Feser's Scholastic Metaphysics: A Book Review, Part II

Feser is a good writer. Like the best analytics, he's clear without sacrificing precision, is on occasion entertaining, and spices his abstract discussions with plentiful concrete examples, something non-analytic philosophy could use a lot more of. Scholastic Metaphysics: A Contemporary Introduction was by and large a pleasure to read. Keeping in mind that it is an introduction and that we shouldn't expect an exhaustive treatment of any given topic, there's a lot in the book I would recommend to any amateur looking to get a handle on the issues covered, for example: the "prolegomena" chapter against scientism and conceptual analysis, the discussion of causal powers in chapter 1, the discussion of the Principles of Non-Contradiction and of Sufficient Reason in chapter 2, the discussion of essentialism in general in chapter 4. I learned a good deal from Feser's survey of analytic positions and his responses to them. To take one example, section 3.3.2.1, "Against four-dimensionalism", presented a position I knew about but frankly had not taken seriously since I was a teenager, and broke down the reasons not to take it seriously in a way I wasn't familiar with (as I said in part I of my review, my familiarity with the state of the art in contemporary metaphysics is limited). I particularly approve of the way Feser frequently appeals to retorsion arguments, which I think are underdeployed against stupid philosophies.

There is, then, a lot that's good in the book. There's also plenty I think is less good. One general observation is a tendency throughout to present the Thomist position on a topic while putting off actually arguing for it. Over and over again the reader encounters remarks to the effect that "my position is this, but the reasons for it depend on something I'm going to say in a later chapter"; this gives the impression of getting the run-around, as though the good deep arguments are always just around the corner. I emphasize that Feser does not always do this; but he does it enough for it to be frustrating.

Sometimes the arguments are not put off, they're simply omitted. The Thomist doctrine is instead handed to the reader on a plate without justification. An early glaring example is section 1.1.3, "Divisions of act and potency". Feser writes on page 38, "Given the distinction between act and potency, quite a few sub-distinctions can be made and commonly are made by Scholastic writers." He then spends the next few pages unloading a boatload of distinctions on the reader without bothering to establish the reality of any of them or show that they are more than ad hoc. I'm not saying it can't be done, Feser doesn't do it and I have a hard time seeing how anyone who doesn't already consider his traditional conceptual apparatus as authoritative is going to be motivated to absorb it.

In my opinion there are also some structural problems. For instance, in my opinion the treatment of causality is pretty seriously defective. Anyone who's read much Feser knows that one of his biggest concerns is to defend the reality of final causality against reductionists who want to eliminate all but efficient causality. This is a project I am fully on board with. Unfortunately this concern leads him to begin his discussion of causality with final vs efficient causes, which is a misstep. Material and formal causality are put off until the following chapter, under the discussion of substance. The result is that the nature and force of the reasons for accepting the reality of final causality always remain somewhat obscure, because final causality is unintelligible without formal causality. Feser is quite right to bemoan the elimination of final causality by the moderns, but why did they do it? Because they eliminated formal causality first, as the most casual reading of, say, Hume (the early modern Feser devotes the most time to, with good reason) will confirm. A huge part of the revolution in modern philosophy and science was to replace the total formal cause of a thing with that dimension of it which is subject to mathematical formalization, and, as Aristotle pointed out, mathematical objects precisely as such prescind from the good and the end. Reducing form to quantity wipes out the notion of finality. The proper way to get back to final causality is to reinstate the robust notion of form; and this is, by the way, the order the causes are treated in in the standard neo-Thomist manuals I'm familiar with. In taking things backwards I think the clarity and rigor of Feser's exposition suffers.

Now let's look at some of the book's content from a specifically Scotist perspective. Consider 1.3.1, "The Scholastic theory of distinctions". Feser gives the definitions of real vs logical distinctions. A real distinction "reflects a difference in extra-mental reality" and a logical distinction "reflects only a difference in ways of thinking about extra-mental reality". He then subdivides logical distinctions into pure and virtual. A purely logical distinction is merely verbal while a virtual distinction "has some foundation in reality". And so on. It's Thomist boilerplate. As expected, then, when Feser presents Scotus' formal distinction, he gives it short shrift. According to Feser, to take his example, rationality and animality in man are virtually distinct. This is a logical distinction because the two are not separable and not really distinct in the thing, and so are really identical. Feser says that a virtual distinction "may appear at first glance to be hard to distinguish from a real distinction. But the key to understanding the difference between any logical distinction and a real one is this: IF the intellect's activity is essential to making sense of a distinction, it is logical; if not, it is real." Good. But the whole point of the formal distinction is that it picks out realities which are inseparable in the thing but are distinct aside from the activity of the intellect considering it; that's why Scotists say that it is a distinction ex parte rei or ex natura rei and deny that it's a logical distinction: because according to them, rationality and animality in Socrates are really not identical in Socrates, even before anyone thinks of them, and even though they are inseparably united in him.

