Study Guide

Week Seven: Dualism and the Mind-Body Problem

Materials Assigned for the Week
The Central Point of This Week's Material
Other Concepts and Points you are Expected To Master This Week
Miscellaneous Comments and Clarifications

I. Materials Assigned for the Week

Reading:

Lecture 18: Dualism and the Mind/Body Problem
Readings: Descartes, Meditations One and Six (again), esp. pp. 217-220, 236, 242-251

Exercises:

Review Questions: pg. 274, esp. #3 & 4
Problems for Further Thought: pg. 274, esp. #3

II. The Central Point of This Week's Material

Mind-body dualism is the view that the mental side of a human being is not at all the same as the bodily side. That is, a human being (and maybe a few other inhabitants of the universe) are made up of two kinds of thing, rather than just one. Dualists argue that several well-known facts require a conclusion as radical as this (amongst them the fact that everything mental seems to pass Descartes' Method of Doubt test while nothing physical does). Sober sets out some of these arguments and assesses them. Putting aside their strengths or weaknesses, dualism faces one huge problem which continues to elude solution: if minds are so radically unlike bodies, how is it possible that there is such smooth causal interaction between the two as we observe in our own case?

III. Other Concepts And Arguments You Are Expected To Master This Week

The difference between the theory of mind-body dualism and the theory of mind-body identity
The difference between minds and brains; or the alleged difference anyway, for this is one of the sticking points between dualism and identity theory
What "Leibniz's Law" is, and how to apply it as a test to decide whether we are dealing with two things or just one thing in some case - be able to use it on some controversial as well as uncontroversial examples
The difference between a thing and the referring expression which refers to that thing or property
The two theories at loggerheads this week agree that we do have both a complex mental terminology and a complex physical terminology in our language; but the theory of mind-brain identity insists that both sorts of terms describe the very same items in the world; the theory of mind-body dualism says that our mental terminology and our physical terminology describe very different items in the world
What a so-called "propositional attitude" is and how it shows (or seems to show) the dualism of minds and bodies rather than the identity of minds and bodies
What the so-called "intensional fallacy" is
What Descartes' spatial argument for dualism is - and be able to assess its merits
The difference between the sense of a referring expression and the reference of a referring expression - and why the difference matters here
What causal interaction is
Why causal interaction is (supposedly) such a problem for any dualist story of minds and bodies but not for any identity story
What is meant by the derogatory, but accurate, label "Ghost in a Machine" (actually first used on pg. 275)

IV. Miscellaneous Comments and Clarifications

Sober touches on each of the issues which follow. But they are so close to my heart that I have not been able to resist adding my two cents worth. (You needn't be patient)

IV.A "Leibniz' Law"

This is the standard philosopher's test for deciding matters of identity and non-identity. So if the issue is to decide whether minds are identical or not identical to bodies (or to certain parts of bodies like brains), then this is the test to use. How does it work?

Consider a very simple example (the best sort). Suppose you know that X is a piece of chalk, and that Y is a piece of cheese: that is, Y is a not piece of chalk. In that case X must clearly be a distinct object from Y. (For if X were the very same object as Y, then one cannot be cheese and the other chalk.) Or suppose that both X and Y are pieces of chalk, but that X is longer than Y. In this case too we know there must be two pieces of chalk. (For if X were the very same object as Y, then X and Y would have to have the same length.) Or suppose that X and Y are both pieces of chalk, and in addition they are the same length; but in this case, X occupies a different spatial position from that occupied by Y. Again, X must be a distinct object from Y. (For if X were the very same object as Y, then X and Y would have to occupy the very same position.)

The principle we have been appealing to in these three simple cases is this:

if X has a property which Y does not have, then X is a distinct thing from Y.

The same principle can be stated in a slightly different form:

if X is the same thing as Y, then X has exactly the same properties that Y has.

The central idea of this identity principle may seem breathtakingly obvious. But its importance can hardly be over-estimated.

Leibniz' Law of Identity

The identity of things is a matter always to be decided by the identity of properties of those things. If the properties on the property lists for each thing match up exactly, then the things are identical. If there is the slightest mismatch between their property lists, then the things are different.

Here's how to apply Leibniz' Law to decide a specific case.

