Hacker News Reader

> Am I correct in my understanding that these alternative theories are trying to prove that there's a way to explain these discrepancies without inventing something unobserved?

I would say MOND is just postulating a different "unobserved" thing--an alternate theory of gravity, or an additional "field" that looks like an alternative theory of gravity, instead of dark matter. I don't think MOND has any advantage in the Occam's Razor department.

Or, to put it another way, the word "unobserved" is not quite correct. We have actual observations that can't be accounted for, as best we can tell, with our current theories of physics, without adding some new element. Dark matter and MOND are just two different choices for what new element to add. Neither one can say it's not adding anything new.

> Why is that so controversial? To me, again as an outside observer, it feels so counter-intuitive to _invent_ a new type of matter you can't observe than to just say that your calculation is close but not right and to start over

MOND is not just doing a different calculation that is claimed to be more accurate within existing physics. It is adding new physics. See above.

"Starting over" would be something like finding errors in the calculations based on existing physics (GR and the observed distribution of visible matter) that, when corrected, removed the discrepancy between observations and theory. Nobody has done that, and the calculations based on existing physics have been checked every which way, so it seems highly unlikely that there is an error lurking there that hasn't been found.

> why the scientific community came to accept the one that is accepted

My understanding is that MOND, in general, does not close the gap between observation and theory as well as dark matter does. The paper referenced in this article appears to be claiming that its version of MOND "catches up" with dark matter in terms of closing that gap. I haven't had a chance to read the paper in detail so I can't say how credible I think that claim is at this point.

Dark matter (at least any of the non cold baryonic matter and standard model WIMP candidates) adds new physics too it’s just adds it in different field of physics that often is much less related to cosmology than gravity is.

> Dark matter (at least any of the non cold baryonic matter and standard model WIMP candidates) adds new physics too

I never said it didn't. The point is that both MOND and dark matter add new physics.

> I don't think MOND has any advantage in the Occam's Razor department.

It has one: there's no need to worry about finding a dark matter particle when there's no room in the standard model for one. I'm surprised how astrophysicists take for granted that such a particle will show up in a collider eventually, if they even think that far ahead.

>It has one: there's no need to worry about finding a dark matter particle when there's no room in the standard model for one.

This is a weird argument. The standard model is incomplete, we know it's incomplete, we don't expect it to be complete, and MOND doesn't fit in the standard model either.

Even if you limit it to things we have observed, the standard model simply doesn't explain gravity. It doesn't explain why neutrinos have mass. It doesn't explain matter/antimatter asymmetry.

We know gravity exists. We know neutrinos have mass. We know there is more matter than antimatter in the universe. These aren't remotely controversial, and the standard model either doesn't incorporate them or gets them wrong.

I said there is one advantage MOND has in terms of parsimony. It may not be a decisive advantage, but it is an advantage.

> I said there is one advantage MOND has in terms of parsimony.

It's not an advantage at all. Parsimony would mean not having to postulate as many new entities. MOND postulates one new entity; so does dark matter.

> I don't think MOND has any advantage in the Occam's Razor department.

MOND has a much more homogeneous behavior, that requires many less variables added to your theory. So I'll disagree, it has the Occam's Razor preference - for the same reason that ever instance of it has been quickly falsified.

> MOND has a much more homogeneous behavior, that requires many less variables added to your theory.

How so?

Dark matter theories can fit any mass distribution without a change to the theory.

Thanks! I think this helps clarify it quite a lot. The use of the word "observed" was a critical part of my misunderstanding, as to me it felt like it was saying that by not being observed it was "made up" instead, when really it's more accurate to say that it's part of the model and we know it exists but have a hard time describing it?

