Bad Astronomy from Lawrence Krauss

[Rocky1775]

I was watching a show on black holes a few weeks ago (I believe it was an episode of How the Universe Works), and the show raised the question "Why doesn't the accretion disk just fall into the black hole?". Their answer, complete with video of Lawrence Krauss reiterating their bad astronomy, was that "centrifugal force" acting on the orbiting matter counterbalances the gravity of the black hole. My astronomy professor pounded into the heads of his students that there is no outward force acting on orbiting objects.

I suppose offering a "centrifugal force" explanation of orbits is easier to convey than the reality, but this just confirms my observation that you can't trust an explanation that authorities on physics give to laymen. Somehow generating an answer that is quick and satisfying to the listener is more important than the truth of the matter. I suppose they tell their kids that babies are delivered by storks.

[grant hutchison]

Some day I'm going to write an article entitled, "Saying 'centrifugal' doesn't mean you're a bad person."
Centrifugal force pops up naturally if you try to retain Newton's Laws in a rotating reference system. It's often called a pseudoforce, because it appears only in a non-inertial frame of reference. So when Krauss says "centrifugal", he's tacitly adopting the rotating reference frame familiar to all of use from fairground rides.
As usual xkcd has something suitable to say on the topic.

Grant Hutchison

[swampyankee]

A long time ago, one of my physics profs said something like "it acts like a force; it's a force." I think he was talking about the Coriolis effect, but the same logic applies.

[John Mendenhall]

V

Some day I'm going to write an article entitled, "Saying 'centrifugal' doesn't mean you're a bad person."
Centrifugal force pops up naturally if you try to retain Newton's Laws in a rotating reference system. It's often called a pseudoforce, because it appearys only in a non-inertial frame of reference. So when Krauss says "centrifugal", he's tacitly adopting the rotating reference frame familiar to all of use from fairground rides.
As usual xkcd has something suitable to say on the topic.

Grant Hutchison

Duh-uh. Of course, non-inertial reference frame. Thank you for the nice ah ha wave, Grant.

[Rocky1775]

I'm afraid Mr. Bond isn't in orbit. His ride, and the ones at the fairgrounds, actually are pushing on the passanger, and the normal force the passanger exerts back is centrifugal, and very real. Pseudoforce sounds a little like pseudoscientific geocentrism to me. You can mathematically put the Earth in the middle, and have everything flying around it, but its not such a good way to understand what is happening.

[grant hutchison]

I'm afraid Mr. Bond isn't in orbit. His ride, and the ones at the fairgrounds, actually are pushing on the passanger, and the normal force the passanger exerts back is centrifugal, and very real. Pseudoforce sounds a little like pseudoscientific geocentrism to me. You can mathematically put the Earth in the middle, and have everything flying around it, but its not such a good way to understand what is happening.

In Newtonian terms, Bond is in orbit. A central force is deflecting his straight-line inertial trajectory, just as under Newtonian gravity. And Krauss is clearly framing this as a Newtonian problem (not invoking General Relativity), because he's talking about a balance of forces.
Here's A.P. French, in his classic Newtonian Mechanics textbook for the M.I.T. Introductory Physics series. This is Chapter 12, "Inertial Forces and Noninertial Frames", in which he's building towards a discussion of orbits in which he invokes the "centrifugal potential energy" curve calculated from centrifugal force in the rotating frame.

We shall now consider a particular kind of inertial force that always appears if the motion of a particle is described and analyzed from the standpoint of a rotating reference frame. This force--the centrifugal force--is familiar to us as the force with which, for example, an object appears to pull on us if we whirl it around at the end of a string.

In Chapter 13, "Motion Under Central Forces" he begins his treatment of orbital mechanics and derives the centrifugal potential energy of a rotating frame:

The term l²/2mr² is often referred to as the "centrifugal" potential energy, because the force represented by the negative gradient of this potential energy is given by ...

