II. Identity, Consciousness, and Psychology

The questions of personal identity and of the nature of mind have, I would argue, been skewed in recent years by the thought that if there is a residual puzzle that has not yet been solved by the twentieth-century's onslaught of materialism, naturalism, and physicalism, it must be the problem of consciousness. Hence the attention that David Chalmers attracted when he published The Conscious Mind,⁶

a book that for many people summed up what has come to be known as the “hard problem.” If there really is something that materialists cannot successfully grapple with, it is the phenomenology of conscious experience, the felt quality of our interaction with the world. Everything else about the mind, according to Chalmers, can be captured within a physicalistic functionalist model. To be sure, there is still the problem of explaining how to identify the correct functional analysis of human psychological operation; but that there is one, and that it is at least in principle realizable in inorganic systems such as computer models, is something already taught to us by cognitive science.

This bifurcation of the question of the nature of the mind—into a question about human cognition on the one hand, and a separate question about the special “problem of consciousness” on the other—and then the subsequent focus on the “problem of consciousness” as the outstanding conceptual issue in the quest for a total naturalistic theory, is, in my view, the biggest wrong turn in the recent history of the subject. First, however, I should explain what I am not claiming. I do not deny that there is indeed a “problem of consciousness,” and that many of the central claims of the nonreductionists, including so-called “naturalistic dualists” like Chalmers, are correct: principally, that there is no explanation of the subjective nature of conscious experience in physicalistic terms. What I do deny, however, is that this is not a problem affecting the psychological in general. For it is at least plausible to claim that there is also a phenomenology of psychology as much as of conscious experience, and the typical responses to such a claim look, as they do in respect of conscious experience, to be question-begging.

By a phenomenology of psychology I mean simply the “what it is like” of ordinary psychological operations such as judging, reasoning, and calculating. There is, I claim, even “something that it is like” to calculate that two plus two equals four. It may not be qualitatively identical for all people, but then neither is the taste of strawberry ice cream exactly the same for all people, one might suppose, while at the same time noting that our similar physiological structures imply that the individual experiences for each kind of act should be highly similar. Indeed, one might assert that these experiences contain a certain phenomenological core, and that the class of such experiences is such that its members are all more similar to each other, all things being equal, than they are to any experience of a different mental act, state, or process.

It might be objected that the phenomenology of calculating that two and two make four, if there were such a thing, would hardly be different from that attending the calculation that four and four make eight, thus reducing the idea to absurdity—a distinction without a difference. Yet this would be as misplaced as denying the distinct phenomenologies of seeing reddish yellow and yellowish red because they are so similar. That there are such phenomenological differences in calculation is not something for which there is nonintrospective proof any more than there is for the standard kinds of qualia to which nonreductionists (such as Chalmers and Frank Jackson) draw attention.⁷

Yet introspection does, I believe, make apparent the qualitative character of calculation, a character easily heightened by comparing, say, the experience of doing algebra with that of doing calculus. Again, there is a conscious experience of performing a piece of deductive reasoning that differs from that attending the judgment of a single proposition. I cannot offer here a taxonomy of such experiences, nor anything like a catalogue of dimensions of similarity such as can be done, to some degree, for the usual perceptual experiences on which the debate always settles. All I propose for consideration is that there is a phenomenology of psychology, whatever the details.

It will not do to respond (as would most defenders of the idea that artificial intelligence captures the essence of human cognition) that since computers can do arithmetic, and by their very nature have no conscious experience, it must be the case that what I claim to exist for people is an illusion. For the response assumes that what we do and what computers do when they calculate that two and two make four is the same in the first place. As a matter of scientific sociology, for what it is worth, no one has the faintest idea of what humans do when they do arithmetic, specifically, what goes on in the brain when even the simplest of calculations is carried out. Ipso facto there is no agreement on what physical system best models what we do.⁸

But the logical point is that one may not assume that what humans and computers do is fundamentally the same; rather, this is a proposition that has to be proven. Moreover, the phenomenological evidence in the human case is so strong that we have a priori reason for thinking that whatever physical model is proposed, it will not capture what we do. One could, of course, seek to show that some physical model captures what we do if one took there to be no problem concerning the reduction of conscious experience in the first place. However, this is a claim that dualists of all stripes deny, so minimizing the problem will gain no traction. Nor, again, is it of any force to claim that since humans can perform unconscious calculation, such an activity can have no phenomenology. For the question is not about what we can do unconsciously. Similarly, if unconscious perception were a genuine phenomenon (a matter of dispute),⁹ this would not disprove the existence of subjective experience during conscious perception. Thus, one cannot neutralize the claim that there is a phenomenology of psychological activity by appealing to unconscious kinds of the same or similar activity.

It might seem to be a somewhat exotic if not irrelevant claim to assert that there is a problem of consciousness for psychology as much as for sensory experience. Yet it is important for our purposes, since it highlights the error for the theory of personal identity of corralling consciousness into a corner of the mind, particularly that corner associated with the mind's lowest function, namely perception. It is no more than a perpetuation of the Cartesian error of identifying the soul with awareness. It positively invites a dichotomizing of the human being into a conscious self plus the physical add-ons, which for the Cartesian dualist means identifying the person with the soul, and, for the reductionist reacting in a perfectly understandable way to the ontological split, means doing away with the Cartesian soul as a piece of obscure metaphysical baggage and reducing the person to some collection or other of physical states of whatever complexity.¹⁰

Dualists must resist both errors, and they can only do so by insisting on the essential unity of the person. To point to the fact that human psychology is shot through with phenomenology is but one way of emphasizing that unity; and it is that unity which is at the heart of the kind of dualism I will set out and defend.

