2. Beginning With an Old Issue: Is the Self a Substance or a Mode?

Whether the self is a substance or a mode is one of the issues that divided Descartes and Leibniz (who took selves to be substances) from Locke and Spinoza (who took selves to be modes). Nowadays, philosophers tend not to think in classical metaphysical terms. The issue may sound both quaint and obscure to the modern ear. I believe, however, that starting with this issue (thinking in classical metaphysical terms) may have a salutary clarifying effect.

3. Substance Defined

There are at least two conceptions of substance that must be distinguished: the Aristotelean conception of substance, and the Early Modern conception of substance.

Aristotle thinks of an individual substance as a ‘this,’ an object to which one can point. He seems to think that biological organisms (Bucephalus the horse) and artifacts (the knife I used to cut my orange this morning) are the best examples of individual substances. Each individual substance, for Aristotle, is a combination of matter and form. Matter is some combination of the four ‘elements’ – earth, air, fire, and water. (Perhaps there is an underlying kind of matter, ‘prime matter,’ more basic than the elements; Aristotle at least flirted with that idea.) Form is a combination of shape or structure, and function. According to Aristotle, a thing's form is closely associated with its causal powers (what it can do). Thus, an Aristotelean individual substance is a complex object, a bearer of properties, including causal powers. With regard to the individual substance that is a particular human being, Aristotle evidently takes this substance to have its own original causal power, and free choice as well. As he puts it, some things (our voluntary acts) are up to us.

In the Early Modern period, Aristotelean metaphysics fell out of favor. The ancient idea of atomism was revived. The search was on for basic elements out of which everything in the physical world is composed. In the Early Modern thought of Locke, Galileo, and Newton, these basic physical substances are tiny corpuscles. They are conceived to be the bearers of all genuine (‘primary’) physical properties or qualities, and such physical properties or qualities confer causal powers. Each basic physical substance (atom or corpuscle) is taken to be independent (it needs nothing else to exist), and such basic physical substances are taken to be the things that maximally persist through changes in the material world.

Some early modern thinkers, such as Descartes and Locke, also believed in mental or spiritual substances; these were taken to be the bearers of experiential or conscious states. Locke, to his credit, admitted that our understanding of how an immaterial substance can think is as dim as our understanding of how a material substance might think. Our ideas of both kinds of substance are equally obscure.Footnote ⁴ Locke, at least in passing, considers the possibility that conscious states might be modes of matter.Footnote ⁵

Not all Early Modern thinkers were atomists with regard to the physical world. Spinoza thought there was only one substance, everything-there-is, which he (perhaps misleadingly) called ‘God’. (Maybe we should say that Spinoza was an atomist, but he believed in just one big Atom.) Spinoza insisted that everything-there-is was the only truly independent thing, therefore, the only true substance. It was also the only thing with independent (‘free’) causal power.

Leibniz, curiously pre-figuring Kant and other transcendental idealists, took the material world to be a dream of the true substances.Footnote ⁶ The true substances were a plurality of independent elementary minds or selves he called ‘monads’. Leibniz denied that monads have causal power. Given their absolute independence from each other, monads cannot interact causally. Instead of having causal power, the monads ‘unfold’ their ‘inner notions’ in ‘pre-established harmony’ with each other.

4. Simple and Complex Substances; Their Properties and Causal Powers

An individual substance, for Aristotle, is obviously a complex entity. Early Modern philosophers often divide substances into basic and non- basic (or simple and complex). If something like atomism turns out to be true, it is said, the atoms will be the basic or simple substances; non-basic or complex substances will be the ‘thises’ made up of arrangements of these atoms. If we choose to talk this way, it seems we must count as non-basic substances both ‘stuffs,’ like chemical elements and compounds, and objects, like Bucephalus the horse and my fruit-knife.

Not everybody accepts complex substances. John Heil, for example, considers ordinary middle-sized physical objects to be substances only ‘by courtesy’. Here is Heil's reason for adopting the view that the only true substances are simple: substances are the bearers of properties. Consider a complex object such as the fruit-knife. It is, presumably, an arrangement of property-bearing substances (sub-atomic particles, or whatever). Now, suppose we ascribe a property to the complex object (sharpness, for example). If sharpness is a genuine property, what bears it? Not the little individual substances that make it up. But that's all it is!Footnote ⁷

Most philosophers disagree with Heil about this. Intuitively, most people want to say that the knife is sharp because its elementary particles (whatever they turn out to be) are arranged in such-and-such a way, to form atoms and molecules; and in turn, the atoms and molecules are arranged in such-and-such a way, to yield a hard substance with a sharp edge. This seems to be an example of a complex substance having a causal power of its own. To find a bearer for sharpness, one has to consider the knife a complex substance. The knife is something over and above the particles of which it is composed, because it has properties (and hence, causal powers) not possessed by any of its compositional elements.

