On Philosophy

September 12, 2006

The Murder of the Self

Filed under: Self — Peter @ 12:03 am

A common question that comes up when thinking about personal identity is: what makes the person I am now the same as the person who was me some time in the past? Clearly we aren’t identical, many things change over time, but we do want to claim that there is some “personhood” that is shared by us.

A common candidate for the property of “personhood” is psychological continuity. The me now and the me yesterday share many of the same desires, attitudes, memories (including new memories of being me yesterday), ect. Dualists would reject this view of “personhood”, of course, but fortunately we are not dualists, for reasons I have outlined elsewhere. The only other alternative would be to subscribe to some kind of physical continuity as defining who we are, but since our cells, and the proteins that constitute them, are constantly in a process of being replaced, without any noticeable effect on our “personhood”, I think we can safely rule it out as an alternative.

Two immediate consequences spring from defining personal identity in this way. One is that two people-instants (a person at a particular point in time) are not simply the same person or not. Psychological continuity comes in degrees, and thus our judgments as to whether two people-instants are the same person must as well. The me now and the me yesterday are the same person to a high degree, while the me now and the me several years ago are the same person to some lesser degree. The second consequence is that it is possible to kill someone by preventing them from sharing a psychological continuity with future people, even if their body lives on. For example, if we had a machine that replaced a person’s desires, attitudes, memories, ect with those of another person to use the machine on someone would be to kill them.

Now let’s consider a slight modification the person-replacing machine. Instead of replacing the person’s psychological attributes all at once it transforms their existing ones into those of another person over some short period of time. Clearly this process is equally destructive to the original person. But, as the title of this post foreshadowed, this is just what happens in ordinary life, since we change over time. For some people there is likely a point in their lives where the only thing they have in common with the “them” at some distant earlier point in their lives are perhaps some vague memories of that time. And, even if these vague memories provide some psychological continuity, it is no more than I have with someone else who, let us say, shares some of my attitudes and goals. And if the psychological continuity I share with the stranger isn’t enough to make us the same person then clearly the vague memories we might share with our earlier selves aren’t enough either.

Have we then killed our earlier selves? (Or perhaps our earlier selves committed suicide, since they didn’t object to the changes that turned them into us.) And even if we have it certainly wouldn’t make sense to fight this process, since that would force us to lock ourselves into the same attitude, opinions, and desires that we have now, definitely not a healthy idea. Now some might object to this conclusion, claiming that because there were intermediate stages that shared the necessary psychological continuity with our earlier selves then somehow our “personhood” has been transmitted intact through them. If this were the case then we shouldn’t resist having the person-replacing machine, modified to transform us over the space of an hour, used on us, since in that case too there are intermediate stages to preserve our “personhood”. Of course one might still resist if they disliked the person it changed you into, so let me stipulate that the person the machine changes people into is something that everyone would desire, and in fact strive for (intelligent, wise, sensitive, moral, ect). Even so, I think most people would resist having the machine used on them, because they feel that in some way it would eliminate the person who they are now, and if this is the correct interpretation then my original reading of psychological continuity holds, and thus the people some of us were in the past really have been replaced by new people. I don’t know exactly how we should read the moral implications of this, so I will leave them as an exercise for the reader.



  1. Without putting too fine an edge on it, what’s missing from your discussion of person-hood is a more or less clear idea of what constitutes a self.

    I would posit a definition of person-hood that equated the self with the sum total of all the thoughts, memories, emotions, and experiences in which a sentient being has participated. Such a definition does not seem unreasonable and captures much of what one ordinarily means when speaking of the self. Moreover such a definition tacitly allows a self to grow (aka change) as it would not be reasonable to read into such a definition an inability of the self to add to the totality of its experiences. Because if one did not allow for such additions, no self would consist of more than that with which it was created.

    It should be noted that what makes any person the person they are at any particular moment is in fact the sum of all their experiences and their knowledge events of their past. It is the function of the central nervous system to integrate these new experiences with past experiences into an integrated whole. And while the resulting “new” self is not identical with the “old” and in that very limited sense the “old” self has died, a more apt metaphor would be that the “old” self has grown and lives on as a more nuanced and informed, “new” self.

    If the person I was yesterday is not the person I am today, then who was that person that is not me?

    Comment by Max Puber — September 12, 2006 @ 10:18 am

  2. That doesn’t solve the problem max, because then the natural process of forgetting leads to the same conclusions. And not all of a persons experiances contribute to who a person is, many have little or no effect (do you really remember and change every time you brush your teeth?).

    Comment by Peter — September 12, 2006 @ 1:31 pm

  3. Whether I remember (i.e. form persistent conscious memories) the details of my past or not, does not preclude past events from contributing to the person I am today. The central nervous system processes many of my experience at a level below my conscious awareness but nevertheless these experiences may have profound effect on my behavior. It is problematic as to whether small repetitive, nearly unconsciously performed tasks have any lasting effect on ones performance or behavior. Ask anyone who’s ever failed to learn to play a musical instrument.

    Comment by Max Puber — September 12, 2006 @ 8:10 pm

  4. So if this is your defintion of self then you would argue that the personality changing machine wouldn’t be death if some faint echo of the old person was left behind, or the new person had some of the old person’s memories? I find this hard to buy, since it might be possible to give person A some of person B’s memories, but it wouldn’t make them person B.