That the problem of personal identity is not primarily a problem about consciousness—at least in the narrow sense that dominates current debate—is also shown by the fact that consciousness does not constitute personhood; rather, it presupposes and reveals it. The point is well known from the classic objections of Thomas Reid and Joseph Butler to the Lockean theory of personal identity:¹¹

there is a vicious circularity in trying to analyze personal identity, as Locke does, in terms of memory or of consciousness in general, since these phenomena presuppose identity (i.e., that it is the same person who remembers or is conscious). Yet it is a point that cannot be repeated often enough. A person is not merely aware—he is aware of something, and that something is, fundamentally, himself. There has, of course, been an attempt to get around the problem by invoking non-identity-presupposing relations such as “quasi-memory,” but such notions are of doubtful coherence at best.¹² Any attempt to synthesize personal identity out of a manifold of conscious states will founder on the task of specifying just what the content of those states is supposed to be, and I take this to be a point extendable beyond persons to the identity of any conscious being, such as an animal. More generally, the circularity objection is a special case of the general one against all attempts to give a non-identity-presupposing, and hence noncircular, theory of diachronic identity (identity over time) for any kind of object—about which I will soon have more to say.

It would be specious to deny that either phenomenology or consciousness in general were relevant to the problem of personal identity: any plausible theory must, for example, account for a person's sense of self as an enduring entity, capacity for higher-order conscious states, and awareness of itself as a being endowed with freedom and responsibility. What I am denying, however, is that the problem of person identity is primarily one about phenomenology or consciousness. Rather, it is about psychology in general, taken in the broad, traditional sense: the problem concerns the specific mental operation of the human being in particular, and of any person at all, whether there be angels, animals that are persons, or other disembodied minds. To broach the problem, we must begin with the concept of form, since this will take us directly to the concept of identity by focusing consideration on the nature, function, and operation of substances.

III. Form and Identity

Here is a standard definition of form: “The intrinsic incomplete constituent principle in a substance which actualizes the potencies of matter and together with the matter composes a definite material substance or natural body.”¹³

It is “intrinsic” because it is a constituent of the substance and solely of the substance. It is a “constituent” in the sense of being a real part or element of it, though not on the same level as the substance's natural parts, for example, the branch of a tree or the leg of a dog; rather, it is a radical or fundamental part of the substance in the sense of constituting it as the kind of substance it is. It is a “principle” in the sense of being that from which the identity of the substance is derived—that in virtue of which the substance is what it is. It is “incomplete” in the sense that it does not and cannot exist apart from its instantiation by a particular individual, contra Platonism. (This does not, however, contradict the possibility of a certain kind of form's existing independently of present instantiation in matter, as we shall see.) It “actualizes the potencies of matter” in the sense of being the principle that unites with matter to produce a finite individual with limited powers and an existence circumscribed by space and time. Together with matter, it composes the distinct individual substance.

These ideas will be expanded as we proceed. For the moment, it is the question of identity itself that needs clarification. The problem of personal identity is a problem about identity over time. Since there is no non-identity-presupposing analysis of diachronic identity in general, there is no non-identity-presupposing analysis of personal identity.¹⁴

The most popular current proposal for analyzing identity over time is the four-dimensionalist account, according to which every persisting object is taken to be a four-dimensional “space-time worm.” Inspired (if not necessarily justified) by contemporary relativistic physics and the supposed amalgamation of the three spatial dimensions and that of time into a “four-dimensional manifold,” this theory has it that persisting objects are really complexes of “temporal parts,” more or less momentary “slices” or “stages” of matter across space-time. What we think of as three-dimensional objects persisting through time are, on this view, four-dimensional objects “smeared out” across the space-time manifold.¹⁵ Yet four-dimensionalism, whatever its version, suffers from many flaws,¹⁶ one of the fundamental ones being that there is no way of analyzing temporal parts that does not either invoke the very phenomenon of identity that is supposed to be analyzed, or else reduce to absurdity by the invocation of literally instantaneous object-stages that cannot give rise to any temporally extended object.

One way out that has gained a little in popularity is to take identity to be primitive.¹⁷

Yet there is a right way and a wrong way of interpreting this. The right way is to take the phenomenon of identity per se to be primitive. In other words, there is no way of defining identity across time in other terms: it is a basic, unanalyzable phenomenon. The wrong way is to take it as meaning that the identity of specific material substances themselves is primitive: in other words, it would be incorrect to claim that when it comes to identity, nothing further can be said about why it is that an object of a certain kind, existing at a given time, is numerically identical to an object of a certain kind identified at a later time; or why an object at one time is identical to this object rather than that object at a later time. It would, to elaborate a little, be wrong to claim that when it comes to kinds of things, the criterion of identity for a given kind is primitive, that nothing further can be said about why, say, objects of kind K continue to exist in certain conditions but cease to exist in others—other than that's just how things are for things of kind K. But even if a sympathizer with nominalism were to say that there are no real kinds of objects, that every object is purely an individual, it would still be wrong to assert that nothing further can be said about why individuals persist in these circumstances rather than those.

The reason the wrong way is wrong is that it simply ignores self-evident truths of identity. We can explain why it is, for instance, that Bessie the cow seen at t₁ is not identical to Rover the dog observed at t₂, and why Rover at t₂ is not the same as Fido at t₃—and why, say, a Lego house at t₃ is distinct from the pile of Lego bricks at t₄ that constituted it at t₃. In all such cases, we do not rest content with saying that Bessie is Bessie and Rover is Rover, that Fido and Rover are just not the same, and that a Lego house is something different from its Lego bricks. Even if the criteria of identity invoked are quite simple, they are informative: a cow and a dog are different kinds of animals; this cow and this dog have different properties; the two dogs are of different breeds, or else differ otherwise in their properties; a pile of Lego bricks does not make a house; and so on. The notion of primitive substance identity does not explain what we do when we account for the identity of substances.