F_centrifugal = mr(dθ/dt)²

which is indentical with the centrifugal force mω²r in a frame rotating at an angular velocity ω equal to the instantaneous value of dθ/dt.

Sitting yourself down in the rotating frame of an orbiting object is actually quite a fruitful way to come at Newtonian orbital mechanics - it's certainly how I learned it. More significantly, it's how French taught it at M.I.T., which is where Krauss got his Ph.D. in physics.

ETA: "Pseudoforce", by the way, is a term introduced by Richard Feynman in The Feynman Lectures on Physics. He wasn't prone to pseudoscientific geocentrism, as far as I'm aware.

Grant Hutchison

[grapes]

I'm afraid Mr. Bond isn't in orbit. His ride, and the ones at the fairgrounds, actually are pushing on the passanger, and the normal force the passanger exerts back is centrifugal, and very real. Pseudoforce sounds a little like pseudoscientific geocentrism to me. You can mathematically put the Earth in the middle, and have everything flying around it, but its not such a good way to understand what is happening.

If you can't understand it both ways, you probably don't understand it, though.

Why would you say that the fairground forces are very real but not the orbital ones? Gravity pulls on the passenger in orbit, and the force the passenger exerts back is centrifugal, no?

[grant hutchison]

If you can't understand it both ways, you probably don't understand it, though.

And there will be a lot of things you simply can't understand if you don't understand the Newtonian central force model of gravity.
Unlike the "stork" hypothesis for childbirth, Newtonian central-force gravity is a hugely successful and productive way of understanding the world.

Grant Hutchison

[Ken G]

What will be very refreshing to me is if there is ever a time when people on this forum start to recognize that the language we use to describe physical situations is just that-- the language we use, to gain understanding. There is not a "correct language", there is only using the language correctly, whatever language we are choosing. We only need to be clear what we are doing, because no language is going to be "what is really happening"-- science doesn't do language like that. There's no better example of this than the language involved in "centrifugal forces."

Indeed, we can consider even more controversial situations to make the point-- we can imagine being run over by a car. Almost anyone, in that situation, would invoke the language that we are being squashed by a force coming from the car. That would be like calling the force that kills Bond in the centrifuge the "centripetal force." However, there is actually nothing at all wrong with saying that the force that squashes us is not a force coming from the car, but rather the "inertial force" that comes from our own inertia. This is also why you can kill a person with a 30 mile-per-hour car, but you cannot kill a fruit fly that way. We can use language that simply enters our own reference frame, and says that we are "squashed" between a force from the car and our own "inertial force" (which is the equivalent way d'Alembert framed all of Newton's laws). We can say that a force from the car alone cannot squash us-- forces don't squash, what is fatal is being caught between two forces. Although this is certainly a nonstandard language, it is not only correct when used correctly, it is actually rather insightful-- the trauma to our bodies will be, after all, not the trauma of a single force, but the trauma of a crush between two forces. Of course, as Grant said, one of the those forces will be a "pseudoforce", meaning it will not obey Newton's three laws about forces. But it will be a useful force to contemplate anyway-- so long as we contemplate it correctly.

[publiusr]

Also, there is zero gravity and microgravity. You may be in free fall--but you can still feel enough of a tug so that a keel of a satellite--or a face of a body tidally locked--still point to the body you orbit.

If you have any mass at all--you never really have zero gravity.

[Rocky1775]

If you can't understand it both ways, you probably don't understand it, though.

I think I understand it pretty well.

Why would you say that the fairground forces are very real but not the orbital ones?

I don't. There is a very real centripital force in both cases.

Gravity pulls on the passenger in orbit, and the force the passenger exerts back is centrifugal, no?

In this case, the centriufugal force the passanger exerts on the planet is trivial. Thats why the Earth doesn't orbit the astronaut, and the Sun doesn't orbit the Earth, mostly. Well, they do an itty bit, but not like, say, Pluto and Charon.

