SUBSTANCE DUALISM AND DISEMBODIED EXISTENCE

 

 

Nicholas Everitt

 

 

 

 

 

University of East Anglia, Norwich, England

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

SUBSTANCE DUALISM AND DISEMBODIED EXISTENCE

 

 

 

 

I

Substance dualism, that most unpopular of current theories of mind, continues to find interesting and able defenders.1 I shall focus on one set of arguments supplied by one of the current defenders, and I shall argue that these arguments fail. That in itself is a matter of some interest, since it is always reassuring to be able to demonstrate that unpopular doctrines are rightly unpopular. But I hope that a further interest will attach to the refutation, in that it will invoke some relatively unfamiliar thoughts about the nature of perception.

My stalking horse in this exercise will Richard Swinburne. Swinburne has defended substance dualism on a number of occasions, and very often by appealing to quite different sets of arguments. In The Coherence of Theism, for example, substance dualism is defended by an argument by elimination: it emerges as the only theory which will enable us to avoid crippling deficiencies in all its rivals. In Is There A God? and The Evolution of the Soul, the theory is defended by an appeal to thought experiments about brain transplants. In Personal Identity, and again in The Evolution of the Soul, it is by an appeal to the possibility of disembodied consciousness. I will focus on the argument from the possibility of disembodied consciousness, as this seems to be the line of argument which Swinburne finds most convincing. The original argument of Personal Identity, for example, is repeated in his 1996 paper 'Dualism Intact', again (in slightly revised form) in the second edition of The Evolution of the Soul, and is endorsed again in his most recent publication on the topic, his 1998 paper 'The Modal Argument Is Not Circular'.

 

II

The central argument in PI for substance dualism can be simply stated:

(A) It is logically possible for me to continue to exist without my body or any part of it such as my brain

so (B) I must (now, here) already have some non-bodily component, whose persistence is the persistence of me

so (C) I must have a non-bodily mind, a mental substance.

Suppose this intended is intended as an argument desgined to convert to dualism someone who is initially neutral on the issue, or is even a committed anti-dualist. The argument then seems open to an immediate objection: surely the premise is too strong. Someone who is uncertain between some form of materialism and substance dualism is unlikely to regard (A) as uncontroversially true; for if they were certain of (A), they might think that (B) and (C) would obviously follow, and they would thus, contrary to hypothesis, not be uncertain betweeen materialism and substance dualism. I shall argue later that this line of thought would be a mistake, i.e. that (B) and (C) do not follow from (A) at all. But there is a good deal of initial plausibility in the thought that (A) is too rich a premise: why would anyone who has not already decided that substance dualism is true accept (A)?

Swinburne implicitly recognises the force of this objection, and seeks to show that (A) is true, by using considerations which he thinks have independent strength. Although all that his initial argument needs is the premise that a person can exist without a body, he seeks to show that this premise is acceptable by arging for the stronger claim that a person can perceive without a body. He offers an analysis of those relations between me and my body in virtue of which we say that one particular body is mine; and then argues that it is possible for me to continue to exist even when those relations do not hold between me and any body at all, i.e. when I do not have a body at all. I will produce in sections III and IV two arguments challenging the possibility of disembodied perception. I will claim that although these are plausible, there is a convincing reply to the second. Then in section V, I will argue that even if neither of those arguments works, and hence that we have no reason to reject premise (A) above, neither (B) nor (C) will follow. In section VI, I will consider a possible objection to the argument of section V, and then in section VII I will reply to that objection. finally, in section VIII, I will relate my criticism to the most recently published criticism of the argument and to Swinburne's reply.

 

III

First, then, what are the relations between me and a particular human body in virtue of which that body is mine? Swinburne invokes two sets of characteristics, one to do with intentional action, the other to do with perception. First, there is one and only one body in the world which I can control without having to perform any other intentional action. I can of course control the movement of many objects in the world. I can push you down the stairs, I can steer my car, I can make a bullet speed through space. But when I make you or my car or a bullet move, I make it move by making my own body move. I move my body against you, and thereby push you. In the car, I move my hands to make the steering wheel move, which makes the wheels move, which makes the car move in one direction rather than another. When I make the bullet move, I do so by moving my finger against the trigger, which makes hammer snap and thereby makes the detonator explode and the bullet speed out of the barrel of the gun. But when I move the physical object that is my body, I do not have to move it by making some other physical object move. I have, in one sense of the word, direct control over at least some of what my body does. So I am uniquely related to one and only one body in the world in terms my power of voluntary agency.