Similarly, we want intuitively to say that water and other liquids are wet, but none of their tiny substantial components has the property of wetness. Wetness is ‘emergent from’ or ‘supervenient on’ certain arrangements of molecules (and whatever tinier things make up the molecules.) Lots of people want to say similar things about the mental properties of beings with large, complicated brains – consciousness, qualia, will, and so forth.

The problem is that the properties and causal powers of so-called complex substances are seemingly completely explained by and dependent on, hence reducible to, the properties and causal powers of their substantial components. There is nothing left over for so-called ‘higher-level’ properties to do – no additional causal powers they can confer. Those who advocate ‘strong emergence’ for mental properties, and other sorts of ‘higher-level’ properties,Footnote ⁸ run into intractable problems about causal over-determination and causal-explanatory exclusion.Footnote ⁹

5. Mode Defined, and Self as Mode

Let us confine ourselves, for the moment, to the Early Modern idea of substance. It is this Early Modern notion of substance that is typically contrasted with the notion of a mode. The simplest definition of ‘mode’ is a way some substance is. We need to understand what it would be to take the self to be a mere mode as opposed to a substance.

If, as Locke believed, there are many simple physical substances (corpuscles), and they organize themselves into loose clouds with vague boundaries and persistence conditions, then we have a decision to make. What should we call these clouds of little substances? Shall we call them complex substances, or should we call them modes? Locke himself often calls chemical elements and ordinary middle-sized objects ‘substances’.Footnote ¹⁰ But perhaps Locke is being careless when he talks this way. What he really seems to think is that gold, a horse, etc., are modes, ways the basic physical substances (corpuscles) organize themselves. So, he should have called them ‘modes’.

Recall John Heil's view that the only genuine substances are simple. The reading of Locke under consideration is precisely the Heil view. The more one thinks about this view, the more problematical it becomes. If the only true physical substances are the corpuscles, the only true modes of physical substance must be ways the corpuscles are, in other words, the primary qualities of the corpuscles that confer causal powers on the corpuscles. When multiple corpuscles organize themselves into complex ‘things’, (such as what Locke calls a ‘man’Footnote ¹¹) the result may be said to be a mode (just as it may be said to be a complex substance) only ‘by courtesy’, only as a facon de parler.

Heil seems to accept this startling view – not only am I not a substance, I am not even a mode. I am a ‘mode by courtesy’.

Spinoza's one-substance view is a bit easier to swallow – ordinary physical objects such as trees, rocks, and our own bodies turn out to be genuine modes – ripples or thickenings in the One Substance, God.

6. Modes, Properties, and Tropes

A question arises as to the relationship between modes and properties. Can modes have properties, or should we say that modes are properties?

The question of the nature of properties is a vexed one – metaphysically central, and notoriously confusing and obscure.

One option is to conceive of properties as Universals. Universals are weird entities – not in space and time, exemplifiable by multiple particulars, and so forth. Some philosophers even say they are ‘wholly present’ in each of their instances. This is rather difficult to understand.

Alternatively, we may think of properties as particulars – as particular as the things that have them. Properties are particular ways particular things are. In other words, properties are… modes! This illuminating and fruitful suggestion is endorsed by D.C. Williams, among others. Williams suggested that we call such individualized properties ‘tropes,’ and he pointed out that such tropes are precisely what the Early Modern philosophers called ‘modes.’ They are ways that things (substances) are.Footnote ¹² The term ‘trope’ has become commonplace now for individualized properties. As D.M. Armstrong has noted, resemblance-classes of tropes seemingly can do all the metaphysical work traditionally done by Universals.Footnote ¹³

Perhaps ‘trope’ and ‘mode’ are not exactly equivalent, as ordinarily used by philosophers (some trope theorists don't even believe in substances, for which they substitute bundles of tropes; believers in modes, by contrast, generally take modes to be dependent on substances) but the two notions are quite similar.

