X-Message-Number: 25549
Date: Sat, 15 Jan 2005 01:53:32 -0700
From: Mike Perry <>
Subject: Duplicates Problem Again
References: <>

Here I respond again to Richard.

>1. Suppose I fall asleep, am destructively scanned, and two
>duplicates of me are created, one looking at the sun, and the other
>looking at the moon. While you may be able to say, neither one of
>them remains me for very long (as you said in your last message),
>you still have yet to answer the question, when I fall asleep, and
>wake up, what do I see? The moon or the sun? I cannot possibly see
>both.

To me, "survival" of a past individual occurs because some present 
individual has suitable brain information, memories and such, which when 
scanned or accessed produce an appropriate effect, that the present 
individual *was* (once) that past individual. (In extreme situations there 
could be dislocations so that the "past" as perceived may actually be the 
future, as Robert Ettinger pointed out the other day. I think this problem 
is manageable, and I devote space to it in my book--but for now let's 
assume the past is really what it seems, as I think would be far more 
likely in practice.) My notion of survival is rather different from 
Richard's, and it is clear that neither would have to imply the other. In 
particular a duplicate could also have the same experience of feeling 
he/she *was* the past individual, but Richard does not allow two versions 
of the "same" person to survive. As he understands "survival" this is quite 
reasonable and unavoidable, but of course is not the only way to view the 
matter. To me Richard's form of survival is something that may not occur at 
all, or may be very brief. I can think of no objective means of resolving 
the matter--how do I *really* know, when I lay down to sleep, that I will 
not simply experience eternal oblivion while someone who only thinks they 
are me wakes up? In any case, if this survival does occur, it could be 
unavoidably brief. It cannot occur if the original splits into two or more 
copies that have to be treated on an equal footing rather than identifying 
one as an unchanged or little-changed "original" and the other as a 
newly-produced "copy." So (as one possibility) if the many-worlds scenario 
is correct (and it has significant support from the physics community, 
albeit some opposition too) we clearly don't survive long in this way. We 
must split--and die--every time a random (unpredictable) event is 
perceived, because there will be some other version of ourselves that sees 
it another way.

In addition to my skepticism about whether survival in Richard's sense is 
ever likely to be for very long (for I do take many-worlds seriously), I 
also question whether it would even be meaningful or really have the 
subjective effect it is supposed to have, or whether its absence would 
truly amount, subjectively, to oblivion. (If you can't remember your past 
self, then it seems to me that the past self does not survive, 
notwithstanding whatever it is that does survive. But if you can remember 
then a reasonable case can be made the past self lives on in you, even if 
it might also live on in someone different from you--that is to say, 
different from what you have now become.) But once the differences between 
my notion of survival and Richard's are understood, it should not be 
difficult to see how I would resolve the main problems with duplicates. 
Much of the difficulty, as I see it, centers around the issue of saying I 
*was* that past individual (my viewpoint) versus I *am* that past 
individual (Richard's if I understand correctly). Of course you cannot have 
two separate, subjectively different people who both *are* (identical to) 
something, for then (by transitivity of identity) they must be one and the 
same. Hence the paradox if one sees the sun and the other sees the moon. 
But if you only allow that they *were* some one thing, they can now be 
different and there is no paradox.

>2. Imagine someone making a duplicate of you while you are awake.
>The duplication process takes only 1 hour. After the duplicate is
>made, you talk with him for a few hours, recounting childhood
>memories, and then take him to dinner and a movie.

Okay, but now we are separate individuals, who only *were* some one entity, 
not presently *are* that entity. So if you kill one, the one does not 
survive, only the other.

>Now answer this question carefully, because IN YOUR VIEW, this
>scenario is identical to a different one, in which an
>anesthesiologist puts you to sleep, you have some operation, and
>then you wake up.
>
>Why? Because the drugs the anesthesiologists use frequently include
>an amnesiac ('just in case'), so you will forget everything that
>happened approximately six hours prior to the operation.
>
>In the first scenario, you and your duplicate were identical 6
>hours ago, but then diverged. In the second scenario, you lose all
>your short-term memories, thus reverting to an earlier self.
>
>If you want to be consistent, you would have to say that the
>duplicate DOES constitute your survival, even though I stabbed you
>so many times, and you died a horrible death.

Ah, the duplicate continues the "me" that existed before the duplication 
occurred, but not the version of me that was knifed.

>  In fact, if presented
>with the situation, if I asked to stab you as a mere academic
>exercise, you should have no problem with that (provided I gave you
>some strong painkiller), because after all, you survive in your
>duplicate, so there is no need to be concerned with the particular
>hunk of matter in your head.

No, the version getting the knife might could value its own, independent 
existence that started after the duplication occurred (reasonable, in 
fact). Again, only the "me" that existed prior to the duplication will 
survive in the other version. If you want to consider making duplicates who 
remain unconscious the whole time, then one is destroyed while the other is 
awakened, it seems reasonable to me that no great loss occurs, if I can 
assume that this is exactly what happens (with high probability). For 
anything you do, of course, there is some chance of things going 
differently than intended or expected, and that has to be taken into 
account too.

>Here I think many people will diverge from you. The survival of the
>duplicate cannot be considered your survival in any useful way.

Again, we have two distinct notions of survival, and their differences must 
be understood. My version is testable, in principle, I think, while I 
submit that the other version is not, despite some brave attempts to argue 
otherwise.

>You are that hunk of matter in your head we call a brain. Protect
>it well.

I will protect it because it has the only copy (in this universe) of my 
(present) identity-critical information, even though I don't accept that it 
"is" me or must also become a future "me," if the present "me" is to survive.

Best wishes,
Mike Perry

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=25549