Secondly, the existence and nature of my perceptions of the external world depends on the position and condition of one and only one body and its sensory organs. Trivially, what I can see depends on where in space this body is located; on whether there are any opaque objects between this body's eyes and the putative object seen; on whether this body's eyes are open and focused in that direction; on whether light waves impinge on this body's eyes, and produce impulses in this body's optic nerve and this body's visual cortex; and on. Analogously with the other senses: if I am to hear anything, then this body's ears must be affected by sound waves which cause impulses in this body's auditory nerve; whether I can feel any surfaces depends on contact between other objects and this body; whether I can smell anything depends on molecules stimulating the nose of this body; and so on. Other bodies can of course make a difference to what I perceive. I can change what I see by turning this body's head (and hence this body's eyes and line of sight); but equally you can change what I see, e.g. by moving your body into my field of vision, or more dramatically by punching me on the head and disruptng my visual powers. But the sort of dependence which I have on this body for my perception of the world around me is different in kind from the kind of dependence on all other bodies.

So, considerations about intentional activity relate me uniquely to one body; and considerations about perceptions of the external world relate me uniquely to one body. In the normal case, there is a single body to which I am uniquely related in each of these two ways: the body which I can move directly is the very same body as the one on whose position and condition my perceptions of the world depend. Swinburne now claims that to say that a body is mine is to say I am related to it in these two sorts of ways.

Swinburne next asserts that the set of relations which obtain between me and the body which I call my own could cease to hold between me and that body, and come to hold between me and another body. This would amount to my changing bodies. It is also logically possible (he urges) that the relations which hold between me and my body might cease, and yet I remain in existence.

Suppose [he writes] that [a person] finds himself able to operate on and learn about the world within some small finite region, without having to use one particular chunk of matter for this purpose. He might find himself with knowledge of the position of objects in a room (perhaps by having visual sensations, perhaps not, and able to move such object just like that, in the ways in which we can know about the positions of our limbs and move them...The person would be in no way limited to operating and learning through one particular chunk of matter. Hence we may term him disembodied.3

The transformation of a normal embodied person into a disembodied person, as here envisaged, has four stages. Since two are unproblematic and two are problematic, it is worth listing them separately. First, we have to imagine that I cease to have any direct control over the body which I call mine. Secondly, I find that what goes in the body I call mine makes no difference to my perceptual experiences. Light waves impinge on these eyes, but no visual experience results for me; sound waves strike the ear drum but produce no auditory experience for me; and so on. Thirdly, I find that I can, for example, see and hear you speaking in one room, when what used to be called my body is in another room down the corridor, or in another building, or in another country, or even completely destroyed. Fourthly, I find that I can directly control the movements of other objects in my environment, in the way in which I used to be able to control at least some of the movements of my own body.

The reason for separating Swinburne's thought experiment into four steps is that there is nothing puzzling about either of the first two. Someone can indeed lose the power to move their own limbs: they can be overtaken by paralysis. And even if the paralysis were absolutely total, so that none of their bodily movements was under their control, that would surely not be any evidence that perhaps the body was beginning not to be 'theirs'. Equally, there is nothing puzzling about the thought that someone can lose the functioning of all their sensory systems, so that they can no longer perceive the external environment at all. We are familiar with the fact that people go blind, go deaf, lose their senses of taste and smell, etc. Clearly, there is no reason why these afflictions should not strike someone simultaneously. And if this were to happen, we would surely not be tempted at all to say that the body was no longer 'theirs'. Even if the twin disasters of total paralysis and total sensory deprivation were present at the same time, there would still be no temptation to say that the victim no longer had a body. The body would remain as before the obvious focus of our concern about the individual.

But it is with the third and fourth changes that deep puzzles emerge about the idea that it is possible for us to exist without a body, and here I want to focus on the third change, that is, on my discovery that I perceive the environment but not from (what I used to call) my body, and not from any other body either. There is, I think, an attractive line of argument which concludes that there are hidden difficulties in the scenario which Swinburne envisages - but it is a line of argument which ultimately fails. So before I come on to what I believe is the really fatal weakness in his position, I want first to explore the appealing but ultimately meretricious line of thought.