[Rocky1775]

Indeed, we can consider even more controversial situations to make the point-- we can imagine being run over by a car. Almost anyone, in that situation, would invoke the language that we are being squashed by a force coming from the car. That would be like calling the force that kills Bond in the centrifuge the "centripetal force." However, there is actually nothing at all wrong with saying that the force that squashes us is not a force coming from the car, but rather the "inertial force" that comes from our own inertia. This is also why you can kill a person with a 30 mile-per-hour car, but you cannot kill a fruit fly that way. We can use language that simply enters our own reference frame, and says that we are "squashed" between a force from the car and our own "inertial force" (which is the equivalent way d'Alembert framed all of Newton's laws). We can say that a force from the car alone cannot squash us-- forces don't squash, what is fatal is being caught between two forces. Although this is certainly a nonstandard language, it is not only correct when used correctly, it is actually rather insightful-- the trauma to our bodies will be, after all, not the trauma of a single force, but the trauma of a crush between two forces. Of course, as Grant said, one of the those forces will be a "pseudoforce", meaning it will not obey Newton's three laws about forces. But it will be a useful force to contemplate anyway-- so long as we contemplate it correctly.

Also note that Bond is getting squashed, and an astronaut in orbit isn't. In fact, unless he looks out the window, he can't even tell that he is in orbit. (Assume he has zero length, and his spacecraft has zero volume, so the tiny tides don't give it away.)
(And no cheating with a gyroscope, either. ;-)

[Ken G]

Also note that Bond is getting squashed, and an astronaut in orbit isn't. In fact, unless he looks out the window, he can't even tell that he is in orbit.

That's true, but the point is, there are (at least) three very different ways to explain an orbit of a point-particle (our approximation to the astronaut), and the typical physicist learns all three:
1) Newton: There is exactly one real force on the orbiting astronaut, and it is gravity, and it causes the astronaut to accelerate, which he/she needs to do in order to go in a circle.
2) Einstein: There is exactly zero force on the orbiting astronaut, so he/she does not accelerate at all, but rather follows the inertial path determined by gravity and their orbital speed, and that path is a circle.
3) d'Alembert: There are exactly two forces on the astronaut, one gravity and one the inertial force (a specific case of which we call the centrifugal force), and those forces balance in the frame of the astronaut. Everything around them will appear to accelerate in a circle around them, but the astronaut cannot perceive acceleration of their own frame, it is their own frame.

Each of these pictures has its strengths and weaknesses. Only #2 is accurate for strong gravity, but all three work well for weak gravity. Above all, though, we must not imagine that any of those three languages is "correct", for each are known to fail in some context.

[Rocky1775]

That's true, but the point is, there are (at least) three very different ways to explain an orbit of a point-particle (our approximation to the astronaut), and the typical physicist learns all three:
1) Newton: There is exactly one real force on the orbiting astronaut, and it is gravity, and it causes the astronaut to accelerate, which he/she needs to do in order to go in a circle.
2) Einstein: There is exactly zero force on the orbiting astronaut, so he/she does not accelerate at all, but rather follows the inertial path determined by gravity and their orbital speed, and that path is a circle.
3) d'Alembert: There are exactly two forces on the astronaut, one gravity and one the inertial force (a specific case of which we call the centrifugal force), and those forces balance in the frame of the astronaut. Everything around them will appear to accelerate in a circle around them, but the astronaut cannot perceive acceleration of their own frame, it is their own frame.

Each of these pictures has its strengths and weaknesses. Only #2 is accurate for strong gravity, but all three work well for weak gravity. Above all, though, we must not imagine that any of those three languages is "correct", for each are known to fail in some context.

In what context does Einstein fail?

[Amber Robot]

I was watching a show on black holes a few weeks ago (I believe it was an episode of How the Universe Works), and the show raised the question "Why doesn't the accretion disk just fall into the black hole?".

It *is* falling into the black hole. It just keeps missing!!

[Rocky1775]

It *is* falling into the black hole. It just keeps missing!!

