[pdonis]

> I always thought we could put a black box around a black hole, and it would be indistinguishable from any other mass - that is, any other mass that can be treated as a point mass.

Yes, that's what the standard theory of black holes says.

> this result implies that the black box with the black hole will gain mass over time, even without adding any mass into the black box?

Sort of. First, it's important to note that the paper is talking about a special type of "black hole", an object that has "vacuum energy" inside it (which means something that acts like a cosmological constant in the Einstein Field Equation)--which isn't a standard black hole (those have zero stress-energy inside). The claim is basically that the total vacuum energy inside such an object can increase as the universe expands.

However, this does not mean that the ordinary "mass" of the black hole would increase. Vacuum energy doesn't work like ordinary mass. The effect that this model is claimed to account for is the accelerated expansion of the universe due to dark energy; basically this model is supposed to provide a mechanism for how dark energy could come into existence as a result of black hole formation (but, again, it's a special kind of "black hole", not the ordinary kind).

[pmontra]

If I understood the paper [1] correctly, the idea is that all black holes don't contain a singularity. They have vacuum energy instead and that leads to the increase of mass and dark energy.

[1] https://iopscience.iop.org/article/10.3847/2041-8213/acb704/...

[pdonis]

> the idea is that all black holes don't contain a singularity

More precisely, theoretically, we can construct models of compact objects that look like standard black holes, but don't have a singularity (and also don't have an event horizon, they only have apparent horizons). Any such compact object must contain "vacuum energy" or something equivalent, i.e., something that looks similar to a cosmological constant in the Einstein Field Equation--that is the only way to evade the conclusions of the various singularity theorems that apply to standard black holes. That type of compact object is what is being hypothesized in the paper under discussion.

[dtgriscom]

> all black holes don't contain a singularity.

Do you mean "not all black holes contain a singularity"?

[kahnclusions]

An important thing to keep in mind is that the singularity is more of a mathematical dead end than a real thing that’s supposed to exist.

The singularity existing in the math suggests that our theories are incomplete, and I would say it’s not surprising that new theories of black holes would do away with the singularity.

[raattgift]

> The singularity existing in the math suggests that our theories are incomplete

No, it's a feature of a mathematically complete model (the Schwarzschild solution) that crops up in extensions (with angular momentum; with electric charge; formed through gravitational collapse rather than eternal) pretty reliably. There is no incompleteness in the Schwarzschild, Kerr, etc. exact solutions. They may not correspond well with things in our universe though, and do not correspond fully to them because our universe (or at least its population of stellar-black-hole-generating galaxies) as far as we can tell is not already infinity years old or full of only vacuum.

(Further efforts which describe somewhat more physically plausible compact objects which grow as matter falls inwards and which are well behaved in deep inter-galaxy-cluster space where expansion is relevant also tend to have singularities if they form by gravitational collapse. Some of these only non-exactly solve the Einstein Field Equations (see the weak <https://en.wikipedia.org/wiki/Non-exact_solutions_in_general...> or the numrel link further below)).

The problem with the singularity is that given a 3d hypervolume (e.g. a set of every point where one would measure an identical average temperature of the cosmic microwave background) which contains all the positions and momenta and other values at every point in the 3d space, one cannot recover the whole set of values from earlier slices, and in particular not the whole set from before the singularity arose.

There was some hope that a collapsed star's singularity would last into the infinite future, or that (since that may not be the case) Hawking radiation would not be thermal noise, so that one could recover all the values of an arbitrary 3d volume after the singularity arose, or at least excise/not-care about the relevant values (see <https://en.wikipedia.org/wiki/Numerical_relativity#Excision> for example). However, now a merely extremely long-lived singularity means that one cannot recover a whole values surface in the far future either.

This causes problems when using the very handy <https://en.wikipedia.org/wiki/Initial_value_formulation_(gen...>.

It is in that sense a model with an evolving black hole is incomplete if it has a singularity. But we know that because General Relativity is a mathematically complete theory, with basically the only open-ended questions living in the mechanisms that generate the stress-energy tensor (i.e. the microscopic behaviour of matter).

The discussion's topic article P.R.s the latest installment in a programme that hopes nature will always generate stress-energy in the interior of a collapsed star in a way that evades the formation of a singularity while (the authors and fellow-travellers hope) preserving the external features of a more standard singularity-containing collapsar. Their model isn't mathematically complete in that they do not have an exact solution to the Einstein Field Equations (§4.6, <https://iopscience.iop.org/article/10.3847/2041-8213/acb704>).

Another mathematically complete theory which may admit non-eternal singularities which frustrate everywhere-determined values is Navier-Stokes. And it's the microscopic behaviour of the fluid matter which may let one recover the missing values.

Mathematical completeness, everywhere-uniquely-determined values, and reasonable physical relevance are three different things.

