James Webb Space Telescope finds evidence for alternate theory of gravity

Another comment linked to https://tritonstation.com/new-blog-page/, which is an excellent read. It makes the case that GR has never been tested at low accelerations, that is might be wrong. But we know for a fact MOND is wrong at high accelerations. Unless your theory can cover both, I don't see how it can be pitched as an improvement to GR.

Edit: this sounds a bit hostile. to be clear, I think modified gravity is absolutely worth researching. but it isn't a silver bullet

There are relativistic versions of MOND, for example, TeVeS [1], but they all still have some problems.

[1] https://en.m.wikipedia.org/wiki/Tensor%E2%80%93vector%E2%80%...

And yeah, that seems like pretty terrible cheating. It's one thing to hang a big theory on a single conjecture, but you still need to be trying to prove the conjecture.

[citation needed]

The consensus is that gravity - outside of extreme mass/energy environments - works just as Newton described it to many many decimal places.

Emphasized part added because people in the replies thought that I literally think that General Relativity is somehow wrong. Don't be dense. All I'm saying is that gravity at galactic scales works as Newton described it. General Relativity has extremely tiny effect at those scales.

Newtonian gravity is an approximation. A perfectly acceptable one in many contexts, but still measurably incorrect.

Also Newton's laws famously could not account for Mercury's orbit. Mercury is just an ordinary planet orbiting an ordinary star. Nothing extreme is involved. He knew his laws were incomplete. But they were so dead-on in basically every other scenario that could be physically observed at the time that he figured there was some small tweak missing (or maybe another planetary body that hadn't been spotted yet).

But ok, let me put it this way: outside of extreme energy/mass environments, gravity is described by Newton's law of gravitation with very high precision. If you look very hard, you may notice differences on the order of 10e-MANY. But for all intents and purposes, gravity is Newtonian in 99.99999% of the universe.

If we are asking whether MOND is useful, then the answer is probably yes. You might use it for simulations of galaxy formation where Newtonian gravity is considered a reasonable approximation today. But MOND is not a correct model of the universe. There is no place in the universe that Newtonian gravity applies, only places where the error is an acceptable trade-off for simpler calculation.

Galactic dynamics is governed by gravity, which is Newtonian at those scales.

All I'm saying is that the effect of general relativity at galactic scales is so minuscule, that galactic dynamics is - for all intents and purposes - governed by the Newtonian limit of gravity.

If you propose that gravity doesn't behave like the Newtonian limit at those scales, then you're contradicting general relativity as well, since the far-field limit of the Schwartzschild metric is literally Newton's inverse square law.

In layman terms, modified Newtonian gravity, that the article talks about, is an attempt to explain why galaxies don't rotate the way they should according to Newton (and Einstein, because at those distances the two are the same!!!).

But sure, newton is good enough to handle most ground based scenarios where we only care about forces at low precision.

The article is about modified Newtonian dynamics (MOND), which is a theory that modifies Newtonian gravitation to fix some observed differences in galaxies' motion, without invoking dark matter. The original commenter then proclaims "haha, MOND cannot be right, because we know that Newtonian gravity is incorrect". Yeah, no shit Sherlock; it is "incorrect" because it is just a limiting case of general relativity. But that's completely besides the whole point of MOND, which tries to "fix" gravity at galactic scales, which is a Newtonian regime even with general relativity. MOND is trying to tweak the Newtonian formula at those extreme distances, and if it works, then maybe it can be worked out to be a limiting case of a "modified general relativity", just as Newtonian gravity is a limiting case of GR. Got it?

For precession of perihelion of Mercury we mostly noticed because any error is cumulative over time and we could integrate over an arbitrarily wide timebase. The relativistic effects are <10^-8 of the total, around 1/10th of the change imparted by Newtonian gravity of planets much, much further away. The BepiColombo orbiter should allow us to correct for the relativistic effects of other planets' pull on Mercury, but it's expected to be a change of <10^-12.

So I guess "many, many decimal places" is in the ballpark of 6-12.