To lay the ground for the falsely tempting argument, we need to remind ourselves first of how our perception of the enviroment is tied to the physical contingencies of our bodies, in much more detailed ways than Swinburne has alluded to. For example, I have two eyes and not one, and the two eyes are separated in space. Because of this, I am able to perceive depth. Again, because I have two spatially separated eyes, there is a difference in my visual perception when I look at a scene with my head held erect, with my head held at 90o to my right, with my head held held upside down, with my head held at 90o to my right, and with my head at 180o to the vertical. If I am thought of as able to see things when I have no body, is the idea that I see things from a mere geometrical point in space, or that somehow although I have no physical embodiment, my visual powers will still be distributed across a spatial region as they are now? Again, because I have eyes at the front of my body but not at the back, I can see what is in front of me, but have to turn round if I am to see what is behind me. But if I have no body, what could determine in which direction I am looking? If I am supposed to be seeing from a geometrical point in space, does it make sense to suppose that that point can rotate to reveal what is behind me? What could 'in front' and 'behind' mean to a perceiver without a body? Does it make sense to speak of variation in the orientation of a geometrical point?

Puzzles like these are not peculiar to the sense of sight. Consider smell. Humans can perceive smells when they inhale: inhaling surely requires a nose or similar organ by means of which to sniff. And if I have a nose, but hold my breath, I cannot smell any odours that are objectively there to be smelt. A creature with no body cannot sniff or inhale anything. How then could it smell anything?

Problems of a different kind arise in connection with touch. When we use the sense of sight, we can be relatively passive: we do not have to do much in order to see. But in relation to our tactual sense, we are typically much more active. If I want to feel the shape, or the temperature, or the texture of an object, I have to bring my body into contact with it and touch it. With smallish objects, I can (e.g.) pick them up in my hands and turn them about. Because I have a spatially extended body, I can simultaneously feel that a block of wood is rough on top and smooth underneath, or hot at one end and cold at the other. But if I do not have any hands or similar physical parts with which to feel two different surfaces, what would determine which surface I was feeling as smooth, and which as rough; which end as hot and which as cold?

For reasons of this kind, the idea that we could continue to perceive our environment even if we had no body is a good deal more puzzling than Swinburne allows. At the very least, we need to be given much more detail about what is being envisaged before we should grant that it is even prima facie possible. Can we go beyond asking these rhetorical questions, and find something incoherent about disembodied perception? In the next section, I will present a plausible argument for saying that nobody can ever have any reason to think that they or anybody else were having disembodied perceptions.

 

 

IV

 

We can start by noting first that if a person were disembodied and were perceiving his environment, no one else could know that it was happening. Since ex hypothesi the person would be non-physical, they would be undetectable by anyone else. So if the occurrence of disembodied perception is to be detectable at all, it will have to be detectable by the person to whom it is supposed to be happening. But here is an argument for saying that that is not possible. In what follows, I will focus on the sense of sight, but the argument will generalise to the other senses as well.

The idea that I am seeing my environment presupposes that I have a location from which I do my seeing. If I am in the kitchen, for example, the sink but not the television in the living room would be within my visual range, whereas if I am in the living room, the television but not the sink would be within my visual range. So if I am in the kitchen, and I have visual experiences as of seeing the sink, then maybe I really am seeing the sink. But if I am in the kitchen, and I have visual experiences as of seeing the television, then I am not really seeing the television: I am hallucinating it, or misperceiving as the television some other object in the kitchen. Similarly, if I am in the living room and I have visual experiences as of seeing the television, then maybe I am really seeing the television. But if I then have visual experiences as of seeing the kitchen sink, then I am hallucinating or misperceiving.

So in order for there to be a distinction within my experience between genuinely seeing on the one hand, and having visual hallucinations and misperceptions on the other, I must have a determinate position in space. It is my position in space which partly determines for me at each moment what is then genuinely visible by me; and it gives content to the idea that visual experiences which are of items that not currently visible by me must be hallucinations or misperceptions, not genuine seeings. In the case of a normal embodied perceiver, her position is given by the position of her body in space. If her body is in the kitchen, that restricts what she can really see to what is visible from within the kitchen. So if she has experiences as of something in the living room, it follows at once that they are not genuine seeings. The problem with Swinburne's disembodied perceiver is that in order to distinguish between seeing and hallucinating, she must have a spatial position; and there is no way of knowing what her spatial position is independently of what her experiences of her environment are. All that she can say (or rather think) to herself is something of the form 'I am having some visual experiences of the sink. So if I am in the kitchen, maybe I am really seeing the sink; whereas if I am in the living room, then I must be hallucinating/misperceiving'. But she has no way of determining which of these 'if' clauses is the right one; and hence no way of determining whether she is really perceiving or hallucinating.

