Maverick Philosopher

The abandonment of final causes has also crucially contributed to the creation of the “mind-body problem,” something that did not exist, certainly not in anything like the form familiar to contemporary philosophers, prior to the moderns’ rejection of the Aristotelian-Scholastic metaphysical framework. For to insist that the material world is utterly devoid of final causes – devoid, that is to say, of anything that inherently “points to” or is “directed toward” anything beyond itself – is implicitly to deny that intentionality could possibly be material, for intentionality, of course, is just the mind’s capacity to point to or be directed towards something beyond itself, as it does in thought. [. . .] Hence to insist that the material world is devoid of any inherent final causes while at the same time acknowledging the existence of intentionality is implicitly to commit oneself to [Cartesian] dualism. [. . .]

The only way to hold on to the mechanistic conception of nature while rejecting [Cartesian] dualism is thus to deny the existence of intentionality. And that is why, as John Searle has argued, all extant forms of materialism do indeed implicitly deny its existence, and thus (I would say) amount to disguised forms of eliminative materialism. This is halfway admitted by Jerry Fodor when he writes, as he does in Psychosemantics, that “if aboutness [i.e. intentionality] is real, it must be really something else.” That is to say, intentionality per se simply cannot be real given the mechanistic conception of the material world that Fodor, like all materialists, has inherited from the early modern philosophers. Hence the most the materialist can do is try to substitute for it some physicalistically “respectable” ersatz. But this is simply eliminative materialism in “folk psychological” drag; and eliminative materialism, however you dress it up, is simply incoherent.

I agree entirely with Ed that the project of substituting something 'physicalistically respectable' for intentionality is doomed and that contemporary materialism is hopeless. One cannot deny that there are intentional states. (It is worth noting that the very denial exhibits intentionality!) Eliminative materialism is indeed incoherent, and indentifications collapse into eliminations. If a token mental state is identical to a token brain state, then there are no mental states, only brain states. Intentionality is irreducibly real.

What is less clear to me is whether intentionality strictly speaking is to be found in the material world, as Ed suggests in the first paragraph, or whether it is only something analogous to intentionality, a sort of proto-intentionality, that is found in the material world. My recent defense of the Potentiality Argument against the moral acceptability of abortion commits me to holding that there are irreducible potentialities in nature below the level of conscious mind. And perhaps this should soften me up for hylomorphic dualism, which is Ed's preferred solution to the mind-body problem. But I find some difficulties with hylomorphic dualism. See Three Dualisms: Simple, Compound, and Hylomorphic and related posts.

Joel Feinberg formulates the Strict Potentiality Criterion (SPC) of Moral Personhood as follows:

All and only those those creatures who either actually or potentially possess c (that is, who either have c now or would come to have c in the natural course of events) are moral persons now, fully protected by the rule against homicide. (Matters of Life and Death, ed. Regan, p. 193)

By 'c' Feinberg means the set of individually necessary and jointly sufficient conditions for what he calls "commonsense personhood." See my #2 above.

One of the objections that Feinberg brings against the (SPC) is "the charge that merely potential possession of any set of qualifications for a moral status does not logically ensure actual possession of that status." (p. 193) He calls this "the logical point about potentiality" and he finds it unanswerable. (p. 194) Feinberg provides an example he borrows from Stanley Benn: A potential president of the United States is not on that account Commander-in-Chief of the armed forces.

It seems to me that this objection is easily answered. Let me begin by conceding something that is perfectly self-evident, namely, that inferences of the following form are invalid:

5. X is a potential possessor of qualifications for a certain moral or legal status
ergo
6. X is an actual possessor of of qualifications for a certain moral or legal status.

This is a glaring non sequitur. But what does this obvious point have to do with (SPC)? (SPC) is a statement, not an argument or argument-form, whereas (5)-(6) is an (invalid) argument-form. Does Feinberg think that the only possible support for (SPC) is an argument of the above invalid form? I hope not.

Here is a simple point of Logic 101: If a proposition p is true, and there is an invalid argument that has p as its conclusion, it does not follow that there is no valid argument for p. Furthermore, if a proposition p is true, it is true whether or not you or anyone can give a valid argument for p. Not every proposition can be supported by argument. (Circular 'support' is not support.) Not every proposition needs argumentative support. It is true that I am thinking now. I intuit this truth. My epistemic access to it is direct, not mediated by any discursive process: I don't know it by argument.

I accept (SPC) and I accept that (5)-(6) is an invalid argument-form. These two propositions that I accept are plainly logically consistent. So what are Feinberg and Co. trying to tell us? If their point is that the only support for (SPC) is an invalid argument, then they are wrong. For one could accept (SPC) on the basis of a valid argument, or on the basis of no argument at all. Either way, Feinberg's "logical point about potentiality" evaporates as an objection to (SPC).

I assert that the potential possession of c by x directly confers upon x moral status. This is an assertion, not an argument. Since it is not an argument, it cannot be an invalid argument. Maybe there is a good argument for this assertion, maybe there is not. But as I said, not every proposition needs or can receive argumentative support.

Consider a neonate, a fresh-born infant human being. It does not actually possess moral status-conferring properties. And yet we consider it to have moral status: we consider the neonate to fall under the prohibition against homicide. What then is the ground of its moral status? Not its mere membership in the species homo sapiens, for reasons I won't rehearse. And it does not have moral status because others (the neonate's relatives and well-wishers, say) ascribe it moral status. It has its moral status intrinsically.

I say the neonate has moral status in virtue of its potentiality to possess the properties on Feinberg's c-list or some relevantly similar list. The neonate's right to life is grounded directly in its potentiality to develop into what what we indisputably consider to be a rights-possessor. What holds for the neonate holds for the fetus in the broad sense to include zygote, embryo and fetus in the narrow sense.

It looks like I gave an argument for (SPC) after all, an argument that needs further development and examination. But the main point I am making is that the oft-bruited "logical point about potentiality" does absolutely nothing to discredit (SPC).

What say you, Peter Lupu? Have I now convinced you that the Potentiality Argument is a good argument?

When I first read this, I thought it had been written by an undergraduate. Note the grammar mistakes in the first and last bolded passages. Note also the absurdity of the criticism lodged in the third bolded passage. It turns out, however, that the author is not an undergraduate but the chairman of the Stanford philosophy department!

I will be told that it is only a blog post. Indeed, but blogging is no excuse for egregious errors and misrepresentations. To make good on these charges, I will now comment on three of the bolded passages.

1. In the first and last of the bolded passages we are told that substance dualism is dead and that there are no rational grounds for accepting it. This is plainly false. A philosophical theory cannot be dead if distinguished contemporary academic philosophers defend it in works published by prestigious presses. The following titles come to mind right off the bat: Richard Swinburne, The Evolution of the Soul, Oxford UP, 1997; John Foster, The Immaterial Self: A Defence of the Cartesian Dualist Conception of the Mind, Routledge 1991; Geoffrey Madell, Mind and Materialism, Edinburgh 1988; W. D. Hart, The Engines of the Soul, Cambridge UP, 1988. Many other books and articles could be cited.

As for the rational grounds in support of substance dualism, they are provided in the rigorously argued works just cited. This is not to say that substance dualism is true; it is to say that a strong rational case can be made for it. It is simply incompetent for the author to claim that substance dualism is dead and that there are no good reasons in support of it.

2. The stock objection, according to which mind-body interaction is inexplicable on substance dualism, is scarcely as strong as its proponents think it is. But this is too large a topic to be discussed here. See the works cited.

3. The author blunders in breathtaking fashion in the following specimen of strawman reasoning:

Descartes believed that the mind was an indivisible simple, that it could not be broken down into an organized collection of interacting parts. But the mind obviously has a vast diversity of its possible states.

It is true that for a substance dualist, the self or mind or person is a simple individual substance. It is simple in that it is not composed of substances. It has no parts that are themselves substances. It is not a bundle or collection, but an irreducible unity of its states. But a substance's being simple is plainly consistent with its being in a wide variety of actual and possible states. For Descartes, an individual res cogitans or thinking thing has the attribute cogitatio (thought or mentality in a broad sense), modes of which are particular perceivings, rememberings, imaginings, etc.

To suppose that a substance's being simple or noncomposite entails its having no states is such an egregious blunder that one wonders how any trained philosopher could make it. But then all too many contemporary analytic philosophers tend to be historically uninformed. They read the passage in Ryle where he speaks of the "ghost in the machine" and that is the extent of their acquaintance with substance dualism. Their materialist prejudice does the rest.

In the grip of this prejudice, they cannot think of Cartesian thinking substances as other than ghosts, when it should be obvious that (i) thinking substances cannot be spatially extended, unlike ghosts which are, and (ii) ghosts are composed of some subtle stuff (spookstuff?) whereas purely spiritual substances are composed of no stuff at all. See Lycan, Dennett, and Spookstuff.

4. Can one be a substance dualist in the 21st century? Of course one can. If there are some scientific facts that disprove dualism, I should like to know what they are. The author alludes to such facts at the outset of his post, but neglects to mention them. Materialists who think that science entails materialism are trading in scientism, not science.

It is difficult to see how items such as sensations can be called surds in nature when they are precisely not in nature as physicalistically conceived, but I shall let that pass the better to get on with spookstuff.

There are philosophers who seem to think that doctrines held by great philosophers and outstanding contemporaries don't need to be studied and refuted but can be shamed or ridiculed or caricatured out of existence. Daniel Dennett is an example:

Dualism (the view that minds are composed of some nonphysical and utterly mysterious stuff) . . . [has]been relegated to the trash heap of history, along with alchemy and astrology. Unless you are also prepared to declare that the world is flat and the sun is a fiery chariot pulled by winged horses — unless, in other words, your defiance of modern science is quite complete — you won't find any place to stand and fight for these obsolete ideas. (Kinds of Mind, Basic Books, 1996, p. 24)

This is an amazing passage in that it compares the views of distinguished dualist philosophers such as Richard Swinburne to the views of astrologers, alchemists, and flat-earthers. It would be very interesting to hear precisely how the views of Swinburne et al. are in "defiance of modern science" — assuming one doesn't confuse science with scientism. But let's look at what Dennett has to say in his more substantial (511 page!) Consciousness Explained (1991).

Dennett there (mis)characterizes dualism as the doctrine that minds are "composed not of ordinary matter but of some other, special kind of stuff. . . ," and materialism as the view that "there is only one sort of stuff, namely matter — the physical stuff of physics, chemistry, and physiology — and the mind is some nothing but a physical phenomenon." (33) "In short, the mind is the brain." (33)

But as I have said before,

A substance dualist such as Descartes does not hold that minds are composed of some extraordinarily thin intangible stuff. The dualism is not a dualism of stuff-kinds, real stuff and spooky stuff. 'Substance' in 'substance dualism' does not refer to a special sort of ethereal stuff but to substances in the sense of individuals capable of independent existence whose whole essence consists in acts of thought, perception, imagination, feeling, and the like. Dennett is exploiting the equivocity of 'substance.'

The point is made very well by the prominent idealist, T. L. S. Sprigge:

It is often difficult to get people to realize that the non-physical mind of which Cartesians speak is not, as some have thought it, 'a ghost in the machine' of the human body, since ghosts and 'spirits' such as might appear in a seance are, in contrast to it, as physical, if made of a finer stuff, as our ordinary bodies. When we speak of the mental we do so mostly or entirely in metaphors (more or less sleeping) of a physical kind: we grasp ideas and have thoughts in our minds. Whatever the real source of this materialism which is endemic to most of our thinking, it is not surprising that there should be a theory of existence which follows its leadings. As thinkers we are subjects, but the natural object of thought is objects and it is only with effort that the subject turns its thoughts upon its own un-object-like nature. (Theories of Existence, pp. 46-47, bolding added.)

The materialist cannot understand how anything could be real that is not physical in nature. So if mind is real, which it obviously is, then it too must be physical in nature. But to treat mind physically is precisely to miss its "un-object-like nature." Mind is subjectivity, and no matter how the materialist twists and turns, he cannot grasp this subjectivity objectively. The materialist project is fundamentally absurd in that it tries to treat as objective what is irreducibly subjective, all the while presupposing subjectivity without which there would be no access to objects at all.

Searle writes 'Intentional' rather than 'intentional' to underscore the fact that 'intention' as philosophers use it is a terminus technicus wider in meaning than 'intention' as when one says, for example, 'Her intention is to retire in Florida.' My intending to run a half-marathon in February is an instance of Intentionality, but so is my noting that the sun is setting -- despite the fact that Old Sol's habits don't fall within the purview of my will. I Intend the sun's setting, but I don't intend it. Get it? Now that the point has been made, I will drop the capital 'I.'

Let's think about Searle's example of an itch, one 'in' the scalp, say. Not every mental item is a conscious item, but this itch I feel right now is a conscious mental item. Attending to this datum, a distinction suggests itself: there is the qualitative character of the itch, its sensory quale or raw feel, and there is its unpleasantness. Since I am attending to the itch, it is the intentional object of a series of acts of phenomenological reflection. At the same time of course, I am 'living' the itch, not just reflecting on it and analyzing it. The itch is an experience, an Erlebnis in something like Husserl's sense; it is something man erlebt, one lives through.

Now the question is whether the itch itself is directed to an object or 'takes an accusative.' Please don't confuse this question with the question whether the itch can itself be made the intentional object of acts of reflection. Of course it can, and I just did. But that is not to say that the itch has an intentional object. Attending carefully to the itch as it presents itself to me, I discern no object to which it points. Surely Searle is right: itches and the like are not about anything in the way a desire to drink a cold beer is about drinking a cold beer, or the seeing of a bobcat is about a bobcat and not a tire iron or nothing at all.

Can one scrounge up some sense of 'about' according to which itches ans tickles and pains are about something? I suppose so. But any sense of 'about' so scrounged would have nothing to do with intentional aboutness which alone is at isssue here. Thus if you said that the itch is about its physical cause [insert complicated dermatological story here], then I would accuse you of equiviocating and/or changing the subject.

The point is a simple phenomenological one. Unfortunately, people in the grip of naturalism and scientism often blind themselves to elementary phenomenological distinctions. Their reductionist zeal disposes them simply to eliminate what they cannot cram into their reductionist scheme. Suppose I inspect my scalp in a mirror. The inspecting is an intentional state of mind, an intentional Erlebnis. It presents something distinct from itself. The itch too is an Erlebnis but it presents nothing to me beyond itself.

That is a simple phenomenological difference. Either you 'see' it or you don't.

We have a subject S in a mental state M. M is an intentional state if and only if all of the following hold:

1. M takes an accusative or is directed upon an object O.

2. M's having an object, and having the very object it has, is not a contingent fact about M, but essential to M. O 'enters into the description' of M. Thus if I am imagining at time t, then this imagining is of a definite object, a unicorn, say, and I cannot describe the act of imagining without using the word 'unicorn.' It is constitutive of the act that it have the object it has: an imagining of a unicorn could not have been an imagining of a flying spaghetti monster; nor could the imagining of a unicorn of such-and-such description have been objectless.

3. From the fact that M has O as it object it does not follow that O exists. I cannot own X unless X exists. But I can want X whether or not X exists. 'Wanting' picks out a type of intentional state; 'owning' does not.

4. From the fact that M has O1 as its object, it does not follow that M has O2 as its object even if O1 is the same as O2. Suppose I am in Iceland and I see an unkempt, bearded old man in a bookstore. Suppose, unbeknownst to me, that this man is the chess grandmaster who defeated Boris Spassky in 1972. The intentional object of visual perceiving has all and only the properties revealed in the visual perceiving: the properties of being unkempt, bearded, old, etc., but not the property of having defeated Boris Spassky.

As John Searle, puts it, "All intentionality is aspectual." (Rediscovery of the Mind, 131) Seeing is seeing-as. A thing is seen under certain aspects and not others. And what holds for seeing holds for every other intentional state.

We are now in a position to answer the question whether the felt pain described above is an intentional state. It is only if all of the four conditions specified above are met. I will argue that none of them are met.

First, does the felt pain have an object or accusative? One might say that the object of the felt pain is the mean molecular kinetic energy of the water. But this can't be right. For one thing, the felt pain does not present or reveal molecules or the collision of molecules in the way seeing an old man presents or reveals an old man. This is a crucial phenomenological difference and by itself is enough to refute the notion that felt pains are intentional. Or to be precise: it refutes the notion that felt pains have as their intentional objects their physical causes.

Now consider condition (2). The seeing of an old man cannot be what it is unless it is precisely the seeing of an old man. 'Old man' enters of necessity into the description of the act (intentional experience). This seeing could not have been the seeing of a sick dog or of a flat tire or of nothing at all. An intentional state must have an object and it must have the very object it has. So if felt pain is an intentional state and its object is water molecules or else the mean molecular kinetic energy of water molecules, then the felt pain oculd not exist without exactly that object. But surely there is no conceptual absurdity in the supposition that the felt pain in question exists in a world in which temperature is not mean molecular kinetic enegy. (Consider Lavoisier's world in which caloric has pride of place.) 'Mean molecular kinetic energy' (MMKE) does not enter of necessity into the description of a felt pain even when said pain is caused by MMKE of a certain degree.

