Re: The copy paradox

Hal Finney (
Fri, 14 Nov 1997 11:09:51 -0800

Harvey Newstrom, <>, writes:
> I'm not sure this is the same copy
> scenario I was considering, since the original and the copy must share
> the same thoughts and cannot diverge or be disconnected. Wouldn't
> killing the original create a disconnection or a divergence in
> experience that would raise the same objections?

You mean, like one's dead and the other is alive? That does seem a bit
different, doesn't it? <g>

I am assuming there is no afterlife, and so there is no experience of being
dead. Therefore there is no divergence, simply that one of them stops
while the other keeps going.

> The only complication
> with killing the original is that the only witness who would object to
> the procedure is silenced, and there is no longer an original reference
> with which to validate the copy.

As long as the copy was in sync up until the moment the original lost
consciousness, that's good enough for me.

> This merely makes it harder to prove
> they are not the same, but does not make the procedure any more
> functional than if the original were allowed to live. If the original
> is allowed to live, you agree that the copy and the original are not the
> same.

Yes, if they are having different experiences. They are the same up until
the moment they become different.

> If you kill the original, I still maintain the same position,
> only now the copy is the closest thing to the original now remaining.
> It seems that you would still exist in your original body. You are just
> experiencing a real-time simulation of what the copy is experiencing.

We're slipping into saying what "you" are here. Do "you" exist in (only)
your original body, or are "you" defined by a pattern of information
processing, irrespective of its time or place?

> When I disconnect the real-time update in preparation to kill the
> original, wouldn't the original object that it has lost contact with its
> second body and is now existing only in the original body?

You must maintain the link until the original no longer has any conscious
thoughts not shared by the copy. If you disconnect the link and the
original is still conscious, then he is now different from the copy.

> Wouldn't the
> original have the same desire and curiosity to want to go continue
> experiencing life?

Again, you shouldn't have disconnected the link. Now if you kill him his
recent memories, experiences, etc. will be lost.

Here is an interesting thought experiment, proposed last year by list
member Lee Corbin, who was really into all these copying paradoxes.

Imagine that you can push a button and erase the last 5 minutes of
your memories. You will instantly be transformed back into the mental
state that you were in five minutes ago. Now, you can make a case that
this is effectivelly the same as the situation where you had a copy made
five minutes ago, it has been suspended, and you can push a button to
(instantly, painlessly) kill yourself and revive the copy. In either
case the last 5 minutes of your life are lost.

How do you feel about this situation? Would you be willing to push the
button, or would it feel like dying? Pushing the button will terminate
your current stream of consciousness just like killing yourself. The
difference is whether the five-minutes-ago state will resume in your
own brain or an identical duplicate.

> It almost seems that the copy argument for extending
> lifespan focusses on continuing information and services to other
> people. If they do not perceive a change, the experiment is declared a
> success. It does not focus on information and services still be
> provided to the original. All such information and service is halted.
> The original ceases to function, but none of his external contacts care.

It's really not that. It's all analyzed from the p.o.v. of the person who
walks away from the experiment. If he remembers everything that happened
to all versions of himself, then nothing has been lost. The changes
were harmless; it's no worse than if he'd had all the atoms of his brain
replaced but the neural function was maintained (which happens anyway
over the course of years).