[rbanffy]

From the article I understood it as “no black holes contain singularities”.

[zwkrt]

The way that I understood it is that what you call an 'ordinary' black hole this paper claims is actually a 'naive' black hole, and that more nuanced solutions to general relativity allow for things that act like black holes that we know but that don't contain singularities.

I only point this out because we haven't been inside a black hole, so we don't really know what an 'ordinary' one looks like.

[rbanffy]

> we haven't been inside a black hole

It’s kind of like that, starting from where we are, black holes have no “inside”, since it takes an infinite amount of time to cross the event horizon.

[lamontcg]

No, if you fall into a black hole, it happens in a finite amount of time to you.

It is the observer at infinity that never sees you fall into the black hole, but real physics is local, you have to use the coordinate system of the person falling into the black hole to determine what happens to them.

[raattgift]

> It is the observer at infinity that never sees you fall into the black hole

We don't even need an observer to be at infinity, thanks to the expansion of the universe. With some future telescope our descendants may observe something on a trajectory to enter a black hole in an early-universe galaxy that is just crossing that observer's (cosmological) horizon.

I think it's relevant to raise this since the article at the top is about embedding black-hole-like collapsed stars in an expanding universe and the research which directly discusses the observable consequences.

> real physics is local

Yes, absolutely. You still get spaghettified if you fly into a black hole which is the only other appreciable mass left in the far far future of our universe. Nobody needs to see your last moments.

> you have to use the coordinate system of the person falling into the black hole to determine what happens to them

No, you can use any coordinates you want (or no coordinates at all), but you have to be aware that there are quantities which are invariant under changes of coordinates (e.g. the curvature scalars) and quantities which are coordinate-dependent, and that some systems of coordinates make the latter difficult or even impossible to calculate.

Indeed the infaller can use any set of coordinates she or he wants. Some time coordinate (wristwatch? distant pulsars?) and spatial spherical coordinates with the infaller always at the spatial orgin, East-North-Up coordinates originating on the (spinning) black hole, etc. are all (pardon the pun) attractive in these circumstances.

Also, defining exactly where "falling in" happens is tricky, even for the infaller. Visser 2014 on horizons: <https://arxiv.org/abs/1407.7295>, second sentence third paragraph of the Introduction section ("These distinctions even make a difference when precisely defining what a "black hole" is -- the usual definition in terms of an event horizon is mathematically clean, leading to many lovely theorems [20], but bears little to no resemblance to anything a physicist could actually measure.")

[rbanffy]

Good point, but what would that observer perceive as they cross the horizon after the end of time?

[raattgift]

First two preliminaries:

The crossing is not at a straightforward conception of "the end of time" in an expanding universe, since most possible observers are carried away from the final fall-in by the expansion of the universe, so there's nobody orbiting "at infinity" who could in principle see the infall take "an infinite time".

Horizons are part of the causal structure of the entire universe, black holes, planets, toads, warts, and all. The horizon is dominated by the central mass and spin, but not fully determined by it. The horizon in a close black hole binary (or triple) gets very complicated. ("The horizon" is not even necessarily physically measurable, and with black hole evaporation might not even exist, although there are other features which can be indicative of the point of no return for an infaller).

Preliminaries done, there is the "no drama" conjecture. Given a large enough black hole in a quiet enough setting a freely-falling infaller will not know she or he has passed the point of no return, perhaps for several minutes according to his or her wristwatch.

That's because the tidal curvature at the point of no return gets very small as we take the mass of a slowly-spinning black hole above millions of stellar masses, and that's the curvature that's relevant in spaghettification, the leading cause of death of astronauts entering isolated black holes.

Of course, most of the black holes we have found are far from isolated (otherwise we probably wouldn't see them with current equipment), so an infaller is likely to be blasted apart by hard X-rays and superhot gas instead of falling straight in.

The observables for something strongly accelerating into a black hole for a faraway orbiting observer can be quite different; unlike for speed there is no maximum acceleration in relativity. One would have to find a limit to acceleration in the behaviour of matter. An astronaut is not going to survive anything like the acceleration needed to make much difference to the distant orbiting observer though.

The distant observer in the not-really-our-universe Schwarzschild model and seeing the infinitely-prolonged final infall is at rest with respect to the central mass. Different observers, e.g. ones shooting themselves into the same black hole, or hovering just above a different black hole, can see qualitatively different things.

Generically, outside observers will see a dimming and shrinking of (practically) any infaller closer to the black hole than the observer. Many such observers will lose sight of the infaller before the infaller has truly hit a point of no return. Consequently some observers could find themselves seeing a presumed-lost astronaut grow brighter and bigger again, and leave the vincinity of the black hole. (Substitute gas, dust, and parts of stars for astronaut in the previous sentence, and that is what the Event Horizon Telescope collaboration, among others, searches for.)