7. The Heil/Martin View

The late C.B. Martin defended an ontology of (simple) substances, existing in a ‘power net’ created by their mutual dispositions or causal powers (their tropes).Footnote ¹⁴ John Heil, influenced by Martin, has defended a similar ontology.Footnote ¹⁵

For Heil, the only genuine properties (tropes) are the ‘powerful qualities’ of the genuine or simple substances. Heil thus rejects the ‘layered’ view of reality so popular at present, according to which novel properties ‘emerge’ or ‘supervene’ as matter becomes more complex. This has important implications for the mind-body problem. As Heil sees it, we may truly predicate mental states (propositional attitudes as well as qualitative states like pain) to persons, but it is a mistake to suppose that these predicates correspond to a special set of mental properties. So-called ‘property dualism’ or ‘non-reductive materialism’ (currently the most popular position on the mind-body problem) is metaphysically confused.

For Heil and Martin (neither of whom, so far as I know, accepts the existence of spiritual substances) a human being is a mode-by-courtesy of simpler physical substances. It is the simpler substances that possess all the causal power. This is comparable to Jaegwon Kim's view that any causal powers seemingly possessed by complex objects, including human beings, reduce to the causal powers of their physical components. According to Kim, the so-called ‘multiple realization’ of mental properties by many different physical properties in different substances makes no metaphysical difference; multiple realization is just a complex variety of reduction.Footnote ¹⁶

This may seem to be a disturbing view of the nature of human beings, because it apparently leaves no room for a robust conception of agency. Those who want to hold out for free will in the libertarian sense, it seems, must eschew the Heil–Martin ontology, as well as Kim's reductionist view.

8. Agency and Freedom

Causal powers might or might not exist in a deterministic web. One could take the self to be a substance, and still not be a libertarian about free will, if one believes that the self's causal powers are always elicited in a deterministic wayFootnote ¹⁷ by the conditions in which the self is embedded. Those who want to defend libertarian free will must (apparently) do two things: defend the idea that the self is a substance, and also defend the idea that the self is a very special kind of substance, a rational substance with a ‘real choice’ about how it behaves. That is, whenever this substance chooses an action, it must be the case that it is also in the substance's power not to choose it.

What if one takes the self to be a mere mode of physical substance, embedded in a deterministic web? Then, it seems, libertarian free will is out of the question. Spinoza (always logical and consistent) bites the bullet and adopts compatibilism rather than any sort of libertarian version of free will. According to Spinoza, it may feel as if we are free to choose or not to choose, but this is an illusion based on ignorance. Our choices, and the actions that follow upon them, are caused in the same way as all other events within God; they are part of the necessary unfolding of the One Substance. Nobody really can do otherwise than what he in fact does. To be ‘free’ as opposed to a victim of ‘human bondage’ is just to have clear ideas about one's environment and interests, and hence to act in a rational way that succeeds in serving one's true interests. Whether a given person will achieve this blessed state is, alas, not ‘up to him’ (except in the sense that it depends partially on his character, which is itself causally determined). Over the course of their lives, some people become free and others don't; it is all part of the necessary unfolding of God.

Locke's views on freedom of the will are rather more difficult to understand. As we have seen, Locke takes the self or person to be a mode of a spiritual substance. He takes the man, however, to be a mode of physical substances (a cloud of corpuscles), or perhaps a Cartesian union of mind and body that somehow unites the cloud of corpuscles with the spiritual substance. Since he insists it is the man that is properly said to be free or unfree, and he certainly accepts determinism as a feature of the physical world, it seems Locke should have unambiguously adopted compatibilism. However, Locke seems torn between compatibilism and libertarianism.

Much of what Locke says sounds compatibilist. For example, he insists that it is nonsensical to ask whether the will is free. The will is a power, and to ask whether the will is free is to ask whether a power has a power, which makes no sense.Footnote ¹⁸ It is the man who is free or not free, and lack of freedom in the case of a man means something like paralysis or external constraint. Freedom is just the ability to act, or not act, according to the preference or direction of one's own mind.Footnote ¹⁹ Locke shows, using the famous example of the man in the locked room who nevertheless stays there of his own free will, that being able to do otherwise than one in fact does is not a necessary condition for acting freely.Footnote ²⁰

The libertarian-sounding aspect of Locke concerns his claim that the self has the power to act, or not act, according to the direction of the mind. Locke clarifies this in later sections of the Essay by saying that a human being has the power to suspend the willing of any action until he has considered the matter according to reason, and decided whether the action is really conducive to happiness or not. Then, he can choose, or not choose, according to what reason decides.Footnote ²¹ When Locke adopts this libertarian-sounding view, it is not clear to what substance he takes this power of choosing to belong. Is it the spiritual substance (the self or person)? Is it the complex substance made up of the union of this spirit with the physical body (the cloud of corpuscles)? Since Locke uses the term ‘Man’ in the relevant passages, it seems reasonable to conclude that he is talking about the whole human being, the union of spirit and body. Locke never, to my knowledge, addresses the notorious interaction problem: how does the spiritual self communicate its decision (to choose or not to choose) to the physical body, so that the body either moves or doesn't?