So, the disembodied perceiver cannot draw a distinction between really seeing and just hallucinating/misperceiving. The most that she can know is that she has some visual experiences, now of the sink, now of the television. But she cannot sort those experiences into cases of genuine seeing and of hallucination/misperception. For all she can tell, all of them could be hallucinatory.

Why should this matter to Swinburne's case? The reason is this: if the supposedly disembodied perceiver's experiences might all be hallucinations, there is no reason to think that the perceiver is located outside her erstwhile body. For although I cannot really see the sink unless the sink is within my visual environment, clearly I can hallucinate the sink from any position in space. So if a perceiver is having a set of visual 'sink' experiences, and there is no reason to think that any of these experiences is non-hallucinatory, the experiences cannot provide any grounds for thinking that the perceiver is located outside her body. And since we agreed at the outset that a disembodied perceiver's supposedly perceptual experiences could not give anyone else a reason for saying that saying she was located outside her body, it follows that her perceptual experiences give no reason to anyone to suppose that she is located outside her body.

If we try to draw together the threads of the above argument, we get the following:

1 The experiences of the supposedly disembodied person (the SDP) do not give anyone else a reason to say that she is located outside her erstwhile body (premise)

2 To draw a contrast between really seeing and merely hallucinating/misperceiving, there must be an independent check on where the perceiver is located (i.e. independent of the phenomenal content of the perception) (premise)

3 For the SDP, there is no independent check on where she is located

so 4 The SDP cannot divide her experiences into cases of really seeing and cases of hallucination (from 2 and 3)

so 5 She has no reason to reject the thought that all her experiences are hallucinations (from 4)

6 No matter how remote the objects which you hallucinate, you can hallucinate them all while remaining 'in' your body (premise)

so 7 The SDP's experiences give her no reason to think that she is located outside her erstwhile body (from 5 and 6)

so 8 Her experiences give no one any reason to think that she is perceiving without a body (from 1 and 7)

so 9 Her experiences give no one any reason to think that she is existing without a body

We have then two plausible lines of argument against the possibility of disembodied perception. The first says that our understanding of what perception is, is closely tied to the way in which human perception is dependent on facts about human bodies. So close is the tie that we cannot, even in a thought experiment, make sense of perception when all these ties have been thought away. The second says that even if the first argument is mistaken, and we could still make sense of perception under the envisaged conditions, it would be impossible to tell which of our experiences were (veridical) perceptions and which were hallucinations.

But it seems on reflection that there must be a reply to the second argument.

It requires that for the person to know that she is genuinely perceiving something, she must (a) know where she is located, and (b) what is perceivable from her position in space; and further that she must know both of these things independently of the content of her current sensory state. For it is being supposed to be a necessary condition of her treating her current sensory state as a perception, that if so treated it would be a perception of something that has been independently determined to be perceivable. But once we put it like this, the assumed preconditions can be seen to be unreasonably demanding. For they are not met even by embodied perceivers. In general, if I am required to abstract from the content of my current sensory states, I cannot tell where I am located in the world. I do not first decide where I am located, and then as an epistemologically independent exercise, decide which of my sensory states is a perception and which a hallucination. Rather, there is a dialectical interplay: I decide where I am on the basis of my current perceptions, and I decide which of my current sensory states are to count as perceptions on the basis of where I am. It is a circle, but a virtuous circle.

But if this circle is virtuous for normal embodied perceivers, there is no reason why it should not be virtuous for abnormal disembodied perceivers. And in that case, the argument from section IV fails. However, that leaves in place the argument from section III, which still stands as a cogent objection to the intelligibility of Swinburne's thought experiment.

V

Suppose that the conclusion of the last section is wrong, and that Swinburne's premise (A) is true, and indeed can be known or reasonably believed to be true. Suppose, in other words, that it is at least possible for me to have good reason to believe that I am continuing to exist without a body. Swinburne argues that it follows from such an assumption that I must now have a non-bodily soul.

From the mere logical possibility of my continued existence [without a body] there follows the actual fact that there is now more to me than my body; and that more is the essential part of myself.4

Swinburne spells out the details of the argument in the following way:

p =I am a conscious person and I exist in 1984'

q = my body is destroyed at the end of 1984

r = I have a soul in 1984

s = I exist in 1985

x ranges over all consistent propositions compatible with (p.q) and describing 1984 states of affairs...

