The Place of Mind in Nature

Stephen Voss

March 2004


We dance round in a ring and suppose,
But the Secret sits in the middle and knows.[1]



            1. How do you and I -- beings with minds -- find a place in the natural world? Mindful that our understanding of both mind and nature is surely very small and that the truth is sure to surprise us, let us consider a line of reasoning which leads from two points inspired by Galen Strawson and William James to an unexpected answer. The points seem to conflict, but they are consistent with one another, and each expresses an intuition worth taking seriously. My  aim here is to explore the consequences of accepting them both.


            Strawson articulates an intuition perhaps first expressed by Aristotle, about the priority of actual mind to the potential for mind. James articulates an intuition earlier voiced by Kant, about the mereological irreducibility of mind. Thomas Nagel articulates intuitions of both sorts. I draw from their writings two theses which I shall call Ultimate Subjecthood and Integral Subjecthood.


            I see the line of reasoning we are about to examine as a narrow path through a jungle overgrown with my ignorance and preconceptions. I see the enterprise of considering it, not as thinking about whether a set of sentences is true, or thinking about whether the fact that p entails q suggests q more than it suggests not-p, but rather as walking patiently back and forth along the path with machete in hand, cutting back the foliage in an effort to discern the contours of the natural world.


            I do not look on the reasoning as a proof. It calls attention to connections among ideas, and to aspects of the world. Its value will be shown in the results of a person’s efforts to widen its pathways and set up camps in the clearings. Naturally the most interesting clearing is the one at the end of the trail, and I devote some time to a sketch of the theory that emerges. But I believe that only after a great deal of that kind of work, and other kinds as well, will it be possible to be clear how sound the reasoning is and how reasonable the theory is.


            2. The fundamental subject matter of philosophy of mind is subjects, that is, beings with minds, that is, beings that are capable of having mental states; and subjects will be our topic.


            I suppose that subjects really are present in nature. For Descartes my question does not even arise, since he takes subjects to comprise a realm alien to nature. But subjects certainly appear to have nonpsychological states, and I take it that possession of nonpsychological states is sufficient for a substance to be present in nature. In particular, subjects appear to be animals.


            3. I will suppose that subjects are substances. I’ll call a particular modification of a substance a state of it. Substances have (particular) states in virtue of having corresponding (universal) properties. They also stand in particular and universal relations to one another, though for expository ease I shall sometimes speak simply of their states or properties.


            4. Physics provides a conception of “the building blocks of nature” -- the entities that make up the substances found in nature. Familiar candidates are atoms (yesterday’s physics), mass-energy packets (today’s physics), and space-time (tomorrow’s physics?). I’ll call them the ultimate entities of nature. These entities are themselves substances. They appear to be extremely tiny. They certainly do not appear to be subjects.


            5. The line of reasoning I will pursue diverges dramatically from the standard picture of both mind and nature: Subjects are not animals and ultimate entities are not tiny. Being a subject is equivalent to being an ultimate entity, and there is only one of them.


I. Subjects


            6. Let us stipulate that a subject is a being that is capable of (at least a small range of) mental states, and that subjecthood is the property of being capable of mental states.


            What is the nature of subjecthood? What is it to be a subject? How is it that beings that have this property are capable of a range of mental states? These are difficult and little explored questions. We need not try to answer them here.


            7. On a strong conception of subjecthood, to be a subject is to have a certain kind of nature, and so to be a certain kind of being. On such a conception, the reason why the disparate mental states compose a distinct metaphysical class for the philosophy of mind is that the very capacity to have any one of them depends on a property specific to the beings that have them. We might mark it by speaking with Leibniz and Nagel of having a point of view, or with Leibniz of being an I, or with ordinary people of having a mind.


            On a weak conception, there need be no interesting property common to beings capable of mental states. For example, it may be thought that subjecthood requires nothing but a level of functional complexity sufficient for some sort or other of mental life, and that what supports one sort may have nothing interesting in common with what supports another. Some kinds of subjects will enjoy a point of view, but others will not.


            On a strong conception, however, one must have a point of view in order to have a sense impression or a feeling -- but equally to calculate, deliberate, wonder, or doubt. One must have a point of view in order to want or feel obligated, in order to know or believe, in order to have a virtue or vice or other character trait, in order to have an inclination to feel something or think of something or do something, in order to be capable of any of the above. In short, one must have a point of view in order to have any of the states that deserve to be called mental. The primordial metaphysical requirement for having any mental state at all is to have a point of view.


            On such a conception of subjecthood the most basic reason a thermostat does not think the room is too cold, or want it to be warmer, is that it does not have a point of view. It is not an I. Pointing to the thermostat, we might make the point by saying that no one is home. None of these ways of speaking is satisfactory. ‘I’ is not a general term; ‘there’s someone home’ is metaphorical; and so are both ‘point of view’, extrapolated from the domain of vision, and ‘point’, extrapolated from geometry. Within the mental states that require a point of view, there is nothing sacred about viewing or spatial positioning. Perhaps the best thing is to say simply that a thermostat is not a subject.


            I have already observed that the fundamental subject matter of philosophy of mind is subjects. But if subjects are a genuine kind of being, the fundamental question in philosophy of mind is what it is to be a subject. The fundamental task for the philosophy of mind will then be the elucidation of what it is to be that kind of being; and with it the elucidation of how it is that being a subject enables one to have mental states, and begins to make possible experience and thought, feeling and desire, action and contemplation.


            8. My motive for focusing here on subjecthood, rather than on experience or knowledge or even mental states generally, reflects my belief that the strong conception is correct, and therefore that the fundamental question in philosophy of mind is the nature of subjecthood.[2] But nothing I shall say here hangs on this belief. We may leave all such questions about the nature of the property open and simply regard subjecthood as the capacity to have mental states. That will suffice to fix the kind of reference to subjects and subjecthood that we shall require here.


            9. We may call a property psychological if the attribution of it to a substance implies that the substance is a subject, and nonpsychological otherwise. A state of a substance is psychological if the substance has it in virtue of having a psychological property, and is nonpsychological otherwise.


II. Ultimate Supervenience


            10. Ultimate Supervenience is the thesis that in the nature of things, every state of a substance supervenes on -- is determined by -- states of ultimate entities. It rules out one form of emergence.


            The thesis says nothing about the kinds of states ultimate entities have. I assume that ultimate entities have nonpsychological states, but the thesis is silent on whether they also have psychological states and on the related question whether they have nonphysical states. It is not a physicalist thesis.


            Some states of ultimate entities themselves may be primitive -- not determined by other states. That would not violate the spirit of Ultimate Supervenience. To allow for the possibility I shall take ‘determine’ in this context to be reflexive, so that every state of a substance will trivially determine itself. This stipulation is harmless and not entirely unintuitive, since every state is necessary and sufficient for itself. As Section VI below suggests, this trivial form of determination may turn out to characterize the very heart of mind’s place in nature.


            Ultimate Supervenience might be true. Let us suppose that it is.


