janw@inmet.UUCP (08/08/86)
[kaufman@nike.UUCP ] [drastic changes of the past] >It's a good way to tell if you live in a parallel universe, but >it's not a good way to tell if you live in a serial one, since >the experiment would have a high probability of causing the ex- >perimenter never to have existed, or at least never to have con- >ducted the experiment. Better to conduct the experiment on a >smaller scale, then you can be sure that you'll be around to see >the results. It seems that the main thing is not the scale of the past event you change, but whether you know the consequences of it in the present, or not. If you do, then *any* experiment should destroy itself. E.g., suppose you want to change the winning lottery number from M to N. If the experiment succeeds, then N has always been the number, and you have never conceived the experiment of changing M to N. You could have planned to change it from N to P, but that is another experiment, also self-destructive. If, in spite of the number being N already, you conduct the procedure that would change it to N, there is no telling if it works or not. If, however, you *don't* know what the winning number is, and change the past so as to fix it as N, and it *is* N, then it *probably* works, and the universe is *probably* serial. Thus a past-changing experiment ought to create an improbable condition in the present. If you know about the condition in ad- vance, it is not improbable but certain. If you know the oppo- site, then you will not have made the experiment, or will not have succeeded. E.g., plant in the past a letter addressed to yourself, but *don't* look for it before you plant it.