Are Black Holes Unfalsifiable and Therefore Not Science?

[twofish-quant]

As Chronos said, we have observed curious things in the universe -- immense sources of energy, jets, accretion discs, large gravitational effects on other objects, even gravitational lensing -- which can only be understood as the consequences of extremely massive (and dense) objects.

In the case of accretion disks they can only currently be understood in terms of massive and dense objects that seem to behave according to how black holes should behave under the rules of GR. For some of the accretion disks, it's not merely a massive and dense object, but a massive and dense object with no apparent surface.

The actual nature of these super dense bodies could be radically different than anything we think we know today.

Possible but not likely. The thing about accretion disks is that a lot of them have been studied enough to reduce the likelihood of "weird physics." If there is something radically different than what we think we know then we have to explain why all that radical stuff only seems to affect the core object and not anything else.

Personally, I think that the observational evidence for black holes is roughly equal to that of exoplanets, and a finding that "black holes don't exist" would be as shocking as "exoplanets don't exist."

 

[twofish-quant]

Please provide a world line that Contains a coordinate singularity.

Someone already put out a diagram that has one. The physics problem with singularities is that if you fall into a black hole, you'll hit it in finite time. If you never hit the singularity then there would be no point in worrying about what they are.

 

[Dmitry67]

http://www.valdostamuseum.org/hamsmith/DFblackIn.gif
It shows a worldline of infalling observer.
You can see his proper time over that line.

 

[twofish-quant]

One curious thing about black holes is that the thing that kills you (i.e. the tidal forces) really are large only for "ordinary" small black holes. If you have a large enough black hole, the tidal forces diecrease, and nothing particularly strange happens to you once you cross the event horizon.

 

[Dmitry67]

Yes. ANother interesting thing about the proper time of freely falling observer is that to have a longest lifespan you need to relax and enjoy the view. If you try to resist, pushing the pedal of your spaceship to the metal, then you decrease your proper time. But this a quite obvious property of pseudo-euclidean space.

 

[George Jones]

Please provide a world line that Contains a coordinate singularity.

I'm not sure what you mean. Relativity models spacetime as a semi-Riemannian differentiable manifold. What is the definition of "semi-Riemannian differentiable manifold?" Also, what is a "coordinate singularity?" I am a little unsure of the (general, not specific to this case) answer to the second question, but I do have something in mind.

 

[atyy]

I read an interesting article which asserts that the existence of black holes cannot be falsified, and therefore they do not qualify as science. Has anybody heard this argument before? Any comments?

http://arxiv.org/abs/gr-qc/9808035

"All the methods for finding black holes described above are indirect. They essentially
say that there is a lot of mass in a small volume. Direct proof that a candidate object is a black hole requires a demonstration that the object has the spacetime geometry predicted by Einstein’s theory. For example, we would like to have evidence for an event horizon, the one feature that is unique to a black hole."

"Black holes connect to a wide variety of fields of physics. They are invoked to explain high-energy phenomena in astrophysics, they are the subject of analytic and numerical inquiry in classical general relativity, and they may provide key insights into quantum gravity. We also seem to be on the verge of verifying that these objects actually exist in nature with the spacetime properties given by Einstein’s theory. Finding absolutely incontrovertible evidence for a black hole would be the capstone of one of the most remarkable discoveries in the history of science."

http://arxiv.org/abs/astro-ph/0701512

"Probing the properties of Massive Dark Objects: black holes?
spinning?"

"Observers cannot yet definitively confirm the form of the metric in the strong-gravity region, in order to prove that BHs are indeed described by the Kerr metric."

"Gravitational waves from an EMRI can be used to map the spacetime of the central massive dark object. The resulting ’map’ can tell us if the standard picture for the central massive object, a Kerr BH described by general relativity, holds."

 

[Phrak]

Someone already put out a diagram that has one. The physics problem with singularities is that if you fall into a black hole, you'll hit it in finite time. If you never hit the singularity then there would be no point in worrying about what they are.