            11. Then each of a subject’s mental states will be determined by states of the relevant ultimate entities. So will its very subjecthood. I will say that an ultimate entity E helps constitute a substance S a subject (helps constitute S, for short) if E is one of the ultimate entities whose states determine that S is a subject. A little more precisely, if the states of one or more ultimate entities E1, E2, ..., En throughout a time interval t determine that a substance S is a subject during t, while the states of E2, ..., En by themselves do not, then E1 helps constitute S a subject over t. (In the limiting case in which n is 1, and the states of E1 on their own determine that S is a subject during t, we may say more simply that E1 constitutes S a subject over t.) No doubt at least some of the ultimate entities that help constitute S a subject will be among those that make S up.


            12. Then ultimate entities will be capable of helping to constitute subjects, when standing in proper relations to other ultimate entities. Not only that; as Nagel points out, if an ultimate entity is capable of helping to constitute a subject, it will also have the following remarkable property. It will be capable of helping to constitute every kind of subject there can be (not only human beings, but bats and the rest): “Not only can a single organism have different experiences, but its matter can be recombined to form other organisms with totally different forms of experience. The mental properties of all matter, therefore, would have to be not species-specific but universal, since they would underlie all possible forms of consciousness.” (Nagel 1979, p. 194)


            13. It is plausible on scientific grounds to conjecture that if some ultimate entities have this property, they all do. For one thing, one ultimate entity of a given kind is as good as any other of the same kind, and current science suggests that most or all of the currently recognized kinds help constitute you and me subjects. For another, science also suggests that one kind of ultimate entity is convertible into another. So I will suppose that every ultimate entity has this property -- namely, the capacity to help constitute any possible kind of subject. This is a weak (because dispositional) form of panpsychism.


            Nevertheless, the argument requires only that some ultimate entities have the property. That is remarkable enough. In the end, weak panpsychism will be a corollary; 30 will independently ground all the claims made in 13-27 about “each” and “every” ultimate entity.


III. The Constitution of Subjects. Strong Panpsychism


            14. What is the nature of this property of ultimate entities? I would like to see what the search for the best answer to this question might disclose about the nature of ultimate entities, the nature of subjects, and the cardinality of each of these classes of substances.


            15. One place where we find a substance’s intrinsic properties is in its nonrelational powers and propensities, those that do not imply the existence of specific other substances. Being capable of bearing a child, for example, unlike being a mother, does not imply the existence of a child, and flammability, unlike being set on fire, does not imply the existence of an igniting object. Childbearing capability and flammability are in this way nonrelational, and therefore intrinsic, properties.


            16. Every ultimate entity has the nonrelational power to help constitute a subject, just as every U-235 atom has the nonrelational power to help cause a self-sustaining nuclear reaction. Then the capacity to help constitute a subject will be an intrinsic property of each ultimate entity.


            Here is how Nagel makes the point.


It is of course obvious that what is going on in my brain causes my mental state, just as it is obvious that when I touch a hot pan it causes pain. There must be some kind of necessity here. What we cannot understand is how the heat, or the brain process, necessitates the sensation. So long as we remain at the level of a purely physical conception of what goes on in the brain, this will continue to appear impossible. The conclusion is that unless we are prepared to accept the alternative that the appearance of mental properties in complex systems has no causal explanation at all, we must take the current epistemological emergence of the mental as a reason to believe that the constituents have properties of which we are not aware, and which do necessitate these results.


       The demand for an account of how mental states necessarily appear in physical organisms cannot be satisfied by the discovery of uniform correlations between mental states and physical brain states, though that is how psycho-physical laws have traditionally been conceived. Instead, intrinsic properties of the components must be discovered from which the mental properties of the system follow necessarily. This may be unattainable, but if mental phenomena have a causal explanation such properties must exist, and they will not be physical. (Nagel 1979, p. 187)


            A corollary is worth noting, even though it is not part of the argument we are examining. It extends the remarkable fact we observed in Section II. If you wish to list the distinct intrinsic properties of ultimate entities, you will have to mention, in addition to mass, charge, and other properties familiar from physics, the capacity to help constitute subjects. (Nagel’s reasons for thinking that this capacity is not a physical property are relevant to the corollary, but not to the argument we are considering. The notion of a physical property is probably not of great importance for the philosophy of mind.)


            17. A U-235 atom has the power to help cause a self-sustaining reaction in virtue of the fact that when bombarded with a neutron it will split in two and give off more neutrons and some energy. Within atomic theory, its power to help cause a self-sustaining reaction is theoretically derivative from its capacity to react in this way to a neutron collision. As the example suggests, there is a way in which, when we theoretically derive a property from some other property, we learn something about the nature of the derivative property. We learn something about how or why, in the nature of things, the property is actually instantiated.

            At some point we will find properties whose nature cannot be better understood via such a derivation and must be accepted as theoretically basic. Some properties seem to cry out for further explanation. Some seem good candidates for basic status. To find a property basic is to accept one kind of theoretical defeat: the impossibility of gaining one kind of understanding of its nature.

            18. Our question is whether an ultimate entity’s power to help constitute subjects is theoretically derivative from some more basic property. This is in one respect the fundamental question concerning the place of mind in nature. We have seen good reason to think we must attribute this property to ultimate entities. It is time to ask whether its nature can be better understood or whether it should simply be accepted as basic. Should we hope to create a theory and locate a property P such that within the theory the power to help constitute subjects is derivative from P?  Or is the power to help constitute the range of possible subjects a theoretically basic property of ultimate entities?


            It may appear to us that something in an ultimate entity’s nature must account for its power to help constitute subjects, and therefore that this power cannot be basic and must be theoretically derivative. We may wish to conjecture that the fundamental question must have a positive answer. That conjecture in fact follows from the thesis which we shall now consider.


            19. The only hypothesis capable of sustaining a positive answer that readily comes to (my) mind is the strong panpsychist position that every ultimate entity is itself a subject. This hypothesis follows from the conjunction of 13’s weak panpsychism with the thesis of Ultimate Subjecthood that (in rough terms) only a subject can help constitute a subject. I shall quote Strawson’s argument for a similar position, laying the groundwork for the more precise formulation of the thesis that I give in 20 and 21 below. Strawson writes:


I think many who call themselves physicalists or materialists really are committed to the thesis that


N-E  physical stuff is, in itself, in its fundamental nature, something wholly non-experiential.


I think they take it, for a start, that ultimates are in themselves essentially non-experiential phenomena. And they are hardly going out on a limb in endorsing N-E, for it seems to be accepted by the vast majority of all human beings.


       And yet I do not really see how physicalists can leave this commitment unquestioned, if they are remotely realistic in their physicalism, i.e. if they really do subscribe to the defining thesis of realistic physicalism that


RP  experience is real and everything real is physical.


For if they are real physicalists, realistic physicalists, they cannot deny that when you put physical stuff together in the way in which it is put together in brains like ours, it constitutes -- is -- experience like ours; all by itself. All by itself: there is on their own physicalist view nothing else, nothing non-physical, involved. ...


       Is this a possible position? I don’t think so, but one defence goes like this.


Experiential phenomena are emergent phenomena. Consciousness properties, experience properties, are emergent properties of wholly non-conscious, non-experiential phenomena. Physical stuff in itself, in its basic nature, is a wholly non-conscious, non-experiential phenomenon. Nevertheless when parts of it interact in certain ways, experiential phenomena ‘emerge’. Ultimates in themselves are evidently wholly non-conscious, non-experiential phenomena, nevertheless when they interact in certain ways, experiential phenomena ‘emerge’.