Step One: Take a referring expression; that is, a word or phrase which refers to some thing or purported thing. In Descartes' own case, for instance, we have the referring expression, "René Descartes" and also "René des Cartes" (another form of his proper name). In Descartes' theory of the mind, we have several referring expressions, "thing that thinks", "soul", "mind".
Step Two: Now write down the property list for the thing referred to by one of those referring terms. Think of God taking a cosmic inventory of the universe: a page for each different thing and a line for each different property. The property list will have a separate entry for each of the properties which we can say that that thing has. It will also have a entry for each property the thing doesn't have - after all, its not having so and so property is itself a property of the thing: for Santa Claus, God would enter "actually non-existent" as one entry just as readily as S/he would enter "believed by small earthlings to bring their Christmas presents".
Step Three: Write down the property list for the thing referred to by the second of the two referring terms you are testing.
Step Four: If every entry on the one property list is matched by the same entry on the other property list, and if there are no properties left over, then the first thing is identically the same as the second thing. (This is put a bit sloppily, because of course two different things can't be the same one thing. Better to say that when the property lists match we have two referring terms which refer to just one item in the universe.) If the property lists don't match - even the tensest bit, then we are dealing with two distinct things.

Example: Take Descartes' referring terms "soul" and "mind". Here are some of the properties to be found on their property lists (abridged):

Soul	Mind
weightless	weightless
odourless	odourless
intangible	intangible
invisible	invisible
does the thinking	does the thinking
survives bodily death	survives bodily death
known for certain to exist	known for certain to exist
introspectible by me	introspectible by me
not a toothbrush	not a toothbrush
etc.	etc.

Identity of property lists. Therefore identity of things with those property lists. Souls, at least according to Descartes' theory, are identical to minds.

This principle of identity is known as "Leibniz' Principle" or "Leibniz' Law" because it was first formulated in exactly this way by the seventeenth-century German philosopher and mathematician Gottfried Leibniz. Although Leibniz was the first to formulate it clearly, it was in wide-spread use long before his formulation. Plato used something like it for instance in the Republic. More to the point, Descartes certainly used it often.

IV.B The General Form of Any Argument for Mind-body Dualism

Most of the arguments for dualism have roughly the same logical form, relying heavily on Leibniz' principle of identity. The mind has a property which no physical object has (e.g. we have privileged access to mental states); or physical objects have a property which the mind clearly lacks (e.g. an exact spatial position). This means that the property list for the mind doesn't match up entry by entry to the property list for a physical object. At least one of the properties on one of the lists isn't going to appear anywhere on the other list. Therefore, by Leibniz' Law, it follows that the mind and the body must be distinct items.

Thus in its simplest form, an argument for dualism will look something like this:

Premise 1: My mind has property P.

Premise 2: My body does not have property P.

therefore

Conclusion 3: My mind /= my body.

Note that this is a deductive argument, not an induction or abduction.

Since Leibniz' Law has to be employed in order to derive that conclusion from those premises, it is usually best to make it into an explicit premise itself. (When you are "stretching out" an argument, try to include every proposition or assumption which actually does some work in that argument. It's the unexamined, because unexpressed, assumptions which usually send things awry.) As well, especially when the properties start to get a little weird or complex, it is useful to make it explicit too that what has just been shown in the first two premises is that minds and bodies therefore have different property lists, and so are suitable for the application of Leibniz' Law. Indeed, in cases of the so-called "Intensional Fallacy" [Sober, pp. 269-270 and Week 7 IV.D below], it will be precisely this seemingly innocent step which will be up for challenge. An easy example (compare Smart, pp. 349-350):

Premise 1: The Noble Savage knows that water quenches his thirst.

Premise 2: The Noble Savage does not know that H2O quenches his thirst.

therefore

Conclusion 3: Water and H2O have different properties on their property lists.

Premise 4: Leibniz' Law applies.

therefore

Conclusion 5: Water /= H2O.

Something has obviously gone very wrong here. But it will be impossible to figure out what if Step 4 hasn't even made explicit. More "stretching out" is obviously helpful then.

Thus the canonical form you should try to set things out in:

The structure of a typical argument for dualism

Premise 1: My mind has property P.

Premise 2: My body does not have property P.

therefore

Conclusion 3: My mind and my body have different properties on their property lists.

Premise 4: Leibniz' Law of identity applies: thing X and thing Y are different things whenever X has properties on its property list which Y does not have on its property list or vice versa.

therefore

Conclusion 5: My mind /= my body.

