D Nagel's Brain
D Nagel's Brain
Nagel believes that what he is, essentially, is his brain. He gives three arguments for his belief. These are contained in an unpublished rough draft, which he may revise; but, even in this rough form, they demand attention. I shall quote from this draft.
Two of Nagel's arguments appeal to a view about meaning, reference, and necessity. The objects we refer to have some essential properties: properties that these objects must have, since, if they lacked them, they could not exist. Some properties are essential because of the meanings of our words. Thus, because of what ‘triangle’ means, it is an essential (p.469) property of triangles that they have three sides. But on the view that I shall now discuss, there is a different way in which objects can have essential properties. These properties are not essential because of the meanings of our words. We discover these essential properties when we discover facts about what it is that we are referring to. On this view, for example, we have discovered that an essential property of gold is having the atomic number 79. Every substance with this number must be gold, and no substance without this number could be gold. This was not part of the meaning of the word ‘gold’.9
What are we referring to when we use the word ‘person’ and the word ‘I’? Nagel writes: ‘What I am is whatever in fact makes it possible for the person TN to identify and reidentify himself and his mental states’. What I am is whatever explains the psychological continuity of my mental life. And Nagel similarly claims that this leaves it open what the explanation is. If the carrier of continuity is a Cartesian Ego, that is what I really am. He continues: ‘If on the other hand certain states and activities of my brain underlie the mental capacity, then that brain in those states . . . is what I am, and my survival of the destruction of my brain is not conceivable. However, I may not know that it is not conceivable, because I may not know the conditions of my own identity.’ As he later writes, ‘in trying to conceive of my survival after the destruction of my brain, I will not succeed in referring to myself in such a situation if I am in fact my brain’.
Nagel is a Reductionist. He agrees that personal identity does not involve the ‘further fact’, which in every conceivable case either holds completely or not at all. Personal identity just involves physical and psychological continuity. But, though he is a Reductionist, Nagel's view differs in two ways from the view that I and some others defend. On this view what fundamentally matters is Relation R: psychological continuity and connectedness. On Nagel's view, personal identity is what matters. And, because he believes that he is his brain, he believes that what fundamentally matters is the continued existence of this brain.
This may seem to be a disagreement only about imaginary cases. There are no actual cases where there is psychological continuity without the continued existence of the same brain. If the disagreement was only about imaginary cases, it would be hardly worth discussing. But this disagreement also covers actual cases, and our own lives. On my view, one of the two relations which matter, psychological connectedness, holds over time to reduced degrees. This is an essential premise of my argument, in Chapter 14, against the Self‐interest Theory. This argument would be undermined if Nagel's view is true. The continued existence of the same brain, in our actual lives, is not a matter of degree.
This claim needs one qualification. Nagel leaves an important question open. Reconsider Williams's Example, where the surgeon tampers with my brain so as to remove all psychological continuity. Would the resulting (p.470) person be me? Though my brain has been tampered with, it is clearly the very same brain. If I am my brain, I shall still exist. But in one of the remarks quoted above, Nagel suggests that what I am is not just my brain, but my brain in certain states. Perhaps these are the states which provide psychological continuity. On this version of his view, I would not exist at the end of Williams's Example.
The two versions of Nagel's view might be re‐expressed as follows. On the simpler version, what I am is what normally causes my psychological continuity. But I would be this thing even when it does not cause psychological continuity. We have discovered that what I am is my brain. At the end of Williams's Example the surgeon has removed all psychological continuity. But since my brain will still exist, I shall still exist.
On the less simple version of Nagel's view, what I am is what causes my psychological continuity, in the particular states which make it be this cause. On this version of the view, my identity does not just involve the continued existence of my brain. It also involves psychological continuity. This version coincides with a view discussed above: the Narrow Psychological Criterion, which appeals to psychological continuity with its normal cause.
Only the first version of Nagel's view disagrees in actual cases with the view that I defend. Should we accept this version of the view? This partly depends on whether Nagel correctly describes the meaning of the words ‘person’ and ‘I’. There is another complication. Nagel makes two claims about what he and others mean by the word ‘I’. One is that he uses ‘I’ intending to refer to whatever explains his psychological continuity. The other is that he uses ‘I’ intending to refer to ‘the unobserved subject’ of his experiences.