Condition (3) is also not satisfied by felt pain. If the intentional object of felt pain is the extramental cause of felt pain, then one would be able to infer the existence of the object from the existence of the pain. But one cannot infer the existence of the intentional object of an intentional state from the existence of the state. Ergo, etc.

The major premise is true because, if X causes Y, then both X and Y must exist. (Causation is a relation, and every relation is such that, if one relatum exists, then so does the other.) But if X is the intentional object of Y, then, while Y must exist, X need not exist. For this reason, Brentano denied that intentionality is a relation, maintaining instead that it is relation-like, ein Relativliches.

Finally, condition (4) is not satisfied. Suppose I see Big Joe walking towards me. Unbeknownst to me, Joe has just been elected sheriff. Even though Big Joe = the sheriff, from the fact that Big Joe is my intentional object it does not follow that the sheriff is. I see the man as Big Joe and not as the sheriff even though Joe is the sheriff. (The as-structure or aspectuality or perspectivity of intentionality is essential to it.) But if the intentional objects of felt pains were their physical causes, then, from the water's physical temperature being the inetntional object of the felt pain it would follow that a certain degree of MMKE would be the intentional object of the felt pain, since temperature = MMKE. But this does not follow. Therefore, the intentional objects of felt pains are not their physical causes.

But there is no other reasonable candidate for the office of being the intentional object of a felt pain. So I conclude that felt pains are non-intentional mental states.

The crucial argument here is: "The torn tendon is the cause of the pain sensation, but it is not the intentional object of the pain sensation. The torn tendon is hidden from the pain experience in the way the foot is not hidden from the visual experience. So what is the intentional object of the pain sensation?"

But it does not follow from the fact that details of something are hidden from an experience that the experience is not of it. Consider: Suppose I feel a pressure in my heart--I am taking that to be an uncontroversially intentional experience. I may be quite unaware that the pressure is in my heart--I might not even know I have a heart (people ignorant of anatomy may not know about hearts at all). I am aware that there is pressure somewhere, and vaguely locate the pressure somewhere in my chest. Likewise, the pain sensation presents me with the fact of damage, and a vague location. It does not seem to present me with details of the damage (perhaps with experience I would be able to recognize them just from the feel--or not).

Moreover, an object may have more than one description. When I feel that a knife is warm, the experience is about the knife's molecules having moderately high kinetic energy. But I can feel that a knife is warm without knowing about molecules at all, and then I simply don't know (non-circularly) what the experience is about.

Respondeo:

Our disagreement is about whether physical pains — felt physical pains — have intentional objects or not. You appear to be claiming that all such pains have intentional objects whereas I am claiming that some felt physical pains do not have intentional objects. Do you agree that this is the question under discussion?

You may be using 'intentional object' in an idiosyncratic way, or (to put it nontendentiously) differently than I do. The intentional object of an act is the object as it appears in the act, with all and only the properties it appears to have. Example. I look out my window and see a yellow wall. The seeing is an intentional experience: it is an act of consciousness directed upon an object. This intentional object has the property of being yellow. But it doesn't have the property of being colored, even though in reality everything yellow is colored. For what is present to my consciousness is simply the wall as yellow, not as colored. It also has the property of being sat upon by a fly, and other properties besides, properties that are not present to my consciousness.

Necessarily, awareness of something (intentional awareness) is always awareness of something as something. I am not aware of bare objects, but of objects 'clothed' in properties. Furthermore, given the finitude of our minds, even if every F is a G, I can be aware of x as F without being aware of x as G. Indeed, this is so even if necessarily (whether metaphysically or nomologically) every F is a G. Thus I can be aware of a moving object as a cat, without being aware of it as spatially extended, as an animal, as a mammal, as an animal that cools itself by panting as opposed to sweating, as my cat, as the same cat I saw an hour ago, etc.

This has been called the aspectuality of intentionality. More in Brentano Notes III: Intentionality Without Mind?

Now suppose I grab hold of a hot poker. (I am trying to get the poker away from Wittgenstein who seems about to clobber Professor Popper.) I feel a painful sensation. Does the felt pain have an intentional object? Is it about anything? I take you to be saying that in cases like this the felt pain is about "molecules having moderately high kinetic energy." Now this makes no sense from my point of view. It is bad phenomenology. For the felt pain does not present to my consciousness molecules or any property of an aggregate of molecules.

There is no question of knowing or not knowing what felt pains are about since they are not about anything, though they do have causes.

"Suppose I feel a pressure in my heart--I am taking that to be an uncontroversially intentional experience." It sounds like you are simply begging the question against me. What you say is "uncontroversially intentional" I deny is such. There is felt pressure on the left side of my chest. I wonder with a bit of anxiety what this is a sign of, what it indicates, what its cause might be. Heart burn? Something worse? My point is that the felt pressure as such does not present anything to my consciousness and surely not its cause.

Our disagreement appears to rest on a broader disagreement about intentionality and intentional objects.

That anyone should be a mortalist does not surprise me, but it does surprise me that anyone should consider it an "obvious fact" that death is the "irrevocable end" of a person. But this is what Heinegg holds: "Everybody knows that the soul dies with the body, but nobody likes to admit it." (11) Priests and metaphysicians may prate about immortality, but deep down in the bowels of the body we all know that we are mortal to the core:

As surely as the body knows pain or delight, the onset of orgasm or vomiting, it knows that it (we) will die and disappear. We have a foretaste of this every time we fall asleep or suffer any diminution of consciousness from drugs, fatigue, sickness, accidents, aging, and so forth. The extrapolation from the fading of awareness to its total extinction is (ha) dead certain. (13)

This is as close as Heinegg comes to an argument in his personal statement, "Why I am a Mortalist." The argument has but one premise:

1. We experience the increase and diminution of our embodied consciousness in a variety of ways.

Therefore

2. Consciousness cannot exist disembodied.

But surely (2) does not follow from (1). If (2) followed from (1), then it would be impossible for (1) to be true and (2) false. But it is easy to conceive of (1) being true and (2) false. It might be like this: as long as the soul is attached to the body, its experiences are deeply affected by bodily states, but after death the soul continues to exist and have some experiences albeit experiences of a different sort than it has while embodied.

Now consider near death experiences. A man has a massive heart attack and has a profoundly blissful experience of a white light at the end of a tunnel. Would any mortalist take such an experience as proving that there is life after bodily death? Of course not. The mortalist would point out that the man was not fully dead, and would use this fact to argue that the experience was not veridical. The mortalist would point out that no conclusions about what happens after death can be drawn from experiences one has while still alive. By the same token, however, a consistent mortalist should realize that this same principle applies to his experiences of the waxing and waning of his consciousness: he cannot validily infer from these experiences that consciousness cannot exist disembodied.

What puzzles me about Heinegg is not that he is a mortalist, but that he is so cocksure about it.

1. Atheism and Dualism. Atheism is the denial of the existence of God. Since the only God worth discussing is the God of a sophisticated monotheism, the only atheism worth discussing amounts to the claim that nothing in reality corresponds to the conceptions of deity championed by the leading philosophers and theologians of such religions as Judaism, Christianity, and Islam. 'Dualism' is elliptical for 'substance dualism in the philosophy of mind.' Dualists make a claim about human persons to the effect that (i) the mental states of a human being are states of an irreducibly mental substance; (ii) the physical states of a human being are states of an irreducibly physical substance; and (iii) mental and physical substances are ontologically separate. A substance is an entity that (i) is logically capable of independent existence, and (ii) maintains its numerical identity over time.

Given these definitions, it seems that that an atheist could also be a mind-body dualist. The dualist is saying something about human persons: they have a mental 'part' and a physical 'part' and these two parts are mutually irreducible. There is nothing in this about the existence or nonexistence of God. It is safe to say that most dualists are theists, and that this has always been the case; but there is nothing in dualism to entail theism or to entail the negation of theism. Hence dualism is consistent with both theism and atheism. Or at least this is the lay of land at first blush.

2. Theism and Materialism. This is a very large topic. One aspect of it is explored in my World Congress of Philosophy paper, Classical Theism and Global Supervenience Physicalism. Another side of the problematic is treated in "Could a Classical Theist be a Physicalist?" Faith and Philosophy, vol. 15, no. 2 (April 1998), pp. 160-180.

Suppose we distinguish between universal materialism and materialism about human persons. The universal materialist claims that everything, or at least everything concrete, is either a material entity or supervenient upon material entities. If this is what is meant by materialism, then no materialist could be a classical theist. The God of classical theism is a purely spiritual, hence immaterial, being. But materialism about persons seems prima facie combinable with theism. Such a materialist claims that a human being is just a very complicated animal organism in principle wholly explainable in material terms. Now could it be true that all of my sensing, perceiving, thinking, etc. is just complex processes transpiring in my brain and central nervous system but that there is a being, God, who thinks without having a brain? On the face of it, this seems like a possibility.

My own view, however, is that, although this makes prima facie sense, deeper analysis reveals that it makes no sense for reasons I present in the Faith and Philosophy article above cited.

. . . Descartes took the interaction between mind and body to be so close that the two together constituted a third, unique substance, with its own distinctive properties: while shape, mass, and the like are confined to the body, and pure intellectual activity confined to the mind, sensation — pains, itches, feelings of thirst or hunger — is a feature strictly attributable only to the substance comprised of mind and body interacting together. Moreoever, it is this composite substance, rather than the mind alone, with which a person or human being is to be identified. (Philosophy of Mind: A Short Introduction, pp. 17-18)

Substances are entities capable of independent existence. The capability in question is broadly logical. If a thing is a substance, then it is possibly such that there is no other thing in which it exists or on which it depends for its existence. A substance exists in se, not in alio. So if my mind is a substance, it is metaphysically capable of existing on its own in a state of separation from my body. But Feser tells us that the mind-body composite is also a substance and that this composite is identical to the person. I find this puzzling. I do not understand how the composite can be a "third unique substance."

If a substance, by definition, is an entity capable of independent existence, then it is difficult to see how the mind-body composite could also be a substance. For this composite is composed of two substances without which it cannot exist. It is obvious that I cannot exist as a soul-body composite unless both my soul and my body exist. So it is obvious that a soul-body composite cannot be a substance, strictly speaking. The fundamental point here is that a whole depends for its existence on the existence of its parts. Anything of which this is true cannot be a substance.

If soul and body are substances, then the composite composed of them cannot be a substance. If, on the other hand, the composite is a substance, then soul and body cannot be substances in their own right; they can at most be 'principles' invoked in the analysis of substances as on Thomistic hylomorphic dualism. What is impossible is that the composite be a "third unique substance" in addition to the other two. For if the composite is a third substance, then it is something over and above soul and body and so precisely not the composite of the two. One must choose. Either the composite is a substance, in which case soul and body are non-substances as on the hylomorphic approach, or soul and body are substances, in which case the composite is a non-substance. It is not an option to say that there are three substances.

And since the composite cannot be a "third unique substance," it is hard to understand how it could have "its own distinctive properties." Moreover, the properties Feser cites, pains, itches, and feelings, are clearly examples of cogitationes in Descartes' broad sense. It is well-known that cogitatio (thinking, thought) in Descartes is not restricted to the higher intellectual operations but covers all manner of mental activity. Thus for Descartes it is not a composite entity that senses, but the mind that senses, imagines, feels and so on.

Since the composite is not a substance on its own, it has no properties of its own. The properties of the composite are either properties of the mind or properties of the body. Thus the property of feeling depressed is a property of the mind, while the property of weighing 180 lbs is a property of the body. But don't I weigh X lbs? Not exactly. I weigh X lbs (derivatively) in virtue of my body's weighing X lbs (nonderivatively). The ordinary language 'I weigh X lbs' is a loose way of talking. This is no different in principle than the situation in which I say with the vulgar, 'I am moving at 65 mph' while meaning with the Cartesianly learned that the car I am driving is moving at 65 mph.

Now who or what am I? If I am a soul-body composite, then my body is a proper part of me. But if so, I cannot exist without my body, since a whole depends for its existence on the existence of its parts. This, however, flies in the face of the Cartesian insight that the I that thinks can exist disembodied.

And if I am a soul-body composite, then my soul, too, is a proper part of me. This has the unacceptable consequence that my thinking is done for me by something numerically distinct from me. For as we saw earlier, if my soul is a proper part of me, then it is numerically distinct from me. But surely it is absurd to say that my thinking is done for me by something numerically distinct from me.

If the reply is that it is the composite that thinks, then the response is that this would make disembodied thought impossible. But this contradicts a central tenet of substance dualism.

My conclusion is that a substance dualist ought to be a simple substance dualist.

1. It is possible that there exist disembodied persons

to

2. Some actual person can exist disembodied.

(1) is de dicto: it says of a dictum or proposition that it is possibly true. (2) is de re: it says of a res or thing that disembodiment is one of its possibilities. To see the difference, consider that (1) can be true even if every actual person is a purely material being, and thus a being for whom disembodied existence is not a possibility. On that supposition, (2) is false. So the truth of (2) does not follow from the truth of (1).

Roger Bannister broke the four-minute mile on 6 May 1954. Before his birth it was surely possible that someone run a mile in under four minutes. But it does not follow that before Bannister's birth there was a runner for whom breaking the four minute barrier was a possibility. The possibility that there exist an F that is G does not logically guarantee that there exists an F that is possibly G. For it might be that every existing F is necessarily not G.

Another example. I cannot imagine a chiliagon (a regular polygon having 1,000 sides) but I can conceive one. Suppose conceivability entails possibility. Then it is possible that there be a chiliagon. But it doesn't follow that there is anything that actually exists anywhere in the universe that could have the shape of a chiliagon.

So how can I get to the conclusion that some actual person, such as dear old me, can survive his physical dissolution and exist disembodied? Well, can't I imagine my existing disembodied as opposed to some person in general's existing disembodied? Isn't there such a thing as de re imagination? But how do I know that when I seem to be imagining my existing disembodied, I am really doing so as opposed to imagining some person's existing disembodied? How do I know that I am not imagining merely the instantiation of disembodied personhood as opposed to some definite actual person's disembodied existence?

Here we enter waters as murky as the Charles. I can imagine myself taking a walk through Boston's North End in search of an old haunt of mine, Piccola Venezia. The imaginability of this state of affairs is pretty good evidence of its possibility. But is this a de re possibility involving me or a de dicto possibility concerning some possible person?

Since no one will claim that I am incapable of revisiting the North End, there is no reason to doubt that when I imagine myself walking its streets, the possibility for which this imagining provides evidence is a de re possibility involving me, not some possible person in general. I am such that it is possible that I revisit the North End. So what reason could there be for denying that I am such that I possibly exist disembodied? Apparently, the only reason is a question-begging one: All actually existing persons are necessarily embodied.

My interim conclusion is the de dicto objection is nothing to worry about.

I exist, and I have a body. But it is conceivable that I exist without a body. 'Conceivable' in this context means thinkable without logical contradiction. My disembodied existence is conceivable in the way that my body's disembodied existence is inconceivable. But the question is this: does the conceivability of my disembodied existence entail the possibility of my disembodied existence? If yes, then a very interesting argument appears sound:

1. If a = b, then necessarily a = b. (Necessity of Identity)
2. If possibly ~(a = b), then ~(a = b). (From 1 by Contraposition)
3. If possibly I am not identical to my body, then I am not identical to my body. (From 2 by Instantiation)
4. Possibly I am not identical to my body. (Premise)
Therefore
5. I am not identical to my body. (From 3, 4 by Modus Ponens)

This is a fascinating argument since it seems to establish a very significant metaphysical result on the basis of cogent reasoning from very plausible premises. The move from (2) to (3) presupposes that the first-person pronoun is a referring term, something denied by Elizabeth Anscombe under the (dubious) influence of Wittgenstein. I would say that Miss Anscombe is mistaken about this, but I don't want to discuss it now.

But what reason do we have to accept (4)? We can argue for (4) as follows: Conceivability entails possibility. It is conceivable that I am not identical to my body (because I can conceive my disembodied existence). Therefore, possibly I am not identical to my body.

But why should we accept that conceivability entails possibility? The following argument sees to show that conceivability does not entail possibility:

6. It is conceivable that I am not identical to my body.
7. It is conceivable that I am identical to my body.
8. If conceivability entails possibility, then it is possible that I am not identical to my body and it is possible that I am identical to my body.
9. It is possible that I am not identical to my body iff I am not identical to my body.
10. It is possible that I am identical to my body iff I am identical to my body.
11. If conceivability entails possibility, then I am not identical to my body and I am identical to my body.
12. The consequent of (11) is a contradiction.
Therefore
13. Conceivability does not entail possibility.

Let me sum up the argument as follows. If we take seriously the Necessity of Identity, then either it is a necessary truth that I am identical to my body, or a necessary truth that I am not identical to my body. If the first is necessarily true, then the second is necessarily false, and vice versa. Thus neither are contingent. Now both my identity and my non-identity with my body are conceivable. Therefore, if conceivability entails possibility, then it is possible that I am not identical to my body and possible that I am. But this is a contradiction. So if we accept the Necessity of Identity, then we must deny that conceivability entails possibility.