Fwiw, we know for a fact also that for edge cases GR is wrong because it doesn't agree with quantum mechanics (unless QM is wrong), so it's maybe not right to take GR as gospel, especially for a theory that only seems to also change GR in edge cases, and the only reason why "it doesn't agree" might amount to "the math is hard and the physicists haven't put enough work in yet"

To wit, accepting a mond-ified GR is probably not going to change how GPS works so the claim that "GR has withstood the test of time and engineering" is not a totally solid refutation of MOND

There's zero mention of MOND being a rejection of general relativity.

OF COURSE, any tweaking of Newton's formula at galactic scales will necessarily invalidate general relativity, since general relativity predicts Newton's formula at those scales! But MOND tries to work backwards: they propose a modification of the far-field Newtonian formula, and the belief is that it can eventually be worked out to be a limiting case of a "modified general relativity", for lack of a better name. Just how Newtonian gravity was eventually worked out to be a limiting case of a theory called general relativity.

If someone emerges with a proof that the two systems are irreconcilable then yeah you have an argument that it's "shitting on GR"

It absolutely does not. Newtonian gravity occurs instantly. It has no notion of information taking time to propagate. But we know gravitational waves happen, so Newtonian gravity is wrong _at even very large scales_. If the sun disappeared Newton tells us we'd know immediately. In GR we'd know about 8 min later.

The bigger problem is not that the quantitative effect is large, but that the _qualitative_ difference of going from the instantaneous effect to one that needs to propagate is enormous. It's the whole point of relativity as a concept.

Even going to GEM as a true, non-singular linear approximation of GR would be a step up from Newton's laws, at least there we can have gravitational waves and causal flow of information.

Mass of Sun: Ms = 1.99e30 kg

Distance to Mercury from Sun: Rm = 5.83e10 m

Mass of Milky Way galaxy: Mg = 6e42 kg

Q: At what distance R from the Milky Way would something have to be to experience the same gravitational field strength from the Milky Way that Mercury feels from the Sun?

A: We want R such that Ms/Rm^2 = Mg/R^2 or R = Rm sqrt(Mg/Ms) = 1.0e17 m.

Let's convert that to lightyears. There are 9.46e15 m/ly. The final result is 10.75 ly. Note that everyplace that close to the center of mass of the Milky Way is inside the galaxy. Anything actually outside the galaxy would be at least 5000 ly away and feel a gravity field at most 1/200000th as strong as what Mercury feels.

For Earth use the same calculation from above but replace Mg with the mass of the Earth, 5.97e24 kg. That gives that the distance from Earth where something would feel the same field strength from Earth that Mercury feels from the Sun is 1.0e9 m. That's a little over 4x the radius of the orbits of GPS satellites, so GPS satellites are feeling a little under 16x the field strength from Earth that Mercury feels from the Sun.

I don't think it doesn't agree. It's just that we never managed to neither formulate quantum mechanics on 4 dimensional space time nor quantize gravitational force. So we simply have no idea what happens in small scale in significant gravitational fieldd.

The fact that MOND fits a lot of the data troubled cosmologists, because they know that a General Relativistic theory is needed to explain pretty much the rest of gravity.

TeVeS is an extension to General Relativity that reduces to MOND in the non-relativistic limit. For comparison, General Relativity reduces to Newtonian gravity in the non-relativistic limit. The non-relativistic limit is when speeds and spacetime curvature are small.

What "edge cases" are you talking about? I agree that GR is not a quantum theory, but it's not established that that has to be a problem, nor is it a matter of "edge cases".

Similarly there are papers that tries to explain the effects attributed to dark matter on the scale of tenths and hundreds megaparsecs using just proper accounting of GR effects. They are rather speculative, but still they show that even on very huge distances Newtonian approximation may not be valid.

The word "mass" is used in physics in three different general contexts. First, we have mass in mass-energy, as in "how much energy can I get for trading in this mass?" Mass-energy is the coin paid as the price of existence. If it exists, it has mass-energy. Mostly mass for us. Mostly. We can skip that one for now.

The second context of mass is inertial. Mass has the property of inertia, of resisting a change in its direction or speed. It resists stopping if it is motion, and if it is stopped, it resists moving. The degree of the resistance is also called mass. Put a pin in this one.