Compatibilists believe it is never true that a person ‘could have done otherwise’ than what he in fact did, except in the hypothetical sense that ‘he could have done otherwise if he had chosen otherwise’. And, in the given circumstances, he was caused to choose as he did, so he really couldn't have chosen otherwise. Thus, compatibilists must apparently deny that persons really have the Lockean rational power to choose or not to choose.

I have never considered Spinoza's compatibilism to be incoherent. It has always seemed to me the reasonable position to adopt on the free will question, if one thinks the self is a mere mode of physical substance. E. J. Lowe, however, thinks compatibilism is incoherent, and defends the idea that the self is an emergent, rational, simple substance. Let us examine Lowe's views on freedom and the nature of the self.

9. Self as Emergent Simple Substance: E.J. Lowe's View

Lowe agrees with Locke that persons have a libertarian power of choosing, or not choosing, their voluntary actions. He agrees with Aristotle that some things are up to us. He is also impressed by the evident fact that the self is a unified subject of experience. His view is subtle and takes some spelling out.

A substance for Lowe, is ‘an individual object or bearer of properties – where these properties include various distinctive and independent causal powers.’Footnote ²² Lowe does not object to complex substances. Indeed, his view of substance is more Aristotelean than Lockean.

Lowe remarks (and I agree) that the terms ‘mental property’ and ‘physical property’ are thrown around in current philosophy of mind with insufficient attention paid to exactly what ‘mental’ and ‘physical’ mean. (What ‘property’ means is also usually left obscure.) However, Lowe puts such problems aside, and defines a mental substance as ‘a bearer of mental or psychological properties’ and a physical substance as ‘a bearer of physical properties.’ As paradigm examples of mental properties, Lowe mentions pain and desire. As paradigm examples of physical properties, he mentions mass and velocity.Footnote ²³

It is apparently not the case, according to Lowe, that one and the same substance bears both kinds of property. You have mass only indirectly, in virtue of having a body with mass m. You are simple, and your body isn't. Your body has parts, each with a definite mass, but you have no parts.

Lowe calls his view ‘non-Cartesian substance dualism’ – hereafter, NCSD. The position is a variety of substance dualism because it contends that certain bearers of mental properties are distinct from (not to be identified with) certain bearers of physical properties. The bearers of the relevant mental properties are subjects of experience. The bearers of the relevant physical properties are bodies (or parts of bodies.) Lowe holds that a person or self is not to be identified with his or her body, or with any part of it, such as the brain.Footnote ²⁴

As Lowe thinks about these matters, a person (not the person's body or brain) feels pain and has desires. This is true even though, probably, the person feels pain or has desires only because his body or brain is in a certain physical state. The relevant physical or neural states are not to be identified with the mental states, which are states of the self, a distinct substance. Neither is the bearer of the mental states (the subject of experience) to be identified with any physical object.

Lowe says that his view differs from Descartes' in holding that a person or subject of experience could have both mental and physical properties. Descartes held the view that the mental substance could have only mental properties. I find this baffling and difficult to square with other things Lowe says about NCSD. Here is how Lowe puts it:

But what physical properties could I bear, that were not properties of my body? And I, according to Lowe, am not my body. Thus, I could be said to bear physical properties, such as mass, only as a loose manner of speaking. I would have such physical properties only ‘by courtesy,’ to use Heil's locution. But then, why couldn't my body have mental properties, in an analogous sense? Perhaps Lowe's position is not really distinguishable from Descartes' in the manner he suggests.

Lowe offers an argument for NCSD's central claim that a person is not identical with his or her organized body. He calls it ‘the unity argument’. It goes as follows:

1. 1. I am the subject of all and only my own mental states.

2. 2. Neither my body as a whole nor any part of it could be the subject of all and only my own mental states.