I start with the second claim. This is hard to deny. I am not a series of thoughts, acts, and experiences. I am the thinker of my thoughts, and the doer of my deeds. I am the subject of all of my experiences, or the person who has these experiences.
Nagel claims that, when I use the word ‘I’, intending to refer to myself, the subject of my experiences, I am in fact referring to my brain. Should we accept this claim?
We should first note that an attempted reference may fail. Call what we are trying to refer to our intended referent. There may be some object which fits one of our beliefs about our intended referent. But this may not be enough to make this the object that we are referring to. We may have too many other beliefs about our intended referent, which would be false when applied to this object. We would then not be referring to this object. And we may be referring to nothing.10
One example would be this. The ancient Greeks believed that the God Zeus was the cause of lightning and thunder. Zeus did not exist, and the Greek word ‘Zeus’ referred to nothing. We should not claim that, since the Greeks believed that Zeus was the cause of lightning and thunder, and this (p.471) cause is an electrical condition in the clouds, Zeus is such a cloud‐condition, and this is what the Greek word ‘Zeus’ referred to. A cloud‐condition is too unlike a God to be the referent of the Greek word ‘Zeus.’
In using the word ‘I’, I intend to refer to myself, the subject of my experiences. Nagel believes that, when using ‘I’, most of us have false beliefs about our intended referent. Even if we are not aware of this, most of believe that our identity must be determinate. We believe that we are entities whose continued existence must be all‐or‐nothing. This belief would have been true if each of us had been a Cartesian Ego. But Nagel believes that there are no such entities. There are no entities with the special properties that we believe to be had by the subject of our experiences. In this respect, the case is like that of the Greek word ‘Zeus.’ But Nagel claims that, in using ‘I’, we do not fail to refer. What ‘I’ in fact refers to is my brain. Nagel admits that our brains do not have the special properties which we believe to be had by what ‘I’ refers to. But, while a cloud‐condition is too unlike what the Greeks believed that ‘Zeus’ referred to, Nagel claims that our brains are not too unlike what we believe that ‘I’ refers to. As he writes, this is ‘one of those cases where some of our most important beliefs about the referent of one of our concepts may be false, without its following that there is no such thing’.
Should we accept this view? Nagel believes that we are not separately existing entities, distinct from our brains and bodies, and our experiences. And he seems to believe that, if the word ‘I’ does not refer to my brain, there is nothing else that it could refer to. My brain must be the subject of my experiences, since, in the absence of Cartesian Egos, there is nothing else that could be the subject of my experiences. Thus, after denying that we are separately existing entities, he asks (1) ‘why not go all the way with Parfit and abandon the identification of the self with the subject of the mental. . . ?’ And he answers (2) ‘that the actual subject is what matters’, even if it is not the kind of entity in which we are inclined to believe. (1) assumes that, on the Reductionist View that I defend, we cease to believe that there are subjects of experiences. (2) assumes that the subject of experiences is the brain.
I deny both of these assumptions. On the Reductionist View that I defend, persons are not separately existing entities. The existence of a person just involves the existence of his brain and body, and the doing of his deeds, and the occurrence of his mental states and events. But though they are not separately existing entities, persons exist. And a person is an entity that is distinct from his brain or body, and his various experiences. A person is an entity that has a brain and body, and has different experiences. My use of the word ‘I’ refers to myself, a particular person, or subject of experiences. And I am not my brain.
It may help to return to Hume's analogy. We can be Reductionists about nations, but still believe that nations exist, and can be referred to. A nation is not a separately existing entity, something other than its citizens, and the land they inhabit. A nation's existence just consists in the existence of its (p.472) citizens, acting together in various ways on its territory. Though this is all there is to the existence of a nation, we can refer to nations, and claim that they exist. Thus we we can truly claim that France exists, and that France declared war on Germany in 1939. In contrast, there is no nation called Ruritania. We can make the same claims about people. Some people exist, and can be referred to, while others never exist, and cannot be referred to. I and Thomas Nagel are two of the people who ever exist, and can be referred to. But we cannot refer to my non‐existent Roman ancestor, Theodoricus Perfectus.
