Conservation
There is a difficulty in the previous section. You said that continuity over time is a sufficient condition for transference, but I can think of a counterexample to this. Let us go back to the example of the backed-up and restored humans. Now suppose that the restored human's body is merely a clone of the backed-up human's body. This provides ample continuity over time between two different person-stages. But one would never conclude from this that both person-stages must belong to the same token of person.
Okay, from this, I think I can infer a necessary condition for the transference of tokens. It is conservation of matter and energy. Notice that this happens to be a scientific principle of matter/energy transference in general--that the matter/energy is conserved in the process. Matter/energy is conserved in the special-place example. But it is not conserved in the backed-up and restored humans example. In the special-place example, the overall amount of matter/energy in the candidate for transference remains constant. But in the backup example, the amount of matter/energy has doubled. That is how you know that a transference has not occurred: because matter/energy has not been conserved in the process. I am not convinced that matter/energy conservation is a sufficient condition for the transference of tokens. It is a sufficient condition for the transference of the matter/energy of tokens. But to be a sufficient condition for the transference of tokens, the form must be transferred as well. A necessary and sufficient condition for the transference of tokens would therefore be twofold: It would be the conservation of matter/energy and form.
In fact, one doesn't even need to speak of the conservation of matter/energy. What one really means by matter/energy is simply content. In other words, conservation of form and content is the necessary and sufficient condition for the transference of tokens. Current science tells us that content is made up of matter/energy. But our account of transference will survive even if this current scientific account is refuted in the future.
We had previously said that backed-up humans were different tokens from restored humans. We were also looking for an account of what makes a token the same token over time. We have now found it above. It gives us a solid argument that the backed-up humans are different tokens from the restored humans. We have also argued previously that continuity is all that is necessary to establish that two types are the same. Because the backed-up and restored humans have the same continuity, we can also say that they are of the same type. The question finally becomes, which are you, your token or your type?
In my opinion, you simply must be your token. For a start, I would not consent to being killed just because I had made a backup of myself. It would not comfort me to know that the same type of person as I am would continue after I had died. I naturally want my token to survive, whether or not my type survives in another token.
But I can see a difficulty in that argument, at least from an evolutionary perspective. In nature, the token always sacrifices itself to the type. You will eventually die in order to make way for your progeny. You may kill yourself in order to save the life of your children, or other relatives, or your countryfolk, or even your fellow human beings. The tokens of genes never survive long, but they enable the genotypes to survive, which is the most important thing. This suggests that, at least over time, the type view of personal identity will prevail. The people who identify with their tokens will die out, and the people who identify with their types will survive. Indeed, this is part of the premise of Down and Out in the Magic Kingdom. When backups were first invented, there was argument about whether your backup really was the same person as you. The philosophical issues behind this debate never really did get resolved. It was simply that the people who didn't identify with their backups died out, and the people who did, survived. In a way, that premise is rather frightening.
In other words, even I individually do not have to identify with my type. This only means that I will die while my dissenters survive. On the other hand, if I do identify with my type, then I (or people like me) will survive while my dissenters die. So the basic argument is that I should identify with my type as the key to immortality.
Unfortunately, I still consider this to be a false immortality. The right view is right independently of whether the majority of people believe it at any one time. It is also right independently of whether it tends to manifest in the majority of the population in the long term. I might be persuaded that my view would not be viable in a world where backups exist. I might be persuaded that the dominant view would in such a world would be that I was my type. But this is still not the reason for saying, "My body can die now, because I've just backed myself up." The reason must simply be that I think that that backup is I. And the reason for thinking that someone is you is unrelated to how many other people think that someone else is they. What other people think is, after all, incidental.
So what is the reason for thinking that someone is you? I think that this question is best left to be answered in the next section...
Okay, from this, I think I can infer a necessary condition for the transference of tokens. It is conservation of matter and energy. Notice that this happens to be a scientific principle of matter/energy transference in general--that the matter/energy is conserved in the process. Matter/energy is conserved in the special-place example. But it is not conserved in the backed-up and restored humans example. In the special-place example, the overall amount of matter/energy in the candidate for transference remains constant. But in the backup example, the amount of matter/energy has doubled. That is how you know that a transference has not occurred: because matter/energy has not been conserved in the process. I am not convinced that matter/energy conservation is a sufficient condition for the transference of tokens. It is a sufficient condition for the transference of the matter/energy of tokens. But to be a sufficient condition for the transference of tokens, the form must be transferred as well. A necessary and sufficient condition for the transference of tokens would therefore be twofold: It would be the conservation of matter/energy and form.
In fact, one doesn't even need to speak of the conservation of matter/energy. What one really means by matter/energy is simply content. In other words, conservation of form and content is the necessary and sufficient condition for the transference of tokens. Current science tells us that content is made up of matter/energy. But our account of transference will survive even if this current scientific account is refuted in the future.
We had previously said that backed-up humans were different tokens from restored humans. We were also looking for an account of what makes a token the same token over time. We have now found it above. It gives us a solid argument that the backed-up humans are different tokens from the restored humans. We have also argued previously that continuity is all that is necessary to establish that two types are the same. Because the backed-up and restored humans have the same continuity, we can also say that they are of the same type. The question finally becomes, which are you, your token or your type?
In my opinion, you simply must be your token. For a start, I would not consent to being killed just because I had made a backup of myself. It would not comfort me to know that the same type of person as I am would continue after I had died. I naturally want my token to survive, whether or not my type survives in another token.
But I can see a difficulty in that argument, at least from an evolutionary perspective. In nature, the token always sacrifices itself to the type. You will eventually die in order to make way for your progeny. You may kill yourself in order to save the life of your children, or other relatives, or your countryfolk, or even your fellow human beings. The tokens of genes never survive long, but they enable the genotypes to survive, which is the most important thing. This suggests that, at least over time, the type view of personal identity will prevail. The people who identify with their tokens will die out, and the people who identify with their types will survive. Indeed, this is part of the premise of Down and Out in the Magic Kingdom. When backups were first invented, there was argument about whether your backup really was the same person as you. The philosophical issues behind this debate never really did get resolved. It was simply that the people who didn't identify with their backups died out, and the people who did, survived. In a way, that premise is rather frightening.
In other words, even I individually do not have to identify with my type. This only means that I will die while my dissenters survive. On the other hand, if I do identify with my type, then I (or people like me) will survive while my dissenters die. So the basic argument is that I should identify with my type as the key to immortality.
Unfortunately, I still consider this to be a false immortality. The right view is right independently of whether the majority of people believe it at any one time. It is also right independently of whether it tends to manifest in the majority of the population in the long term. I might be persuaded that my view would not be viable in a world where backups exist. I might be persuaded that the dominant view would in such a world would be that I was my type. But this is still not the reason for saying, "My body can die now, because I've just backed myself up." The reason must simply be that I think that that backup is I. And the reason for thinking that someone is you is unrelated to how many other people think that someone else is they. What other people think is, after all, incidental.
So what is the reason for thinking that someone is you? I think that this question is best left to be answered in the next section...
Comments