3. 3. Therefore, I am not identical with my body or with any part of it.Footnote ²⁶

Lowe's sub-argument for the second premise goes like this:

1. 1. My body as a whole does not need to exist in order for me to have every one of the mental states that I do in fact have. (If, for instance, I were to lack the tip of one of my little fingers, I might as a consequence lack some of the mental states that I do in fact have, but surely not all of them. I might perhaps lack a certain mildly painful sensation in the finger tip – a sensation that I do in fact have – but many of my other mental states could surely be exactly the same as they actually are.)

2. 2. No entity can qualify as the subject of certain mental states if those mental states could exist in the absence of that entity. (After all, I certainly qualify as the subject of my mental states…mental states must always have a subject… But, as we have seen, very many and quite possibly all of my own mental states could exist even if my body as a whole were not to exist – that is to say, even if certain parts that my body actually possesses were not to exist.)

3. 3. Therefore, my body as a whole cannot be identified with me (the subject of my mental states).Footnote ²⁷

Lowe contends that the same argument can be run with ‘my brain as a whole’ substituted for ‘my body as a whole’ throughout. The crucial insight is that ‘…my mental states do not all depend on my body as a whole or on any part of it in the unified way in which they all depend on me as their subject.’Footnote ²⁸

Recall that Lowe's notion of a substance is a bearer of properties, including causal powers. Lowe's view is that where there is a live human body with an appropriately functioning brain, we don't just get emergent mental properties; we get an emergent bearer of those properties, an emergent (simple) substance. It has its own uniquely mental properties (it is the subject of its own experiences) and its own unique powers (it can perform acts of will.) It is crucial to Lowe's view that he doesn't take this will-power to be causal; it is a special, non-causal power.

An example of a causal power would be the power of water to dissolve sugar. If water comes into contact with sugar, the dissolving happens; the water lacks the power to decide not to dissolve the sugar. Causal powers are like that – their exercise consists in producing a certain sort of effect. Will-power, by contrast, can be exercised even when the desired result doesn't come about. For example, I can will to raise my arm (I can try or make a mental effort) even if paralysis or some external constraint prevents my arm from going up.Footnote ²⁹ Furthermore, it is ‘up to me’ whether or not to exercise the power of will. Because of this ‘up to me’ aspect, Lowe calls the will a ‘spontaneous’ power. Like the decay of a radioactive element, the exercise of will is (Lowe says) uncaused. But unlike radioactive decay, it is exercised in the light of reason. It is a rational power as well as a spontaneous power.Footnote ³⁰

Lowe firmly rejects the compatibilist view that the choices of an individual are as determined as anything else in the empirical world. According to Lowe, rational choice is free in a strong, libertarian sense that is incompatible with its being a causal power:

It is hard not to be persuaded by this, even for an old compatibilist like myself. Certainly, it feels as if we have the power to choose or not to choose. A great deal seems to ride on whether we do in fact possess such a power. Aristotle said that we can improve our own characters by choosing to practice the right sorts of actions until they become habitual. But how is that possible if I don't really have the power to stop myself from choosing the old maladaptive behaviors I strive to overcome? Furthermore, liberal political theory rests on the conception of a person as a free agent with rational self-control. I am a great believer in liberalism, and the kind of free society that is founded on it. It is rather uncomfortable to have to say that it rests on a fiction.

But how is one to square this inner certainty of the self's freedom with what physical science tells us about ourselves? Science recognizes no ‘ghost in the machine’, nothing over and above the physical body with its physical causal powers. If one is going to be scientific about the self and its powers, it seems one is stuck with compatibilism, however unsatisfactory it may seem in certain moods.

Furthermore, I find it hard to make sense of Lowe's idea of a rational power that is not a causal power. Unless the non-physical self can cause the physical body to move, in accordance with its rational decisions, this non-physical self would seem to have no effect on the world we live in. The interaction problem seems as intractable for Lowe as it was for Descartes and Locke.

Lowe's view seems akin to Kant's. According to Kant, the noumenal self (outside space, time, and causality) somehow makes it the case that the world contains a causal chain that includes a free action. Lowe's emergent, simple, rational substance plays a role similar to that of Kant's noumenal self.Footnote ³² I wish that I did not find this so difficult to understand, since I would like to believe in a robust conception of human agency. As I grow older and more thoughtful, I see that the (difficult) exercise of self-determination and autonomy is what makes a human life uniquely worth living. And I doubt more and more that such self-determination and autonomy is compatible with determinism.

Lowe has done us a great favor in showing what sort of metaphysics believers in libertarian free will must embrace. In a more personal vein, he has ‘awakened me from my dogmatic slumbers’ and forced me to think harder about the possibility and desirability of a Kantian sort of free will.Footnote ³³