Perhaps the problem is in semantics. That was the diagram of the Schwarzschild metric. We need a diagram or metric for an in falling observer. Without conflagrating coordinate systems, we can use the observer's Riemann normal coordinates, good throughout the entire in falling trajectory (and excluding the central singularity, it seems).

There will be finite time elapsed on the in falling observer's clock, and a singularity free metric, up to and not including the central singularity.

 

[piareround]

Perhaps the problem is in semantics. That was the diagram of the Schwarzschild metric. We need a diagram or metric for an in falling observer. Without conflagrating coordinate systems, we can use the observer's Riemann normal coordinates, good throughout the entire in falling trajectory (and excluding the central singularity, it seems).

There will be finite time elapsed on the in falling observer's clock, and a singularity free metric, up to and not including the central singularity.

While there are something things I disagree about some of your points, I do agree this debate is gradually deviating from the original question:

Whether or not black holes can be falsified.

I read an interesting article which asserts that the existence of black holes cannot be falsified, and therefore they do not qualify as science. Has anybody heard this argument before? Any comments?

Many people have mention observed objects that one group scientists thinks they are black holes, other scientists thinks they are are not black holes, and sadly another group of scientists who refuse to answer. We call these things http://en.wikipedia.org/wiki/Stellar_black_hole#Candidates" (here I am simply pointing to the list rather than the definition). Regardless of you stance on black holes, these black hole candidates (BHC's) have at least two primary properties of:

1. being themselves in a section of observable space or being in a section of observable space that does not emit/re-emit and/or reflect/refract light
2. have stuff orbiting around them that does emit/re-emit and/or reflect/refract light

1st Answer) So to give one answer to "can black hole be falsified", I would like to point out that that BHC's themselves can be falsified as not being black holes by these checking these two properties. However, I do not know of any examples of things we have confirmed are not BHC's. So I propose this as a question to the Relativistic Physics researchers out there:

• Does anyone know of a historical example of a specific former BHC that has confirmed as a not being a BHC?

Now if really take a look at experimental detection of black holes you will find that there is a parallel with experimentation of gravity waves. This parallel is no coincidence because, historically, the gravitational waves have been tied to astrophysics and to large massive objects including black holes.

2nd Answer) So it would seem to me that if we are to talk about the validity of detecting black holes, we should talk about the validity of Gravitational Waves and thus the accuracy http://en.wikipedia.org/wiki/Gravitational_wave_detector" . However I do not know enough about Gravitational Wave detectors to even begin talking about whether or not gravity waves are detectable. Yet, there are some particular questions about these detector's precision, that I think should be answered such as:

• How does one calibrate something like a LIGO? How do you calibrate something to detect gravity waves before you start looking for them?
• How do you calibrate it so that you know that the gravity waves you do detect come from a particular source?
• How far out and to what frequency ranges does theory say these detectors can detect to?

I firmly believe that answers the above questions, will lead toward an answer of the validity of experimental detection of black holes from other massively large objects better than any debate of semantics of black hole theory can.

I also hope they will show that a single black hole can indeed be falsifiable, but it is, like most modern research, expensive.

 

[twofish-quant]

Many people have mention observed objects that one group scientists thinks they are black holes, other scientists thinks they are are not black holes, and sadly another group of scientists who refuse to answer.

I think this misstates the opinion and consensus in the field. All of the candidates for black holes are objects which pretty much everyone in the field think are black holes. The question is "how sure" are people that these objects are black holes. but that's a different question.

 

[Chronos]

The 'sure' thing has been an issue in this thread. No need to deal with that right now. Nothing is 'sure' in science. I am agreeable with 'highly probable' for the time being.

 

[Phrak]

My uninformed argument is fairly simple; that there are no black holes to be observed because it would require an infinite time for an event horizon to form, where time is being measured by Earthly observers.

My argument might be fairly easy to falsify: Locating any article in the literature that would validly support the formation of an event horizon of a collapsing mass in an outside observer's proper time. Any given distribution of charge, mass, or momentum, or displacement in distance or state of motion of the original mass would suffice.

Barring this, I doubt that a black hole can be measurably distinguished from a proto black hole. By proto black hole, I mean a collapsing mass, where there is (according to the proper time of a distant observer) mass remaining above the projected event horizon, however closely it resides.