Feser says that it's hard to see how the formal distinction can avoid collapsing into either a real distinction or a virtual or logical distinction. The short answer to this is that Thomists play a shell game with the notion of real distinctions: sometimes they act as though separability is an obvious criterion and sometimes as if it isn't. The Scotist position is that a fully real distinction in general is one to which the separability criterion applies (with a very few special exceptions), and that the formal distinction is a species of lesser real distinction to which the separability criterion does not apply. It's not a virtual or logical distinction because, to take Feser's example, animality and rationality are really non-identical prior to and aside from any consideration of the intellect. It's not a "fully" real distinction because Socrates is one animal and is one rational thing, and both of those are one real thing, not two. You can't separate the animal and the rational thing in Socrates the way you could separate his arms from his trunk or his substance from one of his accidents (say, his location).

Feser's account of the formal distinction doesn't address this sort of consideration. Instead it begs the question, and avoids the fact that the various varieties of virtual distinction he lays out were formulated by Thomists specifically in order to jimmy out of admitting the formal distinction as a lesser variety of real distinction, because to do so would threaten other areas of Thomist metaphysics. Feser writes, "For either the intellect plays some role in the distinction or not." The Scotist replies that it does not play a role in making the formaliter distincta, the things which the formal distinction distinguishes, distinct; but it does play a role in how we articulate the distinction, since we typically distinguish formalities by their formal ratios or definitions, i.e. by their formal contents with reference to whether they can be defined and thus apprehended independently of each other. The non-identity of inseparably united formalities is identified by our ability to conceive them independently, not constituted by it.

According to Feser, "Whereas if they are distinct because the intellect separates out the [formalities] ... then we have a logical distinction with a foundation in reality, namely a virtual distinction. There just doesn't seem to be some third, 'formal' distinction." Again this is question-begging, for it is precisely this Thomist tendency to handwave away the "foundation in reality" without explaining it that leaves them unable or unwilling to account for formalities - not incidentally, since the motivation for all this is the Thomist commitment to preserve the unicity and simplicity of all forms. The Scotist insight that really disturbs Thomists is not the formal distinction itself but what it's there to cope with, the layers of a kind of ontological complexity in even "simple" beings. Feser suggests that perhaps the Thomist and Scotist theories differ primarily in emphasis: "Scotus, on this interpretation, is merely concerned to emphasize ... the fact that virtual distinctions are grounded in mind-independent 'formalities'". If the dispute is really about terminology and you prefer "virtual" to "formal, there's no real argument. But it isn't. The whole question is: is fA distinct from fB when inseparably united in x even when we're not thinking of it? In x, does fA=fB? We say no, that when either a) I can define them independently, showing that they have nonidentical formal content, or b) when I can show that A and B even if they can't be separated in x can exist independently of each other in y and z, then this indicates that even in x they are non-identical, distinct among themselves formally and not merely in how they can be considered by us, logically.

As I mentioned earlier, Thomists are unable to maintain the separability criterion for a real distinction (when, for instance, it comes to essence and existence), and they're happy to abandon it all over the place precisely in order to avoid endorsing the formal distinction where we think it obtains. Feser says on page 76, "If every real distinction entailed separability, then there would have to be some intermediate, 'formal' distinction between a real distinction and virtual distinction; but there is no such distinction, since the formal distinction collapses on analysis into either a real distinction or a virtual distinction; so not every real distinction entrails separability." I hope it's clear by this point that this is arguing in a circle. I could just as easily respond: since a real distinction entails separability, and A and B are distinct aside from any operation of the intellect, therefore we must accept some distinction besides real and logical or virtual ones.

The argument Feser takes from Oderberg on the same page doesn't help him, since it confuses properties in the sense of predicates with properties in the sense of real ontological items. Oderberg argues that "there is obviously a real distinction between the properties having a radius and having a circumference" but that having a radius can never exist in a circle without having a circumference. But "having a radius" etc is not a real being in the sense that the radius is. In a given circle the circumference and the radius have a necessary ratio to one another, but as two lines they are really distinct: I can draw the circumference but not draw the radius, and vice versa. In a circle in which the circumference but not the radius is drawn you might say that the radius has a virtual existence in the circle, since its length is necessarily determined whether it's drawn or not (Scotists talk about virtual distinctions in a different sense than the Thomist one Feser gives here: for instance, to describe the way an effect exists in its cause). But they are really distinct: one can actually exist without the other. And they are formally distinct: they have different essential properties: the circumference is necessarily curved and the radius is necessarily straight, etc.

Now some of these disagreements are merely terminological and some involve obfuscation or misunderstanding. Feser sometimes sounds like they would evaporate if the formal distinction is "really" a real distinction, and sometimes that they would do so if it is "really" a logical or virtual one. This waffling itself suggests some Thomist blurring over for the sake of simplicity. Feser does not seem to consider the position that the formal distinction is diminished variety of real distinction, one in which the distincta can remain ontologically inseparable and so not really distinct in the fullest and most perfect sense.

