima (bk. 3, c. 2, 425b26-28), and in Aquinas’s commentary on this passage (In De Anima, bk. 3, lc. 2, nn. 590-593.), the two authors are talking about the numerical identity and conceptual diversity between an act of cognizing and the corresponding act of being cognized. In fact, both of them understand this claim as being a direct consequence of the general principle in bk. 3. of the Physics (bk. 3, c. 3, 202a12-b29; Aquinas, In Phys. bk. 3, lc. 5.) concerning the numerical identity and conceptual diversity between the act of the agent and the act of the patient. But, of course, this claim is not the theory under discussion, which concerns the formal identity, along with the numerical diversity, between the cognizing subject and the object of its act of cognition. So the discussion concerning this claim, which both authors endorse, and which both of them treat as distinct from the formal unity doctrine, is simply irrelevant from the point of view of the principal thesis of the paper.]

What seems to be the main argument for the author’s thesis crops up first on the fifth page of the paper, and it is repeated on page 9, in the context of a discussion of "representationalism" vs. "direct realism"—issues which I will consider later. The point of the argument seems to be that any reasoning trying to use the theory of formal unity against skepticism is necessarily question-begging. However, I think this charge is based on a misunderstanding of the dialectical situation, which in its turn is based on a misunderstanding of the point of both the skeptical challenge and the Aristotelian-Thomistic rebuttal of the challenge. So, let us see first this "dialectical situation".

It is the skeptic who starts the trouble. The philosophical tales about the perfectly life-like dream, about the perfectly deceptive demon, or, for that matter, about the perfectly mad scientist keeping our brains in vats, are designed to establish the skeptic’s principal premise that it is possible that all our acts of awareness (be they perceptual or intellectual) are non-veridical. But since knowledge could be attained only if we had at least some veridical cognitive acts, the skeptic concludes that it is possible that no knowledge is attainable by us.

However, Aristotle’s theory of formal unity entails the epistemological consequence that all our simple cognitive acts are necessarily veridical in respect of their proper objects. This is a conclusion drawn explicitly both by Aristotle and by Thomas, with respect to simple perceptual and intellectual acts, and their corresponding cognitive powers (nulla virtus cognoscitiva circa proprium obiectum decipitur, SCG lb. 3, c. 108, n. 4.). But then, if the theory of formal unity is true, the main premise of the skeptical argument is false.

This simple line of reasoning, as far as I can see, constitutes a perfect example of using the theory of formal unity against the skeptical argument. Indeed, a Thomist who puts the theory to such use does not have to beg the question against the skeptic at all. Of course, the Thomistic rebuttal would constitute question begging if the only basis of the theory of formal unity were the bland rejection of the skeptic’s main premise, namely, the simple, unjustified assertion that at least some of our cognitive acts are necessarily veridical. But this is not the case. The theory of formal unity is based on entirely independent grounds, namely, on the Aristotelian analysis of what a simple act of perception of a proper sensible is (actus sensus proprii), and what a simple act of intellectual apprehension is (indivisibilium intelligentia), distinguishing these acts from other mental acts, such as dreaming, imagining, remembering, judging, etc. The Aristotelian analysis of simple perception (and the analogous analysis of simple intellectual apprehension) as the receiving of the form of the object without its matter is based on the understanding of the actualization of the cognitive subject in respect of its object, but without actually becoming that kind of object in esse naturale. This understanding, in turn, is based on the Aristotelian analysis of the principles of change, establishing the basic conceptual relationships between the notions of being, form, matter, actuality, potentiality, etc. This analysis in the first approach has nothing to do with the skeptic’s concerns. (In fact, it could be carried out in entirely phenomenalistic terms). Thus, the epistemological consequences of this analysis come into play only much later, when we realize the fact that this analysis establishes an essential relationship between a simple cognitive act and its object. For an important consequence of the doctrine of formal unity is that it is part and parcel of the identity conditions of such a simple cognitive act what it is the cognition of. Therefore, that such an act is necessarily veridical is the consequence of the fact that what it is for something to be such a cognitive act is for it to be the form of such and such an object received by the cognitive subject, and so anything that is not so related to this object, as for example an act of hallucination induced by the demon, is not such a cognitive act.

Of course, at this point the skeptic could argue further, shifting his position, by saying that, well, perhaps, some cognitive acts may thus be necessarily veridical, but we may never have those, or, in any case, even if we have them sometimes, we can never tell the difference. But that would be a matter of further dispute, leaving the first argument of the skeptic destroyed, precisely on account of the falsity of its main premise, shown to be false on the basis of the theory of formal unity. In any case, whatever these further moves might be, the paper is not moving in this direction. Rather, it addresses the merits and demerits of this theory in establishing a "direct realist" position, as opposed to a "representationalist" position.