 

[Dmitry67]

My uninformed argument is fairly simple; that there are no black holes to be observed because it would require an infinite time for an event horizon to form, where time is being measured by Earthly observers

So, your argument is, if we express it more formally, the lightcone from the horizon will never ever reach an observer on Earth?

Then Cosmology is not science either; because when we say something like 'universe is flat' or 'spacetime is open' we talk about the regions behind the cosmological horizons. They have the same property.

Looks like you insist on the very strong version of falsibiability; we have no choice but to relax these requirements. We have no choice but
1) to develop the theory that fits observable data
2) apply the same theory to the regions we can (never) observe or experiments we can never perform (like the inside regions of black holes, Big Bang, Planks energies, behind the cosmological horizon).
3) We assume that if theory is valid in 1) then it is valid in 2)
4) If there are other alternative theories we chose the simplest/the most beautiful.

 

[Phrak]

So, your argument is, if we express it more formally, the lightcone from the horizon will never ever reach an observer on Earth?

No, not at all. Outward propagating signals would be a matter for observational evidence. I am referring to the time that would be required for matter moving in the other direction--the collapse of infalling matter to form an event horizon, as defined in an external observers spacetime coordinates, as measured in the proper time of the external observer.

I’ve been trying to make this as broad as possible.

How long, per an outside observer, does it take collapsing matter to develope an event horizon per the outside observer?

 

[Dmitry67]

Now you are saying nonsense.

Imagine 2 roads. When they are parralel, 1 mile on one of then is equal to 1 mile on the other.

If they are not parralel, then there is a 'dilation' - say, 1.5 on one of then is equivalent to 1 mile on another

Now there is a fork and one raod turns at 90 degrees and hides in a tunnel. Now it does not make any sense to compare the distance!

The same in BH: even in SR there is no absolute meamaing of simultaneously of events in different location, and no absolute meaning of age unless objects can return sooner or later to the same point.

If one observer falls into the horizon and another stays on Earth, they will never meet So your claims about 'as defined in an external observers spacetime coordinates' does not make any physical sense.

It is exactly like asking 'but on what of 2 moving ships time is moving slower'? Yes, in some coordinate system events freeze at the horizon. But the coordinate system you are talking about is impared because it can not cover events inside the BH. In another coordinates system your claim is false. So what?

The following pictures use coordinate systems where there is nothing special at the horizon, time inside the BH becomes spacelike, and proper time (line length) is limited.

http://www.valdostamuseum.org/hamsmith/DFblackIn.gif
http://nrumiano.free.fr/Images/lightcones_E.gif
http://www.pitt.edu/~jdnorton/teaching/HPS_0410/chapters/black_holes/bh_lightcones_st.gif
http://physics.syr.edu/courses/modules/LIGHTCONE/pics/bh2.gif

 

[Dmitry67]

How long, per an outside observer, does it take collapsing matter to develope an event horizon per the outside observer?

I repeat, your question IS NOT PHYSICAL

You can ask 'when I receive the signal from infalling observer' - but we both agree that these signals are infinitely redshifted

There is no 'when' in GR in curved spacetime for the events in different locations.

'How long' can be defined as 'proper time' for the same observer.

'How long per outside observer' can be defined as:
* proper time of some of the observers when thy finally meet - it is not the case here;
* time calculated based on the arrival of the signals
* time calculated based on some coordinate system. But it depends on the coordinate system and dos not have physical meaning (example: in twin paradox, when one spaceships stops and starts to go back, it has to 'switch' its coordinate system which leads to assymetry and resolves the twin paradox)

 

[Jonathan Scott]

I repeat, your question IS NOT PHYSICAL

You can ask 'when I receive the signal from infalling observer' - but we both agree that these signals are infinitely redshifted

There is no 'when' in GR in curved spacetime for the events in different locations.

'How long' can be defined as 'proper time' for the same observer.

