Imagine a big crunch scenario in which everything (or almost everything as for I have other doubts, eg about entropy...) rolls back.
Shouldn't the contracting universe reach a point of density at which a black hole forms? Or, perhaps, as high density regions are and will be present, a point at which it fragments into several BHs?
In other words, shall singularities form before reaching the original one (or reaching a point very close to the singularity that our future description should avoid, of course)?
I think this might help understanding the expanding phase even if a big crunch isn't a realistic scenario. If the answer is yes, it would point out that a simple roll back as often presented is a naive approach and unsymmetric processes were at work.