Clearly what we do is more than simply make assertions about what is identical with what. And what emerges is that the criteria we invoke all, whether directly or indirectly, refer back to the forms of things, and, pace the nominalist, to those forms considered as universal entities instantiated in particular cases. The identity of the substance is primitive in this sense—that it cannot be decomposed into elements that do not themselves presuppose either the identity that is the subject of analysis in the first place or the identity of other things on which the identity in question is dependent. So the identity of Rover, for instance, is evidenced by those features we typically point to as features of Rover—Rover's bark, Rover's bite, Rover's characteristic way of chasing postmen. But it would be patently circular to claim that Rover's identity consisted in these things. Or, in the case of a bare natural formation, say, such as a river, the identity is evidenced by typical features of that thing—its characteristic shape or flow. Aggregates such as a pile of bricks have an identity wholly dependent on the identity of their constituents, which need not commit us to mereological essentialism—the idea that even the slightest addition to or replacement of parts destroys a thing—even though it is notoriously difficult to say just how many bricks need to stay the same for the pile to be the same pile. We refer to evidence, and evidence is all we have to go on. Even the much-vaunted phenomenon of spatiotemporal continuity only gives us evidence rather than an analysis.¹⁸

The sorts of features to which we point, however, when we try—impossibly—to analyze identity (as distinct from the actual practice of reidentification, which we do successfully all the time) are notable for having this in common: they are all features referable back to, and deriving from, the form of the object in question. In general, what matters are the congeries of powers, operations, activities, organization, structure, and function of the object, whether it be something as bare as shape in the case of the diachronic identity of a circle drawn on a piece of paper, or something as complex as character in the case of the identity of a relatively higher animal such as a dog. Hence, it is Rover's special way of barking at dinner time that is of more relevance than his color—after all, he could have been swapped for a twin from the litter—and it is his mournful mien when refused a walk in the park that is of more relevance than his enthusiasm for chasing postmen. There seems to be a hierarchy of attributes to which we attach relative importance in grasping a thing's identity; it is better, perhaps, to think of it as a series of concentric circles, moving from the periphery where certain attributes—perhaps (but not necessarily) color, shape, posture, having been at a certain place at a certain time—have a fairly transitory importance, toward the center where, in the case of, say, a higher animal, features such as manner of behavior and characteristic function assume dominance. The closer we get to the center, the nearer we approach what we think of as the essence of the thing.

Why can we not simply refer identity criteria back to spatiotemporal characteristics? Apart from the impossibility of an analysis in terms of spatiotemporal continuity, and apart also from the well-known Max Black–style counterexamples to the Identity of Indiscernibles,¹⁹

the possibility of exact spatiotemporal coincidence of objects precludes any analysis in terms solely of such characteristics. I have argued elsewhere that coincidence is impossible for substances of the same kind because of the problem of individuation, and that for non-substances (at least of certain kinds) it is possible since individuation is effected by appeal to the identity of the coincident objects' ontological sources, since non-substances are ontologically dependent entities. For instance, coinciding objects such as two shadows or two beams of light, one on top of the other, are individuated by their sources (the distinct occluding objects and the distinct light sources, respectively).²⁰ Again, for substances of different kinds, if coincidence is possible it will be referred back to distinct identity criteria for those substances, and this may include modal features, that is, features concerning how things might have been with respect to one or both objects (these being genuine features of objects as much as their nonmodal features such as shape or size). For instance, if a statue is a substance then it is distinct from the lump of marble constituting it because of the different identity criteria for statues and lumps of marble; one could have existed without the other, say, if the lump had been rearranged into a differently shaped object.²¹ In all cases where coincidence is possible, reference to distinct identity criteria entails reference to the distinct forms possessed by the entities in question, substances or not. (In the case of non-substances of the same kind, such as property instances, or such entities as shadows and beams of light, reference is to the identity criteria for the substances on which they are ontologically dependent.)

The moral of the story is that form is the root cause of identity: another way of putting it is that identity has a formal cause. Since, however, substances are individuals and form is not of itself individual, we have to posit a material cause of identity as well: in other words, the identity of a substance is given by the form as instantiated in matter. That the matter is not the root cause of identity is shown by the fact that most macroscopic objects can and often do change all their matter without ceasing to persist.²²

No substance can change its form—that is, its substantial form—and continue to exist. Another way of expressing the proposition that identity has a formal cause is to say that form is the bearer of identity. For a substance to persist is for it to possess this substantial form: not merely a substantial form, but a form instantiated by this matter—where this matter is not identified by there necessarily being a single parcel of atoms or other stuff, since as noted this may itself change over time. The matter is simply the matter of the persisting substance. Only if this were offered as an analysis of identity would there be a problem of circularity; rather, what is offered is an analysis of the causes of identity, and seen as such there is no circularity: a substance persists because it consists of a form instantiated in matter, the form being the actualizing principle in virtue of which the substance is what it is, and the matter being the limiting principle of that form in virtue of which the substance is individual.²³

Since a person is, following the classic definition of Boethius (480–524 a.d.), an individual substance of a rational nature,²⁴

it follows that a person persists in virtue of its form. The form just is the person's rational nature; it is also called the person's soul. Anyone who objects to the term “soul” as metaphysically or theologically loaded can simply use the term “rational nature” wherever “soul” appears in what follows. It is now necessary to understand exactly of what the soul is the form.

IV. Body, Unicity of Form, and Primordial Matter

We can only grasp what the soul is the form of, however, via a defense of two central doctrines of hylemorphism, namely, those of the unicity of substantial form and of the existence of primordial (or prime) matter. Unicity of form means that for any substance, there is one and only one substantial form that it possesses. This is because a substance is one kind of thing, and substantial form determines the kind of thing it is. Hence, when a substance comes into being, it does so in virtue of acquiring a single substantial form, and when it loses that form it ceases to exist altogether as that kind of thing, even if something else is left over that is not that kind of thing. Thus, when a lump of clay is smashed to pieces it ceases to exist altogether even though other, numerically distinct lumps of clay may come into existence in virtue of the persistence of clay material that is not itself a lump of any kind but rather the referent of the mass term “clay.” Suppose, on the contrary, that the lump of clay possessed two substantial forms, that of lump and that of clay. Then we would have to say that if the lump form were removed, say by smashing, the clay form would remain and the lump of clay, not having been completely destroyed, would continue to exist. But how could it exist? One might think it exists as the clay itself. But this is absurd: in what sense has the lump of clay persisted—as clay? But a lump of clay is not mere clay. Or suppose it exists as in some respect “partially identical” to the clay. Yet this is unintelligible, whatever the proponents of “degrees of identity” or peddlers of the idea of “survival” (a kind of persistence short of full identity) may think. Further, it would then seem impossible even to destroy a lump of clay without removing the clay form as well, which would require disintegrating it into its atomic or subatomic parts—but surely destroying a lump of clay cannot be that difficult.