Do I have good reason to think that it is logically possible that I exist without a body? Or is it necessarily the case that if you destroy all my physical parts, you destroy me? Well, the following is true and known to be true:

5. It is conceivable that I exist without a body.

By 'conceivable,' I don't just mean thinkable — round squares are thinkable -- I mean thinkable without logical contradiction. And surely it is thinkable without logical contradiction that I exist without a body. Read your Descartes. I had a student once, a hopeless materialist, albeit otherwise alert and intelligent, who just could not appreciate the point. He kept repeating, "But if I shoot you dead, then you cease to exist!" What I found bizarre was that his religious upbringing hadn't even softened him up for the conceivability of post-mortem existence. It was as if he was so sunk into his bodily existence that the mere thought of not being identical to his body was unavailable to him. So I branded him a Cave-dweller and gave up on him.

But now comes the hard part. How do we move validly from conceivability to possibility, from (5) to (2)? (2) affirms the (broadly) logical possibility of disembodied existence. But it is not clear why being able to conceive a state of affairs should guarantee its logical possibility. Note that it would serve no purpose to stipulate that logical possibility just is conceivability. That would have all the advantages of theft over honest toil. Broadly logical possibility is a species of real possibility, and one cannot just assume that what one can conceive without contradiction is possible in reality. (On the other hand, one can be certain that a concept harboring a contradiction cannot have anything answering to it in reality.)

Consider the floating bar of iron (FBI). If my thought about the FBI is sufficiently abstract and indeterminate, then it will seem that there is no 'bar' to the FBI's logical possibility. If I think of the FBI as an object that has the phenomenal properties of iron (e.g., hardness) but also floats, then those properties are combinable in my thought without contradiction. There seems to be no logical contradiction in the thought of a hard metal that floats.

But if I know more about iron, including its specific gravity, and I import this information into my concept of iron, then the concept of the FBI will harbor a contradiction. The specific gravity of iron is 7850 kg/cu.m, which implies that it is 7.85 times more dense than water, which in turn implies that it will sink in water. For someone with this richer understanding, the FBI is a bar of iron that both floats and does not float — which is a contradiction.

What this example seems to show is that my failing to find a contradiction in my concept of X does not entail that X is logically possible; for it may be that my concept of X is insufficiently determinate, and that if I had a sufficiently determinate concept of X, then I would see from the concept alone that X is logically impossible. Now let's apply this to our problem. My disembodied existence is conceivable. But it might well be that my identity with my body is hidden from my powers of conception in a way similar to the way the logical impossibility of floating iron is hidden from someone whose concept of iron is inadequate. It may be that my belief in the possibility of disembodied existence feeds on ignorance. How can I rule out this possibility?

If the only way to rule it out is by assuming the truth of (4), then the modal argument begs the question. I am not now saying that it does beg the question. I'll leave that for subsequent posts.

Also, the action of no body is self-reflexive. For it is proved in [Aristotle's] Physics [VIII, 5, 256a2-33] that no body is moved by itself except with respect to a part, so that one part of it is the mover and the other the moved. But in acting the intellect reflects on itself, not only as to a part, but as to the whole of itself. Therefore, it is not a body.

One will be tempted to dismiss this quaint verbiage as just so much medieval mumbo-jumbo, but I think there is an argument here that has a serious claim on our attention. Taking some minor liberties, I would present the argument as a nice, neat syllogism:

1. No body is wholly and non-formally self-reflexive.
2. Every intellect in act is wholly and non-formally self-reflexive.
Therefore
3. No intellect in act is a body.

It is easy to see that the conclusion follows from the premises. But are the premises true? More importantly, what do they mean?

Ad (1). A body in this context is any material thing of whatever size, artificial or natural, animate or inanimate. What the first premise says is that no body can act upon itself as a whole. If body B acts upon itself, then some proper part of B acts upon some other proper or improper part of B. Thus when I shave myself, one of my hands drags a razor across small portions of my epidermis: the whole of BV does not shave the whole of BV. In the case of a self-inflating mattress, a proper part of the mattress inflates the whole of the mattress. But it would be a fabulous mattress indeed the whole of which inflated the whole of it.

Self-cleaning ovens and cats clean themselves, but that is not to say that the whole oven or cat cleans the whole oven or cat; the oven doesn't clean its outside, and the cat does not clean its inside. A boiler that is self-monitoring is monitored by a proper part of itself, not by the whole contraption. And so on for eating, biting, touching, feeding, fueling. What about a machine whose sole function, once turned on, is to turn itself off? The whole of it gets turned off, but the whole of it does not do the turning off.

But everything is self-identical, and so one may be tempted to say that bodies stand in the relation of identity to themselves and that therefore there are self-reflexive bodies. To block this silly move, I added 'non-formally' to (1).

So premise (1) looks to be true. Or can you think of a counterexample? If (1) is true, then presumably (1) is necessarily true: there is something about the very ontological structure of bodies that disallows their being wholly and non-formally self-reflexive. If the body is simple (non-partite), then there is no possibility of a distinction between agent and patient; and if the body is partite, then the part that is the agent cannot be the whole.

Perhaps the closest we can come to a body that is wholly and non-formally reflexive is a light source that illuminates itself along with other things in its environment. A lamp illuminates a room and itself, but not every part of itself, the inside of the lampstand, for example. Now imagine a bare light bulb, emitting light, hanging from the ceiling. It does not illuminate the inside of the wire from which it hangs and which supplies it with current. Imagine a light magically on without a source of current. Still, the inside of the filament would go unilluminated, and the glass would do no illuminating.

Ad (2). The second premise says in effect that every mental act, such as an occurrent episode of thinking that 7 is prime, stands as a whole in some non-formal relation to the whole of itself. What might that relation be? As Aquinas says, the intellect reflects upon itself. But this is perhaps too strong. I can think about something without explicitly reflecting upon my thinking about it. I can think about X without thinking: I am thinking about X. Many philosophers and psychologists have held, however, that acts (occurrent episodes) of thinking are implicitly self-aware. Thinking about X, I am not only conscious of X, I am simultaneously albeit secondarily conscious of being consious of X. This is what Brentano called inner perception and distinguished from inner observation, and what Sartre called the pre-reflective cogito. Intentional consciousness, consciousness of an object, is always simultaneously albeit secondarily consciousness of itself as conscious of the object.

In the concrete act of consciousness, consciousness as a whole is conscious of itself as a whole. Suppose I am conscious of X. The idea is that I am not only conscious of X, a mountain say; I am also simultaneously albeit secondarily conscious of myself as conscious of X. Thus there is a distinction between C1 of X and C2 of (C1 of X). But although one can and must draw a distinction between C1 and C2, the two are inseparable: they form a unity. To distinguish is not to divide. Thus the whole of C2 is directed to the whole of C1.

This is one aspect of the unity of consciousness. Consciousness exhibits both diachronic and synchronic unity, and what is under discussion here is one aspect of synchonic unity: awareness of X is simultaneously awareness of (awareness of X).

Ad (3). I have just made a case for (1) and (2), a case that could be strengthened in several ways in a longer exposition. But I think I have said enough for it to be clear that something like Aquinas' argument above has a serious claim on our attention. The premises are plausible and the conclusion follows from them. And note that one needn't be a hylomorphic dualist to makes use of the argument.

Three Dualisms Distinguished

A. Simple or Pure Substance Dualism. This is the view that I am identical to my soul or mind. In Cartesian terms, I am identical to a res cogitans, a thinking thing or thinking substance. If so, my body is not a part of me. I am not a compound of soul and body; I am a simple substance.

B. Compound Substance Dualism. This is the view that I am a composite of soul and body, which implies that my body is a proper part of me. Thus I have two proper parts, soul and body, the first essential to me, the second accidental. I cannot exist without my soul, but I can exist without my body, where 'can' expresses broadly logical possibility.

(A) and (B) are mutually exclusive. Assuming the truth of Cartesian dualism, they are also jointly exhaustive. So if you are a Cartesian dualist, you must choose between saying that you are your soul (mind) or that you are not your soul but a composite of soul and body.

C. Hylomorphic Dualism. This is the view that I am a compound of soul and body, but with the difference that the soul is the form of the body. The idea is not that the soul is like a form or analogous to a form, but that the soul is a form. Hylomorphic dualism is not a species of Compound Substance Dualism for the simple reason that form and matter are 'principles' invoked in the analysis of primary substances but not primary substances themselves. But it is dualistic in that mind and body are mutually irreducible.

A Problem with Compound Substance Dualism

Suppose we ask a simple question. Using 'think' in the broad Cartesian way, to cover all manner of intentional states including perceiving, imagining, remembering, etc., who or what is it that thinks when I think? Who or what is the subject of thinking? There are only two possibilities given the above. Either a soul thinks when I think, or a soul-body composite thinks when I think.

The Compound Substance Dualist must say that a soul-body composite thinks when I think. Now soul = mind = thinking substance (res cogitans). So CSD implies that when BV thinks there are really two thinkers, the thinking substance which is a proper part of BV, and the whole BV as soul-body composite. (The point is made in a different way by Eric T. Olson in "A Compound of Two Substances" in Soul, Body, and Survival, Corcoran ed., Cornell UP 2001, p. 75.) Here is the way Olson puts it:

. . . compound dualism entails that there are at least twice as many thinking things as we thought there were. You are a compound of a body and a soul. But that soul is itself rational and conscious. So there are two thinking things sitting in your chair, a soul and a compound, reading an essay that was co-written by simple and a compound philosopher.

Obviously, this won't do. Well, why not just say that the soul does not think, that only the compound thinks? One might say that soul and body are each sub-psychological, and that to have a psyche and psychic activity (thinking), soul and body must work together. Soul and body in synergy give rise to thinking which qualifies the whole man. But this makes hash of substance dualism. For one of the reasons for being a substance dualist in the first place is the conceivability of disembodied thinking. (We'll have to look at Kripke's argument one of these days.) Disembodied thinking is obviously inconceivable if it is a soul-body composite that thinks. Second, if it takes a soul and a body working together to produce thinking, then the soul is not a mind or thinking substance -- which again makes hash of substance dualism.

For these and other reasons, CSD is to be rejected, and simple or pure dualism is to be preferred if one is to be a substance dualist.

Hylomorphic Dualism Faces a Similar Problem

Now let's confront the hylomorphist dualist with the question about the subject of thinking. Who or what thinks when thinking occurs in BV? It is presumably the composite of soul and body, form and matter. Thus the soul by itself is subpsychological and so does not think: the subject of thinking, that which thinks, is the soul-body composite. But when we turn to the 'king' of the hylomorphic dualists, Thomas Aquinas, we find him saying things about the intellect that run directly counter to this.

In the Summa Contra Gentiles, Book II, chapters 49-51, we find a variety of arguments to the conclusion that the intellect is a subsistent form and so not dependent for its existence on matter. This is not the place to examine these arguments, some of which are defensible. Now since the intellect is that in us which thinks, the same ambiguity we found in Cartesian dualism, as between pure dualism and compound dualism, is to be found in Aquinas. Is it the composite that thinks, or a part of the composite? The answer must be that it is a part of the composite that thinks, the res cogitans on the Cartesian view, the intellectus on the Thomist view. And note that both must be viewed as substances, as capable of independent existence. Aquinas expresses this by referring to the soul as a subsistent form.

But 'subsistent form' smacks of contradiction. How can a form be subsistent? To say that a form is subsistent is to say that is is a primary substance. But a form is precisely not a primary substance but a 'principle' invoked in the analysis of primary substances. Aquinas cannot do justice to his own insight into the independence of the intellect from matter from within the hylomorphic scheme of ontological analysis he inherits from Aristotle. His metaphysica generalis is at war with his special-metaphysical insight into the independence of intellect from matter.

So I cannot see that hylomorphic dualism is any improvement over pure substance dualism. It is rather a step backward. And as I tried to show in the my last post, the notion that HD can solve the interaction problem -- assuming as I do not that it is a genuine problem -- is chimerical.

Cut a zombie open, and you find exactly what you would find were you to cut a human being open. And in terms of linguistic and nonlinguistic behavior, there is no way to tell a human being from a zombie. (So don't think of something sleepy, or drugged, or comatose.) When a zombie sees a tree, what is going on internal to the zombie's brain is a 'visual' computational process, but the zombie lacks what a French philosopher might call interiority. There is no irreducible subjectivity, no qualitative feel to the 'visual' processing; there is nothing it is like for a zombie to see a female zombie or to desire her. (What's it like to be a zombie? There is nothing it is like to be a zombie.)

Surely, zombies are conceivable. To conceive one, start with yourself. You, dear reader, are not a zombie. You have feelings, and you know you do. Right now, perhaps, you feel bored or puzzled by what you are reading. Slap yourself across the face to wake up. You felt something, a stinging sensation. Do it again just to be sure. Now subtract off (in thought) the conscious experiences leaving behind your body, its behavior, its environment and all the causal processes disporting themselves between body and environment. What you are now conceiving is a zombie.

The reader says that he can't conceive of a zombie who is not hungry but who nonetheless want to eat a chocolate bar. But the zombie in question is hungry; it is just that his hunger and its linguistic and behavioral manifestations have a purely objective existence. Salivation, for example, is an objective, 3rd person verifiable process. Objectively viewed, hunger is just the sum total of all such processes.

I said I wouldn't worry about cat zombies, but I just changed my mind. My old and spoiled female cat makes it very clear to me when she is hungry. I don't reckon she's a zombie, but she might be for all I know. Surely her being a zombie is conceivable. Descartes thought of non-human animals as machines, and thus as insentient. If I am not mistaken, this is equivalent to conceiving of non-human animals as zombies. Although I don't believe that non-human animals are zombies, I find this quite conceivable. And vivisectionists who justified their practice on Cartesian grounds found it not only conceivable but believable as well -- so believable that they were prepared to act upon their belief.

But I should not make it sound as if the conceivability of zombies is uncontroversial. Competent philosophers deny that they are conceivable. I attempt to refute one such philosopher in Jim Ryan on Consciousness and Zombies.

Aquinas discusses this question in Summa Theologica, Q. 75, art. 3. He attempts to navigate a middle course between a form of materialism according to which sense and intellect are both "referred to a corporeal principle," and the view he takes to be Platonic according to which both sense and intellect are "referred to an incorporeal principle." On the Platonic view, both sensing and understanding "belong to the soul as such." Thomas points us to Theaetetus 184 c. This is a fascinating passage which, according to Paul Natorp, anticipates Kant's transcendental unity of apperception.

In this passage, Socrates puts the following question to Theaetetus: ". . . which is more correct — to say that we see or hear with the eyes and with the ears, or through the eyes and through the ears?" Theatetus obligingly responds with through rather than with. Socrates approves of this response:

Yes, my boy, for no one can suppose that in each of us, as in a sort of Trojan horse, there are perched a number of unconnected senses which do not all meet in some one nature, the mind, or whatever we please to call it, of which they are the instruments, and with which through them we perceive the objects of sense. (Emphasis added, tr. Benjamin Jowett)

The issue here is the unity of consciousness in the synthesis of a manifold of sensory data. Long before Kant, and long before Leibniz, the 'divine' Plato was well aware of the problem of the unity of consciousness. (It is not for nothing that Whitehead described Western philosophy as a series of footnotes to Plato.)

Sitting before a fire, I see the flames, feel the heat, smell the smoke, and hear the crackling of the logs. The sensory data are unified in one consciousness of a selfsame object. This unification does not take place in the eyes or in the ears or in the nostrils or in any other sense organ, and to say that it takes place in the brain is not a good answer. For the brain is a partite physical thing extended in space. If the unity of consciousness is identified with a portion of the brain, then the unity is destroyed. For no matter how small the portion of the brain, it has proper parts external to each other. Every portion of the brain, no matter how small, is a complex entity. But consciousness in the synthesis of a manifold is a simple unity. Hence the unity of consciousness cannot be understood along materialist lines.

This argument from the unity of consciousness, which of course needs to be more rigorously developed, is present in nuce in Plato in the passage cited. According to Aquinas, if this argument is sound, ". . . it follows that even the souls of brute animals are subsistent." This seems to be a correct inference.

Aquinas hopes to block the inference with the help of Aristotle's De Anima 429 a 24. Thomas gives an argument that I interpret as follows:

1. Although understanding alone is performed without a corporeal organ, as Aristotle maintains, sensation and the operations of the sensitive soul are accompanied by changes in the body at our sensory receptors.

Therefore

2. The sensitive soul has no per se operation of its own, and every operation of the sensitive soul belongs to the soul-body composite.

Therefore

3. The souls of brute animals, having no per se operations, are not subsistent.

Unfortunately, I cannot see that this is a good argument. The gist of the argument is that while there are specific corporeal organs for seeing, hearing, smelling, tasting, and touching, namely, eyes, ears, etc., there is no corporeal organ for understanding. From this it is concluded that there is a crucial difference between sensing and understanding: it is broadly logically possible that there be understanding without a body; but it is not broadly logically possible that there be sensing without a body. And since we alone among animals are rational, or capable of understanding, we alone have subsistent souls.