This argument is deductively valid. It is also wonderfully simple. But some of you may have noticed that it is not all that clear that the conclusion is going to be strong enough to power full-blooded Cartesian dualism. After all, dualism is not the theory that minds and bodies are merely different things. (We can get that much from Leibniz' Law even with two differently sized pieces of chalk.) It's the theory that they are different kinds of things. (Two different pieces of chalk aren't two different kinds of anything.) According to Descartes' story, that is, the mental and the physical constitute two distinct ontological classes of things, minds and bodies belong to separate ontological categories. Again, picture God keeping a ledger of all the things that exist in the cosmos. The theory of dualism says S/he will need at least two sheets of paper, not one: one sheet for all the mental things, another completely different sheet for all the physical things.

To capture this idea, we need an argument with a stronger conclusion:

Premise 1: My mind has property P.

Premise 2: My body does not have property P.

therefore

Conclusion 3: My mind and my body have different properties on their property lists.

Premise 4: Leibniz' Law of identity applies.

therefore

Conclusion 5: My mind and my body are fundamentally different kinds of things.

This stronger conclusion is certainly strong enough to support mind-body dualism, as against mind-brain identity for instance. But now it's not so clear that the premises are strong enough to support that conclusion! (Swings and roundabouts.) Delivering on both demands - premises which are strong enough to justify their conclusion, together with a conclusion which is strong enough to support dualism - is a key issue in evaluating the plausibility of every dualist argument.

IV.C First Argument for Dualism: The Argument from Doubt

Descartes' "Method of Doubt" delivers the indubitability of "I exist" at the beginning of Meditation Two. But it delivers it from the indubitability of "I think" ("I am conscious") - that is, from the indubitability of a certain mental item. This is a general characteristic of the mental for Descartes: everything mental is indubitable; everything physical is dubitable. An obvious candidate for Leibniz' Law treatment.

The Deception Version of Descartes' Argument

Treat the "Evil Genius" hypothesis which Descartes introduces towards the end of Meditation One as a thought-experiment entirely on a par with thought-experiments used in physics and mathematics. So imagine an entity is intent on deceiving you in all the ways it is logically possible for you to be deceived. He can deceive you into thinking you are reading a Study Guide when you are actually gazing blankly out the window. He can make you believe that you are up and about when really you are in bed sound asleep and dreaming. He can trick you into making you feel embarrassed even when you haven't done anything embarrassing. He can fool you into thinking there is a pool of water at the end of the road when there's no water at all there, only smooth asphalt. He can make you feel confident that you remember things which never in fact happened. Indeed, he can make you think falsely about practically anything. However, there is one signal one exception to all this trickery. No matter how hard he tries, no Evil Genius can deceive you into thinking that you are thinking when really you aren't thinking. For a simple reason. In order to fool you into thinking wrongly, you would have to be doing some thinking in the first place!

Now simply extend this reasoning to any mental event or operation or state. For instance, the Evil Genius will have no more success fooling me into thinking I am embarrassed when I'm not embarrassed than he has fooling me into thinking I'm thinking when I'm not thinking. He may well trick me into feeling embarrassed when there is no good reason for me to feel embarrassed (I haven't made a blue). But He can't fool me about whether I am feeling embarrassed when I am feeling embarrassed; he can't whisper persuasively in my ear: "Oh that's not embarrassment you are feeling right now, it's actually a pain in your left toe"; when I feel embarrassed, I can't be tricked into thinking I am feeling something else. According to the dualist, this is characteristic of everything mental. The Evil Genius can't make me think falsely that I'm bored out of my skull when I'm actually so excited I can hardly sit still. Neither can he fool me into believing falsely that I am feeling excitement when what I am actually feeling is boredom. When I'm any of these mental states I know I am, and that's the end of the matter. This means that there are strict limits on what it is logically possible for me to be deceived about, even by a being who has the power to deceive me in all possible ways. Such considerations deliver (or seem to deliver) a tidy argument indeed:

Premise 1: Even an all-powerful all-malevolent Evil Genius doing his worse cannot make me doubt or deceive me about the existence and character of my mind.

Premise 2: The Evil Genius can easy as pie make me doubt and deceive me about the existence and character of my body.

Restatement 3: On the property list for my mind is the property: "the Evil Genius cannot make me doubt it or deceive me about it".