Whatever one might say about the substantiality of such objects as lumps of clay—and some recent writers have cast doubt on it²⁵

—the unicity doctrine is even more apparent in the case of objects over whose substantiality there is no dispute, such as living things. Let us go back to Fido. If substantial forms were multiple in Fido, the multiplicity theorist would have to say either that one substance, Fido, instantiated two substantial forms, or that there were actually two substances where it looked as if there were only one. Take the first alternative. Suppose we say that Fido, being both a living thing and a dog, falls under the two substantial kinds living thing and dog. These being distinct forms, why could they not come apart, with Fido instantiating one but not the other? One scenario is that Fido goes the way of all doggy flesh, leaving behind a canine corpse. It might be said, pointing at the corpse, “There is Fido,” meaning that Fido is still a dog, albeit a dead one. But a dead dog is not a kind of dog any more than the proverbial rubber duck is a kind of duck, or, to change the analogy, than a dead parrot is anything other than an ex-parrot. A substantial form, as defined earlier, supplies the proper functions and operations of its instances. Since no such functions and operations take place in a dead dog²⁶—indeed, the processes undergone by and taking place in a corpse are in general the very reverse of those undergone by and taking place in a functioning dog—clearly a dead dog does not fall under the substantial kind dog.

Another scenario is that Fido acquires the powers of Proteus and morphs into various other kinds of substance, while retaining the form of living creature. Does this indicate that Fido would have ceased to fall under the substantial form dog while continuing to instantiate the separate form of a living creature? No, because in the case of Protean change the transient forms are not substantial but accidental: they do not determine the kind of thing Protean Fido is in his essence or nature, but merely the diversity of forms which that essence or nature allows him to take on. Observing Protean Fido in his canine form, we do not behold a substance that is essentially a dog and a Protean living thing, but an essentially Protean living thing that has taken on the form of a dog. Therefore, neither of the scenarios just described gives us a way of positing distinct substantial forms possessed by a single substance.

Might there, taking the other alternative, be two substances where there only appeared to be one? We can easily dispense with this thought in respect of Protean Fido, because we cannot plausibly say, observing the living creature in its canine form, that here there are two things, namely, a dog and a Protean organism: rather, there is one thing, a Protean organism appearing as a dog. For the organism, the sortal “dog” is as much a phase sortal²⁷

as the sortal “teenager” is for a thirteen-year-old person, in which latter case there do not exist two things, a human being and a teenager. More plausibly, however, it might be argued in the case of normal Fido that there are two substantial forms, namely, those of dog and of body, and that either there are two substances (for example, a certain body constituting a dog) or there is one substance instantiating the forms of both body and dog. The basic confusion at the root of both proposals is that they misunderstand the concept of substantial form. Substantial forms do not make up a hierarchy within a substance—the canine form is not an add-on to the inferior corporeal form, for example. For how would one specify exactly what kind of body the canine form was superadded to? We can eliminate the idea that the canine form is the form of a certain kind of corpse. It is tempting to think that a living dog just is a dead dog plus something extra, and one might imagine dead Fido's being miraculously brought back to life and call that the re-addition of canine form to canine matter. But dead flesh is not a formally impoverished kind of living flesh: in dead flesh, from the moment death occurs, not only is the substantial organic canine form absent but it is replaced by the very form of a dead thing, in which new functions of decay and disintegration immediately begin to occur. The reanimation of dead Fido by means of the re-addition of the organic canine form would involve not merely the super-addition of something to a corpse, but the actual reversal of disintegrative processes already commenced. In other words, Fido's form qua living dog is the form of living flesh; that is, the living flesh has a formal cause in Fido's substantial form. There simply is no metaphysical space for another kind of flesh to which the organic canine form is added to produce a living, breathing dog.

Another way of putting the point is to say that substantial form permeates the entirety of the substance that possesses it, not merely horizontally in its parts—there is as much dogginess in Fido's nose and tail as in Fido as a whole²⁸

—but also vertically, down to the very chemical elements that constitute Fido's living flesh. To use the traditional scholastic terminology, the chemical elements exist virtually in Fido, not as compounds in their own right but as elements fully harnessed to the operations of the organism in which they exist, via the compounds they constitute, and the further compounds the latter constitute, through levels of compounds, proteins, the DNA the latter code for, the organelles that make up the cells, the organs made up of the cells, and so on.²⁹ Supposing there to be elementary particles (a proposal I deny),³⁰ and supposing these to be quarks, it does not follow from the fact that every material substance is made of quarks that every substantial form is the form of a bundle of quarks, because in the existing substance the quarks have no substantial identity of their own, their behavior having been fully yoked to the function and operations of the substance in which they exist. The substantial forms of the particles exist virtually in the substances they constitute. In other words, the quark is ontologically dependent on the whole of which it is a part, but its causal powers persist, albeit in a way radically limited by the whole. The substantial form is what determines the permissible and impermissible behavior of the quarks in the body, which is why some chemical reactions typically occur, others rarely, and others not at all. Nor is there any particular bundle of quarks of which the form could even be the form, given the familiar fact that every body loses and gains quarks all the time. Again, it is the form that determines the when, how, and how much of the loss and gain that may occur, with external circumstances merely operating upon predetermined possibilities.³¹