    Secondly your argument, that a person is defined by the experiances that make them up, would indeed mean that I am not the same person that I was in the past, because the person in the past is only defined by a smaller subset of the experiances that define me now. If you don’t accept that distinction then the following dilemma results: since personhood is transitive, if A is B, and B is C, then C is A, then if we made an exact duplicate of a person at some time (t), and then allowed each duplicate to go its sperate way, you would be forced to conclude that the duplicates, no matter how different they became, would still be the same person, since they are both the same person as that unitary person who existed before the duplicate was made (at time t), and thus by transitivity they are the same person. Clearly this is unacceptable. And you can’t argue that such duplicates are impossible, since if the multiple worlds interpretation of quantum physics is correct they are being made all the time.

    Comment by Peter — September 12, 2006 @ 11:07 pm

  5. Firstly, let me sharpen the the definition I suggested for the self. I equated self with the sum of one’s past experiences. More precisely, I would equate the self to the integrated whole produced by the central nervous system as a result of its processing of those experiences. I don’t believe this materially changes our discourse but I did want to correct this defect in my original statement.

    Secondly and with respect to the person duplication example, I would argue that it is logically impossible to created an exact duplicate, B, of person, A since, at the very least, at the moment of B’s creation, B would have had the experience of having been created as a duplicate and A would have the experience of having been duplicated and therefore at no time would B be an exact duplicate of A. This would, naturally, preclude the use of the transitivity argument.

    Thirdly, there is no denying that the person I am today, is not the same as the person I was yesterday. And if you choose to define personhood as depending on conditions existent at a particular instant t, then certainly person A(t) is not equal to person A(t + delta), delta > 0. But suppose we were to adopt this same definition for mathematical functions, e.g. f(x) = mx + b, no one would argue that f(x) = f(x + delta), delta > 0. But wouldn’t the function, f, be the same? Thus it seems to me that an appropriate definition of selfhood would accommodate change. One could object that to this analogy and claim that personhood is not at all like f(x) = mx + b and say that personhood is more like f(x,t) = mx + b, for a

    Comment by Max Puber — September 13, 2006 @ 6:57 am

  6. Please excuse, I appear to have exceeded some limit. Continuing then

    … that personhood is not at all like f(x) = mx + b and say that personhood is more like f(x,t) = mx + b, for a

    Comment by Max Puber — September 13, 2006 @ 7:29 am

  7. My sincere apologies for this mess, Peter. I see now it must have been the less than symbol I used that caused the break in my post.

    Continuing then

    … that personhood is not at all like f(x) = mx + b and say that personhood is more like f(x,t) = mx + b, for a less than t less than b and f(x) = x*t for b less than t less than c … . But even here a mathematician would say there is but one function with an arbitrary set of descriptions defining how its mapping is to be carried out.

    Finally, I would say that the problem with defining an object that changes over time is that one would like to capture some overarching or invariant property of the object on which to base the definition. That is easy for mathematical functions in which the overarching property is simply the desire to refer to some arbitrary collection of mapping descriptions by a common term. For personhood, something more transcendent may be desired. But in the end, the definition one chooses depends on one’s application.

    Comment by Max Puber — September 13, 2006 @ 7:36 am

  8. Well, you can’t argue againt duplicates by saying they are impossible, or one would rember being duplicated, as you could always make the original unconsious, and of course as I mentioned earlier duplicates occur all the time in the multiple worlds interpretation of quantum physics, and it’s always a bad idea to dany a possible theory of physics because it contradicts your philisophical leanings.

    Otherwise you seem to be advocating the type-token theory or personal identity. The problem with this is that for a type-token theory of anything to work you must identify a set of properties that are essential to the type. Presumedly these are psychological attributes. Thus anyone who falls within some range of mental variation is me. And of course this leads to the problems mentioned in my paper, that the “same person” over a long enough period of time won’t belong to the same type. If we attempt to avoid this by defining the type in a looser way then there is the possibility of strangers being the same person as me. And we can’t fix this by adding the requirement that tokens of the type be physically continuous, as we do with simple objects, for obvious reasons. Additionally the type-token theory of personal identity falls into the same problems with the possiblity of multiple tokens of the same person type existing at the same time. For these reasons we reject the type-token theory of personal identity.

    Comment by Peter — September 13, 2006 @ 1:31 pm

  9. Look, just think about the analogy of stepping in the same river twice, You just cant do it, hopefully that doesnt require an explanation. If we are like the river in that we are in a constant state of flux, then any attempt to point to a static state of personhood will fail. We ought not search for some first principal of personhood. Instead go with what we know, in the everyday sense of know, there seems to be some being that is concious of experiences, that imposes some sort of continuity on what we could call a constant state of evolving personhood. Just remember what Hume said about not jumping out windows…

    Comment by ad hoc — September 19, 2006 @ 8:40 pm

  10. “there seems to be some being that is concious of experiences” = psychological continuity != specialperson property. It seems intuitively right that the sun goes around the earth, but it is also very wrong. Our intuitions about personhood could be equally flawed, and thus we are not justified in appealing to them.

    Comment by Peter — September 19, 2006 @ 10:17 pm

RSS feed for comments on this post.

Create a free website or blog at WordPress.com.

%d bloggers like this: