Saibal Mitra wrote:
I more or less agree with Jesse. But I would say that the measure of similarity should also be an absolute measure that multiplied with the absolute measure defines a new effective absolute measure for a given observer.
Given the absolute measure you can define effective conditional
probabilities, except in cases where branches lead to death. In these cases,
the ''conditional probability'' of there being a next experience at all
would be less than 1.
Would you apply the same logic to copying a mind within a single universe that you would to the splitting of worlds in the MWI? If so, consider the thought-experiment I suggested in my post at http://www.escribe.com/science/theory/m4805.html --
But you should no more expect to end up in a branch where you died than in a branch where you were never born in the first place. Consider, instead of a branching multiverse, a Star-Trek-style transporter/duplicator in a single universe, which can deconstruct you and reconstruct exact copies atom-by-atom in distant locations (assuming the error introduced by the uncertainty principle is too small to make a difference--if you don't want to grant that, you could also assume this is all happening within a deterministic computer simulation and that you are really an A.I.). To use Bruno Marchal's example, suppose this duplicator recreates two identical copies of you, one in Washington and one in Moscow. As you step into the chamber, if you believe continuity of consciousness is "real" in some sense and that it's meaningful to talk about the probabilities of different possible next experiences, it would probably make sense to predict from a first-person-point of view that you have about a 50% chance of finding yourself in Moscow and a 50% chance of finding yourself in Washington.
On the other hand, suppose only a single reconstruction will be performed in Washington--then by the same logic, you would probably predict the probability of finding yourself in Washington is close to 100%, barring a freak accident. OK, so now go back to the scenario where you're supposed to be recreated in both Washington and Moscow, except assume that at the last moment there's a power failure in Moscow and the recreator machine fails to activate. Surely this is no different from the scenario where you were only supposed to be recreated in Washington--the fact that they *intended* to duplicate you in Moscow shouldn't make any difference, all that matters is that they didn't. But now look at another variation on the scenario, where the Moscow machine malfunctions and recreates your body missing the head. I don't think it makes sense to say you have a 50% chance of being "killed" in this scenario--your brain is where your consciousness comes from, and since it wasn't duplicated this is really no different from the scenario where the Moscow machine failed to activate entirely. In fact, any malfunction in the Moscow machine which leads to a duplicate that permanently lacks consciousness should be treated the same way as a scenario where I was only supposed to be recreated in Washington, in terms of the subjective probabilities. Extending this to the idea of natural duplication due to different branches of a splitting multiverse, the probability should always be 100% that my next experience is one of a universe where I have not been killed.
So if the machine accidentally creates a copy of me missing a head, do you agree that doesn't lessen the probability that I will continue to have conscious experiences, that in this case I could be confident I'd end up as the other copy that was created with head intact? If so, is this any different from a situation where someone is shooting at me, and there is a branch of the multiverse where my head gets blown off and another where the bullet misses?