'How long per outside observer' can be defined as:
* proper time of some of the observers when thy finally meet - it is not the case here;
* time calculated based on the arrival of the signals
* time calculated based on some coordinate system. But it depends on the coordinate system and dos not have physical meaning (example: in twin paradox, when one spaceships stops and starts to go back, it has to 'switch' its coordinate system which leads to assymetry and resolves the twin paradox)

I understand your point, but I reckon that's cheating. For example, time rates near a massive object vary with potential, but from the point of view of an observer using a specific background coordinate system (say isotropic) we can use that as a valid model of what "really" happens.

According to almost any coordinate system EXCEPT that of a free-falling observer falling with the material into the forming black hole, such events take an infinite time and therefore do not complete within the lifetime of the universe. From the point of view of a falling observer, things happen in a finite proper time, but any attempt to map that time back to external space-time gets very mixed up, as time and space have swapped roles.

We know from experience with ordinary levels of gravity that this "slowing" of proper time is a physical effect; if you go very close to a black hole and manage to get away again, a lot of extra time will have elapsed. On those grounds, it seems that the infinite delay should be considered an equally physical effect.

So what happens from the point of view of a falling observer and how can this be consistent with the external view? I don't know, and I've not seen any explanation I can believe.

 

[Dmitry67]

1
using a specific background coordinate system (say isotropic) we can use that as a valid model of what "really" happens.

2
According to almost any coordinate system EXCEPT that of a free-falling observer falling with the material into the forming black hole, such events take an infinite time and therefore do not complete within the lifetime of the universe. From the point of view of a falling observer, things happen in a finite proper time, but any attempt to map that time back to external space-time gets very mixed up, as time and space have swapped roles.

1
And by changing that 'specific' coordinate system we can change what 'really' happens :) It is well known that just by walking in different directions in your room you can change what "now" "really" happens in the Adromeda galaxy by several years :)

2
Then you are violating the very central idea of the General Relativity - that ALL coordinate systems are equal in rights. Observers point of view on Earth is not more 'valid' in any sense than the point of view of a free falling observer.

Finally, it is not only "except the free falling observer" but a wide class of falling systems. Not only freely falling: if you resist falling into a black hole, but your engines are not powerful enough, then you are not freely falling but still you reach the singularity in finite proper time.

 

[Phrak]

I repeat, your question IS NOT PHYSICAL

Yes, well, I'll be sure to let Misner, Thorne and Wheeler know your assessment of their NONSENSICAL use of the language.

 

[Dmitry67]

Any arguments except the names?

 

[Jonathan Scott]

1
And by changing that 'specific' coordinate system we can change what 'really' happens :) It is well known that just by walking in different directions in your room you can change what "now" "really" happens in the Adromeda galaxy by several years :)

I disagree; I was talking about sequences of events which may or may not happen. These are the same in all coordinate systems. They can be described consistently in anyone of them. Time, space and directions can of course vary.

2
Then you are violating the very central idea of the General Relativity - that ALL coordinate systems are equal in rights. Observers point of view on Earth is not more 'valid' in any sense than the point of view of a free falling observer.

There's no difference in validity within the range in which the coordinate system is valid, but if part of it corresponds to a non-existent region of the universe, that doesn't necessarily count as valid.

Finally, it is not only "except the free falling observer" but a wide class of falling systems. Not only freely falling: if you resist falling into a black hole, but your engines are not powerful enough, then you are not freely falling but still you reach the singularity in finite proper time.

True, which is why I said "almost".

 

[Dmitry67]

I disagree; I was talking about sequences of events which may or may not happen. These are the same in all coordinate systems. They can be described consistently in anyone of them. Time, space and directions can of course vary.

No, because some observers cross the horizon

Let me explain my point of view in a different way.

"you-now" is a point in spacetime. There are infinitely many spacetime trajectories crossing that line. Each line represents some observer.

Some spacelines never hit the BH: for example, "you-remaining-on-EARTH". Some crosses the horizon: "you-decided-to-go-into-the-BH"

in GR all coordinate systems are equally vaid. So in some coordinate systems BH forms in finite time. In the others we lose communication with the inner parts of BH.

But if at least in some coordinate systems crossing "you-now" BH is formed, how can you say "it will never form"?