I have a background in physics (but not astrophysics) and to me MOND seems much more ad hoc than dark matter does. Dark matter posits that there is some kind of particle we have not observed directly that has a large effect on cosmological scales. We already believe that the standard model of particle physics is incomplete, so this is at least plausible. These particles are not in anyway "unobservable" in an absolute sense, they must at least interact via gravity in order to do what we need them to. They may even interact via the nuclear forces. They just don't interact via the electromagnetic force. We can make hypotheses about these particles, put constraints on their properties from observational data, use particle physics theory to come up with candidates, and do experiments to find them. They are not "unobservable", the are simply hard to observe and require advances in experimental technology. MOND on the other hand was discounted in the early days exactly because it was a fudge factor in Newton's second law. We saw some anomaly in galaxy rotation curves, MOND fixed it by adding a fudge factor to Newton, a theory which we already knew at the time to be incomplete! That is why most physicists preferred dark matter to MOND. Since then, other observational data has increased the evidence for the existence of dark matter (most famously the Bullet Cluster mentioned by others), making MOND that much more unpopular. There has been some work, like this RelMOND theory, that work to integrate the original MOND theory with general relativity. Critically, however, the way in which RelMOND is doing this is by positing the existence of an additional field that can act in the "clumpy" way required by observation. Basically, they have just reinvented dark matter. So I do not doubt this new theory fits the observation, but it does so by adding a new degree of freedom which plays the part of dark matter. This talk (https://www.youtube.com/watch?v=iu7LDGhSi1A) is somewhat technical, but it goes in to some detail as to why we can't explain current observation by messing around with gravity.

The Bullet Cluster (https://en.wikipedia.org/wiki/Bullet_Cluster) is a big reason why: it's a collision of two clusters of galaxies. We can see where the gas and stars are, but gravitational lensing indicates the bulk of the mass is elsewhere.

It is not that dark matter has not been observed. It has. It is called "matter" because its gravitational influence has been observed. Therefore at one level we know it exists because something is impacting gravity at large scales, we are just unsure about its other properties. We cannot see it directly but there are lots of things we don't see directly and that doesn't mean we aren't sure they exist. So it isn't a total invention to make the math work. It is a 'best fit' for a gap in our knowledge.

Dark energy is perhaps a different story.

IIRC, dark matter is supposed to be a different kind of matter that interacts "less than normal" with ordinary matter.

Or is there a version in which it's ordinary matter, just "hidden" by something or in some exotic "other dimension", black hole or parallel universe thingy?

It is thought to reacted less with normal matter because we aren't seeing it react with normal matter. So one potential explanation is a particle that is so tiny/fast, with no electrical charge, that it is can fly through planets without hitting anything: a "ghost" particle like the neutrino. The term used is WIMP (Weakly Interacting Massive Particle). Are we know is that it has mass, because it influences gravity. That's one possible explanation for dark matter.

Ideas based on extra dimensions are becoming less popular. We have evidence that gravity experiences the same dimensions that light does. Gravity waves from things like neutron stars merging seem to arrive at earth at the same time as light. So it is doubtful that extra dimensions can explain dark matter.

> To me, again as an outside observer, it feels so counter-intuitive to _invent_ a new type of matter you can't observe than to just say that your calculation is close but not right and to start over. Is it not a crutch?

Physicist here. If you're doing applied physics or engineering, this certainly would be a crutch. But when we're talking about fundamental physics, talking about new kinds of matter that nobody has seen before is not a crutch -- it's literally the core thing we do. That's what makes it fundamental!

Saw a track in the bubble chamber curving the wrong way? Invent a new kind of matter: antimatter.

Saw short-lived particles in the bubble chamber that shouldn't have made it there? Invent a new kind of matter: mesons that decay into the observed particles.

Problems with getting solar reactions to work out right? Invent a new kind of matter: neutrinos.

Amount of neutrinos detected not quite right? Invent multiple neutrinos and neutrino oscillations.

Saw some weird long-lived particles? Invent a new kind of matter: "strange" mesons and baryons.

Want to explain the pattern of mesons and baryons? Invent a new particle: "quarks", along with the stipulation that they can never be observed, even in principle.

Standard Model seems a little off-balance at this point? Invent a new particle: "charm" quarks to balance out the strange ones, at an energy high enough that nobody has seen them yet.

But the mesons and hadron patterns still aren't consistent with the Pauli exclusion principle! Invent a new force: color charge, carried by "gluons", which are also postulated to be unobservable.

Some particular meson and baryon decays acting weird? Invent a new force: the weak force, carried by "weak bosons", which are too heavy to be observable at the time.

Can't get the weak bosons to have mass? Invent a new interaction, the Higgs interaction, carried by an invented new field, the Higgs field, which gets a vev from an invented new function, the Higgs potential, whose elementary excitations are an invented new particle, the Higgs boson.

