Re: another paradox and a solution

From: Wei Dai <weidai.domain.name.hidden>
Date: Mon, 23 Feb 1998 22:30:40 -0800

On Mon, Feb 23, 1998 at 07:37:28PM -0800, Hal Finney wrote:
> I think we discussed this somewhat on extropians, but I don't remember
> the resolution.

We talked about a different related, but easier problem. There one of the
questions was given that you may have been duplicated depending on the
result of a coin toss, what is the probability that you're a clone. I
think we agreed that it is 1/3. In that problem the probabilities do add
up to one, because you're asking the questions after you're already
(potentially) duplicated.

> It seems to me that the experimenter is rational to make the bet, because
> her expected return, given that she is alive afterwards, is positive.
> And she presumably doesn't care, if she is dead. She can reason this
> way regardless of whether other universes exist.
>
> This would not necessarily be true in all cases, for example if the
> experimenter had an estate and heirs whose welfare was important to
> her, she might not take the attitude that nothing matters if she is dead
> (likewise if she believes in an afterlife).
>
> But if she only care about her own life, then wouldn't it be rational to
> only consider outcomes where she is alive?

To avoid the question of whether she cares about what happens when she is
dead, we should assume that the experimenter has no savings and normally
spends all of her income. She was planning to spend her last $2 before
performing the experiment until her assistant offered her the bet.

She faces a trade-off between consuming $2 with probability 1 and
consuming $3 with probability 1/2, so she should choose the former.

You seem to be talking about a scenario where the $2 would come out of the
experimenter's savings account. In that case, if she doesn't care about
what happens when she's dead, she faces a trade-off between consuming $2
with probability 1/2 and $3 with probability 1/2, so she should choose the
latter. In any case, I think the paradox is resolved.

(I will add that most people do care about what happens after they die,
because that's the way evolution made them. Those who don't care would
leave less inheritence to their children and so they would have less
reproductive success.)
Received on Mon Feb 23 1998 - 22:31:02 PST