Hal (
Thu, 7 May 1998 11:44:14 -0700 (PDT)

Anders Sandberg, <>, writes:
> Actually, I think it is more clever than that. Agents still have their
> free will, it is just that they or the world will not behave in such a
> way that a paradox results. Suppose you were trying to make a paradox
> on the Novikov table by shoving the ball around. When you are planning
> to snatch any balls emerging from the exit and throw them into the
> wastebasket, none will appear.

It seems clear to me that the Novikov table could not exist and be
consistent with free will. (BTW do you have a reference to Novikov's
paper?) It's the classic paradox. You watch the hole, and when
nothing comes out, you throw a ball in. Oops, it was supposed to have
come out one second ago. What happened?

Obviously you will not succeed in throwing the ball into the hole.
Maybe you'll have a heart attack when you try, or maybe a random
fluctuation of air molecules will bat the ball away at the last
minute. It's Asimov's story all over again.

A few years ago it was thought that wormhole paradoxes would be
prevented because, at least in some geometries, virtual particles could
take a closed timelike path. This led to an infinite amplification of
the particle flux, choking off the wormhole.

However I understand that these results did not generalize and that it
is now thought that no physical phenomenon will prevent the paradoxes.
I still suspect that some variation on the virtual particle problem
will arise and prevent wormhole loops, but it remains to be seen.