What this ignores is that sensing is not merely a mechanical or material process. When my cat simultaneously sees, smells, and tastes her food (or my food for that matter) there is a unity of consciousness in her no less than there is one in me when I lay into my food. There is no specific corporeal organ that does this unifying of sensory data, neither in me nor in my cat. The unity-of-consciousness argument against materialism can be 'run' both for man and cat. If it works for me it should work for her. So if the possibility of my disembodied existence follows from there being no physical organ that unifies my consciousness, then we get the same result for my cat, and the difference between man and brute in respect of subsistence cannot be maintained.

To sum up. Thomas wants to say that men, but no brutes, have subsistent souls. This is because men, but no brutes, understand. But sensing is a form of consciousness, and consciousness cannot be understood in materialist terms. Sensing is not a mere collision of atoms in the void. Sensory consciousness, besides displaying unity across its several modalities, reveals qualia. And qualia are a well-known stumbling block to materialism. It is difficult to see why, if understanding supports the possibility of disembodied existence, sensing should not also support this possibility. There is after all only one soul which both senses and understands. The phrases 'sensitive soul' and 'intellective soul' are not to be taken to refer to distinct souls.

I concede that some of the operations of the rational soul show that it is capable of existence apart from matter. And perhaps it is also true that most of the rational or intellective soul's operations require the soul's embodiment for their operation. But these two points are consistent with a modified Platonic/Cartesian substance dualism according to which the person is identified with the soul, and not with a soul-matter composite. On such a scheme, the soul is indeed a substance, and thus capable of independent existence; it is just that its full operation requires embodiment.

Now the difficulty, for me at least, is that of combining this notion of the soul as incomplete substance with the view that the rational soul is the substantial form of the human body. On the face of it, the form of a material substance is no more a substance than the matter of a material substance is a substance. The constituents of substances are not themselves substances; they are non-independent ontological factors revealed by the analysis of substances.

I will be told that the substantial form of the human body, however, is an exception to the rule: it is a subsistent substantial form. I see two problems with this. The first concerns the compatibility of subsistent forms of material substances with hylomorphic analysis. I take it that hylomorphic analysis is broadly Aristotelian in the sense that it does not reify or hypostatize (as Platonism arguably does) the principles in things that account for them ontologically. If this is right, then it sounds incoherent to say that there are substantial forms of material substances that can exist apart from matter.

But suppose I am wrong about this and that there is no inconsistency whatever between hylomorphism and the subsistence of some substantial forms of material things. Suppose there are subsistent forms and that the human soul is a subsistent substantial form of those materials being called human beings. That does not show that this subsistent form is the subject of cognition. But Feser's point (a) above is supposed to show that the soul is the subject of cognition. There is a subject of cognition, no doubt, and I concede that it can exist apart from matter: it is not the brain or any part of the body. But why should anyone think that the subject of cognition is a subsistent form? The mere fact that the substantial form of the human being is called 'soul' proves nothing. For what that term means is life-principle. Is it self-evident that the life-principle of a human being is the subject of cognition 'in' a human being? Not at all.

What's more, it is obvious to me that nothing belonging to the ontological category of forms (subsistent or otherwise) could be a subject of cognition. At the very most it could enter into an analysis of the subject of cognition as one of its factors. Thus if the subject of cognition is identified with the whole human being, then the soul would enter into the analysis of the human being. But if the subject of cognition can operate apart from matter, then of course it cannot be identified with the whole human being as soul-body composite.

Why can't a form be the subject of cognition? Because a form cannot be conscious. A form is an object of consciousness; it cannot be a subject of consciousness. A form cannot be a subject of consciosuness any more than a property can. I am conscious of such-and-such; but whatever the ego is, it can't be a form, even it is a subsistent from. Does it maje sense to think of a Platonic From as conscious of something? Note even the Platonic Form of consciousness is conscious of anything.

Let me sum it up as follows. The very considerations that move Feser and me to deny that consciousness is a brain-function simply cannot be accommodated within Thomas' objectivistic framework. Admittedly much more needs to be said to make this clear. But it is Christmas eve, and time to punch the clock.

People 'vegetate,' sense, and think. So does a human being have three souls or one soul? It has one soul, a rational soul. For the soul is the substantial form of the body, and a substantial form is that in virtue of which a material composite entity is a member of the species to which it belongs. Each individual substance has but one substantial form. Substantial forms, and thus individual substances, emerge. Though the human individual develops from plant and animal precursors, it is a new individual, not the same individual in a new form.

At a certain level of complexity of its underlying matter, a human individual substance emerges. This is at the same time the emergence of its rational (intellective) soul and its substantial form. Talk of emergence suggests that the human soul originates 'from below,' from matter. But Thomas expressly states that the soul is a direct divine creation and so comes 'from above.' The reason is given at Summa Theologica Ia, q. 118, art. 2. Very roughly, the operations of the intellect cannot be understood in material terms, and so the 'seat' of these operations, the intellective soul, must come from God. (I should examine the relevant arguments in detail in separate posts.)

This brings us to the problem of this post: How is Thomistic emergentism compatible with the human soul's being a direct divine creation?

Leftow's solution is intriguing, though he does not elaborate on it in the article cited above. (Hence I will be extrapolating beyond what Leftow actually says.) He reads Aquinas an an occasionalist in this one instance. Of course, Aquinas is not an occasionalist across the board, but a concurrentist. (My article, "Concurrentism or Occasionalism?" American Catholic Philosophical Quarterly, Summer 1996, pp. 339-359 may be worth a look.)

Roughly, occasionalism is the view that God alone is a 'true' or productive cause, and that 'secondary causes,' i.e., natural causes, are mere occasional causes. Holding a flame to a dry piece of paper (in standard conditions), I observe that the paper ignites. But as Hume famously noted, one does not observe any causing. Invoking this insight, which one can find earlier in Malebranche and indeed in al-Ghazzali, an occasionalist will say that what one observes is consistent with the supposition that God creates the second event on the occasion of the occurrence of the first. For the occasionalist, all the causal (productive) work is done by God, and at the level of secondary 'causes' what one has, at the most, is contiguous succession and instantiation of law-like regularities.

Now let's apply this to our problem. In the transition from zygote to human being, a non-rational animal soul emerges, and then a rational soul, and with the emergence of these substantial forms, there concomitantly emerge a non-rational animal individual and a rational animal individual. No divine agency is involved in the emergence of the non-rational soul (except in the general sense in which God sustains in being everything distinct from himself). But special divine agency is involved in the emergence of the human soul. Does the divine 'implantation' of a rational soul contradict its emergence?

Well, if all we mean by 'emergence' is that a new form suddenly (as opposed to gradually) appears in a law-like manner on the occasion of a certain uptick in material complexity, then emergence of substantial forms and individual substances is consistent BOTH with divine 'implantation' AND the absence thereof. Accordingly, there is nothing in the concept of emergence to require origination 'from below' or to disallow origination 'from above.' Emergence is neutral as between these options.

So it looks as if Thomist emergentism is compatible with the human soul's divine origin. But there is something puzzling about (what I am taking to be) Leftow's solution. Presumably, in any given case, emergence must either be from above or from below. Either the form emerges due to divine implantation or to some kind of bottom-up vertical causation. One can form a neutral concept of emergence, but in reality, in any given case, the neutrality must be broken.

If this is right, how can it be from below in most cases and from above in others? If the emergence of Socrates' merely animal fetal precursor is due to bottom-up vertical causation (the inverse of formal causation?), how can the emergence of Socrates the rational animal be due to top-down vertical divine implantation? Given that we are talking about substantial forms occurrent in nature, why do we need to bring God into the picture in the case of the second emergence if we don't have to being God in in the case of the first?

"Because the intellective operations of the human soul cannot issue from, or reside in, or be explained in terms of, any material organ such as the brain." But then in what sense is the human soul a substantial form occurrent in nature? How can it be so radically different from other substantial forms? Indeed, in what sense is it a form at all, when it is rather a subject of intellectual operations? And if it is a subject of intellectual operations, then it is not the soul-body composite that thinks and cognizes, but the soul by itself, and Thomistic dualism seems to collapse into Platonic/Cartesian dualism.

This is clearly not a version of substance dualism since soul and body on the hylomorphic scheme are not (primary) substances in their own right. We might define a (primary) substance as anything logically (as opposed to causally) capable of independent existence. Fido is then a substance but his soul is not inasmuch as his soul cannot exist on its own. And the same goes for Fido's body: it cannot exist on its own. Fido's corpse can exist on its own, but it is not his body. A dead dog is not a dog: 'dead' in 'dead dog' is an alienans adjective. It functions like 'decoy' in 'decoy duck.' Strictly speaking, a body is a body only when animated by a soul, and a soul is a soul only when animating a body. The composite is what lives and dies, death being the separation of soul and body. So far, so good.

Now comes the tricky part. For Thomas, the soul of a human being — or the intellective part of the soul of a human being — is not merely a substantial form, but also a subsistent form, a form that can exist on its own. This is the element of Platonism that remains in Thomas's Aristotelianism. This subsistent form can survive separation from the body. The theological motivation for this is perhaps clear: there must be something that bridges the temporal gap between death (separation of soul from body) and resurrection of the body and concomitant reunification of body with soul. That which grounds personal identity over the temporal gap is the soul as subsistent substantial form. Whether there is a need for such a ground is a question that cannot be discussed at the moment.

So although Thomist dualism is distinct from Platonic or Cartesian dualism, it is still a rather robust form of dualism, more robust than the dualism of the epiphenomenalist, say. As long as we don't confuse dualism with substance dualism, there is no reason that I can see for not describing Thomas's hylomorphic theory of mind as dualistic.

So much for a brief sketch of the hylomorphic position. There will be more to say in explanation of it later. I wish I could report that I find it unproblematic. But I don't. I'll mention one problem now, others later.

Problem One. On some contemporary interpretations, the mind (the intellective part or aspect of the human soul) is an emergent property. Now an emergent property, precisely because it is emergent, does not exist prior to its emergence from its emergence-base. How then can it exist after its emergence from its emergence-base? How can a substantial form exist apart from that of which it is the form? Is it not necessarily tied to that of which it is the form? After all, it is so tied in the case of non-humans like Fido. Fido is a composite the components of which cannot exist on their own. Why should it be any different in the case of the humna soul if the human soul is indeed the form of the human body?

The problem here, in short, is that there is a tension between soul as substantial form and soul as substantial subsistent form. Ontologically, one wants to protest, a form is not the sort of entity that could be subsistent. (Further tantalizing wrinkle: Aquinas describes God as forma formarum, form of all forms, but also as ipsum esse subsistens, self-subsistent existence. So God is a self-subsistent form. More grist for the mill.)

Eating is physical, wanting intentional: the object wanted need not exist without prejudice to its being wanted. The object eaten, however, must exist. But it is better to place the modal operator to the front: Necessarily, if S eats X, then X exists.

So much for eating and wanting. What about needing? It is tempting to assimilate needing to wanting. Robert Stalnaker (Inquiry, p. 11) writes, "One can need food . . . without there being any particular food that one needs."

I'm not sure about this. Although it is true that my need for food in a given situation does not entail the existence of the food I need in that situation, there is an important difference between needing and wanting. If I need vitamin C, it does not follow that vitamin C is available, but the very fact that I need the vitamin shows that it once existed, or exists elsewhere.

For if I need something to exist or be healthy, then the very fact that I need the thing or stuff shows that it is necessary for my existence or my being healthy. Given that I do exist and am healthy, it seems to follow that the thing or stuff in question either did exist or exists elsewhere.

Take water. Dying of thirst in the desert, I need it badly. My needing it does not entail the existence of a potable portion within my reach. But it does seem to entail that water once existed within my reach or does now exist outside my reach. For if there had never been any water I would not exist as the indigent critter I am.

This suggests that in conceptual space needing is 'between' eating and wanting. Eating is physical and perhaps in some cases purely physical. Think of the intravenous ingestion of nutrients by a comatose individual. That would count as 'eating' in an extended sense of the term. Although eating is necessarily eating of something (where the genitive is both subjective and objective), there needn't be any consciousness of the object eaten. And when there is such consciousness, it is irrelevant to the physical transfer of stuff into the fresser's gut. Wanting, by contrast, is essentially an affair of consciousness. There is no necessary connection to existence. I can want what does not now exist, what never did exist, what will never exist, and what cannot exist, whether nomologically or logically. I can want a mountain of gold or a fountain of youth. I can even want a round square. Kierkegaard wanted what he took to be an absurdity, the Incarnation, to be true.

Needing would appear to be an intentional phenomenon (unlike eating) but one necessarily grounded in what exists (unlike wanting). Hence the urgency of needs, especially natural (as opposed to artifically induced) needs. The need for water in all of us is more urgent that the need for heroin in the heroin addict.

To be explored in future posts: the connection between needing and valuing, and whether the objectivity of values can be inferred from the groundedness of natural needs in objective facts. And also: do spiritual needs point to the existence of that which would fulfill them? Is there an 'argument from need' waiting to be worked out?

BV: To put it a bit more precisely: if the pairing problem (PP) can arise in a purely physical situation, then EITHER the PP constitutes an objection to mental-physical causation if and only if it constitutes an objection to physical-physical causation, OR the PP shows that the theory of causation that gives rise to the PP is false. Either way, the PP does not amount to a valid objection to ISD.

In summary (with small change in notation): We have experiment X1 with cause C1: bulb B1 is on for t seconds, and effect E1: temperature rises by d degrees; similarly experiment X2 with cause C2: bulb B2 is on for t seconds, and effect E2: temperature rises by d degrees. We can run X1 and X2 repeatedly and get reproducible results. We then run experiment X* with cause C*: both B1 and B2 on for t seconds simultaneously, and effect E*: temperature rises by 2d degrees. We view C* as a 'sum' of causes C1+C2 and E* as a 'sum' of effects E1+E2, so that C1+C2 ---> E1+E2. PP then asks do we still have C1 ---> E1 or could we have C1 ---> E2.

I think there are two problems with this. The first relates to the sum of the effects and the second to the sum of the causes.

Firstly, to what extent can we discern E1 and E2 within E* in order to ask the PP question? I don't see how we can. We can't examine an air molecule and ask how much of its extra energy came from B1 and how much from B2, as if B1 emitted red energy and B2 green. (Kinetic) energy is just motion and can't be 'labelled' in the required way. We can picture molecule M1 gaining energy by interacting with a photon from bulb B1, and likewise M2 from B2, but when M1 subsequently collides with M2, it makes no sense to ask how much of its 'B1 energy' is transfered to M2 and vice versa. So E1 and E2 are just not discernible within E* and we can't formulate PP.

BV: I agree that the effects E1 and E2 are not discernible within the total effect E*. But how does it follow that we cannot ask the PP question? Your words 'discern' and 'discernible' and 'label' have an epistemological flavor: they pertain to how we know or come to have reason to believe that such-and-such is the case. But in reality -- 'ontologically' if you will -- C1 has a definite effect, namely, E1, and C2 has a definite effect, namely E2. So I can ask the Pairing Question: what accounts for the fact that C1 pairs with E1 and C2 with E2 and not rather C1 with E2 and C2 with E1?

Now the causes and the effects all occur within the same spatial volume. So spatial contiguity cannot be used to distinguish the C1-E1 sequence from the C2-E2 sequence. Temporal contiguity and temporal precedence cannot be used to distinguish them either. Furthermore, both sequences instantiate the same law of nature. Yet C1 pairs with E1 and C2 pairs with E2. Since this pairing cannot be accounted for on the relevant theory of causation -- according to which an event-sequence is causal in virtue of its instantiation of a causal law -- this theory of causation is to be rejected. If we reject it, however, the PP fails to amount to an objection either to mental-physical or to physical-physical causation.

Let me try a simpler example. A bucket is supplied with ball bearings from two tubes T1 and T2. The ball bearings are all of the same size, composition, etc. T1 supplies only enough bearings to fill the bucket halfway. The same goes for T2. Suppose both tubes are emitting ball bearings starting at the same time, and at the same rate, and over the same interval of time. C1 = emission of n bearings from T1. C2 = emission of n bearings from T2. E1 = filling of bucket with n bearings from set S1 of bearings. E2 = filling of bucket with n bearings from set S2 of bearings.

Can one tell which bearings came from which emitter? No, but that is a merely epistemological point. Surely half of the bearings came from one emitter and the other half from the other. (Compare: if you have an indiscernible twin, and I can't tell you two apart, it doesn't follow that that there is no fact of the matter about who is who. If one of you must be executed, you will want it to be your twin, and he will want it to be you.)

So we have distinct causes and distinct effects, with C1 paired with E1 and C2 with E2. But can we deduce this pairing from the theory of causation in question? No, so the theory is defective, and the Pairing 'Problem' is not a problem.

