Re: Measure, madness, and Max

From: Gilles HENRI <Gilles.Henri.domain.name.hidden>
Date: Thu, 21 Jan 1999 08:25:56 +0100

>On Mon, Jan 18, 1999 at 10:48:08PM -0800, hal.domain.name.hidden wrote:
>> I disagree with the reasoning of Jacques Mallah regarding quantum suicide.
>> Jacques claims that in a Many-Worlds universe, someone has no reason
>> to expect the results of a suicide attempt to be different from what
>> they would be if there were just one world. I and many other people
>> argue, instead, that in a MW universe, suicide would eliminate you from
>> some successor "branches", but that you would go on living in others.
>> Since you cannot experience your own nonexistence, those successor
>> branches where you do not exist are irrelevant to you (just as are
>> branches where all life is impossible). Your personal experience is
>> that the suicide never works.
>
>Let's call this the quantum suicide decision argument (as opposed to the
>quantum suicide information argument, which is the argument that quantum
>suicide can provide information about whether MWI is true). I think the
>crucial problem with QSDA is that the successor branches where you do not
>exist are not necessarily irrelevant to you. In fact because of natural
>selection it must be relevant to most people. More generally, natural
>selection will cause most people to act as if their utility function is
>defined on global states of the entire MW universe, not just on their
>subjective experiences.

I'd like to insist again that there is no precise meaning of the question
whether "you" exist or not in some world. The notion of identity exists
only because of your memory, and implies a sufficiently classical evolution
of the world for this memory to be conserved. This is a problem for QS,
which I did not realize at once. The branching event following a quantum
measurement is by far the most rapid event we can conceive, since its
duration decreases exponentially with the number of particles. This is much
more rapid than any "mechanical" process like the triggering of a gun
(unless you could realize a vacuum instability that would propagate at the
speed of light and destroy the whole observable universe. Does anybody know
how?). So before tempting any suicide, you are already split in two (or
more) components. For the part of you that has been branched on the wrong
way, the suicide attempt has almost no chance to transport" or couple you
to the "right " world. It will couple you most probably to a universe
compatible with
1) your own survival and 2) with the greatest measure. It may be a universe
where you survive miraculously without sequels, or one where you are
severely injured and where you may not like to live!. It may also be one
where "you" lost totally the memory of your past, because the worlds where
this condition is satisfied have a measure null or negligible. so in some
sense "you" is no more "you". So if you know that the gun is going to be
triggered, for example by some alarm; it would be a good idea to stop
immediately this experiment...If you are not convinced, what do you think
of a "delayed" suicide experiment where you wait for one day before
triggering the gun?

Gilles
Received on Wed Jan 20 1999 - 23:27:40 PST