The argument may now be set out as follows:

p Premise (1)

(x) (p.q.x.s) Premise (2)

~(p.q.~r.s) Premise (3)

Therefore ~r is not within the range of x

But since ~r describes a state of affairs in 1984, it is not compatible with (p.q). But q can hardly make a difference to whether or not r. So p is incompatible with ~r

Therefore r. 5

To show what is wrong with this argument, I will produce an argument with the same logical form as Swinburne's in which it will be clear that the conclusion does not follow from the premises. This will show that the argument presented by Swinburne fails, at least in the sense that it is not, as he supposes, formally valid. I will then consider a possible criticism of the parallel argument, based on the claim that a key phrase in it can be interpreted in two ways. And I will then show that even if we recognise the possibility of these two interpretations, the fact cannot be used to rescue the argument for dualism.

The line of attack to be pursued here is thus significantly different from that in earlier critiques. As Swinburne himself notes (DI p.70), all earlier critiques have conceded that the argument is valid. I will suggest that the argument must be either invalid or question-begging.

Here, then, is the parallel argument Suppose that I have a chess set, and I raise the Swinburnean question whether it is true that

(1) It is logically possible for my chess set to survive even in a world in which there is no plastic

Remember that when Swinburne says that something is logically possible, he means only that a certain proposition is not self-contradictory. So to ask whether (1) is true is equivalent to asking whether (2) is true:

(2) 'My chess set survives although there is no plastic in the world' is not self-contradictory

And surely, we might think, (2) is true: it is not self-contradictory to say that my chess set exists even although there is no plastic in the world. It could be true if e.g. my chess set is made of wood - and my chess set could have been made of wood. So even if it would be false to say that my chess set continues to exist if all the plastic in the world were destroyed, it is not self-contradictory. But it clearly does not follow from (2) that

(3) My chess set now has some non-plastic part which continues to exist when all the plastic in the world is destroyed, and whose continued existence is the continued existence of my chess set.

Of course, if you know that my chess set is in fact made of plastic, you can deduce that it will be destroyed in a world in which all plastic is destroyed. And similarly, if you know that it is made of wood, you can deduce that it can survive in a world in which all the plastic has been destroyed. But that means that you have to know what it is made of before you can decide whether it can exist in a world in which all the plastic has been destroyed.

Suppose we now apply these reflections to Swinburne's argument. We are granting him the premise that

(1a) It is logically possible for me to survive in a world in which there are no human bodies

where that premise means

(2a) 'I survive in a world with no human bodies' is not self-contradictory

But that does not imply

(3a) I now have some non-bodily part which continues to exist when all the bodies in the world are destroyed, and whose continuance is the continuance of me

If you know that I am in fact a human body, you can deduce that I will be destroyed in a world in which all human bodies are destroyed. And similarly, if you know that I am a non-bodily 'soul', you can deduce that I can survive in a world in which all the human bodies have been destroyed. But that means that you have to know what I am (what I am made of) before you can decide whether I can exist in a world in which all human bodies have been destroyed.

If we wanted to cast the objection is a form that parallels Swinburne's own formal statement of the argument, we could put it like this:

p1 = my chess set exists in 1984

q1 = all the plastic in the world is destroyed at the end of 1884

r1 = my chess set has a non-plastic component in 1984

s1 = my chess set exists in 1985

x1 ranges over all consistent states of affairs compatible with (p1.q1) and describing 1984 states of affairs.

And the argument then goes:

p1 Premise (1)

(x)(p1.q1.x1.s1) Premise (2)

~(p1.q1.~r1.s) Premise (3)

But since ~r1 describes a states of affairs in 1984, it is not compatible with (p1.q1). But q1 can hardly make a difference to whether or not r1. so p1 is incompatible with ~r1

Therefore r1.

VI

It might be objected to this criticism of Swinburne's argument that it overlooks the fact that 'my chess set' and similar terms can be interpreted in two ways. On the first interpretation, 'my chess set' is a description which applies to anything which satisfies the two conditions (a) of being a chess set, and (b) of being mine. Since something could meet these two conditions whether it was made of plastic or not, the phrase 'my chess set' could have true application in a plasticless world even if in the world as it currently is, the phrase applies only to a set of wholly plastic objects. My chess set could still exist in a plasticless world only because the reference of 'my chess set' could be different in that world from what it is in the actual world. In other words, on the first interpretation, 'my chess set' is (in Kripke's terms) a non-rigid designator.