I think issues concerning summation of causes and summation of effects are red herrings -- though it is my fault for not making this clear in my original post. What is relevant here is the possibility of purely physical set-ups in which the right causal pairings cannot be established by invoking spatial differences -- because there are none. Since summation is not the issue, I don't think your second point about summation of causes is relevant. So I won't respond to it now. In any case, brevity is the soul of blog and I have already gone on too long.

The Traditional Objection

It is a given that mind affects body, and body mind. Suddenly remembering the bottle of single-malt Scotch that Harry gave me for my birthday, I am inspired to get out of my chair and pour myself some. After consuming a generous portion, I note that my mental state has altered appreciably. The mental event of remembering causes the physical event of drinking which in turn elicits the mental state of elation. But how are such interactions possible if mind and body/brain are distinct types of substance, one spatial the other nonspatial? As Elliot Sober put it recently:

If the mind is immaterial, then it does not take up space. But if it lacks spatial location, how can it be causally connected to the body? When two events are causally connected, we normally expect there to be a physical signal that passes from one to the other. How can a physical signal emerge from or lead to the mind if the mind is no place at all?

It is difficult to see what this objection is supposed to be. If Sober is saying that all causation must involve physical contact between cause and effect, then he simply begs the question against the dualist. He simply asserts something that the dualist denies. If I assert proposition p, and you assert either the negation of p or a proposition equivalent to the negation of p, then you have not refuted me, but merely opposed me. It is essential to see that if you merely contradict what I say, you have not given me a reason to abandon what I say.

If interactionist dualism is true, then there are cases of causation that do not operate by physical contact. One cannot refute this by saying, or implying, that all cases of causation operate by physical contact. At this point I will be told that dualist interaction is unintelligible. But why? If you say that it is because dualist interaction does not fit the physical-physical pattern, then again you beg the question. If you say that every case of causation involves an intervening mechanism, but there isn't one in the dualist case, then you beg the question in a slightly more subtle way. For what the dualist holds is that there are cases of direct causation. It is not clear that all physical-physical causation is indirect, but even if it were, this putative fact is not a reason to think that all cases of causation are indirect.

But how does mental-physical causation work? If this is a demand for a specification of the intervening mechanism, then the question rests on a false presupposition, namely, that there must be such a mechanism. If, on the other hand, the demand is for a specification of the causally relevant properties and covering laws, then I see no reason why this demand cannot be met.

If one thinks clearly about this, one sees that the traditional objection is quite worthless: it gives the interactionist dualist no reason to abandon his view.

The Pairing Objection

The Pairing Objection is much more serious, and much more interesting. Let m1 and m2 be mental tokens of type M and b1 and b2 brain tokens of type B, and suppose that M-type events cause B-type events. Suppose m1 and m2 both occur at time t, and b1 and b2 both occur at a slightly later time t*. Suppose further that m1 is in Tim's mind, m2 in Tom's mind, b1 in Tim's brain and b2 in Tom's brain. What makes it the case that m1 causes b1 rather than b2, and that m2 causes b2 rather than b1? What insures that m1 is paired with b1 and m2 with b2? How, on dualist interactionist assumptions, can we insure that the picture looks like this:

m1 --> b1
m2 --> b2

and not like this:

m1 --> b2
m2 --> b1?

Note that one cannot appeal to spatial contiguity to establish the right pairings for the simple reason that m1 and m2 are not in space. Thus one cannot say that m1 pairs with b1 because it is nearer in space to b1 than to b2. It is also clear that neither temporal contiguity nor temporal precedence can establish the right pairings. For both mental events occur at t, and both brain events at the later time t*.

One will be tempted to say that a mind controls the body it is embodied in, not some other body. This is true, but doesn't solve the pairing problem since embodiment is a causal notion: Tim is embodied in that body over which he has (some) causal control, but what insures that this is Tim's body? It is circular to say that Tim's mind causes changes in Tim's body rather than Tom's because Tim's mind is embodied in Tim's body — given that Tim's mind is embodied in that body over which he has causal influence.

The Pairing Objection can be met if we reject an assumption on which it is based. The assumption is that an event-sequence is a causal sequence in virtue of its falling under a law of nature. The assumption, in other words, is that what makes m1 the cause of b1 is the fact that the m1-b1 sequence instantiates the law that every M-event causes a B-event. On this assumption, there is nothing to distinguish the m1-b1 sequence from the m1-b2 sequence. For they are alike in being instances of the law, and as noted, nothing else distinguishes them.

Borrowing from Michael Tooley (Sosa and Tooley, p. 173), the assumption can also be put as follows:

Weak Reductionism with Respect to Causal Relations. Any two worlds that agree with respect to all of the non-causal properties of, and relations between, particular events or states of affairs, and with respect to all causal laws, must also agree with respect to all of the causal relations between states of affairs.

Spatial contiguity, temporal contiguity, and temporal precedence are examples of non-causal relations. Clearly, on the above assumption there is no way to distinguish the two different pairings displayed above. This implies that the Pairing Problem may be construable as an objection to Weak Reductionism with respect to Causal Relations rather than as an objection to interactionist dualism. In other words, an interactionist dualist who understands his position will not subscribe to Weak Reductionism as just stated; he will adopt a realist theory of causation according to which certain event-sequences are intrinsically causal.

What we need, however, is an independent reason to reject Weak Reductionism, a reason independent of the mind-body problem. If the Pairing Problem arises in a purely physical situation, then we have an independent, non-question-begging reason to abandon Weak Reductionism with respect to Causal Relations. So consider the following set-up. (What follows was suggested to me by John Foster's "A Defense of Dualism" in The Case for Dualism, pp. 14-15.)

There are two light bulbs, B1 and B2, in an enclosed space. They are indistinguishable: they are of the same wattage and manufacture, each is connected to a 120V AC power source, etc. When either of the bulbs is on, it causes the temperature in the enclosure to increase by X degrees in an interval of time I. Suppose that this is because of a law of nature according to which running a current through a filament causes it to emit heat. Suppose B1 and B2 are both switched on at the same time, t, and kept on for interval I. During I, the temperature in the enclosure increases by 2X degrees. Clearly, each light's being on causes its own temperature increase in the enclosed space. There are two distinct physical causes c1 and c2 and two distinct effects e1 and e2.

But what makes c1 the cause of e1 rather than the cause of e2, and c2 the cause of e2 rather than the cause of e1? The causal pairings are not fixed by the causal law and the non-causal conditions. This shows that there must be more to causation than instantiation of a law. It shows that Weak Reductionism with respect to Causal Relations is false, or at least not obviously true.

If we abandon the assumption that causal relations can be completely explained in terms of non-causal properties/relations and causal laws, then the problem of psychophysical causal pairings no longer arises. If, on the other hand, we hold fast to Weak Reductionism, then, since the Pairing Problem arises both in the mental-physical and the physical-physical case, this problem cannot be taken to be an objection to interactionist dualism.

Sober is here urging a standard objection, endlessly repeated by Dennett, Searle, et al., that is supposed to blow the substance dualist out of the water. To be clear, what we are talking about is interactionist substance dualism. One can be a substance dualist in the philosophy of mind without being an interactionist by being either a parallelist or an occasionalist. Note also that one can be dualist in the philosophy of mind without being a substance dualist by being a property dualist. Note finally that one can be a dualist without being a dualist in the philosophy of mind. If, to save bytes, I write 'dualist,' that's elliptical for interactionist substance dualist in the philosophy of mind.

Now what exactly is the objection? It seems to be this. If mind and body belong to mutually irreducible ontological categories, and yet minds and bodies interact causally, then this violates conservation principles. For example, if my intention to paint the bathroom is an irreducibly mental state that causes the states of the brain that control the motions of my limbs, then there is presumably a transfer of energy into what is supposed to be a closed physical system in violation of the principle of conservation of energy.

The trouble with this objection is that it blatantly begs the question against the dualist by presupposing a transfer theory of causation that makes dualist interaction impossible from the outset. Obviously, there could be a violation of conservation principles only if causation is being viewed as the transfer of some physical magnitude.

But it is not at all clear that causation can be identified with such transfer even among physical causes and physical effects. As Curt Ducasse once wrote:

. . . if "energy" is meant to designate something experimentally measurable, then "energy" is defined in terms of causality, not causality in terms of transfer of energy. That is, it is not known that all causation, or, in particular, causation as between psychical and physical events, involves transfer of energy. (Curt Ducasse, "In Defense of Dualism" in Sidney Hook, ed., Dimensions of Mind, Collier 1961, pp. 88-89)

We should also note that there are theories of causation that do not identify the causal link with any transfer of a physical magnitude. An empirically minded philosopher might subscribe to a version of the Regularity Theory, the gist of which is this:

Event-token e (directly) causes event-token f =df (i) e and f are spatiotemporally contiguous; (ii) e occurs earlier than f; (iii) e and f are subsumed under event-types E and F that are related by the de facto generalization that all events of type E are followed by events of type F.

On this theory, what distinguishes a causal event-sequence from a non-causal one is essentially nothing more than the former's instantiation of a regularity. Causation 'in the objects' is just regular succession, just one event after another. Accordingly, nothing gets transferred or transmitted from cause to effect. On this theory of causation, the above objection to mental-physical and physical-mental causation collapses.

Of course, there are powerful objections to the Regularity Theory. But there are other theories on which mental-physical and physical-mental causation are unproblematic. On a counterfactual theory,

e causes f =df if e had not occurred, f would not have occurred.

Here too, there is no need for any transfer of a physical magnitude and hence no threat to conservation principles.

But suppose that, in the physical world, causation is a process that involves physical contact and the transfer of energy, momentum, or whatever from the cause to the effect. If causation is such a physical process, then it will be a spatiotemporally continuous one and one can attempt to trace the mechanism whereby the cause brings about the effect. But mental-physical causation is direct: there is no intervening mechanism. To demand that there must be one in the mental-physical case as in the physical-physical case is just to rule out by fiat mental-physical causation.

Why then should there be any problem with a mental state directly causing a physical state? Once one has a specification of the relevant causal properties and the covering law, one is on the way to understanding mental-physical causation. Admittedly, though, the connection remains mysterious. But the issue is whether Sober et al. have proven its impossibility. My point is that they haven't.

In sum, the above objection unwarrantedly assumes that causation must in every case involve transfer of some physical magnitude. But it may be that causation never involves such transfer, or it may be that it involves such transfer only in the physical-physical cases. To make the above objection stick, therefore, one must do a lot of work; one must articulate a tenable transfer theory of causation. What one cannot do is simply repeat the canard given at the outset.

When Mach got on the bus he saw himself, but not as himself. His first thought was one expressible by 'The man who just boarded is a shabby pedagogue.' 'The man who just boarded' referred to Mach. Only later did Mach realize that he was referring to himself, a thought that he might have expressed by saying, 'I am a shabby pedagogue.'

Clearly, the thought expressed by 'The man who just boarded is shabby' is distinct from the thought expressed by 'I am shabby.' And this despite the fact that the very same property is being ascribed to the very same person. The difference emerges quite clearly if we alter the example slightly. Suppose Mach sees that the man who has just got on the bus has his fly open. He thinks to himself: The man who has just boarded has his fly open, a thought that leads to no action on Mach's part. But from the thought, I have my fly open, behavioral consequences ensue: Mach buttons his fly. Since the two thoughts have different behavioral consequences, they cannot be the same thought, despite the fact that they attribute the very same property to the very same person.

But what exactly is the difference?

Linguistically, the difference is that between a definite description and the first person singular pronoun. Since the referent (Bedeutung) is the same in both cases, namely Mach, one will be tempted to say that the difference is a difference in sense (Sinn)or mode of presentation (Darstellungsweise, in Frege's terminology). Mach refers to himself in two different ways, a 3rd-person objective way, and a 1st-person subjective way.

If this is right, then although there are two different thoughts or propositions, one indexical and the other non-indexical, it might seem that there need only be one fact in the world to serve as truth-maker for both, the fact of Mach's being shabby. It might seem that reality is exhausted by non-indexical facts, and that there are no such indexical or perspectival facts as those expressed by 'I am shabby' or 'I am BV' or 'I am the man who just got on the bus.' Accordingly, indexicality is merely a subjective addition, a projection: it belongs to appearance not reality. When BV says or thinks 'I,' he refers to BV in the first-person way with the result that BV appears to BV under the guise of 'I'; but in reality there is no fact corresponding to 'I am BV.'

2. But is this right? There are billions of people in the world and one of them is me. Which one? BV. But if the view sketched above is correct, then it is not an objective fact that one of these people is me. That BV exists is an objective fact, but not that BV is me. Geoffrey Maddell strenuously disagrees:

If I am to see the world in a certain way, then the fact that the world seen in this way is apprehended as such by me cannot be part of the content of that apprehension. If I impose a subjective grid on the world, then it is objectively the case that I do so. To put it bluntly, it is an objective fact about the world that one of the billions of people in it is me. (Mind and Materialism, 1988, p. 119.)

Maddell's point is that the first-person point of view is irreducibly real: it itself cannot be a subjective addition supplied from the first-person point of view. It makes sense to say that secondary qualities are projections, but it makes no sense to say that the first-person point of view is a projection. That which first makes possible subjective additions cannot itself be a subjective addition. Consider the phenomenal redness of a stop sign. It makes sense to say that this secondary quality does not belong to the sign itself in reality, but is instead a property the sign has only in relation to a perceiver. In this sense, secondary qualities are subjective. But to say that subjectivity itself, first-person perspectivity itself, is a subjective projection is unintelligible. It cannot belong to mere appareance, but must exist in reality. As Madell puts it, "Indexical thought cannot be analysed as a certain 'mode of presentation', for the fact that reality is presented to me in some particular way cannot be part of the way in which it is presented." (p. 120)

3. Trouble for materialism. According to materialism, reality is exhausted by non-indexical physical facts. But we have just seen that indexical thoughts are underpinned by indexical facts such as the fact of BV's being me. These facts are irreducibly real, but not physically real. Therefore, materialism is false: reality is not exhausted by non-indexical physical facts.

The soundness or otherwise of the argument obviously rests on (2), (1) being beyond dispute. So we need to examine the considerations Nielsen adduces in support of (2). Let's start with Nielsen's claim that "We have never encountered a bodiless person." (p. 96). I take him to mean that no one has ever encountered a bodiless person. One way to rebut this is by citing the numerous religious and mystical experiences that people through the ages have had as of bodiless persons. Once in very deep meditation I had the strong sense of being loved by something or someone distinct from myself. It was an intense sensation of being the object of a loving regard. On another occasion, there flashed across my mind the interior word, 'I am with you.' It is logically and epistemically possible that those were encounters with a bodiless person. Saul of Tarsus on the road to Damascus had an experience as of a bodiless person. And so on.

Experiences like these do not of course prove anything. No experience proves its own veridicality. Not even the ongoing perceptions of the middle-sized dry goods in my immediate environment prove the existence of coffee cups and books and computers. For these perceptions are consistent with the nonexistence of their intentional objects. (Do I know that I am not a brain in a vat? And if I do not know that I am not, can I claim to know that there is a coffee cup on the desk?) So let's leave religious and mystical experience to one side.

Since we do not encounter bodiless persons, says Nielsen, we do not know how to identify them or re-identify them. We do not know what could count as evidence for or against bodiless persons. We don't know what 'bodiless person' refers to. (97) "We literally do not know what we are talking about" (97) if we say that there is or could be a bodiless person.

What Nielsen is serving up here is something very close to the old logical positivist notion that meaning or cognitive significance is tied to empirical verifiability in principle. Nielsen is not saying that statements like 'There are bodiless persons' are false, but that they are meaningless or unintelligible. As such, they are neither true nor false. We don't know what we are talking about when we talk about bodiless persons.

Remember Cactus Ed? Ed said, "Is there a God? Who knows? Is there an angry unicorn on the dark side of the moon?" Cactus Ed's (ultimately unsustainable) point was that belief in God is just as unmotivated as belief in an irate 'lunicorn' (lunar unicorn.) But 'There is an irate lunar unicorn' is at least meaningful in that it is empirically verifiable in principle, whereas 'There is a bodiless person' does not even get the length of meaningfulness.

So Nielsen is recycling a radically positivist claim: not that 'There are bodiless persons' is false, but that it is meaningless. This claim is as shaky as the Verifiability Principle on which it rests. But Nielsen also seems to think that meaninglessness and self-contradictoriness amount to the same thing. 'Person,' he says, refers to a flesh-and-blood reality; 'bodiless person' has no referent in our experience, hence has no meaning. 'Bodiless person' is like 'temperatureless heat' or 'round quadrangle.' (96)

There is some sleight of hand going on here. If 'bodiless person' is like 'round quadrangle,' then there cannot be a bodiless person: it is logically impossible. But how does Nielsen arrive at this modal conclusion from the nonmodal premise, "We have never encountered a bodiless person"? From this premise one cannot infer that there are no bodiless persons, let alone that there cannot be.

I am afraid that Nielsen just assumes that a person is identical to an animated human body. If this were true, then of course it would be impossible for there to be a bodiless person. But the fact, if it is a fact, that bodiless persons are not encountered is not a reason to think that a person is identical to an animated human body.