Restatement 4: On the property list for my body is the property: "the Evil Genius can make me doubt it and can deceive me about it".

therefore

Conclusion 5: My mind and my body have different property lists.

Premise 6: Leibniz' principle of identity applies: X and Y are different things whenever X has properties on its property list which Y does not have on its property list.

therefore

Conclusion 7: My mind /= my body.

Restatement 8: My mental states belong to a fundamentally different ontological category than my bodily states.

Everyone has heard of Descartes' stroke of genius cogito ergo sum, "I think, therefore I am". This argument is just one way of putting the cogito to the service of dualism. Obviously, it depends heavily on the element of doubt and deception introduced by the Evil Genius thought-experiment. But everything here can be put in more "neutral" terms.

The Incorrigibility Version of Descartes' Argument

Just go back to what cogito ergo sum was supposed to deliver. It was supposed to deliver the first of Descartes' "foundational" truths: "I exist". Now, what was so special about foundational truths? That they must be true no matter what the circumstances are; that they are immune from all logical possibility of error, that it is logically impossible - and not merely "technically impossible" or "circumstantially impossible" (Sober, pp. 179-180) - for us to be mistaken about them. This gives another candidate for Leibniz' Law.

Premise 1: It is logically impossible for me to have false beliefs about the existence and character of any of my mental states.

Premise 2: It is logically possible for me to have false beliefs about the existence and character of any of my bodily states.

Restatement 3: On the property list for my mental states is the property: "logically impossible for me to have false beliefs about it".

Restatement 4: On the property list for my bodily states is the property: "logically possible for me to have false beliefs about it".

therefore

Conclusion 5: My mental states and my bodily states have different property lists.

Premise 6: Leibniz' principle of identity applies: X and Y are different things whenever X has properties on its property list which Y does not have on its property list.

therefore

Conclusion 7: My mental states /= my bodily states.

Restatement 8: My mental states belong to a fundamentally different ontological category than my bodily states.

A Modern Version of Descartes' Argument: The "Privileged Access" We Have to Our Own Minds

There is a more contemporary way to run the basic lesson Descartes was trying to run with talk of "the Evil Genius" and "logical possibility". Focus your attention on the question: "Why is everything mental undoubtable and everything physical doubtable?" It is because there are two entirely different methods for arriving at knowledge of factual matters and not just one. Descartes was merely the first philosopher to show us this enduring truth. But enduring truth it is, and we can use it for dualism nowadays as powerfully as before.

Method One is what is sometimes called introspection. It is the method we use in order to find out whether it's regret or relief that we are feeling as someone walks out the door. It's the method we use to access what colour this curtain swatch looks and whether it looks the same colour as that book-cover. It's the method we employ when the dentist asks us whether the anaesthetic is beginning to kick in yet, and especially how we tell when it hasn't! Here are some distinguishing features of this Method One.

It has special limits. I can use it to find out stuff about myself, but you can't use it to find out about me and I can't use it to find out about you. Thus if you want to find out whether I have a headache, you have to ask me (no introspection there). If I want to find out whether you have a headache I have to ask you (no introspection again). It is only if I want to find out whether I myself have a headache that this method comes into play - I introspect my own headaches.
Introspection somehow by-passes the normal physical sense-organs of my body. I don't see my headache with my eyes; I don't touch it with parts of my skin, nor hear it with my ears; I certainly don't taste it by sticking my tongue in it. I don't even use any of the kinaesthetic senses (the sense-organs in my muscles which tell me where my limbs are). It is hard to know how introspection manages to pull off getting knowledge without using any of the body's physical sense-organs. But it's still a plain fact that it does. We do know perfectly well how terrible our headaches are, when they are starting, whether aspirin is beginning to work, whether the headache is more throbbing or dull, and the like.
Because of the privacy and immediacy of its method of access, introspection delivers results which are uncorrectable by anyone else but myself. I may lie, but provided I am not lying, my self-reports are unimpeachable. I can't be wrong. After all, you can't get inside my mind to have my headache.; you have to ask me how it feels, remember; so how could you ever gain a right to say more about its existence or character than I have? If I walk into the doctor's and say: "I've been having this throbbing headache off and on for two weeks. What's wrong?", the doctor can help a lot. But the one thing s/he can't do is correct me: "Sorry friend, you don't have a throbbing headache at all. I can find nothing wrong with you. You must be mistaken. You're not in pain; you're just bored with daytime television." This is where Descartes' cogito ergo sum starts to bite. When I say I have a throbbing headache, no doctor is in a position to override those introspective reports. S/he isn't in there with me, having my headache but diagnosing it more skilfully than I!