According to the hylemorphic theory, the unique substantial form of any material substance must be united to something to produce that substance, since in itself it is only an actualizing principle. What does it actualize? It does not actualize anything whose actuality already presupposes the existence of the substantial form. Here it is useful to distinguish between two senses of “of” in the expression “x is the form of y.” In one sense, the substituend for “y” is simply that whose identity depends on the substituend for “x,” as when we say that a father is the father of his son (“He is his father's son”). In the other sense, the substituend for “y” is the object whose identity does not so depend, the object with its own real existence apart from that to which it is functionally related, as when we say that a father is the father of a person. In the first sense, then, we can say with Aristotle, when speaking about life, that the soul, understood as the organic principle, is the first actuality of a natural body with organs.³²

In other words, the soul is the form of an organism, that which makes the organism an organism; we could also say that the soul is the form of a body that has these kinds of properties. In terms of the real unity relation, however, the soul is the form of something else, something not itself shot through by the very soul to which it is united—and this is what the hylemorphist calls primordial matter. There is no space here to enter into a detailed explanation and defense of primordial matter: for our purposes, it is enough to know that although I have called it a something, it is, in the well-worn phrase, not a something but not a nothing either. It is the closest there is in the universe to nothingness without being nothingness, since it has no features of its own but for the potential to receive substantial forms. There has to be something to which form unites, and primordial matter is the only thing that can fill that role.

To return now to our main concern, which is persons, the situation is this. A person, like any other substance, is actualized by a substantial form. For human beings, the kind of person that is our focus, the substantial form is that principle in virtue of which the person is a person, and that means the principle of life, of consciousness, and of rationality. These are all one principle since the doctrine of unicity applies as much to persons as to any other substance. The fact that persons are also sentient and alive does not mean that there are three forms, the form of life, the form of sentience, and the form of rationality, for what could this mean? There are not three distinct substances—the organism, the animal, and the rational creature. There is one substance, a person, who is both living, and sentient, and rational. There is not one substance instantiating three distinct substantial forms—life, sentience, and rationality—because they are all constituents of one set of powers. What gives the person life is precisely what makes the person sentient, and what makes him sentient is just what makes him rational, even though canine sentience, by contrast, does not give rise to canine rationality. The reverse also holds: for instance, what makes the person rational is also what makes him organic, since the sort of rationality persons have essentially involves the use of sensation, and sensation requires life. There may be—indeed are—kinds of rationality that do not require sensation, but they are irrelevant to consideration of the human person.

The person, then, like any substantial kind of thing, is an essential unity manifesting a multiplicity of operations: one nature, many manifestations of that nature. The nature is called by hylemorphists the soul, the term having been traditionally used for all living things, even plants, but now restricted to human beings. In what sense, then, is hylemorphism a kind of dualism?

V. Soul and Knowledge

The hylemorphic theory is dualistic with respect to the analysis of all material substances without exception, since it holds that they are all composites of primordial matter and substantial form. When it comes to persons, however, the theory has a special account. The soul of Fido, for instance, is wholly material—all of Fido's organic and mental operations are material, inasmuch as they have an analysis in wholly material terms. The soul of a person, on the other hand, is wholly immaterial, the argument for this being that a person has at least some mental operations that are not wholly explicable in material terms—and we can deduce what a thing's nature is from the way it necessarily acts or behaves. If, however, some such operations are not wholly materially explicable, the soul itself cannot be anything other than wholly immaterial because there is no sense in postulating a soul that is a mixture of the material and the immaterial.³³

To take the last point first, if the soul were a mixture of the material and immaterial it would be subject to contrary properties: qua material it would have spatiotemporal characteristics, qua immaterial it would not; qua material it would have parts, qua immaterial it would not; qua material it would be divisible, qua immaterial it would not.³⁴

Although very much imperfect, the analogy with abstract objects is useful: the color red, for instance, though wholly dependent on material tokenings for its existence, is in its own nature an immaterial, abstract object, not a mixture of the material and the immaterial. Its very immateriality is what allows it to be wholly instantiated in more than one place at one time, which is not possible for material objects. But if it is true of immaterial objects wholly dependent on material instantiation that they are not a mixture of the material and the immaterial, how much more will it be true of immaterial objects that are not wholly materially dependent? (We will see this lack of dependence later.) Note also that this point does not exclude the following. (i) The person, being a compound of matter and form, is a compound of the material and the immaterial. In this sense one can speak loosely of the person's being a “mixture” of the material and the immaterial. The soul, however, does not have parts and thus is not itself a compound object (this I assume rather than argue for in the present essay): so it would really possess contradictory properties were it to be both material and immaterial. (ii) The soul, although immaterial in itself, can be described as having a certain essential relation to matter, in that its complete operation requires embodiment. Again, however, this does not mean that the soul has contradictory properties.

Now, if the soul is immaterial, it follows that human nature is immaterial, since the soul of a person just is that person's nature. We can see this by understanding the concept of a hierarchy of capacities. Although some may balk at the idea of such a hierarchy, in fact the idea is easily explained by saying that F-type capacities are superior to G-type capacities just in case the former entail the latter but not vice versa. It follows that sentience is superior to nutrition because sentient operations require nutritive ones but not vice versa—we have abundant examples of such. Hence, the nature of an object that has sentience and nutrition as capacities is sentient, and by implication nutritive, but not merely nutritive. In other words, the nature of a thing is defined in terms of its highest capacities. Human rationality is superior to both human sentience and human nutrition according to the definition given, so human nature is defined in terms of the rational capacity. If the rational capacity is immaterial, however, it follows that human nature, that is, the substantial form of the human person, is immaterial. (This does not imply that nutrition, say, is an immaterial process, only that human nature, being essentially immaterial, contains a power of nutrition that can exist apart from any embodiment. But in the absence of the requisite material conditions—embodiment and objects upon which to act—that power cannot be exercised.)