Does this conception of emergence make sense? I think that it is very, very hard to understand what it is supposed to involve. I think that it may in fact be incoherent, and that this general way of talking may have acquired an air of plausibility, or at least possibility, simply by having been repeated many times in the face of a seeming mystery.


       In order to discuss it I will take it that any position that combines RP with N-E must invoke some notion of emergence, ... and I will start on familiar ground. Liquidity is often proposed as a shiningly clear example of an emergent property. The facts are clear. Liquidity is not a property of individual H2O molecules. Nor is it a property of the ultimates of which H2O molecules are composed. And yet when you put many H2O molecules together they constitute a liquid, they constitute something liquid (at certain temperatures, at least). So liquidity is an emergent property of certain groups of H2O molecules. ...


       Can we hope to understand the alleged emergence of experiential phenomena from non-experiential phenomena by reference to such examples? I don’t think so. The emergent character of liquidity relative to its non-liquid constituents is indeed shiningly easy to grasp. We can easily make intuitive sense of the idea that certain sorts of molecules are so constituted that they don’t bind together in a tight lattice but slide past or off each other (in accordance with van de Waals molecular interaction properties) in a way that gives rise to -- is -- the phenomenon of liquidity. ... We can say that the phenomenon of liquidity reduces without remainder to shape-size-number-position-mass-motion-charge-etc. phenomena. We can see -- to put it another way -- that the phenomenon of liquidity arises naturally out of, and is indeed wholly dependent on, phenomena that do not in themselves involve liquidity at all. ...


       It is this that causes the problem when it comes to relating the supposedly emergent property of experience to the supposedly wholly non-experiential phenomena from which it supposedly emerges. For it now seems that if experiential phenomena -- colour-experiences, for example -- really are somehow (wholly) dependent on non-experiential phenomena in such a way as to be truly emergent from them, emergent from them and from them alone, then there must be (to quote myself)


a correct way of describing things ... that allows one to relate [the experiential phenomenon of] color-experience, considered just as such, to the nonexperiential phenomena on which it is supposed to depend in such a way that the dependence is as intelligible as the dependence of the liquidity of water on the interaction properties of individual molecules. The alternative, after all, is that this total dependence is not intelligible or explicable in any possible physics, not intelligible or explicable even to God, as it were.


You can get liquidity from non-liquid molecules as easily as you can get a cricket team from eleven things that are not cricket teams. In God’s physics, it would have to be just as plain how you get experiential phenomena from wholly non-experiential phenomena. But this is what boggles the mind -- or should do. ...


       Here I resort to the argumentum ad visceris. My gut tells me that ... experiential phenomena cannot be emergent properties of wholly non-experiential phenomena. If it really is true that Y is emergent from X then -- to repeat -- it must be the case that Y is in some sense wholly dependent on X, and it is, I take it, built into the heart of the notion of emergence that emergence cannot be brute in the sense of having, objectively, no explanation, no reason behind it or to it. ... It cannot be spooky, to use another term of art. Spooky emergence is an empty rattle, a philosopher’s game. ... for Y truly to emerge from X is for Y to arise from or out of X given how X is. It is for it to arise out of X in some sort of essentially and indeed wholly non-arbitrary way. That’s how liquidity arises out of non-liquid phenomena. There must be some sort of rich and indeed full connection between how Y is and how X is. The arising must be potentially developmentally intelligible, potentially explicable. Emergence cannot be brute. ...


       Some may still protest that they find nothing intolerable in the idea that [experiential phenomena can be emergent properties of something wholly non-experiential] ... because these non-experiential phenomena are intrinsically suited to constituting experiential phenomena in certain situations, and are ‘proto-experiential’ in that sense, although ultimately non-experiential in themselves.


       ... Clearly -- necessarily -- for X to be intrinsically suited to or for Y is for there to be something about X’s nature in virtue of which X is so suited. So ‘intrinsically suited’ entails possible intelligibility, intelligibility in principle -- although we do not really need to bring in the quasi-epistemological notion of intelligibility. For if potential Y-constitution is part of X’s nature then it is necessarily present for inspection -- present for inspection in principle at least, graspable at least by God -- in the very existence of X, X with the nature that it has.[3]


            Nagel similarly suggests that weak panpsychism without strong panpsychism may not be intelligible. “It is difficult to imagine how a chain of explanatory inference could ever get from the mental states of whole animals back to the proto-mental properties of dead matter. It is a kind of breakdown we cannot imagine, perhaps it is unintelligible.” (Nagel 1979, p. 194) But perhaps Aristotle was the first to anticipate Ultimate Subjecthood. In a notoriously cryptic stretch of De Anima he remarks, “Actual knowledge is identical with its object: potential knowledge in the individual is in time prior to actual knowledge but absolutely it has no priority even in time; for all things that come into being arise from what actually is.” (Aristotle, De Anima III.7, 431a1-3, repeating and amplifying III.5, 430a20-22) We may detect in all three philosophers the same conviction: the psychological cannot arise from a potential toward it that is itself wholly nonpsychological.


            20. I believe that the original form of this conviction concerns, not experience or knowledge or even mental states in general, but subjecthood itself, so I shall formulate the thesis of Ultimate Subjecthood which I wish to examine here in terms of the idea of a subject. On this thesis the capacity to help constitute a subject can be present only in a substance that is itself a subject: a being’s subjecthood must reflect that of the ultimate entities that help constitute it a subject.


            Moreover, on this thesis subjecthood in a subject S must “arise naturally out of” subjecthood in the ultimate entities that help constitute S. It is in that way that S’s subjecthood must be “intelligible” or “explicable.” The real requirement is not epistemic, but metaphysical: it is not that we must be able to understand or explain how S’s subjecthood depends on that of ultimate entities, but that S’s subjecthood must arise naturally out of theirs. The intelligibility of an explanation of how this is so is simply a test of whether that explanation adequately describes the dependence.


            21. Ultimate Subjecthood, then, as I shall understand it, is the thesis that an ultimate entity E capable of helping to constitute a subject must itself be a subject, in such a way that its capacity to help constitute a subject derives from its subjecthood -- “arises from” it, as Aristotle puts it. I shall express this requirement by saying that E’s subjecthood must “contribute to” the subjecthood of any subject it helps constitute. What sort of subjecthood could this be, and what sort of constitution? As we shall see in Section IV, the requirement has far-reaching consequences -- consequences which might seem so strong as to reduce it to absurdity. I do not accept this conclusion.


IV. The Mereological Irreducibility of Subjecthood


            22. William James presents an argument whose conclusion seems to threaten Ultimate Subjecthood. James formulates his point in terms of thoughts and feelings:


Take a sentence of a dozen words, and take twelve men and tell to each one word. Then stand the men in a row or jam them in a bunch, and let each think of his word as intently as he will; nowhere will there be a consciousness of the whole sentence ... Where the elemental units are supposed to be feelings, the case is in no wise altered. Take a hundred of them, shuffle them and pack them as close together as you can (whatever that might mean); still each remains the same feeling it always was, shut in its own skin, windowless, ignorant of what the other feelings are and mean. There would be a hundred-and-first feeling there, if, when a group or series of such feelings were set up, a consciousness belonging to the group as such should emerge. And this 101st feeling would be a totally new fact; the 100 original feelings might, by a curious physical law, be a signal for its creation, when they came together; but they would have no substantial identity with it, nor it with them, and one could never deduce the one from the others, or (in any intelligible sense) say that they evolved it.(James 1890, p. 160)


But again I believe that the original form of the point concerns subjects: subjecthood in the parts of a substance cannot contribute to subjecthood in the substance itself. The instance we care about concerns the relation between subjects and the ultimate entities that help constitute them -- whose subjecthood, if Ultimate Subjecthood is correct, must contribute to theirs.