That is a much better answer than Krauss gave.

[grant hutchison]

Also note that Bond is getting squashed, and an astronaut in orbit isn't.

Which tells us something deep and remarkable about the force of gravity - it acts like a pseudoforce!
So we can think of the astronaut's comfortable state as arising from the balance of two pseudoforces, gravity and centrifugal, whereas Bond is suffering the balance between one "real" force (centripetal) and one pseudoforce (centrifugal).

(Scare quotes because all these forces are "real" - they're just real in different ways.)

Grant Hutchison

[Ken G]

In what context does Einstein fail?

The Planck scale.

[Rocky1775]

The Planck scale.

A very small failure, indeed.

[grant hutchison]

A very small failure, indeed.

Or an absolute huge one, if you consider that it's a disagreement with our other most fundamental theory of the Universe.

Grant Hutchison

[Jeff Root]

If general relativity only fails at the Planck scale because it
disagrees with quantum mechanics, then I see no reason to
presume that it fails at all. Quantum mechanics is *certainly*
incomplete. Something added to it could require that no real
matter or radiation or anything can be as small as the Planck
scale, in which case there would not and could not be any
actual conditions under which GR could fail.

-- Jeff, in Minneapolis

[Ken G]

A very small failure, indeed.

No one said it wasn't small, from a human perspective-- only that it is a failure. It cannot be "correct", and that was the only point being made about it-- in science, we select the idealization we like, and we try to tailor its complexity and accuracy to our needs. There is often a tradeoff there, but none of the choices are "correct" in any absolute sense. So it is with the "centrifugal force."

[Ken G]

If general relativity only fails at the Planck scale because it
disagrees with quantum mechanics, then I see no reason to
presume that it fails at all.

Actually, I take rather the opposite perspective. I see no reason to presume it should succeed! In fact, I see no reason to presume it does not break down well before you get to the Planck scale, that is only the place that it becomes inconsistent with another great theory that is also likely to break down before you get to that scale: quantum mechanics. What amazes me most about physics is not that its theories break down, but rather that they often successfully extrapolate well outside the domains where they have been tested. But there's often a limit to how much extrapolation they can support. It seems to me we must always resist the fallacy of believing we are right until proven wrong, that approach has a habit of forcing us to eat more "humble pie" than is strictly necessary.

[Jeff Root]

GR might fail at any point beyond our observations, but I
see no reason to *presume* that it fails, except as a means
to explore or hedge against the possibility that it does fail.

If a relation fails when a value goes beyond some limit, but
it is fundamentally impossible for the value to ever reach the
limit, does the relation fail?

-- Jeff, in Minneapolis

[Ken G]

GR might fail at any point beyond our observations, but I
see no reason to *presume* that it fails, except as a means
to explore or hedge against the possibility that it does fail.

Then you do see a reason to presume it will fail! In science, this is called "skepticism." I might add one more reason: the track record of history. But of course we cannot know if something will fail until we look for it to fail, and to help motivate that search, skepticism is useful, though other techniques are available as well.

If a relation fails when a value goes beyond some limit, but
it is fundamentally impossible for the value to ever reach the
limit, does the relation fail?

It sounds like you are asking about the issue of "burden of proof." Is the burden of proof on the person who claims a relation succeeds to show that it succeeds, or on the person who claims it fails to show that it fails? I would say both-- so lacking any such demonstration, the safe stance is agnosticism. But when talking about what we expect, rather than what we claim, skepticism says to expect it to fail. Expectations are simply strategies for forming testable hypotheses, they are judged by how they advance science, not whether they are right or wrong. The expectation that the speed of light would be variable as the Earth orbits the Sun was a very valuable expectation that was responsible for a Nobel prize-- when found to be wrong. The important thing is to not let your expectations cause you to fail to do the test-- ergo skepticism.

[Reality Check]

If general relativity only fails at the Planck scale because it
disagrees with quantum mechanics, then I see no reason to
presume that it fails at all.