My next claim has, in this discussion, special importance. When we use the word ‘France’ to refer to a nation, we are not referring to something other than a nation. We are not referring to this nation's government, or to its citizens, or to its territory. This can be shown as follows. If ‘France’ referred to the French government, France would cease to exist if the government resigned and there was a period of anarchy. But this is false. Nations continue to exist during periods when they have no government. Similarly, if ‘France’ referred in 1939 to those who were then French citizens, France would cease to exist when these citizens cease to exist. This is also false. And if ‘France’ referred to these citizens, it must have been these citizens that declared war on Germany. This is also false. There is a use of the word ‘France’ which refers, not to the nation but to the country, or this nation's territory. When we claim that France is beautiful, we are referring to its land and its buildings. But, on the other use, ‘France’ refers to the nation, not to its territory. If ‘France’ referred to French territory, France could not cease to exist unless that territory ceased to exist. This is also false. What was once the territory of the nation Prussia still exists. But Prussia has ceased to exist.
As the case of nations shows, we can refer to something though it is not a separately existing entity. And in referring to such a thing we are not referring to the various other entities that are involved in its existence. If we are Reductionists about persons, we can make similar claims about our use of the word ‘I’. This can refer to a person, or subject of experiences, even though a person is not a separately existing entity. And when we decide that a person is not a separately existing entity, we are not forced to conclude that a person must be either his brain, or his whole body. Though nations are not separately existing entities, we are not forced to conclude that a nation must be either its government, or its citizens, or its territory, or all three. A nation is none of these three. And we can refer to nations. Similarly, we are not forced to conclude that a person is his brain, or his whole body. And we can refer to persons.11
I agree with Nagel that most of us have false beliefs about the intended referent of the word ‘I’. Most of us believe that we are entities whose continued existence must be all‐or‐nothing. It may be objected that, since there are no such entities, we ought to conclude that, as used by most of us, the word ‘I’ fails to refer, just as ‘Zeus’ fails to refer. Like Nagel, I can reject (p.473) this claim. ‘Zeus’ does not refer because a cloud‐condition is too unlike a God. Persons on the Reductionist View are unlike persons on the Non‐Reductionist View. But they are much more similar than Gods and cloud‐conditions. I can therefore claim that persons exist. And since persons exist, though in a different way from that in which we are inclined to believe, our attempts to refer to persons can be claimed to succeed. Like Nagel, we can claim that this is one of the cases where we have false beliefs about our intended referent, without its following that we are not referring to this thing.
Since I am a person, who exists, I seem to be the best candidate for what my use of ‘I’ refers to. Nagel might reply as follows. It is true of most of us that we believe that we are separately existing entities. On the Reductionist View, this belief is false. But this belief would be true when applied to a person's brain. This may be claimed to make a person's brain a better candidate for what his use of ‘I’ refers to.
There is something in this claim, but not, I think, enough. If I use the word ‘X’ trying to refer to the object called X, and X exists, the natural assumption is that I do refer to X. This assumption might not be justified if it was both true that X lacks most of the properties that I believe it has, and true that some other object Y has most of these properties. This might justify the claim that, though I am trying to refer to X, I am in fact referring to Y. When someone uses the word ‘I’, his intended referent—what he is trying to refer to—is himself. This person may believe that he is a separately existing entity, distinct from his brain and body. It would then be true, as Nagel claims, that this person's intended referent lacks one of the properties that he believes it has, while another entity, his brain, has this property—being a separately existing entity (though not, of course, distinct from itself). But a person's brain does not have most of the properties that most of us believe we have. It does not, for example, have a continued existence that must be all‐or‐nothing; nor is it indivisible. When most of us use the word ‘I’, our brains are not very similar to what we believe to be our intended referents. This counts against the claim that, when we use ‘I’, we are in fact referring to our brains. What counts in favour of this claim that is that our brains have one of the properties that we mistakenly believe we have: that of being separately existing entities. But I believe that this is not enough to justify Nagel's claim. It is not enough to falsify the natural answer that we are referring to our intended referent. When we use ‘I’ we are trying to refer, not to our brains, but to ourselves. Our brains have one property that we mistakenly believe ourselves to have; but this is not enough to show that, when we try to refer to ourselves, we fail. We can retain our natural belief that we can refer to ourselves.