But is there a positive reason to think that a person is not identical to an animated body? Consider the phenomenon of the out-of-body experience (OBE) that many people report. In such an experience a person perceives, or seems to perceive, his body from a point of view outside of it. Thus, during caridac arrest a man might look down on his body on an operating table and note certain features of the room and the people in it that are not visible from the perspective of the table. The occurrence of OBEs seems to show at least the possibility of veridical perceptions of one's body from a position external to it. Such a dismebodied perspective would involve a seeing that was not routed though one's physical eyes and would seem to be evidence of at least the possibility of a bodiless person.

Nielsen considers something like this but dismisses it by saying that "We cannot intelligibly imagine our whole body being amputated from our self and our observing this whole body amputation." (99) This is more sleight of hand, it seems. Obviously, there cannot be a whole body amputation: one's body as a whole cannot be physically separated from itself. But this obvious fact has no bearing on the possibility of bodiless personal existence unless one just begs the question by assuming that a person just is a living human body.

Nielsen is quite sure that the notion of observing one's own funeral is unintelligible. "That makes no more sense than it makes sense to speak of Flew's sleeping faster than Plantinga, or Alston's being a prime number." (99) But why exactly? Earlier, we saw Nielsen wax positivist, but now he makes an Ordinary Language move: "The 'I' here, that is supposed to do the observing, has been given no sense. It has no use in our language . . . Language in such circumstances, in Wittgenstein's famous phrase, has gone on holiday." (99)

But again, why should we accept this if we don't just assume that a person is identical to an animated body? And why accept the OLP notion that a term is meaningful only if it has an ordinary use?

2. To help you understand what I have just said, I offer an analogy.

Even though every valve-lifter is an engine part, it is quite clear that not every engine part is a valve-lifter. So what distinguishes the engine parts that are valve-lifters from the parts that are not?

The distinguishing feature cannot be anything intrinsic to engine parts qua physical objects. The mechanical, chemical, electrical, metallurgical and other terms appropriate to the description of engine parts are of no help in specifying what makes an engine part a valve-lifter. A metallurgist might tell us everything there is to know about the physical properties of those engine parts that are valve-lifters. But knowing all of that, I do not yet know what makes the part in question a valve-lifter. Similarly, I don't know what makes a certain heavy object under my hood a battery just in virtue of knowing all the electrochemistry involved in its operation.

3. The obvious thing to say at this point is that what make an engine part a valve-lifter or a battery or a generator or a transmission is its function. Physical composition is irrelevant. What makes a part a valve-lifter is the causal role it instantiates within the 'economy' of the engine. A thing is a valve-lifter in virtue of the job it does when properly connected to valves, cams, etc. Its being a valve-lifter is not intrinsic to it. Its being is its function within a system whose parts are causally interrelated.

I stress that physical composition is irrelevant. Anything that does the job of a valve-lifter is a valve-lifter. The function is 'multiply realizable' as we say in the trade. Of course, not every physical substratum supports the function: not even in Eskimo land could valve-lifters be made of ice.

Another important point is that a particular thing that functions as a valve-lifter can assume other functions, that of paper-weight for example.

I think we are all functionalists when it comes to things like valve-lifters, screwdrivers, switches, and modems. Anything that modulates/demodulates is a modem regardless of the stuff inside the box that realizes or implements the function. For all we care, there is a colony of leprechauns inside the box that chop up the analog input into digital bits. If it does the job of a modem, it IS a modem.

Can we apply this functionalist model to the mind?

4. If there is nothing intrinsic to brain states that explains why some of them are mental states, then the naturalist must look to the extrinsic features of brain states. How do they function? What role do they play? How do they stand in relation to inputs and outputs? How did they come into being? What are they good for?

One answer is the functionalist theory that causal role is what makes a brain state a mental state. What makes a mental state mental is just the causal role it plays in mediating between sensory inputs, behavioral outputs and other internal states of the subject whose state it is. The idea is not the banality that mental events have causes and effects, but that it is causal role occupancy, nothing more and nothing less, that constitutes the mentality of a mental state. The intrinsic nature of what plays the role is relevant only to its fitness for instantiating mental causal roles but not at all relevant to its being a mental state.

That's the basic idea. What makes a brain state a desire is the causal role that state plays. There is nothing intrinsic to the brain state itself that could tell you that it was a desire for a beer rather than an intention to paint the bathroom, or a memory of a trip to the Grand Canyon.

5. Here is one problem. It seems clear that my intention to clear brush could not have been a desire for a cold beer. Nor could it be an intention to paint the bathroom. The act of intending is individuated by its intentional content (to clear brush; to pain the bathroom): the content enters into the description of the act. This entails that the act could not have been an act having a different content.

But if it is causal role occupancy that makes brain state B an intention to clear brush, then B could have been an intention to paint the bathroom. Since this is absurd, it cannot be causal role occupancy that makes B an intention to clear brush.

Compare the valve-lifter. A particular engine part is a valve-lifter in virtue of the causal role it plays in the engine. But that part might not have been in the engine; it might have been on my desk weighing down papers, in which case it would have been playing a different causal role. There is no problem in this case because valve-lifters lack content, or directedness to an object. A valve-lifter is not about anything. But an intention is. And this aboutness is intrinsic to it, which is why it cannot be captured extrinsically in terms of functional role.

Mental states such as wanting, imagining, perceiving, etc. exhibit intentionality, and they do so by their very nature: they are essentially (not accidentally) of or about or directed to objects. I cannot want without wanting something, a dog for instance. What I want may or may not exist without prejudice to my wanting it. My wanting a three-headed dog does not entail the existence of a three-headed dog. And the nonexistence of such a dog leaves my thought untouched: My thought is what it is, and has the content it has, despite the nonexistence of three-headed dogs. Compare wanting to owning: if I own a three-headed dog, then there exists a three-headed dog. So if there are no three-headed dogs, then I cannot own one. Ownership is held hostage to the existence of things owned.

Owning is a relation: the nonexistence of one its terms (relata) entails the relation's nonobtaining. Wanting is not a relation -- though it is relation-like, ein Relativliches, as Brentano observed in 1874 -- since the nonexistence of the thing wanted leaves the wanting unaffected. The same goes for all cases of intentionality. None of them are relations strictly speaking.

Now compare wanting a dog, a beer, a ham sandwich, a matter transmitter. What Searle calls the "psychological mode" is the same in all four cases: all four intentional states are tokens of the type, wanting. But the intentional contents are different.

One question is this: In virtue of what does an intentional state possess content? A related question: In virtue of what does an intentional state possess the specific content that individuates it? For Fodor a state gets its content from external relations to the world. Things in the external world cause our intentional states to have the content they have.

Although it is entirely 'natural' for a naturalist to say something like this, the suggestion is fraught with difficulties.

1. First of all, some thoughts are about the nonexistent. But what does not exist cannot cause anything. So it cannot be the case that all thoughts derive their content from things in the external world. Since Cerberus does not exist, he cannot be causally responsible for my Cerberus-thoughts. The thought has the content it has despite the fact that nothing answering to it exists in the external world.

2. But even in the case of thoughts about existent items, causal relations cannot suffice to explain content. Suppose I am thinking about my cat Caissa. What makes this cat thought about Caissa? The idea is that there is a causal chain, commencing with Caissa, that terminates in my having the thought in question. But this is hopeless for reasons spelled out in a previous post. Briefly, all sorts of things in the external world are involved in the etiology of my thought; but my thought is precisely about Caissa. And as Putnam has argued, any attempt at specifying the salient cause, the cause that brings about the cat thought, will presuppose intentional notions.

3. Suppose I am thinking about Hal, who exists, but I attribute to him a property he does not possess. The content of my thought is: Hal is hypertensive, when in truth he is hypotensive. Since hypertensive Hal does not exist, how can he cause my false thought about him to have the content it has?

Naturalists have answers to these sorts of objections. My claim, however, will be that their answers just make things worse.

Now if propositions are the primary truth-bearers, then propositions, though distinct from the declarative sentences used to express them, must possess an internal structuration isomorphic to that of a declarative sentence. Alan Rhoda questioned this in a comment, citing Robert C. Stalnaker. According to Stalnaker (Inquiry, MIT Press, 1984, p. 3), "propositions lack structure of the kind that reflects the semantic structure of the sentences that express them." So we need to examine Stalnaker's theory of propositions.

1. We can introduce the notion of a proposition by saying that propositions are the vehicles or bearers of the truth-values. We can also say that they are the objects (accusatives) of such propositional attitudes as belief and desire. So far, agreement reigns. But to say the foregoing is just to say that propositions are entities that play a certain role, the role of truth-bearer, and the role of objects of the attitudes. One can presumably ask the further question as to what exactly is the nature of the entities that play those roles. This allows people to propose rather wild theories about what propositions are. Stalnaker's is a wild theory in my opinion.

2. According to Stalnaker, "a proposition is a function from possible worlds into truth-values." (p. 2) Note that S. says that a proposition is a function, not that it can be represented as one. The latter claim is innocuous; the former is not. But could a proposition be a function? Well, what is a function? A function has an 'argument' and a 'value.' If 2 is the argument of the squaring function, then 4 is the value. If -2 is the argument, 4 is the value. Essential to the notion of a function is that it assigns a unique output or value for each input or argument. A function can be thought of as a rule whereby one assigns a unique member of one set — call it the 'range' of the function to each member of another set — call it the 'domain' of the function.

3. Since everything in mathematics is a set, functions are sets. A binary function is a set of ordered pairs, and in general an n-ary function is a set of ordered n-tuples. So the squaring function is the set of all ordered pairs which set has as members <-3, 9>, <3, 9>, <-2, 4>, <2, 4>, <-1, 1>, <1, 1>, . . . .

A proposition, then, is a set of ordered pairs each member of which is of the form [W, TV] where W is a possible world and TV is a truth-value. Thus the proposition Socrates is wise is the set of ordered pairs which would have the following as members, where 'A' stands for the actual world: [A, T], [W1, F], [W2, T], . . . . Accordingly, a proposition is a monstrous set each element of which is an ordered pair whose first member is a possible world and whose second member is a truth-value. (Read the brackets as angle brackets.)

4. Sets are extensional entities: Two sets are the same iff they have all the same members. Since functions are sets, they too are extensional entities. And since propositions are functions, they also are extensional entities. This implies that two propositions are identical if they are true and false in all the same possible worlds. In other words, necessarily equivalent propositions are identical. As Stalnaker appreciates, this is highly counterintuitive. Surely (to use an example of my own), the Axiom of Choice is not the same proposition as Zorn's Lemma: one can accept the first while doubting the second. To prove their necessary equivalence is a nontrivial problem! One wants to insist that they differ in their very meaning or content. But on Stalnaker's extensionalist approach, this difference evaporates and they must be regarded as the very same proposition.

But there is worse to come. Stalnaker's extensionalism implies that there is only one necessary truth. Let P and Q be necessary truths. Since each is true in all possible worlds, there is nothing to distinguish them extensionally, and so they are identical.

5. But even apart from these difficulties, the very notion that a proposition is a set is absurd on the face of it. While there may be some latitude as to what propositions are, surely there are some things they cannot be, sets for example. Propositions are either trut or false. But it borders on incoherence to say that sets are either true or false. But that is what Stalnaker's view implies.

But much more needs to be said — in the next installment for which you will no doubt be waiting with bated breath. But now I'm off with wifey to the Five and Diner for some greasy American grub in a '50's environment.

The materialist, then, holds that everything is material: not just tables and rocks and black holes and their parts and microconstituents all the way down to the subatomic, but also human beings, their thoughts and feelings, and indeed anything that could be an existent object of their thinking.

If we think of all entities as divisible into three mutually irreducible disjoint categories, the material, the mental, and the abstract, then what the materialist is claiming is that the second two categories are empty. So the materialist is committed to denying the existence of God (as traditionally conceived), disembodied and unembodied minds/souls, unexemplified universals, and all 'Platonic' entities including Bolzano's Saetze an sich, Frege's Gedanken, and the like.

But what do we do with truth? Does it have a place in the materialist's world? Of particular interest is the truth of the materialist thesis itself, the truth of the proposition expressed by 'Everything is material.' If everything is material, then the truth or falsehood of this thesis must also be material. But if we cannot make sense of this, then materialism will be inconsistent with the conditions of its own truth.

There is a difference between truth and a truth. Truth is what is common to particular truths. Perhaps it is a property, the property of being true. (Frege thought otherwise, but his theory is so bizarre that I hesitate to bring it up.) A particular truth, on the other hand, is a true proposition, or, to put it more generally, a true truth-bearer.

So let's inquire into truth-bearers. A truth-bearer (truth-vehicle) is whatever can be appropriately characterized as either true or false. Not everything can be so characterized. A banana cannot be a truth-bearer, since no banana is either true or false. (If you say that a banana that is not fake is true, then you are equivocating on 'true.') But Poindexter's belief, of a particular banana, that it is ripe, is appropriately characterizable as either true or false, whence it follows that beliefs are candidates for the office of truth-bearer.

Other candidates: sentences-tokens, sentence-types, utterance-events, judgments, Fregean propositions. Clearly, nothing can be either true or false unless it has a sentence-like structure. Thus the set {Socrates, wisdom} cannot have a truth-value. And the same goes for {Socrates, Exemplifies, Wisdom}.

Now if a you are a materialist, then your options narrow: you cannot admit as candidate truth-bearers anything irreducibly mental or irreducibly abstract. You have to say that truth-bearers are material objects, whether simple or complex.

Herein lies the rub. It is difficult to understand how any material item, or collection of material items, could be true or false. A belief can be true or false because of its intentionality: it is of or about something distinct from itself; it is a representation. As such, it can represent things as they are, or the opposite. But materialists cannot allow that beliefs (occurrent believings) are irreducibly mental items; the materialist must either go eliminativist and deny the very existence of beliefs, or (more plausibly) attempt to identify beliefs with states of the brain. But a brain state is just a physical state of a physical organ.

Such states can be assigned semantic properties ab extra, but it is difficult to see how they could intrinsically possess semantic properties. How could a physical state be intrinsically such as to represent Bush, say, as hopelessly inarticulate, or Ted Kennedy as irremediably confused?

A similar argument can be deployed for any material item that is advanced as a candidate truth-bearer. Here on this screen are dark characters against a light background. You see these characters with your eyes. But you interpret them as meaningful with your mind. Now the meaning of these meaningful marks is not intrinsic to them as physical scribbles. For this reason it makes no sense to say that a declarative sentence-token is true or false qua string of physical marks.

It seems that we have driven the materialist into a corner. Qua materialist, he cannot admit as existent anything other than material items. But we have just seen that no material item can be either true or false. So there is no room for truths in the materialist's world. There is no room for truth since there is no room for truth-bearers.

It follows that there is no room for the putative truth that everything is material. There is no room for the falsehood of this thesis either. So we ought to conclude that materialism is inconsistent with the conditions of its own truth or falsity. Materialism as defined above is a self-refuting thesis. So it cannot be that the category of material items is the only nonempty category. If it is nonempty, at least one of the others has to be nonempty as well.

However matters stand with beliefs, it is hard to see how anyone could take an eliminativist line with respect to desires and pains. I want to protest: my lust is a DATUM — a given in plain English — and neither a theoretical posit nor a product of conceptualization. Suppose you and I are looking at the same plant. You say, "That's a cactus!" I reply, "That is not a cactus, but a succulent; every cactus is a succulent, but not every succulent is a cactus." In this example, conceptualization is at work: one and the same physical object is being conceptualized differently by me and you. My conceptualization is more accurate than yours and functions as a replacement for it. But although my desire for a beer falls under the concept desire, my desire is not external to me in the manner of the plant; I live my desire. It is an Erlebnis as Husserl would say. And though it is directed to something transcendent of my consciousness, the act with its qualitative character is a real part of my consciousness. As immanent to my consciousness, it is not something I have a theory about. It is something I live through. I endure my desire; I don't gawk at it theoretically. It is not something about which I could be substantially mistaken, so mistaken that a day might come when I would say, "I used to think I desired women, but meanwhile I have become convinced that there are no desires at all."

One cannot replace folk psychology (FP) with a better theory (one couched in neuroscientific terms perhaps) unless it is a theory. What I have been suggesting, however, is that it is not a theory. If this is right, then there is just no chance that in the future we will re-conceptualize that in us which we now folk-psychologically conceptualize as desire in some radically different way that makes no mention of desire or cognate mentalistic notions.

But don't we give folk-psychological explanations, and doesn't the fact that we do this show that folk psychology is a theory? Suppose Sally is angry, and we want to figure out why. Talking to her we find out that she is in dire financial straits and needs a substantial raise in pay. Her boss promised her one, and in expectation of the fulfillment of the promise, Sally ran up some credit card debt. But it turns out that the boss was unable to deliver on his promise. Hence, Sally's anger. This suggests an explanation involving a folk-psychological 'law':

L. If a person P is promised X confidently by a person Q whom P trusts, and P needs/desires X and expects the fulfillment of the promise, and Q is unable to deliver on the promise, then, ceteris paribus, P will become angry at Q.