            Integral Subjecthood, then, is the thesis that the capacity of an ultimate entity E to help constitute a subject of which E is a proper part cannot arise naturally from E’s subjecthood. On this thesis, subjecthood is integral to a subject: it cannot arise from subjecthood in parts that help constitute the subject.


            23. The point constitutes a familiar objection to homuncular philosophy of mind, which can be traced back to Kant and forward to Nagel.[4] Kant expresses it in terms of thought: if you teach each member of a regiment one word of a poem, this does nothing to endow the regiment itself with a thought of the entire poem. Transposing Kant’s point into the terms of subjecthood, a soldier’s subjecthood contributes nothing toward subjecthood in the soldier’s regiment. Applying this point to the case we care about, subjecthood in an ultimate entity that is part of a soldier contributes nothing to the soldier’s subjecthood.


            24. In Book IV of the Republic Plato presents the archetypal homuncularism that runs counter to the spirit of Integral Subjecthood. Here Socrates says that a human being is wise because the rational part is wise and rules; courageous because the spirited part preserves the declarations of the rational part about what is to be feared; moderate because all the parts believe that the rational part should rule; and just because each part does its proper work. (Plato, Republic 441c-444e) As Bruce Aune points out in correspondence, Plato conceives these virtues as mental traits, and thus holds that sometimes a human being’s mental state arises from its parts’ mental states. According to James, thinking and feeling never arise from the parts’ thoughts or feelings, and according to the generalized version of Integral Subjecthood, subjecthood never arises from subjecthood in the parts.


            Plato’s picture provides a good place to begin considering whether such homuncular theories are credible. We need to consider how subjecthood and mental states might arise from the parts of Platonic human beings (or of Platonic souls). Notice that Plato does not hold that a human being desires something just when one of the parts does; he holds that the parts often desire opposite things but that the whole cannot. Then when do the mental states of the parts generate mental states in the whole? And how do the mental states of the parts generate a mental state in the whole when they do? How can subjecthood in a plurality of parts of a substance contribute to subjecthood in the whole? If appetite says to act and reason says not to act, and the deed is done, what makes anything the action of the human being? Is it just to punish the human being, or only its parts?


            Plato supports his picture with an analogous picture of the state and its parts. There may be much that is attractive in the analogy, but it fails us here: Plato does not suppose the state to be a subject or to have mental states. How then can we understand the claim that when three subjects have certain beliefs or do certain kinds of work, there arises in a subject composed of them a new mental state identifiable with what we know as moderation or justice? Plato offers no account either of how subjecthood in the three gives rise to subjecthood in a whole composed of them or of how a new mental state, owned by that new subject, arises out of the old ones. We may begin to suspect that no intelligible account is possible. We begin to suspect that insofar as we are serious about parts as subjects, we should cease to be serious about the whole as a subject whose subjecthood arises from theirs.


            Such reflections are the beginnings of a case for James against Plato. The question how good a case can be made requires much more machete work. But these few reflections may begin to suggest that Integral Subjecthood is correct. Let us suppose that it is.


            25. The theoretical hope is that Ultimate Subjecthood will help us understand the nature of the power ultimate entities have to help constitute familiar subjects, but the notion that these subjects have parts that are also subjects seems to lead nowhere. There seems no way to close the metaphysical gap this notion leaves open. Ultimate Subjecthood requires that the subjecthood of an ultimate entity that helps constitute S a subject actually contribute to S’s subjecthood. Integral Subjecthood implies that subjecthood in an ultimate entity that is a part of S could not make that contribution. James’s suggestion that it could not is the start of an answer to the question we raised in 21; more will emerge.


            James in effect observes that even if a part P of S is a subject, and even if by a “curious” law of nature P’s subjecthood is a “signal” for the creation of subjecthood in S -- from which it would follow that if P is an ultimate entity, P helps constitute S a subject --, that is not enough for P’s subjecthood to make the contribution to S’s that Ultimate Subjecthood requires.


            James is surely right, but his observation masks the really startling news. We have been tacitly supposing that the ultimate entities that help constitute familiar subjects are (proper) parts of them. James’s news is that if Ultimate Subjecthood holds, an ultimate entity E that helps constitute S a subject cannot be a part of S.


            For Ultimate Subjecthood implies that if E helps constitute S, then E’s subjecthood must contribute to S’s subjecthood, and Integral Subjecthood implies that if E is a part of S, then E’s subjecthood cannot contribute to S’s. Together they imply that if E helps constitute S then E cannot be a part of S. Therefore the relation between an ultimate entity and a subject it helps constitute cannot be anything like that of part to whole.[5]


            26. It might be urged that familiar subjects are parts of some ultimate entity making them up, rather than the other way around -- perhaps condensations of space-time, or modifications of some cosmic field. But in just the same way, an ultimate entity’s subjecthood cannot contribute to subjecthood in its parts, and so an ultimate entity cannot help constitute its parts subjects.


            The startling news, then, is that if an ultimate entity E helps constitute S a subject, there cannot be anything like a part-whole relation between E and S.


V. Monism


            27. Must we choose between Ultimate Subjecthood and Integral Subjecthood? My question is whether it is possible to reconcile them. Let us ask what mind and nature would have to be like for reconciliation to be possible. How could these things each be true?


(1)        Any ultimate entity that helps constitute a familiar subject like me is itself a subject.


(2)        Ultimate entities and familiar subjects do not stand to one another in anything like a part-whole relationship.


In his monadological account of ultimate entities and familiar subjects, Leibniz accepts both,[6] and that suggests that reconciliation might be possible.


            28. We have tacitly taken seriously two levels, that of animals and that of the ultimately tiny. The problem is insoluble if we confine ourselves to these levels. (a) Tiny ultimate entities are something like parts of animals. (b) The quasi-Leibnizian thesis that familiar subjects are also tiny is untenable. There is no tiny entity with which a familiar subject like me can plausibly be identified. The contribution of every tiny entity to my subjecthood is trivial and dispensable, as the sound made by any single drop of water is dispensable to the sound of the ocean wave. Familiar subjects seem to be at least as large as animal bodies.[7] (c) Finally, the thesis that animals are ultimate entities is unbelievable in the light of modern physics.[8]


            29. Modern physics suggests opening up a new level, that of the ultimately large -- a level at which there can be no more than one substance. It allows the possibility that all of nature is made up of a single ultimate entity, which might be understood as a field, of which tiny entities are condensations or abstractions, or as space-time itself. I suggest that the problem can be solved only on this level. (The Appendix aims to display some teeth behind the suggestion.)


            30. The solution is this.


(3) There is only one ultimate entity, and it makes up everything in nature.


