> So which way is it? Do black holes exist or not? Civilian's answer is extremely good but it's worth expanding on. Physicists have a weird relationship with

> So which way is it? Do black holes exist or not? Civilian's answer is extremely good but it's worth expanding on. Physicists have a weird relationship with the world, often mediated through theory but deeply grounded in fact. As such, we often attach "theoretical names" to raw empirical phenomena that have a relatively high level of contingency associated with them, and we then talk about those names as if they were the real thing, dragging the whole baggage of the theory behind us. This is really useful as it lets us live for a time in the ideal world of theory without bothering too much about the messy experimental details. We can get away with it because we are committed to always coming back to reality in the end. To outsiders this creates confusion, because we take a bundle of empirical data and label it "black hole" or "dark matter" as if we were referring to the reality rather than an explicitly and carefully constructed theory (not model) of reality. We then reason about that theory and try to find ways to extend it, test it, etc. If it passes enough tests (including tests of internal consistency, consistency with other well-thought-of theories, etc) we treat it as "good enough for going on with", which to outsiders looks like "true", but because science is Bayesian we cannot ever assign a theory a plausibility of unity (which is what "true" would mean, if it could mean anything). There is always room for doubt [0]. So when you ask "do black holes exist or not"? there is no simple answer, because we are in a mixed state, holding contingent beliefs but talking about them in non-contingent ways (because we simply don't have the language or mental capacity to do that.) GR has solutions that would explain a great deal of stuff we see in the sky. We call those solutions "black holes". There turn out to be some consistency issues with those solutions that have recently lowered the plausibility of the proposition "GR solutions with singularities explain compact stellar objects greater than 1.4 solar masses" (1.4 solar masses is about the limit for neutron star stability, so we infer anything larger is a black hole.) So where we are today is that the proposition "Standard GR and QM give an extremely good account of a large range of astronomical and cosmological phenomena that go under the name 'black holes'" is still very plausible, but that is different from "black holes exist" or "black holes don't exist", and always will be. There is a very important sense in which your question simply doesn't address the issue that science is concerned with, which is finding propositions that are maximally plausible, not "true". [0] This is true of all propositions whatsoever, including Bayes' rule itself.