But on the second interpretation, 'my chess set' has an indexical element. It means something like 'this very chess set, which is as a matter of fact is mine'. The phrase is then a rigid designator: it picks out the same object in all possible worlds.6 And if it is a rigid designator, it could not pick out the object which is my chess set (this very set) in a plasticless world because this very set would not exist in a plasticless world. This remains true even if in the plasticless world of the future I come to own a new (wooden) chess set, and hence can truly apply the phrase 'my chess set' in that world.

What the above objection to Swinburne's argument presupposes is that 'my chess set' is a non-rigid designator. But (so the criticism of the objection would go) this is an implausible interpretation. Someone who speaks of their chess set almost certainly means to designate something indexically, not merely to designate some chess set or other which satisfies the two properties of being theirs and being a chess set. When we adopt the more plausible interpretation of 'my chess set', we see that (2) does imply (3). For if this very thing, my chess set can survive in a world in which there is no plastic, then this very thing must have a non-plastic part (and we might add, a part of sufficient importance to ensure that its survival is the survival of the chess set). In a parallel way, so the criticism will continue, once we adopt the 'rigid designator' interpretation, we can see that (2a) does imply (3a). For if this very thing that I designate by 'I' survives in a world without any human bodies, this very thing, I, must now have a part which is non-bodily.

Furthermore, it might be urged, there is a crucial difference between the chess set argument and Swinburne's original argument. For the phrase 'my chess set' could be either a rigid or a non-rigid designator'; and hence there are two interpretations of (2), on one of which (2) does not imply (3) and on the other of which it does. But by contrast, 'I', which is the relevant referring expression in Swinburne's argument, can never be anything but a rigid designator. There is therefore no interpretation of Swinburne's argument in which (2a) does not imply (3a).

Consequently, the criticism concludes, the objection to Swinburne fails. Once we draw the rigid/non-rigid distinction, we can see that the chess argument can initially seem an objection to the dualism argument only because it is open to a non-rigid reading. But by contrast, the dualism argument can receive only a rigid reading and that guarantees that the argument is valid.

 

VII

But this criticism of the objection to Swinburne's argument in fact fails. Suppose that we grant that 'I' cannot have a non-rigid reading, and we therefore compare the rigid reading of the chess argument with a rigid reading of the dualism argument. On the rigid reading, (2) means:

(2b) 'This actual object (my chess set) exists in a world in which there is no plastic' is not self-contradictory

We can agree that if this is true, it implies that (to paraphrase Swinburne) there is more to this actual object than plastic, and that more is the essential part of the chess set. But if I am to know that (2b) is true, then I have to know what my chess set is made of; and hence I cannot non-question-beggingly use (2b) to prove what my chess set is made. In particular, I cannot use (2b) as a non-question-begging way of establishing (3).

In a parallel way, on the rigid (and only possible) reading of (2a), it means

(2c) 'This actual object, I, exists in a world in which there is no human body' is not self contradictory

As with the chess argument, we can agree that if this is true, it implies that there is more to this actual object than a human body, and that the more is the essential part of the I. But if I am to know that (2c) is true, then I have to know what the I is made of; and hence I cannot non-question-beggingly use (2c) to prove what the I is made of. In particular, I cannot use (2c) as a non-question-begging way of establishing (3a).

VIII

This conclusion (that the argument is either invalid or question-begging) converges on a recent criticism of Swinburne's argument by Hasker. Hasker argues that Swinburne's argument relies on a premise which Swinburne summarises as follows:

[it] says, loosely in words, that it is logically possible that I who am conscious in 1984 should go on existing in 1985, even if my body is destroyed at the end of 1984 - whatever else might be the case in 1984 compatible with my conscious existence (and the subsequent destruction of my body).7

Call this proposition X. Hasker says in effect that since X is not self-evident, the only way to test it is to assess putative counter-examples. He puts forward as a counter-example 'It is not the case that I have a soul in 1984' (call this proposition Y), and seeks to argue that one could not have a reason for rejecting Y that did not consist in a prior commitment to X.