Of course, not every weird thing is explained by a new type of matter; many anomalies fade away after careful checking. But the anomalous observations that motivate dark matter persisted for almost a century, they're been only building in strength as we get more data, and all attempts we've made to explain them in terms of "normal" physics have failed. So the case for explaining it in terms of something new is at least as strong, in fact far stronger, than the examples I gave above.

I guess what is different about dark matter is that it has to outmass regular matter by a large factor. It feels unparsimonious to invent four-five times the mass of the known universe just to patch a discrepancy between observations and a theory of gravitation. It feels like the theory would better be adjusted to match observation than to patch observations to match theory.

Today I learned that the mass of the neutrinos we know about (which were similarly invented, though since detected) about matches the mass of all the stars.

Actually, in the context of astrophysics, that exact objection has been employed many times. For example, the most famous argument against heliocentrism was that it would require the stars to be ridiculously far away and ridiculously big to patch away the lack of parallax, which felt unparsimonious. Similarly, people believed that galaxies weren't galaxies, because it seems unparsimonious to expand the universe far beyond the Milky Way just to patch up some weird features of fuzzy nebula. And even in our galaxy, the mass in dust and interstellar gas exceed that in stars.

Literally all progress in fundamental physics is "just" "invented". Each time it must triumph against the objections of the same, thousand-year-old philosophical arguments.

Agreed.

Inventing a new particle to explain a discrepancy has worked really well before[0]. Past results are never a guarantee, obviously! But there are lots of ways that a dark matter particle might exist but be really hard to detect; it would not at all be surprising that another particle or class of particles exist that are just really, really hard to see.

[0] https://en.wikipedia.org/wiki/Neutrino#Pauli's_proposal

>A cursory search shows that "dark matter" hasn't been observed at any point in time. It looks like it's a sort-of mathematical placeholder for the "something" that's missing that we can't see (or can't see _yet_.)

Define "see" - that's the scope of the problem. We can see the effects of dark matter. We can measure the effects. What is or isn't seeing is hard to define on this sort of scale.

>Am I correct in my understanding that these alternative theories are trying to prove that there's a way to explain these discrepancies without inventing something unobserved?

More or less.

>Why is that so controversial? To me, again as an outside observer, it feels so counter-intuitive to _invent_ a new type of matter you can't observe than to just say that your calculation is close but not right and to start over. Is it not a crutch?

Because despite many thousands of attempts to provide alternatives, none have passed muster. We couldn't "see" air for quite some time - did that mean it wasn't there? Would a theory positing its existence be a crutch?

No modified theory of gravity has come to close to explaining things like galactic mergers, dwarf galaxies with huge dark matter content, or the galaxies without dark matter content.

You have a situation where we can see gravities and measure their mass. We can figure out the non-dark matter content of galaxies. And galaxies react in a way that really only makes sense if there's additional matter we can't detect - what theory of gravity will explain why two galaxies with similar amount of observable stuff in them have massively different masses? Is the most logical explanation not that there is more stuff in them that we can't see, representing the additional mass? If I have two boxes and they look to be the same size, but when I go to pick them up, one is significantly heavier than the other, the logical explanation is there is more mass inside the other box, despite being unable to see it.

>If I were to take this approach into another field that has an equal amount of controversy where there is a huge gap in my proof, and say that the cause is an "unseen force" that I just invented to explain how my theory works despite the fact that I can't prove in any way that the key part of my theory exists, would I not be laughed out of a room? How did dark matter become accepted while being invented instead of observed?

It's really disingenuous to act like people did a bunch of math, it didn't work as expected, so they just made up some stuff to make it work. The fact of the matter is, you would be laughed out of the room if it was that simple. There has been a massive amount of work done to attempt to prove and disprove dark matter, and there's a reason that it's the most commonly accepted explanation for things. There's also a reason it's one of the theories that people have spent the most time attempting to come up with alternative explanations - and none of them to this day have been able to explain things.

Dark matter might not exist. But it's the best explanation we have at current. People should keep trying to come up with alternative explanations, and continue trying to prove its existence. That's how science works. But you can't expect an incomplete explanation that fails to account for a wide variety of other factors to dethrone one that does account for all of those factors.