The third context of mass is gravitational. Two masses, attracting one another because a force between them, a force which is not based on charge or the relatively nearby exchange of some more exotic bosons, no, just attraction based on how much mass is present. Nothing more special.

Now, curiously, values of each one of these seem to agree!

Einstein's absolute core concept in general relativity, the idea from which all else is built, is that inertial mass is identical to gravitational mass, not merely in number, but so fundamentally intertwined that there is no real difference between them, other than being two faces of the same coin. Now, that does not sound like much, but it gives birth to experiments such as an elevator which is falling toward versus an elevator floating far from gravitational sources, and that they are, from the inside of the elevator, impossible to differentiate.

Einstein then constructs general relativity from this, that the "m" in "F = ma" is identical to the first m in "F = -G m1 * m2 / r^2"

In MOND, the two ms are not identical, they only appear close most places, and so you cannot construct general relativity atop it. You will get most correct approximations but you're missing out in some cases.

Wrong. GR says that gravitation can be modeled as acceleration.

Do you have some references handy for this? Or are you talking about the work of Deur?

I meant it in the sense that "most of the cosmos runs on Newtonian gravity, therefore we can ignore GR" is similar to "most of the visible matter in the cosmos is hydrogen/helium, so we can ignore chemistry".

The interesting part is in the 0.0000001% that isn't like the others.

This is a wild take, given all the issues QM and QFT have.

It’s not really about one model being correct. GR is not a perfect model because its predictions don’t match what’s observed on the scale where QM gives predictions which do.

"Wrong" is overall a poor way of thinking about models. People would like a model which is both general and elegant, not simply a model which is "right". A large and very general model with a lot of parameters which can be well tuned to fit all the observations we have would be "correct" but I am not convinced it would be very useful.

you know what N in MOND stands for, right?

huh?!? there are GR corrections to Newton which include terms like 1/r^3 iirc

I see the same simplification in the most advanced writings. Namely 1) matter out to a radius can be treated as a point mass in the center and 2) we can ignore gravity from mass outside a radius because it all cancels.

These simplifications work for spherical shells or solids of uniform density. They do not apply to disks or rings (galaxies). Period.

More precisely, GR allows spacetime solutions which are geodesically incomplete.

> It's definitely wrong in those regimes.

No, that's too strong a claim. Most physicists believe that the solutions that are geodesically incomplete will turn out not to be valid in the regimes close enough to the endpoints of the incomplete geodesics. But that is a belief, not a proven fact. The solutions themselves are perfectly consistent mathematically.

The References in your "speculative paper" include at least five citations of the same author's previous work, at least one of which didn't even find its way into SCIRP's OALJ, and does not cite the Ludwig paper.

The full text also has such writing and editing gems, in the published version, as "the disk, the bugle and the halo of dark matter" immediately before eqn 23.

The paper's central argument is not obviously worth untangling, because the decomposition into the g and k fields (eqn 10) isn't Lorentz-invariant which raises questions about higher speed observables like cosmic rays, lensed background, "kicked" post-merger BHs, and even stars flung out of globular star clusters. There is no general transform avaiable in his equations of motion between two subsystems (e.g., outer stars and inner stars) related by a Lorentz boost. As far as I can tell the notational approach (and even the expression "gravitic field" to stand for the the gravitomagnetic field B_g) is unique to the author. It's so atypical (for quite ordinary equations) that I'd be surprised if there was any sort of reviewer or editor at all.

The author <https://people.epfl.ch/stephane.lecorre/> is a computer engineer in the university's architecture department, and claims a master's degree in theoretical physics <https://www.researchgate.net/profile/Stephane-Le-Corre-2>. I admire his continuing interest in and even investigations of "Astrophsics" (sic), but would not point to him as a persuasive expert as you have.

The Ludwig paper (EPJC 2021) is by comparison cited by 60, only a couple of which are self-cites. Whatever take one might have on Springer's approaches to open access journals, EPJC has an IF of almost 5.

Ludwig is an electrical engineer and plasma physicist. With the many cites on his set of related papers, it's clear he was not ignored by virtue of not being an astrophysicist or relativist. So we can't blame Le Corre's background for the lack of published engagement with his no-dark-matter-needed papers.