Nagel's second argument appeals to his other claim about the meaning of the word ‘I’. This is the claim that, in using ‘I’, we intend to refer to whatever explains our psychological continuity. I believe that we can reject (p.474) this claim. There is a contrast here between Nagel's arguments. Each involves one claim about meaning, and one claim about the facts. I accept the claim that I use ‘I’ intending to refer to the subject of my experiences. But I have denied the claim that the subject of my experiences is my brain. In considering this argument for Nagel's View, I accept his claim about meaning but deny his claim about the facts. In considering his other argument I accept his claim about the facts. What explains my psychological continuity is the continued existence of my brain. But I deny his claim about meaning. I deny the claim that I use ‘I’ intending to refer to whatever explains my psychological continuity.
Nagel's third argument appeals to an imaginary case, in which what seems to him to matter is the survival of his brain. He describes a case like that of Teletransportation. In this case, many people would accept Nagel's claim. They would believe that what matters is the survival of their brains.
I shall now describe two cases where this is harder to believe. Remember first that an object may continue to exist even if all of its components are replaced. The standard example is that of a ship, which has a piece of wood replaced after every journey. We might believe the same about a brain. We have learnt that the cells in the rest of our bodies are all gradually replaced. Even though this is not true of our brains, it might have been true. Our brains might have continued to exist even if, like the rest of our bodies, they had all of their components naturally and gradually replaced. And we may believe that our brains would continue to exist if we ourselves caused such a gradual replacement. I shall here assume that this is true. Since we believe that the rest of a person's body does continue to exist, though its components are gradually replaced, why should we take a different view about our brains?
Suppose next that I need surgery. All of my brain cells have a defect which, in time, would be fatal. But a surgeon can replace all these cells. He can insert new cells that are exact replicas of the existing cells except that they have no defect. We can distinguish two cases.
There is a real difference between these cases. In Case One, each of the new parts of my brain is for a time joined to the rest of my brain. This enables each new part to become part of my brain. When the first new part is inserted, and joined to the rest of my brain, it wins the title to be as much part of my brain as the old parts. When the second new part is inserted, it too becomes a part of my brain. This is true of every new part, because there is a time when this part is joined to what then counts as the rest of my brain.
In Case One, the surgeon performs a hundred operations. In each of these, he removes a hundredth part of my brain, and inserts a replica of this part. In Case Two, the surgeon follows a different procedure. He first removes all of the parts of my brain, and then inserts all of their replicas.
(p.475) In Case Two, things are different. There are no times when each new part is joined to the rest of my brain. Because of this, the new parts do not count as parts of my brain. My brain ceases to exist.
Something similar might be true about the existence of a club. Consider a club that is limited to fifty members. All of the existing members want to resign. Fifty other people want to join this club. There is a rule that a new member cannot be admitted except in the presence of forty nine existing members. Because of this rule, this club continues to exist only if what happens is like Case One. What happens must be this. One member resigns and a new member is admitted. Another member resigns and a new member is admitted. A third member resigns and a new member is admitted. At the end of this series, this club would still exist, with entirely new members. Suppose instead that what happens is like Case Two. All of the old members resign. Because of the rule, the new members cannot now be admitted. The club ceases to exist.
Return to Cases One and Two. I am assuming that a brain might through a process of gradual replacement become composed of new components. On this assumption, it is clear that, in Case One, my brain continues to exist, and that, in Case Two, it does not. Nagel suggests that identity is what matters, and that I am my brain. On this view, Case One gives me life, and Case Two death.
Is this plausible? Though there is a real difference between these two cases, it is less than the difference Nagel had in mind. He considered a case in which his brain would be destroyed, and a Replica created. And he compared this case with ordinary survival, where his brain continues to have all of the same existing cells.