General relativity cannot fail because it disagrees with QM because QM is a different theory that is not about gravity and so can never disagree with GR. An equally incorrect statement would be to say solid state physics fails because it does not describe gases. General relativity fails because it has singularities. These are below the Planck scale so GR "fails at the Planck scale" by itself.

[Jeff Root]

GR might fail at any point beyond our observations, but I
see no reason to *presume* that it fails, except as a means
to explore or hedge against the possibility that it does fail.

Then you do see a reason to presume it will fail! In science,
this is called "skepticism."

Are you equating "scientific skepticism" with the null
hypothesis? I won't argue against that ... but ...

The comment I was responding to was your assertion that
GR fails at the Planck scale. Not that it could fail or that it
might fail or that we have no reason to presume it doesn't
fail. You asserted that it *does* fail at the Planck scale.
I'm not aware of any reason to think or presume that.

If a relation fails when a value goes beyond some limit, but
it is fundamentally impossible for the value to ever reach the
limit, does the relation fail?

It sounds like you are asking about the issue of "burden of
proof." Is the burden of proof on the person who claims a
relation succeeds to show that it succeeds, or on the person
who claims it fails to show that it fails?

I don't think that's what I'm asking about at all. I'm saying
that you claimed GR fails in a situation that may not be able
to exist, for reasons other than GR failing. If a situation
cannot exist, does it make any sense at all to say that a
theory fails in that situation? I personally think it does not.
If a conflict with QM in a hypothetical situation is the only
reason to say that GR fails (which seems to be the case),
and the hypothetical conflict might possibly be resolved by
a change or addition to QM alone (which also seems to be
the case, even if people who want to quantize GR often
disagree), then I think there is no reason to claim that GR
ever fails.

I would say both-- so lacking any such demonstration, the
safe stance is agnosticism.

I don't disagree with that. I'm certainly not claiming that
GR doesn't fail at the Planck scale, and even more certainly
not claiming that it doesn't fail somewhere. I just don't
see any reason to think it *does* fail at the Planck scale.

And I'm trying to get confirmation or refutation of the idea
that the reason you specify "Planck scale" is that the Planck
scale is where GR and QM come into conflict.

But when talking about what we expect, rather than what
we claim, skepticism says to expect it to fail. Expectations
are simply strategies for forming testable hypotheses, they
are judged by how they advance science, not whether they
are right or wrong.

I see no reason to claim that GR fails at the Planck scale.
It *might* fail anywhere it hasn't been tested.

-- Jeff, in Minneapolis

[Jeff Root]

If general relativity only fails at the Planck scale because it
disagrees with quantum mechanics, then I see no reason to
presume that it fails at all.

General relativity cannot fail because it disagrees with QM
because QM is a different theory that is not about gravity
and so can never disagree with GR.

It appears that GR and QM conflict when GR predicts an
energy density greater than QM can handle. That may
occur at the Planck scale. If actual observational evidence
of the conflict can be had, I expect that it will show up well
before the Planck scale.

An equally incorrect statement would be to say solid state
physics fails because it does not describe gases. General
relativity fails because it has singularities. These are below
the Planck scale so GR "fails at the Planck scale" by itself.

By "below the Planck scale" I presume you mean "at an even
smaller scale than the Planck scale". I agree that zero is
"below the Planck scale". The singularity in GR is when the
scale goes to zero. At that point, the energy density becomes
infinite.

I would not say that GR fails when the scale is zero.
Instead I say that zero is the limiting value. In addition,
I say that there may be a smallest possible value which
is greater than zero.

-- Jeff, in Minneapolis

[Swift]

This thread seems to have gotten completely off the original topic and doesn't seen to have anything to do with either Lawrence Krauss or Bad Astronomy. Before someone gets themselves infracted, I'm closing this thread.

If someone has a very good reason to reopen it, Report this post and 'splain.