Clearly to go on like this and show every place in Feser's book where he handwaves away non-Thomistic scholastic alternatives to his position without really giving them a hearing would require a book-length commentary, and I'm not going to do that. Suffice it to say that whether or not you accepts the formal distinction has a huge ramification on the rest of your metaphysics, and all of those areas are dismissed with cavalier breeziness in Feser's book without exception. If I were so inclined I could write at length about my problems with his presentation of the notion of substance, the unicity of form, individuation, the nature of matter, and so on - not simply that he presents a view controversial among scholastics as "the Scholastic position", but also that other positions are grossly mis- or under-represented to the extent that it would be better not to mention them at all than present a caricature. But I'll confine myself to one more example, the crown jewel of Thomist metaphysics, the "real distinction" between essence and existence.

Feser says on page 241, talking about the distinction "commonly drawn in Scholastic metaphysics" between essence and existence, that "Considered by itself, a contingent thing's essence is taken to be a kind of potency, and its existence a kind of actuality." It's important to note, and Feser seems to have no inkling of the fact, that in actual scholasticism this claim in incredibly controversial. Anyone conversant with the "A" side of so-called "A-T" should know that for Aristotle, form is actuality and the essence of a thing is (or includes) its form. Form is the act of which matter is the potency, and saying that the essence is in potency to existence struck many scholastics wrongheaded and as a basic confusion of the role of form in the constitution of a substance. The essence is not like a quasi-matter waiting around to be actualized by existence in a quasi-formal role. Aquinas' doctrine that on one level the essence is the actuality of a substance, while on another level it is a potency to another really distinct actuality, existence, was deeply troubling to many, perhaps most, non-Thomist scholastics. They argued that it reduces existence to a kind of quasi-accident of the form and reified the abstracted essence which, apart from its existence, isn't really in potency to anything. Suffice it to say that a Scotist would distinguish between possibility and potency, and say that an essence considered as abstracted from its existence is a possible, but not that it has potency, since whatever has no existence has no real being, and whatever is in potency to further actualization is real being and something that exists. The whole Thomistic presumption that essence and existence are related as potency and act and analogously as matter to form is very widely rejected in the broader scholastic tradition.

Feser casually lumps Scotus and Suarez together as rejectors of the "real distinction" and says that "in Scotus' view it is merely a formal distinction", since for Scotus potency and act are only formally distinct. He claims this several times throughout the book but never gives a reference for the claim. In fact Scotists do not regard the distinction between act and potency, or between essence and existence, as formal distinctions, for the obvious reason that neither act nor potency nor existence precisely as such can in any sense be considered formalities. The distinctions that obtain between these items are modal distinctions, but of course the whole notion of intrinsic modes (another sort of ex parte rei distinction other than a real or logical distinction), a central pillar of Scotist and thus of vast swaths of scholastic metaphysics, is entirely ignored in Feser's book. But never mind that. The point at present is that the very presumption that the relation of essence and existence is obviously an instance of the relation of potency to act, (where potency and act are distinct modes of really existing being) is suspect from the default scholastic point of view - i.e. starting on a basis of Aristotelianism - and needs at least to be justified and objections to it dealt with. But the only such objections Feser gives are those from analytic philosophers, making it a case of analytic philosophy vs scholasticism, rather than what it really is, an issue in scholastic philosophy with many possible resolutions. Neither scholasticism nor Thomism nor their engagement with modern thought is served by the pretense that the Thomistic position is the obvious default one from within a scholastic framework - it isn't. And frankly, I wonder whether analytic philosophers might not be more inclined to give scholastics a hearing if they were ever told that some of their objections to Thomism, or analogues to them, had been anticipated from within his own tradition, rather than being presented with what seems to be more or less a monolith of thought.

This is only the barest thumbnail sketch of an enormous topic. (I dealt with essence and existence and how differing conceptions of them play out in different fields at much greater length in my doctoral dissertation, if anyone cares to look that up - it's available on this blog.) I'm going to end this review by repeating that Feser's book has much to commend it, but many defects as a presentation of scholastic metaphysics as such. He makes very little attempt to even acknowledge alternative possible viewpoints within his own tradition, and when he does he makes very little attempt to get them right, instead presenting the most airy caricatures in order to blow them away with the merest breath. The most disappointing thing about this is that this is exactly what he frequently complains of contemporary philosophers doing to scholastic (i.e. Thomist) views. For example, a very recent post showed Feser complaining of the casual dismissal of the cosmological argument by philosophers who clearly had no idea what they were talking about, an all-too-common phenomenon. Sadly I see Feser and nearly all Thomists doing the exact same thing to their closest philosophical relatives, casually dismissing non-Thomist arguments and positions while making it abundantly clear that he hasn't bothered to study them at all. If he wants to be given the courtesy of a hearing and the benefit of the doubt by his enemies, who are both totally ignorant of Thomism and unwilling to consider the possibility that it has something to offer, perhaps he and other Thomists might extend the same courtesy to their friends.