There are various ways of establishing the immateriality of human reason, or the human intellect, and one of these does indeed appeal to consciousness. But as I have claimed, an excessive focus on consciousness is deleterious both to the debate about personal identity (and the mind-body problem) and to our very conception of human nature. Instead, hylemorphists take their primary cue from Aristotle, who asserts that the intellect has no bodily organ.³⁵

In other words, intellectual activity—the forming of ideas or concepts, the making of judgments, and logical reasoning—is an essentially immaterial process, a process that is intrinsically independent of matter, however much it may be extrinsically dependent on matter for its normal operations in the human being.³⁶ Aristotle's position, it must be emphasized, is not that hylemorphism of itself entails the immateriality of the intellect, but that within the hylemorphic conception, considering the specific function of the human person, the intellect must be immaterial. The central theses of hylemorphism in general then tell us in what manner and to what extent the human person is immaterial, as will be explained in due course.

The reason for the proposition that the intellect is immaterial is that there is an essential ontological mismatch between the proper objects of intellectual activity just mentioned and any kind of potential physical embodiment of them: we might call this the embodiment problem, but looked at in a slightly narrower way, in cognitive-scientific terms, it might be called the location or storage problem. Concepts, propositions, and arguments are abstract; potential material loci for these items are concrete.³⁷

The former are unextended; the latter are extended.³⁸ The former are universals; the latter are particular. Nothing that is abstract, unextended, and universal—and it is perhaps hard to see how anything abstract could be other than unextended and universal—could be embodied, located, or stored in anything concrete, extended, and particular. Therefore, the proper objects of intellectual activity can have no material embodiment or locus.

To complicate the problem even more for the materialist, consider those concepts that are not only universal, unextended, and abstract, but also semantically simple. Suppose, per impossibile, that the materialist could overcome the problem of the first three features of concepts, adding that those that are semantically complex, such as the concept of a black dog, had their locus in the brain spatially distributed in a way that mirrored their complexity: thus, the concept black had location A, the concept dog had location B, and some kind of structural relation between A and B constituted the relation between these concepts as elements of the unified concept of a black dog. (Whether it is even right to analyze complex concepts in this way is another matter that cannot be discussed here.) Now what about simple concepts such as the concept of unity, or of being, or of identity? Such concepts do not admit of analysis into semantic parts, though it is possible to explicate the notions contextually, illustrate them, and so on. They are, nevertheless, semantically simple. So there is not even a prospect of finding a material locus for such concepts, assuming all the other difficulties could be overcome, unless the putative locus were materially simple, in the sense of being material and yet metaphysically indivisible. But the very idea of a material simple makes no sense. If a material object were simple it would be unextended—but then in what sense would it be material? An extensionless point is not a something but a nothing, and thus cannot be a locus for concepts, which are something. Further, extensionless points cannot have any constitutive relation to the extended, which is why Aristotle was adamant that the infinite divisibility of space is only potential, not actual. Suppose, however, we could make sense of the idea of a material simple—could it be the candidate locus for simple concepts? Well, are we to postulate a simple located in the brain? If so, is it the same simple that embodies all simple concepts? It would have to be if we were to postulate a single mind having those concepts. But it is hard to make sense of the idea of multiple simple concepts in one materially simple location—about as hard as making sense of many dimensionless points located at one dimensionless point. Yet if we proposed multiple material loci, we would have to account for the mental unity by which one mind has many concepts. All of this without yet having accounted for the possibility of complex concepts, like that of a black dog, in a material simple—how could that be? Yet if there were a non-simple location for these, how again could we account for mental unity given that the simple concepts had simple locations? All in all, the existence of simple concepts merely aggravates the already immense difficulty of smoothing over the fundamental mismatch between concepts and their putative material embodiment.

Needless to say, one of the fundamental problems of cognitive science, in its ubiquitously materialistic contemporary guise, has been to explain the storage of concepts. And needless to say, again, most of the research is either beside the point insofar as it attempts to solve the embodiment problem, or else yields precious little knowledge. For example, one recent paper notes: “A common feature of all concrete objects is their physical form [note the use of the term ‘form’, which in the context of the paper means something more than shape]. Evidence is accumulating that suggests that all object categories elicit distinct patterns of neural activity in regions that mediate perception of object form (the ventral occipitotemporal cortex).”³⁹

The authors go on to describe how functional brain-imaging techniques show that representations of different object categories are located in discrete cortical regions that are “distributed and overlapping,” embedded in a “lumpy feature-space.” To be sure, functional imaging may well reveal correlations between certain intellectual activities and certain cortical activities: for the hylemorphic dualist, such correlations are only to be expected, since persons as embodied beings require corporeal activity in order to interact with the world. Persons are not pure spirits capable of immediate intellectual apprehension or action upon the environment (assuming such things to exist for the purpose of contrast). Nevertheless, the substantial form is what directs and controls corporeal activity, whether by acting upon physical inputs or producing physical outputs.

The authors of the paper go on, prudently, to say: “Clearly, it would be difficult, as well as unwise, to argue that there is a ‘chair area’ in the brain. There are simply too many categories, and too little neural space to accommodate discrete, category-specific modules for every category. In fact, there is no limit on the number of object categories.”⁴⁰

Indeed, this latter observation points again to the ontological mismatch between concepts and their putative material embodiment. The intellect is capable of grasping a potential infinity of concepts, but no corporeal organ can harbor a potential infinity of anything.⁴¹ In particular, the intellect is distinguished by this feature: that it can grasp a potentially infinite number of categories of concepts, and within each category a potentially infinite number of exemplars. In other words, there is no limit to the number of kinds of things the intellect can recognize, and no limit to the number of examples of each kind that it can grasp. By contrast, the eye or ear, for instance, can only receive colors and sounds, respectively; and within each kind of sense datum, they can only receive a limited number of examples—hence, we cannot naturally see certain colors or hear certain sounds. The very physical finiteness of the organs of sight and hearing means they are bounded with respect to what kinds of information they can take in. This is patently not so for the intellect—and it does not exclude the fact that the intellect, being finite in its own way, cannot discover certain things. There is a difference between the intellect's not being able to reach certain truths by its own operation, and its suffering an intrinsic material limitation on the kind of information it can take in. The absence of such a material limitation, again, is consistent with its being extrinsically limited in respect of the physical information it can take in: for example, not having the concept of a color that is beyond the visual spectrum available to the eye. But if the sort of limitation I have been talking about applies to the eye and the ear, it must apply to any proposed organ for embodying concepts. The features of the eye and ear that make them singularly unsuitable for intellectual operation apply equally to the brain, the nervous system, or any other proposed material locus. It is the very materiality of such a locus that prevents it from embodying the proper objects of intellectual activity.

