martinbn said:
That's the whole point, there was no origin in a singularity. The thing, that's not even defined, and we want to call singularity is not part of space-time. In space-time everything is finte and perfectly well behaved.
That's in the mathematics, where the singularity can be regarded as not in the spacetime. That was the point of that nice article you cited. But note that science is not just mathematical theories, it is taking mathematical theories to try to tell a story about what is happening, and how it could be predicted. So if the mathematics can avoid the singularity, the physics cannot-- there are frames that could be populated by hypothetical observers, as we always do in physics to say what is happening, and those frames will get unbounded observables like kinetic energy density and temperature. What's more, they only comprise a finite duration in proper time, with no accounting of t=0 itself. The mathematics can simply not extend to t=0, but in a physical description, that's an incompleteness. It is that incompleteness that causes people to make wholly unsubstantiated statements like "time itself began at t=0," or "time itself began with the Big Bang", and worse, to claim that this claim is part of the Big Bang model, when it certainly is not.
But I think we are basically agreeing here, because we are both saying, you from the mathematical perspective and me from the perspective of observational support and testing, that the t=0 instant is not in the model, is not tested by any observations, and no scientist or GR mathematician really has any basis for making any claims about it whatsoever, other than that it is a kind of flag or milestone worth noticing. In fact, I would say that milestone is the most important aspect of the Big Bang model, but it is not something that the Big Bang model actually models. The unfortunate part is that it is often one of
only two things that appear in pop sci renditions of the Big Bang model: the origin point, and the ensuing expansion. The model is actually nothing about the former, and all about the latter, because only the latter is peppered with a vast array of observational evidence and "risky predictions" that proved true.
The cosmological principle is not the problem, that's what the singularity theorems say. Under physically reasonable conditions Lorenzian manifolds are geodesically incomplete.
OK, I was wondering about that. But let me stop you and ask, what do you mean by "physically reasonable conditions", and how do we know those are not the problem, rather than GR itself? Maybe the physically reasonable conditions are wrong, and since we have no observations to say those conditions hold at arbitrarily early times, we cannot then claim that a singularity is a prediction of GR, it is merely among the things we would like to test. I'm sure I don't need to remind you of all the "physically reasonable conditions" in the history of physics that turned out to not test out at all!