My argument of sections V to VII goes beyond Hasker's in showing in detail how in one sense, the logical possibility of the survival of an entity E in the face of the destruction of stuff S1 does not imply that E consists of a different kind of stuff, S2 (this is the burden of the plastic chess set example construed non-rigidly). It also shows how, if you can know that E will survive the destruction of stuff S1, you can infer that E is made of something other than S1 - but that this requires you to know antecedently what E is made of (this is the burden of the plastic chess example construed rigidly). The first argument is non-question-begging but invalid; the second is valid but question-begging. My objection, then, to Swinburne does not merely argue, as Hasker does, that Swinburne's argument is question-begging: it also provides a possible explanation of why Swinburne should nevertheless think that it is sound, by showing that the question-begging argument which he accepts is very similar to (and can indeed be expressed in the same words as) a non-question-begging kind of argument.

Swinburne in his reply to Hasker rejected the claim that his argument was question-begging; for he said, it was possible to accept the the key premise (i.e. X above) and not even understand the conclusion, let alone accept it as true. And the grounds for accepting the key premise are

...the coherence of various thought experiments described in two pages of my text [ES pp.151-2 - the ones described in section III above] including ones easily graspable by seven year old religious believers or readers of fairy stories.8

But this reply surely confuses two ways in which we can take the appeal to thought experiments. Of course a scenario that is not described in much detail and which occurs as part of an engaging narrative can be 'understood' and 'easily grasped by a seven year old'. It is in this sense that I might be able to 'grasp' narratives involving time travel, the splitting of one person into two or more people, one person 'changing bodies' with another, or a person 'proving' some mathematical hypothesis that is in fact self-contradictory. Imaginary scenarios of this kind (thought experiments, in one sense of the term) obviously tell us nothing at all about what is logically possible. A thought experiment which will tell us about possibilities and necessities must at the least be one in which the envisaged scenario can be spelled out in as much detail as anyone cares to demand, with no impossibilities emerging. In section III, I argued that Swinburne's thought experiment (of imagining my life continuing without a body) did not meet this condition: impossibilities did emerge when we tried to make sense in detail of what Swinburne was envisaging, even if they were not the sort which would prevent an enjoyment of the imagined scenario e.g. for fictional purposes. But even if that argument is wrong, there remains the argument of sections V to VII, which explains why, even if Swinburne's thought experiment is possible, it cannot non-question-beggingly be used to support his dualist conclusion.

 

 

 

 

NOTES

 

1. Howard Robinson, Matter and Sense (Cambridge: CUP 1982); Richard Swinburne, Is There A God? (Oxford: OUP 1996); John Foster, The Immaterial Self (London: Routledge 1991). See also the collection of papers in John R.Smythies and John Beloff (eds), The Case For Dualism (Charlottesville: University of Charlottesville Press 1989).

2. The Coherence of Theism (Oxford: OUP 1977) henceforth CT; Personal Identity (with Sydney Shoemaker) (Oxford: Blackwell 1984) henceforth PI; ; The Evolution of the Soul (Oxford: OUP 1986, revised edition 1997) henceforth ES. See also his response to some earlier criticisms in 'Dualism Intact', Faith and Philosophy 13, (1996): 68-77 (henceforth DI); in 'Reply to Stump and Kretzmann', loc. cit. 413-414; and in 'The Modal Argument Is Not Circular', Faith and Philosophy 15 (1998): 371-372 (henceforth MNAC). For some criticisms of these Swinburnean arguments, see Dean W. Zimmerman, 'Two Cartesian Arguments for the Simplicity of the Soul', American Philosophical Quarterly 28 (1991): 217-226; Paul Moser and Arnold van der Nat, 'Surviving Souls', Canadian Journal of Philosophy 23 (1993): 101-106; William Alston and Thomas Smythe, 'Swinburne's Argument for Dualism', Faith and Philosophy 11(1994): 127-133; Eleonore Stump and Norman Kretzmann, 'An Objection to Swinburne's Argument for Dualism', Faith and Philosophy 13 (1996): 405-412; and William Hasker, 'Swinburne's Modal Argument for Dualism', Faith and Philosophy 15 (1998): 366-370.

3. PI: 23-24

4. Op.cit. 30

5. Ibid. fn.16. The argument is repeated more-or-less verbatim in DI and in ES.)

6. Strictly, it is what Kripke calls a semi-rigid designator. 'My chess set' does not designate my chess set in all possible worlds, because my chess set does not exist in all possible worlds. But it does designate my chess set in all worlds in which my chess set (this very set) exists.

7. Swinburne, MNAC: 371.

8. Ibid.