(4) There is only one subject, and this subject has every mental state.


(5) That ultimate entity is identical with this subject.


Spinoza accepts all three theses, and that suggests that this solution is worth taking seriously. We may call it Monism. Monism answers both of 14’s cardinality questions, and even the cardinality question concerning the set whose members are all the ultimate entities and all the subjects, with a single answer: ‘One’.


            31. Monism respects both Ultimate Subjecthood and Integral Subjecthood, since it entails both (1) and (2). It respects the familiar reasons for accepting Ultimate Supervenience. All of that together counts in its favor.


            Monism turns from Leibniz to Spinoza. It is consistent with Spinoza’s metaphysics[9] and also with Advaita Vedantist metaphysics (whose ‘tat tvam asi’ (4), and 43 and 50 below, mirror),[10] though that in itself counts neither for nor against it.


            Monism lets us speak more simply. For example, instead of saying that E helps constitute S we may always say that E constitutes S, and we may take terms like ‘E’ and ‘S’ as variables or names as we like, since as variables they range over a single substance.


VI. The Capacity to Help Constitute a Subject


            32. We began this line of thought, in 13, with the thesis that each ultimate entity has the property of being capable of helping to constitute any possible kind of subject, and we asked in 14 what the nature of this property is. We can take some steps toward an answer. The first do not presuppose Monism and so gain in generality; the final step takes Monism into account.


            33. Consider any subject S. We observed in 11 that at least some of the ultimate entities that help constitute S a subject will be among those that make S up. Let E be one of these. Given Integral Subjecthood, E does not stand in anything like a part-whole relation to S. Given Ultimate Subjecthood, E is itself a subject. E, it seems, can only be S itself. Then the states of E by themselves determine that S is a subject, in the trivial sense marked off in 10, so by 11 no other ultimate entity E* helps constitute S a subject. We may conclude that E helps constitute S a subject if and only if E = S.


            34. But that in turn leaves no room for any other E* to help constitute S a subject. The capacity to help constitute subjects is not the capacity to constitute subjects of other substances: an ultimate entity E has the capacity to help constitute S a subject only if E = S. Hence for E to be capable of helping to constitute S a subject is for E by its nature actually to constitute itself a subject -- namely S. It is for E by its nature to be a subject -- namely S.


            35. It follows that there can be no unrealized “protosubjecthood” -- no capacity in a nonsubject to become a subject. For suppose that N is a nonsubject capable of becoming a subject. A corollary of 33 is that every subject is an ultimate entity. Then in order for N to be capable of becoming a subject it must be an ultimate entity, with the capacity to help constitute itself a subject. But then by Ultimate Subjecthood it must be a subject, contrary to the supposition.


            36. Finally, given monist theory, to say that E has the property of  being capable of helping to constitute any possible kind of subject is theoretically equivalent to saying that it has the property of being a subject that is capable of assuming every kind of subjecthood that can be found in nature. It is the latter property from which the former is theoretically derivative.


VII. Explaining Subjecthood


            37. Monist theory begins to answer questions about how subjecthood might be determined, how it might be caused, and generally to what extent it might be explained. I leave detailed derivations of the answers here as machete exercises for the reader.


            First, how might the subjecthood of an ultimate entity that helps constitute S help to determine or explain S’s subjecthood? -- Well, if E helps constitute S, E’s subjecthood cannot explain S’s, since the two are the same. E’s subjecthood determines S’s only in the trivial reflexive sense.


            38. Next, how might S’s nature determine or explain its subjecthood? -- Under the head of S's nature, we may consider (a) its psychological properties, (b) its nonpsychological properties, and (c) relations between the two. First, the fact that S has (a) some psychological property P cannot explain the fact that S is a subject. For example, the fact that S has a belief or a desire already implies that S is a subject, and so cannot explain it. Nor can the fact that S has P nontrivially determine that S is a subject, since S must possess the metaphysical status of subject in order to have any other psychological property at all.


            In case S’s possession of some nonpsychological property N determines that it is a subject, it may be possible to propose an explanation of S's subjecthood in terms of (b) and (c): S has N, and by a law of nature whatever has N is a subject. But now what calls for explanation is that law. Why in the nature of things should the possession of N lead to subjecthood?


            Recall that S is our present universe. Consider a Twin Universe S* which is not a subject but is similar to S in every respect except that (i) S* has no psychological properties and (ii) no law that whatever has N is a subject holds of S*.[11] Though the natural laws governing S* differ from those governing S, the notion of such an S* is perfectly intelligible. (The intelligibility of a mindless automaton that is similar in just this way to a subject is a necessary condition of the intelligibility of the classical problem of other minds.) Since the fact that S* has N does not determine that S* is a subject, we face a new request for explanation. Why does the fact that S has N determine that S is a subject, while the fact that S* has N leaves it a nonsubject?


            39. Finally, how might S’s subjecthood originate? -- Since there is no protosubjecthood, this reduces to the question what might cause the subjecthood of a substance at the moment it comes into existence. S may have always existed. But if there is a moment t when S begins to exist, its subjecthood then must be caused either by nothing at all or by states of previous ultimate entities, which must all cease to exist at t.


            Suppose that states of E1, E2, ..., En at t cause a new subject to come into being, and take the list of Ei to be minimal in the style of 11. We cannot say that E1 helps constitute S a subject at t, since 11 construes constitution as determination of subjecthood over an interval. But we may say by analogy that E1 “helps cause” S to come to be at that moment.


            E1 has the following remarkable property: it is capable, when properly related to other ultimate entities, of helping to cause a subject to come into being. And then a plausible analog of Ultimate Subjecthood implies that it must itself be a subject. (But then does some plausible analog of Integral Subjecthood block E1 from helping to cause a new subject to come to be?) In case there is at least one subject that is not tiny, Monism will govern E1’s universe, and E1 will be its only subject (see Appendix). Analogy with the present universe might also suggest this consequence. But if other ultimate entities than E1 help cause S to come to be, they will each be subjects.


            If E1 acts alone, then at t it has some property P such that by a law of nature whenever a substance has P a new subject comes into being. (If E1 does not act alone, the Ei have properties and relations to the same effect.) Then S’s subjecthood will be caused by the fact that E1 has P, and it will be explained by that fact plus that law of nature. Again it is that law that now calls for explanation. Why in the nature of things should a subject be extinguished and replaced by a new subject upon acquiring P?


            40. Suppose now that the states of a single subject cause S to come to be at t. We may be imagining t as a moment in the distant past. But consider now whether t might be very recent, for example a couple of minutes ago as you were reading the previous page. In the first-edition version of the Third Paralogism, Kant asks whether it is possible to know whether a subject persists through a given moment or instead ceases to exist then and is replaced by a similar subject, and he argues that “we cannot even judge from our own consciousness whether as soul we are persisting or not ....” (Kant, Critique of Pure Reason, A361-66) Monist theory suggests that whenever something causes a new subject to come to be, it is an old subject that is passing away. Kant poses for Monism (though not for Monism alone) the question whether there is any constraint on how often this sort of thing might take place.