If researchers into functional imaging have shown anything, then, it is merely that category-specific object recognition is correlated with activity in certain distinct, if highly diffuse and non-discrete, regions in the brain. But this sort of research, as interesting and as potentially useful for brain-damaged patients as it might be, goes no way to even beginning to provide a theoretical or empirical foundation for the idea that concepts, judgments, and inferences themselves have a physical location.

There are, of course, many kinds of challenges that might be leveled against the defense of the immateriality of the intellect I have given. One might level a Rylean-style charge of illegitimate reification against the very idea of concepts as things.⁴²

One might object that an appeal to immateriality to solve the embodiment problem is a classic case of obscurum per obscurius. One might deny that there are concepts in any meaningful sense at all, and claim that there are only distinct, particular acts of representation. There is no space here to canvass these and other objections. But as a general reply we should emphasize that a refusal to reify concepts means an inability to explain fundamental semantic and logical phenomena: not merely the fact that the concept of a black dog is a function of the concept of black and the concept of dog, but that the concept-possessor understands this, which is more than saying he can recognize a black dog only if he can recognize black things and dogs. Rather, it means that if he has those concepts, he can see how one is derived from the others. Mutatis mutandis for judgments and for inferences. And if a person grasps a certain concept, and if that concept is an object (pace Gottlob Frege's worries about the concept horse),⁴³ then the person grasps an object. Since this is a mental act, his mind must take hold of something, and if it takes hold of a thing then that thing must make a kind of contact with it—which means, since there is no other plausible way of understanding it, that the concept must somehow be in its possessor's mind. But if the concept is not the sort of thing that can be physically inside the possessor's brain, his mind cannot be his brain, and moreover must be immaterial since only an immaterial thing can be suited to laying hold of the concept.

VI. Soul, Identity, and Material Dependence

What, then, of the complex relationship between the soul, the person, and the matter the soul informs to produce the person? The first thing to note is that the soul is not the person.⁴⁴

The person is the human being, the substantial compound of matter and form. A person is an individual substance of a rational nature, but the soul is not such a substance—for it is the rational nature, not a substance with a rational nature. Hence, the fundamental flaw in the Cartesian conception of the person is the illegitimate identification of the person with the soul, taking them to be one and the same substance. It might with good reason be said that Descartes, having given up on the notion of substantial form,⁴⁵ yet eager to preserve personal immortality, had nowhere else to go. Yet the mistake is basic, and leads to so many of the problems that have dogged Cartesian dualism ever since.

Next, given the unicity of substantial form, one cannot take there to be separate, lower orders of soul or nature in the human person. Growth, nutrition, reproduction, sentience, perception—all of the operations of the organism belong to the unique human nature of the person. A human being is an essential unity, not a plurality. Some of those operations, however, depend essentially on matter—such as reproduction and sensation—and others, such as the operations of the intellect, as we have seen, do not. But if the person is not to be broken down into a plurality, how do we reconcile the partial dependence and partial independence of matter that we find in human nature? We have to say something like the following. The person, being essentially embodied, depends for its existence and identity on embodiment, as also for some of its operations. Whether it exists at all depends upon its having a human nature individualized in matter; and which person it is depends on which material individualization it is. Again, this is not proposed as an analysis of identity in other terms, but rather as an account of the causes of that identity.

To say, however, that the person is existence- and identity-dependent on its embodiment does not entail that all of its parts depend for their existence on being united in the embodied person. As an imperfect analogy, we observe that a broom cannot exist without a brush but the brush can exist without the broom to which it belonged. That is, it is not a universal truth that if an F cannot exist without a certain part P, then P cannot exist without F: it depends on the kind of thing one is talking about. In the case of nonrational animals, we can say that the animal cannot exist without its soul, but neither can the soul exist without embodiment in the animal since all of the animal soul's operations are wholly material, not rising beyond sensation and perception of the concrete particular. On the other hand, since some of the operations of the intellectual soul are not material, it can exist without its embodiment in matter. The principle at work here is the following: x can exist without y if and only if x can operate without y. The first half is that if x can exist without y then x can operate without y: if x exists without y, then x's nature is actualized without y; but if x's nature is actualized, then x possesses the very operations given to it by its nature, and thus can operate according to that nature without y. It might be the case that x operates in an imperfect way because of the lack of y, but its essential nature and the functions proper to that essential nature will not in themselves be destroyed. Fido can exist without his tail, so he can function without his tail even though the lack of a tail impairs that function. He cannot exist without a head, however, and so cannot function without a head.

