RE: duck me!

From: Dan Fabulich (dfabulich@warpmail.net)
Date: Mon Oct 14 2002 - 14:50:05 MDT


gts wrote:

> I can see no reason that your act of self-vaporization would make you
> and your alleged duplicate into a single thawed-out person with a
> continuous sense of self. Can anyone explain the mechanism by which one
> individual's sense of self should pass to another individual upon the
> death of the first individual? It seems to me that belief in such a
> thing is tantamount to a belief in witchcraft.

The argument is that "your" (past) sense of self passed into the duplicate
*ahead* of time, as the duplicate was made... the confusion here has to
do with the use of the indexical pronouns "I" and "you", especially as it
refers to the present, or the past, or the whole history.

Allow me to suggest a formalism for a moment: to say that any given
something exists is to say that there exists an X which has the relevant
property that defines that something. Thus, "there's something red" would
be written: \exists(x): red(x). Standard Russell.

OK. Well, extend this to personal existence. "I exist" would be
re-written as: \exists(x): me(x). There's some X which has the "me"
property.

Now suppose you split into two. I'd argue that both of the split persons
have your property me(), *where me() is the property you had when you were
one person.*

However, they now also each have their own indexical property that was
co-extensive with me() when there was only one person: one split has
me[1](), the other has me[2](). You *were* me[1]() as well as me[2]().
Now, each split is one or the other, and not both.

Now, as you point out, when when the Nth split talks and thinks about
himself in the present, he'll be thinking about the property me[N](), not
the original property me(). That's as may be, but there's something
interesting about me[N+1](), even if you're just the Nth split; it's not
merely the case that you're "very similar", but you used to be the same
person.

That person me(), who both of you used to be, will survive so long as at
least one me[N]() property continues. So when any Nth split thinks about
the person he was in the past, he'll think about me()... for example, all
splits will remember being the same person, etc.

We can make time-indexed synonyms from the perspective of the Nth split:
me[now]() is synonymous with me[N](), and me[past]() is synonymous with
me(), the original property. And then what you notice is this:

  If I push that button, me[now]() will die, but me[past]() will survive.

This is an extraordinary situation in which to find yourself; if you're at
all like me, you have no relevant moral intuitions with which to settle
this matter. I'm used to the idea that I should try to keep me() alive,
and I'm not at all used to the idea that me[now] might die while me[past]
might live. I want them both to live, I guess, but if one goes, it's not
obvious to what degree "I" have lost anything.

If you want to reject this notion, you'll have to give some kind account
of why I would pay more attention to one or the other: why I'd pay
attention to me[now]() over me[past](), *or vice versa*.

Certainly, if you think that me[now]() is the only interesting property,
[perhaps because "I" is automatically present tense?] then you'll reject
the claim that "you will survive the vaporization." Similarly, if you
think that me[past]() is the only interesting property, [perhaps because
we are defined by our memories?] you'll accept the claim. But I see no
sense in "picking" one over the other: I am both me[now] and me[past].

It just makes no sense to say "I will/will not survive the vaparization"
without specifying whether what is meant is me[now] or me[past], in much
the same way that it doesn't make sense to say that my car is 10 long,
without specifying units. Conversationally, we can get in the habit of
choosing one or the other, but morally, neither seems to be any less
interesting than the other.

-Dan

      -unless you love someone-
    -nothing else makes any sense-
           e.e. cummings



This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 09:17:33 MST