Method Two is better called inspection or observation than introspection. It's the method which the budding doctor learns when he learns all about how to probe, palpitate and poke people. The dentist uses it to discover which teeth are rotten, regardless of whether they happen to be hurting me right now or not. It's the method by which we find out that our skin is inflamed, that our fingers are bleeding, that our glasses have a crack in them. The features of such ordinary sensory observation or inspection are familiar.

Method Two has limits too, but the opposite limits to the method of introspection. Anyone can use it, and whatever information it provides to one person using it, it will provide the same information to anyone else, including myself. It does not give anyone any privileged access to anything. Thus you are not forced to rely on my testimony about, say, the nosebleed I am having. You can look and see for yourself, just as well as I can. (In fact, I have to use a mirror!)
It employs the body's sense-organs and cannot avoid using them. What I, or anyone, can look and see, can be looked at and seen only with the physical eyeballs. We are all in the same boat here too. There are no privileged exemptions from the use of ordinary physical sense-organs for finding out about my nosebleed.
Finally, there is no privileged status to the results gained by inspection. For instance, I can look and see my nosebleed. You can look and see my nosebleed. But I can also go wrong when I say I have a nosebleed (the mirror was cracked, someone splattered me with ketchup). And you can correct me by looking and seeing more carefully. When I make a self-report on the basis of inspection, "My nose is bleeding", you have equal right with me to make corrections on the basis of further inspection.

What the theory of mind-body dualism wants to do with these commonplaces is simple. We are supposed to realise that not only do we have two different methods of acquiring knowledge here. As well, those methods are methods for acquiring knowledge about different kinds of things. We use introspection to access mental kinds of things. We use inspection to access physical kinds of things. This claim generates a tidy argument:

Premise 1: I have privileged access to the existence and contents of my own mind.

Premise 2: Everyone, including myself, has exactly the same access to the existence and character of my body.

Restatement 3: On the property list for my mind is the property: "known by a method privileged to me and with a privileged degree of secureness".

Restatement 4: On the property list for my body is the property: "known by the same methods anyone else can use and with the same degree of secureness anyone else can have".

therefore

Conclusion 5: My mind and my body have different property lists.

Premise 6: Leibniz' Law of identity applies: X and Y are different things whenever X has properties on its property list which Y does not have on its property list.

therefore

Conclusion 7: My mind /= my body.

You are probably sick to death of my inserting Leibniz' Law of Identity as a separate premise in all these arguments. I do it deliberately: Leibniz' Law in these cases is sometimes attacked as a false premise! More accurately: some philosophers argue that it is false to claim Leibniz' Law applies to the properties used in such arguments as the argument from doubt. And they have a point.

IV.D Criticism: Do These Arguments Commit "The Intensional Fallacy"?

After all, the properties that are going on the property lists in steps 3 and 4 of the arguments above are all slightly special. They involve what are called "intensional contexts", in which we use "intensional" or "epistemological" words such as:

"believe that...",
"doubt whether…",
"know that…",
"want to know that...",
"wonder whether…",
"can't figure out if…",
"assume that…",
"mistake for…",
"be deceived about..." and so on.

It's not hard to find arguments which are exactly parallel to Descartes' ones, using exactly the same sorts of epistemological or intensional verbs, but which deliver ludicrously false conclusions. (Remember the "Noble Savage"?) Where these arguments go sour, the finger of blame usually points to using Leibniz' Law in circumstances where it shouldn't be used. What I believe or don't believe about X, what I know or don't know about X, what I fear or don't fear about X, and so on, should never be facts which go on thing X's property list. Rather, since each of those facts is only telling us something about me, not about the thing X in itself, if it's going to go on any item's property list, it should go on the property list for me, not for X.

The "Intensional" Fallacy

Premise 1: An intensional or epistemological property gets assigned to item X (e.g. I believe so and so about X, I know X in a certain way, I can be deceived about X) when that property is a property properly to be assigned only to me (i.e. my believing, my knowing, my being deceived about something are properties of me not of item X).

Premise 2: The opposite intensional property gets assigned to item Y (e.g. I don't believe so and so about Y, I know Y in a different way, I can't be deceived about Y) when that property, again, is properly to be assigned only to me (i.e. it's my believing and so on again).

Premise 3: Leibniz' Law is assumed to apply to the case. (That is, it is assumed that Leibniz' Law is correct to count the intensional properties assigned to items X and Y as among the properties genuinely belonging on item X's property list and on item Y's property list.)

therefore

Conclusion 4: Item X /= item Y.

The fallacy is that Leibniz' Law delivers the Conclusion only by being applied to wrongly constructed property lists in the Premises.

Example:

Premise 1: I know that aspirin relieves my headache.

Premise 2: I do not know that acetylsalicyclic acid relieves my headache.

Restatement 3: On the property list for aspirin is the property: "I know it relieves my headache".

Restatement 4: On the property list for acetylsalicyclic acid is the property: "I do not know it relieves my headache".

therefore

Conclusion 5: Aspirin and acetylsalicyclic acid have different property lists.

Premise 6: Leibniz' Law applies in this case.

therefore

Conclusion 7: Aspirin /= acetylsalicyclic acid.

The epistemological verbs which make this an "intensional context" are underlined. The conclusion of this little argument is patently false; the key Premises 1 and 2 look to be just as patently true; and the logical structure seems valid enough. So where's the little stinker? There are not many places left. What's wrong is that Premise 6 is in fact false: Leibniz' Law does not apply in this case. What I know and what I do not know, especially about the drugs in my cabinet, are facts about me. They are not facts about the chemicals.

So the property "known by me to relieve my headache" ought not to go on aspirin's property list (as in Restatement 3), nor should "not known by me to relieve my headache" go on acetylsalicyclic acid's property list (as in Restatement 4). Both go only on the property list for me. When Leibniz' Law is invoked at Premise 6 and the property lists for aspirin and acetylsalicylic acid are allowed to include such intensional properties, then of course the whole affair will come to grief. We have in effect allowed Leibniz' Law to deliver its judgement on the basis of improperly inflated property lists (Conclusion 5). Little wonder then that it delivers an incorrect judgement.

This happens pretty obviously in the case of aspirin and acetylsalicylic acid. The very same thing happens in the case of mental states and bodily states. It's a bit harder to see the fallacy when you have Descartes' razzle-dazzle about the Evil Genius and cogito ergo sum to contend with. (Neither Descartes nor any of his contemporaries ever saw it for instance.) But it's the same basic mistake anyway.

It's even more obviously a mistake if we use the chemical formula:

Premise 1: I know that aspirin relieves my headache.

Premise 2: I do not know that CH3COOC6H4COOH relieves my headache.

Premise 3: Leibniz' Law applies in this case.

therefore

Conclusion 4: Aspirin /= CH3COOC6H4COOH.

Whole libraries contain what I don't know about chemistry. But none of that is relevant to deciding what's what in chemistry. If a chemist determines (on the basis of their chemical property lists) that aspirin = CH3COOC6H4COOH, who am I to balk at that identity, purely on the basis that it's news to me?!

A final less technical example of the intensional fallacy:

Premise 1: Lois Lane wants to marry Superman.

Premise 2: Lois Lane does not want to marry Clark Kent.

Premise 3: Leibniz' Law applies in this case.

therefore

Conclusion 4: Superman /= Clark Kent.

(This is the example Sober works on, pp. 269-272.) No one in her right mind wants to marry Clark Kent, the wimp of all time. But that fact is not actually a fact about Clark Kent. The fact about Clark Kent is that he behaves like a wimp when he is in his disguise as a reporter for the Daily Globe. The fact that Lois Lane doesn't want to marry a wimp is a fact about Lois Lane. So is the fact that she does want to marry a hunk. Belonging on her property list only, we can say either of two things. We can say that Leibniz' Law does apply, but that it doesn't deliver Conclusion 4 (because Premise 1 and Premise 2 don't put different properties on Superman's and on Clark Kent's property lists, since they put all the Lois Lane stuff on her property list only). Or we can say that Leibniz' Law doesn't apply, at least not in the intended way - to the intended but incorrectly inflated property lists (because Leibniz' Law never applies to property lists which include intensional properties); in which case Conclusion 4 does follow but Premise 3 is simply false. Which we say is six of one, half dozen of the other. Either way, the argument for the non-identity of Clark Kent and Superman fails to be sound.