The second half of the biconditional says that if x can operate without y, then x can exist without y. If x can operate without y, albeit perhaps imperfectly, then x must have a nature that can be actual without y's being actual. But for x to be actual is for x to exist, and for y not to be actual is for y not to exist. So x can exist without y. I can function without ten fingers; so I can exist without ten fingers. I might not be able to hold a baseball bat without ten fingers, but holding a baseball bat—indeed, being able to hold anything—is not essential to my functioning as a human being. By contrast, I cannot function without a heart, or without something that fulfills the role of a heart; hence, I cannot exist without a heart. Whether or not the biconditional is true for any x and y or only for material substances, the hylemorphist only needs it to be true for living things in order to make his point about human souls as opposed to other souls. Since the human soul can operate without matter, it can exist without matter. It might exist in an imperfect state, since it cannot, for instance, perform acts of sensation that require material stimuli and the formation of mental images, but it can still exist apart from matter.⁴⁶

Although the soul of the person is not existence-dependent upon matter, in the way I have claimed (it does not require material embodiment to exist), it is not plausible to deny that its existence depends upon matter in the following sense: that it must be embodied at some time during its existence. This is a weaker form of existence-dependence, and it follows from the fact that the human soul just is the rational nature of an individual substance belonging to a certain kind. Human persons just are embodied creatures, and thus not only must their souls be attached to their bodies—at least at some time in their history—for them to exist, but also their souls, in order to be souls of persons, that is, in order to be what they are, must be at some time the forms of bodies. This means that the idea of a human person disembodied throughout its history is incoherent. Such a being might be a disembodied person, but it would not be a disembodied human person because human persons are just not that kind of thing. In which case, if the human soul has a disembodied existence, that existence can only be made possible by its once having been the form of a body.⁴⁷

Further, it is also identity-dependent on its once having been the form of a body.⁴⁸ In other words, to be the particular soul that it is, it must once have been the form of a particular body making a particular individual substance of a rational nature; just as, in its embodied state, the soul's identity depends on whose, that is, which person's, soul it is.⁴⁹ In short, the principle of individuation for persons must be cross-temporal.⁵⁰

As I claimed earlier, the form is the bearer of the identity of a substance, in the sense that it is the primary part of the substance responsible for the substance's being the substance it is over time. The soul, as form of the body, is therefore also the bearer of personal identity. From the subjective point of view, when I reflect upon my own identity as a person it is my soul that exercises that intellectual operation, recognizing itself as the bearer of my identity as a person. This does not mean that the first-person pronoun is ambiguous, only that it refers to me as a person by means of referring to that person's chief part, which is the soul, just as, when I say “I am in pain” after I stub my toe, “I” refers to me as a person by means of one of my parts, in this case my toe: I am in pain because my toe is in pain. I take the primary reference of the first-person pronoun, as used by me, to be myself as a person; but I propose tentatively that the reference to my soul (in the case of thought) or my toe (when I stub it) is a kind of secondary or instrumental reference.

In the disembodied state, I continue to exist—that is, the person that is me persists despite my physical death, which is the separation of my form from my matter—even though one of my constituents, namely my body, does not. What this means, then, is that my death results in the person that I am continuing to exist as my chief part, namely the part in virtue of which I am specifically different, or different in kind, from any other kind of animal. When the body my soul informs ceases to exist, as surely it does at some time, then the person I am dies but does not thereby cease to exist; hence, death and cessation of existence, for entities like us, are not the same event.⁵¹

I persist both as a person and as the form that once was the form of the body that was a part of that person. My soul is the bearer of my identity as a person, but I am not, and was never, strictly numerically identical with my soul.⁵² Another imperfect analogy helps to make the point. Suppose it were technically possible to reduce my organic existence to that of a head.⁵³ Then I would exist as a head, but I would not be numerically identical with a head any more than I would have been numerically identical with my whole body—there being no reason to affirm one and deny the other, transitivity of identity would be violated. And yet in some sense I am a head: perhaps, to use a much-discussed concept, we can say that I am constituted by a head, as I was once constituted by a whole body (let us leave aside the soul for the moment—the point should be graspable by materialists as well). Although the concept of constitution is not well understood, I think that the best way of interpreting it in this context is to say that my existing as a head just means my being reduced to one of my parts, my existing in a radically mutilated state.

Finally, the consequences for personal responsibility must be something like the following. If persons die when their souls leave their bodies—which is no more than a special case of the general truth that substances cease to exist when their form and matter are no longer united—then can any sense be made of a soul's bearing any responsibility for the acts of the whole person of which it once was a constituent? To pursue the gory analogy of the bodiless head, there does not appear to be anything repugnant to reason in the idea that a person existing solely as a head should be punished for crimes committed while the head was connected to a body. Yet perhaps intuitions differ strongly on this question. I think we can accommodate any divergence by considering generally whether sense can be made of the idea that a part of an F can be held responsible for the acts of a whole F. To see that such an idea is not only coherent but has real-world application, consider the case of a corporation (a legal and moral person) whose chief executive is held responsible for that corporation's illegal actions. Considering the corporation as a kind of aggregate or collectivity, and its directors as constituent parts of that collectivity, we can see that the chief executive as a part of the corporation can be held responsible for the latter's transactions, as can the directors in general.⁵⁴

It is true but irrelevant that the courts have traditionally been reluctant to impute such responsibility in a blanket fashion, their reluctance being motivated not by metaphysics but by a recognition of the disincentive such blanket responsibility would be to anyone thinking of becoming a director, let alone CEO, of a corporation. All we need to see is that it is coherent to suggest that a part might be held responsible for the actions of the whole—moreover, that not any part will do, but only that part (or those parts) which are, as it were, in the driver's seat. The soul, if it is part of the person at all, certainly is in the driver's seat, so if any part of the person can be held responsible, it must be the soul. But since, as I have argued, the person I am continues to exist as a soul (even though I am not numerically identical with a soul), it must be me who is responsible precisely for what I did when my soul informed a certain body.⁵⁵ But doesn't this imply a twofold responsibility, and hence a twofold punishment or reward (if there be such after death)? No, because the soul is held responsible solely in virtue of its being the chief part whereby I, the person, did whatever I did that incurred responsibility. As Aquinas puts it, I am rewarded or punished “in the soul” for what I did when my soul informed my body. To return to the example of my being reduced to a bodiless head, if it is true to say that the head suffers a punishment, it does not do so qua head, but qua the part that now constitutes me: if there is to be any punishment at all of bodiless me, the only way it can be carried out is by punishing my head.