 

[Jonathan Scott]

No, because some observers cross the horizon

Let me explain my point of view in a different way.

"you-now" is a point in spacetime. There are infinitely many spacetime trajectories crossing that line. Each line represents some observer.

Some spacelines never hit the BH: for example, "you-remaining-on-EARTH". Some crosses the horizon: "you-decided-to-go-into-the-BH"

in GR all coordinate systems are equally vaid. So in some coordinate systems BH forms in finite time. In the others we lose communication with the inner parts of BH.

But if at least in some coordinate systems crossing "you-now" BH is formed, how can you say "it will never form"?

The region within the falling observer coordinate system which corresponds to falling past the horizon has no equivalent in the universe, as the time coordinate within the more conventional coordinate systems of its junction with the universe is "at infinity" (and the time coordinate on the other side is space-like).

Basically, as the observer falls, we see his clocks slowing to a halt, although he doesn't notice. This is just like the "stasis chamber" of science fiction. From the point of view of the observer, he doesn't know that he's been frozen in time, and if he is rescued and brought back to life, no time has elapsed. Even though from his point of view everything was normal up to the point where he was frozen, that doesn't necessarily mean that one can extrapolate to the assumption that he will be unfrozen again!

Some people insist that this is "not a problem", and insist that mathematical arguments or numeric simulations have shown that event horizons can form in finite time and things can fall inside them. I am personally yet to be convinced. Of course, in standard black hole theory, once collapse has started, it is inevitable, so the effect is the same anyway as a black hole.

 

[Dmitry67]

There is nothing to be convinced or not. Just look at the diagrams with the lighcones (I provided 4 links). What happens inside and around the BH is very simple. An attempt to think it terms of 'time dilations; leads to many misconceptions.

 

[Jonathan Scott]

There is nothing to be convinced or not. Just look at the diagrams with the lighcones (I provided 4 links). What happens inside and around the BH is very simple. An attempt to think it terms of 'time dilations; leads to many misconceptions.

I'm quite familiar with various different coordinate systems. If you take the falling observer's point of view there does not appear to be any problem with passing the event horizon (and hitting the singularity) in finite time. However, as I explained with the "stasis" example, that does not necessarily mean it's actually possible.

If you consider paths which fall very close to the horizon but then turn round and return, you see that the closer they go, the longer they take. This means that in the limit, as observed externally, the falling observer takes infinite time to cross the horizon, and could still in theory be rescued at any time during the entire future of the universe! So when do they actually fall in?

I find myself somewhat frustrated with books which show that in the falling observer's coordinate system, the horizon coordinate is passed in finite proper time and therefore conclude that the fact that this takes an infinite time as seen by an external observer is somehow just an irrelevant illusion. This isn't just a matter of different a point of view; it's a contradiction, and requires more than hand-waving to explain it.

 

[Dmitry67]

This isn't just a matter of different a point of view; it's a contradiction, and requires more than hand-waving to explain it.

I don't see any contradictions. The problem is that many people are trying to 'map' falling observer time to 'external' time, thinking in terms of

t' = x * t

where x is some variable. Obviously, you get into a problem when x becomes 0 or infinite. But who garanteed you that there is ONE 'river' of time and all times can be 'mapped' into each other?

Talking about the handwaving, what is NOT explained by the spacetime diagrams I provided? Let's talk about the physical things (what is observed, when signals arrive, etc) and avoid non-physical questions (when I am here on Earth, has black hole already formed? etc)

 

[Dmitry67]

If you consider paths which fall very close to the horizon but then turn round and return, you see that the closer they go, the longer they take. This means that in the limit, as observed externally, the falling observer takes infinite time to cross the horizon, and could still in theory be rescued at any time during the entire future of the universe! So when do they actually fall in?

I can ask you a similar question - without black holes.

Spaceship flies toward the Andromeda at very high speed. Then it turns back and returns back to Earth.

The trip took only few years measured by the clock on the spaceship, while on Earth it took millions. So while the austranaut on the spaceship aged only few years, his twin brother on Earth had died a long time ago.

The question you are asking "when actually (on the spaceship clock) the brother on Earth died?" Do you agree that this question is not physical?