            41. According to Monism there is exactly one subject. But it is equally intelligible that this subject has always existed and that it has come into being. It is equally intelligible that its coming to be is caused by the states of prior substances and that it is not; and that it has come to be recently or long ago. And it is equally intelligible that a substance’s subjecthood should be determined by one of its nonpsychological states and that it should not. It may be that not even a god who knew everything about the nature of things could explain why such things are as they are. It may be, once again as Spinoza holds, that there is no explaining the connection between Thought and Extension and no explaining Thought itself.


VIII. Some Implications of Monism


            42. Monism finds in the facts of mind strong evidence about where physics should look for ultimate entities: not in the very small, but in the very large. Already the panpsychisms considered by Nagel and Strawson had found in the same facts evidence about the nature of ultimate entities. Monism accepts, along with these two philosophers, the anti-Cartesian proposition that the fact that nature generates mind can tell us important things about nature.


            43. Monism has equally strong implications for our understanding of mind. For example, if we understand by ‘you’ the subject who has those mental states and by ‘I’ the subject who has these ones, I = you. As things are, probably only a few dozen people take themselves to be President of the US; most of us believe we are not. But given Leibniz’s Law, Monism implies that either everyone is or no one is. Probably the second disjunct is the most sensible option for a monist to take. Everyone who believes himself or herself to be President will be convicted of some kind of insanity. Monism agrees with the Vedantist view that my usual understanding of myself, as a subject linked to this body in particular, is radically misguided. Instead, I am a much larger and longer-lived subject than I used to think I was. Longer lived, since as long as any being capable of mental states remains, no one really dies.


            44. Monism consists of three very strong theses and must be tested in the light of their implications. Thesis (5), the identity of subject and ultimate entity of nature, is confirmed by all the evidence for Ultimate Supervenience that has accumulated since Descartes.


            45. Thesis (3), the thesis that there is only a single ultimate entity in nature, is a hypothesis in physics which I am hardly competent to discuss. But I suggest that we will find evidence for it whenever we find evidence that the tiny things which physicists now deal with are more like abstractions than actual substances. The properties of quarks and the like are said to be incomprehensible, or comprehensible only in the sense that the mathematics gives the right answers. But perhaps part of what generates the incomprehensibility is the idea that quarks and so on are substances, rather than condensations, abstractions, or the like, from some genuine substance. If that is true, Monism may remove a roadblock to comprehension.


            46. If physics one day uncovers grounds for (3), they might also be grounds for (4) (though they might not be). The grounds that imply that animals are made up of a single ultimate entity might also imply that we should cease to attribute mental states to those animals, with their particular sense organs, brain states, and bodily movements, and attribute them instead to that entity. They might imply that we should cease to attribute actions to animals, and take that ultimate entity to be the agent actually responsible for them.


            47. Thesis (4), the thesis that there is only a single subject, is one which we are all richly competent to discuss. It gives rise to a fertile research project: how to show the consistency of the familiar facts of mind with the thesis that your mental states and mine -- the states mainly determined by states of that body and those mainly determined by states of this body -- are states of a single subject.


            48. That thesis is so radical that even the questions raised by the research project will need careful reformulating. For we immediately want to ask: Why is it so hard for me to learn about your mental states, if I am you? If you believe that p and desire x, and I am sure that not-p and abhor x, what sort of chaos must be attributed to my/your/our/its mind?


            49. Chaos of these kinds does not seem conceptually problematic, since we already sensibly enough take each of them to be present in our own minds. Socrates’ very mission gets its bite from the ways in which every Athenian harbors internal hiddenness, internal inconsistency. “Do I contradict myself? Very well then .... I contradict myself; I am large .... I contain multitudes.”[12]


            50. Monism must allow a kind of modularity in the mind of the universe. (a) First, the systematicity and rationality among its mental states seem to be imposed primarily, though not exclusively and not especially forcefully, within certain groups, each group corresponding to what we customarily call a mind, each group causally linked with occurrences in a particular animal body. (b) Secondly, there is a notable lack of access within one group to the mental states in another group. The universe, it appears, has a rather special form of Multiple Personality Disorder. It also appears to have delusions of lack of grandeur. Every now and then the delusions begin to lift, as they did for Mevlana Jelaluddin Rumi.


I am dust particles in sunlight. I am the round sun. To the bits of dust I say, Stay.

To the sun, Keep moving.


I am morning mist, and the breathing of evening. I am wind in the top of a grove,

and surf on the cliff. Mast, rudder, helmsman, and keel; I am also the coral reef they founder on.


I am a tree with a trained parrot in its branches. Silence, thought, and voice. The musical air coming through a flute, a spark of stone, a flickering in metal. Both candle, and the moth crazy around it. Rose, and the nightingale lost in the fragrance.


I am all orders of being, the circling galaxy, the evolutionary intelligence, the lift, and the falling away. What is, and what isn’t. You who know Jelaluddin, You the one in all, say who I am. Say I am You.[13]


            51. We find evidence for thesis (4) whenever we find evidence that animals, including the human ones, are not really subjects and that subjects construed as animals are more like abstractions than real things. Evidence of this kind can be found in Nagel 1971 and in Part III of Parfit 1987, which present cases (of brain bisection and other kinds of fission, replication, and disruption of psychological and physical connectedness) in which there seems to be no way to answer the question which subject, so construed, has a certain mental state. The question to ask is what such cases are really evidence for.


            52. Nagel expresses additional qualms about attributing mental states to animals. “It seems absurd to try to discover the basis of the point of view of the person in an atomistic breakdown of the organism, because that object is not a possible subject for the point of view to which the person’s experiences appear. ... I simply record this feeling of impossibility because I have no more to say about it. When a mouse is frightened it does not seem to me that a small material object is frightened.” (Nagel 1979, p. 189) Nagel is no happier attributing mental states to souls, or espousing a Wittgensteinian denial that there is anything of which they are states. “Where does this leave us? I have now expressed dissatisfaction with three alternative interpretations of mental states: that they are states of the body, that they are states of the soul, and that all we can say about their essence is to give criteria or conditions for their ascription. But what is left? If they are real states of something in the world, if they depend on what is going on in the creature’s body, if they are intimately connected with stimuli and behavior, and if the creature does not consist of a body plus something else, what can experience be but states of the organism?” (p. 193) I am not clear about the grounds of Nagel’s dissatisfaction. Perhaps they are also the source of Wittgenstein’s question -- “Where in the world is a metaphysical self to be found?” (Wittgenstein 1922, 5.633) But the best therapy might be Monism: Mental states are real states of something, all right. But they are not states of something in the world. They are states of the world itself.


            53. Monism may contain seeds for resolving questions in ethics and philosophical psychology. Questions concerning Egoism and Altruism will have to be reformulated along with those mentioned in 48, and the reformulation may help answer them. If I am you, my interest is not distinct from yours. The familiar dubious and inefficacious ethical arguments that everyone should be nice may not be necessary.


IX. Monism and God


            54. Monism suggests considering the hypothesis that the single subject is God. For one thing, Monism entails a significant step toward the longevity traditionally attributed to God. Since the ultimate entity is incapable of mere protosubjecthood, it has been a subject as long as it has existed. Subjecthood may in fact be an essential property of subjects, in which case the ultimate entity will necessarily be a subject as long as it exists.[14] Both Spinoza and Vedantists accept the hypothesis, though again that does not count either way. The result would be not simply a monist form of panpsychism but a monist form of pantheism. The monist research program ought to give special attention to this hypothesis, though Monism is not committed either to theism or to pantheism.