Some of you will have noticed that this is not how Sober treats his Lois Lane case. He doesn't go in for "intensional predicates", but rather "propositional attitudes". Lay these out as alternatives.

In both of the statements "X wants to marry Y" and "X does not want to marry Z", the verb or predicate expression "wants to marry" is an intensional predicate and therefore does not mark properties which properly belong on Ys or Zs property list (Superman's property list versus Clark Kent's property list), but only on Xs property list (Lois Lane's). "Wants to marry" says something about the person who wants to do the marrying, not about the person she wants to marry. So differences here can't be racked up on the property lists for Superman and Clark.
In both of the statements "X wants to marry Y" and "X does not want to marry Z", the verb phrase "wants to marry" is actually a propositional attitude. That is, both sentences are just English shorthand for what would be more accurately expressed as "X wants that the proposition p be true" and "X wants that the proposition q be false". In the case at hand, then, we have the two propositional attitudes: "Lois Lane wants the proposition Lois Lane marries Superman to be a true proposition" versus "Lois Lane wants the proposition Lois Lane marries Clark Kent to be a false proposition".

Now the funny thing about propositional attitudes is this. They are attitudes to those propositions themselves. They are not attitudes to what those propositions are about. Thus the way in which the propositions themselves get expressed in the minds of the person with the "attitude" to them - in this case the person with the attitude to them uses the concept of Superman versus the concept of Clark Kent - matters more to the identity or non-identity of the propositional attitudes than does the fact that, as it happens, both the expression "Superman" and the expression "Clark Kent" eventually anchor at the same entity. The propositional attitude is thus an attitude to the sense of the propositions involved rather than an attitude to the reference of those propositions. And for sense and reference Leibniz' Law breaks down too: characterisations with different senses can still have the same reference, so non-identity of senses is no guarantee by itself to non-identity of references. (Compare Sober, pp. 270-272.)

Hard exercise!
Why are these basically the same diagnoses?

IV.E Second Argument for Dualism: The Spatial Argument

The spatial argument is due to Descartes and is the second discussed by Sober (pp. 272-273). The argument occurs in Meditation Six (pg. 249). Again, the basic idea is very simple. Whereas it makes sense to attribute spatial properties to bodies, it does not make sense to attribute them to minds. With each body it makes sense to ask where it is, how big it is, what volume it takes up, how densely it is crammed into the space it takes, whether its occupancy excludes anything else occupying the same spatial position, and so on. But these questions applied to the mind seem to be nonsense - unless one is already convinced that the mind is identical to the brain, or something like that.

The Crude Form of the Argument

At its most basic, the argument looks like this:

Premise 1: My physical body and each part of my physical body is extended in space. (It has length, breadth and height and therefore volume; it occupies space and excludes anything else from occupying that space; it is spatially to the right or to the left of other things with a spatial location, and so on.)

Premise 2: My mental states are not extended in space. (Certainly none have the spatial properties above.)

Restatement 3: On the property list for my mental states is the property: "not extended in space".

Restatement 4: On the property list for my physical body and each part of my physical body is the property: "extended in space".

therefore

Conclusion 5: My mental states and my physical body (and each part of my physical body too - such as my physical brain) have different property lists.

Premise 6: Leibniz' principle of identity applies: X and Y are different things whenever X has properties on its property list which Y does not have on its property list.

therefore

Conclusion 7: My mental states /= my physical body or a part of my physical body.

A Less Crude Version: The Argument From Divisibility

The argument from the divisibility of physical bodies is closely related to the argument above. (This is the form of the spatial argument actually to be found in Descartes' Meditation Six.)

Premise 1: Every physical body is (in principle) spatially divisible. (For instance, you can cut a physical object into two physical objects, and they will then occupy different volumes of space, and one will be at a certain distance in a certain direction from the other and so on.)

Premise 2: My mind is not (even in principle) spatially divisible.

Restatement 3: On the property list for my mind is the property: "not spatially divisible".

Restatement 4: On the property list for physical bodies is the property: "spatially divisible".

therefore

Conclusion 5: My mind and physical bodies have different property lists.

Premise 6: Leibniz' principle of identity applies: X and Y are different things whenever X has properties on its property list which Y does not have on its property list.

therefore

Conclusion 7: My mind is not a physical body.