 

[chronon]

But who garanteed you that there is ONE 'river' of time and all times can be 'mapped' into each other?

Unfortunately most people tend to think that the different times can be mapped into each other. For instance in the webpage mentioned earlier in the thread http://antwrp.gsfc.nasa.gov/htmltest/gifcity/bh_pub_faq.html#evaporate it says

I won't experience that cataclysm myself, though; I'll be through the horizon, leaving only my light behind.

implying that the light he leaves behind is seen after he has crossed the horizon

 

[Jonathan Scott]

Unfortunately most people tend to think that the different times can be mapped into each other. For instance in the webpage mentioned earlier in the thread http://antwrp.gsfc.nasa.gov/htmltest/gifcity/bh_pub_faq.html#evaporate it says implying that the light he leaves behind is seen after he has crossed the horizon

Do you agree that the external time coordinate shows that he could in theory be rescued at any time in the future of the universe (although of course nothing we know is strong or fast enough to do so)?

I reckon that makes it reasonable to say that he hasn't yet fallen in.

 

[chronon]

Do you agree that the external time coordinate shows that he could in theory be rescued at any time in the future of the universe .

No, after a certain amount of time has passed it will be impossible for light to catch up with him before he crosses the event horizon, and so it would definitely be impossible to send a rescue mission (which would travel slower than light). On the other hand, if he had powerful enough rockets, then he could at decide to turn round at any time before he has crossed the horizon.

 

[Dmitry67]

oops, chronon was faster...

Do you agree that the external time coordinate shows that he could in theory be rescued at any time in the future of the universe (although of course nothing we know is strong or fast enough to do so)?

I reckon that makes it reasonable to say that he hasn't yet fallen in.

No, I don't agree.

for me, the point of no return is at the event horizon.

for you, when I am too close to the horizon it is too late to decide to flight to me to save me: when you approach the BH trying to 'save' me you see how I 'unfreeze' and sink deeper and deeper BEFORE you approach

To simplify, let's say that a simple signal from you can save me: if you send a signal 'please return, I forgive you :)' and I receive it I turn around and return. But if I am too deep inside the black hole then it would take a while for the light signal to cover the distance to the black hole, and it would be too late!

Will you see the universe end?
If an external observer sees me slow down asymptotically as I fall, it might seem reasonable that I'd see the universe speed up asymptotically-- that I'd see the universe end in a spectacular flash as I went through the horizon. This isn't the case, though. What an external observer sees depends on what light does after I emit it. What I see, however, depends on what light does before it gets to me. And there's no way that light from future events far away can get to me. Faraway events in the arbitrarily distant future never end up on my "past light-cone," the surface made of light rays that get to me at a given time.

observer falling into the BH will not see how the Universe ends, and even won't see any signals sent too late!

 

[Jonathan Scott]

No, after a certain amount of time has passed it will be impossible for light to catch up with him before he crosses the event horizon, and so it would definitely be impossible to send a rescue mission (which would travel slower than light). On the other hand, if he had powerful enough rockets, then he could at decide to turn round at any time before he has crossed the horizon.

I don't remember anything about a time limit from when I studied this area (which was admittedly long ago); I thought that it was simply necessary to get closer and closer to the speed of light to catch up the later you started. Can you quote a specific reference, please?

 

[Dmitry67]

Just take one of the diagrams and draw the worldlines there.

 

[Jonathan Scott]

Just take one of the diagrams and draw the worldlines there.

That's an example of what I mean by handwaving.

What I'd like to see is a bit of maths showing that there's a limit incoming light cone beyond which signals cannot reach the falling observer's geodesic. I don't remember having seen that calculation, but I should be able to work that out myself; it would just be easier if someone could point me to it.

 

[Ich]

What I'd like to see is a bit of maths showing that there's a limit incoming light cone beyond which signals cannot reach the falling observer's geodesic.

Look https://www.physicsforums.com/showthread.php?p=2417483#post2417483". There's a finite redshift at the EH, disproving the "he sees the future of the universe" thing. This could also be a starting point for your calculations.