            55. Monist pantheism seems to contain the seeds for a significant attack on the Problem of Evil, since it would be God, not anyone else, who undergoes all the suffering.


            56. Still, Monism must take seriously ignorance, error, suffering, and imperfection, if not actual injustice, and it seems actually to exacerbate the degree of mental chaos in the world. It will be debatable whether those attributes can be squared with the attributes proper to a god. Might God be a deceiver, if the deception is only self-deception? Might God be at play in a cosmic game, as the Vedantic myth suggests?


            57. Mind and nature sit in the center of a circle, its periphery made up of people like you and me at work on ways to understand them. The ways exemplified in the paper and in your responses to it may yield some increment of understanding. There are certainly better ways, some already found, some still to be found. Reflexive observation of our own philosophical work is one of them. It may be relevant that our work has an aspect of play to it. It may be relevant that our minds can meet as we work at understanding. For we are not only at work on the periphery of the circle. In some way we do not yet understand, we are also at rest in the center.


Appendix. The argument for Monism


            There are two kinds of premises. The first govern a basic notion in philosophy of mind, that of a subject; the second govern a basic notion in what we might call philosophy of nature, that of an ultimate entity. The first group of premises contains most of our work and matters most here, so the paper falls within the philosophy of mind.


            Ultimate Supervenience (US), the principle that every state of a substance is determined by states of ultimate entities, is involved in the argument only via its corollary M1. Since it is so much stronger than M1, I will regard M1 rather than US as a premise here.


            I have not tried to characterize the idea of an ultimate entity, but I have assumed that US and M1, as well as N1, N4, and N5, govern it. We might consider defining ultimate entities as the smallest class of substances whose states determine the states of all the substances. That would render US, a condition that has often been held to govern ultimate entities, true by definition.


            The suggestion does not come for free. First, this is really not what is normally meant by an ultimate entity. Then there may be technical obstacles: given all the problems about overdetermination, the definition may not pick out a unique class of substances. Most importantly for us, in case there is a lot of indeterminacy in nature, the class of ultimate entities will balloon. Conceivably every substance could qualify. Even if things are not that indeterminate, animals might qualify, in which case we would have bought US at the price of N4. All things considered, I think it is best to retain a more traditional conception of an ultimate entity and regard US and M1 as substantive theses.


            But in that case the argument might be thought to reduce to absurdity the thesis that there is a class of substances governed by M1, N1, N4, and N5 (whether or not they qualify as traditional ultimate entities). Or is some other assumption to blame? But wait a minute! The really interesting question is whether the argument should be regarded as a reductio at all.


I. Mind: Principles concerning the constitution of subjects


M1       Ultimate Supervenience of Subjecthood. For every subject there are ultimate entities that help constitute it a subject. [from 11]


M2       Ultimate Subjecthood. An ultimate entity capable of helping to constitute a subject must itself be a subject, in such a way that its capacity to help constitute a subject arises naturally from its subjecthood. [from 21]


M3       Integral Subjecthood. The capacity of an ultimate entity E to help constitute a subject of which E is a proper part cannot arise naturally from E’s subjecthood. [from 22]


M4       Analog of Integral Subjecthood. The capacity of an ultimate entity E to help constitute a proper part of E a subject cannot arise naturally from E’s subjecthood. [from 26]


II. Nature: Principles concerning size and parthood


            By the end of 26 the substantive philosophy of mind has been done. I believe it is more reasonable to grasp the inference to Monism in the way 28 presents it than to look for a formal derivation. The premises of a derivation, which are likely to require restriction, and will be stronger than the derivation requires, may lend the inference a misleadingly doubtful air. Still, a derivation is not hard to create. Each of the seven premises I record below seems sensible or sensibly modifiable. A full-scale attack on any of them is apt to entail janitorial and restorative work leading to a metaphysics at least as bizarre as Monism itself, though some such variants may be worth investigating. Alternative sets of metaphysical premises yielding Monism are also conceivable. But this is one way to get from here to there.


Principles concerning tiny and animal-sized substances [from 28]


N1       If some ultimate entities are tiny then all ultimate entities are tiny.


N2       If all ultimate entities are tiny and some subjects are animals, and if for every subject there are ultimate entities that help constitute it a subject, then some ultimate entities that help constitute subjects are proper parts of them.


N3       Some subjects are not tiny.


N4       No animals are ultimate entities.


Principles concerning maximally large substances (Although N7 entails N6, N6 is what makes N7 plausible, and they do different kinds of work, so I mention them both.) [from 30]


N5       If an ultimate entity or a subject is not tiny or an animal it is maximally large.


N6       There is at most one maximally large substance.


N7       For any maximally large substance S, every other substance is a proper part of S.


The argument


            Suppose some ultimate entities are tiny. Then by N1 all ultimate entities are tiny. Suppose some subjects are animals. By M1, for every subject there are ultimate entities that help constitute it a subject. Then by N2 some ultimate entities that help constitute subjects are proper parts of them. By M2 they must themselves be subjects, in such a way that their capacity to help constitute subjects arises naturally from their subjecthood, but then by M3 they cannot be proper parts of any of the subjects they help constitute. So if some ultimate entities are tiny then no subjects are animals, and then by N3 some subjects are neither tiny nor animals, so that by N5 they are maximally large. Again if some ultimate entities are tiny, so that by N1 all ultimate entities are tiny, then by M1 some of them help constitute the maximally large subjects. But by N7 any tiny ultimate entities that help constitute any maximally large subject must be proper parts of it, and by M2 and M3 as before this is impossible. So no ultimate entities are tiny, and by N4 and N5 they are all maximally large. Then by N6 there is just one ultimate entity, and by M1 it helps constitute every subject on its own. More simply, that ultimate entity by itself constitutes every subject.


            Now if any subject is not maximally large, by N7 it is a proper part of the ultimate entity that constitutes it. But by M2 the capacity of that entity to constitute a subject arises naturally from its own subjecthood, and by M4 no subject so constituted can be proper part of it . Therefore every subject is maximally large. Since every ultimate entity and every subject is maximally large, by N6 there is only one of each and they are identical. This is Monism.[15]





1. Robert Frost, “The Secret Sits,” in The Complete Poems of Robert Frost (New York: Henry Holt and Company, 1949), p. 495.


2. This is something that philosophers as diverse as Aristotle in De Anima, Descartes in the Meditations, Leibniz in the New System and the Monadology, and Kant in the Critiques also accepted. It is something that Wittgenstein, under the spell of Moore’s piecemeal method and the Vienna Circle’s antipathy to metaphysics, and then Ryle, under the spell of the tradition stretching from Moore to Wittgenstein, began to bypass. The question whether subjecthood is baby or bath water seems to me to be one of the primary questions in the philosophy of mind. One of the primary metaquestions is surely why no term for subjects has become entrenched either in ordinary language or in philosophical discourse.


3. Strawson forthcoming.. The included quotation is from Strawson 1994, p. 69. I’ve tinkered with his ‘X’ and ‘Y’ for the sake of consistency. Though I have sought to present the core of Strawson’s thinking, it really is necessary to work through the entire article, which he describes as “a studied stumbling in a small maze of linked notions, not a linear argument.”