Now it is true enough, of course, that we talk about our minds being divided about something, being in two minds about something, and so on. But this sort of idiom need not be taken seriously as an objection to Premise 2 of this divisibility argument. Firstly, it is not spatial division that is at issue. Secondly, when a mind is divided in the sense in which we speak of "being in two minds about something", nonetheless it is one and the same mind which is aware of the attractions of two different alternatives; we aren't literally "in two minds".

A Modern Version of the Spatial Argument

A more up-to-date form of this same line of reasoning will stress the oddness of spatial talk more strenuously. Reporting that "My anger at George is three millimetres tall" or "My regret at having to give you a mark of C is an inch to the right of my wish to see you do well on the exam" seems just plain loony, as loony as saying "Wednesday is fat" or "Colourless green ideas sleep furiously". The oddness in such cases is sometimes expressed in terms of committing a "category mistake" (an expression coined by Gilbert Ryle, pp. 278-280). Thus:

Premise 1: It is always a nonsense or a category mistake to ascribe spatial properties to a mental state. (I mean properties such as having a length and breadth and height, or being on top of or underneath or behind or in front of some other mental state.)

Premise 2: It is always perfectly in order and never a category mistake to ascribe spatial properties to any physical state. (Everything physical does have some length, height and breadth, and is located on top of or underneath ... some other physical state.)

Restatement 3: On the property list for mental states is the property: "ascribing spatial properties to them commits a category mistake".

Restatement 4: On the property list for physical states is the property: "ascribing spatial properties to them does not commit a category mistake".

therefore

Conclusion 5: Mental states and physical states have different property lists.

Premise 6: Leibniz' principle of identity applies: X and Y are different things whenever X has properties on its property list which Y does not have on its property list.

therefore

Conclusion 7: Mental states /= physical states.

IV.F Criticism: Do These Arguments Beg the Question?

Each version of the spatial argument is formally valid (the premises deductively entail the conclusion). Hence someone who denies the conclusion (that is, who does not want accept dualism) must deny one of the premises. But the first premise in each argument (the first premise is the key one) seems compelling. Has the dualist finally made their case then? Maybe. But maybe not. In a subtle way, the three arguments we have been looking at all seem to beg the question at issue. Go back to Descartes' Meditation Two:

"But what then am I? [That is, what is this I whose existence has been proven indubitable?] A thing that thinks. What is that? A thing that doubts, understands, affirms, denies, wills, refuses, and which also imagines and senses.

"It is truly no small matter if all of these things pertain to me. But why should they not pertain to me? Is it not I who now doubt almost everything, I who nevertheless understand something, I who affirm that this one thing is true, I who deny other things, I who desire to know more things, I who wish not to be deceived, I who imagine many things against my will, I who take note of many things as if coming from the senses? Is there anything in all of this which is not just as true as it is that I am, even if I am always dreaming or even if the one who created me tries as hard as possible to delude me? Are any of these attributes distinct from my thought? [Obviously, no.]" (pg. 220)

This, and similar passages, amount to a definition of mind:

The mind is that thing which has thoughts, sensations, emotions, desires, doubts.

Here is the point. Unless we are covertly dragging in non-spatiality by the back-door, there is nothing in that particular definition to rule out the possibility that a mind (the thing which does the thinking, sensing, emoting, desiring, doubting) does indeed have spatial dimensions. (Compare Sober, pg. 273.) For there is nothing in that definition, nor in any of the premises of any of the spatial arguments, to rule out the possibility that brains are the things which do the thinking, sensing, emoting, desiring, doubting and the rest. If the thing which does the thinking, sensing, emoting and so on is just the brain, then the mind, on this account, would have exactly the same physical dimensions as the brain - because it would be the brain. A mind would have a spatial dimension and it would be physically divisible. We shall turn to this rival theory of mind-brain identity in Week Nine.

Probably the reason Descartes doesn't twig to this, is that he keeps switching between it and a subtly different definition of mind:

The mind is that thing whose existence and character is beyond all logical possibility of doubt.

This item is what the Method of Doubt delivers as indubitable, incorrigible, immune to the best efforts of the Evil Genius to deceive us about and so on. Anticipating Week Nine, what will the mind-brain identity theory have to say about the mind if it is defined this way? That there is no such thing of course! And that the Method of Doubt is a rotten way of making it seem there is! See the Intensional Fallacy.

Contents
Previous Week: Hume's Problem of Induction
Next Week: Behaviourism and "Scientific Method"