Pasnau’s argument may be reconstructed as follows. Thomists argue that the theory of formal unity entails direct realism. But the Thomistic argument is not valid. Therefore, the theory of formal unity cannot be used to establish direct realism, and hence, possibly to avoid skepticism. (Well, this latter implication is not stated in the paper, but what else can be the point in relation to the main thesis?)

There are several flaws in this line of reasoning. First of all, even if the alleged Thomistic argument that Pasnau criticizes is invalid, it does not follow that some other argument could not establish direct realism on the basis of the theory of formal unity. Second, it is not clear at all that even if direct realism could not be established in any way on the basis of the theory of formal unity, it would be impossible to establish on this basis some form of representationalism which could still be effective against skepticism. Third, the argument Pasnau considers is not relevant from the Thomistic point of view, so any refutation of it is irrelevant. Finally, the whole discussion is marred by the lack of a clear and proper distinction between an immediate and the ultimate object of a cognitive act. So, to substantiate the foregoing claims I have to begin with this distinction. Then, having the distinction at our disposal, I will argue for these claims in the reverse order.

When you look into the mirror to fix your tie, you see your tie only through seeing its reflection. Still, of course, you do not fiddle with the reflection to fix your tie. Instead, you reach for your tie, because what you see by looking into the mirror is your tie, the ultimate object of your act of sight, which you see through its immediate object, the reflection. Indeed, for the reflection to be this immediate object is for it to function only as something that directs your act of sight to its ultimate object. That is to say, to be this immediate object is to be recognized only as something through which [quo] you see the object you want to see, and, at the same time, not to be recognized as that which [quod] you want to see, as the ultimate object, to which your intention, attention and action are directed through or by the former.

Now, in view of this distinction, if the species of a thing is to be regarded by Thomists as the immediate object, whereas the thing itself as the ultimate object of a cognitive act, then it should be clear that Thomists may not be particularly interested in numerically identifying the species with the thing. In fact, Thomists might recognize as many numerically distinct immediate objects as they like, provided these all are thought to function only as through which the ultimate object is recognized, just as two reflections would function as such intermediary objects if you used two mirrors, say, to check the back of your jacket. What is more, given St. Thomas’s own distinction between species intelligibilis and intentio intellecta, the fact of the matter is that for Thomas himself there are two such intermediaries involved in any intellectual act. Nevertheless, through these immediate objects, despite the numerical distinctness of their immediate and ultimate objects, our cognitive acts are essentially connected to their ultimate objects, in virtue of their formal unity. So pointing out the invalidity of an argument that is supposed to show the numerical identity between a species and its ultimate object is quite irrelevant from a Thomistic point of view. On the other hand, if we understand the principal premise of the same argument, namely, "if X=Y then apprehending Y just is apprehending X", as using the notion of formal identity, and as taking Y for the immediate, and X for the ultimate object of an act of cognition, then it states a trivial truth. For, of course, checking your tie in the mirror is just one act of seeing by which you see your tie through seeing its reflection; so seeing the reflection is seeing the tie, precisely because of the formal unity of the tie and the reflection, and because of the fact that the reflection is an immediate, while the tie is the ultimate object of the same act. In this case, however, the observation that St. Thomas insists that the species is not the ultimate object is again irrelevant, since the argument interpreted in this way does not deny this at all.

Secondly, even if admitting a species as an immediate object should make the Thomist into a "representationalist", this type of representationalism, coupled with the theory of formal identity will entail the veridical character of the species, and so it will provide grounds for rejecting the principal premise of the skeptic’s argument. So no skeptical harm should come to a Thomist, even if he or she is proven guilty of "representationalism". Indeed, the label of "representationalism", without a proper understanding of the distinction between immediate and ultimate objects, and without considering whether they are essentially or merely accidentally connected, does not carry any serious theoretical significance.

Finally, as we could see, the theory of formal unity guarantees an essential connection between the species and the ultimate objects of our simple cognitive acts. These ultimate objects, however, are nothing but the things of the sensible and intelligible reality. So through the species our simple cognitive acts necessarily "reach right up to" things of the sensible and intelligible reality. But since, therefore, in virtue of the theory of formal unity our simple cognitive acts essentially "reach right up to" things, and hence they are necessarily veridical, it is entirely irrelevant from the point of view of rebutting the skeptical challenge that they do so by the mediation of the species (and of the intentiones, for that matter). Therefore, I see no reason why we could not say that this theory can serve as the foundation of a certain form of "direct realism"—well, whatever extra advantages there might be in obtaining that label.

A.P.A. Colloquium:

Medieval and Renaissance Philosophy

Chicago, 17 E. Monroe Street

Palmer House Hilton, Private Dining Room #6

April 25, 1996, 2:45 p.m.