To get an explanation in the familiar deductive-nomological style, we simply instantiate the variables in (L) with 'Sally,' 'boss,' and 'raise' and add the premise that Sally is promised a raise by the boss.

But is this the only way to explain Sally's angry behavior? Must we assume that FP is a theory? There is also a mental simulation view of FP. Very roughly, on this model we would understand Sally's angry behavior by 'putting ourselves in her shoes,' by imagining ourselves in her situation and then running a simulation 'off-line' as it were. For details click on the above link.

The main point is that eliminativism in the philosophy of mind requires the notion that FP is a theory; so any considerations that speak against this assumption are considerations that speak against eliminativism.

I find it strange that while behaviorism in the philosophy of mind is pretty much dead, eliminative materialism is not. For while the former borders on the merely preposterous, the latter is positively loony. What the (analytical) behaviorist does it to identify mental states with behavior (linguistic or non-linguistic) and/or with dispositions to behave. Thus my belief that it is about to rain is nothing other than my rummaging for an umbrella in the closet, and the like. This theory is so implausible that an old joke suffices to refute it: After having sex, one behaviorist says to the other, "It was good for you, how was it for me?" (I confess to being quick and Searle-y, but this is a weblog!)

The (analytical) behaviorist, however, does not deny that they are beliefs and desires, thereby retaining a grip on his sanity. His project is not eliminativist but identitarian. There are beliefs and desires, he thinks; it is just that what they are are bits of behavior and/or behavioral dispositions. But there is a wrinkle here that Collins ignores, namely, the position of eliminative behaviorism. This is the view that there are no beliefs, desires, and the like. What there is is behavior, the description of which, in the fullness of time, will not be couched in mentalistic terminology, but perhaps in the austere jargon of kinematics, or perhaps in something more robust, but shy of the irreducibly mental.

Both the eliminative materialist and the eliminative behaviorist deny the existence of beliefs and desires, sweeping them into the dustbin wherein reside phlogiston, caloric, and the luminiferous ether; the difference, however, is that the 'successor theory,' the theory that is to replace folk psychology is different in the two cases. For the eliminative behaviorist, it doesn't matter what is going on inside the skull. For the eliminative materialist, however, that is where the action is.

The eliminative materialist will reply that that argument makes the same mistake that an ancient or medieval person would be making if he insisted that he could just see with his own eyes that the heavens form a turning sphere, or that witches exist. The fact is, all observation occurs within some system of concepts, and our observation judgments are only as good as the conceptual framework in which they are expressed. In all three cases — the starry sphere, witches, and the familiar mental states — precisely what is challenged is the integrity of the background conceptual frameworks in which the observation judgments are expressed. To insist on the validity of one's experiences, traditionally interpreted, is therefore to beg the very question at issue. For in all three cases, the question is whether we should reconceive the nature of some familiar observational domain.

Even if we grant that "all observation occurs within some system of concepts," is the experiencing of a pain a case of observation? If you know your Brentano, you know that early on in Psychology From an Empirical Standpoint he makes a distinction between inner observation (innere Beobachtung) and inner perception (innere Warhnehmung). Suppose one suddenly becomes angry. The experiencing of anger is an inner perception, but not an inner observation. The difference is between living in and through one's anger and objectifying it in an act of reflection. The act of inner observation causes the anger to subside, unlike the inner perception which does not.

Reflecting on this phenomenological difference, one sees how crude Churchland's scheme is. He thinks that mental data such as pains and pleasures are on a par with outer objects like stars and planets. It is readily granted with respect to the latter that seeing is seeing-as. A medieval man who sees the heavens as a turning sphere is interpreting the visual data in the light of a false theory; he is applying an outmoded conceptual framework. But there is no comparable sense in which my feeling of pain involves the application of a conceptual framework to an inner datum.

Suppose I feel a pain. I might conceptualize it as tooth-ache pain in which case I assign it some such cause as a process of decay in a tooth. But I can 'bracket' or suspend that conceptualization and consider the pain in its purely qualitative character. It is then nothing more than a sensory quale. I might even go so far as to abstract from its painfulness.

Now the existence of this rock-bottom sensory datum is indubitable and refutes the eliminativist claim. For this datum is not a product of conceptualization, but is something that is the 'raw material' of conceptualization. The felt pain qua felt is not an object of observation, but an Erlebnis, something I live-through (er-leben). It is not something outside of me which I subsume under a concept, but a content (Husserl: ein reeller Inhalt) of my consciousness. I live my pain, I don't observe it. It is not a product of conceptualization -- in the way a distant light in the sky can be variously conceptualized as a planet, natural satellite, artificial satellite, star, double-star, UFO, etc. -- but a material or object of conceptualization.

So the answer to Churchland is as follows. There can be no question of re-conceptualizing fundamental sensory data since there was no conceptualization to start with. So I am not begging the question against Churchland when I insist that pains exist: I am not assuming that the "traditional conceptualization" is the correct one. I am denying his presupposition, namely, that there is conceptualization in a case like this.

Most fundamentally, I am questioning the Kantian-Sellarsian presupposition that the data of inner sense are in as much need of categorial interpretation as the data of outer sense. If there is no categorization at this level, then there is no possibility of a re-categorization in neuroscientific terms.

Glad we agree. Let me spell out the common argument in a bit more detail.

If EM is true, then there are no beliefs. If so, then there is no knowledge since every instance of knowledge is an instance of belief: necessarily, if subject S knows that p, then S believes that p. (The necessity here is analytic.) So if EM is true, then it cannot be known to be true.

Could the eliminative materialist accept this consequence with equanimity? I don't see how he could.

The concepts of truth, belief, and knowledge are members of a family: they cannot be understood except in relation to one another, so that it is incoherent to suppose that one could have one of them without the other. Take knowledge. Knowledge by definition is knowledge of truth (objective genitive). There can be no false knowledge. Of course, there might be truths we do not know, truths we will never know, and perhaps even truths we can never know, where 'can' picks out some species of real possibility. But the concept of truth makes no sense apart from the concepts of knowledge and mind.

If EM is true, not only are there no beliefs, there are also no conjectures, no hypotheses, no arguments, no evidence. What is an argument? Concretely viewed, an argument is a series of steps in real time executed by some mind, a sequence of inferential moves from premises to conclusions. In working through an argument, a mind has to remember the previous steps and integrate them in the unity of one consciousness with the present step so as to move toward the conclusion. But if EM is true, then there is no memory, no inference, and no argument. It then follows that there can be no evidence for the truth of EM! For the evidence must consist of arguments, and I have just shown that the existence of arguments (concretely as opposed to Platonically viewed) is incompatible with the truth of EM.

Now let's consider the suggesstion that a future neuroscience will allow us to replace such terms as 'belief,' 'memory,' 'pain,' 'headache,' etc. Suppose the suggestion is to replace the latter term with 'delta-A fiber activation of such-and-such a sort.' But surely replacing one word or phrase with another does not eliminate the DATUM that the first word or phrase denotes. The felt pain is what it is, and has the phenomenal features it has, no matter what you call it. Suppose a migraine-sufferer calls his pain 'Charley.' When he says, 'Charley's back,' his friends know what he means. Unfortunately, such a re-christening will do nothing to cause his pain to cease. A Meinongian migraine-sufferer might name his pain 'nonexistent object' but it would continue to exist all the same.

EM is so patently absurd that I must ask myself: Is it a good use of my time to beat up a cripple or roll a drunk?

1. 'Every generalization is false' is self-referential: it refers to itself or comes under its own scope. Hence it refutes itself. If true, then false; if false, then false; so, necessarily false. But 'There are no beliefs' is not self-referential. So we must abandon hope of a quick knock-out along self-referential lines.

If it is true that there are no beliefs, then there is no belief the content of which is the proposition expressed by 'There are no beliefs.' So far, no problem. But if 'There are no beliefs' is true, then no one believes it. But also: no one doubts it, rejects it, entertains it, or practices epoche with respect to it. For none of these propositional attitudes can stand if belief falls. In other words, if 'folk psychology' is a false theory, and 'belief' is vacuous, then the same goes for 'doubt' and other mental terms. One could then neither accept nor reject a proposition. One could not even 'entertain' a proposition, i.e., just consider it without either affirming it or denying it. Nor could one withdraw one's intellectual assent from a proposition that one had previously affirmed. Furthermore, no one could seek reasons for its truth, nor deduce consequences of its being true. To deduce a consequence is to enact an inference. But if there are no beliefs, then there are no inferences either.

Now if you have any common sense, this should smack of lunacy. (How could there be no inferences? We draw inferences day and night: I hear the sound of the garage door opener, and I infer that wifey is back from the store.) But philosophy does not genuflect before anything, not even common sense, so let's proceed in our search for an argument against EM.

2. The proposition expressed by 'There are no beliefs' is supposed to be true. But what are propositions? Suppose we identify propositions with the accusatives of the propositional attitudes: thus a proposition is the content of a believing or a desiring or a wishing or a willing, etc. Now if EM is true, then there are no propositional attitudes. It follows that there are no contents of such attitudes and no propositions. And if there are no propositions, then there is no proposition expressed by 'There are no beliefs.' If so, then 'There are no beliefs' is neither true nor false. For propositions are the vehicles of the truth-values.

We appear to be approaching a knock-down refutation: if 'There are no beliefs' is true, then nothing is either true or false; hence 'There are no beliefs' is neither true nor false.

3. But there is trouble even if we don't identify propositions with the contents of the propositional attitudes. Suppose we just say that propositions are the vehicles or bearers of the truth-values, and we leave propositional attitudes out of it. Propositions, then, are the entities that are appropriately characterizable as either true or false. Among these propositions is the one expressed by 'There are no beliefs.'

Now could truth be out of all relation to belief and knowledge? Truth is a property of (some) propositions. If there are no beliefs, then no proposition is believable (able to be believed), and no true proposition is knowable. For one cannot know what one cannot believe. Knowledge entails belief, though not vice versa. So I argue as follows:

A. If a proposition is true, then it is possibly such that it is believed by someone.
B. If EM is true, then there are no beliefs and no proposition is possibly such that it is believed by someone.
C. EM is true. (Assumption for reductio ad absurdum) Therefore
D. There are no beliefs and no proposition is possibly such that it is believed by someone. (From B and C by Modus Ponens)
E. No proposition is possibly such that it is believed by someone. (From D by Simplification)
F. Some proposition is possibly such that it is believed by someone. (From A and C by Universal Instantiation and Modus Ponens)
Therefore
G. It is not the case that EM is true. (From E and F by reduction ad absurdum: the conjunction of E and F is a contradiction)

The crucial premise in this reductio is (A). What it says in effect is that the concept of truth is the concept of something that cannot be coherently conceived except in relation to the concepts of belief and knowledge. Whether or not you agree that this is self-evident, it is surely more evident than the well-nigh crazy view that there are no beliefs. Since EM entails that there are no beliefs, the rational course is to reject EM.

1. An eliminativist about X is simply one who denies the existence of X. Atheists are eliminativists about God; hard determinists are eliminativists about free will; Hume (on one interpretation) and occasionalists (al-Ghazzali, Malebranche) are eliminativists about (productive event) causation; nominalists are eliminativists about universals, and so on. Eliminative materialism (physicalism) is a doctrine in the philosophy of mind according to which some or all of the items we phenomenologically classify as mental simply do not exist. It is the most radical or extreme position in the philosophy of mind.

2. Example. Right now I am thinking about a can of beer chilling in the reefer. I desire to drink it. Desire is a mental state, indeed an intentional (object-directed) mental state. Beer is a physical stuff, and drinking a physical act; but wanting or desiring a beer is mental in nature. This prima facie distinction between the mental and the physical cannot be allowed to stand on a materialist scheme. For a materialist, nothing concrete in ultimate reality is mental; all concreta are physical. Let's not worry about abstracta.

3. The materialist has two main options. One is eliminativist, the other identitarian. Given the notorious difficulties involved in fitting mental items into the materialist world-picture, the eliminativist simply denies their existence in the way people deny the existence of witches and goblins. It is a bit like cutting the Gordian knot, only more extreme. Can't untie it? Then consign it to the flames! Simply deny the problematic entities. Do unto them what Dennett does unto qualia: 'quine' them.

'Belief,' 'desire' and other mental terms on this approach do not refer to anything; they are terms of 'folk psychology,' which, according to the eliminativist, is a false theory. Thus 'belief' and 'desire' are theoretical terms; it is just that the theory to which they belong is false and slated for replacement, in the fullness of time, by a mature cognitive science.

The identitarian approach is less extreme. One concedes the existence and datanic status of mental items — as one would think that any sane person must — but then goes on to identify them with physical items. Thus my present occurrent desiring of a beer is said to be numerically identical to a certain complex state of my brain or central nervous system. Every identity theorist in the philosophy of mind is a token-token identity theorist, but some adopt the stronger type-type identity theory which entails, but is not entailed by, the token-token theory. But we needn't enter into this complexity for present purposes.

The eliminativist is an ontological radical: he flatly denies the existence of mental states, or what what comes to the same thing, he denies that mental terms have referents: they are like 'phlogiston' or 'caloric,' obsolete theoretical terms belonging to obsolete theories. The identity theorist, whose aim is not to deny the mental but to reduce it to the physical, is by contrast an ontological conservative: he want to hold onto the mental but render it respectable by reducing it to the physical.

4. On the face of it, there is a difference between saying that X does not exist and saying that X = Y. For the latter claim seems to presuppose that X does exist. But when X and Y belong to disparate categories, the difference appears to vanish. For example, is there any difference between saying that God does not exist and taking the Feuerbachian line that God is identical to a unconscious anthropomorphic projection? Suppose someone says, "God exists all right — it is just that God is an anthropomorphic projection." The proper response to such a person is to dismiss such obfuscatory rhetoric as tantamount to the claim that God does not exist. For God is not the sort of thing that could be a projection. Whether or not God exists, the concept of God is the concept of something that is a se, from itself, whence it follows that God cannot have the status of an anthropomorphic projection. So a claim that God does have this status is a claim that God does not exist.

Similarly, is there any difference between saying that mental state M is token-identical to physical state P, and saying that M does not exist? If M reduces to P, and M has all and only theproperties possessed by P, then all there is is P, and the reductionis tantamount to an eli9mination.

Or if a Humean reduces causation to regular succession, then that arguably boils down to a denial of causation.

At this point I would argue:

The token-token identity theory collapses into eliminativism
Eliminativism is false
-----
The token-token idenity theory is false.

5. But why is eliminativism false? I will give one quick argument now, and others later.

Mental terms are not theoretical. It strains credulity to claim that terms like 'pain,' 'desire,' 'belief,' recollection,' etc. are theoretical terms belonging to a supposed theory called 'folk psychology,' a theory that is supposed to be analogous to astrology or alchemy. When I am in pain, I know with certainty that I am in pain: the pain is immediately given. As such, the pain is a datum, not a theory or a theoretical posit. I am not theorizing about anything, but reporting an immediate phenomenologically evident datum. 'Pain' refers directly to that immediate datum, a datum that is indubitable. So not only is 'pain' not theoretical, it obviously has a referent and is known with certainty to have a referent. The same goes for desire and other states of mind. Our direct acquaintance with them shows that they exist and that words referring to them are nonvacuous.

The subtitle is "Philosophical Obstacles to a Science of Consciousness." That raises the question of how there could even be philosophical obstacles to such a science. I am not aware that philosophers control the sources of funding for neuroscience projects. And what could a philosopher say that could stymy brain science?

But let's look at a passage:

If we are are to explain the conscious Subject, one way or another the transition from clueless cells to knowing organizations of cells must be made without any magic ingredients. This requirement presents theorists with what some see as a nasty dilemma . . . . If you propose a theory of the knowing Subject that describes whatever it describes as like the working of a vacant automated factory — not a Subject in sight — you will seem to many observers to have changed the subject or missed the point. On the other hand, if your theory still has tasks for a Subject to perform, still has a need for a Subject as witness, then . . . you have actually postponed the task of explaining what needs explaining.

To me, one of the most fascinating bifurcations in the intellectual world today is between those to whom it is obvious — obvious — that a theory that leaves out the Subject is thereby disqualified as a theory of consciousness (in Chalmer's terms, it evades the Hard Problem), and those to whom it is just as obvious that any theory that doesn't leave out the Subject is disqualified. I submit that the former have to be wrong. . . . (p. 145)

Dennett has done a good job of focussing the issue. On the one side, the eliminativists who hold that the only way to explain the conscious Subject is by explaining it away. On the other side, those who are convinced that one cannot explain a datum by denying its existence.

What we have here fundamentally is a deep philosophical dispute about the nature of explanation, and not a debate confined to the philosophy of mind. What's more, it is not a debate that is going to be resolved by further empirical research.

To me it is self-evident that any explanation that consigns the explanandum to the status of nonexistence is a failure as an explanation. Eliminativist moves are confessions of failure. Any genuine explanation of X presupposes (and so cannot eliminate) the existence of X.