In my pair of cases, the difference is smaller. In both of my cases, there will later be a person whose brain will be exactly like my present brain, except for the defects. As a result, this person will be fully psychologically continuous with me. And, in both cases, this person's brain will be composed of the very same new components, each of which is a replica of some part of my brain. The difference between the cases is merely the way in which these new parts are inserted. It is a difference in the ordering of removals and insertions. In Case One, the surgeon alternates between removing and inserting. In Case Two, he does all the removing before all the inserting.
Can this be the difference between life and death? Can my fate depend on this difference in the ordering of removals and insertions? Can it be so important, for my survival, whether the new parts are, for a time, joined to the old parts? This could make all the difference if it produced some further fact. This would be so if my survival was like some sacred power, which one priest could give to another only by a ritual involving touch. But there is no such further fact. There is merely the fact that, if the new parts are for a time joined to the old parts, we describe the resulting brain as the same brain. If the new parts are not so joined, we describe the resulting brain as a different brain.
Nagel does not believe that the grounds for his view are decisive. And he (p.476) admits that ‘it is hard to internalize a conception of myself as identical with my brain’. He adopts his view partly because, in the pair of cases that he considered, his survival seemed to him to depend on whether his brain continued to exist. In my pair of cases, the difference in what happens is much less. If he considered these cases, Nagel might change his view. He suggests both (1) that identity is what matters, and (2) that he is his brain. But he admits that (2) is hard to accept. It is hard to think of oneself as being one's brain. When I consider Cases One and Two, I find it impossible to believe both (1) and (2). I cannot believe that what would matter for my survival is whether, over some period, the replicas of parts of my brain would be inserted in one of these two ways. I cannot believe that, if the surgeon alternates removing and inserting, this will be just as good as ordinary survival, while if he does all the removing before all the inserting, this will be nearly as bad as ordinary death. If this difference between the two cases is not what really matters, there are two alternatives. Either identity is not what matters, or I am not my brain.
The first alternative is supported by the imagined case where I divide. In this case, two resulting people each have half my brain. And there is no replication. These halves will be composed of my existing brain cells. As I have argued, it is very hard to believe that I should regard division as equivalent to death. My relation to each of the resulting people contains everything that would be needed for ordinary survival. And this remains true even if what I am is my brain. Each half of my brain will continue to exist, and to support conscious life. Each of the people with half my brain will seem to remember my whole life, and be in every other way psychologically continuous with me. If I am my brain, this is not a case in which I die because my brain ceases to exist. My brain continues to exist, and, because it is divided, it supports life more abundantly. It supports not just one but two lives.
Suppose that Nagel agrees that my relation to each resulting person contains what matters. Could he then defend his assumption that identity is what matters? We have seen that he could, by making grotesque distortions in our concept of a person. I assume that Nagel would reject these distortions. He would agree that, after I divide, there will be no one living who is me. If he also agrees that my relation to each resulting person is as good as survival, he must drop the assumption that identity is what matters. Without this assumption, I am not forced to conclude that replication would be as bad as death. I can agree that my Replica is not me, but claim that my relation to him contains what fundamentally matters. I can claim that what matters is psychological connectedness and/or continuity, with any cause.
Nagel might reply as follows. He might agree that there is one special case where identity is not what matters. This is the case where two future people each have half my brain. But what matters here is simply the continued existence of my divided brain. In every other case, personal identity is what matters.
(p.477) This reply may have some force. But if we believe that identity is what matters, it is natural to believe that identity is always what matters. If we admit one exception, it may be hard to justify rejecting others. Given the small difference between my Cases One and Two, we can claim that, here too, identity is not what matters. If this claim is justified in the case where I divide, why can it not be justified here? It is hard to believe that my fate depends on the difference between these cases. Unlike the pair of cases that Nagel considered, this pair suggests that I am not my brain.
I have tried to answer Nagel's arguments. My answers do not show that his view is false. But I believe that they show that we can, defensibly, take a different view. The question remains open. This is why, in Section 98, I offer a quite different response to Nagel's view.
(9) See KRIPKE, throughout. This view has also been developed in several articles, by H. Putnam.
(10) What we are referring to often depends on the causal history of some part of our language. But this is not always so; and causal considerations of this kind seem not to be relevant to the particular view that Nagel advances.