4. “Every composite substance is an aggregate of many, and the action of a composite, or of that which inheres in it as such a composite, is an aggregate of many actions or accidents, which is distributed among the multitude of substances. Now of course an effect that arises from the concurrence of many acting substances is possible if this effect is merely external (as, e.g., the movement of a body is the united movement of all its parts). Yet with thoughts, as accidents belonging inwardly to a thinking being, it is otherwise. For suppose that the composite were thinking; then every part of it would be a part of the thought, but the parts would first contain the whole thought only when taken together. Now this would be contradictory. For because the representations that are divided among different beings (e.g., the individual words of a verse) never constitute a whole thought (a verse), the thought can never inhere in a composite as such. Thus it is possible only in one substance, which is not an aggregate of many, and hence it is absolutely simple.” (Kant, Critique of Pure Reason, A351-52; my italics. Kant immediately rejects this argument for the simplicity of the soul, but he does not criticize the premise I have italicized.) See also Nagel 1979, p. 194.


5. I say ‘anything like’ to take account of Gordon Brittan’s argument that a part-whole relation requires a common metric, whereas the substances whose nature physics aims to explain seem increasingly not to share measurable properties with ultimate entities. Brittan concludes that the relation between those entities and those substances may be only analogous to that of part to whole. (Brittan 1970)


6. See for example Monadology 16-19 in Leibniz 1989, p. 215.


7. There is no clash between 11 and 28. In order to consider the reasoning in 28b, imagine that E1 is one of a host of tiny ultimate entities E1, ..., Em in S’s brain each of which helps constitute S a subject, and imagine in accordance with 11 that the states of E1, E2, ..., En determine that S is a subject while the states of E2, ..., En do not. Relative to E2, ..., En, the role of E1 is certainly not trivial. But it is easy to conjecture that n will be much smaller than m, that the states of E2, ..., Em will with no help from E1 determine that S is a subject, and that relative to the entire collection E2, ..., Em of constitutive ultimate entities other than E1, the role of E1 will indeed be trivial and dispensable. (Imagine a case where m = 7 and n = 3: S’s brain contains seven ultimate entities each helping to constitute S a subject, because the states of any three of E1, ..., E7 determine that S is a subject, though the states of no two suffice. In the presence of E2, ..., E7, the role of E1 in constituting S a subject, though real, is thoroughly dispensable.) By contrast, the quasi-Leibnizian thesis would require not simply that the role of E1 be nontrivial in absolute terms but that m be 1. Leibniz’s view that a single ultimate entity can be distinguished from the infinity of entities making up an animal body, as the subject governing that body, is an unfortunate Cartesian legacy. As we are about to see, m can be 1 only if Em is not tiny.


            I say ‘quasi-Leibnizian’ here because although Leibniz takes both me and ultimate entities to be monads, he does not think monads possess tininess or any other nonpsychological property. “Indeed, considering the matter carefully, we must say that there is nothing in things but simple substances, and in them, perception and appetite.” (Leibniz to De Volder, 20 June 1703, in Leibniz 1989, p. 181) Roderick M. Chisholm defends the kind of quasi-Leibnizian thesis I have in mind. He maintains that I am a proper part of this human body, probably a microscopic part of the brain. (Chisholm 1978, 1986) Are there quasi-Leibnizians today who would agree?


8. Are there Aristotelians today who would disagree?


9. Ethics II, corollary to Proposition 11. See Wilson 1980, pp. 104-110, and Donagan 1988, pp. 125-27.


10. See Deutsch 1969, esp. chs. 1 and 4.


11. John Perry has shown, in discussing “zombie” versions of the problem of other minds, that it would not be coherent simply to imagine an S* that is not a subject but is similar to S in every nonpsychological respect. Hence the need for clause (ii). See Perry 2001, pp. 72-77.


12. Walt Whitman, Leaves of Grass: The First Edition 1855, edited by Malcolm Cowley (New York: Barnes & Noble, 1997), p. 85.


13. Maulana Jal­al al-Din Rumi, “Say I am You,” in The Essential Rumi, translated by Coleman Barks (San Francisco: HarperSanFrancisco, 1996), pp. 275-76.


14. Does the ultimate entity always exist? Is it the only ultimate entity ever to exist? The Third Paralogism is a good place to begin investigating such liminal temporal questions; see 40 above.


15. I am grateful to Ayça Boylu, Simon Bowes, Robert Nola, Adriano Palma, Sun Demurely, Justin Smith, Keith Peterson, Melis Erdur, Abraham Anderson, Vere Chappell, İlhan İnan, and especially Harry Lewis, Bruce Aune, Tomis Kapitan, and Galen Strawson.





Aristotle, 1984. De Anima. Translated by J. A. Smith. In The Complete Works of Aristotle, vol. 2. Edited by Jonathan Barnes. Princeton: Princeton University Press.


Brittan, Gordon, 1970. “Explanation and Reduction.” Journal of Philosophy  67:446-57.


Chisholm, Roderick M., 1978. “Is There a Mind-Body Problem?” The Philosophic Exchange 2:25-34.


Chisholm, Roderick M., 1986. “Self Profile.” In Roderick M. Chisholm. Edited by Radu J. Bogdan. Dordrecht: D. Reidel Publishing Company.


Deutsch, Eliot, 1969. Advaita Vedanta. Honolulu: University of Hawaii Press.


Donagan, Alan, 1988. Spinoza. Chicago: University of Chicago Press.


James, William, 1890/1950. The Principles of Psychology, vol. 1. New York: Henry Holt and Co., reprinted New York: Dover.


Kant, Immanuel, 1781/1997. Critique of Pure Reason. Translated by Paul Guyer and Allen W. Wood. Cambridge: Cambridge University Press.


Leibniz, G. W, 1989. Philosophical Essays. Translated by Roger Ariew and Daniel Garber. Indianapolis: Hackett Publishing Company.


Nagel, Thomas, 1971. “Brain Bisection and the Unity of Consciousness.” Synthese 22:396-413.


Nagel, Thomas, 1979. “Panpsychism.” In Mortal Questions. Cambridge: Cambridge University Press.


Parfit, Derek, 1984, reprinted with corrections 1987. Reasons and Persons. Oxford: Oxford University Press.


Perry, John, 2001. Knowledge, Possibility, and Consciousness. Cambridge, MA: MIT Press.


Plato, 1997. Republic. Translated by G. M. A. Grube, revised by C. D. C. Reeve. In Plato, Complete Works. Edited by John M. Cooper. Indianapolis: Hackett Publishing Company.


Strawson, Galen, 1994. Mental Reality. Cambridge, MA: MIT Press.


Strawson, Galen, forthcoming. “Experience and Emergence.” Unpublished draft.


Wilson, Margaret, 1980. “Objects, Ideas, and ‘Minds’: Comments on Spinoza’s Theory of Mind.” In The Philosophy of Baruch Spinoza. Edited by Richard Kennington. Washington, DC: Catholic University of America Press.


Wittgenstein, Ludwig, 1922/1966. Tractatus Logico-Philosophicus. Translated by D F. Pears and B. F. McGuinness. London: Routledge & Kegan Paul.