The thing about consciousness is that the only way to eplain it in terms satisfactory to a materialist is by denying its existence. But it doesn't bother the materialist that the entities he invokes in his explanations themselves go unexplained.

Why the double standard? Why the demand that consciousness be exhaustively explained in terms involving no tincture of consciousness, without demanding that the material world be explained in terms involving no tincture of materiality?

I develop this line of thought in more detail here. See also the other entries in the category, Dennett Critique.

To answer this question we first need to define naturalism. This is not particularly easy, but in earlier posts I have plumped for David Armstrong's definition:

It is the contention that the world, the totality of entities, is nothing more than the spacetime system. . . . The positive part of the thesis, that the spacetime system exists, is perhaps not very controversial . . . . The negative thesis, that the spacetime system is all there is, is more controversial. (A World of States of Affairs, p. 5)

If we accept Armstrong's definition — and I see no reason not to accept it — and if naturalism so defined is true, then the following do not, and presumably cannot, exist: God as classically conceived, disembodied minds/souls, unexemplified universals, and a whole range of objects variously characterizable as ideal, Platonic, or abstract, including Fregean propositions, Fregean senses in general, numbers, irreducible sets, and the like.

In sum, naturalism (N) is the thesis that reality is exhausted by the space-time system. So the question before us can be formulated thusly:

Q. Is there some proposition P such that (i) we know that P, and (ii) P entails ~N?

There are unexemplified universals entails ~N. But do we know that there are unexemplified universals? There are various arguments one can give, and some of them are valid and have plausible premises. But that is not to say that we know their conclusion. Similarly for arguments for Fregean propositions and other items on the above list.

Those of you who would return an affirmative answer to (Q) need to specify some values for 'P.'

Consider the true sentence '2 + 1 = 3' standardly interpreted as a truth of base-10 arithmetic. I say that this sentence is not about physical objects, but about abstract or ideal or 'Platonic' objects. (Forget about the historical Plato and his actual doctrines.) Such objects are neither spatial nor temporal nor mental. To say that they are abstract is to say that they are (i) necessarily existent (existent in all metaphysically possible worlds) and (ii) causally inert: they do not enter into causal relations as either causes or effects. The term 'abtract' is less than felicitous, suggesting as it does a product of a mental act of abstraction; but that infelicity is no part of its meaning in this context.

One way to support this 'platonism' is by considering the defects of alternative views. By 'support,' I mean make a rational case for. I doubt that one can provide 'knock-down' arguments for it. One alternative could be called physicalism. Bill Keezer in an earlier thread made this suggestion if I understood him correctly.

How might one interpret '2 + 1 = 3' physicalistically? One might interpret it as about piles of physical things. Whenever one has a pile of two things, and adds a third, the result is a pile of three physical things. Accordingly, the arithmetical sentence is not about certain platonic objects subsisting in splendid and serene isolation from the flux and shove of the spatiotemporal order; the sentence is an empirical generalization about physical piles. But surely this leaves something to be desired.

Here is a quick consideration. Set-theory is essential to modern mathematics. But it is difficult to see how sets could be identified with piles of physical things. For there are many more sets than there are piles. I've got a pile of two bricks, A and B. Corresponding to that pile is the set, {A, B}. That set has four subsets: {A, B}, {A}, {B}, { }. But they form a set, which is called the power set of the original set, namely: { {A, B}, {A}, {B}, { }}. But there is also the power set of that power set, and so on ad infinitum.

Clearly, there are no physical piles corresponding to all these sets. So sets cannot be identified with physical piles. Sets are mathematical objects. Therefore there are mathematical objects that cannot be identified with physical objects. Numbers are construable as sets. Therefore numbers cannot be identified with physical objects whence it follows that not everything is a physical object!

The author is introducing a discussion of the imaginary number i = the square root of -1. His point is that we are free to introduce this number since all numbers are inventions. So we can make up any number we like. The actual argument given is self-contradictory: The point of saying that numbers do not grow on trees is that they do not occur in nature. But if they live in our heads, then they are part of nature, because what is in our head is part of nature.

But let's be charitable. The argument the author is trying to give is something like this:

1. Numbers are not physical objects
Therefore
2. Numbers are mental constructions.

That this is a non sequitur should be obvious. For there is a third possibility: numbers are abstract or ideal or Platonic objects. This third possibility is seen to be plausible once one considers the difficulties with the view that numbers are mental constructions.

Note first that an abstract object is not one produced by a mental act of abstraction. For present purposes we can say that an abstract object is any entity that necessarily exists but is causally inert.

Note second that a number is not the same as a numeral. One and the same number can be represented by different numerals. Thus the same number is denoted by the Arabic '9' and the Roman 'IX.' Numerals are physical, while numbers are not. So no number is a numeral.

We also note that 9 in a base-10 or decimal system is equivalent to 1001 in a base-2 or binary system. When I speak of the number 9 I am referring to the denotatum of the numeral '9' as this numeral functions within our ordinary base-10 system. That denotatum is the say as the denotatum of '1001' as the latter functions within a base-2 system.

One and the same proposition can be expressed by different indicative sentences. Thus the binary sentence '1 + 1 = 10' expresses the same true proposition as is expressed by the decimal sentence '1 + 1 = 2.' But if the two sentences are both interpreted relative to the decimal system, then they express different propositions, one true and the other false.

Our question is whether numbers themselves are mental constructions, not whether numerals are mental constructions. This is connected with the question of whether mathematics is in any sense conventional. No doubt notation systems are conventional, i.e. decided upon by human beings (or whatever other intelligent critters there might be elsewhere); but it doesn't follow that numbers or other mathematical objects are.

If numbers themselves are mental constructions, then they depend on our existence for their existence. (Forget about extraterrestrial intelligences for the nonce.) The same goes for the truths in which they are involved. (Thus 7 and 9 and 16 are involved in the truth expressed by '7 + 9 = 16'.) But we didn't always exist. Consider a time before any minds existed, some time after the Big Bang and before the emergence of life on earth, say.

During that interval, the speed of light and the speed of sound were the same as they are now, and during that time the former was greater than the latter, as is the case now. Let 'c' denote the speed of light in a vacuum. C is identical to some number, which number depending on the units of measurement one employs. So c = 186,000 miles/sec (approximately). In the metric system, c = 300,000 km/sec (approximately). The point is that once the system of measurement is fixed -- which of course is conventional -- then some definite number is the SOL. Similarly with the speed of sound, SOS. Now

1. SOL > SOS

is true now and was true at the time when no humans existed. Of course, at that time the concept or notion or idea greater than (taken in its mathematical sense) did not exist since concepts (notions, ideas) cannot exist except 'in' a mind. ('In' here not to be taken spatially.) But the mathematical relation picked out by '>' existed. For if it did not, then (1) could not have been true at the time in question. And the same goes for the relational fact of SOL's being greater than SOS. That fact obtained at the time when no minds existed. So its constituents (the numbers and the greater than relation) had to exist at that time as well.

Therefore, mathematical objects cannot be our mental creations.

. . . The objection to interactionism that causation, in either direction, as between psychical [mental] and physical events is precluded by the principle of the conservation of energy (or of energy-matter) is invalid for several reasons.

A. One reason is that the conservation which that principle asserts is not something known to be true without exception but is . . . only a defining-postulate of the notion of a wholly closed physical world, so that the question whether psycho-physical or physico-psychical causation ever occurs is (but in different words) the question whether the physical world is wholly closed. And that question is not answered by dignifying as a "principle" the assumption that the physical world is wholly closed.

B. Anyway, as C. D. Broad has pointed out, it might be the case that whenever a given amount of energy vanishes from, or emerges in, the physical world at one place, then an equal amount of energy respectively emerges in, or vanishes from that world at another place.

C. And thirdly, if "energy" is meant to designate something experimentally measurable, then "energy" is defined in terms of causality, not causality in terms of transfer of energy. That is, it is not known that all causation, or, in particular, causation as between psychical and physical events, involves transfer of energy. (Curt Ducasse, "In Defense of Dualism" in Sidney Hook, ed., Dimensions of Mind, Collier 1961, pp. 88-89)

I will now proceed seriatim through these points, supplying my own interpretation of them.

Ad (A). Any physicalist worth his salt will uphold the causal closure of the physical domain. That is just part (though not the whole) of what it means to be a physicalist. Borrowing from Jaegwon Kim, the principle may be stated thusly: Any physical event that has a cause at time t, has a physical cause at t. You QM enthusiasts out there will please note that this does not imply that every physical event has a cause. Note also that Kim's formulation seems to allow for irreducibly mental events as causes of physical events. But if a physical event e is a sufficient cause of a physical event f, then any mental cause m of f will just be epiphenomenally along for the ride, if you catch my drift. In other words, m won't do any work.

Adding to Kim's formulation the notion that physical causes are sufficient for their physical effects yields a robust notion of causal closure. Robustly understood, the causal closure of the physical domain amounts to the thesis that all the causal work that gets done in the physical domain is done by physical events; if there are any irreducibly mental events, they don't do any work in the physical domain.

Ducasse's first argument, then, may be understood as follows. Appeal to conservation of energy is equivalent to appeal to causal closure of the physical. But one who objects to interactionist dualism on this basis begs the question against it by assuming the truth of a principle (causal closure) that immediately entails the falsity of interactionism.

The objection from conservation/closure is therefore not decisive against the interactionist. It would be decisive if the closure principle were known to be true. But that would be tantamount to knowing that physicalism is true. But we don't know it to be true. Of course, physical science proceeds by searching for physical causes. That is the kind of game it plays. Its procedure is methodologically naturalistic. But there is a logical gap between methodological and substantive naturalism.

Ad (B). I don't find Ducasse's second point all that impressive. Assume conservation of energy. Then, if causation involves transfer of energy, and some mental events are causally efficacious in the physical domain, then energy must enter the physical domain at some point, call it the locus of interaction. Does this violate conservation of energy? Only if it is assumed that energy does not vanish at some other point. Since this is logically possible, the objection is not decisive. In other words, there could be a net conservation of energy or matter-energy in a system in which energy arose and vanished in different places.

Ad (C). Assume the causal closure of the physical domain. One could still be an interactionist dualist by denying that mental-physical causation involves transfer of a physical magnitude such as energy. Of course, if we know that every instance of causation is an instance of energy transfer, then of course we know (via some simple inferences) that mental-physical causation is impossible. But we don't know this. Therefore, to object to interactionism on the ground that all cauasation involves energy transfer is to beg the question against the interactionist.

Note the difference between the first and third objections. The first objection begs the question against the interactionist by assuming that it is known that the physical domain is causally closed. The third objection begs the question by assuming that it is known that causation always involves energy transfer.

It looks like we ought to distingish between the interactionist who accepts a transfer theory of causation but rejects causal closure, and the interactionist who accepts causal closure but rejects a transfer theory of causation.

Ducasse also alludes to the following point. Isn't transfer itself a causal notion? How then can causation be analyzed in terms of transfer? Even if every instance of causation were an instance of energy transfer, that would not entail that causation consists in energy transfer. But this is a subtle point best reserved for a separate post.

I'm glad you asked, Jimmy. My main target in a series of posts over the past weeks and months has been metaphysical naturalism. (As opposed to epistemological and ethical naturalism which I also reject.) I haven't positively advocated substance dualism, but I maintain that it is a serious contender, along with idealism. I have been concerned to show how weak and dogmatic virtually all contemporary rejections of substance dualism are, especially those of top philosophers such as Dennett and Searle. For the moment, you could put me down as either a substance dualist or an idealist, with a preference for idealism. Idealism receives almost no discussion currently.

I would recommend the following contemporary defenses of substance dualism in no particular order: W. D. Hart, The Engines of the Soul (Cambridge 1988); John Foster, The Immaterial Self (Routledge 1991); Richard Swinburne, The Evolution of the Soul (2nd ed. Oxford 1997); Beloff and Smythies, ed. The Case for Dualism (U. of Virginia Press, 1989). You can find further references in Edward Feser, Philosophy of Mind: A Short Introduction (Oneworld Publications, 2005, p. 37). Ed is a colleague of mine over at Right Reason, and a very good philosopher. I suggest you begin by reading his chapter on dualism before plunging into the more technical material.

And thanks for reminding me that I need to get a copy of Jaegwon Kim's new book.

Now what exactly is the objection? It seems to be this. If mind and body belong to mutually irreducible ontological categories, and yet minds and bodies interact causally, then this violates conservation principles. For example, if my intention to paint the bathroom is an irreducibly mental state that causes the states of the brain that control the motions of my limbs, then there is presumably a transfer of energy into what is supposed to be a closed physical system in violation of the principle of conservation of energy.

The trouble with this objection is that it blatantly begs the question against the dualist by presupposing a transfer theory of causation that makes dualist interaction impossible from the outset. Obviously, there could be a violation of conservation principles only if causation is being viewed as the transfer of some physical magnitude.

But it is not at all clear that causation involves such transfer even among physical causes and physical effects. There are several theories of causation. Many empirically-minded philosophers, following in the footsteps of Hume, adopt some version of the Regularity Theory, the gist of which is this:

Event-token e (directly) causes event-token f =df (i) e and f are spatiotemporally contiguous; (ii) e occurs earlier than f; (iii) e and f are subsumed under event-types E and F that are related by the de facto generalization that all events of type E are followed by events of type F.

On this theory, what distinguishes a causal event-sequence from a non-causal one is nothing more than the former's instantiation of a regularity. Causation 'in the objects' is just regular succession, just one event after another. Accordingly, nothing gets transferred or transmitted from cause to effect. On this theory of causation, the above objection to mnetal-physical and physical-mental causation collapses.

Of course, there are powerful objections to the Regularity Theory. But there are other theories on which mental-physical and physical-mental causation are unproblematic. On a counterfactual theory,

e causes f =df if e had not occurred, f would not have occurred.

Here too, there is no need for any transfer of a physical magnitude and hence no threat to conservation principles.

But suppose that, in the physical world, causation is a process that involves physical contact and the transfer of energy, momentum, or whatever from the cause to the effect. If causation is such a physical process, then it will be a spatiotemporally continuous one and one can attempt to trace the mechanism whereby the cause brings about the effect. But mental-physical causation is direct: there is no intervening mechanism. To demand that there must be one in the mental-physical case as in the physical-physical case is just to rule out by fiat mental-physical causation.

Why then should there be any problem with a mental state directly causing a physical state? Once one has a specification of the relevant causal properties and the covering law, what more could one ask for?

In sum, the above objection unwarrantedly assumes that causation must in every case involve transfer of some physical magnitude. But it may be that causation never involves such transfer, or it may be that it involves such transfer only in the physical-physical cases. To make the above objection stick, therefore, one must do a lot of work; one must articulate a tenable transfer theory of causation. What one cannot do is simply repeat the canard given at the outset.

Ad 1. There are certain data that no one will dispute, whether materialist, dualist, or idealist. Among these data are the various correlations to which Malcolm is referring: stimulate this portion of the visual cortex in such and such a way and the subject experiences phenomenal blue, etc. Intelligent dualists have always been aware of such basic facts as that drinking alcohol alters the quality of one's qualia, that a blow to the head can cause unconsciousness, and the like. It is important to realize that dualists are not in the business of denying obvious facts. The questions are not about the gross facts, but about their interpretation, about what they mean and what they entail. Hence dualists cannot be refuted by citing any obvious facts. Indeed, if dualism could be refuted by citing empirical facts, it would not be a philosophical thesis at all.

I stress this, because many don't understand it. They think that substance dualists deny facts that are well-known or scientifically established. One commenter, for example, compared substance dualists to flat-earthers — which of course shows total misunderstanding.

"Why is this the case, if our minds aren't simply something the brain is doing?" Because it could be the case even if our minds are not simply something the brain is doing. If substance dualism is true, then the mind is a substance. But note the following definition:

D1. X is a substance =df X is metaphysically capable of independent existence.

(D1) lays down what is meant by 'substance' in discussions about substance dualism in the philosophy of mind. That and that alone is what is meant by the term.

Note also that 'substance' has a half-dozen or so meanings and that in this context, 'substance' does not mean stuff. Thus the dualist cannot be blown out of the water by some such cheapshot as saying that he is committed to something self-contradictory like immaterial matter. (Not that Malcolm would reach for such a cheapshot.)

So for the dualist, the mind can exist without being embodied. But my mind, with which I am rather well acquainted, is an embodied mind. It is embodied as a matter of contingent fact, though not as a matter of metaphysical necessity. So it is not surprising that what goes on in my mind affects and is affected by what goes on in my brain and central nervous system. It is not surprising that the states of an embodied mind will be affected by alcohol in the bloodstream. In general, it is not suprising that (some) changes in the brain will bring about changes in the mind.

Since the facts that Malcolm adduces can be explained both materialistically and dualistically,, his adducing of said facts does not support materialism over dualism. Since the facts are consistent with both schemes, they do not entail either scheme.

So what Malcolm says in #1 is not a good reason to reject dualism. Of course, what I said in rebuttal does not provide a good positive reason to accept dualism over materialism. What I have done is merely remove a threat to the rationality of dualist belief.