I don't think that Ludwig's gravitomagnetic vortex model is particuarly interesting in galaxy rotation curves because the fall-off off of the Lorentz force pulling outer margins of the galaxy inwards must have some arbitrary per-galaxy cutoff that also suppresses wild lensing effects at the cutoff point; we're interested mainly in doppler corrections on the HII spectrum rather than luminous stars (we don't necessarily need DM to explain flat rotations for the outer stars - we do need DM for rotating HI gas well beyond those outer stars) so the cutoff point is beyond the optical limb (meaning we should see wild lensing even in HST/WFC3); the gravitomagnetic effects must be smaller than the gravitoelectric effects (and capturing that somewhat in (v/c)^2 terms corrections to Newtonian/Keplerian orbits (v ~ 0.001 c in Andromeda-like galaxies) should be on the order of 10^-6 whereas in this approach we'd need corrections on the order of 10^-5 and higher for lower-mass lower-v dwarfs); and because the formulation does not work well with elliptical and irregular galaxies (both of which can have low circumferential rotational support - blobs of gas move radially in and out) without treating them differently from discoids (and when you do that in this approach you get divergences at galactic cores); and even for discoids there must be a minimum rotational support. More prosaically, the problem with the model is to avoid having to stabilize satellite dwarfs around a galaxy: you have to make the attractive Lorentz force not pull them right into the parent's middle and you have to avoid having satellites tear the crap out of the outer orbits of the parent galaxy's HI gas.

The paper's central idea certainly does not succeed as a general theory for flat rotation curves of HI dust as opposed to stars in circular orbits in a thin-disc plane.

However Ludwig's wasn't an obviously misguided idea, the paper's arguments are pretty clear, he's done follow-on work that is interesting, and the academic dialogue it produced is well deserved. But to say that anyone could use this paper to point to which mathematical object in GR (or which physical aspect of GR) stabilizes the relevant HI and dwarf orbits is, I wager, a huuuuge stretch.

Finally, quoting you:

> For example, properly accounting for GR effects is enough to explain the observed rotational curve for our Galaxy without the need for any dark matter hypothesis

This is not at all borne out by your choice of papers. Ludwig's text doesn't even mention the Milky Way.

Anyways, to claim that failing equivalence principle is disqualifying is begging the question since support of the equivalence principle depends on the observations... And already we observe the rotation curves are "messed up". If that means EP is violated, so be it?

You wouldn't argue against a symmetry violation like CP because "it makes the cute rule fail"

Every physical theory with singularities has has broken down in that regime. It's not even clear what it would mean for reality to permit singularities. That's a bit more than just a belief.

Can you give some examples? Note that GR has not even been tested anywhere close to the regime you are talking about.

> It's not even clear what it would mean for reality to permit singularities.

GR doesn't "permit singularities" in the sense I think you are using that phrase. "Singularity" in GR actually does not mean what I suspect you think it means, that things like spacetime curvature "become infinite". Notice that in my previous post I was careful to use the term "geodesic incompleteness", since that's what "singularity" actually means in the GR literature. And even in particular cases where there are invariants that increase without bound along incomplete geodesics, the limit points, such as r = 0 in Schwarzschild spacetime, are not actually part of the spacetime in GR. All invariants are finite at every point in the actual spacetime.

See Baez's paper, Struggles with the Continuum, https://arxiv.org/abs/1609.01421

The UV catastrophe is probably the most well known.

Most physicists believe that our best current theories, GR and quantum field theory, are approximations anyway ("effective theories" is the term often used in the literature), so that in itself is not a new idea. Baez's paper points at one fairly common hypothesis for why they are approximations and what the underlying theory they are approximations to might look like. I don't have an issue with that as a hypothesis; it's just something we aren't going to be able to test by experiment any time soon, since the most likely scale for where the approximation will break down, the Planck scale, is some twenty orders of magnitude away from the scales we can currently probe with experiments.

I don't think MOND dosen't agree with GR. It's just that we never managed to formulate MOND in a spacetime compatible with GR. So we simply have no idea what happens on the galactic scale in a significant gravitational field.

No, we don't know. Most physicists believe it, but that's not the same as knowing. We won't know unless and until we are able to actually do experiments in the relevant regime.