Hi, I´m new here. Please accept this source of extra noise in your mailbox in the hope to be useful
Federico Marulli wrote: > So we can try to reason upon some examples which has a meaning from a > physical point of view. For instance, we can think about the second > principle of thermodynamics, according to which the entropy of a closed > system necessarly has to increase. That means that, for instance, a gas > put into a container of volume V will tend to spread by occupying all the > available volume. This way we get the most possible disorder and the state > is the most probable. Anyway the state in which all the gas is firmly in a > v < V volume is not forbidden by thermodynamics; it is just a rather > improbable state. But this event, having some chances to take place, has > to happen in infinite places and times in our multiverse. So there will be > infinite Hubble spheres in which everything happens exactly as in our own > sphere, but in which any time you put a gas into a container, it will > never occupy the whole volume. At the same time, there will be infinite > spheres in which some day the gas will occupy all the volume and some > others not. And so on. The estrange behaviours like this can be prohibited by physical laws we don't already know or new consequences discovered from already known laws. In particular the case above, I think we may have a law that has consequences that prohibits such entropic behaviour: Statistical mechanics. As the gas is made by particles which moves in all directions it is easy to see that in the absence of any obstacle, the particles would go straight ahead in all directions to fill all the available space with no exceptions. I´m almost sure that it is impossible to avoid the expansion by means of "well designed" particle collisions. A serious study of this question may be a good statistical mechanic work to be done. I know that this argument is rather simple but I think that no serious work has been done to discard it. My guess is that simply we do not know the reality well enough to be sure that these magical worlds are possible. I confess that, it is more difficult to argue against other probabilistic events mentioned by Federico. May be that each one of the mentioned examples are clearly different and requires a separate study....